ECOSYSTEM ECOLOGY
This page intentionally left blank
ECOSYSTEM ECOLOGY
Editor-in-Chief
SVEN ERIK JØRGENSEN
Copenhagen University,
Faculty of Pharmaceutical Sciences,
Institute A,
Section of Environmental Chemistry, Toxicology and Ecotoxicology,
University Park 2,
Copenhagen Ø, 2100,
Denmark
AMSTERDAM BOSTON HEIDELBERG LONDON NEW YORK OXFORD
PARIS SAN DIEGO SAN FRANCISCO SINGAPORE SYDNEY TOKYO
Elsevier B.V.
Radarweg 29, 1043 NX Amsterdam, The Netherlands
First edition 2009
Copyright Ó 2009 Elsevier B.V. All rights reserved
The following article is US government works in the public domain and is not subject to copyright:
SWAMPS
No part of this publication may be reproduced, stored in a retrieval system
or transmitted in any form or by any means electronic, mechanical, photocopying,
recording or otherwise without the prior written permission of the publisher.
Permissions may be sought directly from Elsevier’s Science & Technology Rights
Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333;
email: permissions@elsevier.com. Alternatively you can submit your request online by
visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting
Obtaining permission to use Elsevier material
Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons
or property as a matter of products liability, negligence or otherwise, or from any use
or operation of any methods, products, instructions or ideas contained in the material herein.
Because of rapid advances in the medical sciences, in particular, independent
verification of diagnoses and drug dosages should be made.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Catalog Number: 2009929111
ISBN: 978 0 444 53466 8
For information on all Elsevier publications
visit our website at books.elsevier.com
Printed and bound in Italy
09 10 11 12 13 10 9 8 7 6 5 4 3 2 1
CONTENTS
v-vi
Contents
vii-ix
Contributors
xi
Preface
ECOSYSTEMS AS SYSTEMS
INTRODUCTION
ECOSYSTEM ECOLOGY
6
B D Fath
ECOLOGICAL SYSTEMS THINKING
ECOSYSTEMS
3
S E Jørgensen
12
D W Orr
16
A K Salomon
ECOSYSTEM SERVICES
26
K A Brauman and G C Daily
FUNDAMENTAL LAWS IN ECOLOGY
33
S E Jørgensen
ECOSYSTEM PROPERTIES
AUTOCATALYSIS
41
R E Ulanowicz
BODY-SIZE PATTERNS
44
A Basset and L Sabetta
CYCLING AND CYCLING INDICES
50
S Allesina
ECOLOGICAL NETWORK ANALYSIS, ASCENDENCY
U M Scharler
ECOLOGICAL NETWORK ANALYSIS, ENERGY ANALYSIS
ECOLOGICAL NETWORK ANALYSIS, ENVIRON ANALYSIS
INDIRECT EFFECTS IN ECOLOGY
EMERGENT PROPERTIES
SELF-ORGANIZATION
B D Fath
98
106
J L Casti and B Fath
GOAL FUNCTIONS AND ORIENTORS
76
91
D G Green, S Sadedin, and T G Leishman
HIERARCHY THEORY IN ECOLOGY
64
81
V Krivtsov
F Müller and S N Nielsen
ECOLOGICAL COMPLEXITY
EXERGY
R A Herendeen
57
T F H Allen
114
H Bossel
120
128
S E Jørgensen
OVERVIEW OF ECOSYSTEM TYPES, THEIR FORCING FUNCTIONS, AND MOST IMPORTANT PROPERTIES
S E Jørgensen
140
ECOSYSTEMS
AGRICULTURE SYSTEMS
145
O Andrén and T Kätterer
ALPINE ECOSYSTEMS AND THE HIGH-ELEVATION TREELINE
C Körner
150
v
vi
Contents
ALPINE FOREST
BIOLOGICAL WASTEWATER TREATMENT SYSTEMS
BOREAL FOREST
181
183
M Soderstrom
190
F G Howarth
CHAPARRAL
195
J E Keeley
CORAL REEFS
DESERTS
201
D E Burkepile and M E Hay
DESERT STREAMS
DUNES
166
M Pell and A Wörman
D L DeAngelis
BOTANICAL GARDENS
CAVES
156
W K Smith, D M Johnson, and K Reinhardt
214
T K Harms, R A Sponseller, and N B Grimm
222
C Holzapfel
P Moreno-Casasola
241
R F Dame
247
ESTUARIES
FLOODPLAINS
253
B G Lockaby, W H Conner, and J Mitchell
FOREST PLANTATIONS
FRESHWATER LAKES
270
S E Jørgensen
FRESHWATER MARSHES
274
P Keddy
GREENHOUSES, MICROCOSMS, AND MESOCOSMS
LAGOONS
264
D Zhang and J Stanturf
281
W H Adey and P C Kangas
296
G Harris
LANDFILLS
303
L M Chu
MANGROVE WETLANDS
MEDITERRANEAN
PEATLANDS
308
R R Twilley
319
F Médail
330
D H Vitt
POLAR TERRESTRIAL ECOLOGY
RIPARIAN WETLANDS
339
T V Callaghan
342
K M Wantzen and W J Junk
RIVERS AND STREAMS: ECOSYSTEM DYNAMICS AND INTEGRATING PARADIGMS
RIVERS AND STREAMS: PHYSICAL SETTING AND ADAPTED BIOTA
K W Cummins and M A Wilzbach
M A Wilzbach and K W Cummins
351
363
ROCKY INTERTIDAL ZONE
P S Petraitis, J A D Fisher, and S Dudgeon
374
SALINE AND SODA LAKES
J M Melack
380
SALT MARSHES
SAVANNA
J B Zedler, C L Bonin, D J Larkin, and A Varty
L B Hutley and S A Setterfield
STEPPES AND PRAIRIES
SWAMPS
W S Currie and K M Bergen
TEMPORARY WATERS
TROPICAL RAINFOREST
UPWELLING ECOSYSTEMS
WIND SHELTERBELTS
INDEX
405
417
E A Colburn
427
R B Waide
439
443
R Harmsen
URBAN SYSTEMS
394
414
C Trettin
TEMPERATE FOREST
TUNDRA
J M Briggs, A K Knapp, and S L Collins
384
T R Anderson and M I Lucas
T Elmqvist, C Alfsen, and J Colding
J-J Zhu
450
461
468
479
LIST OF CONTRIBUTORS
W H Adey
Smithsonian Institution, Washington, DC, USA
C Alfsen
UNESCO, New York, NY, USA
T F H Allen
University of Wisconsin, Madison, WI, USA
S Allesina
University of Michigan, Ann Arbor, MI, USA
T R Anderson
National Oceanography Centre, Southampton, UK
L M Chu
The Chinese University of Hong Kong, Hong Kong SAR,
People’s Republic of China
E A Colburn
Harvard University, Petersham, MA, USA
J Colding
Royal Swedish Academy of Sciences, Stockholm,
Sweden
S L Collins
University of New Mexico, Albuquerque, NM, USA
O Andrén
TSBF-CIAT, Nairobi, Kenya
W H Conner
Baruch Institute of Coastal Ecology and Forest Science,
Georgetown, SC, USA
A Basset
Università del Salento – Lecce, Lecce, Italy
K W Cummins
Humboldt State University, Arcata, CA, USA
K M Bergen
University of Michigan, Ann Arbor, MI, USA
W S Currie
University of Michigan, Ann Arbor, MI, USA
C L Bonin
University of Wisconsin, Madison, WI, USA
G C Daily
Stanford University, Stanford, CA, USA
H Bossel
University of Kassel (retd.), Zierenberg, Germany
R F Dame
Charleston, SC, USA
K A Brauman
Stanford University, Stanford, CA, USA
D L DeAngelis
University of Miami, Coral Gables, FL, USA
J M Briggs
Arizona State University, Tempe, AZ, USA
S Dudgeon
California State University, Northridge, CA, USA
D E Burkepile
Georgia Institute of Technology, Atlanta, GA, USA
T Elmqvist
Stockholm University, Stockholm, Sweden
T V Callaghan
Royal Swedish Academy of Sciences Abisko Scientific
Research Station, Abisko, Sweden
B D Fath
Towson University, Towson, MD, USA and
International Institute for Applied System Analysis,
Laxenburg, Austria
J L Casti
International Institute for Applied System Analysis,
Laxenburg, Austria
J A D Fisher
University of Pennsylvania, Philadelphia, PA, USA
vii
viii
List of Contributors
D G Green
Monash University, Clayton, VIC, Australia
D J Larkin
University of Wisconsin, Madison, WI, USA
N B Grimm
Arizona State University, Tempe, AZ, USA
T G Leishman
Monash University, Clayton, VIC, Australia
R Harmsen
Queen’s University, Kingston, ON, Canada
B G Lockaby
Auburn University, Auburn, AL, USA
T K Harms
Arizona State University, Tempe, AZ, USA
M I Lucas
National Oceanography Centre, Southampton, UK
G Harris
University of Tasmania, Hobart, TAS, Australia
M E Hay
Georgia Institute of Technology, Atlanta, GA, USA
R A Herendeen
University of Vermont, Burlington, VT, USA
F Médail
IMEP Aix-Marseille University, Aix-en-Provence,
France
J M Melack
University of California, Santa Barbara, Santa Barbara,
CA, USA
C Holzapfel
Rutgers University, Newark, NJ, USA
J Mitchell
Auburn University, Auburn, AL, USA
F G Howarth
Bishop Museum, Honolulu, HI, USA
P Moreno-Casasola
Institute of Ecology AC, Xalapa, Mexico
L B Hutley
Charles Darwin University, Darwin, NT, Australia
F Müller
University of Kiel, Kiel, Germany
D M Johnson
USDA Forest Service, Corvallis, OR, USA
S N Nielsen
Danmarks Farmaceutiske Universitet, Copenhagen,
Denmark
S E Jørgensen
Copenhagen University, Copenhagen, Denmark
W J Junk
Max Planck Institute for Limnology, Plön, Germany
P C Kangas
University of Maryland, College Park, MD, USA
T Kätterer
Department of Soil Sciences, Uppsala, Sweden
P Keddy
Southeastern Louisiana University, Hammond, LA, USA
D W Orr
Oberlin College, Oberlin, OH, USA
M Pell
Swedish University of Agricultural Sciences, Uppsala,
Sweden
P S Petraitis
University of Pennsylvania, Philadelphia, PA, USA
K Reinhardt
Wake Forest University, Winston-Salem, NC, USA
J E Keeley
University of California, Los Angeles, CA, USA
L Sabetta
Università del Salento – Lecce, Lecce, Italy
A K Knapp
Colorado State University, Fort Collins, CO, USA
S Sadedin,
Monash University, Clayton, VIC, Australia
V Krivtsov
University of Edinburgh, Edinburgh, UK
A K Salomon
University of California, Santa Barbara, Santa Barbara,
CA, USA
C Körner
Botanisches Institut der Universität Basel, Basel,
Switzerland
U M Scharler
University of KwaZulu-Natal, Durban, South Africa
List of Contributors
S A Setterfield
Charles Darwin University, Darwin, NT, Australia
D H Vitt
Southern Illinois University, Carbondale, IL, USA
W K Smith
Wake Forest University, Winston-Salem, NC, USA
R B Waide
University of New Mexico, Albuquerque, NM, USA
M Soderstrom
Montreal, QC, Canada
R A Sponseller
Arizona State University, Tempe, AZ, USA
J Stanturf
Center for Forest Disturbance Science,
Athens, GA, USA
C Trettin
USDA, Forest Service, Charleston, SC, USA
R R Twilley
Louisiana State University, Baton Rouge, LA, USA
R E Ulanowicz
University of Maryland Center for Environmental Science,
Solomons, MD, USA
A Varty
University of Wisconsin, Madison, WI, USA
K M Wantzen
University of Konstanz, Konstanz, Germany
M A Wilzbach
Humboldt State University, Arcata, CA, USA
A Wörman
The Royal Institute of Technology, Stockholm,
Sweden
J B Zedler
University of Wisconsin, Madison, WI, USA
D Zhang
Auburn University, Auburn, AL, USA
J-J Zhu
Institute of Applied Ecology, CAS, Shenyang, People’s
Republic of China
ix
This page intentionally left blank
PREFACE
S
ystems ecology, also called ecosystem theory, offers today a complete theory about how ecosystems are working as
systems. The theory will inevitably be improved in the coming years, when it hopefully will be used increasingly to
explain ecological observations and to facilitate environmental management including the use of ecotechnology. The
theory is, however, sufficiently developed today to be presented as a complete theory that offers a wide spectrum of
applications. Only through a wider application of the theory – or let us call what we have today propositions of a theory –
it will be possible to see the shortcomings of the present theory and propose improvement of the theory.
The book consists of three parts. The part Ecosystems as Systems emphasizes the system properties of ecosystems
including the presentation of basic scientific propositions to a theory in the chapter Fundamental Laws in Ecology, while
the part Ecosystem Properties gives a more comprehensive overview of the holistic properties of ecosystems, which of
course – not surprisingly – are rooted in the system properties and covered by the propositions. The part Ecosystems
gives an overview of different types of ecosystems, how they function due to their characteristic ecosystem properties,
and how the scientific propositions can be applied to understand and illustrate their characteristic properties.
It is my hope that this book will be utilized intensively by ecologists and system ecologists to gain a deeper
understanding of ecosystems and their function and to initiate the development of ecology toward a more theoretical
science that can explain and predict reactions of ecosystems. By such a development, it will be possible to replace many
measurements that are often expensive to perform with sound theoretical considerations.
The book is based on the presentation of
I.
II.
systems ecology as an ecological subdiscipline and
a very comprehensive overview of all types of ecosystems with many illustrations of their characteristic properties
in the recently published Encyclopedia of Ecology.
Due to an excellent work by the editor of the Ecosystem Section, Donald de Angelis, and the editor of the Systems
Ecology Section, Brian Fath, in the Encyclopedia of Ecology, it has been possible to present a comprehensive and very
informative overview of all types of ecosystems and an updated ecosystem theory. I would therefore like to thank Donald
and all the authors of ecosystem entries and Brian Fath and all the authors of systems ecology entries for their
contributions to the Encyclopedia of Ecology, which made it possible to produce this broad and up to date coverage of a
very important subdiscipline in ecology.
Sven Erik Jørgensen
Copenhagen, May 2009
xi
This page intentionally left blank
ECOSYSTEMS AS SYSTEMS
This page intentionally left blank
Introduction
S E Jørgensen, Copenhagen University, Copenhagen, Denmark
ª 2009 Elsevier B.V. All rights reserved.
According to the definition by Tansley (1935), an ecosys
tem is an integrated system composed of interacting biotic
and abiotic components. It is important in this definition
that an ecosystem is a system, which implies that it has
boundaries and that we can distinguish between the sys
tem and its environment – environment in principle
means the rest of the world beyond the boundaries of
the system. The components – biotic as well as abiotic –
are interacting, which means that they are connected
directly or indirectly. All systems that encompass inter
acting biotic and abiotic components may be considered
as an ecosystem. A drop of polluted water may for
instance be considered an ecosystem, because it contains
microorganisms, organic matter, and inorganic salts and
these components are interacting. Usually, our ecosystem
research and management is interested in a larger area of
nature characterized by its function and properties, for
instance a lake, a forest, or a wetland. All these three
examples of ecosystems have very characteristic functions
and have several unique properties that are different from
other types of ecosystems. The scale that is applied for the
definition of an ecosystem is dependent on the function of
the ecosystem and is determined by the addressed
problem.
Because an ecosystem has interacting and con
nected biotic and abiotic components, it has system
properties in the sense that the components work
together to give the system emerging properties and
make the system more than just the sum of the com
ponents. A living organism is much more than the
cells and the organs that make up the organism.
Similarly, a forest is more than just the trees – it is
a cooperative working unit with emerging unique
properties characteristic of a forest.
It is important to understand fully the function and the
reactions of ecosystems in both ecological research and
environmental management. The two basic questions in
this context are
1. Which
fundamental
properties
characterize
ecosystems?
2. Is it possible to formulate basic scientific propositions
that are able to explain the functions of ecosystems?
It is attempted to answer these two core questions in the
parts Ecosystems as Systems and Ecosystem Properties of
this book, while the part Ecosystems gives an overview of
different types of ecosystems, how they function due to
their characteristic ecosystem properties, and how the
scientific propositions can be applied to understand and
illustrate their characteristic properties. The part
Ecosystems as Systems emphasizes the system properties
of ecosystems and also presents basic scientific proposi
tions, while the part Ecosystem Properties gives a more
comprehensive overview of the holistic properties of
ecosystems, which of course – not surprisingly – are
rooted in the system properties.
The chapters Ecosystem Ecology, Ecological System
Thinking, and Ecosystems in the part Ecosystems as
Systems focus on the most fundamental system properties
that are derived from the above presented definition of
ecosystems. The definition is repeated in all three chap
ters with slight modifications. The system properties
presented in these three chapters may be summarized as
follows:
1. Ecosystems cycle energy.
2. Ecosystems cycle matter.
3. Life and environment are connected, which implies
that the environment of an ecosystem influences the
ecosystem. This influence determines the prevailing
conditions of the ecosystems, or expressed differently
the external variables (also called forcing functions)
determine the conditions for the internal variables
(also called state variables) of an ecosystem. The
wide spectrum of different ecosystems (the part
Ecosystems gives an overview) is the result of an over
whelmingly large number of different conditions
(combinations of external variables).
4. Ecosystems are whole systems and studies of ecosys
tem dynamics therefore require holistic views.
The human society is very dependent on the proper
functioning of ecosystems, because humans are using a
wide spectrum of services offered by the ecosystems. It is
therefore important to understand the ecosystem proper
ties on which these services are based. The chapter
Ecosystem Services and partly the chapter Ecosystems
present the ecosystem services, which may be classified
into three groups:
– production services as we know them from agriculture,
fishery, forestry, and so on;
– regulation services due to cycling, filtration, transloca
tion, and stabilization processes;
– cultural services such as recreation, spiritual inspiration,
and esthetic beauty.
3
4 Introduction
The chapter Fundamental Laws in Ecology gives a brief
summary of the ecosystem properties that are rooted in
the system properties of ecosystems:
– Ecosystems are complex (many steadily varying inter
acting components).
– Ecosystems are open.
– Ecosystems are hierarchically organized.
– Ecosystems are self organizing and self regulated due
to a very large number of feedback mechanisms.
These properties are discussed in more detail in the part
Ecosystem Properties.
The chapter Fundamental Laws in Ecology proposes 10
fundamental laws of ecosystems that are consistent with the
system properties presented in the other chapters of the
part Ecosystems as Systems. The 10 propositions are able
to explain ecosystem behavior and properties. The funda
mental tentative laws presented in this chapter are
furthermore able to explain many ecological observations
and rules, which is a great advantage of having a good
theory. By use of the theory, it is possible to conclude,
without the need for observations, how an ecosystem will
react to different impacts. It is therefore indeed possible to
improve research plans and develop environmental man
agement plans on the basis of theoretical considerations.
The 10 propositions (tentative laws) can be shown to be
rooted in five basic ecological system properties.
The part Ecosystem Properties gives more information
on the basic properties of an ecosystem. The chapter
Autocatalysis focuses on autocatalysis, which frequently
increases the efficiencies and rates of ecological processes.
The chapter Body Size Patterns discusses the body size
pattern of ecosystems. The rate of biological processes
such as growth, metabolism, mortality, generation time,
and respiration is dependent on the size of the organisms.
The spectrum of conditions in an ecosystem determines the
spectrum of these fundamental ecological processes, which
would allow the best utilization of the resources in ecosys
tems. It implies that the conditions also determine the body
size pattern. Different ecosystems at different conditions
may therefore have a different body size pattern, which
therefore becomes a characteristic property of an ecosystem.
All ecosystems cycle the elements that are essential for
the living matter, and thereby the growth and development
of ecosystems can continue, because the essential elements
are steadily recovered with a certain rate. The living mat
ter needs about 22 different elements, of which the cycling
of nitrogen, carbon, phosphorus, sulfur, silica, calcium,
sodium, and magnesium is of utmost importance. The
cycling is possible due to the ecological networks that are
formed in all ecosystems. The network may be considered
a ‘map’ of the connections of abiotic and biotic components.
The network indicates the possibilities for interactions
among the components of the ecosystem. Obviously,
cycling is very important for ecosystems, because without
cycling the growth and development of biological compo
nents would stop due to the lack of one or more essential
elements. The chapter Cycling and Cycling Indices covers
cycling and cycling indices, which quantify the network’s
possibilities to support the cycling processes.
The chapters Ecological Network Analysis, Ascendancy;
Ecological Network Analysis, Energy Analysis; Ecological
Network Analysis, Environ Analysis; and Indirect Effects in
Ecology present different aspects of the ecological network.
Network analysis, ENA (Ecological Network Analysis), uses
network theory to study the interactions between organisms
or populations within their environment. Ascendancy,
which is covered in the chapter Ecological Network
Analysis, Ascendancy, quantifies the efficiency of the net
works on the basis of the actual flows. Development of an
ecosystem will usually imply that the ascendancy is increas
ing. The chapter Ecological Network Analysis, Energy
Analysis analyzes the ecological network by use of the
energy flows, while the chapter Indirect Effects in Ecology
uses the so called environ analysis. Each object in the system
has two ‘environs’, one receiving and one generating inter
actions in the system. It is by analyzing these flows that it is
possible to deduce network properties such as network
mutualism and network synergy. Cycling – the topic of the
chapter Cycling and Cycling Indices – may of course also be
considered a network property. The chapter Indirect Effects
in Ecology focuses on perhaps the most important network
property: the presence of a strong indirect effect that in many
cases may even exceed the direct effect.
The chapter Emergent Properties deals with the topic of
emergent properties – the ecosystem as an integrated sys
tem is more than the sum of the components. The emergent
properties are the result of all the system properties. Due to
the synergistic effect of the network, autocatalysis, cycling,
self regulation and self organization, and so on, an ecosys
tem acquires a number of very useful, holistic properties as a
system – properties that are often called emergent proper
ties. Self organization itself is perhaps the most clear
example of an emergent property. The chapter Self orga
nization looks into the emergent property of self
organization and how it is rooted in complex adaptive
ecosystems. This chapter discusses how the spatial patterns,
persistence, stability, and ability to develop and evolve can
be explained as a result of the self organization. The differ
ences between ecosystems at an early stage and mature
ecosystems can also be explained by self organization.
Ecosystems are very complex systems. They have a
large number of components with a large diversity, hier
archical organization, and nonlinear behavior. The
chapter Ecological Complexity presents various aspects
of ecological complexity, while the chapter Hierarchy
Theory in Ecology presents the application of hierarchy
theory in ecology. The hierarchical organization makes it
possible to overview the complexity. It is also possible to get
a better overview of the complex behavior of ecosystems by
Introduction
5
Table 1 The five basic properties that are rooted in the 10 tentative fundamental laws encompass all the system properties presented
Basic property
Derived system properties
1. Ecosystems are open
2. Ecosystems have directionality
The forcing functions (external variables) determine the ecosystem conditions
Ecosystems show autocatalysis
Ecosystems grow and develop
Ecosystems have the propensity to maximize exergy storage and power
Ecosystems have a body size pattern
The biotic and abiotic components of an ecosystem are connected in a network
The network gives the ecosystem mutualism and synergy
The indirect effect is significant due to the network and may even exceed the direct
effect
Ecosystems are self-organizing and self-regulated
Ecosystems cycle energy, matter, and information
Ecosystems are organized hierarchically
Ecosystems grow and develop by increasing the biomass, the network, and the level of
information
Ecosystems are adaptive systems
Ecosystems grow and develop and can cope with disturbances by a propensity to
increase the exergy storage and the power
Ecosystems, particularly under natural conditions, often have a large diversity, which
gives the ecosystems a wide spectrum of different buffer capacities
Ecosystems have high buffer capacities as a result of the complex dynamics
Ecosystems recover usually rapidly and effectively after disturbances
3. Ecosystems have connectivity
4. Ecosystems have emergent hierarchies
5. Ecosystems have complex dynamics
use of goal functions and orientors that are presented in the
chapter Goal Functions and Orientors. They are able to
quantify the development of ecosystems as a result of the
complex dynamics of ecosystems. One of the most useful
orientors is exergy, which is presented in the chapter
Exergy. The complex dynamics of ecosystems determine
how they are able to develop and cope with disturbances.
The exergy or energy that can do work of ecosystems – we
cannot calculate exergy for an ecosystem due to its enor
mous complexity but we can calculate exergy for a model of
the ecosystem – will have the tendency to be as high as
possible under the prevailing conditions. Disturbances may
of course cause a reduction in the ecosystem exergy, but the
organisms try to organize themselves by their network and
interactions to get the best out of the situation – it means in
the Darwinian sense most survival, which may be expressed
by exergy, as it covers the product of biomass and informa
tion of the ecosystem.
The five fundamental properties (see chapter
Fundamental Laws in Ecology) cover all the ecosystem
properties that are presented in the parts Ecosystems as
Systems and Ecosystem Properties. An overview of the
five basic properties and the derived additional system
properties can be obtained from Table 1. Some of the
properties are derived from more than one of the five
fundamental properties, but to simplify the overview the
derived system properties are associated with one of the
basic properties. Particularly, the basic property that eco
systems have connectivity, which means that they form a
network, and have a complex dynamics has been used to
derive several system properties that could also be derived
partly from one of the four other basic properties.
The chapter Overview of Ecosystem Types, Their
Forcing Functions, and Most Important Properties, which
is the last chapter in the part Ecosystem Properties, gives an
overview of the 39 different types of ecosystems that are
presented in the part Ecosystems. For all the 39 ecosystem
types, the most important forcing functions are indicated,
that is, the forcing functions (impacts) that may be consid
ered a threat to the ecosystem or the forcing functions that
most frequently determine the ecosystem function. It is
possible to classify the forcing functions of the 39 ecosystems
into four groups. The most basic properties of the four
ecosystem classes are presented. They are the result of the
prevailing conditions that are determined by the forcing
functions. The most important properties are those that
need to be maintained for the ecosystem to be able to meet
the threats or those that are particularly important for the
maintenance of the ecosystem function in spite of the impact.
The part Ecosystems has 40 chapters covering 39
different types of ecosystems. Most of the Earth’s ecosys
tems are covered by the 39 types of ecosystems. A few
rare types of ecosystems are not included, but all ecosys
tems frequently represented in nature are included. The
ecosystems that are not included will however have prop
erties close to one or more of the 39 types covered.
See also: Autocatalysis; Body Size Patterns; Cycling and
Cycling Indices; Ecological Complexity; Ecological
Network Analysis, Ascendancy; Ecological Network
Analysis, Energy Analysis; Ecological Network Analysis,
Environ Analysis; Ecosystem Ecology; Ecosystem
Services; Ecological System Thinking; Ecosystems;
Emergent Properties; Exergy; Fundamental Laws in
6 Ecosystem Ecology
Ecology; Goal Functions and Orientors; Hierarchy Theory
in Ecology; Indirect Effects in Ecology; Overview of
Ecosystem Types, Their Forcing Functions, and Most
Important Properties; Self-Organization.
Further Reading
Jørgensen SE (2004) Information theory and energy. In: Cleveland CJ (ed.)
Encyclopedia of Energy, vol. 3. pp. 439 449. San Diego, CA: Elsevier.
Jørgensen SE (2006) Eco Exergy as Sustainability. 220pp.
Southampton: WIT Press.
Jørgensen SE (2008b) Evolutionary Essays. A Thermodynamic
Interpretation of the Evolution, 210pp.
Jørgensen SE (ed.) (2008a) Encyclopedia of Ecology, 5 vols. 4122pp,
Amsterdam: Elsevier.
Jørgensen SE and Fath B (2007) A New Ecology. Systems
Perspectives. 275pp. Amsterdam: Elsevier.
Jørgensen SE, Patten BC, and Straskraba M (2000) Ecosystems
emerging: 4. growth. Ecological Modelling 126: 249 284.
Jørgensen SE and Svirezhev YM (2004) Towards a Thermodynamic
Theory for Ecological Systems. 366pp. Amsterdam: Elsevier.
Ulanowicz R, Jørgensen SE, and Fath BD (2006) Exergy, information
and aggradation: An ecosystem reconciliation. Ecological Modelling
198: 520 525.
Ecosystem Ecology
B D Fath, Towson University, Towson, MD, USA and International Institute for Applied System Analysis,
Laxenburg, Austria
ª 2008 Elsevier B.V. All rights reserved.
Introduction
History of the Ecosystem Concept
Defining an Ecosystem
Energy Flow in Ecosystems
Biogeochemical Cycles
Ecosystem Studies
Human Influence on Ecosystems
Summary
Further Reading
Introduction
abstracting to energetic or material units. The advantage
of this abstraction, of course, is that energy and mass are
conserved quantities, whereas number of individuals is
not. Therefore, using conserved units it is possible to
construct balance equations and input–output models.
In fact, dimensionally, ecosystem ecology has more in
common with organismal ecology in which the thermo
regulation and physiology of a single organism is
studied, which also often relies on energetic units.
Indeed, all scales of ecological study have a role to
contribute to general scientific understanding and have
been developed to address a wide range of interesting
and relevant questions regarding the natural world and
the impact humans have on it.
Ecology is a broad and diverse field of study. One of the
basic distinctions in ecology is between autecology and
synecology, in which the former is considered the ecology
of individual organisms and populations, mostly concerned
with the biological organisms themselves; and the latter, the
ecology of relationships among the organisms and popula
tions, which is mostly concerned with communication of
material, energy, and information of the entire system of
components. In order to study an ecosystem, one must have
knowledge of the individual parts; thus, it is dependent on
fieldwork and experiments grounded in autecology, but the
focus is much more on how these parts interact, relate to,
and influence one another including the physical environ
mental resources on which life depends. Ecosystem
ecology, therefore, is the implementation of synecology.
In this manner, the dimensional units used in ecosystem
studies are usually the amount of energy or matter moving
through the system. This differs from population and com
munity ecology studies in which the dimensional units are
typically the number of individuals (Table 1). This simple
dimensional difference has served as an unfortunate divide
between research conducted at the different ecological
scales. While ecosystem ecologists maintain that it is always
possible to convert species numbers into biomass or nutri
ent mass, population and community ecologists often feel
that too much unique biological detail is discarded by
History of the Ecosystem Concept
Systems concepts of the environment have long played a
role in the development of ecology as a discipline, but
these came to a head in the early twentieth century.
During this period, the two dominant and competing
ecological paradigms were the organismic (e.g.,
Clements) and individualistic (e.g., Gleason) views. The
organismic approach held that communities and ecosys
tems were discernible objects that had an inherent and
organized complexity resulting in a cybernetic and self
governing system, similar in ways to how an organism
Ecosystem Ecology
Table 1 Typical dimensional units of study at different
ecological scales
Ecological scale
Dimensions
Organismal ecology
Population ecology
Community ecology
Ecosystem ecology
dE/dt
dN/dt
dN/dt
dE/dt
dE/dt ¼ change in energy over time; dN/dt ¼ change in number over time.
regulates itself. The individualistic approach held that
communities had observer dependent boundaries and
internal development was stochastic and individual. In
this paradigm, the internal relations were synergistic, but
not cybernetic since the individual parts functioned inde
pendently. The organismic ideas grew out of the
functional understanding of whole systems such as lakes,
and also out of the discussions involving how communities
changed over time during succession. These ideas were
influenced by philosophers of the day such as Jan Smuts.
This was particularly true of German holists, such as the
limnology group at the Kaiser Wilhelm Instituts in Plön
led by Thienemann, and others such as Leick (plant ecol
ogy) and Friedrich (zoology). Table 2 shows a summary of
some of the main ecosystem and related concepts. This
dialog between the holists and reductionists affected the
main currents of ecological thought during this period, and
it was in part resolved by the introduction of ‘ecosystem’,
which is both physical in nature and also systemic.
The term ecosystem, which is ubiquitous today, both as
scientific terminology and in common vernacular, grew out
of this climate. It was first used by Arthur Tansley in 1935 in
a seminal paper in the journal Ecology, entitled ‘The use and
abuse of vegetational concepts and terms’. In fact, his reason
for coining the term ‘ecosystem’ was in response, as the title
says, to a perceived abuse of community concepts by some
such as Clements and Cowles. While Tansley himself
brought a systems perspective, the community as organism
metaphor bothered him to the extent that he wanted to
provide a more scientific footing for the processes and inter
actions occurring during community development. Tansley
7
describes the ecosystem thus, ‘‘. . . the fundamental concep
tion is . . . the whole system, including not only the
organism complex, but also the whole complex of physical
factors forming what we call the environment of the biome –
the habitat factors in the widest sense.’’ The definition he
proposed over 70 years ago sounds fresh today, since it has
changed little if at all. The major tenets of this approach are
the explicit inclusion of abiotic processes interacting with
the biota – in this sense it is more along the Haeckelian lines
of ecology than the Darwinian, with an additional emphasis
on the system. The latter tied the field closely to the bur
geoning disciplines of general system theory and systems
analysis.
While the conceptual underpinning of the ecosystem was
now established, the introduction of this term was theoreti
cal, lacking guidance as to how it might be applied as a field
of study. There were around this time several whole system
energy budgets being developed, particularly for lake eco
systems by North American ecologists such as Forbes, Birge,
and Juday in Wisconsin, and which were ideal test cases for
the ecosystem concept. Building on this work, in 1942,
Lindeman’s study of Cedar Bog Lake also in Wisconsin
was published, providing, for the first time, a clear applica
tion of the ecosystem concept. In addition to constructing the
food cycle of the aquatic system, he developed a metric –
now called the Lindeman efficiency – to assess the efficiency
of energy movement from one trophic level to the next based
on ecological feeding relations. His conceptual model of
Cedar Bog Lake included passive flows to detritus, but
these were not included in the trophic enumeration. Since
then numerous additional studies have followed this same
approach and it has been applied to many habitats such as
terrestrial, aquatic, and urban ecosystems.
Defining an Ecosystem
An ecosystem, as a unit of study, must be a bounded
system, yet the scale can range from a puddle, to a lake,
to a watershed, to a biome. Indeed, ecosystem scale is
defined more by the functioning of the system than by any
checklist of constituent parts, and the scale of analysis
Table 2 Ecosystem and related concept
Year
Term
Author
Concept
1887
1914
1928
1930
1935
1939
1944
1944
1948
1950
Microcosm
Ecoid
Ökologisches system
Holocoen
Ecosystem
Biosystem
Geobiocönose
Bioinert body
Biochore
Landschaft
Forbes
Negri
Woltereck
Friedrich
Tansley
Thienemann
Sukacev
Vernadsky
Pallmann
Troll
Broadening of the biocoenosis concept
Unholistic, based on Gleasonian ideas
Still being used to avoid argument
Holistic, biologistic
Antiholistic, physicalist
Stressing functional organization
Geographic, landscape ecological
Biogeochemical
Landscape ecological
Holistic, ‘Gestalt’ viewing
Modified from Wiegleb G (2000) Lecture Notes on The History of Ecology and Nature Conservation.
8 Ecosystem Ecology
Photosynthesis
Ecosystem boundary
Respiration
Plant
respiration
Herbivore
Carnivore
Primary
production
system
Respiration
Detritus and
decomposers
Input transfers
Immigration/
emigration
Available
nutrients
Soil surface
Output transfers
Leaching
Figure 1 Conceptual diagram of a simplified ecosystem. Clear arrows, energy; dark arrows, biomass; blue arrows, water.
should be determined by the problem being addressed.
Whereas individuals perish over time and even popula
tions cannot survive indefinitely – none can fix their own
energy and process their own wastes – every ecosystem
contains the ecological community necessary for sustain
ing life: primary producers, consumers, and decomposers,
and the physical environment for oikos (Figure 1 shows a
simple ecosystem model). It is this feature of ecosystems,
that they are the basic unit for sustaining life over the
long term, which provides one of the main reasons for
studying them for environmental management and con
servation. The two main features of the ecosystem, energy
flow and nutrient biogeochemical cycling, comprise the
major areas of ecosystem ecology research.
Energy Flow in Ecosystems
The thermodynamic assessment of an ecosystem starts
with the recognition that an ecosystem is an open system,
in the sense of physics, such that it receives energy and
matter input from outside its borders and transfers output
back to this environment. Thus, every ecosystem must
have a system boundary and must be embedded in an
environment that provides low entropy energy input and
can receive high entropy energy output. In addition to
the external resource source–sink, there is another inter
nal, within system boundary environment with which
each organism directly and indirectly interacts. Patten
proposed the concept of these two environments, one
external and mostly unknowable (other than the input–
output interactions), and the second internal and measur
able (i.e., external to the specific organismal component
but within system boundary) as a systems approach to
quantify indirect, yet within system interactions. This
approach – called environ analysis – relying on the
methodologies of input–output analysis has developed
into a powerful analysis tool for understanding complex
interactions and dependencies in ecological networks. For
now though, let us concern ourselves more generally with
what occurs within the ecosystem boundary.
Energy flow in ecosystems begins with the capture of
solar radiation by photosynthetic processes in primary
producers (eqn [1]). Note, there are also chemoautotrophs
that capture energy in the absence of sunlight, but while
biologically fascinating, contribute negligible energy flux
to the overall global ecological energy balance
Energy þ 6CO2 þ 6H2 O ! C6 H12 O6 þ 6O2
½1
The accumulated organic matter, first as simple sugars
then combined with other elements to more complex
molecules, represents the gross primary production in the
system, some of which is released and used for the primary
producers’ growth and maintenance through respiration:
C6 H12 O6 þ 6O2 ! 6CO2 þ 6H2 O þ Energy
½2
The remainder, or net primary production, is available
for the rest of the ecosystem consumers including decom
posers. Secondary production refers to the energetic
availability of the heterotrophic organisms, which
accounts for the energy uptake by heterotrophs and the
energy used for their maintenance. Overall ecosystem
production is supported by the primary producers,
whereas ecosystem respiration includes the metabolic
activity of all the ecosystem biota (Table 3). In this
manner, plants provide the essential base for all ecological
food webs. Since it is often difficult to make direct mea
surements of ecological production, the change in
biomass measures growth, which can be used as represen
tative of production.
The captured energy moves through a reticulated net
work of interactions forming the complex dependency
Ecosystem Ecology
9
Table 3 Ecosystem energetics defined by net and gross production
Net primary production gross primary production respiration (autotrophs)
Net secondary production gross secondary production respiration (heterotrophs)
Net ecosystem production gross primary production ecosystem respiration (autotrophs þ heterotrophs)
Net production biomass (now) biomass (before)
patterns known as food webs. In a simplified food chain,
and as first described by Lindeman, the trophic concept is
used to assess the distance away from the original energy
importation, but in reality the multiple feeding pathways
found in ecological food webs make discrete trophic
levels a convenient yet inaccurate simplification. Elton
observed that one typically finds a decreasing number of
organisms as one proceeds up the food chain from pri
mary producers to herbivores, carnivores, and top
carnivores – leading him to propose a pyramid of num
bers. One can control for the individual variation in body
size by considering the biomass at each trophic level
rather than the number of individuals – resulting in a
pyramid of biomass. The trophic pyramid is a thermo
dynamically satisfying view of interactions since
according to the second law energy must be lost during
each transformation step; in addition, energy is used at
each level for the maintenance of that level. Under this
paradigm, the trophic levels apparently cap out around
five or six levels. Fractional trophic levels have been
employed to account for organisms feeding at multiple
levels, but even these do not usually account for the role
of detritus and decomposition, which extend the feeding
pathways to higher numbers. However, instead of linking
detritus as a source compartment in the ecosystem con
ceptual model, the standard paradigm is to envision two
parallel food webs one with primary producers as the base
and the other with detritus as the base without any input
from the rest of the web. If detritus were properly linked
as both a source and sink in the ecosystem, then it would
be clear that higher order trophic levels are possible, if
not common. The higher observed trophic levels
observed in some studies are not in conflict with the
laws of thermodynamics, but they show that ecosystems
are more thorough at utilizing the energy within the
system, mostly by decomposers, before it is lost as
degraded, unavailable energy.
Energy resources flowing through the ecosystem are
necessary to maintain all growth and development activ
ities. Organisms follow a clear life history pattern, and
while the timescales differ depending on the species, early
stage energy availability is generally used for growth,
while later energy surplus is used for maintenance or
reproduction. A similar pattern is visible in ecosystem
level growth and development. Net primary production is
used to build biomass and physical structure of the eco
system. The additional structure of photosynthetic
material allows for the additional import of solar energy
until saturation is reached at about 80% of the available
solar radiation. At this point the overall growth of the
ecosystem begins to level off because although gross
primary production is high, the overall system supports
more and more nonphotosynthetic biomass both in terms
of nonphotosynthetic plant material and heterotrophs.
When the average gross production is entirely utilized
to support and maintain the existing structure, net pro
duction is zero and the system has reached a steady state
regarding biomass growth. However, the ecosystem con
tinues to develop both in terms of the network
organization and in the information capacity. In addition
to being a dynamic steady state, it does not persist indefi
nitely because disturbances afflict the system setting it
back to earlier successional stages in which the growth
and development processes begin anew, possibly with
different results. In this manner, the disturbance acts
according to Holling’s creative destruction providing the
system the opportunity to develop along a different path
way. Recent work on ecosystem growth and development
has focused on the orientation of thermodynamic
indicators such as energy throughflow, energy degrada
tion, exergy storage, and specific entropy. These orientors
provide good system level indicators of development
during succession or restoration of impaired ecosystems.
Biogeochemical Cycles
Another major focus of ecosystem ecology is understanding
how the chemical elements necessary for life persist and
translocate in pools and fluxes within the ecosphere. The
biosphere actively interacts with the three abiotic spheres
(hydrosphere, atmosphere, and lithosphere) to provide the
available concentration of each for life. This action has a
significant impact on the relative distribution of these ele
ments. The simple sugar products of photosynthesis,
C6H12O6, are the base for organic matter, so carbon, hydro
gen, and oxygen dominate the composition of life, and
while oxygen is available in the lithosphere, and hydrogen
in the hydrosphere, carbon is actually quite scarce in the
environment, making the disproportionate amount of car
bon in biomass a hallmark of life. In fact, there are about 20
elements used regularly in living organisms, of which nine
called the macronutrients are the major constituents of
organic matter: hydrogen, oxygen, carbon, nitrogen,
10
Ecosystem Ecology
Table 4 Percentage atomic composition of the biosphere, hydrosphere, atmosphere, and lithosphere for first 10 elements
Biosphere
H
O
C
N
Ca
K
Si
Mg
P
49.8
24.9
24.9
0.073
0.046
0.033
0.031
0.030
0.017
Hydrosphere
H
O
Cl
Na
Mg
S
Ca
K
C
65.4
33.0
0.33
0.28
0.03
0.02
0.006
0.006
0.002
calcium, potassium, silicon, magnesium, and phosphorus.
Some of these elements are readily available in the abiotic
environment, in which case conservation through cycling
of the elements is not paramount; however, those in scarce
supply, such as nitrogen and phosphorus (Table 4), are
reused many times before being released from the system.
These biogeochemical cycles provide the foundation to
understand how human modification leads to eutrophica
tion (N and P cycles) and global climate change (C cycle).
Therefore, much effort has been made to study and under
stand these cycles, particularly the carbon, nitrogen, and
phosphorus cycles, details of which are addressed else
where in this encyclopedia.
Ecosystem Studies
The ecosystem perspective achieved footing in the eco
logical academic community since it was central to E. P.
Odum’s seminal textbook Fundamentals of Ecology first pub
lished in 1953. An early implementation of this approach
at the institutional scale was attempted was in the
International Biological Program (IBP), which was run
from 1964 to 1974. The program had many successes in
assessing and surveying the Earth’s ecosystems, but faced
the difficulty of compelling a top down, holistic research
paradigm on individual scientific endeavors. As a result of
this conflict, the program did not deliver as much as had
been hoped, but set the stage for the next generation of
ecosystem scale research. One feature of the IBP that did
continue was the use of computer simulation modeling as
a tool to understand the complex ecological interrelations.
The journal Ecological Modelling and Systems Ecology started
in 1975 continues as an active repository for mathematical
and computer based ecosystem research.
Subsequent to the IBP, the US National Science
Foundation officially established the Long Term
Ecological Research Sites (LTER) in 1980 but research
at several of the sites dates much earlier. Currently, there
are 26 such sites ranging from the Coweeta Hydrological
Lab in North Carolina, Hubbard Brook Ecosystem Study
in New Hampshire, Sevilleta National Wildlife Refuge in
Atmosphere
N
O
Ar
C
Ne
78.3
21.0
0.93
0.03
0.002
Lithosphere
O
Si
Al
H
Na
Ca
Fe
Mg
K
62.5
21.22
6.47
2.92
2.64
1.94
1.92
1.84
1.42
New Mexico, to the Baltimore Urban Ecosystem Study.
These projects rely on a vast team of scientists to study
the many interactions at this spatial scale. Still, the diffi
culty lies in putting together all the pieces into an
integrated whole picture of the ecosystem.
Smaller scale, individual led ecological research is
commonly conducted using microcosm and mesocosm
experiments. A mesocosm experiment uses designed
equipment or enclosures in which environmental factors
can be controlled and manipulated to approximate
natural conditions. The prevalence of this approach
created a wealth of small scale experimentation but at
the expense of larger observational studies, which sparked
a fierce debate in the 1990s between the ‘field’ versus
‘bottle’ approach. Indeed, the usefulness of microcosm
experiments for ecosystem ecology was brought into
question, but the resolution has been that a multiplicity
of approaches is useful to address ecological questions.
Human Influence on Ecosystems
Humans have greatly altered and impacted the global
biosphere. We recognize now the importance of main
taining functioning ecosystem services both out of our
own necessity and for the obligation we have to the eco
sphere. In 2000, the United Nations Secretary General
called for a global ecological assessment, which was
recently published as the Millennium Ecosystem
Assessment (MEA) (www.mawed.org). The report com
piled by over 1350 experts from 95 countries found that
humans have changed ecosystems more rapidly and
extensively over the last 50 years than in any comparable
period of time in human history, resulting in a substantial
and largely irreversible loss in the diversity of life on
Earth (other highlights from the report are presented
in Table 5). The MEA operated within a framework
that identified four primary ecosystem services needed
by humans: supporting (nutrient cycling, primary produc
tion, soil formation, etc.), provisioning (food, water,
timber, fuel, etc.), regulating (climate, flood, disease,
etc.), and cultural (esthetic, spiritual, educational,
Ecosystem Ecology
11
Table 5 A few of the trends identified in the Millennium Ecosystem Assessment
50% of all the synthetic nitrogen fertilizer ever used has been used
since 1985
60% of the increase in the atmospheric concentration of CO2 since
1750 has taken place since 1959
Approximately 60% of the ecosystem services evaluated are being
degraded or used unsustainably
20% of the world’s coral reefs were lost and 20% degraded in
the last several decades
35% of mangrove area has been lost in the last
several decades
Withdrawals from rivers and lakes doubled since 1960
Table 6 Ecosystem Approach principles of the Convention on Biological Diversity
1
2
3
4
5
6
7
8
9
10
11
12
The objectives of land, water, and living resource management are a matter of societal choices
Management should be decentralized to the lowest appropriate level
Ecosystem managers should consider the effects (actual or potential) of their activities on adjacent and other ecosystems
Recognizing potential gains from management, there is usually a need to understand and manage the ecosystem in an economic
context. Any such ecosystem-management program should
(a) reduce those market distortions that adversely affect biological diversity;
(b) align incentives to promote biodiversity conservation and sustainable use; and
(c) internalize costs and benefits in the given ecosystem to the extent feasible
Conservation of ecosystem structure and functioning, in order to maintain ecosystem services, should be a priority target of the
ecosystem approach
Ecosystem must be managed within the limits of their functioning
The ecosystem approach should be undertaken at the appropriate spatial and temporal scales
Recognizing the varying temporal scales and lag-effects that characterize ecosystem processes, objectives for ecosystem
management should be set for the long term
Management must recognize the change is inevitable
The ecosystem approach should seek the appropriate balance between, and integration of, conservation and use of biological
diversity
The ecosystem approach should consider all forms of relevant information, including scientific and indigenous and local
knowledge, innovations, and practices.
The ecosystem approach should involve all relevant sectors of society and scientific disciplines
The 12 principles mentioned above are complementary and interlinked.
recreational, etc.). All have shown signs of stress and
human pressures during the past century. One positive
trend was the increase in food production (crops, live
stock, and aquaculture), but this occurred with a
concomitant loss of wild fisheries and food capture,
along with a substantial increase in the resource inputs
required to maintain the high agricultural production.
While these observed changes to ecosystems have con
tributed to substantial net gain in human well being and
economic development, they have come at an increasing
cost to the ecosystem health. The loss of this natural
capital is typically not properly reflected in economic
accounts.
Since the ecosystem provides the necessary functions
for life, environmental management principles being
devised and implemented today use the ecosystem con
cept as foundation. In particular, there have been several
high profile international efforts such as with the
Convention on Biological Diversity (CBD), a treaty
initiated in 1992 and signed by 150 government leaders
with the expressed aim to protect and promote biological
diversity and sustainable development. The ‘ecosystem
approach’ adopted within this convention uses scientific
methodologies regarding ecological interactions among
organisms, their environment, and human activity to pro
mote conservation, sustainability, and equity for
managing natural resources. The approach deals with
the complex socioecological–economic systems by pro
moting integrated assessment and adaptive management.
The ecosystem approach of the CBD is outlined below in
12 principles (Table 6). Note particularly principles 5–8
that deal with ecosystem functioning, and taken in the
context of the other principles assert how this ecological
functioning provides opportunities and constraints for
economic and social well being. Research in ecosystem
ecology today is directed toward improved understanding
of key issues such as ecosystem services, resilience, spatial
and functional scale, time lags, dynamics, and indirect
effects.
Summary
Ecosystem ecology deals with the functioning at the sys
tem level of the ecological community with its abiotic
environment, primarily in terms of the energy flow and
nutrient cycling. Research in ecosystem ecology has given
us a much better understanding of the processes and
12
Ecological Systems Thinking
functions necessary to sustain life. The work in natural
sciences has outpaced the ability of the social institutions
to adapt and implement this knowledge. However, there
is reason to be optimistic because the recent focus on the
ecosystem approach in major international efforts recog
nizes that humans, with their cultural diversity, are an
integral component of ecosystems.
See also: Ecological Network Analysis, Environ Analysis;
Ecosystem Services; Ecosystems; Goal Functions and
Orientors.
Further Reading
Chapin III FS, Matson PA, and Mooney HA (2002) Principles of
Terrestrial Ecosystem Ecology. New York: Springer.
Fath BD, Jørgensen SE, Patten BC, and Straškraba M (2004)
Ecosystem growth and development. Biosystems 77: 213 228.
Golley FB (1993) A History of the Ecosystem Concept in Ecology. New
Haven: Yale University Press.
Likens GE, Borman FH, Johnson NM, Fisher DW, and Pierce RS (1970)
Effects of forest cutting and herbicide treatment on nutrient budgets
in the Hubbard Brook watershed ecosystem. Ecological
Monographs 20: 23 47.
Lindeman RL (1942) The trophic dynamic aspect of ecology. Ecology
23: 399 418.
Odum EP (1969) The strategy of ecosystem development. Science
164: 262 270.
Odum HT (1957) Trophic structure and productivity of Silver Springs,
Florida. Ecological Monographs 27: 55 112.
Patten BC (1978) Systems approach to the concept of environment.
Ohio Journal of Science 78: 206 222.
Tansley AG (1935) The use and abuse of vegetational concepts and
terms. Ecology 16: 284 307.
Weigert RG and Owen DF (1971) Trophic structure, available resources
and population density in terrestrial versus aquatic ecosystems.
Journal of Theoretical Biology 30: 69 81.
Wiegleb G (2000) Lecture Notes on ‘The History of Ecology and Nature
Conservation’. http://board.erm.tu cottbus.de/index.php?id¼5&
no cache¼1&file¼33&uid¼14 (accessed May 2007).
Relevant Website
http://www.maweb.org
Millennium Ecosystem Assessment
Ecological Systems Thinking
D W Orr, Oberlin College, Oberlin, OH, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Applied Systems Thinking
Environmental Education
Summary
Further Reading
Introduction
sometimes morph into other forms and processes. In Earth
systems, small changes can have large effects somewhere
else and at some later time. Natural systems and the world
made by humans are intertwined in more ways than we can
possibly imagine. The result is less like a machine than it is
like a web stretching across all life forms and back through
time. The effects of human actions millennia ago still ripple
forward, intersect with other changes sometimes amplify
ing, sometimes diminishing in intensity. Some human
wrought changes, such as deforestation and saline soils
throughout much of the Middle East, are permanent as
we measure time.
Nothing in the preceding paragraph is particularly
new or controversial. But the idea of interrelatedness has
yet to take hold of us in a deep way. We still live in thrall
to a world created by Descartes, Bacon, Galileo, and their
heirs who taught us to dissect, divide, parse, and analyze
by reduction but not how to put things back together or
see the world as systems and patterns. The results were
The greatest discovery of the past century had nothing to
do with nuclear physics, or computer science, or genetic
engineering. Rather it was the discovery of the essential
connectedness of life and environment. The primary dis
cipline of interrelatedness is ecology beginning with the
work of Ernst Haeckel in the nineteenth century. The
discovery of evolution extended the awareness of our con
nections to life in time and more extensively to the story of
life on Earth. Fields such as ecology, general systems
theory, systems dynamics, operations research, and chaos
theory added details and theoretical depth, but with each
advance in the precision and extent of knowledge the
larger story remained the same. Living systems are linked
in food webs and ecological processes into larger systems
whether called the noosphere, biosphere, ecosphere, or
Gaia. The boundaries between life forms and between
what we take to be living and nonliving things shift and
Ecological Systems Thinking
intellectual power without perspective so that, in time,
overspecialization became a kind of a cultural disease.
There are many reasons why things do not change long
after their deficiencies are apparent: the inertia of habit,
economic inconvenience, the preservation of reputation,
and intellectual laziness. But the most important barrier to
change remains simply that science and the technology it
spawned works and is a powerful presence in our daily
lives. Automobiles, airplanes, the cornucopia evident in
every supermarket, miracle cures, and the wonders of
computers and communications are a constant reminder
of the powers of a particular kind of science and a promise
of things to come. That much of our technology also ‘bites
back’ and incurs costs that we do not see is mostly lost on
us. Many live in what has been called a ‘consensus trance’,
believing that things will go well for us, which is to say
that progress will continue indefinitely. Beneath such
ideas is the faith that nature does not ‘‘set traps for unwary
species,’’ as biologist Robert Sinsheimer once put it or that
progress itself is not a self made trap.
There have always been skeptics, however. Toward the
end of his life, H. G. Wells could see no grounds for hope.
More recently, Joseph Tainter, Martin Rees, and Jared
Diamond have expressed doubts about our longevity
based in no small part on their views of scientific progress.
Rees, for example, believes that our odds of making it to
the year 2100 are no better than fifty fifty. Diamond has
cataloged the reasons why past societies have collapsed and
they bear more than a passing resemblance to our present
behavior. James Lovelock, coauthor of the Gaia hypothesis,
believes that we are approaching a climate tipping point
somewhere between 400 and 500 ppm CO2 in the atmo
sphere after which ‘‘nothing the nations of the world do will
alter the outcome and the Earth will more irreversibly to a
new hot state.’’ In various ways, each of these attributes our
vulnerability to the failure to see systems, patterns, and to
exercise foresight. As a result, we stumble toward a time of
severe climate destabilization, biotic impoverishment, and
ecological surprises.
The failure of ecological knowledge to penetrate very
deeply into the larger society and its decision making
systems ought to be a matter of grave concern. The
early work of ecologists Howard and Eugene Odum on
the productivity of salt marshes, for example, may have
slowed but certainly did not stop the juggernaut of devel
opment that has severely damaged coastal ecosystems
virtually everywhere. Similarly, we know a great deal
about the services of natural systems and the impossibility
of duplicating these by human means. Yet the drawdown
of natural capital and the destruction of ecosystems are
still trumped by narrow short term concerns of profit and
economic expansion. Sometimes the costs of ecological
folly become starkly apparent as they did following hur
ricane Katrina in the fall of 2005 in which the damage
done by a class III hurricane (at landfall) was amplified by
13
the removal of mangroves and coastal forests that would
otherwise have absorbed much of its energy and dam
pened the destructive effects. That, too, was known in
many circles but did not have much effect on the policies
that prevailed along the Gulf Coast, where oil extraction,
commerce, and gambling ruled the day.
Public attitudes toward science are often undermined by
poor education, inadequate public funding, and, sometimes,
religious dogma. In the USA, evolution, once thought to be
an established part of science, is hotly contested as just
another ‘theory’ by advocates of ‘intelligent design’. The
scientific evidence about human driven climate change is
indisputable, but ignored or underestimated even when
alternatives are economically advantageous. The results
are evident in the considerable data describing ecological
deterioration virtually everywhere and the failure to seize
better alternatives as well. Law based on ecological knowl
edge and the hope that we might calibrate our public
business with the way the world works as a physical system
is under constant assault. Evidence about the health and
ecological effects of toxins is downplayed. Public access to
information about the release of toxics is restricted. The
result is a significant gap between what is known about how
the world works as a physical system and the public policy
in every country. The cumulative result is that we are much
more vulnerable to ecological ruin and extreme events than
we might otherwise be.
What can be done with ecological knowledge? One
answer is that ecology as a science ought to do what it
has been doing, which is to say document the deterioration
of ecosystems in ever finer detail. Ecology, the argument
goes, is a science and its practitioners ought to maintain
their credibility as scientists and not assume the role of
advocates and risk losing their credibility even when they
recognize folly disguised as public policy. If that is the
future of the discipline it will, I think, flourish for a time
while the human prospect withers.
There is, however, another perspective on the uses of
ecology. Paul Sears in 1964 and later Paul Shepard and
Daniel McKinley in 1969 once called the discipline ‘‘the
subversive science.’’ They proposed ecology as an integra
tive discipline, ‘‘a kind of vision across boundaries’’ and a
‘‘resistance movement’’ – an alternative to being ‘‘man
fanatic.’’ Ecology in their view ‘‘offers an essential factor
. . . to all our engineering and social planning.’’ In their
perspective, the world needs to know what ecologists
know and needs to take that knowledge seriously enough
to transform the ways by which we provision ourselves
with food, energy, materials, shelter, and livelihood.
Ecology as a subversive science would be integrated with
building, industry, agriculture, landscape management,
economics, and governance. In short, the idea of interre
latedness would move from the pages of obscure scientific
journals out to the main street, and into board rooms,
editorial offices, courtrooms, legislatures, and classrooms.
14
Ecological Systems Thinking
It would progress from being just one more interesting but
obsolete idea to become the design principles for a better
world – the default setting for everyday behavior.
Applied Systems Thinking
In this regard, the news is guardedly optimistic. The art
and science of high performance building is growing. The
result is a new generation of buildings that require a
fraction of the energy of conventional buildings, use mate
rials screened for environmental effects, minimize water
consumption, and are landscaped to promote biological
diversity, moderate microclimates, and grow foods. The
best of these are highly efficient, powered substantially by
sunlight and feature daylight, water recycling, and interior
green spaces. They are a finer calibration between our five
senses and the built environment and tend to promote
higher user satisfaction and productivity. The costs of
building green, as it turns out, are not necessarily higher
than conventional buildings while having lower operating
costs. The goal is to design buildings as whole systems, not
as disjointed components. The green building movement
is now a worldwide movement and is transforming the
practice of architecture, landscape architecture, and engi
neering. It could, in time, transform the design of
communities and cities as well.
Business, too, is beginning to go green. The best exam
ple of a well run environmentally sensitive business is
that of Interface, Inc., a global manufacturer of carpet
tiles and raised flooring. In the mid 1990s, company
founder and CEO, Ray Anderson, decided to transform
the company to eliminate waste and carbon emissions.
Interface launched a pioneering effort to develop carpet
products that were returned to the company as a ‘‘product
of service’’ not otherwise discarded in a landfill. Interface
now leases carpet to its customers and takes it back to be
remade into new products, thereby eliminating much of
the petrochemical sources at one end and waste at the
other. In the past decade, the company has eliminated
56% of its carbon emissions and is on track to becoming
carbon neutral. The model for the company is consciously
that of ecology all the way down to carpet products that
mimic a forest floor. Interface is not alone. Other compa
nies like Wal Mart and DuPont are beginning to
transform themselves as well. Some day, perhaps, all
business will be powered by sunlight with materials cycles
that mimic the circular flow of nutrients in ecosystems.
In agriculture, Wes Jackson, co founder of the Land
Institute, is pioneering the development of natural
systems agriculture. The goal is to model agriculture
on ecological systems such as forests and prairies. If
successful, the end product will be agricultural polycul
tures of high yield perennials, long thought to be a
biological impossibility. The early results, however,
have confirmed Jackson’s hypothesis that the two can
be stitched together, thereby eliminating a great deal of
fossil energy and soil erosion.
Materials science is a fourth area in which ecology is
being taken seriously. Nature, as chemist Terry Collins
has noted, uses only a relatively few ingredients while
industrial chemistry uses virtually the entire periodic
table, creating ecological havoc. The field of biomimicry
has grown in response by studying how nature works in
fine detail. Natural systems are a carnival of color, for
instance, but nature does not use paints. To answer such
questions, Janine Benyus, author of Biomimicry, is devel
oping a database of the ways nature works to filter,
reduce, recycle, color, purify, form, and join – all done
without the use of toxics and fossil fuels and all of it
biodegradable. The result could be a transformation of
materials and industry that dramatically reduce pollution
and energy use.
In these examples and elsewhere, the science of applied
ecology has begun to seriously influence decisions and
behavior and the evolution of architecture, engineering,
materials science, agronomy, urban planning, and econom
ics. The driving force is partly economic (to reduce the
costs of unnecessary energy, materials and water use) and
partly a matter of conviction (that it is wrong to leave a
legacy of ruin behind us). While promising, such measures
are necessary but insufficient. Ecological thinking, in one
way or another, must become a more central part of global
society and this is the task of education.
Environmental Education
The idea of specifically environmental education entered
the public discourse in the late 1960s. Among the recom
mendations of the Stockholm Conference in 1972 was to
‘‘establish an international programme in environmental
education.’’ UNESCO and UNEP subsequently under
took to prepare curricular materials, establish priorities,
develop pilot projects, and organize meetings. The result
was a UN sponsored Conference at Tbilisi, Georgia, in
1978 that produced a consensus statement including the
words:
Environmental education . . . should constitute a compre
hensive lifelong education . . . it should prepare the
individual for life through an understanding of the
major problems of the contemporary world, and the pro
vision of skills and attributes needed to play a productive
role towards improving life and protecting the environ
ment with due regard given to ethical values. By adopting
a holistic approach, rooted in a broad interdisciplinary
base, it recreates an overall perspective which acknowl
edges the fact that natural environment and manmade
environment are profoundly interdependent . . ..
Ecological Systems Thinking
The Tbilisi Conference produced 41 recommendations
spanning the needs for environmental education between
developed and less developed countries. In the subsequent
decades, initiatives, including those spawned by Agenda 21
and discussions about the Earth Charter, have advanced
the discussion of environmental education into a major part
of the dialog about the role of education relative to the
human prospect. There is no serious discussion about the
transition to sustainability launched by the Brundtland
Report in 1987 that does not include changing the goals
and methods of education. From Tbilisi (1978), Talloires
(1990), and subsequent international gatherings, a strong
consensus about the importance of environment in higher
education is clearly apparent.
Despite considerable progress, both conceptually and
practically, there are serious differences about the goals
and methods of environmental education that reflect and,
in some ways, amplify larger disagreements about educa
tion. At the lowest level, there is a general consensus that
the young ought to know something about how nature
works as a physical system – the rudiments of biology and
planetary science. There is considerably less agreement
about how this should be incorporated into the standard
curriculum or at what level. Most elementary schools
include curricular components such as ‘Project Learning
Tree’ or ‘Wet and Wild’ that introduce children to what
was once called natural history along with some field
experience and practical outdoor skills. But the later
inclusion of values or discussion about the causes of
environmental ills has often been controversial, especially
when it has led to questions about conventional economic
or political wisdom.
In important respects, all education is environmental
education, that is, by what is included or excluded stu
dents are taught that they are part of or apart from
ecological systems. The standard, discipline centric cur
riculum may have contributed to a mindset that helped to
create environmental problems by separating subjects
into boxes and conceptually by separating people from
nature. As a result, graduates are often ignorant of ecolo
gical relationships or why they are worthy of
consideration. Not surprisingly, the first response to pro
posals for environmental education attempted to
accommodate environmental issues and ecology into for
mal education as a kind of add on. More radical critics
proposed that formal education ought to be reformed
along ecological lines, raising another and no less con
tentious issues. From either perspective, environmental
mismanagement and the larger discussion of sustainability
raise questions about the meaning of human mastery over
nature, or more accurately as C. S. Lewis once put it: what
does it mean for some men to control other men through
the mastery of some parts of nature? What is the core
knowledge of the environment that ought to be standard
in an educational curriculum? At the heart of such
15
questions are important differences about what it means
to be human, what part of that definition ought to remain
inviolable, and about the manipulation of natural systems
through technological means such as genetic engineering.
Is the problem, in other words, one in education or one of
education?
What can be said with certainty is that public school
ing and higher education have been underachievers in the
task of inculcating essential knowledge about the envir
onment. Public opinion surveys show high levels of
support for environmental quality but little ecological
knowledge. In the words of one typical survey, people
have acquired a ‘‘substantial familiarity with environmen
tal issues, but [have] a long way to go in developing a
working environmental/energy knowledge.’’ Much of
what people know about the environment is derived
from television in bits and pieces and not through direct
experience with nature or through cultural transmission.
One particularly encouraging aspect is the develop
ment of environmental education in institutions of
higher education. Stemming from innovations in the
1980s, a vibrant campus ecology movement has emerged
in Europe, Australasia, and the USA, along with a wide
discussion of sustainability of educational institutions.
Beginning with the studies of college food, energy use,
and pollution, the movement has grown in subsequent
decades to a worldwide scale. Hundreds of colleges and
universities globally have organized efforts to system
atically reduce energy use, water consumption, and
material flows. Campus sustainability and climate stabi
lity have come to the center of institutional planning,
purchasing, and construction. Beginning in the late
1990s with the advent of means to promote and measure
environmental performance of buildings, the construc
tion of academic facilities is undergoing a rapid
revolution. Green or high performance building stan
dards are increasingly regarded as necessary to reduce
energy and maintenance costs as well as laboratories for
research and education. Many of the problems of sus
tainability – ecological design, applications of solar
energy, water purification, food production, ecological
restoration, and landscape management – can be studied
in buildings and adjacent landscapes at a scale that is
both significant yet manageable. Given recent develop
ments on many campuses, it is not inconceivable that
educational institutions at all levels will one day become
models of ecological design mirroring the larger solu
tions necessary to the transition to sustainability.
Summary
In the decades since the Stockholm Conference in
1972, environmental education has emerged as a sig
nificant component of education virtually everywhere
16
Ecosystems
in the world. It has, for the most part, flourished at all
levels of education. There are magazines and journals
such as Sustainability in Higher Education, professional
associations, and regular conferences. It is not difficult
to imagine all of this as the start of something like an
ecological enlightenment emerging in the decades or
centuries ahead. But no such thing is certain. If edu
cation is to be midwife to a deeper, broader, and
sustainable transformation, it will have to surmount
serious challenges.
Further Reading
Barlett P and Chase G (eds.) (2004) Sustainability on Campus.
Cambridge, MA: MIT Press.
Benyus J (1998) Biomimicry. New York: William Morrow.
Bowers C (1993) Education, Cultural Myths, and the Ecological Crisis.
Albany, NY: SUNY Press.
Bowers C (1995) Educating for an Ecologically Sustainable Culture.
Albany, NY: SUNY Press.
Corcoran P and Wals A (eds.) (2004) Higher Education and the Challenge
of Sustainability. Dordrecht, The Netherlands: Kluwer Academic.
Coyle K (2005) Environmental Literacy in America. Washington, DC: The
National Environmental Education & Training Foundation.
Creighton S (1998) Greening the Ivory Tower. Cambridge, MA: MIT
Press.
de Chardin T (1965) The Phenomenon of Man. New York: Harper
Torchbooks.
Fischetti M (2001) Drowning New Orleans. Scientific American
(October, 2001): 76 85.
Kuhn T (1963) The Structure of Scientific Revolutions. Chicago:
University of Chicago Press.
Lovelock J (2006) The Revenge of Gaia. London: Penguin Books.
Lovelock J The Gaia Hypothesis. New York: Oxford University Press.
Lovins A (2005) Winning the Oil Endgame. Snowmass, CO: Rocky
Mountain Institute.
Oakeshott M (1989) The Voice of Liberal Learning. New Haven, CT: Yale
University Press.
Orr D (1992) Ecological Literacy. Albany, NY: Suny Press.
Orr D (1994) Earth in Mind. Washington, DC: Island Press.
Orr D (2006) Design on the Edge. Cambridge, MA: MIT Press.
O’Sullivan E (2005) Millennium Ecosystem Assessment Report,
vols. 1 5. Washington, DC: Island Press.
Rees M (2003) Our Final Hour. New York: Basic Books.
Sears P (1964) Ecology A subversive subject. BioScience
14(7): 11 13.
Shepard P and McKinley D (eds.) (1969) The Subversive Science.
Boston: Houghton Mifflin.
Sinsheimer R (1978) The Presumptions of Science. Daedalus
107: 23 36.
Sobel D (1996) Beyond Ecophobia. Great Barrington, MA: The Orion
Society.
Steffen W, Sanderson A, Jager J, et al. (2004) Global Change and the
Earth System. Berlin: Springer.
Tenner E (1996) Why Things Bite Back: Technology and the Revenge of
Unintended Consequences. New York: Knopf.
Union of Concerned Scientists (1992) World Scientists Warning to
Humankind. Boston: Union of Concerned Scientists.
US Department of Health, Education, and Welfare (1978) Toward an
Action Plan: A Report on the Tbilisi Conference on Environmental
Education. Washington, DC: US Government Printing Office.
Vernadsky V (1998) The Biosphere. New York: Springer.
Washburn J (2005) University INC: The Corporate Corruption of Higher
Education. New York: Basic Books.
Wright R (2005) A Short History of Progress. New York: Carroll &
Graf.
Wright T (2004) Evolution of sustainability declarations in higher
education. In: Corcoran PB and Wals AEJ (eds.) Higher Education
and the Challenge of Sustainability, pp. 7 19. Dordrecht, The
Netherlands: Kluwer Academic.
Ecosystems
A K Salomon, University of California, Santa Barbara, Santa Barbara, CA, USA
ª 2008 Elsevier B.V. All rights reserved.
What Is an Ecosystem?
Studying Ecosystem Dynamics
Ecosystem Function and Biodiversity
Ecosystem Perspectives in Conservation Science
Further Reading
What Is an Ecosystem?
Typically, boundaries between ecosystems are diffuse. An
‘ecotone’ is a transition zone between two distinct ecosys
tems (i.e., the tundra–boreal forest ecotone).
Coined by A. G. Tansley in 1935, the term ‘ecosystem’
refers to an integrated system composed of a biotic com
munity, its abiotic environment, and their dynamic
interactions. A diversity of ecosystems exist through the
world, from tropical mangroves to temperate alpine lakes,
each with a unique set of components and dynamics
(Figure 1). Ecosystems can be classified according to
their components and physical context yet their classifi
cation is highly dependent on the spatial scale of scrutiny.
History
Over 70 years ago, Sir Arthur Tansley (Figure 2) pre
sented the notion that ecologists needed to consider ‘the
whole system’, including both organisms and physical
factors, and that these components could not be separated
or viewed in isolation. By suggesting that ecosystems are
Ecosystems
(a)
(b)
(c)
(d)
(e)
(f)
17
Figure 1 (a) Kelp forest, (b) subarctic alpine tundra, (c) tropical coastal sand dune, (d) tropical mangrove, (e) alpine lake, and
(f) temperate coastal rain forest. Photos by Anne Salomon, Tim Storr, and Tim Langlois.
system of biotic and abiotic components. He considered
how the lake food web and processes driving nutrient
flux affected the rate of succession of the whole lake
ecosystem, a significant departure from traditional inter
pretations of succession.
By the late 1950s and early 1960s, system wide
energy fluxes were quantified in various ecosystems by
E. P. Odum and J. M. Teal. In the late 1960s, Likens,
Bormann, and others took an ecosystem approach to
studying biogeochemical cycles by manipulating whole
watersheds in the Hubbard Brook Experimental Forest to
determine whether logging, burning, or pesticide and
herbicide use had an appreciable effect on nutrient loss
from the ecosystem. This research set an important
precedent in demonstrating the value of conducting
experiments at the scale of an entire ecosystem (see the
section entitled ‘Whole ecosystem experiments’), a sig
nificant advancement which continues to inform
ecosystem studies today.
Figure 2 Sir Arthur G. Tansley coined the term ecosystem in
1935. From New Phytologist 55: 145, 1956.
Ecosystem Components and Properties
dynamic, interacting systems, Tansley’s ecosystem con
cept transformed modern ecology. It led directly to
considerations of energy flux through ecosystems and
the pathbreaking, now classic work of R. L. Lindeman in
1942, one of the first formal investigations into the func
tioning of an ecosystem, in this case a senescent lake,
Cedar Creek Bog, in Minnesota. Inspired by the work of
C. Elton, Lindeman focused on the trophic (i.e., feeding)
relationships within the lake, grouping together organ
isms of the lake according to their position in the
food web. To study the cycling of nutrients and the
efficiency of energy transfer among trophic levels over
time, Lindeman considered the lake as an integrated
Ecosystems can be thought of as energy transformers
and nutrient processors composed of organisms within a
food web that require continual input of energy to balance
that lost during metabolism, growth, and reproduction.
These organisms are either ‘primary producers’ (auto
trophs), which derive their energy by using sunlight to
convert inorganic carbon into organic carbon, or ‘second
ary producers’ (heterotophs), which use organic carbon as
their energy source. Organisms that perform similar types
of ecosystem functions can be broadly categorized by
their ‘functional group’. For example, ‘herbivores’ are
heterotophs that eat autotrophs, ‘carnivores’ are hetero
trophs that eat other heterotrophs, while ‘detritivores’
are heterotrophs that eat nonliving organic material
18
Ecosystems
(detritus) derived from either autotrophs or heterotrophs
(Figure 3). Herbivores, carnivores, and dertitivores are
collectively known as ‘consumers’.
Classifying organisms according to their feeding rela
tionships is the basis of defining an organism’s ‘trophic
level’; the first trophic level includes autotrophs; the
second trophic level includes herbivores and so on.
Ecosystem components that make up a trophic level are
quantified in terms of biomass (the weight or standing
crop of organisms), while ecosystem dynamics, the flow of
energy and materials among system components, are
quantified in terms of rates.
Typically, ecologists quantifying ecosystem dynamics
use carbon as their currency to describe material flow and
energy to quantify energy flux. Material flow and energy
flow differ in one important property, namely their ability
to be recycled. Chemical materials within an ecosystem
are recycled through an ecosystem’s component. In con
trast, energy moves through an ecosystem only once and
is not recycled (Figure 3). Most energy is transformed to
heat and ultimately lost from the system. Consequently,
the continual input of new solar energy is what keeps an
ecosystem operational.
Heat
Heat
4
Detritivores
Detritus
Solar energy is transformed into chemical energy by
primary producers via photosynthesis, the process of con
verting inorganic carbon (CO2) from the air into organic
carbon (C6H12O2) in the form of carbohydrates. Gross
primary production is the energy or carbon fixed via
photosynthesis over a specific period of time, while net
primary production is the energy or carbon fixed in
photosynthesis, minus energy or carbon which is lost via
respiration, per unit time. Production by secondary pro
ducers is simply the amount of energy or material formed
per unit term.
A careful distinction needs to be made between
production rates and static estimates of standing crop
biomass, particularly because the two need not be
related. For example, two populations at equilibrium,
in which input equals output, might have the same
standing stock biomass but drastically different pro
duction rates because turnover rates can vary
(Figure 4). For example, on surf swept shores from
Alaska to California, two species of macroalgal pri
mary producers grow in the low rocky intertidal zone
of temperate coastal ecosystems (Figure 5). The rib
bon kelp, Alaria marginata, is an annual alga with high
growth rates, whereas sea cabbage, Hedophyllum sessile,
is a perennial alga with comparatively lower growth
rates. Although they differ greatly in their production
rates, in mid July, during the peak of the growing
season, these two species can have almost equivalent
stand crop biomasses.
Trophic level
Heat
Carnivores
3
Heat
Herbivores
2
(a)
(b)
High
production
Low
production
Heat
Heat
1
Primary producers
Standing
crop
biomass
(high turnover)
Standing
crop
biomass
(low turnover)
High input
Figure 3 Energy flows and material cycles in an ecosystem.
Materials move through the trophic levels and eventually cycle
back to the primary producers via the decomposition of detritus
by microorganisms. Energy, originating as solar energy, is
transferred through the trophic levels via chemical energy and is
lost via the radiation of heat at each step. Adapted from
DeAngelis DL (1992) Dynamics of Nutrient Cycling and Food
Webs. New York, NY: Chapman and Hall.
Low input
Figure 4 Standing crop biomass is not always correlated to
production rates. Here, two hypothetical species with
populations at equilibrium, where input equals output, have an
equivalent standing crop biomass but differ in their turnover
rates. Population (a) has high input, high production, and high
turnover rates, whereas population (b) has low input, low
production, and low turnover rates. In reality, populations are
rarely at equilibrium so standing crop biomass fluctuates
depending on input rates and the amount of production
consumed by higher trophic levels. Adapted from Krebs C (2001)
Ecology: The Experimental Analysis of Distribution and
Abundance, 5th edn. San Francisco: Addison-Wesley
Educational Publishers, Inc.
Ecosystems
(a)
19
(b)
(c)
Figure 5 (a) In the low intertidal zone of temperate coastal ecosystems, (b) the ribbon kelp, Alaria marginata, is an annual alga with
high growth rates, whereas (c) the sea cabbage kelp, Hedophyllum sessile, is a perennial alga with lower growth rates. During the peak
of the growing season, these two species can have a similar stand crop biomass but differ greatly in their production rates because one
is an annual and the other is a perennial. Photo by Anne Salomon and Mandy Lindeberg.
Ecosystem Efficiency
The efficiency of energy transfer within an ecosystem can
be estimated as its ‘trophic transfer efficiency’, the fraction
of production passing from one trophic level to the next.
The energy not transferred is lost in respiration or to
detritus. Knowing the trophic transfer efficiency of an
ecosystem can allow researchers to estimate the primary
production required to sustain a particular trophic level.
For example, in aquatic ecosystems, trophic transfer
efficiency can vary anywhere between 2% and 24%, and
average 10%. Assuming a trophic efficiency of 10%,
researchers can estimate how much phytoplankton produc
tion is required to support a particular fishery. Consider the
open ocean fishery for tuna, bonitos, and billfish. These are
all top predators, operating at the fourth trophic level.
According to world catch statistics recorded by the Food
and Agriculture Organization, in 1990, 2 975 000 t of
these predators were caught, equivalent to 0.1 g of carbon
per m2 of open ocean per year. To support this yield of
tuna, bonitos, and billfish, researchers can calculate the
production rates of the trophic levels below, assuming a
trophic efficiency of 10% and equilibrium conditions.
Essentially, to produce of 0.1 gC m 2 yr 1 of harvested
predators (tuna, bonitos, and billfish) requires
1 gC m 2 yr 1 of pelagic fish to have been consumed by
the top predators, 10 gC m 2 yr 1 of zooplankton to be
consumed by the pelagic fishes, and 100 gC m 2 yr 1 of
phytoplankton. Note that these values represent the pro
duction that is transferred up trophic levels. They do not
represent the standing stock of biomass at each trophic
level. Knowing the net primary production of the
photoplankton allows researchers to estimate the propor
tion of this production that is taken by the fishery.
It has been estimated that 8% of the world’s aquatic
primary production is required to sustain global fisheries.
Considering continental shelf and upwelling areas speci
fically, these ecosystems provide one fourth to one third
of the primary production required for fisheries. This
high fraction leaves little margin for error in maintaining
resilient ecosystems and sustainable fisheries.
Large-Scale Shifts in Ecosystems
A growing body of empirical evidence suggests that
ecosystems may shift abruptly among alternative
states. In fact, large scale shifts in ecosystems have
been observed in lakes, coral reefs, woodlands, des
serts, and oceans. For example, a distinct shift
occurred in the Pacific Ocean ecosystem around
1977 and 1989. Abrupt changes in the time series of
fish catches, zooplankton abundance, oyster condition,
and other marine ecosystem properties signified con
spicuous shifts from one relatively stable condition to
another (Figure 6). Also termed ‘regime shifts’, the
implications of these abrupt transitions for fisheries
and oceanic CO2 uptake are profound, yet the
mechanisms driving these shifts remain poorly under
stood. It appears that changes in oceanic circulation
driven by weather patterns can be evoked as the
dominant causes of this state shift. However, compe
tition and predation are becoming increasingly
recognized as important drivers of change altering
20
Ecosystems
1977 Regime shift
Ecosystem state
1.0
0.5
0
–0.5
–1.0
1965
1970
1975
1980
1985
1990
1985
1990
1995
2000
Ecosystem state
1989 Regime shift
1.0
0.5
0
–0.5
–1.0
1975
1980
Figure 6 Distinct shifts in ecosystem states, also referred to as ‘regime shifts’, occurred in the Pacific Ocean ecosystem around 1977
and 1989. The ecosystem state index shown here was calculated based on the average of climatic and biological time series. From
Scheffer M, Carpenter S, Foley JA, Folke C, and Walker B (2001) Catastrophic shifts in ecosystems. Nature 413: 591–596.
oceanic community dynamics. In fact, fisheries are
well known to affect entire food webs and the trophic
organization of ecosystems. Therefore, one could
imagine that the sensitivity of a single keystone spe
cies to subtle environmental change could cause
major shifts in community composition. Given this
interplay between and within the biotic and abiotic
components of an ecosystem, resolving the causes of
regime shifts in oceanic ecosystems will likely require
an understanding of the interactions between the
effects of fisheries and the effects of physical climate
change.
Studying Ecosystem Dynamics
Stable Isotopes
Important insights into ecosystem dynamics can be
revealed through the use of naturally occurring ‘stable
isotopes’. These alternate forms of elements can reveal
both the source of material flowing through an ecosystem
and its consumer’s trophic position. This is because dif
ferent sources of organic matter can have unique isotopic
signatures which are altered in a consistent manner as
materials are transferred throughout an ecosystem, from
trophic level to trophic level. Consequently, stable
isotopes provide powerful tools for estimating material
flux and trophic positions.
The elements C, N, S, H, and O all have more than
one isotope. For example, carbon has several isotopes, two
of which are 13C and 12C. In nature, only 1% of carbon is
13
C. Isotopic composition is typically expressed in
values, which are parts per thousand differences from a
standard. For carbon,
13 C ¼
13
C=12 Csample
13 C=12 C
standard
1 103
Consequently, values express the ratio of heavy to light
isotope in a sample. Increases in these values denote
increases in the amount of the heavy isotope component.
The standard reference material for carbon is PeeDee
limestone, while the standard for nitrogen is nitrogen
gas in the atmosphere. Natural variation in stable isotopic
composition can be detected with great precision with a
mass spectrometer.
Stable isotopes record two kinds of information. Process
information is revealed by physical and chemical reactions
which alter stable isotope ratios, while source information is
revealed by the isotopic signatures of source materials.
When organisms take up carbon and nitrogen, chemical
reactions occur which discriminate among isotopes, thereby
altering the ratio of heavy to light isotope. This is known as
Ecosystems
‘fractionation’. Although carbon fractionates very little
(0.4‰, 1 SD ¼ 1‰), the mean trophic fractionation of
15N is 3.4‰ (1 SD ¼ 1‰), meaning that 15N increases
on average by 3.4‰ with every trophic transfer. Because the
15N of a consumer is typically enriched by 3.4‰ relative to
its diet, nitrogen isotopes can be used to estimate trophic
position. Stable isotopes can provide a continuous measure
of trophic position that integrates the assimilation of energy
or material flow through all the different trophic pathways
leading to an organism. In contrast, 13C can be used to
evaluate the ultimate sources of carbon for an organism
when the isotopic signatures of the sources are different.
Stable isotopes can track the fate of different sources of
carbon through an ecosystem, because a consumer’s iso
topic signature reflects those of the key primary producers
it consumes. For example, in both lake and coastal marine
ecosystems, 13C is useful for differentiating between two
major sources of available energy, benthic (nearshore)
production from attached macroalgae, and pelagic (open
water) production from phytoplankton. This is because
macroalgae and macroalgal detritus (specifically kelp of
the order Laminariales) is typically more enriched in 13C
(less negative 13C) relative to phytoplankton due to
boundary layer effects. Researchers have exploited this
difference to answer many important ecosystem level
questions. Below are two examples.
During the late 1970s and early 1980s, in the western
Aleutian Islands of Alaska, where sea otters had recovered
from overexploitation and suppressed their herbivorous
urchin prey, productive kelp beds dominated. There,
transplanted filter feeders, barnacles and mussels, grew
up to 5 times faster compared to islands devoid of kelp
where sea otters were scarce and urchin densities high.
Stable isotope analysis revealed that the fast growing
filter feeders were enriched in carbon suggesting that
macroalgae was the carbon source responsible for this
magnification of secondary production.
In four Wisconsin lakes, experimental manipulations
of fish communities and nutrient loading rates were con
ducted to test the interactive effects of food web structure
and nutrient availability on lake productivity and carbon
exchange with the atmosphere. The presence of top pre
dators determined whether the experimentally enriched
lakes operated as net sinks or net sources of atmospheric
carbon. Specifically, the removal of piscivorous fishes
caused an increase in planktivorous fishes, a decrease in
large bodied zooplankton grazers, and enhanced primary
production, thereby increasing influx rates of atmospheric
carbon into the lake. Atmospheric carbon was traced to
upper trophic levels with 13C. Here, naturally occurring
stable isotopes and experimental manipulations con
ducted at the scale of whole ecosystems illustrated that
top predators fundamentally alter biogeochemical pro
cesses that control a lake’s ecosystem dynamics and
interactions with the atmosphere.
21
Whole Ecosystem Experiments
Large scale, whole ecosystem experiments have contrib
uted considerably to our understanding of ecosystem
dynamics. With its beginnings in wholesale watershed
experiments in the 1960s, ecosystems are now being stud
ied experimentally and analyzed as system of interacting
species processing nutrients and energy within the con
text of changing abiotic conditions. This is particularly
relevant these days given the effects of anthropogenic
climate forcing and pollution in both terrestrial and ocea
nic ecosystems.
A classic series of whole lake nutrient addition experi
ments conducted in northwestern Ontario by David
Schindler and his research group illustrated the role of
phosphorus in temperate lake eutrophication. To separate
the effects of phosphorus and nitrate, the researchers split a
lake with a curtain and fertilized one side with carbon and
nitrogen and the other with phosphorus, carbon, and nitro
gen. Within 2 months, a highly visible algal bloom had
developed in the basin in which phosphorus had been
added providing experimental evidence that phosphorus
is the limiting nutrient for phytoplankton production in
freshwater lakes. Certainly, algae may show signs of nitro
gen or carbon limitation when phosphorus is added to a
lake; however, other processes often compensate for these
deficiencies. For instance, CO2 is rarely limiting because
physical factors such as water turbulence and gas exchange
regulate its availiblity. Further, nitrogen can be fixed by
blue green algae. These species, which are favored when
nitrogen is in short supply, increases the availability of
nitrogen to algae, and the lake eventually returns to a
state of phosphorus limitation. The practical significance
of these results is that lake europhication can be prevented
with management policies that control phosphorus input
into lake and rivers.
Using Management Policies as Ecosystem
Experiments
It has become increasingly common to use management
policies as experiments and test their effects on ecosystem
dynamics. An excellent example of this approach is the
use of marine reserves to investigate the ecosystem level
consequences of fishing. Essentially, well enforced mar
ine reserves constitute large scale human exclusion
experiments and provide controls by which to test the
ecosystem effects of reducing consumer biomass via fish
ing at an ecologically relevant scale. Dramatic shifts in
nearshore community structure have been documented in
well established and well protected marine reserves in
both Chile and New Zealand. In northeastern New
Zealand’s two oldest marine reserves, the Leigh Marine
Reserve and Tawharanui Marine Park, previously fished
predators, snapper (Pagrus auratus) and rock lobster
22
Ecosystems
(Jasus edwardsii), have increased in abundance by 14 and
3.8 fold, respectively, compared to adjacent fished waters.
Increased predation leading to reduced survivorship and
cryptic behavior of their herbivorous prey, the sea urchin
(Evechinus chloroticus), has allowed the macroalga (Ecklonia
radiata) to increase significantly within the reserves, a
trend that has been developing in the Leigh reserve for
the past 25 years (Figure 7). Although this provides
evidence that fishing can indirectly reduce ecosystem
productivity, the trophic dynamics described above are
context dependent and vary as a function of depth, wave
exposure, and oceanographic circulation (Figure 8). For
example, both in the presence and absence of fishing,
urchin densities decline to nearly 0 individuals per m2
below depths greater than 10 m due to unfavorable con
ditions for recruitment, despite the presence or absence of
snapper and lobster, while at depths above 3 m, wave
surge can preclude urchin grazing both inside and outside
the reserves. Furthermore, where oceanic conditions
hinder urchin recruitment, the effects of fishing on
macroalgae become less clear cut. These physical con
straints highlight the importance of abiotic context on
biotic interactions. Ultimately, one can gain a lot of infor
mation by using management policies as experiments.
Although policy experiments have played an impor
tant role in elucidating ecosystem dynamics, in many
cases, it is politically intractable or logistically impossible
(a)
to experiment with whole ecosystems. Under such cir
cumstances, researchers have used alternative techniques
to explore ecosystem dynamics. Models in ecology have a
venerable tradition for both teaching and understanding
complex processes. Ecosystem models are now being used
to gain insight into the ecosystem level consequences of
management policies, from fisheries to carbon emissions.
For more information on ecosystem models and using
management policies as experiments, see the section
entitled ‘Social–ecological systems, Humans as key eco
system components’.
Ecosystem Function and Biodiversity
Accelerating rates of species extinction have prompted
researchers to formally investigate the role of biodiversity
in providing, maintaining, and even promoting ‘ecosys
tem function’. Typically, studies experimentally modify
species diversity and examine how this influences
the fluxes of energy and matter that are fundamental to
all ecological processes. In many cases, studies are
designed to document the effects of species richness on
the efficiency by which communities produce biomass,
although the effects of species diversity on other ecosys
tem functions such as decomposition rates, nutrient
retention, and CO2 uptake rates have also been examined.
(b)
Fished
3
Snapper
Nonfished
Snapper
Lobster
Lobster
Trophic level
–
–
2
Sea
urchin
+
–
Sea
urchin
–
Kelp
Kelp
1
Figure 7 (a) In nearshore fished ecosystems in northeastern New Zealand, snapper and lobster densities have been reduced due to
fishing pressure resulting in high sea urchin densities, urchin barrens, and reduced kelp production. (b) In marine reserves, where
previously fished snapper and lobster have recovered, sea urchins that have not been consumed by these predators behave cryptically,
hiding in crevices. Consequently, kelp forests of Ecklonia radiata dominate. Photos by Nick Shears, Hernando Acosta, and Timothy
Langlois.
Ecosystems
(a )
East Auckland
Current
(b )
Spring
(c)
Summer
23
Chlorophyll a (mg m–3)
0
0.1
0.2
0.5
10
3.0
Figure 8 The effects of fishing on nearshore ecosystems are influenced locally by wave exposure and regionally by oceanographic
circulation. (a) In northeastern New Zealand, ocean circulation patterns influence nutrient delivery and thus (b) spring and (c) summer
pelagic primary production. Satellite images: SeaWiFs Project, Ocean Color Web.
Several seminal studies report a positive relationship
between biodiversity and ecosystem function. Yet, the
generality of the results, and the mechanisms driving
them, have provoked considerable debate and several
counterexamples exist.
At the crux of the debate lies a question with deep
historical roots: do some species exert stronger control
over ecosystem processes than others? Imagine two dis
tinct positive relationships between biodiversity and
ecosystem function (Figure 9). In type A communities,
every single species contributes to the ecosystem function
measured, even the rare species. By contrast, in type B
High
Ecosystem function
Type B
Type A
Low
High
Biodiversity
Figure 9 Type A communities: every single species contributes equally to ecosystem functioning. Type B communities:
ecosystem function is provided by only a few species.
communities, almost all of the ecosystem function meas
ured can be provided by relatively few species,
suggesting that many species are in fact redundant.
Few empirical studies support type A relationships,
rather, empirical evidence points to the prevalence of
type B relationships. In fact, a recent meta analysis of
111 such studies conducted in multiple ecosystems on
numerous trophic groups found that the average effect
of decreasing species richness is to decrease the biomass of
the focal trophic group, leading to less complete depletion
of resources used by that group. Further, the most
species rich polycultures performed no differently than
the single most productive species used in the experi
ment. Consequently, these average effects of species
diversity on ecosystem production are best explained by
the loss of the most productive species from a diverse
community. These results could be considered consistent
with what has become known as the ‘sampling effect’.
Critics argue that a positive relationship between
species diversity and ecosystem function is a sampling
artifact rather than a result of experimentally manipu
lated biodiversity per se. Such a ‘sampling effect’ can arise
because communities comprising more species have a
greater chance of being dominated by the most produc
tive taxa. Yet, controversy surrounding the ‘sampling
effect’ itself exists given the duality in its possible inter
pretation: is this a real biological mechanism that operates
in nature or is it an experimental artifact of using random
draws of species to assemble experimental communities?
To add to the ecosystem function–biodiversity debate is
the critical issue that many of these studies focus on a
24
Ecosystems
single trophic level and neglect or dismiss multiple
trophic level interactions, such as herbivory and other
disturbances well known to alter ecosystem processes,
calling into question the generality of these results.
Despite the controversy, these studies generally rein
force the notion that certain species exert much stronger
control over ecological processes than others. However,
identifying which species these are in advance of extinc
tion remains a challenge. Nonetheless, identifying the
mechanisms driving ecosystem functioning is an impor
tant conservation priority given that human well being
relies on a multitude of these functions.
Ecosystem Perspectives in Conservation
Science
Ecosystem Services
Humans have always relied on nature for environmental
assets like clean water and soil formation. Today, these
assets are receiving global attention as ‘ecosystem ser
vices’, the conditions and processes by which natural
ecosystems sustain and fulfill human life. Natural ecosys
tems perform a diversity of ecosystem services on which
human civilization depends:
1. regulating services – purification of air and water,
detoxification and decomposition of wastes, moderation
of weather extremes, climate regulation, erosion control,
flood control, mitigation of drought and floods, regula
tion of disease carrying organisms and agricultural pests;
2. provisioning services – provision of food, fuel, fiber,
and freshwater;
3. supporting services – formation and preservation of
soils, protection from ultraviolet rays, pollination
of natural vegetation and agricultural crops, cycling
of nutrients, seed dispersal, maintenance of biodiver
sity, primary production; and
4. cultural services – spiritual, esthetic, recreational.
Although critical to human existence, ecosystem services
are often taken for granted or at best, greatly undervalued.
This is ironic given that many ecosystem services are very
difficult and expensive to duplicate, if they can be dupli
cated at all. Normally, ecosystem services are considered
‘free’ despite their obvious economic value. For example,
over 100 000 species of animals provide free pollination
services, including bats, bees, flies, moths, beetles, birds,
and butterflies (Figure 10). Based on the estimate that
one third of human food comes from plants pollinated by
wild pollinators, pollination has been valued at US$4–6
billion per year in the US alone. Globally, the world’s
ecosystem services have been valued at US$33 trillion a
year, nearly twice as much as the gross national product of
all of the world’s countries.
The idea of paying for ecosystem services has been
gaining momentum. Yet, because ecosystem services are
typically not sold in markets, they usually lack a market
value. Given the value of natural capital, nonmarket
valuation approaches are being developed by economists
and ecologists to account for ecosystem services in
decision making processes. The notion being that eco
nomic valuation gives decision makers a common
currency to assess the relative importance of ecosystem
processes and other forms of capital.
Figure 10 Pollination services, provided by bees, bats, butterflies, and birds to name a few, have been valued at US$4–6 billion per
year in the US alone. Consider the global value of this important ecosystem service. Photos by Steve Gaines, Heather Tallis.
Ecosystems
Yet, assigning value to ecosystem services is tricky and
some analysts object to nonmarket valuation, because it is
a strictly anthropogenic measure and does not account for
nonhuman values and needs. Yet, in democratic countries,
environmental policy outcomes are determined by the
desires of the majority of citizens, and voting on a pre
ferred policy alternative is ultimately an anthropogenic
activity. A second objection to nonmarket valuation is a
disagreement with pricing the natural world and dissatis
faction with the capitalistic premise that everything is
thought of in terms of commodities and money. The
point of valuation, however, is to frame choices and
clarify the tradeoffs between alternative outcomes (i.e.,
draining a wetland may increase the supply of develop
able land for housing but does so at the cost of decreased
habitat and potential water quality degradation). Finally,
a third objection to nonmarket valuation stems from the
uncertainty in identifying and quantifying all ecosystem
services. Advocates argue that economic valuation need
not cover all values and that progress is made by captur
ing values that are presently overlooked.
Despite the uncertainties, valuing ecosystem services
can sometimes pay off. When New York City compared
the coast of an artificial water filtration plant valued at
US$6–8 billion, plus an annual operating cost of US$300
million, the city chose to restore the natural capital of the
Catskill Mountains for this watershed’s inherent water
filtration services and for a fraction of the cost (US$660
million). Ultimately, the valuation of ecosystem services,
even if flawed, may get ecosystem processes on the deci
sion making table and lead to more sustainable policies in
light of ever expanding human populations.
Ecosystem services are threatened by growth in the scale
of human enterprise (population size, per capita consump
tion rates) and a mismatch between short term needs and
long term societal well being. With a global population
soon to number 9 billion people, ecosystem services are
becoming so degraded, some regions in the world risk
ecological collapse. Many human activities alter, disrupt,
impair, or reengineer ecosystem services such as overfish
ing, deforestation, introduction of invasive species,
destruction of wetlands, erosion of soils, runoff of pesticides,
fertilizers, and animal wastes, pollution of land, water, and
air resources. The consequences of degrading ecosystem
services on human well being were examined in the
Millennium Ecosystem Assessment (MA) 2005, which con
cluded that well over half of the world’s ecosystems services
are being degraded or used unsustainably. The MA devel
oped global ecological scenarios as a process to inform
future policy options. These scenarios were based on a
suite of models that were designed to forecast future change.
The MA based its scenario analyses on ecosystem services.
Specifically, scenarios were developed to anticipate
responses of ecosystem services to alternative futures driven
by different sets of policy decisions. Following the
25
completion of this ambitious ecological study, there is now
a growing movement to make the value of ecosystem ser
vices an integral part of current policy initiatives.
Social–Ecological Systems, Humans as Key
Ecosystem Components
Humans are a major force in global change and drive eco
system dynamics, from local environments to the entire
biosphere. At the same time, human societies and global
economies rely on ecosystem services. As such, human and
natural systems can no longer be treated independently
because natural and social systems are strongly linked.
Accumulating evidence suggests that effective environmen
tal management and conservation strategies must take an
integrated approach, one that considers the interactions and
feedbacks between and within social, economic, and
ecological systems. As a result, the concept of coupled
‘social–ecological systems’ has become an emerging focus
in environmental and social science and ecosystem manage
ment. Social–ecological systems are considered as evolving,
integrated systems that typically behave in nonlinear ways.
The concept of resilience – the capacity to buffer change –
has been increasingly used as an approach for understanding
the dynamics of social–ecological systems. Two useful tools
for building resilience in social–ecological systems are struc
tured scenario modeling and active adaptive management.
Models of linked social–ecological systems have
been developed to inform management conflicts over
water quality, fisheries, and rangelands. These models repre
sent ecosystems coupled to socioeconomic drivers and are
explored with stakeholders to probe the management deci
sion making processes. Alternative scenarios force
participants to be absolutely explicit about their assumptions
and biases, thereby improving communication between sta
keholders and exposing the ecological consequences of
various management policies.
Adaptive management is an approach where manage
ment policies themselves are deliberately used as
experimental treatments. As information is gained, poli
cies are modified accordingly. This approach helps isolate
anthropogenic effects from sources of natural variation
and, most importantly, considers the consequences of a
human perturbation on the whole ecosystem. In contrast,
basic research on various parts of an ecosystem leads to
the challenge of assembling all the data into a practical
framework. Yet, biotic and abiotic ecosystem components
are not additive, they interact. Due to these interactions,
the dynamics of an ecosystem cannot be extrapolated
from the simple addition of an ecosystem’s components.
Adaptive management examines the response of the sys
tem as a whole rather than a sum of its parts. Furthermore,
this approach involves adaptive learning and adaptive
institutions that acknowledge uncertainties and can
respond to nonlinearities. In sum, structured scenario
26
Ecosystem Services
modeling and policy experimentation are tools that can
be used to examine the resilience of social–ecological
systems to alternative management policies and conserva
tion strategies.
Ecosystem-Based Management
Recognizing the need to sustain the integrity and resili
ence of social–ecological systems has led to calls for
‘ecosystem based management’, a management approach
that considers all ecosystem components, including
humans and the physical environment. With the overall
goal of sustaining ecosystem structure and function, this
management approach:
on key ecosystem processes and their responses
• focuses
to perturbations;
ecological, social, and economic goals and
• integrates
recognizes humans as key components of the ecosystem;
management based on ecological boundaries
• defines
rather than political ones;
the complexity of natural processes and social
• addresses
systems by identifying and confronting uncertainty;
adaptive management where policies are used as
• uses
experiments and are modified as information is gained;
multiple stakeholders in a collaborative pro
• engages
cess to identify problems, understand the mechanisms
•
driving them, and create and test solutions; and
considers the interactions among ecosystems (terres
trial, freshwater, and marine).
Ecosystem based management is driven by explicit goals,
executed by policies and protocols, and made adaptable
by using policies as experiments, monitoring their out
comes and altering them as knowledge is gained.
Traditionally, management practices have focused on
maximizing short term yield and economic gain over
long term sustainability. These practices were driven by
inadequate information on ecosystem dynamics, ignorance
of the space and timescales on which ecosystem processes
operate, and a prevailing public perception that immediate
economic and social value outweighed the risk of alterna
tive management. Seeking to overcome these obstacles,
ecosystem based management relies on research at all levels
of ecological organization, explicitly recognizes the
dynamic character of ecosystems, acknowledge that
ecological processes operate over a wide range of temporal
and spatial scales and are context dependent, and
presupposes that our current knowledge of ecosystem func
tion is provisional and subject to change. Ultimately,
ecosystem based management recognizes the importance
of human needs while addressing the reality that the capa
city of our world to meet those needs in perpetuity has
limits and depends on the functioning of resilient
ecosystems.
See also: Ecosystem Ecology.
Further Reading
Cardinale BJ, Srivastava DS, Duffy JE, et al. (2006) Effects of biodiversity
on the functioning of trophic groups and ecosystems. Nature
443: 989 992.
Daily GC (ed.) (1997) Nature’s Services: Societal Dependence on
Natural Ecosystems. Washington, DC: Island Press.
DeAngelis DL (1992) Dynamics of Nutrient Cycling and Food Webs.
New York, NY: Chapman and Hall.
Krebs C (2001) Ecology: The Experimental Analysis of Distribution and
Abundance, 5th edn. San Francisco: Addison Wesley Educational
Publishers, Inc.
Millennium Ecosystem Assessment (2005) Ecosystems and Human
Well Being: Synthesis. Washington, DC: Island Press.
Pauly D and Christensen V (1995) Primary production required to
sustain global fisheries. Nature 374: 255 257.
Scheffer M, Carpenter S, Foley JA, Folke C, and Walker B (2001)
Catastrophic shifts in ecosystems. Nature 413: 591 596.
Ecosystem Services
K A Brauman and G C Daily, Stanford University, Stanford, CA, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Defining Ecosystem Services
Examples of Ecosystem Services
Capturing the Value of Ecosystem Services
Conclusions
Further Reading
Introduction
The world’s ecosystems yield a flow of essential services
that sustain and fulfill human life, from seafood and tim
ber production to soil renewal and personal inspiration.
Although many societies have developed the technologi
cal capacity to engineer replacements for some services,
such as water purification and flood control, no society
can fully replace the range and scale of benefits that
Ecosystem Services
ecosystems supply. Thus, ecosystems are capital assets,
worthy of at least the level of attention and investment
given to other forms of capital. Yet, relative to physical,
financial, human, and social capital, ecosystem capital is
poorly understood, scarcely monitored, and, in many
cases, undergoing rapid degradation and depletion.
Recognition of ecosystem services dates back at least
to Plato. This recognition of human dependence on eco
systems, in the past and today, is often triggered by their
disruption and loss. Direct enjoyment of services, such as
the extraction of timber, fish, and freshwater, can reduce
the quantity and quality produced. The provision of eco
system services can also be affected indirectly and
inadvertently. Deforestation, for instance, has exposed
the critical role of forests in the hydrological cycle –
mitigating flooding and reducing erosion. Release of
toxic substances has uncovered the nature and value of
physical and chemical processes, governed in part by
microorganisms that disperse and break down hazardous
materials. Thinning of the stratospheric ozone layer has
sharpened awareness of the value of its service in screen
ing out harmful ultraviolet radiation.
Defining Ecosystem Services
Simply put, ecosystem services are the conditions
and processes through which ecosystems, and the
biodiversity that makes them up, sustain and fulfill
human life. Ecosystem services are tightly interrelated,
making their classification somewhat arbitrary.
The Millennium Ecosystem Assessment (MA) – the for
mal international effort to elevate awareness and
understanding of societal dependence on ecosystems –
has suggested four categories.
First, ‘provisioning services’ provide goods such as
food, freshwater, timber, and fiber for direct human use;
these are a familiar part of the economy. Second, and
much less widely appreciated, ‘regulating services’ main
tain a world in which it is biophysically possible for
people to live, providing such benefits as water purifica
tion, pollination of crops, flood control, and climate
stabilization. Third, ‘cultural services’ make the world a
place in which people want to live; they include recrea
tion as well as esthetic, intellectual, and spiritual
inspiration. Fourth, ‘supporting services’ create the back
drop for the conditions and processes on which society
depends more directly. All of these services are provided
by complex chemical, physical, and biological cycles,
powered by the sun, and operate at scales ranging from
smaller than the period at the end of this sentence to as
large as the entire biosphere (Table 1).
27
Table 1 A classification of ecosystem services. Examples of
ecosystem services and how they can be categorized
Provisioning services: Production of. . .
Food
Seafood, agricultural crops, livestock, spices
Pharmaceuticals
Medicinal products, precursors to synthetic pharmaceuticals
Durable materials
Natural fiber, timber
Energy
Biomass fuels, low-sediment water for hydropower
Industrial products
Waxes, oils, fragrances, dyes, latex, rubber
Genetic resources
Intermediate goods that enhance the production of other
goods
Regulating services: Generation of. . .
Cycling and filtration processes
Detoxification and decomposition of wastes
Generation and renewal of soil fertility
Purification of air and water
Translocation processes
Dispersal of seeds to sustain tree and other plant cover
Pollination of crops and other plants
Stabilizing processes
Coastal and river channel stability
Control of the majority of potential pest species
Carbon sequestration
Partial stabilization of climate
Protection from disasters:
regulation of hydrological cycle (mitigation of floods and
droughts)
moderation of weather extremes (such as of temperature
and wind)
Cultural services: Provision of. . .
Esthetic beauty, serenity
Recreational opportunities
Cultural, intellectual, and spiritual inspiration
Supporting services: Preservation of. . .
Processes underlying services in the classes above
Options
maintenance of the ecological components and systems
needed for future supply of the goods and services above
and others awaiting discovery
Tradeoffs in Managing the Flows of Ecosystem
Services
Biophysical constraints on human activities, such as lim
ited supplies of energy, land, and water, typically manifest
themselves as tradeoffs between different uses. Thus,
managing ecosystem services involves difficult ethical
and political decisions about which services to develop
and how to do so. At local scales, allocation of limited
resources to alternative activities typically involves a
zero sum game, illustrated by the widespread redirection
of water from agriculture to urban and industrial pur
poses. At global scales, different groups of people compete
for use of Earth’s open access resources and waste sinks,
Ecosystem Services
such as the atmosphere’s capacity to absorb CO2 and
other greenhouse gases without inducing climate change.
Making informed decisions about how to use ecosys
tem goods and services hinges on understanding these
tradeoffs: knowing the joint products – the suite and
level of services – that ecosystems can provide. For exam
ple, an ecosystem managed exclusively for agriculture
may yield a greater return on agricultural products than
one managed for multiple services, but understanding
that diversified management may produce greater
overall returns could influence management decisions
(Figure 1).
Provision of biodiversity is one supporting service that
has historically been discounted when managing for other
ecosystem services. Biodiversity, however, can provide
irreplaceable benefits. Genetic diversity, for example,
allows for both the survival and evolution of the ecosys
tems we depend on for myriad benefits. Recent research
indicates that diverse systems are more resilient, and
therefore provide ecosystem services more reliably in
the long term, than monocultures. While under optimal
conditions managing for a single species may provide
superior timber supplies or nutrient sequestration, given
natural and human caused variability in temperature,
rainfall, and other environmental factors, managing for a
diverse system will more consistently provide services in
an uncertain world.
services, from local to global, and explore the tradeoffs
inherent in their management.
Pollination Services Provided by Bees
Pollination, the movement of genetic material in the
form of pollen grains, is a key step in the development of
most food crops. Even crops that do not rely on insect
pollination – wind pollinated or self pollinated crops –
are sometimes more productive when visited by an insect
pollinator. Bees are a particularly important group of insect
pollinators, responsible for pollinating 60–70% of the
world’s total flowering plant species, including nearly 900
food crops worldwide, such as apples, avocados, cucum
bers, and squash. These crops comprise 15–30% of the
world’s food production, and bees are credited with $4.2
billion in annual crop productivity in California alone.
Bees are especially important pollen vectors in part
because physical adaptations, such as hairs designed to
pick up pollen, and behavioral adaptations, such as fidelity
to a single species of plant on each pollen gathering trip,
ensure good pollen transport and cross pollination.
In the US, most major agricultural enterprises that rely
on bee pollination import managed bees, almost always
the European honeybee Apis mellifera. The available stock
of managed honeybees has declined dramatically, how
ever, dropping by over 50% in the last 50 years, while
demand for pollination services has increased in many
areas. This decline in managed bee populations has many
causes, including increased pesticide use, disease in the
hives, and downsizing of stocks that have hybridized with
Africanized bees, introducing traits that make managed
bees more aggressive and thus a liability to the farmer.
The contribution of native, wild bees to agricultural
pollination was ignored, and assumed to be negligible,
until the early 2000s. Since then, research has shown
that native bees serve an important role in pollination,
picking up slack when managed bee pollination is
Examples of Ecosystem Services
Agriculture
Ecosystem services can be explored by focusing either on
a single service that may be provided by various ecosys
tems or by looking at a single ecosystem that may provide
a variety of services. Here we illustrate both approaches,
considering first pollination services provided by bees
then the suite of services provided by wetlands and for
ests. We highlight the differences in scale of delivery of
$$$
Agriculture
28
$$
Water
purification
Biodiversity
$$
Climate
Biodiversity
Water purification
Climate
stabilization
$$
$$
Figure 1 Joint products of ecosystems. Many ecosystems are currently managed to exploit only one service. Managing for multiple
services can increase ecosystem benefits.
Ecosystem Services
insufficient and enhancing crop production in general.
Farms with generous native bee habitat nearby may be
able to fully or partially replace pollination by managed
bees. In some cases, native bees are more efficient polli
nators than European honeybees. The variety of wild
bees, with distinct physical and behavioral traits, allows
them, as a group, to pollinate a wide variety of flowering
plants. Tomatoes, for example, have pollen that is acces
sible only by vibrating the flower, which bumble bees and
some other native bees can, while honeybees cannot.
Though tomatoes are self pollinating and do not require
an insect vector, native bees promote cross pollination,
which, for example, significantly increases the fruit set
and size of Sungold cherry tomatoes.
The contributions of native bees to crop production
are usually undocumented and underestimated, and they
are always unpaid, at least directly. Though hives of
managed honeybees must be rented or maintained, wild
bees pollinate at no cost to the farmer. Populations of
native bees are under great threat, however, by land
management practices that promote the use of pesticides
and the loss, fragmentation, and degradation of habitat.
Protecting native bees without protecting the ecosystems
in which they live is impossible. Native habitat, unlike
agricultural monocropping, provides the year round sup
ply of blooming plants that wild bees require for
sustenance. Native habitat also provides nesting areas;
most wild bees are solitary, laying a single egg in a nest
cavity dug into the ground or into dead wood, not forming
social hives. In order to reap the benefits of native polli
nators, food resources and nesting habitat must be
available within a short distance of crops, possibly as
hedgerows, in ditches, or around water ponds. A study
of wild bee pollination of coffee in Costa Rica showed that
farms closer to tropical forest remnants were visited by
many more species of wild bees than those further away.
Had the far sites been adequately pollinated, coffee yield
would have been increased by nearly 20% and misshapen
coffee beans reduced by 27%. A lower bound estimate of
the pollination services from these patches is US $62 000
per year (in the early 2000s).
The diversity of the native bee population is one of its
strengths. Many species of bees participate in pollination,
and the abundance of different species varies year by year.
This diversity allows the native pollinator community to
be both resistant, maintaining functionality in the face of
environmental upheaval, and resilient, able to reestablish
itself in the wake of a destructive event. When the popula
tion of Apis declined dramatically in the second year of the
Costa Rica study, sites close to forest fragments showed
minimal loss of pollination while pollinator visits dropped
by nearly 50% further away. Thus, as well as enhancing
pollination services in conjunction with managed bees,
native bee populations provide important insurance against
29
the possibility that managed bee populations could fail
because of disease, hybridization, or other causes.
Services Provided by Wetlands
Areas inundated by fresh, brackish, and salt water are all
considered wetlands; among many wetland types are fens
and bogs, tidal marshes, riparian zones, and lakeshores.
Wetlands, which cover less than 9% of the Earth’s surface,
can be extremely productive and many are disproportio
nately large providers of ecosystem services. Three of the
key services that wetlands provide are flood mitigation,
water purification, and biodiversity support.
In the upper part of a watershed, many wetlands store
water that flows overland toward rivers and streams.
They can release this water into the main channel slowly,
reducing and delaying flood peaks. Downstream, wet
lands can absorb and reduce peak flood levels, providing
area into which flood waters can spread, dissipating flood
energy by slowing water movement, and removing flood
water through transpiration and infiltration.
The same physical characteristics of wetlands that slow
and absorb overland flow related to flooding can also
provide a mechanism for storing and detoxifying urban
and agricultural wastewater before it discharges directly
into a main channel. Wetlands filter out various nutrients,
other pollutants, and sediment: they support anaerobic
bacteria that denitrify waste; the plants take up and store
nutrients; and by slowing and redirecting water flow, wet
lands enhance sedimentation – the accreting sediments
can effectively bury pollutants. While many wetlands
can purify water very economically, their effectiveness
depends on many factors, including rate of inflow, amount
of sediment and organics in the wastewater, residence time
of wastewater in the wetland, and total surface area.
A wide variety of animals rely on wetlands for survival.
Plant species that deliver flood abatement and water
purification can also support biodiversity, providing
varied food and shelter. A riparian wetland, for example,
might provide food plants and underground burrows for
muskrats; seeds, food plants, and nest building materials
for ducks; and food and shelter for fish and invertebrates.
Wetlands provide a variety of other services as well.
Major products associated with wetlands are peat, timber,
and mulch. Regulating services in addition to flood
mitigation and water purification include waste detoxifi
cation, carbon storage, and control of pests and diseases.
Wetlands provide many cultural services as well,
particularly recreation services such as bird watching,
boating, and hunting. Wetlands also provide key support
ing services, such as soil formation and buffering
freshwater aquifers from saltwater intrusion.
Worldwide, wetlands are estimated to provide many
billions of dollars in services each year. They are recog
nized by the international treaty, the Ramsar Convention
30
Ecosystem Services
on Wetlands, and regulated by domestic law in many
countries. Nonetheless, they have historically undergone
widespread losses in favor of other land uses; worldwide,
50% of wetlands are estimated to have been lost since 1900.
While the services provided by wetlands are widely
recognized, simultaneously maximizing multiple services
may not be possible. In some cases this is related to location:
upland watersheds may be very important for flood control
but may be too far upstream to have an impact on water
purification. In other cases one service may thrive to the
detriment of another: a wetland that is absorbing a heavy
nutrient load may be overtaken by a single, aggressive plant
species and thus fail to be an effective reservoir for biodi
versity. Finally, it can be costly to measure function and
hence difficult to judge how effectively a wetland is per
forming a given service or how to manage for that particular
service.
Services Provided by Forests
Forests provide a wide array of services, such as timber
production, climate stabilization, provision of water quan
tity and quality, and cultural benefits, such as recreation.
Some management options increase the supply of several
services, but often one service is enhanced to the detri
ment of others.
Forests are often managed for provisioning services,
particularly for timber. But even within the category of
provisioning services, management options differ. If a
forest is considered exclusively a supplier of timber,
managers will encourage the growth of only certain
kinds of trees, possibly nonnative fast growing trees, and
will cultivate them so that they grow in a uniform way,
typically straight and tall. When the trees are deemed
mature, they will be cut down, often all at once. By
contrast, if a forest is regarded as a supplier of diverse
benefits, it may be managed to nurture a wide array of
valued species that would not be available in the mono
crop forest described above.
Forests also have both short term and medium term
impacts on climate. Temperature regulation happens in
forests when the canopy shades the ground and when
dark colored foliage absorbs heat. Forests can in certain
circumstances also influence precipitation – in cloud for
ests, for example, trees and epiphytes intercept and
condense water directly from the air, and that water
runs down trunks to plants and soil below. On a longer
timescale, forests play a role in carbon cycling and seques
tration; when forest plants, bacteria, and algae respire,
they take CO2 out of the atmosphere. Plants, soils, and
the animals that eat them in forests, grasslands, and other
terrestrial ecosystems store 2000 billion tons of carbon
worldwide, about half the amount of carbon stored in the
ocean and nearly three times that stored in the atmo
sphere. However, if these ecosystems are burned or
destroyed, as happens when timber is harvested, the car
bon they are sequestering is released to the atmosphere.
Although most organic compounds do return to the atmo
sphere as CO2 when living organisms die and decompose,
in a functioning forest ecosystem some is buried and
sequestered. About 25% of the human caused increase
in CO2 concentration in the atmosphere during the
past 20 years resulted from land use change, primarily
deforestation.
Forests in a watershed, on the hillslopes that drain into a
river, influence the water quality in that river. In part this is
because higher intensity uses, such as agriculture input
pollutants like nutrients and pesticides into a system while
forests do not. Forests themselves also reduce sediment and
nutrient runoff. Clearing trees can have an impact as soon as
the next rainy season on sediment and nutrient loads in
streams, as demonstrated in the classic Hubbard Brook
experiment. In some cases, water users have invested in
forests to keep their water supplies clean. New York City
recently invested US$ 250 million to acquire and protect
land in the Catskills watershed that supplies water to the
city. By working with landowners to reduce pesticide and
fertilizer application and to plant buffer strips along water
ways, New York City reduced potential contamination of
its drinking water. In conjunction with related conservation
investments amounting to US$ 1.5 billion, the city
thereby obviated the need to build a filtration plant pro
jected to cost between US$ 6 and US$ 8 billion.
Forests can also play an important role regulating the
timing and quantity of runoff. The economic value
of forests in the watershed of the Yangtze River above
Three Gorges Dam, in western Hubei Province, Central
China, was quantified in a study published in 2000. Here,
the Gexhouba Hydroelectric Power Plant, the largest
hydro facility in China, producing 15.7 billion kW
annually, requires a narrow range of flows on the Yangtze
in order to run at full power. If the water level is too high,
then water must be released through the sluice gates,
causing the water level below the dam to rise, reducing
the amount of power that can be produced; at very high
flows, turbines are drowned and cannot work at all. If the
water is too low, then generators cannot run at full power.
The goal of the hydroelectric facility’s managers is for
the river to have flow depths that vary as little as possible,
as this has been shown to be much more important for
power generation than the total flow. Upstream forests
damp fluctuations in stream flow by reducing runoff in
wet periods through canopy interception, leaf litter
absorption, and soil and groundwater storage; increased
infiltration provides base flow in dry periods through
groundwater discharge. Though water flow regulation is
a function of vegetation, soil type, and slope, which occur
in a heterogeneous mix through the watershed, forests
and even shrubs with all types of soils and slopes consis
tently provided better water regulation than grasses,
Ecosystem Services
orchards, and crop agricultural fields. This study esti
mated the value of electricity produced by the hydro
facility due to water regulation by the forest at over
US$ 600 000 per year (in the early 2000s), or about 2.2
times the income derived from forest product services in
this area. Because trees lose water to the atmosphere
through transpiration, however, the total water available
downstream was decreased by the forest.
Different management regimes will yield different
suites of services. Some services can never be coproduced;
other services will almost always be produced in tandem,
though often to differing degrees. For the hypothetical
forest illustrated in Figure 2, cattle and timber cannot be
produced on the same parcel of land – conversion to
pasture optimizes livestock but reduces timber output dra
matically. Under timber maximization, once trees are
harvested they are not available for climate or hydrologic
regulation, though before harvest those services will be
produced, as well as some habitat and hiking trails.
Carbon sequestration, hydropower, recreation, and preser
vation of biodiversity tend to be coproduced, but there are
tradeoffs in their optimal supply. Maximizing biodiversity,
for example, produces all four to their fullest extent but
allows for no timber supply. Bringing selective logging
back into the management regime reduces supply of the
other services somewhat; maximizing timber yield reduces
them much more dramatically.
Tradeoffs between services are also tradeoffs between
consumers, such as local recreationalists, regional users of
hydropower, and global beneficiaries of carbon sequestra
tion and biodiversity conservation. These tradeoffs
underscore the importance of valuation, making explicit
who benefits from ecosystem services and who pays for
them. Conceiving of ecosystem functions as services and
assigning a monetary value to them provides a tool for
decision makers to weigh different management options.
Convert to pasture
Maximize timber yield
31
Capturing the Value of Ecosystem
Services
Despite their obvious importance to human well being,
people tend to think of ecosystems as being economically
productive in narrow terms, often assigning value only to
the production of conventional commodities or to real
estate development. Provision of ecosystem services is only
rarely considered in cost–benefit analyses, preparation
of environmental impact statements, or other assessments
of alternative paths of development. There is no shortage of
markets for ecosystem goods (such as clean water and water
melons), but the services underpinning these goods (such as
water purification and bee pollination) often have no mone
tary value. This is in part because ecosystem services are
generally public goods, free to any user, and therefore
difficult to value. Because people mostly do not pay for
them, it can be difficult to discern what the supply, demand,
and willingness to pay for services actually are. As a result,
there are no direct price mechanisms to signal the scarcity or
degradation of these public goods before they fail.
While for some goods and services price reflects value
or importance, when ecosystem services are assigned
monetary value they tend to be priced much lower than
their importance suggests. This is true in part because
when supply is much larger than demand, prices are low,
no matter how necessary the good. The pricing of dia
monds and water is illustrative. Lost in the desert, a
traveler would happily trade all the diamonds in the
world for a single cup of water; back in the marketplace,
our traveler would find that diamonds are many, many
times more costly than water. Water is inexpensive or free
because, like many ecosystem goods and services, it tends
to be far more abundant than the volume demanded by
people; when ecosystems are functioning well, even more
is available.
Maximize biodiversity and
recreation
Diverse portfolio, selective
logging
Management regime
Livestock production
Timber production
Carbon sequestration
Hydropower
Recreation
Preservation of biodiversity
Figure 2 Tradeoffs associated with alternative management objectives for a hypothetical forest ecosystem.
32
Ecosystem Services
Ecosystem services are also often undervalued because
prices are based on current supplies and demands, so the
amount we are willing to pay for continued nutrient
retention in a wetland may be low today even if we can
predict that nutrient laden runoff from increased agricul
ture will threaten a downstream fishery tomorrow.
Further, prices are based on marginal utility – for exam
ple, the amount someone would be willing to pay for the
carbon stored in one more tree in a forest. If that forest is
clear cut, we lose all of the carbon storage and, since the
loss of each tree changes the value of the next, we cannot
account for the whole loss using the price of the first tree.
Precise valuation of ecosystem services is often not
required to provide appropriate economic incentives for
protecting the ecosystems that supply them. Incentives
need only make it more economically appealing to a
landowner to maintain hedgerows as habitat for native
pollinators than to cultivate every last square meter of a
field, for instance, or make it pay to preserve a wetland
rather than filling it to build houses. A farm, as illustrated
below, might generate enough income from nonagricul
tural commodities to alter its land management regime
(Table 2). Incentives to protect and maintain ecosystems
can be provided by the government, privately through
markets, or through hybrid institutions such as cap and
trade systems supported by government policy.
A variety of tools for valuing ecosystem services and
creating incentives for their conservation are currently
being developed, including capital markets such as the
Chicago Climate Exchange, wetland mitigation banks,
and outright payments, often involving private–public
partnerships, for services, as is occurring in Australia,
Costa Rica, and Mexico. These market based approaches
provide a much better indication of value than early,
more theoretical attempts to quantify the value of eco
system services. While valuation is not necessarily a
solution or end in itself, it is a powerful way of organizing
information and an important tool in the much larger
process of decision making.
Conclusions
Because ecosystem services explicitly invoke human
beneficiaries, basic scientific understanding of the eco
system processes producing goods and services is
meaningful only in the context of economic valuation
and institutional structures. There is still much to
learn on many fronts. Important questions include:
Which ecosystems supply which services? What levels
and types of ecosystem protection are required to
sustain service supply? Can we develop robust meth
odologies for the valuation of ecosystems? Even if
clear answers are absent to all of these questions,
numerous and diverse efforts are now underway
worldwide to protect vital ecosystem services, often
using innovative economic incentives.
Explicitly identifying and valuing the goods and
services provided by ecosystems has two obvious benefits.
First, understanding the role of ecosystem services
powerfully justifies habitat preservation and biodiversity
conservation as vital, though often overlooked, policy
objectives. While a wetland surely provides existence
and option values to some people, the benefits provided
by the wetland’s nutrient retention and flood mitigation
services are both universal and undeniable. Tastes may
differ over beauty, but they are in firm accord over the
high costs of polluted water and flooded homes. Second, if
given the opportunity, natural systems can in many cases
quite literally pay their own way. Market mechanisms and
institutions that can capture and maximize service values
can effectively promote environmental protection at the
local, regional, national, and international levels. In some
cases, however, protection of ecosystem services will not
justify conservation of natural habitats. In other cases, the
services will be largely irrelevant to environmental pro
tection efforts. While a focus on ecosystem services
provides great potential to promote environmental pro
tection, its practical implications remain largely
unexamined.
See also: Riparian Wetlands.
Table 2 A hypothetical farm business in 15 years
Commodity
Share of farm business (%)
Wheat
Wool
Water filtration
Timber
Carbon sequestration
Salinity mitigation
Biodiversity
40
15
15
10
7.5
7.5
5
In this model, traditional agricultural commodities account for 55% of
revenues, as opposed to 100% today. Nonagricultural income is
supplied by a mature market for ecosystem goods and services.
Further Reading
Brauman KA, Daily GC, Duarte TK, and Mooney HA (2007) The nature
and value of ecosystem services: An overview highlighting services.
Annual Review of Environmental and Resources 32: 67 98.
Chichilnisky G and Heal G (1998) Economic returns from the biosphere
Commentary. Nature 391: 629 630.
Committee to Review the New York City Watershed Management
Strategy (2000) Watershed Management for Potable Water Supply:
Assessing the New York City strategy. Washington, DC: National
Academy Press.
Daily GC (ed.) (1997) Nature’s Services: Societal Dependence on
Natural Ecosystems. Washington, DC: Island Press.
Daily GC and Ellison K (2002) The New Economy of Nature: The Quest
to Make Conservation Profitable. Washington, DC: Island Press.
Fundamental Laws in Ecology
Daily GC, Soderqvist T, Aniyar S, et al. (2000) The value of nature and
the nature of value. Science 289: 395 396.
Findlay SEG, Kiviat E, Nieder WC, and Blain BA (2002) Functional
assessment of a reference wetland set as a tool for science,
management and restoration. Aquatic Sciences 64: 107 117.
Guo Z (2000) An assessment of ecosystem services: Water flow
regulation and hydroelectric power production. Ecological
Applications 10: 925 936.
Heal G (2000) Nature and the Marketplace: Capturing the Value of
Ecosystem Services, Washington, DC: Island Press.
Heal G, Daily GC, and Salzman J (2001) Protecting natural capital
through ecosystem service districts. Stanford Environmental Law
Journal 20: 333 364.
Kremen C, Williams NM, and Thorp RW (2002) Crop pollination from
native bees at risk from agricultural intensification. Proceedings of the
33
National Academy of Sciences of the United States of America
99: 16812 16816.
Millennium Ecosystem Assessment (2005) Ecosystems and Human
Well being: Current State and Trends: Findings of the Condition and
Trends Working Group. Washington, DC: Island Press.
Postel SL and Thompson BH (2005) Watershed protection: Capturing
the benefits of nature’s water supply services. Natural Resources
Forum 29: 98 108.
Ricketts TH, Daily GC, Ehrlich PR, and Michener C (2004) Economic
value of tropical forest to coffee production. Proceedings of the
National Academy of Sciences of the United States of America
101: 12579 12582.
Zedler JB and Kercher S (2005) Wetland resources: Status, trends,
ecosystem services, and restorability. Annual Review of Environment
and Resources 30: 39 74.
Fundamental Laws in Ecology
S E Jørgensen, Copenhagen University, Copenhagen, Denmark
ª 2008 Elsevier B.V. All rights reserved.
The Need for Fundamental Laws
Systems Ecology in the Jet Stream of Scientific
Development
Systems Ecology: Ten Tentative Fundamental Laws –
An Attempt to Formulate an Ecosystem Theory
Other Ecosystem Theories
Summary
Further Reading
The Need for Fundamental Laws
there is an increasing understanding for the need of
knowledge syntheses to a more holistic image. Today
this search for a holistic understanding of complex sys
tems is considered one of the greatest scientific challenges
of the twenty first century by many scientists.
Several important contributions to systems ecology
have attempted to capture the features and characteristics
of ecosystems, their processes, and their dynamics. The
different theories and approaches look inconsistent at
first glance, but when examined more closely, their com
plementarity becomes evident . This commonality and
consensus regarding ecosystem dynamics was asserted by
Jørgensen in the first edition of Integration of Ecosystem
Theories: A Pattern (1992), and later editions (2nd edn.
1997 and 3rd edn. 2002) have only enhanced the percep
tion that the theories form a pattern and that they are
highly consistent. It is clear from recent meetings and
discussions that today we have a general ecosystem the
ory which is rooted in a consensus of the pattern of
ecosystem dynamics. The ecosystem theory presented
here combines the work of several scientists, and provides
a foundation for further progress in systems ecology,
ecosystem theory, and ecology. Furthermore, it may be
feasible to use a few fundamental laws to derive other
laws to explain most observations. We do not know yet to
Humans have always strived toward finding a structure or
a pattern in their observations – to develop a theory.
Science does not make sense without theory. Without
theory, our observations become only a beautiful collec
tion of impressions without explanation or application to
solve problems of human interest. The alternative to
scientific theory is to observe everything which is not
possible. A well developed theory can be used to make
predictions.
Our scientific knowledge has to be coherent in order to
apply the underlying theory and explain our observations.
Ecology has only partially been able to condense the
systematic collection of observations and knowledge
about ecosystems into testable laws and principles.
During the last few decades systems ecologists have
developed hypotheses that together with basic laws from
biochemistry and thermodynamics are proposed as a
first attempt to formulate fundamental laws in ecology.
The inherent complexity of ecosystems means that it
is necessary to break from the long reductionistic scien
tific tradition to a new holistic ecological approach.
Reductionistic science has had a continuous chain of
successes since Descartes and Newton. Lately, however,
34
Fundamental Laws in Ecology
what extent this is possible in ecology, but at least we
propose a promising direction for a useful, comprehensive
ecosystem theory. Only by the application of the theory
can we assess how and where the theory needs
improvements.
Systems Ecology in the Jet Stream
of Scientific Development
Seven general scientific theories have changed our per
ception of nature radically during the last 100 years:
general and special relativity, quantum theory, quantum
complementarity, Gödel’s theorem, chaos theory, and
theory for far from thermodynamic equilibrium systems.
With these seven theories, we understand today that
nature is much more complex than we thought 100
years ago, but we also have tools to understand this com
plexity better, which has entailed that we have a general
ecosystem theory today.
The speed of light is the absolute upper limit for any
transmission of matter, energy, and information according
to the special relativity theory. This has given a comple
tely new meaning to the concept of locality. It has also in
systems ecology brought another meaning of network:
links among components that share a locality and of the
hierarchical organization: networks of smaller and smaller
localities that are linked together on the next level of the
hierarchy. Relativity theory also gives us a clear under
standing of the lack of absolute measures, which was the
governing scientific perception before the twentieth cen
tury. When we use ecological indicators to assess
ecosystem health, we can only apply them relatively to
other (similar) ecosystems; and, when we use thermo
dynamic calculations of ecosystems we know that we
cannot get the absolute value but only an index or relative
value because ecosystems are too complex to allow us to
include all the components in our calculations. Quantum
theory and later chaos theory upended the deterministic
world picture: we cannot determine the future in all
detail, even if we know all details of the present condi
tions. The world is ontically open. In the nuclear world,
uncertainty is due to our inevitable impact on nuclear
particles, while in ecology the uncertainty is due to the
enormous complexity. Ecosystems are middle number
systems. The number of components in such systems is
many orders of magnitude smaller than the number of
atoms in a room but too many to be countable. Further
complicating the situation is that while the atoms are
represented by a few different types all ecosystem com
ponents are different even among organisms of the same
species. A room may contain 1028 components but they
are represented by only 10 or 20 different types of mole
cules with exactly the same properties. An ecosystem
contains in the order of 1015–1020 different components
all with different individual properties and interaction
potentials. It would be impossible to observe all compo
nents and even more impossible to observe all the possible
interactions among these 1015–1020 different components.
Such complexity leads to a nondeterministic picture in
ecology. In accordance with quantum complementarity,
light can only be described by an interpretation as both
waves and particles (photons). An ecosystem is much
more complex than light. Therefore, a full (holistic)
description of an ecosystem will also, not surprisingly,
require two or more complementary descriptions.
Various descriptions suggest ecosystems as dissipative,
self organizing systems that follow a dynamic to increase
energy, emergy, ascendency (see Ecological Network
Analysis, Ascendency), or eco exergy which are not in
conflict, because they cover different aspects of the eco
system. All descriptions help to understand ecosystem
dynamics, but some may be more applicable for addres
sing specific ecosystem questions.
Gödel’s theorem that there are no complete theories –
they are all based on some assumptions – is of course also
valid for ecological theories. We shall not expect a com
plete theory based on no assumptions and which can be
used in all contexts.
Newtonian Physics is based on the reversibility of all
processes. Prigogine’s new interpretation of the second
law of thermodynamics has shown that time has an
arrow. All processes are irreversible and evolution is
rooted in this irreversibility. Einstein’s special relativity
theory, which provides the speed of light as an upper
speed making it impossible to change the light signals
which give information about a previous event, also
supports the principle of irreversibility. We cannot
change the past but only the future. With the enormous
complexity of ecosystems it also implies that the
same conditions will never be repeated . Ecosystems
are always confronted in space and time with new chal
lenges, which explains the enormous diversity that
characterizes the biosphere. Clearly, systems ecology
has not developed in a vacuum, but has been largely
influenced by the general scientific development during
the last 100 years. A summary of a general ecosystem
theory is presented here. The current proposed theory
consists of ten laws.
Systems Ecology: Ten Tentative
Fundamental Laws – An Attempt
to Formulate an Ecosystem Theory
A tentative ecosystem theory consisting of eight basic
laws has previously been presented, but it seems to be
an advantage to split one of the laws into three due to
some recent results, which are presented below with a few
comments.
Fundamental Laws in Ecology
1. All ecosystems are open systems embedded in an environ
ment from which they receive energy (matter) input and discharge
energy (matter) output. From a thermodynamic viewpoint,
this principle is a prerequisite for ecological processes.
If ecosystems could be isolated, then they would be at ther
modynamic equilibrium without life and without gradients.
This law is rooted in Prigogine’s use of thermodynamics far
from thermodynamic equilibrium. The openness explains,
according to Prigogine, why the system can be maintained
far from thermodynamic equilibrium without violating the
second law of thermodynamics.
2. Ecosystems have many levels of organization and operate
hierarchically. This principle is used again and again when
ecosystems are described: atoms, molecules, cells, organs,
organisms, populations, communities, ecosystems, and the
ecosphere. The law is based on the differences in scale of
local interactions. The distance between components
becomes essential because it takes time for events and
signals to propagate. Ecological complexity makes it
necessary to distinguish between different levels with
different local interactions.
3. Thermodynamically, carbon based life has a viability
domain determined between about 250 and 350 K. It is within
this temperature range that there is a good balance
between the opposing ordering and disordering
processes: decomposition of organic matter and building
of biochemically important compounds. At lower tem
peratures the process rates are too slow and at higher
temperatures the enzymes catalyzing the biochemical
formation processes decompose too rapidly. At 0 K
there is no disorder, but no order (structure) can be
created. At increasing temperatures, the order (struc
ture) creating processes increase, but the cost of
maintaining the structure in the face of disordering
processes also increases.
4. Mass, including biomass, and energy are conserved. This
principle is used again and again in ecology and particu
larly in ecological modeling.
5. The carbon based life on Earth has a characteristic basic
biochemistry which all organisms share. It implies that many
similar biochemical compounds can be found in all living
organisms. They have largely the same elementary com
position, which can be represented using around 25
elements. This principle allows one to identify stoichio
metric relations in ecology.
6. No ecological entity exists in isolation but is connected to
others. The theoretical minimum unit for any ecosystem
is two populations, one that fixes energy and another
that decomposes and cycles waste, but in reality viable
ecosystems are complex networks of interacting popula
tions. This reinforces the openness principle at the scale
of the individual component. The network interactions
provide the environmental niche in which each compo
nent acts. This network has a synergistic effect on the
35
components: the ecosystem is more than the sum of
the parts.
7. All ecosystem processes are irreversible (this is probably the
most useful way to express the second law of thermodynamics in
ecology). Living organisms need energy to maintain, grow,
and develop. This energy is lost as heat to the environ
ment, and cannot be recovered again as usable energy for
the organism. Evolution can only be understood in the
light of the irreversibility principle rooted in the second
law of thermodynamics. Evolution is a step wise develop
ment based on previously achieved solutions to survive in
a changing and dynamic world. Due to the structural and
genetic encapsulation of these solutions, evolution has
produced more and more complex solutions. Eco exergy
expressed by Kullbach’s measure of information (see
Exergy) is one way to measure this development.
8. Biological processes use captured energy (input) to move
further from thermodynamic equilibrium and maintain a state of
low entropy and high exergy relative to its surrounding and to
thermodynamic equilibrium. This is just another way of
expressing that ecosystems can grow. It has been shown
that eco exergy of an ecosystem corresponds to the
amount of energy that is needed to break down the
system.
9. After the initial capture of energy across a boundary,
ecosystem growth and development is possible by (1) an increase
of the physical structure (biomass), (2) an increase of the network
(more cycling), or (3) an increase of information embodied in the
system. All three growth and development forms imply
that the system is moving away from thermodynamic
equilibrium and all three are associated with an increase
of (1) the eco exergy stored in the ecosystem, (2) the
energy flow in the system (power), and (3) the ascen
dency. When cycling increases, the eco exergy storage
capacity, the energy use efficiency, and space–time dif
ferentiation all increase. When the information increases,
the feedback control becomes more effective, the animal
gets bigger, which implies that the specific respiration
decreases, and there is a tendency to replace r strategist
with K strategists. Notice that the first growth form cor
responds to the eco exergy of organic matter, 18.7 kJ g 1,
while the increase of the network plus the increase of the
information correspond to the eco exergy calculated as
( 1)c (see Exergy). Notice also that the three growth and
development forms are in accordance with EP Odum’s
trends of ecosystem development (Table 1). A typical
growth and development sequence is present as follows
(Figure 1): increased biomass (form 1) has a positive
feedback allowing even more additional solar energy
capture, until a limit of around 75% of the available
solar energy is reached. Thereafter the ecosystem con
tinues to grow and develop by increasing network
interactions (form 2) and improving energy efficiencies
(form 3).
36
Fundamental Laws in Ecology
Table 1 Differences between initial stage and mature stage according to Odum (1959 and 1969) are indicated
with reference to the three growth forms
Growth form
Properties
Early stages
Late or mature stage
1 (biomass)
Production/respiration
Production/biomass
Respiration/biomass
Yield (relative)
Total biomass
Inorganic nutrients
>> 1 << 1
High
High
High
Small
Extra biotic
Close to 1
Low
Low
Low
Large
Intra biotic
2 (network)
Patterns
Niche specialization
Life cycles
Mineral cycles
Nutrient exchange rate
Life span
Ecological network
Stability
Ecological buffer capacity
Poorly organized
Broad
Simple
Open
Rapid
Short
Simple
Poor
Low
Well organized
Narrow
Complex
Closed
Slow
Long
Complex
Good
High
3 (information)
Size of organisms
Diversity, ecological
Diversity, biological
Internal symbiosis
Stability (resistance to external perturbations)
Ecological buffer capacity
Feedback control
Growth form
Types
Small
Low
Low
Undeveloped
Poor
Low
Poor
Rapid growth
r-strategists
Large
High
High
Developed
Good
High
Good
Feedback controlled growth
K-strategists
Reproduced by permission of Elsevier.
Exergy captured by the ecosystem
100% incoming solar radiation
Growth form 2 and 3: the network,
energy cycling and information
are increasing
Growth form 1: the biomass is increasing
and is able to capture more solar radiation
Exergy stored in the ecosystem
Figure 1 The development of an ecosystem is illustrated by plotting exergy captured from the inflowing solar radiation toward the
exergy stored in the ecosystem. Growth form 1 is dominant in the first phase of the development from an early-stage ecosystem to a
mature ecosystem. By increasing the biomass the percentage of solar radiation captured increases up to about 80% corresponding to
what is physically possible. Growth forms 2 and 3 are dominant in the intermediate phase and when the ecosystem is in a mature stage.
Thereby more exergy is stored without increasing the exergy needed for maintenance. The system becomes in other words more
effective in the use of the solar radiation according to Prigogine’s minimum-entropy principle. The exergy stored is increased for all three
growth forms. Reproduced by permission of Elsevier.
Fundamental Laws in Ecology
10. An ecosystem receiving solar radiation will attempt to
maximize eco exergy storage or maximize power such that if
more than one possibility is offered, then in the long run the one
which moves the system furthest from thermodynamic equilibrium
will be selected. The eco exergy storage and energy flow
increase during all three growth and development forms –
see above. When an ecosystem evolves it can apply all
three forms in a continuous Darwinian selection process.
The nested space–time differentiation in organisms opti
mizes thermodynamic efficiency as expressed in the tenth
law, because it allows the organism to simultaneously
exploit equilibrium and nonequilibrium energy transfer
with minimum dissipation.
II.
III.
IV.
V.
37
possible to make exact predictions on their develop
ment due to their enormous complexity.
Ecosystems have directionality.
Ecosystems have connectivity.
Ecosystems have emergent hierarchies.
Ecosystems have a complex dynamics (growth and
disturbances).
A special issue of Ecological Modelling (vol. 158) was
devoted to use the proposed ecosystem theory to explain
ecological observations that were unexplained in the eco
logical literature.
Steps toward a wider application of a theoretical explana
tion of ecological observations will reinforce the
fundamentals of ecology. The experience from physics
shows that a theory advances through wider use, because
every application will either support the theory or
improve it by demonstrating where it fails. In conclusion,
a tentative ecosystem theory is proposed which has broad
explanatory power today, but will improve with more
experience providing an even stronger theoretical basis
for ecology.
Other Ecosystem Theories
See also: Exergy; Hierarchy Theory in Ecology.
The ten fundamental laws presented above have been
formulated in a slightly different manner in the scientific
literature and other systems ecologists may emphasize
other aspects. For instance, H. T. Odum could emphasize
maximum power more than eco exergy storage; but, as
pointed out, they are two perspectives of the same basic
dynamic. Such perspectives show that a complex system
should be described by several different viewpoints
according to the complementarity theory. However, this
attempt to provide fundamental laws does not mean that
there are no other candidates in the literature. For exam
ple, the allometric principles are fundamental principles
in ecology. Emergent properties are also sometimes con
sidered sufficiently general to be considered as a
fundamental principle. Other ecologists still withhold
that fundamental laws exist, preferring to focus on
descriptions of fundamental properties and processes.
Therefore, the discussion about which laws should be
considered the fundamental laws in ecology and systems
ecology is still open.
Summary
Advancement in our understanding of ecosystem theories
has led to a tentative consensus of the principle laws in
ecology as outlined above. A priority now is to gain wider
application of the theory and to promote, in general,
ecology as a theoretical science. As such the following
synthesis was recently put forth:
I. Ecosystems are physically and ontically open, mean
ing that they can exchange mass, energy, and
information with the surroundings and that it is not
Further Reading
Elsasser WM (1975) The Chief Abstraction of Biology. Amsterdam:
North Holland.
Fath B, Jørgensen SE, Patten BC, and Strakraba M (2004) Ecosystem,
growth and development. BioSystems 77: 213 228.
Fath BD, Patten BC, and Choi JS (2001) Complementarity of ecological
goal functions. Journal of Theoretical Biology 208(4): 493 506.
Ho MW and Ulanowicz R (2005) Sustainable systems as organisms?
BioSystems 82: 39 51.
Jørgensen SE (1990) Ecosystem theory, ecological buffer capacity,
uncertainty and complexity. Ecological Modelling 52: 125 133.
Jørgensen SE (1995) The growth rate of zooplankton at the edge of
chaos: Ecological models. Journal of Theoretical Biology
175: 13 21.
Jørgensen SE (2002) Integration of Ecosystem Theories: A Pattern, 3rd
edn. Dordrecht, The Netherlands: Kluwer Academic Publishing
Company (1st edn. 1992, 2nd edn. 1997).
Jørgensen SE and Fath B (2004) Application of thermodynamic
principles in ecology. Ecological Complexity 1: 267 280.
Jørgensen SE, Fath BD, Bastianoni S, et al. (2007) Systems Ecology:
A New Perspective, 275pp. Amsterdam: Elsevier.
Jørgensen SE, Patten BC, and Strakraba M (2000) Ecosystems
emerging: 4. growth. Ecological Modelling 126: 249 284.
Jørgensen SE and Svirezhev YM (2004) Towards a Thermodynamic
Theory for Ecological Systems, 366pp. Oxford: Elsevier.
Margalef RA (1968) Perspectives in Ecological Theory. Chicago, IL:
Chicago University Press.
Margalef RA (1995) Information theory and complex ecology.
In: Patten BC and Jørgensen SE (eds.) Complex Ecology, pp. 40 50.
Princeton, NJ: Prentice Hall.
Margalef RA (1997) Our Biosphere,178pp. Nordbunte, Oldendorf,
Germany: Ecology Institute.
Margalef RA (2001) Exosomatic structures and captive energies relevant
in succession and evolution. In: Jørgensen SE (ed.) Thermodynamics
and Ecological Modelling, pp. 117 132. Boco Raton, FL: CRC
Press.
Morowitz HJ (1968) Energy Flow in Biology. Biological Organisation
as a Problem in Thermal Physics,179pp. New York: Academic
Press (see also the review by H.T. Odum, Science 164: 683 684).
Odum EP (1959) Fundamentals of Ecology, 2nd edn. Philadelphia, PA:
W.B. Saunders.
Odum HT (1983) System Ecology, 510pp. New York: Wiley Interscience.
38
Fundamental Laws in Ecology
Odum HT (1996) Environmental Accounting Emergy and Decision
Making, 370pp. New York: Wiley.
Odum HT (1998) Self organization, transformity, and information.
Science 242: 1132 1139.
Patten BC (1991) Network ecology: Indirect determination of the life
environment relationship in ecosystems. In: Higashi M and
Burns TP (eds.) Theoretical Studies of Ecosystems: The Network
Perspective, pp. 288 351. Cambridge: Cambridge University
Press.
Schrødinger E (1994) What is Life? Cambridge: Cambridge University
Press.
Svirezhev YM (2001) Thermodynamics and theory of stability.
In: Jørgensen SE (ed.) Thermodynamics and Ecological Modelling,
pp. 117 132. Boco Raton, FL: CRC Press.
Ulanowicz RE (1986) Growth and Development. Ecosystems
Phenomenology, 204pp. New York: Springer.
Ulanowicz RE (1997) Ecology, The Ascendent Perspective. New York:
Columbia University Press.
ECOSYSTEM PROPERTIES
This page intentionally left blank
Autocatalysis
R E Ulanowicz, University of Maryland Center for Environmental Science, Solomons, MD, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Autocatalysis in Ecology
Centripetality and Agency
Further Reading
Introduction
the bladder. This feeding upon microheterotrophs
helps the Utricularia to grow and increase its surface
area (process A). In nature the surface of Utricularia
plants is always host to a film of diatomaceous algal
growth known as periphyton, so that more surface
area encourages the growth of more periphyton (pro
cess B). More periphyton in its turn means more food
to support the growth of any number of species of
small microheterotrophs (process C). The autocataly
tic cycle is closed when it is noted that a greater
density of microheterotrophs provides more resources
for the Utricularia to grow (process A again) by cap
turing and absorbing more abundant zooplankton
(Figure 2b).
Unlike in chemistry, the actors in ecology are more
complex, malleable entities with capabilities to undergo
small, incremental alterations. Such malleability sub
stantially enhances the repertoires of autocatalysis and
enables it to exhibit some very nonmechanical beha
viors. This is especially the case when autocatalysis
involves processes that can change in stochastic and
nonpredictable ways. An important characteristic of
causal cycles (e.g., autocatalysis) is that when random
events impinge upon them, they usually yield nonran
dom results. This is the consequence of the first and
foremost attribute of autocatalysis – its generation of
selection pressure.
To see how autocatalysis generates selection, one
begins by considering a small spontaneous change in
process B. If that change either makes B more sensi
tive to A or a more effective catalyst of C, then the
transition will receive enhanced stimulus from A. In
the Utricularia example, diatoms that have a higher P/
B ratio and are more palatable to microheterotrophs
would be favored as members of the periphyton com
munity. Conversely, if the change in B makes it either
less sensitive to the effects of A or a weaker catalyst of
C, then that perturbation will likely receive dimin
ished support from A. Hence, the response of this
causal circuit is decidedly not symmetric, and out of
this asymmetry emerges a direction. This direction is
not imparted or cued by any externality; its action
resides wholly within the system. As one might expect
from a causal circuit, the resulting directionality is in
In chemistry the term catalysis means the speeding up of a
chemical reaction. It follows that autocatalysis then means
‘‘the catalysis of a chemical reaction by one of the pro
ducts of the reaction.’’ For example, oxalic acid oxidizes
purple permanganate. When a few crystals of MnSO4
are added to a mixture of the chemicals, the conversion
to Mn(II) is sped up. If no MnSO4 is added, then the
reaction will gradually speed up of itself, because Mn(II)
is gradually being created by the reaction, and this pro
duct autocatalyzes the reaction itself. Autocatalysis in
chemistry is usually considered to occur among relatively
simple, fixed, and inflexible reactants. As such it is com
monly regarded as a subclass of general mechanisms.
Autocatalysis in Ecology
In systems ecology, autocatalysis is regarded as a general
ized form of mutualism, that is, an association between
organisms of two different species in which each member
benefits. In systems ecology focus remains more on pro
cesses and less on objects. Hence, an autocatalytic
configuration of two or more ecological processes is one
in which the processes can be arrayed in a closed cycle,
wherein each process in the cycle facilitates the next.
Without loss of generality, one may focus on a serial,
circular conjunction of three processes – A, B, and C
(Figure 1). Thus, any increase in the rate of process A is
likely to induce a corresponding increase in process B,
which in turn elicits an increase in process C, and whence
back to A.
A didactic example of autocatalysis in ecology is
the community that builds around the aquatic macro
phyte, Utricularia (commonly called Bladderwort). All
members of the genus Utricularia are carnivorous
plants. Scattered along its feather like stems and
leaves are small bladders, called utricles (Figure 2a).
Each utricle has a few hair like triggers at its term
inal end, which, when touched by a feeding
microheterotroph, opens the end of the bladder, and
the animal is sucked into the utricle by a negative
osmotic pressure that the plant had maintained inside
41
42
Autocatalysis
A
+
+
+
C
B
with teleology. There is no externally determined or
preexisting goal toward which the system strives.
Direction arises purely out of the immediate response
by the internal system to a novel, random event
impacting one of the autocatalytic members.
Figure 1 Schematic of a hypothetical three-component
autocatalytic cycle.
Centripetality and Agency
(a)
(b)
Zooplankton
Utricle
n
yto
iph
r
layer
Pe
ton
P
Leaflet
er
iph
y
Figure 2 (a) Sketch of a typical ‘leaf’ of Utricularia floridana,
with detail of the interior of a utricle containing a captured
invertebrate. (b) Schematic of the autocatalytic loop in the
Utricularia system. Macrophyte provides necessary surface upon
which periphyton (speckled area) can grow. Zooplankton
consumes periphyton, and is itself trapped in bladder and
absorbed in turn by the Utricularia.
part tautologous, that is, autocatalytic systems respond
to random events over time in such a way as to
increase their degree of autocatalysis. It should be
emphasized that this directionality, by virtue of its
internal and transient nature, should not be conflated
A second important and related directionality emerges
out of autocatalysis – that of centripetality. To see this
one notes in particular that any change in B is likely to
involve a change in the amounts of material and energy
that are required to sustain process B. As a corollary to
selection pressure one immediately recognizes the ten
dency to reward and support any changes that serve to
bring ever more resources into B. Because this condi
tion pertains to any and all members of the causal
circuit, any autocatalytic cycle becomes the epicenter
of a centripetal flow of resources toward which as many
resources as possible will converge (Figure 3). That is,
an autocatalytic loop defines itself as the focus of
centripetal flows. A didactic example of such centripe
tality is coral reef communities, which by their
considerable synergistic activities draw a richness of
nutrients out of a desert like and relatively inactive
surrounding sea.
The centripetality generated by autocatalysis is a
much neglected and essential attribute of the life pro
cess. For example, evolutionary narratives are replete
with explicit or implicit references to such actions as
striving or struggling, but the origin of such directional
behaviors almost always remains unmentioned. Such
actions are simply postulated. Centripetality, however,
appears to be at the very roots of such behaviors. To see
this, one only needs to recognize that it is centripetality
that gives rise to the much vaunted competition, which is
the crux of evolutionary theory. For centripetality guar
antees that, whenever two or more autocatalytic loops
Centripetality
Figure 3 Centripetal action as engendered by autocatalysis.
Autocatalysis
(a)
(b)
A
C
(c)
A
B
43
C
A
B
C
D
D
Figure 4 (a) Original configuration. (b) Competition between component B and a new component D, which is either more sensitive to
catalysis by A or a better catalyst of C. (c) B is replaced by D, and the loop section A–B–C by that of A–D–C.
exist in the same system and draw from the same pool of
finite resources, competition among the foci necessarily
ensues. In particular, whenever two loops share pathway
segments in common, the result of this competition is
likely to be the exclusion or radical diminution of one of
the nonoverlapping sections. For example, should a new
element D happen to appear and to connect with A and
C in parallel to their connections with B (Figure 4), then
if D is more sensitive to A and/or a better catalyst of
C, the ensuing dynamics should favor D over B to the
extent that B will either fade into the background or
disappear altogether. That is, the selection pressure
and centripetality generated by complex autocatalysis
(a macroscopic ensemble) is capable of shaping and repla
cing its own elements. Perhaps the instances that spring
most quickly to mind here involve the evolution of obli
gately mutualistic pollinators, such as yuccas and yucca
moths, which coevolve with the yucca so as to displace
other pollinators.
One notes in passing that the same tendency to replace
B with D could as readily replace a defective or destroyed
B with another similar component B9, that is, autocatalysis
lies behind the ability of living systems to repair themselves.
It becomes obvious that the autocatalytic system is no
longer acting merely at the behest of externalities, but it is
actively drawing ever more resources unto itself. In fact,
the tendency of centripetality to transform as much as
possible into itself lies at the very crux of evolutionary
drive; for absent such striving, there would be no compe
tition at the next level up.
Furthermore, one perceives autocatalytic action as the
agency behind one of a pair of agonistic tendencies that
together account for the patterns of life forms and
functions. On the one hand is the stochastic, entropic
tendency to fall apart, which at the same time generates
new diversities of form and behavior. Arrayed against the
inevitable centrifugal drift toward disorder is the autoca
talytic selection and centripetal pull toward greater
activity and tighter organization. Opposing thrusts though
they are, the continued development of life would be
impossible without the actions of both.
Finally, the focal position that autocatalytic configura
tions of processes occupy in the phenomenon of life is
aptly illustrated by considering what differs between a
living organism (say a deer) and the same entity immedi
ately upon death. The mass of the deer remains the same,
as does its overall form, chemical constitution, embodied
energy, and genomic configuration. What the live deer
had that the dead deer no longer possesses is simply its
configuration of autocatalytic processes.
See also: Ecological Network Analysis, Ascendency;
Ecological Network Analysis, Energy Analysis.
Further Reading
Eigen M and Schuster P (1979) The Hypercycle: A Principle of Natural
Self Organization. Berlin: Springer.
Kauffman SA (1995) At Home in the Universe: The Search for Laws of
Self Organization and Complexity. New York: Oxford University
Press.
Ulanowicz RE (1995) Utricularia’s secret: The advantages of positive
feedback in oligotrophic environments. Ecological Modelling
79: 49 57.
Ulanowicz RE (1997) Ecology, the Ascendent Perspective. New York:
Columbia University Press.
44
Body-Size Patterns
Body-Size Patterns
A Basset and L Sabetta, Università del Salento – Lecce, Lecce, Italy
ª 2008 Elsevier B.V. All rights reserved.
Background
The Problem of Measuring Body Size
Population and Species-Level Patterns
Community Level Patterns
Decoding Mechanisms
Further Reading
Background
trophic links and competitive ranking among co occur
ring species differing in body size. However, the simple,
energy related, ‘taxon free’ explanation emphasizes the
ecological relevance of body size patterns, which are
conceptually independent of species specific resource
requirements and species composition.
Body size patterns include population or species level
patterns, such as the body size range pattern, and com
munity, landscape, or continental level patterns. The
latter include variations with individual body size in
number and biomass of individuals, number of species,
population densities, and energy used by populations.
The body size ratio between co occurring species pairs,
known as the Hutchinson ratio, also shows deterministic
and consistent patterns of variation within the commu
nity level body size patterns.
Life in the biosphere shows an impressive variety of
individual shapes and body sizes. From the smallest
microorganism (approx. 10 13 g) to the largest mammal
(>108 g), living things cover more than 21 orders of mag
nitude of body size. The largest living organisms are
actually plants (giant sequoia, Sequoiadendron giganteum
(Lindl.) Buchholz), but since most of their bodies are
actually dead bark tissues, their living biomass is lower
than that of the largest mammals. Given this impressive
variability of sizes, consistent body size patterns, so com
mon at every scale of observation as to be considered
universal, can be detected.
The first body size patterns to be emphasized were
that there are many small and few large individuals and
species in the biosphere. The range of body sizes from
the smallest to the largest individuals may vary sub
stantially, when moving from marine to brackish water
to freshwater and terrestrial ecosystems, as well as from
tropical to polar ecosystems or from lowlands to high
lands, but the pattern of many small and few large
individuals still holds. This simple and universal obser
vation was reported by Charles Elton in the first half of
the twentieth century in his pivotal book Animal Ecology.
This pattern can be explained by means of simple,
‘taxon free’, energy related arguments: since small indi
viduals require less energy per unit of time than large
individuals for their maintenance and activity, a fixed
productivity will support, at equilibrium, a higher den
sity of small than large individuals. This explanation is
actually an oversimplification of the real world; there
are at least two other components that need to be taken
into account in order to decode the body size abun
dance patterns into a deterministic mechanism of
community organization: a phylogenetic and evolution
ary component, determining the actual diversity of
species and body sizes at continental and global scales;
and an interaction component, selecting the body sizes
and the species best suited to withstand the locally
occurring abiotic conditions and structural habitat
architecture (abiotic niche filtering), and determining
The Problem of Measuring Body Size
Measurements of body size include linear dimensions
(e.g., body length, body width, and the length or width
of some morphological attribute of individuals), body
surface area and biovolume, and weight (e.g., wet
weight, dry weight, and body mass as ash free dry
weight). The energy content of biomass, measured in
energy units, can also be used as a measurement of body
size.
Body size patterns are derived from individual bio
mass data, although the original data may have been
obtained as the body length, morphological attribute
length or size, body wet weight, body volume, or cell
volume of unicellular individuals. Some conversion is
required, because individual biomass cannot always be
measured, although body size in general is an easily
measurable characteristic of individuals. Indeed, in
many cases it is necessary to avoid the destructive analysis
required to measure individual biomass, and in other
cases individuals are simply too small.
Indirect measurements of individual biomass, where
lengths or biovolumes are converted to weights, may
make the body size patterns weaker or harder to detect,
Body-Size Patterns 45
depending on the dimensions measured, on the preci
sion of the allometric relationship used with respect to
the specific set of data, on the precision of biovolume
detection, and on the adequateness of the conversion
equations. As regards the weight per length allometry,
the comparability of the seasonal period, climatic con
ditions, sex ratio and the reproductive status of
individuals, and resource availability all have to be
taken into account as major sources of variation. As
regards biovolume, the complexity of individual or cell
shape, taxon specific weight per unit of biovolume, and
the type of weight unit used (C, biomass) all have to be
taken into account in order to minimize the bias intro
duced by using indirect measurements and conversion
factors.
Population and Species-Level Patterns
Range-Size Patterns
The range of a species is its natural area of geographic
distribution. Considering the overall range of species and
body sizes occurring in the biosphere, there does not seem
to be any simple and deterministic relationship between
body size and species range size: very large species, such
as some cetaceans, and very small species, such as many
microorganisms, can have very wide natural ranges.
However, within much more restricted taxonomic
groups, small bodied species tend to have smaller mini
mum geographic ranges than large bodied species. The
interspecific relationships of body size to geographic
range size commonly exhibit an approximately triangular
form, where species of all body sizes may have large
geographic ranges while the minimum range size of a
species tends to increase with body size.
The relationship between body size and home range
size (i.e., the minimum space needed by an individual to
successfully complete its life cycle) can help to account
for patterns of natural range size. Since home range size
(H ) scales with individual body size (BS) according to an
allometric equation (H ¼ aBSb), in which the slope (b) is
significantly larger than 1, large bodied species may
require a larger total geographic range than small species
in order to maintain minimum viable population sizes in
all local areas. This results in the triangular relationship
between body size and range size, because there is not
necessarily an upper limit on the range size of small
bodied species.
The dependence of a species’ fundamental niche
space and dispersal ability on body size may also help
to explain range size patterns, since species of large body
size are potentially able to maintain homeostasis in a
wider range of conditions and to successfully colonize a
larger proportion of their potential range than small
bodied species.
These mechanistic explanations of the relationship
between range size and body size are not mutually exclu
sive and may be reinforcing.
Community Level Patterns
Body Size–Abundance Distributions
Body size–abundance distributions describe the variation
of some measurements of individual abundance with
individual body mass. The measurements of abundance
used are number and biomass of individuals of each
population within a guild or a community, number or
biomass of individuals in successful populations within a
species range, at the regional, continental, and global
scale, number or biomass of individuals within a com
munity and number or biomass of individuals in base 2
logarithmic body size classes. Whatever criteria for
grouping individuals are selected, within populations,
communities or size classes, at the guild, community,
landscape, continental or global scale, as number or
biomass, a negative relationship between individual
density and body size is generally observed. However,
the shape and coefficient of these relationships, the
mechanisms involved and the ecological significance
vary according to the criteria selected, and each single
body size pattern provides different information, contri
buting to a better understanding of the role of individual
body size in structuring and organizing ecological
communities.
The selection of either species populations or body
size classes as a grouping criterion creates two main
categories of size–abundance distributions: ‘taxonomi
cally based and nontaxonomically based’. The latter are
commonly referred to as ‘size spectra’. Studies of terres
trial ecosystems have preferentially used body size–
abundance distributions based on the taxonomic grouping
of individuals into populations and communities, whereas
studies of aquatic ecosystems have preferentially used
body size–abundance distributions as ‘taxon free’ pat
terns, grouping individuals into logarithmic body size
classes independently of their taxonomy.
Taxonomically based size–abundance
distributions
On average, population densities (PDs) scale with indivi
dual body size (BS) according to the allometric equation
PD ¼ a1 BSb1
where b1 is typically lower than 0 and a1 is the specific
density. a1 expresses the combined action of factors such
as average energy transfer efficiency, average energy
availability, and temperature driven shifts in the meta
bolic rates of the populations in question.
46
Body-Size Patterns
Broadly speaking, taxonomically based size–abundance
distribution derives from the notion that since the energy
requirement of individuals (Met) increases with individual
body size according to a well known allometric equation
Met ¼ a2 BSb2
where b2 has been consistently found to be close to 0.75,
the number of individuals of each population supported
by the available resources must decrease with average
individual body size. Assuming that resource availability
is homogeneous across species and body sizes, the slope of
the body size–abundance distribution (b1) is expected to
be –0.75.
The processes underlying body size–abundance dis
tributions, and hence their information content and
ecological meaning, depend on whether they account
for density values and average body sizes of species on a
regional, continental, or global scale (hereafter, ‘global
scale size–abundance distributions’), for density and aver
age individual body size of co occurring populations
within guilds or communities (hereafter, ‘local scale
size–abundance distributions’), or for average population
densities and individual body sizes of entire guilds or
communities along ecological, climatic, or biogeographic
gradients (hereafter, ‘cross community size–abundance
distributions’).
Global scale size–abundance distributions are among
the most extensively studied. They cover regional, con
tinental, and global scales, and the broadest range of
taxonomic variation, with a bias toward birds and mam
mals, for which more extensive databases of population
densities and body sizes are available at every spatial
scale. Data used to compile global scale size–abundance
distributions typically describe densities of successful
populations within the species’ geographic range, which
may be close to the maximum carrying capacity. Most
commonly, populations included in the global scale size–
abundance distributions do not coexist, and affect each
other through vertical or horizontal interactions. For
large compilations of population densities, population
density generally scales very closely with body size,
with a slope near the value of –0.75. The close agreement
between the slope observed for global scale size–abun
dance distributions and that expected on the grounds of
simple energetic arguments confirms that at the continen
tal and global scales, availability of resources or energy is
not correlated with species body size. The homogeneity
of resource or energy availability across species body sizes
is an interesting, but far from straightforward, aspect of
global size–abundance distributions. It implies that the
advantage for large species arising from their wider niches
(and thus greater availability of resources) with respect to
small species, is counter balanced by the presence of
other body size related factors which compensate.
These include the resource density perceived by indivi
duals and the individuals’ exploitation efficiency, both of
which decrease with increasing body size. Intercepts of
global size–abundance distributions express the average
energy use efficiency of the group of populations consid
ered. Compilations of global size abundance distributions
for ectothermic and endothermic species show different
intercept (a1) values, the former having less negative
intercepts than the latter due to the cost of being homoeo
thermic. Similarly compilations of size abundance
distributions of herbivores have higher a1 values than
those obtained for carnivores, reflecting the overall effi
ciency of energy transfer in food webs.
When size–abundance distributions are compiled at
the local level, where the body size and abundance of
each species (N ) is measured at the same location, body
size generally explains only a small part of the variation in
population abundance, and the regression slope is much
higher than the expected –0.75. The observed deviations
from the expected slope in local size–abundance distribu
tions are suggestive of size biases in resource acquisition
that could be driven by size asymmetry in competition.
An alternative hypothesis to explain the deviation of local
size–abundance distributions from global ones is that the
former typically examines a smaller range of sizes than
the latter. Observing a smaller portion of the overall
relationship accentuates the noise in the local sample.
This could explain why local size–abundance distribu
tions in aquatic environments, covering a larger spectrum
of body sizes than terrestrial ones, are also generally
stronger. In fact, at the local scale, triangular shaped
size–abundance distributions are much more commonly
observed than simple allometric relationships. Triangular
distributions have three major attributes: an ‘upper
bound’, a ‘lower bound’, and a dispersion of points in the
size–abundance space (Figure 1a). The ‘upper bound’ of
the triangular shaped size–abundance distributions is
determined by the body size scaling of the dominant
species’ population densities. The ‘upper bound’ has
been used as a proxy of the complete local size–
abundance distribution, under the assumption that the
ecological role of rare and occasional species, being
weak, is unclear. The procedure may be useful for applied
purposes but since most species are rare the assumption is
not generally acceptable. The body size dependency of
the minimum viable population may explain an expected
‘lower bound’, which is difficult to measure because of the
problems with correctly quantifying the rarity of popula
tions. The density of points between these two bounds is
determined mainly by regional processes and horizontal
and vertical partitioning rules. The ecological informa
tion carried by the intercepts of the size–abundance
distributions is of lower value at local than at global
scale because whenever slopes are different, as often
Body-Size Patterns 47
BS and Ntot. In general cross community size–abundance
distributions tend to be well described by allometric
equations, whose slopes tend to be similar to the inverse
of the scaling exponent of metabolic rates with individual
body size. A similarity between observed and expected
slopes has been also detected in guilds and communities
which are not regulated by self thinning rules, such as
bird and phytoplankton guilds. However, since much
fewer data are available for cross community size–
abundance distributions than for global and local size–
abundance distributions, the underlying mechanisms
remain to be determined.
Species abundance (log(no. m–2))
(a)
10
8
6
4
2
0
–2
–4
–6
0
10
15
5
Body size (log(μg))
20
Number of species
(b)
50
Nontaxonomically based size–abundance
distributions
40
30
20
10
0
1 3 5 7 9 11 13 15 17 19 21
Body size (log(μg))
(c)
185 taxa
46 taxa
19 taxa
Body size (log(mg))
1000.00
100.00
10.00
1.00
0.10
0.01
0.00
0
50
100
150
Species rank
200
250
Figure 1 Body-size patterns of macroinvertebrate guilds of
transitional water ecosystems in the Mediterranean and Black
Sea Eco-regions. Both local size–abundance distributions (a) and
body size–species distributions (b) are triangular shaped. The
‘upper bounds’ of the triangular distributions are reported. The
graph (c) emphasizes that species of transitional water
macroinvertebrates are clumped around the mode of the body
size–species distribution, with 74% of the species being grouped
in 2 out of the 5 order of magnitudes occurring between the size
of the smallest and the largest species.
occurs when comparing local size–abundance distribu
tions, comparisons between intercepts are not possible.
Classifying all the individuals in a population into
guilds or communities and averaging their mass, we may
then describe every guild and community with two sim
ple parameters: mean organism size (BS) and total
community abundance (Ntot). The scaling of total com
munity abundance with mean organism size leads to
cross community size–abundance distributions. Cross
community size–abundance distributions were first stud
ied in self thinning plant and sessile communities, where,
as organisms grow, there is space for fewer and fewer
individuals, determining a negative relationship between
In large aquatic ecosystems, early studies of body size–
abundance distributions focused on energy transfer (i.e.,
how information on productivity and energy transfer may
be gained from body size data, which can be collected
relatively easily). In accordance with this objective, they
dealt with particles rather than with species, dividing
particles suspended in the water column into logarith
mic base 2 size classes, irrespective of species and
including nonliving organic particles. Thus the ni parti
cles in the ith body size class of average mass BSi may
represent more than one species, and every species can
occur in more than one class. Nontaxonomic size–
abundance distributions (hereafter referred to as size
spectra) have been quantified for many different guilds
and communities, including plankton, benthos and fish
guilds, woodland and forest plant guilds, as well as mar
ine, freshwater, and terrestrial ecosystems; however, a
large proportion of the ecological literature addressing
size spectra deal with the pelagic marine environment.
According to the classification reported for taxonomi
cally based body size–abundance distributions, almost all
size spectra are local, being determined at the guild or
community scale. Size spectra can be compiled with two
different types of data, that is, biomass and number of
individuals. Both biomass size spectra and number size
spectra can cover different body size ranges, describing
either entire communities or single guilds.
Regarding biomass size spectra, the amount of biomass
has been shown both empirically and theoretically to be
constant when plankton individuals are organized into
logarithmic size classes. As a result of this equal partition
ing of biomass, the slope of a straight line fitted to
plankton biomass size spectra is expected to be 0; this
relationship is known as the ‘linear biomass hypothesis’,
which has strong experimental support in aquatic pelagic
environments, particularly when a large spectrum of sizes
and trophic levels are considered. Often the data is sub
jected to a normalization procedure, which consists of
dividing the biomass in each size class by the width of
the size class. In normalized biomass size spectra, biomass
48
Body-Size Patterns
in each size class decreases isometrically with the average
class size, the slope being close to –1. The linear biomass
hypothesis implies that in pelagic systems, the number of
individuals within logarithmically increasing size classes
declines linearly with average body size. The slope of the
allometric equation tends to be close to –1; when number
size spectra are normalized, the expected slope is equal
to –2. Nevertheless, within pelagic size spectra, a series of
dome like distributions are typically detected, corre
sponding mainly to different functional guilds within
which there is a poor fit with linear statistical regressions.
‘Dome like’ distributions and gaps in number and
biomass size spectra occur not only between but also
within functional groups, such as phytoplankton and zoo
plankton, even when they are not attributable to
incomplete censuses of species or to systematic under
estimation of intraspecific size variation. Dome like
patterns of biomass distribution have been observed
both in freshwater and marine ecosystems, as well as in
macro zoobenthos and fish. Therefore, by restricting the
range of body size considered and addressing specific
functional groups, size spectra tend to have a shape simi
lar to the triangular shape of local size–abundance
distribution. Most commonly, the maximum number
and biomass of individuals, either partitioned into species
or irrespective of species, occur at some small but inter
mediate body size, rather than at the smallest size.
Two kinds of scaling in the relationship between body
size and abundance within size spectra may be recog
nized. A unique and primary slope reflecting the size
dependency of metabolism (‘metabolic scaling’), and a
collection of secondary slopes which represent the scaling
of numerical or biomass abundance with body size within
groups of organisms having similar production efficien
cies (‘ecological scaling’). Size dependent coexistence
relationships are likely to be representative of the second
ary slopes, leading to a dominance of large cells/species,
and slopes that are less negative than predicted by the
‘linear biomass hypothesis’. Ecological scaling can also
produce dome like patterns in size spectra within the
size range of each functional group.
Body Size–Energy Use Distributions
The body size dependence of both metabolic rates and
population densities makes it possible to evaluate popula
tions’ rates of energy use and how they scale with
individual body size. Indeed, the rate at which energy
flows through a population (E ) can be evaluated as the
product of individual metabolism (Met) and population
density (PD), as follows:
E ¼ Met PD ¼ a2 BSb2 a1 BSb1 ¼ ða2 a1 ÞBSðb2 þb1 Þ
Since b2 has been found to be consistently close to 0.75,
the scaling of energy use rates with individual body size
depends on b1, which is generally expected to be negative,
since, at every spatial scale of ecological organization,
many small and few large individuals occur.
Assuming that resource availability is homogeneous
across species and that species do not limit each other’s
resource availability and have optimized the efficiencies
of resource exploitation and use, then population densi
ties are expected to scale with individual body size with a
slope (b1) of –0.75, and the amount of energy each species
uses per unit of area is expected to be independent of
body size:
E ¼ ða2 a1 ÞBSð0:75 – 0:75Þ ¼ a3 BS0
The independence of energy use per unit area from
body size is known as the energetic equivalence rule
(EER). Whenever b1 is consistently lower, more negative,
than –0.75, small species dominate energy use.
Conversely, if b1 is consistently larger, less negative,
than –0.75, large species make a disproportionately large
use of the available energy per unit of area.
Global size–abundance distributions seem to agree
with the EEF. At the global scale, the energy use of
the most successful populations within the species range
seems to be actually independent of the body size of
individuals within populations. On the other hand, local
size–abundance distributions, which commonly show
scaling exponents higher, less negative, than –0.75 show
that within local guilds and communities large species
normally dominate energy use. Dominance of small spe
cies has also been detected at the local scale usually in
relation to some degree of stress. Therefore, the shape and
slope of local size–abundance distribution, and conse
quently the body size scaling of energy use, can have
practical applications in ecology.
Body-Size–Species Distributions
Understanding biodiversity is a major goal of ecology.
Since many small and few large species occurs in the
biosphere, at every scale, from the community to the
continental and global level, describing and understand
ing the scaling of biodiversity patterns with individual
body size is also a key topic.
Basically, whenever organisms perceive a two
dimensional (2D) habitat, they sample habitats on a grid
proportional to the reciprocal of the square of their linear
dimension (L). Therefore, the likelihood of niches being
opened up to species specializing in particular resources
and habitat patches is proportional to L 2 or to BS 0.67.
Consequently, the number of species (S) is expected to
decrease with individual body size according to L 2.
Whenever organisms perceive a 3D habitat, the species
Body-Size Patterns 49
number is expected to be proportional to L 3 or BS 1.
Considering that the linear dimension of individuals
represents the ‘ruler’ (L) they use to sample the habitat
and that habitats are rarely completely homogeneous at
every scale of perception, individuals perceive the habitat
to be fractal. The perceived 2D habitat scale is L 2D,
where D is the fractal dimension of the habitat as well as
of the resources. The fractal dimension is a habitat prop
erty that in many field studies has been found to be close
to 1.5. This would mean that in 2D habitats the number of
species (S) is expected to be between L D and L 2D, that is,
between L 1.5 and L 3.0, where L 2D is analogous to the
‘upper bound’ of size–abundance distributions. In 3D
habitats the number of species (S) is expected to be
proportional to L 3D, that is, to L 4.5. Assuming D 1.5,
a tenfold decrease in individual size determines a three
fold increase in the perceived length of each habitat edge,
a tenfold increase in apparent habitat surface and a max
imum tenfold increase in S.
Available data on both the full range of taxa and
particular groups of species consistently show that the
species–size distributions are humped, with the mode in
some small but intermediate size class (Figure 1b). The
underestimation of the number of existing small species
may be an explanation of humped distributions covering
the whole scale of size from the smallest to the largest
species. Underestimation of small species is less likely to
explain humped distributions observed within restricted
taxonomic groups, such as invertebrates, birds, and mam
mals. Within restricted taxonomic groups, it seems likely
that an optimal body size exists, where species and indi
viduals perform optimally and tend to be clumped
(Figure 1c). An optimal body size of between 100 g and
1 kg has been proposed for mammals and an optimal body
size of 33 g has been proposed for birds. Two hypotheses
have been proposed to explain the size dependency of
species performance at every scale: the energy conversion
hypothesis, addressing optimal size according to the size
dependency of the efficiency of energy conversion into
offspring; and the energy control hypothesis, addressing
optimal body size according to the species’ performance
in monopolising resources.
Body-Size Ratios
Coexisting species of potential competitors commonly
differ in body size. Using consumer body size as a proxy
of resource size, this difference may explain competitive
coexistence; in his famous paper ‘Homage to Santa
Rosalia: or why are there so many kind of animals’
Hutchinson proposed that in order to coexist species
must be spaced in size with a ratio between their linear
dimensions of at least 1.28 (2.0–2.26 in biomass), which is
commonly referred to as the ‘Hutchinson ratio’. Patterns
of body size spacing between coexisting species pairs
consistent with the ‘Hutchinson ratio’ have been observed
for many groups of animals, including, birds, desert
rodents, and lizards. The ‘Hutchinson ratio’ corresponds
to a limiting similarity threshold; therefore, the average
size ratio between species is expected to vary with
resource limitation. Actually, the average size ratio
between species pairs decreases with increasing richness
and with decreasing guild trophic level.
That co occurring species within guilds tend to have
different body size, with an average size ratio close to the
expected 2–2.26, is a very general observation in ecology.
However, the ecological relevance of the ‘Hutchinson
ratio’ has been questioned, mainly due to two key critical
observations, apparently in contrast with the interpreta
tion that size ratios between species correspond to a low
enough niche overlap between species pairs to allow
interspecific coexistence: (1) many nonliving things in
nature as well as many objects built by humans, from
nails to musical instruments, are scaled in size according
to the ‘Hutchinson ratio’; and (2) size spacing between
species pairs does not always seem to be related to niche
spacing. The latter is the most critical issue. However, a
functional link between the body size ratios of coexisting
species and competitive coexistence conditions may also
be derived independently of any niche spacing. Body
size mediated coexistence between species differing in
size may result from simple energetic constraints on indi
vidual space use regardless of any a priori resource
partitioning: that is, size ratios between species may cri
tically affect species coexistence even if niche spacing is
not detected.
Decoding Mechanisms
At their most simple, body size patterns depend on
phylogenetic and evolutionary constraints, on energetic
constraints, and on interactions with the habitat structure
and with co occurring species.
As regards phylogenetic and evolutionary constraints,
body size patterns are in some way dependent on existing
biodiversity and its evolutionary basis. Each taxon per
forms best under a fixed set of conditions and has
bioengineering constraints on its performance. For exam
ple, insects cannot be too large and birds cannot be too
small; therefore, although both of them can take advan
tage of a 3D space, the complete spectrum from insects to
birds has ‘dome like’ distributions which incorporate the
bioengineering constraints of the two groups of species.
As regards energy constraints, metabolic theory gives a
general explanation of body size patterns in terms of
energy and temperature constraints on metabolism and
the intrinsic properties of energy partitioning. Metabolic
theory sets the theoretical expectations of body size pat
terns, under the assumption that they are basically driven
by simple energy constraints
50
Cycling and Cycling Indices
As regards interactions, clearly populations interact
with their environment and with co occurring species.
‘Textural habitat architecture’ and ‘body size mediated
coexistence’ hypotheses have been proposed to explain
the abiotic and biotic components of interaction regarding
its influence on the observed body size patterns.
Further Reading
Brown JH, Gillooly JF, Allen AP, Savage VM, and West GB (2004)
Towards a metabolic theory of ecology. Ecology 85: 1771 1789.
Brown JH and West GB (2000) Scaling in Biology. Oxford: Oxford
University Press.
Elton C (1927) Animal Ecology. London: Sidgwick and Jackson.
Gaston K (2003) The Structure and Dynamics of Geographic Ranges.
Oxford: Oxford University Press.
Holling CS (1992) Cross scale morphology, geometry and dynamics of
ecosystems. Ecological Monographs 62: 447 502.
Hutchinson GE (1959) Homage to Santa Rosalia, or why are
there so many kinds of animals? American Naturalist 93: 145 159.
Lawton JH (1990) Species richness and population abundance of
animal assemblages. Patterns in body size: Abundance space.
Philosophical Transactions of the Royal Society of London, Series B
330: 283 291.
May RM (1986) The search for patterns in the balance of nature:
Advances and retreats. Ecology 67: 1115 1126.
Peters RH (1983) The Ecological Implications of Body Size. Cambridge,
UK: Cambridge University Press.
Sheldon RW, Prakas A, and Sutcliffe WH, Jr. (1972) The size distribution
of particles in the ocean. Limnology and Oceanography 17: 327 340.
White EP, Ernest SKM, Kerkhoff AJ, and Enquist BJ (2007) Relationship
between body size and abundance in ecology. Trends in Ecology
and Evolution 22: 323 330.
Cycling and Cycling Indices
S Allesina, University of Michigan, Ann Arbor, MI, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Definition of Cycle
Cycles in Food Webs
Structure of Cycles in Ecological Networks: Strongly
Connected Components
Quantifying Cycled Fraction: Finn’s Cycling Index
Limitations of FCI
Number of Cycles in Food Webs
Finding Cycles in Ecological Networks
Removing Cycles in Ecological Networks
Ecological Applications of Cycle Analysis
Further Reading
Introduction
Definition of Cycle
Given the finite amount of chemical compounds in the
biosphere, it is inevitable that the same material will be
utilized repeatedly by different organisms. This phenom
enon is addressed as ‘recycling’ or simply ‘cycling’ of energy
and matter. Familiar examples of recycling of nutrients
involve the so called ‘detritus chain’, which decomposes
organic matter that is unusable for some organism to its
basic compounds that can be recycled into the grazing chain.
This article provides an overview of cycles and cycling
indices in ecosystems ecology. Depending on the way of
modeling ecosystems, cycles assume different meanings.
In what follows, a general definition of cycles, taken from
graph theory will be introduced. Then the concept of
cycling is applied to (1) food webs (description of who
eats whom in the ecosystem) and (2) ecological networks
(weighted, mass balanced versions of food webs). Simple
ways of computing cycling indices and removing cycles
will be provided.
A very common way of describing ecosystems is by means
of graphs. Graphs are constituted by nodes (representing
species or functional groups of species) connected by
arrows (or edges, arcs, links, representing relationships
between species).
The simplest way of sketching ecosystems using
graphs is the food web representation. In this way of
drawing species relations, edges connect prey to their
predators (see Figure 1).
This food web representation can be associated with a
matrix that expresses the relationships between species.
This is the so called ‘adjacency matrix’, A. If the row
species is a food source of the column species then the
corresponding coefficient will be 1. More generally, this
relation is a consumer–resource relation, as nodes can
represent nutrient pools, etc. Elsewhere, the coefficients
will be 0. The food web in Figure 1 can therefore be
represented by this adjacency matrix:
Cycling and Cycling Indices 51
All kinds of paths, other than simple paths, contain at least
one cycle. For example, the graph in Figure 1 contains
just the simple cycle 2 ! 3 ! 4 ! 2. A graph containing
no cycles is said to be acyclic. Every pathway can be
classified according to its length that is given by the
number of nodes involved.
5
4
3
1
2
Cycles in Food Webs
Figure 1 Example of food web containing five species and
seven feeding relations (arrows, edges).
0
0 0 1 1 0
1
C
B
B0 0 1 0 0 C
C
B
C
B
B
A ¼ B0 0 0 1 1 C
C
C
B
B0 1 0 0 1 C
A
@
0 0 0 0 0
The adjacency matrix represents direct interactions
between species. These direct interactions, however,
yield chains of indirect interactions. These will be
sequences of nodes and edges that are called ‘paths’. We
can discriminate between different kinds of paths:
1. Open paths connect two different nodes. They can be
subdivided into ‘simple paths’, containing no repeated
nodes (e.g., A ! B ! C, Figure 2a) and ‘compound
paths’, which contain repeated nodes (e.g.,
A ! B ! C ! B ! D, Figure 2b).
2. Closed paths start and end at the same node. Also closed
paths can be divided into ‘simple cycles’, containing
no repeated nodes except the initial one (e.g.,
A ! B ! C ! A, Figure 2c) and ‘compound cycles’,
representing repeated cycles (e.g., A ! B ! A ! B ! A,
Figure 2d, where double arrows mean that the cycle is
traversed twice).
(a)
A
(b)
B
C
A
B
C
D
(c)
(d)
B
A
C
A
B
Figure 2 Classification of pathways in (a) simple paths (open
pathways start and end at different nodes); (b) compound paths
(open pathways start and end at different nodes, contain
repeated nodes); (c) simple cycles (closed pathways start and
end at the same node); and (d) compound cycles (same cycle
traversed more than once).
Cycles in food webs can be divided into two main classes:
feeding cycles and nonfeeding cycles. The former involve
species and their feeding relations (e.g., species A eats
species B; species B eats species A); cannibalism is a
simple kind of feeding cycle. The latter are typical of
food webs that comprise detritus compartments and
nutrient pools: organic matter is recycled in the system
via mineralization, creating a huge number of detritus
mediated cycles.
Feeding cycles are rare in published food webs. This
is mainly due to the fact that the resolution of food
webs is usually at the species/group of species level.
The number of feeding cycles becomes more significant
when age structured populations are considered, espe
cially in aquatic food webs. Nonfeeding cycles, on the
other hand, are extremely abundant in published net
works, being several billion cycles for highly resolved
ecosystem models.
Structure of Cycles in Ecological
Networks: Strongly Connected
Components
Two nodes A and B are said to belong to the same
strongly connected component (SCC) if they are reach
able from each other, that is to say if we can find a path
going from A to B and a path coming back from B to A.
If A and B belong to the same SCC, then they are
connected by cycles. A graph can be divided into its
SCCs, considering every node that is not involved in
cycles as an SCC by itself. Figure 3a represents the
Baltic Sea ecosystem. One can individuate 6 SCCs: 4 of
them are composed by a single node, while 2 of them
comprise more than 1 node (Figure 3b).
If we compact every SCC into a single node, we
produce an acyclic graph (Figure 3c). Further analysis
shows how one component contains just pelagic species
and the other one just benthic. Acyclic graphs can be
ordered so that all edges point in the same direction
(from bottom to top in Figure 3c) using a procedure
known as ‘topological sort’ (or partial ordering). Acyclic
graphs are therefore intrinsically hierarchical. In this
case, the flows find a sink in the benthic compartment,
while the pelagic compartment acts as a bridge between
52
Cycling and Cycling Indices
(c)
(b)
(a)
12
12
10
Benthic
11
10
9
9
8
8
15
15
2
11
7
7
Pelagic
7
2
14
14
2
13
5
5
6
6
1
4
4
3
3
13
13
1
1
Figure 3 Schematic representation of Baltic Sea ecosystem (a). The boxes define different strongly connected components.
Condensing each box into a single node yields an acyclic graph (b). This graph can be sorted so that all arrows point in the same
direction, showing the underlying straight flow between compartments (c).
the primary producer 1 and the benthic compartment.
The same structure was found for other aquatic net
works as well. Note that this feature depends drastically
on the presence/absence of resuspension of nutrients. If
this is negligible, then the network presents several
SCCs. When remineralization is strong, however, the
process joins the benthic and pelagic components, thus
forming a giant SCC.
describes the flow of energy–matter from the row
compartment (i) to the column compartment (j ). An
example of network and its matrix representation is
given in Figure 4.
In order to show the computation of the Finn’s cycling
index, it is necessary to introduce the concept of power of
adjacency matrices. Take as an example the matrix intro
duced in the first section. If we square it, we obtain
0
Quantifying Cycled Fraction: Finn’s
Cycling Index
Ecological networks are food webs where the edges are
quantified and represent exchanges of nutrients (usually
grams of carbon per m2 per year, but also nitrogen or
phosphorous) or energy. Moreover, inputs to the sys
tem and outputs from the system are explicitly
represented by flows involving ‘special compartments’
(i.e., nodes that act as a source (imports) or sink
(exports and respirations) for the system). Besides the
graph representation, a system can be described using
the so called flow matrix T, where each coefficient tij
0 1 0 1 2
B
B0
B
B
A2 ¼ B
B0
B
B0
@
0
1
C
0 0 1 1C
C
C
1 0 0 1C
C
C
0 1 0 0C
A
0 0 0 0
This matrix shows the pathways of length 2 that connect
to two nodes. For example, there is just one path connect
ing node 1 to node 4 in two steps (the path 1 ! 3 ! 4),
while there are two pathways connecting 1 to 5
(1 ! 4 ! 5 and 1 ! 3 ! 5).
In the same way, if we multiply this matrix with the
adjacency matrix we get A3, which describes all the path
ways connecting two nodes in three steps; A4 will
Cycling and Cycling Indices 53
(a)
860
300
167
2003
–11 184
Carnivores 203
Plants
370
2309
Detritus
feeders
200
8881
1814
Detritus
635
75
5205
Bacteria
255
1600
3275
3109
(b)
635
0
0
0
0
0
8881
0
0
0
300 2003
Plants
0
0
0
0
860 3109
Detritus
0
0
200
0
0
370
0
0
1600
75
0
0
0
0
167
0
0
0
0
203
0
0
0
0
0
0
0
0
Exports
0
0
0
0
0
0
0
0
Dissipations
0 11184
T=
2309 5205
0
Imports
0
0
1814
Detritus feeders
255 3275
Bacteria
Carnivores
Figure 4 Schematic representation of cone spring ecosystem (a). There are two imports (to Plants and Detritus), three exports (from
Plants, Detritus, and Bacteria) and five Dissipations (dashed arrows). The network can be associated with a matrix of transfers (b). The
first row represents imports, the last two columns stand for exports and dissipations, and the internal 5 5 part depicts
intercompartment flows.
similarly contain all the pathways of length 4, and so forth.
The power Ax will contain all the pathways of length x. If
the food web contains no cycles, then for some x < n
(where n is the number of species) the matrix will contain
just zeros. If the food web contains cycles, on the other
hand, the powers never converge to 0. The pathways
enumerated in these matrices belong to all the different
types that we illustrated in the first section. Now we can
see how these considerations apply to quantified
networks.
Dividing each coefficient tij for the row sum produces
the coefficients gij (of matrix G), which describe the frac
tion of flow leaving each compartment:
tij
gij ¼ P
k tik
For example, the G matrix for the network in Figure 4
would be
0
0 0:946 0:054
B
B0
B
B
B0
B
B
B0
B
G¼B
B0
B
B
B0
B
B
B0
@
0
0
0
0
0
0
0
0
0
0
0
0
0
1
C
0:027 0:179 C
C
C
0 0:201 0:453 0 0:075 0:271 C
C
C
0:084 0
0 0:155 0 0:761 C
C
C
0:307 0:014 0
0 0:049 0:629 C
C
C
0:451 0
0
0
0 0:549 C
C
C
0
0
0
0
0
0 C
A
0:794
0
0
0
0
0
0
0
0
0
Multiplying G by itself, one obtains the fraction of flow
leaving the row compartment and reaching the column
compartment in two steps (i.e., passing by an intermediate
54
Cycling and Cycling Indices
compartment). G3 will describe the exchanges in three steps,
and so forth. Summing over all possible powers of G, one
obtains the average number of visits a quantum of matter
leaving the row compartment will pay to the column com
partment. This computation is made possible by the fact
that the power series of G converges to the so called
Leontief matrix L. G0 is defined as the identity matrix I:
I þ G þ G 2 þ G 3 þ G 4 þ ¼ ½I G – 1 ¼ L
The Leontief matrix for the network in Figure 4 would be
0
1 0:946 0:946 0:202 0:440 0:031 0:120 0:880
B
B0
B
B
B0
B
B
B0
B
L¼B
B0
B
B
B0
B
B
B0
@
0
1
0:958 0:199 0:434 0:031 0120
0
1:207 0:251 0:547 0:039 0:117
0
0:186 1:039 0:084 0:161 0:018
0
0:374 0:092 1:169 0:014 0:085
0
0:545 0:113 0:247 1:018 0:053
0
0
0
0
0
1
0
0
0
0
0
0
1
C
0:880 C
C
C
0:883 C
C
C
0:982 C
C
C
0:915 C
C
C
0:947 C
C
C
0 C
A
1
In an acyclic network, the maximum coefficient of L will
be 1 (i.e., a quantum of matter can visit another compart
ment maximum once). This is because a particle of matter
leaving a compartment will never be recycled to the same
compartment again. This is not true when cycles are
present. In fact, when matter cycles in the network, a
particle can be recycled into the same compartment
many times, raising the maximum value of the coefficients
of the Leontief matrix. Therefore, the Leontief matrix of
an acyclic network would contain unitary coefficients on
the diagonal for all compartments (a particle starting at
any compartment will never come back). Consequently, a
simple way of estimating the cycled fraction would be to
see how much these coefficients deviate from 1. This is at
the heart of the so called ‘Finn’s cycling index’ (FCI).
There are various formulations for this index, but here
we present the simplest one, adapted from the one devel
oped in 1980 by J. T. Finn; the reader is referred to the
‘Further reading’ section for a complete account of the
possible variations. The following computation is valid
for steady state network only, that is, for networks where
the input to any node equals the output from the same
node.
We will call Tk the sum of all flows entering the
compartment k:
Tk ¼
X
tik
i
For example, in Figure 4 the sum of the flows to the
‘Plants’ compartment T1 would be 11 184.
A particle entering compartment k will be recycled
lij 1 times. The fraction of flow recycled is therefore
Rk ¼
lkk 1
lkk k
The recycled fraction for ‘Bacteria’ (fourth compartment)
would be (1.018 1)/1.018 ¼ 0.0172. The total flow
cycled C will be
C¼
X
Rk Tk
k
which, computed for the example, will result in
2777.23 units recycled.
The total fraction of recycled flow for the whole sys
tem will therefore be
C
FCI ¼ P
tij
ij
which, for the network in Figure 4, would be 0.0654.
Limitations of FCI
FCI considers only the diagonal coefficients of the
Leontief matrix, accounting therefore only for paths start
ing and ending at the same node.
Using the notation introduced above, we see that
FCI accounts for simple cycles and compound cycles,
but does not consider the contribution of compound
paths, as they never appear on the diagonal.
Compound paths, however, contain cycles that should
be included in the definition of cycling index.
Unfortunately, there is no simple linear algebra tech
nique that can account both for cycles and compound
paths, and counting all the pathways in an ecological
network is computationally very intense.
As an example of the limitation of the FCI, we see that in
Figure 4 the pathway Plants ! Detritus ! Detritus
feeders ! Detritus ! Bacteria will not contribute to any
diagonal coefficient, even if it contains a cycle. Because each
quantum of matter can be recycled into the same compart
ment many times, it will also move around compound paths
many times. This may result in off diagonal coefficients in the
Leontief matrix that are greater than 1, stressing the need for
counting compound paths in the cycling process.
Number of Cycles in Food Webs
In order to quantify the abundance of simple cycles in
food webs, one should know the maximum possible num
ber of simple cycles. The maximum number of simple
cycles will be associated with a completely connected
food web, that is, a food web whose adjacency matrix
contains just 1s.
Cycling and Cycling Indices 55
In order to count the maximum number of simple cycles,
we start from the ones with maximum length (in graph
theory they are called Hamiltonian cycles). In a completely
connected food web composed of n species, the number of
simple cycles of level (i.e., length) n is (n 1)!. This simple
formula can be explained combinatorically using permuta
tions: we can see a cycle of level n as a permutation of the n
labels of the nodes: for example, ABCD will represent the
cycle A ! B ! C ! D ! A. Now, the number of permuta
tions of n elements is n!. We note, however, that every cycle
gives rise to n possible sequences (e.g., ABCD, BCDA,
CDAB, and DABC represent the same cycle of length 4).
Therefore, the total number of simple cycles of maximum
length is n!/n ¼ (n 1)!.
This is an enormous number, as soon as n becomes
large. For example, in a 100 species food web, we can find
almost 10155 simple cycles of level n.
Now that we know the total number of simple cycles of
level n in a completely connected food web, we can easily
derive the number of simple cycles of level (n 1). For
each subgraph containing (n 1) species we will have
(n 1)!/(n 1) ¼ (n 2)! simple cycles of length (n 1).
The number of possible subgraphs containing (n 1)
species is given by the binomial coefficient
n
n 1
The total number of cycles is therefore given by the
following formula:
TotCycles ¼
Finding Cycles in Ecological Networks
Finding cycles in graphs is a computationally difficult
task. Nevertheless, published ecosystems contain a few
hundred nodes at most, and the low connectance (fraction
of realized connections) displayed by these systems
ensures that the number of simple cycles is much lower
than the theoretical case illustrated above, where all pos
sible cycles are present.
The idea behind most algorithms for cycle search is
simple: one should construct a path inside the network
until the same node is found twice. In this case the path is
either a cycle (the initial and final nodes do coincide) or a
compound path (initial and final nodes are different).
Of the various possible ways of searching the cycles,
backtracking based ones, such as ‘depth first search’
(DFS) are surely the easiest to implement.
Removing Cycles in Ecological Networks
Therefore, the total number of simple cycles of level
(n 1) in a completely connected food web composed of
n species is n(n 2)!.
Similarly, we can define the total number of simple
cycles of length k in a completely connected food web of n
species as
C ðk;nÞ ¼ ðk 1Þ!
k
k
!
The first 10 values are represented in Table 1. Note that this
sequence is defined, in combinatorics, as ‘logarithmic numbers’.
!
n
n
X
n
ðk 1Þ!
C ðk;nÞ ¼
k
k 1
1
n
X
We have stated above that it is possible to enumerate all
the cycles in a food web. In an ecological network, how
ever, each cycle will also possess a ‘weight’, given by the
amount of flow passing through the cycle.
Some network analysis applications (e.g., the so called
‘Lindeman spine’) require an acyclic network as an input.
The removal of the cycles therefore becomes an impor
tant topic for network analysis.
The current procedure requires the removal of cycles
according to their ‘nexus’. Two cycles are in the same
!
Table 1 represents the number of cycles of level k (column)
for a completely connected food web of n species (rows).
Table 1 Number of simple cycles of length k (column) in a completely connected food web formed by n species (rows)
k
n
1
2
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
10
1
3
6
10
15
21
28
36
45
3
2
8
20
40
70
112
168
240
4
6
30
90
210
420
756
1260
5
24
144
504
1344
3024
6048
6
120
840
3 360
10 080
25 200
7
720
5 760
25 920
86 400
8
5 040
45 360
226 800
9
40 320
403 200
10
Total
362 880
1
3
8
24
89
415
2 372
16 072
125 673
1 112 083
56
Cycling and Cycling Indices
nexus if they share the same weak arc, defined as the
smallest flow in the cycle. Cycles are then removed divid
ing the flow constituting the weak arc among all the cycles
sharing the same nexus. The resulting amounts are then
subtracted from each edge of the cycles. This process
results in the removal of the weak arc. The procedure is
then repeated until the resulting network is acyclic.
A nice by product of the procedure is the creation of a
network composed of all the cycles in the original network.
This is usually referred to as ‘aggregated cycles’ network in
ecological literature. This network will receive no input,
produce no output, and will be balanced (i.e., incoming
flows equal outgoing ones) for all nodes. If the resulting
aggregated cycle network is composed of several sub
graphs, each subgraph is a strongly connected component.
Note that while some applications require acyclic net
works, most of them are actually based on the fact that
empirical networks contain millions of cycles. As
explained in the next section, in fact, cycles are among
the most important features of ecosystems.
Ecological Applications of Cycle Analysis
The recycling of energy–matter is an important process that
occurs in every ecosystem. Cycling is believed to be a buffer
ing mechanism that allows ecosystems to face shortage of
nutrient inflows. This process, however, has been neglected
in many theoretical models, which concentrated on commu
nities rather than ecosystems, and which usually comprised
just a few species due to constraints of modeling techniques.
Food web ecologists always had an ambivalent attitude
toward cycling. For example, the first collection of food
webs published (which contained poorly resolved food webs
with just a few nodes) showed that cycles are very rare. This
was justified by the fact that cycles are likely to destabilize a
system, because they introduce positive feedbacks. This result
was, however, challenged by the discovery of many cycles in
larger food webs, and the role of cannibalism in age
structured population dynamics. In recent times, the impor
tance of cycles in food webs has been reconsidered, thanks to
the switch of focus from local stability dynamics toward a
more comprehensive approach to ecosystems persistence and
nonlinear dynamics. Moreover, a greater attention has been
devoted to the microbial loop, which, in some aquatic eco
systems, receives more than 50% of the primary production,
remineralizes it and feeds it back to higher trophic levels.
Ecosystem oriented modeling, on the other hand,
included cycles as the very foundation of the discipline.
The first clear reference to the importance of cycling in
ecological network comes from the work of Lindeman
who, in his seminal paper in 1942, described food webs
as cycling material and energy. Odum then included the
amount of recycling as one of the 24 criteria for evaluat
ing if an ecosystem is ‘mature’ (i.e., developed).
The request for a quantification of cycling was then
answered by the FCI illustrated above. Modified versions
of the FCI, including biomass storage, utilizing the so called
‘total dependency and contribution matrices’ were published,
increasing the possibilities for modelers and therefore the
number of applications of such indices to empirical studies.
Recently, it was pointed out how all these calculations
ignore some cycling that involves just off diagonal terms
in the Leontief matrix. Unfortunately, in order to com
pute the exact amount of cycling in an ecosystem one
should utilize a computationally intensive method, which
is therefore unfit to be applied to large ecosystem net
works. Fortunately, studies conducted on many small
networks showed that the total amount of cycling and
the FCI seem linearly related, with the total cycling
being around 1.14 times the FCI.
The relation between cycling and maturity of ecosystems
was challenged by the work of Ulanowicz. He showed how
cycling could be inversely related to the developmental
status of an ecosystem, and how perturbations could be
reflected into a higher cycling index. These considerations
suggest that cycling could be seen as a homeostatic response
to stress: impacts on ecosystems free nutrients from the
higher trophic levels; this freed matter is then recycled into
the system by microorganisms, generating cycles at the lower
trophic levels. In this view, responding to stress ecosystem
would show a decrease in cycle length and an increase in total
cycling. It is therefore important to know the distribution of
cycle lengths together with the total amount of cycling in the
ecosystem when one wants to assess the ecosystem status and
maturity. Ulanowicz also presented important insights on
cycling as autocatalytic processes. The cycling feature of
ecosystems is at the basis of the views of several authors on
ecosystem function and dynamics, such as, for example, the
work of Patten and colleagues.
Another aspect of cycling is represented by the compart
mentalization into SCCs. Although ecosystems comprise
myriad interactions, they still can be divided into a few
subsystems that are connected by linear chains of energy
transfers. In several aquatic food webs, SCC analysis shows a
subdivision into pelagic and benthic components of the
ecosystem. This result is, however, dependent on the way
the ecosystem is modeled, with particular emphasis on the
importance of including several detritus compartments.
Summarizing, cycling is an important aspect of ecosys
tem dynamics. Although cycles seem to be rare in published
community food webs and models, their number is very
large when detritus compartments are considered.
Moreover, it is important to stress that the role of the so
called microbial loop, neglected in studies that concentrate
on larger organisms, can dramatically change the cycling
performance of the system. These considerations lead eco
system ecologists to the formulation of the amount of
cycling in ecosystem networks. The FCI, even though it is
a biased count of the cycling in ecosystems, has found wide
Ecological Network Analysis, Ascendency 57
application in ecosystem studies. The problem of measuring
the exact amount of cycling in an ecosystem is still an open
problem, as it could be possible to ameliorate the algorithms
for finding and removing cycles. Finally, the network build
ing process is likely to determine the outcome in terms of
cycling. It would therefore be important to have shared rules
for network building that would result in the comparability
between different networks and ecosystems.
See also: Autocatalysis; Ecological Network Analysis,
Ascendency.
Allesina S and Ulanowicz RE (2004) Cycling in ecological networks: Finn’s
index revisited. Computational Biology and Chemistry 28: 227 233.
De Angelis DL (1992) Dynamics of Nutrient Cycling and Food Webs,
270pp. London: Chapman and Hall.
Finn JT (1976) Measures of ecosystem structure and functions derived
from analysis of flows. Journal of Theoretical Biology 56: 363 380.
Finn JT (1980) Flow analysis of models of the Hubbard Brook
ecosystem. Ecology 61: 562 571.
Patten BC (1985) Energy cycling in the ecosystem. Ecological Modelling
28: 1 71.
Patten BC and Higashi M (1984) Modified cycling index for ecological
applications. Ecological Modelling 25: 69 83.
Ulanowicz RE (1983) Identifying the structure of cycling in ecosystems.
Mathematical Biosciences 65: 219 237.
Ulanowicz RE (1986) Growth and Development: Ecosystems
Phenomenology. New York: Springer.
Ulanowicz RE (2004) Quantitative methods for ecological network
analysis. Computational Biology and Chemistry 28: 321 339.
Further Reading
Allesina S, Bodini A, and Bondavalli C (2005) Ecological subsystems via
graph theory: The role of strongly connected components. Oikos
110: 164 176.
Ecological Network Analysis, Ascendency
U M Scharler, University of KwaZulu-Natal, Durban, South Africa
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Principle of Ascendency
Ascendency Applications
Further Reading
Introduction
occurring in response to a cause. In ecosystems, it is
believed that no such direct, mechanistic cause and effect
behavior exists due to the interaction with other elements
which in turn influence the patterns of cause and effects
between pairs. Instead of the absolute probability, Popper
introduces the term propensity, which describes a bias
that events might (not will) happen. Popper therefore
calls for a measure of such relative or conditional prob
abilities. Conditional probabilities are denoted by pðai jbj Þ,
and are calculated by dividing the absolute probabilities
p(ai,b j ) by the marginal probability p(ai), or the sum of all
probable effects of one cause (Tables 1–3). The
conditional probability thus describes cause and effect in
the context of other absolute probabilities, considering
that a cause might have more than one effect. This elim
inates the pitfall of disregarding the influence of other
interactions on the one in question. It is, of course, possi
ble to calculate the conditional probability of a
mechanical cause–effect pair, that is, the case of having
one cause and one effect. This turns out to be 1, or in
other words, there is certainty that the effect in question
will follow the cause in question.
Since ecosystems are open, not all causes can be
accounted for. Some of them might originate outside the
In the search for a nonmechanistical explanation of ecosys
tem behavior and development, Ulanowicz developed the
theory of ascendency. Direct cause and effect mechanisms,
as known from the Newtonian world, are believed not to be
sufficient to describe, or predict, the behavior of ecosystems.
Such mechanisms are inherently reversible and are not seen
to be sufficient to explain the behavior of single components
(e.g., species) within the context of the ecosystem.
Ecosystems are believed to behave and evolve in a nonme
chanistic fashion. The theory of ascendency tries to capture
this nonmechanistic behavior in a single index, indicative of
ecosystem state and development, and of ecosystem health.
Principle of Ascendency
Ascendency
Conditional probabilities and ecosystem
complexity
In a mechanistic world, the probabilities of events follow
ing specific causes can be calculated by joint probabilities
p(ai,bj ). These describe an absolute probability of an effect
58
Ecological Network Analysis, Ascendency
Table 1 Frequencies of joint
occurrences of events (total 60)
b1
b2
b3
b4
a1
4
5
7
9
a2
2
4
2
1
a3
6
7
9
4
Table 2 Joint probabilities (p(ai, bj )) and their column/row
sums or marginal probabilities ( p(ai ), p(bj ))
p(ai ,bj)
b1
b2
b3
b4
p(ai)
a1
0.07
0.08
0.12
0.15
0.42
a2
0.03
0.07
0.03
0.02
0.15
a3
0.10
0.12
0.15
0.07
0.43
p(bj)
0.20
0.27
0.30
0.23
1.00
Values are obtained by dividing the number of occurrences (Table 1)
by the total number of observations (60).
Table 3 Conditional probabilities
b1
b2
b3
b4
a1
0.33
0.31
0.39
0.64
a2
0.17
0.25
0.11
0.07
a3
0.50
0.44
0.50
0.29
p(ai |bj)
Values are obtained by dividing the values in the joint probability
matrix by the column sums (p(ai)) (see Table 2).
system. Therefore, an open ecosystem can never evolve
toward a mechanistic behavior of cause and effect.
Ulanowicz states that autocatalysis, or indirect mutualism,
is an important cause in ecosystem growth and develop
ment. Autocatalysis is apparent when members of a
feeding loop positively enhance the following member
of the loop, which eventually leads back to a positive
enhancement of the starting member. Autocatalytic
loops exert a selection pressure on its members in that a
member of the loop might be replaced with a new con
stituent who has a more positive effect. Autocatalytic
loops exhibit a centripetality, which enables them to
attract more resources (available energy). These are rea
sons for the growth and development of, or increase of
order in, ecosystems.
To quantify growth and development, ecosystems are
portrayed as networks of material or energy exchanges.
These networks of feeding transfers are believed to ade
quately describe an ecosystem. It is assumed that other
significant aspects of ecological systems, such as beha
vioral aspects, are in one form or another imprinted on
the amount of energy transferred, through their effect on
population size and predator avoidance.
Ascendency describes both growth and development.
Growth of the ecosystem is measured as any increase
in total system throughput (TST), which is the sum of
all exchanges within the ecosystem and between the
system in question and its outside (imports, exports,
respirations). Total system throughput can rise either
by increasing the extent of the system (more species, or
by extending ecosystem borders) or by an increased
activity of the system (e.g., during phytoplankton
blooms).
Ecosystem development is quantified from the same
networks of material exchanges with the help of informa
tion theory. In autocatalytic loops, the trend for
transferring material is as follows: those linkages which
are most rewarding to the loop will transfer more material
than those which are not (compartments have in general
more than one outgoing link and can thus have pathways
to compartments outside the loop). The latter are not
necessarily discarded, but transfer only a small amount
of material. If a quantum of material sits in a compartment
in an autocatalytic loop, then it is therefore more likely to
be transferred along a route with high material transfer
than along a route with low material transfer. The prob
ability that a quantum of material flows along the highly
frequented routes is, therefore, higher compared to a net
work where all routes transfer the same amount of
material. Conversely, the probability that a quantum of
material flows along the less frequented routes will be
lower compared to a network where all routes transfer
the same amount of material. Such a change in probability
can be quantified with the help of information theory.
Information is defined as the agent that causes a change
in probability. Ulanowicz uses the term information to
describe ‘the effects of that which imparts order and
pattern to the system’.
In the calculation of information, the starting point is
to quantify ecosystem complexity. The complexity of a
system is mirrored in the system configuration (amount of
links and distribution of transfers along those links).
According to Boltzmann, the potential of each configura
tion contributing to systems complexity, s, can be
calculated as the negative logarithm of the probability
that the event (the system configuration) will occur
(s ¼ k log p, where k is a constant of proportionality,
i.e., a scaling factor). If a system configuration (the event)
will occur always (p ¼ 1), then the contribution to com
plexity is diminished (log(1) ¼ 0, uncertainty is at its
Ecological Network Analysis, Ascendency 59
lowest) and the system behavior is simple (i.e., it always
behaves the same way). If a system configuration (or
event) occurs only rarely, then there is a large potential
for complexity (i.e., it can behave in many different ways,
uncertainty is high). Behavior of a truly complex system is
unique each time it functions (uncertainty is highest).
To calculate how much a rare configuration contri
butes to system complexity, it is weighted by the (low)
frequency of its occurrence. The potential contributions
(or events) are averaged by the configurations of the
system by weighting each si by its corresponding
P
pi log pi ). In other
pi (Shannon’s formula: H ¼ – K
i
words, each potential contribution
of occurrence is
weighted by its corresponding probability that it will
occur, which is summed over all system configurations.
A high H corresponds to high uncertainty, complexity,
and diversity.
H¼
The above discussion serves to illustrate how events can
contribute to the complexity of a system. Next to consider
is whether these events contribute to an ordered pattern
in the system, or whether they contribute to random
behavior. If all events are equiprobable, then the average
uncertainty about what event will happen next is the
highest. This hypothetical situation can serve as a starting
point to calculate how much less uncertainty there is
under circumstances where not all events are equiprob
able. The decrease in uncertainty from a situation of
equiprobability to any other is called information. From
an ecosystem perspective, a situation of equiprobability is
one where material flows in equal amounts along all path
ways (Figure 1a). One that is not equiprobable is where
(a)
A
12
12
12
12
C
B
H¼
I ¼
(b)
A
23
1
23
23
B
1
Figure 1 (a) Hypothetical unconstrained network: low AMI.
(b) Hypothetical constrained network: higher AMI.
K log p ai jbj
K log pðai Þ
or
I ¼ K log p ai jbj
or
½2
k log p ai jbj
K log pðai Þ
I ¼ K log p ai jbj =pðai Þ
½3a
½3b
½3c
I is not positive for all pairs of occurrences. The sum of all
I ’s which have been weighted by the corresponding joint
probability turns out always to be positive, however. The
joint probability of each occurrence serves, as in
Shannon’s formula, as a weighting for the frequency of
occurrence of each event (i.e., each co occurrence of ai
and bj). The result is called the average mutual informa
tion (AMI) or
AMI ¼ K
C
½1
The information then is the a priori uncertainty minus the
uncertainty if bj is known or
12
1
K log pðai Þ
and the uncertainty that an event occurs provided certain
information (bj) is available is
Average mutual information
12
more material flows along some pathways, and less mate
rial along others (Figure 1b). Thus, the most
indeterminate network is one where all compartments
are connected with each other and where, in proportion
to the compartmental throughput, equal amounts of
material flow along the ingoing and outgoing pathways.
Quantifying the information which is gained by transfer
ring material along more and less frequented routes thus
gives a clue about the unevenness of material flowing
along pathways.
The change in probability from a situation where a
quantum of material flows along an equiprobable pathway
and along a pathway which is not equiprobable is calcu
lated using conditional probabilities. To start with, the
uncertainty that an event occurs is
XX
i
j
p ai ; bj log p ai jbj =pðai Þ
½4
AMI is the amount of uncertainty reduced by knowing bj .
Results are in units of K.
As in the hypothetical example above, the a priori
uncertainty about where a quantum of material flows
in ecological networks is given by Shannon’s formula.
The additional information (bj) to calculate the condi
tional probability is the knowledge of the outputs from
each compartment in the flow network a time step
earlier.
60
Ecological Network Analysis, Ascendency
Since, from an ecological network point of view, joint
and conditional probabilities refer to transfers of material
from compartment i to compartment j, the above formula
can be rewritten as
AMI ¼ K
X Tij Tij T::
log
T ::
Ti: T:j
i; j
½5
where the joint probability of a quantum of material
(p(ai,bj)) flowing from species i to species j can be denoted
as Tij/T.., remembering that the events in an events table
are material flows in a system. T.. is the total system
throughput, or the sum over all combinations of Tij .
The summation among all rows of the matrix is denoted
by the first dot, while the second dot stands for summation
among columns.
The conditional probability
p ai jbj ¼ p ai ; bj =pðai Þ
can be rewritten as Tij/Ti. and the marginal probability
(sum of all probable outcomes, p(ai)) as T.j/T...
To summarize, the AMI describes the information
gained by knowing the outputs from each compartment
in the flow network a time step earlier (bj) in addition
to the a priori situation describing the flow of a quan
tum of energy or material between two compartments
(ai). The uncertainty of where a quantum flows is
calculated through Shannon’s index of flow diversity.
The uncertainty of where a quantum of material will
flow by knowing bj is calculated by the conditional
probability.
ascendency. Mutualism is furthermore not a result of
events elsewhere in the system’s hierarchy but can arise
at any level. Therefore it is theorized that in the absence
of overwhelming external disturbances, the ascendency of
a system has a propensity to increase, that is, both activity
(TST) and structure (AMI) increase. The theoretical
behavior of mutual information conforms to most of the
24 ecosystem properties originally put forward by Odum
to characterize mature ecosystems.
Ascendency is limited by any constraints on the
increase in either TST or AMI. Limits to TST are set
by the finite imports from outside system boundaries and
by the second law of thermodynamics, which requires that
a portion of the compartmental throughput be lost as
dissipation. Therefore the TST cannot increase indefi
nitely via recycling. The limits to the AMI, or system
development, are set by the flow structure. It limits the
extent to which the flows can be organized without a
change to the structure itself. Further limits to the AMI
in real networks are discussed in the section titled
‘Overhead’.
In theory, ascendency is higher when pathways are
fewer in numbers (more specialization) and more articu
lated (few pathways transport most of the material). The
highest theoretical value of ascendency is achieved when
all players in the system have one input and one output
only, and are thus joined in one big single loop. This
configuration mirrors highest specialization, and in this
case AMI ¼ H (diversity of flows, see below). This situa
tion cannot be achieved in real systems, due to reasons
discussed in the section titled ‘Overhead’.
Ascendency
Development Capacity
The scalar constant, k, has been retained throughout all
calculations. To be able to combine growth and develop
ment into one single index, k is substituted by the ‘total
system throughput’ or TST in order to scale the AMI to
the size of the system in question. The resulting index is
called ascendency and is denoted by
As mentioned above, the limit to development is set by
Shannon’s diversity index pertaining to the material
transfers or flows. MacArthur applied Shannon’s diversity
index to the material flows in an ecosystem to arrive at a
measure for the diversity of flows, H:
XTij Tij T::
A ¼ TST
log
T::
Ti: T:j
i;j
or
A¼
X
i;j
Tij T::
Tij log
Ti: T:j
H¼
½6a
½6b
Besides indirect mutualism there are a number of
influences that can change the ascendency of a system.
These influences are thought to not have any favored
direction of change, whereas indirect mutualism is
believed to drive development toward increased
XTij Tij
k
log
T::
T::
i;j
½7
where k is a scalar constant, and T.. is the TST, or the sum
over all combinations of Tij .
H can, like the AMI, be multiplied by TST to scale the
diversity of flows to the system in question. TST H is
called the development capacity, or limit for develop
ment, C:
C¼
or
TST
XTij Tij
log
T::
T::
i;j
½8a
Ecological Network Analysis, Ascendency 61
C¼
X
Tij log
i;j
Tij
T::
½8b
The development capacity is limited by two factors,
namely TST and the number of compartments. The
limits to TST are the same as in the case of ascendency.
If a certain amount of TST is split between too many
compartments, then some compartments will end up with
a very small throughput. These are, in turn, prone to
extinction should the system undergo disturbances. This
process is believed to reduce the number of compart
ments and thus the number of flows. More stable
systems are thus believed to show a higher C compared
to systems undergoing frequent perturbations.
The initial complexity, H, consists of two elements.
One is the AMI, describing the information gained by
reducing the uncertainty in flow probability. It is an
index of the organized part of the system. The other is
the residual uncertainty, or Hc (also called conditional
diversity). Thus, H ¼ AMI þ Hc.
only one import path, then the overhead due to imports
is minimal and equals zero. From a systems point of view
it is regarded as counterproductive to minimize the mag
nitude of the import, or to import only via one pathway.
The insurance lies in being able to receive imports via
several pathways in case one is lost. In the case of
increased recycling within the system, the imports will
occupy a smaller and smaller part of the TST. In this case,
the development capacity will rise faster than the over
head on imports.
If the imports enter the system via fewer pathways or
compartments, then the ascendency will increase at the
expense of the overhead. Systems are expected to pro
gress toward fewer import pathways. The number of such
pathways can be changed should those links be disrupted
and others become necessary. Overall it is expected that
systems in a more stable environment rely on fewer
import pathways compared to perturbed systems.
The formula for the overhead on imports is as follows:
I ¼
Hc or Overhead
The residual uncertainty Hc, when scaled by TST is also
called the overhead. The overhead represents the unor
ganized, inefficient, and indeterminate part of the flow
structure and is considered an insurance for the system.
Should the system become overly organized (high ascen
dency), it will also be prone to perturbations. The
overhead is split into four components: overhead due to
imports, exports, respiration, and internal pathways.
The combined overhead is denoted by
Hc ¼
k
!
XTij
Tij2
log
Ti: T:j
T::
i;j
½9
Scaling Hc to the system by replacing k with TST
yields
¼
X
i;j
Tij log
Tij2
Ti: T:j
!
The relationship between C, A, and so becomes
C ¼ A: þ .
Imports
The overhead due to imports is dependent on the number
of pathways originating outside the system, and on the
magnitude of the material transferred along those path
ways. If all sustenance is equally distributed among all
import pathways, then the contribution to the overhead
will be maximal. It will decrease when some pathways
import more and others less. It will also decrease if the
overall magnitudes of the imports decrease. If there is
T0j log
j 1
T0j2
T0: T:j
!
½11
where imports are assumed to originate in the fictitious
compartment 0.
Exports
Similar to the overhead on imports, the overhead on
exports depends on the amount of exporting pathways
leaving the system and the amount transferred along
those pathways. The overhead due to export diminishes
whenever there are fewer export pathways, lower magni
tude of transfers, or an uneven distribution of amounts
transferred along the pathways. An increase in exports
becomes beneficial to the system whenever there is posi
tive feedback via another system. The overhead on
exports is denoted by
E ¼
½10
n
X
n
X
i 1
2
Ti;nþ1
Ti;nþ1 log
Ti: T:;nþ1
½12
where exports are assumed to flow into a fictitious com
partment n þ1.
Respiration
Again, the overhead regarding the dissipations depends
on the magnitude lost to the environment, on the number
of pathways, and the distribution of the magnitude trans
ferred. Losses through dissipation are required by the
second law of thermodynamics and are necessary to main
tain metabolisms. The overhead on dissipation is
D ¼
n
X
i 1
Ti;nþ2 log
2
Ti;nþ2
Ti: T:;nþ2
½13
62
Ecological Network Analysis, Ascendency
where respiration is assumed to flow into a fictitious
compartment n þ 2.
or
Redundancy
The fourth part of the overhead is that of internal trans
fers and represents the extent of pathway redundancy.
There are disadvantages to the system in maintaining
redundant, or parallel pathways. For one, there can be
an increase in dissipations, whenever transfers occur not
only along the most efficient route, but also along leakier
pathways. Also, the resource transferred along different
parallel pathways might not always end up at the right
time at the consumer.
An obvious advantage of parallel pathways is the insur
ance of having more than one route of transfer in case of
disturbances of other routes. Redundancy is denoted by
R¼
n X
n
X
i 1 j 1
Tij log
Tij2
Ti: T:j
!
½14
Tij B 2
IB ¼ K log
T:: Bi Bj
½15c
Summing over all realized combinations of i and j and
weighted by the joint probability of occurrence, one
arrives at the biomass inclusive AMI, AMIB:
AMIB ¼ K
½15a
Tij B 2
log
T::
T:: Bi Bj
X Tij
i;j
½16
which is also called the Kullback–Leibler information.
Scaling by the total system throughput gives the biomass
inclusive ascendency, AB:
AB ¼ TST
The above indices were calculated on the trophic flows
between compartments. It is also possible to calculate a
systems ascendency that embraces the connection
between biomass stocks and the trophic flows. This bio
mass inclusive ascendency can be used as a theoretical
basis to derive element limitations for compartments, to
identify limiting nutrient linkages, and to quantify the
successional trend to include larger species with slower
turnover times.
Above, AMI was calculated as the difference between
two flow probabilities, the unconstrained or a priori joint
probability, and the constrained or a posteriori conditional
probability. AMI can also be calculated between a bio
mass (unconstrained or joint) probability and the resulting
flow (constrained or conditional) probability, thereby
calculating a relationship between biomass and flows.
From the principal of mass action, the joint probability
of whether a quantum of biomass leaves compartment i
(Bi/B) and enters compartment j (Bj/B) is BiBj/B 2. This
expression constitutes the unconstrained joint probability
that a quantum flows from i to j. No constraining assump
tions are made about this exchange, with the exception of
the magnitudes of the stocks. The corresponding con
strained distribution is taken as the conditional
probability of the actual flow from i to j or Tij/T. This
constraint is an addition to the probability calculated from
the stocks only, and therefore, structure and function are
tied together. The information gained is calculated as
follows:
IB ¼
½15b
or
Biomass Inclusive Ascendency
Bi B j
Tij
K log
K log
B2
T::
Tij
Bi Bj
K log
IB ¼ k log
T::
B2
Tij B 2
log
T:: Bi Bj
T::
X Tij
i;j
½17a
or
AB ¼
X
i;j
Tij B 2
Tij log
T:: Bi Bj
½17b
AB is sensitive to changes in biomass and can thus show
the sensitivity of the whole system to changes in stock of a
particular compartment.
The above term can be split into the following terms:
AB ¼
X
Tij T::
Ti: B
Ti: log
þ
Ti: T:j
T:: Bi
i;j
i
X
T:j B
þ
T:j log
T:: Bj
j
X
Tij log
½18
The first term is exactly the same as in the above
definition of the flow ascendency. Therefore, also the
biomass inclusive ascendency rises with an increased
number of compartments, increased specialization of
flows, and increased throughput. The second and third
terms become zero whenever the proportional flow
through each compartment is the same as its proportion
of the biomass. Only in this case would AB equal A. In all
other cases, AB will exceed A.
Limiting elements in compartments and limiting
flows
If one is interested in calculating a compartment’s
contribution to the ascendency of a particular element k
(e.g., C, N, P, S, . . .) during a certain time step l, then one
Ecological Network Analysis, Ascendency 63
has to substitute into above equation the element and the
time step:
AB ¼
X
Tijkl log
i;j ;k;l
Tijkl B 2
T:: Bikl Bjkl
½19
where Tijkl is the flow from i to j of element k during time
step l.
To show how the ascendency responds to turnover
times of various elements, the differential of AB regarding
compartment p is given as
qAB
T::
¼2
qBpk
B:
1 T:pk þ Tp:k
2
Bpk
½20
Here the relative contributions of all elements investi
gated to the system’s ascendency can be calculated. Results
will show that the system is most sensitive to the element
with the slowest turnover rate. The element with the slow
est turnover rate is also the element which enters the
compartment in its least relative proportion. The last state
ment accords with Liebig’s law of the minimum for which
ascendency provides a theoretical basis. The same results
could have been obtained by comparing elemental turnover
rates of all compartments. However, ascendency provides
yet another level of information, namely it identifies which
source provides the limiting flow of the controlling element.
To calculate this, the sensitivities of the individual bio
masses can be expanded to include the sensitivities of the
individual flows from source r to predator p. The following
equation calculates the contributions of each flow:
Trp B 2
qAB
¼ log
T:: Br Bp
qTrp
½21
The limiting source of the controlling element is the
one which is depleted fastest in relation to its available
stock, that is, the one with the highest (Trp/Br). Knowing
the sensitivity of the flow for each element and compart
ment, it is thus possible to pinpoint nutrient limitations
and the limiting flows for each compartment in the food
web. In ecosystems, not all species are limited by the same
nutrient. For instance when primary producers are lim
ited by nitrogen, it does not necessarily mean that the
entire food web is limited by nitrogen.
Ascendency Applications
Principles of ascendency, as they have been shown here, have
been applied to compare similar ecosystems (e.g., estuaries),
or the same ecosystems over a period of time including the
response of systems to disturbances. Examples of such appli
cations are the description of spatial and temporal change of
ascendency in marine microbial systems. They revealed that
ascendency is strongly related to the functionality of the
microbenthic loop. Important parameters determining the
value of ascendency were the decomposition activity and
the capacity for resource exploitation. Ascendency was found
to be a useful indicator for the health assessment of marine
benthic ecosystems over space and time.
Ascendency has also been applied to establish ecosys
tem responses to eutrophication and other anthropogenic
system alterations of carbohydrates, proteins, lipids, and
carbon biopolymers in various parts of the globe.
Whereas ascendency is, in general, believed to rise with
eutrophication due to an increase in TST, this is not
always the case. Depending on the extent and frequency
of the eutrophication event, it might disturb the system to
an extent where ascendency reflects a decrease in ecosys
tem stability through a decrease in AMI and TST. Another
case of system perturbation was described for pesticide
perturbed microcosms, using an index called ‘scope for
change in ascendency’ (SfCA). SfCA is an analogy to
scope for growth of an organism and is the balance of the
ascendency of individual compartment inputs and outputs.
SfCA was hypothesized to decrease in the presence of a
disturbance and was ultimately found to be a useful indi
cator for the short term assessment of perturbations in
herbicide treated microcosms.
Ascendency has also been used to assess the whole
ecosystem impacts of severe freshwater abstractions from
an estuarine catchment. The interdecadal comparison
between light and severe freshwater abstraction and the
consequential reduction in sustained and pulsing freshwater
inflow into the Kromme estuary revealed a decrease in
ascendency under the present, freshwater starved condi
tion. The spatial comparison with other, similar, estuaries
that do not have such severe freshwater abstractions in the
catchment shows a higher ascendency in estuaries with
higher freshwater inflow that ensures sustained renewal of
the nutrient pool to fuel primary production.
Since ascendency is very often influenced by a change
in the magnitude of TST, the organization of a system
is frequently reported as a ratio of ascendency/develop
ment capacity (A/C), which cancels out the influence of
TST. Also the AMI is used as an unscaled index in a
comparative way. In general, it is advised to take the
behavior of other indicators of ecosystem health (e.g.,
exergy) into account in combination with ascendency to
arrive at a representative assessment of ecosystem state.
Ascendency has been shown to vary with the degree of
aggregation of the network. In general, ascendency
decreases in highly aggregated networks, even if the
TST is the same. The type of aggregation, that is, which
compartments are aggregated, also significantly affects the
value of ascendency. This is equally true for the aggrega
tion of living and nonliving components of the network.
The biomass inclusive version of ascendency and the
sensitivities of the individual flows were determined for
the Chesapeake Bay system to identify the limiting nutri
ent in the ecosystem and bottlenecks in carbon, nitrogen,
64
Ecological Network Analysis, Energy Analysis
and phosphorus transfers. The comparison over four sea
sons revealed that, in general, the primary producers were
nitrogen limited, which was in concordance with previous
studies on these groups. However, the nitrogen limitation
on the primary producer level was not propagated
throughout the entire web, but all nekton was found to
be phosphorus limited. The type of nutrient limitation
changed over the course of the year for a few primary
producers and invertebrates, but not for the nekton. It is
important to note that nutrient limitations in a trophic
flow network are not determined by the type of limitation
of the primary producer, since the various organisms have
different stoichiometric requirements.
See also: Autocatalysis; Emergent Properties; Goal
Functions and Orientors; Indirect Effects in Ecology.
Further Reading
Baird D and Heymans JJ (1996) Assessment of the ecosystem changes
in response to freshwater inflow of the Kromme River estuary, St.
Francis Bay, South Africa: A network analysis approach. Water SA
22(4): 307 318.
Fabiano M, Vassallo P, Vezzulli L, Salvo VS, and Marques JC (2004)
Temporal and spatial changes of exergy and ascendency in different
benthic marine ecosystems. Energy 29: 1697 1712.
Genoni GP (1992) Short term effect of a toxicant on scope for change in
ascendency in a microcosm community. Ecotoxicology and
Environmental Safety 24: 179 191.
MacArthur R (1955) Fluctuations of animal populations, and a measure
of community stability. Ecology 36(3): 533 536.
Morris JT, Christian RR, and Ulanowicz RE (2005) Analysis of size and
complexity of randomly constructed food webs by information
theoretic metrics. In: Belgrano A, Scharler UM, Dunne JA, and
Ulanowicz RE (eds.) Aquatic Food Webs, vol. 7, pp. 73 85. New
York: Oxford University Press.
Odum EP (1969) The strategy of ecosystem development. Science
164: 262 270.
Patrı́cio J, Ulanowicz RE, Pardal M, and Marques J (2006) Ascendency
as ecological indicator for environmental quality assessment at the
ecosystem level: A case study. Hydrobiologia 555: 19 30.
Popper KR (1982) A World of Propensities, 51pp. Bristol: Thoemmes.
Rutledge RW, Basore BL, and Mulholland R (1976) Ecological stability:
An information theory viewpoint. Journal of Theoretical Biology
57: 355 371.
Scharler UM and Baird D (2005) A comparison of selected ecosystem
attributes of three South African estuaries with different freshwater
inflow regimes, using network analysis. Journal of Marine Systems
56(3 4): 283 308.
Tobor Kaplon MA, Holtkamp R, Scharler UM, Bloem J, and de
Ruiter PC (2007) Evaluation of information indices as indicators of
environmental stress in terrestrial. Ecological Modelling 208: 80 90.
Ulanowicz RE (1986) Growth and Development: Ecosystems
Phenomenology. New York: Springer.
Ulanowicz RE (1997) Ecology, The Ascendent Perspective. New York:
Columbia University Press.
Ulanowicz RE (2004) Quantitative methods for ecological network
analysis. Computational Biology and Chemistry 28: 321 339.
Ulanowicz RE and Abarca Arenas LG (1997) An informational synthesis of
ecosystem structure and function. Ecological Modelling 95: 1 10.
Ulanowicz RE and Baird D (1999) Nutrient controls on ecosystem
dynamics: The Chesapeake mesohaline community. Journal of
Marine Systems 19: 159 172.
Relevant Websites
http://www.dsa.unipr.it Dipartimento di Scienze Ambientali.
http://www.ecopath.org Ecopath with Ecosim.
http://www.cbl.umces.edu Ecosystem Network Analysis.
http://www.glerl.noaa.gov National Oceanic and Atmospheric
Administration, Great Lakes Environmental Research
Laboratory.
Ecological Network Analysis, Energy Analysis
R A Herendeen, University of Vermont, Burlington, VT, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Level of Analysis
Steady-State Analysis: Energy and Nutrient Intensities
Steady-State Analysis: Other Indicators
Indicators in Dynamic Systems
Applications
Further Reading
Introduction
oil. And economists have shown that demand for a shirt
produces a demand for steel. These are all examples of
indirect effects. The techniques used to quantify them
span systems ecology, engineering, and economics – a
compelling example of cross disciplinary fertilization.
Ecologists have long told us that all flesh is grass, which in
turn is sunlight. Thanks in large part to the oil embargo in
1973, we appreciate that bread is not just sunlight, but also
Ecological Network Analysis, Energy Analysis
Understanding them yields insights in diverse applica
tions, from bioaccumulation of pollutants in ecosystems
to labor demand in economies.
In principle, one could discern all aspects of indirect
ness from the full diagram of flows between
compartments in a system. In practice, we often desire,
or accept, summary variables or indicators which are
specific to a particular application and convey the con
cept more concisely, though often with a loss of details.
In this article, such indicators are discussed, often
using explicit calculations applied to a simple, idealized
two compartment ecosystem. The indicators are energy
and nutrient intensities, trophic position (TP), path
length (PL), and residence time. Besides application to
steady state, the concept is also extended to dynamic
ecosystems such as those responding to perturbations.
Finally, calculating energy intensity of goods and services
in economic systems is discussed. The latter is a crucial
step in determining the energy cost of living in a con
sumer society, and has specific application in analyzing
consequences of an energy tax.
Level of Analysis
In a multicompartment system, interactions can be ana
lyzed at three levels of aggregation/detail:
1. Single compartment (isolated). This addresses direct
effects (e.g., an eagle eats mice but no grass).
Traditional population biology often works at this
level, which includes no indirectness at all.
2. Single compartment in system. This addresses direct plus
indirect effects (e.g., by eating mice which eat grass, the
eagle is consuming embodied grass, which is embodied
sunlight).
3. Whole system. This addresses system wide processes
(e.g., to what extent is the entire system recycling
phosphorus vs. leaking it immediately?).
This article concentrates on level 2. The indicators cal
culated are the property of a single compartment
explicitly connected to other compartments in an eco
system. Level 3 is beyond the scope of this article.
Level 2 analysis is the basis for most of the energy
analysis started in the early 1970s. It led to then surpris
ing results such as these:
1. Only c. 60% of the energy to own and operate a car is
the fuel in the tank. Around 15% is required to pro
duce the car, and c. 25% is for parts, maintenance,
insurance, registration, parking, etc.
2. Only c. 10% of the energy to make the car is consumed
at the assembly plant. The remainder is consumed at
the steel mill, glass works, iron mine, rubber plantation,
etc.
65
3. Switching from throwaway to returnable beverage bot
tles saves energy and increases jobs.
A more recent example is that suburban living (‘sprawl’) is
only c. 10% more energy intensive than urban (‘compact’)
living. A biological example is the trophic cascade, exem
plified by consequences of recent wolf reintroduction into
Yellowstone National Park. Adding wolves has sup
pressed elk activity, resulting in increased regeneration
of browse vegetation.
Steady-State Analysis: Energy and
Nutrient Intensities
The bookkeeping of energy analysis can be used to allocate
many other kinds of indirectness. Starting with energy, we
extend the method to other entities. To illustrate this, a
hypothetical two compartment system at steady state is
used (Figure 1). This is complex enough to allow feedback
(recycling), yet simple enough to allow using standard
algebra. All that is done here can be couched in matrix
notation, and shorthand is useful for systems with many
compartments, but algebra is more transparent.
Figure 1 shows the flow of something, say energy, in a
two compartment system containing producers (e.g.,
green plants) and consumers (e.g., herbivores). (See
Table 1 for definitions of terms.) The input to producers
comes from outside the system (the Sun), and the export
from consumers leaves the system. There is a variable
amount of feedback from consumers to producers; it is
possible for some plants to eat animals. In all diagrams of
this type we decide which flows convey the direct and
indirect influences we deem important; our judgement is
required. For example, the standard energy intensity con
cept is that the energy losses (e.g., low temperature heat)
are assumed to be embodied in the remaining flows of
high quality metabolizable biomass. This would give a
modified diagram (Figure 2).
The remaining flows thus convey the input, which we
wish to account for. The missing losses are now implicitly
embodied in the flows that remain, and the energy inten
sities carry this formally. The parallelism and difference
FEEDBACK
INPUTp
Producers
INPUTc
LOSSp
EXPORTc
Consumers
LOSSc
Figure 1 Energy flows (cal/day) in a two-compartment system
at steady state. INPUTp is gross primary production of biomass
by photosynthesis.
66
Ecological Network Analysis, Energy Analysis
Table 1 Definitions of terms
Symbol
Definition
Unitsa
tj
Ej
"j
ECOL
<">
"impj
EXPORTj
FEEDBACKj
gloof
GPP
IMPORTj
INPUTj
LOSSj
Nj
j
impj
OUTPUTj
PLj
Sj
TPj
tj
j
Xij
Xj
Yi
Y
Z
Time step
Energy input to compartment j
Energy intensity of compartment j’s output
Energy cost of living for a household
Household average energy intensity ( ECOL/Y)
Energy intensity of imported material functionally identical to that produced by compartment j
Export from compartment j
Flow from consumers to producers
Generic term for system input which is allocated by method used in this paper
Gross primary production, the energy fixed by plants
Import of material functionally identical to that produced by compartment j
Input to compartment j
Loss from compartment j
Nutrient input to compartment j
Nutrient intensity of compartment j’s output
Nutrient intensity of import of material identical to that produced by compartment j
Output of compartment j
Path length of compartment j
Stock of compartment j
Trophic position of compartment j
Isolated-compartment residence time for compartment j
In-system residence time for compartment j
Flow from compartment i to compartment j
OUTPUTj
Annual household expenditure for consumption category i
Sum of all annual household expenditures
FEEDBACK
day
cal/day
cal/cal
Btu/yr
Btu/$
cal/cal
cal/day
cal/day
?/day
cal/day
cal/day
cal/day
cal/day
g/day
g/cal
g/cal
cal/day
dimensionless
cal
dimensionless
day
day
cal/day
cal/day
$/yr
$/yr
cal/day
a
Grams, calories, and days are arbitrarily chosen as the units of mass, energy, and time for the examples in this article. Other units, as appropriate, are
used for the applications.
FEEDBACK * εc
INPUTp
Producers
INPUTc * εP
EXPORTc * εc
Consumers
Figure 2 Embodied energy flows (cal GPP/day) in a twocompartment system. The energy intensities " (cal GPP/cal)
convert energy flows (cal/day) of Figure 1 to embodied energy
flows (cal GPP/day).
between Figures 1 and 2 crystallize the entire import of
indirectness presented in this article. The intensities are
thus the conceptual, dimensional, and numerical bridge
between system input and flows. In this system, the units
of intensities will be calories of gross primary production
(i.e., the sunlight fixed by photosynthesis, abbreviated
GPP) per calorie of producer biomass or consumer bio
mass, but diverse units are possible depending on the
system and the question asked. For example, in economic
systems, energy intensity is measured in Btu/dollar
(1 British thermal unit ¼ 1055 J).
The key assumption is that embodied energy is
conserved in every compartment: embodied energy in ¼
embodied energy out. The general process is summarized
in Figures 3a and 3b. Because the method can be used to
allocate many things besides energy, the generic term
‘gloof’ is used in Figure 3a. Figure 3b shows the balance
equation for energy. Figures 3a and 3b allow for imports
to inject embodied energy into the system. An example is
Howard Odum’s study of Silver Spring, Florida, where
tourists fed bread to fish whose normal food was plants
growing in the spring. Another example of imported embo
died energy is America’s importing of clothes made in
China.
The energy intensities are obtained by solving the
(linear) equations implied by Figure 3b:
n
X
i 1
"i Xij þ "impj IMPORTJ þ Ej ¼ "j Xj
½1
The sum is over all within system inputs. In Figure 3b E
is the energy from the Earth, or at least external to the
system. The implicit energy intensity of energy itself is
1.0. Imported materials already have an energy intensity
because of their production elsewhere. This is discussed
in more detail in the section on dynamic indicators. Let us
apply this to a specific set of flows shown in Figure 4a.
For conciseness, feedback will be denoted by Z.
Equation [1] gives one equation for each compartment:
100 þ Z"c ¼ 10"p
10"p ¼ ð5 þ ZÞ"c
½2
67
Ecological Network Analysis, Energy Analysis
(a)
Zη c
In-system:
gloof embodied in
incoming in-system
flows
Output:
gloof embodied
in output flow
j
Producers
5η c
Consumers
10η P
1
Source:
gloof generated
internally plus that
embodied in import
flow
(5 – Z )/2 × η c
Figure 5 Embodied nutrient flows (g/day) in system of
Figure 4a. (g/cal) are the nutrient intensities.
(b)
ΣεiXij
i = in-system inputs
εi Xj
j
2Z /(15 – Z )
10/(15 – Z )
Ej + εimpj IMPORTj
Figure 3 (a) Generic scheme for allocating the embodied
generic system input gloof for steady-state system. Gloof may be
energy, nutrient, or even time (for calculating residence time). In
economic applications, gloof may be money, labor, or pollution
assimilation. For each compartment j, embodied gloof
in embodied gloof out, by assumption. (b) Embodied energy
balance equation for compartment j, which follows from (a). Xij
are in-system inputs to j; Xj is j’s total output; Ej is the (direct)
energy input to j.
(a)
Z
100
Producers
10
5
Consumers
90 + Z
5–Z
Producers
Consumers
(15 + Z )/(15 – Z )
1
(5 – Z )/(15 – Z )
Figure 6 Embodied nutrient flows (g/day) in system shown in
Figure 4a as a function of feedback, Z. Both compartments are in
embodied nutrient balance, as is the entire system.
balance, a consequence of the assumption that each indi
vidual compartment is in balance.
Now suppose we want to track the flow of nutrient.
Nutrient intensities () will be expressed in grams of nutri
ent/cal of biomass. Here we will (arbitrarily) assume that
nutrient is taken up only by producers and passed on with
out loss to consumers, but that some nutrient is leaked from
consumers during metabolism. Specifically, assume that the
nutrient intensity of the consumer metabolic loss is half that
of the consumer export flow. This is shown in Figure 5.
The balance equations are then:
(b)
20Z
100
1 þ Zc ¼ 10p
10p ¼ 5c þ Zc þ
100
Producers
10(10 + 2Z )
ð5 Z Þ
c
2
½3
Consumers
Figure 4 (a) Explicit energy flows (cal/day) for a twocompartment system. Feedback, Z, can vary between 0 and 5
cal/day. Example is arbitrary, but the output/input ratio of 0.1 for
producers is appropriate for many real systems. (b) Embodied
energy flows (cal GPP/day) in system shown in (a) as a function of
feedback, Z. Both compartments are in embodied energy
balance, as is the entire system.
The solution is "p ¼ 10 þ 2Z, "c ¼ 20, both expressed in
cal GPP/cal biomass. Substituting these energy intensi
ties in Figure 3b gives the embodied energy flows shown
in Figure 4b. The entire system is in embodied energy
which are solved to yield p ¼ (1/10)(15 þ Z)/(15 Z),
c ¼ 2/(15 Z). Using these intensities yields Figure 6
for embodied nutrient flow.
The system is in embodied nutrient balance, but the
flows have a surprising feature: for Z > 0, the nutrient flow
from producers to consumer (an internal flow) exceeds
1 g/day, the system’s input flow. Critics have called this
apparent contradiction a damning flaw of the method, but
actually it is to be expected. Feedback speeds up a sys
tem’s flows: more molecules pass a given point per unit
time. Because here embodied nutrient is actual nutrient,
the effect could be measured experimentally. This vali
dates the method generally.
Ecological Network Analysis, Energy Analysis
0.25
25
Cons. energy intens.
20
0.2
Cons. nutr. intens.
15
0.15
Prod. nutr. intens.
10
0.1
Prod. energy intens.
0.05
5
Nutrient intensity (g cal)
Energy intensity (cal GPP/day)
68
0
0
0
1
3
2
Feedback, Z
4
5
Figure 7 Energy and nutrient intensities vs. Z.
Energy intensities and nutrient intensities as a function
of feedback, Z, are shown in Figure 7.
As feedback increases (and hence loss from consumers
decreases), the two intensities approach equality. When
Z ¼ 5 cal/day, consumers have no losses and the two
compartments have functionally merged.
Steady-State Analysis: Other Indicators
Below are discussed three other indicators: TP, PL, and
residence time. In Table 2, the equations for each are
listed. The possibility of imports is explicitly allowed for.
An example is American household electronics, most of
which are made abroad. In Table 3, the specific equations
for the example system in Figure 4a and their solutions
are listed.
Trophic Position
Trophic levels apply to a linear chain picture of feeding
patterns: A eats nothing but B; B eats nothing but C, etc. If
there are n compartments in the chain, then there are n
integral trophic levels, and trophic level is the number of
steps from the Sun þ 1. Thus for producers and consu
mers in a chain, trophic levels ¼ 1 and 2, respectively.
With omnivory and the resulting web interactions, this
view breaks down unless nonintegral TPs are allowed.
Simply put, a compartment’s TP is the (energy) weighted
average of the TPs of each of its inputs plus 1. Caution:
trophic interactions are always expressed in energy flows,
so here one must use energy flows only, not nutrients or
other flows. (There is also a dual approach, which results
in an infinite series of integral trophic levels, which is not
covered here.)
The standard convention of setting TPSun ¼ 0 is used.
From Table 3, the TPs are 1 þ 2Z/100 and 2 þ 2Z/100
for producers and consumers, respectively. Feedback
increases TP of both. For Z ¼ 0, TPp ¼ 1, TPc ¼ 2, as
expected for a straight food chain.
Table 2 Explicit forms of the input and outputs in calculating several indicators for compartment j in steady-state analysis. For every
indicator, the equation to solve is Col A þ Col B Col C
Indicator
Energy
intensity
(")
Nutrient
intensity
()
Trophic
position
(TP)
Path
length
(PL)
Residence
time ()
B. Source term:
Internal or
imported inputs
C.
Output
term
P
"impj IMPORTj þ Ej
"jXj
P
impj IMPORTj þ Nj
jXj
1
TPj
None
PLj
tj
j
A. In-system inputs term
i
"i Xij
in-system inputs
i
i Xij
in-system inputs
P
i in-system inputs
TPi Xij þ TPimpj IMPORTj
P
i in-system inputs
i in-system inputs
P
i in-system inputs
Xij þ IMPORTj þ Ej
P
i in-system inputs
P
i in-system inputs
Flows can have different units for different
compartments. Intensities will
correspondingly have different units
Flows can have different units for different
compartments. Intensities will
correspondingly have different units
For trophic position, all flows must be in
energy terms
Xij þ IMPORTj þ Ej
ðPLi þ 1ÞXij
P
Comment
i Xij
Xij þ IMPORTj þ Ej
Xij, flow of i to j; Ej, energy flow to j; Nj, nutrient flow to j.
See Table 1 for additional definitions.
Path length is almost the same as trophic
position. Flows need not be energy but
must have the same units for every
compartment
tj is the isolated compartment residence
time for compartment j ( stock/
throughflow), assumed to be constant.
Flows need not be energy but must have
the same units for every compartment
Ecological Network Analysis, Energy Analysis
69
Table 3 Explicit balance equations and solutions for five indicators
Indicator
Refer to figure
Equations
Solutions
Energy
intensity (")
4a
100 þ Z"c 10"p
10"p ð5 þ Z Þ"c
"p
"c
Nutrient
intensity ()
5
1 þ Zc
Trophic
position (TP)
Path length (PL)
Residence
time ()
4a
4a
4a
TPp 10
þ1
10
TPc
TPc
Z
ðPLc þ 1Þ
100 þ Z
PLp þ 1 PLc
Z
þ tp
100 þ Z
p þ tc c
Path Length
This can be expressed in two ways:
backward in time. A molecule is just now leaving
• Looking
compartment j. How many intercompartment transits
•
ð5 Z Þ
c
2
TPp
c
has it made between its entering the system and now?
Looking forward in time. A molecule is just now leaving
compartment j. How many intercompartment transits
will it make on average before exiting the system?
g nutrient/cal
PLp
PLp
PLc
p
p
c
2þ
2Z
100
2Z
100
2Z
1þ
100
Z
Z
1þ
tc
tp þ
100
100
Z
1þ
t p þ tc
100
day
isolated compartment residence times ti, typically defined
as the ratio of stock to throughflow. (Unlike all the indi
cators so far discussed, this requires that we know the
stocks at steady state.) The system residence time is a
function of these isolated compartment residence times
and the degree of connectedness of the compartments. In
words, residence time for compartment j is the weighted
average of the residence time of each input þ tj. From
Table 3, the residence times are (1 þ Z/100)tp þ (Z/
100)tc and (1 þ Z/100)(tp þ tc) for producers and consu
mers, respectively. Without feedback, the residence time
for producers, p, is just tp because the only input is from
outside the system. Consumer residence time, c, is tp þ tc,
because a molecule leaving consumers has passed exactly
once through producers and consumers. TP and resi
dence time are graphed versus Z in Figure 8.
7
2.5
Cons. res. time
6
2
Cons. trophic pos.
5
1.5
4
3
1
Prod. trophic pos.
2
0.5
1
Prod. res. time
Residence Time ()
0
0
0
As with PL, this can be expressed looking either backward
or forward in time, but here only the former is treated: a
molecule is just now leaving compartment j. How long has
it been in the system? It is assumed that we know the
1
2
3
Feedback, Z
4
5
Figure 8 Trophic position and residence time vs. Z. The
isolated compartment residence times are 1 and 5 days for
producers and consumers, respectively.
Residence time (day)
For a steady state system, these are equally easy to
calculate. For a dynamic system, the backward looking
PL is preferable because it can be calculated without
knowing the future. Therefore we calculate only the
backward looking PL. In words, PL is the weighted
sum of the quantity (PL þ 1) for each input. Imports
do not figure in PL, which is based upon internal flows
only. The input flows need not be energy, but they
must all be in the same units so that the weighted
average can be calculated.
PL is almost the same as TP. For a system with only
sunlight as energy input, TPi ¼ PLi þ 1. If there are other
system energy inputs such as imported feed, the difference
between the two is more significant. As shown in Table 3,
PL ¼ 2Z/100 and 1 þ 2Z/100 for producers and consu
mers, respectively. For Z ¼ 0, PLp ¼ 0 because there are
no in system inputs to producers.
cal GPP/cal
p
5c þ Zc þ
TPc Z
þ1
100 þ Z
10 þ 2Z
20
1 15 þ Z
10 15 Z
2
c
15 Z
2Z
TPp 1 þ
100
10p
Trophic position
10p
Units
70
Ecological Network Analysis, Energy Analysis
Indicators in Dynamic Systems
Calculating Dynamic Indicators
Most of energy analysis and systems ecology stressing
indirectness has assumed a steady state in which flows
and stocks are constant over time. Yet real systems are
almost always dynamic. All the indicators addressed in
this article can have a dynamic interpretation, as long as
we use the back looking form. Any dynamic analysis must
be explicit about stocks, flows, and time steps. The ele
ments of the dynamic view are shown in Figure 9.
Figure 9 summarizes the assumption that in a time
step t, the gloof embodied in the inflows and in the
stock is distributed over the final stock and outflows. At
the end of the time step, there has been the mathematical
equivalent of perfect mixing, so that the energy intensity
Gloof embodied in
incoming in-system
flows during time
step Δt
Gloof embodied in
stock at time t
Gloof embodied in
output flow during
time step Δt
j
Gloof generated
internally plus that
embodied in import
flow during time step Δt
Gloof embodied
in stock at
time t + Δt
of stock and output are the same. Figures 10a and 10b
illustrate this in detail for energy intensity and residence
time.
The flows are multiplied by the time step t for
dimensional commensurateness with the stocks. Output
can include a change in stock (inventory change in eco
nomic terminology), so that the new stock is the old stock
plus this change. Figure 10a shows that energy intensity
at time t þ t is a function of the flows at time t þ t, and
the stocks and energy intensities at time t. If one knows
the initial energy intensities and stocks, and one has a
dynamic model to specify stocks and flows over time, one
can use the equation implied by Figure 10a to calculate
dynamic energy intensities:
X
t þt
IMPORTjt þt t
"ti þt Xijt þt t þ "tj Sjt þ "impj
i in system inputs
þ Ejt þt t ¼ "tj þt Xjt þt t
þ "tj þt Sjt
½4
Similarly, from Figure 10b, one obtains for dynamic
residence time:
X
i¼in system
inputs
it þt Xijt þt t þ jt Sjt þ Sjt t
1
X
C
B
Xijt þt t þ IMPORTj þ Ej A þ jt þt Sjt
¼ jt þt @
0
i¼in system
inputs
Figure 9 Generic scheme for allocating the embodied generic
system input gloof for the dynamic system. In the underlying
dynamics, stock changes over time as Stþt St þ
P
ðOUTPUTtþt
INPUTStþt Þt.
½5
Figures 10a and 10b also demonstrate how intensity
(of anything) is injected into a system as a source term
(a)
εit + Δt X ijt + Δt Δt
i = in-system
inputs
Σ
εit + Δt X jt + Δt Δt
j
εtj Stj
εjt + Δt S jt
+ Δt
εtimpj
IMPORTj t + ΔtΔt + Ej t + Δt Δt
(b)
Σ Xij t + Δt Δt + IMPORTj + Ej ⎛
τjt + Δt ⎛ i = in-system
⎝
⎝ inputs
τit + Δt X ijt + Δt Δt
i = in-system
inputs
Σ
j
τtj Stj
τjt + Δt S jt
Stj Δt
Figure 10 (a) Scheme for calculating dynamic energy intensity. The source term is energy itself plus energy embodied in imports. The
energy intensity of imports of type j, "impj, is specified exogenously. (b) Scheme for calculating dynamic residence time. The source term
is the aging of the existing stock in the time period t.
Ecological Network Analysis, Energy Analysis
and then allocated by internal flows. For energy inten
sity, the source is the embodied energy in imports of
similar entities (competitive imports in economic ter
minology) plus imports of different entities, here just
energy itself.
For residence time, the source term is just the aging
of the stock; external inputs do not contribute to resi
dence time by definition. Similar comments apply to TP
and PL.
Simulations of Dynamic Indicators
Figure 11 shows a two compartment dynamic model
system. Initially the system is at steady state with no
feedback (Z ¼ 0), but feedback is switched on (Z ¼ 3) at
time ¼ 20 days, and then off again (Z ¼ 0) at 500 days.
The details of the underlying model are not important
here; it incorporates a nonlinear ratio dependent feeding
response by consumers to abundance of producers, and
vice versa when feedback is on. Producer output depends
on light level, which is assumed constant, and producer
biomass. Simulation is performed using the modeling
software STELLA.
Figures 12a–12d show dynamic behavior of four of
the indicators calculated here: energy intensity, TP, PL,
and residence time. On all graphs, the stock of producers
and consumers is shown as well. Immediately after the
onset of nonzero feedback, producer stock increases as
more material now enters that compartment and consu
mer stock drops. But then consumer stock increases in
response to increased producer stock, and both stocks
asymptotically increase. This is reasonable, because
along with increased feedback comes decreased loss, as
shown in Figure 11.
Similar to the steady state calculations, energy inten
sity, TP, PL, and residence time all increase for both
producers and consumers. However, the values are not
given exactly by the static equations for Z ¼ 3 cal d 1.
This is because in the dynamic model, all flows and stocks
change when feedback changes, while in the static model
used in the previous two sections all flows except feed
back are assumed to remain constant.
Applications
Ecological Example: Four-Compartment Food
Web
Figure 13 shows steady state energy flows and stocks in
a bog in Russia. The analysts disaggregated this ecosys
tem into four compartments: plants, animals,
decomposers, and detritus. Detritus consists of undiffer
entiated dead material and therefore has no metabolic
losses. All other compartments contribute to detritus.
Additionally, animals and decomposers also eat detritus,
resulting in two feedback flows and a web structure. All
indicators can be calculated from the equations given in
Table 2. The results are given in Table 4. Because
plants have only a solar input, their energy intensity is
quite low and TP ¼ 1. For the other compartments,
however, the energy intensities are higher and the TPs
are high. Decomposers have a TP ¼ 4.9, higher than the
value of 4 which one would expect for a food chain
instead of this web. Decomposers come out on top in
both energy intensity and TP. Because this system has
only one input, PL is just TP 1.
Table 4 shows that residence times are affected dra
matically by web structure. Isolated compartment
residence times (¼stock/throughflow) are long for plants
and detritus, and short for animals and decomposers. The
longest, detritus, is 760 times the shortest, animals. In
contrast, in system residence times differ by only a factor
of 4. Both animals and decomposers, which in isolation
would be fast, have large input flows from detritus (which
in isolation is slow). The consequence is that all three are
comparably slow. This is one aspect of the notion that
detritus links tend to slow down the response of ecosys
tems to perturbations.
Z [0]
f (Sun, Sp )[100]
Producers
SP [100]
71
f(SP, SC) [10]
0.90 * S P
[90]
5
Consumers
SC [100]
0.05 * S C
[5]
Figure 11 Model for dynamic simulation. Figures in square brackets are initial steady-state values, before feedback is started.
Figures within boxes are stocks; others are flows.
72
Ecological Network Analysis, Energy Analysis
(a)
Sun
60
200
987.55
Stock (cal)
150
Cons. stock
100
40
30
Prod. stock
Cons. en. intensity
50
10
Prod. en. intensity
0
0
200
400
600
Time (day)
(b)
800
Cons. troph. pos.
2
1.8
1.6
100
Prod. stock
1.4
1.2
50
Trophic position
Cons. stock
1
Prod. troph. pos.
0
0
200
(c)
400
600
Time (day)
800
Cons. path length
150
Cons. stock
100
1
0.8
0.6
Prod. stock
0.4
Path length
Stock (cal)
0.8
1000
1.2
200
50
0.2
Prod. path length
0
0
200
(d)
400
600
Time (day)
800
Cons. stock
100
Cons. res. time
10
Prod. stock
5
50
Prod. res. time
0
0
200
400
600
Time (day)
800
Residence time (day)
150
Stock (cal)
0
1000
15
200
38.05
0
1000
Figure 12 Indicators in dynamic system. Initially system is at
steady state with feedback ( Z ) 0. Z is increased abruptly to a
steady value of 3 cal/day for days 20–500, and then returned
abruptly to zero. (a) energy intensity; (b) trophic position; (c) path
length; (d) residence time.
Animals
1.25
608
16.74
36.9
58.21
337.4
78.81 584.9
Detritus
8836
2.2
150
4.1
Plants
8490
0
1000
200
Stock (cal)
20
Energy intensity (cal/cal)
50
Decomposers
35.0
279.9
305
Figure 13 Energy flows (g fixed carbon per m2 per year) in a
bog in Russia. Detritus is undifferentiated dead material, and
therefore has no metabolic loss. Numbers in compartments are
stocks (g fixed carbon per m2). From Logofet DO and Alexandrov
GA (1984) Modelling of matter cycle in a mesotrophic bog
ecosystem. Part 1: Linear analysis of carbon environs. Ecological
Modelling 21: 247–258.
Energy/Economic Example: Energy Intensity of
Consumer Goods and Services, Energy Cost of
Living
This topic is included to emphasize and illustrate the
breadth of applicability of energy analysis and the analo
gies between ecological and economic systems. The
question is how much energy is required to support,
directly and indirectly, human household consumption
patterns. The approach is in two steps: (1) determine
how much energy is needed, directly and indirectly, to
produce a product and (2) determine how much of it a
household consumes.
Consider a loaf of bread. The energy to grow the
ingredients, make the bread, and transport and market it
can be determined by a detailed vertical analysis (also
called process analysis), in which one sums:
1.
2.
3.
4.
5.
the energy used in the supermarket;
the energy consumed in the bakery;
the energy consumed at the flour mill;
the energy used on the farm;
the energy for transport at every link; and so on.
This process can even lead to cycles in systems with
feedback (e.g., cars require steel, but the steel industry
uses some cars), but the process usually converges to an
acceptable answer after just several steps.
A vertical analysis is potentially accurate, but expen
sive. Performing it for a wide range of products is
prohibitive. There is, however, a large database on the
interactions of the sectors (c. 350–500) of the US econ
omy. This is the input–output (I–O) table published by
the US Department of Commerce. Many other countries
have similar I–O tables. With a number of fairly
Ecological Network Analysis, Energy Analysis
73
Table 4 Energy intensities, trophic positions, path lengths, and residence times for the Russian bog food web of Figure 13
Compartment
Energy
intensity, "
(cal GPP/cal)
Trophic
position
(TP)
Path length
(PL)
Isolated-compartment residence
time, t (years)
In-system residence
time,
(years)
Plants
Animals
Detritus
Decomposers
2.60
9.56
12.4
23.8
1.00
3.43
3.90
4.90
0.00
2.43
2.90
3.90
8.60
0.017
12.6
0.060
8.60
20.5
32.7
32.8
stringent assumptions, this table can be combined with
direct energy use data for each sector to produce energy
intensities using the equation implied by Figure 3b. One
such assumption is necessitated by the fact that the units
in I–O tables are monetary units per year, so one must
accept dollars as an appropriate allocator of embodied
energy. Because in the American energy industry,
energy is usually measured in Btu, the energy intensity
of goods and services is then expressed in Btu/$. I–O
based determination of energy intensities has been per
formed for c. 35 years. Under further assumptions, the
intensities can be used to evaluate the energy impact of
different expenditure patterns. Doing this for a house
hold yields the so called energy cost of living.
The I–O data are available, but gathering the asso
ciated direct energy data and performing the computation
is tedious, though today’s computers make it increasingly
easier. Solving 500 simultaneous equations of the form in
eqn [1] is done by inverting a 500 rank matrix.
Once we have the energy intensities, we need details
on how households spend their money over the range of
consumer product categories, also known as their market
basket. This information is collected by the US Bureau of
Labor Statistics. Putting the two together yields the
energy cost of living (ECOL):
ECOL ¼
X
"i Yi
i all expenditure categories
½6
where Yi is the household’s annual expenditure for
expenditure category i. Applying eqn [6] allows one to
analyze the effect of overall spending and the mix in the
market basket. The latter will be significant only if the
energy intensities are different for different expenditure
categories.
Table 5 shows I–O based energy intensities deter
mined by Carnegie Mellon University for 1997, and
updated and aggregated by the author into 15 categories
covering all household expenditures. The intensities are
indeed different, especially energy itself and service
industries such as health care.
Figure 14 shows the result of transforming of a house
hold market basket (in dollars/yr) to its energy impact (in
Table 5 Energy intensities for household consumption
categories (Btu/$, 2003 technology, 2003 dollars)
1. Residential fuel, electricity
2. Vehicle fuel
3. Vehicle purchase, maintenance
4. Food
5. Alcohol, tobacco
6. Apparel
7. Communication, entertainment
8. Health, personal care
9. Reading, education
10. Insurance, pension
11. Contributions
12. Public transportation
13. Asset gain
14. Miscellaneous
15. Housing
Direct energy ((1) þ (2))
Nonenergy (sum of (3)–(15))
139 300
94 300
5 400
6 100
3 700
6 500
4 000
2 400
3 000
1 600
3 800
21 200
4 700
4 200
5 100
118 100
4 700
All personal consumption
Energy/GDP
11 100
8 900
Sprawl ((1) þ (2) þ (3) þ (15))
Nonsprawl (sum of (4)–(14))
15 700
4 300
Auto and related ((2) þ (3))
21 200
Shaded categories indicate an intensity greater than the energy/GDP
ratio. ‘Sprawl’ contains housing and auto ownership and operation.
Source: Author’s calculations based on Carnegie Mellon University
data.
Btu/yr) using eqn [6] and intensities such as those shown in
Table 5. In Figure 14a, we see that of the average house
hold’s expenditures of $49 300 in 1973, only 6.4% was
for direct energy (residential fuel and electricity and
auto fuel). After conversion to energy requirements
(Figure 14b), this portion was 63% of the total impact of
604 million Btu. The total is roughly the energy equivalent
of 100 barrels of oil. Figures 15a and 15b show the energy
pie for the lowest expenditure decile ($11 500/yr, 241
million Btu/yr) and highest decile ($140 200/yr, 1233
million Btu/yr). The direct fraction is largest, 79%, for
the lowest decile, and lowest, 47%, for the highest decile.
Figure 16 shows a statistical fit to energy versus
expenditures for a representative sample of several thou
sand American households. It confirms that because the
mix changes, energy is not a linear function of total
74
Ecological Network Analysis, Energy Analysis
(a)
(a)
Asset gain
6.8%
Health care
5.9%
Food/alc./tobacco
12.5%
Apparel
2.5%
Infrastructure
0.9%
Other
19.6%
Other
Public trans. 1.5%
0.6%
Auto purch. maint.
0.5%
Auto fuel
15.3%
Asset gain
0.1% Health care
Food/alc./tobacco
0.8%
6.6%
Apparel
0.6%
Infrastructure
1.6%
Housing
9.1%
Housing
30.7%
Public trans.
0.8%
Auto purch. maint.
13.8%
Auto fuel
2.9%
Residential
energy
3.5%
Residential
energy
63.4%
(b)
Other
4.6%
Public trans.
1.5%
Asset gain
2.5%
Health care
1.0%
Food/alc./tobacco
6.3%
Apparel
1.4%
Infrastructure
0.9%
Housing
12.4%
Auto purch. maint.
6.2%
(b)
Asset gain
10.0%
Health care Food/alc./tobacco
Apparel
1.0%
5.4%
1.7%
Infrastructure
0.5%
Other
7.3%
Housing
14.6%
Public trans.
2.4%
Auto fuel
22.9%
Residential
energy
40.3%
Figure 14 For the average American household in 2003:
(a) expenditures, which total $49 300; (b) energy impacts, which
totaled 604 million Btu. Source: Unpublished calculations by
R. Shammin, R. A. Herendeen, M. Hanson, and E. Wilson.
expenditures, but rather bends down and away from a
straight line through the origin. The reason is that direct
energy (auto fuel and residential fuel and electricity)
tends to level out as household expenditures increase.
Expenditures increase for other products, but these
tend to be less energy intensive. For developed countries,
the shape of Figure 16 seems robust: studies of
Norway, the Netherlands, and Australia have found a
similar result.
Energy/Economic Example: Regressive Effects
of an Energy Tax
Concern with global warming, energy security, and pol
lution strongly implies that fossil energy is too cheap to
compensate for its drawbacks. Energy taxes of various
sorts have been proposed to stimulate more efficient use
and to fund alternatives. In the debate, equity issues
quickly surface. Because direct energy is a larger fraction
Auto purch. maint.
10.1%
Residential
energy
28.2%
Auto fuel
18.8%
Figure 15 2003 household energy impacts for (a) lowest
income decile ($11 500; 241 million Btu), (b) highest income decile
($140 200; 1233 million Btu). Source: Unpublished calculations by
R. Shammin, R. A. Herendeen, M. Hanson, and E. Wilson.
of the total for less affluent households, regressive impacts
of an energy price increase would be expected if one
ignored the indirect portion.
However, because the total energy curve in Figure 16
bends down, some regressiveness is still expected. To
compensate, one could design an income tax rebate to
even out the impacts over income classes. Figure 16 is the
key, as follows.
Suppose that fossil energy at the wellhead or mine
mouth is taxed at rate of p dollars per Btu. Assume that
economic sectors maintain their patterns of using inputs
to produce inputs, that is to say, technology is constant.
Assume further that each sector can successfully pass on
its increased costs to the consumers of its output. Then a
household’s market basket, if unchanged, will now cost an
additional amount ¼ p ECOL. The fractional increase
Ecological Network Analysis, Energy Analysis
Table 6 Consequences of a $4 per million Btu mine mouth
fossil energy tax, based on assumptions in text
2500
Total energy
Direct energy
Indirect energy
2000
Energy in MBtu/yr
75
Expenditure level
5000
1000
500
0
0
50
100
150
200
250
Market basket expenditure
(thousand $/yr)
Total energy (million Btu/yr)
h"i (thousand Btu/$)
Market basket price
increase ($/yr)
Market basket price
increase (%)
Lowest
decile
Average
Highest
decile
11.5
49.3
140.2
241
21.0
964
604
12.3
2416
1233
8.8
4932
8.4
4.9
3.5
Total expenditure in 1000 US $/yr
Figure 16 Household energy impact vs. total expenditures.
Direct energy is auto and residential fuel and electricity.
Total energy direct energy plus energy impact of all other
purchases. Source: Unpublished calculations by R. Shammin,
R. A. Herendeen, M. Hanson, and E. Wilson.
is this quantity divided by the total market basket’s origi
nal cost, denoted by Y. then
Fract:incr:in mkt: basket cost ¼
p ECOL
¼ ph"i
Y
½7
where h"i is the average energy intensity of the market
basket. The average is just energy/expenditure at the
appropriate point on Figure 16; it is the slope of a straight
line connecting the origin and the point.
As an example, consider a tax of $0.50 per gallon of
gasoline equivalent. This is about $4/million Btu, or $24/
barrel of oil. The latter is about 25% of the world crude
oil price as of 7 November 2007.
Using Figures 14–16, we perform the calculation in
Table 6. The increase in market basket price ranges from
3.5% for the highest expenditure decile to 8.9% for the
lowest. For full equity, income tax rebates, or other mea
sures, could address this differential.
Needless to say, a proper calculation is much more
involved than this one, but the idea of indirectness will
pervade it.
See also: Cycling and Cycling Indices; Ecological
Network Analysis, Ascendency; Ecological Network
Analysis, Environ Analysis; Indirect Effects in Ecology.
Further Reading
Bullard C, Penner P, and Pilati D (1978) Net energy analysis: Handbook
for combining process and input output analysis. Resources and
Energy 1: 267 313.
Burns T (1989) Lindeman’s contradiction and the trophic structure of
ecosystems. Ecology 70: 1355 1362.
Fath BD and Patten BC (1999) Network synergism: Emergence of
positive relations in ecological models. Ecological Modelling
107: 127 143.
Finn JT (1976) Measures of ecosystem structure and function derived
from analysis of flows. Journal of Theoretical Biology 56: 115 124.
Hannon B (1973) The structure of ecosystems. Journal of Theoretical
Biology 41: 535 546.
Herendeen R (1989) Energy intensity, residence time, exergy, and
ascendency in dynamic ecosystems. Ecological Modelling
48: 19 44.
Herendeen R and Fazel F (1984) Distributional aspects of an energy
conserving tax and rebate. Resources and Energy 6: 277 304.
Herendeen R, Ford C, and Hannon B (1981) Energy cost of living, 1972
1973. Energy 6: 1433 1450.
Lenzen M, Wier M, Cohen C, et al. (2006) A comparative multivariate
analysis of household energy requirements in Australia, Brazil,
Denmark, India, and Japan. Energy 31: 181 207.
Logofet DO and Alexandrov GA (1984) Modelling of matter cycle in a
mesotrophic bog ecosystem. Part 1: Linear analysis of carbon
environs. Ecological Modelling 21: 247 258.
Odum HT (1996) Environmental Accounting. New York: Wiley.
Ulanowicz R (1986) Growth and Development: Ecosystems
Phenomenology. New York: Springer.
Relevant Website
http://www.eiolca.net Economic Input Output Life Cycle
Assessment, Carnegie Mellon University.
76
Ecological Network Analysis, Environ Analysis
Ecological Network Analysis, Environ Analysis
B D Fath, Towson University, Towson, MD, USA and International Institute for Applied System Analysis,
Laxenburg, Austria
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Theoretical Development of Environ Analysis
Data Requirements and Community Assembly Rules
Methods and Sample Network
Network Properties
Summary
Further Reading
Introduction
fact, one of the three foundational principles in his seminal
paper introducing the environ theory concept. The neces
sary boundary demarcates two environments, the
unbound external environment, which indeed includes
all space–time objects in the universe, and the second
internal, contained environment of interest. This quantifi
able, internal environment for each system object is
termed ‘environ’, and is the focus of environ analysis. An
object’s environ stops at the system boundary, but as eco
systems are open systems, they require exchanges across
the boundary into and out of the system environs.
Therefore, input and output boundary flows are necessary
to maintain the system’s far from equilibrium organiza
tion. Objects and connections that reside wholly in the
external environment are not germane to the analysis.
Another foundational principle of environ analysis
theory is that each object in the system itself has two
‘environs’ one receiving and one generating interactions
in the system. In other words, an object’s input environ
includes those flows from within the system boundary
leading to the object, and an output environ, those flows
emanating from the object back to the other system
objects before exiting the system boundary. This alters
the perception of a system component from internal–
external to receiving–generating. Thus, the object, while
distinct in time and space, is more clearly embedded in
and responsive to the couplings with other objects within
the network. This shifts the focus from the objects them
selves to the relations they maintain; or from parts to
processes (or what Ilya Prigogine called from ‘being’ to
‘becoming’).
The third foundational principle is that individual
environs (and the flow carried within each one) are
unique such that the system comprises the set union of
all environs, which in turn partitions the system level of
organization. This partitioning allows one to classify
environ flow into what have been called different
modes: (1) boundary input; (2) first passage flow received
by an object from other objects in the system (i.e., not
Environ Analysis is in a more general class of methods
called ecological network analysis (ENA) which uses net
work theory to study the interactions between organisms
or populations within their environment. Bernard Patten
was the originator of the environ analysis approach in the
late 1970s, and he, along with his colleagues, has
expanded the analysis to reveal many insightful, holistic
properties of ecosystem organization. ENA follows along
the synecology perspective introduced by E. P. Odum
which is concerned with interrelations of material,
energy, and information among system components.
ENA starts with the assumption that a system can be
represented as a network of nodes (compartments, ver
tices, components, storages, objects, etc.) and the
connections between them (links, arcs, flows, etc.). In
ecological systems, the connections are based on the
flow of energy, matter, or nutrients between the system
compartments. If such a flow exists, then there is a direct
transaction between the two connected compartments.
These direct transactions give rise to both direct and
indirect relations between all the objects in the system.
Network analysis provides a systems oriented perspec
tive because it uncovers patterns and relations among all
the objects in a system. Therefore, showing how system
components are tied to a larger web of interactions.
Theoretical Development of Environ
Analysis
Patten was motivated to develop environ analysis to
answer the question, ‘‘What is environment?’’. In order to
study environment as a formal object, a system boundary is
a necessary condition to avoid the issue of infinite indir
ectness, because in principle one could trace the
environment of each object out in space and back in time
to the big bang origins. The inclusion of a boundary is, in
Ecological Network Analysis, Environ Analysis
boundary flow), which has not cycled; (3) cycled flow,
which returns to a compartment before leaving the sys
tem; (4) dissipative flow that it has left the focal object
not to return, but does not directly cross a system
boundary (i.e., it flows to another within system object);
and (5) boundary outflow. The modes have been used to
understand better the general role of cycling and the flow
contributions from each object to the other, which has had
application in showing a complementarity of several of the
holistic, thermodynamic based ecological indicators.
y2
z1
x2
f21
x1
f42
y4
x4
f32
f43
y1
f54
f31
x3
f53
x5
y3
f15
Data Requirements and Community
Assembly Rules
Network environ analysis could be referred to as a
holistic/reductionistic approach. It is holistic because it
considers simultaneously the whole influence of all sys
tem objects, yet it is reductionistic in that the fine details
of all object transactions are entailed in the analysis. In
other words, it is the opposite of a black box model. The
network data requirements are considerable, which
include the complete flow–storage quantities for each
identified link and node (note flow and storage are
interchangeable as determined by the turnover rate).
Data can be acquired from empirical observations, lit
erature estimates, model simulation results, or balancing
procedures, when all but a few are unknown. This diffi
culty in obtaining data has resulted in a dearth of
available complete network data sets. Due to this lack
of requisite data for fully quantified food webs, research
ers have developed community assembly rules that are
heuristics to construct ecological food webs. Assembly
rules are in general a set of rules that will generate a
connectance matrix for a number of species (N).
Common assembly rules that have been developed are
random or constant connectance, cascade, niche, modi
fied niche, and cyber ecosystem, each with its own
assumptions and limitations. In all but the last case, the
assembly rules construct only the structural food web
topology. The cyber ecosystem methodology also
includes a procedure for quantifying the flows along
each link. It uses a metastructure of six functional groups:
producer (P), herbivore (H), carnivore (C), omnivore
(O), detritus (D), and detrital feeders (F), within which
random connections link species based on these defini
tional constraints. Flows are assigned based on realistic
thermodynamic constraints.
Methods and Sample Network
To demonstrate basic environ analysis, it is best to pro
ceed with an example. Consider the network in Figure 1,
which has five compartments or nodes (xi, for i ¼ 1–5).
77
y5
Figure 1 Sample network with five compartments used to
demonstrate environ analysis notation and methodology.
Compartments are connected by transaction of the
energy–matter substance flowing between them. These
pairwise couplings are the basis for the internal network
structure. A structural connectance matrix, or adjacency
matrix, A, is a binary representation of the connections
such that aij ¼ 1 if there is a connection from j to i, and a 0
otherwise (eqn [1]):
2
3
0
0
0
0
1
6
61
6
6
A¼6
61
6
60
4
0
0
0
0
1
0
0
1
1
0
0
1
1
7
07
7
7
07
7
7
07
5
0
½1
Storage and flows must have consistent units (although it
is possible to consider multiunit networks). Typically,
units for storages are given in amount of energy or bio
mass per given area or volume (e.g., g m 2), and units for
flows are the same but as a rate (e.g., g m 2 d 1). The
intercompartmental flows for Figure 1 are given in the
following flow matrix, F:
2
0
6
6 f21
6
6
F¼6
6 f31
6
60
4
0
3
0
0
0
f15
0
0
0
f32
0
0
f42
f43
0
0
f53
f54
7
0 7
7
7
0 7
7
7
0 7
5
½2
0
Note that the orientation of flow from j to i is used
because that makes the direction of ecological relation
from i to j. For example, if i preys on j, the flow of
energy is from j to i. All compartments experience
dissipative flow losses (yi, for i ¼ 1–5), and here the
first compartment receives external flow input, z1
(arrows not starting or ending on another compartment
78
Ecological Network Analysis, Environ Analysis
represent boundary flows). For this example, these can
be given as
y ¼ ½ y1 y2 y3 y4 y5
½3
and
2
z1
3
6 7
607
6 7
6 7
7
z¼6
607
6 7
607
4 5
0
½4
Total throughflow of each compartment is an impor
tant variable, which is the sum of flows into,
P
P
Tiin ¼ zi þ nj fij , or out of, Tiout ¼ yi þ nj fji the ith
compartment. At steady state, compartmental inflows
and outflows are equal such that dxi/dt ¼ 0, and therefore,
incoming and outgoing throughflows are also equal:
Tiin ¼ Tiout ¼ Ti . In vector notation, compartmental
throughflows are given by
2
T1
3
6 7
6 T2 7
6 7
6 7
7
T¼6
6 T3 7
6 7
6 T4 7
4 5
T5
½5
This basic information regarding the storages, flows, and
boundary flows provides all the necessary information to
conduct environ analysis. Environ analysis has been clas
sified into a structural analysis, dealing only with the
network topology, and three functional analyses (flow,
storage, and utility) – which requires the numerical values
for flow and storage in the network (Table 1).
The technical aspects of environ analysis are explained
in detail elsewhere, so rather than repeat those here, the
remainder of the article highlights some of the important
results from environ analysis. But first, one issue that must
be covered is the way in which network analysis identifies
and quantifies indirect pathways and flow contributions.
Indirectness originates from transfers or interactions that
occur nondirectly, and are mediated by other within
system compartments. These transfers could travel two,
three, four, or many links before reaching the target
destination. For example, the flow analysis starts with
the calculation of the nondimensional flow intensity
matrix, G, where gij ¼ fij/Tj. The generalized G matrix
corresponding to Figure 1 would look as follows:
2
0 0 0 0 g15
6
6 g21
6
6
G¼6
6 g31
6
6 0
4
0
3
7
0 0 0 0 7
7
7
g32 0 0 0 7
7
7
g42 g43 0 0 7
5
0 g53 g54 0
½6
These values represent the fraction of flow along each link
normalized by the total throughflow at the donating com
partment. These elements give the direct, measurable flow
intensities (or probabilities) between any two nodes j to i.
To identify the flow intensities along indirect paths (e.g.,
j ! k ! i ), one need only consider the matrix G raised to
the power equal to the path length in question. For exam
ple, G2 gives the flow intensities along all paths of length
2, G3 along all paths of length 3, etc. This well known
matrix algebra result is the primary tool to uncover system
indirectness. In fact, it turns out that due to the way in
which the G matrix is constructed, all elements in Gm go
to zero as m ! 1. Therefore, it is possible to sum the
terms of Gm to acquire an ‘integral’ flow matrix (called N),
which gives the flow contribution from all path lengths:
N ¼ G0 þ G1 þ G2 þ G3 þ ¼
1
X
m 0
Gm ¼ ðI GÞ – 1
½7
where G0 ¼ I, the identity matrix, G1 the direct flows,
and Gm for m > 1 are all the indirect flows’ intensities.
Note, that the elements of G and N are nondimensional;
to retrieve back the actual throughflows, one need only
multiply the integral matrix by the input vector: T ¼ Nz.
In other words, N redistributes the input, z, throughout
each compartment to recover the total flow through that
compartment. Similarly, one could acquire any of the
direct or indirect flows by multiplying Gmz for any m.
Table 1 Basic methodologies for network environ analysis
Structural analysis
Functional analyses
Path analysis
Enumerates pathways in a network (connectance, cyclicity, etc.)
Flow analysis: gij fij/Tj
Identifies flow intensities along indirect pathways
Storage analysis: cij fij/xj
Identifies storage intensities along indirect pathways
Utility analysis: dij (fij fji)/Ti
Identifies utility intensities along indirect pathways
Ecological Network Analysis, Environ Analysis
A similar argument is made to develop integral storage
and utility matrices:
storage : Q ¼ P0 þ P1 þ P2 þ P3 þ
utility : U ¼ D0 þ D1 þ D2 þ D3 þ
where pij ¼ (fij/xj)t, and dij ¼ (fij
1
X
m 0
1
X
m 0
Pm ¼ ðI PÞ – 1
½8
Dm ¼ ðI DÞ – 1
½9
fji)/Ti.
Network Properties
Patten has developed a series of ‘ecological network proper
ties’ which summarize the results of environ analysis. The
properties have been used to assess the current state of
ecosystem networks and to compare the state of different
networks. Furthermore, while interpreting some of the prop
erties as ecological goal functions, it has been possible to
identify the structural or parametric configurations that posi
tively affect the network property values as a way to detect or
anticipate network changes. For example, certain network
alterations, such as increased cycling, lead to greater total
system energy throughflow and energy storage, so one could
expect that if possible ecological networks are evolving or
adapting to such configurations. This leads to a new area of
research on evolving networks. In this section, a brief over
view is given for four of these properties: dominance of
indirect effects (or nonlocality), network homogenization,
network mutualism, and environs.
79
The implications of this important result are clear in
that each compartment is embedded in and dependent
on the rest of the network for its situation, thus calling for
a true systems approach to understand such things as
feedback and distributed control in the network.
Network Homogenization
The homogenization property yields a comparison of
resource distribution between the direct and integral
flow intensity matrices. Due to the contribution of indir
ect pathways, it was observed that flow in the integral
matrix was more evenly distributed than that in the
direct matrix. A statistical comparison of resources dis
tribution can be made by calculating the coefficient of
variation of each of the two matrices. For example, the
coefficient of variation of the direct flow intensity
matrix G is given by
CVðGÞ ¼
Pn
j 1
Pn
i 1
gij gij
ðn 1Þg
2
½11
Network homogenization occurs when the coefficient of
variation of N is less than the coefficient of variation of G
because this says that the network flow is more evenly
distributed in the integral matrix. The test statistic
employed here looks at whether or not the ratio CV(G)/
CV(N) exceeds 1. The interpretation again is clear that
the view of flow in ecosystems is not as discrete as it
appears because in fact the material is well mixed (i.e.,
homogenized) and has traveled through and continues to
travel through many, if not, most parts of the system.
Network Mutualism
Dominance of Indirect Effects
This property compares the contribution of flow along
indirect pathways with those along direct ones. Indirect
effects are any that require an intermediary node to
mediate the transfer and can be of any length. The
strength of indirectness has been measured in a ratio of
the sum of the indirect flow intensities divided by the
direct flow intensities:
Pn
nij
Pn
i;j 1
gij
i;j 1 gij
ij
½10
where ij, the Kronecker delta, is 1 if and only if i ¼ j and
is 0 otherwise. When the ratio is greater than 1, then
dominance of indirect effects is said to occur. Analysis of
many different models has shown that this ratio is often
greater than 1, revealing the nonintuitive result that indir
ect effects have greater contribution than direct effects.
Thus, each compartment influences each other, often
significantly, by many indirect, nonobvious pathways.
Turning now to the utility analysis, the net flow, utility
matrix, D, can be used to determine quantitatively and
qualitatively the relations between any two components
in the network such as predation, mutualism, or competi
tion. Entries in the direct utility matrix, D, or integral
utility matrix, U, can be positive or negative ( 1 dij,
uij < 1). The elements of D represent the direct relation
between that (i, j) pairing; for the example in Figure 1,
this produces the following:
2
6
6
6
6
6
6
6
6
6
D¼6
6
6
6
6
6
6
6
4
f21
T2
f31
T3
0
f21
T2
0
f32
T3
f42
T4
f31
T3
f32
T3
0
f43
T4
0
f42
T4
f43
T4
0
f15
T5
0
f53
T5
f54
T5
0
3
f15
T5 7
7
7
7
0 7
7
7
f53 7
7
7
T5 7
7
f54 7
7
7
T5 7
7
5
0
½12
80
Ecological Network Analysis, Environ Analysis
Table 2 Direct and integral relations in sample network from Figure 1
Direct
(sd21, sd12)
(sd31, sd13)
(sd41, sd14)
(sd51, sd15)
(sd32, sd23)
(sd42, sd24)
(sd52, sd25)
(sd43, sd34)
(sd53, sd35)
(sd54, sd45)
Integral
(þ, ) ! exploitation
(þ, ) ! exploitation
(0, 0) ! neutralism
( , þ) ! exploited
(þ, ) ! exploitation
(þ, ) ! exploitation
(0, 0) ! neutralism
(þ, ) ! exploitation
(þ, ) ! exploitation
(þ, ) ! exploitation
(su21, su12)
(su31, su13)
(su41, su14)
(su51, su15)
(su32, su23)
(su42, su24)
(su52, su25)
(su43, su34)
(su53, su35)
(su54, su45)
The direct matrix D, being zero sum, always has the same
number of positive and negative signs:
2
0
6
6þ
6
6
sgnðDÞ ¼ 6
6þ
6
60
4
0
0
þ
0
þ
þ
0
0
þ
þ
þ
7
07
7
7
7
7
7
7
5
0
½13
The elements of U provide the integral, system deter
mined relations. Continuing the example, and now
including flow values derived from 10% transfer effi
ciency along each link (gij ¼ 0.10, if aij ¼ 1, and gij ¼ 0
otherwise), we get the following integral relations
between compartments:
2
þ
6
6þ
6
6
sgnðUÞ ¼ 6
6þ
6
6þ
4
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
3
7
þ7
7
7
7
7
7
7
5
þ
distinct environs, there are in fact 2n environs in total.
The output environ, E, for the ith node is calculated as:
E ¼ ðG IÞN̂ i
3
(þ, ) ! exploitation
(þ, ) ! exploitation
(þ, þ) ! mutualism
(þ, þ) ! mutualism
( , ) ! competition
(þ, ) ! exploitation
(þ, þ) ! mutualism
(þ, ) ! exploitation
(þ, ) ! exploitation
(þ, ) ! exploitation
½15
where N̂i is the diagonalized matrix of the ith column of
N. When assembled, the result is the output oriented flow
from each compartment to each other compartment in the
system and across the system boundary. Input environs
are calculated as
E9 ¼ N̂ i9ðG9 IÞ
½16
where g9ij ¼ fij/Ti, and N9 ¼ (I G9) 1. These results
comprise the foundation of network environ analysis
since they allow for the quantification of all within system
interactions, both direct and indirect, on a compartment
by compartment basis.
Summary
½14
Unlike, the direct relations, this is not zero sum.
Instead, we see that there are 17 positive signs (includ
ing the diagonal) and 8 negative signs. If there are a
greater number of positive signs than negative signs in
the integral utility matrix, then network mutualism is
said to occur. Network analysis demonstrates the posi
tive mutualistic relations in the system. Specifically,
here, we can identify two cases of indirect mutualism,
seven of exploitation, and one of competition
(Table 2).
A practical objective of ENA in general, and environ
analysis in particular, is to trace material and energy
flow–storage through the complex network of system
interactions. The network environ approach has been a
fruitful way of holistically investigating ecological sys
tems. In particular, a series of ‘network properties’ such
as indirect effects ratio, homogenization, and mutualism
have been observed using this analysis, which consider
the role of each entity embedded in a larger system.
See also: Cycling and Cycling Indices; Ecological
Network Analysis, Ascendency; Ecological Network
Analysis, Energy Analysis; Emergent Properties; Indirect
Effects in Ecology.
Environ Analysis
Further Reading
The last property mentioned here is the signature prop
erty, the quantitative environ, both in the input and
output orientation. Since each compartment has two
Dame RF and Patten BC (1981) Analysis of energy flows in an intertidal
oyster reef. Marine Ecology Progress Series 5: 115 124.
Fath BD (2007) Community level relations and network mutualism.
Ecological Modelling 208: 56 67.
Indirect Effects in Ecology
Fath BD and Patten BC (1998) Network synergism: Emergence of
positive relations in ecological systems. Ecological Modelling
107: 127 143.
Fath BD and Patten BC (1999) Review of the foundations of network
environ analysis. Ecosystems 2: 167 179.
Fath BD, Jørgensen SE, Patten BC, and Straškraba M (2004)
Ecosystem growth and development. Biosystems 77: 213 228.
Gattie DK, Schramski JR, Borrett SR, et al. (2006) Indirect effects and
distributed control in ecosystems: Network environ analysis of a
seven compartment model of nitrogen flow in the Neuse River
Estuary, North Carolina, USA Steady state analysis. Ecological
Modelling 194(1 3): 162 177.
Halnes G, Fath BD, and Liljenstrom H (2007) The modified niche model:
Including a detritus compartment in simple structural food web
models. Ecological Modelling 208: 9 16.
Higashi M and Patten BC (1989) Dominance of indirect causality in
ecosystems. American Naturalist 133: 288 302.
81
Jørgensen SE, Fath BD, Bastianoni S, et al. (2007) Systems Ecology: A
New Perspective. Amsterdam: Elsevier.
Patten BC (1978) Systems approach to the concept of environment.
Ohio Journal of Science 78: 206 222.
Patten BC (1981) Environs: The superniches of ecosystems. American
Zoologist 21: 845 852.
Patten BC (1982) Environs: Relativistic elementary particles or ecology.
American Naturalist 119: 179 219.
Patten BC (1991) Network ecology: Indirect determination of the life
environment relationship in ecosystems. In: Higashi M and Burns TP
(eds.) Theoretical Ecosystem Ecology: The Network Perspective,
pp. 288 315. London: Cambridge University Press.
Whipple SJ and Patten BC (1993) The problem of nontrophic processes
in trophic ecology: Towards a network unfolding solution. Journal of
Theoretical Biology 163: 393 411.
Indirect Effects in Ecology
V Krivtsov, University of Edinburgh, Edinburgh, UK
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Basics
Examples of Occurrence and Importance of Indirect
Effects
Approaches and Techniques Used to Detect and
Measure Indirect Effects
Problems and Implications for Environmental
Management
Current and Further Directions
Further Reading
Introduction
technological progress. It should also be noted that the
boost of the growing appreciation of indirect effects in
twentieth century was partly initiated by Vernadsky’s
fundamental theories about the ‘biosphere’, the ‘noö
sphere’, and interrelations between biota and
geochemical cycling. Popularization of these views 50
years later (e.g., by Lovelock’s Gaia theory) stimulated
investigations of indirect effects even further.
Interrelations among ecosystem components and pro
cesses can be subdivided into direct (i.e., those which are
restricted to the direct effect of one component/process
on another, and are attributable to an explicit direct
transaction of energy and/or matter between the compo
nents in question) and indirect (i.e., those that do not
comply with the above restriction). The history of natural
sciences is inseparable from the gradually increasing
awareness and understanding of indirect effects. By nine
teenth century the significance of indirect interactions was
well realized, and was (sometimes implicitly) accounted
for in the classic studies of Darwin, Dokuchaiev,
Gumboldt, Engels, and many other scientists. In the twen
tieth century, however, appreciation of indirect effects in
nature received considerable acceleration, predominantly
due to the accumulating interdisciplinary knowledge of
natural ecosystems, the development of appropriate
mathematical techniques, and the urgent necessity to
resolve the growing problems of environmental damage,
resulting, ironically, from the uncurbed expansion of the
human population backed by the advances of the
Basics
There have been many definitions of direct and indirect
effects. Information on indirect interactions is scattered in
the literature, and may appear under various terms. For
example, among ecological phenomena which may
(depending on the exact definition) be regarded as indir
ect effects are exploitative and apparent competition,
facilitation, mutualism, cascading effects, tri trophic
level interactions, higher order interactions, interaction
modification, nonadditive effects, etc.
First of all, it is important to distinguish between direct
and indirect effects. Usually, the interactions between two
82
Indirect Effects in Ecology
components not involving direct transfer of energy and/
or matter are viewed as indirect, while those that involve
an explicit direct transaction are viewed as direct. The
literature is inconsistent on the definitions of indirect
effects, and one way to clarify the problem is to stress
the difference between a transaction and a relation. A
simple transaction between two ecosystem components
is always direct since it is the transfer of matter and/or
energy, whereas a relation is the qualitative type of inter
action. Relations include predation, mutualism,
competition, commensalism, ammensalism, etc. Hence a
direct relationship is the one which is based on a direct
(i.e., unmediated by another ecosystem component) trans
action only. For example, the classic predation (not to be
mistaken with, for example, keystone predation, indirect
predation, etc.) is direct, and so is the nutrient uptake by
plants, algae, and bacteria, whereas mutualism and com
petition are always indirect, as they result from the
combination of a number of simple transactions. It is
worth pointing out that the observed patterns of inter
relations between ecosystem components (e.g.,
correlation between abundance indices) frequently result
from a combination of direct and indirect effects, as each
component is involved in a large number of pathways.
Furthermore, if a direct relationship between two ecosys
tem components (say A and B) is modified by a third
ecosystem component, attribute, or forcing function (the
two latter notions will include, for example, such modi
fiers as sunlight, temperature, pH, external and internal
concentrations of alternative nutrients) then the
indirect relationship between the modifying agent and
the first two components (i.e., A and B) becomes
superimposed upon the direct relationship between the
components A and B. Consequently, the observed pattern
of interrelation (e.g., correlation between the abundance
data) between A and B will in this case result from the
combination of direct and indirect effects.
Examples of factors known to modify the strength of
density mediated indirect interactions include differ
ences in the specific growth rates (important, for
example, for apparent competition), density dependence
of the transmitting compartment, and the possibility of
stochastic physical disruption. On the other hand, issues
important in determining the manifested strength of the
behavior mediated indirect interactions involve ability of
a focal species to detect changes in factors which matter
for energetic costs and benefits of its behavior, sensitivity
of its optimum behavior to these costs and benefits, and
available behavioral options.
For density mediated effects, presence and strength of
indirect interactions can be determined by analyzing par
tial derivatives of the abundance of a species on the
abundances of other (not immediately connected) species.
However, indirect interaction may involve ecologically
important changes other than changes in abundance, for
example, demographic changes in the population struc
ture, changes in the genotypic composition, and changes
in behavior (e.g., searching rates, antipredator behaviors),
morphology, biochemistry (e.g., nutrient content, toxin
concentration), or physiology.
Most Commonly Studied Indirect Effects
Among a plethora of possible indirect effects, there are
five that have been studied most commonly. Their
essence is depicted in Figure 1 and is briefly explained
below.
Interspecific competition
Interspecific competition (also called exploitative compe
tition) takes place whenever two (or several) species
compete for the same resource. In Figure 1a, an increase
in Component 1 will lead to the increased consumption of
the shared resource (Component 2), and consequently to
the decrease in a competitor (Component 3). Examples of
this include, for example, two predators sharing the same
prey, or two microbial species whose growth is limited by
the availability of the same nutrient.
Apparent competition
Apparent competition occurs when two species have
a common predator. In Figure 1b an abundant
population of species 1 sustains a high density population
of predator 2, who, in turn, may limit the population of
another prey species 3. From practical point of view, it is
worth noting here that this situation sometimes happens
as an unwanted result in biocontrol, when a biocontrol
agent (species 2), specifically introduced to control a
target (species 1), may increase the risk of a nontarget’s
(species 3) extinction.
Trophic cascades
Trophic cascades involve propagation of the effect along
a vertical trophic chain consisting of three or more com
ponents connected by grazing or predation. In Figure 1c,
an increase/decrease in Component 4 will lead to the
decrease/increase in Component 3, increase/decrease in
Component 2, and decrease/increase in Component 1.
These effects are particularly well studied in aquatic
food chains (see examples below), but have also been
studied in terrestrial systems.
It is worth pointing out, however, that the structure
of real ecosystems hardly ever fits tidily into the con
cepts of simple trophic levels (e.g., omnivory is
widespread in nature), and trophic cascades, therefore,
are often complicated by the interlinks within and
among trophic levels (e.g., in terrestrial ecosystems
insectivorous birds prey on predatory, herbivorous, and
parasitoid insects, and the resulting effect of birds on the
Indirect Effects in Ecology
(b)
(a) Interspecific competition
1
Apparent
competition
83
(c) Trophic cascade
4
3
2
3
2
1
2
3
1
(d)
Indirect mutualism
involving exploitative
competition
1
5
2
(f) Interaction modification
(e) Indirect mutualism involving
interference competition
1
4
2
3
4
2
3
1
3
Figure 1 Diagrams of the most commonly studied indirect affects. Direct effects are shown using solid lines, while indirect effects
(only the effects relevant to the accompanying discussion are illustrated) using dotted lines. Interaction modification is illustrated using a
dashed line. Numbers in the compartments are used solely for labeling to distinguish between different compartments, and do not relate
to any kind of hierarchy. Likewise, the box sizes do not bear any relevance to the sizes or significance of the compartments drawn, and
the relative size of the arrows relates neither to the effect’s strength no to the preferential directionality. See further explanations in the
text. (a) Interspecific competition; (b) apparent competition; (c) trophic cascade; (d) indirect mutualism involvingb9000
exploitative
competition; (e) indirect mutualism involving interference competition; (f) interaction modification. Modified from Wootton JT (1994) The
nature and consequences of indirect effects in ecological communities. Annual Review of Ecology and Systematics 25: 443–466.
primary producers and their damage by herbivory may,
therefore, depend on the specific species and the condi
tions involved). In particular, proper consideration of
detritus contributions to the energy flows may prove
the ‘trophic cascade’ simplification unsuitable, as the
detritus compartment often has direct links to a number
of trophic levels.
Indirect mutualism and commensalism
Indirect mutualism and commensalism involve a con
sumer–resource interaction coupled with either
exploitative (Figure 1d) or interference (Figure 1e)
competition. For instance, starfish and snails reduce
the abundance of mussels, a dominant space occupier,
and increase the abundance of inferior sessile species.
The presence of grazers on oyster farms in Australia
increases oyster recruitment by removing algae, who
otherwise preempt the available spaces. In Figure 1d,
an increase in species 1 should lead to a decrease in
species 2 and an increase in species 3. The latter posi
tive effect would propagate up the right branch of the
diagram, increasing the abundances of species 4 and 5.
This situation arises when, for example, planktivorous
fish preferentially feeding on large zooplankton indir
ectly increase the abundance of small zooplankton.
Cases involving interference competition are well
known from, for example, the intertidal environment,
where birds increase the abundance of acorn barnacles
by consuming limpets that otherwise dislodge the
young barnacles off the rock.
Interaction modification
Interaction modification occurs when the relationship
between a species pair is modified by a third species
(Figure 1f). Examples include positive effects of macro
algae on zooplankton through interference with the
hunting potential of fish and changing of a chemical’s
bioavailability due to the activity of a species, when the
chemical in question is important for the functioning of
another species (e.g., acids produced by one microbial
population may increase bioavailability of compounds
that are bound or unaccessible for another microbial
population).
It is worth pointing out that ‘interaction modification’
is often, and quite rightly, considered as a principally
different type of indirect effect. By coupling interaction
modifications with other types of relationships (e.g.,
trophic), one may arrive at possibilities of numerous
(including very complex) relationships. One of the more
simple of such combinations may be exemplified
(Figure 2) with an indirect effect of grazers and certain
agricultural practices on the population density of foxes
(Vulpes vulpes) and the rodent Marmota bobac in Eastern
Europe (V. Takarsky, personal communication): lower
84
Indirect Effects in Ecology
Grazing/hay making
Vulpes vulpes
ecosystem components simultaneously take part in a
multitude of interactions, and it is therefore appropriate
to name it an interaction web. In fact, the number of
possible kinds of indirect effects is likely to be limited
only by the number of system components considered.
Classifications of Indirect Effects
Grasses
Marmota bobac
Figure 2 Diagram illustrating a positive indirect effect of
grazing on Marmota bobac population resulting from a
combination of consumer–resource relationships with an
interaction-modification relationship. See further explanations in
the text.
grazing rates lead to a denser and taller grass cover,
enabling more successful hunting of predators.
Conversely, higher grazing rates lead to a lower grass
cover, thus enhancing the detection of predators by the
rodents. As a result, increase in grazing may have an
indirect positive effect on the Marmota bobac population,
and an indirect negative effect on the population of foxes.
It should also be noted that some of the known examples
of ammensalism and commensalism do actually fit in the
description either of a simple interaction modification or
interaction modification coupled with a number of tropic
relationships. For instance, the bioavailability example
described above has been quoted by Atlas and Bartha as
an example of commensalism. If, however, the chemical in
question is not nutritional, but harmful for the second
species, then the relationship fits the criteria for ammens
alism. In a similar vein, protocooperative and mutualistic
relationships are easily envisaged from certain combina
tions of interaction modifications and tropic relationships.
It is worth pointing out that although the indirect
relationships listed above are mainly studied in relation
to pairs of biological species, they are applicable to a
wider range of system components. It should also be
noted that many more types of indirect effects are easily
envisaged from various possible combinations between
interacting compartments, and quite a few have indeed
been observed in nature. For example, Menge distin
guished 83 subtypes of indirect effects. However, an
attempt to exemplify every possible type of indirect
effects would be outside the scope of this article. The
readers could easily construct, for example, many further
types of indirect effects combining the most commonly
studied ones depicted in Figure 1. In a real world,
Although detailed analysis of various possible classifica
tions would be outside the scope of this publication, it is
worth mentioning, however, that indirect effects can be
characterized in a number of ways, related, for example,
to the characteristics of exerting, receiving, and transmit
ting compartments, presence/absence of a lag phase
before the manifestation of a response, strength of the
interaction (particularly in relation to the direct interac
tions) and its directionality (e.g., whether it is isotropic or
anisotropic), dependence on a specific ecosystem context,
importance for the functioning of the compartments
involved, importance for structural (e.g., successional or
evolutionary) changes in the populations involved and
the whole biological community, and significance for
overall ecosystem functioning. In the author’s view, the
different ways to characterize indirect interactions are not
contradictory, but rather complementary, and may con
veniently contribute to the toolbox for comparative
ecosystem analysis.
Indirect Effects
All the relations not restricted to the effects of a direct
transaction of matter and energy between the adjacent
ecosystem components are treated as indirect. Hence, for
the purpose of the forgoing sections, all types of indirect
interactions mentioned above will be considered as indir
ect effects. However, the distinction between directly and
indirectly mediated effects will be made where deemed
appropriate. The terms ‘relationship’ and ‘interaction’ will
be used interchangeably. Furthermore, although it is rea
lized that for the purpose of quantitative assessment the
distinction between the terms ‘effect’ and ‘interaction’
may be helpful, no such distinction has been made in
this article, as in many studies addressing indirect effects
these terms are used interchangeably.
The definition of indirect effects given above is very
encompassing, and will include some of the effects which
may fall into the category of ‘direct’ under a different
definition. For example, it is useful to account for the
distinction between those effects that are directly and
indirectly mediated, since the latter ones are particularly
difficult to observe, especially if the cause and effect are
substantially separated in time.
The directly mediated effects have previously been
regarded as direct (i.e., as regards to the properties of their
propagation). Here, however, the directly mediated effects
Indirect Effects in Ecology
will be treated as indirect, and the definition of indirect
effects will, therefore, include such effects as trophic cas
cades, top down and bottom up controls, etc. The
classification of indirect effects into directly and indirectly
mediated is applicable to a wide range of environmental
processes and bears certain similarities with the distinction
between ‘interaction chains’ and ‘interaction modifications’
earlier recognized for purely biotic relationships.
Examples of Occurrence and Importance
of Indirect Effects
Indirect Effects in Terrestrial Environment
Arguably, the awareness of natural scientists as regards
indirect effects in the terrestrial environment can be
traced back at least to the end of nineteenth century,
when the school of thought founded by Dokuchaiev had
developed a theory that soil was a product of complex
interactions between climate and geological and biologi
cal components of the terrestrial landscape. To date, the
importance of indirect interactions in the terrestrial
environment is well recognized. Indirect effects in terres
trial ecosystems relate, for instance, to the dependence of
plant nutrient supply on mineralization of nutrients by
soil biota, and to the propagation of these effects through
the food chain. Soil fauna may help to disperse microor
ganisms crucial for plant functioning and biogeochemical
cycling, and physically modify the habitat, thus changing
environmental conditions for all the biological commu
nity. Plants, in turn, modify the habitat for other
organisms, for example, by producing litter, providing
shade, shelter, etc. All in all, indirect effects in the terres
trial environment are widespread; below are just a few
examples of their recent studies.
A number of studies conducted in the terrestrial envi
ronment (this includes both field experiments and soil
microcosms) adopted experimental approach focusing
on the density manipulation experiments followed by
analysis of the results obtained using parametric (e.g.,
ANOVA, Tukey’s HSD) and nonparametric (e.g.,
Kruskal–Wallis and Mann–Whitney U tests) statistical
tests. For instance, Miller used exclusion experiments to
elucidate direct and indirect species interactions in a field
plant community. Experimental results were analyzed by
parametric and nonparametric techniques, which yielded
interesting information on the ecological characteristics
of the species involved. Particularly, it was established
that species with a large competitive ability due to direct
effects generally had almost as large indirect effects, so
that the two effects almost cancelled each other.
A number of terrestrial studies used various mathema
tical methods to investigate indirect interactions. In
particular, a good insight into specific indirect effects
was gained using simulation modeling to interpret
85
monitoring or experimental results. For example, Hunt
and co authors found that the increase in net N miner
alization with precipitation is a consequence of not only
the direct effect of moisture supply on decomposition, but
also an indirect effect of changes in substrate supply and
quality. de Ruiter and co authors studied nitrogen miner
alization conducted at a wheat field. The impact of
microfaunal functional groups on N mineralization was
evaluated by calculating the impact of group deletion.
The results showed that the effect of the removal of a
group may exceed the direct contribution of this group to
N mineralization rather considerably, with amoebae and
bacterivorous nematodes having values of 18% and 28%,
and 5% and 12% for, respectively, direct contribution
toward and impact of deletion upon overall N mineraliza
tion. Influence of the transitions of soil microorganisms
between dormant and active stages was studied by
Blagodatsky and co authors. Such transitions were shown
to be important for biogeochemical cycling and the rate of
organic matter decomposition.
A combination of a detailed monitoring program, and
statistical and simulation modeling has been used in a
study of ecological patterns in the Heron Wood
Reserve, located at the Dawyck Botanic Garden in
Scotland. The suite of statistical techniques included
ANOVA, ANCOVA, correlation analysis, CCA, factor
analysis, and stepwise regression modeling. The study
revealed a number of indirect effects resulting from a
complex multivariate interplay among ecosystem compo
nents. For example, the results suggested that both direct
negative and indirect positive effects of the microarthro
pod community on specific fungal groups appeared to
take place. The relatively high local abundances of the
dominant collembolan Folsomia might have caused local
declines in ectomycorrhizal fungi, reflected, in turn, in the
increase in pH (Whist this work was in press, Dr. Peter
Shaw has checked identification of the dominant
Folsomia species (previously referred to as F. candida)
from the Dawyck ecosystem study, and has shown that
it appears to fit the description of F. inoculata.) However,
for those samples where the dominant Folsomia were less
abundant, overcompensatory fungal growth due to graz
ing by mites and other collembola was implicated.
Complex effects were also shown for bacteria, nematodes,
protozoa, plants, and soil properties.
Indirect Effects in Aquatic Systems
Awareness of indirect interactions in aquatic environment
has rather a considerably long history, and clearly pre
sented examples can be found in works (among others)
of, for example, Mortimer, Hutchinson, and Reynolds. In
particular, in an earlier review by Abrams it was even
suggested that most studies specifically addressing behav
ior mediated indirect effects tend to be conducted in
86
Indirect Effects in Ecology
freshwater ecosystems, while many of the early demonstra
tions of density mediated indirect effects were done in
community studies in marine habitats. Likewise, much of
the knowledge related to indirect ecological interactions
has been contributed through the development and appli
cations of the methods of simulation modeling and network
analysis in relation to aquatic environment. Consequently,
simulation models capable of demonstrating indirect inter
actions in aquatic biogeocenoses (e.g., the Lake 2 model of
J. Solomonsen) are widely used for teaching in the educa
tional establishments across the world.
Recent studies of indirect effects in aquatic environ
ment variously involved a combination of the empirical
approach and an application of statistical techniques,
methods of network analysis, simulation modeling using
‘What if’ scenarios, and sensitivity analysis. One of the
perhaps most frequently addressed examples of indirect
effects in aquatic environment relate to trophic cascades,
which involve propagation of the effect along a vertical
trophic chain consisting of three or more components
connected by grazing or predation. For instance, as was
recently investigated by Daskalov, a decrease in the top
predator’s population in the Black Sea due to overfishing
resulted in a ‘trophic casade’, leading to an increase in the
abundance of planktivorous fish, a decline in zooplankton
biomass, and an increase in phytoplankton crop.
The previously made statements regarding the abiotic
components (see above) can be emphasized with exam
ples related to the importance of detritus. For instance,
Carrer and Opitz found that in the Lagoon of Venice
about half of the food of nectonic benthic feeders and
nectonic necton feeders passed through detritus at least
once, while there was no direct transfer of such food
according to the diet matrix. Whipple provided an analy
sis of the extended path and flow structure for the well
documented oyster reef model. Few simple paths and
large number of compound paths were counted. The
study provided structural evidence for feedback control
in ecosystems, and illustrated importance of nonliving
compartments (in this case, detritus) for the ecosystem’s
functioning. Even for the model with a low cycling index
(i.e., 11%) multiple cyclic passage paths provided a con
siderable (22%) flow contribution. Therefore, it was
envisaged that for ecosystems with higher cycling indexes
the patterns observed should be even more pronounced.
Another noteworthy illustration of indirect effects in
aquatic ecosystems relates to the interdependency of bio
geochemical cycles. For example, Dippner concluded that
indirect effect of the silicate reduction in coastal waters
causes an increased flagellate bloom, due to a high avail
ability of riverborne nutrient loads. In a study of lake
Suwa (Japan), Naito and co authors have shown that the
physiological parameters of the diatom Melosira were the
important sources of the cyanobacterium Microcystis’ pro
duction variability. These results agree well with our
work on Rostherne Mere and suggest that the underlying
mechanism might be a common inverse relationship
between spring diatom and summer cyanobacterial
blooms resulting from the fact that the biogeochemical
cycles of Si and P in the aquatic environment are coupled
via the dynamics of primary producers (i.e., increased
concentrations of Si in spring lead to an increase in a
spring diatom bloom, and an increase in the removal of
P, N, and microelements from the water column with
easily sedimenting biomass at the end of the bloom; con
sequently, this may lead to a decrease in the summer
cyanobacterial development).
Role of Abiotic Components
Although the importance of abiotic ecosystem compo
nents is commonly recognized, most of the ecological
studies (including those addressing the indirect effects)
tend to study in detail only relationships among biota.
The restriction of the integrative synthesis to species
interaction only cuts off a plethora of useful environmen
tal studies related, for example, to issues of global climate
change. It should be noted, however, that the science of
ecosystem dynamics is highly interdisciplinary, and the
information relevant to the present discussion can, there
fore, be found not only in ecology and biology, but also
virtually in any section of natural and environmental
sciences, with geography, palaeontology, geoecology,
and climatology comprising the most obvious candidates.
In ecology, it is widely recognized that species inter
action can be mediated by a nonliving resource, and that a
species can potentially exert a selective force on another
species through nontrophic interactions. It should also be
noted that in nature many species are very well adapted
to modify their community and habitat (e.g., beavers by
changing the habitat’s hydrological regime, humans by
initiating dramatic changes in global climate and geo
chemical fluxes, earthworms by increasing aeration and
redistributing organic matter in soil, etc.). Changes in
physical characteristics of a habitat caused by the activity
of so called ‘ecosystem engineers’ may be regarded as an
extreme case of such nontrophic interactions. Often, how
ever, even if abiotic components are considered in terms
of detrital pathways and/or nutrient cycling, the effects
studied in detail are mostly confined to trophic interac
tions only. Furthermore, many indirect interactions occur
between different stages of ecosystem development and
are therefore easily overlooked and understudied. In eco
logical literature these interactions are sometimes called
‘historical effects’, ‘priority effects’, or ‘indirect delayed
regulations’. Consideration of these effects is particularly
important for the correct understanding of an overall
ecosystem functioning. Hence, if one abstracts from the
labels given to different branches of science, the impor
tance of abiotic ecosystem components and physical
Indirect Effects in Ecology
environment for ecosystem dynamics and evolutionary
development becomes increasingly obvious.
Indirect Effects of Global Relevance
Indirect relationships important on the global or subglobal
scale are often separated from their cause spatially and/or
temporally. For example, the dramatic increase in volcanic
activity (possibly caused by the impact of an asteroid) at
the end of the Mesozoic era is thought to have led to the
extinction of dinosaurs, which arguably stimulated the
eventual evolution of mammals (including humans). The
increased production and use of fertilizers in the 1950s led
to the increased phosphate inputs, eutrophication, and
decrease in water quality in many lakes, ponds, and reser
voirs during the subsequent decades. The increased
consumption of fossil fuels in the twentieth century led to
the increased emissions of carbon dioxide, which were
eventually followed by global warming and an apparent
increase in the frequency of natural disasters. This climate
change was probably accelerated by the depletion of the
planet’s ozone layer due to the CFC (chlorofluorocarbon)
containing deodorants and refrigerants.
It should be noted that indirect relationships are not
related just to the activities of humanity, but have been
important throughout the history of our planet. For exam
ple, a gradual development of the modern atmosphere
was largely due to the activity of cyanobacteria, which
were among the first organisms to produce oxygen as a
by product of their metabolism. The indirect implications
of the atmospheric oxygen enrichment were far reaching,
and led not only to profound global biological and geo
chemical changes, but also ultimately enabled the
development of Homo sapiens and its current civilization.
Last century, the line of thought started by Vernadsky
has eventually led to the creation of a new integrative
branch of natural sciences, sometimes referred to as ‘global
ecology’. Essentially, ‘global ecology’ encompasses meth
ods and scope of virtually all other environmental
disciplines, and is predominantly concerned with the
dynamics (including past and future) of the global ecosys
tem – the biosphere. As an example, it is worth mentioning
the now classic climatological research carried out by
Budiko and co workers, which led to the creation of a
half empirical model of the thermal regime of the atmo
sphere. This model was subsequently used to simulate past
and future dynamics of the atmosphere, and changes
between glaciation and interglacial periods. Furthermore,
the results obtained aided interpretation of human evolu
tion, and led to further research aiming to counteract
possible global change, for example, by injecting certain
substances into the stratosphere, and direct and indirect
consequences to which such manipulations may lead.
Currently, global climate change (principally related
to the increased concentrations of greenhouse gases) is
87
still one of the most discussed topics in ecology and
environmental sciences in general. While the detailed
review and the lively controversy of the discussions
related to this topic is outside the scope of this publica
tion, it is worth pointing out that the absolute majority of
studies dealing with it also inevitably deal with indirect
effects (although the exact term is often not mentioned).
Indirect Effects and Industrial Ecology
This article would be incomplete without mentioning of
studies and methods used in ‘industrial ecology’.
Industrial ecology is based on the analogy between
natural and industrial ecosystems, and aims to facilitate
the development of industrial recycling and cascading
cooperative systems by minimizing the energy consump
tion, generation of wastes, emissions, and input of raw
materials. Complex interplay among system components
has been taken into account in a large number of
waste management and industrial ecology studies.
Consequently, throughout the second half of the last and
the beginning of the present century, some substantial
progress has been made in various aspects of industrial
ecology, and in particular in understanding and account
ing for indirect effects.
One of the commonly used methods of industrial ecol
ogy is ‘life cycle assessment’ (LCA). It studies the
environmental aspects and potential impacts throughout
a product’s life (commonly referred to as cradle to grave
approach), from raw material acquisition through produc
tion, use, and disposal, and the same methodological
framework allows analysis of the impacts associated with
physical products (e.g., cars, trains, electronic equipment),
and services such as waste management and energy sys
tems. Similar to LCA, but usually with considerably
narrower system boundaries, are methods of energy
analysis, including, for example, energy footprinting
(which, effectively, constitutes calculations of how much
energy is spent and saved/recovered in all the processes
included within the chosen system boundary) and net
energy analysis (which in addition to the detailed energy
budgeting involves calculation of indicators such as incre
mental energy ratio and absolute energy ratio). For
example, on the basis of the energy budget estimates
for case studies from the UK and Switzerland it has
been argued that increasing recycling rates for plastic
and glass would improve the energy budget of waste
management programmes, and, therefore, benefit the
corresponding industrial ecosystems. Further modifica
tions of the energy analysis methods make fruitful use of
emergy and exergy budgets.
Another method popular in ‘industrial ecology’ is ‘eco
logical footprinting’. Basically, the method estimates the
area necessary to support (i.e., in terms of, for example,
production of food, energy, processing of wastes) current,
88
Indirect Effects in Ecology
past, or probable future functioning of particular geogra
phical (often administrative, for example, countries,
counties, towns) units. Despite numerous logistical pro
blems of interconversions, system boundary definitions,
and coefficient estimates, application of this method is
very useful and illustrative. For example (as illustrated by
Herendeen), out of all Western industrialized countries,
only the ecofootprints of Australia and Canada appear to fit
inside their borders (the rest of the ‘developed’ countries
appear to live on the expense of other territories).
Evolutionary Role of Indirect Effects
It has been postulated by a number of authors, and has
been proved mathematically by Fath and Patten, that
indirect effects often promote coexistence and the role
of indirect effects should, in general, increase in the
course of evolution. For example, in grassland commu
nities containing Rumex spp., insect herbivory (by
Gastrophysa viridula) appears to be a cost inherent in the
development of plants’ resistance to pathogenic fungi
(Uromyces rumicus). Another example relates to the fact
that infection of plants with endophytic fungi often
enhances plants’ competitive abilities via deterring gra
zers by production of toxic compounds (as a result, some
plants might have coevolved together with their endo
phytes, for example, coupled evolution of Festuca and
Acremonium spp.).
It should be noted, that indirect effects are important
for the evolution of not only natural, but also industrial
ecosystems. Traditionally, human society has developed
without the necessary due respect to the rules and pro
cesses governing the stability of its environment.
However, by analogy with natural ecosystems (i.e., as
regards recycling and cascading networks) industrial eco
systems should aim to facilitate the development of
recycling and cascading cooperative systems by minimiz
ing the energy consumption, generation of wastes,
emissions, and input of raw materials.
Approaches and Techniques Used to
Detect and Measure Indirect Effects
Detection and measurements of indirect effects are often far
from straightforward, and are mostly based on the intuition,
common sense, and prior knowledge of any particular sys
tem. Abrams and co authors described two major
approaches adopted in ecological studies, namely theoreti
cal and experimental. They stated that in practice, the
theoretical and empirical approaches may be regarded as
endpoints of a methodological continuum. Recently, how
ever, we have argued that the methodological continuum to
study indirect interactions is best represented by a triangle,
with observational, experimental, and theoretical nodes.
Within the theoretical approach, observations (and/or
carefully considered experimental data) are used together
with theoretical considerations to construct a model cap
able of investigating interactions among the components
incorporated in the model structure. This model is sub
sequently used to examine indirect effects between the
components. There are a number of drawbacks of this
approach, for example, difficulties related to obtaining
sufficient details about the components represented in
the models, unavoidable uncertainty as regards fluxes,
parameters, initial values, etc. This uncertainty may
mask the significance of the relationships studied, includ
ing indirect effects. Furthermore, as it is impossible to
reproduce all the complexity of a real ecosystem, any
model is a simplification of reality. Therefore, some of
the potentially important interactions may be lost just by
defining the model structure, while the importance of the
others may be considerably altered.
Within the experimental approach, densities of indivi
dual species are manipulated (e.g., by total removal) in
microcosms or experimental plots, and statistical analysis
(e.g., ANOVA, ANCOVA) are subsequently applied to
estimate the magnitude of indirect effects of manipula
tions on densities of other species. It has been argued that
this approach is best applied using a factorial design,
where the densities of a number of components (e.g.,
species or trophic groups) are changed both alone and in
combination. If implemented properly, this approach
leads to a straightforward estimation of net effects.
However, there is always a danger that some of the
indirect interactions have not manifested owing to un
avoidable time constraints of any experiment. Also,
partitioning of the registered net effects may be subject
to speculation. Experiments are often costly and by defi
nition are limited by their design and the hypotheses
tested. The simplicity of the experimental design may
mask the significance of the relationships studied for
trait mediated effects; measurements of population abun
dances may need to be supplemented by behavioral
observations, and/or biochemical, physiological, genetic,
and other analyses. Furthermore, there is always a big
question mark how applicable are the results obtained to
the processes happening in the real world.
Among mathematical methods which have been used
in studies of indirect effects in natural ecosystems are
statistical methods (e.g., regression and correlation analy
sis, PCA, factor analysis, CCA, ANCOVA, ANOVA),
simulation modeling (e.g., using ‘what if scenarios’, sensi
tivity and elasticity analysis), and methods of network
analysis. In particular, indirect interactions have often
been analyzed using methods of network analysis. For
example, Fath and Patten used methods of network
analysis to show that, in the ecosystem context, direct
transactions between organisms produce integral effects
more positive than a simple sum of direct effects. This was
Indirect Effects in Ecology
in line with the view that mutualism is an implicit con
sequence of indirect interactions and ecosystem
organization, and that the contribution of positive rela
tionships should increase along the course of evolution
and ecological succession.
It should be noted that all the methods so far applied to
investigations of indirect effects have both advantages and
limitations. Many of these have been previously addressed
and no attempt to discuss the benefits and disadvantages of
the techniques used to investigate indirect interactions has
been done in this article. Neither was it intended to address
any controversy and related discussion resulting from spe
cific applications (and/or implications of such applications)
of any particular method. It should be noted, however, that
the methodological framework of ‘comparative theoretical
ecosystem analysis’ (CTEA) (see below) suggests that the
mathematical techniques may be best used in concert, thus
allowing a detailed complementary insight into complex
patterns of mechanisms underpinning dynamics of natural
ecosystems.
Problems and Implications for
Environmental Management
There are many problems associated with studies of
indirect effects. Here we list the most general ones, in
the author’s view, resulting from the very nature of such
relationships, and the complexity of natural environment.
We also emphasize the potential of using indirect effects
in environmental management and caution as regards
their misuse and careful consideration.
Complexity and Uncertainty
Although the characteristics of indirect effects are
fairly readily established in a controlled laboratory experi
ment involving a very limited number (typically <5) of
interacting components, in the natural environment the
complexity of interactions renders their characterization,
in particular in practical terms as regards the outcome for
any specific ecosystem, rather extremely challenging.
Using the inverse of 100 community matrices (elements
of such an inverse matrix specify how the change in density
of a particular species affects the density of the other
species, with species pairs determined by the element’s
position in the matrix) obtained by randomly changing
the species specific parameters, Yodzis has shown that a
large number of predictions for overall interactions
between separate species were directionally undetermined,
that is, positive in some cases and negative in others. Owing
to such an uncertainty, some theoreticians have even ques
tioned whether credible predictions of the overall outcome
of indirect interactions could be made at all. As a conse
quence, there are numerous cases where the application of
89
seemingly appropriate environmental management meth
ods, and in particular biomanipulation, resulted in failure.
Separation in Time
Another problem related, in particular, to the detection of
indirect effects, is the fact that they often occur after a
considerable time lag. Unfortunately, research grants are
typically no longer than 3 years (often less). Hence it is
likely that in many studies the potential for indirect
effects has been considerably underestimated (if not
totally overlooked).
Not only are cause and effect often separated in time,
but they also may occur at different stages of the system’s
succession. This, however, may be turned to human
advantage, and used as a complementary measure of
environmental control. For example, it has been shown
that in freshwater lakes and reservoirs biogeochemical
cycle of P may be regulated by alterations in the biogeo
chemical cycle of Si, and stimulation of the spring diatom
growth may help to alleviate nasty cyanobacterial blooms
in summer. Clearly, due to inherent problems with eco
system complexity and uncertainty any application of
environmental management measures utilizing time lags
of indirect effects should be done with extreme caution.
Separation in Space
Geographical separation between the cause and the effect is
another inherent problem of indirect effects. For economic
and societal reasons, it is particularly well documented in
studies and practical applications of pollution and biologi
cal control, but numerous examples are readily available
from virtually any areas of ecology and environmental
sciences. As regards pollution damage, the well known
examples relate to coastal and benthic effects of oil spills,
consequences of lake acidification in Scandinavia due to the
transboundary transfer of sulphur dioxide since the begin
ning of industrial revolution, bioaccumulation of
radionuclides in the food webs of, for example, British
pastures following the Chernobyl fallout, etc.
In ecology, there is a growing awareness of placing the
dynamics of a local biological community in the broader
landscape and biogeocoenological context. For example,
in biocontrol applications, failure to anticipate indirect
effects resulting from the community openness and
exchange between localized subpopulations has led to a
number of costly failures and unexpected outcomes.
Following introduction, migrations of biocontrol agents
were shown to be capable of suppressing nontarget spe
cies in spatially removed areas. Certain traits of the agent
were shown to be related to elevated risks of nontarget
species extinction, for example, the risk of extinction of a
nontarget species increases with the decrease of its growth
rate and the increase of the agent’s attack rate compared
90
Indirect Effects in Ecology
to the target species. Hence the threat of extinction
appears to be particularly acute for a nontarget species
occupying habitat fragments co mingled with the habitats
occupied by target species, and when the target species is
relatively productive and the agent is only moderately
efficient at limiting the target species numbers.
Defining System Boundaries
Closely related to the ‘separation’ problem is the problem of
drawing the system’s boundaries. By narrowing the bound
aries of the system considered, many of indirect effects may
be reduced to the effects attributable to the variation in
external forcing. This is often convenient, in particular, if
the aim is to study responses in a controlled environment.
However, the scope for investigations of indirect effects in a
system with narrow system boundaries inevitably becomes
limited; by widening the boundaries and including more
components the potential for indirect effects is greatly
enhanced, but as the system’s complexity increases, so
does the associated uncertainty of any interpretations!
Implications for Environmental Management
Qualitative and quantitative account of indirect effects is
(albeit often implicitly) becoming a common part of
environmental management, and is indispensable for suc
cessful application of, for example, landscape engineering,
biomanipulation, biogeochemical manipulation, strategic
environmental assessment (SEA), and environmental
impact assessment (EIA). In particular, mathematical
methods can be helpful in this respect. For example,
Ortiz and Wolff used Ecopath with Ecosim software to
study benthic communities in Chile. They found that a
simulated harvest of the clam Mulinia generated a com
plex interplay involving direct and indirect effects, and
drastically changed the properties of the whole system.
McClanahan and Sala used a simulation model of the
Mediterranean infralittoral rocky bottom to study possi
ble effects of various management options. Running a
number of ‘what if’ scenarios they concluded that many
of potential changes are likely to be indirect effects caused
by changes in trophic composition. For example, if inver
tivorous fish were removed as part of a management
scenario, sea urchins would reduce algal abundance and
primary production, leading to competitive exclusion of
herbivorous fish. Although similar interactions were
known from tropical seas, these results were not antici
pated by previous field studies in the Mediterranean.
In some cases appreciation and reliance on indirect
effects may form the basis of environmental management
measures. To date, there are numerous relevant examples,
including, for example, biocontrol, bio and biogeochem
ical manipulation or application of chemicals to reduce
sediment P release for subsequent control of blue green
algae, infection of grass cultivars by endophytic fungi in
turf industry, etc. It should be reiterated, however, that a
sound knowledge of the system’s natural history is abso
lutely indispensable for an application of any
environmental control measure. The problems related to
the detection and investigations of indirect effects (with the
major ones listed above) are likely to provide challenging,
and often unexpected complications, frequently after a
considerable time period. Hence, a combination of empiri
cal and theoretical work should precede any practical steps,
and any desk study should be backed up by a thorough
monitoring plus (where necessary) experimental program.
Current and Further Directions
Further investigations of indirect effects are important both
for enhancing our understanding and therefore improving
management of specific ecosystems, and for general
development of ecology. Due to the increasing pace of
technological progress, collection of monitoring data
using automated techniques, and in particular remote sen
sing, is becoming increasingly easy. Combined with rapidly
increasing computing power and progressive development
of mathematical methods, this may provide the necessary
basis for a dramatic acceleration of investigations of indir
ect effects, in particular the ones manifesting on a global
level, and in cases when the geographical separation is an
issue. With progressive accumulation of long term data
sets, it is also likely that the effects occurring after a time
lag will become more readily discernable. To maximize the
benefits (i.e., for investigations of indirect effects) from the
technological development, however, it should be supple
mented by contemporary advances in the methodology of
their investigations.
It has been argued that analysis of indirect relation
ships in ecosystems may be greatly enhanced by the
application of a specialized methodological framework,
called CTEA. CTEA is aimed at bringing together sepa
rate lines of current investigations, hence combining them
in an integrative approach (see the section titled ‘Further
reading’). Further development and systematic applica
tion of CTEA is vital for improving the accuracy of
ecological forecasting, and has, therefore, potential socie
tal benefits related to issues of EIA and sustainable
development. It is suggested that further developments
should pay much attention to similarities and differences
of the indirect effects revealed in various types of ecosys
tems, or at different stages of the ecosystem development,
and that the characteristics (e.g., magnitude, sign, etc.) of
indirect interactions should be increasingly used for
describing differences in ecosystem state, structure, and
overall functioning. For example, analysis of specific
Emergent Properties 91
ecosystems may benefit from answering the following (to
name but a few) questions:
types of indirect effects are important for the overall
• What
functioning of an ecosystem under investigation?
does the importance of indirect effects compare
• How
with the importance of direct effects?
the pattern of indirect effects relatively constant, or
• Issubject
to (system specific) seasonal and longer term
•
•
•
•
changes?
How does the pattern of indirect interactions change
due to pollution, disturbance, and various management
practices?
Do indirect interactions predominant in an ecosystem
help to stabilize this ecosystem?
What is the relative contribution of indirect interac
tions to resistance, resilience, and facilitation of
successional changes?
How have the indirect effects changed during the
evolution of a particular ecosystem, and what was
their contribution toward the driving forces of this
evolution?
Further Reading
Abrams PA, Menge BA, Mittelbach GG, Spiller DA, and Yodzis P (1996)
The role of indirect effects in food webs. In: Polis G and Winemiller K
(eds.) Food Webs: Integration of Patterns and Dynamics,
pp. 371 395. New York: Chapman and Hall (also see other papers in
this book).
Budiko MI (1977) Global’naya Ekologiya (Global Ecology). Moscow:
Misl’(in Russian).
Fath BD and Patten BC (1998) Network synergism: Emergence of
positive relations in ecological systems. Ecological Modelling
107: 127 143.
Fath BD and Patten BC (1999) Review of the foundations of network
environ analysis. Ecosystems 2: 167 179.
Fleeger JW, Carman KR, and Nisbet RM (2003) Indirect effects of
contaminants in aquatic ecosystems. Science of the Total
Environment 317: 207 233.
Herendeen RA (1998) Ecological Numeracy. Quantitative Analysis of
Environmental Issues. Toronto: Wiley.
Kawanabe H, Cohen JE, and Iwasaki K (1993) Mutualism and
Community Organisation: Behavioral, Thoretical and Food Web
Approaches. Oxford: Oxford University Press.
Korhonen J (2001) Industrial ecosystems Some conditions for
success. International Journal of Sustainable Development and
World Ecology 8: 29 39.
Krivtsov V (2004) Investigations of indirect relationships in ecology and
environmental sciences: A review and the implications for
comparative theoretical ecosystem analysis. Ecological Modelling
174(1 2): 37 54.
Menge BA (1995) Indirect effects in marine rocky intertidal interaction
webs Patterns and importance. Ecological Monographs 65: 21 74.
Miller TE and Travis J (1996) The evolutionary role of indirect effects in
communities. Ecology 77: 1329 1335.
Patten BC, Bosserman RW, Finn JT, and Gale WG (1976)
Propagation of cause in ecosystems. In: Patten BC (ed.) Systems
Analysis and Simulation in Ecology, pp. 457 579. New York:
Academic Press.
Schoener TW (1983) Field experiments on interspecific competition.
American Naturalist 122: 240 285.
Strauss SY (1991) Indirect effects in community ecology Their definition,
study, and importance. Trends in Ecology and Evolution 6: 206 210.
Wardle DA (2002) Communities and Ecosystems. Linking the
Aboveground and Belowground Components. Princeton: Princeton
University Press.
Wootton JT (1994) The nature and consequences of indirect effects in
ecological communities. Annual Review of Ecology and Systematics
25: 443 466.
Wootton JT (2002) Indirect effects in complex ecosystems: Recent
progress and future challenges. Journal of Sea Research
48: 157 172.
Emergent Properties
F Müller, University of Kiel, Kiel, Germany
S N Nielsen, Danmarks Farmaceutiske Universitet, Copenhagen, Denmark
ª 2008 Elsevier B.V. All rights reserved.
Introduction
The History of the Concept
Emergence and Hierarchy
How Emergence Emerges
Classification of Emergent Properties
Quantifying Emergence
Further Reading
Introduction
or quantities, beyond easily measurable or predictable phy
sical parameters. The resulting properties have been
described on many levels of the biological hierarchy, from
simple physical systems, like laser beams, to the organization
of the whole biosphere within the Gaia concept of
J. Lovelock. The emergent property at one level is in general
Many biologists will recognize the statement that ‘‘the whole
is more than the sum of the parts’’ as a commonly used (but
hardly understood) phrase. This formulation refers to the
idea that there are systems which possess additional qualities
92
Emergent Properties
finding its causality at the subsystem components and the
interaction between them. For example, an organized form
of cell functions, stemming from self organized transforma
tions of cellular compounds, known as hypercycling may be
considered as an emergent entity; at the physiological levels
we are, for example, dealing with the mating behaviors of
organisms as results of hormone interactions. Similarly,
motion, feelings, or intelligent behavior occur as a conse
quence of special couplings of neurons. In addition, the
patterns in the development of ecosystems may not be
predictable from knowledge of organisms alone. Therefore,
emergent properties are not unusual phenomena, but simply
consequences of hierarchical organizations.
The concept of emergence found its way into ecology
through the proposal of E. P. Odum, who suggested that the
study of emergence should lead to a ‘new integrative dis
cipline’. This idea was due to the fact that studies of
complex systems had shown that the investigations of the
details alone were not adequate in predicting ecosystem
function and behavior. Neither were they sufficient to
explain a more advanced pattern like behavior and perfor
mance of ecosystems, for example, during the succession
from young systems toward more mature states.
bound to happen that are not easy to predict from the
basic knowledge of the system, no matter how extensive
this knowledge is.
Highly relevant to biology and ecology is the question
when an emergent property appears. This leads to
the distinction of ‘primary’ and ‘secondary’ emergence,
primary emergence being the first time an emergent prop
erty appears. To be conserved the property can
be reproduced again and again but in this case it is nomi
nated as a secondary emergence. Recent approaches to
emergence have come up with three further notions
of emergence: ‘computational emergence’, ‘thermodynamic
emergence’, and ‘emergence relative to a model’. The com
putational emergence deals with the patterns produced by
different computer programs, for example, cellular auto
mata systems developing complex distributions out of
simple rules from game theory. Thermodynamic emer
gence covers the establishment of highly complex, self
organized structures and their relations to the nonlinear,
far from equilibrium thermodynamics. Emergence relative
to a model defines emergence as the deviation of the actual
behavior of a physical system in comparison with an obser
ver’s model of it.
Summarizing these historical notions of emergence,
the following features can be stated:
The History of the Concept
properties are properties of a system which
• Emergent
are not possessed by component subsystems alone.
properties emerge as a consequence of the inter
• The
actions within the system.
fundamental types of interactions are found that
• Two
may be characterized as intra and inter connected
The concept of emergent properties originates in the
nineteenth century, finding the primary roots back in
Kantian philosophy. The term was coined by G. H.
Lewis as far back as 1875. A common definition from
that time states that ‘‘emergence is the denomination of
something new which could not be predicted from the
elements constituting the preceding condition.’’
Throughout the last century, several scientists have
addressed the concept from a more philosophical point of
view, resulting in the appearance of different descriptions
and explanations. The definitions have, in general, been
referring to subjective arguments, such as surprise, unex
pectancy, thus being clearly observer dependent. This has
strongly influenced the present approaches, and this com
prehension has often been connected with a flavor of
mysticism. Thus, the seriousness of the concept has
often been underestimated.
During the last decades, the use of the term emergent
properties has found widespread use in biological
sciences, especially, because it is clearly connected with
the growing implementation of the system approach in
ecology. The need for a holistic concept was due to the
failure of the traditional reductionistic research strategies
to explain the properties of ecosystems by the knowledge
of the behavior and the properties of the ecosystem con
stituents alone. Ecosystems are highly complex middle
numbered systems dominated by nonlinear relationships
between their constituents. In such systems, things are
•
•
ness, that is, connections within and between levels,
including controls. This point does not consider the
direction of the intra level interactions. Emergence is
based on both, upward and downward causation.
The historically emerged properties are considered
‘new’ with reference to their primary appearance.
These new properties appear at one level of a sys
tem and are not immediately deducible from
observation of the levels or units of which the sys
tem consists.
Emergence and Hierarchy
Emergence has been described at many levels of the
biological hierarchy. As argued above, the reason for
emergence is to be found in the hierarchical organization
of the system and the quantitative and qualitative char
acters of the ‘linkages’ within the structure. As biological
structures are often complex, this makes it hard to deter
mine the actual cause of emergence.
Hierarchy theory (see Hierarchy Theory in Ecology)
states that middle numbered systems – such as
Emergent Properties 93
ecosystems – can be comprehended if they are investi
gated on different levels of integration. Broad scale
levels can be assigned to high spatial extents, and low
typical frequencies, filtering the signals from lower
levels. These scale levels are spatially smaller and
their typical frequencies are higher. They are not able
to filter constraints from the higher levels, but their
potentials and interactions are building up the material
basis and the coordination functions of the higher level.
Emergent properties are created by both types of non
linear interactions. Therefore, the properties of specific
levels can be termed ‘hierarchical emergent properties’.
Of course, the interesting question is how these proper
ties emerge. Some examples might be helpful to
illuminate this question.
Prebiological Emergence
Several examples of emergent properties can be found
in physical and chemical sciences. They form an impor
tant prerequisite for protobiology and evolutionary
processes. Within the area of physics some examples
are nearly classical: Water (e.g., its wetness), which is a
simple molecule with a rather complex behavior, that is
unpredictable from knowledge about oxygen and hydro
gen alone, has often been used to demonstrate emergent
properties. Similarly, the sense of colors by the eyes is
not predictable by knowing a certain wavelength of
light.
Two famous chemical examples related to self
organized behavior of systems may also be mentioned,
the Bénard cells and the Bhelusov–Zabotinsky (BZ) reac
tion. In the case of Bénard cells, during specific
conditions, hexagonal, convective cells (the emerging
structures) form a fluid when a thermal gradient is
imposed on the experimental setup containment. In the
BZ reaction a special ratio of chemicals causes a mixture
to perform a pulsing pattern in colors with a period of
about one minute. The structure of these physico
chemical processes, gradients resulting in convective
cells, pulsing patterns, together with other observations
like the occurrence of Turing structures in chemical fluids,
spontaneous formation of lipid coacervates, might be cru
cial to our understanding of the emergence of life.
Protobiological Emergence
The appearance of the earliest life forms has often been
referred to as a primary emergence. Although many of the
properties occurring during this phase of evolution have
been repeated, over and over again their appearance
still qualifies them as emergent properties. As examples,
the emergence of life, emergence of animals, or the
emergence of bird feathers from reptile scales can be
mentioned to characterize situations of primary
emergence.
Many examples found in the literature deal with the
formation of the earliest cells. Biochemical cycles, the
organization and exchange of information by DNA or
RNA and the compartmentalization of material within
membranes are but a few examples. Molecular comple
mentarity, defined as ‘‘nonrandom, reversible coupling
of the components of a system,’’ has been argued to be a
widespread mechanism in biological systems and impor
tant for the understanding of the processes lying behind
emergent properties. The seemingly (self)organization of
molecules observed in prebiotic systems, such as
Turing structures and autocatalytic hypercycles (see
Autocatalysis), can be seen as emergent properties already
at a very low level of organization.
Emergence in Biological Systems
Emergent properties really come into play when biologi
cal systems reach higher levels of complexity. This
becomes evident already when cells or groups of cells
communicate with each other as in the case of hormones
and natural neural networks. Organs are composed of
cells, their individual functions are important only to
the organism as a whole. A heart, kidneys or lungs, are
vital but their function is not existent when they are on
their own. Organisms interacting as populations or socie
ties provide properties which cannot be explained by
properties of the individual organisms alone. They all
go together in what we consider as ecosystems and thus
are a part of the biosphere.
Cellular level. In regarding the outcome of interacting cells
many studies have been concentrating on the organization of
neuronal systems, which result in unexpected properties like
the ability to move, to sense, to be intelligent, and to emote.
The sensory systems, being connected to visual, auditory, or
other communicative processes are all playing a major role
in how successful living organisms are in performing specific
life strategies. Reliable senses, and responding the right way
to the received stimuli are crucial to the existence of many
life forms, in processes like finding food, knowing when and
where to escape, or creating bonds with other members of
the species, for example, during reproduction.
Neural networks, like in our brains, consisting of a huge
number of interconnected neurons, are so complex that
unforeseen patterns in responses are bound to occur and
have also been reported to exist. During the evolution of the
brain, emergent properties, together with new cell types,
local and large circuits have added up to the increasing
complexity of brain function. Motor control, the control
and coordination of motor activity are taken care of by our
brain passing on signals to the limbs or organs involved.
Organ level. Numerous cells, often during morphogen
esis differentiated in certain, specialized directions, form
94
Emergent Properties
organs, take up a particular task of the organism, like for
instance liver cells secreting enzymes, or kidney cells
filtering and cleaning the coelom. Although the formal
‘layout’ for this functionality is existent in the genetic
material of all cells, the eventual determination occurs
during the development of the organism and the actual
function of the organs may be viewed as emergent. The
brain as an organ may serve as an example of
this emergence: Here differentiated cells, with highly
specialized physiological properties, go together and cre
ate activity patterns that are far more complex than
expected from knowing the physiology of neural cells
alone. The whole becomes more than the sum of the parts.
Organism level. Complex behavior occurs among the indi
vidual organisms that cannot be determined exclusively by
internal factors. The sending, reception and interpretation of
signals from interagent organisms, the relationship(s) to the
outside, and thus semiotics play an important role, creating
patterns impossible to foresee if only the subsystems are
known. For example, in trees, the formation of branches
and leaf mosaics have been studied in a number of recent
investigations with modeling approaches as well as the allo
cation of resources between above and belowground
biomass and the related physiological mechanisms. A mod
eling study of this problem indicates a ‘complex integrated
growth pattern’ which may only be understood as an emer
gent property as it is claimed to have no direct or indirect
mechanistic basis related to subcellular activities. In a similar
manner it was shown that whole plant behavior is an emer
gent property arising from a rule based model of the system.
Communications between individuals, that is, their social
interactions within a population, are important to the func
tion of the organism as a whole and are indistinguishable
from the emergence of ethological features. Stressing the
importance of communication, may lead to an interpretation
of the communicative process as an emergent interpretation
of signs, which is described within the discipline of semiotics.
Population level. Populations are composed of individual
organisms, interacting in various ways, differing in quan
tity and quality, throughout the biological system. The
interactions may vary in character according to the com
plexity. At the one end of the spectrum, we find the single
cell organisms interacting mostly on a material basis
(matter fluxes). At the other end, there are colonial organ
isms forming complex societies, where brains, senses,
memory, and thus informational interaction become
dominating. Emergent properties as a result of individual
level behavior and interactions in populations of social
insects have been argued in several studies. For instance,
the distribution of food to larvae of the fire ant has been
argued as emerging from interactions between indivi
duals, workers, and larvae. Cellular automata models
were used to study the short time oscillations in ant
colonies. The nonlinear dependencies describing the rela
tionships between, and the movement of, individuals
explain this behavior. The resulting oscillations were
found to be emergent properties of the colony.
Ecosystem level. Ecosystems are inherently complex as they
are composed of an embedded hierarchy of all the pre
viously mentioned subsystems in close interaction with
abiotic factors. Emergence is to be expected, but surprisingly
few reports exist at this level, before all analyzing micro
cosms, forest ecosystems, predator–prey relationships, food
webs, and the organization of aquatic communities.
Ecosystem behavior is often analyzed through model
ing studies. The relation to emergent properties becomes
clear when looking at recent efforts of structural dynamic
modeling, where the changes in ecosystem composition
and structure over time are analyzed. Another example is
the work of B. C. Patten on the propagation of matter–
energy through the ecosystem network, leading to the
discovery of the importance of ‘indirect effects, quantita
tive and qualitative utilities’ of the system, results that
are highly surprising and unexpected, and as such are
emergent properties (see Ecological Network Analysis,
Environ Analysis). Both examples link to higher level
information expressions such as ascendency, different
kinds of entropy or information derived descriptors like
exergy (see Exergy).
The ability of the ecosystem to perform with systema
tic directional changes in some macroscopic characters,
not predictable from knowledge about the single ecosys
tem members alone, has been discussed since 1967 on
the basis of the 24 principles of ecosystem development
during succession in the second edition of E. P. Odum’s
Fundamentals of Ecology. Many other factors, known as
indicators, orientors, or goal functions have been pre
sented since then (Table 1).
How Emergence Emerges
The concept of emergent properties refers very clearly to,
and must be seen in tight connection with, at least two other
concepts often occurring in literature on modern ecosystem
theory, the concepts of hierarchy and self organization.
In connection with hierarchy, the emergent properties
are seen as outcomes of ecosystem organization where
supersystems are formed with subsystems as constituents
and where the properties are observable at the supersystem
level only. Here the emergent property is an outcome of a
certain way of organization. To exemplify this point, we
might look at the following hierarchical features:
1. Individual level: individual nutrition budgets – fora
ging strategies.
2. Population level: species nutrition efficiencies – intras
pecific food competition.
3. Ecosystem level: nutrient cycling – food webs.
Emergent Properties 95
Table 1 Some ecosystem orientors
Immature state
Mature state
Properties of the dominating species
Rapid growth
R-selection
Quantitative growth
Small size
Short life spans
Broad niches
Slow growth
K-selection
Qualitative development
Large size
Long life spans
Narrow niches
Properties of production
Small biomass
High P/B ration
Low respiration
Small gross production
Large biomass
Low P/B ratio
High respiration
Medium gross production
Properties of nutrient flows and cycles
Simple, rapid, and leaky
Small storage
Extrabiotic
Small amounts of detritus
Rapid nutrient exchange
Short residence times
Minor chemical heterogeneity
Loose network articulation
Low diversity of flows
Undeveloped symbiosis
Complex, slow, and closed cycles
Large storage
Intrabiotic nutrient distribution scheme
Large amounts of detritus
Slow nutrient exchange
Long residence times
High chemical heterogeneity
High network articulation
High diversity of flows
Developed symbiosis
Properties of the community
Low diversity
Poor feedback control
Poor spatial patterns
High diversity
Developed feedback control
Developed spatial patterns
Thermodynamic and integrative system properties
Poor hierarchical structure
Close to equilibrium
Low exergy storage
Small total entropy production
High specific entropy production
Small level of information
Small internal redundancy
Small path lengths
Low ascendency
Poor indirect effects
Small respiration and evapo-transpiration
Small energy demand for maintenance
Developed hierarchical structure
Far from equilibrium
High exergy storage
High total entropy production
Small specific entropy production
High level of information
High internal redundancy
High path lengths
High ascendency
Developed indirect effects
High respiration and evapo-transpiration
High energy demand for maintenance
The features, that are optimized throughout natural successions, provide several characteristics of emergent properties: They are only observable at the
ecosystem level (which is the typical and the lowest logical level to describe, e.g., cycling phenomena), and they are based on self organized processes.
They can not be explained on the basis of knowledge of the parts alone, and the emergence creating processual linkages between the subsystems are
nonlinear processes. From the hierarchy based viewpoint also the additive features (e.g., size, biomass, life spans) can be categorized as emergent
properties because their extensions are dependent on the scale of observation and because they also are based on internal system interrelations.
4. Landscape level: lateral nutrient transfers – food webs
including large scale predators.
On the other hand, the ability of biological systems to
arrange themselves in a special manner, for example, in
a hierarchical way, is in itself a property which emerges
as a consequence of the properties of its constituents,
but the organization and the function for sure cannot
always be foreseen. Thus, the capability of self organi
zation can be seen as an emergent property itself
(Figure 1).
The existence of emergent properties is based on the
system’s organization (built up by structures and func
tions) whereby the interrelations (energy, matter, water
and information flows, communications) play an impor
tant role. Some conditions of the system’s state add up to
the increased chances that emergent properties will
appear. For example, instabilities seem to be important
conditions that support emerging processes, especially
referring to evolutionary emergence. Stable periods may
lead to the emergence of new structures through bifurca
tions. As systems move toward the state of minimum
96
Emergent Properties
Inter-level
relation
lntra-level
relation
Figure 1 Biological entities are often organized in a hierarchical
manner, whereby the emergent properties of a certain level are
based on the interrelations between the lower levels, while both
are constrained from the highest level linkages.
dissipation they are, at the same time, moving toward
bifurcation points with possibilities of further evolution
to occur. Similarly broken symmetries, complementarity
has been proposed as a global mechanism.
Classification of Emergent Properties
From the presentation of the concepts above it can be
seen that emergence and emergent properties will not
easily find a clear, consistent and unifying definition for
covering all the cases described. The widespread and
‘loose’ use of the concepts over a vast range of areas at
first glimpse simply shows confusion. However, it is pos
sible to establish some typology of the areas where, and
the ways in which, the concepts have been used, following
Figure 2.
First, emergent properties might appear through evo
lution of the systems, primary emergence, hereafter only
being repeated. This characteristic may be called ‘evolu
tionary emergence’. As structures are integrated, new
organizational forms, as previously mentioned often hier
archical, occur (‘hierarchical emergence’).
Taking the view that emergent properties do exist and
that the reductionistic approach to science will not (dis)
solve the problem so it eventually disappears may allow
us to establish a schematic relationship between the vari
ous categories of ecosystem properties.
One major line follows a direction of research problems,
the search for the unexplained and not understood. This
lies close to using emergent properties as research strate
gies, while the extreme leads to the reductionist approach.
This is more or less the situation at the second line, where
properties are ‘collective’ and additive, that is, that the
properties are the sum of the whole, and may be explained
at subsystem level, provided sufficient knowledge exists. At
the other end, the attitude that only holistic studies will
lead to increased understanding might be taken.
Along the third line, we find the core of emergence,
and following the above points the respective features
may be divided in an evolutionary line and in a hierarchal
line. Here emergence is basically represented as a func
tion of time and space. The evolutionary process was
Global
emergent prop.
Ecosystem
properties
Hierarchical
properties
Research problems
Formalistic
Thermodynamic
emergent prop.
Computational
emergent prop.
Emergent prop.
relative to model
Ecosystem
properties
Interactions
Additive relations
Collective
properties
Emergent
properties
Dynamic
Evolutionary
properties
first time
Primary
emergent prop.
Secondary
emergent prop.
repetition
Figure 2 An attempt to form a typology of emergent properties.
Emergent Properties 97
described above and deals with primary and secondary
emergence. The organizational, hierarchical line includes
four areas described in the previous sections: ‘global
emergent properties’ as a function of local rules and
local interactions, ‘thermodynamic emergent properties’
dealing with emergence as a consequence of mainly the
second thermodynamic law, the emergence of (dissipa
tive) structures as a result of thermodynamic gradients.
‘Computational emergence’ is also based on global pat
terns emerging form local rules. As mentioned above,
emergent properties also appear as a result of models
being used to analyze the problem, which is called ‘emer
gent property relative to a model’.
Quantifying Emergence
Several authors argue that in any attempts to formalize or
quantify the concepts, true emergent properties should
be observer independent. This does not necessarily
mean that emergent properties should be observation
independent. Observations undertaken by different
methods result in differences in acquired knowledge.
This means that emergent properties can be defined as
the differences in knowledge gained by the observation of
a system by two different methods. This is partly reflected
by the computational emergence.
It is this observer dependency that leaves a way open
for the quantification of emergent properties. Emergent
properties could then be expressed in a semiquantitative
way by the use of an ‘index’ derived of Kullback’s measure
of information (Figure 3). This involves moving the
normal reference frame in information theory assuming
the a priori knowledge of the system to be zero, which is
not necessarily the case.
Object/system
Subject/observer
Information
gain
I ***
Xi+1
I **
Emergence
path
I*
I0
Information/level
Xi
Hierarchical
levels
Xi–1
Figure 3 Quantification of emergence, based on Kullback’s
measure of information, might be carried out from quantifying the
difference between actual observed, a posteriori, behavior or
composition of a system and what may be predicted from a priori
knowledge about subsystems. The analysis may be carried out at
various levels of hierarchy, differing in emergence value.
Rather in ecology, we do possess some knowledge
about the system and what we usually refer to are the
deviations in what we observe in the systems or models of
systems compared with our expectations built on previous
knowledge. The way of quantifying emergence has to be
built on the use of computers and models. If our knowl
edge gained hitherto is synthesized and treated in a
computer model (from traditional ecological science) is
p, and the outcome of an experiment or observations of a
system differs by p the emergent properties can be
calculated by the following:
Emergence ¼
X
p ln
p
p
which correlates emergence to the concept of exergy.
Emergence now is a consequence of information gained
between observations.
The question is if emergence in this manner will, at the
end, dissolve itself and disappear as knowledge increases,
which refers to the above debate of reductionism versus
holism.
Many of the concepts used to characterize ecosystems
are based on various numerical treatments of data
observed in the ecosystem. Since the concepts are
immediately deducible (calculable) from certain knowl
edge about the components of the ecosystem, for
example, numbers, species, biomass etc. such concepts
cannot be coined as emergent property but rather as a
‘collective’ property of the system. An interesting corre
sponding analog in this context are the macroscopic
properties from thermodynamics such as entropy and
parallels in formulation of formulas. Reductionism cannot
win the debate since it will be impossible to achieve
enough knowledge. If not for anything else, then
for thermodynamic reasons, since the achievement of
more and more detailed knowledge becomes more and
more expensive in terms of not only energy but also
dissipation.
Meanwhile, what strikes is that such a traditional,
vertical organization of systems is not mandatory in
order to produce emergent behavior. Vertical, here, refers
to levels being either higher or lower in the hierarchy.
Rather only parts are needed, of which none have actual
regulatory functions and therefore should be evaluated or
ranked higher than the other(s). Emergent properties can
occur also in horizontally organized systems, emergence
appearing alone as a consequence of interactions at the
same level. The study of these intra level relationships
and their consequences to the higher levels in the hier
archy may be important to investigate in the future.
See also: Autocatalysis; Ecological Network Analysis,
Environ Analysis; Exergy; Hierarchy Theory in Ecology.
98
Self-Organization
Further Reading
Bhalla US and Iyengar R (1999) Emergent properties of networks of
biological signalling pathways. Science 283: 381 387.
Breckling B, Muller F, Reuter H, Holker F, and Franzle O (2005)
Emergent properties in individual based ecological models
introducing case studies in an ecosystem research context.
Ecological Modelling 186: 376 388.
Cariani P (1992) Emergence and artificial life. In: Langton G, Taylor C,
Farmer JD, and Rasmussen S (eds.) Artificial life II, pp. 775 797.
Redwood City: Addison Wesley.
Conrad M and Rizki MM (1989) The artificial worlds approach to
emergent evolution. BioSystems 23: 247 260.
Emmeche C, Køppe S, and Stjernfelt F (1993) Emergence and the
Ontology of Levels. In Search of the Unexplainable. Arbejdspapir.
Afdeling for Litteraturvidenskab. Copenhagen: University of
Copenhagen.
Morgan CL (1923) Emergent Evolution. Williams and Norgate.
Nielsen SN and Muller F (2000) Emergent properties of ecosystems.
In: Joergensen SE and Muller F (eds.) Handbook of Ecosystem
Theories and Management, pp. 195 216. Boca Raton, FL: Lewis
Publishers.
Salt GW (1979) A comment on the use of the term emergent properties.
American Naturalist 113(1): 145 148.
Wicken JS (1986) Evolution and emergence. A structuralist perspective.
Rivista di Biologia/Biology Forum 79(1): 51 73.
Wieglieb G and Broring U (1996) The position of epistemological
emergentism in ecology. Senckenbergiana maritima
27(3/6): 179 193.
Self-Organization
D G Green, S Sadedin, and T G Leishman, Monash University, Clayton, VIC, Australia
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Historical Comments
Theories of Self-Organization
Self-Organization in an Ecological Setting
Practical Considerations
Further Reading
Introduction
Self organization is the appearance of order and pattern
in a system by internal processes, rather than through
external constraints or forces. Plant distributions provide
examples of both constraints and self organization. On a
mountainside, for instance, cold acts as an external con
straint on the ecosystem by limiting the altitude at which
a plant species can grow. Simultaneously, competition for
growing sites and resources leads to self organization
within the community by truncating the range of altitudes
where plant species do grow. Self organization can also be
seen among individuals within a population (e.g., within
an ant colony or a flock of birds) and within individuals
(e.g., among cells during development) (Figure 1).
A growing understanding of ways in which internal
processes contribute to ecological organization has pro
vided new perspectives on many phenomena familiar
from traditional ecology. Self organization usually
involves interactions between components of a system,
and is often closely identified with complexity. Also asso
ciated with self organization is the idea of emergence:
that is, features of the system emerge out of interactions,
as captured by the popular saying, ‘‘the whole is greater
than the sum of its parts.’’ It is necessary to distinguish
between emergent features and other global properties of
Figure 1 Effect of competition on plant distributions on a
gradient. The two plant species shown are adapted to different
conditions, which are here found at either end of the slope. At left,
there is no competition, so the distributions merge into one
another. At right, competition truncates the distributions, leading
to sharply defined altitudinal zones.
a system. For instance, although biomass production in a
forest is a global property, it is simply the sum total of
production by all the organisms within the forest. A
stampede, on the other hand, is behavior that emerges
when panic spreads from one animal to another within a
herd.
Semantic and philosophical issues sometimes lead to
confusion about self organization. Self organizing sys
tems are usually open systems, that is, they share
information, energy, or materials with their surroundings.
However this does not necessarily mean that the external
Self-Organization
environment controls or determines the way they orga
nize. A growing plant, for instance, absorbs water, light,
and nutrient from its environment, but its shape and form
are determined largely by its genes.
Also, in considering self organization, it is important to
clearly identify the system concerned, and in particular,
what is external and what is internal? This issue arises in
the difference between a community and an ecosystem.
For a community, which consists of the biota of an area,
the effect of (say) soil is an external constraint. However,
for the corresponding ecosystem, which would include
soils, the interactions between plants, microorganisms,
and soil formation are internal processes. Defining the
physical limits of an ecosystem poses similar problems.
A lake, for instance, is not a closed ecosystem. Among
other things, water birds come and go, removing some
organisms and introducing others.
99
role of adaptation in self organization. He suggested that
seven basic elements are involved in the emergence of
order in complex adaptive systems. These include four
properties – aggregation, nonlinearity, flows, and diver
sity – and three mechanisms – tagging, internal models,
and building blocks. In contrast, Stuart Kauffman’s work
on autocatalytic sets within Boolean networks emphasizes
ways in which self organization may structure biological
systems independent of selection. Likewise, embryologist
Brian Goodwin suggested that to understand macroevo
lution, we require a theory of morphogenesis which takes
account of physical, spatial, and temporal dynamics in
addition to selection. The work of James Kay provided
an interpretation of life from a thermodynamic perspec
tive, arguing that self organizing systems maximize the
dissipation of gradients in nature. In particular, Kay
argues that over time, ecosystems evolve to dissipate
energy more efficiently by becoming increasingly com
plex and diverse.
Historical Comments
Self organization as a widespread phenomenon first came
to the attention of researchers during the mid twentieth
century. The interest in self organization comes from
many different fields of study. The biologist Ludwig von
Bertalanffy drew attention to the role of internal interac
tions and processes in creating organization within
biological systems. His ‘general systems theory’ drew
heavily on analogies to highlight the existence of common
processes in superficially different systems. Meanwhile,
W. Ross Ashby and Norbert Wiener explored self
organization from the perspective of communications
and feedback in the control of systems. Ashby introduced
the term self organizing in 1947. Wiener coined the
term cybernetics to refer to the interplay of control
systems and information. In the 1950s, systems ecologist
H. T. Odum collaborated with engineer Richard
Pinkerton to develop the principle of maximum power,
which states that systems self organize to maximize
energy transformation.
During the 1970s and 1980s, increasing computing
power made it possible to use simulation to explore the
consequences of complex networks of interactions. By the
last two decades of the twentieth century, the nature and
implications of biological self organization were increas
ingly being explored as a part of the complexity theory.
The new field of Artificial life (Alife), initiated by pio
neers such as Chris Langton, Pauline Hogeweg, and
Bruce Hesper, has produced a series of seminal models
that demonstrate self organization in a variety of ecolo
gical and evolutionary contexts. Around the same time,
H. T. Odum introduced the systems concept of ‘emergy’
to represent the total energy used in developing a process.
By the 1990s, researchers were looking for broad based
theories of self organization. John Holland stressed the
Theories of Self-Organization
Thermodynamic Basis
In physical terms, the phenomenon of self organization
appears at first sight to be ruled out by the second law of
thermodynamics, which states that in any closed system,
entropy increases with time. In this sense, living systems
seem to fly in the face of thermodynamics by accumulat
ing order. However, self organizing systems need not be
closed. Open systems, including living things, share
energy and information with the outside environment.
In the late 1960s, Ilya Prigogine introduced the idea of
dissipative systems to explain how this happens. He
defined dissipative systems to be open systems that are
far from equilibrium. Dissipative systems have no ten
dency to smooth out irregularities and to become
homogeneous. Instead, they allow irregularities to grow
and spread. Physical examples include crystal formation.
Biological systems, including cells, organisms, and eco
systems, are all examples.
The Network Model
An important source of self organization is provided by
the interactions and relationships between the objects that
comprise a complex system. Patterns of such relationships
are captured by the network model of complexity.
Networks capture the essence of interactions and rela
tionships, which is a fundamental source of complexity. A
graph is defined to be a set of nodes (objects) joined by
edges (relationships) and a network is a graph in which
the nodes and/or edges have values associated with them.
In a food web, for instance, the populations form nodes,
and the interactions between them (e.g., predation) form
100
Self-Organization
the edges. In a landscape, spatial processes and relation
ships create many networks. For instance, the nodes might
be individual plants and the corresponding edges would
be any processes that create relationships between them,
such as dispersal or overshading. In an animal social
group, the nodes would be individuals and the edges
would be relationships such as kinship or dominance.
Nodes that are joined by an edge are called neighbors.
The degree of a node is the number of immediate neigh
bors that it has. A path is a sequence of edges in which the
end node of one edge is the start node of the next edge, for
example, the sequence of edges A–B, B–C, C–D, D–E
forms a path from node A to node E. A cycle is a path that
ends where it starts, for example, A–B, B–C, C–A. A
network is called connected if, for any pair of nodes,
there is always some path joining them (otherwise it is
disconnected). The diameter of a network is the maxi
mum separation between any pair of nodes. Clusters are
highly connected sets of nodes.
The importance of networks stems from their univer
sal nature. Network structure is present wherever a
system can be seen to be composed of objects (nodes)
and relationships (edges). Less obvious is that networks
are also implicit in the behavior of systems. In this respect,
the nodes are states of the system (e.g., species composi
tion) and the edges are transitions from one state to
another.
Sometimes, network structure plays a more important
part in determining the behavior of a system than the
nature of the individual components. In dynamic systems,
for instance, cycles are associated with feedback loops. In
disconnected networks, the nodes form small, isolated
components, whereas in connected networks, they are
influenced by interactions with their neighbors. Self
organization in a network can occur in two ways: by the
addition or removal of nodes or edges, or by changes in
the values associated with the nodes and edges.
Several kinds of network patterns are common and
convey important properties.
•
•
•
•
A random network is a network in which the nodes are
connected at random. In a random network of n nodes,
the degrees of the nodes approximate a Poisson dis
tribution, and the average length (L) of a path between
any two nodes is given by L ¼ log(n)/log(d), where d is
the average degree.
A regular network is a network with a consistent pat
tern of connections, such as a lattice or cycle.
Small worlds fall between random networks and reg
ular networks. They are typically highly clustered, but
with low diameter. A common scenario is a system
dominated by short range connections, but in which
some long range connections are also present.
A tree is a connected network that contains no cycles. A
hierarchy is a tree that has a defined root node. For
•
instance, the descendents of a particular individual
animal (the root of the tree) form a hierarchy deter
mined by birth. Trees and hierarchies are closely
associated with the idea of encapsulation.
A scale free network is a connected network in which
the degrees of the nodes follow an inverse power law.
That is, some nodes are highly connected, but most
have few (usually just one) connections.
Encapsulation
Encapsulation is the process by which a set of distinct
objects combine to act as a single unit. Individual fish, for
example, form a school by aligning their movements with
their neighbors. Because smaller objects usually merge
into larger wholes, encapsulation is often linked to ques
tions of scale. Encapsulation is closely associated with the
idea of emergence. The whole emerges when individuals
become subsumed within a group in relation to the out
side world. There are many examples in ecology.
Ecosystems are communities of interacting organisms;
populations are groups of interbreeding organisms; and
schools, flocks, and herds are groups of animals moving in
coordinated fashion. In all of these cases, the individuals
may not be permanently bound to the group, unlike cells
within the human body. Cellular slime molds present an
intermediate case in which cells sometimes act indepen
dently but at other times aggregate to form a multicellular
individual.
Various ecological theories are based on the assump
tion that encapsulation plays an important role in
ecosystem structure and function. The concept of ecosys
tem compartments implies that a community is formed of
distinct groups (compartments) consisting of mutually
interacting species, but the interactions between the
groups are limited.
Connectivity and Criticality
Criticality is a phenomenon in which a system exhibits
sudden phase changes. Examples include water freezing,
crystallization, and epidemic processes. Associated with
every critical phenomenon is an order parameter, and the
phase change occurs when the order parameter reaches a
critical value. For example, water freezes, when its tem
perature falls to 0 C. A wildfire spreads when fuel
moisture falls below a critical level (else it dies out).
Changes in the connectivity of a network have impor
tant consequences and often underlie critical phenomena.
When a network is formed by adding edges at random to a
set of N nodes, a connectivity avalanche occurs when the
number of edges is approximately N/2. This avalanche is
characterized by the formation of a connected subnet,
called a unique giant component (UGC), which contains
most of the nodes in the full network. The formation of
Self-Organization
the UGC marks a phase change in which the network
shifts rapidly from being disconnected to connected.
Any system that can be identified with nodes and
edges forms a network, so the connectivity avalanche
occurs in many settings and is the usual mechanism
underlying critical phase changes.
The connectivity avalanche has several important
implications. For interacting systems, it means that the
group behaves either as disconnected individuals, or as a
connected whole. Either global properties emerge, or they
do not: there is usually very little intermediate behavior.
Landscape connectivity provides an important ecological
example of critical phase change.
Phase changes in connectivity also underlie criticality in
system behavior. The degree of connectivity between states
of a system determines the richness of its behavior. Studies
based on automata theory show that if connectivity is too low,
systems become static or locked in narrow cycles. If connec
tivity is too high, systems behave chaotically. The transition
between these two phases is a critical region, popularly
known as the ‘edge of chaos’. It has been observed that auto
mata whose state spaces lie in this critical region exhibit the
most interesting behavior. This observation led researchers
such as James Crutchfield, Christopher Langton, and Stuart
Kauffman to suggest that automata need to reside in the
critical region to perform universal computation. More spec
ulative is their suggestion that the edge of chaos is an essential
requirement for evolvability (the ability to evolve) in complex
systems, including living things. Others have proposed that
living systems exploit chaos as a source of novelty, and that
they evolve to lie near the edge of chaos. These ideas are
closely related to self organized criticality (SOC).
Self-Organized Criticality
SOC is a phenomenon wherein a system maintains itself in
a critical or near critical state. A classic example is the
pattern of collapses in a growing pile of sand. Because
information theory suggests that systems in critical states
are most amenable to information processing and complex
ity, self organized criticality has been proposed as a
component of collective behavior in ant colonies, societies,
ecosystems, and large scale evolution. SOC is character
ized by events whose size and frequency distributions
follow an inverse power law. However, it is often difficult
to distinguish genuine cases of SOC from simple cause and
effect processes that exhibit similar distributions.
For example, ecosystems might tend toward critical
states through the following mechanism. If new species or
mutations appear in an ecosystem occasionally, then as
the variation in the ecosystem increases over time, so does
the probability of forming destabilizing positive feedback
loops. Such destabilizing interactions could initiate ava
lanches of extinctions, and the probable size of such
avalanches would be related to the preexisting
101
connectivity of the system. In this way, mutation, migra
tion, and extinction could keep the system near the
critical region, as the addition of new variation drives
the ecosystem out of subcriticality, while extinction ava
lanches prevent supercriticality. Proponents of this idea
point to extinction events, whose distribution follows an
inverse power law, as supporting evidence. However,
other explanations of this pattern, such as cometary
impacts, are also plausible.
Feedback
Feedback is a process in which outputs from a system
affect the inputs. Predator–prey systems are examples of
negative feedback. For instance, any increase in the size of
a predator population means that more prey are eaten, so
the prey population decreases, which in turn leads to a
decrease in the predator population. Reproduction is an
example of positive feedback: births increase population
size, which in turn increases the rate of reproduction,
which leads to yet more births. Feedback loops arise
when a sequence of interactions form a closed loop, for
example, A–B–C–A. Feedback loops play an important
role in food webs and ecosystem stability. Time delays in
the response within a feedback loop often lead to cyclic
behavior (e.g., in predator–prey systems).
Both positive and negative feedback are important in
self organization. By dampening changes, negative feed
back acts as a stabilizing force. It is one of the principal
mechanisms of homeostasis, the maintenance of dynamic
equilibrium by internal regulation. In contrast, positive
feedback magnifies minor deviations. An example is com
petitive exclusion: any small decrease in size of a
competing population is likely to lead to further
decreases, until it dies out (Figure 2).
Stigmergy
Stigmergy is a form of self organization that occurs when
parts of a system communicate by modifying their environ
ment. Many examples of stigmergy occur in the organization
of eusocial insect colonies. For example, in ant colonies,
objects such as food, larvae, and corpses are often stored in
discrete larders, nurseries, and cemeteries. Models show that
this civic order can emerge through interactions between the
ants and their environment. In the model, ants pick up objects
at random, and may drop them when they encounter similar
objects. Over time, this process creates piles of similar objects.
Positive feedback causes larger piles to grow at the expense of
smaller ones (Figure 3).
Synchronization
Synchrony can alter system level behavior by enhancing or
dampening nonlinearities. For example, when predator and
102
Self-Organization
(a)
(b)
–
–
+
+
+
+
+
+
–
–
+
+
+
+
+
+
+
+
+
Figure 2 The role of feedback in self-organization of a food
web. In this diagram, circles represent populations and arrows
indicate the influence (positive or negative) of one population on
another. In a food web, circular chains of interaction between
populations form feedback loops, as in the example shown here.
(a) The initial food web contains both positive and negative
feedback loops. Internal dynamics within the positive feedback
loops leads to the local extinction of several populations. (b) The
resulting food web contains only negative feedback loops, which
stabilize the community.
(a)
(b)
change in temperature or day length. Synchronized breed
ing conveys distinct advantages such as maximal
exploitation of resources and satiation of predators.
Different species often have co adapted simultaneous
seasonal behavior, such as birds that breed when butter
flies emerge. However, both the environmental cues,
and the physiological response, may differ among these
co adapted species. For example, great tits time their egg
laying by photoperiod. Winter moths are an important
food source during the breeding season, and they develop
more quickly at higher temperatures. As a result, recent
warm springs in Europe caused by climate change have
disrupted the synchronization between these species,
reducing food availability for nesting great tits and poten
tially destabilizing populations.
In other cases, synchronous behavior arises through
social contagion, where individuals imitate others. The
dynamics of such behavior are similar to those seen in
epidemiology. Social contagion can lead to coordinated
group behavior such as flocking, as well as disparate phe
nomena such as synchronized flashing in fireflies, and
‘fashions’ in mate choice among birds and fish. The emer
gence of synchronous behavior in these cases is highly
sensitive to the structure of social networks. Synchrony is
easily achieved when networks are highly connected (i.e.,
individuals can perceive a large number of other indivi
duals, or some individuals have very large influence).
However, in loosely connected networks, social contagion
can result in asynchronous waves or chaos.
Complex Adaptive Systems
Figure 3 The emergence of order by stigmergy and
feedback in an ant colony. Given a random scatter of objects (a),
ants sort objects by picking them up and dropping them again
when they find a similar object. This process creates piles, which
grow over time. Large piles grow at the expense of smaller ones
until only a few large piles remain (b).
prey populations are tightly coupled to one another, a stable,
negative feedback relationship can result where an increase
in prey causes increased predators and a subsequent
decrease in prey. In this case, the ecological interaction
acts like a thermostat regulating population size. However,
if the two populations respond at different rates, oscillations
or even chaotic behavior can occur instead. A classic exam
ple of such oscillations occurs in the interaction between
populations of hares and lynxes in the Arctic Circle.
Synchronized breeding behavior is common and includes
mass flowering in plants, mass breeding in birds, and mass
spawning among marine animals such as corals and squid. In
these cases, synchrony is usually achieved by individuals
responding to a common environmental cue, such as a
Complex adaptive systems (CASs) consist of diverse,
locally interacting components that are subject to selection.
Examples include learning brains, developing individuals,
economies, ecosystems, and the biosphere. In such systems,
hierarchical organization, continual novelty and adapta
tion, and nonequilibrium dynamics are known to emerge.
As a result, the behavior of a CAS is characterized by
nonlinearity, historical contingency, thresholds, and multi
ple basins of attraction. A key question in current CAS
research has been the relationship between resilience and
criticality. Some authors suggest that a CAS will generally
evolve toward self organized criticality. By being main
tained near the edge of chaos, such systems might
maximize information processing. In this way, criticality
might enhance the ability of CASs to adapt to changing
environments and efficiently utilize resources, making sys
tems become more resilient over time.
Artificial Life
The field of Alife uses simulation models to understand
biological organization by abstracting crucial features and
examining living systems ‘as they could be’.
Self-Organization
One of the most widespread representations used in
Alife models is the cellular automaton (CA). This is a grid
of cells in which each cell has a state (some property of
interest) and is programmed to behave in identical fash
ion. Each cell has a neighborhood (usually the cells
immediately adjacent to it) and the states of it neighbors
affect changes in a cell’s state. The most famous example
is the Game of Life, in which each cell is either ‘alive’ or
‘dead’ at any time. Despite its extreme simplicity, the
game showed that large numbers of interactions governed
by simple rules lead to the emergence of order within a
system. Cellular automata have been used to model many
biological and ecological systems. In models of fires, epi
demics, and other spatial processes, each cell represents a
fixed area of the landscape and the cell states represent
features of interest (e.g., susceptible, infected, or immune
organisms in an epidemic model).
Other prominent ALife models include Tom Ray’s
Tierra model, which demonstrated adaptation within
self reproducing automata. Craig Reynolds’ boids model
demonstrated that flocking behavior emerges from simple
interactions between individuals. James Lovelock’s
Daisyworld model showed the potential for biotic feed
back and adaptation to stabilize the biosphere.
Self-Organization in an Ecological Setting
Social Groups
Relationships between individuals create several kinds of
organizations within groups of animals.
Coordination between moving animals leads the for
mation of groups. Examples include swarms of insects,
flocks of birds, schools of fish, and herds of mammals.
Coordinated group movements, even in very large
groups, can be achieved by individuals obeying simple
rules, such as ‘keep close, but not too close, to your
neighbors’ and ‘head in the same general direction as
your neighbors.’
Several mechanisms that channel aggressive behavior
create social organization. In social animals, dominance
hierarchies reduce the potential costs of conflict over
mates and food. Adominance hierarchy emerges when
interactions between individuals result in physiological
and behavioral changes: for example, ‘winning’ a contest
may elevate testosterone, causing increased dominance
behavior, and evoking submissive behavior from indivi
duals who have been less successful in the past. In this
way, coherent transitive hierarchies can emerge even
when all individuals were initially equal. Similarly, terri
toriality reduces the costs of conflict over resources by
partitioning a landscape among a population.
Territoriality often generates spatial patterns, such as
regular distances between nests in seabird colonies. In
this case, the distance between nests is defined by the
103
maximum area that a sitting bird can defend without
abandoning her nest. More complex coordinated group
behaviors can emerge when individuals take on different
tasks and roles within groups. For example, within ant and
termite colonies, individuals can develop into a variety of
castes, each with distinct roles such as foraging, nest
defense, and nursing young. In honeybees, individuals
take on different roles at different life stages.
In some cases, upper limits exist on the size that social
groups can attain and depend on interactions between the
animals. In apes, for instance, where social bonds are main
tained by grooming, troop sizes tend to be 30–60 individuals.
Larger troops tend to fragment. Among humans, social
groups are usually much larger. The anthropologist Robin
Dunbar argues that this is a consequence of speech providing
more efficient social bonding than grooming, leading to a
natural group size of 100–150 individuals.
In most cases, group size may be the outcome of
several interacting ecological and social factors. For
example, although lions hunt cooperatively, prides and
hunting groups are usually larger than is optimal for
hunting efficiency. Lionesses cooperate to defend cubs
against infanticidal males by forming crèches. In addition,
hunters are vulnerable to attack by larger groups,
and territories are more effectively defended by larger
prides.
The origin of cooperation among groups of cells and
organisms can also be examined from the perspective of
self organization. The paradox of the evolution of coop
eration is that (by definition) selfish individuals
outcompete altruists, and therefore in a population of
self replicators, a selfish mutant should always spread at
the expense of altruists. Nonetheless, altruism does occur
among humans and cooperative behavior is often seen
among animals. Such cooperative behavior can self orga
nize when the network structure that governs interactions
among individuals results in the same individuals encoun
tering one another repeatedly (e.g., when individuals are
fixed in space, so that their only interactions are with their
neighbors), or when their reproductive fate is very closely
tied to that of others (as is the case for cells within a
multicellular organism). Experimentally, the evolution
of cooperation has been induced in bacterial populations
by production of adhesive, causing individual cells to
clump together. Cooperation can also evolve in marginal
environments, where the evolutionary impact of compe
tition between individuals is outweighed by the need to
survive. Experimental studies of bacteria in marginal
environments show that complex spatial patterns and
signaling behaviors can emerge as a result of this selec
tion. In theoretical models, the inclusion of policing
behavior (punishing nonconformists) can also enforce
high levels of cooperation even when interactions occur
at random in large societies.
104
Self-Organization
Persistence and Stability in Ecosystems
One of the most puzzling topics in systems ecology is how
ecosystems emerge that are at once complex and stable.
Field studies suggest that the most complex (diverse)
ecosystems are also the most stable. However, this obser
vation runs counter to expectation from systems theory. It
shows that the more components a dynamic system has,
the more likely it is that a destabilizing interaction (such
as a positive feedback loop) will cause it to collapse and
lose species. Consequently, systems theory suggests that
simpler ecosystems should be more stable than complex
ones. The paradox implies that the complex, stable eco
systems seen in nature are not random assemblages.
Self organization in this case involves removal of desta
bilizing positive feedback loops.
Communities versus Assemblages
The question of how important self organization is in
ecosystems has long been debated in ecology. Are ecosys
tems communities of co adapted species, or are they
simply random assemblages? Some early theorists, such
as Clements, believed that the groups of species found
together were specialized for living together, whereas
others, such as Gleason, stressed the importance of chance
and individuals.
The idea of succession concerns the patterns and pro
cesses involved in community change, especially after
disturbance. A form of self organization often associated
with succession is facilitation. That is, plants and animals
present in an area can alter the local environment, thereby
facilitating the appearance of populations that replace
them. After a fire, for example, a forest will regenerate
with herbs and shrubs growing back almost immediately.
The first trees to reappear will be ‘pioneer’ (disturbance)
species, which disperse well, grow fast, and can tolerate
open, exposed conditions. These trees create shade and
leaf litter, which favor slow growing, shade tolerant trees.
Recent theoretical work (such as Hubbell’s neutral
theory of biodiversity and biogeography) emphasizes the
role of chance and spatial dynamics in generating ecolo
gical patterns. In these models, self organization is trivial
because all individuals and species are effectively identi
cal, and species abundances are driven by random birth,
migration, and death processes. Both neutral and self
organizing models have been successful in explaining
real relative abundance and species–area curves.
Food Webs
Species interactions lead to the flow of material within an
ecosystem. For animals the most common processes are
eating, respiration, excretion, and egestion. For plants,
they are root uptake of water and nutrients, respiration,
and photosynthesis. The outputs of material from one
organism often become inputs to other. This focus on
‘what eats what’ led Elton to identify several patterns,
notably the food chain and the food web, the food cycle,
the ecological niche, and the pyramid of numbers.
Self organization in ecosystems is evident in the struc
ture in food webs, networks that describe trophic
interactions among species. Within food webs, specific
patterns of interaction may be prevalent. These patterns,
termed ecological motifs, are thought to represent espe
cially stable interactions. The concept of keystone species
supposes that certain species play a crucial role in main
taining the integrity and stability of an ecosystem.
Analysis of food webs suggests that a small world
structure is common. That is, most species interact with
only a small number of other species, but the connectivity
of the web as a whole is maintained by a few species that
interact with a large number of others. This observation
provides a theoretical basis for the idea of keystone spe
cies. Functionally, small world networks are thought to be
robust to random loss of nodes (e.g., species), but vulner
able to attacks that target their highly connected nodes
(e.g., keystone species).
Spatial Patterns and Processes
Spatial processes lead to the formation of distribution
patterns. Seed dispersal, for instance, often produces con
centrations of seedlings around parent plants and leads to
the formation of clumped distributions. When local dis
persal is combined with patchy disturbance, such as fire,
the result is a distribution composed of patches. When
combined with environmental gradients, such as soil
moisture, local dispersal can produce zone patterns, with
different species dominating different areas (Figure 4).
(a)
(b)
(c)
Figure 4 Emergence of spatial patterns from dispersal. This
CA model shows the hypothetical distributions of two plant
populations that result in three different scenarios. (a) Global
dispersal, in which seeds can spread anywhere, results in
random distributions of plants. (b) Dispersal from local seed
sources leads to clumped distributions. (c) The combination of
local dispersal and environmental gradients (from top to bottom)
creates vegetation zones.
Self-Organization
Fragmentation is one of the most important conse
quences of landscape connectivity. When the density of
(randomly located) objects in a landscape falls below a
critical density, they are mostly isolated individuals.
When the density exceeds the critical threshold, they
become connected. The density at which the critical
threshold occurs depends on the size of the neighborhood
of the objects. There are many cases where landscape
connectivity plays an important role. Epidemic processes
require a critical density of resources to spread. Instances
include disease outbreaks (susceptible individuals), fire
spread (fuel), and invasions of exotic plants (suitable
sites). Populations become fragmented if individuals can
not interact with one another. For instance, in wet years the
water bodies of central Australia are essentially connected
for water birds, which can fly from one body to another
almost anywhere in the continent. In dry years, however,
many water bodies shrink or dry up and become too widely
separated for birds to migrate between them (Figure 5).
Self-Organization in the Biosphere
Arguably the most ambitious ecological theory based on
self organization is the Gaia hypothesis, which postulates
that the biosphere itself evolves to a homeostatic state.
Lovelock suggested the Daisyworld model as an illustra
tion of how this process might occur. On the hypothetical
Daisyworld, black and white daisies compete for space.
Although both kinds of daisies grow best at the same
temperature, black daisies absorb more heat than white
daisies. When the Sun shines more brightly, heating the
planet, white daisies spread, and the planet cools again.
When the Sun dims, the black daisies spread, warming the
planet. In this way, competitive interactions between daisies
provide a homeostatic mechanism for the planet as a whole.
The idea behind Gaia is that ecosystems will survive
and spread more effectively if they promote the abiotic
conditions required for their own persistence. If so, eco
systems might gradually evolve to be increasingly robust,
Subcritical
Critical
Supercritical
Figure 5 Critical phase changes in connectivity within a
fragmented landscape. In this CA model, grid cells represent
sites in a landscape. Gray and black cells represent vegetation
and white cells have no cover. The black cells show examples of
patches of vegetation sites that are connected, for example, by
spread of a fire ignited in the center of the grid. Notice that only a
small increase in the density of covered sites makes the
difference between subcritical and supercritical.
105
and if this happened on a global scale, then the biosphere
itself might behave as a self regulating system. However,
evidence for Gaian processes in real ecosystems remains
tenuous and their theoretical plausibility is disputed.
Evolution
Self organization may play a prominent role in evolution,
especially in the context of landscapes, which regulate
interactions between individuals. One consequence is the
evolution of cooperation in marginal and viscous habitat
networks, whereas randomly interacting populations are
more dominated by intraspecific competition and there
fore more likely to behave selfishly.
Landscape structure influences genetic diversity and
speciation. In connected landscapes, genes flow freely
throughout a species and speciation is inhibited.
However, in fragmented landscapes, a species breaks
into isolated subpopulations. Fragmentation increases the
risk of inbreeding and loss of genetic diversity in these
subpopulations. Divergence between population fragments
may also underlie adaptive radiations, in which many
novel species suddenly emerge simultaneously.
As species adapt to their environment, they are often
faced by tradeoffs in allocating resources for different
purposes. These tradeoffs can lead to the evolution of
distinct morphs within a species, or to speciation. For
example, many mangrove species face a conflict between
salt tolerance and competitive ability. Mangroves grow in
estuaries, where salinity varies along the gradient
between land and sea. Mangroves growing landward will
be under strong selection for competitive ability, while
those growing closer to the sea require better salt toler
ance. The tradeoff, combined with local seed dispersal,
can generate discrete banding patterns in the distribution
of mangrove species, where each species is displaced by a
more salt tolerant one closer to the sea.
Contingency also plays a large part in the organization
of spatial distributions. Spatial dominance occurs when a
particular species is overwhelmingly abundant in a local
environment. In this situation, the species can resist inva
sion, even by a superior competitor, because its
propagules are much more numerous locally than those
of any other population. For the same reason, a mutation
that enables a species to exploit a novel environment may
result in it permanently excluding potential competitors
from that environment, even after they have evolved
similar adaptations.
Practical Considerations
The insights provided by theories of self organization
have many practical implications, both for ecology and
for conservation. The sharp end of the conservation debate
106
Ecological Complexity
often hinges on the question of which areas and which sites
to conserve. If ecosystems consist of random collections of
species, then one site in a landscape is as good as another.
All that matters is to preserve representative populations
of each species. However, if the ecosystems consist of self
organized communities, in which the species are adapted
to depend on one another for survival, then whole com
munities need to be conserved.
Closely related to the above issue is that the tendency
for randomly constructed food webs to be unstable raises
questions about the long term viability of artificially cre
ated communities in which translocated species are
introduced into new areas. Self organization is evident
even in artificial ecosystems. In biosphere 2, for instance,
a closed, experimental environment designed to emulate
natural ecosystems, the environment was found to favor
species that collect more energy and internal processes
led to unexpected problems, such as runaway depletion of
oxygen levels.
The need to understand self organization is impor
tant when considering altered ecosystems. For
instance, it is usually not possible to carry out experi
ments to determine the long term effects of current
ecological management practices such as translocation
of populations, controlled burning or allocation of
reserves and wilderness areas. This problem makes
simulation modeling a potentially crucial tool of eco
logical theory and practice. New methods of field
observation are also appearing. For instance, the
need to understand landscape fragmentation has led
to studies of connectivity in landscapes, both field
based, and using data from remote sensing and
geographic information.
See also: Autocatalysis; Ecological Complexity; Emergent
Properties; Hierarchy Theory in Ecology.
Further Reading
Ball P (1999) The Self Made Tapestry: Pattern Formation in Nature.
Oxford: Oxford University Press.
Camazine S, Deneubourg J L, Franks NR, et al. (2003) Self Organization
in Biological Systems. Princeton: Princeton University Press.
Green DG, Klomp NI, Rimmington GR, and Sadedin S (2006)
Complexity in Landscape Ecology. Amsterdam: Springer.
Holland JH (1996) Hidden Order: How Adaptation Builds Complexity.
New York: Addison Wesley.
Levin SA (1998) Ecosystems and the biosphere as complex adaptive
systems. Ecosystems 1(5): 431 436.
Patten BC, Fath BD, and Choi JS (2002) Complex adaptive hierarchical
systems Background. In: Costanza R and Jørgensen SE (eds.)
Understanding and Solving Environmental Problems in the 21st
Century, pp. 41 94. London: Elsevier.
Prigogine I (1980) From Being to Becoming. New York: Freeman (ISBN
0 7167 1107 9).
Rohani P, Lewis TJ, Gruenbaum D, and Ruxton GD (1997) Spatial
self organization in ecology: Pretty patterns or robust reality? Trends
in Ecology and Evolution 12(8): 70 74.
Solé RV and Levin S (2002) Preface to special issue: The biosphere as a
complex adaptive system. Philosophical Transactions of the Royal
Society of London B 357: 617 618.
Watts DJ and Strogatz SH (1998) Collective dynamics of ‘small world’
networks. Nature 393(6684): 440 442.
Ecological Complexity
J L Casti, International Institute for Applied System Analysis, Laxenburg, Austria
B D Fath, Towson University, Towson, MD, USA and International Institute for Applied System Analysis,
Laxenburg, Austria
ª 2008 Elsevier B.V. All rights reserved.
Complexity as a Systems Concept
Surprise-Generating Mechanisms
Emergent Phenomena
Ecological Complexity
‘Would-Be’ Worlds
Conclusions
Further Reading
Complexity as a Systems Concept
are difficult to understand. The behavior of national
economies, the human brain, and a rain forest ecosystem
are all good illustrations for complex systems.
These examples show that there is nothing new about
complex systems. But what is new is that for perhaps the
In everyday parlance, the term ‘complex’ is generally
taken to mean a person or thing composed of many
interacting components whose behavior and/or structure
Ecological Complexity
first time in history, we have the knowledge – and the
tools – to study such systems in a controlled, repeatable,
scientific fashion. So there is reason to believe that this
newfound capability will eventually lead to a viable the
ory of such systems.
Prior to the recent arrival of cheap and powerful
computing capabilities, we were hampered in our ability
to study a complex system like a road traffic network, a
national economy, or a supermarket chain because it was
simply too expensive, impractical, too time consuming –
or too dangerous – to tinker with the system as a whole.
Instead, we were limited to biting off bits and pieces of
such processes that could be looked at in a laboratory or in
some other controlled setting. But with today’s computers
we can actually build complete silicon surrogates of these
systems, and use these ‘would be worlds’ as laboratories
within which to look at the workings – and behaviors – of
the complex systems of everyday life.
In coming to terms with complexity as a systems con
cept, we first have to realize that complexity is an
inherently subjective concept; what is complex depends
upon how you look. When we speak of something being
complex, what we are really doing is making use of every
day language to express a feeling or impression that we
dignify with the label ‘complex’. But the meaning of
something depends not only on the language in which it
is expressed (i.e., the code), the medium of transmission,
and the message, but also on the context. In short, mean
ing is bound up with the whole process of communication
and does not reside in just one or another aspect of it. As a
result, the complexity of a political structure, an ecosys
tem, or an immune system cannot be regarded as simply a
property of that system taken in isolation. Rather, what
ever complexity such systems have is a joint property of
the system and its interaction with another system, most
often an observer and/or controller.
So just as with truth, beauty, good, and evil, complex
ity resides as much in the eye of the beholder as it does in
the structure and behavior of a system itself. This is not to
say that there do not exist ‘objective’ ways to characterize
some aspects of a system’s complexity. After all, an amoeba
is just plain simpler than an elephant by whatever notion
of complexity you happen to believe in. The main point,
though, is that these objective measures only arise as
special cases of the two way measures, cases in which
the interaction between the system and the observer is
much weaker in one direction than in the other.
A second key point is that common usage of the term
‘complex’ is informal. The word is typically employed as
a name for something that seems counterintuitive, unpre
dictable, or just plain hard to understand. So if it is a
genuine ‘science’ of complex systems we are after and
not just anecdotal accounts based on vague personal opin
ions, we are going to have to translate some of these
informal notions about the complex and the
107
commonplace into a more formal, stylized language, one
in which intuition and meaning can be more or less
faithfully captured in symbols and syntax. The problem
is that an integral part of transforming complexity (or
anything else) into a science involves making that which
is fuzzy precise, not the other way around, an exercise we
might more compactly express as ‘formalizing the
informal’.
To bring home this point a bit more forcefully, let us
consider some of the properties associated with ‘simple’
systems by way of inching our way to a feeling for what is
involved with the complex. Generally speaking, simple
systems exhibit the following characteristics.
Predictable behavior. There are no surprises in simple
systems; simple systems give rise to behaviors that are
easy to deduce if we know the inputs (decisions) acting
upon the system and the environment. If we drop a stone,
it falls; if we stretch a spring and let it go, it oscillates in a
fixed pattern; if we put money into a fixed interest bank
account, it grows to a predictable sum in accordance with
an easily understood and computable rule. Such predict
able and intuitively well understood behavior is one of the
principal characteristics of simple systems.
Complex processes, on the other hand, generate coun
terintuitive, seemingly acausal behavior that is full of
surprises. Lower taxes and interest rates lead to higher
unemployment; low cost housing projects give rise to
slums worse than those the ‘better’ housing replaced; the
construction of new freeways results in unprecedented
traffic jams and increased commuting times. For many
people, such unpredictable, seemingly capricious, behav
ior is the defining feature of a complex system.
Few interactions and feedback/feedforward loops. Simple
systems generally involve a small number of components,
with self interactions dominating the linkages among the
variables. For example, primitive barter economies, in
which only a small number of goods (food, tools, weapons,
clothing) are traded, seem much simpler and easier to
understand than the developed economies of industria
lized nations, in which the pathways between raw
material inputs and finished consumer goods follow
labyrinthine routes involving large numbers of interac
tions between various intermediate products, labor, and
capital inputs.
In addition to having only a few variables, simple
systems generally consist of very few feedback/feedfor
ward loops. Loops of this sort enable the system to
restructure, or at least modify, the interaction pattern
among its variables, thereby opening up the possibility
for a wider range of behaviors. To illustrate, consider a
large organization that is characterized by variables like
employment stability, substitution of capital for human
labor, and level of individual action and responsibility
(individuality). Increased substitution of work by capital
decreases the individuality in the organization, which in
108
Ecological Complexity
turn may reduce employment stability. Such a feedback
loop exacerbates any internal stresses initially present in
the system, leading possibly to a collapse of the entire
organization. This type of collapsing loop is especially
dangerous for social structures, as it threatens their ability
to absorb shocks, which seems to be a common feature of
complex social phenomena.
Centralized decision making. In simple systems, power is
generally concentrated in one or at the most a few deci
sion makers. Political dictatorships, privately owned
corporations, and the Roman Catholic Church are good
examples of this sort of system. These systems are simple
because there is very little interaction, if any, between the
lines of command. Moreover, the effect of the central
authority’s decision upon the system is usually rather
easy to trace.
By way of contrast, complex systems exhibit a diffu
sion of real authority. Such systems seem to have a
nominal supreme decision maker, but in actuality the
power is spread over a decentralized structure. The
actions of a number of units then combine to generate
the actual system behavior. Typical examples of these
kinds of systems include democratic governments, labor
unions, and universities. Such systems tend to be some
what more resilient and stable than centralized structures
because they are more forgiving of mistakes by any one
decision maker and are more able to absorb unexpected
environmental fluctuations.
Decomposable. Typically, a simple system involves weak
interactions among its various components. So if we sever
some of these connections, the system behaves more or
less as before. Relocating American Indians to reserva
tions produced no major effects on the dominant social
structure in New Mexico and Arizona, for example, since,
for various cultural reasons, the Indians were only weakly
coupled to the dominant local social fabric in the first
place. Thus, the simple social interaction pattern present
could be further decomposed and studied as two inde
pendent processes – the Indians and the settlers.
Complex processes, on the other hand, are irreducible.
Neglecting any part of the process or severing any of the
connections linking its parts usually destroys essential
aspects of the system’s behavior or structure. You just
cannot start slicing up systems of this complexity into
subsystems without suffering an irretrievable loss of the
very information that makes these systems a ‘system’.
complexity as ‘surprise generating mechanisms’, whose
quite different natures each lead to their own character
istic type of surprise. Let us take a quick look at each of
these mechanisms before turning to a more detailed con
sideration of how they act to create complex behavior.
Paradox. Paradoxes arise from false assumptions about
a system leading to inconsistencies between its observed
behavior and our ‘expectations’ of that behavior.
Sometimes these situations occur in simple logical or
linguistic situations, such as the famous ‘liar paradox’
(‘‘This sentence is false.’’). In other situations, the paradox
comes from the peculiarities of the human visual system,
as with the impossible staircase shown in Figure 1, or
simply from the way in which the parts of a system are put
together, like the developing economy discussed in the
preceding section.
Instability. Everyday intuition has generally been
honed on systems whose behavior is stable with regard
to small disturbances, for the obvious reason that unstable
systems tend not to survive long enough for us to develop
good intuitions about them. Nevertheless, the systems of
both nature and humans often display pathologically sen
sitive behavior to small disturbances, as for example,
when stock markets crash in response to seemingly
minor economic news about interest rates, corporate mer
gers, or bank failures. Such behaviors occur often enough
that they deserve a starring role in our taxonomy of
surprise.
Incomputability. The kinds of behaviors seen in models
of complex systems are the end result of following a set of
rules. This is because these models are embodied in
computer programs, which in turn are necessarily just a
set of rules telling the machine what bits in its memory
array to turn on or off at any given stage of the calculation.
By definition, this means that any behavior seen in such
worlds is the outcome of following the rules encoded in
the program. Although computing machines are de facto
rule following devices, there is no a priori reason to
believe that any of the processes of nature and humans
are necessarily rule based. If incomputable processes
do exist in nature – for example, the breaking of waves
on a beach or the movement of air masses in the atmo
sphere – then we could never see these processes manifest
themselves in the surrogate worlds of their models. We
(a)
(b)
3
1
Surprise-Generating Mechanisms
The vast majority of counterintuitive behaviors shown by
complex systems are attributable to some combination of
the following five sources: paradoxes/self reference,
instability, incomputability, connectivity, and emergence.
With some justification, we can think of these sources of
2
3
2
1
Figure 1 The impossible staircase.
Ecological Complexity
may well see processes that are close approximations to
these incomputable ones, just as we can approximate an
irrational number as closely as we wish by a rational
number. However, we will never see the real thing in
our computers, if indeed such incomputable quantities
exist outside the pristine world of mathematics.
Connectivity. What makes a system a system and not
simply a collection of elements is the connections and
interactions among the individual components of the
system, as well as the effect these linkages have on the
behavior of the components. For example, it is the inter
relationship between biota and abiota that makes an
ecosystem. Each component taken separately would not
suffice. The two must interact for sustainable life to take
place. Complexity and surprise often reside in these
connections.
Emergence. A surprise generating mechanism depen
dent on connectivity for its very existence is the
phenomenon of emergence. This refers to the way the
interactions among system components generate unex
pected global system properties not present in any of
the subsystems taken individually. A good example is
water, whose distinguishing characteristics are its natural
form as a liquid and its nonflammability, both of which are
totally different from the properties of its component
gases, hydrogen and oxygen.
The difference between complexity arising from
emergence and that coming only from connection pat
terns lies in the nature of the interactions among the
various component pieces of the system. For emergence,
attention is not simply on whether there is some kind of
interaction between the components, but also on the
specific nature of that interaction. For instance, connec
tivity alone would not enable one to distinguish between
ordinary tap water involving an interaction between
hydrogen and oxygen molecules and heavy water (deu
terium), which involves interaction between the same
components albeit with an extra neutron thrown in to
the mix. Emergence would make this distinction. In prac
tice it is often difficult (and unnecessary) to differentiate
between connectivity and emergence, and they are fre
quently treated as synonymous surprise generating
procedures. A good example of emergence in action is
the organizational structure of an ant colony.
Like human societies, ant colonies achieve things that no
individual ant could accomplish on its own. Nests are
erected and maintained, chambers and tunnels are exca
vated, and territories are defended. All these activities are
carried on by individual ants acting in accord with simple,
local information; there is no master ant overseeing the
entire colony and broadcasting instructions to the individual
workers. Somehow each individual ant processes the partial
information available to it in order to decide which of the
many possible functional roles it should play in the colony.
109
Recent work on harvester ants has shed considerable
light on the process by which an ant colony assesses its
current needs and assigns a certain number of members to
perform a given task. These studies identify four distinct
tasks an adult harvester ant worker can perform outside
the nest: foraging, patrolling, nest maintenance, and mid
den work (building and sorting the colony’s refuse pile).
So it is these different tasks that define the components of
the system we call an ant colony, and it is the interaction
among ants performing these tasks that gives rise to
emergent phenomena in the colony.
One of the most notable interactions is between
forager ants and maintenance workers. When nest main
tenance work is increased by piling some toothpicks near
the opening of the nest, the number of foragers decreased.
Apparently, under these environmental conditions, the
ants engaged in task switching, with the local decision
made by each individual ant determining much of the
coordinated behavior of the entire colony. Task allocation
depends on two kinds of decisions made by individual
ants. First, there is the decision about which task to per
form, followed by the decision of whether to be active in
this task. As already noted, these decisions are based
solely on local information; there is no central decision
maker keeping track of the big picture.
Figure 2 gives a summary of the task switching roles
in the harvester ant colony, showing that once an ant
becomes a forager it never switches back to other tasks
outside the nest. When a large cleaning chore arises on
the surface of the nest, new nest maintenance workers are
recruited from ants working inside the nest, not from
workers performing tasks on the outside. When there is
a disturbance like an intrusion by foreign ants, nest main
tenance workers will switch tasks to become patrollers.
Finally, once an ant is allocated a task outside the nest, it
never returns to chores on the inside.
The ant colony example shows how interactions
among the various types of ants can give rise to patterns
of global work allocation in the colony, patterns that
could not be predicted or that could not even arise in
any single ant. These patterns are emergent phenomena
due solely to the types of interactions among the different
tasks.
Table 1 gives a summary of the surprise generating
mechanisms just outlined.
Emergent Phenomena
Complex systems produce surprising behavior; in fact,
they produce behavioral patterns and properties that just
cannot be predicted from knowledge of their parts taken
in isolation. These so called ‘emergent properties’ are
probably the single most distinguishing feature of com
plex systems. An example of this phenomenon occurs
110
Ecological Complexity
Environmental change
Task switching
Midden worker
Forager
Patrolling ant
Forager
Nest maintenance worker
Forager
Increase in food availability
Brood-care worker,
nest construction, seed
storage, reserves
Nest maintenance worker
Detritus on surface of nest mound
Nest maintenance worker
Patrolling
ant
Intrusion by foreign ants
Figure 2 Task switching in a harvester ant colony.
Table 1 The main surprise-generating mechanisms
Mechanism
Surprise effect
Paradoxes
Instability
Incomputability
Connectivity
Emergence
Inconsistent phenomena
Large effects from small changes
Behavior transcends rules
Behavior cannot be decomposed into parts
Self-organizing patterns
when one considers a collection of independent random
quantities, such as the heights of all the people in New
York City. Even though the individual numbers in this set
are highly variable, the distribution of this set of numbers
will form the familiar bell shaped curve of elementary
statistics. This characteristic bell shaped structure can be
thought of as ‘emerging’ from the interaction of the com
ponent elements. Not a single one of the individual
heights can correspond to the normal probability distri
bution, since such a distribution implies a population. Yet
when they are all put into interaction by adding and
forming their average, the ‘central limit theorem’ of prob
ability theory tells us that this average and the dispersion
around it must obey the bell shaped distribution.
Ecological Complexity
Complexity research has discovered that many systems
display common structural and behavioral/dynamical
characteristics (Table 2). The interplay of these complex
system characteristics entails systems to exhibit proper
ties such as surprise, emergence, and power law scaling.
Ecological complexity is the observation that ecological
systems exhibit many of the same properties as physical
complex systems, and thus an active research program has
arisen over the analysis of ecological data to see to what
extent ecosystems share these common properties with
other complex systems.
Ecosystems are composed of a large number of highly
diverse components interacting with self stabilizing and
self promoting feedback to produce emergent patterns. As
Ecological Complexity
111
Table 2 Some characteristics of complex systems
Structural characteristics
Behavioral/dynamical characteristics
Large number of components
Large number of components
High diversity of components and connections
Asymmetries
Strong interactions
Hierarchic organization
Nonlinear
Chaotic
Catastrophic
Self-organization
Multiple steady states
Adaptive
such, ecological systems have been described as complex,
adaptive, hierarchical systems (CAHS) or self organized,
hierarchical open (SOHO) systems. Unlike with complex
physical systems, openness is a property that is required of
all ecological systems. This is because ecological systems
are self perpetuating through means of capturing energy,
doing useful work (biochemical reactions, growth, and
maintenance) to persist at least momentarily at a highly
organized state far from thermodynamic equilibrium – this
is the metabolic process. A second ecological defining fea
ture is that organisms are able to replicate themselves such
that the system outlives the constituent parts – this is the
reproductive process. Therefore, it is often said of an
ecosystem that, ‘the whole is more than the sum of the
parts’.
Two of the most pressing issues regarding ecological
complexity are the need to develop appropriate measures
to quantify the structural and behavioral complexity of
ecosystems, and to identify the underlying processes that
generate this complexity, through theory, analysis, mod
eling, and field studies.
A new journal, Ecological Complexity (Elsevier), is one
forum for this research since 2004. The journal considers
papers dealing with biocomplexity related to the envi
ronment with an emphasis on interdisciplinary and
integrated natural and social systems science.
Topics typically found in the journal include:
aspects of biocomplexity in the environment and
• alltheoretical
ecology;
ecosystems
• systems; and biospheres as complex adaptive
organization of spatially extended ecosystems;
• self
properties and structures of complex
• emergent
ecosystems;
pattern formation in space and time;
• ecological
role of biophysical constraints and evolutionary
• the
attractors on species assemblages;
scaling (scale invariance, scale covariance,
• ecological
and across scale dynamics), allometry, and hierarchy
theory;
topology and networks;
• ecological
toward an ecology of complex systems;
• studies
systems approaches for the study of dynamic
• complex
human–environment interactions;
knowledge of nonlinear phenomena to better
• using
guide policy development for adaptation strategies
•
and mitigation to environmental change; and
new tools and methods for studying ecological
complexity.
The emphasis on integrated natural and social systems
addresses the growing interest to understand the role of
human influence on the environment. There is a recent
awareness of the need to alter this influence in some fashion
both for ourselves and our environment. New tools and
approaches incorporating self organization, emergence,
and co adaptation are needed to improve our ability to
manage and restore natural systems. These new approaches
to ecosystem management must also account for the natural
dynamics and integrate concepts of sustainable develop
ment. Advances in ecological complexity science are
essential in successfully navigating this transition.
‘Would-Be’ Worlds
In the past few years, a number of electronic worlds have
been created by researchers associated with the Santa Fe
Institute and elsewhere to study the properties of com
plex, adaptive systems. The authors cite just three such
worlds here as prototypical examples of how to use the
computer as a kind of information laboratory to investi
gate such systems.
Tierra. This world, created by naturalist Tom Ray, is
populated by binary strings that serve as electronic surro
gates for genetic material. As time unfolds, these strings
compete with each other for resources, with which they
create copies of themselves. New strings are also created
by computational counterparts of the real world pro
cesses of mutation and crossover. Over the course of
time, the world of Tierra displays many of the features
associated with evolutionary processes seen in the natural
world, and hence can be used as a way of experimenting
with such processes – without having to wait millions of
years to bring the experiment to a conclusion. But it is
important to keep in mind that Tierra is not designed
to mimic any particular real world biological process;
rather, it is a laboratory within which to study neo
Darwinian evolution, in general.
112
Ecological Complexity
TRANSIMS. For the past 3 years, a team of researchers
at the Los Alamos National Laboratory headed by Chris
Barrett has built an electronic counterpart of the city of
Albuquerque, New Mexico, inside their computers. The
purpose of this world, which is called TRANSIMS, is to
provide a testbed for studying the flow of road traffic in an
urban area of nearly half a million people. In contrast to
Tierra, TRANSIMS is explicitly designed to mirror the
real world of Albuquerque as faithfully as possible, or at
least to mirror those aspects of the city that are relevant
for road traffic flow. Thus, the simulation contains the
entire road traffic network from freeways to back alleys,
together with information about where people live and
work, as well as demographic information about incomes,
children, type of cars, and so forth. So here we have a
would be world whose goal is to indeed duplicate as
closely as possible a specific real world situation.
Sugarscape. Somewhere between Tierra and
TRANSIMS is the would be world called Sugarscape,
which was created by Joshua Epstein and Rob Axtell of
The Brookings Institution in Washington, DC. This
world is designed as a tool to study processes of cultural
and economic evolution. On the one hand, the assump
tions about how individuals behave and the spectrum of
possible actions at their disposal is a vast simplification of
the possibilities open to real people as they go through
everyday life. On the other hand, Sugarscape makes fairly
realistic assumptions about the things that motivate peo
ple to act in the way they do, as well as about how they go
about trying to attain their goals. What is of considerable
interest is the rich variety of behaviors that emerge from
simple rules for individual action, and the uncanny
resemblance these emergent behaviors have to what is
actually seen in real life.
In order to conduct the kinds of repeatable, controlled
experiments that natural scientists take for granted when
trying to understand and create theories of physical and
engineering systems, Epstein and Axtell decided to ‘grow’
a social order from scratch by creating an ever changing
environment and a set of agents who interact with each
other and the environment in accordance with simple
rules of survival. An entire social structure – trade, economy,
culture – then evolves from the interactions of the agents. As
Epstein remarks about social problems, ‘‘You don’t solve it,
you evolve it.’’ Epstein and Axtell call their laboratory in
which societies evolve the ‘CompuTerrarium’. Here is how
it works.
The interacting agents are each graphically repre
sented by a single colored dot on the landscape they
inhabit, which is called the Sugarscape. Every location
in the landscape contains time varying concentrations of
a food resource, called sugar. Each individual has a unique
set of characteristics; some are fixed like sex, visual range
for food detection, and metabolic rate, whereas others are
variable like health, marital status, and wealth. The
behavior of these agents is determined by a set of extre
mely simple rules that constitute nothing more than
common sense rules for survival and reproduction. A typi
cal set of rules might be:
1. Find the nearest location containing sugar. Go there,
eat as much as necessary to maintain your metabolism,
and save the rest.
2. Breed if you have accumulated enough energy and
other resources.
3. Maintain your current cultural identity (set of charac
teristics) unless you see that you are surrounded by
many agents of different types (‘tribes’). If you are,
change your characteristics and/or preferences to fit
in with your neighbors.
With even such primitive rules as this, strange and won
drous things begin to happen. A typical scenario is shown
in Figure 3, where we see the sugar marked by yellow
dots on the Sugarscape. The agents are initially distrib
uted randomly on the landscape, red dots being agents
that have a good ability to see food at a distance, blue dots
representing more myopic agents. It is reasonable to
expect that if no other considerations enter, natural selec
tion would tend to favor good vision over time. Indeed
this is the case, as seen by the center panel in Figure 3,
showing a preponderance of red agents in the population.
However, if the experiment is run again, giving agents the
possibility of passing wealth on to their offspring in the
form of sugar, we find that inheritance has a pronounced
effect on survival. This is shown in the third panel of the
figure, in which many more agents having poor vision are
able to survive by making use of sugar willed to them by
their parents.
Although this simple example is useful in illustrating
the workings of the CompuTerrarium, it hardly suggests
a revolution in our way of thinking about and studying
social structures. For that we need to add a lot more
whistles and bells to the system. Epstein and Axtell have
done exactly this. When they add seasons so that sugar
concentrations change periodically over time, the agents
begin to migrate. When a second resource, spice, is intro
duced, a primitive economy emerges as a result of a new
elementary rule: ‘‘Look around for a neighbor having a
commodity you need. Bargain with that neighbor until
you reach a mutually satisfactory price. Trade at that
price.’’ Figure 4 shows the effect of this type of trading
Initial condition
No inheritance
Figure 3 Evolution on the Sugarscape.
With inheritance
Ecological Complexity
1
Agents forage for ‘sugar and spice’
2
If trade is allowed, they flourish
113
3 Without trade, many starve
Figure 4 The effects of trade in the Sugarscape.
economy. In the first part of the figure, agents are simply
foraging independently for both sugar and spice. In the
middle panel we see the effect of beginning trade; now
lots of agents flourish. Finally, the third panel shows the
effect of turning off the trade. Without trade being
allowed, many of the agents cannot survive.
There is certainly much more that can be said about
the social laboratory constructed by Epstein and Axtell.
Issues involving the emergence of cultural groups, com
bat, institutional structures, and the like can all be
introduced to study myriad questions of interest to
social scientists. The interested reader will certainly
want to consult the monograph by Epstein and Axtell
that details these and many other matters. It is cited in
the references for this article. Here we must content
ourselves with simply noting that the CompuTerrarium
offers a platform to study society from the bottom up.
With this view, we can explore social behavior that is
dynamic, evolutionary, and locally simple. What could
be better than to have a laboratory like this in which to
do such experiments?
The main point in presenting these discussions of
Tierra, TRANSIMS, and Sugarscape is to emphasize
two points: (1) we need different types of would be
worlds to study different sorts of questions, and (2) each
of these worlds has the capability of serving as a labora
tory within which to test hypotheses about the
phenomena they can represent. And, of course, it is this
latter property that encourages the view that such com
putational universes will play the same role for the
creation of theories of complex systems that chemistry
labs and particle accelerators have played in the creation
of scientific theories of simple systems. Gleick has given a
fuller account of the technical, philosophical, and theore
tical problems surrounding the construction and use of
these silicon worlds.
Conclusions
The key components in each and every complex, adap
tive system and a decent mathematical formalism to
describe and analyze them would go a long way toward
the creation of a viable theory of such processes. These
key components are given as follows.
A medium sized number of agents. In contrast to simple
systems – like superpower conflicts, which tend to involve
a small number of interacting agents – or large systems –
like galaxies or containers of gas, which have a large
enough collection of agents that we can use statistical
means to study them – complex systems involve what
we might call a medium sized number of agents. Just like
Goldilocks’s porridge, which was not too hot and not too
cold, complex systems have a number of agents that are
not too small and not too big, but just right to create
interesting patterns of behavior.
Intelligent and adaptive agents. Not only are there a
medium sized number of agents, these agents are intelli
gent and adaptive. This means that they make decisions
on the basis of rules, and that they are ready to modify the
rules they use on the basis of new information that
becomes available. Moreover, the agents are able to gen
erate new rules that have never before been used, rather
than being hemmed in by having to choose from a set of
preselected rules for action. This means that an ecology of
rules emerges, one that continues to evolve during the
course of the process.
Local information. In the real world of complex sys
tems, no agent knows what ‘all’ the other agents are
doing. At most, each person gets information from a
relatively small subset of the set of all agents, and
processes this ‘local’ information to come to a decision
as to how he or she will act. In the Sugarscape, for
instance, what the traders adjacent to a given individual
in the market are doing constitutes the local informa
tion that the individual has available to help decide
what to do next.
So these are the components of all complex, adap
tive systems like the Sugarscape, TRANSIMS, or
Tierra situations – a medium sized number of intelli
gent, adaptive agents interacting on the basis of local
information. At present, there appears to be no known
mathematical structures within which we can comfor
tably accommodate a description of ‘any’ of these
worlds. This suggests a situation completely analogous
to that faced by gamblers in the seventeenth century,
114
Hierarchy Theory in Ecology
who sought a rational way to divide the stakes in a
game of dice when the game had to be terminated
prematurely (probably by the appearance of the police
or, perhaps, the gamblers’ wives). The description and
analysis of that very definite real world problem led
Fermat and Pascal to the creation of a mathematical
formalism we now call probability theory. At present,
complex system theory still awaits its Pascal and
Fermat. The mathematical concepts and methods cur
rently available were developed, by and large, to
describe systems composed of material objects like pla
nets and atoms. It is the development of a proper
theory of complex systems that will be the capstone
of the transition from the material to the informational.
See also: Emergent Properties; Hierarchy Theory in
Ecology; Self-Organization.
Further Reading
Casti J (1992) Reality Rules: Picturing the World in Mathematics. I The
Fundamentals, II The Frontier. New York: Wiley (paperback edition,
1997).
Casti J (1994) Complexification. New York: HarperCollins.
Casti J (1997) Would Be Worlds. New York: Wiley.
Epstein J and Axtell R (1996) Growing Artificial Societies. Cambridge,
MA: MIT Press.
Gleick J (1987) Chaos. New York: Viking.
Jackson E (1990) Perspectives of Nonlinear Dynamics. Cambridge:
Cambridge University Press, vols. 1 and 2.
Mandelbrot B (1982) The Fractal Geometry of Nature. San Francisco,
CA: W.H. Freeman.
Nicolis J (1991) Chaos and Information Processing. Singapore: World
Scientific.
Peitgen H O, Jurgens DH, and Saupe D (1992) Fractals for the
Classroom, Parts 1 and 2. New York: Springer.
Ray T (1991) An approach to the synthesis of life. In: Langton CG,
Taylor C, Farmer JD, and Rasmussen S (eds.) Artificial Life II, SFI
Studies in the Science of Complexity, vol. X, pp. 371 408. Reading,
MA: Addison Wesley.
Schroeder M (1991) Fractals, Chaos, Power Laws. New York: W.H.
Freeman.
Stewart I (1989) Does God Play Dice? Oxford: Basil Blackwell.
Hierarchy Theory in Ecology
T F H Allen, University of Wisconsin, Madison, WI, USA
ª 2008 Elsevier B.V. All rights reserved.
Need for Hierarchy Theory
Hierarchy and Hypothesis
Hierarchical Levels
History of the Field
Scale and Type
Further Reading
Need for Hierarchy Theory
the onset of hostilities when Nichols attacked Gleason’s
paper at the 1926 International Congress of Plant
Sciences. Hierarchy theory’s focus on level of analysis
offers such clarification. It lays subtle distinctions bare,
so that definitions work for the ecologist instead of ecol
ogists working for their definitions.
If big, slow things were always on top, such that hier
archical levels were only a matter of scale, the problem
would reduce to a straightforward technical scaling issue.
Not to underestimate the challenges of scaling in engi
neering, but that technical setting does not need
something as grand as a theory to deal with hierarchies.
But in ecology, scaling is complicated by higher ecologi
cal levels giving lower levels meaning. In ecology, the
move upscale to be more inclusive often changes signifi
cance more than it invokes a change of size, and so we do
indeed need a special body of theory to deal with differ
ence of quality, not just quantity. Differences in scale
quickly become large enough to cause qualitative change
In ecology we need a body of theory to address relation
ships that are consequences of changing levels of analysis,
which call for altered definitions. For instance, the con
temporary fracas over definitions of plant competition
could benefit from recognizing differences in the scope
and type of the framework used by the respective parti
sans. With distinctions between levels of analysis made
clear, each school may test their respective hypotheses in
peace aware of which theories actually compete and when
they merely address some other level of discourse. The
contentious literature surrounding overcompensation of
plants in response to losses to grazing was significantly a
matter of pulse versus press consumption in relation to
different timescales for assessing recovery. The
Clements/Gleason debate over the proper definition of
plant community might not have lasted the body of the
twentieth century had hierarchy theory been available at
Hierarchy Theory in Ecology
in perception, which forces a change in the level of
analysis. Thus scale is soon embroiled in values, judg
ment, and arbitrary choices, not just as an inconvenience,
but as a necessity for proper understanding. While scaling
as an engineering technicality actively ignores such messy
issues, hierarchy theory explicitly includes value based
decisions of the observer in creating hierarchies.
To control for observer values, technical measurement
and analysis in science keeps its criteria constant across the
local discourse. But large discoveries precisely amount to
the recognition of a change in value. New scientific ideas
indicate a specific change in the preanalytic stage, before
deciding what might be relevant data. In the terms of
Russell and Whitehead (made accessible by Gregory
Bateson), new scientific ideas amount to the definition of
new logical types. Hierarchy theory’s central activity is
recognizing logical type. Logical types are tied to some
new level of inclusivity, a new hierarchical level with its
own meaning. Notice how left and right sides are possessed
by organisms at their own level of existence. Meanwhile,
the notions of up and down refer to a larger discourse that
includes an environment, which is shared by many organ
isms. As a result, a mirror switches the image left and right,
but with no switch in up versus down. The larger scope
invoked by the idea of up introduces a new logical type,
even though left and right may often be simply at right
angles to up and down. If left and right contrasted with up
and down can be problematic, ecosystems are a nightmare.
While exquisitely holding criteria constant in formal scien
tific calibration will help, it is insufficient for large
discoveries, which turn on recognizing when a new type
is necessary to solve some puzzle.
Ecology in particular invites many logical types because
its hierarchies are so rich. A new type invokes new aggre
gation criteria, which come explicitly from observer
decisions. Consider the difference between a community
conception of vegetation as opposed to the process
functional conception that prevails in ecosystem modeling.
A forest can be considered as a collection of trees on a tract
of land. Alternatively those same tree trunks may be aggre
gated as a separate class from the leaves (Figure 1). If
leaves in a forest are the production system independent
of species, then the boles are part of the carbon storage
function. This assignment has the peculiar effect of unify
ing the tree trunks with soil carbon in a single carbon
storage compartment. A community focus aggregates
trees set in an environment of soil and atmosphere.
Meanwhile a flux process conception splits the trees into
at least two parts, one of which aggregates with the soil. But
the soil was part of the environment in the community
conception. Thus the same pieces of soil and plant biomass
are aggregated into different higher units, depending on
the type of system that is recognized as being in the fore
ground by the observer. Note how forests under either
conception may be called forest ecosystems, suggesting
that one use of hierarchy theory is to untangle alternative
meanings in commonplace ecological terminology. The
difference between a process focused ecosystem and a
community is a change in logical type.
Hierarchy and Hypothesis
Hierarchy theory is a body of thought that relates chosen
levels of analysis to defined levels of organization, all in
Process
ecosystem
Biota and
environs
Community
Environs
Carbon
storage
Species
Species
Soil
Bole
Air
Tree
bole
Tree
Leaves
115
Soil
carbon
Carbon
capture
Leaves
Roots
Figure 1 A community conception leads to an expected situating of whole trees in an environment. But a process-functional
ecosystem conception of that same forest can lead to a hierarchy where tree boles are separated from the leaves, and then united with
soil elements in a carbon storage compartment. Each respective hierarchy takes its form from the purpose for which it is intended.
116
Hierarchy Theory in Ecology
the context of scaling. It advises the scientist of subtle but
crucial distinctions that follow from observing and mak
ing analytical decisions. A significant part of hierarchy
theory is observation in relation to conceptions of order in
complex systems that would otherwise invite confusion. It
is a metatheory that guides the generation, fine tuning
and testing of other bodies of thought, themselves more
easily recognizable as theories in the conventional sense.
Some theories in ecology are associated with answering
questions taken as given. Such theory is validated
in testing hypotheses. Other ecological theories may
fine tune questions, perhaps clarifying what is meant by
competition, so that worthwhile hypotheses may be gen
erated. By contrast, hierarchy theory applies in the
preanalytical stage, when the questions are being framed
rather than when clarified or answered. In the preanaly
tical stage, the boundaries of things are established, and
structures are assigned to types or classes. With the dis
course laid out unambiguously by hierarchy theory, other
theories may come into play, testing explicit hypotheses
with measurement and models. Thus hierarchy theory
does not have its own hypotheses per se, but rather opens
the way for subsequent testing of specific hypotheses.
Like multivariate description in ecology, hierarchy the
ory focuses on hypothesis generation and clarification.
From a small number of first principles, it highlights
what would be otherwise taken for granted and then
forgotten in the muddle that ensues. Hierarchy theory is
explicit, as it positions the tacit next to the focal. Its
precision is in thought and choosing definitions, more
than action in quantitative experimentation.
because of inequivalent size. Host versus parasite is the
basis of a hierarchy employing levels subtly different from
those in the population/organism distinction. Hierarchy
theory places entities in levels, taking care to be explicit
about the definitions that lead to those levels, and the
criteria that create order and linkage.
Scale versus definition has potential for generating
different sorts of hierarchies. Some hierarchies focus on
size and containment, while others are control hierarchies
where upper level entities simply control lower levels. A
Watt governor may be placed at a higher level in a control
hierarchy, while being smaller than the whole steam
engine it controls. Whether it is a scalar or a control
hierarchy depends on the use for which the hierarchical
conception is intended, something for which the observer
must take responsibility. Time against space plots are
popular in landscape ecology. But such hierarchies can
miss out on the interesting situations where large space
maps onto short time spans, or long time spans map onto
small places (Figure 2). The globe is large enough to be
the context of continental movement over hundreds of
millions of years, but at the same time the rotation of the
globe is also responsible for diurnal phenomena, at the fast
end of ecological happenings. Surfaces arise when narrow
space applies to large differences in time constants (strong
temporal connection within, but weak connections across
surfaces). Ecotones would be a case in point, because
there is rapid exchange and fast process inside the abut
ting ecosystems or communities areas, while the
exchanges across the narrow ecotone may be remarkably
slow. Thus ecotones are spatially small, while
Hierarchical Levels
Entities in a hierarchy are recognized as belonging to
levels. Levels are sets, but the sets become levels because
of robust asymmetry between them in a hierarchy.
Mathematically, the asymmetry between levels makes
hierarchies partially ordered sets. Hierarchy theory is
the set theory that may precede network theory (see
Ecological Network Analysis, Environ Analysis). A level
of analysis assigns entities to levels, and is often explicit
about their relationship to other entities assigned to other
levels. There is a distinction between levels of observation
and levels of organization. Levels of observation are
ordered relative to each other on matters of size and
scale. Meanwhile relationships between levels of organi
zation follow from definitions chosen by observers,
sometimes as a prelude to actual observation. For
instance, organisms are subsumed by populations only
by a definition. Hidden in the definition is a requirement
for equivalence between population members.
Meanwhile, a host and its parasite, while both organisms,
are generally not assigned to the same population, in part
Long
wide
Conventional linear plot
time–space for biota and
climate
Not often plotted
but common and
important
n
tatio
’s ro 00 km
h
t
r
Ea
35 0
Ice age
&
24 h
Biome
rs
rido
Community
r
Co
Canopy
50
Tree
Short
narrow
Leaf
ce
Fen
Fast
ne
oto
Ec
6
0
1
6×
ars
ye
&
ait
lia
km s Str stra
e
r
Au
r
To lates
iso Not often
plotted but
still common
and important
Slow
Figure 2 A common graph appearing in landscape ecology
plots increases in time against space, focusing on the
quasi-linear pattern of larger things seen as behaving over longer
time periods. But such plots ignore potential control systems and
their hierarchies. Local intransigence can control large entities
(Wallace’s realms where Australia’s fauna is isolated from Asia by
the narrow Torres Strait, separating millions of years of
evolution). Barriers and surfaces occur at the lower right, while
communication channels and corridors appear upper left.
Hierarchy Theory in Ecology
representing slow exchanges that might cause the ecotone
to move in a process of gradual encroachment.
Conversely, in a communication channel, small differ
ences in time constants apply along the long connection.
Corridors would be an example here, where there is rapid
movement along the extended length of the corridor. In
ecology, these special places, such as ecotones and corri
dors, are at least as interesting as situations where time
and space widen in concert. Complexity in hierarchies
arises from the challenge of mapping between levels, as
scale and definition entwine.
History of the Field
Hierarchy theory has its roots in economics and business
administration of the 1960s, suggesting that the world
appears nearly decomposable. We can decompose wholes
into parts, but only to a degree, in that parts communicate
and leak onto each other. Complete decomposability
would deny upper level structures’ existence. Completely
decomposed, the parts would not to be able to commu
nicate with each other in making the larger whole. Parts
have strong connections within, but weak connections
between, and those weak connections may be precisely
what links hierarchical levels (Figure 3).
While some practitioners in subsequent studies have
sought real hierarchies in an external world, much of the
early literature of business administration hierarchies is
agnostic about the ultimate reality of hierarchical structure.
In this spirit, hierarchy theory in social organizations
Levels and strength of connection
N+3
N+2
N+1
N
N + 3 High
Weak connection
Weak relative to
N + 2 Middle strong below
N+3
But strong relative
N+2
to weak above
N + 1 Low Strong connection
Figure 3 In nested hierarchies, the bonds that unite members
of the lowest level N are strong, as they make entities of level
N þ 1. The bonds that create level N þ 2 are weaker.
Nevertheless, these weaker bonds appear as the strong bonds
making the entities at level N þ 2 when seen from level N þ 3. If
the two largest units in the figure are molecules, then the entities
at N þ 1 are atoms. Breaking their atomic bonds releases huge
amounts of atomic energy as subatomic particles, N, are freed.
Atomic bonds are stronger and release more energy when
broken, compared to breaking the chemical bonds that make
molecules, the N þ 2 entities.
117
operates largely in the realm of epistemology, as a theory
of observation and analysis. The discourse generally takes
the position that hierarchies appear somewhere between
the material world and human understanding. If there are
complex material systems that are not hierarchic, we might
expect to have great difficulty in observing or understand
ing them. There appear to be points of passage of
information up hierarchies, where details are explicitly
lost. A military command is a favorite hierarchic example
of a human organization. There, details of how an indivi
dual soldier observed local enemy concentrations fall away
as the intelligence passes up the command. Setting the
detail aside allows the top brass to make sweeping deci
sions, without being encumbered by a blizzard of local
happenings. Not only must the general in command let
go of details of grains so as to get a handle on the wide
extent, but so too must the observers of hierarchical struc
ture. To understand what a general is doing, the observer
of a command structure needs to integrate away the details
inside the army. By the late 1960s, the notion of hierarchy
had moved beyond administration systems, and was being
taken up across a range of disciplines.
In the following decade, hierarchy theorists from phys
ics addressed hierarchical complexity after Heisenberg,
invoking dualities, uncertainty, and complementarity
between dynamics versus structure. Important develop
ments have turned on the tension across the dilemmas
presented by dual structures, as in the holon, a general
ized entity in a hierarchy. Holons have been equated with
the concept of system, with the advantage that holon does
not appear in common parlance. The holon can therefore
escape the reification and slovenly usage in the vernacular
where the model is mistaken for the materiality.
Conceptual developments suggest that what is inside a
given holon is chosen by an observer. This emphasizes
that holons are abstractions more than material objects; a
point forgotten when ‘system’ is used for ‘holon’. In the
holon, the tension is between system and subsystem. But
the subsystem is a system in its own right, thus offering
some sort of dual existence that invites contradiction.
The concept of the holon takes the whole to be a
surface that integrates the parts to give a unified signal
to the rest of the universe. At the same time, the holon is
the surface that integrates the external environment for
the parts to experience. In ecology, the environment falls
away at the level of the holon when viewed from the
perspective of the parts. A forest raises humidity and
lowers temperature, thus allowing survival of some of its
parts, tree seedlings that are its future. The parts are
protected. Conversely, viewed from the context of the
hot, dry environment surrounding the forest, the contri
butions of each tree to the water vapor inside the forest
are lost in a more general flux of water from the canopy.
Thus, loss of information occurs with movement both up
and down the hierarchy. While the environment is too
118
Hierarchy Theory in Ecology
large to catch the details of the working of the parts, the
parts themselves cannot span wide enough to see large,
slow differences in the larger context. In all this, we see
again the tension embodied in hierarchical discussions
between scale, organization, and uncertainty in
observation.
Earlier, five general principles for ordering ecological
hierarchies were recognized:
(1) As to frequency of behavior, higher level holons
operate at a lower frequency, taking longer to exhibit
returns in behavior than holons at lower levels.
(2) Higher levels in a hierarchy constrain lower levels by
displaying intransigent constancy. Deans constrain
faculty by not changing the budget, except once a
year.
(3) Higher levels in a hierarchy will be contextual to
lower levels. The environment would be seen as
operating at a higher level.
(4) With regard to bond strength, higher level holons are
held together by weaker forces than those that inte
grate lower level holons (e.g., chemical vs. nuclear
bonds) (Figure 3).
(5) As to containment, if higher level holons consist of
lower holons, which they contain, then the hierarchy
is said to be nested. Not all the criteria apply to all
hierarchies, but all five principles may apply
simultaneously.
The distinction between nested and non nested hierar
chies matters (Figure 3). In nested systems, upper level
entities contain and consist of lower level entities. In non
nested hierarchies, containment is not a criterion, but
principles (1)–(3) can still apply. In nested hierarchies,
containment applies even if aggregation criteria between
levels change type. Western medicine generally uses
nested hierarchies for the human condition. Thus orga
nelles may be aggregated into cells by biochemical
interaction. Meanwhile, nesting of organs inside the
whole body may invoke fluid mechanics as a principal
on which parts make the whole person. When whole
humans nest inside groups, relationships may be in epi
demiological terms. In Western medicine, there are
regular changes in aggregation criteria from biochemical,
through fluid dynamic, to epidemiological. Despite
inconstant criteria for linking levels, the nesting keeps
such hierarchies straight. But in non nested hierarchies,
such as food chains or pecking orders, the top dog neither
contains nor consists of the subordinate individuals.
Because there is no nesting to maintain order, non nested
hierarchies embody only one specific rule for moving
between levels. As a result, the criteria for moving up a
food chain must be consistently ‘is eaten by’, or conver
sely going down it is ‘eats’. In this way, the hierarchy is
consistent top to bottom. Because of their robustness to
changes in aggregation criteria, nested hierarchies are
particularly useful for exploration before firm criteria
connecting levels have been established. Concomitantly,
non nested hierarchies are for mature ideas, where
focused sets of relationships are organized and abstracted
in a control system.
In thermodynamic studies of ecological emergence,
nested hierarchies are essential, because otherwise the
bookkeeping of energy flow between the system and its
environment would not sum. In complexity theory,
self organized emergence is a matter of thermodynamic
gradients being applied to material systems that are
pushed away from equilibrium. Thus, nested hierar
chies apply when self organization is invoked, when
holons emerge at a new level without any plan.
Planned systems often yield to a non nested concep
tion. A surprising and important new turn in applied
ecology of human management systems links non
nested human socioeconomic hierarchies to nested
thermodynamic hierarchies. The whole system is embo
died in energy flow and control through the twinned
social and biogeochemical hierarchies.
These thermodynamic approaches develop self
organizing holarchic open systems (SOHOs), using the
term holarchy for nested hierarchies. The word holar
chy appears in part to sidestep the political
unacceptability of hegemonic hierarchical control.
Using the SOHO approach, the full power of hierarchy
theory in solving real time problems has been developed
by Waltner Toews and colleagues at NESH, a Canadian
centered, complex systems group. They solved some
critical problems in Peru, Kenya, and Nepal. For
instance, a Kathmandu sewer had children playing
around slaughterhouse waste. By linking the social hier
archy to the ecological process hierarchy, NESH
identified that a street cleaner caste was being blamed
for things out of their control. Blaming scapegoats had
led to inaction and paralysis, but once the street cleaners
were no longer held responsible, the SOHO thermody
namic methodology achieved significant rehabilitation
as the social and ecological hierarchies began to function
in concert.
The earliest explicit introduction of hierarchy the
ory into ecology in the 1970s spoke of decomposability
as an issue in some of the biomes studied in the
International Biological Program (IBP). At that time,
terms, such as ‘environ’, ‘creaon’, and ‘genon’ were
coined as extensions of the concept of holon. Environ
addresses the environment acting as an integrated
whole for its residents (see Ecological Network
Analysis, Environ Analysis). The inward direction
toward the holon pertains to the creaon, whereas the
outward direction pertains to the genon generating new
things and experiences for the environment and its
residents. Holon remains the central concept. The first
fully integrated treatments of hierarchies in ecology
Hierarchy Theory in Ecology
turned on epistemological implications of scale and
dynamics. Following shortly, evolutionary ideas focused
on the structural elements in hierarchies, in a more
ontological spirit. The structural elements were cast as
a triadic view of holons, where the level above and the
level below, as well as the level of the holon in
between, are all required for an adequate treatment.
Recently, two more crucial levels were added: the
level above the context keeps the context of the
holon stable, while the level below the parts provides
stability for the material of which the parts are made.
Scale and Type
Scale problems invited hierarchy theory into the disci
pline. Ecologists have long been aware of scale,
investigating the properties of quadrats in obtaining esti
mates of vegetation in the 1950s. Then change in variance
across quadrat size was used to measure aggregation of
plants on the ground. Hierarchy theory remains asso
ciated with scale today. The observation protocol brings
attention to a universe of a certain extent, while making a
second distinction, the finest grain at which observation
units are distinguished from one another. Grain and
extent together characterize the scalar level in question
in many ecological hierarchies. Grain and extent are
connected. Wider extents require coarser grains, if the
mass of data are to be remembered, analyzed, and under
stood. Modern computational power has widened the gap
between grain and extent, where remotely sensed areas
are captured in billions of pixels. Even so, explicitly link
ing items in the grain across the extent becomes
difficult, and generally impossible as the extent widens
by much.
In contrast to linking across scales, it is possible to
unify ecology across types of ecological system that
correspond to the main subdisciplines of ecology:
organism; population; community; ecosystem; land
scape; biome; and biosphere. These types for
ecological subdisciplines are explicitly not scale based,
and so are not required to be assigned to level in the
order given in the previous sentence. When scale is
parsed away from type, the various approaches to ecol
ogy achieve a sharper depth of focus, offering clear
relief between types of investigation. The subdisciplines
of ecology are not scalar levels. If they are levels at all,
they are type based levels of organization, with the
different types related to one another by asymmetric
relationships made explicit in the definitions. As a
separate issue, a typed level of organization itself con
tains scale based hierarchies, as in fractal landscapes. In
that scaled universe, the ecosystem modeling strategy
may apply across a range of sizes, where local processes
are part of more global processes. Communities too
119
may be variously inclusive of species across narrow or
wider areas. Under the organism criterion, examples are
found from redwood trees to mites. An ecological hier
archy may change the scale and type at the same time,
but it is fraught with conceptual danger. Indeed, hier
archy theory is often invoked to clean up the mess in
the aftermath of scale and type being mixed together.
There is no prohibition changing both together, but
only so long as the relationships at each new level
are explicit. This matters because most descriptions of
ecological material precisely do change type across
widening scalar levels, although most of them do not
follow the textbook ordering from organism to bio
sphere. For instance, in a forest community, a rotting
tree trunk may be considered an ecosystem, whose
upper surface is landscape, on which grows a commu
nity of bryophytes.
The copious variety of materials, entities, and sizes in
ecology invites hierarchy theory into ecology. Indeed, it is in
ecology that hierarchy theory has been used most often to
significant effect, as in the NESH studies mentioned above.
Hierarchy theory can capture a rich set of scaled examples
across a mixture of types. Ecology is a multiple scaled labyr
inth of types. Hierarchy theory is the ball of string that
we can trail behind, so that ecological scientists do not get
lost.
See also: Ecological Network Analysis, Environ Analysis.
Further Reading
Ahl V and Allen TFH (1996) Hierarchy Theory, A Vision Vocabulary and
Epistemology. New York: University of Columbia Press.
Allen TFH and Hoekstra TW (1992) Toward a Unified Ecology.
New York: University of Columbia Press.
Allen TFH, O’Neill RV, and Hoekstra TW (1984) Interlevel relations in
ecological research and management: Some working principles from
hierarchy theory. General Technical Report R.M.110. Fort Collins:
USDA Forest Service (republished in 1987 in Journal of Applied
Systems Analysis 14: 63 79).
Allen TFH and Starr TB (1982) Hierarchy: Perspectives for Ecological
Complexity. Chicago: University of Chicago Press.
Kay J, Regier H, Boyle M, and Francis G (1999) An ecosystem approach
for sustainability: Addressing the challenge of complexity. Futures
31: 721 742.
Koestler A (1967) The Ghost in the Machine. Chicago: Gateway.
O’Neill RV, DeAngelis D, Waide J, and Allen TFH (1986) Monographs in
Population Biology 23: A Hierarchical Concept of Ecosystems.
Princeton: Princeton University Press.
Overton WS (1975) Decomposability: A unifying concept? In: Levin S
(ed.) Proceedings of the SIAM SIMS Conference on Ecosystems
Analysis and Prediction, pp. 297 299. Philadelphia: Society for
Industrial and Applied Mathematics.
Patten BC (1978) Systems approach to the concept of environment.
Ohio Journal of Science 78: 206 222.
Pattee HH (ed.) (1973) Hierarchy Theory: The Challenge of Complex
Systems. New York: Braziller.
Salthe SN (1985) Evolving Hierarchical Systems. New York: Columbia
University Press.
Simon HA (1962) The architecture of complexity. Proceedings of the
American Philosophical Society 106: 467 482.
120
Goal Functions and Orientors
Waltner Toews D, Kay JJ, Neudoerffer C, and Gitau T (2003)
Perspective changes everything: Managing ecosystems from
the inside out. Frontiers in Ecology and the Environment
1(1): 23 30.
Webster JR (1979) Hiearchical organization of ecosytems. In: Halfon E
(ed.) Theoretic Systems Ecology, pp. 119 131. New York: Academic
Press.
Whyte LL, Wilson AG, and Wilson D (1969) Hierarchical Structures.
New York: Elsevier.
Relevant Websites
http://www.nesh.ca James Kay Web Page, Network for
Ecosystem Sustainability and Health (NESH).
http://www.nbi.ku.dk Stanley N. Salthe Web Page, Center for
the Philosophy of Nature and Science Studies (CPNSS),
Niels Bohr Institute.
Goal Functions and Orientors
H Bossel, University of Kassel (retd.), Zierenberg, Germany
ª 2008 Elsevier B.V. All rights reserved.
Introduction
System Concepts
System Orientation in a Complex Environment
Simulation of the Evolution of System Orientation
Further Reading
Introduction
When we talk about a viable system, we mean that this
system is able to survive, be healthy, and develop in its
particular environment. In other words, system viability has
something to do with both the system and its properties, and
with the environment and its properties. And since a system
usually adapts to its environment in a process of coevolu
tion, we can expect that the properties of the system’s
environment will be reflected in the properties of the sys
tem; for example, the form of a fish and its mode of motion
reflect the laws of fluid dynamics of its aquatic environment.
Systems are termed complex if they have an internal
structure of many – qualitatively different – processes,
subsystems, interconnections, and interactions. Besides
assuring their own viability, the individual systems that
are part of a complex total system specialize in certain
functions that contribute to the viability of the total sys
tem. Viability of subsystems and the total system requires
that subsystem functions and interactions are organized
efficiently (or at least effectively). In the evolution of
complex systems, two organizing principles in particular
have established themselves: hierarchy and subsidiarity.
They can be found in all successful complex systems:
biological, ecological, social, political, technological.
Hierarchical organization means a nesting of subsys
tems and responsibilities within the total system. Each
subsystem has a certain degree of autonomy for specific
actions, and is responsible for performing certain tasks
contributing to the viability of the total system. For
example, body cells are relatively autonomous subsys
tems, but contribute specific functions to the operation
The global ecosystem is made up of an ensemble of inter
acting local and regional ecosystems, each composed of
biotic and abiotic subsystems. The evolution of these
systems is constrained by physical and system laws and by
the basic properties of their environment, including the
constraints of exergy (energy that can be usefully trans
formed into work), material, and information flows.
Sustainability (persistence) of a system in its environment
therefore requires respecting these constraints. Conversely,
the very fact of its persistence demonstrates that a system
has successfully adapted to its operating conditions.
Evolution has forced it to respect physical and system laws
and the basic properties of its environment. To an observer,
the system’s behavior appears to be guided by a particular
attractor state, or by attention to a number of orientors.
System Concepts
System Organization
‘System’ is anything that is composed of system elements
connected in a characteristic system structure (Figure 1).
This configuration of system elements allows it to per
form specific functions in its environment. These
functions can be interpreted as serving a distinct system
purpose. The system boundary is permeable for inputs
from, and outputs to, the environment. It defines the
system’s identity and autonomy.
Goal Functions and Orientors
System environment
Feedbacks
S1
System output
S2
System
elements
S4
System
System inputs
S3
System structure
System boundary
Figure 1 System notation.
of particular body organs, which in turn contribute to the
viability of an organism.
Subsidiarity means that each subsystem is given the
responsibility and the means for keeping its own house in
order, within the range of its own abilities and potential.
Only if conditions occur that cannot be handled by the
subsystem would the suprasystem step in and help. The
principles of hierarchical organization and subsidiarity
require that each subsystem has a certain measure of
autonomy. In its particular environment, each subsystem
must be viable. The total system can only be viable if each
of the subsystems supporting it is viable. Each subsystem
reflects the properties of its individual environment; its
behavior is informed (oriented) by that environment.
Note that this way of looking at complex systems is
recursive. If necessary, we can apply the same system/
subsystem dichotomy of viable systems again at other
organizational levels. For example, a person is a subsys
tem of a family; a family is a subsystem of a community; a
community is a subsystem of a state; a state is a subsystem
of a nation, etc.
It is not enough to be concerned with the viability of
individual systems. There are no isolated systems in the
real world; all systems depend in one way or another on
other systems. Hence their viability, and ultimately the
viability of the total system are also preconditions for
sustainable development. This means that a holistic sys
tem view must be adopted.
Evolution of Systems, and Emergence of
Orientors and Goal Functions
The adaptation of a system to its environment is reflected
in its structure, including its nonmaterial, cognitive struc
ture. This system structure determines its behavior, and
hence the adaptive response to its particular environment.
System structures of material systems are dissipative; they
require exergy and material flows for their construction,
maintenance, renewal, and reproduction.
The dissipative structures of the global ecosystem are
constructed and maintained by a finite rate of exergy
121
input (mostly solar energy) and a finite stock of materials.
The global ecosystem is therefore forced to recycle all of
its essential material resources. The development of local
ecosystems is constrained by the local rate of exergy flux
(solar radiation input) and by the local rate of material
recycling (weathering rate, absorption rate, decomposi
tion rate, etc.) that it produces.
Evolution favors those species or (biotic) subsystems of
the ecosystem that have learned to use available resources
more efficiently and effectively than their competitors.
This learning is embedded in their genetic code, and it is
manifest in the dissipative structures they construct. Both
will increase in complexity as a species evolves. At the
ecosystem level, species evolution will cause increasingly
better use of (exergy and material) resources. Species as
well as ecosystems as a whole therefore tend to progress
toward more complex dissipative structure producing
more complex behavior.
Interacting species in a common ecosystem coevolve
in the direction of increasing fitness of each individual
species. Evolution of ecosystems therefore proceeds in the
direction (arrow of time) of specialization, speciation,
synergy, complexification, diversity, maximum through
flow of exergy, and more efficient use of material
resources. This development becomes manifest in the
corresponding emergent properties: exergy degradation,
recycling, minimization of output, efficiency of internal
flows, homeostasis and adaptation, diversity, hetero
geneousness, hierarchy and selectivity, organization,
minimization of maintenance costs, storage of available
resources. These properties can be viewed as orientors,
propensities, or attractors guiding system evolution and
development. They are not limited to ecosystems; they
are a general feature of living systems, including human
organizations. When quantified and used in models, we
refer to them as goal functions.
In particular, ecosystems will therefore build up in the
course of their development as much dissipative structure as
can be supported by the available exergy gradient. Available
opportunities will eventually be found out by the processes
of evolution, and will then be utilized. The ability to
respond successfully to environmental challenges can be
‘interpreted’ as intelligent behavior, although it is strictly
the result of nonteleological evolutionary development.
System Orientation in a Complex
Environment
Basic concepts can be introduced by visualizing a simple
animal with limited vision in a simple environment. The
animal requires exergy for self organization, motion, har
vesting food, and maintenance. The environment provides
122
Goal Functions and Orientors
food in certain locations, usually associated with obstacles
that must be avoided since they have an exergy cost.
In a stable environment where sufficient (regenerat
ing) food is distributed in a completely regular pattern,
evolutionary adaptation would eventually lead to optimi
zation of an animal’s movements in a regular grazing
pattern, with a single objective, optimum exergy uptake
and use. The regular grazing pattern reflects the complete
certainty of the next step, which the animal learns by
accumulating and internalizing experience in a cognitive
structure aiding its limited vision.
In more complex and diverse environments the ani
mal, because of its limited vision, may not know for
several steps which situation it will encounter next. It
will therefore have to develop decision rules that have
greater generality and are applicable to (and will be
reinforced by) different motion sequences with different
outcomes. In addition to the requirement of harvesting
and using exergy resources effectively and efficiently,
another objective is now implicitly added, to secure food
under the constraint of incomplete information, that is, a
security objective. Note that this is an emergent property
that is not explicit in the reward system (which still
rewards only food uptake). Failure to heed this implicit
security objective will reduce food uptake and may
endanger survival. On the other hand, the pressure to
play it safe will occasionally mean giving up relatively
certain reward. With other words, efficiency is traded for
more security, and both are now prominent normative
orientations (goals, values, interests) incorporated in the
cognitive structure.
Orientation theory deals in a more general way with
the emergence of behavioral objectives (orientors) in self
organizing systems in general environments. The propo
sition is that if a system is to survive in a given
environment – characterized by a specific normal envi
ronmental state, sparse resources, variety, unreliability,
change, and the presence of other systems – it must be
able to physically exist in (be compatible with) this envi
ronment, effectively harvest necessary resources, freely
respond to environmental variety, protect itself from
unpredictable threats, adapt to changes in the environ
ment, and interact productively with other systems.
These essential orientations emerge in the course of the
system’s evolution in its environment.
(explicit or implicit) normative concepts that direct beha
vior and development of systems in general. In the social
context, values and norms, objectives and goals are impor
tant orientors. Ecosystems and organisms tend toward
certain attractor states whose specific characteristics can
be viewed as orientors. Orientors exist at different levels
of concreteness within an orientor hierarchy. The most
fundamental orientors, the basic orientors, are identical
for all complex adaptive systems. Orientors are dimen
sions of concern; they are not specific goals. Their
satisfaction can be determined by observation of corre
sponding indicators, which can also be used to define goal
functions for model studies.
In addition to the physical constraints of exergy and
material flows, ecosystem and species development
is determined by the ‘general properties of the environment’:
1. Normal environmental state . The actual environmental
state can vary around this state in a certain range.
2. Scarce resources. Resources (exergy, matter, informa
tion) required for a system’s survival are not
immediately available when and where needed.
3. Variety. Many qualitatively very different processes
and patterns occur in the environment constantly or
intermittently.
4. Reliability. The normal environmental state fluctuates
in random ways, and the fluctuations may occasionally
take it far from the normal state.
5. Change. In the course of time, the normal environmen
tal state may gradually or abruptly change to a
permanently different normal environmental state.
6. Other systems. The behavior of other systems changes
the environment of a given system.
Basic Orientors
Properties of Environments
If evolution enforces fitness of (natural) systems, then
persistent systems must reflect the properties of their
environment in their structure. More generally, the
basic properties of the environment require correspond
ing basic system features. Since the basic environmental
properties are independent of each other, a similar set of
independent system features must exist, and it must find
expression in the concrete features of the system
structure.
There is a one to one relationship between the prop
erties of the environment and the ‘basic orientors of
systems’ (Figure 2):
There is obviously an immense variety of system envi
ronments, just as there is an immense variety of systems.
But all of these environments have some common general
properties. These properties will be reflected in systems.
These reflections, or basic orientors, orient not just struc
ture and function of systems, but also their behavior in the
environment. The term orientor is used to denote
1. Existence. Attention to existential conditions is necessary
to insure the basic compatibility and immediate survival
of the system in the normal environmental state.
2. Effectiveness. In its efforts to secure scarce resources
(exergy, matter, information) from, and to exert influ
ence on its environment, the system should on balance
be effective.
Goal Functions and Orientors
Normal
environmental
state
Environmental
reliability
Scarce
resources
Existence
Security
Efficiency
Environment
System
Adaptability
Freedom
Coexistence
Environmental
change
Environmental
variety
Other
systems
Figure 2 A tentative typology of emergent properties.
3. Freedom of action. The system must have the ability to
cope in various ways with the challenges posed by
environmental variety.
4. Security. The system must have the ability to protect
itself from the detrimental effects of variable, fluctu
ating, unpredictable, and unreliable environmental
conditions.
5. Adaptability. The system should be able to change its
parameters and/or structure in order to generate more
appropriate responses to challenges posed by changing
environmental conditions.
6. Coexistence. The system must modify its behavior to
account for behavior and interests (orientors) of other
systems.
Obviously, the system equipped to secure better overall
orientor satisfaction will have better fitness, and will
therefore have a better chance for long term survival
and sustainability. In persistent systems or species, these
orientors will be found as emergent objectives (or system
interests).
Properties of Orientors
Each of the basic orientors stands for a unique require
ment. Attention (conscious or unconscious) must
therefore be paid to each of them, and the compensation
of deficits of one orientor by over fulfillment of other
orientors is not possible. Fitness forces a multicriteria
response, and comprehensive (conscious or unconscious)
assessments of system behavior and development must
also be multicriteria assessments.
123
In the assessment and orientation of system behavior,
we deal with a two phase assessment process where each
phase is different from the other.
Phase 1. First, a certain minimum satisfaction must be
guaranteed separately for each of the basic orientors.
A deficit in even one of the basic orientors threatens
long term survival. The system will have to focus its
attention on this deficit.
Phase 2. Only if the required minimum satisfaction of
all basic orientors is guaranteed is it permissible to try to
raise system satisfaction by improving satisfaction of indi
vidual orientors further.
Adequate satisfaction of each of the basic orientors
requires, on a lower level, system and environment
specific satisfaction of thermodynamic, structural, func
tional, ecophysiological, and system orientors. Network
analysis suggests complementarity of different formula
tions of extremal principles as orientors describing
ecosystem development.
Characteristic differences in the behavior of otherwise
very similar systems (animals, humans, political, or cultural
groups) can often be explained by differences in the rela
tive importance attached to different basic orientors
(i.e., emphasis on freedom, or security, or effectiveness, or
adaptability) in phase 2 (i.e., after minimum requirements
for all basic orientors have been satisfied in phase 1).
The basic orientor proposition has three important
implications:
1. If a system evolves in a normal environment, then that
environment forces it to implicitly or explicitly ensure
minimum and balanced satisfaction of each of the basic
orientors (and of lower level orientors contributing to
this satisfaction).
2. If a system has successfully evolved in a normal envi
ronment, its behavior will exhibit balanced satisfaction
of each of the basic orientors.
3. If a system is to be designed for a normal environment,
proper and balanced attention must be paid to satisfac
tion of each of the basic orientors.
The third implication has particular relevance for the crea
tion of programs, institutions, and organizations in the
sociopolitical sphere, among other things. Note that for a
specific system in a specific environment, each orientor will
have a specific meaning. For example, security of a nation is
a multifacetted objective set with very different content
from the security of an individual particular organism.
However, the systems theoretical background for satisfac
tion of the security orientor is the same in both cases.
Orientors as Implicit Attractors
Better orientor satisfaction (better fitness) for more parti
cipants in a system requires more dissipative structure,
which requires more exergy throughput as well as exergy
124
Goal Functions and Orientors
accumulation. Since the exergy flow of ecosystems is
limited (capture of solar radiation by photoproduction),
increasingly better utilization is to be expected in the
course of system development. This saturates at maxi
mum exergy flow utilization for the ecosystem as a whole.
Ecosystems as a whole therefore move in the direction of
using all available exergy gradients. For organisms in the
ecosystem, this implies development tendencies (orien
tors, propensities, attractors) toward specialization (using
previously unused gradients), more complex structure
(greater use efficiency), larger individuals (less mainte
nance exergy required per biomass unit), mutualism, etc.
For species development, this translates into a principle of
maximum exergy use efficiency. On the basis of these
principles, prediction of development trends in ecosys
tems is possible.
The selection for better fitness in evolutionary pro
cesses favors systems (organisms) with better coping
ability. Aspects of the behavioral spectrum of a system
that improve coping ability (basic orientors) can be
understood as implicit goals or attractors: existence,
security, effectiveness, freedom, adaptability, coexistence.
In the developmental stage of ecosystems, emphasis is on
the basic orientors: existence, effectiveness, and freedom;
in the mature stage it shifts to security, adaptability, and
coexistence (see Table 1, where orientor concepts have
been linked to E. P. Odum’s classical model of ecological
succession).
The existence of these implicit goals does not imply
teleologic or teleonomic development toward a given
Table 1 Orientor concepts in the context of ecological
succession
Developmental
stage
Mature stage
Basic orientor emphasis
Existence
Coexistence
Freedom
Security
Effectiveness
Adaptability
Ecosystem orientor
Growth and change
Life cycle
Biomass
Energy conservation
Nutrient conservation
Nutrient recycling
Specialization
Diversity
Organization
Symbiosis
Stability; feedback
control
Structure
Orientor emphasis (goal function)
High
Low
Short, simple
Long, complex
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Information
Entropy
Low
High
Linear, simple
Network,
complex
High
Low
goal (where the final state is specified). These attractors
do not determine the exact future states of the system at
all; they only pose constraints on choices (or evolutionary
selection). The process and its rules are known, the
product is unknown. The spectrum of (qualitatively dif
ferent) possible future development paths and sustainable
states remains enormous. The shape of the future, and of
the systems that shape it, cannot be predicted this way. All
one can say with certainty, however, is that (1) all possible
futures must be continuous developments from the past,
and (2) paths with better orientor satisfaction are more
likely to succeed in the long run (if options to change
paths have not been foreclosed).
In many systems, in particular ecosystems, specific
attractors or functional orientors are often more imme
diately obvious than the basic orientors that cause the
emergence of these orientors in the first place. These
orientors can be viewed as appearing on a level below
the basic orientors in the hierarchical orientation system
(see Table 1). They translate the fundamental system
needs expressed in the basic orientors into concrete
attractor states linking system response to environmen
tal properties. In models and ecosystem analyses,
measures of ecosystem integrity can be based on corre
sponding ecosystem goal functions. Ecosystem attractor
states emerge as general ecosystem properties in the
coevolution of ecosystem and environment. They can
be viewed as ecosystem specific responses to the need
to satisfy the basic orientors. Major ecosystem orientors
are optimization of use of solar radiation, material, and
energy flow intensities (networks); matter and energy
cycling (cycling index); storage capacity (biomass accu
mulation); nutrient conservation, respiration, and
transpiration; diversity (organization); hierarchy (signal
filtering).
The emergence of basic orientors in response to the
general properties of environments can be deduced from
general systems theory, but supporting empirical evidence
and related theoretical concepts can also be found in
such fields as psychology, sociology, and the study of
artificial life.
Orientor Guidance in System Development,
Control, Adaptation, and Evolution
Environmental influences partially determine system
behavior. The magnitude of their effect on behavior
depends on the influence structure of the system.
Sometimes systems can be controlled by controlling
the inputs from their environment. However, the feed
backs in the system itself are usually more important for
system control and adaptation of behavior to environ
mental conditions. Feedback means that the system state
influences itself. Behavior changing internal feedbacks
Goal Functions and Orientors
are possible on several hierarchical levels in complex
systems with different typical response characteristics
and time constants (typical response times). These pos
sibilities are also shown in Figure 3.
Response time
Level
Response
Immediate
Short
Medium
Long
Very long
Always
Process
Feedback
Adaptation
Self-organization
Evolution
Basic orientors
Cause–effect
Control
Parameter change
Structural change
Change of identity
Maintaining integrity
The simplest type of system response is the cause–
effect relationship. It occurs at once as in, for example,
stimulus–response reflex. It is the only type of system
behavior which can legitimately be described by relating
the output directly to the input. Unfortunately, it is often
assumed that the same simple relationship is also applic
able to other types of system response (such as the
following), and this erroneous assumption often leads to
fundamental mistakes.
On the next higher level we find responses which are
generated by feedback in the system, involving at least
one state variable or delay – such as an empty stomach
causing hunger and the search for food. Control processes
belong to this category. The response time is short, and
influence structure and system parameters remain
invariant.
On the next higher level we find processes of adapta
tion. In this case the system maintains its basic influence
structure, but parameters are adjusted to adapt to the
situation, possibly changing the response characteristics
in the process. For example, a tree may adapt to the
gradual lowering of the groundwater level by growing
its roots to greater depth. This constitutes a parameter
change (root length and root surface). The fundamental
system structure of a tree, in particular, the function of the
roots, has not changed in this case.
On the next higher level we find processes of self
organization in response to environmental challenges.
This means structural change in the system. Processes
of this kind have a longer response time and can only be
conducted by systems having the capability for self
organization. Adult organisms or technical systems
rarely or never belong to this category; on the other
hand, this characteristic is often found in the develop
ment of organisms, social systems, organizations, and
ecosystems.
A system may also change its identity in the course of
an evolutionary process. This means that its functional
characteristics, and hence its system purpose, change with
time. Adaptations of this kind take place as a result of
reproduction and evolution of living organisms. It is char
acteristic of this process that the system change coincides
with a possibly drastic shift in system identity (change of
goal function and of system purpose). An evolutionary
example is the development of flying animals (birds) from
water dwelling reptiles.
All of these system responses to challenges from the
environment in essence constitute attempts to maintain
system integrity (possibly over many generations and
over a long time period) even if it means changing system
identity, that is, system purpose. From this observation it
can be deduced that a system must orient its development
with respect to certain basic criteria (basic orientors) to
assure its long term existence and development in an
often hostile environment. This orientation may be impli
cit (forced upon the system) or explicit (actively pursued
by the system). It does not require conscious decision or
Integrity
Orientors
Evolution
Purpose
Self-organization
Structure
Parameters
Adaptation
System input
125
Discrepancy
Control
System state
System output
Feedback
Stimulus response
Figure 3 System response can be caused by different processes with very different time constants: stimulus–response, feedback
control, adaptation, self-organization, evolution, maintaining system integrity.
126
Goal Functions and Orientors
even cognitive ability, although resulting action may
appear to an observer as intelligent or even goal or
value oriented behavior.
Simulation of the Evolution of System
Orientation
Animats and Genetic Algorithms for Orientation
Orientation theory is not just a conceptual framework for
understanding system evolution and behavior under the
exergy availability constraint. It also allows quantitative
and comparative analysis of system performance under
different environmental conditions.
Genetic algorithms are models of biological adaptive
processes that are being widely and successfully applied
to a wide spectrum of adaptation and optimization pro
blems. In particular, these algorithms have been used to
simulate learning and adaptation of artificial animals (ani
mats) in simulated environments containing food and
obstacles. They can be used to demonstrate the emer
gence of basic orientors in self organizing systems having
to cope with complex environments.
The animat model incorporates essential features of a
simple animal in a diverse environment. Being an open
system, an animal depends on a flow of exergy from the
environment. In the course of its (species) evolution, it has to
learn to associate certain signals from the environment with
reward or pain and to either seek or avoid their respective
sources (exergy gain or exergy loss). This learning phase (of
populations) will eventually lead to the establishment of
cognitive structure and behavioral rules which are approxi
mately optimal in the particular environment (with respect
to maximization of reward, minimization of pain, and secur
ing survival). These behavioral rules incorporate knowledge
which enables intelligent behavior.
The animat is designed to simulate this process. It can
pick up sensory signals from its environment (containing
food and obstacles), and classify them with available rules
to determine an appropriate action (direction of move
ment). After a successful move, the strength of rules
leading up to it is increased by sharing in the reward
(i.e., exergy gain). New rules are occasionally generated
by either random creation, or by genetic operations
(crossing over and recombination). They are added to
the existing rule set, and compete with the other rules
for reward. Unsuccessful rules are not reinforced and lose
strength and influence in the rule set.
The training process consists of placing the animat at a
random empty location in an environment with specific
environmental properties, and allowing it to move around
searching for food. A collision with an obstacle causes a loss
of exergy and throws the animat back to its previous posi
tion. Rules leading to success are rewarded. A genetic event
of rule generation may occur with a prescribed probability.
Random rules are created in unknown situations. The pro
cess is repeated for a large number of steps (typically 10 000).
Eventually, a set of behavioral rules develops which allows
optimal behavior under the given set of conditions.
Note that this optimal behavior has not been defined in
terms of an objective function guiding the evolution of the
set of behavioral rules. The rule set develops solely from
the reinforcement of rules which lead to food or avoid
collisions. An explicit exergy balance accounts for all
exergy losses associated with movement, collisions with
obstacles, and rule generation, and exergy gains due to
uptake of food. The development of the rule set is then
driven by the requirement to optimize exergy pickup in
the given environment (with specific resource availabil
ity), while allowing for environmental variety, variability,
and change specific for that environment. Neglect of these
properties is penalized by lack of fitness, and threat to
survival, and causes disappearance of deficient rules.
Other criteria besides efficiency will therefore be
reflected in the set of behavioral rules. Since these were
not expressly introduced, we must recognize them as
emergent value orientations or objective functions.
The animat experiment contains all components
necessary for a study in the basic orientor framework.
Animat fitness depends on the ability to maintain a posi
tive exergy balance in the long term. This exergy balance
is therefore at the core of the orientor satisfaction assess
ment. At each step, exergy uptake (by food consumption)
and exergy losses (by collisions with obstacles, motion,
and learning of rules) are recorded and used to compute
the momentary exergy balance. Attention to all orientors
is mandatory to ensure a positive exergy balance even
under adverse environmental conditions.
Quantitative measures must be defined for character
izing the different properties of the environments used in
the animat experiments. Animat performance in different
environments is compared by using measures of orientor
satisfaction. These have to be defined using relevant
parameters of animat performance.
Emergence of Basic Value Orientations,
Anticipation, and Individual Differences
Since the animat’s training depends on a number of ran
dom factors, each animat develops a different cognitive
system (classifier set and decision rules), even though
final performance may be similar. In order to show gen
eral tendencies despite these individual differences, mean
values over large populations were obtained. These dealt
with (1) results of the training process in two (otherwise
identical) environments having different variety and
variability, and with (2) performance of animats after
transfer from their training environment to environments
challenging them with more variety, or variability, or
change.
Goal Functions and Orientors
One remarkable result from these experiments is that
individuals achieve comparable performance in a given
training environment with very different cognitive sys
tems, and in particular with different orientor emphasis.
While this may not provide any particular advantage in
the training environment, it may provide distinct fitness
advantages if the animat is moved to a different environ
ment. Three particular types of individuals stand out:
generalists (type F) stressing freedom of action, specialists
(type E) focusing on effectiveness, and cautious type
(type S) emphasizing security. Figure 4 shows the
different orientor stars for these three types.
The ability to develop a cognitive system reflecting its
environment makes the animat a suitable vehicle for inves
tigating goal function emergence and value orientation.
Genetic algorithms are very effective processes that seem
to capture the essentials of real processes found in the
evolution of organisms and ecosystems. In the animat,
they very effectively build up a cognitive model (or goal
function) that enables anticipatory behavior; since rewards
flow back to earlier rules leading to later pay off, the
activation of the initial rules in a pay off chain means
that the system suspects possible pay off and anticipates
the near future, that is, it has an internal model of the
results of its actions under the given circumstances.
In the animat experiments (and similarly, in real life),
implicit and (more or less) balanced multidimensional
attention to the basic orientors emerges from the simple
one dimensional mechanism of rewarding success in the
given environment. Thus, in the course of its evolutionary
development in interaction with its environment, the
system evolves a complex multidimensional behavioral
objective function from the very unspecific requirement
of fitness. Conversely, this also means that balanced atten
tion to the emergent basic orientors is necessary for
2
1
Security
0
Freedom
Adaptability
Generalist (F)
Specialist (E)
system viability and survival – they would not have
emerged unless important for the viability of the system.
Balanced attention still leaves room for individual
differences in the relative emphasis given to the differ
ent orientors. Individuals belonging to the populations
used in the animat experiments evolve significant dif
ferences in value emphasis (e.g., specialist, generalist,
cautious type). These individual variations, while not
significantly reducing performance in the standard
training environment, provide comparative advantage
and enhanced fitness when resource availability, vari
ety, or reliability of the environment change. They also
result in distinctly different behavioral styles. However,
pathological behavior will follow if orientor attention
becomes unbalanced (e.g., dominant emphasis on one
orientor).
Training of animats in different environments, the per
formance of animat individuals in environments that differ
from their training environments, and the simulation of
adaptive learning in a changing environment, lead to
some general conclusions that are in full agreement with
everyday observations and general systems knowledge:
have a better survival chance than others if
• Generalists
moved to an environment of greater variety.
types have a better survival chance than
• Cautious
others if moved to a less reliable environment.
in more unreliable and/or more diverse
• Training
environments increases satisfaction of the security
•
•
•
Effectiveness
3
127
and/or freedom of action orientors at the cost of the
effectiveness orientor.
Training in an uncertain environment teaches caution
and improves fitness in a different environment.
Learning caution (better satisfaction of the security
orientor) takes time and decreases effectiveness, but
increases overall fitness.
Investment in learning (exergy cost of learning in the
animat) pays off in better fitness; the learning invest
ment is (usually) much smaller than the pay off gain.
Animat individuals not only develop behavior that can be
interpreted as intelligent, they also develop a complex
goal function (balanced attention to basic orientors), or
value orientation. Serious attention to basic values (basic
orientors: existence, effectiveness, freedom, security,
adaptability, coexistence) is therefore an objective
requirement emerging in, and characterizing self organi
zing systems. These basic values are not subjective human
inventions; they are objective consequences of the process
of self organization in response to normal environmental
properties.
Cautious (S)
Figure 4 In an identical training environment, different lifestyles
may evolve. Generalists stress freedom of action, specialists
focus on effectiveness, while cautious types emphasize security.
See also: Ecological Network Analysis, Ascendency;
Ecological Network Analysis, Environ Analysis; Exergy;
Fundamental Laws in Ecology.
128
Exergy
Further Reading
Ashby WR (1962) Principles of the self organizing system. In: von
Foerster H and Zopf GW (eds.) Principles of Self Organization,
pp. 255 278. New York: Pergamon.
Bossel H (1977) Orientors of nonroutine behavior. In: Bossel H (ed.)
Concepts and Tools of Computer Assisted Policy Analysis,
pp. 227 265. Basel: Birkhauser.
Bossel H (1999) Indicators for Sustainable Development: Theory,
Method, Applications. Winnipeg: IISD International Institute for
Sustainable Development.
Bossel H (2001) Exergy and the emergence of multidimensional system
orientation. In: Jørgensen SE (ed.) Thermodynamics and Ecological
Modelling, pp. 193 209. Boca Raton, FL: Lewis.
Fath BD, Patten BC, and Choi JS (2001) Complementarity of ecological
goal functions. Journal of Theoretical Biology 208(4): 493 506.
Holland JH (1992) Adaptation in Natural and Artificial Systems.
Cambridge, MA: MIT Press.
Jantsch E (1980) Self Organizing Universe: Scientific and Human
Implications of the Emerging Paradigm of Evolution. New York:
Pergamon.
Jørgensen SE (2001) A tentative fourth law of thermodynamics.
In: Jorgensen SE (ed.) Thermodynamics and Ecological Modelling,
pp. 305 347. Boca Raton, FL: Lewis.
Krebs F and Bossel H (1997) Emergent value orientation in self
organization of an animat. Ecological Modelling 96: 143 164.
Mayr E (1974) Teleological and teleonomic: A new analysis. Boston
Studies in the Philosophy of Science 14: 91 117.
Mayr E (2001) What Evolution Is. New York: Basic Books.
Miller JG (1978) Living Systems. New York: McGraw Hill.
Odum EP (1969) The strategy of ecosystem development. Science
164: 262 270.
Muller F and Leupelt M (eds.) (1998) Eco Targets, Goal Functions, and
Orientors. Berlin/Heidelberg/New York: Springer.
Wilson SW (1985) Knowledge growth in an artificial animal.
In: Grefenstette JJ (ed.) Proceedings of the First International
Conference on Genetic Algorithms and Their Applications,
pp. 16 23. Pittsburgh PA and San Mateo: Lawrence Earlbaum and:
Morgan Kaufmann.
Exergy
S E Jørgensen, Copenhagen University, Copenhagen, Denmark
ª 2008 Elsevier B.V. All rights reserved.
Definition: Exergy
Definition: Eco-Exergy
Exergy and Information
Exergy and the Dissipative Structure
How to Calculate Eco-Exergy of Organic Matter and
Organisms?
Why Do Living Systems Have Such a High Level
of Eco-Exergy?
Losses and Gains of Eco-Exergy by Human Activities
Included Pollution
Formulation of a Thermodynamic Hypothesis for
Ecosystems
Support to the Maximum Eco-Exergy Hypothesis
Further Reading
Definition: Exergy
S, U, V, N1, N2, N3,...
Energy is defined as the amount of work (¼ entropy free
energy) a system can perform when it is brought into
thermodynamic equilibrium with its environment
(Figure 1). The considered system is characterized by the
extensive state variables S, U, V, N1, N2, N3, . . . , where S is
the entropy, U is the energy, V is the volume, and N1, N2, N3,
. . . are moles of various chemical compounds, and by the
intensive state variables, T, p, c1, c2, c3, . . . , where T is the
temperature, p the pressure, and symbolizes the chemical
potential of the components 1,2,3,. . .. The system is
coupled to a reservoir or reference state by a shaft, together
forming a closed system. The reservoir (the environment) is
characterized by the intensive state variables To, po, oc1,
oc2, oc3, . . . , and as the system is small compared with the
reservoir the intensive state variables of the reservoir will
not be changed by interactions between the system and the
reservoir. The system develops toward equilibrium with the
T, p, μc , μc , μc ,...
1
2
3
Toward thermodynamic
equilibrium with the
environment
S, Uo, V, N1, N2, N3,...
To, po, μoc , μoc , μoc ,...
1
2
3
Figure 1 Definition of exergy is shown. The work is symbolized
by the gain in potential energy of the weight.
reservoir and is simultaneously able to release entropy free
energy to the reservoir. During this process the volume of
the system is constant as the entropy free energy (i.e., work
energy) must be transferred through the shaft only.
Exergy
According to the definition of exergy, Ex, we have
Ex ¼ U ¼ U Uo
½1
As
U ¼ TS pV þ
X
c Ni
c
½2
(when we only consider the three energy forms:
heat, spatial energy (displacement work), and chemical
energy; see any textbook in thermodynamics), and
correspondingly
Uo ¼ To S po V þ
X
co Ni
c
½3
we get the following expression for exergy, when in this
case kinetic energy, potential energy, electrical energy,
radiation energy, and magnetic energy are excluded we
have
Ex ¼ S ðT To Þ V ðp po Þ þ
X
ðc co ÞNi
c
½4
The total transfer of entropy free energy in this case is
the exergy of the system. It is seen from this definition
that exergy is dependent on the state of the total system
(¼ system þ reservoir) and not dependent entirely on the
state of the system. Exergy is therefore not a state variable.
This definition of exergy is used in engineering to
express the efficiency of power plants. The energy effi
ciency of power plants is of course 100%, according to the
first law of thermodynamics, while the interesting effi
ciency is the exergy efficiency: how much of the chemical
energy (exergy) in the applied fossil fuel if fossil fuel is the
energy source is converted to useful work (exergy)? What
is not converted to exergy in form of electricity is lost as
heat to the environment at the temperature of the envir
onment – it contains therefore no work potential.
Notice that the exergy of the system is dependent on
the intensive state variables of the reservoir. Notice that
exergy is not conserved – only if entropy free energy is
transferred, which implies that the transfer is reversible.
All processes in reality are, however, irreversible, which
means that exergy is lost (and entropy is produced). Loss
of exergy and production of entropy are two different
descriptions of the same reality, namely, that all processes
are irreversible, and we unfortunately always have some
loss of energy forms which can do work to energy forms
which cannot do work (heat at the temperature of the
environment). So, the formulation of the second law of
thermodynamic by use of exergy is ‘all real processes are
irreversible which implies that exergy inevitably is lost’.
‘Exergy is not conserved’, while energy of course is con
served by all processes according to the first law of
thermodynamics.
129
The efficiency of concern is the ratio of useful energy
(work) to total energy which always is less than 100% for
real processes, which always are irreversible. This effi
ciency expresses that a part of the energy cannot be
utilized as work and that all processes are irreversible
because exergy is lost by all energy transfer processes as
heat to the environment.
Exergy efficiency, defined as work performed divided
by the total exergy available, is also of interest, particu
larly in technology. It expresses how much of the work
capacity we are able to utilize.
All transfers of energy imply that exergy is lost because
energy is transformed to heat at the temperature of the
environment. It has therefore been of interest to set up for
all environmental systems an exergy balance in addition
to an energy balance. Our concern is exergy loss because
it means that ‘first class energy’ which can do work is
converted to ‘second class energy’ (heat at the tempera
ture of the environment) which cannot do work. So, the
particular properties of heat and temperature are a meas
ure of the movement of molecules, given limitations in
our possibilities to utilize energy to do work. Due to these
limitations, we have to distinguish between exergy which
can do work and anergy which cannot do work, and all
real processes imply inevitably a loss of exergy as anergy
(see also the next section).
Exergy or rather the loss of exergy as heat, which
means production of entropy, seems more useful to
apply than entropy to describe the irreversibility of real
processes. It has the same unit as energy and is an energy
form, while the definition of entropy is more difficult to
relate to concepts associated to our usual description of
reality. In addition entropy is not clearly defined for
‘far from thermodynamic equilibrium systems’,
particularly for living systems. Moreover, it should be
mentioned that the self organizing abilities of systems
depend strongly on the temperature. Exergy takes the
temperature into consideration as the definition shows,
while entropy does not. It implies that exergy at 0 K is 0
and at minimum. Negative entropy is not expressing
the ability of the system to do work (we may call it
‘the creativity’ of the system as creativity requires
work), but exergy becomes a good measure of ‘the
creativity’, which is increasing proportional with the
temperature. Furthermore, exergy facilitates the differen
tiation between low entropy energy and high entropy
energy, as exergy is entropy free energy.
Information contains exergy. Boltzmann showed that
the free energy of information (it means exergy) that we
actually possess (in contrast to the information we need to
describe the system) is kT ln I, where I is the information
we have about the state of the system, for instance,
that the configuration is 1 out of W possible ones and k
is Boltzmann’s constant ¼ 1.3803 10 23 J/(molecules
deg). It implies that one bit of information has the exergy
130
Exergy
equal to kT ln2. Transformation of information from one
system to another is often almost an entropy free energy
transfer. If the two systems have different temperatures,
then the entropy lost by one system is not equal to the
entropy gained by the other system, while the exergy lost
by the first system is equal to the exergy transferred and
equal to the exergy gained by the other system, provided
that the transformation is not accompanied by any loss of
exergy. Also, in this case, it is obviously more convenient
to apply exergy than entropy.
System at T, p
Displacement
work, not useful
Work (exergy)
Reference environment
at T, p
Definition: Eco-Exergy
In ecology, technological exergy is not so useful
because the reference state, the environment, would
be the adjacent ecosystem and we would like to find
an expression that can measure how developed an
ecosystem is, that is, how far it is from thermodynamic
equilibrium. For a reservoir or reference state, it is
therefore advantageous in ecology to select the same
system but at thermodynamic equilibrium, that is, that
all components are inorganic and at the highest oxida
tion state, if sufficient oxygen is present (nitrogen as
nitrate, sulfur as sulfate, etc.). The reference state will
in this case correspond to the ecosystem without life
forms and with all chemical energy utilized or as an
‘inorganic soup’. Usually, it implies that we also con
sider T ¼ To, and p ¼ po, which means that the exergy
becomes equal to the difference of Gibb’s free energy
of the system and the same system at thermodynamic
equilibrium, or the chemical energy content included
the thermodynamic information (see below) of the
system. Gibb’s free energy is defined according to
the following equation:
dG ¼ dE þ p dV S dT
where dV is the change in volume and dS is the change
in entropy. T and p are the temperature and pressure,
respectively. The exergy becomes by this definition
clearly a measure of how far the ecosystem is from thermo
dynamic equilibrium, that is, how much (complex)
organization the ecosystem has build up in the form of
organisms, complex biochemical compounds, and com
plex ecological network. Here, we use the available work,
that is, the exergy, as a measure of the distance from
thermodynamic equilibrium.
This description of exergy development in an ecosys
tem makes it pertinent to assess the exergy of ecosystems.
It is not possible to measure exergy directly – but it is
possible to compute it by eqn [4]. Figure 2 illustrates the
definition of ‘eco exergy’. As the chemical energy embo
died in the organic components and the biological
structure contributes far most to the exergy content of
Figure 2 The exergy content of the system is calculated in the
text for the system relatively to a reference environment of the
same system at the same temperature and pressure, as an
inorganic soup with no life, biological structure, information, or
organic molecules.
the system, there seem to be no reason to assume a
(minor) temperature and pressure difference between
the system and the reference environment. Under these
circumstances we can calculate the exergy content of the
system as coming entirely from the chemical energy:
Ex ¼
X
ðc co ÞNi
c
½5
This represent the nonflow chemical eco exergy. The
difference in chemical potential (c – co) between the eco
system and the same system at thermodynamic equilibrium
determines the eco exergy. This difference is determined by
the concentrations of the considered components in the
system and in the reference state (thermodynamic equili
brium), as it is the case for all chemical processes.
We can measure the concentrations in the eco
system, but the concentrations in the reference state
(thermodynamic equilibrium) can only be based on the
usual use of chemical equilibrium constants. If we have
the process
Component A $ inorganic decomposition products
It has a chemical equilibrium constant, K:
K ¼ ½inorganic decomposition products=½Component A ½6
The concentration of component A at thermodynamic
equilibrium is difficult to find, but we can find the
concentration of component A at thermodynamic equili
brium from the probability of forming A from the
inorganic components.
We find by these calculations the exergy of the system
compared with the same system at the same temperature
and pressure but in form of an inorganic soup without any
life, biological structure, information, or organic mol
ecules. As (c – co) can be found from the definition of
Exergy
the chemical potential replacing activities by concentra
tions, we get the following expressions for eco exergy:
Ex ¼ RT
i n
X
ci ln ci =ci;o
i 0
½7
where R is the gas constant (8.314 J K 1 mol 1 ¼
0.08207 l atm. K 1 mol 1), T is the temperature of the
environment (and the system; see Figure 2), while ci is
the concentration of the ith component expressed in a
suitable unit, for example, for phytoplankton in a lake
ci could be expressed as mg l 1 or as mg l 1 of a focal
nutrient. ci,o is the concentration of the ith component at
thermodynamic equilibrium and n is the number of com
ponents. ci,o is of course a very small concentration (except
for i ¼ 0, which is considered to cover the inorganic com
pounds), corresponding to a very low probability of
forming complex organic compounds spontaneously in an
inorganic soup at thermodynamic equilibrium. ci,o is even
lower for the various organisms because the probability of
forming the organisms is very low with their embodied
information, here represented by the genetic code.
By using this particular exergy based on the same
system at the thermodynamic, chemical equilibrium as
reference, the eco exergy depends only on the chemical
potential of the numerous biochemical components that
are characteristic for life. It is consistent with Boltzmann’s
statement that life is a struggle for free energy. Eco
exergy has a definition close to the free energy, but unlike
free energy, eco exergy is not a state variable. It will
depend on the reference state that will vary from
ecosystem to ecosystem. Furthermore, it is difficult to
apply the classic state variables in thermodynamics far
from thermodynamic chemical equilibrium. Classic
thermodynamics presumes that the system is close to
equilibrium, which makes it possible to show that for
instance free energy is a state variable that gives the
same result independent on the pathway. We want to
use eco exergy far from thermodynamic equilibrium
and can therefore not use free energy in this context.
As we know that ecosystems due to the throughflow
of energy have the tendency to move away from
thermodynamic equilibrium losing entropy or gaining
exergy and information, we can at this stage formulate
the following proposition of relevance for ecosystems:
‘ecosystems attempt to develop toward a higher level of
exergy’.
Exergy and Information
Information means ‘acquired knowledge’. The thermody
namic concept of exergy is closely related to information.
A high local concentration of a chemical compound, for
131
instance, with a biochemical function that is rare else
where, carries exergy and information. On more complex
levels, information may still be strongly related to exergy
but in more indirect ways. Information is also a conveni
ent measure of physical structure. A certain structure is
chosen out of all possible structures and defined within
certain tolerance margins.
It is possible to distinguish between the exergy of
information and the exergy of biomass. pi defined as
ci/A, where
A¼
n
X
ci
i 1
½8
is the total amount of matter in the system, is introduced
as new variable in eqn [7]:
Ex ¼ A RT
n
X
i 1
pi ln ðpi =pio Þ þ A ln A=Ao
½9
As A Ao, exergy becomes a product of the total biomass
A (multiplied by RT) and Kullback’s measure:
K ¼
n
X
i 1
pi ln ðpi =pio Þ
½10
where pi and pio are probability distributions, a posteriori
and a priori to an observation of the molecular detail of the
system. It means that K expresses the amount of informa
tion that is gained as a result of the observations.
If we observe a system, which consists of two connected
chambers (see Figure 3), then we expect the molecules to
be equally distributed in the two chambers, that is, p1 ¼ p2
is equal to 1/2. If, on the other hand, we observe that all
the molecules are in one chamber, then we get p1 ¼ 1 and
p2 ¼ 0. Let us presume that the chamber to the left con
tains 1 mol of a pure ideal gas, while the chamber to the
right is empty. If we open the valve between the two
chambers, the loss of eco exergy (and also of technological
exergy) for the system would be RT ln2 in accordance
with eqns [7], [9], and [10]. We could utilize at least a part
of the exergy by installation of a small propeller in the
valve. The system will by this process increase its entropy
R ln2. This is in accordance with eqn [8].
... .. .
... . . .. . .
. ...
Figure 3 The left chamber contains 1 mole of a pure ideal gas,
while the right chamber is empty. If we open the valve, the
system will loose eco-exergy (or technological exergy) RT ln2,
which we could utilize by installation of a propeller in the valve.
The entropy of the system will simultaneously increase by R ln2.
132
Exergy
Exergy and the Dissipative Structure
As an ecosystem is nonisolated, the entropy changes dur
ing a time interval, dt, can be decomposed into the
entropy flux due to exchanges with the environment,
and the entropy production due to the irreversible pro
cesses inside the system such as diffusion, heat
conduction, and chemical reactions. It can also be
expressed by use of exergy:
e
Ex=dt ¼ de Ex=dt þ di Ex=dt
½11
where d Ex/dt represents the exergy input to the system
and di Ex/dt is the exergy consumed (is negative) by the
system for maintenance, etc. e is used to indicate an
external source and i to indicate the internal exergy
change.
Equation [11] shows among other things that sys
tems can only maintain a nonequilibrium steady state
by compensating the internal exergy consumption
with a positive exergy influx (de Ex/dt > 0). Such an
influx induces order into the system. In ecosystems,
the ultimate exergy influx comes from solar radiation,
and the order induced is, for example, biochemical
molecular order. If de Ex > di Ex (the exergy con
sumption in the system), the system has surplus
exergy input, which may be utilized to construct
further order in the system, or as Prigogine calls it,
dissipative structure. The system will thereby move
further away from thermodynamic equilibrium.
Evolution shows that this situation has been valid for
the ecosphere on a long term basis. In spring and
summer, ecosystems are in the typical situation that
de Ex exceeds di Ex. If de Ex < di Ex, the system
cannot maintain the order already achieved, but will
move closer to the thermodynamic equilibrium, that
is, it will lose order. This may be the situation for
ecosystems during fall and winter or due to environ
mental disturbances.
the number of components. cio is very low for living
component because the probability that living compo
nents are formed at thermodynamic equilibrium is very
low. It implies that living components get a high eco
exergy. cio is not zero for organisms, but will correspond to
a very low probability of forming complex organic com
pounds spontaneously in an inorganic soup at
thermodynamic equilibrium. cio on the other hand is
high for inorganic components, and although cio still is
low for detritus, it is much higher than for living
component.
The exergy of structurally complicated material can
be estimated based on the elementary composition. This
has, however, the disadvantage that a higher organism and
a microorganism with the same elementary composition
will get the same exergy which is in complete disagree
ment with the lower probability to form a more complex
organism, that is, the lower concentration of cio in the
equation. The composition will not account for the con
tribution of Kullbach’s measure of information, which is
often the major part of the eco exergy, as it is shown
below.
The problem related to the assessment of cio has been
discussed and a possible solution proposed. For dead
organic matter, detritus, which is given the index 1, it
can be found from classical thermodynamics.
For the biological components, 2,3,4, . . . ,N, the prob
ability, pio, consists at least of the probability of producing
the organic matter (detritus), that is, p1o, and the prob
ability, pi,a, to find the correct composition of the enzymes
determining the biochemical processes in the organisms.
Living organisms use 20 different amino acids and each
gene determines in average the sequence of about 700
amino acids. pi,a, can be found from the number of permu
tations among which the characteristic amino acid
sequence for the considered organism has been selected.
It means that
pi;a ¼ a Ngi
How to Calculate Eco-Exergy of Organic
Matter and Organisms?
The following expression for what we could call the
ecological exergy per unit of volume has been presented;
see eqn [7]:
Ex ¼ RT
i n
X
i 0
ci lnðci =cio Þ ML – 1 T – 2
½12
where R is the gas constant, T is the temperature of the
environment, ci is the concentration of the ith component
expressed in a suitable unit, and cio, is the concentration of
the ith component at thermodynamic equilibrium and n is
½13
where a is the number of possible amino acids ¼ 20, N is
the number of amino acids determined by one gene
¼ 700, and gi is the number of non nonsense genes. The
following two equations are available:
pio ¼ p1o pi;a ¼ p1o a Ng p1o :20 700g
½14
The exergy contribution of the ith component can be
found by combining eqns [12] and [14]:
Ex ¼ RT ci ln ci =ðp1o a Ngc0o Þ ¼ ð1 1o Þci ci ln pi;a
¼ ð1 1o Þci ci ln ða Ng
iÞ
¼ 18:7ci þ 700ðln20Þ ci gi ML – 1 T – 2
½15
The total eco exergy can be found by summing up the
contributions originated from all components. The
Exergy
contribution by inorganic matter can be neglected as the
contributions by detritus and even to a higher extent from
the biological components are much higher due to an extre
mely low concentration of these components in the
reference system (thermodynamic equilibrium for the sys
tem). The contribution by detritus, dead organic matter, is
18.7 kJ g 1 times the concentration (in gram per unit of
volume) corresponding to the composition of detritus,
namely lipids, carbohydrates, and proteins mainly, while
the eco exergy of living organisms with approximations
consists of
Ex1chem ¼ 18:7 kJ g – 1 times the concentration ci
ðgram per unit of volumeÞ
and
Exibio ¼ RT ð700 ln20Þ ci gi ¼ RT 2100 gi ci
½16
R ¼ 8.314 J mole 1 and if we presume a molecular weight
of an average 105 for the enzymes, we obtain the follow
ing equation for Exibio at 300 K:
Exibio ¼ 0:0529 gi ci
½17
where the concentration now is expressed in g per unit
of volume and the exergy in kilojoules per unit of
volume.
For the entire system the eco exergy, Ex total ¼
exergy chemical þ exergy biological can be found as
Ex total ¼ 18:7
N
X
ci 0:0529
i 1
N
X
i 1
ci gi ML – 1 T – 2
½18
where g for detritus (i ¼ 1) of course is 0. Table 1 shows
the weighting factor, , which is introduced to be able to
cover the exergy for various organisms in the unit detritus
equivalent or chemical exergy equivalent:
Ex total ¼
N
X
i 1
i ci ðas
detritus equivalentÞ
½19
The calculation of eco exergy accounts for the chemi
cal energy in the organic matter as well as for the
(minimum) genetic information embodied in the living
organisms. The latter contribution is measured by the
extremely small probability to form the living compo
nents, for instance algae, zooplankton, fish, mammals,
etc., spontaneously from inorganic matter. Weighting fac
tors defined as the exergy content relatively to detritus
(see Table 1) may be considered quality factors reflecting
how developed the various groups of organisms are and to
what extent they contribute to the exergy due to their
content of information which is reflected in the computa
tion. The values in Table 1 are found on basis of latest
133
knowledge of the genome size and the complexity of
different organisms. A value of 2.0 means that the
eco exergy embodied in the organic matter and the infor
mation are equal. As the values in Table 1 are much
bigger than 2.0 (except for virus, where the value is
1.01) the information eco exergy is the most significant
part of the eco exergy of organisms.
The eco exergy due to the ‘fuel’ value of organic
matter (chemical energy) is about 18.7 kJ g 1 (compare
with coal: about 30 kJ g 1 and crude oil: 42 kJ g 1). It can
be transferred to other energy forms for instance mechan
ical work directly, and be measured by bomb calorimetry,
which requires destruction of the sample (organism),
however. The information eco exergy ¼ (
1) c is
taken care of by the control and function of the many
biochemical processes. The ability of the living system to
do work is contingent upon its functioning as a living
dissipative system. Without information eco exergy, the
organic matter could only be used as fuel similar to fossil
fuel. But due to the information eco exergy, organisms
are able to make a network of the sophisticated biochem
ical processes that characterize life. The eco exergy (of
which the major part is embodied in the information) is a
measure of the organization. This is the intimate relation
ship between energy and organization that Schrødinger
was struggling to find.
As calculated here, eco exergy is a result of evolution
and of what Elsasser calls re creativity to emphasize that
the information is copied and copied again and again in a
long chain of copies where only minor changes are intro
duced for each new copy. The energy required for the
copying process is very small, but it has of course required
a lot of energy to come to the ‘mother’ copy through the
evolution for instance from prokaryotes to human cells.
Kullback’s measure of information covers the gain in
information when the distribution is changed from pion to
pI. Note that K is a specific measure (per unit of matter).
Expressed by the Kullbach’s measure of information, we
get the following equation for eco exergy:
Ex organism ¼ cRTK
½20
is therefore RTK.
The total eco exergy of an ecosystem cannot be cal
culated exactly, as we cannot measure the concentrations
of all the components or determine all possible contribu
tions to exergy in an ecosystem. If we calculate the exergy
of a fox for instance, then the above shown calculations
will only give the contributions coming from the biomass
and the information embodied in the genes, but what is
the contribution from blood pressure, sexual hormones,
network interactions, etc.? These properties are at least
partially covered by the genes but is that the entire
story? We can calculate the contributions from the domi
nant components, for instance by the use of a model or
134
Exergy
Table 1 Eco-exergy of living organisms
Early organisms
Plants
Detritus
Virus
Minimal cell
Bacteria
Archaea
Protists (Algae)
Yeast
1.00
1.01
5.8
8.5
13.8
20
17.8
33
39
43
61
76
91
92
97
98
109
120
133
133
143
143
165
158
163
164
174
167
191
221
232
246
275
314
310
322
393
499
688
833
980
2127
2138
2145
2173
Fungi, molds
Rhodophyta
Prolifera,sponges
Mustard weed
Seedless vascular plants
Moss
Rice
Gymnosperms (incl. Pinus)
Flowering plants
Animals
Mesozoa, Placozoa
Protozoa, amoebe
Phasmida (stick insects)
Nemertina
Cnidaria (corals,sea anemones, jelly fish)
Gastroticha
Brachiopoda
Plathyhalminthes (flatworms)
Nematoda (round worms)
Annelida (leeches)
Gnathostomulida
Kinorhyncha
Rotifera (wheel animals)
Entoprocta
Insecta (beetles, flies, bees, wasps, bugs, ants)
Coleodiea (Sea squirt)
Lepidoptera (butterflies)
Crustaceans
Chordata
Molluska, bivalvia, gastropodea
Mosquito
Fish
Amphibia
Reptilia
Aves (birds)
Mammalia
Monkeys
Anthropoid apes
Homo sapiens
values ¼ exergy content relatively to the exergy of detritus (Jørgensen et al.).
measurements that cover the most essential components
for a focal problem. The ‘difference’ in eco exergy by
‘comparing’ two different possible structures (species
composition) is decisive here. Moreover, eco exergy
computations give always only relative values, as the
eco exergy is calculated relatively to the reference
system.
Eco exergy calculated using the above equations has
some clear shortcomings:
1. We have made some although minor approxima
tions in the equations presented above.
2. We do not know the genes in all details for all
organisms.
3. We calculate only in principle the eco exergy embo
died in the proteins (enzymes), while there are other
components of importance for the life processes. These
components are contributing less to the exergy than the
enzymes and the information embodied in the enzymes
control the formation of these other components, for
instance hormones. It can however not be excluded that
these components will contribute to the total exergy of the
system. The life processes are of course considered indir
ectly as the enzymes determine the life processes.
Exergy
4. We do not include the eco exergy of the ecological
network. If we calculate the exergy of models, the net
work will always be relatively simple and the contribution
coming from the information content of the network is
considerably less than the exergy contribution from the
organisms. The real ecological network may contribute
much more to the total exergy. When network models are
compared it may also be relevant to compare exergy of
different networks.
5. We will always use a simplification of the ecosys
tem, for instance by a model or a diagram or similar. This
implies that we only calculate the exergy contributions of
the components included in the simplified image of the
ecosystem. The real ecosystem will inevitably contain
more components which are not included in our
calculations.
It is therefore proposed to consider the eco exergy found
by these calculations as a ‘relative minimum eco exergy
index’ to indicate that there are other contributions to the
total exergy of an ecosystem, although they may be of
minor importance. In most cases, however, a relative
index is sufficient to understand the reactions of ecosys
tems because the absolute exergy content is irrelevant for
the reactions. In most cases, the change in eco exergy is of
importance to understand ecological responses.
The weighting factors presented in Table 1 have been
applied successfully to calculate eco exergy applied as an
indicator to assess ecosystem health and in several struc
turally dynamic models to express the model goal
function, and furthermore in many illustrations of the
maximum eco exergy principle, that is presented below.
Structural dynamic models are able to take a shift of
species composition into account: which combinations of
properties are able to offer most survival? Further infor
mation about structural dynamic models is given in
Structural Dynamic Models. The relatively good results
in applying the weighting factors in this context, in spite
of the uncertainty of their assessment, seems only to be
explicable by the robustness of the application of the
factors in modeling and other quantifications. The differ
ences between the factors of the microorganism, the
vertebrates, and invertebrates are so clear that it does
not matter if the uncertainty of the factors is very high –
the results are influenced slightly.
On the other hand, from a theoretical point of view it
would be an important progress to get better weighting
factors but also because it would enable us to model the
competition between species which are closely related.
The key to find better values maybe the proteomes
(the total compositions of the proteins that as enzymes
determine the life processes). Our knowledge about the
composition of proteomes in various organisms is, how
ever, more limited than for the number of the genes.
135
Why Do Living Systems Have Such a High
Level of Eco-Exergy?
A frog of 20 g will have an eco exergy content of
20 18.7 688 kJ 257 MJ, while a dead frog will have
only an exergy content of 374 kJ, although they have the
same chemical composition, at least a few seconds after
the frog has died. The difference is rooted in the informa
tion or rather the difference in the useful information.
The dead frog has the information a few seconds after its
death (the amino acid composition of the proteins has not
yet been decomposed), but the difference between a live
frog and a dead one is the ability to utilize the enormous
information stored in the genes and the proteomes of the
frog.
The amount of information stored in a frog is really
surprisingly high. The number of amino acids placed in
the right sequence is about 200 000 000 and for each of
these 200 000 000 amino acids there are 20 possibilities.
This information is again repeated in billions of cells that
are cooperating to make up the frog. This enormous
amount of information is able to allow reproduction and
is transferred from generation to generation which
implies that the evolution can continue because what is
already a favorable combination of properties is con
served through the genes.
The information in living organisms applies ‘nano
technology’ in the sense that the weight of 200 000 000
amino acids is for an average amino acid molecular weight
of 125 g moles 1 2.5 1010 g/A ¼ 4 10 14 g, where A is
Avogadro’s number (A ¼ 6.2 1023). A book with the
same amount of information would weigh several hun
dreds of kilograms.
Because of the very high number of amino acids, about
200 000 000, it is not surprising that there will always be a
minor difference from frog to frog in the amino acid
sequence. It may be a result of mutations or of a minor
mistake in the copying process. These variations are
important because they give possibilities to ‘test’ which
amino acid sequence gives the best result with respect to
survival and growth. The best – representing the most
favorable combination of properties – will offer the high
est probability of survival and give the most growth and
the corresponding genes will therefore prevail. Survival
and growth mean more exergy, resulting in a bigger
distance from thermodynamic equilibrium. Exergy could
therefore be used as a thermodynamic function which
could be used to quantify Darwin’s theory. In this context,
it is interesting that it has been demonstrated that eco
exergy also represents the amount of energy needed to
tear down the system. It means that the more exergy the
system possesses the more difficult it becomes to degrade
the system and the higher is therefore the probability of
survival. Consequently, eco exergy can be applied as a
136
Exergy
measure of sustainability. The crucial question is there
fore: do we hand over the Earth to our children and
grandchildren with the same distance from the thermo
dynamic equilibrium, that is, the same exergy, as we
received it from our ancestors?
also ash, but let us not consider it in our calculations), the
exergy loss due to the dispersion can be determined by
the following calculations:
Losses and Gains of Eco-Exergy by
Human Activities Included Pollution
where 50 10 9 and 4 10 4 represents concentrations
(expressed as ratios, i.e., no units) of sulfur dioxide S and
carbon dioxide C in a typical town atmosphere. The
chemical exergy content of 1 g coal is about 32 kJ. The
loss of exergy by dispersion is therefore only 5% of the
loss directly of chemical exergy by burning coal. As all
our calculations will have a higher uncertainty than 5%,
and the quality of coal may vary more than 5%, it seems
acceptable not to include the dispersion exergy loss by
use of fossil fuel or as alternative to multiply all exergy
losses due to our consumption of fossil fuel by a factor
of 1.05 to compensate approximately for the exergy
loss due to the dispersion of the formed gases in the
atmosphere.
The deterioration of ecosystems. By the use of eqn [20] it is
possible to find the eco exergy of an ecosystem or rather
of the ecosystem corresponding to our model of the
ecosystem. Consequently, the loss of eco exergy due to
deterioration and pollution of ecosystems can be found by
calculation of the eco exergy before and after deteriora
tion. The difference will directly yield the loss.
The use of renewable resources. The formation of renew
able resources are found separately by multiplication of
the annual consumption of the various resources by the
exergy content of each renewable resource. If, for
instance, the annual fishery in the North Sea has the last
many years been in the order of 100 000 t, which implies
that the eco exergy of the North Sea has been
reduced 1011 499 18.7 kJ ¼ 9.3 1017 J, then 499 is
the value for fish (Table 1). Sustainability requires
that the growth of fish biomass compensate for this loss
of eco exergy. It has, unfortunately, for a couple of dec
ades in many marine ecosystems, including the North
Sea, not been the case due to over fishing.
Dispersion of waste. This to a certain extent can be
calculated parallel to eqn [21]. This is often named the
external costs of our activities including the industrial
and agricultural activities, but it is actually as the other
four points just a question about the loss of eco exergy.
It is not surprising that the cost of treating waste is
increasing as the environmental agencies require a
more and more complete elimination of these exergy
losses. Or expressed differently: we are coming closer
and closer to the carrying capacity of the Earth for
man made production. In this context it should not be
forgotten that also the treatment of waste costs eco
exergy.
When contaminants, for instance heavy metals, are
widely dispersed, eco exergy is lost. When leaded gaso
line was used to obtain a higher octane number, on the
order of 400 000 t of lead were dispersed annually around
the globe. Lead was even found in the ice pack of
Greenland! A typical concentration in lead ore is about
5% or 0.05 kg kg 1ore, while a typical concentration in
the environment after the dispersion is 1 mg kg 1 soil. If
we presume 300 K, the annual eco exergy lost can be
found by eqn [7] as
Ex lost ¼ 8:314 0:300 4 1011 =207
ln 0:05=10 – 9 85 000 GJ yr – 1
½21
where 207 is the atomic weight of lead. The consumption
of lead has decreased due to shifts to other additives in the
gasoline in most countries. This loss of eco exergy by the
use of lead as additive to gasoline is therefore today only
estimated to be around 40 000 GJ yr 1.
‘The loss of exergy due to dispersion of resources in
general’ can be calculated parallel to the application of
eqn [7] as shown in eqn [21]. In addition to lead (only the
dispersion of lead as gasoline additive is considered in this
context), the loss of exergy by dispersion of other non
renewable resources is shown in Table 2.
The ‘loss of eco exergy due to the consumption of
fossil fuel’ is found by addition of the chemical free
energy (the work capacity) of the fossil fuel and the loss
of eco exergy due to the dispersion of the gases resulting
from the chemical processes. The exergy loss due to
dispersion of the components of fossil fuel is found by
the following calculations: if we consider 1 g of coal that
contains 1% of sulfur and 99% of carbon (coal contains
Table 2 Loss of eco-exergy due to dispersion of nonrenewable
mineral resources
1
Element
GJ yr
Chromium
Nickel
Zinc
Copper
Mercury (included fossil fuel)
Lead (today)
32 000
15 000
80 000
18 000
27 000
40 000
Calculations based upon principles shown in eqn [21].
0:01ð8:314 0:300=32Þ ln 0:01=50 10 – 9 Þ
þ 0:99ð8:314 0:300=12Þ ln 0:99=4 10 – 4 Þ
¼ 1617 J 1:6 kJ
Exergy
Consumption of nonrenewable fuel, including fossil fuel and
nuclear fuel. The annual loss of eco exergy is found by
multiplication of the exergy content and the annual
consumption.
We calculated above the loss of eco exergy by disper
sion of waste due to consumption of nonrenewable
resources. This is, however, not the entire story, as energy
is required in producing various materials from ore and
raw source materials. The energy requirement when the
material is produced from scrape is included. As it can be
seen, the energy requirement is less when scrap if used for
the production. Reuse and recycling gives therefore dou
ble benefits: we save to draw on the limited resources and
we save energy. Notice, particularly for aluminum, that
the energy requirement by the use of scrap instead of ore
is considerable. As energy consumption explains one of
our major losses of eco exergy, the latter benefit is of
great importance.
Formulation of a Thermodynamic
Hypothesis for Ecosystems
If an (open, nonequilibrium) ecosystem receives a bound
ary flow of energy from its environment, it will use what it
can of this energy, the free energy or the exergy content,
to do work. The work will generate internal flows, leading
to storage and cycling of matter, energy, and information,
which move the system further from equilibrium. Self
organizing processes get started. This is reflected in
decreased internal entropy and increased internal
organization.
The open question of this section is which of many
possible pathways will an ecosystem take in realizing its
three forms of growth? The answer given is that an
ecosystem will change in directions that most consistently
create additional capacity and opportunity to achieve
increasing deviation from thermodynamic ground,
that is, the exergy stored in the ecosystem will increase.
Abundant and diverse living biomass represents abundant
and diverse departure from thermodynamic equilibrium,
and both are captured in this parameter. If multiple
growth pathways are offered from a given starting state,
those producing greatest exergy storage will tend to be
selected, for these in turn require greatest energy
dissipation to establish and maintain, consistent with the
second law. Energy storage by itself is not sufficient, but it
is the increase in specific exergy, that is increased
exergy/energy ratio, that reflects improved usability,
and this represents the increasing capacity to do the
work required for living systems to continuously evolve
new adaptive ‘technologies’ to meet their changing
environments.
These considerations lead to a thermodynamic hypo
thesis which is able to explain the growth and
137
development of ecosystems and the reactions of ecosys
tems to perturbations: ‘‘If a system receives an input of
exergy, it will utilize this exergy after the maintenance of
the system far from thermodynamic has been covered to
move the system further from thermodynamic equili
brium If more than one pathway to depart from
equilibrium is offered, the one yielding the most gradi
ents, and most exergy storage (dEx/dt is maximum) under
the prevailing conditions, to achieve the most ordered
structure furthest from equilibrium, will tend to be
selected.’’
Just as it is not possible to prove the first three laws of
thermodynamics by deductive methods, so can the above
hypothesis only be ‘proven’ inductively. In the next sec
tion we do examine a number of concrete cases which
contribute in a general way to the weight of evidence in
favor.
This tentative law may be considered a translation
of Darwin’s theory into thermodynamics. Exergy mea
sures survival: the biomass and the network,
information, and organization that imply that the
resources are used in the best possible way to gain
most survival. The question is which of the possible
combinations of properties by the entire spectrum of
organisms in an ecosystem will be able to store most
exergy (obtain most survival)? The organisms with the
properties that make it possible to gain most survival
(exergy) will win in accordance to Darwin and in
accordance to the tentative fourth law of thermody
namics. Notice that the resources are always limited
relatively to the number of possible offsprings.
Therefore there will always be a competition about
the resources – and this competition explains together
with the huge variation of properties even by the same
species, that an evolution has taken place.
Support to the Maximum Eco-Exergy
Hypothesis
Eight supporting arguments for the hypothesis presented
above. More evidence has been provided; but the eight
supporting evidences presented here give a good idea of
the theoretical support for the hypothesis.
1. The exergy storage hypothesis might be taken
as a generalized version of ‘Le Chatelier’s Principle.’
Biomass synthesis can be expressed as a chemical reaction:
Energy þ nutrients ¼ molecules with more free energy
ðexergyÞ and organization
þ dissipated energy
According to Le Chatelier’s Principle, if energy is put into
a reaction system at equilibrium the system will shift its
equilibrium composition in a way to counteract the
138
Exergy
change. This means that more molecules with more free
energy and organization will be formed. If more pathways
are offered, those giving the most relief from the distur
bance (using most of the inflowing energy) by using the
most energy, and forming the most molecules with the
most free energy, will be the ones followed in restoring
equilibrium.
2. The sequence of organic matter oxidation takes place
in the following order: by oxygen, by nitrate, by
manganese dioxide, by iron (III), by sulfate, and by carbon
dioxide. This means that oxygen, if present, will
always outcompete nitrate which will outcompete
manganese dioxide, etc. The amount of exergy stored as a
result of an oxidation process is measured by the
available kJ mole 1 electrons which determine the number
of adenosine tri phosphate molecules (ATPs) formed. ATP
represents an exergy storage of 42 kJ mole 1. Usable energy
as exergy in ATPs decreases in the same sequence as
indicated above. This is as expected if the exergy storage
hypothesis was valid (Table 3). If more oxidizing agents are
offered to the system, the one giving the resulting system
the highest storage of free energy will be selected.
3. Numerous experiments have been performed to imi
tate the formation of organic matter in the primeval
atmosphere on Earth 4 billion years ago. Energy from
various sources were sent through a gas mixture of carbon
dioxide, ammonia, and methane. Analyses showed that a
wide spectrum of compounds, including several amino
acids contributing to protein synthesis, is formed under
these circumstances. There are obviously many pathways
to utilize the energy sent through simple gas mixtures, but
mainly those forming compounds with rather large free
energies (high exergy storage, released when the com
pounds are oxidized again to carbon dioxide, ammonia,
and methane) will form an appreciable part of the mixture.
4. There are three biochemical pathways for photo
synthesis: (1) the C3 or Calvin–Benson cycle, (2) the C4
pathway, and (3) the Crassulacean acid metabolism
(CAM) pathway. The latter is least efficient in terms of
the amount of plant biomass formed per unit of energy
received. Plants using the CAM pathway are, however,
able to survive in harsh, arid environments that would be
inhospitable to C3 and C4 plants. CAM photosynthesis
will generally switch to C3 as soon as sufficient water
becomes available. The CAM pathways yield the highest
biomass production, reflecting exergy storage, under arid
conditions, while the other two give highest net produc
tion (exergy storage) under other conditions. While it is
true that one gram of plant biomass produced by each of
the three pathways has different free energies, in a general
way improved biomass production by any of the pathways
can be taken to be in a direction that is consistent, under
the conditions, with the exergy storage hypothesis.
5. Givnish and Vermelj observed that leaves optimize
their size (thus mass) for the conditions. This may be
interpreted as meaning that they maximize their free
energy content. The larger the leaves the higher their
respiration and evapotranspiration, but the more solar
radiation they can capture. Deciduous forests in moist
climates have a leaf area index (LAI) of about 6%. Such
an index can be predicted from the hypothesis of highest
possible leaf size, resulting from the tradeoff between
having leaves of a given size versus maintaining leaves
of a given size. Size of leaves in a given environment
depends on the solar radiation and humidity regime, and
while, for example, sun and shade leaves on the same
plant would not have equal exergy contents, in a general
way leaf size and LAI relationships are consistent with the
hypothesis of maximum exergy storage.
6. The general relationship between animal body
weight, W, and population density, D, is D ¼ A/W,
where A is a constant. Highest packing of biomass depends
only on the aggregate mass, not the size of individual
organisms. This means that it is biomass rather than
population size that is maximized in an ecosystem, as
density (number per unit area) is inversely proportional
to the weight of the organisms. Of course the relationship
is complex. A given mass of mice would not contain the
same exergy or number of individuals as an equivalent
weight of elephants. Also, genome differences (example 1)
and other factors would figure in. Later we will discuss
exergy dissipation as an alternative objective function
proposed for thermodynamic systems. If this were maxi
mized rather than storage, then biomass packing would
follow the relationship D ¼ A/W 0.65 0.75. As this is not the
case, biomass packing and the free energy associated with
this will lend general support for the exergy storage
hypothesis.
Table 3 Yields of kJ and ATPs per mole of electrons, corresponding to 0.25 moles of CH2O oxidized
Reaction
kJ/(mol e )
ATPs/(mol e )
CH2O þ O2 $ CO2 þ H2O
CH2O þ 0.8 NO3 þ 0.8 Hþ $ CO2 þ 0.4 N2 þ 1.4 H2O
CH2O þ 2 MnO2 þ Hþ $ CO2 þ 2 Mn2þ þ 3 H2O
CH2O þ 4 FeOOH þ 8 Hþ $ CO2 þ 7 H2O þ Fe2þ
CH2O þ 0.5 SO42 þ 0.5 Hþ $ CO2 þ 0.5 HS þ H2O
CH2O þ 0.5 CO2 $ CO2 þ 0.5 CH4
125
119
85
27
26
23
2.98
2.83
2.02
0.64
0.62
0.55
The released energy is available to build ATP for various oxidation processes of organic matter at pH ¼ 7.0 and 25 C.
Exergy
2.5
R
+
2.0
+
+
+
1.0
+
+
0.5
0.5
1.0
1.5
2.0
log (N /P )
Figure 4 Log–log plot of the turnover rate ratio of nitrogen to
phosphorus, R, at maximum exergy versus the logarithm of the
nitrogen/phosphorus ratio, log N/P. The plot is consistent with
Vollenweider (1975).
7. If a resource (for instance, a limiting nutrient for
plant growth) is abundant, it will typically recycle faster.
This is a little strange, because a rapid recycling is not
needed when a resource is nonlimiting. A modeling study
indicated that free energy storage increases when an
abundant resource recycles faster. Figure 4 shows such
results for a lake eutrophication model. The ratio, R, of
nitrogen (N) to phosphorus (P) cycling which gives the
highest exergy is plotted versus log (N/P). The plot in
Figure 4 is also consistent with empirical results. Of
course, one cannot ‘inductively test’ anything with a
model, but the indications and correspondence with data
do tend to support in a general way the exergy storage
hypothesis.
8. Dynamic models whose structure changes over time
are based on nonstationary or time varying differential or
difference equations. We will refer to these as ‘structu
rally dynamic models’. A number of such models, mainly
of aquatic systems, have been investigated to see how
structural changes are reflected in free energy changes.
The latter were computed as exergy indexes.
Time varying parameters were selected iteratively to
give the highest exergy index values in a given situation
at each time step. Changes in parameters, and thus system
139
structure, not only reflect changes in external boundary
conditions, but also mean that such changes are necessary
for the ongoing maximization of exergy. For all models
investigated along these lines, the changes obtained were
in accordance with actual observations (see references).
These studies therefore affirm, in a general way, that
systems adapt structurally to maximize their content of
exergy. It is noteworthy that Coffaro et al., in their struc
tural dynamic model of the Lagoon of Venice, did not
calibrate the model describing the spatial pattern of var
ious macrophyte species such as Ulva and Zostera, but used
exergy index optimization to estimate parameters deter
mining the spatial distribution of these species. They
found good accordance between observations and
model, as was able by this method ‘without’ calibration
to explain more than 90% of the observed spatial distri
bution of various species of Zostera and Ulva.
See also: Fundamental Laws in Ecology.
Further Reading
Fath B, Jørgensen SE, Patten BC, and Strakraba M (2004) Ecosystem
growth and development. BioSystem 77: 213 228.
Jørgensen SE (2002) Integration of Ecosystem Theories: A Pattern, 3rd
edn., 432pp. Dordrecht, The Netherlands: Kluwer Academic
Publishing Company (1st edn. 1992, 2nd edn. 1997).
Jørgensen SE and Fath B (2004) Application of thermodynamic
principles in ecology. Ecological Complexity 1: 267 280.
Jørgensen SE and Svirezhev YM (2004) Towards a Thermodynamic
Theory for Ecological Systems, 366pp. Oxford: Elsevier.
Jørgensen SE, Patten BC, and Strakraba M (2000) Ecosystems
Emerging: 4. Growth. Ecological Modelling 126: 249 284.
Jørgensen SE, Ladegaard N, Debeljak M, and Marques JC (2005)
Calculations of exergy for organisms. Ecological Modelling
185: 165 176.
Morowitz HJ (1968) Energy Flow in Biology. Biological Organisation as a
Problem in Thermal Physics, 179pp. New York: Academic Press.
(See also the review by Odum HT (1969) Science 164: 683 84).
Schrødinger E (1944) What is Life? Cambridge: Cambridge University
Press.
Svirezhev YM (2001) Thermodynanics and theory of stability.
In: Jørgensen SE (ed.) Thermodynamics and Ecological Modelling,
pp. 117 132. Boco Raton, FL: CRC Press, LLC.
Ulanowicz RE (1986) Growth and Development. Ecosystems
Phenomenology, 204pp. New York, Berlin, Heidelberg, Tokyo:
Springer.
140
Overview of Ecosystem Types, Their Forcing Functions, and Most Important Properties
Overview of Ecosystem Types, Their Forcing Functions,
and Most Important Properties
S E Jørgensen, Copenhagen University, Copenhagen, Denmark
ª 2009 Elsevier B.V. All rights reserved.
The 39 ecosystems that are described in the part
Ecosystems may be classified into four groups according
to their forcing functions.
Class I consists of ecosystems that are completely or
almost completely managed by man. This class encom
passes agriculture systems, biological waste water
systems, botanical gardens, green houses, microcosms
and mesocosms, landfills, forest plantations, urban sys
tems, and wind shelterbelts. The properties of these
nine types of ecosystems are to a large extent deter
mined by the management strategy and plan. The
properties of agriculture systems are for instance very
much dependent on the management plan: is the sys
tem an industrialized agriculture system as it is known
in the industrialized world, an integrated agriculture
system as it is used in China and to a lesser extent in
Europe, or is the agriculture based on the principles of
organic farming. The last type of agriculture is increas
ingly practiced in industrialized countries, although the
percentage of organic farming in countries where it is
practiced mostly is only slightly more than 10%.
Class II comprises ecosystems that are highly dependent on
human activities, which may cause pollution and dete
riorate water quality, soil quality, or the quality of other
important ecological properties. In other words, class II
ecosystems are usually strongly affected by the impacts
of pollution. Ecosystems that belong to this class are
estuaries, floodplains, freshwater lakes, freshwater
marshes, lagoons, mangrove wetlands, Mediterranean
ecosystems, riparian wetlands, rivers and streams, salt
marshes, and temporary waters. Discharge of pollutants
in these ecosystems is usually – but not necessarily
always – the most important forcing function.
Class III encompasses ecosystems where pollution may play
a role while climate change may also be an important
forcing function. For a few of the ecosystem types in this
class, the conservation of the ecosystems and their func
tion may be very important globally. It is the case for the
rain forests and the temperate forests, which are ecosys
tems of importance for the global ecological balance, the
global carbon cycle, and the climate. These two types of
ecosystems may not be threatened by pollution but it is
important that the areas occupied by forests – rain forests
as well as temperate forests – are not reduced because of
the significant role of these ecosystems in the global
ecological cycling of carbon and nutrients. There is of
course a gradual transition from class II to class III, usually
dependent on how important the discharge of pollutants is
as a forcing function for the ecosystem. The following
ecosystems belong to this class: boreal forests, chaparrals,
peatlands, savannas, swamps, steppes and prairies, rain
forests, temperate forests, and upwelling ecosystems.
Class IV includes ecosystems that are very little
influenced by man and it has been possible up to now
to maintain them in almost entirely natural conditions.
Some of these ecosystems have extreme conditions; for
example, caves have no or almost no light and polar
terrestrial ecosystems are characterized by relatively low
temperatures. This class includes the following ecosys
tems: alpine ecosystem, alpine forest, caves, desert
streams, dunes, polar terrestrial ecosystems, rocky inter
tidal ecosystems, saline and soda lakes, and tundra.
For the four classes of ecosystems, the properties of the
ecosystems determine whether the ecosystems can
remain in ecologically healthy conditions in spite of the
forcing functions. In Table 1, the four ecosystem types
and their characteristic forcing functions are listed
together with the ecosystem properties that are most
important for the maintenance of a good ecosystem
health. The ecosystem properties mentioned in the table
are the properties that must be kept at a stable level to
ensure that the ecosystem health is not deteriorated. It is
therefore the ecosystem properties that should be cur
rently recorded to follow the development of the
ecosystems as a result of the forcing functions.
The classification of the 39 ecosystem types is as
follows:
Class I: agriculture systems, biological waste water
systems, botanical gardens, green houses, microcosms
and mesocosms, landfills, forest plantations, urban sys
tems, and wind shelterbelts.
Class II: estuaries, floodplains, freshwater lakes, freshwater
marshes, lagoons, mangrove wetlands, Mediterranean
ecosystems, riparian wetlands, rivers and streams, salt
marshes, and temporary waters.
Class III: boreal forests, chaparrals, coral reefs, peatlands,
savannas, swamps, steppes and prairies, rain forests,
temperate forests, and upwelling ecosystems.
Class IV: alpine ecosystem, alpine forest, caves, desert
streams, deserts, dunes, polar terrestrial ecosystems,
rocky intertidal ecosystems, saline and soda lakes, and
tundra.
Overview of Ecosystem Types, Their Forcing Functions, and Most Important Properties
141
Table 1 Ecosystem classes according to their forcing functions with specification of the most important ecosystem properties
Ecosystem class
Forcing functions
Important properties
I1
II2
III3
IV
Almost completely managed by man
Discharge of pollutants (and climate)
Discharge of pollutants, climate changes
Almost only natural forcing functions
Determined by the management
Buffer capacity, diversity, adaptation
Buffer capacity, ability to adapt and recover
Buffer capacity, diversity, health of the ecological network.
Often vulnerable due to extreme conditions
Further Reading
Jørgensen SE (2004) Information theory and energy. In: Cleveland CJ
(ed.) Encyclopedia of Energy, vol. 3. pp. 439 449. San Diego, CA:
Elsevier.
Jørgensen SE (2006) Eco Exergy as Sustainability. 220pp.
Southampton: WIT Press.
Jørgensen SE (2008b) Evolutionary Essays. A Thermodynamic
Interpretation of the Evolution, 210pp.
Jørgensen SE (ed.) (2008a) Encyclopedia of Ecology, 5 vols. 4122pp,
Amsterdam: Elsevier.
Jørgensen SE and Fath B (2007) A New Ecology. Systems
Perspectives. 275pp. Amsterdam: Elsevier.
Jørgensen SE, Patten BC, and Straskraba M (2000) Ecosystems
emerging: 4. growth. Ecological Modelling 126: 249 284.
Jørgensen SE and Svirezhev YM (2004) Towards a
Thermodynamic Theory for Ecological Systems. 366pp.
Amsterdam: Elsevier.
Ulanowicz R, Jørgensen SE, and Fath BD (2006) Exergy, information
and aggradation: An ecosystem reconciliation. Ecological Modelling
198: 520 525.
This page intentionally left blank
ECOSYSTEMS
This page intentionally left blank
Agriculture Systems
O Andrén, TSBF-CIAT, Nairobi, Kenya
T Kätterer, Department of Soil Sciences, Uppsala, Sweden
ª 2008 Elsevier B.V. All rights reserved.
Introduction
The Agroecosystem
Further Reading
Introduction
27% of the total land area is under permanent pasture,
grazed by cattle, goats, sheep, camels, etc. Clearly, we are
actively managing a considerable part of our planet for
agricultural purposes, and to this one can add other simi
lar systems, such as intensively managed forest systems
(planted and harvested, sometimes fertilized), etc.
Ecological research performed in agricultural systems
has many advantages compared with research in most
natural ecosystems. For example, there are a number of
long term field experiments running, although originally
designed for, for example, crop production response to
fertilizer dose, that can give us a 30 year integration of
what has happened, for example, to organisms in the soil
under different conditions. Further, agricultural fields are
‘homogenized’, that is, trees, larger stones, etc. are
removed and regular soil cultivation evens out differences
in topsoil properties over time. However, even after many
years of cultivation, a fairly high variability in soil proper
ties remain, which is the incentive for ‘precision farming’,
where soil and crop properties are measured at high
resolution (m2), and management is based on these meas
urements. For ecological research, this is an opportunity,
since any given hectare will yield numerous observation
points, each helping us to answer questions such as: Why
does this particular location yield more wheat, or why is
more water present at that location?
Another advantage is that agricultural crops often have
a short lifespan and a small size, compared with, for
example, forest trees. Often, an experiment can be started
when the soil is bare, and a single crop can be followed
from sowing, through harvest, and finally when the stub
ble is plowed down at the end of the growing season. This
life cycle can take a century for a tree in a northern forest,
which, to add insult to injury, also may contain several
other plant species. Therefore it is not surprising that a
considerable part of modern ecological theory (predator–
prey interactions, general soil ecology, above and below
ground plant growth dynamics, organic matter
decomposition, nutrient mineralization, etc.) is based on
work performed in agricultural land, and that the reluc
tance of ecologists to work in agricultural systems that
was obvious 30 years ago seems to have vanished.
An agricultural ecosystem is an ecosystem managed with
a purpose. This purpose usually is to produce crops or
animal products. Agricultural ecosystems are designed by
humans, and current agroecosystems are products of a
long chain of experimental work. These experiments
have been performed by individual farmers as well as
research institutions, and when results were positive for
the purpose, the methods have been adopted.
The purpose has, however, changed with time. In highly
productive regions, for example, Western Europe, the
emphasis has changed from maximum productivity to envir
onmental considerations, such as reduction of nutrient losses
to groundwater and maintaining an open landscape with
high biodiversity, etc. In less productive regions, where
resources such as water or fertilizers are scarce and produc
tion is too low to properly feed the farmer, environmental
considerations have low priority. This is a major global
problem, since this leads to land degradation and even
lower production, etc. in a downward spiral.
Agroecosystems are conceptually fairly similar to man
aged forests and grasslands, and whether extensively cattle
grazed natural grasslands should be included under the
category of agroecosystems is a matter of choice in the
individual case. Arable land is defined as land that is soil
cultivated regularly, but also here the boundaries are not
sharp (seminatural grasslands, permanent crops, etc.). At
the other end, agroecosystems border horticultural sys
tems, that is, vegetable cropping. Alternatively,
horticulture can be viewed as a subset of agriculture.
Production of cabbage in a field can be considered as
agriculture, but hydroponic (soil less) production of toma
toes in a greenhouse under artificial light can perhaps not
be included. However, in many respects even an artificial
ecosystem such as this can be considered as an agricultural
ecosystem. It is designed for production of a crop and is just
managed to a higher extent than an arable field.
According to FAO statistics for 2002, agricultural eco
systems comprise almost 40% (5 Gha) of the total land
area of the Earth. About 11% of the total land area is
arable land (cultivated with crops), and approximately
145
146
Agriculture Systems
The Agroecosystem
Figure 1 is an attempt to summarize the characteristics of
an agroecosystem as compared to most natural ecosys
tems. Note that this comparison is between a typical
natural ecosystem and a typical, high production
agroecosystem.
Abiotic Constraints
Just like natural ecosystems, agroecosystems are con
strained by climate and soil properties – maize does not
grow in Northern Sweden. However, climate can be
modified, that is, in dry climates one can irrigate (with
surface or groundwater), and soil properties can be
modified through, for example, liming, organic matter
amendments, and fertilization. Too high water tables
can be lowered through ditching or tile draining.
Nutrients
Highly productive agroecosystems need high inputs of
plant nutrients (nitrogen, phosphorus, potassium, and
other elements) to replenish the nutrients removed with
the exported products. These inputs can be delivered
either as commercial fertilizer, recirculated sewage sludge
and ash from garbage burning, or manure from cattle,
pigs, poultry, etc. All sources have their advantages and
disadvantages. Commercial fertilizers are well defined,
low in pollutants such as heavy metals (although excep
tions exist), hygienically safe, and are concentrated, easy
to transport, and rapidly available to the plant when
applied in the field. However, production and long
range transport of fertilizers is energy consuming, and
the concentrated product increases the risk for too high
doses, leading to environmental pollution. An even
greater problem is that a large part of the farmers of the
world cannot afford to buy enough fertilizer to maintain
soil fertility and obtain good yields. In all, world N ferti
lizer production in 2001 was slightly less than 90 Mt, very
unevenly distributed. In sub Saharan Africa, only 1.1 kg
fertilizer nitrogen is used per person and year, whereas in
China the corresponding value is 22 kg.
In theory, recirculation of nutrients from waste of the
exported products seems to be ecologically sound. In
practice, there are a number of problems. First, sewage
sludge mainly consists of water, which either must be
removed (requires energy) or transported, which is
expensive and impractical. Second, sewage sludge con
tains harmful bacteria, human parasites, etc. and has to
undergo hygienic treatment. Third, and most severe, is
the problem with contaminants, such as heavy metals and
organic toxins. Therefore recirculation of sewage sludge
and garbage incineration ash is strictly regulated in most
countries. In this perspective, replacement of nutrients
using newly produced fertilizer can be a better solution
from an environmental viewpoint.
Naturally, animal manure produced on the farm
should be and is recycled to soil as much as possible.
Compared with fertilizers, manure has the advantage of
containing organic matter, which improves soil structure.
On the other hand, manure contains mostly water (expen
sive storage and transportation, heavy machinery needed
for spreading), and it will lose nitrogen through ammonia
emission, both at storage and spreading.
Crops, Varieties, and Cropping Systems
The vegetation found in an agroecosystem is usually
divided into crop and weeds, where weeds are unwanted
Natural
Solar energy
Rainfall
Nutrient input
Seed input
Pesticide input
Migration
Human control
Plant biodiversity
Soil biodiversity
Plant production
Soil nutrient status
Soil cultivation
Decomposition rates
Primary consumers Plant disease
Potential for nutrient loss
Solar energy
Rainfall
Nutrient input
Seed input
Pesticide input
Migration
Human control
Plant biodiversity
Soil biodiversity
Plant production
Soil nutrient status
Soil cultivation
Decomposition rates
Primary consumers Plant disease
Potential for nutrient loss
Agricultural
Carbohydrates
proteins
Other ecosystem
services
Carbohydrates
proteins
Other ecosystem
services
Figure 1 Similarities and differences between typical natural and high-production agricultural ecosystems. Inputs of energy, mass,
and control (left), comparison of selected ecosystem properties (center), outputs (right). Note that cattle, etc. are not included as primary
consumers here. (Bold markedly higher value than in the other ecosystem type. Italic very low.)
Agriculture Systems 147
trespassers, which traditionally have been regarded only
as negatives. More recently, this view has been modified,
and weeds, particularly weedy border zones can be
accepted to some extent, as biodiversity enhancers and
refuges, for example, for beetles.
The crops used today are products of many years (in
some cases millennia) of plant breeding, and properties
selected for are usually productivity, product quality, pest
resistance, etc. This directed selection, in recent years
augmented by direct manipulation of DNA, is one of
the main differences between agro and natural ecosys
tems. Crop species and varieties are being redistributed
all over the world; maize, a staple food in Africa, comes
from Central America, common West European and
North American cereals such as wheat come from the
Middle East, etc. This breeding and distribution of
improved crops, together with improved cultivation/
fertilization techniques probably is the main reason for
the global success of the human species (three billion in
1960, probably nine billion in 2050). For example, world
grain production was 631 Mt in 1950, and in 2000 it had
increased to 1840 Mt.
Herbicides, Pesticides, and Fungicides
To reach the goal of high production of crops of good
quality, weeds (unwanted plants), pests (unwanted ani
mals), as well as fungal, bacterial, and viral diseases must
be kept in check. A monoculture crop is vulnerable to
attacks, since one (or a pair) of the pests that enter a field
will have a high concentration of food with no transport
stretches in between. Potential predators may be absent,
since they may need a litter layer on the ground for
reproduction, which does not exist in the field, etc.
Repeated monocultures may build up specialized pests,
such as plant parasitic nematodes. Crop rotations (switch
ing crops from year to year according to a predetermined
pattern) can successfully deal with many pests and dis
eases, and careful soil cultivation can reduce weed
problems. Intercropping (growing two or more crops
together, such as barley/clover) may also help.
However, most fields will benefit from occasional
chemical (or biological) pesticide/herbicide treatment.
These types of agrochemicals have a somewhat dubious
reputation among laymen and perhaps also ecologists
(DDT, Agent Orange, mercury, etc.). Three things should
be kept in mind, though. First, the substances and formu
lations used today are thoroughly tested before approval,
and their side effects and the fate of their decomposition
products are well known. Second, chemical warfare is
common in natural systems – all successful plant species
present today have at least some chemical defense against
microorganisms and pests. Third, which alternatives do
we have? A failed crop in a well fertilized field will lead to
high risks for nutrient losses to the environment. A failed
crop in poorer conditions may lead to starvation for the
farmer and her family.
Alternative methods, such as increased cultivation,
hand weeding, or biological pest reduction by introduc
tion of predators all have their advantages and
disadvantages, but there is no ‘silver bullet’ available. In
summary, an integrated approach with a combination of
methods is the solution, and modern agriculture has
moved and is moving in this direction. Of course, for
commercial reasons it can be profitable to cultivate, for
example, ‘organic’ crops (without fertilizer or pesticides)
to obtain a higher price, but from an ecological or envi
ronmental viewpoint this approach is not necessarily
better.
Agriculture can thus be classified according to the use
of agrochemicals, for example, biodynamic, organic, inte
grated, and industrialized farming. Biodynamic farming
forbids the use of conventional agrochemicals and
replaces them with exotic homemade concoctions, and
organic farming a priori forbids conventional agrochemi
cals. None of these farming systems is firmly based on
scientific evidence; instead they are based on a green view
of nature that leads to the banning of certain chemicals.
Integrated and industrial farming can also be called
‘conventional’, where economic, legal, and environmental
constraints limit the end goal, maximum productivity,
and profitability. The main difference between the latter
two is that integrated is more environmentally concerned
(reduced pesticide use, use of biological pest reduction
methods, etc.), and industrialized is more leaning to maxi
mum production with whatever means available, with a
minimum of environmental concerns. It should be noted
that ‘conventional’ and particularly ‘industrialized’ are
somewhat derogatory terms, mainly used by those nega
tive to these approaches.
Migration
Natural ecosystems, for example, East African savannas,
can be subjected to major migrations of large herbivores
that annually move long distances, following the seasonal
changes in rainfall and consequential grass growth. Most
natural ecosystems are less subjected to migrations, but,
for example, in Northern forests at least migratory birds
occur seasonally.
In agroecosystems, migration is usually kept to a mini
mum. Measures are taken to keep large or small grazers out
from the cropped field. In some regions, wild grazers are
exterminated (or close to extinction – Western European
agricultural regions) and in other regions crop fields are
guarded or fenced. However, migration is a component in
animal husbandry; cattle is often shifted between pastures,
which are given time to recover. Nomadic herding of cattle
(Sami people, Masai) is similar to the savanna migrations
148
Agriculture Systems
mentioned above; the cattle and herdsmen follow the
annual cycles in grazing opportunities.
Biodiversity
In a cereal monoculture, plant biodiversity is extremely
low – if weed control is successful there may be only one
species present, a highly specialized and genetically
homogeneous wheat variety. This is not common in natu
ral ecosystems, although it can occur in extreme
environments. As mentioned above, this means that a
pest can have a field day if it can reproduce in the field
(or migrate into the field at a large scale).
However, agricultural monocultures still are common
and continue to produce good yields. There are several
reasons for this. First, there is no simple relation between
biodiversity, productivity, or ecosystem stability. A plant
monoculture that is well adapted, grows under good con
ditions, and has a reasonable resistance to pests and diseases
can survive and produce well. This is exactly what a highly
productive agricultural field is – a well adapted monocul
ture. The crop variety has been selected for high
production under a number of years with different weather
(and on different soils) in a region. A variety that would
demand intensive treatment with herbicides, pesticides,
and fungicides will not be economical and will be rejected.
Second, the low plant diversity reduces animal diver
sity in the stand, but perhaps less than one would expect.
In a cereal monoculture stand, there can be hundreds of
species of insects, mites, springtails, snails, slugs, etc. In
the soil under a monoculture the biodiversity is almost
always extremely high, though usually lower than in
natural systems. Thousands, perhaps millions of bacterial
species, tens to hundreds of species of earthworms, enchy
traeids, soil insects, springtails, mites, spiders, millipedes,
flagellates, amoebae, blue green algae, etc. can be found.
There are no consistent indications that soil functions
such as organic matter decomposition is hampered by a
low biodiversity under monocultures – a given plant
residue will decompose at the same rate under a mono
culture as under mixed plants, if soil temperature and
moisture are the same.
Third, the last line of defense is the crop protection
measures that the farmer takes. For example, in several
countries there is a sophisticated monitoring and predic
tion system for aphid outbreaks. Aphids suck the sap from
the crop leaves, but they are also vectors for crop diseases.
Therefore their hibernating stages are enumerated,
weather is monitored, and if the conditions are ‘right’
the farmers are recommended to spray the fields with an
insecticide (or a more specific aphicide) with dose x at
date y. In less technically developed regions, experience
and skill is a substitute for the model projections, but the
principles are the same. It should also be mentioned that
in spite of these defenses, pest insects, pathogens, and
weeds still reduce worldwide crop yields considerably,
and there is a great potential for improvements.
Other Ecosystem Services
The main ecosystem service from agricultural systems is
simply to ‘feed the world’. This simple fact is easily
forgotten in the richer parts of the world. However,
even in Europe, which for centuries has been thoroughly
under agriculture, there are other ecosystem services that
are appreciated. In the forest dominated northern
Europe, agriculture actually contributes to biodiversity
and landscape diversity. Without agriculture, the forest
would cover all land area – the only open areas at lower
altitudes would be the lakes and rivers (and the newly
clear cut forest areas, rapidly covered by shrubs). The
European rural landscape in general, that is so refreshing
for the city dweller, is an agricultural product.
In other areas of the world, where the agricultural land
is not sufficient to properly feed the population, other
ecosystem services become relatively less important.
However, if agricultural productivity can be increased,
some agricultural land can be returned to savanna, forest,
or other natural or seminatural states – which would be
another type of service from the agroecosystem.
Since the agroecosystem is managed, and more or less
sophisticated machinery and management skills are in
place, it can easily be converted according to new
demands from the society. If the quality requirements
are met, agricultural fields can be used for recycling
organic waste and ashes, and even for drawing nutrients
out of sewage water. Conversion to energy crops is
not too difficult (grasses, sugar beet, willow, sugarcane,
etc.). Another demand from society, to sequester carbon
in the soil to reduce CO2 in the atmosphere, has recently
received much attention. Increasing soil carbon content
usually has beneficial effects for soil structure,
water holding capacity and general fertility, and C
sequestration, perhaps even with direct payments
per ton C sequestered to the farmer, is a new potential
service.
The Intelligent Choices
As mentioned in the introduction, an agroecosystem has a
purpose. It is designed to obtain certain goals, and the state
of the system at any given point in time is a consequence of
an array of intelligent choices by the farmer, complement
ing the border conditions set up by weather and soils, etc.
The following decision matrix (Figure 2) illustrates how
decisions made by a maize farmer in sub Saharan Africa
can be supported by basic science knowledge. Note that the
chemical analyses are not necessary for every farmer and
decision. Instead, typical values for the different organic
Agriculture Systems 149
N > 2.5%
YES
NO
Lignin < 15%; phenol < 4%
Lignin < 15%
YES
NO
YES
Incorporate directly
with annual crops
Mix with fertilizer
of high quality
organic matter
Mix with fertilizer
or add to compost
NO
Surface application
for erosion and
water control
For N on maize
- Need 2 t /ha as
minimum to have
an effect (50 kg
N/ha)
- Need at least 2
t /ha; 5 t /ha
preferable
- Apply more than
5 t /ha to give a
reasonable
response (1–2 t
grain per hectare)
- Need 40 kg N/ha
of fertilizer N for
each ton of
potential yield
- Do not apply
more than 10 t /ha
as much of the N
may be lost
Open kraal
Stall-fed cattle,
roofed, hard
floor
- Feed to livestock
and recycle
manure
- Need enough for
effective soil cover
OR
- Compost with
manure or
household waste
- May cause
problems of N
availability
OR
- Add to
kraal/boma to
trap urine N
- Need 40 kg N/ha
fertilizer N per
ton potential yield
- Need 40 kg N/ha
fertilizer N per
ton potential yield
Cattle manure
Figure 2 Example of farmer’s decisions regarding N management for a maize crop in sub-Saharan Africa, using a decision
support system for organic N management depending on resource quality, expressed as N, lignin, and soluble polyphenol content.
General decision matrix (top), more detailed for N economy in a maize cropping system (bottom). Modified from Vanlauwe B, Sanginga
N, Giller K, and Merckx R (2004) Management of nitrogen fertilizer in maize-based systems in subhumid areas of sub-Saharan Africa.
In: Mosier AR, Syers JK, and Freney JR (eds.) Agriculture and the Nitrogen Cycle. 124p. SCOPE 65. Washington Island Press.
resources are estimated, and the individual farmer uses the
rule of the thumb based on these estimates.
In the upper part of the Figure 2, the general decision
matrix is shown. Let us assume that we have leaves from a
tree, which we know have a low N content and less than
15% of lignin. Then we should mix the leaves with
fertilizer or add to compost. Now, in the lower part of
Figure 2 we can see that if we look in more detail at the N
economy of a maize system, we have other options –
maybe add the low N material to the cattle corral
(kraal/boma) to trap urine N or feed to livestock to
produce higher quality organic inputs. Organic resources
belonging to the third column from the left could be fed to
livestock and the manure thus produced could belong to
the first or the second organic resource class, depending
on the management of that manure.
All over the world, farmers make these kinds of
choices, based not only on biophysical knowledge and
constraints, but also on economic and sociopolitical
opportunities and constraints. An agroecosystem is not
only controlled by farmers, but also by the society the
farmer operates in. Subsidies can make growing products
that have no market an intelligent choice for the farmer;
lack of money can make fertilization impossible, even if it
would be profitable in the long run, or real or imaginary
environmental concerns from the society can force a
farmer to, for example, abandon fertilizer use, cereal
cropping, or pig farming.
Summing up, the agroecosystem, although limited by
climatic constraints, is a product of decisions made by
generations of farmers, supported by advice from agro
nomists and extension workers – all within a societal
context of values, traditions, and legislation. In fact, the
present and future agroecosystems are at least equally
dependent on the societal context as on the climate and
soil. However, the organisms involved are, as in any
ecosystem, products of millions of years of evolution,
and crop and animal breeding has only contributed
with small, although important changes to the
germplasm.
150
Alpine Ecosystems and the High-Elevation Treeline
Further Reading
Andrén O, Lindberg T, Paustian K, and Rosswall T (eds.) (1990)
Ecological Bulletins 40: Ecology of Arable Land Organisms, Carbon
and Nitrogen Cycling. Copenhagen: Ecological Bulletins.
Brussaard L (1994) An appraisal of the Dutch program on soil ecology of
arable farming systems (1985 1992). Agriculture, Ecosystems and
Environment 51(1 2): 1 6 and following papers.
Clements D and Shrestha A (eds.) (2004) New Dimensions in
Agroecology, 553p. Binghamton: The Hawort Press, Inc.
Eijsackers H and Quispel A (eds.) (1988) Ecological Bulletins 39:
Ecological Implications of Contemporary Agriculture. Copenhagen:
Ecological Bulletins.
Kirchmann H (1994) Biological dynamic farming an occult form of
alternative agriculture? Journal of Agricultural and Environmental
Ethics 7: 173 187.
Mosier AR, Syers JK, and Freney JR (eds.) (2004) Agriculture and the
Nitrogen Cycle, 124p. SCOPE 65 Washington: Island Press.
New TR (2005) Invertebrate Conservation and Agricultural Ecosystems,
354p. Cambridge: Cambridge University Press.
Newman EI (2000) Applied Ecology and Environmental Management,
2nd edn. Blackwell Science.
Vanlauwe B, Sanginga N, Giller K, and Merckx R (2004) Management of
nitrogen fertilizer in maize based systems in subhumid areas of
sub Saharan Africa. In: Mosier AR, Syers JK, and Freney JR (eds.)
Agriculture and the Nitrogen Cycle. 124p. SCOPE 65. Washington
Island Press.
Woomer PL and Swift MJ (1994) The Biological Management of Tropical
Soil Fertility. Chichester: Wiley.
Relevant Websites
http://www.cgiar.org Consultancy Group on International
Agricultural Research.
http://www.fao.org Food and Agriculture Organization of the
United Nations.
Alpine Ecosystems and the High-Elevation Treeline
C Körner, Botanisches Institut der Universität Basel, Basel, Switzerland
ª 2008 Elsevier B.V. All rights reserved.
Definitions and Boundaries
The Alpine Treeline
Alpine Plants Engineer Their Climatic Environment
Alpine Ecosystem Processes
Biodiversity in Alpine Ecosystems
Alpine Ecosystems and Global Change
Further Reading
Definitions and Boundaries
scattered flowering plants also grow above the snow line,
in favorable, equator facing, and sheltered places. The
uppermost part of the alpine belt, where closed ground
cover by vegetation is missing, is often termed ‘nival’,
referring to sparse vegetation in rock and scree fields.
The highest place on Earth where flowering plants have
been found is in the Central Himalayas at 6200–6350 m
above sea level.
Depending on latitude, the climatic treeline and hence
the lower limit of the alpine belt can be anywhere between
close to sea level in subpolar regions (>70 N, >55 S) and
close to 5000 m in subtropical continental climates (trees
>3 m at 4800 m in Bolivia and at 4700 m in Tibet). In the
cool temperate zone (45–50 N), the alpine belt may start
anywhere between 1200 and 3500 m (in the European Alps
at 2000 m, the Colorado Rocky Mountains at 3400 m); that
is, it is lower under strong oceanic influence and higher in
the inner parts of continents. The common natural treeline
altitude near the equator is 3600–4000 m. The altitudinal
width of the alpine belt above treeline is roughly 1000 m.
It covers c. 3.5% of the globe’s terrestrial area, if cold
and hot deserts are disregarded (Antarctica, Greenland,
Sahara, etc.).
Ecosystems above the upper climatic limit of trees are
termed ‘alpine’. Scientifically, the alpine life zone is an
altitudinal belt defined by climatic boundaries
(Figure 1) and the term ‘alpine’ does not refer to the
European Alps, but refers to treeless high elevation
biota worldwide (mostly grassland and shrubland).
‘Alpine’ supposedly roots in the pre Indogermanic
word alpo for steep slopes, still used today in the
Basque language. By contrast, in common language,
‘alpine’ is often used for places anywhere in mountai
nous terrain, irrespective of altitude (e.g., alpine village,
even alpine cities). If a city were truly alpine it would
have to be above the climatic treeline, but no such city
does exist worldwide. Hence, a distinction must be
made between the scientific, biogeographic meaning
of alpine (the issue of this text) and common (often
touristic) jargon.
The upper limit of the alpine life zone or alpine belt is
reached where flowering plants have their high altitude
limit. This is often close to the snow line (the altitude at
which snow can persist year round), but commonly a few
Alpine Ecosystems and the High-Elevation Treeline
Nival
Alpine
Treeline
Treeline
ecotone
Timberline
Montane
forest
Figure 1 The altitudinal belts of mountain ecosystems. With
increasing altitude these belts become fragmented and
topography (exposure) plays an increasing role. (Example from
the Swiss Central Alps with Pinus cembra forming the treeline at
2350 m.)
Given this convention on the two boundaries of the
alpine belt, it is important to note that these boundaries
are not sharp lines, but are centered across gradients
which change from place to place and depend on topo
graphy and region. Usually these boundaries are obvious
from great distance (an airplane), but hard to depict on the
ground, hence depend on scale.
The Alpine Treeline
Since, by definition, the alpine belt is naturally treeless, the
mechanisms by which trees are restricted from growing
beyond a certain altitude are key to any understanding of
alpine ecosystems. The so called treeline marks the upper
limit of the life form ‘tree’ irrespective of the tree species
involved (see Alpine Forest). Generally, species which form
treelines are Pinus, Picea, Abies, Juniperus, and Larix among
conifers, and Betula, Alnus, Erica, Polylepis, Sorbus, Eucalyptus,
and others among non coniferous families. Because tree
occurrence does not stop abruptly, and trees gradually get
smaller and finally become crippled, any definition of ‘a line’
is a convention. The forest line or timberline represents the
edge of the closed upper montane forest (note, ‘montane’ is
the biogeographic term for the next lower belt, not to be
confused with ‘mountain’), the zone of gradual forest open
ing near the treeline is often termed treeline parkland, and
the uppermost position where tree species can survive as
151
small saplings or shrubs among other low stature vegetation
is called the tree species line, with the ‘treeline’ holding a
middle ground, used for the line connecting the uppermost
patches of trees >3 m. The whole transition zone from
montane forest to alpine heathland is termed treeline eco
tone, across which alpine vegetation gains space yielded by
the thinning forest. The altitudinal range of the treeline
ecotone may be 20–200 m, often <50 m.
Where moisture is permitting tree growth at these
altitudes (a minimum of 250–300 mm of precipitation
per year), the position of the natural climatic treeline
matches with a mean growing season temperature of
6.6 0.8 C worldwide. The duration of the growing
season may vary from 10 weeks at high latitude to a
full year in the tropics and its onset and end are
defined by a weekly mean air temperature of 0 C
(corresponding to c. 3 C in 10 cm soil depth, where
most roots occur). This isotherm sets the lower climatic
threshold for alpine vegetation, which can be close to
5 C in dry subtropical mountains and 7.5 C in cool
temperate mountains, which is a surprisingly narrow
range, given the great difference in season length across
latitudes.
It is very important not to confuse this climatic (phy
siological) limit of trees with a multitude of other natural
or anthropogenic causes for the local absence of trees such
as fire, avalanches, logging and pasturing, loose or missing
substrate, waterlogging, or the regional lack of cold
adapted tree species (as is the case for instance in
Hawaii or New Zealand). In the last case, the treeline
observed is a specific tree species line, not representative
of the climatic limit of the life form tree, as can easily be
demonstrated by the success of introduced tree species
which grow well at much higher altitudes in such regions.
Open ‘alpine looking’ grassland and shrubland may thus
occur several hundreds of meters below the climatic tree
line; among the most famous of these are the Andean
Páramo grasslands with their spectacular giant rosette
plants (Figure 2).
Alpine Plants Engineer Their Climatic
Environment
Why is there lush alpine vegetation but trees cannot
grow? Are alpine plants physiologically superior, able to
cope with those low temperatures which otherwise are
harming trees? There is good evidence that thermal con
straints for growth, that is, building new tissue, are the
same for alpine plants, cold adapted trees, and winter
crops (winter rape and winter wheat), all being comple
tely halted when tissue temperatures drop below 5 C,
and growth is close to zero at 6–7 C. In contrast, all these
species reach 30–50% of maximum rates of photosynth
esis at these same temperatures; thus the provision of raw
152
Alpine Ecosystems and the High-Elevation Treeline
Figure 2 Fire and grazing (both naturally and under human
influence) can replace the montane forest, leading to
‘alpine-looking’ vegetation below the climatic treeline. Here is an
example of the Ecuadorian páramos at 3600 m altitude, c.
400–500 m below the potential climatic treeline (Páramos El
Angel). Giant rosettes of the genus Espeletia are the prominent
feature of this landscape, with similar vegetation also found in
African highlands.
material for growth (sugar) cannot be decisive. Neither
are there critical differences in freezing resistance
between alpine plants compared to trees. Hence, at tissue
level, there is no physiological reason why alpine grasses,
herbs, and shrubs should grow at a given low temperature
and trees should not.
There are two reasons for alpine plant success above
treeline:
1. By low stature and dense stand structure, alpine
plants restrict aerodynamic exchange with the atmosphere,
which causes heat to accumulate during periods with solar
radiation and permit plants to operate at comparatively
warm temperatures, much unlike those experienced by
upright, ventilated trees. The life form ‘tree’ does not
permit any escape from the gradually declining ambient
temperatures, whereas alpine plants engineer their micro
climate and aircondition their meristems close to the
ground so that they can build new tissue at otherwise
cold air temperatures above the plant canopy (Figure 3).
2. By developmental flexibility and morphological
adaptation, alpine plants are able to make use of short
favorable weather conditions, they sprout rapidly, pro
duce only a few, mostly short lived leaves (c. 60 days), and
have their meristems positioned very close to the ground,
in the case of many grasses, sedges, or rosette plants, often
1–2 cm below ground, where the solar heated soil pro
vides a thermally buffered environment. In contrast trees
operate at longer leaf duration (mostly >120 days, in
evergreen treeline conifers 4–12 years) and leaves take
longer to mature, and their aboveground meristems are
fully exposed to the cold air temperatures.
Figure 3 Trees are coupled to air temperature and thus,
appear ‘cool’ on this infrared thermograph taken at 10 a.m. on a
bright midsummer morning in the Swiss Alps near Arolla. Alpine
grassland and shrub heath accumulate heat by decoupling from
atmospheric conditions (low stature, dense structures). So the
treeline can clearly be depicted as a thermal boundary driven by
plant architecture.
The transition from trees to alpine vegetation is thus
dictated by plant architecture and not by tissue specific
inferiority of trees compared to alpine plants. This close
coupling of trees to atmospheric conditions also explains
the surprisingly uniform leveling of treelines across
mountain valleys which reminds one of the level of a
water reservoir. In contrast, the climate in alpine vegeta
tion varies with compactness and height of the leaf canopy
and exposure to the sun. A sun exposed, sheltered micro
habitat at 3000 m of altitude may be warmer than a shaded
microhabitat at 1800 m. Altitude per se, or data from a
conventional climate station, thus, tell us little about the
climate actually experienced by alpine plants. It had long
been known that mutual sheltering among alpine plants
or leaves/tillers within a plant is very beneficial (‘facilita
tion’), and removing this shelter effect by opening the
plant canopy can be disastrous.
Alpine Ecosystems and the High-Elevation Treeline
Alpine plants are small by design (genetic dwarfs); they
are not forced into small stature by the alpine climate
directly, though evolution had selected such morpho
types. What seems like a stressful environment is not
really stressful for those well adapted. However, there is
some additional modulative, direct effect on size by low
temperature. Alpine plants that survive in low altitude
rock gardens indeed grow taller than their relatives in the
wild. But plants grown in such rock gardens are com
monly of montane origin, because most typical alpine
plants fade at such high, low altitude temperatures, pos
sibly because of overshooting mitochondrial respiration.
Alpine Ecosystem Processes
Almost everything gets slower when it gets cold, but slow
production of biomass and slow recycling of dead biomass
(litter) go hand in hand, so that the carbon and nutrient
cycles remain in balance. Recycling of organic debris is
responsible for most of the steady state nutrient provision
and thus controls vigor of growth. When mineral nutrients
are added, all alpine vegetation tested had shown immedi
ate growth stimulation, but this holds for most of the world’s
biota and is not specific to alpine ecosystems. On the other
hand, nutrient addition had been shown to make alpine
plants more susceptible to stress (softer tissue, reduced
winter dormancy) and pathogen impact (e.g., fungal infec
tions) and causes nitrophilous grasses and herbs to overgrow
the best adapted slow growing alpine specialist species.
It comes as a rather surprising observation that alpine
plant productivity – at least in the temperate zone – is only
low when expressed as an annual rate of biomass accumula
tion, but is not low at all when expressed per unit of growing
season length. In a 2 month alpine season in the temperate
zone alpine belt, the biomass production (above plus below
ground) accumulates to c. 400 g m 2 (range 200–600 g m 2).
A northern deciduous hardwood forest produces 1200 g m 2
in 6 months and a humid tropical forest 2400 g m 2 in a
12 month season, all arriving at c. 200 g m 2 per month.
Time constraints of growth are thus the major causes of
reduced annual production in closed alpine grass and shrub
land and not physiological limitations in what seems to a
human hiker like a rather hostile environment. Acclimation
to lowtemperature, perfect plant architecture, and develop
mental adjustments can equilibrate these constraints on a unit
of time basis. It makes little sense to relate productivity to a
12 month period when 9–10 months show no plant activity
because of freezing conditions and/or snow cover.
Similar to carbon and nutrient relations, alpine ecosys
tem’s water relations are largely controlled by seasonality.
During the growing season in the humid temperate zone,
daily water consumption during bright weather hardly
differs across altitude (c. 3.5–4 mm evapotranspiration).
However, because of the short snow free season at
153
such latitudes, annual evapotransiration may be only
250–300 mm compared to 600–700 mm at low altitude,
hence runoff is much higher in alpine altitudes. Given
that precipitation often increases with altitude in the tem
perate zone (a doubling across 2000 m of altitude is not
uncommon), annual runoff may be 3–5 times higher in the
alpine belt, with major implications for erosion in steep
slopes.
In many tropical and subtropical mountains, moisture
availability drops rapidly above the condensation cloud
layer at 2000–3000 m altitude, causing the alpine belt to
receive very little water, often not more than 200–400 mm
per year (e.g., the high Andes, Tenerife, East African
volcanoes). The resulting sparse vegetation is often
termed alpine semidesert, but because of wide spacing
of plants and very little ground cover, those plants which
are found in this semiarid alpine landscape were found to
be surprisingly well supplied with water even at the end
of the dry season (Figure 4). As a rule of thumb, alpine
plants are thus better supplied with moisture (even in dry
alpine climates) than comparable low altitude vegetation.
True physiologically effective water stress is quite rare in
the alpine belt, but moisture shortage in the top soil may
restrict nutrient availablity periodically, which restricts
growth.
Biodiversity in Alpine Ecosystems
For plants and animals to become ‘alpine’ they must pass
through a selective filter represented by the harsh cli
matic conditions above treeline. It comes as another
surprise that alpine ecosystems are very rich in organis
mic taxa. It was estimated that the c. 3.5% of global land
area that can be ascribed to the alpine belt hosts c. 4% of
Figure 4 High-altitude semideserts (near Sajama, Bolivia,
4200 m) are often dominated by sparse tussock grasses, shrubs,
and minor herbs in the intertussock space, all together preventing
soil erosion, while being used for grazing. The wide spacing
mitigates drought stress in an otherwise dry environment.
154
Alpine Ecosystems and the High-Elevation Treeline
all species of flowering plants. In other words, alpine
ecosystems are on average similarly rich or even richer
in plant species than average low altitude ecosystems.
This is even more surprising if one accounts for the fact
that the available land area above treeline shrinks rapidly
with altitude (on average a halving of the area in each
successive 170 m belt of altitude). A common explanation
for this high species richness is the archipelago nature of
high mountains (a fragmentation into climatic ‘islands’),
the high habitat diversity as it results from gravitational
forces (topographic diversity, also termed geodiversity),
and the small size of alpine plants, which partly compen
sates for the altitudinal loss of land area . The altitudinal
trends for animal diversity are similar to plants, but some
animal taxa decline in diversity with altitude more
rapidly (e.g., beetles, earthworms, butterflies) than others
(e.g., vertebrates, birds). Often animal diversity peaks at
mid altitudes (close to the treeline ecotone) and then
declines.
The four major life forms of flowering plants in the
alpine belt are graminoids (grasses, mostly forming tus
socks, sedges, etc.), rosette forming herbs, dwarf shrubs,
and cushion plants (Figure 5). In most parts of the world,
bryophytes and lichens (a symbiosis between algae and
fungi) contribute an increasing fraction of biodiversity as
altitude increases. Each of these life forms can be subdi
vided into several subcategories, mostly represented by
different forms of clonal growth. Clonal (vegetative)
spreading is dominant in all mountains of the world
and it secures long term space occupancy by a ‘genet’
(a single genetical individual) in a rather unpredictable
environment. Because of the topography driven habitat
diversity, rather contrasting morphotypes and physio
types may be found in close proximity, as for instance
succulent (water storing) plants such as alpine cactus or
some leaf succulent Crassulaceae (Sedum sp., Echeveria sp.)
next to wetland or snowbed plants.
Alpine ecosystems are known for their colorful
flowers, and it was often thought that this may be a
selected for trait, because it facilitates pollinator visita
tion. There is also morphological evidence that alpine
plants invest relatively more in flowering, given that
plant size (and biomass per individual) declines by nearly
tenfold from the lowland to the alpine belt, whereas the
size of flowers hardly changes. Futhermore, flower dura
tion increases and so does pollinator visiting duration, and
there is no indication that there is a shortage in alpine
pollinators. The net outcome is a surprisingly high
genetic diversity in what seems like highly fragmented
and isolated habitats. Despite the successful reproductive
system at the flower pollinator scale and well adapted
(fast) seed maturation, the real bottleneck is seedling
establishment (the risk to survive the first summer and
winter), which explains why most alpine plants also pro
pagate clonally.
Overall, mountain biodiversity (the montane belt, the
treeline ecotone, and the alpine belt) is a small scale
analog of global biodiversity, because of the compression
of large climatic gradients over very short distances.
Across a vertical gradient from 1200 to 4200 m in the
Tropics one may find a flora and fauna with a preference
for climates otherwise only found across several thousand
kilometers of latitudinal distance. This is why mountains
are ideal places for biodiversity conservation as long as
the protected mountain system is large and has migration
corridors to prevent biota from becoming trapped in ever
narrowing land area should climatic warming induce alti
tudinal upward shifts of life zones.
Figure 5 The four major life-forms of flowering plants in alpine ecosystems: cushion plants (Azorella compacta, Silene exscapa), herbs
(small: Chrysanthemum alpinum, tall: Gentiana puncata), dwarf shrubs (Loiseleuria procumbens, Salix herbacea), and tussock-forming
graminoids (Carex curvula, diverse tall grass tussock).
Alpine Ecosystems and the High-Elevation Treeline
Alpine Ecosystems and Global Change
‘Global change’ includes changes in atmospheric chemis
try (CO2, CH4, NxOy), the climatic consequences of these
changes, and the manifold direct influences of humans on
landscapes. All three global change complexes affect
alpine biota, either directly or indirectly.
Elevated atmospheric carbondioxide (CO2) concentra
tions affect plant photosynthesis directly, although late
successional alpine grassland in the Alps was found to be
carbon saturated at ambient CO2 concentrations of the
early 1990s. The effect of doubling CO2 concentrations
over four consecutive seasons on net productivity was
zero. However, not all species within that sedge grass
herb community responded identically, hence there is a
possibility of gradual shifts in species compositition in the
long run, with some species getting suppressed and others
gaining.
In contrast, even very moderate additions of soluble
nitrogen fertilizer at rates of those received today by
mountain forelands in Central Europe with rains
(40–50 kg N ha 1a 1) doubled biomass in only 2 years.
Even 25 kg ha 1a 1 had immediate effects on biomass
(þ27%), again favoring some species more than others.
Atmospheric nitrogen deposition is thus far more impor
tant for alpine ecosystems than elevated CO2. Just for
comparison, in intense agriculture, cereals are fertilized
with >200 kg N ha 1 a 1.
Consequences of climatic change for alpine ecosys
tems are hard to predict because of the interplay of
climatic warming with precipitation. A warmer atmo
sphere can carry more moisture; hence increasing
precipitation had been predicted for temperate moun
tain areas. Greater snowpack can shorten the growing
season at otherwise higher temperatures. While the
temperate zone has seen more late winter snow in
recent years, the uppermost reaches of higher plants
seem to have profited from climatic warming over the
twentieth century. Several authors documented a clear
enrichment of summit floras, accelerated in recent
decades.
Treeline trees respond to warmer climates by faster
growth, but whether and how fast this would cause the
treelines of the world to advance upward depends on tree
establishment, which is a slow process. Hence treelines
always lagged behind climatic warming during the
Holocene by centuries, as evidenced by pollen records.
Current trends are largely showing an infilling of gaps in
the treeline ecotone, but upward trends await larger scale
confirmation. Eventually any persistent warming will
induce upward migration of all biota. By contrast, recent
climatic warming has caused the tropical upper montane/
alpine climate on Kilimanjaro to become drier, facilitating
devastating fires, which depressed the montane forest by
155
several hundred meters with a downslope advance
(expansion) of alpine vegetation following.
Land use is still the most important factor for changes
in alpine ecosystems. Around the globe, alpine vegeta
tion is used for herding or uncontrolled grazing by
lifestock. Much of the treeline ecotone has been con
verted into pasture land, with both overutilization and
erosion (mainly in developing countries) and abandon
ment of many centuries old, high elevation cultural
landscapes (mainly industrialized countries) causing pro
blems. The question is not whether there should be
pasturing, but how it should be done. Sustainable grazing
requires shepherding and observation of traditional prac
tices, which largely prevent soil damage and erosion.
Traditional alpine land use has a several thousand
years history and was optimized for maintaining an
intact landscape for future generations as opposed to
land hungry newcomers faced with the need of feeding
a family today, rather than thinking of sustained liveli
hood in a given area. All other forms of land use (except
mining), as dramatic their negative effects at certain
places may look, are less important, because their impact
is rather local (e.g., tourism, road projects). Agriculture is
by far the most significant factor in terms of affected land
area.
Mismanagement of alpine ecosystems has severe con
sequences (e.g., soil destruction, sediment loading of
rivers) not only for the local population, but for people
living in large mountain forelands, which depend on
steady supplies of clean water from high altitude catch
ments. Almost 50% of mankind consumes mountain
resources, largely water and hydrolectric energy, hence
there is an often overlooked teleconnection between
alpine ecosystems and highly populated lowlands.
Highland poverty is thus affecting the conditions and
the economic value of catchments, which goes far beyond
the actual agricultural benefits. This insight should lead to
better linkages between lowland and highland commu
nities and also include economic benefit sharing with
those that perform sustainable land care in alpine
ecosystem.
See also: Alpine Forest.
Further Reading
Akhalkatsi M and Wagner J (1996) Reproductive phenology and seed
development of Gentianella caucasea in different habitats in the
Central Caucasus. Flora 191: 161 168.
Bahn M and Korner C (2003) Recent increases in summit flora caused
by warming in the Alps. In: Nagy L, Grabherr G, Korner C, and
Thompson DBA (eds.) Ecological Studies 167: Alpine Biodiversity in
Europe, pp. 437 441. Berlin: Springer.
Barthlott W, Lauer W, and Placke A (1996) Global distribution of species
diversity in vascular plants: Towards a world map of phytodiversity.
Erdkunde 50: 317 327.
156
Alpine Forest
Billings WD (1988) Alpine vegetation. In: Barbour MG and Billings WD
(eds.) North American Terrestrial Vegetation, pp. 392 420.
Cambridge: Cambridge University Press.
Billings WD and Mooney HA (1968) The ecology of arctic and alpine
plants. Biological Reviews 43: 481 529.
Bowman WD and Seastedt TR (eds.) (2001) Structure and Function of
an Alpine Ecosystem Niwot Ridge, Colorado. Oxford: Oxford
University Press.
Callaway RM, Brooker RW, Choler P, et al. (2002) Positive interactions
among alpine plants increase with stress. Nature 417: 844 848.
Chapin FSIII and Korner C (eds.) (1995) Arctic and Alpine Biodiversity:
Patterns, Causes and Ecosystem Consequences. Ecological Studies
113. Berlin: Springer.
Dahl E (1951) On the relation between summer temperature and the
distribution of alpine vascular plants in the lowlands of Fennoscandia.
Oikos 3: 22 52.
Fabbro T and Korner C (2004) Altitudinal differences in flower traits and
reproductive allocation. Flora 199: 70 81.
Grabherr G and Pauli MGH (1994) Climate effects on mountain plants.
Nature 369: 448.
Hemp A (2005) Climate change driven forest fires marginalize the
impact of ice cap wasting on Kilimanjaro. Global Change Biology
11: 1013 1023.
Hiltbrunner E and Korner C (2004) Sheep grazing in the high alpine
under global change. In: Luscher A, Jeangros B, Kessler W, et al.
(eds.) Land Use Systems in Grassland Dominated Regions,
pp. 305 307. Zurich: VDF.
Kalin Arroyo MT, Primack R, and Armesto J (1982) Community studies
in pollination ecology in the high temperate Andes of central Chile.
Part I: Pollination mechanisms and altitudinal variation. American
Journal of Botany 69: 82 97.
Korner C and Larcher W (1988) Plant life in cold climates. In: Long SF
and Woodward FI (eds.) Symposium of the Society of Experimental
Biology 42: Plants and Temperature, pp. 25 57. Cambridge: The
Company of Biology Ltd.
Korner C (2003) Alpine Plant Life, 2nd edn. Berlin: Springer.
Korner C (2004) Mountain biodiversity, its causes and function. AMBIO
13: 11 17.
Korner C (2006) Significance of temperature in plant life. In: Morison JIL
and Morecroft MD (eds.) Plant Growth and Climate Change,
pp. 48 69. Oxford: Blackwell.
Korner C and Paulsen J (2004) A world wide study of high altitude
treeline temperatures. Journal of Biogeography 31: 713 732.
Mark AF, Dickinson KJM, and Hofstede RGM (2000) Alpine vegetation,
plant distribution, life forms, and environments in a perhumid New
Zealand region: Oceanic and tropical high mountain affinities. Arctic
Antarctic and Alpine Research 32: 240 254.
Messerli B and Ives JD (eds.) (1997) Mountains of the World: A Global
Priority. New York: Parthenon.
Meyer E and Thaler K (1995) Animal diversity at high altitudes in the
Austrian Central Alps. In: Chapin FS, III, and Korner C (eds.) Ecological
Studies 113: Arctic and Alpine Biodiversity: Patterns, Causes and
Ecosystem Consequences, pp. 97 108. Berlin: Springer.
Miehe G (1989) Vegetation patterns on Mount Everest as influenced by
monsoon and fohn. Vegetatio 79: 21 32.
Nagy L, Grabherr G, Korner C, and Thompson DBA (2003) Ecological
Studies 167: Alpine Biodiversity in Europe. Berlin: Springer.
Pluess AR and Stocklin J (2004) Population genetic diversity of the
clonal plant Geum reptans (Rosaceae) in the Swiss Alps. American
Journal of Botany 91: 2013 2021.
Rahbek C (1995) The elevational gradient of species richness: A uniform
pattern? Ecography 18: 200 205.
Sakai A and Larcher W (1987) Ecological Studies 62: Frost Survival of
Plants. Responses and Adaptation to Freezing Stress. Berlin:
Springer.
Spehn EM, Liberman M, and Korner C (2006) Land Use Change and
Mountain Biodiversity. Boca Raton, FL: CRC Press.
Till Bottraud J and Gaudeul M (2002) Intraspecific genetic diversity in
alpine plants. In: Korner C and Spehn E (eds.) Mountain Biodiversity:
A Global Assessment, pp. 23 34. New York: Parthenon.
Yoshida T (2006) Geobotany of the Himalaya. Tokyo: The Society of
Himalayan Botany.
Alpine Forest
W K Smith, Wake Forest University, Winston-Salem, NC, USA
D M Johnson, USDA Forest Service, Corvallis, OR, USA
K Reinhardt, Wake Forest University, Winston-Salem, NC, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Alpine Forest Biogeography
The Abiotic Environment
Altitude versus Microclimate
The Treeline Ecotone – Tree Distortion,
Clustering, and Spacing
Mechanisms of Treeline Formation
Summary
Further Reading
Introduction
Observations of the spatial patterns of the tree species do
insinuate some successional character, although the long
term encroachment of the subalpine forest into the alpine
zone, or vice versa, is a slow process that is detectable only
after centuries of change, at least. Although the alpine forest
has a well defined, characteristic vegetation pattern that
contrasts with the subalpine forest and alpine zones, animal
The forest of the alpine zone occurs near mountain tops and
forms a transition zone between the subalpine forest below
and the alpine zone above (Figure 1). Whether this zone of
overlap represents a definable, stable community with its
own inherent structure and stability is open for debate.
Alpine Forest 157
Figure 1 Alpine forest landscape (3200 m altitude) in the
treeline ecotone of the Snowy Range, Medicine Bow Mountains,
southeastern Wyoming (USA). Alternating snow glades
(long-lasting snow pack) and ribbon forest are characteristic of
this alpine forest, along with the potentially extreme distortion of
individual tree structure and form (see Figure 2). Prevailing
winds are from the right in this photo.
species are often viewed as community members of either or
both. This boundary ecotone between two contiguous com
munities is often referred to as the upper (or cold) treeline
(or timberline) ecotone where the treeline limit is reached.
This limit is defined as the highest occurrence of a tree
species in any form, or for a tree species that has a certain
minimum tree stature (e.g., greater than 2 m vertical height).
The latter definition is necessary because this upper limit of
tree occurrence is often composed of disfigured (flagged
branching) and stunted (krummholz mat) tree forms that
Seedling/sapling
are more shrub like than tree like in appearance (Figure 2).
This upper (cold) treeline ecotone can vary in altitude and
width according to latitude and proximity to maritime
influences, as well as the degree of slope and azimuth at a
given location. In addition, plant demographics such as tree
size, age, spacing, and clustering among individual trees,
plus the structural distortion and disfigurement of individual
trees further from the timberline, can vary dramatically.
Regardless of the latitude or altitude of mountain areas,
excessive steepness of the slope and, thus, poorly developed
soils, will prevent tree establishment and result in sharp
boundaries between the timberline and alpine community.
Above the timberline, individual trees or patches occur
sporadically associated with less wind exposed microsites
where aeolian soil and snow accumulate. These character
istics of the alpine forest landscape can also vary according
to the proximity to oceans or other large bodies of water
(e.g., ‘lake effect’ weather patterns). In general, greater lati
tudes result in a decrease in the altitude at which alpine
forest is found, as does a closer proximity to oceans or other
large bodies of water. In contrast, the dryer continental
mountains tend to have timberlines and treelines at the
highest altitude for a given latitude.
Alpine Forest Biogeography
Most of the ecological research focusing on the alpine
forest has involved vegetation studies, although many
animal species use this zone seasonally, especially later
in summer when lower elevations have dried from the
longer summer. This area is a prolonged green zone
Forest tree
Flagged tree
Abundance and
wind direction
15 m
Timberline
Treeline
Krummholz mats
Intact forest Snow
glade
Ribbon forest
Flagged trees
with mats
Mats with few
flagged trees
Mats only
Figure 2 Schematic representation of Figure 1 showing the relative size and spacing of individual tree forms and tree clusters making
up a typical alpine forest within the treeline ecotone of a dry, continental mountain range. See text for further explanation.
158
Alpine Forest
where food for herbivores, especially, is still in abundance
compared to lower elevations where most annual plants
have completed their life cycle, and the perennial species
have undergone a seasonal senescence due to accumulat
ing summer drought. Alpine forest is found on all
continents except Antarctica, as well as several oceanic
islands. The mountain regions of the Western
Hemisphere form large, N–S cordilleras that connect
polar regions to the subtropics. For example, the
Cascades, Rocky, and Sierra Nevada Mountains of the
western US extend from the most northern boreal forest
to southern Mexico where very high volcanic mountain
ranges occur, while the high ranges of the Andes connect
the full latitudinal extent of South America along its
western seaboard. In contrast, the high ranges of the
Alps of Central and Southern Europe, as well as the
Himalayas of the Eurasia, are formed along an E–W axis
and are much more discontinuous between the boreal and
subtropical latitudes. Further south, high mountains of
southern and eastern Africa represent much more isolated
ranges compared to the more continuous cordilleras of
the Western Hemisphere. In the Southern Hemisphere
where there is much less land mass, alpine forests are less
extensive and found in only a few mountain regions that
tend to be close to coastlines and, thus, have a strong
maritime influence (e.g., Andes, Australian Alps, New
Guinea, and New Zealand).
The question of why treelines across the globe occur at
specific altitudinal limits, and no higher, has been a focus
of research and discussion for over a century and a half.
Although it is well known that the altitude of upper
treelines have been strongly influenced by anthropogenic
causes (e.g., grazing and fires), the primary focus of these
studies has been on identifying the abiotic factors that are
most limiting to the growth and survival of trees.
However, there is also evidence that certain seed
dispersing bird species (e.g., Clark’s nutcracker and the
gray jay) may play a crucial role in the distribution of
certain species in the high altitude treeline (e.g., limber
and whitebark pine of the western US). The high altitude
environment involves particularly extreme values of cold
temperature, high wind, high and low (clouds) sunlight
levels, low air humidity, high long wave energy
exchange, and rapid mass diffusion due to low ambient
pressure. On wetter tropical mountaintops, forests may be
cloud immersed for much of the year. In general terms,
the temperature lapse rate (dry adiabatic) associated with
altitude generates a maximum decrease in air tempera
ture of approximately 1 C per 100 m of increasing
altitude. Thus, this environmental factor alone is a domi
nant environmental factor influencing differences in the
alpine forest located at dryer continental versus more
moist coastal mountain ecosystems. Coastal mountains
experience much lesser lapse rates (<0.3 C per 100 m)
because substantial condensation of moisture with greater
altitude transfers heat to the thinning atmosphere. In
addition to this extreme abiotic environment, the total
length of the growth season is severely curtailed (often
<90 days), and even growth during summer is severely
limited for short, but often frequent, time periods that
occur periodically during the entire summer growth per
iod. Because of these factors alone, adaptation and
survival of alpine forest species is most often perceived
as driven by abiotic pressures.
The distribution and species composition of alpine
forests on a global scale vary strongly according to both
latitude and longitude. In general, the Western
Hemisphere of North and South America has an N–S
and S–N cordillera that extends from the boreal forests
of northern Canada (Canadian Rocky Mountains) all the
way across North and Central America to the southern
most portions of South America (southern Andes). In
addition, this long expanse of alpine forest may be
strongly influenced by nearby oceanic influences. In con
trast, the major mountain ranges of Europe and Asia have
a more E–W distribution and in the case of the Asian
provinces, are much further away from strong oceanic
influences. There are also high mountains with treelines
located on volcanic islands in all the major oceans. The
mountains of New Guinea, southeastern Australia,
Tasmania, and New Zealand are examples of island like
alpine forests with strong oceanic influence and that also
extend from the subtropics to the extreme south tempe
rate zone of the Southern Hemisphere. There are only a
few Antarctic treelines that occur on small islands rela
tively close by. The distribution of different treeline tree
species on a global scale and according to plant type and
latitude is summarized in Table 1 and Figure 3.
Treelines of the northernmost temperate zones of the
Northern Hemisphere are dominated in the eastern
hemisphere by white birch at the highest latitudes of the
Scandinavian, Ural, and eastern Siberia ranges, followed
by the Scotch pine and European spruce treelines of
central Norway and Sweden. In the more interior, con
tinental areas of central Europe, the Swiss mountain pine
(Pinus mugo) forms the alpine forest, while European larch
(Larix decidua) and stone pine (Pinus cembra) form the
treelines of the central Alps. In the maritime mountains
of the western and southern Alps, European beech (Fagus
silvatica) is dominant. In the Western and Northern
Hemisphere, evergreen conifers dominate the alpine
forest (e.g., larch, bristlecone pine, subalpine fir, and
Engelmann spruce), while deciduous conifers and broad
leaf species are found less frequently, along with the rare
occurrence of evergreen broadleaf species in South
America (Table 1 and Figure 4). Many of these distribu
tion patterns appear influenced by differences not only in
abiotic factors, but also in historical factors related to
dispersal mechanisms and historical factors related to
continental drift over geological timescales.
Alpine Forest 159
Table 1 Biogeography of alpine forest areas worldwide
Latitude
Mountain range
Western Hemisphere
50–60 N
Northern Rockies,
USA/Canada
Scottish Highlands,
55–58 N
UK
Northern
45–50 N
Appalachians, USA
Pacific Coast Mtns,
30–60 N
USA, Canada
Middle Rockies, USA
40 N
Sierra Nevadas, Spain
37 N
Sierra Nevadas, USA
35–40 N
Southern Rockies,
30–35 N
USA
Sierra Madres, Mexico
18–25 N
Talamancas, Costa
9–11 N
Rica
Tropical Andes, Costa
10 N–20 S
Rica–Peru
South America,
23 S–50 S
Temperate Andes
Altitude (m)
Climate
Life form
Dominant tree genera
2600–2900
Continental
DN, EN
Abies, Picea, Pinus, Larix
600–800
Oceanic
DB
Pinus, Juniperus, Betula
1500
Continental
EN
Abies, Picea
To 3300
Oceanic
EN
2900–3300
1950
3000–3500
3300–3800
Continental
Continental
Oceanic
Continental
EN
DB, EN
EN
EN
Tsuga, Pinus, Abies, Picea,
Chamaecyparis
Abies, Picea, Pinus
Quercus, Pinus
Pinus, Tsuga, Juniperus
Abies, Picea, Pinus
4000
3000
Oceanic
Oceanic
EN
DB
Pinus
Quercus
3150–4700
Oceanic
DB
Polylepis
1100–1500
Oceanic
EB, DB
Podocarpus, Nothofagus
Eastern Hemisphere
Skanderna,
60–70 N
Scandinavia
Altai, Mongolia
50 N
Tien Shan, Asia
44 N
Alps, Europe
46–43 N
700–900
Continental
DB
Betula
2000
3000
1600–2300
Continental
Continental
Continental
Pinus, Larix
Picea
Abies, Picea
43 N
Caucasus, Georgia
2200
Continental
38 N
35–36 N
Pamir, Tajikistan, Asia
Japanese Alps/Fuji,
Japan
Great Atlas, Northern
Africa
Himalayas, Asia
3000
1950–2400
Continental
Oceanic
EN, DN
EN
EN, DN,
DB
DB, EB,
EN
EN
EN, DN
EN
Cedrus, Juniperus
3800–4500
Continental
Maoke Mtns
East Africa Highlands,
Africa
Australian Alps,
Australia
Tasmania
3000–3600
4050
Oceanic
DB, EB,
DN, EN
EN
EB
Betula, Rhododendron, Picea, Larix,
Juniperus, Tsuga
Podocarpus
Erica
1800–1950
EB
Eucalyptus
1200–1260
EN, EB,
DB
DB
Athrotaxis, Eucalyptus, Nothofagus
31 N
28 N
2 S
3 S
36 S
42 S
43 S
Southern Alps,
New Zealand
2850
1200–1500
Betula, Rhododendron, Pinus
Picea
Abies, Larix, Pinus, Tsuga
Nothofagus
DN, deciduous needleleaf; EN, evergreen needleleaf; DB, deciduous broadleaf; EB, evergreen broadleaf.
The Abiotic Environment
Aboveground
Abiotic factors have traditionally been viewed as domi
nating the ecology of high altitudes, including the alpine
forest. Sunlight, temperature, water, and gas phase nutri
ents (e.g., CO2 and O2) can vary substantially with
altitude, regional climate, and orographics (e.g., maritime
vs. continental mountain ranges). In addition, many fac
tors influencing leaf energy balance and temperature may
also vary with elevation, including solar and long wave
radiation, wind, and ambient humidity. Probably, the
best known abiotic change with increasing elevation is
the decline in air temperature in response to lower ambi
ent pressure. Ambient pressure decreases by over 20% at
2 km and over 50% at 6 km, leading to a maximum, dry
adiabatic lapse potential of 1.0 C/100 m. Simulated dry
(8.0 C km 1) versus wet (3.0 C km 1) lapse conditions
resulted in a more rapid decline in air temperature with
altitude for both winter and summer temperatures. Also,
160
Alpine Forest
Rocky
Mountains
Pacific Coast Range
Scottish
Highlands
Northern Appalachians
Sierra Nevadas
Sierra Nevadas
Sierra Madres
Talamancas
Tropical Andes
Temperate Andes
Deciduous needleleaf (DN)
Evergreen needleleaf (DN)
Deciduous broadleaf (DB)
Evergreen broadleaf (DB)
Skanderna
Alps
Atlas Mtns
Urals
Sayans
Altais
Tien Shan
Caucasus
Pamirs
Himalayas
Japanese Alps
East Africa Highlands
Maoke Mtns
Australian Alps
Tasmania Southern Alps
Figure 3 Global biogeography of alpine forest areas. Italicized mountain ranges have no corresponding information in Table 1.
Alpine Forest 161
(a)
Within herbaceous cover
cold nights, intermediate water stress, shade
90% survival *
(b)
Figure 4 Individual alpine forest tree in the high-altitude
treeline (3306 m) at a wind-exposed site. At this high-altitude limit
of tree growth, extreme distortion in tree structure results in the
classic krummholz mat growth form showing the presence of
flagged branches on the downwind edge. The prevailing winds
are from the right in this photo.
dry lapse conditions in summer generated similarly cold
air temperatures at higher elevations (>4 km) that were
very near values computed for wet lapse conditions dur
ing winter (Figure 5b). Similar dry and wet lapse
rates of 7.5 and 5.5 C km 1, respectively, have been
used previously to evaluate transpiration potential for
plants growing on mountains of temperate and tropical
zones.
Another fundamental change in abiotic factors of
increasing altitude is the unique and colligative property
of decreasing atmospheric pressure and, thus, the partial
pressures of gas phase molecules such as CO2 and O2.
In contrast, the amount of water vapor in the air at satura
tion is dependent only on temperature and, thus, strongly
influenced by the lapse rate in air temperature described
above. Because ambient CO2 concentration can have a
strong, direct influence on plant photosynthesis via the
leaf to air concentration gradient (driving force for diffu
sion), it has often been assumed to be a limiting factor for
carbon gain and growth at high elevation. For plants, where
the diffusion process is the primary mode of gas exchange,
a lower ambient CO2 concentration with altitude could
result in a corresponding decrease in the leaf to air gradi
ent, assuming a constant CO2 concentration inside the leaf.
For this reason, mountain ecosystems have been consid
ered as natural field models for evaluating the effects of
natural differences in atmospheric CO2 concentrations.
However, because molecular diffusion is more rapid at
lower ambient pressure, a substantial compensatory effect
on CO2 uptake potential occurs with greater elevation. In
agreement with this physiochemical property, little evi
dence has been found supporting the idea that lower partial
pressures result in diffusion limitations at higher altitudes,
at least for systems depending on the diffusion process
for physiological gas exchange. Although quantitative
Exposed, on bare soil
warmer nights, least water stress, full sunlight
44% survival
(c)
In opening in herbaceous cover
cold nights, greatest water stress, full sunlight
19% survival
Figure 5 Microsite alteration experiment showing effects of
facilitation vs. competition on survival of new (first-year)
seedlings of Picea engelmannii Parry ex. Engelm. (Engelmann
spruce) in an alpine treeline ecotone, southeastern Wyoming.
Greatest survival (90%) occurred for seedlings growing in
vegetative ground cover that resulted in low sky exposure and
incident sunlight the following morning, intermediate water
stress, and relatively cold nights. Removing all vegetation well
away from a seedling reduced competition for soil water (higher
xylem water potentials), but increased sky exposure, resulting in
significantly lower survival (44%). The highest mortality occurred
when only proximal vegetation was removed to increase sky
exposure, while maintaining boundary layer effects, lower
minimum needle temperatures, and competition for water
(as validated by higher water potential values). Higher
photosynthetic carbon gain due to less low-temperature
photoinhibition of photosynthesis was also associated with
greater survival. Thus, facilitated reduction in sky exposure
(day and night) appeared to have a greater influence on
photosynthesis and survival, compared to low temperatures or
competition for water with neighbors, although all three stress
factors had significant impact. From Germino MJ, Smith WK, and
Resor C (2002) Conifer seedling distribution and survival in an
alpine-treeline ecotone. Plant Ecology 162: 157–168.
evaluations showing these compensating effects on photo
synthetic CO2 uptake exist in the literature, there are few
comprehensive studies incorporating all of the potentially
important factors influencing diffusional gas exchange at
higher altitudes. Similar concerns for animal O2 uptake at
high altitude form a vast literature, although animals,
depend primarily on bulk supply mechanisms for enhan
cing gas exchange. However, diffusion effects on animal
ecophysiology at high elevations (e.g., eggs, burrowing and
subnivian animals) are not well studied, except for a large
162
Alpine Forest
literature dealing with human physiology under hypoxic
conditions.
Other abiotic factors such as the known increases in
sunlight due to a thinner, unpolluted atmosphere, lower
ambient humidities, high wind regimes, and decreased
long wave radiation from the sky (downwelling) have
been studied less thoroughly, and for only a few mountain
systems. In particular, the decrease in downwelling radia
tion can result in lower minimum temperatures at night
that are often freezing even in summer. The influence of
snow accumulation has been shown to be critical for
winter survival of evergreen plants, preventing poten
tially lethal wind damage and desiccation via cuticle
abrasion, as well as exposure to the cold sky and lower
air temperatures above the snowpack. Though most stu
dies have considered changes in single, or a few, abiotic
factors, none have considered the concerted influence of
multiple stress factors on the different habitat types of the
alpine forest environment; for example, only a few studies
have incorporated multiple abiotic factors to evaluate
effects of high elevation on such important physiological
processes as evapotranspiration, even though water dif
fuses rapidly from all evaporating surfaces, both plants
and animals, compared to sea level.
Belowground
The soil environment of the alpine forest, as for many
other communities and ecosystems, is strongly depen
dent on the prevailing moisture regime, including the
seasonal timing and physical nature of the precipitation
(rain vs. snow). Although the winter season can result in
snow even in tropical mountains, there can also be
important impacts due to the occurrence of dry seasons,
sometimes twice per year. Regardless, rainfall of a melt
ing snowpack results in quite different impacts on soil
nutrients based on the accompanying temperature
regimes. Warmer periods with rainfall will result in the
release of soil nutrients, while colder periods with snow
accumulation will involve dormant periods for surface
soil organisms and, thus, decomposition and nutrient
release. In addition, the growth activities of the impor
tant mycorrhizal fungi of plant root systems are strongly
limited by soil temperatures well above freezing (up to
7 C). Because more tropical alpine forests receive the
majority of precipitation as snow during a relatively brief
winter season, or wet season, snowmelt often occurs
quickly and is followed by a rapid drying of surface
soils where roots are found. Thus, plants must take up
soil nutrients during a very abbreviated time period
which can be limited by persistently cold soils that lag
behind air temperatures on a daily and seasonal basis.
Soils’ freeze and thaw cycles at high altitudes can also
create distinct patterns in the microtopography of the
soil surface (e.g., polyhedrons), providing important
microsites for seedling establishment and small scale
differences in plant distribution patterns.
Altitude versus Microclimate
An important message concerning changes in abiotic fac
tors with elevation is the realization that decreasing
ambient pressure is the only physical property unaffected
by microclimatic effects. All others (e.g., temperature,
sunlight, wind, long wave radiation, water and nutrient
relations) can be strongly influenced by topography,
microsite, and plant form at any altitude. Natural varia
bility in these factors can substantially lower, or raise, the
effective altitude of a microsite at any given altitude. In
the Northern Hemisphere, south facing, wind sheltered
microsites can effectively match conditions at an altitude
thousands of meters lower, while similar north facing
microsites, sheltered from sun but not the cold night
sky, could generate increases in effective altitude. Even
smaller microsites around a fallen tree stem or an exposed
boulder can result in effectively different altitudes based
on differences in sunlight exposure and temperatures.
Additionally, changes in leaf orientation of the plant
can create different levels of sky and wind exposure,
two primary factors influencing microclimate at any alti
tude. Leaf and plant aggregation (close spacing) and
height patterns can also influence microclimate due to
the potentially strong boundary layer effects on tempera
ture and ambient gas concentrations. Thus, microclimate
effects can significantly impact fundamental gas exchange
processes at any altitude, with the exception of ambient
pressure effects on molecular diffusion. In contrast to the
potential effects of microsite and plant form on effective
altitude, individual plants cannot escape the ambient
pressure of their respective altitudes (only negligible
changes in ambient pressure due to weather fronts).
Thus, lower ambient pressure and more rapid molecular
diffusion are the only immutable abiotic factors associated
with increasing altitude, one that is not dependent on
microsite/microclimate effects.
The Treeline Ecotone – Tree Distortion,
Clustering, and Spacing
On a geographic scale, the appearance of an alpine forest
community can vary considerably, depending primarily on
latitude and the distance from oceanic influences. Alpine
forests occur at higher altitudes at lower latitudes, but with
stronger maritime impacts. However, similar variation at a
particular site is also associated with the steepness of the site,
along with sun and wind exposure. Typically, the altitude of
the treeline falls as these site specific factors increase.
A typical alpine forest and treeline ecotone community of
Alpine Forest 163
a dry, continental mountain range is shown in Figures 1 and
2, both of which represent the most extreme changes in tree
form and landscape found in alpine treeline ecotones (alpine
forest). As one progresses from the treeline toward the intact
subalpine forest, individual trees occur as small (<2 m in
width and a meter in height) krummholz mats, larger mats
with flagged trees at their leeward edge, still larger tree
islands (>10 m diameter) with more intensely flagged trees
near the windward side, and, finally, ribbon forest alternat
ing with snow glades (>10 m across) just prior to edge of the
intact subalpine forest (although some flagging at tree tops is
still noticeable). The density of these various structures also
increases closer to the forest edge, along with the occurrence
of young seedlings and saplings.
Mechanisms of Treeline Formation
Investigators have been interested for over a century in the
question of why trees do not occur above certain altitudes.
Ecological studies have shown that the occurrence of both
timberlines and treelines decline steeply and almost linearly
in altitude as latitude increases between about 30 N and S
latitude and over 60 N and S latitude. This linear relation
ship results in an estimated change in timberline altitudes
of 100 m per degree of latitude. However, between
about 30 on each side of the equator, there is a relatively
constant, maximum altitude of occurrence that is near
3.5–4.0 km. Little information exists concerning the dif
ference in altitude between the timberline and treeline, or
the width of the treeline ecotone (alpine forest) as related
to geography or any specific environmental factor.
Although most of these studies have associated this alti
tude of occurrence to the colder temperature regimes at
higher latitudes, the actual ecophysiological mechanisms
are still being debated, and may involve a large number of
abiotic and biotic factors. In addition, major changes in
tree habit occur within this life zone, including dramatic
alterations in plant height and crown features such as
branching pattern. This change in growth form becomes
more dramatic as distance from the forest edge (timber
line) increases toward the ultimate treeline limit (Figures
1 and 2). Across this ecotone, the full tree stature of a
typical forest tree becomes twisted and distorted, forming
ultimately a small, shrub like habit commonly referred to
as the ‘krummholz’ mat at treeline. During this transition,
trees also become more and more flagged in appearance,
with stems occurring only on the downwind side of trunks
and main stems (Figure 3).
Because temperature data have been mostly available
for the longest period of time and for most locations
worldwide, a host of studies have attempted to correlate
measured temperature regimes with the highest altitude
of tree occurrence. Within these myriad studies, the
occurrence of minimum temperatures and the amount
and physical nature of the prevailing snowfall has been a
central focus. For example, more continental (noncoastal)
mountain ranges of both hemispheres have dryer, colder
climates characterized by ‘powder snow’ conditions. This
type of snow is strongly influenced by wind driven snow
that can generate strong abrasive forces due the sharp
edged, crystalline nature of these snow particles. These
systems also have distinct snow accumulation patterns
across the landscape that are the result of the strong
turbulent and eddy flow characteristics. Moreover, snow
burial and avoidance of excessive exposure to wind and
colder temperatures may be critical for the winter survi
val of both plants and animals in this alpine forest belt. In
contrast, coastal ranges with lower altitudes of treeline
also have higher air humidity levels, snowfall of high
water content, and low abrasive power of softer ice crys
tals that are relatively uncoupled from the influence of
wind patterns. This wetter, heavier snow can accumulate
on exposed branches, creating severe mechanical forces
that can bend, break, and distort stems due to snow and
ice loading above the snowpack surface and freeze–thaw
compression forces beneath the snow surface. Snow accu
mulation in the dryer continental alpine forest is much
more dependent on drift mechanics and eddy flow
dynamics (e.g., burial of krummholz mats), while the
wetter snows of more coastal systems generate a more
uniform depth and homogeneous distribution pattern
across the treeline ecotone. For the dryer powder snow
of the continental mountain tops, severe abrasive proper
ties can lead to abrasion of leaf cuticles, removal of paint
from highway traffic placards, and the common windburn
suffered by skiers on windy days and powder like snow.
Thus, these differences in the physics of snow particles
and spatial distribution dynamics also play a major role in
the distortion and disfiguration effects on individual trees
of the alpine forest (stunting, flagging, and krummholz
tree forms), as well as the spatial patterns of tree spacing.
These differences in the basic physical make up of snow
have not been considered systematically in terms of their
influence on the vegetation patterns and distortions in
growth form observed for individual trees within different
alpine forests (appearance of krummholz and flagged
growth forms). These effects for more maritime versus
continental mountain ecosystems need further elucida
tion, in particular, the impact on the altitude at which
trees can no longer regenerate.
The ecophysiological mechanisms regulating the
upper elevational limits of treelines across the globe
have been contemplated by plant ecologists, biogeogra
phers, and biometeorologists for over a century. A recent
review concluded that the elevation limits of the upper
treelines on a global scale is the result of (1) the
inability of alpine plants to metabolically process the
carbon gained from daytime photosynthesis because of
164
Alpine Forest
cold temperature limitations (e.g., respiratory limita
tions), and (2) the large size of conifer trees which
prevents adequate warming of the soil due to soil surface
shading by the closed, overstory canopy. Thus, low soil
temperatures due to self shading was proposed as a major
abiotic determinant of the elevational limits of upper
forest treelines. However, other studies have provided
evidence of strong limitations to resource acquisition at
high altitudes, specifically the photosynthetic uptake of
CO2 by alpine forest trees. Many other investigators have
also questioned conclusions (1) and (2) above.
Despite the longstanding interest in the environmental
and physiological mechanisms generating observed alti
tudinal patterns in the formation of alpine forests and
their respective treelines, virtually all of this research
has focused on the ecophysiological effects measured for
adult trees, even though they may show distortions in
form and greatly diminished stature, for example,
krummholz mats and stunted, flagged trees. Very little
research has focused on the establishment of new seed
lings away from the forest edge into the treeline ecotone.
Yet, it is this life stage within the treeline ecotone that
appears critical for migration to a higher altitude and
formation of new subalpine forest. The formation of new
subalpine forest at higher elevation is dependent on seed
ling regeneration into the ecotone, whereas the migration
of the forest timberline to a lower altitude would require
both the mortality of older trees and the successful seed
ling regeneration at the new altitude of occurrence.
However, any mortality of the overstory trees could also
introduce an important impact – a decrease in the ecolo
gical facilitation of seedling establishment. Likewise, a
lack of establishing seedlings in the forest understory at
the forest edge, in combination with the death of the
overstory trees, would result in a lowering of the timber
line and, most likely, the treeline as well. An important
component of this process is the ecological facilitation of
new seedling survival and growth that results from a more
mature forest structure (Figure 2). In other words, the
development of trees with forest like stature (no flagging
or krummholz distortion) requires the formation of an
intact forest and the resulting amelioration of a host of
extreme abiotic factors outside the forest. Thus, the alti
tudinal movement of timberline and treeline boundaries
begins with new seedling establishment, either below of
above the existing timberline that will act, ultimately, to
facilitate further seedling establishment and the gradual
development of new subalpine forest either above or
below the altitude of the existing timberline. For example,
the mechanisms involved in the migration of a timber
line/treeline to a higher altitude must initially depend
upon new seedling establishment above the existing tim
berline, into the treeline ecotone. Moreover, greater
seedling/sapling abundance must follow to provide the
ultimate facilitation required for continued growth to full
forest tree stature and, thus, the formation of new sub
alpine forest at higher altitudes. At high elevation, this
migration of timberline is possible only with the protec
tive, mutual facilitation provided by neighboring trees
and surroundings, similar to that found within intact sub
alpine forest. Thus, growth to forest tree stature without
structural distortion may require, to some degree, ‘the
forest before the tree’. In the Rocky Mountains of south
eastern Wyoming (USA), the establishment of new tree
seedlings into a treeline ecotone appears also to involve
considerable microsite facilitation (Figure 5) by either
inanimate objects (e.g., rocks, fallen logs, microtopogra
phy due to freezing and thawing of the soil surface), or by
intra and interspecific spatial associations generating
ecological facilitation of microsites. Structural self facil
itation (e.g., cotyledon orientation and primary needle
clustering, krummholz mats) may also act to enhance
the growth and survival at all structural scales from the
seedling to mature trees (Table 2). Increased seedling
establishment and abundance is followed subsequently
by even greater facilitation, which leads to even greater
seedling establishment and sapling growth, and so on
(Table 3). Thus, increased seedling/sapling abundance
will lead to the same ‘sheltering’ effect that is necessary
for the formation of the forest ‘outposts’, or islands, known
to be important shelters for improved seedling establish
ment. In addition, the ultimate development into a forest
tree (nondistorted growth form) is analogous functionally
to the biophysical ‘escape’ of vertical stems from the sur
face boundary layer of a krummholz mat (Figure 3).
Subsequently, continued facilitation of the sapling stage,
approaching a similar level as found within the intact
subalpine forest at lower elevation, is required before an
establishing sapling can reach the stature of a subalpine
forest tree.
Table 2 Factors identified as important for explaining the
altitudinal occurrence of alpine forest and its maximum altitude of
occurrence as an alpine treeline ecotone
1. Seedling/sapling establishment – seed germination, growth,
and survival
2. Mechanical damage – wind abrasion of needle cuticles, apical
bud damage, snow loading, and frost heaving cause tissue
and whole-tree mortality
3. Physiological tissue damage – low temperature and
desiccation limits growth and survival
4. Annual carbon balance – photosynthetic carbon gain minus
respiratory demands is less than that needed for successful
growth and reproduction
5. Biosynthesis and growth limitationa – greater cold
temperature limitation to growth processes than to
photosynthetic carbon gain
a
Cold soil temperature due to the large size of conifer trees and
consequential soil shading have been hypothesized as a primary
environmental factor limiting the processing of assimilated carbon and,
thus, maximum altitude of alpine treelines.
Alpine Forest 165
Table 3 The importance of ecological facilitation for seedling
establishment, growth, and survival in the alpine forest
Source
Biotic: Inanimate (rocks, dead wood, microtopography)
Abiotic: Plant structure (clustering), intraspecific and
interspecific facilitation of microsites
Benefits
Winter
Snow burial – prevents ice crystal abrasion and desiccation;
warmer and less extreme diurnal temperature
differences; no excessive sunlight exposure
Clustering at the shoot-to-landscape scale – increased
snow deposition and burial
Flagging – prevents damage from snow loading and rime ice
accumulation
Summer
Less sky exposure
Day: Less sunlight and cooler temperatures
Night: Higher minimum temperatures and less LTP; less
dew and frost accumulation
Less wind exposure – warmer needles in sun
Possible adaptive tradeoffs
Less sun sky exposure due to burial and mutual shading
Day: Less sunlight for photosynthesis and lower
temperatures
Less wind exposure
Day: Warmer temperatures and greater transpiration
Night: Colder minimum temperatures and greater LTP
Inanimate, intraspecific, interspecific, and structural facilitation can all
generate protective snow burial, as well as amelioration of subsequent
growth limitation factors within and just above associated ground cover.
LTP represents low temperature photoinhibition of photosynthetic
carbon gain.
Summary
The alpine forest represents a transitional zone separating
the alpine tundra and subalpine forest communities. This
treeline ecotone is also the highest altitude at which trees
are found to occur, although the exact environmental
factors and mechanisms limiting this occurrence are just
beginning to be unraveled. These treelines are composed
of evergreen conifer species most often, although decid
uous conifers and broadleaf species also occur, as well as
evergreen broadleaves at lower latitudes. There is also a
strong correlation between higher latitudes and a lower
treeline altitude, as well as with more continental versus
maritime mountains. Ecological facilitation of seedling
microsites by inanimate structures and microtopography,
along with intra and interspecific facilitation, is a funda
mental property of timberline migration up or down the
mountain and, thus, the formation of new subalpine for
ests at a different altitude. This facilitation of microsites
involves environmental parameters such as avoidance of
wind exposure, wind /snow abrasion, and exposure to
sunlight and the cold nighttime sky. In addition, the
ability to survive in exposed microsites appears coupled
to developmental capabilities for forming krummholz and
flagged forms that enable wind protection, including ade
quate snow collection and burial to prevent damage from
the abiotic environment. As new seedling and sapling
cover increases, facilitation of growth processes by micro
climate amelioration leads to the ultimate growth of trees
to a forest tree stature, culminating in the protective
environment of a new subalpine forest.
See also: Alpine Ecosystems and the High-Elevation
Treeline; Boreal Forest.
Further Reading
Arno SF and Hammerly RP (1990) Timberline: Mountain and Arctic
Forest Frontiers. Seattle, WA: The Mountaineers.
Callaway RM (1995) Positive interactions among plants. Botanical
Review 61: 306 349.
Choler P, Michalet R, and Callaway RM (2001) Facilitation and competition
on gradients in alpine plant communities. Ecology 82: 3295 3308.
Germino MJ, Smith WK, and Resor C (2002) Conifer seedling
distribution and survival in an alpine treeline ecotone. Plant Ecology
162: 157 168.
Grace J, Berniger F, and Nagy L (2002) Impacts of climate change on
the treeline. Annals of Botany 90: 537 544.
Holtmeier FK (1994) Ecological aspects of climatically caused timberline
fluctuations: Review and outlook. In: Beniston M (ed.) Mountain
Environments in Changing Climates, pp. 223 233. London: Routledge.
Innes JL (1991) High altitude and high latitude tree growth in relation to
past, present and future climate change. Holocene 1: 168 173.
Jobbagy EG and Jackson RB (2000) Global controls of forest line
elevation in the Northern and Southern hemispheres. Global Ecology
and Biogeography 9: 253 268.
Korner C (1998) A re assessment of high elevation treeline positions and
their explanation. Oecologia 115: 445 459.
Smith WK, Germino MJ, Hancock TE, and Johnson DM (2003) Another
perspective on the altitudinal limits of alpine timberline. Tree
Physiology 23: 1101 1112.
Smith WK and Knapp AK (1985) Montane forests. In: Chabot BF and
Mooney HA (eds.) The Physiological Ecology of North American
Plant Communities, pp. 95 126. London: Chapman and Hall.
Stevens GC and Fox JF (1991) The cause of treeline. Annual Review of
Ecology and Systematics 22: 177 191.
Sveinbjornsson B (2000) North American and European treelines:
External forces and internal processes controlling position. AMBIO
29: 388 395.
Tranquillini W (1979) Physiological Ecology of the Alpine Timberline.New
York: Springer.
Walter H (1973) Vegetation of the Earth in Relation to
Climate and Ecophysiological Conditions. London: English
University Press.
Wardle P (1974) Alpine timberlines. In: Ivey JD and Barry R (eds.) Artic
and Alpine Environment, pp. 371 402. London: Meuthuen
Publishers.
166
Biological Wastewater Treatment Systems
Biological Wastewater Treatment Systems
M Pell, Swedish University of Agricultural Sciences, Uppsala, Sweden
A Wörman, The Royal Institute of Technology, Stockholm, Sweden
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Life and Nutrient Transformation Processes
Biological Wastewater Treatment Systems
Perspective on Biological Wastewater Treatment
Further Reading
Introduction
Hence, globally, WWT probably is the most common
biotechnological process.
Though the same biological processes are the basis
for most WWT systems, the number of technological
solutions for achieving the goal probably is innumer
able. The numbers of techniques are as many as there
are sanitary engineers. However, the techniques may be
categorized as follows: (1) soil filters and wetlands –
terrestrial ecosystems working as natural filters; natural
water courses, lakes, and wetlands; soils receiving irri
gated wastewater; constructed wetlands and ponds; soil
or sand absorption systems; and trickling filters; and
(2) treatment plants – rotating biological contactors;
fluidized beds; and activated sludge systems including
sequencing batch reactors (SBRs). This array of techni
ques describes the systems on a scale from natural
ecosystems at one end to high technology solutions at
the other end. In the choice of WWT system to be
used many factors have to be considered like influent
water characteristics, desirable effluent water quality,
costs for building and maintenance, and population
density and dimensioning.
In this article we have chosen first to give a general
background on the microbial cell and biological processes
important in all WWT and, second, to focus on the impor
tance of understanding the interaction between hydraulic
performance and microbial processes to achieve effective
nitrogen removal, and third, to outline the function of two
common systems: the constructed wetland, requiring in
depth knowledge on hydraulic properties, and the activated
sludge process, relying on advanced control and optimiza
tion. Finally, we give some perspectives on the future
development of biological WWT systems and their use.
Eutrophication of water courses, lakes, and marine envir
onments is a major issue in most parts of the world.
Looking back 150 years the urban situation in the emer
ging industrial part of the world led to the introduction of
water based systems for conveying and discharge of sew
age. At first the wastewater was disposed into nearby
watercourses and lakes. As the populations grow, this
was not a sustainable solution – the natural wetlands
became overloaded as evident from the odors. This
untenable situation led to the development of more active
treatment systems like shallow ponds and sand filters. In
1914 the activated sludge technique was introduced by
Arden and Lockett, a technique that still probably is the
most common technique for wastewater treatment
(WWT) in the industrial part of world. In the 1960s
eutrophication became evident due to the high amounts
of plant nutrients discharged from sewage treatment
plants. The first and maybe the simplest solution was to
remove phosphorus by chemical precipitation. The
European Commission and national authorities have gra
dually over the latest couple of decades sharpened the
treatment demands, especially with regard to nitrogen, in
order to avoid further eutrophication in the sea. Hence,
WWT today probably is more focused on removing
phosphorus and nitrogen than pathogens. It is still argued
whether phosphorus or nitrogen is limiting for the eutro
phication process, that is, should either one or both of
these elements be eliminated.
Simply put, biological WWT can be defined as a
natural process in which organisms assist in environ
mental cleanup simply through their own life
sustaining activities. By studying the organisms in
natural ecosystems the biologists have explored their
function and capacity to degrade organic matter and
transform nutrients. Such information has then been
used by engineers to design effective WWT systems,
that is, the biological processes have been concentrated
into well regulated units. In addition, knowledge of
geochemistry, hydrology, etc., is essential component
of a successful system for treating polluted waters.
Life and Nutrient Transformation
Processes
The Cell
The cell is the smallest independent unit in all living
organisms. The cell can also form an individual organism
itself. Such organisms are referred to as microorganisms as
Biological Wastewater Treatment Systems 167
they are not visible to the naked eye. Examination of the
internal structure of the microbial cells reveals two struc
tural types: the prokaryote (Bacteria or Archea) and the
eukaryote (Eukarya) (Table 1). The previous group
includes the bacteria while the latter contain protozoa,
fungi, algae, plants, and animals. Prokaryotic cells have a
very simple structure. They lack a membrane enclosed
nucleus and they are very small, typically being from less
than 1 mm up to several micrometers. Eukaryotic cells are
generally larger and structurally more complex. They con
tain a membrane enclosed nucleus, and several membrane
enclosed organelles specialized in performing various cell
tasks. The morphological differences between the two cell
types have profound effect on their capacities to absorb and
transform nutrients and energy. The prokaryotes have a
large surface in relation to their volume meaning short
transportation distances within the cell not hindered by
complex membrane systems. Their potential to transform
and take up nutrients as well as to grow is very high; hence,
they can be said to be tailor made for high metabolic rates.
Some bacteria may under optimal conditions multiply by
binary division every 20 min. This will result in a rapid
exponential increase in cells.
For its growth the cell needs energy, carbon, and
macronutrients like nitrogen and phosphorus, and several
elements in minor amounts. In addition, an adequate
environment is needed, with oxygen, water, temperature,
and pH being the most important regulators. Most micro
organisms are heterotrophs and organotrophs meaning
that they derive their energy and carbon, respectively,
from organic molecules (Table 2). Other energy options
available are inorganic chemicals (lithotrophs) and light
Table 1 Cell types and some typical characteristics
Prokaryotic
Characteristic
Morphology and genetic
Cell size
Cell wall components
Cell membrane lipids
Membrane-enveloped
organelles
DNA
Plasmids
Biochemistry and
physiology
Methane production
Nitrification
Denitrification
Nitrogen fixation
Chlorophyll-based
photosynthesis
Fermentation end
products
Eukaryotic
Bacteria
Archaea
Eukarya
Small, mostly
0.5–5 mm
Peptidoglucane
Larger, mostly
5–100 mm
Absent, or cellulose or chitin
Ester-linked
Absent
Small, mostly
0.5–5 mm
Protein,
pseudopeptidoglucane
Ether-linked
Absent
One chromosome,
circular, naked
Yes
One chromosome,
circular, naked
Yes
No
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
No
No
No
No
Yes
Diverse
Diverse
Lactate or ethanol
Ester-linked
Mitochondrion, chloroplast, endoplasmatic
reticulum, Golgi apparatus
Several chromosomes, straight, enveloped
Rare
Table 2 Characterization of chemotrophic organisms according to their need of carbon and energy
Type
Carbon
source
Examples of primary
electron donors
Energy metabolism
Lithotrophs
–
Respiration: O2, NO3 , NO2 , S0, SO24 , CO2
Organotrophs
–
NH3, NO2 , H2S, S0,
Feþ
2 , H2
Organic
Carbon metabolism
Autotrophs
Heterotrophs
CO2
Organic
–
–
–
–
, not relevant to this term.
Examples of terminal electron acceptors
Respiration: O2, NO3 , NO2 , SO24 , Fe3þ, CO2,
organic; fermentation: organic
168
Biological Wastewater Treatment Systems
(phototrophs). It is not uncommon that bacteria, like
plants, can use carbon dioxide as the carbon source (auto
trophs). Though the most common trait of living is
organo heterotrophic, virtually all combinations above
of energy and carbon derivation exist.
Classical taxonomy of microbes is based on phenotypic
characters like shape and size, and their relation to oxy
gen, as well as way of utilizing the carbon and energy
source. Two classical shapes of bacteria are the rod and
coccus, but filamentous and appendaged forms are also
common. In addition to the shape, production of different
enzymes is an important parameter in grouping and iden
tifying bacteria. Recent developments within the nucleic
acid based molecular biology have provided invaluable
tools in the systematic of life by genotypic characters. By
comparing nucleotide sequences of not known organisms
with the emerging database of sequence information,
unknown organisms can be identified and/or classified.
The Microbial Community
Aggregated microbial communities called flocs or biofilms
are the backbone of most WWT processes (Figures 1a
and 1b). The source of microorganisms is soil and sewage
coming in with influent wastewater. In the WWT system
the organisms are subjected to high selective pressure.
Those tolerating the new environment will develop and
even thrive to form the basis for an effective WWT pro
cess. In any system organic molecules due to their
chemical/energetic properties will accumulate at
(a)
Mineral particle
Bacteria
Protozoa
Filamentous Air bubble
bacteria
Organic
fiber
Polysaccharide matrix with
oxygen and chemical gradients
(b)
Fixed carrier
Bacteria
Protozoa
Filamentous Air bubble
bacteria
Organic
fiber
Figure 1 Structure of (a) activated floc and (b) biofilm on solid surface.
Polysaccharide matrix with
oxygen and chemical gradients
Biological Wastewater Treatment Systems 169
interfaces (gas/liquid or liquid/solid). Hence, these niches
will be the first to be colonized and microorganisms with
features for keeping the community tightly together, for
example, production of extracellular polysaccharides act
ing as glue, will dominate. The microbial community so
formed will consist of a web of different species of bacteria,
protozoa, and metazoa. Though present, fungi, algae, and
virus probably play a less important role. The communites
can be observed as sludge flocks or biofilms. Another
advantage of living in dense communities is that environ
mental gradients, for example, of oxygen and substrate, are
formed, allowing many types of organisms to share the
space. From the WWT point of view the cooperation of
micoorganisms will result in an effective degradation and
mineralization of organic matter.
Investigation of activated sludge flocs and biofilms
concerns the following issues: (1) morphology, that is,
size and shape; (2) composition, that is, internal structure;
(3) identification of microbial species; and (4) spatial
arrangement of microorganisms. Traditionally, the detec
tion of bacteria in wastewater is restricted to the ability to
culture them. However, it has become evident that most
organisms are unculturable which is the reason for our
limited knowledge of the microbial actors in WWT pro
cesses. Recent advances in molecular techniques have
supplied the means for examination community structure
and detecting specific organisms in complex ecosystems
without cultivation. Most techniques are based on nucleic
acid fingerprinting after amplification by the polymerase
chain reaction (PCR) of extracted DNA or RNA.
Examples of techniques used are amplification of riboso
mal DNA restriction analysis (ARDRA), denaturing
gradient gel electrophoresis (DGGE), and terminal
restriction fragment length polymorphism (T RFLP).
Microarray technology seems to be promising in cap
turing the taxonomical or functional structure of complex
ecosystems. In this technique a vast number of oligonucleo
tide probes of known genes can be attached (spotted) to the
surface of a glass slide. Extracted DNA or RNA from an
unknown sample is then applied to the microarray plate.
After hybridization the presence of target organisms will
appear as radiant or fluorescent spots. Moreover, the inten
sity reflects the concentration of the sequence. By
constructing a DNA microarray containing probes target
ing the 16S rRNA of several groups of nitrifying bacteria
the presence of Nitrosomonas spp. has been detected without
need for PCR amplification prior to analysis. However, the
technique failed to detect Nitrospira and Nitrobacter, but its
future potential was clearly demonstrated. Fluorescence in
situ hybridization (FISH) is an effective technique to detect
specific bacteria in complex microbial communities. By use
of confocal laser scanning microscopy (CLSM) FISH
images of nitrifying bacteria in biofilms of domestic waste
water have been analyzed. Where the C/N ratio of the
substrate was high, heterotrophic bacteria occupied the
outer part of the biofilm while ammonium oxidizing bac
teria were distributed in the inner part. As the C/N ratio
gradually decreased, the nitrifying bacteria began to colo
nize the outer layer.
The use of the molecular approaches discussed above
has drastically widened our knowledge on bacterial diver
sity in WWT systems. Until 2002 more than 750 16S
rRNA gene sequences derived from wastewater had been
analyzed and sequences affiliated to the Beta , Alpha , and
Gammaproteobacteria as well as the Bacteroidetes and the
Actinobacteria were most frequently retrieved. Many new,
previously unrecognized, bacteria have been detected, and
many more, without doubt, await identification. Although
some of the newly identified organisms can be attributed to
the flocculation process as well as the biological nitrogen
and phosphorus removal processes, most of them possess
unknown functions. Not until it is fully understood can the
potential of the biological component of the WWT system
be fully utilized.
Microbial Carbon and Phosphorus Processes
Respiration
Respiration is probably the process most closely asso
ciated with life and in WWT systems it is attributed to
a wide range of microorganisms such as bacteria and
protozoa. Respiration is the aerobic or anaerobic
energy yielding process where reduced organic or inor
ganic compounds in the cell serve as primary electron
donors and imported oxidized compounds serve as term
inal electron acceptors (Figure 2). During respiration the
energy containing compound descends a redox ladder
commonly consisting of the glycolysis, citric acid cycle
(CAC), and finally the electron transport chain. The
ultimate aim is to convert energy into proton gradients
and ATP. During the metabolic pathway various inter
mediate organic molecules are withdrawn to enter the
anabolic route, that is, building blocks incorporated into
new cell material. Roughly, in actively growing hetero
trophic cells, 50% of the substrate carbon will form new
cells while the other 50% will be released as mineralized
carbon dioxide (CO2). In a less strict sense respiration can
be defined as the uptake of oxygen while at the same time
Org-C (100%)
Biomass-C
(50%)
ADP
Carbon flow
ADP ATP
ATP
O2
e–
Electron flow
H2O
CO2 (50%)
Figure 2 Carbon and electron flow in aerobic respiration. Box
represents the microbial cell.
170
Biological Wastewater Treatment Systems
carbon dioxide is released. However, in the ecosystem,
CO2 is also formed by other processes such as fermenta
tion and abiotic processes, for example, CO2 release from
carbonate. In addition, several types of anaerobic respira
tion can take place where, for example, nitrate or sulfate
are used by microorganisms as electron acceptors; hence,
O2 is then not consumed as in aerobic respiration.
Precipitation and cellular uptake of phosphorus
Removal of phosphorus from the wastewater stream is a
common strategy to control eutrophication. The idea is to
limit this element in the ecosystem and hence starve the
organisms to avoid growth and increase in biomass. In all
cases, phosphorus is removed by converting the phos
phorus ion into a solid fraction.
Chemical orthophosphate (PO34 ) removal uses the
property of metal ions like Al3þ, Ca2þ, Fe2þ, Fe3þ, or
Mg2þ to effectively react with phosphorus and form
stables precipitates, under specific sets of pH. These
ions may be naturally present in some soils and, hence,
the phosphate will be adsorbed to surfaces. Alternatively,
chemicals containing these ions can be added to the
WWT system and the precipitate formed mechanically
removed after having settled. Not only phosphorus is
affected by the chemical addition, the pH may also
change and the content of organic matter in the water
may be reduced. Both these events will affect the micro
bial activity in the system.
An alternative to chemical precipitation is to employ
plants, macrophytes, microalgae, or bacteria, or combina
tions of these, to concentrate the phosphorus. All cells
need phosphorus and the uptake of this element is part of
the natural metabolism. Phosphorus is an essential com
ponent of nucleic acids and phospholipids are located in
the various cell membrane systems. In addition, the pH of
the cell is regulated by a phosphate buffer system.
Therefore, phosphorus is needed in high quantities and
the cell normally constitutes 1–3% phosphorus per gram
dry matter. To achieve real removal of phosphorus the
produced biomass must be harvested.
PHA
Poly-P
In the activated sludge process under certain condi
tions, it may be possible to enhance the storage capacity of
highly energy rich polyphosphate by the bacterial bio
mass. Under anaerobic conditions in the WWT reactor
principally acetate, but also other volatile fatty acids
(i.e., fermentation products) are taken up and incorpo
rated in biopolymers like poly
hydroxyalkanoate
(PHA) or glycogen (Figure 3). In the anaerobic stage
the level of polyphosphate in the cell decreases while at
the same time soluble phosphate is released. When con
ditions are changed to aerobic and carbon poor, the
stored reserve of PHA is used as an energy and carbon
source for uptake of even larger amounts of phosphorus
than previously released to the system. The concentration
of phosphorus in polyphosphate accumulating (PAO)
bacteria can then be increased up to >15%. In the end
of a successful process the buoyant density of the sludge
should have increased. The polyphosphate forms dense
granules that can be stained and easily observed under the
microscope. The ecological mechanisms selecting for
polyphosphate accumulating organisms are not clearly
understood. Originally strains of Acinetobacter were
thought to be the key players in the process. The role of
Acinetobacter has been argued against as recent molecular
biology based tools for identification of bacteria have
demonstrated that other bacteria, for example,
Rhodocyclus spp., Dechloromonas spp., and Tetrasphaera
spp.,
may
dominate
the
polyphosphate
accumulating community.
Nitrogen Transformation Processes
In microbial ecosystems nitrogen is of special interest, as
it can exist in several oxidation levels ranging from
ammonium/ammonia ( III) to nitrate (þV). Moreover,
the transitions of nitrogen, both oxidation and reduction,
are mediated mostly by microorganisms and, in particu
lar, bacteria. When transformed, the nitrogen compounds
may serve as building blocks in the cell, as energy sources,
or as a way of dumping electrons.
PO43–
Acetate
–O2
PHA
Poly-P
Energy
Energy
PO43–
Acetate
+O2
Figure 3 Release and uptake of phosphorus by polyphosphate-accumulating bacteria under varying oxygen status. Shaded boxes
represent bacterial cells, Poly-P is polyphosphate, and PHA is poly -hydroxyalkanoate.
Biological Wastewater Treatment Systems 171
Mineralization and immobilization
Virtually all microorganisms can mineralize and immobi
lize nitrogen, and the processes are more or less
independent of oxygen. Proteins and nucleic acids, being
the two dominating macromolecules in the cell, contain
nitrogen as an essential component. Thus, most organic
matter contains at least some nitrogen. By predation or
after cell death and lysis the nitrogen containing mole
cules will be released (Figure 4a). However, due to their
molecular size, they cannot be directly taken up and
immobilized by new bacteria. Growing bacteria exudes
so called exoenzymes that attack and degrade the macro
molecules into smaller portions: amino acids and ammonia
that can be transported through the cell membrane. The
fate of the nitrogen part will depend on the nitrogen and
carbon status of both the cells and the environment. In a
carbon rich environment with high ratios of carbon to
nitrogen (>20) all nitrogen will be assimilated, that is,
immobilized in the cell. If the ratio is low (<10), nitrogen
will be mineralized and released to the environment. The
release of ammonium might also lead to an increase in pH
due to the alkaline properties of ammonia (NH3).
two groups of specialists within the bacterial family
Nitrobacteriaceae. Earlier Nitrosomonas europaea and
Nitrobacter spp. were thought to be the ammonium and
nitrite oxidizers in WWT systems. Molecular tools for
analysis of the amoA gene present in all ammonia oxidizers
have lately revealed a wide variety of different proteo
bacterial ammonia oxidizers (AOBs, ammonia oxidizing
bacteria) in nitrifying WWT processes. In addition, it
seems that Nitrospira like microorganisms and not
Nitrobacter spp. are the dominating nitrite oxidizers.
Through oxidation of the mineral nitrogen the bacteria
derive energy for growth, that is, fixation of carbon diox
ide into their biomass. In addition, most nitrifiers are strict
aerobes, that is, they are completely dependent on oxygen
in their respiration. Being both lithotrophic and auto
trophic, they have a complex cell machinery including
an extensive system of internal membranes, leading to
both slow growth and sensitivity to environmental dis
turbances. One consequence of disturbance on the AOBs
can be the formation of nitric or nitrous oxide that will be
emitted to the atmosphere. In addition, AOBs produce
protons which will lower the pH somewhat.
Nitrification
Denitrification
In lithotrophic nitrification, ammonia is stepwise oxidized
first via hydroxylamine (NH2OH) to nitrite (NO2 ) and
then further to the end product nitrate (NO3 )
(Figures 4b and 4c). The two steps are carried out by
Denitrification is the anaerobic respiration process in
which nitrogenous oxides, principally nitrate and nitrite,
are used as terminal electron acceptors and, hence, reduced
into the gaseous products nitric oxide (NO), nitrous oxide
(a) N-mineralization–N-assimilation
Lysis and release Org(3)-N
Org(1)-N
NH4+
Org(2)-N
Mineralization
(c) Nitrite oxidation
NH4+
Org(3)-N
Assimilation
Release
NH4+/ NH3
(b) Ammonia oxidation
CO2
Org-C
NO3–
NO2–
NO2–
Ammonia
monoxygenase
+O2
NH3
NH2OH
Hydroxylamine Org-C
oxidoreductase
Nitrite
oxidoreductase
±O2
CO2
NO2–
NO3–
Org-C
Nitrate
reductase
N2O
NO
reductase reductase
NO3–
Org-C
NO(g) N2O(g) N2(g)
NO2–
Nitrite
reductase
NO(g)
(d) Heterotrophic denitrification
N2O(g) N2(g)
CO2
Org-C
NO2–
NH4+
NO3– (10%)
N2
–O2
N2(g)
(e) Anammox
Figure 4 Five microbial nitrogen transformation processes leading to nitrogen removal in the ecosystem. For each process its relation
to the environmental oxygen status is given. Shaded boxes represent microbial cells.
172
Biological Wastewater Treatment Systems
(N2O), and dinitrogen (N2) (Figure 4d). The process is
controlled by factors such as pH, supply of organic carbon
and mineral nitrogen, and aeration. Normally, dinitrogen
dominates the end product but under conditions not opti
mal for the organisms nitrous oxide can constitute a
considerable fraction. Nitrous oxide contributes to the
global warming as well as to the depletion of the ozone
layer in the stratosphere. Denitrifying capacity is repre
sented within most taxonomical and physiological groups
of bacteria. Denitrifiers are facultative anaerobes, that is,
they prefer oxygen as terminal electron acceptors in their
respiration but upon depletion of oxygen they rapidly
switch into the use of a nitrogenous oxide. Denitrifiers, of
which most are organotrophs and heterotrophs, are known
to use only the easily available fraction of the organic
matter. It should also be mentioned that some bacteria
like Paracoccus denitrificans have been found to denitrify
under aerobic conditions. In addition, the AOB
Nitrosomonas europea may, under certain conditions, deni
trify aerobically, resulting in dinitrogen gas and nitrite as
end products. Conventional nitrogen removal techniques
are based on a combination of autotrophic nitrification and
denitrification.
Anaerobic ammonium oxidation
The nitrogen pathways in bacteria are more complex than
earlier thought. It was recently discovered not only that
some ammonium oxidation bacteria can denitrify, but also
that lithoautotrophic bacteria belonging to the phylum
Planctomycetes can perform anaerobic ammonium oxida
tion (anammox); they oxidize ammonium with nitrite as
the electron acceptor to yield dinitrogen gas (Figure 4e).
Lipids of anammox bacteria contain a combination of
ester linked (typical of the Bacteria and Eukarya) and
ether linked (typical of the Archea) fatty acids. Like the
autotrophic nitrifiers, internal lipid membranes of anam
mox bacteria are essential to create proton gradients.
While the membranous compartmentalization in nitrifiers
is arranged as stacked lamellae, in the anammox bacteria
the compartmentalization involves a large single vesicle,
called anammoxozome.
Coupling between Microbiology and Water
Circulation
Although they rely on the same biological principles,
treatment plants and wetlands for WWT operate very
differently regarding flow of water through the system.
Treatment plants can generally be considered to act as a
well mixed reactor in which the contact between solutes
and the microbial flocs is sufficiently effective not to limit
the reactions. The mixing is assured by mechanical stir
ring in the water circulating through relatively narrow
and long stretched basins. Because wetland systems rely
on natural mixing of the water, it is essential to have an
appropriate planning of the wetland shape, bottom bathy
metry, and placement of vegetation – factors that control
the water circulation on its way through the wetland.
A canalized flow through the center of a wetland pond
with vegetation on the side along the flow channel implies
a separation of the main flowing water from the host
environment for biofilms, that is, the vegetation stems.
Introducing deep zones in the bottom or vegetation zones
transverse to the flow direction can counteract this. Both
measures even out differences in pressure head of the
water flow, which should provide a more uniform flow
and better utilization of the entire wetland volume.
Generally, it is also better to place the in and outlet in
such a way that the water flow is as stretched out as far as
possible. This can be arranged also by introducing hang
ing separation walls or dams.
Reaction Kinetics in Biological Treatment
Systems
Coupling of water circulation and microbiology
The reactions of phosphorus and nitrogen in treatment
systems involve kinetic processes related both to the
timescale (on the order of seconds) for circulation of
solutes and water as well as the kinetics of the biological
processes and sorption. This section gives a general the
oretical basis by which we can describe the interactions of
processes in a manner that principally links the scientific
basis of the individual mechanisms with the gross system
response.
As a representation of the solute and water circulation
we use the probability density function (PDF) of the
residence time, , for water between the inlet and the
outlet, f (). Models where residence time distribution and
equations for chemical transformations are coupled have
been used in many applications before, especially for
modeling chemical reactors, but also in natural stream
systems and wetlands. Basic assumptions include that
each flow path is characterized by a unique residence
time, , and no mixing occurs between the paths. Hence,
in the steady state case we can evaluate the average
response of the exit concentrations of several pathways
of different residence time :
hCðt Þi ¼
Z
0
1
Cðt ; Þf ðÞ d
½1
where C(t, ) is the concentration of nitrogen dissolved in
water at time t (i.e., [N]) and at the exit point of the
wetland defined by the residence time .
Recognizing eqn [1] as an approximation of the inte
grated system concentration response, we have divided the
problem in a multidimensional water flow problem to define
water residence time PDF f () and a one dimensional
problem of the nitrogen concentration response along
each flow path C(t, ). For a well mixed reactor of a
Biological Wastewater Treatment Systems 173
wastewater treatment plant (WWTP, see below), the all
water parcels stay equal time expressed by the ‘nominal’
water residence (detention) time, that is, the ratio between
flow volume V and discharge Q. Mathematically, this
means that f () ¼ ( V/Q), where is the Dirac delta
function. Hence, eqn [1] yields hC(t)i ¼ C(t, V/Q), sug
gesting that the average response is given by a single value
of the response concentration curve associated with a resi
dence time ¼ V/Q. Specifically, for a steady state reactor
with constant input and external constraints, we have
hC i ¼ C( V/Q).
Because of the complicated mixing conditions of
treatment wetlands, the nominal residence time is often
a poor approximation. Open water flow in two dimen
sions basically follows the so called Saint Venant
equations, that is, the depth averaged form of the
momentum equation. However, generally scientists and
engineers neglect inertia effects in the analysis of wetland
flows, which yields a significantly simpler mathematical
statement for the flow problem. A main problem is to
relate the friction losses to, for example, the distribution
of vegetation, in particular, since this controls the flow
pattern and solute mixing in wetlands. In addition to a
passive advection with the flow, solutes undergo mixing
such as dispersion and exchange with stagnant zones in
vegetation and bottom sediments. This leads to a calcula
tion procedure, demonstrated in Figure 5, that results in
a physically mathematically based estimation of the flow
residence times.
As a simpler alternative to represent the flow residence
times utilizes the fact that the residence time distribution
for treatment wetlands has been found to generally follow
(b)
450
350
where hi is the expected value of the residence time.
The lack of physical basis of the model implies that the
number of compartments M needs to be determined
explicitly from comparison of the model prediction and
results of tracer injections in the wetland such as those
shown in Figure 5d. Generally, it has been found that the
M value falls around 3. A main advantage of the functional
form is that it can easily be utilized in eqn [1].
Coupled enzyme kinetics and bacterial growth
Both nitrogen and phosphorus removal processes
described earlier involve several reaction steps and con
trolling factors that are nested in a complex manner. This
is why the treatment process often is described using
system analysis and automatic control. A strong charac
teristic in most biologically catalyzed reactions, however,
is the fundamental limitation on the process of the sub
stances carbon, phosphorus, and nitrogen. For both
traditional treatment plants and treatment wetlands, the
total nitrogen in water is often assumed to follow a reduc
tion with a type of Michaelis–Menten enzyme kinetics
according to
dC ð Þ
¼
dt
½3
in which q is the specific bacterial activity, X is the number of
bacteria, and K is a saturation coefficient or a critical
Streamlines
200
150
qX
CðÞ
K þ CðÞ
0.3
Inlet canal
Position (m)
250
½2
Simulated
flow residence
times
400
300
ðM=hiÞM M – 1 – M=hi
e
ðM 1Þ!
(d)
(c)
Outlet
f ðÞ ¼
Numerical
particles
released
here
A B C
100
50
Probability density (day s–1)
(a)
that of an idealized system consisting of M tanks in a
series with the same residence time in each tank:
0.25
Simulated
β=0
0.2
Simulated
β = 0.1
0.15
Simulated
β=1
0.1
Tracer experiment
0.05
Inlet
0
0
50
100 0
Position (m)
50
100
Position (m)
0
50
Position (m)
100
0
0
2
4
6
8
10
12
14
Residence time (days)
Figure 5 Figures exemplifying the procedure with which a physically based flow model can be used to derive the water residence time
distribution in a treatment wetland pond. (a, b) Measured bathymetry in the pond as well as an estimated vegetation distribution. This
information is used as input for the flow modeling that consists of two steps, one in which the water surface elevation and flow velocities
are calculated (not illustrated here), and one in which numerical particles are released at the pond inlet to determine the streamlines and
flow residence times (c). The simulated flow residence times can be compared with flow residence times determined from tracer
experiments (d). Partly reprinted after Kjellin J, Wörman A, Johansson H, and Lindahl A (2007) Controlling factors for water residence
time and flow patterns in Ekeby treatment wetland, Sweden. Advances in Water Research 30(4): 838–850.
174
Biological Wastewater Treatment Systems
concentration representing conditions when the bacteria are
in a transition between limiting and nonlimiting state. The
original Michaelis–Menten equation assumes that qX is con
stant, whereas here we assume that the number of bacteria
can vary with time. For a constant number of bacteria, K can
be seen as a constant saturation concentration. Equation [3]
approaches the zero order form dC/dt ¼ qX for nonlimit
ing conditions (i.e., C >> K) under which the nitrogen
concentration is not important for the reaction. Only the
amount of enzymes (number of bacteria) controls the reac
tion. Under limiting conditions (i.e., C << K), which is the
main interest for treatment wetlands, eqn [3] approaches the
first order form dC/dt ¼ (qXC )/K that reflects a control on
the reaction of the nitrogen concentration. The ratio (qX )/K
is the denitrification rate coefficient.
If there is steady state in reaction controlling factors,
like the number of bacteria, and we have nitrogen limiting
conditions, the nitrogen concentration decays with resi
dence time as
C ð Þ ¼ C0 e – k
½4
where C0 is the initial concentration.
The activated sludge process in a WTP is run under
carbon excess. By controlling the removal of the activated
sludge, the concentration of bacteria is kept on a fairly
constant level. Further, the circulation of wastewater
ensures approximately a constant residence time that bal
ances the nitrogen reducing processes between nitrogen
limiting and nonlimiting conditions. In treatment wetlands
with carbon excess due to sufficient carbon production in
plants, we can normally assume nitrogen limited conditions.
In the activated sludge there is just a single residence time
defined by the ratio between reactor volume and discharge,
that is, ¼ V/Q (see section ‘Respiration’). In a treatment
wetland, the average response needs to be evaluated for a set
of water pathways with specific residence times. The use of
eqn [1], in which we insert eqns [2] and [4], yields a com
monly used design formula for treatment wetlands:
Cout ¼
Z
1
0
C0 e – k f ðÞ d ¼ C0
M
k <> þ M
M
½5
where k is the first order volumetric coefficient for total
nitrogen reduction. From this formula we can see that the
most effective treatment is obtained when M approaches
infinity for so called plug flow.
The reduction kinetics expressed by eqn [3] is a simpli
fication of the several reaction steps (with multiple reaction
kinetics) that accounts for the growth and decay of the
microbial communities. In 1949 Monod proposed that the
bacterial growth rate unlimited by environmental factors
other than the substrate, such as concentration [N], follows
dX
S
X
¼ max
dt
S þ Ks
½6
in which S is substrate concentration (carbon concentra
tion), Ks is the substrate saturation coefficient, and max is
the maximum population specific growth rate constant. The
linear, first order differential equation for bacterial number
means that the bacteria grows exponentially with time as
long as the growth process is unlimited. A change of the
nitrogen load to a biological treatment system, thus, leads to
a change in the bacterial community and in the denitrifica
tion rate coefficient. Sometimes one can see variants of the
Monod growth rate formulation involving the nitrogen
concentration in another factor similar to the factor in
which substrate concentration is included in eqn [6].
Measurements of denitrification rates
Laboratory experimental techniques for determination of
potential denitrification activity (PDA) are based on inhi
biting the final denitrification step in which N2O transforms
to nitrogen gas. The PDA assay is prepared with an excess
of carbon (e.g., glucose) and nitrate sources as well as
inhibiting acetylene (C2H2). The preparation of the experi
ment implies, however, a growth of the bacterial population
according to eqn [6] as long as this growth is not limited by
the availability of nutrients. For relatively high nitrogen
concentration in the substrate the production in terms of
[N2O N] is not limited by the solute concentrations, but
only by bacterial growth. The corresponding zero order
reaction can be written as
d½N2 O N
¼ qX
dt
½7
where [N2O N] denotes the nitrogen concentration in the
form of N2O. Since the rate of mass production of N2O N is
assumed to be the same as the rate of mass reduction of total
N available in dissolved phases in water provided in the
assay, the qX factor in eqn [7] is the same as that in eqn [3].
Figure 6 shows how the production of N2O increases
during an initial phase (before phase marked as ‘1’), which is
more or less nonlinear depending on the relative increase of
the bacterial activity or population growth. Between the
phases ‘1’ and ‘2’, bacterial growth is for some reason limited
or much slower, but not due to limited nitrogen concentra
tion. Because of the nitrogen excess, the reactions zero
order up to phase ‘2’ where nitrogen becomes limiting. As
the nitrogen availability of the sample decreases, the reac
tion is successively first order controlled by the nitrogen
concentration according to eqn [4] with the number of
bacteria X existing at that time. In Figure 6, this first order
reaction phase is relatively short. In biological treatment
systems, the bacteriological composition reaches such an
equilibrium in the bacteria populations after a short while
due to limitation of carbon and/or nitrogen and follows
the second phase of the experimental results shown in the
figure. The PDA associated with the treatment system is
defined from the initial slope of the N2O production versus
Biological Wastewater Treatment Systems 175
900
2
800
700
Unlimited (zero-order)
nitrogen reaction
ppm N2O
600
First-order nitrogencontrolled reaction
500
1
400
300 Increase of bacterial activity or
200 growth
(NO3) = 0.3 mM
100
Slope = qX0
0
0
20
40
60
Time (min)
80
100
Figure 6 Example of laboratory measurement of kinetics in N2O
production with initially unlimited conditions in nitrate and carbon
involving a first-order growth of the bacterial population and
enzyme kinetics. The initial phase – before dashed line marked
with ‘1’ – is controlled by increased bacterial activity or growth.
Thereafter, the number of bacteria is constant due to the linear
increase of N2O production. As nitrogen becomes limiting, the
reaction is succeeded by nitrogen deficiency and a first-order
production rate. This result was obtained from a sediment sample
taken in Ekeby wetland, Eskilstuna, Sweden.
time relationship, since the initial amount of bacteria X0 in
the sample is believed to represent the state of the treatment
system in which the sample was taken.
At the transition to nitrogen limited conditions, the
product qX is the time rate of reaction of nitrogen in
units mass per unit time and volume, so the ratio qX/K
is the denitrification rate coefficient, where K is a limiting
nitrogen concentration at the transition.
A batch reactor undergoes all stages represented in
Figure 6, whereas the continuous flow system is generally
kept at steady state (see section entitled ‘Continuous flow
systems and SBRs’). In treatment wetlands with sufficient
carbon supply from moldering vegetation, denitrification
is nitrogen limited and first order controlled by the nitro
gen concentration. The denitrification rate coefficient is
given by qX0/K.
Biological Wastewater Treatment
Systems
Treatment Wetlands
Classification of wetlands
There are several types of natural wetlands such as
swamps, fens, bogs, marshland, and tidal freshwater
areas. Swamps and marshes have open flowing water
and are distinguished in terms of vegetation, soil type,
and wild life. Mires such as fens and bogs are mainly
subsurface wetlands with little open water. Bogs are iso
lated hydrological units that receive water only through
precipitation, whereas fens have through flowing water.
As water goes through such wetland areas it undergoes
great chemical transformation. Both nutrients and ele
ments like heavy metals that attach chemically (sorb) on
solid surfaces are effectively removed such that water
reaches a status corresponding to ‘natural’ water quality.
Constructed wetlands or treatment wetlands are
usually built where natural wetland conditions can be
found and are, therefore, to some extent modifications of
a natural system. By introducing dams and canals, how
ever, it is possible to provide proper water depth for
carbon providing vegetation species, like common reed
(Phragmites), and a separation of oxygen conditions. In
some cases, wetlands can be built in clay strata or artifi
cially sealed using clay even if there is no natural
groundwater reaching the ground surface. Hence, leakage
through infiltration is an essential problem that needs to
be accounted for in the design.
Constructed wetlands are commonly divided into
subsurface flow (SSF) and surface flow or free water sur
face (FWS) wetlands. Both types are used for treating
domestic, municipal, and industrial wastewater. In parti
cular, these systems can be useful for treating landfill
leachates, agricultural runoff, and wastewater from
minor communities. SSF wetlands are commonly used
as a polishing step after conventional treatment plants
for municipal wastewater. SSF systems are often favored
in minor communities due to the soil cover of possibly
contagious wastewater. Subsurface systems require
separation of solid material in the wastewater before
solute fractions are led into a sand filter or other soil
layer in which phosphorus is removed through sorption
to the particulate matrix and nitrogen to denitrification
supported by soil bacteria (Figure 7).
Because of the large discharge capacity surface flow
wetlands are usually preferred as polishing step for muni
cipal wastewater. Phosphorus is generally effectively
removed in the treatment plant, whereas nitrogen treat
ment requires longer detention times that are provided in
the wetland.
Both FWS and SSF wetlands used for treating muni
cipal and industrial wastewater are designed with an area
of c. 5–10 m2 per person equivalent.
Functionality of FWS wetlands
Vegetation is important in surface flow wetlands to pro
vide carbon for supporting denitrification, to offer host
environment for biofilms that grow on stems, and to cause
friction losses for flow water, which can be utilized to
provide a beneficial flow pattern. Submersed vegetation
also controls the oxygen level in the water.
Vegetation in a recently established wetland changes
with time and this leads to a relatively long period (years)
to approach equilibrium in the wetland ecosystem and the
interacting treatment processes. In cold climates, the effec
tiveness of treatment wetlands also varies over the year,
but there is a notable effect even during the winter.
176
Biological Wastewater Treatment Systems
Effluent wastewater
Septic tanks
Recipient water
Sand filter
Figure 7 Typical subsurface treatment system for a single household including septic tanks and downstream sand filter.
The shores of wetland ponds in cold climates can be
populated by reed sweet grass (Glyceria maxima L.), com
mon reed (Phragmites australis L.), and cattail (Typha
latifolia L.). Examples of submersed vegetation include
slender waterweed (Elodea nuttallii L.), sago pondweed
(Potamogeton pectinatus L.), coontail (Ceratophyllum demersum
L.), and spiked watermilfoil (Myriophyllum spicatum L.).
Coontail forms dense layers of vegetation that can be
considered to be a porous medium for the water flow
with a large inner surface available for biofilms.
A main role of treatment wetlands that are constructed
as a polishing step after a conventional treatment plant is
to remove nitrogen through denitrification in biofilms.
Generally, the ordinary treatment process has included
oxygenation of the water, which transforms most nitrogen
fractions, like ammonium and nitrite, to nitrate before it
enters the wetland. Biofilms grow both on vegetation
stems and in the bottom sediments. Therefore, an impor
tant factor is the exchange rate for solute substances
between flowing channels in the wetland with bottom
sediments and vegetation zones. The potential for deni
trifications in the host environments for biofilms, like
bottom sediments, is usually significantly higher than
actual rates on the scale of the entire system, because of
the difficulty to arrange an effective contact between
water and biofilms.
Functionality of SSF wetlands
The advantage of SSF wetlands is that the water is pre
sent below the ground surface, which decreases odors and
the risk for public exposure for possible contagious bac
teria. The construction usually includes a sand and/or
gravel bed with supporting emergent vegetation such as
cattail (Typha) and reeds (Phragmites). The systems are
designed with aspect ratio (L:W) of about 15:1 and a
flow velocity in the order of centimeters to decimeters
per day.
A typical design layout for a single household is shown
in Figure 7. A first step usually involves separation of
coarse fractions of the wastewater in deposition basins, or,
such as in this case, in septic tanks. This produces waste
water that can percolate and flow through the sand filter
without rapidly clogging the pores of the filter and end its
lifespan too fast.
The active processes include mechanical filtering of
particulate (organic) matter in the porous material, sorp
tion of phosphorus and heavy metals to the solid matrix,
and nitrogen decomposing reactions caused by nitrifying
and denitrifying bacteria in the upper soil layer. Good
performance is commonly reported for the removal of
biological oxygen demand (BOD), total suspended solids
(TSS), phosphorus, and nitrogen.
WWTPs – The Activated Sludge Process
General
Generally, WWT systems containing compartmenta
lized reactors (basins or tanks) for their performance
often are termed a WWTP. In addition, the flow of
wastewater through such systems is thoroughly con
trolled and optimized. The WWTP may consist of a
mechanical, chemical, and biological step. In the
mechanical step, heavy solid particles are allowed to
settle at the bottom and light material floating on the
water surface is removed. In the chemical step metal
salts are added to precipitate phosphorus. Phosphorus
removal can be performed at different stages in the
treatment process: prior to, simultaneous with, or after
the biological step and are hence called preprecipitation,
coprecipitation, or postprecipitation, respectively. The
biological step can be performed according to either of
two basic principles. The reactor may contain solid
surfaces to support bacterial growth and the develop
ment of a biofilm (trickling filters, rotating biological
contactors, and fluidized beds) (Figure 1b). The other
approach is to allow bacterial growth in the water body
supported by natural occurring suspended solids (acti
vated sludge process) (Figure 1a).
The activated sludge process can be designed as either
a continuous flow system or as SBRs. Both systems nor
mally include an aerated biological nutrient removal step
followed by settlement of produced sludge. The differ
ence between the systems is that in the continuous
Biological Wastewater Treatment Systems 177
process these processes take place in two different reac
tors, whereas in the SBR process they occur sequentially
in the same reactor.
Continuous flow systems and SBRs
In the conventional continuous flow system primary trea
ted wastewater is conveyed into an aerated basin
(Figure 8a). The feed of wastewater and supply of com
pressed air can be done in many ways, from being
introduced at one end leading to gradients of oxygen
and substrate throughout the basin to being introduced
at several points giving a more homogeneous environ
ment. Moreover, during its way through the basin, the
water may be led through more or less open compart
ments. In a completely mixed process, the water is also
circulated within the basin. The effluent is led to a clari
fier to allow particles to settle before the clear phase
leaves the process. The settled excess sludge containing
a viable biomass is removed and treated separately; how
ever, some is recycled to reinoculate the process. This
procedure will ascertain stable function of the unit. The
whole concept is not unlike the continuous culturing of
microorganisms in the laboratory or in many industrial
processes.
The operation of one or more SBRs in a series consists
of a sequence of fill and draw cycles. Each cycle typically
consists of a number of separate operational phases of fill,
react, settle, draw, and idle. Hence, after the react phase,
that is, growth phase, the produced biomass is allowed to
settle and the clear treated supernatant can be removed.
The process resembles that of batch culturing of bacteria
in the laboratory.
The biological process
In the biological step of the activated sludge process,
suitable mixing of the water is necessary to allow sus
pended solids, air, nutrients, and microorganisms to make
intimate contact. Based on the mixing regimes, plug flow
or completely mixed systems can be differentiated. Large
volumes of air are blown into the reactor tanks from
beneath to achieve effective mixing and support the aero
bic microorganisms with sufficient oxygen for respiration.
The concentration of dissolved oxygen (DO) should be
kept at approximately 2 mg l 1. As described above, the
(a)
Influent
Effluent
Aerobic
zone
Return sludge
Excess sludge
(b)
Influent
Aerobic
zone
Anoxic
zone
Return sludge
Effluent
Excess sludge
(c)
Influent
Anaerobic
zone
Anoxic
zone
Recirculation
Aerobic
zone
Effluent
Excess sludge
Return sludge
Figure 8 Examples of conventional activated sludge processes (a) without and (b) with nitrogen removal capacity, and (c) with
capacity of both biological nitrogen removal and enhanced biological phosphorus removal.
178
Biological Wastewater Treatment Systems
phase transitions in the system will support an optimal
environment for microbial growth. During the react
phase, three dimensional aggregates of highly active
microbial communities, called flocs, are formed
(Figure 1a). Flocs typically are 100–500 mm in diameter.
New microscopic techniques such as epifluorescence
and CLSM in combination with image analysis have
been used to analyze the aggregates of activated sludge.
Generally, four main structures can be discriminated in
the flocs: (1) active and inactive microbial cells, mainly
bacteria, protozoa, and metazoa; (2) extracellular poly
meric substances like carbohydrates and proteins;
(3) inorganic particles (sand); and (4) water. From a tech
nical point of view, the sludge properties are essential.
The forming of dense flocs with good properties for
settlement will provide good operational conditions.
Filamentous bacteria will always be present in a healthy
process which operates normally, and which shows no
symptoms of problems with bulking or foaming. Several
sludge indices have been suggested to describe and char
acterize the sludge property.
Occasionally the proportion of filamentous bacteria will
increase causing flocs of loose structure that settle only
slowly and compact poorly. This phenomenon called bulk
ing leads to uncontrollable loss of solids, including, for
example, the active nitrifying biomass. Though most fila
mentous bacteria are heterotrophs, they are shown to be
extremely difficult to cultivate. Approximately ten types of
filamentous microbes seem to be involved in most bulking
events. Microthrix parvicella seems to be especially important.
The bacterium is long and thin, and its coiled appearance
makes it easy to distinguish by microscopy of activated
sludge samples. Only metabolic characteristics of a few
isolates have been reported and the results are not always
concordant. Microthrix parvicella seems to be negatively
affected by DO concentrations above >6 mg l 1 but grows
well at 0.4 mg l 1 and should therefore be considered a
microaerophile. It prefers a somewhat alkaline environment
and optimum growth is reported at 25 C, though some
growth was still observed at 8 C. The range of maximum
growth rates (max) reported is 0.38–1.44 d 1. The bacter
ium cannot utilize glucose but seems to prefer long chained
fatty acids like oleic acid. It can store intracellular PHA and
lipids. No reliable control strategy exists for bulking caused
by increased amounts of filamentous organisms in the acti
vated sludge process. Based on the physiological properties
of the bacterium, the following alteration of the process has
been suggested to reduce its abundance: shorten the sludge
retention time, increase the DO to >2 mg l 1, removal of
high lipid contents by flotation.
Another widespread problem also leading to solids
separation problems is foaming. Stable foams will bring
the sludge to the surface of the clarifier and carry over
of solids from the clarifier. The foam most often consists
of a dense matrix of filamentous bacteria and air bubbles.
Foaming may have several causes. Microthrix parvicella
seems to be more hydrophobic than most other bacteria
in the activated sludge process and are frequently asso
ciated with foaming problems. Another group of bacteria
identified in activated sludge foams is mycolic acid
producing actinomycetes. The most commonly methods
for controlling foaming are the same as those for con
trolling bulking problems. However, the magnitude of
the problem has forced the development of both physical
and chemical short term measures to control these
situations.
Nutrient removal capacity
A properly controlled activated sludge process can
remove very effectively the content of organic carbon,
and mineralize and nitrify nitrogen. Typically, the che
mical oxygen demand (COD) and BOD removal
capacities for municipal wastewater are higher than
85 and 95%, respectively. The reduction of carbon is
due to aerobic respiration losses, removal of settled sludge
produced by biomass growth, as well as flocculation of
dissolved and particulate organic matter. In addition,
some 20–30% each of influent phosphorus and nitrogen
will be trapped in the settled sludge; however, most
phosphorus and nitrogen will leave the system as dis
solved phosphate and nitrate. Thus, the basic design of
the activated sludge process is less effective in reducing
nitrogen and phosphorus. By introducing chemical pre
cipitation and combined nitrification–denitrification the
total removal capacities for phosphorus and nitrogen may
be improved to >90 and 70%, respectively.
The high amounts of sludge produced by activated
sludge systems are problematic. Although sludge is a
potential ‘organic fertilizer’, since it is rich in plant nutri
ents, due to the risk of occurrence of pathogens and
chemical toxicants, such as heavy metals in the sludge,
there are problems associated with recycling the sludge to
arable land. Therefore, efforts are made to reduce the
sludge production. Increasing the periods of aeration
will lead to higher sludge residence time which will
extend the periods of endogenous metabolism, that is,
microbial consumption of internal cell material as well
as mineralization of lysed cells and particulate matter.
Application of aquatic predatory oligichaetes has been
suggested as means to reduce excess sludge production.
One common means to reduce the amounts of sludge
from WWTPs is to treat the sludge in an anaerobic
reactor to produce biogas (CO2 and CH4).
Enhanced nitrogen and phosphorus reduction
The combination of nitrification and denitrification has
since long been known as an effective biological solution
to achieve nitrogen removal in wastewater. The obvious
way to arrange suitable environments for the two groups
of bacteria is to connect an aerobic compartment or zone
Biological Wastewater Treatment Systems 179
prior to an anoxic in a so called post denitrification pro
cess (Figure 8b). However, since most organic matter is
consumed in the aerobic zone, this setup may experience
low effects due to lack of easy available energy to
the denitrifiers. A more effective solution can be to
place the anoxic zone prior to the oxic zone and circulate
the water between the two zones. In this design, called
pre denitrification, the denitrifiers will meet both anoxic
conditions and fresh organic material from the influent.
Another solution is to support the denitrification with an
external organic energy source. Effective denitrification
has been reported with, for example, acetate, ethanol, and
methanol. The response to acetate and ethanol is immedi
ate as these molecules are part of the normal metabolic
pathways of organotrophic bacteria. For effective
denitrification with methanol a long period of adaptation
is needed, typically several months. Only a few slow
growing specialists, for example, Hyphomicrobium sp., can
use one carbon compounds (CH3OH) and the metabolic
pathways are complex.
Recent developments in biological nitrogen removal
techniques in combination with the discovery of novel
bacteria have resulted in some new methods. By combining
partial nitrification with the anammox process some nitro
gen removal techniques have been set up that may consume
lower resources (Figures 4b and 4e). In the partial nitrifica
tion process a shortcut is taken by preventing the oxidation
of nitrite to nitrate by nitrite oxidizing bacteria. Instead the
nitrite is removed directly by heterotrophic denitrification.
In the single reactor system for high ammonium removal
over nitrite (SHARON), incomplete nitrification is
achieved by use of the slower growth rate of nitrite oxidi
zers than ammonium oxidizers at higher temperatures
(>26 C). By applying higher hydraulic retention times,
the nitrite oxidizer will be washed out. The nitrite thus
accumulated can be removed by the anammox process in a
succeeding reactor. In the anammox process nitrite is oxi
dized with ammonia as the electron donor. In the partial
nitrification process, half the ammonium is converted into
nitrite. One advantage with the process is that no extra
organic energy is needed for the denitrification step.
Another variation is to let nitrifiers oxidize ammonia to
nitrate in a single reactor and consume oxygen to create
the anoxic conditions needed by the anammox bacteria.
This process is called CANON, the acronym for ‘comple
tely autotrophic nitrogen removal over nitrite’.
As both biological nitrogen removal and enhanced bio
logical phosphorus removal need alternating cycles of
aerobic and anoxic conditions, it seems logical to combine
the two processes in the same WWTP. However, this is not
as easy as it seems to be. In addition to alternating anoxic
and aerobic regimes, the anoxic zone must be maintained
completely anaerobic to provide fermentation end pro
ducts like fatty acids to select for PAO bacteria. The level
of nitrate in the anaerobic zone must be low; otherwise the
heterotrophic denitrifiers will consume the organic mole
cules needed by the PAO bacteria. In the so called three
stage PHOREDOX process, influent water is fed to an
anaerobic reactor, and then conveyed to an anoxic reactor
also fed with recycled activated sludge from the last aero
bic reactor (Figure 8c). In this way less nitrate is returned
with sludge from the clarifier to the head of the system.
Thus, both phosphorus and nitrogen removal are accom
plished by this design.
Regulation and simulation models
The activated sludge process does not only involve com
plex elements but also the influent wastewater
characteristics vary temporarily. This emphasizes the
need for thorough control and optimization to maintain
and fine tune the process performance. To describe the
actual WWTP, a general model including the ensemble
of an activated sludge model, hydraulic model, oxygen
transfer model, and sedimentation tank model can be
used. The activated sludge model describes the biological
reactions occurring in the process by a set of differential
equations. In addition to use in control and optimization, a
WWTP model can be used to simulate different scenarios
for learning or to evaluate new alternatives for design.
Strengths and weaknesses of WWTP
In its basic design the activated sludge process has a high
capacity to biologically oxidize carbon and nitrogen.
In addition, this is achieved in comparable small units,
that is, less space is needed, which most often is a pre
requisite for WWT in urban areas. By modifying the
design also high amounts of nitrogen and phosphorus
can be removed by biological processes. The SBR process
is both a stable and flexible activated sludge process. The
biomass cannot be washed out and the possibility to
handle shifts in organic and hydraulic loads is good. In
addition, less equipment and operator attention are
needed to maintain the SBR process.
WWT by the activated sludge process must be regarded
as a highly technological process, that is, much knowledge
and experience are needed to operate a system based on
this technique. In the process design of activated sludge
processes, much focus has been put into efficiency in
nutrient removal. Although generally pathogens are accep
tably removed, most WWTPs are not designed for treating
pathogenic microorganisms. Moreover, the environmental
selective pressure on the microbial communities probably
leads to highly specialized ecosystems. Consequently, the
treatment process may be sensitive to disturbances due to
environmental variations such as sewage load and compo
sition as well as influent toxicants. The costs for
maintenance and care are high. The nitrogen removed
from the system is left as gaseous emissions instead of
using such a valuable plant nutrient in crop production.
In addition, the plant nutrient rich sludge may contain
180
Biological Wastewater Treatment Systems
heavy metals as well as anthropogenic organic pollutants
that may pose a risk to the ecosystem and must therefore
most often be deposited or possibly incinerated.
Finally, the activated sludge process most likely is a
WWTP technique that will also prevail in the foreseeable
future. Process designs are continuously evolving to meet
the demands of upcoming wastewater types, improved
performance, and less resource consumption.
the ecosystem as well as to create easy accessible
recreational and educational meetings between urban
citizens and the ecosystem. Most importantly, this
would create awareness of the waste stream as a
resource and probably encourage the citizen to con
tribute to this idea.
Further Reading
Perspective on Biological Wastewater
Treatment
Originally, organized WWT was introduced for sanitation
reasons. Today, in the industrialized world, WWTPs and
arable land contribute with a substantial proportion to the
anthropogenic nitrogen load to the marine recipients, which
severely enhances eutrophication of aquatic environments.
Most natural ecosystems are controlled by a deficiency in
macronutrients like phosphorus and nitrogen, which means
that eutrophic level often directly controls ecosystem
responses. This interplay stresses the importance that
WWT systems are adapted to natural biogeochemical
cycles and are aligned with a vision of a durable society.
An important question is to what extent wastewater, for
example, municipal wastewaters and sewage sludge, should
be considered a waste or valuable resource and recycled as
plant nutrients in crop and in energy production. Key
constraints for the growing global population are due to
food and energy. Today, both extraction of phosphorus and
production of mineral nitrogen fertilizers consume exten
sive resources of fossil fuels. Hence, one important future
aim must be to create a sustainable loop of plant nutrients
through food production and refinement, urban consump
tion, waste handling, and back to arable land. To achieve
this, the effluent wastewater stream must contain as much
phosphorus and nitrogen as possible in addition to minimal
amounts of organic and inorganic toxicants.
Such global aims have to be linked with the ability to treat
a growing amount of wastewater. Not only is it important to
select specific solutions for specific treatment situations, but
it will also be essential to be able to optimize treatment with
account to the broad scientific basis involving both water
dynamics and biological processes. The coupled scientific
basis is essential for an in depth understanding of the key
microbiological processes involved in nitrogen removal and
for optimizing biological treatment systems.
Another future perspective is the contribution of
treatment wetlands to maintain biological diversity in
Ahn Y H (2006) Sustainable nitrogen elimination biotechnologies.
Process Biochemistry 41: 1709 1721.
Bolster CH and Saiers JE (2002) Development and evaluation of a
mathematical model for surface water flow within Shark River Slough
of the Florida Everglade. Journal of Hydrology 259: 221 235.
de Bashan L E and Bashan Y (2004) Recent advances in
removing phosphorus from wastwater and its future use as
fertilizers (1997 2003). Water Research 38: 4222 4246.
Garnaey KV, van Loosdrecht MCM, Henze M, Lind M, and
Jørgensen SB (2004) Activated sludge wastewater treatment plant
modelling and simulation: State of the art. Environmental Modelling
and Software 19: 763 784.
Gilbride KA, Lee D Y, and Beudette LA (2006) Molecular techniques in
wastewater: Understanding microbial communities, detecting
pathogens, and real time processes. Journal of Microbiological
Methods 66: 1 20.
Hughes J and Heathwaite L (1995) Hydrology and Geochemistry of
British Wetlands. London: Wiley.
Juretschko S, Loy A, Lehner A, and Wagner M (2002) The microbial
community composition of a nitrifying denitrifying activated sludge
from an industrial sewage treatment plant analyzed by the full
cycle rRNA approach. Systematic and Applied Microbiology
25: 84 99.
Kadlec RH and Knight RL (1996) Treatment Wetlands. New York: CRC
Press LLC.
Kelly JJ, Siripong S, McCormack J, et al. (2005) DNA microarray
detection of nitrifying bacterial 16S rRNA in wastewater treatment
plant samples. Water Research 39: 3229 3238.
Kjellin J, Worman A, Johansson H, and Lindahl A (2007) Controlling
factors for water residence time and flow patterns in Ekeby
treatment wetland, Sweden. Advances in Water Research
30(4): 838 850.
Levenspiel O (1999) Chemical Reaction Engineering. New York: Wiley.
Liwarska Bizukokc E (2005) Application of image techniques in
activated sludge wastewater treatment processes. Biotechnology
Letters 27: 1427 1433.
Rossetti S, Tomei MC, Nielsen PH, and Tandoi V (2005) ‘Microthrix
parvicella’, a filamentous bacterium causing bulking and foaming in
activated sludge systems: A revew of current knowledge. FEMS
Microbiology Reviews 29: 49 64.
Schmidt I, Sliekers O, Schmidt MS, et al. (2003) New concepts of
microbial treatment processes for the nitrogen removal in
wastewater. FEMS Microbiology Reviews 27: 481 492.
Seviour RJ and Blackall LL (eds.) (1999) The Microbiology of Activated
Sludge. Dordrecht: Kluwer Academic Publishers.
Van Niftrik LA, Fuerst JA, Sinninghe Damsté JS, et al. (2004) The
anammoxosome: An intracytoplasmic compartment in anammox
bacteria. FEMS Microbiology Letters 233: 7 13.
Wagner M and Loy A (2002) Bacterial community composition and
function in sewage treatment systems. Current Opinion in
Biotechnology 13: 218 227.
Boreal Forest 181
Boreal Forest
D L DeAngelis, University of Miami, Coral Gables, FL, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Climate and Soils
Forest Structure and Species
Animals
Biodiversity
Ecosystem Dynamics
Conservation and Global Issues
Further Reading
Introduction
leading to waterlogged soils. The soil decomposition
rate in the taiga is slow, which leads to the accumulation
of peat.
Several soil types characterize the boreal forest. The
soils of a major part of the boreal forest, under a dense
coniferous canopy, are heavily podzolized where the soil
is permeable, and so it consists largely of Spodosols.
Intense acid leaching forms a light ash colored eluvial
soil horizon leached of most base forming cations such
as calcium. Thus taiga soils tend to be nutrient poor.
Gelisols are common in the north, where permafrost
occurs. These are young soils with little profile develop
ment. Histosols, which are high in organic matter, form in
non permafrost wetlands, where decomposition is slowed
by hypoxic conditions. These are often referred to as
peatlands.
The boreal forest biome is also referred to as the ‘taiga’
(Russian for ‘swamp forest’). Geographically, the boreal
forest is located between latitudes 45 and 70 N, and
virtually all of it is in Canada, Alaska, and Siberia, with
portions in European Russia and Fennoscandia. The
boreal forest is bordered on the north by treeless tundra
and on the south by mixed forest. The boreal forest is
termed a ‘biome’ by ecologists, a term that refers to a
biogeographic unit that is distinguished from other
biomes by the structure of its vegetation and dominant
plant species. A biome is the largest scale at which
ecologists classify vegetation. All parts of a biome tend
to be within the same climatic conditions, but because
local conditions differ, a biome may encompass many
specific ecosystems (e.g., peatlands, river floodplains,
uplands) and plant communities. Despite this diversity
within a biome, in referring to the boreal forest we will
here use the terms ‘biome’ and ‘ecosystem type’
interchangeably.
Climate and Soils
The climate of the boreal forest is continental and, impor
tantly, for the growing season, there tends to be
between 30 and 150 days of temperatures above 10 C.
Temperature lows can fall below 25 C. Average annual
precipitation is 38–50 cm, with the lowest amounts in the
northern boreal forest, and greater frequency of precipi
tation during the summer season. Water is seldom
limiting because of the generally flat topography and
low rate of evaporation.
Permafrost can occur in the northern parts of this
zone, the southern limit coinciding roughly with a
mean air temperature of 1 C and snow depth of
about 40 cm. The zone of permafrost generally starts at
depths ranging from 1.5 to 3 m in the areas of the boreal
forest where it occurs. Its occurrence limits soil processes
to an upper active layer and impedes water drainage,
Forest Structure and Species
Because many hardwood trees are both sensitive to low
winter temperatures and require a long and warm sum
mer, the true boreal forest begins where the few
remaining hardwoods become a minor part of the forest.
Four coniferous genera dominate a major part of the taiga:
Picea (spruce), Abies (fir), Pinus (pine), and Larix (larch).
The hardwoods, which largely occur in dwarf form,
include Alnus (alders), Populus (poplars), Betula (birches),
and Salix (willows). The hardwoods tend to be early
successional species following disturbances such as fires
or erosion/deposition processes on riverbanks, which are
eventually shaded out by slower growing spruces and firs.
Much of the main boreal forest is dominated by a few
spruce species. These form a dense canopy in the central
and southern taiga, with a ground cover of dwarf shrubs,
such as cranberries and bilberries, and mosses and lichens.
In northern Siberia, huge areas are covered almost solely
by larch, and the canopy is much less dense. Pine species,
which can withstand a range of harsh conditions, grow in
light, sandy soils and other dry areas. As the boreal forest
tundra boundary is approached, conifers thin out to a
182
Boreal Forest
woodland, with lichen and moss dominating the ground.
Trees become more and more stunted.
The standing stock of biomass of the boreal forest
ranges is estimated at 200 (range 60–400) metric tons
per hectare (t ha 1). This compares with an estimate of
350 t ha 1 for the temperate deciduous forest and
10 t ha 1 for the tundra ecosystems. The boreal forest
differs from the temperate forest in having a much higher
percentage of its total biomass as photosynthetic foliage
(7% vs. 1%). It differs from the tundra in having a lower
percentage of root biomass (22% vs. 75%).
Animals
Animal life in the boreal forest is far less diverse than in
most temperate zone ecosystems. One component of the
taiga fauna, conspicuous for its frequent devastating
effects on thousands of hectares of forest, is that of phy
tophagous insects. Populations of these insects, which
include pine sawflies, spruce budworms, bark beetles,
and many others that attack conifers, are capable of escap
ing natural enemies and building up to huge population
densities. The large monospecific stands of the boreal
forest may be especially vulnerable. The high numbers
of insects during the warm months is a main explanation
for the large numbers of birds that migrate from the south
to breed in the taiga, especially large numbers of species
of warblers and thrushes. A number of bird species are
adapted to being residents of the taiga. Grouses such as
the capercaillie of the Old World, are adapted to year
round life in the taiga, as are some owls, woodpeckers, tits,
nuthatches, crossbills, and crows. Small mammal herbi
vores of the boreal forest include the squirrels,
chipmunks, voles, and snowshoe hares. These provide
food for a small number of predator species, including
the red fox (Vulpes vulpes) and members of the weasel
family. The moose (Alces alces) (called elk in the Old
World) has a wide geographic distribution in the taiga.
They are prey for wolves (Canis lupus) and occasionally
the brown bear (Ursus arctos).
Biodiversity
Tree species richness is far smaller than that in the tem
perate forests to the south, where more than 100 species
are typically observed in 2.5 2.5 quadrats in eastern
United States. Species richness clearly declines from
south to north in the taiga. Whereas 40 or more tree
species can be found in the southern taiga in Canada,
this declines to 10 or so species near the tundra boundary.
Animal species also show strong gradients. Reptile and
amphibian species are almost nonexistent above 55 .
Mammal species richness declines from close to 40
species to about 20 going northward in the boreal forest
biome in North America, while bird species decline from
about 130 to less than 100.
Ecosystem Dynamics
In keeping with its position between much warmer cli
mate of the temperate zone and colder climate of the
tundra, the boreal forest’s indices of production are inter
mediate between those two ecosystem types. Annual net
primary production in the boreal forest has been esti
mated at 7.5 t ha 1 yr 1 (range 4–20). This compares
with 11.5 t ha 1 yr 1 for temperate forest and 1.5 t ha 1
yr 1 for tundra ecosystems. Mean boreal forest litterfall is
estimated to be 7.5 t ha 1 yr 1 compared with 11.5 and
1.5 t ha 1 yr 1 for the temperature forest and tundra,
respectively. Because low temperatures slow decomposi
tion, the rate of litterfall decomposition in the boreal
forest, 0.21 t ha 1 yr 1, is also intermediate between 0.77
and 0.03 t ha 1 yr 1 for the temperate forest and tundra.
This means that it takes roughly 3 (1/0.21) ¼ 14 years
for 95% of a pulse of litter to decompose.
Fire is an inherent factor in the ecosystem dynamics of
the boreal forest. Lightning caused fires occur on a given
area at intervals of 20–100 years in drier areas to 200þ
years in wetter areas such as floodplains. Because nutrients
tend to be tied up in slowly decomposing organic matter,
fire may be important for maintaining tree growth by
releasing pulses of nutrients periodically. Many taiga
plant species have adaptations to fires, such as serotinous
cones and early sexual maturity of some conifers, and
resprouting capacity of hardwood trees and many herbs
and shrubs. Fires also reset the successional cycle, allowing
shade intolerant species like birch and aspen to invade.
Conservation and Global Issues
The boreal forest represents the single largest pool of living
biomass on the terrestrial surface (more than 30% of the
total terrestrial pool), and is therefore critically important
in global carbon dynamics. Much of the carbon is stored in
the ground layer. Currently, the taiga is thought to act as a
net sink of carbon. However, global climate change, in the
form of higher temperatures, may cause significant changes
in the carbon dynamics by increasing decomposition rates
faster than photosynthetic rates. Fire frequencies may also
increase with temperature, as precipitation is not expected
to rise, which will further increase the release of carbon
stored in the ground layer. According to some studies, the
boreal forest will be a net contributor to CO2 in the atmo
sphere under the projected climate changes.
Climate induced changes in the boreal forest will also
have an impact on migrant birds that use the region for
Botanical Gardens 183
reproduction. Changes in tree species composition may
challenge the capacity of birds to adapt, as has already the
increasing fragmentation of the forest due to clear cutting
in many areas within the biome.
See also: Tundra.
Further Reading
Danell K, Lundberg P, and Niemala P (1996) Species richness in
mammalian herbivores: Patterns in the boreal zone. Ecography
19: 404 409.
Henry JD (2003) Canada’s Boreal Forest. Washington, DC:
Smithsonian.
Hunter ML, Jr. (1992) Paleoecology, landscape ecology, and
conservation of neotropical migrant passerines in boreal forests.
In: Hagan JMIII and Johnston DW (eds.) Ecology and Conservation of
neotropical Migrant Landbirds, pp. 511 523. Washington, DC:
Smithsonian Institution Press.
Knystautus A (1987) The Natural History of the USSR. New York:
McGraw Hill.
Krebs CJ, Boutin S, and Boonstra R (2001) Ecosystem Dynamics of the
Boreal Forest: The Kluane Project. New York: Oxford University
Press.
Larsen JA (1980) The Boreal Ecosystem. New York: Academic Press.
Oechel WC and Lawrence WT (1985) Taiga. In: Chabot BF and
Mooney HA (eds.) Physiological Ecology of North American Plant
Communities, pp. 66 94. New York: Chapman and Hall.
Botanical Gardens
M Soderstrom, Montreal, QC, Canada
ª 2008 Elsevier B.V. All rights reserved.
Gardens for Systematic Study
The Gardens of the Ancients
Recreating Eden
The Gardens of Discovery
Botanic Gardens in Colonies
The Intrinsic Value of Biodiversity and Nature
Education and the Future
Further Reading
Gardens for Systematic Study
According to Botanic Garden Conservation
International, at the beginning of the twenty first century
some 2000 botanic gardens in 148 countries harbored
representatives of more than 80 000 plant species, or
about one third of the vascular plant species in the world.
The gardens range from large ones like Kew and the New
York Botanical Garden, where gorgeous plant displays are
coupled with scientific research, to much smaller ones like
Nezahat Gokyigit Memorial Park, near Istanbul, Turkey
and Bafut Botanic Garden in northwest Cameroon which
concentrate on safeguarding and studying local biosystems.
Botanic gardens are gardens where plants are gathered
together for systematic study. Often they imitate a num
ber of naturally occurring ecosystems: the San Francisco
Botanical Garden has created a cloud forest section while
the basement of the Palm House (Figure 1) in the Royal
Botanical Gardens at Kew (Figure 2) features marine and
intertidal habitats, for example. But in botanic gardens the
term ecology means far more than imitation, and the
gardens’ ecological impact has changed as philosophies
and world views have evolved.
Originally, interest was directed toward collecting and
studying plants themselves, with little care taken in
recording details of the plants’ habitats or in safeguarding
the ecosystems. Later, during the period of what might be
called the imperial botanic garden, Western countries
used botanic gardens to transfer plants from one part of
the world to another, with sometimes devastating conse
quences for the ecosystems receiving the foreign plants.
Most recently, botanic gardens have begun to play a
major role in conserving endangered plants and preser
ving threatened habitats. Nearly 2500 botanic gardens are
listed with Botanic Gardens Conservation International.
To search for gardens by country, refer to http://
www.bgci.org.uk/. Table 1 lists a few selected gardens.
The Gardens of the Ancients
The idea of the modern botanic garden dates from the
Renaissance, but it is possible that gardens which
resembled them existed long before. Certainly plants
valued for their medicinal properties were collected,
grown, and studied in gardens in many parts of the world.
Chinese tradition says that the emperor Shen Nung
experimented to find the medicinal properties of plants as
early as the twenty seventh century BCE. But since no
writing existed at the time and his materia medica Shen
Nung Pen Ts’ao Ching dates only from the seventh century
CE, the possibility of a garden somewhat like a botanic
garden in ancient China is only that, a possibility.
184
Botanical Gardens
Table 1 Selected botanic gardens
Early botanical gardens
Orto Botanico at Pisa, Italy: founded c. 1545
Orto Botanico at Padua, Italy: founded c. 1545
Hortus Botanicus, Leiden, Netherlands: founded 1590
Le Jardin des plantes de la Université Montpellier, France:
founded 1593
Oxford Physic Garden, Oxford University, UK: founded 1621
Le Jardin des plantes, Paris, France: founded 1626
Figure 1 The Palm House at Kew is one of its most distinctive
features, and inspired many other glasshouses in other botanic
gardens. Photograph by M. Soderstrom.
Some other notable European gardens
Botanischer Garten und Botanisches Museum Berlin-Dahlem,
Berlin, Germany
Linnaean Garden, Botaniska trädgården, Uppsala, Sweden
Jardı́n Botánico de Madrid, Spain
Jardim Botanico, University of Coimbra, Portugal
The Royal Botanic Gardens at Kew, London, UK
The Royal Botanic Garden, Edinburgh, Scotland
Eden Project, Cornwall UK
The National Botanic Garden of Wales, Llanarthne,
Carmarthenshire, Wales, UK
Some gardens with colonial roots
Amani Nature Reserve, Tanzania
Bogor Botanical Gardens, Bogor, Indonesia
Indian Botanical Gardens, Shibpur, Kolkata, India
Pamplemousse Botanic Gardens, Mauritius
Rimba Ilmu Botanic Gardens, Kuala Lumpur, Malaysia
Royal Botanic Gardens, Trinidad
Singapore Botanic Gardens, Singapore
Figure 2 Bluebells growing under trees in the Conservation
Area of The Royal Botanic Gardens at Kew. Photograph by
M. Soderstrom.
The systematic garden developed by the Greek scholar
Theophrastus (372–288 BCE) is much better documented.
The author of two major works on plants and botany
Historia de Plantis (History of Plants or Inquiring into
Plants) and De Causis Plantarums (The Causes of Plants),
he was a trusted associate of Aristotle, who bequeathed to
him his library, garden, and the leadership of his school.
Among the students was Alexander the Great who appears
to have sent back plants from his campaigns through Central
Asia, which were then planted in Theophrastus’s garden.
Other illustrious gardens featuring plants gathered for
study were established in pre Spanish conquest Mexico.
The Mexican emperor Montezuma’s garden brought
together plants from tropical regions as well as Mexico’s
highlands. Hernando Cortez was impressed by them
when he and his men overran Mexico in the 1520s. He
described the great gardens he found there as unlike
anything known in Europe at the time.
Things would soon change, however, in part because of
the plants brought back to Europe by explorers like Cortez.
Some notable New World gardens
USA
Boyce Thompson Arboretum. Superior, AZ
Brooklyn Botanic Garden, New York
Chicago Botanic Garden, Chicago, IL
Fairchild Tropical Botanic Garden, Fairchild, FL
Hawaii Tropical Botanical Garden, outside Hilo, Hawaii
Missouri Botanical Garden, St. Louis, MO
New York Botancial Gardens, New York
San Francisco Botanical Garden at Strybing Arboretum, CA
Canada
Jardin botanique, Montréal, QC
Royal Botanical Gardens, Hamilton, OM
UBC Botanical Garden and Centre for Plant Research,
Vancouver, BC
Latin America
Belize Botanic Garden, San Ignacio, Belize
Jardin Botanico Francisco Javier Clavijero, Xalapa, Veracruz,
Mexico
The UNAM Botanical Garden, Mexico City, Mexico
Jardim Botânico de São Paulo – São Paulo, Brazil
Jardim Botânico do Rio de Janeiro – Rio de Janeiro, Brazil
Some Asian gardens
Maharashtra (Mahim) Nature Park in Mumbai, India
Narayana Gurukula Botanical Sanctuary, North Wayanad,
Kerala, India
Beijing Botanical Garden, Beijing, China
Lijiang Botanic Garden & Research Station, Yunnan Province,
China
Nanjing Botanical Garden, Nanjing, China
Koishikawa Botanical Gardens, Tokyo, Japan
(Continued )
Botanical Gardens 185
Table 1 (Continued)
Some Southern Hemisphere gardens
Kirstenbosch National Botanical Garden, Cape Town, South
Africa
Royal Botanic Gardens – Melbourne, Victoria, Australia
Royal Botanic Gardens – Sydney, New South Wales, Australia
Alice Springs Desert Park and Olive Pink Botanic Garden,
Northern Territory, Australia
Bafut Botanic Garden in northwest Cameroon
Recreating Eden
Records from the Middle Ages testify to the interest of
Europeans in studying plants for their medicinal properties.
By the time of the Renaissance the five volumes of herbal
lore prepared by the second century pharmacist doctor
Dioscorides were used throughout Europe to teach about
plants useful for medicine. Many monasteries had little
plots of loosestrife and mints, of St. John’s wort and cha
momile, while untold numbers of midwives and lay healers
cultivated medicinal herbs. One record of such a garden is a
decree by Pope Nicolas V who in 1447 set aside part of the
Vatican grounds as a garden where medicinal plants could
be grown and botany taught as a branch of medicine.
A 100 years later Italy saw the establishment of the first
botanic gardens in the modern sense at two universities,
Padua and Pisa. The two dispute which was first. The Orto
Botanico at Padua was established by decree of the Senate
of the Venetian Republic in May 1545 and in July the
monastery of S. Giustina ceded about 20 000 square meters
to the republic and the University of Padua. No such
decrees exist for the Orto Botanico of Pisa, but a letter
written in early July 1545 by Lucca Ghinni, founder of the
garden, suggests that it was already in existence then. What
is clear is that these two gardens were places where plants
were grown for systematic study, and which were organized
to make that study easier. Nor were Pisa and Padua alone:
in 1590 the University of Leiden established its botanic
garden, while a year later the Jardin des Plantes of the
Université Montpellier in southern France was begun.
Today a glimpse of what these gardens were like can
be enjoyed at Leiden where a walled garden set apart
from the rest of the university’s botanic garden, the
Hortus Botanicus, is laid out as it was in about 1594 by
the pioneer botanist, Clusius. His career also gives a sense
of the inquiring spirit which was developing among
observers of the natural world. A native of the part of
Flanders now in France, he spent his life collecting and
describing plants all over Europe. He wrote treatises on
the flora of Spain, Austria, and Hungary, corresponded
with every botanist of note in Europe, collected and
distributed plants and bulbs widely, and wrote the first
monographs on both the tulip and the rhododendron.
Behind all this lay a belief that the beauty of plants was
a reflection of the wonders of God’s creation and the
harmony of the universe.
The religious impulse was extremely important during
this period. Practically no one in Christendom in the six
teenth and seventeenth century doubted that Eden as
described in the Bible had once existed. Many hoped
that it still did. Part of Portugal’s explorations were fired
by the desire to find the lost paradise, while Christopher
Columbus included a converted Jew in his first crew. The
man spoke Hebrew, Arabic, and Aramaic, and so, it was
thought, would be able to converse to the inhabitants of
Eden, should that splendid garden be discovered on the
westering voyages.
Eden, of course, was not found, and many botanists,
both religous and secular, began to wonder if Eden might
be recreated simply by bringing together all the plants
which must have grown in it. Some thought that even if a
latter day Eden were impossible to create, much good
would be done by studying as much of God’s creation as
possible: each plant was a facet of God, so that knowing all
plants would mean knowing an important part of God.
There were built in contradictions in this effort, however,
since it coincided with the great age of exploration when
plants and animals unimagined by Europeans were brought
back from the Americas for study. Questions arose: Were
they created at the same time as all the familiar flora and
fauna? Or were there perhaps two Creations, or parts of the
world which had escaped the Flood? Opinions varied, but
one thing was clear: the theological ideas behind the efforts to
bring plants together for study would have to be modified.
The Gardens of Discovery
Indeed one of the foremost gardens of the age was located
where it was in order to escape the influence of the Roman
Catholic church and its educational institutions. The Jardin
des Plantes (Figure 3) in Paris was chartered in 1626 by
Figure 3 The Jardin des plantes of the Muséum de l’histoire
naturelle is now surrounded by Paris, but when it was opened in
the seventeenth century it was outside the city’s walls.
Photograph by M. Soderstrom.
186
Botanical Gardens
Louis XIII on land a short way outside the wall encircling
the city which put it beyond the reach of the Université de
Paris and its Faculté de médecine. For the next 150 years
during the high tide of French exploration and coloniza
tion and throughout the French Enlightenment, Paris’s
botanic garden was the world’s main center for plant col
lection and study as well as home to sometimes audacious
research into other aspects of the natural world.
In England, several medicinal, or physic gardens, were
also established in the seventeenth century. The first was
the Oxford Physic Garden, set up in 1621 ‘‘for the advance
ment of medicine . . . the promotion of learning and the
glorification of the work of God.’’ Spain and Portugal
began their royal botanic gardens somewhat later. The
Jardı́n Botánico de Madrid was established in 1755 while
the Jardim Botanico of the University of Coimbra dates its
roots to 1775. The small botanic garden which was to
become the Royal Botanical Gardens at Kew was started
a few years before them as the pet project of Frederick,
Prince of Wales, and his wife on the royal country estates
upstream from London on the Thames. Frederick’s son,
George III, expanded the garden and saw to it that British
explorers under the aegis of Sir Joseph Banks were given
mandates to bring back plants for the Royal Gardens.
Botanic Gardens in Colonies
Britain and other colonial powers began not only to increase
the size of foreign plant collections in their botanic gardens
at home, but also to establish gardens in the countries they
were colonizing. The Dutch set up gardens in southern
Africa and on Java in what is now Indonesia both to provi
sion their ships and to study and to acclimatize plants which
might be useful either at home or in other colonies. The
French followed suit on Mauritius in the Indian Ocean and
on Martinique in the Caribbean. The British had their own
botanic gardens at Calcutta, Singapore, and in what is now
Sri Lanka. The Germans, who were late comers to the
colonial game, set up botanic gardens in Africa in what is
now Cameroon and Tanzania in the late nineteenth century.
In all cases the gardens maintained close ties with the home
country, and the home gardens.
There are a number of ways that this network of botanic
gardens have had ecological effects. By introducing plants
into the home country, they paved the way for exotics to
become established in new habitats. Two examples of intro
ductions which appeared initially to have few negative
effects are plants brought back to the Jardin des Plantes in
Paris. The black locust, a large tree originally found in a
relatively limited area of the Appalachians of North
America, now grows freely in forests and woodlots all over
Europe as well as far beyond its home range in the United
States and Canada. Its scientific name Robinia pseudoacacia L.,
honors Jean Robin who was the King’s gardener even before
the Jardin des Plantes was opened. A tree Robin planted in
the early 1600s was transplanted by his son to the Jardin, and
still grows there, the oldest tree in the center of Paris.
Another plant which migrated via the Jardin des
Plantes is the butterfly bush, Buddleia davidii. This native
of China was sent back to France by Abbé Armand David
in the nineteenth century. It now thrives in cultivated
gardens but also grows wild along railroad lines and in
disused land in Europe and North America.
Both of these plants are today considered undesirable
alien invaders in some parts of their adopted countries.
The black locust can produce thick plantations whose
shade does not allow other, native plans to grow, while
buddleia frequently forms dense thickets, forcing out
native plants along streambeds and in old pastures.
Other transplants produced consequences which took
less time to become apparent. Among them is breadfruit, a
native of the South Pacific, which Sir Joseph Banks, then
director of Kew, thought would be good food for the
slaves who worked in the sugarcane plantations in the
Caribbean. After a false start in 1791 – the first shipment
from Tahiti was on the Bounty when its crew mutinied
against Captain William Bligh – breadfruit and the plan
tain, another import, helped make plantation agriculture
profitable by providing cheaply and easily grown food.
Coffee first arrived in the Caribbean directly from a
botanic garden. In 1714 Louis XIV obtained a plant from
Amsterdam and sent it to the Jardin des Plantes. The inten
dant of the day had the Jardin’s first heated greenhouse
constructed for it, where it did very well. By 1721 enough
new plants had been propagated from it to risk sending the
first offspring to the botanic garden at Martinique in hopes
that after acclimatization there, the plants could be estab
lished in the French possessions around the Caribbean. It
worked: the coffee plantations of the French Antilles as well
as of Brazil, Jamaica, Columbia, and Mexico were all initi
ally planted with descendants from that one coffee tree.
Another example is that of rubber. Many plants in
tropical Asia, Africa, Central America, and Brazil, produce
latex: Columbus may have been the first to mention ‘white
milk’ oozing from the bark of some trees while the French
explorer La Condamine brought the first specimens of
caoutchouc to Europe in the eighteenth century. But it was
not until 1839 when Charles Goodyear discovered a pro
cess which produced rubber suitable for hoses and other
industrial uses that demand increased dramatically.
The only commercial source of rubber for most of the
nineteenth century was the wild rubber tree in Amazonia,
Hevea brasiliensis. So intense was the demand that a direct
steamship line ran from Manaus more than 1800 km (1100
miles) upstream on the Amazon to Liverpool, carrying
trading goods one way, and latex the other. In 1876 Henry
Wickham, a plant collector engaged by Kew’s director
Joseph Hooker, chartered a ship on the line to rush some
70 000 seeds across the Atlantic. He got permission from
Botanical Gardens 187
Brazilian authorities for the transfer by convincing them
of the need to release ‘‘exceedingly delicate botanic speci
mens specially designated for delivery to Her Britannic
Majesty’s own Royal Garden at Kew.’’
Hooker arranged for a night freight train to meet the
ship when it docked at Liverpool and cleared space in
Kew’s glasshouses for the seeds. Within 2 weeks of their
arrival in England, some 7000 seedlings had begun to grow,
and a year later 1900 plants were sent to the Perdeniya
Garden in what is now Sri Lanka. From there, seedlings
were distributed to several other tropical botanic gardens.
The Singapore Botanic Gardens (Figure 4) got 22 seed
lings, 11 of which it used for propagation in the garden. By
1917 it is estimated that the Singapore garden and its
director Henry ‘Rubber’ Ridley had distributed seven mil
lion seeds and by 1920 the Malaysian peninsula was
producing more than half the world’s rubber. There is no
way of estimating how many native plants disappeared
during the rapid transformation of jungle into rubber plan
tations. Indirectly the cultivation of rubber had other
effects on habitats also, since it made the development of
trucks and cars – and therefore of the industrialized world’s
sprawling, petroleum powered society – much easier.
Figure 4 View of the Palm Valley, the heart of the Singapore
Botanic Gardens. Photograph by M. Soderstrom.
Those who undertook these transfers of plants felt no
guilt at the massive reworkings of ecosystems which ensued.
Most people in the nineteenth and early twentieth century
believed that God made the world for humans to enjoy so
that making plants serve humans was doing God’s work.
At the same time, however, many botanic gardens by
accident or design preserved part of the native vegetation
in the gardens themselves. For example, the New York
Botanical Garden (Figure 5) includes 16 ha (40 acres) of
first growth, mixed hardwood forest. This remnant is a
unique reminder of the forest which covered most of what
is now the city of New York before Europeans wrested
control from the indigenous population.
Other examples of habitat conservation include the
Singapore Botanic Gardens’ small jungle enclave amid
the city’s myriad high rises as well as the
Conservation Area at Kew. There a part of the garden
is being conserved as British farmland, with upkeep
and interventions following traditional British agricul
tural practices.
Figure 5 The hemlock forest in the New York Botanical Garden
preserves a remnant of the forest which once covered much of
the New York City region. Photograph by M. Soderstrom.
188
Botanical Gardens
The Intrinsic Value of Biodiversity and
Nature
The philosophical framework in which most botanic gar
dens operate now places a high value on maintaining what
measure of biodiversity exists today. This can be seen as a
direct outgrowth of concerns about the natural world
which began to develop during the early twentieth cen
tury as the damage resulting from industrialization,
population growth, and uncontrolled exploitation of nat
ural resources became apparent. Rather than being
motivated primarily by a desire to study God in nature,
scientists and others began to think that nature was intrin
sically valuable, over and above whatever link it might
have with a deity or what economic advantage it might
bring to human society. Often work begun by botanic
gardens directly led to better understanding of the won
derful interplay of organisms in ecological systems, with
far reaching philosophical and scientific repercussions.
Take for example the huge water lily Victoria amazo
nica, grown by many botanic gardens today. When it was
first described in the early nineteenth century by plant
explorers who found it in British Guyana, the accounts
caused a sensation: in addition to having lovely flowers, its
leaves grew up to 6 ft in diameter, and were strong enough
to support an adult man. Seeds were sent back to Europe
several times, but it was not until 1849 that Kew was able
to raise plants to the stage where they could be set out in
ponds. Of these, three flowered, the first being one in
Duke of Devonshire’s garden at Chatsworth. The one at
Kew bloomed the next year after being installed in a
special glasshouse, and some 30 000 visitors came to mar
vel at the flowers and the leaves. The craze was not
confined to England: the Hortus Botanicus (Figure 6) at
Leiden succeeded in getting a plant to flower in 1872, and
kept one alive during the coldest days of World War II
when there was only enough fuel to keep the water lily
greenhouse heated. Early pictures from both the
Figure 6 The Clusius Garden in the Hortus Botanicus at the
University of Leiden is arranged much as it was in 1594.
Photograph by M. Soderstrom.
Singapore Botanic Gardens and the Missouri Botanic
Garden feature the plant too.
But the story of Victoria amazonica does not end with
special ponds and crowds of visitors. One of the oddities
of the flowers is that when they are dissected in the wild, a
particular sort of beetle (Cyclocephala hardyi) is often found
inside. For a long time botanists suspected that the beetles
pollinated the plant but were not sure how. The mystery
was only unraveled in the 1970s when Ghillean Prance,
Kew’s director from 1988 to 1999 but then a research
biologist for the New York Botanical Garden, spent nights
standing hip deep in Brazilian ponds, watching the flow
ers open and beetles flie in and out. He found that the
beetles were attracted to the fragrance of the opening
flowers, crawled inside to feed, and were trapped there
when the flowers closed as dawn approached. The next
evening the beetles, sticky from feeding, crawled back out
as the flowers opened again, picking up a load of pollen as
they passed. Then they flew away to repeat the process in
another flower, and incidentally pollinate it. In so doing
they demonstrated the complexity of ecosystems and the
intricate way plants and animals are interrelated.
Much of botanic gardens’ present day work in the
field, the laboratory, and in the gardens themselves is
designed to study these kinds of relationships. Botanic
gardens today also actively work for conservation of spe
cies by propagating plants and collecting and storing
seeds. According to the World Conservation Union,
34 000 taxa around the world are considered threatened
with extinction. Of them 10 000 threatened species – or
about a third – are growing in one or more botanic
gardens. In some cases, the collection of plant specimens
comes just in the nick of time. The canyon in Chiapas,
Mexico where botanists in the late 1990s found Deppea
splendens, a shrub with lovely two inch orange flowers
hanging in long clusters, has since been cleared for corn
fields: the plant is thought to be now extinct in the wild.
But seeds from the shrub flourished in the San Francisco
Botanical Garden, and cuttings from the plants are even
sold by the Friends of the Garden at their annual sale.
Perhaps the biggest conservation project is Kew’s
Millennium Seed Bank. The £ 80 million (US$ 160 mil
lion) undertaking is housed in new facilities at Kew’s
Wakehurst auxiliary garden south of London. Its aim is
to collect seeds from 24 000 species of plants from all of
the world by 2020, and to keep them in secure locations so
that they can be used at a later date. When seeds are dried
so that they contain only 5% moisture, about 80% of
them can successfully be held at –20 C for periods of
up to 200 years. A portion of all seeds will be held in their
country of origin in order to avoid repeating ‘theft’ of
plants like that which occurred during the great period of
European colonialism. Fortunes were made then by
European exploiters of coffee, rubber, and other plants
but the countries of origin received no compensation.
Botanical Gardens 189
As part of the effort to compensate for damage done in the
past and to preserve remaining biodiversity, Botanic
Gardens Conservation International has set a series of tar
gets to be met by 2010. The overall aims are general – things
like protection of plant diversity, conservation of endan
gered species in botanic gardens and in their native habitat,
and public education about the importance of plant diver
sity. Specific goals in the 20 item to do list are quite specific,
though. For example, at least 10% of each of the world’s
ecological regions are to be effectively conserved, and the
number of trained botanic garden staff working in conserva
tion, research, and education should be doubled. In addition,
international databases of such things as which endangered
species are cultivated in what botanic garden and what plant
introduction has become invasive in what range are under
development. Many botanic gardens are already promoting
awareness of the problems posed by invasive species through
such things as the St. Louis Declaration on Invasive Plant
Species, developed at a conference organized by the
Missouri Botanical Garden in 2001.
Several new botanic gardens have also been estab
lished recently with the principal aim of protecting
unique and relatively untouched environments. One of
them is the Alice Springs Desert Park in central Australia,
opened to the public in 1997, which preserves a section of
that continent’s desert. Another is the Bafut Botanic
Garden in northwest Cameroon, also opened in 1997, a
savanna botanic garden and forest reserve.
In addition, two other new botanic gardens point the
way to reclaiming landscapes destroyed by human care
lessness and greed. The first is the Eden Project in
Cornwall UK, where a former clay pit has been trans
formed into a botanic garden with several distinct
ecosystems represented in geodesic buildings sunk into
the former mine landscapes. The other is the Maharashtra
(Mahim) Nature Park in Mumbai, India, where 15 ha of
former garbage dump have been reclaimed. The recon
structed forest is now home to 380 varieties of plants, 84
varieties of birds, and about 34 kinds of butterflies.
Figure 7 The Bog and Marsh Garden at the Jardin botanique in
Montreal presents wetlands plans – many of them endangered –
in a series of basins and ponds. Photograph by M. Soderstrom.
in winter, Montreal’s Jardin botanique (Figure 7) offers twi
lights full of Chinese lanterns in the fall, gardens everywhere
advertise their spring flowers and their summer splendor to
lure people to see their plants, and hear their message. The
effectiveness of these educational efforts may mean the dif
ference between governments setting ecologically sound
policy or not. Without public recognition that habitat pro
tection and biodiversity are important, governments in
democratic countries may drag their feet while in countries
where decisions are made from the top down, those in power
would not be convinced of the need to do the same.
Further Reading
Brockway L (1979) Science and Colonial Expansion: The Role of the British
Royal Botanic Gardens. New York and London: Academic Press.
Hyams E (1969) Great Botanical Gardens of the World (with
photographs by Macquitty W). London: Bloomsbury Books.
Laissus Y (1995) Le Muséum national d’histoire naturelle. Paris:
Découvertes Gallimard.
McCracken DP (1997) Gardens of Empire: Botanical Institutions of the
Victorian British Empire London: Leicester University Press.
Prest J (1981) The Garden of Eden: The Botanic Garden and the
Re Creation of Paradise. New Haven: Yale University Press.
Soderstrom M (2001) Recreating Eden: A Natural History of Botanical
Gardens. Montreal: Véhicule Press.
Education and the Future
Relevant Websites
Kings and religious authorities are no longer the patrons
behind botanic gardens, so those in charge of them must
convince the public, governments, and industry to support
the gardens and their work. This is why botanic gardens
today devote so much effort to education and public infor
mation projects. Some are high tech like the interactive rain
forest displays in the Climatron at the Missouri Botanic
Garden. Others, like the 12 acre adventure site in the New
York Botanical Garden, introduce children to ecological
concepts through activities full of action. Still others aim to
make botanic gardens places where pleasure goes hand in
hand with research and learning. Kew has an ice skating rink
http://www.bgci.org Botanic Gardens Conservation International.
http://www2.ville.montreal.qc.ca Jardin botanique in Montreal.
http://www.mnhn.fr Jardin des Plantes of the Muséum de
l9histoire naturelle.
http://www.nybg.org New York Botanical Garden.
http://www.sbg.org.sg Palm Valley, Singapore Botanic Gardens.
http://www.kew.org Royal Botanic Gardens, Kew.
http://www.sbg.org.sg Singapore Botauic Gardeus.
http://www.centerforplantconservation.org The Saint Louis
Declaration on Invasive Plants.
http://www.hortus.leidenuniv.nl The Hortus Botanicus of
Leiden, University of Leiden.
190
Caves
Caves
F G Howarth, Bishop Museum, Honolulu, HI, USA
ª 2008 Elsevier B.V. All rights reserved.
Caves
Cave Environments
Food Resources
Cave Communities
Adaptations to Cave Life
Other Cave-Like Habitats
Case Study: Hawai‘i
Perspective
Further Reading
Caves
dimensional (3D) mazes, in which food and mates may be
difficult to find. In addition, the water can stagnate, locally
becoming hypoxic with high concentrations of toxic gases
including carbon dioxide and hydrogen sulfide.
Caves are defined as natural subterranean voids that are
large enough for humans to enter. They occur in many
forms, and cavernous landforms make up a significant
portion of the Earth’s surface. Limestone caves are the
best known. Limestone, calcium carbonate, is mecha
nically strong yet dissolves in weakly acidic water. Thus
over eons great caves can form. Caves form in other
soluble rocks, such as dolomite (calcium magnesium car
bonate), but they are usually not as extensive as those in
limestone. Volcanic eruptions also create caves. The most
common are lava tubes that are built by the roofing over
and subsequent draining of molten streams of fluid basal
tic lava. In addition, cave like voids form by erosion (e.g.,
sea caves and talus caves) and by melting water beneath or
within glaciers. Depending on their size, shape, and inter
connectedness, caves develop unique environments that
often support distinct ecosystems.
Cave Environments
The physical environment is rigidly constrained by the
geological and environmental settings and can be defined
with great precision because it is surrounded and buffered
by thick layers of rock. Caves can be water filled or aerial.
Aquatic Environments
Aquatic systems are best developed in limestone caves
since water creates these caves. Debris laden water in
voids in nonsoluble rock will eventually fill caves.
A significant exception is found in young basaltic lava
that has flowed into the sea. Here, subterranean ecosystems
develop in the zone of mixing freshwater and salt water
within caves and spaces in the lava. The system is fed by
food carried by tides and groundwater flow. Frequent
volcanism creates new habitat before the older voids fill
or erode away. Aquatic cave environments are dark, three
Terrestrial Environments
The terrestrial environment in long caves is buffered from
climatic events occurring outside. The temperature stays
nearly constant, fluctuating around the mean annual sur
face temperature (MAST); except passages sloping down
from an entrance tend to trap cold air and remain a few
degrees cooler than MAST. Passages sloping up are often
warmer than MAST. The environment is strongly zonal
(Figure 1). Three zones are obvious: an entrance zone
where the surface and underground habitats overlap; a
twilight zone between the limit of photosynthesis and the
zone of total darkness. The dark zone can be further
subdivided into three distinct zones: a transition zone
where climatic events on the surface still affect the atmos
phere, especially relative humidity (RH); a deep zone
where the RH remains constant at 100%; and an inner
most stagnant air zone where air exchange is too slow to
flush the buildup of carbon dioxide and other decomposi
tion gasses. The boundary between each zone is often
determined by shape or constrictions in the passage. In
many caves, the boundaries are dynamic and change with
the seasons.
The subterranean aerial environment is stressful for
most organisms. It is a perpetually dark, 3D maze with a
water saturated atmosphere and occasional episodes
of toxic gas concentrations. Many of the cues used by
surface animals are absent or operate abnormally in caves
(e.g., light/dark cycles, wind, sound). Passages can flood
during rains, and crevices might drop into pools and
water filled traps. If the habitat is so inhospitable, why
and how do surface animals forsake the lighted world and
adapt to live there? It is the presence of abundant food
resources that provides the impetus for colonization and
adaptation.
Caves 191
Entrance
zone
Twilight
zone
Transition
zone
Deep
zone
Figure 1 Schematic profile view of the cave habitat showing the location of principal zones.
Food Resources
The main energy source in limestone caves is sinking
rivers, which carry in abundant food not only for aquatic
communities but also via flood deposits for terrestrial
communities. Rivers are less important in nonsoluble
rock, such as lava, but percolating runoff washes surface
debris into caves through crevices. Other major energy
sources are brought in by animals that habitually visit or
roost in caves, plants that send their roots deep under
ground, chemoautotrophic microorganisms that use
minerals in the rock and accidentals that fall or wander
into caves and become lost.
Generally in surface habitats, accumulating soil filters
water and nutrients and holds these resources near the
surface where they are accessible to plant roots and sur
face inhabiting organisms. However, in most areas with
underlying caves, the soil is thin with areas of exposed
bare rock because developing soil is washed or carried
into underground voids by water or gravity. Soil forma
tion is limited, and much of the organic matter sinks out of
the reach of most surface animals.
Except for guano deposits, flood deposits, scattered
root patches, and other point source food inputs, the
defining feature of cave habitats is the appearance of
barren wet rock. Visible food resources in the deep cave
are often negligible, and what food deposits there are
would be difficult for animals to find in the 3D maze.
Food resources in the system of smaller spaces is difficult
to sample and quantify, but in theory, some foods may be
locally concentrated by water transport, plant roots, or
micro point source inputs such as through cracks extend
ing to the surface. These deposits would be more easily
exploited than would widely scattered deposits.
In each biogeographic region, a few members of the
surface and soil fauna have invaded cave habitats and
adapted to exploit this deep food resource. The colonists
usually were pre adapted; that is, they already possessed
useful characteristics resulting from living in damp, dark
habitats on the surface.
Cave Communities
Guano Communities
Many animals live in or use caves. Cave inhabiting verte
brates are relatively well known. Cave bats, swiftlets
(including the edible nest swiftlet of Southeast Asia),
and the oil bird in South America use echolocation to
find their way in darkness. Pack rats in North America,
along with cave crickets and other arthropods also roost in
caves. Large colonies of these cave nesting animals carry
in huge quantities of organic matter with their guano and
dead bodies. This rich food resource forms the basis for
specialized communities of microorganisms, scavengers,
and predators. Arthropods comprise the dominant group
of larger animals in this community, and like their verte
brate associates, most species are able to disperse outside
caves to found new colonies.
Deep Cave Communities
In the deeper netherworld, communities of mysterious,
obligate cave animals occur. Most are invertebrates, but a
few fishes and salamanders have colonized the aquatic
realm. Crustaceans (shrimps and their allies) dominate
in aquatic ecosystems, and insects and spiders dominate
terrestrial systems. Although a few species are specialists
on living plant roots or other specific resources, most are
generalist predators or scavengers. The relatively high
percentage of predators indicates the importance of acci
dentals as a food resource. However, many presumed
predatory species, such as spiders, centipedes, and ground
beetles, will also scavenge on dead animals when
192
Caves
available. It is not advantageous to have finicky tastes
where food is difficult to find. Thus, the food chain,
which normally progresses from plants through plant
feeders, scavengers, and omnivores to predators, more
closely resembles a food web with most species interact
ing with most of the other species in the community.
into the clutches of a predator. Small insects are often too
heavy or are unable to climb the meniscus at the edge of
rock pools and will eventually drown. However, many
cave adapted insects have unique knobs or hairs near the
base of each elongated claw and modified behavioral traits
that allow them to climb the meniscus and escape. Some
of the latter are predators or scavengers, who wait on
pools for victims.
Adaptations to Cave Life
Animals roosting or living in caves must adapt to cope
with the unusual environment. Paramount for the cave
roosting vertebrates is the ability to find their way to and
from their roosts at the correct time. Not surprisingly,
the birds and bats display uncanny skill in memorizing the
complex maze to and from their cave roosts. Pack rats use
trails of their urine to navigate in and out of caves. Species
using the twilight and transition zones can use the daily
meteorological cycle for cues to wake and leave the cave.
Those roosting in the deep zone may rely on accurate
internal clocks to know when it is beneficial to leave their
roost.
Organisms that adapt to live permanently under
ground must make changes in behavior, physiology, and
structure in order to thrive in the stressful environment.
They need to find food and mates and successfully repro
duce in total darkness. Their hallmark is the loss or
reduction of conspicuous structures such eyes, bodily
color, protective armor, and wings. These structures are
worthless in total darkness, but they can be lost quickly
when selection is relaxed because they are expensive for
the body to make and maintain. How such losses could
happen quickly is demonstrated by the cave adapted
planthoppers (Cixiidae). The nymphs of surface species
feed on plant roots and have reduced eyes and bodily
color whereas their adults have big eyes, bold colors, and
functional wings. The cave adapted descendents main
tain the nymphal eyes, color, and other structures into
adulthood, a phenomenon known as neoteny.
The high relative humidity and occasional episodes of
elevated CO2 concentrations are stressful to cold blooded
organisms. The blood of insects and other invertebrates
will absorb water from saturated atmosphere, and the
animals literally will drown unless they have adaptations
to excrete the excess water. High levels of CO2 force
animals to breathe more, which increases water absorp
tion. Cave adapted insects often have modified spiracles
to prevent or cope with their air passages filling with
water.
Most lava tube arthropods have specialized elongated
claws to walk on glassy wet rock surfaces. Many have
elongated legs to step across cracks rather than having
to descend and climb the other side. Jumping or falling
might land a hapless animal in a pool or water filled pit or
Other Cave-Like Habitats
Cavernous rock strata contain abundant additional voids
of varying sizes, which may not be passable by humans.
These voids are interconnected by a vast system of cracks
and solution channels. The smaller capillary sized spaces
are less important biologically because their small size
limits the amount of food resources they can hold and
transport. Voids larger than about 5 cm can transport large
volumes of food as well as serve as habitat for animals. In
terms of surface area and extent, these intermediate size
voids are the principal habitat for specialized cave ani
mals. Many aspects of their life history may occur only in
these spaces. Some cave species (such as the earwig,
Anisolabis howarthi (Figure 2), and sheet web spiders,
Linyphiidae, in Hawaiian lava tubes) prefer to live
in crevices and are only rarely found in caves. In addition,
cave adapted animals have been found living far
from caves in cobble deposits beneath rivers, fractured
rock strata, and buried lava clinker in Japan, Hawai‘i,
Canary Islands, Australia, and Europe. These discoveries
corroborate the view that cave adaptation and the devel
opment of cave ecosystems can occur wherever there is
suitable underground habitat.
Because these smaller voids are isolated from airflow
from the surface, the environment resembles the stagnant
air zones of caves. Caves serve as entry points and win
dows in which to observe the fauna living within the voids
Figure 2 The Hawaiian cave earwig, Anisolabis howarthi
Brindel (family Carcinophoridae). Photo by W. P. Mull.
Caves 193
in the cavernous rock strata. The view is imperfect
because the environment is so foreign to human
experience.
Case Study: Hawai‘i
Food Web
The main energy sources in Hawaiian lava tube ecosystems
are tree roots, which penetrate the lava for several deca
meters; organic matter, which washes in with percolating
rainwater; and accidentals, which are surface and soil animals
blundering into the cave. Both living and dead roots are
utilized, and this source is probably the most important.
Furthermore, both rainwater and accidentals often use the
same channels as roots to enter caves, so that root patches
often provide food for a wide diversity of cave organisms.
The importance of roots in the cave ecosystem makes it
desirable to identify the major species. This has become
possible only recently by using DNA sequencing technol
ogy. The most important source of roots is supplied by the
native pioneer tree on young lava flows; Metrosideros polymor
pha. Cocculus orbiculatus, Dodonaea viscosa, and Capparis are
locally important in drier habitats. Several different slimes
and oozes occur on wet surfaces and are utilized by scaven
gers in the cave. They are mostly organic colloids deposited
by percolating groundwater, but some may be chemoauto
trophic bacteria living on minerals in the lava. Cave roosting
vertebrates do not occur in Hawai‘i. Native agrotine moths
once roosted in caves in large colonies, but the group has
become rare in historic times. The composition of the com
munity their colonies once supported is unknown. Feeding
on living roots are cixiid planthoppers (Oliarus). Their
nymphs suck xylem sap with piercing mouthparts. The
blind flightless adults wander through subterranean voids
in search of mates and roots. Caterpillars of noctuid moths
(Schrankia) prefer to feed on succulent flushing root tips, but
they also occasionally scavenge on rotting plant and animal
matter. Tree crickets (Thaumatogryllus), terrestrial amphipods
(Spelaeorchestia), and isopods (Hawaiioscia and Littorophiloscia)
are omnivores but feed extensively on roots. Cave rock
crickets (Caconemobius) are also omnivorous as well as being
opportunistic predators. Feeding on rotting organic material
and associated microorganisms are millipedes (Nannolene),
springtails (Neanura, Sinella, and Hawinella), and phorid flies
(Megaselia). Terrestrial water treaders (Cavaticovelia aaa) suck
juices from long dead arthropods. Feeding in the organic
oozes growing on wet cave walls are larvae of craneflies
(Dicranomyia) and biting midges (Forcipomyia pholeter). The
blind predators include spiders (Lycosa howarthi, Adelocosa
anops (Figure 3), Erigone, Meioneta, Oonops, and Theridion),
pseudoscorpions (Tyrannochthonius), rock centipedes
(Lithobius), thread legged bugs (Nesidiolestes), and beetles
(Nesomedon, Tachys, and Blackburnia). Most of the cave pre
dators will also scavenge on dead animal material.
Figure 3 The no-eyed big-eyed hunting spider, Adelocosa
anops Gertsch (family Lycosidae) from caves on the island of
Kaua‘i. Photo by the author.
Nonindigenous Species
Several invasive nonindigenous species have invaded cave
habitats and are impacting the cave communities. The
predatory guild is the most troublesome, with some spe
cies being implicated on the reduction of vulnerable native
species. Among these, the nemertine worm (Argonemertes
dendyi) and spiders (Dysdera, Nesticella, and Eidmanella) have
successfully invaded the stagnant air zone within the
smaller spaces. The colonies of cave roosting moths dis
appeared from the depredations of the roof rat (Rattus
rattus) on their roosts and from parasites purposefully
introduced for biological control of their larvae. Many
non native species (such as Periplaneta cockroaches,
Loxosceles spiders, Porcellio isopods, and Oxychilus snails)
survive well in larger accessible cave passages, where
they have some impact, but they appear not to be able to
survive in the system of smaller crevices. A few alien tree
species also send roots into caves, creating a dilemma for
reserve managers trying to protect both cave and surface
habitats since their roots support some generalist native
species but not the host specific planthoppers.
Succession
Inhabited Hawaiian lava tubes range in age from 1 month
on Hawai‘i Island to 2.9 million years on O‘ahu Island. On
Hawai‘i Island colonization and succession of cave eco
systems can be observed. Crickets and spiders arrive on
new flows within a month of the flow surface cooling.
They hide in caves and crevices by day and emerge at
night to feed on windborne debris. Caconemobius rock
194
Caves
crickets are restricted to living only in this aeolian (wind
supported) ecosystem and disappear with the establish
ment of plants. The obligate cave species begin to arrive
within a year after lava stops flowing in the caves. The
predatory wolf spider, Lycosa howarthi, arrives first and
preys on wayward aeolian arthropods. Other predators
and scavenging arthropods – including blind, cave
adapted Caconemobius crickets – arrive during the next
decade. Under rainforest conditions, plants begin to
invade the surface after a decade, allowing the root feed
ing cave animals to colonize the caves. Oliarus
planthoppers arrive about 15 years after the eruption
and only 5 years after its host tree, Metrosideros polymorpha.
The cave adapted moth, Schrankia species, and the under
ground tree cricket, Thaumatogryllus cavicola, arrive later.
The cave species colonize new lava tubes from neighbor
ing older flows via underground cracks and voids in the
lava. Caves between 500 and 1000 years old are most
diverse in cave species. By this time the surface rainforest
community is well developed and productive, while the
lava is still young and maximal amount of energy
is sinking underground. As soil formation progresses,
less water and energy reaches the caves, and the commu
nities slowly starve. In highest rainfall areas, caves support
none or only a few species after 10 000 years. Under desert
conditions, succession is prolonged for 100 000 years or
more. Mesic regimes are intermediate between these two
extremes. New lava flows may rejuvenate some buried
habitat as well as create new cave habitat.
Perspective
The fauna of a large percentage of the world’s cave
habitats remain unknown to science, and new species
continue to be discovered in well studied caves.
Additional biological surveys are needed to fill gaps in
knowledge and improve our understanding of cave eco
systems. Improved methods for sampling the inaccessible
smaller voids are needed. The cave environment is a
rigorous, high stress one, which is difficult for humans
to access and envision because it is so foreign to human
experience. Working in caves can be physically challen
ging. However, recent innovations in equipment and
exploration techniques allow ecologists to visit the
deeper, more rigorous environments.
In spite of the difficulties of working in the stressful
environment, several factors make caves ideal natural
laboratories for research in evolutionary and physiologi
cal ecology.
Since cave habitats are buffered by the surrounding
rock, the abiotic factors can be determined with great
precision. The number of species in a community is
usually manageable and can be studied in total.
Questions that are being researched are how organisms
adapt to the various environmental stressors; how com
munities assemble under the influence of resource
composition and amount; and how abiotic factors affect
ecological processes. For example, a potential overlap
between cave and surface ecological studies occurs in
some large pit entrances in the tropics. The flora and
fauna living in these pits frequently experience CO2
levels 25–50 times ambient.
See also: Rocky Intertidal Zone.
Further Reading
Camacho AI (ed.) (1992) The Natural History of Biospeleology. Madrid:
Monografias, Museo Nacional de Ciencias Naturales.
Chapman P (1993) Caves and Cave Life. London: Harper Collins
Publishers.
Culver DC (1982) Cave Life. Cambridge, MA: Harvard University Press.
Culver DC, Master LL, Christman MC, and Hobbs HH, III (2000) Obligate
cave fauna of the 48 contiguous United States. Conservation Biology
14: 386 401.
Culver DC and White WB (eds.) (2004) The Encyclopedia of Caves.
Burlington, MA: Academic Press.
Gunn RJ (ed.) (2004) Encyclopedia of Caves and Karst. New York:
Routledge Press.
Howarth FG (1983) Ecology of cave arthropods. Annual Review
Entomology 28: 365 389.
Howarth FG (1993) High stress subterranean habitats and evolutionary
change in cave inhabiting arthropods. American Naturalist
142: S65 S77.
Howarth FG, James SA, McDowell W, Preston DJ, and Yamada CT
(2007) Identification of roots in lava tube caves using molecular
techniques: Implications for conservation of cave faunas. Journal of
Insect Conservation 11(3): 251 261.
Humphries WF (ed.) (1993) The Biogegraphy of Cape Range, Western
Australia. Records of the Western Australian Museum, Supplement
no.45. Perth: Western Australian Museum.
Juberthie C and Decu V (eds.) (2001) Encyclopaedia Biospeologica Vol
III. Moulis, France: Société de Biospéologie.
Moore GW and Sullivan N (1997) Speleology Caves and the Cave
Environment, 3rd edn. St. Louis, MO: Cave Books.
Wilkins H, Culver DC, and Humphreys WF (eds.) (2000) Ecosystems of
the World, Vol. 30 Subterranean Ecosystems. Amsterdam: Elsevier
Press.
Chaparral
195
Chaparral
J E Keeley, University of California, Los Angeles, CA, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
The Ecological Community
Community Succession
Allelopathy
Fire
Community Recovery from Wildfires
Seed Germination
Seed Dispersal
Regional Variation in Fire Regime
Future Threats and Management
Further Reading
Introduction
The Ecological Community
Chaparral is the name applied to the evergreen sclero
phyllous (hard leaved) shrub vegetation of southwestern
North America, largely concentrated in the coastal zone
of California and adjacent Baja California. It is a dense
vegetation often retaining many dead spiny branches
making it nearly impenetrable (Figure 1). It dominates
the foothills of central and southern California but is
replaced at higher elevations by forests. On the most
arid sites at lower elevations evergreen chaparral is
replaced with a lower stature summer deciduous ‘soft
chaparral’ or sage scrub.
Chaparral owes much of its character to the
Mediterranean climate of winter rain and summer
drought. The severe summer drought, often lasting 6
months or more, inhibits tree growth and enforces the
shrub growth form. Intense winter rains coincide with
moderate temperatures that allow for rapid plant
growth, producing dense shrublands. These factors
combine to make this one of the most fire prone
ecosystems in the world. This Mediterranean climate
is the result of a subtropical high pressure cell that
forms over the Pacific Ocean. During the summer, this
air mass moves northward and blocks water laden air
masses from reaching land, and in winter this high
pressure cell moves toward the equator and allows
winter storms to pass onto land. On the Pacific
Coast it is wettest in the north, where the effect of
the Pacific High is least, and becomes progressively
drier to the south, and consequently chaparral dom
inates more of the landscape in the southern part.
Interestingly, these synoptic weather conditions form
globally at this same latitude (30–38 north or south)
and on the western sides of continents. As a result
similar Mediterranean climate shrublands occur in the
Mediterranean Basin of Europe, central Chile, South
Africa, and southern Australia.
Chaparral is a shrub dominated vegetation with other
growth forms playing minor or temporary successional
roles after fire. More than 100 evergreen shrub species
occur in chaparral, although sites may have as few as one
or more than 20 species, depending on available moisture,
slope aspect, and elevation. The most widely distributed
shrub is chamise (Adenostoma fasciculatum), ranging from
Baja to northern California, occurring in either pure cha
mise chaparral or in mixed stands. It often dominates at
low elevations and on xeric south facing slopes. The short
needle like leaves produce a sparse foliage, and soil litter
layers are poorly developed and result in weak soil hor
izons. Chamise often forms mixed stands of vegetation
with a number of species. These include the bright smooth
red barked manzanita (Arctostaphylos spp.), the sometimes
spiny ceanothus, also known as buckbrush or California
lilac (Ceanothus spp.). On more mesic north facing slopes
chaparral is commonly dominated by broader leaved
shrubs, including the acorn producing scrub oak (Quercus
spp.), the cathartic coffeeberry (Rhamnus californica), red
berry (R. crocea), the rather bitter chaparral cherry (Prunus
ilicifolia), and chaparral holly (Heteromeles arbutifolia), from
whence the film capital Hollywood derives its name.
The most common shrub species and the majority of
herbaceous species have fire dependent regeneration,
meaning that seeds remain dormant in the soil until
stimulated to germinate after fire (see the section titled
‘Fire’ below). These include chamise, manzanita, and
ceanothus shrubs, which flower and produce seed most
years but seldom produce seedlings without fire. Some
ceanothus species are relatively short lived or are easily
shaded out by other shrubs and die after several decades.
They, however, persist as a living seed pool in the soil. In
addition, a large number of annual species live most of
their life as dormant seeds in the soil, perhaps as long as a
century or more. Also, many perennial herbs with
196
Chaparral
Figure 1 Chaparral shrubland in California. Photo by J. E. Keeley.
underground bulbs, known as geophytes, may remain
dormant for long periods of time between fires.
All of the other shrub species listed above are not fire
dependent and produce seeds that germinate soon after
dispersal; however, successful reproduction is relatively
uncommon. This is because their seedlings are very sen
sitive to summer drought and because there are a number
of herbivores that live in the chaparral understory and
prey on seedlings and other herbaceous vegetation. These
include deer mice (Peromyscus maniculatus), woodrats
(Neotoma fuscipes), and brush rabbits (Sylvilagus bachmani).
Both rodents (mice and rats) are nocturnal; however,
evidence of woodrats, or packrats as they are sometimes
called, is very evident in many older chaparral stands
because of the several foot high nests of twigs they make
under the shrub canopy. These animals not only affect
community structure by consuming most seedlings and
herbaceous species, but also are important vectors for
disease and other health threats. For example, deer mice
are host to the deadly hanta virus and woodrats are a
critical host for kissing bugs (family Reduviidae) that
can cause lethal allergic responses in humans. All animals
including reptiles act as hosts for Lyme disease carrying
ticks (Ixodes pacificus). The browser of mature shrubs is the
black tailed deer (Odocoileus hemionus), although many are
attacked by specific gall forming wasps and aphids. Often
scrub oak will have large fruit like structures produced by
gall wasp (family Cynipidae). The adult wasp oviposits on
a twig, leaf, or flower and the developing larvae hijack the
metabolic activities of the plant cells and force it to
produce a highly nutritious spongy parenchymous tissue
for the developing wasp larva.
These shrubs that reproduce in the absence of fire have
successful seedling establishment largely restricted to more
mesic plant communities such as adjacent woodlands, or to
very old chaparral with deep litter layers that enhance the
moisture holding capacity of the soil. When seedlings do
establish under the shrub canopy, they typically persist for
decades as stunted saplings in the understory. These saplings
are heavily browsed by rodents and rabbits and often will
produce a swollen woody basal burl that survives browsing
and continually sprouts new shoots. If these saplings survive
until fire, they are capable of resprouting from their basal
burl after fire and exhibit a growth release that enhances
their chances of recruiting into the mature canopy during
early succession. Thus, in some sense these shrubs may be
indirectly fire dependent for completion of their life cycle.
Chaparral has a number of herbaceous or woody (lia
nas) vines, including manroot (Marah macrocarpus) and
chaparral honeysuckle (Lonicera spp.). These vines over
top the canopy of the shrubs and flower on an annual or
near annual frequency. The former produce fleshy spiny
fruits with very large seeds that are highly vulnerable to
predation and the latter dry capsules with light seeds that
may be wind borne. Both have weak seed dormancy and
often establish seedlings in the understory.
Yucca (Yucca whipplei) is a fibrous leaved species that
persists as an aboveground rosette of evergreen leaves. It
often survives fire because it prefers open rocky sites with
very little vegetation to fuel intense fires. Because they
are monocotyledonous species they have a central mer
istem that is protected by the outside leaves, which can
withstand severe scorching. This species flowers prolifi
cally after fire and exhibits a remarkable mutualism with
the tiny yucca moth (Tegiticula maculata). Moth pupae
survive in the soil and emerge in the growing season as
adults that fly to yucca flowers where they collect pollen.
They then instinctively fly to another yucca plant and
pollinate the flower, ensuring cross pollination, and then
oviposit an egg in the base of the ovary. This egg soon
hatches and the larva feeds on the developing seeds.
Yucca moths only reproduce on yucca flowers and yuccas
apparently require the pollinator services of this moth for
successful seed production, a classic example of symbiosis.
Community Succession
Chaparral succession following some form of distur
bance such as fire is somewhat different than in many
other ecological communities. Generally all of the spe
cies present before fire in chaparral will be present in
the first growing season after fire, and thus chaparral
has been described as being ‘auto successional’, mean
ing it replaces itself. In the absence of disturbance
chaparral composition appears to remain somewhat sta
tic with relatively few changes in species composition
or colonization by new species. In part because of the
rather static nature of chaparral, old stands have been
described with rather pejorative terms such as ‘senes
cent’, ‘senile’, ‘decadent’, and ‘trashy’, and considered to
be very unproductive with little annual growth. This
Chaparral
notion derives largely from wildlife studies done in the
mid twentieth century that concluded, due to the
height of shrubs in older stands, there was very little
browse production for wildlife. However, if total stand
productivity is used as a measure, very old stands of
chaparral appear to be very productive and are not
justly described as senescent. Also, these older commu
nities appear to retain their resilience to fires and other
disturbances, as illustrated by the fact that recovery
after fire (see below) in ancient stands (150 years old)
recover as well as much younger stands.
197
Fire
foliage in the summer and fall and spread by the dense
contiguous nature of these shrublands. Fires have likely
been an important ecosystem process since the origin of
this vegetation in the late Tertiary Period, more than
10 Ma, if not earlier. Until relatively recently the primary
source of ignitions was lightning from summer thunder
storms. Fires would largely have been ignited in high
interior mountains and coastal areas would have burned
less frequently and only when these interior fires were
driven by high winds with an offshore flow. In many
parts of California such winds occur every autumn and
are called Santa Ana winds in southern California and
Diablo winds or Mono winds in northern California.
When Native Americans colonized California at the end
of the Pleistocene Epoch around 12 000 years ago, they too
became a source of fires, and as their populations greatly
increased over the past few thousand years humans likely
surpassed lightning as a source of fire, at least in coastal
California. Today humans account for over 95% of all fires
along the coast and foothills of California.
Chaparral fires are described as crown fires because
the fires are spread through the shrub canopies and
usually kill all aboveground foliage. Normally, following
a wet winter, high fuel moisture in chaparral shrubs makes
them relatively resistant to fire. The amount of dead
branches is important to determining fire spread because
they respond rapidly to dry weather and combust more
readily than living foliage. As a consequence, fires spread
readily in older vegetation with a greater accumulation of
dead biomass. However, there is a complex interaction
between live and dead fuels, wind, humidity, tempera
ture, and topography. In particular, wind accelerates fire
spread primarily by heating living fuels and often can
result in rapid fire spread in young vegetation with rela
tively little dead biomass. Fires burning up steep terrain
also spread faster for similar reasons.
The marked seasonal change in climate is conducive to
massive wildfires, which are spawned by the very dry shrub
Community Recovery from Wildfires
Figure 2 Bare zone between chaparral and grassland. Photo
by J. E. Keeley.
Rate of shrub recovery varies with elevation, slope aspect,
inclination, degree of coastal influence, and patterns of
precipitation. Recovery of shrub biomass is from basal
resprouts (Figure 3) and seedling recruitment from a
dormant soil stored seed bank. After a spring or early
summer burn, sprouts may arise within a few weeks,
whereas after a fall burn, sprout production may be
delayed until winter. Regardless of the timing of fire,
seed germination is delayed until late winter or early
spring and is less common after the first year. Resilience
of chaparral to fire disturbance is exemplified by the
marked tendency for communities to return rapidly to
prefire composition.
Shrub species differ in the extent of postfire regenera
tion from resprouting versus reproduction from dormant
Allelopathy
The lack of shrub seedlings and herbaceous plants in the
understory of chaparral and related shrublands has led to
extensive research on the potential role of allelopathy,
which is the chemical suppression by the overstory shrubs
of germination (known as enforced dormancy) or growth
of understory plants. Often this lack of growth extends to
the edge where these shrublands meet grasslands, and
forms a distinct bare zone (Figure 2). The importance
of allelopathy has long been disputed, with some scientists
arguing that animals in the shrub understory are the
primary mechanism limiting seedlings and herbaceous
species from establishing. While research has not comple
tely ruled out the possibility of chemical inhibition, it is
known that for a large portion of the flora, allelopathy has
no role in seed dormancy but rather dormancy is due to
innate characteristics that require signals such as heat and
smoke to cue germination to postfire environments rich in
nutrients and light.
Chaparral
Figure 3 Postfire resprouts from basal burl of chamise with
meter stick. Photo by J. E. Keeley.
seed banks. Most species of manzanita and ceanothus have
no ability to resprout from the base of the dead stem and
thus are entirely dependent on seed germination. Such
shrubs are termed ‘obligate seeders’. A few species of
manzanita and ceanothus as well as chamise resprout
and reproduce from seeds, and these are referred to as
‘facultative seeders’. The majority of shrubs listed above,
however, regenerate after fire entirely from resprouts and
are ‘obligate resprouters’.
In the immediate postfire environment the bulk of
plant cover is usually made up of herbaceous species
present prior to the fire only as a dormant seed bank
or as underground bulbs or corms. This postfire com
munity comprises a rich diversity of herbaceous and
weakly woody species, the bulk of which form an
ephemeral postfire successional flora. This ‘temporary’
vegetation is relatively short lived, and by the fifth
year shrubs will have regained dominance of the site
and most of the herbaceous species will return to their
dormant state. These postfire endemics arise from
dormant seed banks that were generated after the
previous fire and typically spend most of their life as
dormant seeds. These are termed ‘postfire endemics’
and they retain viable seed banks for more than a
century without fire until germination is triggered by
heat or smoke of a fire. Postfire endemics are highly
restricted to the immediate postfire conditions and if
the second year has sufficient precipitation may persist
a second year but usually disappear in subsequent
years.
Not all of the postfire annuals are so restricted, rather
some are quite opportunistic, taking advantage of the
open conditions after fire but persisting in other openings
in mature chaparral. Such species often produce poly
morphic seed pools with both deeply dormant seeds that
remain dormant until fire and nondormant seeds capable
of establishing in or around mature chaparral. These
species fluctuate in relation to annual precipitation pat
terns, often not appearing at all in dry years.
Herbaceous perennials that live most of their lives as
dormant bulbs in the soil commonly comprise a quarter of
the postfire species diversity. Nearly all are obligate
resprouters, arising from dormant bulbs, corms, or rhi
zomes and flowering in unison in the first postfire year.
Almost none of them produce fire dependent seeds; how
ever, reproduction is fire dependent because postfire
flowering leads to produce nondormant seeds that readily
germinate in the second year.
Diversity in chaparral reaches its highest level in the
first year or two after fire. It is made up of a large
number of relatively minor species and a few very
dominant species and is illustrated by dominance–
diversity curves (Figure 4). Dominance in chaparral is
driven by the fact that a substantial portion of resources
are taken by vigorous resprouting shrubs and much less
is available for the many annual species regenerating
from seed.
Plants are not the only part of the biota that has
specialized its life cycle to fire. Smoke beetles
700
600
500
Cover
198
400
300
200
100
0
0
5
10
15
20
Sequence
Figure 4 Dominance–diversity curve based on cover of species
in sequence from highest to lowest from postfire chaparral.
Chaparral
(Melanophila spp.) are widely distributed in the western
US and are attracted by the infrared heat given off by
fires. Often while stems are still smoldering they will bore
into the scorched wood and lay their eggs.
Seed Germination
Many chaparral species have fire dependent regenera
tion, meaning that dormant seeds in the soil require a
stimulus from fire for germination. A few species have
hard seeds that are cracked by the heat of fire and this
stimulates germination. Ceanothus seeds are a good
example of this germination mode. However, for the
majority of species, seeds do not respond to heat but
rather to chemicals generated by the burning of plant
matter. This can result from exposure to smoke or charred
wood. In many of these species seeds will not germinate
when placed at room temperature and watered, unless
they are first exposed to smoke or charred wood. In
natural environments the seeds remain dormant for dec
ades until fire. There is evidence that a variety of
chemicals in smoke and charred wood may be responsible
for stimulating germination of postfire species, and both
inorganic and organic compounds may be involved.
Seeds of many species have a requirement for cold
temperatures (<5 C), which is interpreted as a seasonal
cue, but in these chaparral species this requirement is not
like the cold stratification requirement of many species from
colder climates, where the seeds require a certain duration
of cold in order to prevent winter germination. In California
species just a short burst of cold often will trigger germina
tion; thus, cold is not a cue that winter is over (as with more
northern latitude species) but rather that winter has arrived,
which is consistent with the winter germination behavior of
these Mediterranean climate species.
Seed Dispersal
Shrubs can be divided into those with temporal dispersal
versus those with strong spatial dispersal. The former are
the fire dependent species that accumulate dormant seed
banks, which in essence disperse these shrubs in time,
from one fire cycle to the next. Within this group there
is limited spatial dispersal. Ceanothus have explosive
capsules that shoot seeds a short distance of a meter or
two from the parent shrub. Manzanitas drop most of
their seeds beneath the parent plant because their small
dry fruits are not attractive to birds, although a small
number of the seeds are distributed further by coyotes
(Canis latrans) and bears (Ursus americans, historically also
included U. horribilis). Chamise produces small light fruits
that may be carried tens of meters or more by the wind
199
but it appears that most are distributed around the parent
shrub.
The postfire endemic annuals also have seeds that are
largely dispersed in time rather than in space. Most do
not have characteristics suggestive of widespread disper
sal. For example, the fire following whispering bells
(Emmenanthe penduliflora) derives its name because the
flowers and fruits are pendulous and drop seeds directly
beneath the parent plant. Postfire endemics in the sunflower
family (Asteraceae), a family noted for well developed dis
persal with dandelion like pappus, commonly have
deciduous pappus, which ensures dispersal in time rather
than in space.
Shrub species that exhibit fire free (nonfire dependent)
reproduction have fruits highly attractive to birds and
mammals, and the bulk of the seed crop appears to be
dispersed by these vectors. Seedling recruitment is sensi
tive to desiccation and thus it is of some significance that
one of the main dispersers of these fruits, the scrub jay
(Aphelocoma californica), preferentially caches seeds in the
shade.
Regional Variation in Fire Regime
California chaparral exhibits regional differences in burn
ing patterns and largely due to regional variation in
winds. In much of coastal California autumn winds create
severe fire conditions. These occur every year and result
in 5–10 days of strong offshore flow with windspeeds of
100 kph or more. These winds result from a high pressure
system in the interior West, and are known as Santa Ana
winds in southern California and Diablo or Mono winds
in northern California. As these air masses move from the
high pressure cell in the interior to a low pressure trough
off the coast, the air descends and dries adiabatically,
resulting in relative humidity below 10%. The fact that
these winds occur every year and arrive at the end of an
extended drought results in one of the most severe fire
conditions in the world. As a consequence only a small
portion of southern California landscape has escaped fire
during the last century, and much of the lower elevation
chaparral has burned at an unnaturally high frequency.
In contrast, Santa Ana winds are absent from the south
ern Sierra Nevada and parts of the central coast, in part to
mountain barriers that fail to funnel these winds coast
ward. This, coupled with lower human population
density, has resulted in many fewer fires. As a conse
quence nearly half of the landscape in the southern
Sierra Nevada has not had a fire for well over a century.
This condition places these landscapes at the upper end of
the historical range of variability. Nonetheless, these
older stands of chaparral appear to maintain natural eco
system processes and exhibit no sign of dying out or
replacement by other vegetation types. This is
200
Chaparral
particularly evident, following fires in ancient stands of
chaparral from the region, that it exhibits vegetative
recovery in cover and diversity indistinguishable from
postfire recovery in younger stands.
Future Threats and Management
Degradation and type conversion of native shrublands to
alien dominated grasslands has been noted by numerous
investigators, some of whom contend that increased fre
quency of disturbance is the primary factor that favors
non native annuals over woody native species. In the
absence of fire, seeds of non natives have a low residence
time in the soil; thus, the presence of these species on
burned sites is more often due to colonization after fire.
Typically a repeat fire within the first postfire decade is
sufficient to provide an initial foothold for aliens. In
addition to outcompeting native plant species, non native
grasses alter the fire regime from a crown fire regime to a
mixture of surface and crown fires, where highly com
bustible grass fuels carry fire between shrub patches. This
increases the likelihood of fires and ultimately increases
fire frequency. As fire frequency increases there is a
threshold beyond which the native shrub cover cannot
recover.
Fire management practices potentially conflict with
natural resource needs. These landscapes currently
experience an unnaturally high frequency of fire; thus,
much of it is at risk for alien invasion. When fire managers
add to this by using prescription burning and other fuel
manipulations, they open up these shrublands and expose
them to invasion and potential type conversion to non
native grasslands. In managing these landscapes it might
be helpful to consider the fact that the vast majority of
alien species in California are opportunistic species that
capitalize on disturbance. Adding additional disturbance
through prescription burning (or grazing) will only
exacerbate the alien problem.
Very little chaparral landscape is protected in parks or
wilderness areas. Much of it is in private hands or under
federal jurisdiction. Historically, it has largely been man
aged as rangeland by frequent burning to destroy the
chaparral cover, or burned to reduce fuels perceived to
be hazardous to more desirable forests or urban environ
ments. Today the expansion of urban development has
resulted in large portions of urban communities being
juxtaposed with watersheds of potentially dangerous cha
parral fuels. Historical studies show that large high
intensity crown fires are a natural part of this ecosystem
and there is little reason to believe there will not be more
such fires in the future. Fire management has always
worked under the philosophy that they can change
the vulnerability of communities to wildfires through
manipulation of fuels. However, over the past century of
such management, every decade has been followed by one
of increasing losses to wildfires. Californians need to
embrace a different model of how to view fires on these
landscapes. Our response needs to be tempered by
the realization that these are natural events that cannot
be eliminated from the southern California landscape. We
can learn much from the science of earthquake or other
natural hazard management. No one pretends they can
stop earthquakes; rather, they engineer infrastructure to
minimize impacts. In the future, living safely with fire is
not going to be achieved solely by fire management
practices, but will require close integration with urban
planning.
See also: Mediterranean.
Further Reading
Arroyo MTK, Zedler PH, and Fox MD (eds.) (1995) Ecology and
Biogeography of Mediterranean Ecosystems in Chile, California and
Australia. New York: Springer.
Christensen NL and Muller CH (1975) Effects of fire on factors
controlling plant growth in Adenostoma chaparral. Ecological
Monographs 45: 29 55.
Halsey RW (2004) Fire, Chaparral, and Survival in Southern California.
San Diego, CA: Sunbelt Publications.
Halsey RW (2005) In search of allelopathy: An eco historical view of the
investigation of chemical inhibition in California coastal sage scrub
and chamise chaparral. Journal of the Torrey Botanical Society
131: 343 367.
Keeley JE (2000) Chaparral. In: Barbour MG and Billings WD (eds.)
North American Terrestrial Vegetation, pp. 203 253. Cambridge:
Cambridge University Press.
Keeley JE and Fotheringham CJ (2003) Impact of past, present, and
future fire regimes on North American mediterranean shrublands.
In: Veblen TT, Baker WL, Montenegro G, and Swetnam TW (eds.)
Fire and Climatic Change in Temperate Ecosystems of the Western
Americas, pp. 218 262. New York: Springer.
Mooney HA (ed.) (1977) Convergent Evolution of Chile and California
Mediterranean Climate Ecosystems. Stroudsburg, PE: Dowden,
Hutchinson and Ross.
Odion DC and Davis FW (2000) Fire, soil heating, and the formation of
vegetation patterns in chaparral. Ecological Monographs
70: 149 169.
Rundel PW, Montenegro G, and Jaksic FM (eds.) (1998) Landscape
Disturbance and Biodiversity in Mediterranean Type Ecosystems.
New York: Springer.
Wells PV (1969) The relation between mode of reproduction and extent
of speciation in woody genera of the California chaparral. Evolution
23: 264 267.
Coral Reefs
201
Coral Reefs
D E Burkepile and M E Hay, Georgia Institute of Technology, Atlanta, GA, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
The Coral–Algal Mutualism
Ecological Interactions on Coral Reefs
Replenishment of Coral Reefs: The Role of
Reproduction and Recruitment in the Ecology of
Reefs
Landscape Ecology of Coral Reefs: Connections of
Coral Reefs to Mangrove and Seagrass Systems
Geographic Distribution of Coral Reefs
Threats to Coral Reefs
Summary
Further Reading
Introduction
abundance and also record chemical signals of past cli
matic events, like temperature and sea level changes.
Thus, reefs not only feed and protect humans and other
species, but also provide a valuable window into our past,
including how our present activities may be changing our
environment, and possibly our future.
In this article, we review the major ecological interac
tions that shape coral reef ecosystems. We pay particular
attention to (1) the dynamic relationship between corals
and the symbiotic algae living within their tissues, (2) the
role of reef herbivores in protecting corals from being over
grown by seaweeds, (3) the numerous ecological processes
such as predation, competition, recruitment of juvenile reef
organisms, and disturbance that influence the structure of
coral reefs, and (4) the dynamic ecological connections
between reefs and nearby ecosystems such as seagrass
beds and mangrove forests. Finally, we review the current
dangers to coral reefs and how these threats undermine the
ecological integrity of these diverse ecosystems.
Corals are simple, clonal invertebrates that serve as ecosys
tem engineers, building living structures (reefs) so large that
they can be seen from space. These structures, which rival
the greatest feats of human engineering, are powered
through symbiosis with single celled algae that are housed
within the coral animal. This coral–algal cooperation facil
itates a productive ecosystem that can grow in the nutrient
poor ‘desert’ of isolated tropical seas. The rich structural
complexity provided by the coral’s hard bodies gives shelter
to many other species of plants and animals making coral
reefs among the Earth’s most biologically diverse ecosys
tems, harboring hundreds of thousands to millions of
species worldwide (Figure 1).
Coral reefs also support human societies by providing
critical sources of protein, protecting coasts from dama
ging waves, attracting tourists, and serving as the
backbone of the economies for many tropical islands. In
addition, coral reefs are crucial in the fight against
human diseases, as many of the plants and animals that
live on coral reefs produce chemicals that are useful as
pharmaceuticals. Reefs have also fascinated naturalists
and scientists for centuries. Before publishing his
groundbreaking work on natural selection, Charles
Darwin published a treatise on reefs in 1842 hypothesiz
ing that coral atolls (rings of reefs in the deep tropical
Pacific) were formed around mountain tops as these
mountains sank back into the Earth’s crust under their
own weight. It was more than 100 years before drilling
technologies developed to the point where this hypothesis
was tested – like with so many other aspects of Darwin’s
writings, he proved to be correct. For modern ecologists,
reefs serve as a model ecosystem for developing basic
hypotheses about the ecology and evolution of population
structure, of community organization, and about how spe
cies diversity evolves and is maintained. In addition, reefs
give us a glimpse of the spectacular record of Earth’s
history because the hard skeletons of corals fossilize to
provide a long record of changes in coral distribution and
The Coral–Algal Mutualism
Corals are ecosystem engineers in that the growth of their
calcium carbonate skeleton creates the biogenic structure
on which the entire ecosystem depends. The calcification
and growth of reef corals depends on a mutualism
between corals and their intracellular, photosynthetic
dinoflagellates, Symbiodinium spp., also known as zoo
xanthellae. Both corals and zooxanthellae benefit from
this relationship, as corals can receive up to 95% of
their carbon from the zooxanthellae’s photosynthesis,
while the zooxanthellae acquire the nitrogen and other
inorganic nutrients in coral excretion products for their
growth. Photosynthesis by zooxanthellae enhances calci
fication in corals and increases coral growth rates,
ultimately leading to reef accretion and the massive
reef framework found in many tropical seas. Thus, the
physical structure of live and dead corals created by the
coral–zooxanthellae mutualism provides heterogeneity
202
Coral Reefs
Figure 1 Coral reefs, like this one in the Indo-Pacific, harbor
hundreds of thousands to millions of species worldwide. Photo
credit M.E. Hay.
and habitat complexity, facilitating the coexistence of
diverse plant and animal assemblages.
Although zooxanthellae were initially assumed to
represent one species, recent molecular evidence shows
that there are at least seven distinct types or clades
(referred to as clades A–G). Many corals house multiple
clades of zooxanthellae, setting the stage for possible com
petition among symbionts and for symbiont selectivity by
the host. Clades of zooxanthellae differ in their photosyn
thetic capacity and their tolerance of light, temperature,
and other stressors, making them differentially useful to
their hosts under changing environmental conditions.
When corals are stressed by increasing light levels or
temperatures, they often expel their zooxanthellae and
become pale in color (called coral bleaching). This process
of bleaching may allow corals to take up new clades of
zooxanthellae that are better adapted to the new envi
ronmental conditions. However, corals that fail to re
acquire zooxanthellae or acquire the wrong clades may
ultimately die from the stress, suggesting that a failure of
corals to acquire appropriate symbionts can be fatal under
changing environmental conditions. Such alterations in
the coral–zooxanthellae mutualism may allow corals
greater flexibility in adapting to global climate change,
which is a major threat to the health of coral reefs and
the integrity of the coral–zooxanthellae mutualism.
Ecological Interactions on Coral Reefs
Competition
Competition for limiting resources such as nutrients, space,
light, or food is often a strong mechanism limiting the
distribution and abundance of species in communities. On
many coral reefs, the limiting resource for most benthic
organisms is space or light, as most of the reef structure is
often occupied (Figure 2). Consequently, corals have
Figure 2 Competition for space is often an important
ecological force structuring coral reefs as corals and other
invertebrates cover most of the benthos on healthy coral reefs.
Photo credit M.E. Hay.
evolved a variety of competitive mechanisms including
sweeper tentacles, digestive filaments, and rapid growth
rates that allow them to fight neighbors for new space or
protect the space they already occupy. Slow growing, mas
sive corals often have the most potent direct competitive
mechanisms (i.e., sweeper tentacles and digestive filaments
that can sting and directly harm neighboring corals) while
branching corals such as many Acropora spp. rely on their
high growth rates to overtop and shade competitors. Other
reef invertebrates such as sponges exude chemical com
pounds that are toxic to their neighbors, essentially using
chemical warfare (termed allelopathy) to gain new space.
Although most early studies of competition on reefs
focused on coral–coral competition, more recent studies
have examined coral–seaweed competition because reefs
are now more commonly overgrown by seaweeds that
periodically seem to be killing corals. The conventional
wisdom is that seaweeds are competitively superior and
can overgrow and kill most corals. Although not all sea
weeds are harmful to corals, most studies of coral–algal
competition show that direct competition from seaweeds
Coral Reefs
reduces the growth, survivorship, fecundity, and recruit
ment of many corals. Contact with the calcareous green
seaweed Halimeda opuntia has even been shown to induce
black band disease in some corals. Small, filamentous
seaweeds, which are not as directly harmful to corals as
are larger, foliose seaweeds, often trap sediments next to
coral tissue, and this can smother and kill corals. Thus,
even competition with typically innocuous filamentous
seaweeds can be harmful on reefs that receive high sedi
ment loads. However, corals are not uniformly susceptible
to competition from seaweeds, and competitive outcomes
may vary with coral morphology. Foliose corals such as
Agaricia spp. are more susceptible to seaweed overgrowth
than massive corals such as Montastrea spp. In addition,
seaweeds have disproportionately high negative effects on
smaller coral colonies, particularly newly recruited corals,
and large stands of seaweed can prevent juvenile corals
from recruiting to reefs at all.
Competition on coral reefs is not limited to sessile inver
tebrates; mobile animals also compete. Because herbivores
are abundant on undisturbed coral reefs and standing bio
mass of seaweeds is generally low in these conditions,
competition between herbivores would be expected. When
the herbivorous sea urchin Diadema antillarum was removed
from Caribbean reefs by either purposeful experimental
manipulations or by disease outbreak, feeding rates of her
bivorous fishes increased, as did the densities of some
species, suggesting that fishes and urchins competed for
food. Diadema also competed intensely with each other for
food. However, competition for limiting algal resources
generally did not result in a decrease in the size of Diadema
populations but an increase in the size of their mouthparts
(called the Aristotle’s lantern) relative to the size of their
body. Basically, Diadema bodies would shrink in size when
food was limiting as a tradeoff between growth and survival.
203
seaweeds replace corals and the biogenic structure of
the reef degrades. Both reductions in coral structure
and increases in seaweeds are associated with losses of
herbivorous reef fishes. Interestingly, large scale manual
removals of seaweeds from reefs have resulted in only
temporary increases in herbivorous fish abundance with
seaweeds becoming the dominant benthic organism once
again after several months. Thus, reductions in seaweeds
without recovery of corals may inhibit the recovery of
many reef fishes, leading to the continued degradation of
coral reefs.
The main herbivorous fishes on coral reefs are generally
surgeonfishes (Acanthurdiae) and parrotfishes (Scaridae)
with rabbitfishes (Siganidae), chubs (Sparidae), and dam
selfishes (Pomacentridae) also responsible for considerable
herbivory in some locations (Figure 3). Surgeonfishes
typically feed on turf algae and foliose seaweeds with
some species feeding primarily on detrital material.
Parrotfishes have robust jaws with teeth fused into a
beak like formation (hence the name parrotfish), which
allows them to feed on tough, calcified seaweeds in addi
tion to algal turfs and foliose seaweeds. Although the
important role of herbivores in influencing reef community
is well established, less is known about the importance of
individual herbivore species or the role of herbivore diver
sity in affecting coral reef health. Herbivore diversity
should benefit reefs because a more diverse herbivore
assemblage should include herbivores with varied attack
strategies, which in turn should increase the efficiency of
seaweed removal because particular seaweeds are unlikely
to be well defended against all types of herbivores.
Experimental manipulations of herbivorous fish diversity
demonstrate that species richness is important for reef
function because complementary feeding by different
herbivorous fishes suppresses upright seaweeds, facilitates
crustose corallines and turf algae, reduces coral mortality,
Herbivory
Because seaweeds can overgrow and kill corals, herbi
vores are critical for coral reef function because they
keep reefs free of seaweeds, thus facilitating the recruit
ment, growth, and resilience of corals. Fishes and urchins
are typically the dominant herbivores on coral reefs with
fishes in some reef areas biting the bottom at rates of
>100 000 times per m2 every day. When in sufficient
numbers, either fishes alone or sea urchins alone can
remove greater than 90% of the daily primary production
on reefs. By feeding on seaweeds that are competitively
superior to corals, herbivorous fishes both clear the sub
strate for settling coral larvae and prevent seaweed
overgrowth of established corals. In return, the biogenic
structure and topographic complexity of reef corals ben
efit herbivorous reef fishes and urchins by providing food,
habitat, and refuges from predation. When herbivores
are removed by experimentation, overfishing, or disease,
Figure 3 Herbivores, like this mixed-species school of
parrotfishes in the Caribbean, are important to coral reef health
because they remove seaweeds that would otherwise overgrow
and kill corals. Photo credit M.E. Hay.
204
Coral Reefs
and promotes coral growth. Hence not only are herbivores
critical for coral reefs, but herbivore species richness is also
essential as a range of feeding strategies and physiologies
allows efficient removal of seaweeds and promotes coral
health.
Predation
Predation is often a strong top down force in ecosystems
mediating coexistence of lower trophic level species by
preventing competitive exclusion among ecologically
similar organisms. In fact, predators often maintain spe
cies diversity in ecological communities by preventing
expansions of certain prey that would otherwise outcom
pete competitively inferior organisms and come to
dominate the community. If important predators are
removed from a food web, the absence of their strong
effects can ripple throughout the system, fundamentally
altering a variety of predator–prey interactions.
The effects of the largest predators on reefs such as
sharks, jacks (Carangidae), and large groupers
(Serranidae) are virtually unknown due to the logistical
problems of studying such large creatures and the fact that
the majority of these species were rare before ecologists
began studying reef ecology in situ (Figure 4). Although
rigorous study of the roles that these fishes play in com
munities has been limited, a recent model of a Caribbean
reef food web suggests that sharks are often the most
strongly interacting species in these webs indicating that
their removal may have had strong cascading effects on
reefs. Further, surveys of lightly fished reefs in the north
western Hawaiian Islands showed that large apex
predators such as sharks and jacks represented >50% of
the total fish biomass as compared to <3% on heavily
fished reefs from the main Hawaiian Islands. These abun
dant apex predators on lightly fished reefs surely exert a
strong top down force on the community structure of
these reefs.
However, the human exploitation of medium sized
predatory fishes has given us the best insight into how
predation influences reef communities. On many
Pacific coral reefs, outbreaks of the crown of thorns
starfish, Acanthaster planci, cause loss of many square
kilometers of coral reefs. These starfish are voracious
coral predators that forage in large groups of up to
hundreds of thousands of individuals that can deci
mate large stands of coral, and their outbreaks have
become more frequent since the 1960s when they
were first documented. Research in the Fiji islands
has shown that outbreaks of Acanthaster are correlated
to fishing pressure on reefs. Acanthaster are 1000 times
more dense around islands that have high fishing
pressure and low predatory fish abundance than they
are on reefs that have light fishing pressure and high
predator abundance. High densities of Acanthaster
Figure 4 Apex predators like sharks (top) and groupers
(bottom) are now rare on many coral reefs due to overfishing.
Photo credit M.E. Hay.
decrease cover of reef building corals and crustose
coralline algae while increasing cover of filamentous
algae. Thus, the removal of large predators is asso
ciated with explosions of Acanthaster populations that
then have strong cascading effects on the organization
of reef communities.
A similar situation exists on many reefs in eastern
Africa where intense fishing of triggerfishes and large
wrasses allows population explosions of sea urchins such
as Echinometra mathaei. Reefs unprotected from fishing
have six times more urchins than protected reefs, and
feeding by these dense urchin populations physically
erodes the reef structure once most of the algal biomass
has been consumed. This intense grazing decreases coral
cover and diversity as well as increases bioerosion rates
up to 20 fold compared to reefs that are protected from
fishing and have abundant predators. When formerly
fished areas were protected from fishing, urchin eating
fishes became more abundant and predation on urchins
increased, suggesting that recovery of predatory fish
populations should lead to lower urchin populations and
a recovery of reef structure over time.
Coral Reefs
Positive Interactions
Ecologists now realize that positive interactions between
species can have strong, cascading effects on natural com
munities and are no less important than negative
interactions (i.e., predation or disturbance) in affecting com
munity structure. On reefs, the most obvious positive
interaction is the mutualism between corals and their sym
biotic algae. Another is the positive feedback between
herbivores and corals that maintains a coral dominated
7m
10 m
15–20 m
35 m
80
60
Coral cover (%)
Although biotic interactions (e.g., competition and her
bivory) are emphasized as having important consequences
for coral reef structure, abiotic disturbances such as hur
ricanes, temperature fluctuations, sedimentation stress,
and sea level change also produce long lasting effects on
reefs. Coral reefs are one of the hallmark ecosystems
strongly influenced by disturbance as the frequency and
intensity of hurricanes or disturbance events determines
how many species of corals coexist on reefs. If disturbance
is very frequent or very intense, then only species that can
recolonize disturbed areas quickly or that can withstand
intense disturbances will persist. If disturbance is infre
quent and mild, then the most competitive species
eliminate the less competitive species and come to domi
nate. However, if disturbance is of an intermediate
frequency and intensity, then species with different life
history characteristics (i.e., good colonizers vs. good com
petitors) can coexist because the disturbance intolerant
species are not displaced frequently and the poor compe
titors are not outcompeted.
Reefs often recover from acute disturbances such as
storms but infrequently recover from chronic distur
bances. The coupling of acute natural disturbances with
chronic anthropogenic disturbances often leads to preci
pitous declines in coral reef health. One of the best
examples of compounded disturbances driving coral reef
decline is from the reefs of Jamaica. Chronic overfishing
of herbivorous fishes compounded with two hurricanes
and the mass mortality of the herbivorous sea urchin
Diadema antillarum acted synergistically to force these
once coral dominated reefs into an alternate state of sea
weed dominance (Figure 5). In more than two decades,
these reefs have shown few signs of recovery. In fact, the
episodic effects of natural physical disturbances, coral
disease, and coral bleaching along with the constant
anthropogenic disturbances of overfishing and pollution
have combined to decrease coral cover an average of 80%
on reefs throughout the Caribbean over the past few
decades. Although disturbance is a natural and integral
part of coral reef ecosystems, the compounding of many
disturbances over short timescales is often more than reefs
can withstand.
(a)
40
20
0
(b)
100
80
Algal cover (%)
Disturbance
205
60
40
20
0
1975
1980
1985
Year
1990
1995
Figure 5 Degradation of Jamaican coral reefs over two
decades. Changes in (a) coral cover and (b) seaweed cover at
four depths in Discovery Bay, Jamaica. This decline in corals and
increase in seaweeds was the result of synergistic interactions of
natural and anthropogenic disturbances including overfishing,
hurricanes, disease, and eutrophication. A similar decrease in
coral cover occurred throughout Jamaica with coral cover
nationwide declining from about 60% to about 4%. From Hughes
TP (1994) Catastrophes, phase shifts, and large-scale
degradation of a Caribbean coral reef. Science 265: 1547–1551.
ecosystem. Other crucial positive interactions come from
species that are normally thought of as competitors but can
mutually benefit each other under the right conditions.
Sponges, for example, compete with each other, but can
also interact positively. It is more common to find morpho
logically similar species of sponge growing intermingled in
multispecies groups than it is to find a sponge colony grow
ing alone. When growing in these groups, the growth rates
of the different species of sponge are often greater than what
they would be if these sponges were growing by themselves.
This enhancement of growth rates may stem from differ
ences among the species in their susceptibility to predation,
pathogens, and physical disturbance. The summed traits of
206
Coral Reefs
the sponge consortia may enable participants to survive
environmental challenges that would be insurmountable
for any of them growing alone. Further, sponges are impor
tant to the stability and integrity of the reef itself. Sponges
actually act as a type of cement that binds the reef together
and holds corals in place. When sponges are removed from
reefs, storms displace and kill more corals from these reefs
than from reefs that have an abundance of sponges.
Net positive interactions may also occur even
between consumers and their prey. Herbivorous dam
selfishes often form mutualisms with some seaweeds on
tropical reefs. Through aggressive defense of the algal
mats on which they feed, damselfish create patches of
species rich algae on reefs where these algae would
normally be grazed to near local extinction by large
herbivorous fishes. Although the rapidly growing fila
mentous algae in the damselfish’s territory are its prey,
they are also dependent on the territorial behavior of
the fish for their persistence at high density. If the
territorial fish is removed, its algal lawn is consumed
within hours. However, the positive interactions
between damselfishes and their algal gardens can be
overridden by cooperation among other species of her
bivorous fishes that forage in large schools. While
schooling would appear to increase the competition
for resources among herbivores, parrotfishes and sur
geonfishes often form feeding schools to overrun
territories of pugnacious damselfishes. For these large
herbivores, increasing school size allows for more bites
per individual fish when foraging in and around dam
selfish territories. The benefits of acting mutually to
overwhelm damselfish and gain access to resource rich
habitats must outweigh the potential for competitive
interactions between the fishes using these schools.
Similarly, piscivorous fishes such as grouper and
moray eels often hunt in mixed species or cooperative
foraging groups. For these fishes that forage coopera
tively, many mouths may be better than one in terms of
overall prey yield to each predator when summed over
a lifetime of hunts.
Finally, small wrasses (Labridae) and gobies
(Gobiidae) act as cleaner fishes on tropical reefs and
remove parasites, mucus, and dead or infected tissue
from larger fishes. Reef based cleaner fish are found at
specific cleaning stations, usually situated on prominent
portions of the reef. These fish can clean up to 2300
individuals of 132 different species in a day, and some
client fish visit cleaners over 100 times a day. If these
cleaner fish are removed from reefs, diversity of reef
fishes declines, especially for large, transient fishes that
may visit reefs specifically to be cleaned. Cleaner fishes
can, thus, have a strong effect on parasite loads of their
client fishes and on fish usage patterns across patchy reef
environments.
Replenishment of Coral Reefs: The Role of
Reproduction and Recruitment in
the Ecology of Reefs
Most reef organisms are sessile (corals, sponges, seaweeds) or
use only a small portion of a much larger reef habitat (most
reef fishes). Thus, colonization of new habitats is achieved
through recruitment of juvenile organisms that may drift for
long distances in the plankton before settling onto, and using
a small area of reef. Consequently, the production and
recruitment of juvenile organisms is a key factor in the
ecology of reefs as the replenishment of plant and animal
populations is integral to the resilience and recovery of reefs
in the face of natural and anthropogenic disturbances.
Coral species differ considerably in their modes of
reproduction and in the ability of their larvae to disperse
to new reefs. Many corals reproduce both asexually through
fragmentation and sexually by the production of gametes.
Important reef building corals such as acroporids are extre
mely successful at reproducing asexually and are dispersed
when storms break apart parent colonies and spread the
fragments to new portions of a reef where they can reattach
and grow. Sexual reproduction in corals is also variable in
that corals are typically either brooders or spawners.
Brooders release fertilized larvae into the water column
while spawners release sperm and eggs into the water
column, where they fertilize and disperse with the ocean
currents. These fertilized larvae will eventually exit the
plankton and return to reefs as newly recruited juvenile
corals. Research from the Great Barrier Reef, Australia has
shown that there is large variation in the abundance of coral
recruits across both large and small spatial scales. The best
predictor of differences in recruitment rates among reefs
was the fecundity, not abundance, of adult corals and
explained 72% of the variation in recruitment for acroporid
corals. Recruitment rates decreased dramatically as the
fecundity of adults decreased, but this decrease was not
linear; a small decrease in the fecundity of adults resulted
in a dramatic decrease in juvenile recruitment (Figure 6).
These results suggest that processes such as sedimentation,
eutrophication, and competition with seaweeds, all of which
reduce the fecundity of adult corals, could dramatically
affect the replenishment of coral populations.
Recruitment of juveniles is also important to the
replenishment of fish populations and considerable
research has gone into determining how recruitment pro
cesses affect the assembly of reef fish communities. Most
reef fishes, like corals, have planktonic larvae that can
disperse wide distances from their point of origin. One
of the key questions in the ecology of reef fishes is how
the recruitment of juvenile fishes is related to the density
of fishes already on the reef (i.e., whether local patterns of
recruitment are density dependent or density indepen
dent). Despite considerable research on the subject, little
Coral Reefs
250
No. of recruits
200
150
100
50
0
0
20
40
60
Percentage gravid
80
100
Figure 6 Relationship between the percentage of coral colonies
(Acropora hyacinthus) that were gravid and the number of coral
recruits. Each point is a separate reef on the Great Barrier Reef,
Australia. From Hughes TP, Baird AH, Dinsdale EA, et al. (2000)
Supply-side ecology works both ways: The link between benthic
adults, fecundity, and larval recruits. Ecology 81: 2241–2249.
consensus has been reached and studies have shown that
recruitment rates can be either positively or negatively
correlated with adult abundance (positively or negatively
density dependent), or show no correlation at all
(density independent). These relationships may vary with
the species being studied, with location, or with the currents
and physical processes prevailing at the time of the test.
Continued research is needed to generalize how recruit and
adult densities are related and how environmental and
biological variables change these relationships.
A key component to the replenishment of populations
of coral reef organisms is the extent to which reefs are
connected to other reefs (i.e., whether juveniles recruit to
reefs from local or distant sources). Coral reefs, and marine
ecosystems in general, differ from many terrestrial systems
in that juvenile organisms have the potential to ride ocean
currents and be dispersed over wide distances potentially
connecting geographically distant populations. However,
the actual extent that marine populations are connected to
each other is still a topic of vigorous debate. This knowl
edge is crucial to the protection and management of reefs
as the connectivity of populations of coral reef organisms
will determine whether local populations can be managed
with efforts based close to the target population (if the
system is fairly closed and recruitment from local popula
tions is frequent) or if management of local populations
will necessitate international cooperation (if reefs are fairly
open and recruitment is driven by larvae from distant
reefs). Thus, solving the question of connectivity among
reefs is critical to the preservation of reef health.
207
Initial models of connectivity for fish populations in the
Caribbean suggested that many of the populations were
very open and well connected to other populations hun
dreds of kilometers away. However, these models were
based on passive dispersal of fish larvae and simple models
of surface currents and did not account for the effect of
larval behavior on dispersal or for the effects of fine scale
oceanographic processes. Thus, viewing larvae as passive
dispersal agents may overestimate the actual dispersal of
larvae and the connections among reefs. Recent models of
connectivity in the Caribbean that account for larval
behavior suggest that fish populations are less connected
than assumed under passive dispersal models and that
different regions of the Caribbean are essentially isolated
from each other, at least on an ecological timescale.
However, there are considerable differences among regions
of the Caribbean as to the extent that reefs are connected to
distant areas as some regions appear to import a large
portion of recruits while other regions are primarily self
seeding. The differences in the relative importance of local
and long distance recruitment of juveniles among regions
of the Caribbean underscores the role that careful planning
will play in the implementation of marine reserves for
protection of coral reefs as understanding how reefs are
connected to one another will influence how large reserves
should be and where they should be located.
Landscape Ecology of Coral Reefs:
Connections of Coral Reefs to Mangrove
and Seagrass Systems
Coral reefs are typically found in close proximity to other
coastal ecosystems, particularly seagrass beds and man
grove forests. These different ecosystems are often
connected to reefs via the movement of animals and nutri
ents across their boundaries. For example, carnivorous
grunts (Haemulidae) forage in seagrass beds at night but
school around large coral heads on reefs during the day as a
refuge from predation. Coral heads that harbor fish schools
receive nutrient supplements from fish excretion, grow up
to 23% faster, and have more nitrogen and zooxanthellae
per unit area than do corals without resident fishes. Thus,
fishes that have no direct trophic link with corals collect
nutrients from other ecosystems (seagrass beds) and con
centrate these near their host coral. This facilitates coral
growth, enhancing the coral’s value as a refuge for these
fishes and for other reef organisms.
Mangroves and seagrass beds also serve as nursery
grounds and provide refuge from predators and an abun
dance of food for many juvenile fishes that are typically
found on coral reefs as adults. Grunts (Haemulidae),
snappers (Lutjanidae), barracuda (Sphyraena barracuda),
and some parrotfishes (Scaridae) are particularly
208
Coral Reefs
Mangroves present
F
B
A
C
E
D
Mangroves
Patch reef
Seagrass beds
Shallow forereef
Montastraea reef
Mangroves absent
G
Patch reef
Seagrass beds
Shallow forereef
Montastraea reef
Figure 7 Schematic illustrating the connection between mangroves and coral reefs. Ecosystem connectivity is stylized for Haemulon
sciurus (gray and black fish) and Scarus guacamaia (orange and green fish) although other parrotfish (scarid), grunt (haemulid), and
snapper (lutjanid) species also exhibited similar ontogenetic shifts in habitat use. H. sciurus show a substantial shift in size frequency from
seagrass (A) to mangroves at approximately 6 cm. On reaching a given size in seagrass beds, juvenile fish then move to mangroves (B)
which serve as an intermediate nursery habitat before migrating to patch reefs (C). If mangroves are not present, H. sciurus move directly
from seagrass to patch reefs, appearing on patch reefs (G) at a smaller size and at lower density (260 ha 1 compared to 3925 ha 1 in
mangrove-rich systems). In the presence of mangroves, the biomass of H. sciurus is significantly enhanced on patch reefs, shallow
forereefs, and Montastraea reefs (C, D, E). S. guacamaia (F) has a functional dependency on mangroves and is not seen where mangroves
are absent. Illustration describes findings from Mumby PJ, Edwards AJ, Arias-Gonzalez JE, et al. (2004) Mangroves enhance the biomass
of coral reef fish communities in the Caribbean. Nature 427: 533–536. Schematic and description courtesy of Peter J. Mumby.
dependent on the presence of nearby mangroves. In
Belize, reefs closely associated with mangroves have up
to 26 times more biomass of some species of fish than
reefs not associated with mangroves. A common species
on these reefs, the bluestriped grunt (Haemulon sciurus),
typically goes through an ontogenetic change in habitat
use as it migrates from seagrass beds to mangroves to
patch reefs to the forereef as it ages (Figure 7). In areas
where mangroves are absent, bluestriped grunts move
from seagrass beds directly to patch reefs and are typi
cally smaller than grunts that inhabit patch reefs with
nearby mangroves. Thus, mangroves provide important
habitat where juvenile grunts feed and increase in size
before moving to patch reefs which may subsequently
decrease the threat of predation once they move to these
reefs. Further, the rainbow parrotfish (Scarus guacamaia),
the largest herbivorous fish in the Caribbean, is func
tionally dependent on mangroves for shelter; juveniles of
this species live primarily in mangroves, and the species
goes locally extinct on reefs when nearby mangroves are
removed (Figure 7). Interestingly, density of fishes that
have no direct link to mangroves at any stage of their life
history can still be influenced by the proximity of
mangroves, probably via interactions with mangrove
dependent fishes. Thus, the composition of the fish com
munity on reefs is greatly influenced by the proximity of
mangroves, and the rapid removal of mangrove forests
from coastlines worldwide will certainly have drastic
negative impacts on the ecology coral reefs.
Geographic Distribution of Coral Reefs
Coral reefs exist in tropical areas worldwide (Figure 8). In
general, reefs are abundant in areas with shallow coastlines
and clear, warm water where riverine discharge of
Coral Reefs
209
Figure 8 Worldwide distribution of coral reefs. Coral reefs (red dots) cover roughly 250 000 km2 of the Earth’s surface, but
zooxanthellate corals inhabit a far wider range (blue shading). From Veron JEN (2000) Corals of the World, vols. 1–3. Townsville, QLD:
Australian Institute of Marine Science.
sediments is low. Large coral reefs are rarely found in areas
above 29 latitude where ocean temperatures fall below
18 C for extended periods as this slows coral growth and
their capacity to build large reefs; however, zooxanthellate
corals can be found in areas with water temperatures as low
as 11 C. In addition, herbivory is often less intense in
cooler waters meaning that seaweeds are more abundant
in temperate areas and that competition between corals
and seaweeds is more intense. The combination of cooler
water temperatures and more intense competition with
abundant seaweeds likely interact to limit the latitudinal
range of large coral reefs. However, when the physical and
ecological criteria are met, the results can be phenomenal.
For example, the most biologically diverse reefs occur in
the tropical Indo Pacific in the areas around Indonesia and
the Philippines and house over 550 species of coral and
thousands of species of fish. The Great Barrier Reef off
northeastern Australia is the largest reef in the world with
more than 2800 individual reefs occupying over 1800 km
of the Australian coastline and can be seen from outer
space.
Threats to Coral Reefs
Coral reefs are imperiled around the world because of the
compounding effects of multiple stressors such as over
fishing, pollution, climate change, and change in coastal
land use. The decline of reefs is particularly evident in the
Caribbean where coral cover has decreased by 80% in
recent decades and may drop further as reefs fail to
rebound following continued coral bleaching, overfishing,
disease outbreaks, and other disturbances. The causes of
coral reef decline are many and frequently act synergis
tically to drive coral reefs to alternate states such as
seaweed dominated reefs or sea urchin barrens
(Figure 9). We review the major threats to coral reefs
and the role that marine reserves can play in stemming
the tide of coral reef decline.
Coral Bleaching
Coral bleaching occurs when corals degrade or expel their
dinoflagellate symbionts in response to environmental stres
sors such as elevated sea surface temperature and increased
UV radiation. Although corals can reacquire symbionts and
recover in weeks to months, recovered corals may grow
slower and have reduced fecundity as compared to pre
viously unbleached corals, giving bleaching resistant corals
an ecological advantage after bleaching events. In severe
cases, bleaching may occur on the scale of hundreds to
thousands of kilometers and radically alter coral cover and
composition with coral mortality from bleaching events
approaching 100% in extreme cases. Branching corals such
as acroporid and pocilloporid corals are often more suscep
tible to bleaching and mortality than are massive corals,
allowing the slower growing massive corals to be more
persistent on reefs after bouts of strong bleaching.
Bleaching events not only decrease live coral cover but also
provide large areas for seaweed colonization, and these sea
weeds can prevent corals from reestablishing if herbivores
are not present in sufficient numbers to suppress seaweed
colonization and growth. Additionally, large scale bleaching
and mortality of branching corals can suppress fish popula
tions that are dependent on live coral for shelter and food.
Analyses of coral bleaching on Caribbean reefs over
the past two decades suggests that small increases in
regional sea surface temperature (0.1 C) result in large
increases in the geographic extent and intensity of coral
bleaching events. Given that climate change models sug
gest an increase in sea surface temperatures of 1–3 C
over the next 50–100 years, coral bleaching events may
become an intense, annual stress on coral reefs through
out the Caribbean and even the world. Although corals
may adapt and their bleaching thresholds may increase
over time as sea surface temperatures rise, the threat of
repeated, intense bleaching events over the next several
decades is a significant concern. If even the conservative
predictions of global climate models are realized, these
climate changes could result in the fundamental reorgani
zation of the ecology of coral reefs.
210
Coral Reefs
(A)
(C)
(a)
(b)
(c)
(d)
(e)
(f)
Slime f
Extra nutrients
Heterotrophic e
Macro algae c
‘Stressed’ b
or
turf d
‘Healthy’
reef a
Sea urchin
barren f
Rock
Fishing pressure
(B)
Overfishing of herbivores
Extra nutrient
‘Healthy’
state
Overfishing of echinoid
predators
Macro algae
state
Sea urchin
barren state
Rock
Figure 9 Alternate states in coral reef ecosystems. (A) A conceptual model showing human-induced transitions between alternate
ecosystem states based on empirical evidence of the effects from fishing and excess nutrients. The ‘stressed’ state illustrates loss of
resilience and increased vulnerability to phase-shifts. (B) A graphic model depicting transitions between ecosystem states. ‘Healthy’
resilient coral-dominated reefs become progressively more vulnerable owing to fishing pressure, pollution, disease, and coral
bleaching. The dotted lines illustrate the loss of resilience that becomes evident when reefs fail to recover from disturbance and slide
into less-desirable states. (C) Pictoral representation of the different reef states shown in (A). From Bellwood DR, Hughes TP, Folke C,
and Nystrom M (2004) Confronting the coral reef crisis. Nature 429: 827–833.
Disease and the Structure of Coral Reef
Communities
The impact of diseases on coral reefs has been realized
over only the past two decades. Two of the most exten
sive disease outbreaks have been on reefs in the
Caribbean and have fundamentally changed the ecology
of Caribbean reefs. In 1983–84 an unknown pathogen
swept through the Caribbean and killed approximately
99% of the then abundant sea urchin Diadema antillarum.
In many areas of the Caribbean, D. antillarum had been
the dominant herbivore keeping reefs free of most fleshy
seaweeds and facilitating recruitment and growth by
corals. After the mass mortality, levels of herbivory
plummeted and standing crop of seaweeds dramatically
increased on many reefs. D. antillarum populations are
recovering in some areas of the Caribbean, and in these
‘urchin zones’, seaweeds cover 0–20% of the reef as
opposed to 30–79% of the reef. Juveniles corals are ten
times more abundant in some urchin zones. The poten
tial recovery of this critical herbivore gives hope to
Caribbean reefs many of which are still enveloped in a
blanket of seaweed.
The other outbreak that altered the structure of
Caribbean reefs was the epidemic of white band disease
among acroporid corals in the mid to late 1980s. This
disease attacked two of the major reef building corals in
the Caribbean Acorpora palmata and A. cervicornis. These
two corals were once so abundant on Caribbean reefs
that early coral reef ecologists named characteristic
zones on reefs for these dominant corals (i.e., the ‘pal
mata zone’ and the ‘cervicornis zone’). These corals that
had dominated Caribbean reefs for at least a half million
years are now rare to absent on most reefs and have
declined so dramatically that they are both being listed
as threatened species under the Endangered Species Act
in the United States. If the prevalence and severity of
coral diseases is linked to pollution and climate change
as has been demonstrated for some studies, then a con
tinued increase in the effects of diseases on the ecology
of reefs can be expected.
Coral Reefs
Shifting Baselines, Overfishing, and the Altered
Food Webs of Coral Reefs
In many regions of the world, coral reefs are mere rem
nants of what they were only a few decades ago. These
changes to reefs are not adequately appreciated due to the
problem of the ‘shifting baseline syndrome’ – reefs that are
deemed ‘normal’ today are not what was ‘normal’ only a
few decades ago, much less a century or more ago. Each
new generation of divers or marine ecologists suffers from
reduced expectations of what a healthy coral reef should
be. For example, as a graduate student and post doc, one of
us (M.E. Hay) dove on Caribbean reefs dominated by
luxuriant stands of elkhorn and staghorn coral (Acropora
palmata and A. cerviconis) the size of football fields and saw
reefs abundant with grouper, large herbivorous fishes, and
Diadema urchins that formed ‘fields’ of gigantic black pin
cushions on regions of some reefs. In contrast, the younger
author here (D.E. Burkepile) has never seen a live stand of
elkhorn coral more than a few m2 and is lucky to see one
Diadema on most dives. However, both of us dive on reefs
that are vastly different from those that the first European
colonists would have experienced. Because of this problem
of shifting baselines, it is informative for ecologists to
explore the history and paleoecology of reefs in order to
deduce how reef communities have changed over hun
dreds, or thousands, of years.
Caribbean reefs were once dominated by sea turtles,
crocodiles, manatees, large predatory fishes such as
sharks and large groupers, and the now extinct monk
seal. Reefs with such a diversity of charismatic mega
fauna scarcely exist today anywhere in the world.
Centuries of overfishing have made many of these spe
cies ecologically extinct and altered the strong trophic
211
interactions that once dominated Caribbean food webs
(Figure 10). Including humans into the ecological equa
tion began a process of ‘fishing down the food web’
where large consumers such as sharks and manatees
were the primary targets of human harvesting. After
the larger animals were depleted, fisheries switched to
smaller predators such as groupers and then to herbi
vores such as parrotfishes. The changes in the
connections of these food webs have fundamentally
altered the dynamics of these ecosystems and have
resulted in cascading effects such as the decline of corals
and increase of seaweeds and sponges.
Although the largest megafauna are now largely gone
from Caribbean reefs, we have some idea of their histor
ical populations. For example, green turtles were once so
abundant that ships’ naturalists from the sixteenth to
seventeenth centuries remarked that they could navigate
to the Cayman Islands via the sounds of turtles swim
ming and that congregations of turtles seemed so thick as
to confound a ship’s path. One estimate puts the total
pre Columbian population of green turtles in the
Caribbean at greater than 30 million as opposed to the
tens of thousands today. Green turtles typically eat sea
grasses and seaweeds, but the top down force that this
historical population would have exerted on seagrass
production and biomass is unrivaled by any current
estimates of herbivory in seagrass beds. Because the
biota of coral reefs has changed so dramatically over
the past few hundred years, Jeremy Jackson writes that
scientists trying to understand the ecological processes
that structure coral reefs are ‘‘. . .trying to understand the
ecology of the Serengeti by studying the termites and
the locusts while ignoring the elephants and the
Before fishing
After fishing
People
Birds
Sharks
Monk
Crocodiles
seals
Pred
fish
Pred
inverts
Grazing
inverts
Grazing
fish
Macroalgae
Corals
Birds
Sea
cows
Seagrass
Sea
turtles
Sponges
Sharks
Pred
inverts
Grazing
inverts
Macroalgae
Grazing
fish
Monk
Crocodiles
seals
Pred
fish
Sea
cows
Corals Seagrass
Sea
turtles
Sponges
Figure 10 Simplified coastal food web for coral reefs showing changes in some important top-down interactions due to overfishing;
before (left side) and after (right side) fishing. Bold font represents abundant; normal font represents rare; ‘crossed-out’ represents
extinct. Thick arrows represent strong interactions; thin arrows represent weak interactions. Modified from Jackson JBC, Kirby MX,
Berger WH, et al. (2001) Historical overfishing and the recent collapse of coastal ecosystems. Science 293: 629–638.
212
Coral Reefs
wildebeest.’’ Basically, the biotic forces that impact coral
reefs today are mere shadows of what they once were,
and humans have radically changed the ecological and
evolutionary trajectories that have influenced coral reef
ecosystems for millennia.
Protection and Resurrection of Coral Reefs
One of the saving graces of coral reefs over the next few
decades may be the creation and enforcement of marine
reserves that protect reefs from overfishing. Overfishing is
one of the most devastating threats to reefs, as fishers
preferentially remove the large bodied fishes that are
the strongest interactors in these ecosystems, resulting in
fundamental changes to the food webs of reefs. The estab
lishment of marine reserves limits or prevents the
harvesting of fishes and invertebrates from areas of reef
and theoretically allows populations of overharvested
species to rebound, reestablishing viable populations of
fishes and crucial ecosystem processes on reefs. Recent
studies of the efficacy of marine reserves show that redu
cing fishing pressures on reefs allows increases in the
(a)
1400
Predator biomass
Parrotfish biomass
Predator biomass (g 200 m–2)
5000
1200
1000
4000
800
3000
600
2000
400
1000
0
Parrotfish biomass (g 120 m–2)
6000
200
South
Park
North
All other
0
(b)
1.4
100
90
1.2
1.0
70
60
0.8
50
0.6
40
30
0.4
Mean macroalgal cover (%)
Mean grazing intensity (% hr–1)
80
20
0.2
0
10
Non–park
Park
0
Figure 11 (a) Patterns of parrotfish biomass and their predators (SE) within the Exuma Cays (Bahamas) and for all other survey areas
combined. ‘Park’ denotes the Exuma Cays Land and Sea Park which is 456 km2 and was established in 1959. ‘South’ and ‘North’
represent reef systems that are near the southern and northern borders of the park. (b) Mean macroalgal cover (gray bars) (SE) and
grazing intensity of parrotfishes (black bars) inside and outside the Exuma Cays Land and Sea Park. Reserve impacts are significant
(P < 0.01) for each variable. From Mumby PJ, Dahlgren CP, Harborne AR, et al. (2006) Fishing, trophic cascades, and the process of
grazing on coral reefs. Science 311: 98–101.
Coral Reefs
density, biomass, individual size, and diversity of fishes
and invertebrates inside the reserves and that these effects
occur rapidly and are longlasting. In addition, these
reserves not only allow increases in fish density and
biomass within the protected areas but also result in the
‘spillover’ of fishes as they migrate from the reserves into
unprotected areas. Thus, marine reserves may subsidize
fish populations on reefs that are not directly
protected from fishing, although the extent to which this
spillover effect will actually affect unmanaged reefs is
equivocal.
Marine reserves can also restore trophic linkages
that enhance the recovery of coral reefs. For some
reefs in the Bahamas, long term protection from fish
ing (i.e., roughly 50 years of enforcement) has led to
increases in the abundance of medium sized predatory
fishes such as the Nassau grouper (Epinephelus striatus)
(Figure 11). Increases in grouper abundance resulted
in increased predation rates on small herbivorous par
rotfishes, which would seemingly decrease the rate of
herbivory on reefs. However, the protection from fish
ing also allowed large parrotfishes to recover and
actually increased the overall rate of herbivory in
the reserve despite increased predation on smaller
herbivores (Figure 11). These increased rates of her
bivory decreased macroalgal abundance and may
increase coral abundance and cover over time if this
balance between predation and herbivory can be
maintained. Although the benefits of reserves to con
servation and fisheries are promising, one of the main
challenges to the success of marine reserves is the
enforcement of no harvesting policies once the reserve
is established. In many areas, reserves are ‘paper parks’
or parks in name only as there is insufficient money
or political will to achieve the enforcement necessary
for the reserves to succeed. However, if marine
reserves can be implemented and enforced they will
be one of the best tools that conservation science
currently has to protect, and hopefully resurrect,
many coral reefs.
Summary
Fossil evidence shows that corals have dominated many
reefs for over 10 000 years. However, the balance
between the ecological forces of predation, competition,
213
disturbance, and recruitment that allowed thousands of
years of uninterrupted reef growth have now been
grossly altered by human activities. Consequently,
healthy and dynamic reefs have declined dramatically
in the last two decades as a result of overfishing,
climate change, pollution, and other anthropogenic
insults. Although the ecological future of reefs seems
bleak, we hope that creative management of these
ecosystems has the potential to protect them for future
generations.
Further Reading
Bellwood DR, Hughes TP, Folke C, and Nystrom M (2004) Confronting
the coral reef crisis. Nature 429: 827 833.
Birkeland C (ed.) (1997) Life and Death of Coral Reefs. New York:
Chapman and Hall.
Burkepile DE and Hay ME (2006) Herbivore versus nutrient control of
marine primary producers: Context dependent effects. Ecology
87: 3128 3139.
Cowen RK, Paris CB, and Srinivasan A (2006) Scaling and connectivity
in marine populations. Science 311: 522 527.
Dulvy NK, Freckleton RP, and Polunin NVC (2004) Coral reef cascade
and the indirect effects of predator removal by exploitation. Ecology
Letters 7: 410 416.
Gardner TA, Cote IM, Gill JA, Grant A, and Watkinson AR (2003) Long
term region wide declines in Caribbean corals. Science
301: 958 960.
Halpern BS (2003) The impact of marine reserves: Do reserves work
and does reserve size matter? Ecological Applications
13: S117 S137.
Hay ME (1997) The ecology and evolution of seaweed herbivore
interactions on coral reefs. Coral Reefs 16: S67 S76.
Hughes TP (1994) Catastrophes, phase shifts, and large scale
degradation of a Caribbean coral reef. Science 265: 1547 1551.
Hughes TP, Baird AH, Dinsdale EA, et al. (2000) Supply side ecology
works both ways: the link between benthic adults, fecundity, and
larval recruits. Ecology 81: 2241 2249.
Jackson JBC, Kirby MX, Berger WH, et al. (2001) Historical overfishing
and the recent collapse of coastal ecosystems. Science
293: 629 638.
Knowlton N and Rohwer F (2003) Multispecies microbial mutualisms
on coral reefs: The host as a habitat. American Naturalist
162: S51 S62.
McClanahan TR and Mangi S (2000) Spillover of exploitable fishes from
a marine park and its effect on the adjacent fishery. Ecological
Applications 10: 1792 1805.
McCook LJ, Jompa J, and Diaz Pulido G (2001) Competition between
corals and algae on coral reefs: A review of evidence and
mechanisms. Coral Reefs 19: 400 417.
Mumby PJ, Dahlgren CP, Harborne AR, et al. (2006) Fishing, trophic
cascades, and the process of grazing on coral reefs. Science
311: 98 101.
Mumby PJ, Edwards AJ, Arias Gonzalez JE, et al. (2004) Mangroves
enhance the biomass of coral reef fish communities in the Caribbean.
Nature 427: 533 536.
Veron JEN (2000) Corals of the World, vols.1 3. Townsville, QLD:
Australian Institute of Marine Science.
214
Desert Streams
Desert Streams
T K Harms, R A Sponseller, and N B Grimm, Arizona State University, Tempe, AZ, USA
ª 2008 Elsevier B.V. All rights reserved.
Distribution and Physical Template
Temporal Dynamics
Biota
Energetics
Nutrient Dynamics
Human Modifications
Further Reading
Distribution and Physical Template
streams via overland flow during storms or via infiltration
of permeable low order channel sediments followed by
subsurface flow. Resulting flash floods scour the
streambed, resulting in downstream export of sediments
and aquatic organisms and creating a wide channel. Large
floods also deposit alluvial materials in riparian zones and
may remove riparian vegetation. These effects vary
depending on the scale of the event (see the section titled
‘Temporal dynamics’).
The boundaries of a stream ecosystem in any climate
region extend beyond the wetted channel and comprise a
stream–riparian corridor (Figure 1). The aquatic ecosys
tem encompasses surface water as well as the alluvial
sediments beneath the streambed where surface and
groundwater mix, termed the hyporheic zone. The para
fluvial zone is defined by the region of the active channel
over which water flows only during floods and in desert
streams this region can be much wider than the stream
itself. Finally, the riparian zone is the land area surround
ing the stream that is significantly influenced by the
stream. The availability of water contrasts starkly among
these subsystems in desert streams making each subsys
tem more distinct than in mesic streams. In deserts, the
Desert streams occupy arid and semiarid regions defined
by low annual precipitation. Semiarid and arid climate
zones are found on all continents and include both hot
and cold deserts. Although the range of temperatures
varies across desert regions, summer temperatures may
exceed 40 C in hot deserts. Annual precipitation ranges
from <100 to 300 mm yr 1 and combined with high tem
perature can result in high rates of evapotranspiration.
Higher precipitation in the mountains (up to 1000 mm
yr 1) can feed streamflow in the low deserts, often sup
porting perennial flows in large basins. Arid and semiarid
regions are characterized by distinct seasons defined by
precipitation and/or snowmelt and the amount of preci
pitation that falls during these seasons shows high
interannual variability. This results in extreme seasonal
and interannual variation in stream discharge. Indeed,
streams in some desert regions flow in response to rain
events that occur only once in several years or even less.
Arid and semiarid lands account for over one third of
global lands, making desert streams prominent among aqua
tic ecosystems. The large geographic area covered by deserts
results in a wide variation in temperature and precipitation
regimes as well as in geomorphology. Thus, the hydrogeo
morphic templates and resulting ecological characteristics of
desert streams exhibit a great diversity of patterns. Despite
this extensive distribution of desert streams, the vast major
ity of ecological studies of desert streams have occurred in
the southwestern United States, Australia, and Antarctica.
Our discussion thus draws from results of studies in these
ecosystems. Future studies of desert stream ecosystems in
other regions are likely to add new dimensions to the state of
our understanding presented here.
Desert stream hydrographs are punctuated by events
when discharge may exceed baseflow by several orders of
magnitude. Precipitation falling on the catchment rapidly
reaches the stream and stream discharge rapidly dissipates
following floods. Infiltration of desert soils is minimal and,
at the scale of whole basins, much of the water that is not
returned to the atmosphere by evapotranspiration reaches
Surface
stream
Riparian
Channel
Parafluvial
Hyporheic
Figure 1 Schematic drawing of a desert stream–riparian
corridor.
Desert Streams
Temporal Dynamics
Desert streams are highly variable over time, at a range of
temporal scales. In addition to the seasonality that typifies
many streams, temporal dynamics of desert streams are
strongly influenced by disturbances at two extremes
(flash flooding and drying) of a hydrologic spectrum.
Researchers have largely focused on temporal dynamics
at decadal and lower scales; however, decade to century
scale channel change establishes a geomorphic template
upon which these higher frequency dynamics play out.
Our discussion will consider temporal change from low
to high frequency events (Figure 2).
The concept of disturbance has various meanings, but in
stream ecology disturbance is usually associated with
hydrologic extremes that change ecosystem structure and
processes. Using terms from disturbance ecology, we can
characterize a disturbance regime, which has features such
3
Temporal scale (log years)
hyporheic zone often contains water and sustains biolo
gical activity in the subsurface even in the absence of
surface flow. The parafluvial zone contains surface
water only during floods and flow quickly recharges the
subsurface through coarse sediments or gravel in this
zone, leading to short periods of surface flow but sus
tained subsurface flow. The riparian zone contrasts
starkly with desert uplands owing to the presence of
shallow groundwater that is accessible to plants.
Desert streams may contain sections of both gaining and
losing hydrologic templates. Gaining sections of streams
are those in which the water table is sloped toward the
stream channel such that groundwater discharges into the
surface stream. Losing reaches are characterized by a water
table that slopes away from the stream causing surface flow
to recharge groundwater. Along losing reaches, the
predominate direction of water flow in desert catchments
is from the uplands directly into the stream before rechar
ging riparian groundwater; whereas along gaining reaches,
water flowing overland into the riparian zone recharges
groundwater there before discharging into the surface
stream. These contrasting hydrologic templates can have
significant effects on nutrient dynamics, water storage, and
biota within stream–riparian corridors. Losing reaches, for
example, often have no surface flow during dry seasons,
whereas gaining reaches are a more permanent source of
surface water. Due to permeable sediments, interactions
between surface and subsurface water are dynamic. For
example, water moves into the hyporheic zone in regions
of downwelling and from the hyporheic zone to the surface
in regions of upwelling. Water within the hyporheic zone
moves through the interstitial spaces of sediments slowing
water velocity compared to the surface stream. Such pat
terns in hydrologic flows have important implications for
nutrient cycling and stream productivity.
215
Climate change
2
Climate variability
1
Disturbance
Succession
life cycle
0
–1
–2
Physiology
–4
–3
–2
–1
0
1
2
Particle Subreach
Section
Assemblage
Reach
Catchment
Spatial scale (log km)
Figure 2 Temporal scales considered in this section (ordinate)
are correlated with spatial scales (abscissa), and each
phenomenon discussed is associated with a characteristic range
of time and space scales.
as interannual (or interdecadal) variability, seasonality, and
timing, frequency, and magnitude of individual events.
Disturbance is intimately connected to succession, which
is most simply defined as the change in ecosystem proper
ties on a site following disturbance. Ecosystem components
that are affected by a disturbance are those that undergo
succession after the disturbance; for example, a flash flood
that removes algae and invertebrates but does not affect
streamside vegetation initiates succession in the stream but
not in the riparian zone.
Successional patterns depict the temporal changes in
stream and riparian communities and processes that are
superimposed upon a larger temporal scale of variability.
For longer lived riparian vegetation, successional patterns
and time frames may be similar to those of terrestrial
communities but for stream biota, succession often plays
out against a seasonal backdrop. Thus, successional pat
terns differ between seasons, with faster increases in
biomass during warmer months. Successional patterns
also vary depending upon the size and nature of the initi
ating disturbance as well as antecedent conditions (which
are themselves influenced by the disturbance regime –
timing and clustering of individual events). Whereas dis
turbances that occur over short timescales may produce
predictable recovery sequences, biota recover from longer
term, infrequent disturbances with less regularity. Effects
on biota of pronounced interannual variation that charac
terizes deserts include shifts in community composition of
invertebrates and differences in the relative importance of
nitrogen fixing cyanobacteria versus nonfixer algae in the
primary producer assemblages.
Flash flooding and drying are the primary disturbances
that characterize desert streams. Flood magnitude is
usually described by the peak discharge, but other aspects
216
Desert Streams
of a flash flood hydrograph – the steepness of its rise and
the length of its tail – also determine flood effects. Floods
are important geomorphic agents, shaping channel form,
as well as disturbances that initiate biotic succession. In
deserts, floods connect elements of the landscape, from
ridgetops to large rivers to groundwater, which are other
wise isolated and disconnected. Drying is at the opposite
hydrologic extreme but is more difficult to treat as a
discrete disturbance because it represents a protracted
reduction and ultimately loss of stream flow. As drying
progresses, there is first a concentration of mobile biota
that precedes a concentration of dissolved materials
(through evaporation); there may be isolation of sections
of a stream and distinctive patterns of surface water loss;
direction of surface–subsurface water exchange may flip;
organisms may move into sediments; and, ultimately,
surface flow is lost entirely. Drying ends when surface
flow resumes, either during a flood or as a gradual
increase in discharge.
At the scale of centuries, events that occur only every
50–100 years or so can shape channel form and initiate
riparian succession. For example, in the southwestern
USA, a period of erosion occurred forming arroyos or
gullies and draining the riverine wetlands that were once
characteristic of these desert environments. This period left
a geomorphic structure that persists today in many south
western river–riparian ecosystems. Dramatic changes such
as these can affect groundwater–surface water interactions
and change species composition of the riparian vegetation.
Indeed, such large scale changes have repercussions for
many stream characteristics, underscoring the importance
of the hydrogeomorphic template in establishing structure
and function of stream–riparian ecosystems.
Decadal variability resulting in relatively wet and dry
periods in the southwestern USA is related to quasi cycles
of the Pacific Decadal Oscillation and El Niño Southern
Oscillation (ENSO). For the southwestern USA, a strong
ENSO signal is seen in decadal patterns of winter runoff
from the Puerco and Grande Rivers in New Mexico and
Sycamore Creek in Arizona. During wetter periods, frequent
high discharge events remove active channel vegetation,
leaving open gravel bars (parafluvial zone). Although these
particular characteristics may be unique to desert streams of
the southwestern USA, the important point is that larger
scale forcing from global climatic patterns can result in
decadal shifts in near stream riparian vegetation that have
profound consequences for stream ecosystem function.
Given the high degree of interannual variability of
desert environments, annual averages often carry little
information and long term trends are masked. Years vary
not only in the total amount of runoff but also in the
temporal distribution and timing of individual events or
clusters of events. During the five wettest years of a
30 year period for Sycamore Creek, frequent floods
meant the ecosystem was in an early successional state
most of the year, whereas stream organisms experienced
severe drying conditions over much of the year during the
five driest years. Furthermore, years with the same total
annual runoff may differ substantially in seasonal
distribution of that runoff with consequences for the sea
sonal patterns of drying. A single, large late summer flood
in 1970 in Sycamore Creek carried the same total volume
of water as nine total floods distributed more evenly across
the spring season in 1988, with the result that much of the
stream was dry during the hottest months in 1970 but was
undergoing succession during summer 1988 (Figure 3).
Although seasonality in desert streams may be strongly
dependent upon the distribution of disturbance events,
other variables in addition to discharge, such as tempera
ture and day length, can influence the biota of desert
streams. Deserts are defined only by low precipitation;
thus, there is a broad range across the world’s deserts in
both flow seasonality and annual temperature distributions.
Flow seasonality may vary from highly unpredictable and
episodic events to a relatively predictable and sustained
increased in discharge associated with a distinct wet season,
with consequences for successional patterns. Temperatures
of stream water in cool deserts may fluctuate seasonally
from near freezing to 20 C; in hot deserts, daytime stream
temperature can reach >30 C but may be ameliorated by
extensive evaporative cooling.
Temperature variation over the course of a single 24 h
period may be nearly as great as seasonal variation. High
albedo and low heat capacity of desert land surfaces cause
extensive diel fluctuations in air temperatures, leading to
wide (though comparatively muted) ranges in stream tem
perature. Particularly in summer when evapotranspiration
rates are very high, streamflow varies measurably over
24 h, causing stranding at stream margins. At points
where drying streams sink into the sediments, the end of
the stream can migrate up and down the channel by
several meters! Desert streams of Antarctica show extreme
diel variation in discharge, but by a very different mecha
nism. Streamflow is generated by solar melting of the
vertical walls (ice cliffs) of glaciers. During summer,
when the Sun circles around the horizon at a low angle,
melting (and streamflow) stops when the cliffs are in
shadow.
Biota
Desert stream ecosystems support a diverse assemblage of
riparian plants and stream biota. A unifying characteristic
of desert stream organisms is the shared evolutionary
history in a hydrologically extreme environment. The
consequences of this extreme physical template are evi
dent from the variety of adaptations that allow species to
thrive in systems prone to flash flooding and prolonged
drought. Rather than providing a list of taxonomic names
Desert Streams
Mean daily discharge (m3 s–1)
1.1
9.4 8.1
2.9
217
0.5 57
0.4
0.3
0.2
0.1
Oct
Dec
Feb
Apr
Jun
Aug
1970 (solid)
1988 (dashed)
Figure 3 Contrasting seasonal discharge patterns in 2 years with nearly identical total annual discharge. Bars at bottom show time
periods likely to be influenced by postflood succession (hatched), drying (open), or neither (solid). Redrawn from Grimm NB (1993)
Implications of climate change for stream communities, pp. 293–314. In: Kareiva P, Kingsolver J, and Huey R (eds.) Biotic Interactions
and Global Change. Sunderland, MA: Sinauer Associates.
for each group, we place emphasis on life history, behav
ioral, and morphological adaptations for living in
hydrologically variable ecosystems.
Desert stream ecosystems house diverse periphyton
communities, which include a variety of filamentous
green algae, epilithic, epiphytic, and episammic diatoms
(attached to rocks, plants, and sediments, respectively),
and nitrogen fixing cyanobacteria. Both flash flooding
and prolonged drought decimate algal biomass in desert
streams. Rapid drying is particularly lethal, and algae
typically die within hours of exposure to the hot, dry
desert environment. Algal species often have physiolo
gical adaptations that allow for some resistance to
drying, however, and can withstand periods of gradual
drying. Such adaptations include the production of
extracellular mucilage that increases cellular water
retention, and intracellular osmoregulatory solutes that
also prevent water loss in drying sediments. In addition
to these mechanisms, at the onset of drying, algae may
also produce spores, cysts, or zygotes that can reactivate
upon rewetting. Benthic algae rapidly recolonize stream
sediments following floods, whereas recovery following
drought is variable and depends on the degree and
modes of drought resistance. In Antarctic desert streams,
for example, glacial melt is the primary source of
streamflow, and primary producers (cyanobacterial and
other microbial assemblages) are activated by higher
temperature and renewed flows which may occur sea
sonally or even on a diel basis. However, these
organisms are also able to persist for decades in the
absence of liquid water.
A productive and diverse invertebrate fauna charac
terizes many desert streams, consisting of insect and
crustacean taxa residing in both benthic and hyporheic
habitats. Life history characteristics of desert stream
invertebrates reflect an evolutionary history in a hydro
logically variable ecosystem, and are shaped by both
flooding and drying disturbances (Table 1). Most stream
invertebrate larvae have few mechanisms that confer
resistance to either type of hydrologic disturbance.
Instead, many species have short developmental times
(e.g., 1–3 weeks) that increase the probability of offspring
surviving to reproductive maturity in ephemeral environ
ments, and ensure that some aerial adults are available for
recolonization following floods or upon rewetting pre
viously dry channels. In addition, organisms with longer
life cycles exhibit an array of avoidance behaviors to
minimize the effects of flooding and drying disturbance.
These include timing reproductive activity to periods of
low flood probability, as well as ovipositing eggs in sec
tions of stream that are likely to retain water for longer
periods of time (e.g., deep pools, riffles). Finally, air
breathing insects (e.g., coleopterans, hemipterans) may
exhibit more direct avoidance behaviors, including
the use of rainfall as a cue for leaving aquatic habitats
before floods.
Relative to mesic counterparts, fish assemblages of arid
river systems are species poor, and are composed of taxa
that also have specific adaptations to life in hydrologically
variable systems. These adaptations include large repro
ductive efforts, multiple clutches per year, and short
developmental times. Such life history features, along
with the ability to migrate long distances during periods
of sufficient flow, allow native desert fish to rapidly
colonize habitats after disturbances, and result in dramatic
temporal fluctuations in population size. In addition,
218
Desert Streams
Table 1 Invertebrate colonization/recolonization characteristics of desert streams in relation to floods in different physiographic
regions
Colonization
sourcesa
Colonization
distancesb
Pathwaysc
Refugia
Species diversity
Resilienced
Flood occurrence
Spatial extent of
flood
Severity of flood
Mesic
Hot desert
Endorheic cold
desert
Exorheic cold
desert
Chapparal
Glacial
Numerous
Few
Intermediate
Intermediate
Intermediate
Few
Close
Far
Intermediate
Intermediate
Far
DD, um, S, O,
H
Abundant
High
DD, um, S, O
S
dd, um, S, o, h
Limited
Low
Intermediate
Intermediate
High
Spring/
summer
Extensive
Limited
Intermediate/
high
High
Winter/
summer
Variable
Low
Winter
Extensive
High
Winter/spring/
summer
Extensive
Intermediate/
far
DD, um, S, O,
h
Intermediate
Intermediate/
high
High
Winter
Intermediate
High
High
Intermediate
dd, um, d, O,
h
Limited
Low
Extensive
Unknown
Spring/
summer
Extensive
High
Intermediate
a
Refers to sources separate from the perturbed stream.
Refers to distance from other unaffected water bodies.
c
Status at time of spate: DD/dd, downstream drift; um, upstream migration; S/s, survivors; O/o, oviposition; H/h, hyporheic. Upper and lower case
indicates major or lesser importance, respectively.
d
Refers to number of taxa, not individuals, following recovery.
Reproduced from Cushing CE and Gaines WL (1989) Thoughts on recolonization of endorheic cold desert spring streams. Journal of the North
American Benthological Society 8: 277 287.
b
while intense flash floods can decimate fish populations,
many desert fish have morphological adaptations that
allow for some resistance to high flows. These include
depressed skulls, keeled or humped napes, buttressed
fins, narrow caudal peduncles, slim bodies, and reduced
scales – all of which act to reduce drag and improve
swimming ability in turbulent flow.
As desert stream ecosystems contract during drought,
fish become isolated in pools where the physical environ
ment can fluctuate dramatically. Although complete water
loss is lethal, as streams contract, individuals of many fish
species can survive in small pools, as well as beneath logs,
stones, and within beds of algae. As a consequence, native
desert fish are able to tolerate a broad range of tempera
tures (7–37 C); indeed, desert pupfish of western North
America can survive in temperatures that exceed 40 C.
Similarly, most desert fish are able to tolerate high salinity
and low dissolved oxygen concentration. Others still, like
the African lungfish, can burrow into the stream substrate
during dry periods and survive for months by breathing
atmospheric air with primitive lungs.
Streamside forests, or riparian zones, stand out as hot
spots for aboveground primary productivity in arid land
scapes. Arid riparian zones include assemblages of
phreatophytic deciduous trees capable of accessing
groundwater, as well as shrubs and annual grasses. The
overall taxonomic composition of riparian zones is typi
cally in striking contrast to that of the surrounding desert
landscape. Deeply rooted riparian trees are well suited to an
environment where the water table is temporally variable.
Obligate wetland species appear in desert riparian areas
with permanent access to shallow groundwater, whereas
those found in areas with strong seasonal fluctuations in
water table elevation have structures such as tap roots or
root architecture that maximizes water capture during
precipitation events. Many riparian tree species in arid
landscapes actually require over bank flooding at specific
times of the year to induce germination. Riparian vegeta
tion is thought to play an important role in the overall
cycling of nutrients in arid landscapes by taking up nutri
ents present in shallow groundwater and building organic
matter pools in riparian soils. Finally, riparian vegetation
serves as critical habitat for invertebrate, vertebrate, and
avian taxa within arid landscapes.
Energetics
In contrast with streams of temperate and tropical
biomes and because of their flood shaped channel mor
phology, desert streams generally are not shaded by
adjacent riparian vegetation. As a consequence, inci
dence of photosynthetically active radiation (PAR)
reaching desert streams is high, and rates of instream
primary production, the process by which energy is
captured and organic matter is produced in ecosystems,
are among the highest documented for benthic algae.
The accrual of algal biomass in turn represents the
energetic basis for stream food webs, and is central to
the overall ecosystem dynamics of arid streams. For
example, the abundance of high quality benthic algae,
together with warm temperatures and selection for rapid
growth, result in among the highest rates of secondary
production reported for benthic invertebrates.
Moreover, owing to high standing stocks and growth
rates, invertebrates play an important role in organic
matter dynamics and nutrient cycling in desert streams.
Indeed, the quantity of organic matter ingested by
stream invertebrates can be 2–6 times greater than pri
mary production. Finally, the emergence of desert
stream insects represents an important resource for pre
dators in adjacent terrestrial habitats.
At the ecosystem level, high rates of algal productivity
in desert streams set them apart from streams of forested
regions with respect to the relative rates of production (P)
and respiration (R). Specifically, desert streams are often
autotrophic (P > R). This is in striking contrast to streams
of other biomes that receive the bulk of organic matter
from outside the stream ecosystem and are often highly
heterotrophic (P << R). Productivity of desert streams is
also influenced by the disturbance regime. Flash floods
scour stream channels, decimate existing organisms, and
initiate a suite of algal and macroinvertebrate successional
processes that correspond to temporal changes in photo
synthesis and respiration (Figure 4). Post flood recovery
of heterotrophs is enhanced by availability of organic
matter that was stranded or deposited on the stream
margins and in the riparian zone during dry periods.
In addition to metabolic changes associated with flash
floods, spatial patterns of photosynthesis and respiration
and post flood successional dynamics are further influ
enced by hydrologic exchange between hyporheic and
parafluvial subsystems and the surface stream.
Specifically, rates of photosynthesis and the speed of
post flood recovery are greatest where nutrient rich
water from hyporheic and parafluvial sediments enters
the surface stream. Conversely, rates of respiration are
100
Bare sand
Diatoms
Cladophora
Bluegreens
Mat
Percent cover
80
60
40
20
10
20
30
40
Days after flooding
50
60
Figure 4 Recolonization of Sycamore Creek, AZ by primary
producers. Redrawn from Fisher SG, Gray LJ, Grimm NB, and
Busch DE (1982) Temporal succession in a desert stream ecosystem
following a flash flood. Ecological Monographs 52: 93–110.
Chlorophyll a (mg m–2)
Desert Streams
219
Upwelling
600
400
200
Downwelling
20
40
60
80
Days after flooding
100
Figure 5 Comparison of algal colonization in zones of
upwelling and downwelling following floods in Sycamore Creek,
AZ. Valet HM, Fisher SG, Grimm NP, Camill P (1994) Vertical
hydrologic exchange and ecological stability of a desert stream
ecosystem. Ecology 75(2): 548–560.
greatest where oxygen and organic matter from the sur
face stream enters subsurface and lateral sediments
(Figure 5). When both surface and hyporheic processes
are taken into account, desert streams may more closely
approximate a balanced metabolism (P ¼ R), which high
lights the connection between the two subsystems.
Nutrient Dynamics
Various factors can limit the rate of primary production if
demand (requirements of autotrophs) exceeds availability.
Limiting factors in streams are typically light and nutri
ents. Because desert streams often have open canopies and
receive abundant light, nutrients are the primary con
straint on algal growth. Lack of precipitation in arid and
semiarid regions leads to very slow rates of weathering of
parent materials, which can lead to phosphorus (P) lim
itation as rocks are the ultimate source of P in ecosystems.
However, in many well studied arid and semiarid water
sheds of the US southwest, volcanic derived parent
materials yield highly dissolved P. Primary production
in these desert streams is thus limited by nitrogen (N).
Nutrients enter streams via inputs from upstream,
from groundwater and overland flow, in plant materials
deposited from the riparian zone, and in the case of
N, via fixation of atmospheric N2 by cyanobacteria.
Unidirectional flow of water results in continual input
and output of nutrients in dissolved and particulate
forms, although inputs of limiting nutrients may be low
due to processing that occurred upstream. Nutrient spir
aling theory, a set of hypotheses that describe how
nutrients move between water column, subsurface, and
biotic compartments while being transported down
stream, predicts that nutrient uptake should be more
efficient under conditions of nutrient limitation. In
streams limited by N, for example, inorganic N is rapidly
removed from the water column by biota. For hot desert
streams, rates of nutrient uptake can be particularly rapid
due to high temperatures and light availability, which
increase rates of biological reactions. Concurrent with
rapid uptake by algae, excretion of inorganic N by inver
tebrate consumers can represent nearly 30% of the total
N delivered to the ecosystem. Over successional time,
export of rafting algal mats or stranding of algae on the
stream banks during dry periods results in loss of organic
N from the stream ecosystem. Budgets of organic N for
desert streams thus conform with the successional trajec
tory of terrestrial ecosystems wherein ecosystems at late
successional stages tend to lose nutrients. In contrast to
terrestrial ecosystems, however, net primary productivity
may continue to be positive during late successional
stages of desert streams, resulting in continued uptake of
inorganic nutrients by primary producers.
In surface water, nutrient cycling is dominated by
uptake of nutrients by algae and benthic biofilms. The
dominant pathway of nutrients in the surface is therefore
from inorganic to organic forms. Regeneration (minerali
zation) of inorganic nutrients in the subsurface may in
turn resupply dissolved inorganic nutrients. Processes
occurring in the hyporheic and parafluvial zones thus
contribute strongly to patterns of nutrient availability in
desert streams. Water flowing through sediments slows in
velocity allowing for greater interactions between sedi
ment surfaces and materials delivered in water. Microbes
inhabiting the interstitial areas of sediments transform
nutrients present in these downwelling zones, influencing
the spatial distribution of nutrients in the stream channel.
In coarse sediments where dissolved oxygen remains
relatively high as water moves through the hyporheic
zone, mineralization often dominates N transformations,
resulting in a localized increase in streamwater dissolved
inorganic N concentrations at locations of upwelling.
Increased nutrient availability in zones of upwelling is
often associated with hot spots of algal biomass. These
patterns are typical of alluvium dominated reaches where
algae are the predominant primary producers. In patches
where macrophytes colonize gravel bars and parafluvial
zones or in patches of fine sediment deposition, dissolved
oxygen concentration in the subsurface is decreased due
to root respiration and decomposition of plant derived
organic matter, and hyporheic flows are slowed, all of
which lead to hypoxic or anoxic conditions. Hot spots of
denitrification are associated with anoxic conditions in
the hyporheic zone and water upwelling downstream of
such patches is therefore depleted of inorganic N
(Figure 6).
Because of these differences in nutrient processing
between surface and subsurface flowpaths, streams that
undergo drying may exhibit marked spatial variability in
nutrient availability. Sections of the stream that dry may
continue to harbor subsurface flows and rapid
Nitrate concentration (mg N L–1)
Desert Streams
0.25
0.20
0.15
0.10
0.05
0.00
0.0
0.4
0.8
1.2
1.6
Distance (m)
2.0
2.4
0.20
Denitrification
(mg N gAFDM–1 h–1)
220
0.16
0.12
0.08
0.04
0.00
1
2
3
Site along flowpath
4
Figure 6 Changes in concentration of nitrate in water flowing
through plant-colonized gravel bars (top). Note precipitous drop
in concentration when water encounters a plant patch (denoted
by branch figure). Denitrification is a likely mechanism
accounting for the drop in concentration of nitrate; in situ rates of
denitrification increase in plant patches (bottom). Redrawn from
Schade JD, Fisher SG, Grimm NB, and Seddon JA (2001) The
influence of a riparian shrub on nitrogen cycling in a Sonoran
desert stream. Ecology 82: 3363–3376.
transformation of nutrients for some time after surface
flows are depleted. Nutrient inputs and outputs from dry
reaches may show strong contrasts in forms or concentra
tion of nutrients. In contrast, reaches characterized by
perennial flow tend to show dampened upstream–down
stream contrasts due to the homogenizing effects of
processes occurring in surface flows.
As with nearly all aquatic ecosystems, the surrounding
terrestrial landscape influences nutrient dynamics in
desert streams. In deserts, however, hydrologic connec
tivity between the stream and terrestrial portions of the
catchment, including the riparian zone, are variable in
time. Deposited nutrients and those stored by plants and
microbes may accumulate in the riparian zone and
uplands during dry periods. When precipitation or snow
melt events occur, water carries these particulate and
dissolved nutrients overland from the uplands to the
stream, and between the riparian subsurface and the sur
face stream. This creates pulses of nutrient transport
between desert streams and their watersheds. Pulsed
inputs of nutrients result in hot moments of nutrient
processing, short time periods with rapid rates of nutrient
Desert Streams
transformations. Hot moments may account for a signifi
cant fraction of annual nutrient processing within riparian
zones of desert streams.
Connectivity between terrestrial and aquatic portions
of desert stream–riparian corridors may also occur from
the stream to the riparian zone. Riparian plants can access
water and nutrients from the hyporheic zone as well as
from shallow groundwater. Access to these more perma
nent sources of water and nutrients leads to high
productivity in riparian zones relative to desert uplands.
Stream biota may transfer nutrients between streams and
riparian zones of desert streams. Owing to high rates of
primary productivity, insect emergence from desert
streams can result in significant exports of nutrients out
of the wetted stream. Emerging aquatic insects may thus
provide a significant source of nutrients to riparian food
webs.
Human Modifications
Desert streams present important challenges to under
standing and management of water resources. They
exemplify the resource that is most precious to humans
inhabiting arid and semiarid regions, yet they are threa
tened by increasing pressures of human exploitation,
agricultural expansion, and urbanization. Direct appro
priation of streamflow to support human activities is the
most serious threat to desert streams. This takes the form
of diversion, interbasin water transfer, and groundwater
withdrawal (which reduces baseflow); for example,
groundwater withdrawals over the past century have con
verted the Santa Cruz River in Tucson, Arizona from a
perennial to an ephemeral stream. In the Salt River of
central Arizona, river diversion into a system of canals
that feed agricultural and domestic/industrial demand in
Phoenix has left a dry riverbed throughout the metropo
litan region. Water extraction, primarily for irrigation, has
also resulted in salinization of streams in much of the
world and has triggered shifts in the composition of biotic
communities.
In their appropriation of water for a variety of uses,
people also modify the form and hence the function of
desert streams. For example, the creation of canals that
are straightened and lined with concrete replaces struc
turally complex streams with ecosystems that are unlikely
to support the ecosystem functions characteristic of
unmodified desert streams. Furthermore, impoundment
and flow regulation can have profound effects on riparian
ecosystems, for example, through the colonization and
persistence of exotic plant species that outcompete native
221
species under conditions of lowered water tables and
reduced flow variability.
See also: Ecosystem Ecology; Rivers and Streams:
Ecosystem Dynamics and Integrating Paradigms; Rivers
and Streams: Physical Setting and Adapted Biota.
Further Reading
Boulton AJ, Peterson CG, Grimm NB, and Fisher SG (1992) Stability of
an aquatic macroinvertebrate community in a multi year hydrologic
disturbance regime. Ecology 73: 2192 2207.
Cushing CE and Gaines WL (1989) Thoughts on recolonization of
endorheic cold desert spring streams. Journal of the North American
Benthological Society 8: 277 287.
Fisher SG, Gray LJ, Grimm NB, and Busch DE (1982) Temporal
succession in a desert stream ecosystem following a flash flood.
Ecological Monographs 52: 93 110.
Fisher SG, Grimm NB, Marti E, Holmes RM, and Jones JB (1998)
Material spiraling in stream corridors: A telescoping ecosystem
model. Ecosystems 1: 19 34.
Fountain AG, Lyons WB, Burkins MB, et al. (1999) Physical controls on
the Taylor Valley Ecosystem, Antarctica. Bioscience 49: 961 971.
Grimm NB (1993) Implications of climate change for stream
communities. In: Kareiva P, Kingsolver J, and Huey R (eds.) Biotic
interactions and Global Change. Sunderland, MA: Sinauer
Associates pp. 293 314.
Grimm NB and Fisher SG (1989) Stability of periphyton and
macroinvertebrates to disturbance by flash floods in a desert stream.
Journal of the North American Benthological Society 8: 293 307.
Grimm NB, Arrowsmith RJ, Eisinger C, et al. (2004) Effects of
urbanization on nutrient biogeochemistry of aridland streams.
In: DeFries R, Asner G, and Houghton R (eds.) Ecosystem
Interactions with Land Use Change. Geophysical Monograph
Series 153, pp. 129 146. Washington, DC: American
Geophysical Union.
Hastings JR and Turner RM (1965) The Changing Mile: An
Ecological Study of Vegetation Change with Time in the Lower
Mile of an Arid and Semi Arid Region. Tucson: University of
Arizona Press.
Holmes RM, Jones JB, Fisher SG, and Grimm NB (1996) Denitrification
in a nitrogen limited stream ecosystem. Biogeochemistry
33: 125 146.
McKnight DM, Runkel RL, Tate CM, Duff JH, and Moorhead DL (2004)
Inorganic N and P dynamics of Antarctic glacial meltwater streams as
controlled by hyporheic exchange and benthic autotrophic
communities. Journal of The North American Benthological Society
23: 171 188.
Minckley WL and Melfe GK (1987) Differential selection by flooding
in stream fish communities of the arid American southwest.
In: Matthews WA and Heins DC (eds.) Ecology and Evolution of
North American Stream Fish Communities, pp. 93 104. Norman,
OK: University of Oklahoma Press.
Schade JD, Fisher SG, Grimm NB, and Seddon JA (2001) The influence
of a riparian shrub on nitrogen cycling in a Sonoran desert stream.
Ecology 82: 3363 3376.
Stanley EH, Fisher SG, and Grimm NB (1997) Ecosystem expansion and
contraction in streams. Bioscience 47: 427 435.
Stromberg J and Tellman B (eds.) (in press) Ecology and Conservation of
Desert Riparian Ecosystems: The San Pedro River Example. Tucson:
University of Arizona Press.
Valet HM, Fisher SG, Grimm NP, and Camill P (1994) Vertical hydrologic
exchange and ecological stability of a desert stream ecosystem.
Ecology 75(2): 548 560.
222
Deserts
Deserts
C Holzapfel, Rutgers University, Newark, NJ, USA
ª 2008 Elsevier B.V. All rights reserved.
System Ecology (Ecosystem and Communities)
Human Ecology
Further Reading
Geography
Biogeography and Biodiversity
Ecophysiology and Life Strategies
What makes the desert beautiful is that somewhere it
hides a well. Antoine de Saint Exupéry
Based on geographic location and a combination of
temperature and geographical causes of aridity, deserts
can be separated into five classes:
deserts. They are found in the hot dry lati
• Subtropical
tudes between 20 and 30 , both north and south.
Geography
Definition of Deserts
It is common belief that all deserts are hot and sandy
places. While this is generally not true, a common factor
of deserts is aridity, the temporal and/or spatial scarce
ness of water. True deserts can be delineated from other
biomes based on their aridity. Of the following groups,
only the first two are considered as true deserts here.
Aridity can be divided into four groups:
arid: less than 60–100 mm mean
• extreme
precipitation;
from 60–100 to 150–250 mm;
• arid:
semiarid:
from 150–250 to 250–500 mm; and
• nonarid (¼mesic):
above 500 mm.
•
annual
Since evaporation depends largely on temperature, bio
climatic aridity cannot be defined solely by the amount of
precipitation. Therefore, the higher limits given above
refer to areas with high evaporativity in the growing
season (e.g., in subtropical areas with rainfall in warm
seasons). This is taken into consideration in UNESCO’s
‘World Map of Arid Regions’ that defines bioclimatic
aridity by P/ET ratios (annual precipitation/mean
annual evapotranspiration). P/ET ratios smaller than
0.03 qualify for hyperarid zones (roughly corresponding
to the extreme arid zone above) and a ratio of 0.03–0.20 as
arid zone (thereby corresponding to the arid zone men
tioned above).
Another common way of delineating deserts is based
on their vegetation pattern and optional land use.
Extreme arid zones typically show contracted vegetation
restricted to favorable sites or lack vegetation altogether.
Arid zones are characterized by diffuse vegetation.
Semiarid zones mostly are characterized by continuous
vegetation cover (if edaphic conditions allow for it) and
only very locally dry land farming (without irrigation) is
possible. Farming without irrigation becomes a reliable
option at larger scale in nonarid zones only.
•
•
•
•
These deserts lie within the subtropical high pressure
belt where the descending part of the Hadley’s cell air
circulation causes general aridity.
Rain shadow deserts. They are found on the landward
side of coastal mountain ranges.
Coastal deserts. Found along coasts bordering very cold
ocean currents that typically wring moisture as preci
pitation from the air before it reaches the land, these
deserts are often characterized by fog.
Continental interior deserts. They are found deep within
continents and far from major water sources.
Polar deserts. They are found both in the northern and
southern cold dry polar regions.
This article focuses on extreme/hyperarid and arid zones
and on the first four of geographic desert classes listed above.
Where Are Deserts Found?
True deserts are found on all continents except the
European subcontinent (see Figure 1). Altogether about
20% of landmass can be classified as desert, making it the
largest biome on Earth. Table 1 gives an overview of the
largest deserts. In addition to these major deserts, many
smaller, separately named deserts exist; many of these can
be classified as local rain shadow deserts. All desert areas
of the world border on land semiarid zones. These are
either Mediterranean type climate and vegetation, or dry
temperate or tropical grasslands/savannas. The vicinity
to these areas is important as many desert organisms are
either shared with these transitional biomes or evolved
from similar more mesic organisms.
Desert Landforms
According to relief type, two general groups of desert
landforms can be distinguished: (1) shield platform deserts
and (2) mountain and basin deserts. The shield platform
Deserts
Figure 1 Map of world distribution of deserts. Shown are the
arid and hyperarid desert region. Borders are somewhat tentative
as a clear separation from semidesert scrublands is often not
readily possible. Polar deserts are excluded.
deserts are most common in Africa, the Middle East,
India, and Australia and are characterized by tablelands
and basin lowlands. Mountain and hill slopes in this type of
deserts are restricted to ancient mountains or areas with
more recent volcanic activity. The geologically younger
mountain and basin deserts (also called mountain and
range) are predominant in the Americas and Asia and
consist typically of mountain ranges separated by broad
alluvially filled valleys. Within the two groups of desert
landforms, there are several dominant geomorphological
landscape types that are described here briefly.
223
Desert mountains consist chiefly of sheer rock out
crops and tend to rise abruptly from desert plains. The
slopes of these mountains differ according to geological
origin of the parent material. Igneous rock mountains
tend to be characterized by large debris (boulder fields),
while softer sedimentary rocks tend to lack these.
Desert mountains (Figure 2) dominate the desert of
the USA (38% of desert area), the Sahara (43%), and
Arabia (45%).
Piedmont bajada formations (Figure 3) cover roughly
a third of the arid Southwest USA but do less so in other
desert areas of the world. These formations are built up
from alluvial material that tends to accumulate in fans at
the mouth of mountain canyons. Individual alluvial fans
often coalesce and form large scale graded slopes called
piedmont bajadas (often only ‘bajadas’). Depending on
deposition age and location along the bajada, the fill
material is very diverse and differs strongly in alluvial
particle size and soil structure, thus creating complex
gradients and mosaics of distinct geological landforms.
These gradients have been studied extensively in the
Sonoran and Mojave Deserts and it had been shown that
predominantly, the age and consequent erosion of the
alluvial material within these mosaics determine the bio
logical communities that can establish on it.
Table 1 List of the major desert areas of the world (larger than 50 000 km2)a
Name
Size
(km2)
Type
Temperature
Countries
Sahara Desert
Kalahari Desert
Namib Desert
Arabian Desert
8 600 000
260 000
135 000
2 330 000
Subtropical
Subtropical
Coastal
Subtropical
Hot
Hot
Hot
Hot
Syrian Desert
Kavir Desert
Thar Desert
Gobi Desert
Taklamakan
Karakum Desert
Kyzyl Kum
Great Victoria Desert
Great Sandy Desert
Gibson Desert
Simpson Desert
Great Basin Desert
Chihuahuan Desert
Sonoran Desert
Mojave Desert
260 000
260 000
200 000
1 300 000
270 000
350 000
300 000
647 000
400 000
155 000
145 000
492 000
450 000
310 000
65 000
Subtropical
Subtropical
Subtropical
Continental
Continental
Continental
Continental
Subtropical
Subtropical
Subtropical
Subtropical
Continental
Subtropical
Subtropical
Subtropical/rain
shadow
Coastal
Rain shadow
Hot
Hot
Hot
Cold
Cold
Cold
Cold
Hot
Hot
Hot
Hot
Cold
Hot
Hot
Hot (cold)
Egypt, Libya, Chad, Mauritania, Morocco, Algeria Tunisia.
Botswana, Namibia, South Africa
Namibia
Saudi Arabia, Jordan, Iraq, Kuwait, Qatar, United Arab
Emirates, Oman, Yemen, Israel
Syria, Jordan, Iraq
Iran
India, Pakistan
Mongolia, China
China
Turkmenistan
Kazakhstan, Uzbekistan
Australia
Australia
Australia
Australia
United States
Mexico, United States
United States, Mexico
United States
Hot
Cold
Chile, Peru
Argentina
Polar
Very cold
Antarctica
Atacama Desert
Patagonian and
Monte Deserts
Antarctic Desert
a
Various sources.
140 000
673 000
1 400 000
224
Deserts
Figure 2 Desert mountains: the Cambrian sandstone
formations rise almost vertically from the valley floor filled deeply
by sands that locally eroded from the mountain fronts. Wadi
Ram, Jordan, October 2003. Photograph by C. Holzapfel.
Figure 3 A piedmont bajada in the Mojave Desert: alluvial fan
deposits stemming from a nearby mountain range vary in age
and structure (here a mixture of Pleistocene and Holocene
deposits). The position along the fan and the composition and
structure of the deposits determine hydrology and plant growth
(here the common desert shrubs Larrea tridentata and Ambrosia
dumosa). Fremont Valley, California, USA, March 2006.
Photograph by C. Holzapfel.
Desert flats (basins) are another common landscape
type (about 20% in the USA and 10–20% in other
regions). Often these flats have rather fine textured soils
and with sufficient rainfall vegetation is diffuse and rather
evenly spaced across the landscape (Figure 4). In more
arid regions and when rainfall redistribution is patchy due
to minor relief differences, distinct banded vegetation
patterns can arise. These bands exist mostly in Africa
and Australia but are also present in restricted areas in
the Middle East and North America. The open areas
produce the runoff of rainfall that accumulates in the
Figure 4 Desert flats: this large desert basin in the Atacama
Desert has very little surface dynamics and fine-textured soil
materials are overlain by rocks forming a partial pavement. Among
the harshest deserts on Earth, the Atacama receives very little to
no rainfall and plant growth is lacking in most years. South of
Antofagasta, Chile, October 1994. Photograph by C. Holzapfel.
Figure 5 Sand dunes: ergs are seas of sands that are
constantly on the move. Vast sand deserts are typical for the
Sahara (shown here) and Arabian Desert. Douz, Southern
Tunisia, March 1986. Photograph by C. Holzapfel.
bands, supporting the growth of vegetation. Another
type of flat desert region can be differentiated as hamma
das (bedrock fields). These bedrock fields develop in situ
and depending on the size of rock fragments, can build
dense pavements (regs) consisting of densely packed sur
face stones resting on finer textured subsoil. This desert
landscape type is common in the Sahara and the Middle
East and accounts for 40% of the area.
Sand dunes (Figure 5), known as ‘ergs’ in Arabic
speaking countries, are dominant desert landscapes
only in extreme arid desert areas (25% of Sahara and
Arabian Deserts, less than 1% in arid Southwest USA).
Characterized by moving sands, they depend on sources
of sand, sufficient wind energy, and favorable
accumulation areas. Depending on these factors and pre
vailing wind direction, different dune types arise.
Deserts
Figure 6 Desert playas are often prehistoric lake beds with
fine-textured, alkaline soils. Depending on current rainfall and
temperature, playas can be flooded and then resemble the
former lakes, as this playa on the altiplano of the Andes at an
elevation of 4400 m. Plant life on playas is typically sparse but
when microorganisms and invertebrates are active, birds such as
these Andean flamingos (Phoenicopterus andinus) assemble in
large numbers. East of San Pedro de Atacama, Chile, November
1994. Photograph by C. Holzapfel.
Crescent shaped dunes (barkhan dunes) form perpendi
cular to the prevailing wind direction and tend to be
highly mobile. Linear dunes form in the direction of the
wind and therefore do not move over the desert land
scape. This distinction is of biotic importance as the edges
of dunes are favorable to plant growth, while the dune
crest and upper slopes, due to sand drift and fast erosion,
are usually devoid of vegetation. In sandy flats, individual
shrubs tend to accumulate sand deposits and eventually
form phytogenic hummocks (so called nebkhas).
Playas (Figure 6) are depressions with very fine
textured, often saline soils. Playas are the beds of former
lakes that can be flooded in years of abundant rainfalls.
These depressions are known under various names
(North Africa and Middle East: chotts, sebkas). Even
though individual playas can be large, worldwide they
cover only 1% of desert.
Badlands form in areas with clay rich soils and are
typically located at the margins of arid lands, although
they are found locally in arid regions as well. Depending
on the strength of water forced erosion, badlands are areas
with extremely high surface relief, typically forming fan
tastic ‘lunar landscapes.’
Dry river beds of ephemeral streams (Figure 7) are of
little importance with respect to land cover in deserts
(only 1–5% worldwide), but are of immense biological
importance. In extreme arid regions, these are the only
places with vascular plant growth, and almost all animals
depend at least at times during their life on the primary
production here. This biological importance of ephemeral
streams is therefore a foremost feature in deserts, and
even though small in size, these landmarks were always
225
Figure 7 Ephemeral stream: in arid areas vegetation
concentrates along and in the bed of temporal stream beds. Due
to available water in the subsoil that is plant extractable even long
after temporal surface flow ceased, most of the primary
production and species diversity in extreme deserts is restricted
to these habitats. Nahal Zin, Negev Desert, Israel, April 1987.
Photograph by C. Holzapfel.
distinctly named by human desert dwellers (washes in
North America, wadi/oued in Arabic speaking regions,
arroyo seco in Spanish speaking regions).
What Is Special about Desert Climates?
Life in desert is limited by the scarceness of water.
Secondary limiting factors are correlated to the main fac
tor: the dearth of nutrients for producers and food energy
for consumers and for both – at least temporally – high heat
stress. Precipitation typically is so low that water becomes
the controlling factor for biological processes. Precipitation
is also highly variable throughout the year and typically
occurs in infrequently defined events (discontinuous
input). To make things even worse, precipitation varies
randomly between years and is therefore not predictable.
The coefficient of variation between years in arid areas
is typically larger than 30% of the long term average
(and ranges in some extreme deserts to 70%). For compar
ison, temperate zones and tropical areas typically have
coefficient of variation of less than 20%. Individual pre
cipitation events in deserts can be tremendously large (for
instance, 394 mm in a single rainstorm in the Peruvian
desert that receives a long term annual precipitation aver
age of 4 mm) and due to surface runoff, large scale flash
flooding can occur (Figure 8). Even though such sudden
floods will replenish needed water to desert systems, ero
sion and direct damage to desert plants are the
consequence. Because of the small and temporally highly
variable rainfall amounts, deserts have been described by
Noy Meir as ‘‘water controlled ecosystems with infre
quent, discrete, and largely unpredictable water inputs.’’
Adding to and interacting with the pronounced tem
poral variation is the high spatial variation of rainfall in
226
Deserts
Figure 8 A flash flood obstructs traffic on a desert road.
Depressions and stream beds quickly flood after strong rainfall
events due to the high surface runoff in deserts. As in the case
here, the precipitation source of the water can be remote and
floodwater travels far distances. Due to high stream velocity and
carried erosion material, such sudden floods can be disruptive to
biotic communities and dangerous to humans. Sedom, Israel,
March 1991. Photograph by C. Holzapfel.
deserts. This variation is caused by: (1) oreographic fea
tures (e.g., increase with altitude), (2) differences in
degree and direction of slopes, and (3) the typically
small size of precipitation fronts (often less than1 km in
diameter).
Depending on the latitudinal geographical location and
the origin of rain fronts, deserts receive either precipita
tion in the cooler season (cyclonic/frontal rainstorms) or
in the warmer season (tropical, convective rainstorms).
Some transitional desert regions receive both. The season
ality of rainfall is of great bioclimatological importance as
evapotranspiration is larger during the warmer season and
rainfall therefore tends to have a smaller biological effect.
On the other hand, cold deserts receive precipitation dur
ing the cold season mostly in the form of snow and
biological activity is then limited both by low tempera
tures and aridity. Snowmelts in spring create deep
reaching wetting fronts that will hold water available for
plant uptake during the warmer growing season. Locally
important other water inputs are condensations of atmo
spheric moisture as dew. These are crucial for plant
production in the coastal fog deserts that otherwise do
not receive direct precipitations. It is less common inland
but can be noticeable in high desert areas as well (for
instance in the Negev Desert of Israel). Fog water inputs
are directly usable for cryptogamic organisms (e.g.,
lichens) and many arthropods. Foliar uptake of fog by
vascular plants has been demonstrated but its relative
importance in the water balance remains controversial.
Water vapor tends to move along temperature gradients
and can be important in dry soils with strong diurnal
radiation. An upward movement of water vapor at night
causes formation of dew close to the surface. Such water
might sustain germinated plants until they produce roots
long enough to reach deeper and wetter soil depths.
Deserts usually experience an extreme diurnal tem
perature range, with high daylight temperatures (up to
50 C, the highest temperature recorded in Death Valley
was 56.7 C), and extremely low nighttime temperatures
(often dropping below 0 C). This is caused by very dry
air that is transparent to infrared (heat) radiation from
both the sun and the ground. Thus during daylight all of
the sun’s heat reaches the ground. As soon as the sun sets,
the desert cools quickly by radiating its heat into space.
Clouds reflect ground radiation and desert skies are
usually cloudless, thereby increasing the release of heat
at night. With intense sun radiation, surface temperatures
can be extreme and depending on the color and type of
surface can exceed 80 C.
Desert Soils
The main features of desert soil that affect water and
nutrient availability include texture, content of organic
matter, pH, and orientation within the landscape. Desert
soils show typically little development from parent mate
rial and some authors even state that typical developed
soils do not exist in deserts. Most desert soils are classified
as Aridisols and are differentiated into soils with a clay
(argilic) horizon (Argids) and soils without such horizons
(Orthids). Other soils, less common in deserts, are molli
sols, soils with dark A horizons, and Vertisols, cracking
clay soils. Accumulated subsurface horizons with either
clays or calcium carbonate (calcic horizons) have clear
implication as impediments to water infiltration.
Most desert soils tend to be slightly to highly basic.
Such reactivity can negatively affect phosphorous and
micronutrient availability as these are generally not in
solution at pH > 7.0. Organic matter helps to increase
infiltration and via decomposition adds to nutrient avail
ability. It is often distributed unevenly in desert soils
(see below).
Soils in deserts have important effects on water inputs
as they act as short term water stores and modify water
availability by a number of regulation processes. These
regulation processes include direct infiltration and often
more importantly runoff and horizontal redistribution of
water. Redistribution by runoff tends to be of crucial
importance in deserts and contributes to spatially very
patchy distribution of water. Relatively impermeable sur
faces (e.g., biotic or physical crust in clay rich soils) create
runoff areas that result in catchments that are water rich.
Such water redistribution enables patchy plant produc
tion even in extreme arid zones, where plant growth
would not be possible since evenly distributed sparse
rainfalls would not exceed the threshold needed for
plant life. Because of sparse plant growth, soil created
Deserts
redistribution of water is more important than precipita
tion interception through plant surfaces. However, locally
such interception combined with stem flow can create
water rich spots under shrub or tree canopies. In contrast,
smaller precipitation events can be locally intercepted
and lost by evaporation. This is the reason that soils in
the understory of desert shrubs or trees can be either
wetter or dryer than the surrounding soil.
Soil texture is of large importance as it affects both
infiltration and the movement of wetting fronts. Fine
textured soils that are high in clay and silt fraction tend
to impede infiltration, in which wetting fronts move only
very slowly, and surface evaporation after rainfalls can be
very high. More coarse textured soil rich in sand frac
tions, as for instance sandy loams, is characterized by high
infiltration rates and rapid percolation. For this reason,
coarse textured soils are often better for plant growth.
As this is in contrast to soils in mesic areas where fine
textured soils are commonly considered to be superior for
plant production, this is called the ‘inverse texture effect’.
Clearly, the orientation and dynamics of soil surfaces
within the landscape plays a large role in arid ecosys
tems. Exposed southern (or northern, depending on the
hemisphere) slopes receive high solar radiation and
therefore due to higher evapotranspiration, tend to be
drier than opposite slopes (Figure 9). These inclination
differences are observable on large scale landscape level
or small scale microtopography level. An example is the
sun exposed sides of shrub hummocks that are often
only raised by a few centimeters, but can be bioclimati
cally and ecologically very different from the less
227
exposed side. Slope exposition also plays a role when
rainfall directions due to prevailing winds are constant.
Rain exposed slopes can receive up to 80% more water
than other slopes.
Biogeography and Biodiversity
General Diversity
The casual observer often assumes that deserts support
only low species richness and diversity because of the
harsh environmental conditions prevalent in arid areas,
but among plants and animals almost all taxa are repre
sented (even aquatic groups like fishes and amphibians)
here, and their species richness may be comparable to that
of more mesic environments. Even though detailed com
parative data are lacking, it has been argued that the
diversity in North American deserts is comparable to
some grasslands and even temperate forests. In general,
however, evidence based on correlations along climate
gradients indicates a decrease of species richness in plants
and animals with increasing aridity. Regardless of this,
specific taxa can be more species rich in deserts than in
bordering less arid systems and regionally show negative
relationship of richness with increasing precipitation.
Examples for these are reptiles and birds in North
America, and ants in Australia. Taxonomic groups that
are generally species rich in deserts are rodents, reptiles,
some insect groups (e.g., ants and termites), solpugids
(camel spiders), and scorpions. In the following, an over
view of typical desert taxa is given, and some emphasis is
given on the ecological role of these groups in deserts.
More specific treatment of ecophysiological adaptations
follows in the next section.
Ecological Role and Diversity of
Microorganisms
Figure 9 Marked phenological and plant composition
differences due to slope exposition (southeast facing slope on
the left and northwest facing slope on the right). The southeastfacing slope is subject to higher evaporational water losses and
receives less direct rainfall compared to the northwest-facing
slope. Such abiotic differences result in clear biotic contrast in
arid environments, making these systems ecological model
cases. Judean Desert, Palestine, December 1989. Photograph
by C. Holzapfel.
Even though obviously not readily observable, micro
organisms inhabit all desert areas and in the extreme
arid zones are often the only life forms present.
Relatively little is known about the diversity within
the lower three kingdoms (Fungi, Protista, Monera) in
general and even less is known about the species rich
ness of these groups in deserts. A recent survey that uses
‘DNA fingerprinting’, aiming at resolving bacterial ribo
somal DNA, indicated that soils of semiarid sites can
harbor higher bacterial richness then mesic sites. Since
factors other than water availability are more important
(chiefly soil pH) in determining microbial diversity,
it can be assumed that true desert can be quite rich
as well.
Mycorrhizal fungi seem to be quite important in desert
ecosystems, as in more mesic ecosystems. It appears that
228
Deserts
mycorrhizae of desert plants not only supply the plants
with nutrients but also supply moisture during the dry
season, at times taking the place of root hairs. Studies
conducted in the Chihuahuan Desert indicated that
most dominant, perennial species have high arbuscular
(AM) fungal infection rates in their coarse roots system,
while fine rooted annual species in comparison show
much lower infection rates and are also much less depen
dent on mycorrhizal associations in general. Worth
mentioning are mycorrhizal desert truffles (Terfezia
and Tirmania: Ascomycetes), that are host specific to
Helianthemum species in the arid region of the Middle
East and the Mediterranean zones of the Old World.
The desert of the American West supports an elusive
community of aboveground observable fungi in which
the Gasteromycetes (puffballs and allies) figure predomi
nately. Another common example is Podaxis psitillaris
(desert shaggy mane), a species most common in sandy
deserts.
Except for their crucial part in mycorrhizal associa
tions, desert microorganisms are noteworthy for their role
in three typical desert phenomena: desert crusts, desert
varnish, and interstitial communities. Desert crusts are
microbiotic communities composed of drought and
heat tolerant algae, cyanobacteria, fungi, lichen, and
mosses. These often species rich communities are held
together by sticky polysaccharide secretions and thus
form surface crusts. Desiccated crusts are often indiscern
ible until rainfall or dew moistens the surface and
microbial communities become active and green. Under
extreme conditions, such crusts can form below the sur
face. This is possible especially under the protection of
semitransparent calcareous or siliceous stones (quartz is a
good example) that enables transmission of light up to a
depth of 5 cm. The most common life form in crusts (and
in some areas also in hot deserts in general) is cyanobac
teria. Among their roles in the desert ecosystem are
atmospheric fixation of nitrogen and the binding of soil
particles. Together with mineral reducing bacteria, the
cyanobacteria are important in soil fertilization and soil
formation and thereby have clearly important effects on
vascular plants and dependent animal consumers. In hot
deserts, cyanobacterial crusts often form smooth surfaces,
while in cold deserts, where crust forming interacts with
frost heaving, a very rough surface is typical. These
different surface types clearly affect vascular plants
differently.
Even exposed desert rocks can support life. Clearly the
most visible organisms are crustose lichens. However,
when conditions become too extreme for growth of
lichens, bacteria can still survive on the surface of rocks.
Desert varnish, the dark and shiny surface found on sun
exposed, porous stones in hot deserts, is the result of
bacterial activity. These bacterial colonies obtain energy
from inorganic and organic substances and trap
submicroscopic, wind borne clay particles. These parti
cles accumulate in a thin layer and act as sun protection.
Over very long time periods, estimated at thousands of
years, these bacterial communities oxidize wind blown
manganese and iron particles and when baked together
with clay particles form the dark desert varnish. The color
of desert varnish varies depending on the relative propor
tion of oxidized manganese (dark black) to iron (reddish).
Environmental conditions even more extreme than
those that support surface bacterial growth can still
allow the formation of interstitial communities. These
communities consist mostly of algal species that inhabit
the matrix of sedimentary rocks in depth up to 4 mm.
These communities can stay dormant for long periods of
time and inhabit hot and cold deserts alike (they are
known to exist on exposed rocks in Antarctica).
Desert Flora
Even though the geological record indicates that arid
conditions existed for a long time (since the Devonian),
the current modern desert flora might have originated in
the Miocene, expanded in the Pliocene (after restrictions
during moist periods in the Cretaceous and Tertiary), and
reached its current distribution only during the Pleistocene.
Specifically, the deserts in the North American Southwest
are relatively young. Overall richness and uniqueness of
desert floras reflect size, age, and isolation of desert areas,
with larger deserts typically hosting larger numbers of
endemic species. Smaller desert regions and edges of
larger regions are often characterized by species that
evolved in adjacent more mesic areas and partially
adapted to arid conditions. A good illustration is the
high incidence of Mediterranean plants in desert areas
bordering regions with semiarid Mediterranean climates
in all parts of the world. In general, desert floras tend to
have high affinity to bordering semiarid climate zones,
such as Mediterranean climate type regions and semiarid
grassland. Taxonomical studies of many species groups
revealed that desert species have evolved (recently) from
nondesert species. Biogeographically, strong floristic links
exist between old deserts in North Africa, Middle East,
and Asia. Floristic similarities among desert regions
stretching from North Africa to Central Asia are particu
larly obvious since no wide barriers of ocean or humid
vegetation exist to restrict plant migration; these floristic
similarities are present despite strong climatic contrasts
ranging from hot environments in North Africa to the
much colder, arid Central Asia deserts. Apparent links
between the North American Great Basin and Central
Asian deserts might be explained by plant migration
across the Beringian land bridge. Clear affinities between
the deserts in both Americas can be explained by the
Panamanian land bridge. In this respect, the distribution
of Larrea shrubs is remarkable. The two recognized
Deserts
species – Larrea divaricata in South America and L. triden
tata in North America – are taxonomically and
phenotypically very close. It appears that the genus
Larrea evolved in South America and migrated only tens
of thousands of years ago (bird assisted?) to North
America where it quickly became the dominant shrub in
all warmer desert areas. Corresponding to the isolation of
the Australian continent, the flora of the Australian desert
is very different from all other deserts of the world.
Dominant plant life forms in deserts reflect water
stress conditions typical for deserts (for a treatment of
drought adaptations see the following section). While
trees are relatively rare and restricted to more mesic
microsites, a wide range of plant life forms can be found
that include many short lived and seasonal active plants
(e.g., annual or ephemeral plants and bulbous plants/
geophytes). The dominant life forms that visually shape
the plant formations are perennial woody plants (mostly
shrubs) and fleshy succulent plants (cacti and others).
Large succulent species can be dominant in some of the
hot desert regions (e.g., the saguaro cactus in the Sonoran
Desert). A few plant families are predominant in desert
areas. The aster family (Asteraceae) is the most diverse
plant family in deserts overall; it is especially numerous in
Australia, southern Africa, the Middle East, and North
America. Some deserts can be dominated by grass species
(Poaceae). Some plant families have their global center of
diversity in deserts and most likely evolved here. Notable
examples are the chenopods (Chenopodiaceae) that are
diverse in arid and semiarid regions of Australia, North
America, and from the Sahara to Central Asia. The New
World cacti (Cactaceae) are another example of a group
of species rich in deserts but relatively sparse in other
biomes.
Deserts are home to some of physiognomically extre
mely unusual plant types. Worth mentioning in this
respect are plant characters as the Joshua trees of the
Mojave Desert (Yucca brevifolia), the famous Welwitchia
of the Namib (Welwitchia mirabilis), and the boojum tree
(Figure 10) of the Sonoran Desert in Baja California
(Fouquieria columnaris). Exactly why and how deserts host
these exceptional plant types is not clearly understood
and such ‘Dr. Seussification’ of the desert flora deserves
systematic study.
Desert Fauna
The faunas of deserts are often biogeographically more
distinct between regions than the desert floras are. Despite
this, many similarities exist between the different desert
regions. Such phylogenetic similarities typical for the
African–Asian deserts are explained by the lack of dispersal
barriers, and similarities between North American and
Asian regions on one hand and North American and
South American regions on the other are likely the result
229
Figure 10 Boojum trees (Fouquieria columnaris) with
associated shrubs, agave, and cacti on a bajada in the Sonoran
Desert of Baja California. Cataviña region, Mexico, October
1997. Photograph by C. Holzapfel.
of existing land bridges. The Australia desert fauna, as its
desert flora, is very distinct. As mentioned earlier on, almost
all animal taxa are present in deserts, but some groups are
more diverse than others, with the major deciding factor for
this being the general aridity.
Relative to other insect groups, ants and termites are
very diverse in deserts. However, their species richness is
lower than it is in the Wet Tropics, where these groups
originated. These groups reach high population densities
and ecological importance is high. With up to 150 species
per hectare, the highest species richness for ants is found
in Australian deserts. Most desert arthropods are either
detrivores (termites, beetles, etc.) or granivores (mostly
ants), or are predators feeding on these (scorpions, spi
ders, etc.). Due to the lack of constant plant production,
herbivores are relatively sparse or show pronounced,
often dramatic temporal–spatial fluctuations (e.g., mass
flights of desert locusts). Species rich substone commu
nities consisting of protozoa, nematodes, mites, and other
microarthropods are typical for deserts, creating a micro
cosm where grazers and predators feed on bacteria, algae,
fungi, and detritus.
Fishes live in almost every aquatic habitat on the globe
and small, permanent desert water sources are no excep
tions. Obviously richness is extremely low, but species
often live in very restricted areas and often under extreme
conditions. The desert pupfishes (Cyprinodon sp.) in the
deserts of North America are among the most species rich
groups in deserts. Some species live at temperatures of
45 C and salt regimes 4 times that of seawater, while
some species are restricted to an area as small as 20 m2
(e.g., the Devil’s Hole pupfish in Nevada). These fishes
are opportunistic omnivores.
Likewise, desert amphibian communities are depaupe
rate since at least the juvenile stages depend on water.
230
Deserts
Only a small fraction of the world’s amphibians, mainly
anurans, are able to occupy deserts.
Reptiles are common and widespread in all deserts
and, with the exception of crocodilians and amphisbae
nians (worm lizards), all orders are represented in deserts.
Relatively few tortoises occur in deserts since they are
restricted due to their plant diet. Snakes and lizards are
well represented (especially in Australia). The extreme
high diversity of reptiles in Australian deserts has been
explained by low diversity of mammal and birds which
resulted in lower competition for food and lesser preda
tion pressure than in other desert regions. It appears that
reptiles as endothermic consumers enjoy an advantage
over other ectothermic consumers in the deserts of
Australia that are characterized by low quality plant
production.
Even though birds have basic adaptation to cope with
dry climates, diversity in deserts worldwide is relatively
low and a clear positive relationship between rainfall and
bird diversity is typical. Despite this, few desert specialist
species developed among the avifauna: sand grouse, lark,
parrots, etc.
Likewise, mammals are not very diverse in comparison
to other biomes, but some taxa evolved to be true desert
groups. Among smaller mammals are the heteromyds in
North America, the jirds and gerbils in the African–Asian
deserts, and the dayurid marsupials in Australia. Some of
the desert mammals are rather large and therefore have
advantageous low surface to volume ratios (see next sec
tion). The ‘flagships’ for this are clearly the camel species
(Camelidae) that originated in the Americas in the
Miocene and are now naturally found in desert regions
of the Old and New Worlds; they are clearly the largest
animals in all desert regions. It is of significance that most
large herbivorous mammals, including camels, donkeys,
goats, sheep, and horses, have been domesticated histori
cally in deserts and semiarid regions and are common as
domesticated livestock today. Other large, nondomesti
cated ungulates such as gazelles, ibexes, and oryxes are
generally extinct or at least rare and endangered.
Convergence of Desert Life Forms
Most desert plants and animals initially evolved from
ancestors in moister habitats, an evolution that occurred
mostly independently on each continent. Despite this
phylogenetic divergence, a high degree of similarity of
body shape and life form exists among the floras and
faunas of different desert regions. Since desert environ
ments are defined by their water limitation and have
similar landscapes worldwide, it is not surprising that
many organisms show convergent evolution and are mor
phologically and functionally alike. Similar pressures of
natural selection have resulted in similar life forms.
In fact, many of these analogous species groups became
textbook examples of evolutionary convergence:
and leaf succulence is found in nonrelated plant
• Stem
taxa: cacti in New World, milkweeds and Euphorbia
•
•
•
species in the Old World (however, this form is lacking
in Australia).
Bipedal locomotion is found in unrelated small rodent
groups: jerboa (family Dipodidae) in the Old World,
kangaroo rats (family Heteromyidae) in the New
World.
Bipedal locomotion is shared in a few larger mammals:
African springhare (genus Pedetes), desert living kan
garoos (Macropodidae) in Australia.
North American horned lizards (genus Phrynosoma) and
the Australian thorny devil, the unrelated agamid
lizard Moloch horridus, share similar grotesque spiny
body armors. This has been explained as an adaptive
suit that facilitates their need of having a large body
due to their specialization on ants. Ants as eusocial
insects present a clumped however low digestible
source of food (formic acid, chitin). Both lizard groups
are in need of a larger digestive system and therefore
large bodies that in turn makes them slow moving and
in need of protection.
Many adaptations that are discussed in the following
sections are typical for all desert regions of the world.
A combination of these traits creates the ‘typical’ desert
life form that to some extent is similar worldwide.
Ecophysiology and Life Strategies
Strategies for Coping with Drought
All life originated in the sea and all organisms that have
left their ancestral home depend on an ‘inner sea’, high
internal water content. This phylogenetic inheritage
restricts life in many habitats, and obviously deserts are
among the harshest in this respect. Even though deserts
are not only water limited (they are also low in nutrients
and energy resources), adaptations to cope with the spa
tiotemporal scarcity of water are predominant of most
(if not all) true desert organisms.
All desert life forms, animals, plants, and microorgan
isms alike, employ one or more of three basic strategies to
cope with the dearth of water: (1) drought evasion, a
strategy of avoiding water stress temporarily in inactive
states; (2) drought endurance, a suit of adaptations that
reduce actual stress and enable being active during
drought; and (3) drought resistance, a suit of adaptations
evolved to avoid water stress altogether. Note that water
and heat stresses are coupled, thus many of the adapta
tions mentioned below can be understood as strategies to
cope with both.
Deserts
1. Drought evading organisms ‘choose’ to pass exceed
ingly dry periods in dormant stages. Predominant
examples are short lived (ephemeral) plants that survive
the dry season or longer periods of drought in the dor
mant seed stage. Such annual plants are indeed very
common in many deserts of the world and compose a
large portion of the plant diversity in many areas (up to
80% of species richness). An equivalent for animals can be
found in cryptobiosis of invertebrate eggs and larvae.
Such aridopassivity can be found in fully developed
organisms as well; examples are bulbous geophytes and
desert animals that pass dry season belowground inactive
(estivation). Choosing of less arid microsites is another
way of avoiding drought. In animals, these are typically
behavioral space choices (e.g., permanent habitation or
temporary use of stress protected microsites: below
shrubs or stones, rock fissures, litter, below tree and
shrub canopies, or even soaring in high air). Likewise,
many plants are restricted to favorable microsites (e.g.,
under tree and shrub canopies, runon microsites, algae
growing under stones). Some organisms, mostly plants,
are able to lose water almost completely and ‘resurrect’
once water becomes available again (poikilohydry:
Selaginella species, algae, lichens, and moss species).
2. Drought endurance is a main strategy common
among the dominant desert organisms worldwide. A suit
of ecophysiological, morphological, and behavioral adap
tations work together to reduce the most detrimental
impacts of water stress.
Reducing water expenditure. Evergreen desert shrubs are
capable of fine tuned regulation of stomatal movement.
Specialized photosynthetic pathways evolved in desert
plants that minimize water loss and maximize carboxyla
tion. C4 and crassulacean acid metabolism (CAM)
pathways are adaptations to hot temperatures, compared
to the C3 pathway adapted to colder conditions. Animals
of arid regions are able to regulate and restrict water loss
by concentrating urine. Birds and reptiles excrete urinary
waste as uric acid that can be concentrated and allow
reabsorption of water in the urinary tract, a trait not
available to mammals. Desert mammals and most other
taxa excrete dry feces and reduce the urine flow rate.
Water loss through surfaces is reduced in plants through
an increase in thick lipid cuticulae, epidermal hair cover,
sunken stomata, small surface/volume ratio (leafless
plants with photosynthesizing stems – xenomorphic).
Animals employ a variety of adaptations that reduce
water loss: impermeable integuments (e.g., in arthropods),
changes of lipid structure in the epidermis that create
diffusion barriers to water vapor (some desert birds),
denser hair or feather cover, and small surface to volume
ratios (common in large mammals).
Prevention of overheating. High temperature stress is
closely connected to water stress as many of the ways of
coping with higher temperatures involve expenditure of
231
water, thereby exacerbating water stress. Examples are
transpiration cooling in plants and evaporative cooling
in animals (including humans; see below). Desert organ
isms typically have high heat tolerance and capability
to function at high temperatures. The comparatively
high temperature optima and temperature compensation
points of photosynthesis in plants and high body tempera
tures and high lethal temperatures in animals attest that.
Among the most thermotolerant species are desert
dwelling ants that forage on extremely hot surfaces.
A Saharan desert ant species (Cataglyphis bicolor) is noted
to hold the record with a critical thermal maximum of
55 1 C.
Apart from tolerating high temperatures, an array of
mechanisms evolved to decrease or dissipate heat loads
both in plant and animals. The formation of sheltering
boundary layers, employment of insulating structures,
and increase of reflection (white color, glossiness) are
among these mechanisms. Behavioral space and temporal
choices are a contribution to the prevention of overheat
ing. Seeking of sheltered microhabitats and nocturnal
activity of many (if not most) desert animals are obvious
examples. The nocturnal CO2 uptake in CAM plants is an
interesting analog to this.
3. Drought resisting organisms employ adaptations
that allow them to pass dry periods in an active state
without experiencing physiological water stress. The suc
culence of many typical desert plants worldwide is a form
of water storage that enables these plants to use water
during dry periods. Examples for taxa that are rich in
succulent species are the cacti (Cactaceae) and yuccas
(Agavaeae) in the New World and some members of
Euphorbiacea and Crassulaceae in the Old World.
Succulent plants typically cannot become dormant and
therefore require at least periodically predictable preci
pitation, a requirement that explains the general lack of
succulent plants in extreme arid environments where
prolonged droughts are common. Most succulent plants
have fairly shallow root systems that react very quickly
following larger rainfall events. An analog to plant succu
lence in animals can be found in desert snails that can
store large amount of water. The ability of desert mam
mals (notably the camel) to store large amount of water in
the blood is another analogous trait. The accumulation of
fat tissue that can be metabolically transformed into water
(see below) as a water storage mechanism is somewhat
controversial and is more universally understood as being
merely an energy source (e.g., fat reserves in desert reptile
tails, body of rodents, and the famous camel’s hump).
Water Uptake in Deserts
Animals
Vertebrates are able to obtain water from three sources:
(1) free water, (2) moisture contained in food, and
232
Deserts
(3) metabolic water formed during the process of cellular
respiration. Some are able to receive water from all three
sources, while others are able to exploit only one or two
methods. Highly mobile animals tend to be restricted to
the use of open water sources that are often sparse and
far between. Typical examples are desert birds that fly in
regular intervals to the few bodies of water available. To
mention are the desert adapted orders of sand grouse
(Pteroclidiformes) and some doves (Columbiformes) that
tend to visit standing water in large flocks at dawn and/or
dusk. The former are even known to transport water
soaked in their specialized belly feathers to their flightless
chicks. Many desert animals are able to use available
water opportunistically by drinking large quantities in
short time. This ability is proverbial in the camel that
can take up to 30% of its body weight in a few minutes.
Camels and other desert mammals have resistant blood
cells that can withstand osmotic imbalance. Animals liv
ing in more mesic environments (including humans)
would destroy their red blood cell at such high water
content in their blood. Much of the free available water
has high salinity, and so it is not a surprise that many
desert animals show high salt tolerance, for instance by
employing salt excreting glands. Other animals, mostly
the ones that are restricted in their mobility (e.g., mam
mals, reptiles, and insects), rely on water obtained from
their food. Carnivorous and insectivorous animals typi
cally receive enough water from their prey. Herbivores
do so as well, as long as the moisture content of the
consumed plant material is relatively high (>15% of
fresh weight: fresh shoots and leaves, fruits, and berries).
The ultimate desert adapted method however is the
extraction of metabolic water. Especially seed eating
(granivorous) animals are able to metabolically oxidize
fat, carbohydrate, or protein. Rodents and some groups
of desert birds (e.g., larks, Old World and New World
sparrows) are able to convert these energy sources into
water: 1 g of fat produces 1.1 g of water, 1 g of protein
produces 0.4 g of water, and 1 g of carbohydrates pro
duces 0.6 g of water. Schmidt Nielson has shown that
kangaroo rats (genus Dipodomys) are able to obtain 90%
of their water balance from metabolic water derived
from consumed seeds. The remaining 10% is obtained
from moisture stored in seeds. The use of already stored
body fat as source of water is controversial. It has been
argued that metabolizing fat and other storage sources
into water requires increased ventilation and therefore
increases water loss by transpiration from lung tissue. At
the most, no net gain of water will be the result.
According to this, the camel’s hump might function
simply as a fat energy storage facility, one that is situated
in one place in order to reduce isolation and allow
dissipation of heat.
In areas with high humidity, animals are able to
receive water from dew. Such direct uptake as the main
Figure 11 Desert sand rat (Psammomys obesus). As the
scientific name implies, this day-active desert rodent can store
large amounts of body fat as reserves during unproductive
seasons. Like other desert rodents, it obtains all of its needed
water through its plant diet. Negev Desert, Mitzpe Ramon, Israel,
May 2003. Photograph by C. Holzapfel.
source of water is probably restricted to arthropods and
some mollusks (snails). There is some evidence that
rodents can utilize condensation by water enrichment of
stored food (Figure 11).
Plants (and microorganisms)
Plants, with few exceptions, depend on water uptake by
their roots from the soil. Due to low soil matrix water
potentials and high salinity in arid regions, such soil water
is often not readily available. One way for desert plants to
overcome this restriction physiologically is to osmoregu
late the plant cell water potentials to overcome the low
potentials of desert soils, a mechanism that also aids them
in extracting water from saline solutions. Indeed, some
of the lowest water potentials have been measured in
desert shrubs ( 8 to 16 MPa (mesic plants rarely go
below 2 to 3 MPa)) and salt tolerant (halophytes)
desert perennials (as low as 9 MPa). In general, many
desert plants tend be deep rooted and are therefore able to
exploit water reserves that tend to be available in the
deeper soil layers. Due to the need of desert plants to
forage extensively for water, root to shoot ratio of desert
plants is typically high and rooting depths are larger than
in other ecosystems. In extreme cases, as in phreatopytes,
rooting depth can exceed 50 m. This was found for mes
quite trees (genus Prosopis) that are practically
independent from local precipitation and are able to
maintain very high transpiration rates for prolonged per
iods. In contrast and as mentioned before, many succulent
plants that store water in their tissues tend to be shallowly
rooted and are able to intercept even light summer rains
that do not cause a deeper recharge of soils and would
otherwise be lost to evaporation. Annual plants and most
grasses also benefit from being shallowly rooted. In
Deserts
general, many desert plants can react quickly to available
water by deploying fast growing ‘water roots’ from spe
cial dormant root meristems. Shallow rooting plants show
temporally intensive water exploitation patterns while
plants with deeper root systems are characterized by
spatially extensive water exploitation patterns.
Some deep rooted perennial plants exhibit hydraulic
redistribution from deeper soils to shallow soils. Water is
absorbed from the soil at greater depth during the day and
moves via the transpiration stream upward into shallower
roots and the aboveground parts of the plant. At night
when the air is more humid and plant stomata are closed,
plants become often fully hydrated and water may be
exuded from the root into the dry shallow soil. This
pattern, described as hydraulic lift, may have nutritional
benefits for the perennial plant itself, as it enables it to
utilize the nutrients from what would have otherwise
been dry soil. Released water – on the other hand –
might become available for competing plants. Hydraulic
lift has been described in almost all of the dominant
shrubs of the arid Western US (e.g., Artemisia tridentata,
Larrea tridentata, Ambrosia dumosa) and might be prevalent
all over the world’s arid zones.
Plants of saline habitats, halophytes, must be able to
acquire water with high salt concentrations. They need to
overcome the high osmotic pressure of saline solutions
and need to avoid the potential toxicity of some ions
(Naþ, Cl ). In order to achieve such a high salt tolerance,
halophytes employ strategies as osmoregulation, dilution
of inner cell salt concentration by succulence, and use
specialized salt excreting glands.
Special water rich habitats within deserts, for instance,
permanent stream sides and springs, attract extrazonal
plants that often possess only few aridity adaptations.
Found in these oases are wetland plants and some salt
tolerant tree species that can be characterized as ‘water
spenders’. Good examples are palm trees (the date palm
Phoenix dactylifera and the Californian palm Washingtonia
filifera) and salt cedars (Tamarix species).
Direct uptake of condensed atmospheric water (dew
and fog) and water vapor is generally possible only for
some specialized poikilohydrous vascular plants, but is of
much greater importance for microbiotic organisms such
as lichens and cyanobacteria.
233
can establish, reproduce, and eventually send their own
diaspores onto other favorable microsites. Such life cycles
are typical for short lived plants (annuals, ephemerals)
and some invertebrates. These diaspores typically remain
viable for long periods and can ‘sit and wait’ for years with
sufficient precipitation. Most annual desert plants form
such extensive seed banks. Seeds within such seed banks
tend not to germinate equally and even after strong pre
cipitation events, a fraction of the seed will remain
dormant. Such fractional dormancy might serve as avoid
ance of sibling competition as it will reduce densities, but
more importantly has been explained as a bet hedging
adaptation in order to cope with rainfall stochasticity.
When no supplemental rainfall follows an initial germi
nation triggered by a rainfall event, at least a fraction of
the seeds will be available in the following years, thereby
ensuring the long term survival of the population.
In addition to dormancy, many desert plants develop
some water sensing adaptation (so called ‘water clocks’)
that controls both dispersal and germination. Dry
inflorescences of the famous rose of Jericho (Anastatica
hierochuntica) and other annual plants (e.g., the New
World Chorizanthe rigida) open up only after abundant
rainfall and release only some of their seeds (Figure 12).
Many desert plants have morphologically different seeds
that differ in dispersal ability and have different germina
tion requirements (amphicarpic plants). In general, a high
proportion of desert plants suppress seed dispersal alto
gether (atelechory). This has been interpreted as an
adaptation to remain on the mother site, as it has already
been proved to be a favorable location.
Most perennial plants suppress flowering (aridopassive
shrubs) or sprouting altogether (e.g., geophytes) in drought
Strategies to Cope with Unpredictable Water
Resources
A wealth of adaptations arose in desert organisms that
allows them to utilize the pronounced spatiotemporal
stochasticity of water availability typical to deserts. As
detailed before, mobile organisms are able to use spatially
patchy water sources that are not available to less mobile
organisms. These sessile organisms often have dormant
dispersal units that can reach good microsites where they
Figure 12 Dry dead plant of the rose of Jericho (Anastatica
hierochuntica – mustard family Brassicaceae). Seed pods of this
annual plant are contained within curled branches forming a ball
that opens when moistened and seeds are released only after
rainfall events. Dead Sea region, Israel, March 1987. Photograph
by C. Holzapfel.
234
Deserts
years. This is analogous to many desert animals that shift
sexual maturity and mating to synchronize with favorable
conditions. Similar to plants, sterility is typical for extreme
drought years and dispersal and migration (nomadism) are
triggered by precipitation regimes. There is some indica
tion that insects and desert shrubs can shift their sex
expression with changing rainfall regimes. Especially,
monoecious shrubs, plants that have male and female
reproductive units on the same individual, can shift their
sex ratio with water availability. The male function
requires fewer resources from the plant (‘cheaper sex’),
and is typically the predominant sex in dry years.
Many desert shrubs tend to break apart into separate
shoot sections over time (axial disintegration). This so
called ‘clonal splitting’ is very common for desert shrubs
worldwide and has been explained as a risk spreading
adaptation. In time of severe drought, instead of the
death of the whole original individual, some segments of
the original shrub may survive. The consequence of this
growth strategy is often the formation of shrub rings that
grow outward and have a dieback zone in the center. Age
estimations have been made based on this growth form.
Large creosote bush (Larrea tridentata) rings in the Mojave
Desert, for instance, have been determined to be of an age
exceeding 11 000 years (e.g., the famous ‘King Clone’
located by Vasek in 1980).
roots, and stems – reserves). This pulse–reserve concep
tual desert model is clearly too simplistic; however, it
provides an important framework for the description of
major ecological components of deserts.
In contrast to this basic view of deserts, two major
alternative hypotheses have been developed in regard to
the driving factors defining communities and popula
tions in deserts. One hypothesis states that only the
primary producers are water limited and all other
trophic levels (consumers) are determined by the mag
nitude of this water dependent primary production.
Another hypothesis postulates that water shortage
affects organisms only individually and has no direct
effect on higher order species interactions. According
to this view, aridity effects on ecosystems and commu
nities are rather the indirect outcomes of direct
physiological and behavioral responses of individual
organisms (and their populations) to scarcity of water.
Despite the fact that the temporal and spatial lack of
water is clearly the driving force behind the individual
ecologies of desert species, current research makes it
clear that species interactions, including both negative
and positive ones, can be strong in deserts. The follow
ing sections strive to provide a brief summary of the
types of interactions typical to deserts.
Production
System Ecology (Ecosystem and
Communities)
The leading question in desert ecology is whether aridity
alone can explain all aspects of biological systems. If so,
desert environments could be understood simply by char
acterizing the harsh, abiotic environmental factors that
prevail in desert systems. Thus, desert systems do not
follow the typical ecosystem view and can be described
as simplified systems that react to discrete rain events
(triggers) by short term growth production (pulse), inter
spersed by long term storage of organic material (seeds,
Net annual primary production (NPP) is lower in deserts
than in most major biomes. However, when taking into
account that deserts typically are also characterized by
low amounts of permanent plant mass (standing phyto
mass), relative primary production (the ratio of NPP/
standing phytomass) is among the highest worldwide
(see Table 2). As rainfall fluctuates strongly within and
between years, it is no wonder that there is a tremendous
spatiotemporal variation in the amount of primary pro
duction. However, due to the lack of responsive
vegetation structure and typically low levels of soil ferti
lity, deserts are somewhat limited in their biological
Table 2 Phytomass and primary production of deserts in comparison to some other major biomes of the worlda
Plant formation
Tropical forests
Deciduous forest
Boreal forest
Savanna
Temperate
grassland
Tundra
Deserts
Phytomass of mature stands
(t ha 1)
Net annual primary production
(t ha 1 yr 1)
Relative primary
production
60–800
370–450
60–400
20–150
20–50
10–50
12–20
2–20
2–20
1.5–15
0.004–0.05
0.03–0.06
0.03–0.05
0.1–0.14
0.08–0.3
1–30
1–4.5
0.7–4
0.5–1.5
0.09–0.1
0.33–0.5
a
Modified from Evenari M, Schulze E D, Lange O, Kappen L, and Buschbom U (1976) Plant production in arid and semiarid areas. In: Lange OL,
Kappen L, and Schulze E D (eds.) Water and Plant Life Problems and Modern Approaches, pp. 439 451: Berlin: Springer; and other sources.
Deserts
potential to react to extremely wet years. Semiarid grass
lands, rich in very plastic perennial plant structures and
therefore exhibiting high potential growth rates, show
much larger fluctuations in response to changing water
availability (Figure 13). Also water use efficiency (NPP
divided by annual water loss) in deserts is lower than it is
in dry grasslands (0.1–0.3 g per 1000 g water in deserts
compared to up to 0.7 g in dry grasslands and 1.8 g in
forests).
During brief periods when water is available in excess,
the typically short supply of nitrogen (and other plant
macronutrients) is limiting. Even though nitrogen is limit
ing in almost all terrestrial ecosystems, deserts are
typically more limited due to four reasons: (1) plant
growth is triggered by available water faster than nutri
ents can be replenished by decomposition; (2) desert soils
typically have little nutrient holding capacities; (3) the
nutrient rich organic matter is located in the upper layers
of soils, a layer that is typically too dry for root growth
to occur, rendering the nutrients inaccessible; and (4)
detritus and other organic material is deposited and accu
mulated unevenly across the desert surface. Plant debris
235
typically accumulates passively under the canopy of
shrubs or is concentrated in nests of animals such as
harvester ants and termites. Thus the desert is an ‘infertile
sea’ with interspersed islands of fertility.
Resource–Consumer Relationships (Trophic
Interactions)
In contrast to some ecosystems, food chains in deserts can
be characterized by the importance of the link between
producers and consumers via decomposition. Less than in
most mesic environments, plant material is typically not
directly consumed alive; some estimation puts the
amount of energy that moves via decomposition into the
food web as above 90% of total primary production. Since
food resources are unpredictable, many animals can
opportunistically switch from one mode of consumption
to another (e.g., many arthropods are either herbivores or
decomposers).
Decomposition
Microbial decomposition is often limited by low water
availability, resulting in the accumulation of dry plant
material and seeds. For that reason, animal detrivores
are more important in deserts than in more mesic envir
onments. Examples are darkling beetles, termites, and
isopods. Termites are abundant in most of the warmer
deserts and are often the dominant decomposers of dead
plant material (above and belowground) that play an
extraordinarily important role in nutrient cycling. Since
most termites live belowground, they are also important
in the formation of soils. A similar phenomenon is dis
played by scavenging animals, which are comparatively
abundant among the desert fauna. Examples are large
mammals (hyenas, coyotes, and jackals) and many birds
(Old World and New World vultures, ravens, etc.). Like
smaller detrivores in the desert fauna, many of these
scavengers can switch to a predatory diet when needed.
Herbivory
Figure 13 Extreme, 30-fold differences of plant growth on a
rocky desert slope during (a) a dry (precipitation 40 mm, NPP
0.03 t ha–1 yr–1) and (b) an extremely wet year (193 mm, 0.87 t ha–1
yr–1). Northern Dead Sea, Palestine, March 1991 and March
1992. Photographs by C. Holzapfel.
Similar to other ecosystems, deserts host a large variety of
herbivorous animals that potentially utilize every part of
the plants. Some of the drought adaptations of plants,
discussed before, also function to deter herbivores.
Tough outer layers, spines, and elevated leaf chemicals,
all typical for desert plants, can therefore also be under
stood as mechanisms to protect low and therefore costly
primary production. Some plants appear to employ
growth forms that make them less conspicuous for herbi
vores. Remarkable examples are the living stones (Lithops
species, Aizoaceae) of South Africa that blend with the
surrounding rocky desert pavement.
236
Deserts
Predation
Abundant detrivorous arthropods are the most important
prey source in the desert and provide the base for a
relatively large assembly of smaller (e.g., spiders, scorpions)
and larger predatory animals (e.g., reptiles, birds). The
abundance of long term stored seeds and fruits in desert
systems supports an assembly of a diverse guild of grani
vores (seed predators). These granivores are recruited from
taxonomically much differentiated groups (e.g., ants, birds,
rodents), all of them potentially competing for similar food
sources. Carnivorous predators can be abundant as well.
These predators are mammals, birds, and reptiles (mostly
snakes). Because of the relative openness of the desert
terrain, prey organisms rely on a number of predator
avoidance strategies. Examples are general crypsis (camou
flage), ‘freezing behaviors’, and nocturnal activity pattern.
Active deterrents are spines (desert hedgehogs, horned
lizards), hard shells (desert tortoise), and poisons that can
be employed in active predation as well. Strong predator
pressure combined with the need for efficient predation in
a desert environment poor in prey might be the reason that
some of the most poisonous animals we know (e.g., snakes,
scorpions, Gila monster) are true desert animals.
Parasitism
Parasitic interactions are often very conspicuous in desert
environments. Many desert shrubs show abundant signs of
an attack by gall forming insects. For instance, Larrea
tridentata, the dominant shrub in all the hot deserts of
North America, is attacked by 16 specialized species of
gall forming insects. Parasitic plants, stem and root para
sites alike, are common in deserts worldwide. Though
detailed studies are lacking, these parasites seem to have
the potential of reducing host plant production and per
formance (Figure 14).
Nontrophic Species Interactions
Competition among and within species has been recog
nized as an important force that shaped the communities
in all mesic environments and the question whether this is
also true for deserts is a natural one, however one that has
not been answered univocally. Some researchers con
clude that biomass production and densities in desert
are typically below a threshold that would necessitate
competition for resources. Observing the same density
pattern, other researchers state that because such low
densities indicate strong resource limitation in desert,
strong competition should ensue.
Based on studies of spatial plant community structure, it
appears that current competition in deserts is rare; most
studies show clumped or neutral patterns – itself a sign of
the lack of competition – while only few studies show a clear
regular pattern (a sign of past competition). Experimental
removal of individual plants in the Mojave Desert, on the
Figure 14 Heavy infestation of a desert shrub (Ambrosia
dumosa) by an epiphytic parasite (Cuscuta sp.). Parasitic plants
can be common in deserts and their effects can add to the abiotic
stresses of aridity. Panamint Valley, California, USA, April 1995.
Photograph by C. Holzapfel.
other hand, demonstrated interspecific competition among
dominant desert shrubs. Spatial studies that assess the size
distributions in dependence of distance between desert
shrubs typically detect signs of negative association; larger
shrubs tend to be spaced farther from each other than smaller
ones. Removal experiments with granivorous rodents com
monly result in density increase of the remaining species,
thereby indicating current competition. The fact that char
acter displacement, the evolution of divergent body features
in coexisting species, has been demonstrated for desert
rodents is another sign that competition has been of impor
tance at least at one time.
Ecological theory predicts that negative interactions
(such as resource competition) decrease in importance
with increasing abiotic stress, and positive interactions
(such as facilitation) increase. Following this, it should
be possible to observe along a mesic to arid gradient a
waning of competitive interaction and an increase of
facilitative interactions. Indeed, a clear indication of this
has been observed in a survey of positive effects among
plants that resulted in a proportionally large number of
cases from arid regions. In many deserts of the world, one
can easily observe the positive association of either young
perennials with adult perennials or herbaceous plants
with larger perennial plants. Experimentally, it had been
shown that the perennials had net positive effect on the
smaller sheltered plants. Examples for these so called
‘nurse plant effects’ are the associations of young succu
lent plants (often cacti), trees, and shrubs and the
prevalent, close association of annual plants with desert
shrubs. Typically the larger nurse plant provides canopy
shading and increased soil fertility (see above discussion
on islands of fertility), and sometimes protection from
herbivorous animals to the sheltered plants. In accordance
with this prediction, shrub–annual associations tend to be
Deserts
Figure 15 Clear associations of annual plants with shrubs
(here Ambrosia dumosa) are common in deserts. Annual plants
benefit from nutrient enrichment and shade provided by the
shrub canopy and since they usually only provide little benefit to
the shrub (e.g., thatch-induced increase in water infiltration and
lower soil surface evaporation), they can compete with the shrub
for resources. Owens Valley, California, USA, March 1997.
Photograph by C. Holzapfel.
strongly positive in arid sites and less so (or even nega
tive) in less arid sites (Figure 15). As nothing ever in
nature is one sided, these unidirectional facilitative
effects are countered by negative effects as the nursed
plants can have negative, competitive effects on their
benefactor. Competition for water has been shown
between annuals and sheltering shrubs and such negative
effects are typical once sheltered young succulents out
grow the nurse plant.
Tradeoffs in competitive/facilitative interactions are also
found between taxonomically very distant groups. One
example is the complex nature of interaction between
microbial crusts and vascular plants. For one, these crusts
can have very contrasting effects on seed placement. Cold
deserts tend to have very rough crust surfaces that facilitate
seed deposition and establishment, while the smooth crusts
typical to hot deserts decrease such seed entrapment.
Because of these differences, no general effect of desert
crusts on the performance of vascular plants has been recog
nized. Nitrogen fixation by cyanobacteria increases nitrogen
availability, thereby favoring plant growth; however, the
creation of crusts can result in runoff and water redistribu
tion that in turn locally reduces plant performance.
237
many of the physical adaptations of true desert dwellers,
we humans might be a desert species after all. One of the
adaptations humans bring to live in the desert is a rather high
heat tolerance. The combination of upright position that
minimizes direct sun exposition during hottest times of the
day, the profusion of sweat glands all over the body, and the
lack of body hair, together with an energetically conservative
way of movements, contributes to our ability to cope with
hot deserts. As long as water and salt balances are maintained,
humans can perform relatively well under heat stress. This is
evidenced by the success of persistence hunting practices in
desert and semideserts, which involves tracking large ungu
late prey on foot during midday heat. Such persistence
hunting, today only employed by hunter gatherers in the
Kalahari Desert, has been the most successful mode of hunt
ing prior to the domestication of dogs, and uses the relative
heat balance advance that a well hydrated and trained
human can have over animal quadrupeds. Recent data
show that contemporary hunters run for 2–5 h over distances
of 15–35 km at temperatures of 39–42 C until prey items
(mostly antelopes) overheat and can be overcome.
Deserts have been important throughout human his
tory and the first civilizations arose in or close to deserts
(Mesopotamia and Egypt). Agriculture practices, often
involving irrigation, are sometimes interpreted as cultural
ways to deal with the stochasticity of the desert climate.
It is interesting to note that the first written law, the codex
written by the Babylonian King Hammurabi dating back
to 1750 BC, was designed to manage such crucial irriga
tion systems. It is basically the same set of laws that gave
rise to our modern laws. Since ancient history, deserts
have been the cradle of great civilizations on one hand
and the theater of fierce armed conflict on the other
(Figure 16). One wonders whether the nuclear weapons
Human Ecology
Origin and History
Humans have lived at the edge of desert and in the desert
proper for ever and there are some indications that modern
Homo sapiens evolved when the world climate turned to be
more arid at the end of the Pleistocene. Though lacking
Figure 16 Many ancient sites thrived near or in deserts.
The former Nabatean capital Petra is located in a desert valley
surrounded by steep mountains. From here the Nabateans, an
Arabic tribe, controlled the trade through the deserts of the Middle
East. Petra, Jordan, October 2003. Photograph by C. Holzapfel.
238
Deserts
tests that have been conducted in the deserts of New
Mexico and Nevada (among other desert sites worldwide)
symbolize that deserts can foster both the beginning and
the end of civilization.
Desert Economy
For humans, there are traditionally only three basic ways
to sustain themselves in deserts: hunting gathering, pas
toralism, and to some extent agriculture.
Ever since the rise of agriculture in the Neolithic era,
foraging as the exclusive mode of production (hunter
gatherers) became limited to areas that were marginal to
agriculture or animal husbandry. Naturally deserts are
among these zones. Examples of peoples who foraged as
hunter gatherers are the aborigines in Australian deserts
(this practice receded since the European discovery of the
continent), and the !Kung (bushmen) of the Kalahari, who
remain foragers in our times. Recent research on the
!Kung people showed that hunting gathering is a suitable
lifestyle that can sustain healthy populations that are even
able to spend sufficient leisure time, all this as long as
population densities are low. Some Amerindian people
employed hunting gathering in deserts as well. There are
some evidences that a later immigration wave of people,
the Nadene, linguistically distinct from the first Clovis
people, were culturally better adapted to harsh environ
ments and settled first in semiarid grasslands and
eventually in deserts (the Navajo and Apache might be
the descendents of the Nadene).
Pastoralism is a true desert activity that is also typical
for semiarid grasslands. It is obvious that many of the
livestock animals that were and are herded by pastoralists
originated from arid and semiarid areas and therefore are
well adapted to such environments. The ancestors of
horses, sheep, and goats evolved in semiarid environ
ments and donkeys and camels in arid environments.
People who live as pastoralists in deserts often combine
animal husbandry with some scale of horticulture; this
combination is called transhumance. In order to use the
stochastic desert environment optimally, many pastoral
ists have to follow rainfall events and are partly or truly
nomadic, as is exemplified by the traditional lifestyle of
the Bedouin of the Arabian Peninsula (Figure 17).
The use of agriculture most likely did not evolve in the
desert proper, but it has to be mentioned that the first
cultured plants, annual grasses, and legumes were domes
ticated near the edge of the desert in the Middle East
(10 000–8000 BC Natufian culture). Independently, in
likewise semiarid areas in Mexico (Tehuacan Valley,
before 7200 BC), the domestication of Teosinte into
corn (Zea mays) took place. Deserts harbored in historical
times small scale horticulture near springs and elaborately
designed irrigation systems that utilized the effects of run
off and water redistribution. Water harvesting systems in
Figure 17 The nomadic lifestyle is a cultural adaptation of
desert-dwelling people to the unpredictability of the desert
environment. As still seen here in the Sahara Desert, traditionally
camels were essential for transport between grazing areas and
arable oases. Douz, Southern Tunisia, March 1986. Photograph
by C. Holzapfel.
runoff farms have been found and partly recreated in the
Negev Desert (e.g., the Nabatean system in Avdat and
Shifta) and in the arid southwest of North America.
Large scale agricultural enterprises depend on permanent
water courses. As along the Nile in Egypt and along the
Tigris and Euphrates in Mesopotamia, these water sources
originated from areas far beyond the desert region.
Modern, often large scale irrigation projects are mostly
independent from surface water and use deeper aquifers.
In history, many large cities were established in desert
areas (Egypt, Middle East, South America) and there are
many cites in deserts in our times (Phoenix, Tucson, Las
Vegas). Incidentally, the climate and ecology of urban
areas even in the temperate, nonarid zones has many
similarities to true deserts (e.g., water limitation due to
surface sealing and runoff, high temperatures, etc.).
Human Impact on Deserts
As all ecosystems with low productivity, deserts are fra
gile to disturbance. Some ecologists go as far as to state
that no direct succession occurs at all after disturbance but
it is at least obvious that regeneration times after pertur
bations can be very long. The few long term studies
following disturbance, as for instance the vegetation
recovery of ghost towns in the American West, demon
strate these long recovery times that often exceed many
decades. It can be generalized that any human impact that
changes the soil structure will last very long. Unlike in
mesic environments, abandoned agricultural fields in
deserts will recover only very slowly (if at all) to natural
desert vegetation. Additionally, formerly irrigated fields
will have elevated salt concentrations for long periods of
time. Soil surface disturbance caused by off road vehicles
Deserts
inflict severe changes in hydrological characteristics of
soils, which might remain permanently. The increase in
off road vehicles in the North American deserts, and
increasingly also in the Middle East, is a serious threat
to deserts and desert biotas.
Desertification, largely the human caused extension of
the desert, is one the most serious problems facing the
globe. Causes of the growth of the desert regions are
multifaceted and are a combination of natural long term
variation in the weather, climate destabilization, and
human mismanagement due to overpopulation and
land use change. Under the UN Convention to Combat
Desertification, desertification is defined as land degrada
tion in arid, semiarid, and dry subhumid areas resulting
from various factors, including climatic variations and
human activities. The effects of desertification promote
poverty among rural people, and by placing stronger
pressure on natural resources, such poverty tends to rein
force existing trends toward desertification. It is now clear
that in several regions, desert environments are expand
ing. This process includes general land degradation in
arid, semiarid, and also in dry and subhumid areas.
Clearly in areas where the vegetation is already under
stress from natural or anthropogenic factors, periods of
drier than average weather may cause degradation of the
vegetation. If such pressures are maintained, soil loss and
irreversible change in the ecosystem may ensue, so that
areas that were formerly savanna or scrubland vegetation
are reduced to human made desert. To counter this pro
cess that will increasingly endanger lives and livelihood of
millions of people (not to speak of drastic effects on the
biodiversity of the planet), synoptic management
approaches are needed that combine understanding of
the process and investigation into the regional causes of
the process, in order to comprehend the effects on the
Earth’s overall system. It is important to emphasize that
desert border areas that undergo desertification will
not simply convert into natural deserts. Disturbed and
overused semiarid zones are characterized by lower
biodiversity than original, natural deserts. Therefore
desertification will not simply increase the global area of
deserts; it will create large tracts of devastated lands.
Human activity and human caused climate change
will facilitate the migration of ruderal (disturbance
adapted) plant species into locally favorable microsites
within the desert. This has been shown for the vegetation
along roadsides in the Middle East and in the North
American southwest. Even though this might enhance
local, small scale species richness, an overall reduction
in regional diversity and a loss of desert adapted species
might follow. Such a strong mixing of former distinct
biotic zones has been observed along the edges of deserts
in the context of human caused disturbances and climate
change. A wide variety of ‘extrazonal’ plants are crossing
zonal borderlines, a process that will potentially lead to a
239
marked decrease in large scale species diversity. This
migration by species that are native to the general geo
graphic area but are now spreading into new climatic or
biogeographic zones is an overlooked aspect of species
invasion.
Due to typically strong abiotic stress, desert areas
have been in the past remarkably resistant to invasions
by non native organisms. Notable exceptions have been
biological invasions by deliberately introduced organ
isms in Australian deserts (e.g., rabbits and Opuntia
species). However, invasion seems to increase rapidly
worldwide and many desert areas today show a dramatic
increase in the arrival and spread of non native species.
At present, the deserts of the American South West seem
to be affected most. Plants originated from the Old
World, mostly grasses (e.g., annual Bromus species, some
perennial grasses), but increasingly members of other
plant families also have invaded many desert commu
nities and can have strong impacts on native desert
communities. Among the detrimental effects are dra
matic changes in fire regimes and direct competition
with recruiting shrub seedlings and native annual plants,
and even negative effects on adult desert perennials have
been demonstrated. The main reason for these trends is
due to general land use changes in desert and desert
margins. In the Southwestern US, disturbances due to
increasing suburbanization of deserts, besides increases
in nutrient depositions, seem to be central agents of these
changes.
Endangered Species
Many of the larger vertebrate desert species are threa
tened and a number of species have been lost to global
extinction. The openness of the desert habitat and natu
rally small population size makes large mammals and
birds conspicuous and thus very vulnerable to overhunt
ing. Threatened species include the central Asian wild
Bactrian camel (Camelus bactrianus), the onager (Equus
hemionus; a wild ass of southwestern and central Asia),
and large antelope species as the addax (Addax nasomacu
latus) of North Africa and the Arabian oryx (Oryx leucoryx;
Figure 18). Hunting is also the main reason that larger
birds are endangered. Among birds many bustard species
are threatened (e.g., the houbara, Chlamydotis sp.) or are
already extinct (e.g., the Arabian subspecies of the ostrich:
Struthio camelus syriacus). Large predators have been and
continue to be extensively hunted since they are per
ceived to be a threat to livestock (e.g., desert subspecies
of the Old World leopards, Panthera pardus jarvisi).
International efforts to save many of the larger endan
gered animals are currently ongoing; many of these efforts
involve reintroductions.
240
Deserts
research enterprises. The permanent research sites estab
lished worldwide during the International Biological
Program (IBP) are good examples; in the US, many of
these continue to be monitored under the Long Term
Ecological Research (LTER) program.
See also: Mediterranean; Steppes and Prairies;
Temperate Forest.
Further Reading
Figure 18 Many larger desert animals became extinct in the
wild due to hunting pressure. These captive Arabian oryx (Oryx
leucoryx) are part of a breeding effort that led to release
operations into formerly occupied desert ranges (Oman, Bahrain,
Jordan, Saudi Arabia). Wadi Araba, Israel, May 2003. Photograph
by C. Holzapfel.
Invasive species can have detrimental effect on threa
tened species as well. An example is the increased fire
frequency caused by annual, non native grasses, which is
threatening populations of the desert tortoise (Gopherus
agassizii) in the deserts of North America.
Desert Research
One of the major attractions of desert ecosystems for
scientists lies in their simplicity. Spatial patterns of life
are often visible and clear cut and ecologists tend to feel
empowered by the sense of ecological understanding. As
any desert scholar will have to attest though, this simpli
city is only relative. In comparison to more complex
systems, deserts seem to invite ecological questions with
greater ease than for instance tropical rainforests would.
Therefore much of basic ecological knowledge has been
founded in desert research and these dry places more
often than not were used as simplified models for the
green and (forbiddingly) complex world. Thus it is no
wonder that the desert has spawned many research
efforts, notably among them large, coordinated ecological
Belnap J, Prasse R, and Harper KT (2001) Influence of biological soil
crusts on soil environments and vascular plants. In: Belnap J and
Lange OL (eds.) Biological Soil Crusts: Structure, Function, and
Management, pp. 281 300. Berlin: Springer.
Evenari M, Schulze E D, Lange O, Kappen L, and Buschbom U (1976)
Plant production in arid and semi arid areas. In: Lange OL,
Kappen L, and Schulze E D (eds.) Water and Plant Life Problems
and Modern Approaches, pp. 439 451. Berlin: Springer.
Evenari M, Shanan L, and Tadmor N (1971) The Negev. The Challenge
of a Desert. Cambridge, MA: Harvard University Press.
Fonteyn J and Mahall BE (1981) An experimental analysis of structure in
a desert plant community. Journal of Ecology 69: 883 896.
Fowler N (1986) The role of competition in plant communities in arid and
semiarid regions. Annual Review of Ecology and Systematics
17: 89 110.
McAuliffe JR (1994) Landscape evolution, soil formation, and ecological
patterns and processes in Sonoran Desert bajadas. Ecological
Monographs 64: 111 148.
Noy Meir I (1973) Desert ecosystems: Environment and producers.
Annual Review of Ecology and Systematics 4: 25 41.
Petrov MP (1976) Deserts of the World. New York: Wiley.
Polis GA (ed.) (1991) The Ecology of Desert Communities. Tucson, AZ:
University of Arizona Press.
Rundel PW and Gibson AC (1996) Ecological Communities and
Processes in a Mojave Desert Ecosystem: Rock Valley, Nevada.
Cambridge: Cambridge University Press.
Schmidt Nielsen K (1964) Desert Animals: Physiological Problems of
Heat and Water. London: Oxford University Press.
Shmida A (1985) Biogeography of the desert flora. In: Evenari M, Noy
Meir I, and Goodall DW (eds.) Hot Deserts and Arid Shrublands,
pp. 23 88. Amsterdam: Elsevier.
Smith SD, Monson RK, and Anderson JE (1997) Physiological Ecology
of North American Desert Plants. Berlin: Springer.
Sowell J (2001) Desert Ecology: An Introduction to Life in the
Arid Southwest. Salt Lake City, UT: University of Utah Press.
Whitford WG (2002) Ecology of Desert Systems. San Diego: Academic
Press.
Dunes 241
Dunes
P Moreno-Casasola, Institute of Ecology AC, Xalapa, Mexico
Published by Elsevier B.V.
Introduction
Abiotic Factors
Biological Factors
Further Reading
Introduction
reducing erosion. Sometimes salt forms a whitish crust
on the sand surface, also bonding sand grains.
Sand grains come in a wide variety of shapes, colors,
and densities, depending on their origin and on how long
they have been rolling in water currents and wind. Silicate
sand and calcium carbonate sand (formed by fractured
shells and skeletons) are the more common components
of coastal dunes. Sand texture, as well as shape and den
sity, affect transport. Smaller particles are easier to move
than larger ones. Sediment size is measured on the
Wentworth scale. It is harder for angular grains to become
airborne but they may move further once they have.
Denser grains are harder to move and often accumulate
as lag deposits on the upper beach.
Almost all wind blown sand travels quite close to the
ground, through a mechanism called saltation. Individual
grains move in a series of continuous leaps. Once air
borne, a grain describes a curve path, and lands hitting
the ground at a low angle, but with sufficient force to
rebound into the air again. It hits other sand grains that
become airborne and do the same thing. In a short time,
there is a considerable amount of sand in the air. Under
most circumstances, deposition takes place within a short
distance although sometimes sand may be transported
long distances alongshore where the wind blows parallel
to the coast. Deposition is favored by obstacles such as
driftwood, clumps of vegetation, boulders, and plastic
objects which perturb air flow and create a shelter zone.
Small dunes are formed with their tails – called trailing
ridges – stretching downwind.
Changes in wind strength and direction cause rapid
resedimentation. Often a dune’s surface changes by the
hour, creating complex stochastic patterns. Over time,
these processes create recognizable dune bedforms such
as ripples, sand waves, and barchans.
Most coastal dunes form in the presence of vegetation.
An important determinant of dune form is the drag
imposed by the vegetation on the air flow. Dunes can be
classified according to the percentage vegetation cover. At
one extreme are dunes that have been stabilized by their
vegetation cover (fixed, shore parallel ridges and parabolic
dunes) and at the other are the free wind forms (barchan or
sand wave dunes, transverse dunes). Transitional forms are
typified by a fragmented topography (hummock dunes).
Coastal beaches and dunes have a worldwide distribution.
They are common in both temperate and humid tropical
areas, in arid climates, and in regions covered by snow
during the winter. Beaches and dunes are considered two
of the most dynamic systems. They are not permanent
structures, but rather huge sand deposits that move and
have an episodic supply of sand.
They can be found in deserts as well as on dissipative
coasts with a plentiful supply of sediments and where
there are strong onshore winds or winds that are parallel
to the coastline. Sand dunes are eolian bedforms and
beaches are marine geomorphic structures. Dunes form
from marine sand delivered to the beach from the near
shore by waves. The exposed sediment is dried by the sun
and the wind then transports sand inland to form incipient
dunes and foredunes. Tidal range is important in this
process since a high range exposes a large intertidal area
that often dries out between the tides. These sediments
constitute a major source of wind blown sand given that
sand sized sediments are more easily transported by wind.
Dune size varies considerably. Some of the biggest
dunes are found in deserts such as Badain Jaran Desert
in the Gobi Desert in China (approximately 500 m), the
Sossuvlei Dunes, Namib Desert (380 m), and the Great
Sand Dunes National Park Preserve in Colorado, USA
(230 m). Along the coast, on the Bassin d’Arcachon,
France, is Europe’s largest sand dune, the Dune du Pyla,
nearly 3 km long, reaching 107 m in height, and moving
inland at a rate of 5 m yr 1.
Dune Origin and Formation
Wind is the main agent forming sand dunes. There is an
exchange of sediments between beaches and dunes and
this is part of a natural process that maintains both mor
phological stability and ecological diversity. Once
exposed, sand is vulnerable to aerodynamic processes.
Wide beaches are formed in the summer and narrower
beaches during the winter. Storms erode beaches and
transport sand out of the system or to other beaches.
Bonding, both by moisture and chemical precipitates,
may cause surface adhesions, raising thresholds and
242
Dunes
There is a strong interaction between vegetation and
dune form, and there are several patterns of incipient
dunes. Plant form modifies sand deposition, forming a
leading edge (as in the case of Ammophila arenaria), a
trailing edge (Spinifex hirsutus), or intermittent deposition
in clumped vegetation. Perennial grasses such as Agropyron
junceiforme and Elymus arenarius as well as tropical long
branched creepers (Canavalia rosea and Ipomoea pes caprae)
grow laterally and vertically and are able to raise a dune a
meter or two high.
Sand dunes act as a buffer to extreme winds and waves
and they also shelter landward communities. They
replenish the depleted beach and near shore during and
after storms, and are important in the retention of fresh
water tables against salt intrusion. They filter rain water
and are also important habitats for plants and animals.
People have always appreciated their beauty and recrea
tional value.
R. W. Carter wrote that ‘‘Of all the coastal systems,
sand dunes have suffered the greatest degree of human
pressure.’’ Many have been irreversibly altered by human
activities such as tourist developments, golf courses, and
urban growth.
Abiotic Factors
Dune ecosystems may be viewed as a series of gradients
related to various environmental factors, which operate
on different spatial and temporal scales. If we view a
profile from the sea landward, we first have the beach
(near shore and back shore), the embryo or incipient
dunes, and the foredune. The first dune ridge (the next
inland from the foredune) is normally the highest and
forms a continuous sand structure. The second is an
older dune ridge, frequently lower because of the reduc
tion in sand supply and the gradual loss of sand. This
formation occurs when we have a series of parallel ridges,
formed by onshore winds, each ridge lower than the
previous. Sand is trapped by vegetation and saltation
cannot be initiated beneath the vegetation, unless a blow
out forms. Older dune ridges become fragmented when
blowouts and parabolic dunes develop. Parabolic dunes
are formed when prevailing winds blow at right angles to
the dune ridges. Poorly stabilized regions are rapidly
eroded but the more vegetated areas on either side remain
covered by plants for a longer time. As the bare sand of
the central region moves inland, the two horns or tips of
the parabola remain attached to the relatively stabilized
sand of the trailing ridges. A slack (a dune depression
where sand has been blown away until the water table is
exposed) may be formed in the middle, between the
parabola arms. Parabolic dunes can also be formed in
transverse dunes.
Throughout the dune field, there are gradients in
salinity, sedimentation, nutrients, flooding, and shelter.
Dune vegetation forms a complex spatial mosaic, mainly
because of variations in physical gradients which depend
on the distance to the sea and topography. Disturbances
also result in temporal successions that add another
dimension of complexity to the spatial mosaic described.
Sand Movement
Dune movement has only been measured in a few dune
systems and most of the published records are based on
estimates from maps, the height of sand on fence posts,
houses, and trees, etc. The results show that the rate of
movement varies considerably among systems, varying
from a few centimeters per year to 70 m per month, the
latter in New Zealand (personal observation of Patrick
Hesp). Dune formation depends on an adequate supply of
sand and the wind to transport it. The interaction of wind
and vegetation is of primary importance for dune growth.
Colonization by plants accelerates dune growth, because
surface roughness created by vegetation decreases wind
flow and increases sand deposition. Several plants show an
inherent capacity to bind sand and are able to develop
extensive horizontal and vertical rhizome systems. The
growth form and the ecological dynamics of dune plants
are important contributors to foredune growth.
Rhizomatous growth (as in the grass Ammophila) or sto
loniferous growth (as in Ipomoea or Spinifex) can extend the
foredune depositional area by 5–15 m in a few months.
Elymus arenaria (Europe) develops vertical rhizomes
150 cm long and Ipomoea pes caprae (pantropical) can
have 25 m long branches that are buried two or three
times along their length. Figure 1 shows species that are
able to survive and reproduce successfully under high
rates of sand mobility in different parts of the world. In
each region, sand tolerating species have evolved, and
they play a very important role in dune formation. Sand
deposition produces vigorous growth in some of these
species; both plant height and plant cover increase, mak
ing these species excellent dune fixers. Many hypotheses
have been suggested to explain this response of sand dune
plants, but there are few studies in which the explanations
are based on experimental evidence. Changes in soil
temperature, increased space for root development,
higher nutrient and moisture availability, a response to
darkness, meristem stimulation, and interactions with
endomycorrhizae and nematodes are probably factors
that play an important role in this response.
Nutrients
There are great differences in the soil properties of young
dunes (formed by recently blown sand), and those of more
mature dunes in which vegetation has dominated. Newly
Dunes 243
Ammophila breviligulata
Uniola paniculata
Scaevola plumieri
Uniola paniculata
Leymus arenarius
Agropyrum junceum
Ammopbila arenaria
Festuca rubra var. arenaria
Mesembryanthemum aequilaterale
Franseria pinnatifida
Atriplex leucophylla
Abronia maritima
Ixeris repens
Wedelia prostrata
Messerschmitia sibirica
Calystegia soldanella
Blutaparon portulacoides
Panicum racemosum
Spartina ciliata
Scaevola plumieri
Spinifex hirsutum
Palafoxia lindenii
Chamaecrista chamaecristoides
Randia laetevirens
Figure 1 Species that are able to survive and reproduce successfully under high rates of sand mobility in different parts of the world.
Many regions have their own set of species that play important roles in stabilizing sand dunes locally.
blown sand from the beach is low in mineral nutrients.
Dune soils show marked changes as they age. Pioneer
species that initiate dune stabilization are able to live in
very poor soils. On fully vegetated dunes, organic matter
and nutrients accumulate, and the leaching effects of rain
fall decrease. Leaching dissolves carbonate and moves it
downward to the water table. With time, the organic
matter of nutritionally poor soils of younger dunes
increases, and pH decreases. The increase in organic mat
ter content varies among dune systems, depending on the
climate and colonizing species. In high rainfall climates
such as Southport (Lancashire, Great Britain), organic
matter increases slowly at first but much faster after
about 200 years. In Studland, Dorset, the early invasion
of Calluna is largely responsible for a very rapid increase in
organic matter. Primary productivity and the competitive
abilities of coastal plants are frequently limited by nutrient
availability, with nitrogen deficiency the most severe. As
succession advances, plants increase their cover, commu
nities change from grasslands to thickets, and then to
tropical or temperate forests, adding nutrients and organic
matter to soils. In dune depressions, where water is not a
limiting factor for plant establishment, the accumulation of
organic matter is faster. Experiments with dune plants
have shown that many species are slow growing and gen
erally show growth responses characteristic of plants from
infertile habitats.
Salinity
Soil salinity comes from salt spray and foam blown inland,
and the amount of salt usually correlates well with the
distance inland or degree of protection from the wind. In
some regions with a Mediterranean climate, such as
California, soil salinity follows a seasonal progression.
Late summer additions by fog and salt spray result in
high values at this time of the year. Winter rains leach
salt away, salinity decreases, and in early spring reaches
its lowest level. Salinity gradients affect species distribu
tion, especially for those plants sensitive to salinity.
Germination and growth might be difficult when soil
salinity is high. Salts in the soil affect plants by making
water less available, and high salinity is considered a
physiological drought. Frequently, there are no shared
species between the beach and the more sheltered or
inland areas of the dunes. Experiments on sea rockets
(Cakile maritima) and lupines (Lupinus spp.) which were
sprayed with seawater showed that lupine seedlings were
not tolerant of salt spray. The level of salt spray in a
Californian beach may be 1 mg cm 2 d 1 on a calm day,
but is much higher on a windy day. On other beaches and
dunes, where onshore winds are not as strong, airborne
spray is very low and plants are not subjected to these
conditions.
Water
The primary source of water for dune plants is rainfall.
Radiation causes considerable diurnal and nocturnal tem
perature variations. These fluctuations in soil temperature
are sufficient to cause the periodical condensation of water
vapor in the soil. This produces an increase in water
availability from dew that is sufficient to maintain plants
in rainless periods. Fog can be another source of water, but
in some areas it contains salt. Studies in open dune com
munities have shown that soil moisture increases to depths
of about 60 cm below the dune surface and then tends to
fall off. In closed dune communities, where the soil has
244
Dunes
some organic matter, rainfall is absorbed and held near the
surface where it is available to roots. Experiments with
Chamaecrista chamaecristoides seedlings, a species that
thrives in mobile dunes, showed that they had the ability
to withstand total lack of watering for more than 80 days.
This probably allows them to survive during the dry
months of the year in the dunes of the Gulf of Mexico.
In a wet year, there may be widespread flooding in
dune depressions. Blowouts are wind hollows or basins of
exposed sand within dunes, called slacks or depressions.
They frequently arise through the erosion of deflated
areas in poorly vegetated dunes. The deflation limit is
reached owing to the presence of water, algae, or the
accumulation of coarse immovable material. Deposition
occurs around the borders of the blowout and vegetation
may recolonize the area. The water table falls during the
dry season and recovers during the rainy months and the
composition of the plant community reflects this ground
water regime. When the soil is completely flooded, the
prevailing anaerobic conditions can influence its chemical
composition and the concentration of nutrients, affecting
plant survival and growth. Flooding can cause the local
extinction of some non wetland species and facilitate the
establishment of others.
The frequency and duration of slack inundation are
factors that can alter the distribution of vegetation and
plant community composition. When flooding takes place
occasionally, on very wet years, many of the plants die,
and when the water recedes, colonization takes place
again. In wet slacks that flood every year, water loving
plants establish and a completely different set of species is
found in these areas. Thus community composition will
depend on the differential tolerance of plants to the
environmental conditions associated with inundation,
particularly anoxia. Species are good indicators of the
water table depth. In temperate areas, Erica tetralix,
Glyceria maxima, Carex nigra, and Juncus effusus are some
of the more common species. In Europe, slacks are very
important for endemic and rare species. In tropical
regions, Cyperus articulatus, Lippia nodiflora, Hydrocotyle
bonariensis (Mexico) and Paspalum maritimum, Fimbrisitylis
bahiensis, Marcetia taxifolia (Brazil) are frequently found in
these depressions. Thickets are also common and in
Mexico they are formed by Pluchea odorata, Chrysobalanus
icaco, and Randia laetevirens. In Brazil, there is Ericaceae
scrub dominated by Humiria balsmifera, Protium icicariba,
and Leucothoe revoluta.
Temperature
On open sand dunes, there are considerable diurnal and
nocturnal temperature variations. In California, on an
August day, when the air temperature was above 15.5 C
1 m above the ground, the soil surface was at 38 C and
soil 15 cm below the surface was at 19 C. In a Nevada
desert, the soil surface temperature reaches 65.5 C and in
Veracruz, in the coastal dunes in the central Gulf of
Mexico, the soil surface also reaches 65 C. These tem
peratures are critical for seed germination and seedling
establishment. Some species, such as hard coated
legumes, need these temperature oscillations over several
weeks to break the hard seed coat. They lie on the soil
during the dry season, and the temperature fluctuations
break the testa. When the rains come, they are ready to
germinate. Vegetation cover reduces these temperature
oscillations considerably. There are also temperature dif
ferences over short distances because of topography and
orientation. In the dunes of temperate regions, there are
temperature and vegetation differences depending on
dune slope orientation.
Habitats
Coastal dunes are very dynamic systems offering a wide
variety of habitats with different physical and biotic con
ditions, and this allows for the existence of species with
very diverse life history traits. They can be visualized as a
permanently changing environment with distinct degrees
of stabilization that is closely correlated with topography,
the disturbance produced by sand movement, and distance
to the sea. Dune habitats can be classified into three types:
(1) those where sand movement dominates, sea spray is
sometimes important, and nutritionally poor soils prevail
(they are formed by the sandy beach, embryo or incipient
dunes, foredunes, blowouts, and active dunes); (2) humid
and wet slacks or depressions, that is, those habitats which
become inundated during the rainy season when the water
table rises and they sometimes may even form dune lakes
with wetland vegetation; (3) stabilized habitats, which
show no sand movement, conditions are less stressful,
and there is more organic matter in the soil. Vegetation
cover is more continuous – grasslands, thickets, wood
lands, and tropical forests.
Figure 2 shows a beach and dune topographic profile
as well as the intensity of some of the abiotic factors
mentioned and the areas where they affect the dune
system.
Biological Factors
Dune plants are found all over the world, from the frosty
regions of Canada and Patagonia, to the tropical areas of
the Caribbean, Africa, and the South East, and the dry
regions of Australia, Peru, and California. They are sub
jected to very different climatic conditions, they share few
species, and life forms vary. Raunkaier developed an eco
logically valuable system of plant classification, based on
the position of the vegetative perennating buds or the
persistent stem apices in relation to the ground level
Dunes 245
(a) Habitats
Dunes
Stabilized
Slacks
Beach
Incipient dunes
Foredune
Active dunes
Humid slacks
Wet slacks
Sheltered zones
Stabilized dunes
Topographic profile
(b) Intensity of abiotic and biotic factors
Inundation (freshwater)
Inundation (seawater)
Salt spray
Wind
Sand movement
Biotic interactions
Figure 2 (a) Beach and dune topographic profile showing each of the habitats. (b) Intensity, indicated by the width of the line, of some
of the abiotic factors mentioned along with the areas where they affect the dune system. Reproduced with permission from MorenoCasasola P and Vázquez G (2006) Las comunidades de las dunas. In: Moreno-Casasola P (ed.) Entornos veracruzanos: La costa de La
Mancha. Xalapa, Mexico: Instituto de Ecologı́a AC.
during the unfavorable season of the year, which can be
either the cold winter or the dry summer. There is a strong
correlation between the climate of an area and the life
forms of the plants present. This system allows compari
sons to be made between particular areas or regions. The
biological spectrum found in a dune system is an expres
sion of the number of species in each life form class as a
percentage of all the species present. A comparison of the
biological spectrum of a dune system in Braunton Burrows
(North Devon, Great Britain) and one in La Mancha
(Veracruz, Gulf of Mexico) was made. Braunton Burrows
is dominated by hemicryptophytic plants (perennating
buds are at the surface of the sand) and therophytes
(annuals that survive the unfavorable season as seeds); La
Mancha is dominated by phanerophytic types (these grow
continually, forming stems that often have naked buds
projecting into the air, such as in Hippophae rhamnoides or
Chamaecrista chamaecristoides).
Facilitation and Succession
Ecological succession refers to a more or less predictable
and orderly change in the composition or structure of an
ecological community. Facilitation is one of the mech
anisms by which succession takes place. It occurs
when plant establishment is favored or facilitated by
previously established plant communities that ameliorate
environmental extremes. Physical factors in dune envir
onments produce a very harsh environment where few
plants can survive. Several studies show that facilitation
takes place in the early stages of colonization and succes
sion in coastal dunes. As succession proceeds, pioneer
species will tend to be replaced by more competitive
species, the abiotic environment will become less harsh,
and biotic interactions such as competition and predation
will be more common.
Dune succession is comprised of a pioneer (also called
yellow dunes, associated with the most seaward dunes
that are still receiving a significant input of wind blown
sand), intermediate, and mature stages (gray dunes or
inland dunes with little or no sand, a high humus content,
and where soil development has occurred). The rate of
succession varies with the harshness of the environment.
This is related to the abiotic factors mentioned and the
vegetation stock. Detailed studies have been undertaken
in the Lake Michigan dunes, a salt free system, in the
coastal dunes of Newborough Warren, several sites in
Holland, and La Mancha, among others.
Competition, Predation, Disease
Biotic interactions among plants are an important deter
minant of structure and dynamics. Competition is
recognized as one of the most important forces
246
Dunes
structuring ecological communities. Competition is the
interaction of organisms or species such that, for each, the
birth or growth rate is depressed and the death rate
increased by the presence of the other organisms (or
species). Well known examples of competition between
plants growing on coastal dunes are grass and shrub
encroachment and the invasion of exotic species.
Grass encroachment occurs when aggressive and com
petitive grasses spread over dune areas, reducing
biodiversity because of the dominance of a few species.
Grass encroachment is found in many dune areas, where
grasslands become the dominant community type. Among
the more aggressive species are Calamagrotis epigejos,
Ammophila arenaria, and Schizachyrium scoparium. Shrub
encroachment is also common, for example, in the
Caribbean (Coccoloba uvifera).
Species introduction has been a common practice in
dunes, both for dune stabilization and for cattle ranching
activities. European marram grass was widely dispersed in
other regions that were quite different from its native
Europe, mainly to fix sand dunes. Several conifers have
also been used for example in Doñana’s dune system in
southern Spain. African grasses (e.g., Panicum maximum) have
been brought to America and used to replace local grass
species because they have been considered better fodder.
Neither the effects of fauna nor those of grazing ani
mals (especially rabbits) on dunes have received the
attention they deserve. The importance of herbivory by
rabbits was seen in Great Britain during the outbreak of
myxomatosis, a viral disease which infects rabbits. The
disappearance of rabbits led to profound changes in the
structure of the vegetation, mainly the development of
scrub in several dune areas. Rabbits also produce nitrogen
and phosphorus enrichment beneath the scrub species
under which they find shelter, causing N fixing root
nodules to invade.
Lethal yellowing is a specialized bacterium, an obli
gate parasite that attacks many species of palms, including
the coconut palm (which has become the symbol of tro
pical beaches). Extensive coconut plantations in the
Tropics have been abandoned because of coconut die
back, and shrub encroachment has taken place.
Symbiotic Relations
Symbiotic associations involving nitrogen fixation by
microorganisms are frequent in dunes. There are nitro
gen fixing bacteria such as Rhizobium, which form a
symbiosis with numerous forbs and shrubs in temperate
and tropical dune systems. Some of the plants showing
nodules are Ulex europaeous, Trifolium spp., Lupinus arbor
eus, and Hippophae rahmnoides in Europe, Acacia shrubs in
South Africa, and Chamaecrista chamaecristoides in Mexico.
In foredunes and mobile dunes, pioneer grasses such as
Ammophila, Elytrigia, and Uniola show different degrees of
infection by vesicular arbuscular mycorrhizae (VA). The
major benefit to these grasses is probably enhanced phos
phorus uptake under conditions of phosphorus limitation.
They also help in the aggregation of sand particles.
Tropical sand dune plants also frequently show symbiosis
with mycorrhizae.
Sand dunes are harsh environments where abiotic fac
tors act as filters that determine species survival. The
interactions between abiotic and biotic factors in sand
dunes change as dunes mature. They are delicate systems
in which plant cover, formed by different vegetation
structures and species assemblages, maintains the system
in a stabilized condition. Higher diversity is found when
there are different habitats. Today, these fragile systems
are endangered and the urbanization of the coast is
increasing. We must find ways to make our activities
and dune conservation compatible.
See also: Deserts; Floodplains; Landfills.
Further Reading
Barbour MG, Craig RB, Drysdale FR, and Ghiselin MT (1973) Coastal
Ecology: Bodega Head. Los Angeles, CA: University of California
Press.
Carter RWG (1988) Coastal Environments: An Introduction to the
Physical, Ecological and Cultural Systems of the Coastlines. New
York: Academic Press.
Hesp PA (2000) Coastal sand dunes: Form and function. Massey
University Coastal Dune Vegetation Network, New Zealand,
Technical Bulletin No. 4, 28pp.
Lortie CJ and Cushman JH (2007) Effects of a directional abiotic
gradient on plant community dynamics and invasion in a coastal
dune system. Journal of Ecology 95(3): 468 481.
Martı́nez ML and Psuty NP (eds.) (2004) Coastal Dunes: Ecology and
Conservation. Berlin: Springer.
Moreno Casasola P and Vázquez G (2006) Las comunidades de las
dunas. In: Moreno Casasola P (ed.) Entornos veracruzanos: La costa
de La Mancha. Xalapa, Mexico: Instituto de Ecologı́a AC.
Olson JS (1956) Rates of succession and soil changes on southern Lake
Michigan sand dunes. Botanical Gazette 199: 125 170.
Packham JR and Willis AJ (1997) Ecology of Dunes, Salt Marshes and
Shingle. London: Chapman and Hall.
Pilkey OH, Neal WJ, Riggs SR, et al. (1998) The North Carolina Shore
and Its Barrier Islands: Restless Ribbons of Sand. London: Duke
University Press.
Ranwell DS (1972) Ecology of Salt Marshes and Sand Dunes. London:
Chapman and Hall.
Rico Gray V (2001) Encyclopedia of Life Sciences: Interspecific
Interaction. New York: Macmillan Publishers.
Seeliger U (ed.) (1992) Coastal Plant Communities of Latin America.
New York: Academic Press.
Van der Maarel E (1993) (ed.) Dry Coastal Ecosystems, vol. 2A.
Amsterdam: Elsevier.
Van der Maarel E (1994) (ed.) Dry Coastal Ecosystems, vol. 2B.
Amsterdam: Elsevier.
Van der Maarel E (1997) (ed.) Dry Coastal Ecosystems, vol. 2C.
Amsterdam: Elsevier.
Estuaries
247
Estuaries
R F Dame, Charleston, SC, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Definitions of Estuarine Ecosystems
Geomorphological Types of Estuaries
Estuarine Ecosystems and Maturity
Estuaries as Complex Systems
Major Estuarine Subsystems or Habitats
Material Fluxes
Estuarine Ecosystem Resilience and Restoration
Further Reading
Introduction
Remane curve
Marine
species
No. of species
Estuarine ecosystems are among the most complex and
complicated systems in the biosphere. Because they are at
the interface of terrestrial, freshwater, and marine sys
tems, estuaries are subject to massive fluxes of materials
and energy. Further, as a large percentage of the human
population lives in close proximity to estuarine and
coastal environments, anthropogenic impacts and stress
are major driving factors in determining the health and
functional status of estuarine ecosystems. In this section,
the structure and function of estuarine ecosystems are
examined.
Freshwater
species
Salinity
Figure 1 A generalized Remane curve with number of species
plotted versus estuarine salinity gradient.
Definitions of Estuarine Ecosystems
To set the stage for any discussion of estuarine ecosys
tems, a clear working definition is needed. One of the
simplest and most utilized definitions of an estuarine
ecosystem is:
the zone where freshwater from land runoff mixes with
seawater.
Another common definition is:
An estuary is a semi enclosed coastal body of water that
has free connection with the sea where seawater is diluted
by freshwater derived from land drainage.
The preceding definitions focus on the geomorpholo
gical and hydrological aspects of estuaries with no
mention of the abiotic or physical driving sources of
energy, that is, tides and solar insolation. Nor are any
biotic components or processes utilized. Thus, the follow
ing definition is proposed:
An estuarine ecosystem is a system composed of relatively
heterogeneous biologically diverse subsystems, i.e., water
column, mud and sand flats, bivalve reefs and beds, and
seagrass meadows as well as salt marshes. These subsys
tems are connected by mobile animals and tidal water
flows that are embedded in the geomorphological structure
of creeks as well as channels, and together form one of the
most productive natural systems in the biosphere.
Recent quantitative studies indicate that estuaries are
ecoclines that are composed of gradients containing
relatively heterogeneous subsystems that are environ
mentally more stable than ecotones (Figure 1).
Ecoclines are boundaries with more gradual, progressive
change between freshwater and the sea. In this view, the
organisms in the estuary are either from freshwater or
from marine environments; there are no brackish water
species. Each estuarine system will respond to at least a
freshwater and a marine gradient as well as have its own
particular combination of biological and physical compo
nents and processes. Thus, every estuarine ecosystem is
unique.
Geomorphological Types of Estuaries
Bar-Built and Lagoonal
Bar built or lagoonal estuaries form in the areas behind
sandy barrier islands and usually drain relatively small
watersheds. The exchange of water between the estuary
248
Estuaries
and the sea occurs through tidal inlets. Astronomical
tides and winds are the major forces controlling water
circulation and water height. The areas behind barrier
islands are generally subject to less wave action and this
promotes the development of extensive wetlands. Bar
built estuaries are generally smaller than other estuarine
types, suggesting that they have a higher surface area to
volume ratio and, therefore, play a greater role in eco
logical processes than was previously thought. Well
studied examples of bar built estuaries are found along
the temperate and subtropical coasts of eastern North
America, Europe, Asia, and the southern and eastern
shores of Australia.
Riverine Estuaries
There are two fundamentally different riverine systems
(Figure 2). First, are those that arise in the piedmont,
have extensive watersheds, receive substantial freshwater
discharge, but only a small portion of their watershed is
covered by wetlands. Chesapeake Bay and San Francisco
Bay in North America as well as the Eastern Scheldt in
Northern Europe are well studied examples of this type
of riverine system. A second type of riverine system
known as coastal plain estuaries are characterized by a
much gentler slope with proportionally more wetlands
than piedmont estuaries. Generally, these systems are
less studied, smaller and have a lower, more sluggish
flows.
Estuarine Ecosystems and Maturity
In an attempt to place a more ecosystems oriented
emphasis on estuaries, the ‘geohydrologic continuum
theory of Marsh Estuarine ecosystem development’ was
proposed as a scheme for categorizing estuarine ecosys
tems. In this theory, the tidal channels of the estuarine
ecosystem represent a physical or geohydrologic model
of how the ecosystem adapts until there is a change of
state. Mature portions of the system are at the ocean–
estuary interface, mid aged components are intermediate
River
Sea
within the longitudinal distribution of the system, and
young or immature areas are at the land–estuary inter
face. Mature systems export particulate and dissolved
materials, mid aged areas import particulate and export
dissolved materials, and immature systems import both
particulate and dissolved materials. Some estuarine eco
systems may have all three types while others may have
only one or two.
Estuaries as Complex Systems
While it is generally acknowledged that ecosystems are
complex systems, it is appropriate to describe estuarine
ecosystems in the context of the complex systems
approach. Complexity as used here can be defined by
(1) the nonlinear relationships between components;
(2) the internal structure created by the connectivity
between the subcomponents; (3) the persistence of the
internal structure as a form of system memory; (4) the
emergence or the capacity of a complex system to be
greater than the sum of its parts; (5) the reality that
complex systems constantly change and evolve in
response to self organization and dissipation; and
(6) behaviors that often lead to multiple alternative states.
Thus, estuarine ecosystems are open nonequilibrium sys
tems that exchange matter and energy as well as
information with terrestrial and marine ecosystems as
well as internal subsystems. These exchanges not only
connect various components, but are the essential ele
ments of feedback loops that generate nonlinear
behavior and the emergence of structures and behaviors
whose sum is greater than the whole. These systems do
exhibit alternate states, for example, Chesapeake Bay
appears to have a benthic state dominated by oysters
and a water column state dominated by plankton.
Major Estuarine Subsystems or Habitats
The landscape approach to estuarine ecosystems focuses
on subsystems or habitats as major components within
estuaries. Because organisms respond to the amount of
change in the physical (abiotic) environment, their reac
tion to their environment results in subsystems or habitats
composed of specific groups of species that are adapted to
that particular set of abiotic factors. In estuarine ecosys
tems, the major abiotic factors are salinity, water velocity,
intertidal exposure, and depth.
Water Column
Bottom sediments
Figure 2 Generalized material flux patterns in a Riverine
estuary.
Water is the primary medium for the transport of matter
and information in estuarine ecosystems. Freshwater
enters the estuary either as precipitation or as an
Estuaries
accumulation driven by gravity down slope through
streams and rivers to the estuary. Salt water enters the
estuary from the sea via tidal forcing. The gradient of
increasing salt concentration from freshwater to marine
divides the estuary into zones of salt stress and subse
quently into different pelagic subsystems (Figure 2).
Phytoplankton are small chlorophytic eukaryotes that
drift as single cells or chains of cells in estuarine currents.
Diatoms and dinoflagellates are the dominant groups
while species composition of a specific system is usually
determined by salinity, nutrients, and light. They are a
major component of the estuarine water column and
provide food for many suspension feeding animals.
Planktonic primary production is seasonal and varies
from distinct peaks in the arctic to spring and autumn
blooms in temperate systems and almost no peaks in
tropical estuaries. Average annual planktonic primary
production in estuaries is about 200–300 gC m 2 yr 1
and is mainly a function of light, nutrient availability,
and herbivore grazing.
There are two major categories of zooplankton: holo
plankton that in most estuaries are dominated by calanoid
copepods which spend their entire life in the planktonic
state and the diverse meroplankton that only spend their
larval state in the plankton. Most estuarine zooplankton
are believed to be herbivores and play a major role in
connecting carnivores to phytoplankton. They are also
thought to be major sources of inorganic nutrients that are
available to phytoplankton.
The microbial loop in estuaries is composed of micro
and nano planktonic bacteria, protozoans, and flagellates.
Initially, the microbial loop was thought to play a major
role in recycling nutrients with dissolved organic matter
(DOM) a major product. However, the recent finding that
a sizable proportion of DOM is made up of viruses has
forced a major change in the microbial loop model
(Figure 3). The current paradigm of the microbial–viral
249
loop envisions the viruses (1010 l 1) as 10 times more
abundant than bacteria (109 l 1) and controllers of bacter
ial diversity and abundance. The viruses are small
(20–200 nm), ubiquitous particles that use the process of
cell lysis to attack and kill bacteria. As a result, more
bacterial biomass is shunted into DOM and away from
the macroplankton and suspension feeding macro
benthos. The much more rapid viral recycling of
nutrients also has the potential to generate more stability
in the system.
Large mobile animals, birds, terrestrial and aquatic mam
mals, and fish, shrimps and crabs, are common residents as
well as transients in estuarine systems. These animals trans
form and translocate materials both within the estuary and
between the estuary and other systems. The nekton organ
isms, in particular, use the tidally forced water column as a
pathway between deeper channels and intertidal habitats
where they seek refuge, feed, and develop.
Marshes and Mangroves
Emergent vascular plant dominated intertidal wetlands
are major subsystems in most estuaries. The two most
common habitats are geographically zoned latitudinally
with marshes dominating the temperate zone and man
groves the frost free subtropical and tropical zones. Both
are found in low energy wave protected, sedimentary,
high salinity, and intertidal environments near the
mouth of the estuary. While wetlands in the high salinity
portion of estuaries are low in species diversity (almost
monocultures) of vascular plants, diversity is much higher
in the freshwater reaches.
Salt marshes reach their greatest extent and produc
tivity along the Gulf and southeast Atlantic coast of
North America where the cord grass Spartina alterniflora
dominates. This high production is the result of near ideal
conditions of temperature, salinity, light, sediment tex
ture, nutrients, and tidal range. Marsh grasses produce
large quantities of both above and belowground biomass
that accumulates in the surrounding sediments (Table 1).
The stems and leaves of the grasses also provide a struc
tural base for an epiphytic community that further
increases production. Decomposition processes in the
organically rich sediments generate a strongly anaerobic
reducing environment making the salt marsh a major
Table 1 Primary production in estuaries
Figure 3 A simple microbial–viral loop food web for an
estuarine system. D, dissolved organic matter; G, grazers; H,
heterotrophs; N, nutrients; P.P., primary producers; V, viruses.
Primary producer
Annual primary production (g C m 2)
Macrophytes
Spartina
Rhizophora
Microphytobenthos
Epiphytes
Phytoplankton
400–2480
696–2100
50–200
12–260
25–150
250
Estuaries
center for nutrient cycling. The nutrient uptake mechan
isms of vascular plants are poisoned by the reducing
environment; however, air passages in the roots, rhizomes,
and stems of these grasses aerate the surrounding sedi
ments so that nutrient uptake can be maintained. The
vertical stems and leaves of Spartina also serve as a passive
filter that slows water flow, can remove via deposition
suspended sediments from the water column, and allows
many marshes to maintain their elevation with respect to
rising sea level. This same environment provides food and
refuge for many economically important nekton.
Mangroves are intertidal, tropical, and subtropical
woody vascular plants that fill a niche similar to that of
Spartina. In the high salinity portions of the estuary, the
red mangrove, Rhizophora, dominates. Red mangroves
have prop roots that lift the plants above the reducing
environment of the surrounding sediments. There is a
gradient from high production in riverine swamps to
low production in high salinity scrub areas. On a global
scale of increasing light with decreasing latitude, the
closer a system to the equator, the higher the mangrove
productivity. Nutrients have also been implicated as a
major limiting factor on mangrove productivity. There is
evidence that mangrove production is enhanced by flushing
action of storms. In addition to being a nursery for many
fish, shrimps, and crabs, the structural mass of mangroves
may form a protective buffer to the impacts of storm surges
and tsunamis on coastal and estuarine systems.
Seagrasses
Seagrasses are submerged vascular plants that are found
in aerobic, clear water, high salinity systems with mod
erate water flow. Cold water systems are dominated by
eel grass, Zostera, and in the tropics turtle grass, Thalassia,
is the major group. These grasses are not found in estu
aries with high suspended sediment loads, that is, Georgia
and South Carolina where there is insufficient light pene
tration to support their growth. They are also limited
to the upper 20 m of water because water pressure
compresses their vascular tissues. Maximum seagrass pro
duction can approach 15–20 gC m 2 d 1. The high
productivity of the seagrass is almost equaled by the
productivity of the epiphytes on their leaves; however,
the sediment trapping abilities of seagrasses give them an
advantage over phytoplankton and epiphytes in nutrient
limiting conditions. The structure of the seagrasses pro
vides feeding habitat for many mobile animals as well as
deposit feeding and suspension feeding benthos.
Invertebrate Reefs and Beds
Suspension feeding benthic animals are common in most
estuaries because of the high availability of suspended
phytoplankton. A number of bivalves and a few worms
can aggregate in very dense, high biomass beds or reefs.
These structures are found both intertidally and subtid
ally in high to moderate salinities. The eastern oyster,
Crassostrea virginica, in its intertidal form builds some of
the most extensive aclonal reefs known. Intertidal beds of
Crassostrea and Mytilus can have biomass densities exceed
ing 1000 gdb m 2. Depending on the estuary, suspension
feeders such as oysters and mussels have been shown
to control phytoplankton populations in some systems
and influence nutrient cycling by short circuiting plank
tonic food webs and reducing the recycle time for
essential nutrients. There is evidence that the presence
of a significant bivalve suspension feeder component in
estuarine ecosystems enhances system stability.
Mud and Sand Flats
Mud and sand flats are common to the intertidal zone
of most estuaries. The major biotic components of tidal
flats are bacteria, microbenthic algae, small crustaceans,
and burrowing deposit feeders. As in the water column,
the microbial–viral loop is thought to play a major role
in the decomposition of organic matter in tidal flat
sediments. In some estuaries, the microbial–viral loop
utilizing a variety of electron acceptors may represent a
significant sink for matter and energy. Thus, the pre
vailing processes on these flats can potentially redirect
the fluxes of matter and energy away from macrofaunal
food webs to those dominated by microbial processes.
The occurrence of tidal flats was originally attributed
to the hydrodynamics and sediment sources in tidal
creeks; however, with the application of complexity
theory to ecological systems, these flats are also being
described as alternative states of salt marshes and
bivalve beds.
Material Fluxes
Water Fluxes and Residence Times
Interest in the exchange of nonliving materials and organ
isms between estuarine ecosystems and the sea was
initiated by the first quantitative metabolic studies on
the high productivity of marsh dominated estuaries.
These studies were first synthesized in simple energy
budgets that were found to explain less than 50% of the
productivity of estuarine ecosystems. Investigators specu
lated that the unaccounted for energy must be exported
from the estuarine ecosystem by tidal currents. This idea
led to the ‘outwelling hypothesis’ that states that estuarine
ecosystems produce much more organic material than can
be utilized or stored by the system and that the excess is
exported to the coastal ocean where it supports near
coastal ocean productivity. While the energy budget or
mass balance approach is a cheaper and quicker method of
Estuaries
determining the direction of material fluxes, in recent
years the direct measurement of material fluxes is favored
because this approach provides statistically meaningful
results.
Another aspect to the fluxes of materials in estuarine
systems is the time the water mass remains in the system
or residence time (also known as flushing time or turn
over time). Residence time can provide essential
information to resource managers on the retention and
dispersal of toxins, the incubation of invasive species, and
the carrying capacity of a system for benthic suspension
feeders (Figure 4). Recent studies on the physics and
geomorphology of water in estuarine tidal channels sug
gest that the residence time of water may vary greatly
from place to place within some estuaries. Such variations
have been used to explain growth variations in bivalves in
different locations within the same estuary. Traditional
estimates of an estuarine system’s residence time can
be computed from measurements of system volume,
tidal prism, and water input to the system. The advent
of fast computers and numerical models, however, now
allows for much more modeling of these systems with the
potential for more sophisticated spatial and temporal
management strategies.
In riverine systems, river flow is the main physical
cause of material and organismic transport from estu
aries to the sea. Each of these systems are a unique and
changing feature on the present landscape because
rising sea level is drowning their basins and sediments
are gradually filling their channels. For example in
Chesapeake Bay, 35% of the particulate nitrogen and
most of the phosphorus is buried in the sediments of
the bay. Of the nitrogen in the bay water column, 31%
was exported to the sea and 8.9% was removed from
the system as commercial fish harvest. In general, the
nutrients transported and exported by riverine estuaries
are thought to be a significant source for generating
Clearance time (days)
1000
No regulation
Regulation
1
1
10
100
Residence time (days)
1000
Figure 4 A plot of water volume residence time versus bivalve
clearance time showing areas of potential control by suspersionfeeders.
251
new organic production in the coastal ocean. As many
of these systems have dams or have them proposed,
managers must take into account the direct and indirect
effects of these structures on recreational and coastal
fisheries.
In bar built estuaries, tides are usually the major
source of energy for the transport of materials into and
out of the estuary. If the fastest currents are on the flood
ing tides, then the system tends to import suspended
particulate material. In contrast, if ebbing tides have
the fastest currents, then the system usually exports
suspended particulate materials. The Wadden Sea of
Northern Europe is a flood dominated system and
North Inlet in South Carolina is an ebb dominated
system.
In shallow, high insolation, low precipitation, warm
systems, evaporation can dictate the direction of trans
port. This is the case in some small tropical systems where
water loss due to evaporation is replaced by the influx of
water and nutrients from the adjacent sea.
Organismic Transport
In addition to inanimate materials, the larval and adult
stages of many organisms are exchanged between the
estuary and the sea. Some organisms may be passively
carried by estuarine currents while others may actively
swim or take advantage of the direction of tidal flows to
move across the estuary–ocean interface.
Primary producers, including phytoplankton and resus
pended benthic microalge, depend on passive transport
between estuaries and the sea. Most flux studies show that
these organisms have a net seasonal or annual transport into
the estuary from the coastal ocean. This import has been
explained by passive filtration by estuarine wetlands and by
active filtration by suspension feeding animals within the
estuary. Protozoans, bacteria, and viruses are also found in
the estuarine water column and while they most certainly
are passively transported by estuarine currents, the direction
of their net flux is yet to be determined.
The exchange of invertebrate larvae between estu
aries and the coastal ocean has been explained by two
competing schools of thought, the passive and active
hypotheses. In the passive hypothesis, the horizontal
movements of larvae are mainly a function of current
direction and velocity. The active transport school con
tends that invertebrate larvae swim both vertically and
horizontally to take advantage of tidal currents. In one
group that includes oysters, the early stage larvae stay
high in the water column with later stages sinking to
lower depths. This strategy allows downstream move
ment of early larvae with some exiting the estuary to the
sea, while older larvae are entrained in inflowing bottom
currents and effectively retained in the estuary. A second
252
Estuaries
approach is used by larvae that migrate vertically in the
water column in synchrony with tidal cycles. This strat
egy allows larvae to maximize upstream transport and
retention. A third group has larvae that are immediately
transported to the coastal ocean where they stay for
weeks before returning into the estuary using wind and
tidal currents. A final group uses the coastal ocean dur
ing their adult and larval life. In this case, the postlarvae
enter the estuary maintaining their position by swim
ming against tidal currents.
Nekton organisms (fish, crabs, and shrimps) are mobile
links between the various subsystems of estuarine ecosys
tems as well as links between the estuary and the sea.
These animals feed and accumulate biomass while in the
estuary and then move back to the coastal ocean, thus
exporting biomass and inorganic wastes.
Global Climate
While seasonal and latitudinal climatic effects on coastal
and estuarine systems have long been documented, the
impacts of global climate change (warming or cooling) on
estuarine systems have only recently been quantified.
Major storms, El Nino Southern oscillation (ENSO)
events, seismic sea waves, or tsunamis and sea level rise
(SLR) are global effects that can significantly influence
water and material fluxes in estuaries.
Hurricanes and major storms generally influence estu
aries through storm surges and short term increases in
precipitation. These enormous pulsed fluxes of water can
change the geomorphology of estuaries and their water
sheds, massively resuspend sediments, and flush materials
off the landscape and into the estuary. Tsunamis can be
even larger than storm surges and can have similar impacts
to even greater areas of the coastal ocean and estuaries.
However, extensive marsh and mangrove wetlands com
mon to estuaries can buffer these pulses of water and
reduce the damage they can cause to the coastal landscape.
ENSO events only affect some estuaries. The effect is
usually a drought or higher than average precipitation.
For example in some South Carolina estuaries, ENSO
induced precipitation and upland runoff can depress sali
nity up to 75% for as much as 3 months.
SLR is an example of global change on both seasonal
and annual time scales that directly influences estuarine
systems. Seasonal changes in sea level are the result of air
pressure changes at the water’s surface and the expansion
or contraction of water mass due to heating and cooling. In
estuarine systems, these changes are reflected in the depth
of the system, but more importantly in the area and time of
exposure or submergence in the intertidal zone. SLR will
gradually force the transgression of estuaries upslope along
the coastal plain. Eventually, SLR will compete with
human development for the coastal landscape.
Estuarine Ecosystem Resilience and
Restoration
Estuarine ecosystems and subsystems can and do exhibit
alternate or multiple states of existence. The ability of an
ecosystem to absorb disturbance and resist a change in
state is termed ecological resilience, as opposed to engi
neering resilience, which is the time it takes a system to
return to its original state. In the last decades of the
twentieth century, ecologists observed that ecosystems
were not static entities, but appeared to change in
response to external and internal forces. In the
Chesapeake Bay estuary, for example, some of the factors
causing a state change were over fishing, increased sus
pended sediment load, eutrophication, species invasion,
and disease. The bay’s responses to these forces were slow
at first, but with the steady increase in the human popula
tion in the bay watershed and with its adherent
development, the signs of a state change were dramati
cally evident. The oyster reefs, a major benthic subsystem
or habitat that had dominated the bay for centuries, began
to decline rapidly or crash. The benthic dominated food
web was replaced by a planktonic food web. Management
efforts to restore the initial oyster dominated system did
not work, probably because they had a single species focus
and because ecosystems are strongly nonlinear which
means the path to restoration is different from that lead
ing to the initial change of state and many more
components of the ecosystem are involved in addition to
the oysters.
See also: Mangrove Wetlands; Salt Marshes.
Further Reading
Alongi DL (1998) Coastal Ecosystem Processes. Boca Raton, FL: CRC
Press.
Attrill MJ and Rundlle SD (2002) Ecotone or ecocline: Ecological
boundaries in estuaries. Estuarine, Coastal and Shelf Science
55: 929 936.
Dame RF, Childers D, and Koepfler E (1992) A geohydrologic continuum
theory for the spatial and temporal evolution of marsh estuarine
ecosystems. Netherlands Journal of Sea Research 30: 63 72.
Dame RF, Chrzanowski T, Bildstein K, et al. (1986) The outwelling
hypothesis and North Inlet, South Carolina. Marine Ecology Progress
Series 33: 217 229.
Dame RF and Prins TC (1998) Bivalve carrying capacity in coastal
ecosystems. Aquatic Ecology 31: 409 421.
Day J, Hall C, Kemp W, and Yanez Arabcibia A (1989) Estuarine
Ecology. New York: Wiley.
Gunderson LH and Pritchard L (2002) Resilience and the Behavior of
Large Scale Systems. Washington, DC: Island Press.
Lotze HK, Lenihan H, Bourque B, et al. (2006) Depletion, degradation,
and recovery potential of estuaries and coastal seas. Science
312: 1806 1809.
Mann KH (2000) Ecology of Coastal Waters, 2nd edn. Oxford: Blackwell
Science.
Floodplains
253
Floodplains
B G Lockaby, Auburn University, Auburn, AL, USA
W H Conner, Baruch Institute of Coastal Ecology and Forest Science, Georgetown, SC, USA
J Mitchell, Auburn University, Auburn, AL, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Geomorphic Origins
Hydrology
Biogeochemistry
Vegetation Community Structure and Composition
Adaptations of Floodplain Vegetation
Productivity
Anthropogenic Impacts
Africa
Asia
Australia
Europe
North America
South America
Summary
Further Reading
Introduction
climate and geomorphology of a landscape will define
the hydrology and initial chemistry of streams. Stream
characteristics determine the hydrology, soil characteris
tics, flora and fauna, and biogeochemistry of the
floodplain. In turn, biogeochemical feedback from flood
plains to streams helps define the environment seen by
aquatic flora and fauna. Thus, a strong interdependency
exists between aquatic and terrestrial components of
riparian ecotones.
It is critical to understand that land clearing and devel
opment, construction of dams and impoundments,
pollutant export, and other human activities constitute
major influences on streams and floodplains. In some
cases, these will override the original hydrology, biogeo
chemistry, and ecology. As an example, the original
hydrology of a riparian system could be dramatically
altered by the construction of bermed roadways that
cross streams without adequate provision for through
flow. Since hydrology is the primary driver of all flood
plain functions, corresponding changes in net primary
productivity (NPP), species composition of animal and
plant communities, and biogeochemistry could be
expected to follow.
Globally, floodplains may be of greater value to society
than any other ecosystem type. This is because of the
critical role that interactions between floodplains and
associated streams play in maintaining supplies of clean
water. While that role is conceptually simple, the pro
cesses which define interactions (i.e., floodplain functions)
in aquatic–terrestrial ecotones are exceedingly complex.
Consequently, it is necessary to develop some under
standing of the ecological mechanisms behind those
interactions in order to fully appreciate the importance
of floodplain ecosystems. To that end, the goal of this
article is to provide a first iteration overview of flood
plain form and function (Figure 1).
A key concept is that floodplains and associated
streams are both causes and reflections of the other’s
characteristics and functioning. As an example, the
Geomorphic Origins
Figure 1 Panoramic view of the Timpisque River and the Palo
Verde Marsh in Costa Rica.
Streams in steep topography tend to undergo continual
downcutting and, consequently, act as sources of fine and
coarse material with little to no opportunity for deposi
tion. Sediment loads are easily carried downstream
because the high gradient of the channel imparts suffi
cient energy for water to retain particles. In many cases, as
streams emerge from steeper terrain and move into flatter
areas such as coastal plains, the gradient of the channel
Surface 2
Surface
1
Floodplain
Sl
ou
gh
River
base level
Ri
dg
e
Fl
at
Sca
rp
Surface 2
Terrace
p
Hydrologic
floodplain
Scarp
Terrace
Sw
am
Uplands
Surface 3
Surface 3
Sl
ou
g
Ri h
dg
e
Fl
at
(a) Nonincised stream
Na
tu
ra
l le
ve
Fl
at
e
Floodplains
Ba
r
254
Terrace
Figure 3 Cross-sectional view of a floodplain topographic
positions. Adapted from Hodges 1998.
Bankfull channel
(b) Incised stream (early widening phase)
Surface 3
Surface 3
Terrace
ce 2
Surfa
Terrace
Surface 2
Surface
1
Scarp
rp
Sca
Terrace
Sca
rp
rp
Sca
Terrace
Hydrologic
floodplain
Incised, widening channel
(c) Incised stream (widening phase complete)
Surface 4
Surface 4
Ter 3
rac
e
rp
Surfa
Terrace
Surface
3
Terrace
Sca
rp
Sca
Hydrologic
floodplain
ce
arp
Sc
Surface
Surface
2 Surface 2
1
Scarp
Terrace
in gaining an accurate assessment of net changes. The
result of the spatial irregularities is a pattern of swales
and berms that generally runs parallel to the stream course.
The convex and concave microrelief may represent eleva
tion differentials of only a few centimeters. Nonetheless,
those minor differences have major importance in defining
soil environments for vegetation and in influencing the
extent of contact between floodwaters and the floodplain
surface. In many cases, the microtopography of major
floodplains is somewhat predictable (Figure 3) and, simi
larly, drives spatial patterns of species composition and
NPP of vegetation communities. However, changes in
microrelief may be much less apparent on some floodplains
due to either prolonged or infrequent flooding.
Hydrology
Bankfull channel
Figure 2 Terraces in (a) nonincised and (b and c) incised
streams. In Stream Corridor Restoration: Princples, Policies, and
Practices (10/98). Interagency Stream Restoration Working
Group (15 federal agencies) (FISRWG).
decreases and flows may spread and lose energy. This
promotes the occurrence of overbank flow and creation
of deposition surfaces or sediment sinks. However, a
floodplain may shift between being a sediment sink or
source depending on hydrologic changes induced by
climate, anthropogenic activities, or other influences.
Downcutting also occurs as older floodplains are aban
doned by streams and become terraces which resemble
stair steps in cross section (Figure 2).
Sedimentation occurs as particles settle during sheet
flow and is highly variable both temporally and spatially.
Sediment deposition or alluviation makes possible the
high soil fertility that is generally associated with many
floodplains although there are notable exceptions. Rates
of sediment accumulation vary markedly among flood
plains and, in the southeastern United States for example,
range from 1 to 6 mm yr 1.
Deposition and scouring may often occur simulta
neously on different portions of floodplains and at
different times in individual locations. Consequently, the
scale at which sedimentation is assessed is very important
Hydrology is the foremost determinant of vegetation
species occurrence, NPP, biogeochemistry, floral and
faunal habitat, and all floodplain functions and traits.
Consequently, any insights into the nature of floodplain
ecosystems and the basis of their societal value are pre
dicated upon an understanding of hydrology. The ‘flood
pulse’ concept provides one framework within which to
develop this understanding.
In this concept, the river and floodplain are considered
as a single system and the ‘rhythm of the pulse’ (i.e., the
hydroperiod) is the controlling mechanism which regu
lates exchange of energy and material between the river
and floodplain. An influx of sediment and nutrients and
export of organic carbon from the floodplain will occur at
intervals dependent on the pulse rhythm. Examples of
common rhythms include single, long duration and multi
ple, short duration which might be stereotypic of high
order river floodplains and low order headwater streams,
respectively.
In general, flood frequency and duration may decrease
and increase, respectively, as stream order rises. When
headwaters originate in mountainous terrain, narrow
V shaped valleys form and hydroperiods may be charac
terized as flashy (i.e., frequent flooding, with sharp rises
and drops associated with stage levels). Hydroperiods
reflect the integration of rainfall patterns, water storage
capacities, and many other factors across the associated
Floodplains
catchments. Consequently, stage level rises and falls are
slower due to the ‘buffering’ that is provided by high
storage capacities and the greater variability of other
factors. Conversely, small catchments have much less
storage capacity and, consequently, streams respond
rapidly to precipitation events. As a result, floodplains of
large rivers can stay flooded for significant portions of a
year while low order floodplains may be inundated fre
quently but for much shorter periods.
Interchange of water between floodplains and rivers is
very complex and involves mutualistic influences. The
nuances of those interactions form the basis of the role of
floodplains as ecotones and regulators of energy and
nutrient exchange. At low stage levels, water within
swales and depressions may have originated with the
river, precipitation, an upwelling of groundwater, or
some combination. From a biogeochemical standpoint,
the origin is significant in terms of the degree of spatial
and temporal contact with the floodplain. At low stage
levels, there is less opportunity for river water to contact
the floodplain and, consequently, biogeochemical and
dissolved organic carbon exchanges are minimal. As
stage levels rise, the potential for the floodplain to influ
ence the biogeochemistry of sheetflow increases as well.
However, at some point, increasing floodwater volumes
and higher velocities reduce contact with the floodplain.
This is because a decreasing proportion of the sheetflow
volume is in contact with the floodplain as volumes
increase. Similarly, temporal contact is reduced as sheet
flow velocities rise.
There is also significant interaction between the river
and floodplain in terms of groundwater. Channel waters
often generate a head pressure which declines with dis
tance from the stream bank. Groundwater transmittance
will decline as hydraulic conductivity of alluvium
decreases (e.g., clays have reduced conductance compared
to sands). In humid regions, groundwater near the channel
moves under pressure and will contact and mix with
water that has seeped into the alluvium from adjacent
uplands. As a result, groundwater mixing can be quite
active during periods of low evapotranspiration.
Biogeochemistry
Once considered purely as nutrient sinks, floodplains are
now known to play multiple roles from a geochemical
perspective. Based on the type of floodplain, associated
vegetation, and the degree and nature of disturbance,
floodplains may also serve as sources or transformation
zones for nutrients. The widely held perception of flood
plains as fertility hot spots belies the complexity
associated with input–output budgets as well as the bio
geochemical processes within the floodplain ecosystem.
In particular, the impact of hydroperiod on
255
biogeochemical processes sets floodplain biogeochemistry
apart from that of non wetland ecosystems. Periodic
flooding makes possible nutrient exchange across the
aquatic–terrestrial ecotone and controls the nature of
decomposition, nutrient uptake and release by vegetation,
and many other processes. As an example, the process of
denitrification or the anaerobic conversion of nitrate to
gaseous forms of nitrogen is very important on flood
plains. In addition, the interaction of hydrology and
biogeochemistry necessitates the development of unique
approaches to the study of nutrient cycling in these
ecosystems.
As previously mentioned, floodplains may serve as
sinks, sources, or transformation zones for geochemical
inputs of nutrients derived from inflow, precipitation,
nitrogen fixation, and soil weathering. Multiple roles
may proceed simultaneously on the same floodplain if
spatial heterogeneity in hydrology, vegetation, distur
bance, and nutrient influx so dictate. The use of a
geochemical budget allows net inputs to be compared to
net outputs and is based on the perspective of the ecosys
tem as an integrated system.
In general, the factors that promote nutrient sink activ
ity on floodplains include (1) presence of aggrading
vegetation; (2) wide carbon: nutrient ratios in living vege
tation and detritus; (3) topographic positions conducive to
somewhat frequent, short duration, and low energy flood
ing; (4) basin geomorphology that promotes significant
sediment loads in streams (e.g., redwater, brownwater, or
whitewater based on the color of suspended clay); (5) high
occurrence of nitrogen fixers; and (6) until nutrient
saturation is approached, association with a river sub
jected to high anthropogenic nutrient loadings.
Alternatively, rivers draining low gradient basins with
sandy soils are often referred to as blackwater systems
because their waters are stained with organic substances
(Figure 4). These tend to carry low sediment loads and,
consequently, alluviation (i.e., sink activity) is less pro
nounced. Also, floodplains occupied by mature vegetation
communities may act as transformers of nutrients (e.g.,
inorganic inputs of nitrogen converted to organic outputs)
rather than a sink or source. The latter is a key facet of the
‘kidney function’ of these systems and has great signifi
cance for maintenance of water quality. Sink activity, such
as the filtration and accumulation of sediments (and asso
ciated nutrients) from sheetflow also plays a major role in
cleansing water (Figure 5). Finally, floodplains that have
been altered in some way by disturbance may function as
nutrient sources. The longevity of the source activity
could be short term (e.g., a well planned forest harvest
followed by rapid forest regeneration) or long term (e.g.,
conversion to agricultural or urban uses, impoundments,
or climate change).
Similarly, all biogeochemical processes within flood
plain ecosystems reflect the overriding influence of
256
Floodplains
(a)
(b)
Figure 4 Amazon River: (a) a broad and (b) a close-up view. The
formation of the Amazon River at the ‘o encontros das aquas’ or
mixing of the Rio Negro and Rio Solimoes near Manaus, Brazil.
The blackwater Rio Negro is contrasted with the sediment-laden
Rio Solimoes.
Figure 5 Flint River – sediment accumulation on the Flint River
floodplain near Ft. Valley, GA during floodwater drawdown.
hydrology. As an example, the timing of litterfall is heav
ily affected by hydroperiod because different vegetation
communities occur under different hydrologic regimes. In
the southeastern United States, forest species associated
with Nyssa may grow under wetter conditions than com
munities dominated by some species of Quercus. On wetter
sites, Nyssa foliage tends to senesce earlier in the autumn
than other floodplain tree species and, consequently, the
senesced foliage is exposed to a different microenviron
ment than litter that falls later in the year. As a result,
nutrient release and immobilization sequences are likely
to differ among sites.
Mass loss and nutrient dynamics during decomposition
are a function of both litter quality and the decomposition
microenvironment. Litter quality (the biochemical com
position of detritus) is defined by the conditions under
which a plant is growing as well as genetics and has been
shown to be closely linked to variation in hydroperiod.
Also, the frequency and duration of flooding play a dom
inate role in determining biomass and composition of
microbial populations. Key determinants of shifts between
nutrient mineralization and immobilization include
hydroperiod and nutrient inflow. In the southeastern
United States, mass loss rates of foliar litter (with litter
quality held constant) are maximized by moderate dura
tions of flooding followed by several months of
noninundation.
In general, rates of litter mass loss in forested flood
plains exceed those of uplands. Globally, decay constants
for temperate floodplain forests average approximately
1.00 while the mean for all temperate deciduous forests
is less than 0.80. This differential is partly due to
the greater availability of soil moisture (better habitat
for microbial populations) during parts of the year.
However, mass loss, as measured by disappearance of
confined litter, includes both mechanical disintegration
as well as metabolic conversion of organic carbon and,
consequently, periodic inundation offers greater opportu
nities for disintegration and export.
The general perception that floodplains are very fer
tile has led to misconceptions regarding the degree to
which insufficient nutrient availability may constrain
floodplain NPP. In many cases, it is true that floodplain
soils are more fertile than upland counterparts. However,
vegetation species found in many floodplains often have
higher annual nutrient requirements compared to species
adapted to uplands. Consequently, forest vegetation on
many floodplains is likely to be nitrogen deficient and, in
some cases such as blackwater systems, deficient in phos
phorus and base cations as well. An example would be the
nutrient demanding Populus deltoides Batr. plantations that
grow in extraordinarily fertile soils of the Southern
Mississippi Alluvial Valley, USA. In spite of fertile soils
and high aboveground NPP (20–25 t ha 1 yr 1), those
Floodplains
systems would increase in NPP if supplied with addi
tional nitrogen.
The degree to which a floodplain ecosystem is defi
cient or nondeficient for particular nutrients is critical in
regard to that system’s potential to act as a nutrient sink.
As previously mentioned, the kidney function is
enhanced if floodplain vegetation can assimilate incoming
nutrients from sources such as polluted water or atmos
pheric inputs. Once a deficiency is eliminated, it is still
possible for floodplain vegetation to assimilate particular
nutrients such as nitrogen through luxury consumption.
However, a level may be reached after which the vegeta
tion’s capacity to retain nutrients is saturated. The latter
condition reflects a high degree of biotic stress and is a
serious threat to floodplain vegetation associated with
eutrophic streams.
Vegetation Community Structure and
Composition
Vegetation communities in floodplain systems have
developed over hundreds of years as a function of soil
type, topography, and hydrology. The type of vegetation
growing on a particular floodplain will be dominated by
trees or shrubs adapted to the environmental conditions
of that floodplain. Hydroperiod is the most important
local environmental condition determining composition,
and the species found respond to elevation differences
relative to the river’s flooding regime. Typical floodplain
forests begin at the natural levee where coarse grained
deposits result in quickly draining soils and continue as
surface elevations decrease away from the river and
become more poorly drained.
Structural characteristics of floodplain forests vary
depending upon location (Table 1). Stem density and
basal area are generally greater in the southeastern
257
United States and the humid tropics than in arid areas,
but in arid areas basal area can still exceed 50 m2 ha 1.
Basal areas in floodplain forests tend to be as high as or
higher than that of upland forests. Almost without excep
tion, the number of tree species increases as flooding
decreases. The greatest number of tree species occurs
in wet, tropical floodplains such as the Amazon. The
understory of floodplain forests is generally lower in
density and species numbers, probably due to reduced
light levels and the extended flooding conditions.
Adaptations of Floodplain Vegetation
Due to the alternating wet–dry environment experienced
by trees growing on floodplains, they have developed a
variety of physiological and morphological adaptations
that allow survival during flooding. Initially, stimulation
of alcohol dehydrogenase (ADH), enzyme activity may
provide a temporary means to support essential metabolic
functions. The anaerobic pathway is less efficient than the
aerobic pathway (39 moles ATP per mole hexose vs.
3 moles ATP per mole hexose), but provides an energy
resource while anatomical changes are occurring.
The seeds of floodplain tree species require oxygen for
germination, and even those species that can grow in
permanently to nearly permanent flooded conditions
(e.g., Taxodium and Nyssa) require moist, but not flooded,
soil for germination and establishment. Occasional draw
downs are necessary for the survival of tree species. Rapid
stem elongation, such as been observed with Nyssa aqua
tica, allows the seedling to get its crown above the water
surface of subsequent floods. The dispersal and survival of
many wetland tree seeds is dependent upon hydrologic
conditions. Taxodium and Nyssa seeds are produced in the
fall and winter between the periods of lowest and highest
streamflows, giving the seeds the widest possible range of
Table 1 Mean structural and aboveground productivity characteristics of floodplain forests
Aboveground NPP
Area
Southeastern USA
Northeastern USA
North Central USA
Western USA
Central USA
Europe
Central America
Caribbean
South America
Africa
Southeast Asia
Australia
a
No. of
species
Density
(no. ha 1)
Basal area
(m2 ha 1)
Biomass
(t ha 1)
13
10
5
5
12
1242
970
546
310
405
1237
726
3359
687
45.0
26.1
29.5
27.5
33.5
26.5
49.9
42.4
33.0
302
150
5.36
7.78
13.26
290
314
118
224
413
4.20
3.48
11.61
15.55
2.50
17.88
8.70
10
27
89
26
Leaf
9.15
12
493
Total NPP does not always equal leaf plus wood as some sources only report total.
260
Wood (t ha yr 1)
Total a
258
Floodplains
hydrologic conditions. Overall, seed production of many
wetland species seems to be linked to the timing and
magnitude of hydrologic events.
Stem hypertrophy, commonly called butt swell or but
tressing, is characterized by an increase in diameter of the
basal portion of the stem and is common in Taxodium,
Fraxinus, Nyssa, and Pinus species. Basal swelling can
extend from just above the ground level to several meters
depending upon the depth and duration of flooding.
Swelling generally occurs along that portion of the trunk
that is flooded seasonally. Increased air space in the swollen
portion of the stem allows increased movement of gases
within the plant. Ethylene production has been documen
ted to play a regulatory role in altering growth and stem
anatomy of woody plants, and has been found to be higher
in flooded Fraxinus stems with well defined hypertrophy
than those without stem hypertrophy. Lenticel hypertro
phy has long been associated with flooding and acts to
increase internal gas transport from the stem to the roots.
Duration of flooding does not appear to affect the number
of lenticels formed but does affect the size. The formation
of hypertrophied lenticels under anoxic conditions also
appears to be induced by ethylene. Other commonly
observed features in flooded environments include buttress
roots and knees. Buttress roots appear as fluted projections
at the base of mature trees and extend for several feet from
the trunk outward and down into the soil. Because of the
shallow nature of root systems in saturated or flooded soils,
these buttress roots are thought to provide additional sup
port to the tree. Knees are common in Taxodium spp. in the
southeastern United States. Their function has not been
confirmed, although there is some speculation that they
also serve in stability of the tree. In Australia, Melaleuca
trees on floodplain sites have modified bark structures such
as papery bark with internal longitudinal air passages that
allow them to tolerate flooded conditions.
seasonal flooding can be both a subsidy and a stress. In the
southeastern United States, aboveground NPP was similar
for upland hardwood, bottomland hardwood, and Taxodium–
Nyssa forests. The reason for this may be that for some sites,
subsidies and stresses occur simultaneously and cancel one
another. As a result, flood intensity and duration affect soil
moisture, available nutrients, anaerobiosis, and even length
of growing season in a complex and nonlinear ‘push–pull’
arrangement. When hydrology is altered rapidly, above
ground productivity is less than in natural forest
communities with nearly continuous flooding (Figure 6).
Aboveground biomass in floodplain forests ranges
between 100 and 300 t ha 1, although there is one report
of a forest in Florida where biomass exceeds 600 t ha 1.
Leaves account for only 1–10% of the total aboveground
biomass. Belowground biomass has been sampled rarely
and varies greatly, but reported values tend to be somewhat
lower than the 20% of total biomass often cited for upland
species. Total aboveground biomass production (leaves
plus stem wood) ranges from 668 to 2136 g m 2 yr 1,
with leaves accounting for approximately 47% of the
production. Although it has been reported that there are
no latitudinal patterns in NPP, litterfall production of
Taxodium forests in the United States shows a curvilinear
relationship with latitude with a maximum occurring at
about 31.9 N. In northern Australia, litterfall in Melaleuca
forests has been reported to be 2–3 times greater than that
in forests in the southern part of the continent. Changes to
natural hydrologic regimes decrease litter production by
half. As a result of the high productivity, generally
associated with floodplain forests, carbon sequestration
is particularly important there.
1600
Riverine floodplains are typically characterized by high
productivity. Productivity is enhanced in many flood
plain areas by the continued import and retention of
nutrient rich sediments from headwater regions and lat
eral sources, increased water supply (especially in arid
regions), and more oxygenated root zones as a result of
flowing waters. The flood pulse advantage has long been
recognized, with ancient Egyptians setting taxes based on
the extent of the annual flood.
Primary productivity of unaltered, seasonally flooded
ecosystems is generally higher than that of floodplain forests
that are permanently flooded or those with stagnant waters.
Despite the theoretical basis for increased floodplain pro
ductivity due to pulsing, it has been difficult to confirm.
More recent studies tend to point toward the idea that
Aboveground NPP (g m–2 yr–1)
Productivity
–100.00
1200
800
400
0
–50.00
0.00
50.00
Mean growing season water depth (cm)
100.00
Figure 6 Relationship between aboveground net primary
productivity (NPP) of floodplain forests of the southeastern United
States and mean water depth during the growing season.
Floodplains
Anthropogenic Impacts
Rivers and associated floodplains have been vitally linked
to civilization throughout history for food production. In
order to make farming easier and more productive, rivers
have been diverted and floodplains have been deforested
and drained or leveed to provide fertile land. A major
consequence of the widespread use of floodplains and
adjacent uplands for agriculture has been the generation
of large sediment loads in associated streams and
rivers. As a result, much sediment has been deposited on
streambeds and floodplains with negative consequences
for aquatic habitat and floodplain vegetation. More
recently, impoundments have become commonplace for
energy production and water storage and levees continue
to be built to provide space for development as well as for
farming. Globally, it is estimated that, at a minimum, 75%
of total floodplain area has been lost.
Floodplain function is dependent on connectivity
between the river and its riparian area. Unfortunately,
many anthropogenic impacts eliminate or reduce that
connectivity so that key functions such as water filtration
are much reduced at the landscape level. Similarly, altera
tions in hydroperiod caused by human activity often drive
changes in composition and productivity of vegetation
communities as those species adapted to the former con
ditions decline and are replaced by others.
Additional impacts include fragmentation of riparian
vegetation communities and stimulation of invasive non
native plant invasion. Fragmentation often results in
reduced habitat quality while successful invasion by
non native species may cause major alterations in com
munity composition, structure, and function. While
ecological restoration of floodplains has attracted wide
spread interest, economic constraints have primarily
limited restoration applications to localized areas.
However, notable exceptions include restoration of the
portions of the Pantanal River Basin in South America
and the Kissimmee River Corridor in Florida, USA.
More recently, urbanization has led to significant and
growing impacts on floodplains in many parts of the
world. As catchments become developed, the concomitant
rise in impervious surface drives major increases in runoff
volume and velocity. As a result, rising limbs during flood
events become much steeper, a condition that is often
associated with higher in channel velocities. Higher
flow velocity increases the rate of channel incision result
ing in a lowered groundwater table and reduced
connectivity between the stream and floodplain. In addi
tion, urbanization stimulates loadings of nutrients
(particularly, nitrogen) and causes a considerable degree
of water pollution in general.
Further anthropogenic impacts include channelization
of river systems. Channelization has benefited farming
259
and waterborne transportation by reducing flooding and
removing obstacles to barge and other water traffic.
However, water quality has suffered in many instances
since there is again less opportunity for river waters to
contact floodplain surfaces and undergo pollutant
reduction.
Africa
The African continent has approximately 99 large
wetlands, of which 43 are floodplain systems. Some of
the larger floodplain systems include the Zaire Swamps
(200 000 km2), the Inner Niger Delta of Mali (320 000 km2
when flooded), the Sudd of the Upper Nile (16 500 km2 of
permanent swamp and 15 000 km2 of seasonal floodplain),
and the Okavango (14 000 km2 of permanent swamp and
14 000 km2 of seasonal floodplain). These floodplain sys
tems are in dynamic equilibrium with the constant flux of
pulsing events occurring within them at different spatial
and temporal scales. Goods and services resulting from
pulsing events include floodplain recession agriculture,
fish production, wildlife habitat, livestock grazing, eco
tourism, and biodiversity, as well as natural products and
medicine.
In semiarid and arid regions of Africa, floodplains are
often the only source of year round water. As in other
floodplains around the world, vegetation distribution is
strongly related to flooding frequency and duration and
microtopography. Dense evergreen tree growth occurs on
higher well drained areas like levees and termite mounds,
while grasslands tend to dominate lower, more frequently
flooded areas. Typical grasses found growing in these fre
quently flooded areas (called swamp) include Phragmites,
Typha, and Polygonum. Tree and bush genera in less fre
quently flooded areas include Hyphanene, Borassus, Acacia,
Ficus, and Kigelia.
Floodplain areas are centers of high diversity of animal
and plant life. These floodplain areas are of profound
importance for fish production and probably serve as
spawning and recruitment areas. Interannual fluctuations
in fish production have been correlated with the flooding
regime. Numerous bird species (over 400 in some flood
plains) can be found in these areas, including bee eaters,
jacanas, malachite kingfishers, grey herons, egrets, African
fish eagles, and Zaire peacocks. The birds share the flood
plains with antelope (sitatunga, waterbuck, puku, and
lechwe), hippopotamus, zebra, and buffalo; vegetation
ranges from water lilies and papyrus to floodplain forests
with minor topographical variations playing an important
role in distribution of forest and grassland. Climatic vari
ations are also important, with forest only occurring near
rivers in drier areas while in wetter areas forests can
extend for a considerable distance away from the river.
260
Floodplains
River meanders tend to be cut off during flooding periods,
adding diversity to the floodplain topography.
Unfortunately, very few studies of the ecology of many
of the African floodplain systems have been carried out.
The most studied floodplain system in Africa is the
Okavango Delta. Annual floods travel uninterrupted
down the Okavango River and inundate the Okavango
Delta from April to September. River water is character
ized by moderate levels of nutrients, but when it enters
the floodplain it becomes strongly enriched by nutrients
via leaching from soil, detritus, and feces. Organic carbon
enrichment comes from leaching of floodplain leaf litter
and soil, although dissolved organic carbon release from
leaf litter is over 2 orders of magnitude greater than for
leached soils. This nutrient enrichment has a major
impact on aquatic productivity in the delta and illustrates
the strong links between terrestrial and aquatic
ecosystems.
African floodplains face a different set of challenges as
opposed to those in developed countries. In Africa, flood
plains generally occur in semiarid areas to arid regions,
and flooding is the driving force behind the high produc
tivity of these areas. From as early as the 900s, people
have inhabited these areas, and pastoral and agricultural
economies are dependent upon the continued presence of
the floodplains. Continued pressures from agricultural
practices within the floodplains themselves and popula
tion growth that demands the transfer of water to alleviate
shortages outside of the floodplain need to be addressed
to ensure survival of these important ecosystems.
Asia
In northern Asia, there are extensive productive wetlands
along the floodplains of rivers. In western Siberia, the
river Ob extends over 50 000 km2 and supports what is
called the largest waterfowl breeding and moulting area
in Euroasia. The Ob Valley is a labyrinth of intricately
arranged channels and floodplain lakes. As in other sea
sonal floodplains, the region is a land of fluctuating water
levels, with seasonal and annual fluctuations in river dis
charge and flooding patterns. This area avoided any
serious human impacts for centuries, but oil and gas
exploration has resulted in significant pollution and trans
formation of the landscape.
The Indus River has long been the lifeblood of arid
Pakistan. In earlier times, people used the river’s water to
cultivate the floodplain, but during the last 100 years,
the river has been dammed and diverted into one of the
largest and most complex irrigation systems in the world.
In the absence of a drainage system to remove irrigation
water, evaporation leaves salt in the soil. As a result of this
salinization of the soil, combined with waterlogging, over
400 km2 of irrigated land is lost each year.
Many of the large river systems in South Asia display
considerable annual variation in discharge and, during the
rainy season, may flood very large expanses of land (e.g.,
approximately 200 km on each side of the Ganges). In
some cases, entire deltaic areas may be inundated. The
prolonged, monomodal flooding promotes extensive
spatial and temporal contact with floodplains and, conse
quently, dominates the socioeconomics of the large
human populations near those systems. Agricultural
activities often cause significant sediment export from
upper reaches of many rivers and, as a result, delta tribu
taries may become clogged. Due to the subsequent
reduced flow, salinity can increase in soils and alter spe
cies composition in the delta forests.
About 80% of Bangladesh (115 000 km2) is formed by
floodplains of the Ganges, Brahmaputra, and Meghna
rivers. In major floods, 57% of the country can be flooded.
Availability of water during the dry season makes it
possible to grow three crops a year in some areas.
Deposition of waterborne sediments keeps the soils fertile
and algal growth enriches the soil by fixing nitrogen. As in
many parts of the world, forest vegetation of South Asian
floodplains strongly reflects variations in hydroperiod and
soil. In the Ganges and Brahmaputra River Valleys,
within areas with heavy clay soils where flooding occurs
for most of the year or permanently, forest vegetation
may be only 5–10 m in height and occur in conjunction
with numerous vines. However, combinations of similar
flooding regimes and lighter, fertile soils may increase
canopy heights by 10 m or more. Many of these riverine
forests exhibit a prevalence of evergreen or semiever
green species, although at higher altitudes alders may
dominate. Some lowlands, in particular many river deltas
such as those of the Ganges–Brahmaputra and Irrawady,
are occupied by mangrove forests.
In China, 95% of the population is concentrated in the
eastern half of the country, mainly in the vast alluvial
plains of the major rivers, the Yellow and Yangtze Rivers
primarily. High population densities coupled with high
growth rates, rapid urbanization, and industrialization
play a major role in most Asian countries. Water resources
in this region are under increasing pressure as the demand
for domestic supplies, agricultural use, and hydroelectric
power increases. Past water resource and agricultural
management practices have resulted in rapid loss and
degradation of natural wetlands throughout the region.
The regulation of rivers and streams through embank
ments and dams has eliminated floodplains and reduced
groundwater recharge. Changing hydrological regimes
have increased flooding during the rainy season and
reduced availability during dry periods. Water resource
management has often resulted in numerous man made
wetlands such as reservoirs and paddy fields that have
very different functions and values than natural wetlands,
and are in no way a substitute for natural wetlands,
Floodplains
261
particularly floodplain wetlands. In short naturally occur
ring floodplains in these regions are threatened by
numerous human activities, including mining, aqua
culture, unsustainable forestry or fisheries practices, and
conversion of forests to urban or agricultural land.
many floodplains into terrestrial ecosystems. The effect of
this change in flooding has not been well studied and data
exist only for a fraction of the area affected. Floodplain
loss will continue until there is a better understanding of
the long term ecological effects of dams and diversions.
Australia
Europe
Australia is distinctive in that there are few permanent
wetlands due to high evaporation rates and low rainfall.
Most wetlands on the continent are intermittent and seaso
nal. Common features of floodplains are waterholes and
lagoons called billabongs that retain water seasonally or
permanently, providing important habitat for many animals
at different times of the year. Floodplain wetlands tend to
be sites of extraordinary biological diversity of waterbirds,
native fish, invertebrate species, aquatic plants, and
microbes. Key drivers of this biodiversity are the lateral
connectivity to the river of the floodplain wetland and the
unpredictable flows that create wide ranges of temporally
and spatially different aquatic ecosystems.
Humid coastal areas are drained by short, perennial
streams, while much of the streamflow in the rest of the
country is intermittent or nonexistent because of low and
unreliable rainfall, high evaporation, and flat topography.
Even under these conditions, forested wetlands can be
found throughout Australia, but they can only be classed
as true forests in the wettest localities. The largest area of
floodplain forested wetland (over 60 000 ha) occurs on the
Murray River. Floodplain forests are generally composed of
Melaleuca or Eucalyptus species, but they cannot survive very
long periods (>5 months) of flooding. If flooding exceeds
several weeks during the growing season, forest canopy
cover declines to between 10% and 70%, creating open
woodlands.
Tropical floodplain wetlands are found across northern
Australia, covering an estimated 98 700 km2. Vegetation of
these wetlands has been mapped at various scales, but
there are few specific or long term analyses of the distri
bution or successional changes of the plants. The Ord
River floodplain in northern Australia encompasses
approximately 102 000 ha and is a large system of river,
tidal mudflat and floodplain wetlands that supports exten
sive stands of mangroves, large numbers of waterbirds, and
significant numbers of saltwater crocodiles. In southeast
ern Australia, the Murray–Darling river system drains
14% of the continent and contains the greatest amount
of floodplain wetlands on the continent.
In recent years, floodplain areas have undergone con
siderable change because of animal (buffalo, pigs, cane
toads) and plant (mimosa, salvinia, paragrass) invasions,
changes in fire regimes, water resource management, and
saline intrusion. Dams and the cumulative impact of
diversions and upstream river management have turned
Floodplains in Europe have been influenced by humans
for thousands of years. Civilizations often were estab
lished near rivers and frequently utilized floodplain
resources for food (agriculture or hunting), power (wood
or water mills), and shelter. As communities grew there
was an increased need to control the flooding that natu
rally occurred in the floodplains with the use of dams,
dikes, and ditching. These structures altered the hydrol
ogy which in turn has altered the forest composition in
these areas. Furthermore, channel straightening has
caused major hydrologic changes resulting from faster
flow and increased groundwater depth. In some of the
Danube watersheds, there has been an 80% decrease in
first order streams from 1780 to 1980.
In many areas, depth to groundwater has increased due
to the ‘drying’ of the floodplains and this has driven shifts
in the composition of vegetation communities. In parti
cular, species such as Quercus robur, Fraxinus spp., and
Ulmus spp. are becoming rarer due to the altered hydrol
ogy. Forestry practices have induced a further shift from
natural systems to faster growing Populus clones in many
of the floodplains across central Europe. However, rees
tablishment of the more traditional forest composition of
uneven aged oak, ash, and maple mixes has been achieved
in some areas as recently as the past 50 years. Large
portions of the forests remain monocultures of even
aged Acer or Fraxinus.
The Danube Delta represents one of the largest wet
lands in Europe and is undergoing eutrophication as a
result of increasing nutrient inputs, decreased riparian
vegetation, and loss of the filtration function. One major
difference between European floodplains and others
worldwide is that increased flow and flooding often occurs
in the spring as a result of snowmelt in high altitudes.
North America
In the dry climate of the Western United States, water is a
limited resource not only for the wildlife but also for the
human inhabitants. Although wetland areas comprise a
very small portion of total land area (i.e., less than 2%),
over 80% of wildlife is dependent on their presence.
Rainfall in this region varies from less than 15 cm yr 1
in the desert regions to greater than 140 cm in the moun
tains. In the mountainous regions, rainfall and snowmelt
262
Floodplains
are greater than losses and, therefore, wetlands rarely dry
out. However, evapotranspiration in the basin areas is 3–4
times greater than precipitation and, consequently, soil
salinization is a stress to which vegetation must adapt. In
the driest regions soil salinization prevents vegetative
establishment. Ephemeral drains are prevalent in the
intermountain west with snow melt and high rain con
tributing to their flow.
At higher elevations in the United States, where soils are
semipermanently inundated or saturated, associations of
Populus, Salix, and Acer are found. Floodplain areas flooded
or saturated 1–2 months during the growing season are
comprised of a wide array of hardwood trees. Common
species in the United States include Fraxinus spp., Tilia
spp., Ulmus spp., Liquidambar spp., Celtis spp., Acer spp.,
Plantanus spp., and some Quercus spp. At the highest eleva
tions, flooding occurs for less than a week to about a month
during the growing season. Typical tree species include a
variety of Quercus spp. and Carya spp., with some Pinus spp.
Floodplains of the southeastern United States occur
within three physiographic regions: (1) coastal plains,
(2) piedmont, or (3) Appalachian Mountains. Rainfall is
sufficiently prevalent during all seasons except for brief
periods of drought. Successional patterns of southern
forested floodplains are often dictated by hurricanes, tor
nados, catastrophic ice storms, and extended drought.
Soils are typically acidic, with the exception of near
neutral pH soils across much of the Southern
Mississippi Alluvial Floodplain and the Selma Chalk
geologic region of Alabama and Mississippi. In many
floodplains, as one moves in a direction perpendicular to
the river, soil textures range from coarse sands near
stream channels, fine sands in natural levees, to loams
and clays in backwater areas. This separation pattern is
a result of particle size and sheet flow velocity.
The lowest elevation, nearly always flooded sites on
floodplains in the southeastern United States are occu
pied by Taxodium–Nyssa swamps. In other parts of the
world, it appears there are no similar tree species that
can survive permanent or long periods of inundation. As
long as the floodplain channel remains stable and flooding
frequency remains constant, these species should domi
nate the stands indefinitely.
South America
Much of our current knowledge about forested floodplains
has been derived from extensive studies performed in the
sub basins of the Amazon River. In particular, our under
standing of floodplain biogeochemistry, NPP, vegetation
dynamics, geomorphology, and faunal relationships has
been greatly influenced by Amazonian research.
In comparison to river basins in other parts of the
world, the water balance of Amazonia lowlands is roughly
evenly divided between evapotranspiration and runoff.
This contrasts with systems in Asia where runoff domi
nates due to generally steeper terrain and many African
systems where broad floodplains and high potential evapo
transpiration result in low runoff. Floodplain forests in
South America are typically composed of a small number
of fast growing, early successional species capable of
surviving periodic floods and large amounts of sediment
deposition (e.g., Salix and Inga spp.).
The ‘flood pulse’ concept was originally conceptua
lized in relation to the Amazon and similar floodplains
and can be applied worldwide. The major river flood
plains of South America such as the Amazon, Orinoco,
and the Parana display singular, river borne flood pulses
of large amplitude and duration. In contrast, inundation
on floodplains situated within large depressions such as
the Pantanal is generally rainfed (as opposed to overbank
flow from rivers), and also displays a singular periodicity
but with lower amplitude. Finally, multiple flood pulses
that are less predictable in terms of occurrence and ampli
tude are characteristic of floodplains associated with
smaller order streams.
Some of the classic research that defined global vari
ation among floodplains took place in Amazonia and was
associated with contrasts between blackwater versus
brownwater or whitewater rivers. Similar types of flood
plain systems occur in many parts of the world. The color
of the river waters is reflective of the geomorphology
of particular systems and is a strong indicator of flood
plain biogeochemistry, vegetation dynamics, and NPP.
Whitewater rivers in the Amazon Basin derive their
color from white clay sediments that originate in the
Andes. The suspended clays contain higher levels of
nutrients (particularly base cations) which, when depos
ited, often create fertile floodplains labeled varzea.
In contrast, blackwaters are stained by fulvic acids and
other organic compounds and are more acidic than white
water counterparts (pH <5.0 vs. >6.0 for blackwater and
whitewater, respectively). Due to the low sediment loads,
floodplains associated with blackwater streams are often
nutrient poor and are referred to as igapo. Consequently,
forest litterfall production on varzea floodplains is
often considerably higher than that of the igapo (approxi
mately 10 vs. 5 t ha 1 yr 1 for the respective system
types). Also, the standing crop of fine roots is much higher
in igapo soils compared to varzea, a reflection of greater
belowground allocation of biomass as would be expected
in resource poor soils. Such adaptations increase the like
lihood of nutrient capture from decomposing igapo litter.
The distinctions in hydrology and biogeochemistry
between the igapo and varzea also drive major differences
in vegetation species occurrence, root, shoot, and repro
ductive phenology, and community structure.
Distinctions between floodplains types are also impor
tant in regard to animal populations. This is particularly
Floodplains
true for fish which depend on interactions with inundated
floodplains for resource acquisition, reproductive habi
tats, and other factors. As an example, the lower NPP on
igapo floodplains may translate to lower food resources
for fish. While the amount of plant detritus exported from
varzea floodplains is higher, phytoplankton production
also depends on settling of the clay sediments so that
sufficient light can penetrate the waters. Although more
difficult to document in riverine systems, fish catches are
generally much lower in igapo lakes compared to varzea
counterparts.
As is the case in much of the world, South American
floodplain ecosystems are under pressure from an array of
human activities. As an example, the lower reaches of the
Parana’ River have undergone changes in hydrology due
to construction of dams and upstream expansion of agri
culture. The altered hydrology, along with increased
concentrations of sediment and other contaminants have
resulted in heavy impacts to fish populations and conco
mitant economic declines in local fishing communities.
Although there is a growing voice for conservation and
protection of natural resources, it is unclear to what
extent anthropogenic impacts may be curtailed.
Summary
As pathways between aquatic and terrestrial ecosystems,
floodplains perform a myriad of functions that are critical
to humanity and all other components of the biosphere.
Because of the vital need of all organisms for clean water,
the kidney or filtration function is the most important
attribute of healthy floodplain systems.
The filtration function entails sediment and nutrient
deposition and, consequently, has long made floodplains
very attractive for exploitation as agricultural sites. It is
ironic that the very function that makes floodplains so
important attracts major disturbances which, in turn,
result in destruction of the kidney function in those
systems. Globally, that destruction is reflected in the
magnitude of floodplain loss (i.e., 75%).
While the primary cause of floodplain destruction is
shifting from agriculture to urban development, it would
be unrealistic to expect that the general magnitude of
anthropogenic pressures on these systems will abate.
Consequently, an answer to the critical question of
263
whether adequate supplies of clean water exist will
become increasingly uncertain. In order to provide a
positive answer and, subsequently, protect human
health and well being, it is vital that we more clearly
understand how these ecotones operate so that func
tional floodplains can be maintained and integrated
into evolving landscapes.
See also: Ecosystem Ecology; Ecosystems; Riparian
Wetlands; Rivers and Streams: Ecosystem Dynamics and
Integrating Paradigms; Rivers and Streams: Physical
Setting and Adapted Biota; Swamps.
Further Reading
Brinson MM (1990) Riverine forests. In: Lugo AE, Brinson MM, and
Brown SL (eds.) Forested Wetlands, Vol. 15: Ecosystems of the
World, pp. 87 141. Amsterdam: Elsevier Science Publishers.
Cavalcanti GG and Lockaby BG (2005) Effects of sediment deposition
on fine root dynamics in riparian forests. Soil Science Society of
America Journal 69: 729 737.
Groffman PM, Bain DJ, Band LE, et al. (2003) Down by the riverside:
Urban riparian ecology. Frontiers in Ecology and the Environment
6: 315 321.
Hupp CR (2000) Hydrology, geomorphology and vegetation of coastal
plain rivers in the south eastern USA. Hydrological Processes
14: 2991 3010.
Junk WJ (1997) Ecological Studies 126: The Central Amazon
Floodplain: Ecology of a Pulsing System. Berlin: Springer.
Lewis WM, Jr., Hamilton SK, Lasi MA, Rodriguez M, and Saunders JF, III
(2000) Ecological determinism on the Orinoco floodplain. Bioscience
50: 681 692.
McClain ME, Victoria RL, and Richey JE (2001) The Biogeochemistry of
the Amazon Basin. New York, NY: Oxford University Press.
Megonigal JP, Conner WH, Kroeger S, and Sharitz RR (1997)
Aboveground production in southeastern floodplain forests: A test of
the subsidy stress hypothesis. Ecology 78: 370 384.
Messina MG and Conner WH (1998) Southern Forested Wetlands:
Ecology and Management. Boca Raton, FL: CRC Press.
Mitsch WJ and Gosselink JG (2000) Wetlands, 3rd edn. New York, NY:
Wiley.
Naiman RJ and Decamps H (1997) The ecology of interfaces: Riparian
zones. Annual Review of Ecology Systematics 28: 621 658.
National Academy of Science (2002) Riparian Areas. Functions and
Strategies for Management. Washington, DC: National Academy
Press.
Paul MJ and Meyer JL (2001) Streams in the urban landscape. Annual
Review of Ecology and Systematics 32: 333 365.
van Splunder I, Coops H, Voesenek LACJ, and Blom CWPM (1995)
Establishment of alluvial forest species in floodplains: The role of
dispersal timing, germination characteristics and water level
fluctuations. Acta Botanica Neerlandica 44: 269 278.
264
Forest Plantations
Forest Plantations
D Zhang, Auburn University, Auburn, AL, USA
J Stanturf, Center for Forest Disturbance Science, Athens, GA, USA
ª 2008 Elsevier B.V. All rights reserved.
An Overview and Economic Explanation of Global
Forest Plantation Development
Factors Influencing Forest Plantation Development
Forest Plantations and Conservation of Natural Forests
Direct Ecological Effects of Forest Plantation
Further Reading
Between the extremes of afforestation and unaided natural
regeneration of natural forests, there is a range of forest
conditions in which human intervention occurs. Previously,
forest plantations were defined as those forest stands estab
lished by planting and/or seeding in the process of
afforestation or reforestation. Within plantations, there is a
gradient in conditions. At one extreme is the traditional
forest plantation concept of a single introduced or indigen
ous species, planted at uniform density and managed as a
single age class (the so called monoculture). At the other
extreme is the planted or seeded mixture of native species,
managed for nonconsumptive uses such as biodiversity
enhancement. To further complicate matters, many forests
established as plantations come to be regarded as secondary
or seminatural forests and no longer are classed as planta
tions. For example, European forests have long traditions of
human intervention in site preparation, tree establishment,
silviculture, and protection; yet they are not always defined
as forest plantations.
Further refinement of the plantation concept is necessary
in order to encompass the full range of actual conditions. A
useful typology is based on purpose, stand structure, and
composition of plantations. Thus, an industrial plantation is
established to provide marketable products, which can
include timber, biomass feedstock, food, or other products
such as rubber. Industrial plantations usually are regularly
spaced with even age classes. Home and farm plantations are
managed forests but at a smaller scale than industrial planta
tions, producing fuelwood, fodder, orchard, and garden
products but still with regular spacing and even age classes.
A wide range of agroforestry systems exist, distinguishable as
a complex of treed areas within a dominantly agricultural
matrix. Environmental plantations are established to stabilize
or improve degraded areas (commonly due to soil erosion,
salinization, or dune movement) or to capture amenity values.
Environmental plantations differ from industrial plantations
by virtue of their purpose; they may still be characterized as
regularly spaced with even age classes. Efforts to restore forest
ecosystems are increasing and often utilize the technology of
plantation establishment, at least initially.
Recently, FAO defined ‘planted forests’ as forests in
which trees have been established through planting or
seeding by human intervention. This definition is broader
than plantations and includes some seminatural forests
that are established through assisted natural regeneration,
planting or seeding (as many planted forests in Europe
that resembled natural forests of the same species mix)
and all forest plantations which are established through
planting or seeding. Planted forests of native species are
classified as forest plantations if characterized by few
species, straight, regularly spaced rows, and/or even
aged stands. Forest plantations may be established for
different purposes and were divided by FAO into two
classes: protective forest plantations which are typically
unavailable for wood supply (or at least having wood
production as a secondary objective only) and often con
sist of a mix of species managed on long rotations or under
continuous cover; and productive plantation forests which
are primarily for timber production purposes.
Figure 1 shows that, in 2005, some 36% of global forests
(about 4 billion ha, covering 30% of total global land area)
are natural forests, 53% are modified natural forests, 7%
are seminatural forests, and the remaining 4% are forest
plantations. Of these forest plantations, productive forest
plantations account for 78% and protective forest planta
tions account for 22%. While natural forests and modified
natural forests declined between 1990 and 2005, semina
tural forests and forest plantations increased (Figure 2).
7%
3% 1%
36%
53%
Primary forest
Modified natural forest
Seminatural forest
Productive forest plantation
Protective forest plantation
Figure 1 Global forest characteristics 2005. Modified from
FAO (2005) Global forest resources assessment 2005. FAO
Forestry Paper 147. Rome, Italy.
Forest Plantations
265
Primary forest
Modified natural forest
Seminatural forest
Productive forest plantation
Protective forest plantation
0
200
400
600
800
1990
1000
2000
1200
1400
1600
2005
Figure 2 Global trends in forest characteristics 1990–2005 (million ha). Modified from FAO (2005) global forest resources assessment
2005. FAO Forestry Paper 147. Rome, Italy.
This article provides an overview and economic
explanation of global forest plantation development. It
also presents factors influencing global forest plantation
development and lists the usefulness of forest planta
tions, including their roles in the conservation of
natural forests. Finally, it summarizes the impact of
forest plantations on biodiversity and other ecological
functions.
26%
27%
2%
2%
2%
2%
3%
4%
16%
5%
An Overview and Economic Explanation of
Global Forest Plantation Development
Currently, there are about 109 million ha of productive
forest plantations in the world. Productive forest planta
tions represented 1.9% of global forest area in 1990, 2.4%
in 2000, and 2.8% in 2005. The Asia region accounted for
41%; Europe 20%; North and Central America 16%;
South America and Africa 10% each; and Oceania 3%.
Forest plantations have been increasing at an increased
rate. The area of forest plantations increased about 14
million ha between 2000 and 2005 or about 2.8 million
ha per year, 87% of which are in the productive class. The
area of productive forest plantations increased by 2.0
million ha per year during 1990–2000 and by 2.5 million
ha per year during 2000–05, an increase of 23% compared
with the 1990–2000 period. All regions in the world
showed an increase in plantation area, with the highest
plantation rates found in Asia, particularly in China. The
ten countries with the greatest area of productive forest
plantations accounted for 79.5 million ha or 73% of the
China
United States
Russia Federation
Brazil
Sudan
Indonesia
Chile
Thailand
France
Turkey
Remaining countries
11%
Figure 3 Ten countries with largest area of productive forest
plantations in 2005. Modified from FAO (2005) Global forest
resources assessment 2005. FAO Forestry Paper 147. Rome,
Italy.
total global area of productive forest plantations
(Figure 3). China, the United States, and the Russian
Federation together accounted for more than half of the
world’s productive plantations.
Forest plantations, productive or protective, develop in
response to a relative scarcity of timber and other goods and
services associated with forests. In the early part of modern
human history, population was sparse, forests were abun
dant, and survival, economic development, and territorial
control were the primary concerns of governments and
society. As forest resources declined, assuring an adequate
timber supply gradually caught the attention of rulers and
planners and became state policy. Often, the very first
policy implemented would be to regulate timber harvesting
schedule and intensity. Society also responded by moving
266
Forest Plantations
to frontiers farther and farther away from population cen
ters, which in economic terms is called a shift in the
extensive margin of timber production. In a nutshell, the
production and consumption of forest products were all
from natural forests in the early part of human history,
and forest plantations were not needed.
When the increase in timber consumption caught up
with the ability of a country or a region to produce timber
in naturally regenerated forests, citizens and governments
would become interested in tree planting. While tree
planting occurred at least several thousands of years ago
in the Middle East, China, and Europe, and nearly 200
years ago in the Americas, the areas planted with trees
through afforestation (planting land that was formerly in a
nonforest cover) and reforestation (planting land on
which a former forest had been harvested) were relatively
insignificant in size before AD 1800. It was only after the
industrial revolution that timber consumption increased
drastically, due to increasing human population and
industrial use of wood – initially as charcoal, then lumber,
other solid wood products including mine props and rail
road ties, and pulp and paper, and finally for conservation
uses – that large scale forest plantations started to emerge
in Europe, North America, Asia, and other regions in the
last century, especially in the last few decades.
Thus, forest plantations develop primarily in response
to economic necessity. Timber depletion drives the transi
tion of human consumption of natural forests to artificial
forests. Early in the development of North America, for
example, timber prices were low, and forest lands were
more valuable for other uses, especially the production of
food. So trees were removed, forest lands were converted
to other use, and timber inventory declined. As the stand
ing inventory declines, timber becomes increasingly scarce
and timber prices start to rise. As the prices continue to rise
for timber in natural forests, the purposeful husbandry of
planted forests becomes economically attractive, and pro
ductive forest plantations begin to emerge.
Further, timber depletion affects the supply and
demand balance for environmental services from natural
forests, whether or not these services go through formal
markets. Related to this balance is the fact that the demand
for most environmental services such as clean water, clean
air, and esthetics, which are often produced from or pro
tected by forests, is highly correlated with personal income.
As personal income increases, society demands more
environmental services from forests, as well as more
wood commodities. When natural forests are depleted to
the extent that they cannot adequately provide these ser
vices, protective forest plantations emerge. In some
developing countries, subsistence farming requires forests
to protect farming and grass land from potential flooding,
dust storms, soil erosion, and desertification, and trees are
thus planted for protective purposes whether or not their
personal incomes actually grow over time.
Factors Influencing Forest Plantation
Development
As mentioned earlier, rising timber prices, caused by
timber scarcity, lead to forest plantation development.
Thus, timber prices are the primary factor that influences
forest plantation development. Holding everything else
constant, whenever a country or a region experiences a
long period of rising timber prices, forest plantations
would develop quickly.
Tree planting also requires land, labor, and capital.
The cost of these production factors thus influence forest
plantation development. Further, high timber prices, high
land costs, and high labor costs force innovation in tree
growing technologies in conventional silvicultural treat
ments and biotechnologies. A recent report shows that the
growth rate of pine plantations in Alabama, a southern
state in the US increased about 25% in a decade (from
8.20% in the period from 1982 to 1990, to 10.17% in the
period of 1990–2000). This increase in growth rate is
attributed to advancement in tree growing technologies
as well as an increase in management intensity.
Government policies influence forest plantation devel
opments as well. Taxes on land and forest related income,
cash subsidies to plant trees, regulations on land use and
labor, and free education and extension services to forest
farmers all have an impact, positively or negatively, on tree
planting. In general, the primary motivation for the private
sector to plant trees is to generate financial (or other)
benefits from their investment. In some cases, government
policies (positive or pervasive incentives in taxes, subsidies)
provide or take away a significant proportion of the finan
cial benefits from forest plantation development. Where
governments own land, they could conduct afforestation
and reforestation activities directly, for purely financial
reasons or for social and environmental benefits or both.
The US South is perhaps an important region in tim
ber supply as it produces some 18% of the world’s
industrial round wood with just 2% of the world’s forest
lands and 2% of the world’s forest inventory. Some 90%
of forest lands in the southern US are owned by nonin
dustrial private and industrial owners, and timber markets
are competitive. A study of tree planting showed that tree
planting by both forest industry and nonindustrial private
landowners was positively related to the availability
(measured as previous year harvest) and the price of
land. Planting by forest industry and nonindustrial private
landowners was responsive to market signals, positively to
softwood pulpwood prices and negatively to planting
costs and interest rates. Finally, government subsidy pro
grams, which increase the total plantation area, might
have substitution effects on nonindustrial private tree
planting. The federal income tax break for reforestation
expenses promoted reforestation in the southern US.
Forest Plantations
Since forests often have a long production cycle, per
haps the most important government policy in promoting
forest plantation development is to provide long term and
secure property rights (private property or land tenure) to
private landowners or forest farmers. Many theoretical
and empirical studies substantiate that long term and
secure property rights promote tree planting activities in
both developed and developing countries. For example,
in British Columbia, Canada tree planting was done more
often and more promptly following harvest when forest
property rights were secure. In Ghana, reforestation was
significantly influenced by the form of forest tenure, and
more intensive resource management was fostered by
more secure forms of tenure.
Forest Plantations and Conservation of
Natural Forests
Plantation forests can provide most goods and services
that are provided by natural forests. These include tim
ber, nontimber forest products, protection of clean water
and clean air, soil erosion control, biodiversity, esthetics,
carbon sequestration, and climate control. Nonetheless, as
the value of environmental services from natural forests is
higher than that from forest plantations, the demand for
conservation of natural forests is stronger. It is possible
that a division of land, with some land specialized in
timber production and other land in providing environ
mental services, would produce more forest related goods
and services to society. Because forest plantations grow
much faster than natural forests, forest plantations are
seen as an increasingly important source of timber supply.
Should more forest plantations be developed, more nat
ural forests might be saved.
In 1995, natural forests contributed some 78% of glo
bal industrial timber supply, and the remaining was from
forest plantations. With growing concerns about the status
and loss of natural forests, the rapid expansion of pro
tected areas, and large areas of forest unavailable for wood
supply, plantations are increasingly expected to serve as a
source of timber. The general trend of the sector is for
timber supply to shift from natural forests to plantations.
A simple simulation of global timber supply and
demand, allowing forest plantations and their productiv
ity to extend at the current rate, has shown that logging on
natural forests could fall by half, from about 1.3 billion m3
in 2000 to about 600 million m3 in 2025. Thus, forest
plantations will have an increasingly significant role in
substituting products from natural forests, even if they
cannot replace harvests from natural forests for a long
period of time.
One side impact of forest plantation development is
that the supply of large quantities of low cost timber
could perhaps undermine the value of natural forest
267
stands, leading to more rapid destruction, especially
where legal frameworks and law enforcement are inade
quate. Therefore from a global perspective, the transition
from natural forests as the primary source of timber
supply to forest plantations will take a long time.
Nonetheless, the transition has been completed in some
countries such as New Zealand and Chile.
Direct Ecological Effects of Forest
Plantation
Forest plantations have direct ecological effects in addi
tion to the positive impact of reducing pressure on natural
forests. Generalizations are difficult, however, in part
because plantation management regimes are diverse and
the appropriate comparison is not always to unmanaged
natural forests. In worst case scenarios, natural forests or
savannas on fragile soils are converted to plantations of
exotic species that lower groundwater tables, decrease
biodiversity, and develop extreme nutrient deficiencies
in successive rotations. While this scenario overstates the
impact of plantations, their generally monoculture nature
and intensive management raises concerns about the
effect of plantations on biodiversity, water, long term
productivity and nutrient cycling, and susceptibility to
insects and diseases.
Biodiversity illustrates the complicated ecological impact
of forest plantations; although biodiversity encompasses
genetic, species, structural, and functional diversity, much
of the focus in discussions about diversity has been at the
genetic, species, and local ecosystem levels. As has occurred
in agriculture, the introduction of genetically improved
exotic or native species in forestry increases productivity
and carbon fixation efficiency. In some regions, this intro
duction has also increased interspecies diversity at landscape
and regional scales. In France, compared with 70 natural
forest tree species, 30 introduced species that are commonly
used in forest plantations have helped increase the inter
species genetic diversity of forests at the local level. In
Europe, at least, there is no doubt that the introduction of
new tree species has increased the species richness of forests.
Nevertheless, exotic species, even those long naturalized
species such as Douglas fir (Pseudotsuga menziesii) are unac
ceptable in nature conservation schemes.
Exotic species can have negative impacts on native
species and communities. For example, fast growing spe
cies can replace native forest species because of their
natural invasive potential, as have been observed with
Eucalyptus in northwestern Spain and Portugal. As the
introduction of exotic species has potential risks, confir
mation of long term adaptation to local environmental
conditions and pest resistance is necessarily the first step
for the use of exotic species in extensive plantation
programs.
268
Forest Plantations
Plantations tend to be even aged and managed on
relatively short rotations; thus, simple stand structures
are common. When repeated across a landscape, large
areas of similar species and low structural complexity
result in a loss of habitat for taxa that require the kind of
conditions provided by naturally regenerated stands
or old forests. It has been reported that the bird fauna of
single species plantation forests is less diverse than that of
natural and seminatural forests. In other cases, however,
bird species diversity in plantation forests is comparable
with that in naturally generated stands. For example,
cottonwood (Populus deltoides) plantations in the
Mississippi River Valley in the southern United States
are intensively managed (rotation lengths of 10–15 years),
reaching crown closure in 2 years. In comparison to
natural stands, bird species diversity and abundances are
similar for all guilds except cavity nesters.
Where avian diversity is decreased in managed forests
generally, loss of structure following harvest is usually the
cause. In plantations, simplified structure may be exacer
bated further by use of exotic species or by monoculture.
Because plantations are harvested at or near economic
optima, rather than at biological maturity, plantations
seldom develop much beyond the stem exclusion stage
of stand development and do not re establish character
istics of old forests or complex stand structures such as
snags and coarse woody debris. Strategies to compensate
for the simplifying tendencies of plantations and integrate
biodiversity considerations include complex plantations
composed of multiple species, varying planting spacing,
thinning to variable densities, and retaining uncut patches
and snags after harvest. Such biological legacies should
benefit invertebrates such as saproxylic beetles as well as
fungi, small mammals, and birds.
Silvicultural and site management practices of site
preparation, competing vegetation control, and fertiliza
tion may reduce understory and groundcover vegetation
diversity, although the effects of previous land use such as
agriculture may play a larger role. For example, in south
ern United States industrial pine plantations, understory
diversity was correlated with previous land use; lower
diversity of native forest species occurred in plantations
established on former farmland and higher diversity in
plantations on cutover forest land.
Some species can benefit from forest plantations. For
example, clear cutting and short rotations favor the occur
rence of ruderal plant species over some long lived
climax species. Forest plantations accommodate edge
specialist bird species and generalist forest species such
as deer. Some rare and threatened species have been
found to occupy forest plantations, especially when they
lost most of their habitat to agricultural and urbanized
land uses. For example in the UK, the native red squirrel
is out competed in native woodlands by the gray
squirrel introduced from North America but the red
squirrel thrives in conifer plantations, which are poor
habitat for the gray squirrel.
Spatial considerations play a role in maintaining bio
diversity at the landscape scale. Landscape diversity can
meet the habitat needs of wildlife and be achieved by
varying the size and shape of plantations and incorporat
ing adjacency constraints into harvest scheduling models
(i.e., a plantation adjacent to a recently harvested or young
stand cannot be harvested until the adjacent stand reaches
a certain age or crown height). Retaining areas of natu
rally regenerated forest, riparian buffers, or open habitat
creates a landscape mosaic that combined with prescribed
burning in fire affected ecosystems, adds to landscape
diversity. Landscape connectivity that provides dispersal
corridors for mobile species is fostered by careful place
ment of forest roads and firebreaks.
Concerns about plantations and water are as varied as
the issues surrounding biodiversity but generally relate to
water use, water quality, or alteration of natural drainage.
Species of Eucalyptus planted outside their native Australia
have attracted the most negative attention for their puta
tive excessive water use, especially in Africa and India but
Populus species have similarly been accused in China of
lowering local water tables and adding to drought. Species
such as Eucalyptus camaldulensis, E. tereticornis, and E. robusta
(and hybrids of these and other eucalypts) are drought
tolerant and able to transpire even under considerable
moisture stress. On balance they probably do not use
more water than adjacent natural forests but certainly
use more of the available water than grasslands or agri
cultural crops. There is little evidence that they can
abstract groundwater; however, there is no recharge
below the root zone. In the Wheatbelt of Western
Australia, removal of the deep rooted native vegetation
including eucalypts and conversion to cereal crops has
caused water tables to rise with subsequent salinization
of soils and surface water bodies. Plantations of oil
mallee crops (E. polybractea, E. kochii subsp. plenissima, and
E. horistes) are planted to restore natural hydrology and
counteract salinization.
Negative effects of plantations on water quality and
aquatic resources are more due to intensive management
than to use of exotic species. Intensive mechanical site
preparation, especially on sloping sites, can result in sedi
ment movement into streams. Chemical herbicides are
used to control competing vegetation at various stages
in the plantation growth cycle, but usually for site pre
paration in place of mechanical treatments or early in the
life of the stand to release crop species from competitors.
Less intense site preparation, formulations of herbicides
that are not toxic to insects or other aquatic organisms and
break down in soil, careful placement of chemicals to
avoid direct application to water bodies, and designation
of riparian buffers all have contributed to protection of
water quality.
Forest Plantations
Harvesting practices, especially placement and con
struction of harvest roads and layout of skidding trails,
potentially can degrade water quality. In developed
nations, forest practices such as site preparation, har
vesting, use of herbicides, and even choice of species
may be regulated to some extent. In the United States,
best management practices (BMPs) to address non
point source pollution and protect water quality have
been codified by state agencies and landowners follow
them voluntarily. Research shows generally high rates
of compliance. Certification schemes substitute the
coercive power of the marketplace for that of govern
ment; the various certification bodies differ in how
they regard plantations, especially with regard to the
use of herbicides, exotic species, or genetically mod
ified trees.
Use of inorganic fertilizers to overcome fertility defi
ciencies, promote rapid growth, and sustain biomass
accumulation generally has been found to have little
impact on aquatic systems unless fertilizers are applied
directly to streams, lakes, rivers, or adjacent riparian
zones. Greater attention has focused on nutrient
removals in harvests and the potential for intensive
management to reduce site fertility and cause a fall off
in productivity of subsequent rotations. Claims of later
rotation productivity declines have been hard to sub
stantiate, however, as general improvements in seed and
seedling quality, genetic makeup, site preparation and
competition control, and more careful harvesting that
conserves site fertility have raised, rather than lowered
yields. Nevertheless, there exist documented cases of
lowered fertility caused by export of nutrients in the
harvested wood. These localized cases have been caused
by low initial fertility, often of phosphorus, potassium, or
micronutrient deficiencies inherent in the soil parent
material that are easily overcome by application of inor
ganic fertilizers.
In the most intensive management of pine plantations
for pulpwood in the southern United States, some com
panies routinely apply complete nutrient mixes
containing all macro and micronutrients as a precau
tion, despite lack of demonstrated deficiency of most
nutrients except phosphorus and a responsiveness to
added nitrogen. A stand may be fertilized with nitrogen
up to five times in a 25 year rotation, sometimes in
combination with phosphorus. These stands occur
mostly on relatively infertile Ultisols and Spodosols
developed on old marine sediments. On better soils
(Alfisols, Entisols, and Vertisols), cottonwood plantations
managed on 10 year rotations receive only an initial
application of nitrogen at planting to promote rapid
height growth to better compete with herbaceous com
petitors. Management of site nutrients in intensive
plantations is critical to high yields as well as to protect
long term productivity and may require attention to
269
retaining soil organic matter, especially on sandy soils.
Factors to consider include inherent soil fertility (nutri
ent stocks as well as transformations and fluxes), plant
demand and utilization efficiency, and nutrients export
in products removed as well as leakages.
It is common wisdom that monoculture plantations are
more susceptible than natural forests to insect and disease
attacks, yet there is little evidence this is generally true. On
the one hand, single species stands occur naturally and some
of these natural vegetation types are the product of periodic,
catastrophic disturbances such as pine bark beetles or spruce
budworm. On the other hand, one explanation for the often
greater productivity of exotic tree species than attained in
their native habitat is the lack of yield reducing insects and
diseases. But diversity in the abstract is not a guarantor of
lessened risk; diverse, multiple species stands themselves are
not immune to devastating attack by introduced pests, a
situation likely to increase in frequency as a result of globa
lization of trade in timber products.
Often the practices associated with intensive management
are the causes of insect and disease problems. For example,
the desire to maximize wood production may set the level of
tolerable damage from native pests lower than the stable
equilibrium levels for the pest; attempts to control the pest
at lower levels may cause unstable population growth cycles.
The potential risks of plantations stem from their uniformity:
the same or a few species, planted closely together, on the
same site, over large areas. Pests and pathogens adapted to
the dominant species may build up quickly due to food
supply and abundant sites for breeding or infection.
Proximity of the branches and stems in closely spaced stands
may favor buildup of species with low dispersal rates or small
effective spread distances. Conversely, the same uniformity
of plantations that contributed to the risks of insects and
diseases also confers some advantages. Species can be chosen
that have resistance to diseases, for example, the greater
resistance of loblolly pine (Pinus taeda) compared to slash
pine (P. elliottii) to Cronartium rust was one reason loblolly was
favored by forest industry in the US South. The shorter
rotation length of plantations relative to naturally regener
ated stands means trees are fallen before they become
overmature and become infected. The compact shape and
uniform conditions in plantations facilitate detection and
treatment of economically important pests and pathogens.
Plantations may negatively impact adjacent commu
nities – because of invasive natural regeneration of
planted trees in adjacent habitat or alteration of local
and regional hydrologic cycles and poor management
practices may damage aquatic systems. Plantations are
certainly simpler and more uniform than naturally regen
erated stands or native grasslands, and may support a less
diverse flora and fauna. Nevertheless, plantations can
contribute to biodiversity conservation at the landscape
level by adding structural complexity to otherwise simple
270
Freshwater Lakes
grasslands or agricultural landscapes and by fostering the
dispersal of forest dwelling species across these areas.
Further, comparisons of plantations to unmanaged
native forests or even naturally regenerated secondary
forests are not necessarily the most appropriate compar
isons to make. Although the conversion of old growth
forests, native grasslands, or some other natural ecosystem
to forest plantations rarely will be desirable from a biodi
versity point of view, in that forest plantations often
replace other land uses including degraded lands and
abandoned agricultural areas. Objective assessments of
the potential or actual impacts of forest plantations on
biological diversity at different temporal and spatial scales
require appropriate reference points. Forest plantations
can have either positive or negative impacts on biodiver
sity at the tree, stand, or landscape level depending on the
ecological context in which they found. Impacts on water
quantity and quality can be minimized if sustainable
practices are followed; similarly with soil resources and
long term site productivity. Both complex plantations for
wood production and environmental plantations can ben
eficially impact local and regional environments.
Lastly, managing forest plantations to produce goods such
as timber while at the same time enhancing ecological ser
vices such as biodiversity involves tradeoffs; this can be made
only with a clear understanding of the ecological context of
plantations in the broader landscape. Tradeoffs also require
agreement among stakeholders on the desired balance of
goods and ecological services from plantations. Thus, there
is no single or simple answer to the question of whether forest
plantations are ‘good’ or ‘bad’ for the environment.
See also: Boreal Forest; Temperate Forest; Tropical
Rainforest.
Further Reading
Binkley CS (2003) Forestry in the long sweep of history. In: Teeter LD,
Cashore BW, and Zhang D (eds ) Forest Policy for Private Forestry:
Global and Regional Challenges, pp. 1 8. Wallingford: CABI Publishing.
Brown C (2000) The global outlook for future wood supply from forest
plantations. FAO Working Paper GFPOS/WP/03. Rome, Italy.
Carnus J M, Parrotta J, Brockerhoff E, et al. (2006) Planted Forests and
Biodiversity. Journal of Forestry 104(2): 65 77.
Clawson M (1979) Forests in the long sweep of history. Science
204: 1168 1174.
Evans J and Turnbull JW (2004) Plantation Forestry in the Tropics: The
Role, Silviculture and Use of Planted Forests for Industrial, Social,
Environmental and Agroforestry Purposes, 3rd edn. Oxford: Oxford
University Press.
FAO (2001) Global forest resources assessment 2000. FAO Forestry
Paper 140. Rome, Italy.
FAO (2005) Global forest resources assessment 2005. FAO Forestry
Paper 147. Rome, Italy.
Harris TG, Baldwin S, and Hopkins AJ (2004) The south’s position in a
global forest economy. Forest Landowner 63(4): 9 11.
Hartsell AJ and Brown MJ (2002) Forest statistics for Alabama, 2000.
Resource Bulletin SRS 67, 76pp. Ashville, NC: USDA Forest Service
Southern Research Station.
Li Y and Zhang D (2007) Tree planting in the US South: A panel data
analysis. Southern Journal of Applied Forestry 31(4): 192 198.
Royer JP and Moulton RJ (1987) Reforestation incentives: Tax
incentives and cost sharing in the South. Journal of Forestry
85(8): 45 47.
Stanturf JA (2005) What is forest restoration? In: Stanturf JA and
Madsen P (eds.) Restoration of Boreal and Temperate Forests,
pp. 3 11. Boca Raton, FL: CRC Press.
Stanturf JA, Kellison RC, Broerman FS, and Jones SB (2003) Pine
productivity: Where are we and how did we get here? Journal of
Forestry 101(3): 26 31.
Zhang D (2001) Why so much forestland in China would not grow trees?
Management World (in Chinese) 3: 120 125.
Zhang D and Flick W (2001) Sticks, carrots, and reforestation
investment. Land Economics 77(3): 443 56.
Zhang D and Oweridu E (2007) Land tenure, market and the
establishment of forest plantations in Ghana. Forest Policy and
Economics 9: 602 610.
Zhang D and Pearse PH (1997) The Influence of the form of tenure on
reforestation in British Columbia. Forest Ecology and Management
98: 239 250.
Freshwater Lakes
S E Jørgensen, Copenhagen University, Copenhagen, Denmark
ª 2008 Elsevier B.V. All rights reserved.
Introduction
The World’s Freshwater Lakes
Importance of Lakes
Water Quality Problems of Lakes and Reservoirs
Further Reading
Introduction
and permafrost, while all fresh groundwater makes up
0.76% of the global water. It leaves 0.01% only for the
surface freshwater, of which 70% or 0.007% of the global
water is stored in the freshwater lakes. As surface water is
easily accessible water, the storage of water in lakes and
Freshwater lakes and reservoirs are basins filled with
freshwater. Only 2.53% of the global water is freshwater;
1.76% of the global water is stored in ice caps, glaciers,
Freshwater Lakes 271
reservoirs becomes very important for the water supply
and represents a large proportion of the world’s readily
accessible water (see Figures 1 and 2). Lake water is not
only used for human consumption. Other water uses
include industrial applications and processes and trans
portation and generation of hydropower.
The World’s Freshwater Lakes
Table 1 gives an overview of 12 important freshwater lakes,
including the deepest lake, the lake with the largest surface
area, and the lake with the biggest volume. The lakes are not
equally distributed in the world. About 10% of the total land
is occupied by lakes in Scandinavia, while lakes occupy less
than 1% of the land area in Argentina and China.
Importance of Lakes
The lake and reservoir water uses are becoming more
intensive and multipurpose, particularly for lakes in heav
ily populated areas and intensively utilized regions. We
can distinguish nine functions of lakes and reservoirs:
1.
2.
3.
4.
5.
6.
7.
8.
9.
drinking water supply,
irrigation,
flood control,
aquatic production and fishery,
fire and ice ponds,
transportation,
hydropower,
conservation of biodiversity, and
recreation.
The multipurpose and extensive use of lakes and reservoirs
can often lead to abuse and conflicts. There are numerous
examples of such conflicts which are often rooted in inap
propriate and insufficient water management.
Figure 1 Lake Baikal, the deepest lake in the world. The
volume of Lake Baikal corresponds to almost 20% of all global
surface freshwater.
Water Quality Problems of Lakes and
Reservoirs
Nine problems associated with the extensive use of lakes
and reservoirs can be identified.
Table 1 Major freshwater lakes
Lake
Figure 2 Crater Lake, Oregon State, the lake famous throughout
the world for its clarity. The Secchi disk transparency is 42 m.
Lake Baikal
Lake
Tanganyika
Lake Superior
Lake Malawi
Lake Michigan
Lake Huron
Lake Victoria
Lake Titicaca
Lake Erie
Lake
Constance
Lake Biwa
Lake Maggiore
Volume
(km3)
Area
(km2)
Max. depth
(m)
22 995
18 140
31 500
32 000
1 741
1 471
12 100
6140
4920
3540
2700
903
484
48.5
82 100
22 490
57 750
59 500
62 940
8 559
25 700
571
170
706
110
92
80
283
64
254
674
213
104
370
27.5
37.5
272
Freshwater Lakes
Eutrophication
This is the most pervasive water quality problem on a
global scale, being a primary cause of lake deterioration.
Eutrophication (nutrient enrichment) represents the nat
ural aging process of many lakes in which they gradually
become filled with sediments and organic materials over a
typically geologic timescale. Human activities in a drain
age basin can, however, dramatically accelerate this
process. Its primary cause is the excessive inflow of nutri
ents (mainly phosphorus, sometimes nitrogen, sometimes
both) to a water body from municipal wastewater treat
ment plants and industries, as well as drainage or runoff
from urban areas and agricultural fields. Most lakes in
densely inhabited regions of the world suffer from eutro
phication, both in industrialized and developing
countries. The impacts of the eutrophication process
include heavy blooms of phytoplankton in a water body.
These blooms will inevitably result in (1) reduced water
transparency; (2) decreased oxygen concentration in the
water column, particularly in the bottom layer (hypolim
nion), which can cause fish kills and the remobilization or
resuspension of heavy metals and nutrients into the water
column; and (3) significant declines in the biodiversity of
the lakes, including the disappearance of sensitive aquatic
species. In shallow lakes, eutrophication can also cause an
enormous increase in the growth of submerged and emer
gent rooted aquatic plants, as well as floating plants. This
can lead to dramatic changes in the ecosystem structure.
If the sources of nutrients are removed or reduced
significantly, the eutrophication problems can be fully
controlled (see Figures 3 and 4). Lake Constance, also
known as Bodensee, gives very illustrative examples.
Constructed
wetland
Withdrawal of water
Water treatment
Lake
Figure 4 Lake Bled, where restoration by siphoning hypolimnic
water has been applied.
After the Second World War, the phosphorus concentra
tion in the lake was about 0.01 mg l 1 and the lake was
oligotrophic. In the year 1980, the lake was mesotrophic
to eutrophic and the phosphorus concentration was about
0.08 mg l 1. Due to a massive reduction in the discharge of
phosphorus from all sources, wastewater, agricultural
drainage water, and septic tanks, it has been possible
to reduce the phosphorus concentration to about
0.013 mg l 1 today. Lake Biwa, Japan, is illustrative of a
partial solution of the problem (Figure 5). The discharge
of phosphorus from wastewater has been significantly
reduced since the 1970s, but due to almost no reduction
in the phosphorus coming from agricultural drainage
water, it has only been possible to stabilize the eutrophi
cation level at a phosphorus concentration about
0.035 mg l 1. If on the other hand, the phosphorus in
wastewater would not have been reduced, the eutrophi
cation level would have increased.
Removal of phosphorus by precipitation
before discharge
Stream
Stream
Hypolimnion water is
removed by siphoning
Figure 3 Abatement of eutrophication requires often the use of
several methods at the same time, as shown here: removal of
phosphorus from wastewater, construction of wetland to remove
phosphorus from the inflowing tributary, and removal of
hypolimnic (bottom) water by siphoning.
Figure 5 Lake Biwa in Japan is a very important recreational
area for the population. A museum has been erected to present
for the population all aspects of the lake: the culture, the
limnology, the geology, and the history.
Freshwater Lakes 273
Acidification
This process of lake deterioration is caused mainly by acid
precipitation and deposition. The nitrogen and sulfur com
pounds that cause this problem are emitted by industrial
activities and by the consumption of fossil fuels, and fall to
the land surface. The water in a lake can become acidic
over time if its drainage basin does not contain the appro
priate soil and geologic characteristics to neutralize the
acidic water prior to its inflow into the lake. The primary
consequence of acidification of lake water is the significant
reduction of species diversity, the extinction of fish popu
lations, and the disruption of lake ecosystem equilibrium.
Other causes of lake acidification also exist, including
water discharges from mining activities and the direct
discharge of industrial waste effluents containing acidic
components. Natural sources of acidifying substances
include volcanic activities and natural emissions of gases.
This problem, because of the geological characteristics, has
been a major problem in Scandinavia (except the most
southern Scandinavia) and the northeastern United States.
Toxic Contamination
This problem can have direct and dramatic impacts on
both human and ecosystem health. Toxic substances origi
nate not only from industrial activities and mining, but also
as a result of intensive agriculture practices. Identification
of the number of lakes and reservoirs exhibiting toxic
contamination will doubtlessly increase in future years as
we obtain more information on their concentrations in the
environment, particularly in developing countries. Major
impacts of toxic substances include the disappearance of
sensitive species, as well as their accumulation in lake
sediments and biota. The latter can directly and indirectly
impact human health. Because the number of risk assess
ments applied to already existing chemicals is currently
extremely low (500), a complete solution of this problem
will take many years.
Water-Level Changes
Significant changes in water levels, particularly dropping
levels, can be caused by
1. excessive withdrawal of water from lakes and/or their
inflowing or out flowing rivers, and
2. the diversion of the inflowing water.
The consequences of water level changes include:
decrease in lake volume and/or surface area; unstable
shoreline area communities; changes in lake ecosystem
structure; reduced fish spawning areas; increased water
retention time (decreased flushing rate), which can accel
erate other negative lake processes (e.g., eutrophication,
retention of toxic substances); and increased salt concen
tration, leading to reduced water quality for human uses.
Lake Aral is probably the most illustrative example of this
problem. Due to uncontrolled use of the inflowing river
water for irrigation, the water level in the lake was reduced
by almost 20 m. The lake was divided in two lakes, Large and
Small Aral, with together less than half of the original lake
area and with a salinity 10 times what it was 40 years ago.
Salinization
This process is an increase in the concentration of salts (all
ions, not just sodium and chloride) in lake water, caused by
such factors as (1) decreased in lake water levels; (2) overuse
of water in the drainage basin (e.g., cooling water, irrigation);
and (3) global climate change. The effects of water saliniza
tion include (1) dramatic changes in lake biological
structure; (2) lower fish production; and (3) reduced biodi
versity. Human utilization of lake water with a high salt
concentration also can become very problematic. This pro
blem, however, can at least be partly addressed with the
implementation of appropriate environmental management
and agricultural practices in a lake’s drainage basin.
Siltation
Accelerated soil erosion, resulting from such activities
as the overuse or misuse of arable land, mining and/or
deforestation in a drainage basin, can lead to the
excessive loading of suspended solids (sediment) to
lakes. The consequences of these increased loads
include the rapid accumulation of sediment within
the lake basin, and the increased turbidity (decreased
transparency) of the water in the lake. The immediate
impacts can be a significant reduction in the number
of living organisms in a lake, decreased biodiversity,
and reduced fisheries.
Introduction of Exotic Species
The intentional introduction of exotic (nonresident)
species has become an almost common practice in
some fisheries to increase the production of commer
cially important species. The introduction of Nile perch
into Lake Victoria is a primary example. However,
the intentional or unintentional (or sometimes illegal)
introduction of exotic species can cause very serious
problems in a given lake. The accidental introduction
of zebra mussels in Lake Erie and water hyacinths in
several lakes of China provides a dramatic example of
this phenomenon. The introduction of exotic species can
provoke very dramatic changes in the ecosystem struc
ture not only at the biological community level, but also
in a lake’s chemical–physical environment. The major
negative consequences of exotic species include the
274
Freshwater Marshes
(1) disappearance of native species; (2) alteration of
trophic equilibrium; (3) significant reduction in species
diversity; and (4) reduction of water transparency and
changes in algae bloom patterns, via chemical–physical
feedback processes in a lake.
in the early 1990s. It is likely that many deaths in devel
oping countries are due to consumption of dirty lake
water.
Further Reading
Overfishing
Unsustainable fishing practices, sometimes combined
with other problems, can lead to the collapse of fisheries.
It seems to be an increasing problem for many African
lakes, particularly Lake Victoria.
Pathogenic Contamination
This problem is caused by discharge of untreated sewage
or runoff from livestock farms, a problem in both develop
ing and developed countries. A Cryptosporidium outbreak
in Lake Michigan (Milwaukee) sickened 400 000 people
ILEC (2005) Managing Lakes and Their Basins for Sustainable Use: A
Report for Lake Basin Managers and Stakeholders, 146pp.
Kusatsu, Japan: International Lake Environment Committee
Foundation. http://www.ilec.or.jp/eg/lbmi/reports/
LBMI Main Report 22February2006.pdf (accessed October
2007).
ILEC and UNEP (2003) World Lake Vision: A Call to Action, 37pp.
Kusatsu, Japan: World Lake Vision Committee. http://
www.ilec.or.jp/eg/wlv/complete/wlv c english.PDF (accessed
October 2007).
Jørgensen SE, de Bernard R, Ballatore TJ, and Muhandiki VS (2003)
Lake Watch 2003. The Changing State of the World’s Lakes, 73pp.
Kusatsu, Japan: ILEC.
Jørgensen SE, Loffler H, Rast, and Straškraba M (2005) Lake and
Reservoir Mangement, 502pp. Amsterdam: Elsevier.
O’Sullivan PE and Reynolds CS (2004, 2005) The Lakes Handbook,
vols. 1 and 2, 700pp. and 560pp. Blackwell Publishing.
Freshwater Marshes
P Keddy, Southeastern Louisiana University, Hammond, LA, USA
ª 2008 Elsevier B.V. All rights reserved.
Six Types of Wetlands
The Distribution of Marshes
Water as the Critical Factor
Other Environmental Factors Affecting Marshes
Plant and Animal Diversity in Wetlands
Human Impacts
Wetland Restoration
Summary
Further Reading
Wetlands are produced by flooding, and as a conse
quence, have distinctive soils, microorganisms, plants,
and animals. The soils are usually anoxic or hypoxic, as
water contains less oxygen than air, and any oxygen that is
dissolved in the water is rapidly consumed by soil micro
organisms. Vast numbers of microorganisms, particularly
bacteria, thrive under the wet and hypoxic conditions
found in marsh soils. These microbes transform elements
including nitrogen, phosphorus, and sulfur among differ
ent chemical states. Therefore, wetlands are closely
connected to major biogeochemical cycles. The plants in
wetlands often have hollow stems to permit movement of
atmospheric oxygen downward into their rhizomes and
roots. Many species of animals are adapted to living in
shallow water, and in habitats that frequently flood. Some
of these are small invertebrates (e.g., plankton, shrimp,
and clams), while others are larger and more conspicuous
(fish, salamanders, frogs, turtles, snakes, alligators, birds,
and mammals).
Six Types of Wetlands
There are six major types of wetlands: swamp, marsh, fen,
bog, wet meadow, and shallow water (aquatic). These six
types are produced by different combinations of flooding,
soil nutrients, and climate. A seventh group, saline wetlands,
which includes salt marshes and mangroves, is often treated
as a distinct wetland type. Saline wetlands occur mostly
along coastlines (see Mangrove Wetlands), but also occa
sionally in noncoastal areas where evaporation exceeds
Freshwater Marshes
rainfall, such as in arid western North America, northern
Africa, or central Eurasia.
Swamps and marshes have mineral soils with sand, silt,
or clay. Swamps are dominated by trees or shrubs
(see Swamps), whereas marshes are dominated by herbac
eous plants such as cattails and reeds (Figure 1). Such
wetlands tend to occur along the margins of rivers
(Figure 2) or lakes, and often receive fresh layers of sedi
ment during annual spring flooding. Marshes are among the
world’s most biologically productive ecosystems. As a con
sequence, they are very important for producing wildlife,
and for producing human food in the form of shrimp, fish,
and waterfowl.
Fens and bogs have organic soils (peat), formed from
the accumulation of partially decayed plants. Most
275
peatlands occur at high latitudes in landscapes that were
glaciated during the last ice ages. In fens, the layer of peat
is relatively thin, allowing the longer roots of the plants to
reach the mineral soil beneath. In bogs, plants are entirely
rooted in the peat. As peat becomes deeper (the natural
trend as fens become bogs), plants become increasingly
dependent upon nutrients dissolved in rainwater, even
tually producing an ‘ombrotrophic’ bog. The large
amounts of organic carbon stored in peatlands help
reduce global warming.
Wet meadows occur where land is flooded in some
seasons and moist in others, such as along the shores of
rivers or lakes. Wet meadows often have high plant diver
sity, including carnivorous plants and orchids. Examples
of wet meadows include wet prairies, slacks between sand
dunes, and wet pine savannas. Pine savannas may have up
to 40 species of plants in a single square meter, and
hundreds of species in a hundred hectares.
Aquatic wetlands are covered in water, usually with
plants rooted in the sediment but possessing leaves that
extend into the atmosphere. Grasses, sedges, and reeds
emerge from shallow water, whereas water lilies and
pondweeds with floating leaves occur in deeper water.
Aquatic wetlands provide important habitat for breeding
fish and migratory waterfowl. Animals can create aquatic
wetlands: beavers build dams to flood stream valleys, and
alligators dig small ponds in marshes or wet meadows.
The Distribution of Marshes
Figure 1 Marshes occur in flooded areas, such as this
depression flooded by beavers in Ontario, Canada. As the photo
illustrates, marshes form at the interface of land and water.
Courtesy of Paul Keddy.
Wetlands can occur wherever water affects the soil. Not
only are there therefore many kinds of wetlands, but their
size and shape is very variable. Wetlands can include
small pools in deserts and seepage areas on mountain
sides, or they can be long but narrow strips on shorelines
of large lakes (Figure 3), or they may cover vast river
floodplains (Figure 4) and expanses of northern plains.
The two largest wetlands in the world (both >750 000 km2)
are the West Siberian lowland and the Amazon River
basin. The West Siberian Lowland consists largely of
fens and bogs, but marshes occur along rivers, particularly
in the more southern regions (Figure 5). The Amazon is a
tropical lowland with freshwater swamps and marshes
containing more kinds of trees and fish than any other
region of the world.
Water as the Critical Factor
Figure 2 Extensive bulrush (Schoenoplectus spp.) marshes
along the Ottawa River in central Canada. The stalks of purple
flowers indicate the invasion of this marsh by purple loosestrife
(Lythrum salicaria), a native of Eurasia. Courtesy of Paul Keddy.
Water is a critical factor in all marshes. The duration of
flooding is the most important factor determining the kind
of wetland that occurs. Water can arrive as short pulses of
flooding by rivers, as rainfall, or as slow and steady see
page. Each mode of arrival produces different kinds of
276
Freshwater Marshes
Figure 3 Sedges, grasses, and forbs compose this marsh on
the leeward side of a narrow peninsula projecting into one of
the Great Lakes (Lake Michigan), Michigan, USA. Courtesy of
Cathy Keddy.
Figure 4 Extensive marshes of bulltongue (Sagittaria lancifolia)
and American bulrush (Schoenoplectus americanus) now occur
in coastal Louisiana, USA, where logging destroyed baldcypress
forest. Courtesy of Paul Keddy.
wetlands. In order to better understand marshes, let us
consider four examples of wetlands with very different
flooding regimes.
Floodplains. Wetlands along rivers are often flooded by
annual pulses of water (see Riparian Wetlands). These
pulses may deposit thick layers of sediment or dissolved
nutrients that stimulate plant growth. In floodplains
(see Floodplains), animal life cycles are often precisely
determined by the timing of the flood. Fish may depend
upon feeding and breeding in the shallow warm pools left
by retreating floodwaters. Birds may time their nesting to
be able to feed their young on the fish and amphibians left
behind by receding water. Marshes are often intermixed
Figure 5 The largest wetland in the world occurs in the
Western Siberian Lowland. Although much of this is peatland,
marshes occur along the watercourses, particularly in the
southern areas. Courtesy of M. Teliatnikov.
Figure 6 Southern marshes on the coastal plain of North
America may be dominated by a single grass, maidencane
(Panicum hemitomon). This marsh occupies an opening within a
baldcypress swamp, Louisiana, USA. Courtesy of Cathy Keddy.
with swamps, depending upon the duration of flooding
(Figure 6). Early human civilizations developed in this
type of habitat, along the Nile, Indus, Euphrates and
Hwang Ho, where the annual flooding provided fertilized
soil and free irrigation.
Peat bogs. Some peat bogs receive water only as rainfall.
As a consequence, the water moves slowly, if at all, and
contains very few nutrients. Hence, these types of wetlands
often are dominated by slow growing mosses and evergreen
plants (see Peatlands). Most such wetlands occur in the far
north in glaciated landscapes. Humans have developed a
number of uses for the peat – in Ireland, the peat is cut into
blocks and used for fuel. In Canada, the peat is harvested
and bagged for sale to gardeners. In Russia, peat is used to
fuel electrical plants. Marshes may form on the edges of
Freshwater Marshes
bogs where nutrients accumulate from runoff, or along river
courses where nutrients are more available.
Seepage wetlands. In gently sloping landscapes water can
seep slowly through the soil. In northern glaciated land
scapes, such seepage can produce fens, which have
distinctive species of mosses and plants, and may develop
in distinctive parallel ridges. In more southern landscapes,
seepage can produce pitcher plant savannas or wet prai
ries. Often these seepage areas are rather small (only a few
hectares in extent) but are locally important because of
the rare plants and animals they support. Seepage areas
can be larger, and when the water flow is sufficiently
abundant, shallow water can move across a landscape in
a phenomenon known as sheet flow. The vast Everglades,
with its distinctive animals, is a product of sheet flow of
water from Lake Okeechobee in south central Florida
southward to the ocean.
Temporary wetlands. In many parts of the world, small
temporary (or ephemeral) pools form after heavy rain or
when snow melts. These pools can go by a variety of local
names including vernal pools, woodland ponds, playas or
potholes (see Temporary Waters). The aquatic life in
these pools is forced to adopt a life cycle that is closely
tied to the water levels. Many species of frogs and sala
manders breed in such pools, and the young must
metamorphose before the pond dries up. Wetland plants
may produce large numbers of seeds that remain dormant
until rain refills the pond.
Since water has such a critical effect on wetlands,
where water levels change, plant and animal commu
nities will change as well. A typical shoreline marsh will
often show distinct bands of vegetation (‘zonation’), with
each kind of plant occupying a narrow range of water
depths (Figure 7). Most kinds of animals, including
frogs and birds, also have their own set of preferred
Figure 7 Different marsh plants tolerate different water levels.
Hence, as the water level changes from shallow water (left;
seasonally flooded) to deeper water (right; permanently flooded),
the plants appear to occur in different zones. Courtesy of
Rochelle Lawson.
277
water depths. Wading birds (egrets, ibis, herons) may
feed in different depths of water depending upon the
length of their legs. Ducks, geese, and swans can feed at
different water depths depending upon the length of
their necks. Some water birds (Northern Shoveler, fla
mingos) strain microorganism from shallow water, while
others (cormorants, loons) dive to feed further below the
surface. Some ducks prefer wetlands that are densely
vegetated, while others prefer more open water.
Hence, even small changes in the duration of flooding
or depth of water can produce very different plant and
animal communities.
Many marsh plants adapt to flooding by producing
hollow shoots, which allow oxygen to be transmitted to
the rooting zone. The tissue that allows the flow of oxy
gen is known as aerenchyma. Not only can oxygen move
by diffusion, but there are a number of methods in which
oxygen moves more rapidly through large clones of
plants, entering at one shoot and leaving at another.
Consequently, plants can play an important role in oxi
dizing the soil around their rhizomes, allowing distinctive
microbial communities to form. Some marsh plants also
have floating leaves (e.g., water lilies) or even float
entirely on the surface (e.g., duckweeds). The largest
floating leaves in the world (Figure 8) are those of the
Figure 8 The Amazon water lily has the largest floating leaves
of any wetland plant. Note the prominent ribs on the underside of
the leaf. Courtesy of Corbis.
278
Freshwater Marshes
Amazon water lily (Victoria amazonica). The gargantuan
leaves can be 2 m in diameter with an elevated lip around
the circumference. There are two gaps in the lip to allow
water to drain, and large spines to protect the underwater
sections of the foliage.
Other Environmental Factors Affecting
Marshes
Nutrients
The main nutrients that affect the growth of marsh plants,
and plants in general, are nitrogen and phosphorus. As
described above, flood pulses that carry sediment down
river courses can produce particularly fertile and produc
tive marshes. Floodplains can therefore be thought of as
one natural extreme along a gradient of nutrient supply.
At the other end of the gradient lie peat bogs, which
depend partly or entirely upon rainfall, and which there
fore receive few nutrients. Sphagnum moss is well
adapted to peatlands, and often comprises a large portion
of the peat. In between the natural extremes of river
floodplains (high nutrients) and peat bogs (low nutrients),
one can arrange most other types of wetlands. The type of
plants, and their rate of growth, will depend where along
this gradient they occur, but most marshes generally
occur in more fertile conditions.
While nutrients enhance productivity, paradoxically
they can often reduce the diversity of plants and animals.
Often, the high productivity is channeled into a few
dominant species. One finds large numbers of common
species, while the rarer species disappear. Humans often
increase nutrient levels in watersheds and wetlands,
thereby changing the species present and reducing their
diversity. Carnivorous plants are known for tolerating low
nutrient levels, because they can obtain added nutrients
from their prey. Common examples include pitcher
plants (Sarracenia spp.), bladderworts (Utricularia spp.),
and butterworts (Pinguicula spp.). Cattails (Typha spp.)
and certain grasses (Phalaris arundinacea) are particularly
well known for rapid growth and an ability to dominate
marshes at higher nutrient levels.
Disturbance
A disturbance can be narrowly defined as any factor that
removes biomass from a plant. In marshes, sources of
disturbance may include waves in lakes, fire, grazing, or
(in the north) scouring by winter ice. One of the principal
effects of disturbances is the creation of gaps in the vege
tation, allowing new kinds of plants to establish from
buried seeds. Most marshes have large densities of buried
seeds, often more than 1000 seeds m 2. After disturbance,
marsh plants can also re emerge from buried rhizomes.
Hence, cycles of disturbance play an important role in
creating marshes.
Although the presence of fire in wetlands may seem
paradoxical, fire can often occur during periods of
drought. Northern peatlands, cattail marshes on lake
shores, wet prairies, and seepage areas in savannas can
burn under the appropriate conditions. In northern peat
lands, a fire can remove thousands of years of peat
accumulation in a few days, even uncovering boulders
and rock ridges that were buried beneath the peat. In
marshes, fire can selectively remove shrubs and small
trees, preventing the marsh from turning into a swamp.
In the Everglades, burning can create depressions that
then cause marshes to revert to aquatic conditions.
Animals that feed upon plants often cause only small
and local effects. Think of a moose grazing on water lilies,
a muskrat feeding on grasses, or a hippopotamus feeding
on water hyacinth. Often the small patch of removed
foliage is quickly replaced by new growth. But when
herbivores become overly abundant, they can destroy
the marsh vegetation entirely. In northern North
America along Hudson Bay, Canada geese (Branta
canadensis) are now so abundant that they remove all
vegetation from expanses of coastal marsh. In southern
North America, along the Gulf of Mexico, an introduced
mammal, nutria (Myocastor coypus), similarly can strip
marsh vegetation to coastal mudflats. To some extent,
disturbance by herbivores is a natural phenomenon, one
that has occurred cyclically throughout history. However,
in the above two examples, one suspects humans may be
the ultimate cause of the large scale overgrazing (see the
next section).
Periodic droughts may at times function like a nat
ural disturbance by killing adult plants, and allowing
new species to re establish from buried seeds. Vernal
pools and prairie potholes both have plant and animal
species that are adapted to this kind of cyclical
disturbance.
Plant and Animal Diversity in Wetlands
Wetlands are important for protecting biological diver
sity. Their high productivity provides abundant food, and
the water provides an important added resource. Hence,
wetlands often have large populations of animals and
wading birds. The Camargue in Southern Europe, for
example, is considered to be the European equivalent of
the Everglades. Both have species of wading birds such as
storks and flamingos (Figure 9). Large numbers of other
kinds of species including fish, frogs, salamanders, turtles,
alligators (Figure 10), crocodiles, and mammals require
flooded conditions for all or at least part of the year. If the
wetlands are drained, all of the species dependent upon
them will disappear.
Freshwater Marshes
279
Human Impacts
Figure 9 Marshes provide essential habitat for many kinds of
wading birds including flamingos, Jurong Bird Park, Singapore.
Courtesy of Corbis.
Figure 10 Alligators are one of the many species that benefit
from protected marshes such as the Everglades, Loxahatchee
National Wildlife Refuge, Florida, USA. Courtesy of Paul Keddy.
All wetlands, however, do not support the same spe
cies. Often, as already noted in the section entitled
‘Disturbance’, small differences in water level or nutrient
supplies will produce distinctive types of wetlands.
Hence, wetlands that are variable in water levels and
fertility will frequently support more kinds of species
than wetlands that are uniform. Along the Amazon
River floodplain, for example, different kinds of swamp,
marsh, grassland, and aquatic communities form in
response to different flooding regimes, and each has its
own complement of animal species. In the Great Lakes,
different flood durations similarly produce different types
of wetlands, from aquatic situations in deeper water, to
marshes and wet meadows in shallower water. Some types
of frogs, such as bullfrogs, require deeper water, while
others, such as gray tree frogs, require shrubs.
Humans have had, and continue to have, serious impacts
upon wetlands in general, and marshes in particular. Some
human impacts include draining, damming, eutrophication,
and alteration of food webs. Let us consider these in turn.
One of the most obvious ways in which humans affect
wetlands is by draining them. When the wetlands are
drained, the soil becomes oxidized, and terrestrial plants
and animals replace the wetland plants and animals.
Often, drainage is followed by conversion to agriculture
or human settlement, entirely removing the marshes that
once existed. Vast areas of farmland in Europe, Asia, and
North America were once marshes and have now been
converted to crops for human consumption. Many coun
tries now have laws to protect wetlands from further
development, although the degree of protection provided,
and the degree of enforcement, varies from one region of
the world to another. Wetlands are also often included in
protected areas such as national parks and ecological
reserves.
Construction of dams can also have severe negative
effects upon wetlands. The dams may be built for flood
control, irrigation, or generating electricity. The wetland
behind the dam may be destroyed by the prolonged flood
ing, whereas the wetlands downstream are disrupted by
the lack of normal flood pulses. A single dam can there
fore affect a vast area of wetlands. The degree of damage
depends upon the pattern of water level fluctuations in
the reservoir behind the dam, but in general large areas of
marsh are lost both upstream and downstream from the
dam. Sediment that would have expanded and fertilized
wetlands during periodic floods becomes trapped behind
the dam. Most of the world’s large rivers have now been
significantly affected by dams. To protect wetlands, it is
necessary to identify rivers that are still relatively natural
and to prevent further dams from being constructed. In
other cases, it is possible to remove dams and allow
natural processes to resume. An artificial levee can be
considered a special type of dam that is built parallel to
a river to prevent it from flooding into adjoining lands.
Levees harm marshes by preventing the annual flooding,
and by allowing cropland and cities to move into
floodplains.
Humans can also affect wetlands by changing the
nutrients in the water. Sewage from cities provides a
specific ‘point source’ of nutrients, particularly nitrogen
and phosphorus, that enter water courses then spread
into wetlands. Activities such as agriculture and forestry
provide ‘diffuse sources’ of nutrients, where runoff from
large areas carries dissolved nutrients, and nutrients
attached to clay particles, into the water and into
adjoining marshes. The added nutrients can stimulate
plant growth, which may seem to be beneficial – but it
280
Freshwater Marshes
often leads to significant changes in the biota. Rarer
plants and animals that are adapted to low fertility are
replaced by more common plants and animals that
exploit fertile conditions. Rapid growth of algae, fol
lowed by decay, can eliminate oxygen from lakes,
causing fish kills. Protecting the quality of marshes
therefore requires two sets of actions. First, it is neces
sary to control the obvious point sources of pollution
by building sewage treatment plants. Second, it is
necessary to use entire landscapes with care, with the
broad objective of reducing nutrients in runoff. This
can involve carefully timing the fertilization of crops,
maintaining areas of natural vegetation along water
courses, fencing cattle away from stream valleys,
minimizing construction of new logging roads, and
avoiding construction on steep hill sides.
Herbivores are common in wetlands, and a natural
part of energy flow from plants to carnivores. Common
examples of large herbivores include moose, geese,
muskrats, and hippopotamuses. Humans can disrupt
wetlands by disrupting the natural balance between
herbivores and plants. Herbivores can increase to
destructive levels in several ways. When humans intro
duce new species of herbivores, rates of damage to
plants may increase greatly – for example, nutria intro
duced from South America are causing significant
damage to coastal wetlands in Louisiana. When humans
reduce predation on herbivores, they may also increase
to higher than natural levels. Killing alligators may
damage wetlands by allowing herbivores such as nutria
to reach high population densities; similarly, the loss of
natural predators may be one of the reasons that
Canada geese have multiplied to levels where they
can destroy wetlands around Hudson Bay. There is
also evidence that when humans harvest blue crabs,
snails that the crabs normally eat begin to multiply
and damage coastal marshes. These types of effects
are difficult to study, since the effects may be indirect
and take place over the long term.
Road networks are a final cause of damage to wetlands.
The obvious effects of roads include the filling of wet
lands, and the blocking of lateral flow of water into or out
of wetlands. But there are many other effects. When
amphibians migrate across roads to breeding sites, vast
numbers can be killed by cars. In northern climates, the
road salt put on roads as a de icer can flow into adjoining
wetlands. Snakes may be attracted to the warm asphalt
and killed by passing cars. Invasive plant species can
arrive along newly constructed ditches. Overall, roads
change a landscape by accelerating logging, agriculture,
hunting, and urban development. As a consequence, the
quality of the marshes in a landscape is linked to two
factors: the abundance of roads (a negative effect) and the
abundance of forest (a positive effect). Although it may
not be obvious, halting road construction (or removing
unwanted roads) and protecting forests (or replanting new
areas of forest) may have important consequences for all
the marshes in a landscape.
Wetland Restoration
Humans have caused much damage to wetlands over the
past thousand years, and the effects have increased as
human populations and technological power have
grown. We have seen some examples of damage in the
preceding section. In response to such past abuses,
humans have also begun consciously re creating wet
lands. There are a growing number of efforts to create
new wetlands and enhance existing wetlands. Along both
the Rhine River and the Mississippi River, some levees
have been breached, allowing floodwater to return and
marshes to recover. Depressions left by mining, or delib
erately constructed for wetlands, can be flooded to
recreate small marshes in highly developed landscapes.
Construction of dams and roads has been more carefully
regulated.
The future of marshes will likely depend upon two
human activities: our success at protecting existing
marshes from damage and our success at restoring
marshes that have already been damaged. The list of the
world’s largest wetlands in Table 1 provides an important
set of targets for global conservation.
Summary
Marshes are produced by flooding, and, as a conse
quence, have distinctive soils, microorganisms, plants,
and animals. The soils are usually anoxic or hypoxic,
allowing vast numbers of microorganisms, particularly
bacteria, to transform elements including nitrogen,
phosphorus, and sulfur among different chemical states.
Marsh plants often have hollow stems to permit move
ment of atmospheric oxygen downward into their
rhizomes and roots. Marshes are some of the most
biologically productive habitats in the world, and there
fore support large numbers of animals, from shrimps
and fish through to birds and mammals. Marshes are
one of six types of wetlands, the others being swamp,
fen, bog, wet meadow, and shallow water. Humans can
affect marshes by changing water levels with drainage
ditches, canals, dams, or levees. Other human impacts
can arise from pollution by added nutrients, overhar
vesting of selected species, or building road networks in
landscapes.
Greenhouses, Microcosms, and Mesocosms
281
Table 1 The world’s largest wetlands (areas rounded to the nearest 1000 km2)
Continent
Wetland
Description
Area (km2)
Source
1
Eurasia
Bogs, mires, fens
2 745 000
Solomeshch, chapter 2
2
South America
West Siberian
Lowland
Amazon River basin
Savanna and forested floodplain
1 738 000
3
North America
4
Africa
5
North America
6
South America
7
North America
8
Africa
Mississippi River
basin
Lake Chad basin
9
Africa
River Nile basin
North America
South America
Prairie potholes
Magellanic
moorland
Rank
10
11
Hudson Bay
Lowland
Congo River basin
Mackenzie River
basin
Pantanal
Bogs, fens, swamps, marshes
374 000
Swamps, riverine forest, wet
prairie
Bogs, fens, swamps, marshes
189 000
Junk and Piedade,
chapter 3
Abraham and Keddy,
chapter 4
Campbell, chapter 5
166 000
Vitt et al., chapter 6
138 000
Alho, chapter 7
108 000
Shaffer et al., chapter 8
106 000
Lemoalle, chapter 9
Savannas, grasslands, riverine
forest
Bottomland hardwood forest,
swamps, marshes
Grass and shrub savanna, shrub
steppe, marshes
Swamps, marshes
Marshes, meadows
Peatlands
92 000
63 000
44 000
Springuel and Ali,
chapter 10
van der Valk, chapter 11
Arroyo et al., chapter 12
Modified from Fraser LH and Keddy PA (eds.) (2005) The World’s Largest Wetlands: Ecology and Conservation. Cambridge: Cambridge University
Press.
See also: Floodplains; Mangrove Wetlands; Peatlands;
Riparian Wetlands; Swamps; Temporary Waters.
Further Reading
Fraser LH and Keddy PA (eds.) (2005) The World’s Largest Wetlands:
Ecology and Conservation. Cambridge: Cambridge University Press.
Keddy PA (2000) Wetland Ecology. Cambridge: Cambridge University
Press.
Middleton BA (ed.) (2002) Flood Pulsing in Wetlands: Restoring the
Natural Hydrological Balance. New York: Wiley.
Mitsch WJ and Gosselink JG (2000) Wetlands, 3rd edn. New York:
Wiley.
Patten BC (ed.) (1990) Wetlands and Shallow Continental Water Bodies,
Vol. 1: Natural and Human Relationships. The Hague: SPB
Academic Publishing.
Whigham DF, Dykyjova D, and Hejnyt S (eds.) (1992) Wetlands of the
World 1. Dordrecht: Kluwer Academic Publishers.
Greenhouses, Microcosms, and Mesocosms
W H Adey, Smithsonian Institution, Washington, DC, USA
P C Kangas, University of Maryland, College Park, MD, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Physical/Chemical Control Parameters
Biotic Parameters
The Operational Imperative
Case Study: Coral Reef Microcosm
Case Study: Florida Everglades Mesocosm
Case Study: Biosphere 2
Further Reading
Introduction
time stability is implied, although dimensions are
optional, ranging from the biosphere subset, the biome,
to perhaps a field or pond. Ecosystems with their complex
food webs and biotic physical/chemical relationships are
self organizing due to the genetic information existing in
An ecosystem is an assemblage of organisms living
together and interacting with each other and their envir
onment. An element of biodiversity and biogeochemical
282
Greenhouses, Microcosms, and Mesocosms
the genome of each species. Even when spatially well
bounded, ecosystems are not closed. At the very least,
they are subject to energy input and energy and materials
exchange with adjacent ecosystems. Often, ecosystems
demonstrate biotic exchange with adjacent ecosystems
that can be complex and include reproductive and seaso
nal phases.
The development of an ecosystem in a greenhouse
implies that the ecosystem is solar driven, and thus no
deeper than the photic zone of the ocean, although
the basic principles discussed could generally apply to
deep ocean ecosystems. Greenhouse placement generally
requires spatial limitation, and scaling of model to
analog is necessary for many physical and biotic factors.
In some cases, it is intuitive, and in others must be
empirically demonstrated by trial and error in compar
ison with analog function. All of the following case
studies demonstrate aspects of this necessary scaling
exercise. Normal, biogeochemical, and biotic exchange
with adjacent ecosystems must also be simulated.
The reasons for placing or developing ecosystems
within greenhouses for research or educational purposes
have varied enormously and have ranged from the
strongly funded, multiscientist research endeavors, to
the classroom aquarium or terrarium. Even the naming
of the research field of endeavor has varied widely: to
name a few, synthetic ecology, ecological engineering,
controlled ecology, closed systems ecology, ecosystem
modeling, etc. The systems themselves are called living
systems models, microcosms, mesocosms, macrocosms,
ecotaria, living machines, closed ecological life support
systems (CELSS), etc.
In this article, we limit our discussions to those serious
research efforts in which significant effort is expended to
match biodiversity, food web and symbiotic relationships,
as well as biogeochemical function to analog wild ecosys
tems. By definition, such systems cannot be closed;
however, the known interchanges, biotic and biogeo
chemical, with adjacent ecosystems must be known,
studied in the wild, and be simulated so that the essential
functional characteristic of the analog ecosystem can be
maintained.
Hundreds, perhaps thousands, of microcosm studies
of liter or few liter dimensions of a very limited bio
diversity have been undertaken to elucidate component
ecosystem function, often related to toxic compound
effect. Rarely could these studies be regarded as the
modeling of an ecosystem. At the other extreme, per
haps the most complex ecosystem modeling effort ever
undertaken was the Biosphere II project in Arizona
during the 1980s and 1990s. Biosphere II was an
ecologically well conceived collection of interacting
terrestrial marine and freshwater ecosystems. However,
it was intentionally operated as a closed system because
of its planned space station future. Several decades ago,
it was widely regarded among ecologists that even
though greenhouse enclosure provided a critical ele
ment of control over variables, the difficulties inherent
in enclosure and operation were too great to allow
ecosystem modeling. As the examples we provide
below show, this judgment was only minimally correct
and perhaps no more severe than the breakup of wild
ecosystems by development or farming expansion.
Human expansion and perturbation has severely altered
many of the ecosystems on Earth, and has altered all
ecosystems in at least minor ways. The entire biosphere
has been in effect placed in a poorly operated green
house, with the atmosphere serving as its upper ‘glass’
roof. There can be no valid argument against greenhouse
enclosure of ecosystems for research and education. It is
simply one end of a complex spectrum of interacting
biota and biogeochemistry that we seek to understand.
Indeed, in many ways, such model ecosystems may be
‘purer’ than their wild counterparts.
Physical/Chemical Control Parameters
The Enclosure
The shape of an ecosystem relative to its controlling
physical and energy parameters can be crucial. In the
case of aquatic systems, the relative thickness of the
water mass and its relationship to the bottom establish
the basic character of an ecosystem. A large body of water
would be dominated by true plankters, normally living
most of their lives suspended in mid and surface waters,
with little benthic (or bottom) influence, whereas the
shallow stream or narrow lagoon of a few meters in
depth is benthic dominated. Light enters only through
the air–water interface of a water ecosystem, and the
shape of the containing body of water relative to depth,
as well as water turbidity, determines the photosynthetic
versus heterotrophic character of the ecosystem. The
direction of current flow and wave action through an
aquatic system relative to the position and orientation of
its communities is critical to simulate in any model. The
direction, frequency, and strength of wind relative to
forest or field size can also be critical to systems function,
as can be the physical dimension and density of such
ecosystems.
The all glass or acrylic aquarium box, ranging from
about 40 l (10 gallons) to 1000 l (250 gallons), is a standard
piece of equipment in terrestrial and aquatic modeling,
and by drilling holes to attach pipes and linking all glass
tanks in complex arrays, many aspects of wild ecosystems
can be modeled with reasonable accuracy.
Greenhouses, Microcosms, and Mesocosms
The construction of molded fiberglass tanks or poured
concrete or concrete block tanks, sealed with a wide
variety of newer sealants, has considerable advantages
for larger systems.
Ideally, the ecosystem envelope would be like that of the
boundary of the mathematical modeler, a theoretical
boundary allowing the controlling of exchange but not
having any inherent characteristics. Walls, whatever their
nature, unless rather esoteric measures are used to prevent
organisms and organic molecules from using their surfaces,
or blocking wind or current, are intrusions into the model
ecosystem that may or may not be acceptable. For a small
model of a planktonic system, the presence of uncleaned
walls may prevent the system from being plankton domi
nated. To some degree, walls also interact with the water
and atmosphere of the ecosystem they contain. For most
purposes, glass and many plastics are ideal in this respect.
Greenhouse walls and roofs can block ultraviolet light
and, for most ecosystem models, a component of artificial
light is probably essential to achieve both the intensity
and spectral veracity of natural light. Reinforced cement
block or concrete can be valuable construction materials
for large systems; however, concrete interacts with both
water and atmosphere, being one of the limiters of vera
city in of Biosphere II (as we describe below), and must be
sealed with epoxy or other, carefully considered resins
(Figure 1).
Many chemical elements and compounds used in con
struction are toxic. Some of these are only mildly
poisonous and are often required by organisms as ele
ments in small quantities and only become toxic in excess.
Others are always toxic and only concentration deter
mines effect. Glass, acrylics, epoxies, polyesters,
Figure 1 Florida Everglades mesocosm during construction.
The butyl rubber-lined concrete block walls were used to
constrain the entire system as well as to physically separate the
salinity subcomponents thereby creating a salinity gradient. The
plastic box at the lower right contains the tide controller, which
determines the tide level in the estuary (center tank and the four
smaller units behind).
283
polypropylenes, polyethylenes, nylons, Teflon, and sili
cones, among others, are structural materials commonly
used in model/greenhouse construction. When properly
cured these materials are generally inert, nonbiodegrad
able, and nontoxic. Many metals and organic additives
easily find their way into construction processes and must
be avoided or sealed off.
Physical/Chemical Environment
Many of the physical/chemical parameters of ecosystem,
such as temperature, salinity, pH, hardness, and oxygen,
are more or less obvious and generally accepted as crucial.
Others such as light, wind, tides, currents, and wave
action, have often been neglected or at least minimally
considered in their effects on ecosystem models.
Light
Whole communities or parts of ecosystems, where plants
are major components, typically capture a maximum of
6% of the incident light energy in photosynthesis.
Nevertheless, full light is often required to achieve that
peak transfer of energy. Also much higher capture effi
ciencies are possible when forcing energy such as wind
and wave are present. In many cases, if greenhouse roofs
cannot be opened, artificial lighting will have to be intro
duced to achieve the correct spectrum and intensity to
drive the primary production characteristics of an analog
ecosystem.
Water Supply/Water Environment
Whether a terrestrial or aquatic ecosystem is planned, the
supply and internal transfer of water is critical. Air and
water handling systems need to be carefully designed to
prevent water contamination. Since water sequestration
and loss is more or less inevitable, the water quality of
both initial water and later top ups must be carefully
controlled. Rarely would tap water be acceptable. Water
is the universal solvent, whether in liquid or gaseous form,
and often ‘sequesters’ gases. Most ecosystems in green
houses require the dedicated monitoring and control of
atmospheric and water quality. Managed aquatic plant
systems, such as algal turf scrubbers (ATSs), have been
successfully used to manage water quality of adjacent
ecosystems interaction, as we describe in some of the
examples. Such systems can also control atmospheric
quality (Figure 2).
Water and Air Movement
In virtually all water ecosystems, the water flows, and in
most shallow water systems it oscillates (surges) as well.
In models, this flow and surge are developed, at least
284
Greenhouses, Microcosms, and Mesocosms
Biotic Parameters
Ecosystem Structuring Elements
Figure 2 Red mangrove community in the Florida Everglades
mesocosm from the engineering control pad. The large fan in the
upper center provides wind for the mangrove communities. The
box in the lower right is one of a bank of five algal turf scrubbers
(ATSs) that control water quality (nutrients, pH, O2) in the coastal
system.
Some communities of organisms are structured by phy
sical elements – a sandy beach, or rock, for example.
However, in most terrestrial environments and in many
shallow aquatic environments, plants and algae are the
structuring elements. They not only provide the food
and water and atmospheric chemistry but also greatly
increase surface for attachment and cover. In general,
plants also provide a spatial heterogeneity (spatial sur
face) that does not exist in the physical world.
Particularly in the marine environment, where calcifi
cation is enhanced, many animals join plants to provide
a community structure that consists of reef or shell
framework. This framework is calcium carbonate (or
other organic solid such as chiton) instead of (or
along with) plant cellulose. In constructing any living
ecosystem, it is essential that these structuring elements
be first developed as ‘colonial’ stages soon after the
physical environment is formed.
Ecosystem Subunits
Figure 3 Engineering/control pad in the Florida Everglades
mesocosm. The green diagonal tube in the center is an
Archimedes screw that lifts water from the coastal tank
(far right) for distribution to the estuary (back right), the ATS
(left-foreground), and the wave generator (out of view to
lower right).
initially by pumps. However, standard impellor pumps
destroy or damage many plankters, particularly larger
zooplankton and swimming invertebrate larvae. Several
approaches are available to solve this problem, including
using slow moving piston pumps, membrane pumps, and
Archimedes screws (Figure 3). All of these devices can
work well, though relative performance is not fully
quantified.
In terrestrial environments, fans for wind and air
handlers for heat and air conditioning, as well as the
cooling or heating surfaces employed have the same
effect on flying insects and birds. On the other hand, in
the wild, ultraviolet light, wind, and rain have critical
controlling effects on many plant predators. These fac
tors cannot be omitted.
In the construction of greenhouse ecosystems, subunit
installation can be utilized. However, it would be impos
sible to individually extract and emplace the tens to
hundreds of species amounting to hundreds to millions
of individuals that occur in these subunits. Installation of
sub blocks of wild ecosystem includes the microspecies
and keeps their relationships intact. For example, soil
blocks, or in the marine or aquatic habitat, mud or rock
blocks, can be introduced into the preexisting physical/
chemical elements of the model ecosystem.
Repeated efforts must be taken to install rock, soil,
mud, or ‘planktonic blocks’. These injections should be
periodically carried out during system stocking; at com
pletion of development, they should be followed by
several final injections. The process of cutting out, or
otherwise extracting, an ecological block or ecosystem
subunit and transporting it to the waiting model can be
stressful to the community of organisms within the block.
Even in the model, the block meets conditions that at least
initially consist of the raw physical/chemical environ
ment unameliorated by the effects of a functioning
community of organisms. The first block injections are
likely to lose species. However, with each addition, the
diversity of reproductively successful species increases.
All ecological communities are patchy. An island, coral
reef, a large salt marsh, a field, even a forest, all differ from
place to place. Chance factors of organism settlement,
negative and positive interaction between species, the
local effect of environment, and real differences of envir
onment (wave exposure, current, etc.) all lead to
patchiness within a community. The model itself, no
Greenhouses, Microcosms, and Mesocosms
Figure 4 Salt marsh community 1 year after establishment of
the Florida Everglades mesocosm. Young white mangrove tree
to left. After 5 years of self-organization the system followed a
succession to a white mangrove/buttonwood swamp.
matter how accurate, is a patch, or several patches, that the
modeler hopes represents a ‘mean’ of most wild patches.
After the structuring elements are established, and
the entire pool of available species from the type com
munity given a chance at immigration into the model,
the model will self organize. In the form of the genetic
codes of its constituent species, the ecosystem carries a
tremendous quantity of information with regard to its
structure and function. Particularly since we know and
understand only a small part of this information, we
should be loath to subvert ecosystem self organization
(Figure 4).
Care in adhering to wild density levels will help
prevent overstocking, overgrazing, and overpredation
until the model is better understood. Single members
of species guilds can be selected to perform a function,
and thus reproductive density is achieved without
exceeding ecosystem density requirements. In general,
the larger the population of any single species, the more
likely it is that breeding success will be achieved.
Ecosystem Interchange
However, one arbitrarily draws the boundary, no ecosys
tem occurs in total isolation. In many cases when such
boundaries are arbitrary, major survival effects must be
provided by adjacent ecosystems. For example, coral reefs
and most shallow benthic communities are greatly depen
dent on the effects of adjacent open bodies of water for
food, oxygen, and wave and current ‘drive’. Typically,
filters, the core element of aquarium science, have been
devised to fill the need for the larger, less animal dense
body of adjacent open water that has been ‘filtered’ by the
settling or loss of organic particles to deep water.
However, such filters to a large extent usurp the role
285
normally provided by plants in most of the communities
that are modeled. Unfortunately, in so doing they do not
add oxygen as the plants do, and they raise nutrient levels.
Both bacterial and foam fractionation methods remove
organic particulates and swimming plankters, including
reproductive stages that should be part of ecosystem
function. Managed aquatic plant systems, such as ATSs,
have been successfully used to manage water quality of
adjacent ecosystems interaction, as we describe in some of
the examples.
Although terrestrial systems, in general, may be less
difficult in this regard, simulation of biotic interchange
may be crucial. For example, birds and mammals often
change ecosystems seasonally and even diurnally and the
effects may be critical. Many insects are seasonal, some for
very short periods, and often cross ecosystem boundaries.
In some cases, it may be possible to provide these inter
actions through a human manager; however, a refugium,
or alternate ecosystem may be necessary to achieve
veracity.
The Operational Imperative
Successful enclosed ecosystem operation requires the
monitoring of a large number of physical and chemical
factors. To a large extent, this can be automated with
electronic sensors, and the data can be logged and the
system computer controlled. Some chemical parameters
require wet chemistry, though a once a week analysis is
usually sufficient in a well run system. Like any piece of
complex laboratory equipment (a scanning electron
microscope, for example) a dedicated and highly trained
technician is needed to manage the monitoring equip
ment, though in a well tuned system, considerable time
can be available for other duties.
An operational feature that is rarely discussed, and in
practice is mostly anecdotal is that of population instabil
ity. A mesocosm, in effect, is a few square meter patch of
a larger ecosystem. In the wild, ecosystem patches of a few
square meters can be subject to considerable short term
variability, though stability is achieved to some extent by
the smoothing effect of the larger ecosystem that may be
measured in square kilometers.
Microcosms and mesocosms require an ecologist, fully
acquainted with ‘normal’ community structure of the ‘wild
type’ system. Effectively, that ecologist/operator performs
as the highest, and most omnivorous, predator. In the cases
of algal or insect ‘explosions’ the operator’s function is
obvious, a once a week cropping or ‘grazing’ (i.e., hand
harvest) until the explosion tendency subsides. In other
cases, the short term introduction of predator to carry out
the limited cropping or grazing role can be quite successful.
286
Greenhouses, Microcosms, and Mesocosms
These ‘managed predators’ can be kept in a refugium unit
where they are readily available for such service.
(nutrients, oxygen, and pH) of the system was controlled by
algal turf scrubbing.
The mean oxygen concentration of the microcosm as
shown in Figure 6 is very close to that of the analog
St. Croix reef. Net primary productivity (NPP) and
respiration (R) were calculated based on the rate of oxy
gen increase and decrease, respectively, across the point
of saturation (6.5 mg 1 1 O2), to avoid atmospheric fluxes.
This gave a mean gross primary productivity (GPP) of
Case Study: Coral Reef Microcosm
The Caribbean coral reef ecosystem model shown in
Figure 5 received natural sunlight from one side, south
facing at 37.5 N latitude; the metabolic unit had six
160 W VHF flourescent lamps (to match tropical inten
sity), step cycled to bring mid day peak intensity to
approximately 800 uE m 2 s 1 and total incoming light
to 220 Langleys/day (Figure 6). The ATS, lighted at
night, had three 100 W metal halide lamps. The discus
sion presented represents data accumulated throughout
the 9th year of 10 years of operation.
The physical and chemical components of the micro
cosm were measured in the metabolic unit and closely
match those of the St. Croix analog (Table 1). The pH of
the microcosm ranges from 7.96 0.01 (n ¼ 62) in the morn
ing to 8.29 0.10 (n ¼ 39) in the late afternoon. Because of
linked interacting photosynthesis and calcification in the
ecosystem, calcium concentrations and alkalinity continu
ally fall during the day and are stable or rise slightly at night.
Calcium was added each morning as a solution of aragonite
dissolved in HCl at approximately 24 000 mg l 1. To keep
microcosm concentrations above 420 mg l 1, after a full day
of calcification, the mean concentration of calcium in the
system was maintained at 491 6 mg l 1. Bicarbonate, was
added as either NaHCO3 or KHCO3 dissolved in distilled
water. The mean alkalinity was 2.88 meq l 1 (n ¼ 59), in
order to maintain levels above 2.40 meq l 1. Water quality
Microcosm vs. St. Croix
9.5
9.0
Microcosm
8.5
St. Croix
Oxygen (mg l–1)
8.0
7.5
7.0
6.5
6.0
5.5
5.0
0600
1000
1400
1800
2200
0200
0600
0800
1200
1600
2000
2400
0400
Time
Figure 6 Comparison of mean daily oxygen concentration in
the coral reef microcosm in comparison with that over the wild
analog reef (1-year means).
Experimental coral reef microcosm (5.0 m2; 1680 l)
Export of
dried algae
Import mixed
feed (0.41 g d 1)
Bellows
pumps
8l
refugium
from fish
Manual transfer
of organisms
Algal turf
scrubber
Lighting: six 400 W metal halides
Wave
bucket
Algal turf
scrubber
Wave
surge
Reef
Patch
sand
Import seawater: 2 l d
1
Unit for metabolic work
(0.757 m2; 400 l)
Figure 5 Diagram of coral reef microcosm with its refugium.
Manual transfer of
exchange water:
2ld 1
Refugium unit
(4.29 m2; 1280 l)
Export
water:
2ld 1
Greenhouses, Microcosms, and Mesocosms
287
Table 1 Comparison of physical/chemical parameters between coral reef microcosm and the wild analog reef
Temperature ( C) (am–pm)
Salinities (ppt)
pH (am–pm)
Oxygen concentration (mg l 1) (am–pm)
GPP (g O2 m 2 d 1); (mmol O2 m 2 d 1)
Daytime NPP (g O2 m 2 day 1); (mmol O2 m 2 d 1)
Respiration (g O2 m 2 h 1); (mmol O2 m 2 h 1)
N NO2 þ NO23 ðmmolÞ
Calcium (mg l 1); (mmol l 1)
Alkalinity (meq l 1)
Lighte (Langleys d 1)
Microcosm
St. Croix Reefs (fore-reef)a
26.5 0.03 (n 365)–27.4 0.02 (n 362)
35.8 0.02 (n 365)
7.96 0.01 (n 62)–8.29 0.02 (n 39)
5.7 0.1 (n 14)–8.7 0.2 (n 11)
14.2 1.0 (n 4); 444 3 (n 4)
7.3 0.3 (n 4); 228 9 (n 4)
0.49 0.04 (n 4); 15.3 1.3 (n 4)
0.56 0.07 (n 6)
491 6 (n 33); 12.3 0.2 (n 33)
2.88 0.04 (n 59)
220
24.0–28.5
35.5b
8.05–8.35c
5.8–8.5
15.7; 491
8.9; 278
0.67; 20.9
0.28
417.2d; 10.4
2.47b
430 (surface); 220 (5 m deep in
fore-reef)
a
The St. Croix data is from Adey and Steneck (1985).
Tropical Atlantic means from Millero and Sohn (1992); no data available for St. Croix.
Values from Enewetak and Moorea (Odum and Odum, 1955; Gattuso et al., 1997).
d
Tropical Atlantic means from Sverdrup et al. (1942); no data available for St. Croix.
e
The light levels of the system were measured with a pyranograph. All of the physical and chemical components of the microcosm are compared to the
fore reef of St. Croix since light levels are equivalent (Kirk, 1983; Adey and Steneck, 1985).
For references, see Small A and Adey W (2001) Reef corals, zooxanthellae and free living algae: A microcosm study that demonstrates synergy
between calcification and primary production. Ecological Engineering 16: 443 457.
b
c
200
Microcosm
St. Croix
fore-reef
100
0
0
100
200
Respiration (mol C m–2 yr–1)
300
Figure 7 GPP as a function of respiration in the coral reef
microcosm and its wild analog reef in comparison with selected
worldwide reefs.
Total alk. and HCO3 mg l–1
GPP (mol C m–2 yr–1)
300
95
20
93
18
91
16
89
12
87
HCO3–
85
10
83
8
81
6
=
CO3
79
4
CO2
77
2
8.02
80
0
75
60
0
14.2 1.0 gO2 m 2 d 1, as compared to the mean GPP for
the analog fore reef at 15.7 gO2 m 2 d 1. The difference
between the microcosm and reefs in situ can be accounted
for by the difference in spatial heterogeneity; topographic
relief on the St. Croix fore reef typically ranges from 1 to
2 m, while in the microcosm only 10–30 cm is possible.
In Figure 7, GPP versus R for the microcosm and its
analog are plotted, showing that both are well within the
range of typical wild reefs. Even though primary produc
tivity of the microcosm is very close to the wild analog,
the fact that respiration is somewhat lower probably
relates to the proportionally lower spatial heterogeneity
in the microcosm.
Whole ecosystem calcification in the coral reef model,
at 4.0 0.2 kg CaCO3 m 2 yr 1 , is related to its primary
components (stony coral 17.6%, Halimeda 7.4%, Tridacna
9.0%, algal turf, coralline and foraminifera 29.4%, and
miscellaneous invertebrates 36%). Through analysis of
14
Total alkalinity
7.97
0
0
10
CO3 and CO2 mg l–1
Microcosm
400
0
0
12
0
0
14
8.17
0
0
16
8.29
0
0
18
0
0
0
20
8.24
Time pH
Figure 8 Mean daytime carbonate cycle in coral reef microcosm
calculated by namograph from pH and total alkalinity data.
the microcosm’s daily carbonate system, it is demon
strated that bicarbonate ion (Figure 8), not carbonate
ion, is the principal component of total alkalinity reduc
tion in the water column.
This coral reef microcosm contained 534 identified spe
cies within 27 phyla (Table 2), with an estimated 30%
unaccounted for due to lack of taxonomic specialists.
Because of the length of time that this model system was
closed to biotic interchange, virtually all of the biotic com
position of the system (over 95%) had to be maintained by
288
Greenhouses, Microcosms, and Mesocosms
Table 2 Families of organisms, with numbers of species and
genera found in the Coral reef microcosm after 10 years of
operation, 7 years in closure
Plants, algae, and cyanobacteria
Division Cyanophota
Chroococcaceae 6/5
Pleurocapsaceae 4/2
UID Family 4/4
Stephanopogonidae 2/1
Phylum Euglenozoa
UID Family 4/3
Bondonidae 7/1
Phylum Choanozoa
Codosigidae 2/2
Salpingoecidae 1/1
Phylum Rhizopoda
Oscillatoriaceae 8/6
Acanthamoebidae 1/1
Rivulariaceae 4/1
Hartmannellidae 1/1
Scytonemataceae 1/1
Hyalodiscidae 1/1
Mayorellidae 2/2
Phylum Rhodophyta
Goniotrichaceae 2/2
Acrochaetiaceae 2/2
Reticulosidae 2/2
Gelidiaceae 1/1
Thecamoebidae 1/1
Wurdemanniaceae 1/1
Trichosphaeridae 1/1
Peysonneliaceae 3/1
Vampyrellidae 1/1
Corallinaceae 11/8
Allogromiidae 1/1
Hypneaceae 1/1
Ammodiscidae 1/1
Astrorhizidae 1/1
Rhodymeniaceae 3/2
Saccamoebidae 1/1
Champiaceae 1/1
Ceramiaceae 3/3
Ataxophragmiidae 1/1
Delesseriaceae 1/1
Cibicidiidae 1/1
Rhodomelaceae 7/6
Cymbaloporidae 1/1
Phylum Chromophycota
Bolivinitidae 3/1
Discorbidae 5/2
Cryptomonadaceae 2/2
Homotremidae 1/1
Hemidiscaceae 1/1
Peneroplidae 1/1
Miliolidae 10/2
Diatomaceae 6/4
Naviculaceae 9/4
Cymbellaceae 3/1
Planorbulinidae 2/2
Entomoneidaceae 1/1
Soritidae 4/4
Nitzchiaceae 6/4
Epithemiaceae 3/1
Siphonidae 1/1
Textulariidae 1/1
Phylum Ciliophora
Mastogloiaceae 1/1
Kentrophoridae 1/1
Achnanthaceae 9/3
Blepharismidae 2/2
Condylostomatidae 1/1
Gymnodiniaceae 6/4or5
Gonyaulacaceae 1/1
Prorocentraceae 2/1
Folliculinidae 4/3
Zooxanthellaceae 1/1
Protocruziidae 2/1
Ectocarpaceae 2/2
Aspidiscidae 7/1
Phylum Chlorophycota
Peritromidae 2/1
Chaetospiridae 1/1
Ulvaceae 1/1
Discocephalidae 1/1
Cladophoraceae 4/2
Euplotidae 11/3
Keronidae 7/2
Valoniaceae 2/2
Derbesiaceae 3/1
Caulerpaceae 3/1
Oxytrichidae 1/1
Codiaceae 6/2
Ptycocyclidae 2/1
Colochaetaceae 1/1
Spirofilidae 1/1
Phylum Magnoliophyta
Hydrocharitaceae 1/1
Kingdom Protista
Phylum Percolozoa
Vahlkampfiidae 2/1
UID Family 2/2
Psilotrichidae 1/1
Strombidiidae 1/1
Uronychiidae 2/1
Urostylidae 4/2
Cinetochilidae 1/1
Cyclidiidae 3/1
Pleuronematidae 3/1
(Continued )
Greenhouses, Microcosms, and Mesocosms
289
Table 2 (Continued)
Zoanthidae 3/2
Uronematidae 1/1
Cerianthidae 1/1
Phylum Platyhelminthes
Vaginicolidae 1/1
Vorticellidae 2/1
Parameciidae 1/1
UID Family 1/1
Colepidae 2/1
Anaperidae 3/2
Metacystidae 3/2
Prorodontidae 1/1
Nemertodermatidae 1/1
Kalyptorychidae 1/1
Phylum Nemertea
Amphileptidae 3/3
Enchelyidae 1/1
UID Family 2/2
Lacrymariidae 4/1
Micruridae 1/1
Lineidae 1/1
Phylum Heliozoa
Phylum Gastrotricha
Actinophyridae 2/1
Chaetonotidae 3/1
Phylum Placozoa
Phylum Rotifera
Family UID 5
Phylum Porifera
UID Family 2/?
Phylum Tardigrada
Plakinidae 2/1
Batillipedidae 1/1
Geodiidae 5/2
Tetillidae 1/1
Phylum Nemata
Draconematidae 3/1
Suberitidae 1/1
Phylum Mollusca
Pachastrellidae 1/1
Spirastrellidae 2/2
Acanthochitonidae 1/1
Clionidae 4/2
Tethyidae 2/1
Fissurellidae 2/2
Chonrdrosiidae 1/1
Trochidae 1/1
Axinellidae 1/1
Turbinidae 1/1
Agelasidae 1/1
Haliclonidae 4/1
Phasianellidae 1/1
Neritidae 1/1
Oceanapiidae 1/1
Rissoidae 1/1
Mycalidae 1/1
Rissoellidae 1/1
Dexmoxyidae 1/1
Halichondridae 2/1
Vitrinellidae 1/1
Clathrinidae 1/1
Phyramidellidae 1/1
Leucettidae 1/1
Fasciolariidae 2/2
UIDFamily 2/?
Olividae 1/1
Marginellidae 1/1
Acmaeidae 1/1
Vermetidae 1/1
Eumetazoa
Mitridae 1/1
Phylum Cnidaria
UID Family 3/?
Bullidae 1/1
UID Family 4/?
Eudendridae 1/1
Mytilidae 2/1
Olindiiae 1/1
Arcidae 2/1
Plexauridae 1/1
Glycymerididae 1/1
Anthothelidae 1/1
Isognomonidae 1/1
Limidae 1/1
Briareidae 1/1
Alcyoniidae 2/2
Pectinidae 1/1
Actiniidae 3/2
Aiptasiidae 1/1
Chamidae 1/1
Lucinidae 2/2
Stichodactylidae 1/1
Carditidae 1/1
Actinodiscidae 4/3
Tridacnidae 2/1
Corallimorphidae 3/2
Tellinidae 1/1
Acroporidae 2/2
Phylum Annelida
Syllidae 3/2
Caryophylliidae 1/1
Faviidae 3/2
Amphinomidae 1/1
Eunicidae 3/1
Mussidae 1/1
Poritidae 3/1
(
)
(Continued )
290
Greenhouses, Microcosms, and Mesocosms
Table 2 (Continued)
Lumbrineridae 1/1
Dorvilleidae 1/1
Orbiniidae 1/1
Diogenidae 1/1
Xanthidae 2/?
Phylum Echinodermata
Ophiocomidae 1/1
Spionidae 1/1
Ophiactidae 1/1
Chaetopteridae 1/1
Cidaroidae 1/1
Paraonidae 1/1
Cirratulidae 4/3
Toxopneustidae 1/1
Ctenodrilidae 4/3
Capitellidae 3/3
Holothuriidae 1/1
Chirotidae 1/1
Phylum Chordata
Oweniidae 1/1
Ascidiacea UID Fam..1/1
Grammidae 1/1
Terebellidae 2/1
Chaetodontidae 1/1
Sabellidae 14/4
Pomacentridae 5/4
Serpulidae 6/6
Spirorbidae 2/2
Acanthuridae 1/1
Muldanidae 1/1
Dinophilidae 1/1
Phylum Sipuncula
Golfingiidae 1/1
Phascolosomatidae 3/2
Phascolionidae 1/1
Aspidosiphonidae 3/2
Phylum Arthropoda
Halacaridae 1/1
reproduction. Based on standard species/area relationships
(S ¼ kAz, where S ¼ species richness and A ¼ area), the pre
dicted pan tropic coral reef biodiversity calculated from the
model biodiversity (at three million species) exceeds that of
recent estimates for wild coral reefs.
UID Family 2/?
Cyprididae 2/2
Bairdiiaae 1/1
Paradoxostomatidae 1/1
Case Study: Florida Everglades
Mesocosm
Pseudocyclopidae 1/1
Ridgewayiidae 2/1
Ambunguipedidae 1/1
Argestidae 1/1
Diosaccidae 1/1
Harpacticidae 1/1
Louriniidae 1/1
Thalestridae 1/1
Tisbidae 1/1
Mysidae 1/1
Apseudidae 2/1
Paratanaidae 1/1
Tanaidae 1/1
Paranthuridae 1/1
Sphaeromatidae 1/1
Stenetriidae 1/1
Juniridae 1/1
Lysianassidae 1/1
Gammaridae 4/4
Leucothoidae 1/1
Anamixidae 1/1
Corophiidae 1/1
Amphithoidae 2/2
Alpheidae 2/2
Hippolytidae 2/1
Nephropidae 1/1
This greenhouse scale mesocosm is a 98 500 l butyl lined,
concrete block tank divided into seven connected sections
of varying salinity (Figure 9). Each section contains
water, algae, animals, sediments, and wetland–coastal
plants representative of habitats along a transect from
the full salinity Gulf of Mexico through the estuarine
Ten Thousand Islands and into the freshwater Florida
Everglades (Figure 10).
As in the wild analog, the Gulf Shore and estuary
are part of the same dynamic water mass. Here, the
estuarine salinity gradient is created by pump driven
tidal inflow interacting through open weir constrictions
and against downstream freshwater flow. Tank #1, the
Gulf Shore, acts as a tidal reservoir for the estuary,
thereby saving the need for a blank reservoir
(Figure 11). The primary pump is an 800 lpm,
DiscfloTM unit that utilizes a rotating disk, rather
than plankton destructive impellers. The freshwater is
derived from rain and from reverse osmosis extraction
from the Gulf Shore (the equivalent of Gulf evapora
tion and resulting rainfall in the wild). All aquatic
organisms, including adult invertebrates, can move
from the estuary to the Gulf Shore. All organisms
that can survive DiscfloTM pumping (including small
fish) can return to the estuary via tidal inflow. The
freshwater system, at times, flows directly into the
291
Greenhouses, Microcosms, and Mesocosms
Freshwater stream
Upper pool
R. O. filter
Pinus/palm hammock
Taxodium
swamp
#7
Annona/
Chrysobalanus
swamp
Rhizophora
‘island’
ia
nn
Tidal
stream
Av
#3
Cr
s/ Th
beac espia /S
e
h rid
ge suvium
R
hi
z
‘is op
la ho
nd ra
’
Rhizophora
‘island’
Tidal
controller
as
s
Be ost
d rea
gu
sw nc
am ula
p ria
#4
Tidal gate
Rhizophora
‘island’
Freshwater pond
a
ph
Ty rsh
ma
Typha marsh
#6
La
#2
bolu
Spo
ro
Hyp
Calo
Thalassia / Halodule bed
nea
/Ca
Wormulerpa / U
lva
reef
thrix
/She
ll be
ach
Refugia
Pumping sump
#1
Swietenia/
Quercus/Persea
hammock
Ludwigia/Zizaniopis/Pandanus
marsh
Cladium
prairie
Sa ix/Taxodium
swamp
ice
Refugia
Discflo
pump
Wave generator
Cladium
prairie
Juncus/Setaria/
Leersia prairie
Eleochains
prairie
Seawater
distribution
tower
Laguncularia/Myrica
swamp
Scrubber battery
R. O. unit
Freshwater
scrubber battery
Avicennia /Sesuvium
swamp
Laguncularia/
Achrostichum
swamp
#5
Laguncularia /Juncus/
Achrostichum swamp
Figure 9 Plan view of the Florida Everglades mesocosm and its critical engineering components.
Figure 10 Florida Everglades mesocosm approximately
4 years after construction showing salt marsh, black mangrove,
and red mangrove communities (from front to background at left)
and lower freshwater stream at right. At this point the greenhouse
roof is providing a significant constraint to community
succession by limiting vertical growth of mangrove and
hammock trees.
uppermost estuary and technically all organisms can
enter the estuary from freshwater.
The initial stocking of the mesocosm was completed
in mid 1988, and small collections continued to be
injected through 1990. During this period, partial cen
suses for key organisms were undertaken, and, where
required, additional stocking was carried out. From late
1990 to late 1994, the system was operated as a
biotically closed system, with minor human interaction,
functioning as an omnivorous predator.
Major physical/chemical parameters are shown in
Table 3. Dissolved nitrogen was monitored as
nitrite plus nitrate in each of the community units
(tanks); these were typically at levels of 5–8 mM
(NO2 þ NO3) through the middle of the estuary, and
at 3–5 mM (NO2 þ NO3), in the Gulf (#1) system.
Levels average a few mM higher in winter than in
summer. Nutrient flow through is achieved by
algal export, in the ATS banks. When levels drop
below 1–2 mM in the Gulf (#1) system (typically in
summer), the dried scrubber algae are redistributed to
the system.
After 4 years of biotically closed operation, the Florida
Everglades mesocosm was censused for organisms. The
abundance of the principal higher plants, algae, inverte
brates, and fish are shown in Figures 12–14. A total of 369
species (not including bacteria, fungi and the minor
‘worm’ phyla) was tallied. Excluding algae, protists, and
small invertebrates, which could not be censused during
introduction, it can be estimated that approximately
20–40% of the introduced species survived through the
4 years of biotic closure. In most cases these were the
dominating species in the analog ecosystems. At the time
of termination of the system as a carefully monitored
mesocosm, only 15–30% of the originally introduced
species were reproductively maintaining populations.
However, in most cases, as Figures 12–14 show, these
292
Greenhouses, Microcosms, and Mesocosms
‘Optional’
Freshwater return
R. O.
Freshwater
scrubbers
Rain
0.1’ d –1, dry season
–1
0.3’ d , wet season
Distribution tower
Brine
Electronic
tidal control
and gate valve
Upper
pool
Open
weir gates
Spring tidal ranges (cm)
Tide Return
Seasonal
(30)
Impeller
pumps
(4)
(20)
(10)
(20)
Gulf shore
scrubbers
Disc
flo
pump
(26)
Wave
generator
(26)
7
System
6
Freshwater
5
4
Upper estuary
3
2
Lower estuary
1
Gulf Shore
Figure 11 Vertical/longitudinal section through Florida Everglades mesocosm showing water management system and tide levels.
Table 3 Physical/chemical parameters of the Florida Everglades mesocosm
Parameter
Temperature C
Spring
Summer
Fall
Winter
Salinity, ppt
[NO2 þ NO3] mM
Tap H2O as top upa
Milli RO as top upb
Tidal range cm/0.5 day
Hydroperiod cm yr 1
Tank #1
Tank #2
Tank #3
Tank #4
Tank #5
Tank #6
Tank #7
23.4
25.7
22.2
21.0
31.6
23.2
25.5
22.3
20.9
31.2
22.5
25.4
21.8
19.9
30.5
22.6
25.7
22.0
19.6
28.7
22.0
25.6
21.7
19.0
19.7
21.3
25.1
21.3
18.4
0.7
23.2
25.1
22.1
21.9
0.1
7.2
1.4
13–26
0
7.6
1.7
13–26
0
8.2
2.3
13–20
0
6.3
1.8
11–20
0
5.4
0.9
6–10
0
6.6
1.7
0–4
0
6.7
1.4
0
30.5
a
Nutrient levels in system as (NO2 þ NO3) mM when ‘Tap H2O as top up’.
‘Milli RO as top up’ refers to mean system values when reverse osmosis water from Milli ROä is used as evaporative replacement.
b
were the species that provided primary structure and
metabolism in the analog ecosystem.
Case Study: Biosphere 2
Biosphere 2, located near Tucson, AZ, USA, is the
largest greenhouse system ever built with nearly three
acres (1.2 ha) of enclosed space. It is unique in surpass
ing any other greenhouse ecosystem in size,
complexity, and duration of operation. The system
was originally intended as a model of the Earth’s bio
sphere (e.g., biosphere 1) with several tropical and
subtropical ecosystems, an agricultural area, wastewater
treatment wetlands, and a human habitat, along with a
factory sized machinery area for maintaining physical–
chemical conditions. It was built to develop bioregen
erative technology for future space travel, to educate
the public about biosphere scale issues and for basic
ecological research. Atmospheric closure of gas cycles
was part of the system design, which was tested with a
prototype module of 11 000 ft3 (312 m3) from 1988 to
1990.
A number of ecologists were consulted for the
creation of the greenhouse’s ecosystems which included
plots of rainforest, desert, savanna, mangrove estuary,
Greenhouses, Microcosms, and Mesocosms
293
Benthic macroalgae of the Everglades mesocosm
(dominant elements)
100
Cladophora repens
Derbesia spp.
Scytonema
hofmanni
Caulerpa fastigiata
Calothrix sp.
Oscillatoria sp.
Chlorophyta
Caulerpa verticillata
Murreyella periclados
Bostrychia
montagnei
Rhodophyta
Polysiphonia
subtilissima
Acanthophora
spicifera
Callithamnion spp.
Browns and misc. reds
0
System
# Species
1
Gulf
Coast
29
2
Red
mangrove
3
Oyster
bays
4
Black
mangrove
Lower estuary 14
5
Salt
marsh
4
Chlorophyta
Boodieopsis
pusilla
Rhodophyta
Misc. Cladophora Rhizoclonium
crispata
sp.
Rhizoclonium
riparium
Spirogyra
sp.
50
Audouinella
violacea
Relative algal biomass
(%) (Frequency × abundance – fragment/occurs/massive)
Chaetomorpha
gracilis
Cyanobacteria
Microcoleus
calcicola
Chaetomorpha
minima
Misc. greens
7
6
Oligohaline Freshwater
marsh
4
16
Figure 12 Relative biomass of dominant benthic algae in Florida Everglades mesocosm.
and ocean with coral reef. Thousands of species
were added to the greenhouse intentionally and unin
tentionally (i.e., in ecosystem sub blocks, as described
above), from existing tropical systems as distant as
Venezuela and from the local Arizona desert. After
construction the ecosystems self organized and
many of the added species went extinct within the
system as expected. Success, in terms of replication
of the analog ecosystems in nature, has varied among
the different model ecosystems, but most have devel
oped and sustained a significant degree of ecological
integrity.
Two experiments were conducted in Biosphere II
during which humans were enclosed inside the system:
the first for 2 years (1991–93) and the second for 6 months
(1994). These experiments tested concepts of sustainabil
ity at a very basic level since the humans had to rely on
the overall greenhouse system for life support function.
However, changes in the gas cycles within the greenhouse
caused the human experiments to be modified and ulti
mately terminated. During the first human experiment
oxygen concentration in the atmosphere decreased dra
matically because high rates of soil respiration released
more carbon dioxide than was taken up in photosynthesis;
some of the carbon dioxide was absorbed as carbonates in
the concrete of the greenhouse foundation. Oxygen had to
be pumped into the system to maintain the humans so that
the 2 year test could be completed. During the second
294
Greenhouses, Microcosms, and Mesocosms
Macroinvertebrates of the Everglades mesocosm
(dominant elements)
100
Sponges:
Haliclona spp.
Coelenterates:
Viatrix globulifera
cia
tra es
De aoid
ll
u
b
Hydra sp.
Aiptasia pallida
Sipunculids:
Me
lan
oid
es
tu
Ph berc
ys
ell ulata
ac
ub
en
Hypolytus sp.
Phascolion strombi
Mollusks:
sis
la
el
al
Hy
ca
te
az
Codakia orbiculata
(Bivalve)
a
Pa
ell
lch
pu
lae
la
l
ate
ne
gera
te s
ia fili
form
Cirri
ta
cula
a
m
linna
Cerithium lutosum
(Snail)
us
os
ae
lid
pa
arv
id l
mo
no
iro
Ch
Me
Annelids:
Sabella spp.
(filter-feeding polychaete)
50
On detritus and algae
Spirorbus spp.
(filter-feeding polychaete)
Nephtys bucera, Eurythoe complanata
(predatory polychaetes)
Unidentified
Ostracods
Arthropods:
Cypridopsis vidua
On detritus
Relative macroinvertebrate abundance (%)
mo
nc
Tru
Ostracods:
Cypridina sp.
Leptochelia savignyi
Tanaids:
Tanais cavolini
Isopods:
Paracerceis caudata
Cyathura polita
Amphipods:
Gammarus sp.
Corophium sp.
Ampithoe sp.
0
Echinoderms:
Ophiactis sp.
System
# Species
1
Gulf Coast
42
2–4
5
6
Lower
Salt Oligohaline
estuary marsh
29
5
4
7
Freshwater
11
Predominant bottom type:
Hard, including mangrove
Soft
Mixed hard and soft
Figure 13 Relative invertebrate abundance in the Florida Everglades mesocosm.
human experiment, buildup of noxious concentrations of
nitrous oxide in the atmosphere from microbial metabo
lism caused the experiment to be shut down ahead of the
planned schedule.
At least two of the basic principles of ecosystem modeling
discussed in the introduction were violated in this system.
Incoming light was greatly reduced, due to the glass and
significant support structure, resulting in insufficient photo
synthesis and primary productive to balance respiration. This
could have been offset by introducing a subset of highly
efficient photosynthesis (such as provided by an ATS),
using artificial lighting; indeed, some ATS systems were
used, but only as a minor element of control on the ocean
system. Monitored exchange with the external environment
could also have been employed. Also, the concrete as an
atmospheric reactant should have been sealed with a non
reactive material, such as glass or plastic.
Much controversy developed during these human
experiments. Colombia University took over manage
ment of the system from 1996 to 2003. During this time
period the research program changed from human enclo
sure experiments to work on global climate change.
Greenhouses, Microcosms, and Mesocosms
295
Fishes of the Everglades mesocosm
100
Higher predators *
Lepornis spp.
Fundulus grandis
(gulf killi)
(1.0)
ulus
Fund
(2.0)
)
Fundulus chrysotus
s
gno
(lon
similis
(2.0)
illi
ek
Rivulus marmoratus
(rivulus)
Lucania goodei
(bluefin killi)
Cyprinodon variegatus
(sheepshead minnow)
(2.5)
50
Heterandria formosa
(least killi)
Cyprinodontidae
% Individuals in each mesocosm unit
Fundulus confluentus
(marsh killi)
(2.2)
Gambusia affinis
(mosquito fish)
Floridichthys carpio
(goldspotted killi)
(2.0)
(2.5)
Poecilia reticulata
(guppy)
Poeciliidae
Poecilia latipinna (sailfin moity)
(2.3)
(3.3)
Acanthurus coeruleus
0
System
1
Gulf Coast
2, 3, 4
Lower estuary
# Fish in unit
Area of unit (m2)
2
# Individuals per m
# Fish species
94 +
44.7
2.1
8
118 +
34.8
3.4
6
5, 6
Upper estuary
17 +
7.6
2.2
3
11 +
7.6
1.4
2
7
Freshwater
102 +
37.4
2.7
8
* Higher predators
In Gulf:
Opsanus beta (gulf toadfish)
Haemulon macrostomum (spanish grunt)
Lagodon rhomboides (pinfish)
In Lower estuary:
Eucinostomus gula (silver jenny)
Not reproducing
Possibly reproducing or reproduced in past
Ratio of total individuals to reproductive adults
Figure 14 Distribution of fish (% of total) in Florida Everglades mesocosm.
See also: Freshwater Lakes; Freshwater Marshes.
Further Reading
Adey W, Finn M, Kangas P, et al. (1996) A Florida Everglades
mesocosm model veracity after four years of self organization.
Ecological Engineering 6: 171 224.
Adey W and Loveland K (2007) Dynamic Aquaria: Building and
Restoring Living Ecosytems, 3rd edn., 505pp. San Diego: Elsevier/
Academic Press.
Kangas P (2004) Ecological Engineering: Principles and Practice,
452pp. Boca Raton: Lewis Publishers, CRC Press.
Korner C and Arnone J, III (1992) Responses to elevated carbon dioxide
in artificial tropical ecosystems. Science 257: 1672 1675.
Marino BDV and Odum HT (1999) Biosphere 2: Research Past and
Present, 358pp. Amsterdam: Elsevier (Also Special Issue of
Ecological Engineering 13: 3 14).
Osmund B, Aranyev G, Berry J, et al. (2004) Changing the way we think
about global change research: Scaling up in experimental ecosystem
science. Global Change Biology 10: 393 407.
Petersen J, Kemp WM, Bartleson R, et al. (2003) Multi scale
experiments in coastal ecology: Improving realism and advancing
theory. Bioscience 53: 1181 1197.
Small A and Adey W (2001) Reef corals, zooxanthellae and free living
algae: A microcosm study that demonstrates synergy between
calcification and primary production. Ecological Engineering
16: 443 457.
Walter A and Carmen Lambrecht S (2004) Biosphere 2, center as a
unique tool for environmental studies. Journal of Environmental
Monitoring 6: 267 277.
296
Lagoons
Lagoons
G Harris, University of Tasmania, Hobart, TAS, Australia
ª 2008 Elsevier B.V. All rights reserved.
Background
Inputs – Catchment Loads
Fates and Effects – Physics and Mixing
Fates and Effects – Ecological Impacts and Prediction
Nonequilibrium Dynamics
Emerging Concepts – Multifractal Distributions of
Species and Biomass
Further Reading
Background
Nevertheless, two points are worthy of note. First,
there is great functional similarity between systems
despite differing in the actual species involved. Second,
human activity is quickly moving species around the
world so that there are large numbers of what might
be called ‘feral’ introduced species in coastal waters
close to ports and large cities.
Coastal lagoons are ecologically diverse and provide
habitats for many birds, fish, and plants. The interactions
between the species in estuaries and coastal lagoons pro
duce valuable ecosystem services. Indeed, the value of
ecosystem services calculated for such systems by
Costanza et al. was the highest of any ecosystem studied.
Lagoons are also esthetically pleasing and desirable places
to live, providing harbors, fertile catchments, and ocean
access for cities and towns; thus, they have long been the
sites of rapid urban and industrial development. Habitat
change and other threats to lagoons now compromise
these valuable services. All around the world they are
threatened by land use change in their catchments, urba
nization, agriculture, fisheries, transport, tourism, climate
change, and sea level rise. Coastal waters and lagoons are
therefore definitive examples of the problems of multiple
use management. Rapid population growth in coastal
areas is common in many western countries (particularly
the common ‘sea change’ phenomenon, in which there is
a trend toward rapid population growth along coasts), so
the threats and challenges are increasing rapidly. Climate
change and sea level rise are also becoming issues to be
dealt with. In tropical and subtropical regions there is
both evidence of rapid coastal habitat loss and population
growth as well as an increased frequency of severe hurri
canes. Modified systems impacted by severe hurricanes
and tsunamis appear to be more fragile in the face of
extreme events and certainly do not degrade gracefully.
Research and the management of coastal systems
require a synthesis of social, economic, and ecological
disciplines. Around the world there are a number of
major research and management programs which aim to
apply ecosystem knowledge to the effective management
of coastal resources. Current examples include work in
Chesapeake Bay and the Comprehensive Everglades
Coastal lagoons are estuarine basins where freshwater
inflows are trapped behind coastal dune systems, sand
spits, or barrier islands which impede exchange with the
ocean. They are most frequent in regions where fresh
water inflows to the coast are small or seasonal, so that
exchange with the ocean may not occur for months or
years at a time. Many occupy shallow drowned valleys
formed when the sea level was lower during the last ice
age and subsequently flooded by postglacial sea level rise.
The tidal range is usually small. Accordingly, coastal
lagoons are frequently found in warm temperate, dry
subtropical, or Mediterranean regions along moderately
sheltered coasts. Lagoons are infrequent in wetter tempe
rate and tropical regions where freshwater inflows are
sufficient to scour out river mouths and keep them
open. Here estuaries are dominated by salt marshes in
temperate and mangroves in tropical climes. A particu
larly good example is the series of coastal habitats on the
southern and eastern coastline of Australia which change
from open temperate estuaries and salt marshes in the
wetter southern regions of Tasmania, through a series of
coastal lagoons of varying sizes and ecologies along the
south and east coasts, to open subtropical and tropical
estuaries, reefs, and mangroves in the warmer and wetter
north. A similar, although inverted, sequence can be seen
running south along the east coasts of Canada, and the
northeastern, central, and southeastern coasts of the USA.
The resulting lagoons have varying water residence
times, depending on volume, climate, freshwater inflow
volumes, and the tidal prism.
Some lagoons are predominantly freshwater or brack
ish, while others are predominantly marine; so the
dominant organisms in coastal lagoons reflect the bal
ance of freshwater and marine influences. All are
influenced by the local biogeography. Thus, the domi
nant species in Northern Hemisphere lagoons are quite
different from those in their Southern Hemisphere
equivalents. Different coastal regions of the globe differ
in their biodiversity; for example, the endemic biodiver
sity of seagrasses is very high in Australian waters.
Lagoons
Inputs – Catchment Loads
Land use change in catchments changes the hydrology of
rivers and streams and increases nutrient loads to lagoons.
Rivers draining clear catchments, or those with extensive
urbanization, show ‘flashier’ flow patterns with water
levels rising and falling quickly after rainfall. The hydro
logical balance and water residence times of the lagoons
are altered as a result. While nutrient loads are generally
proportional to catchment area (Figure 1), loads from
cleared agricultural or urban catchments are higher than
those from forested catchments, the nutrient loads being
proportional to the amount of cleared land or the human
population in the catchment. Carbon, nitrogen, and phos
phorus loads all increase; C loads from wastewaters may
lead to biochemical oxygen demands (BODs) and anoxia,
while increased N and P loads stimulate algal blooms and
the growth of epiphytes in seagrasses. A further problem
is the fact that forested catchments tend to export organic
forms of N and P (which are less biologically active in
receiving waters), whereas cleared and developed catch
ments tend to export biologically available inorganic
forms of N and P. Thus, both nutrient loads and the
availability of those loads increase when catchments are
cleared and developed.
N is in many cases (particularly in warmer coastal
waters) the key limiting element in lagoons because of
high denitrification efficiencies in sediments and long
water residence times in summer. In temperate waters
N and P may be co limiting or the limitation may vary
seasonally and on an event basis. Overall the climate
regime, geomorphology, and biogeochemistry of coastal
lagoons seem to lead to extensive N limitation and
10 000
R 2 = 0.884 6
1000
TN load (t y–1)
Restoration Plan in the USA. In Italy the lagoon of Venice
is a classic example. In Australia major programs have been
undertaken in coastal embayments and lagoons in Adelaide
(Gulf of St. Vincent), Brisbane (Moreton Bay), and
Melbourne (Port Phillip Bay). (For details on these pro
grams and useful links, see www.chesapeakebay.net,
www.evergladesplan.org, and www.healthywaterways.org.)
Land use change (both urbanization and agriculture) in
catchments, together with the use of coastal lagoons for
transport and tourism, has led to a combination of changes
in physical structures (both dredging and construction of
seawalls and other barriers), altered hydrology and tidal
exchanges, increased nutrient loads, and inputs of toxicants.
The resulting symptoms of environmental degradation
include algal blooms (which may be toxic), loss of biodiver
sity, and ecological integrity (including the loss of seagrasses
and other important functional groups), anoxia in bottom
waters, loss of important biogeochemical functions (deni
trification efficiency), and the disturbances caused by
introduced, ‘feral’ species from ships and ballast water.
297
100
10
1
1
10
100
1000
10 000 100 000
Catchment area (km2)
Figure 1 The empirical relationship between catchment area
(km2) and the total nitrogen load (tonnes per year) to their
associated coastal lagoons. Data from the catchments of 19
coastal lagoons on the east coast of Australia. Details of data
sources are given in Harris GP (1999) Comparison of the
biogeochemistry of lakes and estuaries: Ecosystem processes,
functional groups, hysteresis effects and interactions between
macro- and microbiology. Marine and Freshwater Research 50:
791–811.
denitrification is an important process which determines
many ecological outcomes. The effect of land use change
on N loads is therefore a key area of concern. A consider
able amount of work has been done on the export of N
from catchments around the world. Catchments tend to
retain on average about 25% of the N applied to them and
export about 75%. There are both latitudinal and seaso
nal factors which affect this figure. Catchment exports on
the eastern coast of North America show an effect of
latitude, with warmer, southern catchments with peren
nial vegetation exporting about 10% of applied N and
more northerly catchments with seasonal vegetation
growth exporting as much as 40% of applied N, particu
larly in winter. P exports tend to come primarily from
sewage and other wastewater discharges, and also from
erosion and agricultural runoff. Catchment loads show
evidence of self organized pattern and process in catch
ments – nutrient loads and stoichiometries change over
time at all scales and the distribution of inflowing nutri
ents may be fractal.
Fates and Effects – Physics and Mixing
Water movement and mixing are driven by the effects of
wind and tide on coastal lagoons. The basic hydro
dynamics of coastal systems are well represented by
physics based simulation models of various kinds. A num
ber of two and three dimensional (2D and 3D) models
298
Lagoons
now exist (both research tools and commercially available
products) which can adequately represent wind induced
wave patterns and currents, tidal exchanges and circula
tion, and changes in surface elevations due to tides and
winds. (For an introduction to a variety of models, see
www.estuary guide.net/toolbox or www.smig.usgs.gov;
models by Delft Hydraulics at www.wldelft.nl and DHI
www.dhigroup.com/) Input data required are basic
meteorological data: wind speed and direction, plus solar
insolation, and a detailed knowledge of the morphometry
and bathymetry of the lagoon in question. Based on the
conservation of mass and momentum and various turbu
lence closure schemes, it is possible to adequately model
and predict both velocity fields and turbulent diffusion in
the water column. Calibration and validation data are
obtained from in situ current meters and pressure sensors.
Bottom stress, sediment resuspension, and wave induced
erosion can also be represented. It is thus possible to
model the effects of various climate and engineering
scenarios, everything from sea level rise to construction
projects of various kinds. These models are widely used to
develop environmental impact statements (EIS) for major
projects and to manage major dredging projects around
the world. Only some of these models are capable of
long term predictions of water balance and of water resi
dence times. Such predictions require careful analysis of
long term meteorological records and good predictive
models of inflows and evaporation. Nevertheless such
models also exist.
Fates and Effects – Ecological Impacts
and Prediction
Given the nature of the threats, the value of ecosystem
services delivered, and the importance of ecosystem
management, there have been many studies of ecosys
tems in coastal lagoons. As noted above some of the
ecosystem studies have been in the form of major
multidisciplinary programs. The knowledge obtained
has then frequently been encapsulated in various
kinds of predictive ecological models which attempt
to provide answers to ‘what if ’ questions from environ
mental managers and engineers seeking to implement
catchment works or reductions in wastewater dis
charges. The ecological models are driven by the
hydrodynamic models described above – the physical
setting provides the basic context for the ecological
response. In many cases the knowledge has also been
built into a variety of EIS and risk assessments which
attempt to judge the possible detrimental effects of
land use change, port construction, harbor dredging,
and other engineering developments in urban and
industrial areas.
Empirical Knowledge and Models
Despite the pandemonium of interactions between spe
cies in coastal marine systems (or perhaps because of it),
there are some high level empirical relationships which
can be used for diagnosis and management. Much as
Vollenwieder discovered in lakes there are some predict
able high level properties of coastal marine systems. For
example, the total algal biomass (as chlorophyll a)
responds to N loads just as lakes respond to P loads.
This is further evidence of the importance of N as a
limiting element in marine systems, and for the key of
P as the limiting element in freshwater systems. The differ
ing biogeochemistry of marine and freshwater ecosystems
is explicable on the basis of the evolutionary history and
geochemistry of the two systems. The existence of a rela
tionship between N and algal biomass is evidence for a
kind of ‘envelope dynamics’ of these diverse systems.
N does not limit growth rates of the plankton so much as
the overall biomass. As a result of high growth rates, graz
ing, and rapid nutrient regeneration in surface waters, the
total community biomass reaches an upper limit set by the
overall rate of supply of N. This is a form of ‘extremal
principle’ of these pelagic ecosystems which indicates that
with sufficient biodiversity then an upper limit to max
imum nutrient use efficiencies can be reached. A similar
model of high level ecosystem properties has been devel
oped in which some fundamental physiological properties
of phytoplankton (the slope of the P vs. I curve at low light
and the maximum photosynthetic rate) are used to develop
a production model based on biomass, photosynthetic
properties, and incident light. This amounts to saying
that even in shallow coastal systems it is possible to get
some reasonable empirical predictions of the physiological
(photosynthetic parameters and nutrient uptake efficien
cies) and ecosystem responses to some driving forces
(nutrient loads and incident light).
A second form of empirical determinant of system
function is set by the stoichiometry and biogeochemistry
of these systems. The characteristic elemental ratios in
the key organisms (algae, grazers, bacteria, macrophytes)
and the ratios of elemental turnover set limits on the
overall system performance. The predominant element
ration in pelagic marine organisms is the Redfield ratio
(106C:15N:1P). This aspect of the biogeochemistry of
coastal lagoons has been used in a global comparison of
the biogeochemistry of these systems by the IGBP
LOICZ program. Knowing the loading rates of major
nutrients, the concentrations of nutrients in the water
column, and the rates of tidal exchange allows simple
mass balance models of C, N, and P to be constructed.
The salt and water budget of these systems can be used to
obtain bulk hydrological fluxes. Making stoichiometric
assumptions via the Redfield ratio about fluxes of C, N,
and P (as well as oxygen) in the plankton and across the
Lagoons
sediment interface allows estimates to be made of the
overall autotrophic–heterotrophic balance of the system
as well as nitrogen fixation and denitrification rates
(essentially by estimating the ‘missing N’ based on the
C, N, and P stoichiometry). These techniques have made
it possible to do global comparisons of the biogeochem
istry of lagoons around the world and to examine the
effects of inflows, tidal exchanges, and latitude or climate.
This has been a major contribution to the knowledge of
the ways in which major elements are processed and
transported from the land to the ocean through the coastal
zone.
The overall impression is that pristine lagoons (loaded
by largely organic forms of C, N, and P) are mostly net
heterotrophic and strong sinks for N through denitrifica
tion. More eutrophic systems with higher N and P loads
(and more of those in inorganic forms) tend to be net
autotrophic and, if dominated by cyanobacterial blooms,
net N fixing systems. Decomposition of these blooms may
be sufficiently rapid to cause anoxia in bottom waters and
lead to the cessation of denitrification and the export of
N (as ammonia) on the falling tide. Warm temperate and
subtropical lagoons – with low hydrological and nutrient
loads – seem to have higher denitrification efficiencies
than temperate systems. They are often heterotrophic and
strongly N limited systems. An extreme is Port Phillip
Bay in Melbourne which has low freshwater inflows, high
evaporation, a long water residence time (c. 1 year), high
denitrification efficiency (60–80%) and is so N limited
that it imports N from the coastal ocean on the rising tide.
Temperate lagoons and estuaries have higher freshwater
and nutrient inflows, are more eutrophic (autotrophic),
and are exporters of N. Temperate systems are therefore
more likely to show occasional P limitation. Overall, the
cycling of the major elements is driven by the stoichio
metry of the major functional groups of organisms.
Thus in biodiverse ecosystems it is possible to obtain
some high level state predictors from a knowledge of key
drivers and the basic physiology and stoichiometry of the
dominant organisms. The predictions so produced are not
perfect but they do capture a large fraction of the beha
vior of these systems. At this level these models can be
used for the management of nutrient loads to coastal
lagoons.
Detailed Simulation Models of Ecosystems,
Functional Groups, and Major Species
Many of the questions that are asked of ecologists study
ing coastal systems are of a more detailed nature and
relate to loss or recovery of major species, functions, or
functional groups – ecosystem services and assets if you
like. Examples would be dominant algal groups, sea
grasses, macroalgae, denitrification rates, benthic
biodiversity, fish recruitment, etc. At this level a large
299
number of dynamical ecological simulation models of
shallow marine systems have been constructed. There is
much more uncertainty in the ecological models than
there is in the physical models. Much of the required
ecological detail is unknown, key parameters can be
ill defined, the data are usually sparse in space and time,
and the computational resources are not adequate to the
task of a complete simulation of the entire system.
Ecological models are therefore abstractions which
attempt to represent the major ecological features and
functions of the greatest relevance to the task at hand.
Nevertheless, 30 years of research in lagoons and coastal
systems around the world have uncovered a number of
major functional groups and ecosystem services which,
when coupled together in models, give some guide as to
the overall ecological responses.
The generic models of coastal systems use two basic
functional components. A nutrient, phytoplankton, zoo
plankton (NPZ) model for the water column, and a
benthic model incorporating the necessary functional
groups – macroalgae, zoo and phytobenthos, seagrasses –
with the groups chosen to represent the particular system
of interest. All functional groups are represented by their
basic physiologies and stoichiometries and the inter
connections (grazing, trophic closure, decomposition,
and denitrification rates) are represented by established
relationships. The NPZ models adequately predict the
average chlorophyll of lagoons and, when coupled with
3D physical models, can give predictions of the spatial
distribution of algal biomass in response to climate and
catchment drivers. For reasons which will become clear
below, these models only predict average biomass levels
and cannot predict all the dynamics of the various trophic
levels. The coupling between the plankton and the
benthos in lagoons is nonlinear and results in some
strongly nonlinear responses of the overall system to
changes in nutrient loads. Basically, there is competition
between the plankton and the benthos for light and nutri
ents which can drive switches in system state. Thus,
lagoons, much like shallow lakes, may show state switches
between clear, seagrass dominated states and turbid,
plankton dominated states.
The major driver of the state switches is the high
denitrification efficiencies exhibited by the diverse
phyto and zoobenthos in lagoons with strong marine
influences. As long as there is sufficient oxygen in bottom
waters, diverse zoobenthos burrow and churn over the
sediments causing extensive bioturbation and 3D struc
ture in the sediments. Clams, prawns, polychaete worms,
crabs, and other invertebrates set up a complex system of
burrows and ventilate the sediments through feeding cur
rents and respiratory activity. Given sufficient light at the
sediment surface the phytobenthos (particularly diatoms,
the microphytobenthos, MPB) photosynthesize rapidly
and set up strong gradient of oxygen in the top few
Lagoons
millimeters of the sediment. These gradients, together
with the strong 3D microstructure of the sediments set
up by the zoobenthos, favor the co occurrence of adjacent
oxic and anoxic microzones which are required for effi
cient denitrification. N taken up by the plankton sinks is
actively denitrified by the sediment system. In marine
systems the abundance of sulfate in seawater ensures
that P is not strongly sequestered by the sediments.
Thus, the basis of the LOICZ models lies in the efficiency
of denitrification of N in sediments and the more or less
conservative behavior of P in these systems. These eco
system services are supported by the high biodiversity of
the coastal marine benthos.
In lagoons with higher nutrient loads, the entire eco
system may switch to an alternative state. Increased N
loads stimulate the growth of plankton in the water col
umn and shade off the MPB. The increased planktonic
production sinks to the bottom depleting oxygen and
reducing the diversity of zoobenthos, restricting the com
munity structure to those species resistant to low oxygen
concentrations. Active decomposition in anaerobic sedi
ments together with reduced bioturbation leads to the
cessation of denitrification and the release of ammonia
from the sediments. So instead of actively denitrifying and
eliminating the N load, the system becomes internally
fertilized and algal production rises further. This is ana
logous to the internal fertilization of eutrophic lakes
through the release of P from anoxic sediments. In both
cases the switch is caused by a change in redox conditions
and the change in performance of suites of microbial
populations. Once switched to a more eutrophic state
(algal bloom dominated), these lagoons do not easily
revert to their clear and macrophyte dominated state.
Loads must be strongly reduced to get them to switch
back – something which may not be possible if the catch
ment has been modified by urban or agricultural
development. There is thus evidence for strong hysteresis
in the response of these ecosystems to various impacts.
The overall biodiversity and nutrient cycling perfor
mance of coastal lagoons therefore depends on the
relative influences of marine and freshwaters, the differ
ing biodiversity of marine and freshwater ecosystems, the
relative C, N, and P loads to the plankton and the
benthos, and on seasonality, latitude, and climate drivers.
Nevertheless, at least the broad features of their behavior
can be explained and predicted on the basis of sediment
geochemistry, and the stoichiometry and physiology of
the major functional groups in these ecosystems.
Empirical work on a number of lagoons up the east
coast of Australia allowed Scanes et al. to effectively
determine the response of ‘titrating’ these systems with
nutrients. As the N load to the lagoons was increased,
seagrasses were lost and algal blooms were stimulated.
Even at a crude level of visual assessments it was possible
to rank these systems in order of loading and to show that
6
5
Ecosystem state
300
4
3
2
1
0
1
10
100
1000
N exports (kg ha–1 y–1)
10 000
Figure 2 The empirical ‘ecosystem titration’ relationship
between catchment N exports and the resulting ecosystem state
in 17 coastal lagoons on the east coast of Australia. Ecosystem
state is defined as 1, pristine; 3, showing marked seagrass loss
and the growth of macrophytic algae; 5–6, dominated by
nuisance algal blooms (some of which may be toxic). Data from
personal observations and reworked from Scanes P, Coade G,
Large D, and Roach T (1998) Developing criteria for acceptable
loads of nutrients from catchments. In: Proceedings of the
Coastal Nutrients Workshop, Sydney (October 1997), pp. 89–99.
Artarmon, Sydney: Australian Water and Wastewater
Association.
the pattern of response was entirely similar to that pre
dicted by the models (Figure 2). Thus, despite difference
in biogeochemistry and biodiversity, shallow lakes and
coastal lagoons have broadly similar response to increased
nutrient loads and other forms of human impact. Even
broad indicators of system state reveal consistent patterns
of change.
So oligotrophic lagoons with a Mediterranean climate
(warm temperatures in summer and long water residence
times) and strong marine influences can be strong sinks
for N, whereas cooler, temperate lagoons and estuaries
with larger freshwater inflows and higher productivity
may export N and be frequently P limited. As the
LOICZ program intended, we have managed a broad
understanding of the ways in which the coastal zone
influences the transport of major elements from land to
ocean.
Nonequilibrium Dynamics
If more detailed descriptions and predictions are required
(e.g., the diversity and abundance of individual species
and other specific ecosystem services and assets), then the
predictive ability is less. One of the reasons for this is the
fact, alluded to above, that these are nonequilibrium
Lagoons
systems which respond to individual events (storms and
engineering works) over long time periods. The elimina
tion and invasion of species may take decades and the
responses of freshwater lagoons, for example, to salt
incursions may also take decades. A particularly good
example is Lake Wellington in the Gippsland Lakes sys
tem in Victoria, Australia. The entire system is slowly
responding to the ingress of salt made possible by the
opening of the lagoon system mouth (Lakes Entrance) in
1883. Lake Wellington, the lake farthest inland, remained
fresh until after the 1967 drought when a combination of
high N and P loads from agriculture, the extraction of
water from the inflowing La Trobe River for power sta
tion cooling and irrigation, and the incursion of salt killed
all the freshwater macrophytes in the Lake. In a few years
the lake switched from its previous clear and macrophyte
dominated state to being turbid and dominated by toxic
algal blooms. It does not appear to be possible to switch it
back.
The response of these lagoon systems to climate and
other perturbations is nonlinear and complex because of
the interactions between the major functional groups and
because the timescales of response of the major groups
differ strongly. Phytoplankton may respond to changes in
loads and water residence times in a matter of days,
whereas seagrasses take decades or longer to recover. By
perturbing a simple coupled plankton benthos model
with storm events and ‘spiked’ N loads, Webster and
Harris showed that the threshold load for the elimination
of seagrasses could be altered considerably depending on
the characteristics of the input loads. So the response of
the system was a function of the overall load and the
frequency and magnitude of events. Climate change and
catchment development both alter the overall C, N, and P
load to lagoons as well as the characteristics of that load,
so that ecological responses by lagoons are highly com
plex and change over time depending on a variety of
modifications and management actions. Consequently,
lagoons are always responding to the last storm or inter
vention and the abundance of key species drifts to and
fro over time as the entire plankton–sediment system
responds.
The picture is made more complex by the evidence for
strong trophic cascades in marine as well as freshwater
systems. Coastal ecosystems are frequently over fished;
larger predators and grazers are removed by human hand.
Removal of the ‘charismatic megafauna’ of coastal sys
tems, together with beds of shellfish and other edible
species, has changed the ecology of many lagoons and
estuaries. Coastal ecosystems around the world have also
been strongly modified by the removal of natural physical
structures (mangroves and reefs) which confer resilience
in the face of extreme events. We have removed both
larger fish and benthic filter feeders from many systems
compromising function and the ability to respond to
301
changes in catchment loads. Overall there has been a
consistent simplification of both physical and ecosystem
structures (removal of reefs and macrobiota, simplifica
tion of food chains, etc.) and a trend toward more
eutrophic (nutrient rich) and simplified systems domi
nated by microbiota, especially algae and bacteria. We
know less about the response of ecosystems to changes in
the ‘top down’ trophic structure than we do about the
responses to ‘bottom up’ catchment drivers; nevertheless,
there is good evidence for similar nonlinearities and state
switches in response. A nonequilibrium view of coastal
lagoons changes the way we look at them. Overall there is
a need to pay attention to the ‘precariousness’ of these
systems and manage them adaptively for resilience and
response to natural and anthropogenic impacts. Despite
being over fished and highly modified, there is still a need
for the ecosystem services they produce.
Emerging Concepts – Multifractal
Distributions of Species and Biomass
The underlying complexity of interactions and species
distributions is displayed when detailed (high frequency)
observations are made of the spatial and temporal distri
butions of biomass and species. There is now much
evidence to show that the underlying distribution of the
plankton and the MPB are fractal or multifractal.
Similarly, high frequency observations in catchments
show similar multifractal and even paradoxical properties
of hydrological and nutrient loads. So underlying all the
generalizations discussed above lies a pattern of behavior
which gives strong evidence of self generated complexity
which arises from the pandemonium of interactions
between species and functional groups. Indeed, we can
probably argue that the kinds of general, system level,
responses described above would not occur if it were not
for the underlying complexity. While making high level
statements about ecosystem behavior possible, these
small scale, multifractal properties (and the possibilities
created by emergence) cause problems when we wish to
make predictions at the meso scale level of dominant
species and functional groups. Because of the work that
has been done across the levels of organization, coastal
lagoons are very good examples of a new kind of ecology –
an ecology of resilience and change, rather than an ecology
and equilibrium and stasis.
One fundamental problem that these new insights
reveal is that most of the data we presently use for the
analysis of coastal lagoons are collected too infrequently
to be useful for anything other than the analysis of broad
trends. Data collected weekly or less frequently are
strongly aliased and cannot reveal the true scales of
pattern and process. It is just possible to analyze daily
data for new insights and processes but high frequency
302
Lagoons
data – collected at scales of hours and minutes – reveal a
wealth of new information. Aliased data combined with
frequentist statistical techniques that ‘control error’ actu
ally remove information from multifractally distributed
data and raise the possibility of serious type I and II errors
in ecological interpretations. Most importantly, there is
information contained in the time series of multivariate
data that can be collected from coastal systems. Most
analyses of ecological data from ecological systems use
univariate data and because of the infrequent data collec
tion schedules – including gaps and irregular time
intervals – time series analyses are not possible.
We are just beginning to find new technologies and
techniques to study the high frequency multivariate beha
vior of these systems using moorings and other in situ
instruments. New electrode technologies make on line
access to data possible and throw up new possibilities for
new kinds of observations of system state. We are beginning
to realize that in addition to the ‘top down’ causation of
climate and trophic interactions, there is also a ‘bottom up’
driver of complexity and the strong possibility of the emer
gence of high level properties from the interactions between
individuals. New forms of statistical analyses display infor
mation in time series of complex and emergent systems.
This emerging understanding of complexity and emergent
properties changes the ways in which we should approach
EIS and risk assessments. We now know that interactions
and self generated complexity, together with hysteresis
effects at the system level, can cause surprising things to
happen as a result of anthropogenic change. Coastal lagoons
are now classic examples of this. That means that risk
assessments and EIS cannot look at impacts and changes in
isolation; somehow we must develop integrated risk assess
ment tools that examine the interactive and synergistic
effects of human impacts on coastal ecosystems. A further
level of complexity is contained in the similar complex and
emergent properties of the interactions between agents in
the coupled environmental and socioeconomic (ESE) sys
tem in which all coastal lagoons are set. Multiple use
management decisions are set in a complex web of ESE
interactions across scales. Decisions made about industrial
and engineering developments for financial capital reasons
influence both social capital and ecological (natural capital)
outcomes. Feedbacks ensure that this is also a highly non
linear set of interactions. What we do know is that the
prevalent practices of coastal management and exploitation
are not resilient in the face of extreme events and that they
do not degrade ‘gracefully’ when impacted by hurricanes
and tsunamis. New management practices will be required.
See also: Mangrove Wetlands.
Further Reading
Adger WN, Hughes TP, Folke C, Carpenter SR, and Rockstrom J (2005)
Socio ecological resilience to coastal disasters. Science
309: 1036 1039.
Aksnes DL (1995) Ecological modelling in coastal waters: Towards
predictive physical chemical biological simulation models. Ophelia
41: 5 35.
Berelson WM, Townsend T, Heggie D, et al. (1999) Modelling bio
irrigation rates in the sediments of Port Phillip Bay. Marine and
Freshwater Research 50: 573 579.
Brawley JW, Brush MJ, Kremer JN, and Nixon SW (2003)
Potential applications of an empirical phytoplankton production
model to shallow water ecosystems. Ecological Modelling
160: 55 61.
Costanza R, d’Arge R, de Groot R, et al. (1998) The value of ecosystem
services: Putting the issues in perspective. Ecological Economics
25: 67 72.
Fasham MJR, Ducklow HW, and Mckelvie SM (1990) A nitrogen based
model of plankton dynamics in the oceanic mixed layer. Journal of
Marine Research 48: 591 639.
Flynn KJ (2001) A mechanistic model for describing dynamic multi
nutrient, light, temperature interactions in phytoplankton. Journal of
Plankton Research 23: 977 997.
Gordon DC, Boudreau PR, Mann KH, et al. (1996) LOICZ
biogeochemical modelling guidelines. LOICZ Reports and Studies,
No. 5. Texel: LOICZ.
Griffiths SP (2001) Factors influencing fish composition in an Australian
intermittently open estuary. Is stability salinity dependent? Estuarine,
Coastal and Shelf Science 52: 739 751.
Harris GP (1999) Comparison of the biogeochemistry of lakes and
estuaries: Ecosystem processes, functional groups, hysteresis
effects and interactions between macro and microbiology. Marine
and Freshwater Research 50: 791 811.
Harris GP (2001) The biogeochemistry of nitrogen and phosphorus in
Australian catchments, rivers and estuaries: Effects of land use and
flow regulation and comparisons with global patterns. Marine and
Freshwater Research 52: 139 149.
Harris GP (2006) Seeking Sustainability in a World of Complexity.
Cambridge: Cambridge University Press.
Harris GP and Heathwaite AL (2005) Inadmissible evidence: Knowledge
and prediction in land and waterscapes. Journal of Hydrology
304: 3 19.
Hinga KR, Jeon H, and Lewis NF (1995) Marine eutrophication review.
Part 1: Quantifying the effects of nitrogen enrichment on
phytoplankton in coastal ecosystems. Part 2: Bibliography with
abstracts. NOAA Coastal Ocean program, Decision Analysis Series,
No 4. Silver Spring, MD: US Dept of Commerce, NOAA Coastal
Ocean Office.
Howarth RW (1998) An assessment of human influences on fluxes of
nitrogen from the terrestrial landscape to the estuaries and
continental shelves of the North Atlantic Ocean. Nutrient Cycling in
Agroecosystems 52: 213 223.
Howarth RW, Billen G, Swaney D, et al. (1996) Regional nitrogen
budgets and the riverine N and P fluxes for the drainages to the North
Atlantic Ocean Natural and human influences. Biogeochemistry
35: 75 139.
Lotze HK, Lenihan HS, Bourque BJ, et al. (2006) Depletion, degradation
and recovery potential of estuaries and coastal seas. Science
312: 1806 1809.
McComb AJ (1995) Eutrophic Shallow Estuaries and Lagoons. Boca
Raton: CRC Press.
Mitra A (2006) A multi nutrient model for the description of
stoichiometric modulation of predation in micro and
mesozooplankton. Journal of Plankton Research 28: 597 611.
Moll A and Radach G (2003) Review of three dimensional ecological
modelling related to the North Sea shelf system. Part 1: Models and
their results. Progress in Oceanography 57: 175 217.
Murray AG and Parslow JS (1999) Modelling of nutrient impacts in Port
Phillip Bay A semi enclosed marine Australian ecosystem. Marine
and Freshwater Research 50: 597 611.
Landfills
Nicholson GJ and Longmore AR (1999) Causes of observed
temporal variability of nutrient fluxes from a southern Australian
marine embayment. Marine and Freshwater Research 50: 581 588.
Occhipinti Ambrogi A and Savini D (2003) Biological invasions as a
component of global change in stressed marine ecosystems. Marine
Pollution Bulletin 46: 542 551.
Pollard DA (1994) A comparison of fish assemblages and fisheries in
intermittently open and permanently open coastal lagoons on the
south coast of New South Wales, south eastern Australia. Estuaries
17: 631 646.
Roy PS, Williams RJ, Jones AR, et al. (2001) Structure and function of
south east Australian estuaries. Estuarine, Coastal and Shelf
Science 53: 351 384.
Scanes P, Coade G, Large D, and Roach T (1998) Developing criteria
for acceptable loads of nutrients from catchments. In: Proceedings
of the Coastal Nutrients Workshop, Sydney (October 1997),
pp. 89 99. Artarmon, Sydney: Australian Water and Wastewater
Association.
Scheffer M (1998) Shallow Lakes. London: Chapman and Hall.
Scheffer M, Carpenter S, and de Young B (2005) Cascading effects of
overfishing marine systems. Trends in Ecology and Evolution
20: 579 581.
Seitzinger SP (1987) Nitrogen biogeochemistry in an unpolluted estuary:
The importance of benthic denitrification. Marine Ecology Progress
Series 41: 177 186.
Seitzinger SP (1988) Denitrification in freshwater and coastal marine
systems: Ecological and geochemical significance. Limnology and
Oceanography 33: 702 724.
Seuront L, Gentilhomme V, and Lagadeuc Y (2002) Small scale nutrient
patches in tidally mixed coastal waters. Marine Ecology Progress
Series 232: 29 44.
Seuront L and Spilmont N (2002) Self organized criticality in intertidal
microphytobenthos patterns. Physica A 313: 513 539.
303
Smith SV and Crossland CJ (1999) Australasian estuarine systems:
Carbon, nitrogen and phosphorus fluxes. LOICZ Reports and
Studies, No. 12. Texel: LOICZ.
Sterner RW and Elser JJ (2002) Ecological Stoichiometry: The Biology of
Elements from Molecules to the Biosphere. Princeton, NJ: Princeton
University Press.
Vollenweider RA (1968) Scientific fundamentals of the eutrophication of
lakes and flowing waters, with particular reference to nitrogen and
phosphorus as factors in eutrophication. Technical Report DAS/SCI/
68.27, 182pp. Paris: OECD.
Walker DI and Prince RIT (1987) Distribution and biogeography of
seagrass species on the northwest coast of Australia. Aquatic
Botany 29: 19 32.
Walker SJ (1999) Coupled hydrodynamic and transport models of Port
Phillip Bay, a semi enclosed bay in south eastern Australia. Marine
and Freshwater Research 50: 469 481.
Webster I and Harris GP (2004) Anthropogenic impacts on the
ecosystems of coastal lagoons: Modelling fundamental
biogeochemical processes and management implications. Marine
and Freshwater Research 55: 67 78.
Relevant Websites
http://www.chesapeakebay.net Chesapeake Bay Programme.
http://www.dhigroup.com DHI.
http://www.evergladesplan.org Everglades.
http://www.healthywaterways.org Healthy Waterways.
http://www.estuary guide.net Toolbox, The Estuary Guide.
http://www.wldelft.nl wl delft hydraulics.
Landfills
L M Chu, The Chinese University of Hong Kong, Hong Kong SAR, People’s Republic of China
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Postclosure End Uses
Soil Cover
Vegetation
Fauna
Ecological Approach
Further Reading
Introduction
most common method of municipal solid waste manage
ment worldwide. Landfill leachate is formed when
rainwater infiltrates and percolates through the degrading
waste, while landfill gas is a microbial degradation bypro
duct under anaerobic conditions. Modern landfills are
designed and engineered to restrict the formation and
movement of landfill leachate and gas, and to minimize
environmental nuisance caused by wind blown litter,
pests, and odor during operation. These landfills, either
the containment or entombment type, have buried waste
that is isolated from the environment. Older landfills are
Landfills are seminatural terrestrial ecosystems recon
structed on lands degraded by waste disposal. They are
unique in terms of site formation, nature of stratum, and
biological activities, but vary according to their age, waste
composition, engineering design, and ecological practice.
From an environmental perspective, landfills are deposi
tories for municipal solid wastes (sanitary landfills) and
less frequently hazardous wastes (secure landfills).
Landfills are ubiquitous, as sanitary landfilling is the
304
Landfills
of the dilution and attenuation type that makes use of the
substratum for pollution mitigation; they are unconfined
with no facilities for leachate treatment and gas extrac
tion. With dilution and attenuation landfills, problems
associated with leachate and gas are common. In terms
of environmental biotechnology, landfills can be regarded
as large scale bioreactors in which the organic matter in
the buried waste is anaerobically degraded to produce
landfill gas which is methane rich and can be used for
electricity generation.
Postclosure End Uses
Once fill capacity is reached, landfills are closed for
rehabilitation. With the exception of older landfills that
are left abandoned with minimal human interference,
most postclosure landfills are rehabilitated using engi
neering and ecological approaches. Landfilled wastes are
isolated physically from the biosphere by bottom barrier
layers and surface cap technologies. Barrier systems can
be sophisticated with multiple layers of geotextiles and
impermeable synthetic membranes. The surface is usually
covered with soil of 1–2 m thickness. Subsequent site
development entails the establishment of a vegetation
cover on the landfill soil, with the primary aims of mini
mizing environmental impact and making good value of
its designated afteruse. Technically, closed landfills can
be rehabilitated by either spontaneous ecological develop
ment in the absence of human intervention, manipulated
succession followed by natural development, or habitat
creation which involves intensive and prolonged man
agement. Sole natural development is unreliable and
slow, and lacks control of the ecological outcome. The
aftercare period for a landfill can be as long as 30 years,
but public safety and engineering concerns are usually of
higher priority than the ecological function of the
reclaimed site.
The criteria for selecting afteruses for former landfills
include landuse planning policies, site characteristics,
soil resource availability, social needs, and cost considera
tion. As construction on postclosure landfills is generally
prohibited due to severe subsidence as a result of organic
matter decomposition, and fire hazards associated with
landfill gas, it is a usual practice to reclaim urban sites
for soft end uses in order to provide amenity facilities
such as parks, botanical gardens, golf courses, and playing
fields that are safe for use by the public. Alternatives
end uses for agriculture, nature conservation, and forestry
are also common. Grassland has been one of the most
popular end uses for rural sites, but agricultural conver
sion is not always appropriate because of the lack of
quality topsoil. Nature conservation is sometimes a
more suitable afteruse as it requires less intensive after
care and is more flexible on the postclosure ecological
design, though the transformation to wildlife habitat is
not imperative. End use after closure can be mixed land
scapes as in the Fresh Kills Landfill in New York, USA,
which is converted to an amenity parkland with a range
of landuses which include forests, dry lowlands, tidal wet
lands, freshwater wetlands, waterways, and wildlife
habitats.
Soil Cover
As it is the final soil cover which supports vegetation
establishment and ecosystem development, the quality
of the soil material and the thickness of the soil cover
are of fundamental importance in affecting rehabilitation
success. However, as good soil is usually not available or
expensive, the soil used is usually derived from ex situ
substandard soil or subsoil that is nutrient deficient and
poor structured. There are inevitable problems of soil
compaction and waterlogging for clay soil, drought
when coarse soil is used, as well as infertility, which can
be amended by conventional measures such as plowing,
organic matter amendment, nurse species planting, and
fertilizer application.
Many old landfills have experienced revegetation fail
ure to various extents as a result of leachate seepage,
landfill gas evolution, poor soil management, and minimal
aftercare. Landfill gas is the major cause, among other
constraints such as low fertility, high soil temperature,
drought, and toxicity from leachate contamination.
Unless it is vented to the atmosphere or extracted for
energy production, it will displace oxygen and suffocate
plant roots, which usually results in the death of vegeta
tion, and gas production can last for 75 years after the
deposition of wastes. Even for an engineered site, gas
problems may still exist if an impermeable layer is not
formed for the entire site or the soil cap is cracked by
uneven subsidence of the site. Landfill gas creates a redu
cing soil condition which severely impairs microbial
processes such as decomposition and symbiotic nitrogen
fixation; this together with elevated soil temperature of
over 40 C is detrimental to plants, and plant growth is
impeded under the adverse impact of these landfill
associated factors. Localized pollution hot spots reduce
plant coverage and result in patchy greenness. A thin soil
cover will exacerbate the problem of gas and leachate
contamination.
The revegetation success of closed landfills depends
heavily upon the quality of the soil cover material, adapta
tion of the planted vegetation to the landfill environment,
and aftercare management strategy. In containment and
entombment landfills, contamination by landfill gas and
leachate is usually greatly alleviated, though not necessa
rily eliminated. However, the final soil cover may remain
stressful for plant growth, and there is also concern that the
Landfills
containment design may elevate nutrient and water stresses
on these landfills. Thin soil cover, poor soil quality, and
unfavorable landfill conditions will result in poor vegeta
tion growth, especially in the initial phase of ecosystem
development.
It is important for rehabilitated landfills to develop a
functional soil–plant system, as shortage of nutrients, in
particular nitrogen, is common in most imported soils for
use as the final top layer on completed landfills. This can
be achieved by the addition of chemical fertilizers at the
onset of postclosure revegetation works. However, as
repeated application is costly, revegetated sites are
usually left to nature for the accumulation of nutrients
needed for the establishment of self perpetuating nutrient
cycle. This has to be achieved to allow good vegetation
growth, the establishment of a fully functional soil–plant
system, and ecosystem development. Plant growth during
the early phase of ecosystem rehabilitation is usually
limited by the rate of nutrient turnover, and the use of
poor soil material as the final cover will inevitably result
in rehabilitated sites that are neither productive nor sus
tainable. There is a paucity of information on the nutrient
fluxes and compartmentation in landfill cover soils, and
there is only partial idea of nutrient mobilization and
immobilization as a function of soil status, and stage of
soil development and vegetation succession. Shortage of
mineral nutrients could be either due to a lack of suffi
cient nutrient capital or a failure in mineralization
processes. Therefore, litterfall, litter quality, mineraliza
tion rate, and the level of biological activity are important
determinants of landfill soil quality. Slow decomposition
rate implies that nutrients are trapped in organic matter
and are not available to nutrient transformation.
Nutrients such as nitrogen and phosphorus accumu
late in landfill soil as the ecosystem develops, and their
levels have a positive correlation with vegetation estab
lishment. In abandoned landfills, without much aftercare,
litter from invaded vegetation is the primary source of
organic matter and nutrients in the absence of biological
fixation. However, there is a lack of information regarding
the nitrogen capital of landfill soils. Nitrogen is supplied
from fertilizer application, decomposition, biological fixa
tion, and rainfall. It is susceptible to immobilization on the
youngest sites, and the primary production of newly
established grassy vegetation cannot rely on decomposi
tion, even though the rate is comparatively high for a
sustainable nitrogen turnover. Total amount of nitrogen
mineralized on more mature woodlands is high, but it is
unclear as to how much nitrogen accumulated in the soil
is sufficient to create a self perpetuating ecosystem on
closed landfills.
Within the soil, the microflora, fauna, and the abiotic
components are all important and interrelated compart
ments of the landfill ecosystem. Former landfills support
diverse soil and litter fauna which have an active role in
305
the detritus food web. They comprise of high diversity
and populations of saprohagous arthropods and macroin
vertebrates such as isopods, millipedes, and centipedes
that are tolerant of the landfill environment. Springtails
and mites are abundant in landfills with gas problems.
Earthworms are also adaptive to landfill conditions and
have been inoculated to landfills for soil amelioration, but
natural colonization and soil improvement appear to be
slow, and it takes 3–14 years for earthworm species to
invade landfills. Low accumulation of organic matter and
patchy coverage of vegetation can hinder the recruitment
as well as the mobility of earthworms in landfills.
The best soil cover on landfills should support diverse
communities of soil microflora and invertebrates which
play crucial roles in organic matter decomposition and
nutrient cycling. Active populations of microorganisms
and invertebrates will improve the physicochemical status
of the soil, which in turn encourage the colonization of
plants to support more diverse animal species, thus form
ing a community of a greater structural complexity and
functional stability. This is important not only for the
success of revegetation but also the successional develop
ment afterwards. In the long run, this will facilitate
autogenic change which is the result of the recruitment
of late successional species and the development of eco
system processes on these man made habitats.
Vegetation
Plant cover on landfills contributes to its landscape and
assists in the reduction of leachate discharge through
evapotranspiration. The latter function is particularly
important if the landfill is not capped with an imperme
able layer to control infiltration. Other benefits provided
by the vegetation cover include visual improvement of
the site, creation of wildlife habitat, and the sequestration
of greenhouse gases. The species chosen for revegetation
purpose depends on the afteruse of the site, climatic
conditions, nursery stock availability, and hardiness of
the species.
Despite the tremendous efforts and investment
devoted to site engineering, the inclusion of a soil cover
does not guarantee the successful establishment of vege
tation. The depth and quality of the soil layer affect
revegetation as a thicker soil cover is required for
woody species which have deeper root systems.
Poor vegetation performance is a common feature of
many old landfills. In the US, a nationwide survey con
ducted in the early 1980s showed that the major cause for
plant failure was the high concentration of landfill gas in
the root zone. Negative correlation was found between
landfill gas concentration and plant coverage or tree
growth in municipal landfills because tree growth was
hampered by high landfill gas content, and to a certain
306
Landfills
extent by high soil temperature and drought. In addition,
root development, and hence plant growth at landfills was
also adversely affected by pedoclimatic conditions such as
high underground temperature, drought, soil acidity, and
contamination by leachate. To counteract these problems,
the planting of species tolerant to the above adverse
conditions is recommended. This is why earlier studies
on landfill revegetation focused on the adaptability of
plant species to landfill gas. Leguminous trees are better
than nonlegumes in their tolerance to high landfill gas
that prevailed in old landfills.
Rehabilitation is traditionally initiated by hydroseed
ing with grass species and/or planting of tree species for
erosion control and esthetic improvement. Landscaping
and artificial revegetation are the initial rehabilitation
works, irrespective of the afteruse of the site, as this
accelerates ecological development. The site is revege
tated preferentially using grasses which grow fast and
provide good immediate ground cover to control erosion
and reduce visual impact. Grass swards also survive better
than trees on landfills with gas influence, a feature which
is attributed to their shallow rooting depth. Tree planting
is less popular especially on the top platform of a landfill,
because of the negative effects of tree growth on landfills.
Following initial revegetation, the rehabilitated site is left
for secondary succession to take place.
Grassland can be a versatile habitat option for closed
landfills as it can be established on a wide range of soil
types. While pasture or arable grassland is more demand
ing on soil quality and requires greater fertilizer input,
low maintenance grassland can be established on infertile
soils. A seed mix of more species and the inclusion of
wildflowers can increase the species richness of the vege
tated sites. Open grasslands are good habitats for many
animal species (e.g., butterflies), but others prefer scat
tered scrubs and trees for shelter. Planting trees had not
usually been recommended as it was believed that tree
roots would perforate and crack by drying out the landfill
cap. In addition, tree growth on landfill soil may be
difficult because of poor soil quality. However, as wood
lands have the greatest conservation value, it seems
desirable to plant trees to form woodlands which have
the benefits of increasing forest resources, habitat con
nectivity, wildlife biodiversity, and landscape integration.
Vegetation is an integral part of the landfill ecosystem,
and flora composition of vegetated sites differs with
respect to landfill technology (i.e., gas and leachate con
trol), hydrometeorological conditions, as well as the
quality and depth of the soil cover. Vegetation composi
tion is also directly affected by the species planted,
survival of the planted species, replanting/enrichment
planting, natural invasion of other species, and the seed
bank in the soil cover material. A suitable species will very
much adapt to and survive in the landfill conditions, at
least for a certain period of time, and facilitate the growth
of late successional species. With differential site avail
ability, species availability, and species performance,
rehabilitation can be directed by using different soil and
planting strategies to achieve successional intervention. A
good choice of species for revegetation could enhance the
sustainability of ecosystem development. Nitrogen fixers
and those pioneer species usually outcompete other spe
cies in the first 10–20 years of ecosystem development
after rehabilitation. Nitrogen fixing trees such as the tropi
cal species of Acacia confusa, A. auriculiformis, A. mangium,
Albizia lebbeck, Casuarina equisetifolia, Leucaena leucocephala
and temperate species such as Alnus glutinosa are widely
used for planting on closed landfills. These species assist
in nitrogen accumulation in the landfill soil and are very
important in the successional development of the soil
cover. Therefore, enrichment planting of late succes
sional species is sometimes necessary at a later stage of
development to enhance plant density and maintain
species rich vegetation in secondary forests on closed
landfills.
The establishment of woodland communities is the
result of gradual ecological development, which cannot
be achieved simply by tree planting. The success and
speed of succession rely on the availability of appropriate
seeds with the proper dispersal mechanism and the pre
sence of effective animal dispersers, and species with the
appropriate ecological characteristics. The seed bank, in
the cover soil, supplies the species for the early vegetative
colonization, which resembles the floristic composition of
the areas where the soil is obtained. Soil seed densities
decline with landfill age, a trend similar to the course of
old field succession. Young landfills have more r selected
species, which tend to produce more seeds, whereas older
sites have more K selected species, which produce fewer
seeds but a higher population of perennials. Some woody
plants that are more adaptive can invade gaps and estab
lish slowly. Trees that are either early successional
species or leguminous species should be planted in
greater proportion to accelerate succession in landfills,
preserve the biodiversity of local flora, and provide
more favorable habitats for wildlife conservation.
Planting more native or exotic species has been debated;
natives, though not necessarily fast growing, are adaptive
to local environmental conditions, and provide indigen
ous characters that are not found in artificial revegetation,
but whether natives or exotics are better choices depends
on their adaptation to landfills conditions and the quality
of soil for revegetation.
Postclosure landfills can be a good refuge for rare
species including wild orchids, and are important to the
conservation of native flora. Older sites are better devel
oped in terms of soil quality and vegetation coverage.
Ecosystem development on closed landfills can be rapid
and is accelerated by artificial planting and good manage
ment practices.
Landfills
307
Fauna
Ecological Approach
The landfill cover supports vegetation which serves as a
habitat for native fauna, but ecologically, it is useless if
rehabilitated landfills fail to provide suitable grounds for
faunal colonization. Not much has been done on the
faunal assemblages on closed landfills, but rehabilitated
landfills are potential sites for faunal colonization because
they attract insects and herpetofauna and have an impor
tant role to play in wildlife conservation.
Open grasslands developed on abandoned landfills
are an important insect habitat, and some closed landfills
which have been converted into woodlands or grasslands
provide valuable habitat for butterflies, especially those
species which are declining in population and distribu
tion. However, butterfly community composition and
structure have stronger links with vegetation that are
either a source of nectar or host plants for larvae, and do
not necessarily reflect the successional development of
closed landfills.
Closed landfills could also be colonized by amphibians
and reptiles within a few years after revegetation, and
herpetofaunal diversity and abundance increase with
time after closure. Constructed wetlands, though not a
conventional option for habitat creation on landfills, pro
vide refuges for amphibians and reptiles. An example of
this are ponds that have been designed and constructed on
a landfill in Cheshire, England, specifically for great
crested newts that were originally present on the site
before landfilling.
Birds play a very vital role in the secondary succession
on landfills as seed dispersers. It has been reported that
birds introduced 20 new plant species to a landfill
annually via endozoochory. This increases the floral
diversity and contributes to vegetation development.
However, only species that produce fleshy fruits will be
spread by frugivorous species. It is generally advocated
that more fleshy fruited natives should be planted to
attract birds, and even small mammals such as bats for
full restoration of the ecological function of landfill as a
wildlife habitat.
The reestablishment of faunal communities is closely
related to that of vegetation. Closed landfills are potential
refuges for uncommon and rare species, and it is sug
gested that planting of more natives can aid in the
creation of a more favorable habitat for ecological diver
sity. Rehabilitated landfills may not be as ecologically
diverse as natural areas, but their conservation values
should not be overlooked, as they can be good wildlife
habitat and connecting links to enhance remnant frag
mented areas. Sites with relatively high biodiversity and
rare species records should be designated conservation
areas, especially for those which are not suitable for
other alternative development.
The basic ecological principles of successional development
are totally applicable to rehabilitated landfills, and rehabilita
tion success depends on the reestablishment of biological
activities of surface horizons in the long term. The natural
succession of grassland to woodland ecosystem is slow and
may take up to 50 years. It is generally accepted that inter
vention of ecosystem reconstruction followed by natural
succession is the best practicable option for landfills. If
closed landfills were reclaimed properly, they could pro
vide an attractive source of land for nature conservation,
forestry, and recreation. However, the success of reclama
tion depends much upon the growth of plants and the
efficient cycling of nutrients in the cover material. An
integrated approach which includes gas control, soil man
agement, and directed succession can accelerate the
development of a sustainable ecosystem in terms of struc
ture and function on closed landfills.
See also: Biological Wastewater Treatment Systems.
Further Reading
Chan YSG, Wong MH, and Whitton BA (1996) Effects of landfill factors
on tree cover: A field survey at 13 landfill sites in Hong Kong. Land
Contamination and Reclamation 2: 115 128.
Chan YSG, Chu LM, and Wong MH (1997) Influence of landfill factors on
plants and soil fauna: An ecological perspective. Environmental
Pollution 97: 39 44.
Dobson MC and Moffat AJ (1993) The Potential for Woodland
Establishment on Landfill Sites, 88pp. London: Department of the
Environment, HMSO.
Dobson MC and Moffat AJ (1995) A re evaluation of objections to tree
planting on containment landfills. Waste Management and Research
13: 579 600.
Ecoscope (2000) Wildlife Management and Habitat Creation on Landfill
Sites: A Manual of Best Practice. Muker, UK: Ecoscope Applied
Ecologists.
Ettala MO, Yrjonen KM, and Rossi EJ (1988) Vegetation coverage at sanitary
landfills in Finland. Waste Management and Research 6: 281 289.
Flower FB, Leone IA, Gilman EF, and Arthur JJ (1978) A Study of
Vegetation Problems Associated with Refuse Landfills, EPA 600/2
78 094, 130pp. Cincinnati: USEPA.
Handel SN, Robinson GR, Parsons WFJ, and Mattei JH (1997)
Restoration of woody plants to capped landfills: Root dynamics in an
engineered soil. Restoration Ecology 5: 178 186.
Moffat AJ and Houston TJ (1991) Tree establishment and growth at Pitsea
landfill site, Essex, U.K. Waste Management and Research 9: 35 46.
Neumann U and Christensen TH (1996) Effects of landfill gas
on vegetation. In: Christensen TH, Cossu R, and Stegmann R
(eds.) Landfilling of Waste: Biogas, pp. 155 162. London:
E & FN Spon.
Robinson GR and Handel SN (1993) Forest restoration on a closed
landfill: Rapid addition of new species by bird dispersal.
Conservation Biology 7: 271 278.
Simmons E (1999) Restoration of landfill sites for ecological diversity.
Waste Management and Research 17: 511 519.
Wong MH (1988) Soil and plant characteristics of landfill sites
near Merseyside, England. Environmental Management 12: 491 499.
Wong MH (1995) Growing trees on landfills. In: Moo Young M,
Anderson WA, and Chakrabarty AM (eds.) Environmental
Biotechnology: Principles and Applications, pp. 63 77. Amsterdam:
Kluwer Academic.
308
Mangrove Wetlands
Mangrove Wetlands
R R Twilley, Louisiana State University, Baton Rouge, LA, USA
Published by Elsevier B.V.
Introduction
Ecogeomorphology of Mangroves
Biodiversity
Ecosystem Processes
Impacts of Environmental Change
Management and Restoration
Further Reading
Introduction
with geophysical processes control the basic patterns in
forest structure and growth. These coastal geomorphic
settings can be found in a variety of life zones that depend
on regional climate and oceanographic processes.
Hydroperiod of mangroves resulting from gradients in
microtopography and tidal hydrology (Figure 1) can
influence the zonation of mangroves from shoreline to
more inland locations forming ecological types of man
grove wetlands. Lugo and Snedaker identified ecological
types of mangroves based on topographic location and
patterns of inundation at local scales (riverine, fringe,
basin, and dwarf; Figure 1) that Woodroffe summarized
into basically three geomorphic types (riverine, fringe,
and inland). A combination of ecological types of man
groves can occur within any one of the geomorphic
settings occurring at a hierarchy of spatial scales that
can be used to classify mangrove wetlands.
Various combinations of geophysical processes and geo
morphologic landscapes produce gradients of regulators,
resources, and hydroperiod that control mangrove growth
(Figure 2). Regulator gradients include salinity, sulfide, pH,
and redox that are nonresource variables that influence
mangrove growth. Resource gradients include nutrients,
light, space, and other variables that are consumed and
contribute to mangrove productivity. The third gradient,
hydroperiod, is one of the critical characteristics of wetland
landscapes that controls wetland productivity. The interac
tions of these three gradients have been proposed as a
constraint envelope for defining the structure and produc
tivity of mangrove wetlands based on the relative degree of
stress conditions (Figure 2). At low levels of stress for all
three environmental gradients (such as low salinity, high
nutrients, and intermediate flooding), mangrove wetlands
reach their maximum levels of biomass and net ecosystem
productivity.
Soil nutrients are not uniformly distributed within
mangrove ecosystems, resulting in multiple patterns of
nutrient limitation. Along a microtidal gradient in carbon
ate reef islands, trees were generally N limited in the
fringe zone and P limited in the interior or scrub zone.
Fertilization studies demonstrated that not all ecological
processes respond similarly to or are limited by the
Mangroves refer to a unique group of forested wetlands
that dominate the intertidal zone of tropical and subtropi
cal coastal landscapes generally between 25 N and 25 S
latitude. These tropical forests grow along continental
margins between land and the sea across the entire sali
nity spectrum from nearly freshwater (oligohaline) to
marine (euhaline) conditions. The coastal forests also
inhabit nearly every type of coastal geomorphic forma
tion from riverine deltas to oceanic reefs – another
example of the tremendous ‘biodiversity’ of mangrove
ecosystems. Mangroves are trees considered as a group
of halophytes with species from 12 genera in eight differ
ent families. A total of 36 species has been described from
the Indo West Pacific area, but fewer than ten species are
found in the new world tropics. The term mangroves may
best define a specific type of tree, whereas mangrove
wetlands refers to whole plant associations with other
community assemblages in the intertidal zone, similar to
the term ‘mangal’ introduced by Macnae to refer to
swamp ecosystems. In addition, the habitats of tropical
estuaries consist of a variety of primary producers and
secondary consumers distributed in bays and lagoons that
have the intertidal zone dominated by mangrove wet
lands. These may be referred to as mangrove dominated
estuaries.
There are numerous reviews and books that describe
the ecology and management of mangroves around the
world, including references describing techniques to
study the ecology of mangrove wetlands.
Ecogeomorphology of Mangroves
The environmental settings of mangroves are a complex
behavior of regional climate, tides, river discharge, wind,
and oceanographic currents (Figure 1). There are about
240 103 km2 of mangroves that dominate tropical con
tinental margins from river deltas, lagoons, and estuarine
settings to islands in oceanic formations (noncontinental).
The landform characteristics of a coastal region together
Mangrove Wetlands
309
Global distribution:
temperature
25° N
25° S
Geomorphological type:
environmental settings
Delta
Oceanic islands
km2
Lagoon
Scrub
Estuary
Basin
Ecological type:
hydrology and topography
nd
la
In
ge
ha to km2
in
Fr
Riverine
Habitat units:
mangroves and soil
Fringe
ha
Resource
gradients
Regulator
gradients
Hydroperiod
gradients
Figure 1 Hierarchical classification system to describe patterns of mangrove structure and function based on global, geomorphic
(regional), and ecological (local) factors that control the concentration of nutrient resources and regulators in soil along gradients from
fringe to more interior locations from shore. Modified from Twilley RR, Gottfried RR, Rivera-Monroy VH, Armijos MM, and Bodero A
(1998) An approach and preliminary model of integrating ecological and economic constraints of environmental quality in the Guayas
River estuary, Ecuador. Environmental Science and Policy 1: 271–288 and Twilley RR and Rivera-Monroy VH (2005) Developing
performance measures of mangrove wetlands using simulation models of hydrology, nutrient biogeochemistry, and community
dynamics. Journal of Coastal Research 40: 79–93.
310
Mangrove Wetlands
Benign
Resource gradient
Regulator gradient
Hydroperiod
Stress
Growth
Benign
Growth
Growth
Benign
Stress
Regulator gradient
Stress
Resource gradient
Stress
Hydroperiod
Figure 2 Interaction of three factors controlling the productivity of coastal wetlands, including regulator gradients, resource gradients,
and hydroperiod. The bottom panel defines stress conditions associated with how gradients in each factor control growth of wetland
vegetation. From Twilley RR and Rivera-Monroy VH (2005) Developing performance measures of mangrove wetlands using simulation
models of hydrology, nutrient biogeochemistry, and community dynamics. Journal of Coastal Research 40: 79–93.
same nutrient. It is also apparent that mangrove forests
growing in other ecogeomorphic settings are also prone to
P limitation associated with different geophysical pro
cesses. One of the most critical regulator gradients
(Figure 2) controlling mangrove establishment, seedling
survival, growth, height, and zonation is salinity, depend
ing on their ability to balance water and salt. Interspecific
differential response of mangrove propagules to salinity
occurs at salinities from 45 to 60 g kg 1. The 13C and
15N signatures of mangrove leaf tissue can indicate stress
conditions such as drought, limited nutrients, and hyper
salinity across a variety of environmental settings.
Biodiversity
Mangrove ecosystems support a variety of marine and
estuarine food webs involving an extraordinarily large
number of animal species and complex heterotrophic
microorganism food web. In the New World tropics,
extensive surveys of the composition and ecology of
mangrove nekton have found 26–114 species of fish. In
addition to the marine and estuarine food webs and asso
ciated species, there are a relatively large number and
variety of animals that range from terrestrial insects to
birds that live in and/or feed directly on mangrove vege
tation. These include sessile organisms (such as oysters
and tunicates), arboreal feeders (such as foliovores and
frugivores), and ground level seed predators. Sponges,
tunicates, and a variety of other forms of epibionts on
prop roots of mangroves are highly diverse, especially
along mangrove shorelines with little terrigenous input.
Over 200 species of insects have been documented in
mangroves in the Florida Keys, similar to the richness of
insects and faunal biota observed in other parts of the
Caribbean. One of the most published links between
mangrove biodiversity and ecosystem function may be
the presence of crabs in mangrove wetlands. Crabs can
Mangrove Wetlands
influence forest structure, litter dynamics, and nutrient
cycling of mangrove wetlands, suggesting that they are a
keystone guild in these forested ecosystems.
Ecosystem Processes
Succession
Succession in mangroves has often been equated with zona
tion, wherein ‘pioneer species’ would be found in the fringe
zones, and zones of vegetation more landward would ‘reca
pitulate’ the successional sequence toward terrestrial
communities. Zonation in mangrove communities has
variously been accounted for by a number of biological
factors, including salinity tolerance of individual species,
seedling dispersal patterns resulting from different sizes of
mangrove propagules, differential consumption by grapsid
crabs and other consumers, and interspecific competition.
Snedaker proposed the establishment of stable monospecific
zones wherein each species is best adapted to flourish due to
the interaction of physiological tolerances of species with
environmental conditions. Geological surveys of the inter
tidal zone of Tabasco, Mexico, demonstrated that the
zonation and structure of mangrove wetlands are responsive
to eustatic changes in sea level, and that mangrove zones
can be viewed as steady state zones migrating toward or
away from the sea, depending on its level. Thus, both
monospecific and mixed vegetation zones of mangrove
wetlands represent steady state adjustments rather than
successional stages. Many models of mangrove succession
are based on how gap dynamics influence spatial patches of
community dynamics across the landscape.
Productivity and Litter Dynamics
Tree height and aboveground biomass of mangrove
wetlands throughout the tropics decrease at higher lati
tudes, indicating the constraint of climate on forest
development in the subtropical climates. In addition,
mangrove biomass can vary dramatically within any
given latitude, an indication that local effects of regula
tors, resources, or hydroperiod may significantly limit the
potential for forest development at all latitudes. The
primary productivity of mangroves is most often evalu
ated by measuring the rate of litter fall, as recorded for
other forested wetlands. Regional rates in litter produc
tion in mangroves are a function of water turnover within
the forest, and rank among the ecological types is as
follows: riverine > fringe > basin > scrub.
The dynamics of mangrove litter, including produc
tivity, decomposition, and export, can determine the
coupling of mangroves to the secondary productivity
and biogeochemistry of coastal ecosystems. Patterns of
leaf litter turnover have been proposed to vary among
ecological types of mangroves with greater litter export in
311
sites with increasing tidal inundation (riverine > fringe >
basin). However, several studies in the Old World tropics
in higher energy coastal environments of Australia and
Malaysia have emphasized the influence of crabs on the
fate of mangrove leaf litter, rather than geophysical pro
cesses. In these coastal environments, crabs consume
28–79% of the annual leaf fall. A similar biological factor
was observed in the neotropics where the crab Ucides
occidentaliss in the Guayas River estuary (Ecuador)
processed leaf litter at similar rates observed in Old
World tropics. Differences in litter turnover rates among
mangrove wetlands are a combination of species specific
degradation rates, hydrology (tidal frequency), soil ferti
lity, and biological factors such as crabs.
Nutrient Biogeochemistry
The nutrient biogeochemistry of mangrove wetlands as
either a nutrient source or sink depends on the process of
material exchange at the interface between mangrove
wetlands and the estuary, which is largely controlled
by tides (tidal exchange, TE, in Figure 3). Nutrient
exchanges may occur either with coastal waters (TE) or
with the atmosphere (atmosphere exchange, AE),
depending on whether the nutrient has a gas phase or
not (Figure 3). Substantial amounts of carbon and nitro
gen can exchange with the atmosphere, resulting in very
complex mechanisms both at the interface with coastal
waters and with the atmosphere that influence the mass
balance of these nutrients. In addition, there are internal
processes, including root uptake (UT), retranslocation
(RT) in the canopy, litter fall (LF), regeneration (RG),
immobilization (IM), and sedimentation (SD) (Figure 3).
The balance of these nutrient flows will determine the
exchanges across the wetland boundary.
There are very few comprehensive budgets of carbon,
nitrogen, or phosphorus for mangrove ecosystems.
Mangrove sediments have a high potential in the removal
of N from surface waters, yet estimates of denitrification
have a large range from a low of 0.53 mmol m 2 h 1, to
9.7–261 mmol N m 2 h 1 in mangrove forests receiving
effluents from sewage treatment plants. Small amendments
of 15NO3 followed by direct measures of 15N2 production
have shown that denitrification accounts for <10% of the
applied isotope suggesting that NO3 is accumulated in the
litter via immobilization on the forest floor rather than a sink
to the atmosphere. The other nutrient sink in mangrove
wetlands is the burial of nitrogen and phosphorus associated
with sedimentation. A survey of sedimentation and nutrient
accumulation among five sites in south Florida and Mexico
indicates patterns associated with the ecological types of
mangroves, with rates of about 5.5 g m 2 yr 1. This rate is
higher than nitrogen loss via denitrification, indicating the
significance of burial as nitrogen sink in mangrove ecosys
tems. Intrasystem nutrient recycling mechanisms in the
312
Mangrove Wetlands
CO2, N2, CH4
Tides
Atmosphere
exchange
Allochthonous
input
Tidal
exchange
Immobilization
IN
Uptake
Roots/
stems
River
Litter
Regeneration
Shelf
exchange
Estuary
Sedimentation
Canopy
Litter fall
Retranslocation
Peat
RT
AE
LF
TE
IM
RG
UT
Nutrient resource
SD
Figure 3 Upper panel: Schematic of the various fluxes of organic matter and nutrients in a mangrove ecosystem, including exchange
with the estuary (IN inorganic nutrients). Lower panel: A diagram of a mangrove wetland with soil nutrient resources describing the
various processes associated with intrasystem cycling and exchange. From Twilley RR (1997) Mangrove wetlands. In: Messina M and
Connor W b0665 (eds.) Southern Forested Wetlands: Ecology and Management, pp. 445-473. Boca Raton, FL: CRC Press.
canopy may be a site of nitrogen conservation in mangroves
and, together with leaf longevity, could influence the nitro
gen demand of these ecosystems. The significance of this
ecological process to the nutrient budget of different man
grove wetlands has not been determined.
Surveys of nitrogen exchange demonstrate some of the
principles of determining the function of mangrove wet
lands as a nutrient sink. The largest nitrogen flux of
nitrogen from sites in Mexico and Australia is export of
particulate nitrogen, consistent with organic carbon repre
senting the largest flux from most mangroves (Figure 3).
Compared to other flux studies of mangroves, there seems
to be a pattern of net inorganic fluxes into the wetlands and
corresponding flux of organic nutrients out. The best sum
mary may be that mangrove wetlands transform the tidal
import of inorganic nutrients into organic nutrients that are
then exported to coastal waters. Carbon export from man
grove ecosystems ranges from 1.86 to 401 g Cm 2 yr 1,
with an average rate of about 210 g Cm 2 yr 1. Carbon
export from mangrove wetlands is nearly double the rate
of average carbon export from salt marshes, which may be
associated with the more buoyant mangrove leaf litter,
higher precipitation in tropical wetlands, and greater tidal
amplitude in mangrove systems studied.
Mangrove Food Webs
The function of mangrove wetlands as a source of habitat
and food to estuarine dependent fisheries is one of the
most celebrated values of forested wetlands. There are
Mangrove Wetlands
several excellent reviews that describe the secondary
productivity of tropical mangrove ecosystems. The origi
nal ‘outwelling hypothesis’ of mangroves has been revised
from the original paradigms based on comparisons among
different mangrove estuaries using natural isotope abun
dance to trace mangrove organic matter through
estuarine food chains. There are seasonal and spatial
differences in the amount of mangrove detritus that can
be measured in shrimp and fish that inhabit mangrove
estuaries. If the distance from the source of mangrove
detritus increases, the proportion of carbon in the tissue
of shrimp from mangrove detritus decreases as the signal
of carbon phytoplankton increases. The seasonal timing of
mangrove export of detritus relative to the migration of
estuarine dependent fisheries may also dilute the contri
bution of mangrove detritus from the food webs among
diverse sites. The migratory nature of many of the nekton
communities and the seasonal pulsing of both organic
detritus input and in situ productivity result in very com
plex linkages of mangroves with estuarine dependent
fisheries. In addition, mangrove detritus low in nitrogen
relative to carbon may be modified by the microbial
community and then utilized by higher trophic levels,
masking the direct utilization of this organic matter as
an energy source.
Impacts of Environmental Change
Mangroves are arguably an excellent indicator of how
ecosystems will respond to the manifold impacts of global
environmental change and land use disturbance. Given
present patterns, the combined effects of climate and
land use change will be noticeably evident in reduced
goods and services of mangroves to human systems
throughout the tropics in the twenty first century.
For example, accelerated rates in sea level rise have
been speculated as the most critical environmental
change affecting the continued existence of mangrove
ecosystems. Numerous processes contribute to vertical
accretion of mangroves at a rate that balances the increase
in regional sea level rise. Critical rates in sea level rise
have been estimated above which there is a projected
collapse of mangrove ecosystems. While some speculation
suggests that mangroves cannot sustain existence at
sea level rise >1.2–2.3 mm yr 1, there is evidence that
mangroves located in particular environmental settings
existed through periods of accelerated sea level rise.
Mangroves in Australia can keep pace with changes in
sea level rise with rates ranging from 0.2 to 6 mm yr 1
in the south Alligator tidal river. Also, mangrove forests in
many estuaries in northern Australia tolerated sea level
rise of 8–10 mm yr 1 in the early Holocene. Many of
these mangroves receive terrigenous sediments and exist
in macrotidal environments, with critical rates that are
313
much different than for mangroves in microtidal and
carbonate environments. In addition, mangrove areas
can be sustained along the coastline by migrating inland
under conditions of increased sea level rise. But this
inland migration will depend on whether suitable inshore
landscapes are available. The most significant recent
restriction to mangrove colonization is human land use
of available landscapes.
Mangroves in many coastal regions such as Gulf of
Mexico and Caribbean are distributed in latitudes where
the frequency of hurricanes and cyclones is high, resulting
in strong effect on mangrove forest structure and commu
nity dynamics. Several patterns have been observed in
Florida, Puerto Rico, Mauritius, and British Honduras.
Species attributes and availability of propagules are impor
tant factors along with the severity of storm and sediment
disturbance in projecting recovery patterns. Frequent
storm disturbance tends to favor species capable of constant
or timely flowering, abundant seedling or sprouting, fast
growth in open conditions, and early reproductive matur
ity. Woody debris resulting from these disturbances have
an important role in biogeochemical properties of dis
turbed mangrove forests. Although mangrove trees show
these ‘traits’, it is important to consider the cumulative
impact of human activities on these ecosystems in conjunc
tion with the complex natural cycle of regeneration and
growth of mangrove forests. Cyclonic disturbance in areas
with higher rates of sea level rise has been demonstrated to
cause sediment collapse (drop in surface elevation) that
reduces the ability of mangroves to recolonize disturbed
areas. Yet this potential impact may vary across ecogeo
morphic types of mangroves.
River (and surface runoff) diversions that deprive tro
pical coastal deltas of freshwater and silt result in losses of
mangrove species diversity and organic production, and
alter the terrestrial and aquatic food webs that mangrove
ecosystems support. Freshwater diversion of the Indus
River to agriculture in Sind Province over the last several
hundred years has reduced the once species rich Indus
River delta to a sparse community dominated by
Avicennia marina. It is also responsible for causing signifi
cant erosion of the seafront due to sediment starvation
and the silting in of the abandoned spill rivers. A similar
phenomenon has been observed in southwestern
Bangladesh following natural changes in river channels
of the Ganges and the construction of the Farakka barrage
that reduced the dry season flow of freshwater into the
mangrove dominated western Sundarbans. Freshwater
starvation, both natural and human induced, has had
negative impacts on the biodiversity of mangroves in the
Ganges River delta as well along the dry coastal life zone
of Colombia (the Cı́enaga Grande de Santa Marta
lagoon).
Deforestation of mangrove wetlands is associated with
many uses of coastal environments, including urban,
314
Mangrove Wetlands
agriculture, and aquaculture reclamation, as well as the
use of forest timber for furniture, energy, chip wood, and
construction materials. Two reclamation activities that
have contributed to examples of massive mangrove defor
estation are agriculture and aquaculture enterprises.
Agriculture impacts on mangroves are most noted in
West Africa and parts of Indonesia. Many of the large
agricultural uses are found in humid coastal areas or
deltas where freshwater is abundant and intertidal lands
are seasonally available for crop production. Mariculture
use of the tropical intertidal zones, in the construction and
operation of shrimp ponds, has become one of the most
significant environmental changes of mangrove wetlands
and water quality of tropical estuaries in the last several
decades.
Oil spills represent contaminants to mangroves that
can alter the succession, productivity, and nutrient
cycling of these coastal forested wetlands. These impacts
have been well documented in ecological studies in
Puerto Rico, Panama, and Gulf of Mexico. An oil slick
in a mangrove wetland will cause a certain mortality of
trees depending on the concentration of hydrocarbons
and species of trees, as well as the edaphic stress levels
already existing at the site. Thus, those mangroves in dry
coastal environments may be more vulnerable to oil spills
than those in more humid environments.
River
Coastal
ocean
Land
use
Management and Restoration
Mangroves produce a variety of forest products, support
the productivity of economically important estuarine
dependent fisheries, and modify the water quality in
warm temperate and tropical estuarine ecosystems.
These goods and services lead to increased human utiliza
tion of mangrove resources that vary throughout the
tropics depending on economic and cultural constraints
(Figure 4). Economic constraints are usually in the form of
available capital to fund land use changes in coastal
regions, as well as river basin development. Cultural con
straints are complex and determine the degree of
environmental management and natural resource utiliza
tion. However, the sustainable utilization of coastal
resources, to a large degree, is controlled by these two
social conditions of a region. Human use and value of
mangrove wetlands are therefore a combination of both
the ecological properties of these coastal ecosystems
together with patterns of social exploitation. Therefore,
any best management plan designed to provide for the
sustainable utilization of mangrove wetlands has to con
sider both the ecological and social constraints of the
region. Humans are part of all ecosystems, and manage
ment of natural resources is a combination of policies that
seek to regulate the actions of societies within limitations
Management impacts
Market
forces
Estuarine ecosystem
Tide
Rain
Refugia
Primary productivity
Sedimentation
Nutrient cycling
Intertidal zone
Mangroves
Wind
Refugia
Primary productivity
Sedimentation
Nutrient cycling
Organic matter export
Shrimp ponds
Sun
Fertilization
Feeding
Sedimentation
Pumping
Ecosystem properties
Natural resources
Habitat quality
Water quality
Nutrient sinks
Sediment trap
Shoreline protection
Aesthetics
Timber production
Wildlife protection
Biodiversity
Management
decisions
Agriculture reclamation
Mariculture reclamation
Urbanization
Industrialization
Conservation areas
Exchange
rates
Trade
policies
Monetary
policies
Economic value
Engineered
resources
Habitat quality
Nutrient sources
Sediment trap
Water quality
Goods and services
Mariculture profits
Commercial fisheries
Sport fisheries
Timber products
Charcoal/energy
Salt production
Agriculture
Ecotourism
Real estate
Socioeconomic
properties
Fiscal
policies
Political
forces
Cultural
forces
Figure 4 Conceptual framework constraints of environmental setting and human activities on ecosystem properties, ecological
functions, and uses of mangrove ecosystems that determine management decisions in coastal environments. From Twilley RR,
Gottfried RR, Rivera-Monroy VH, Armijos MM, and Bodero A (1998) An approach and preliminary model of integrating ecological and
economic constraints of environmental quality in the Guayas River estuary, Ecuador. Environmental Science and Policy 1: 271–288.
Mangrove Wetlands
that are imposed by the environment. Recent emphasis has
been placed on comprehensive ecosystem restoration pro
grams that represent changes in management of landscapes
to reduce impacts on natural processes that enhance sys
tem recovery.
There have been several reviews of mangrove restora
tion, which collectively have alluded to the concept that
since these forested wetlands are adapted to stressed
environments, they are relatively amenable to restoration
efforts. The success of mangrove restoration is the estab
lishment of the proper environmental settings that control
the characteristic structure and function of mangrove
wetlands. The goal of ecological restoration is to return
a degraded mangrove site back to either the natural con
dition (restoration) or to some other new condition
(rehabilitation). The rates of change in the ecological
characteristics of mangrove wetlands between natural,
degraded, and some rehabilitated condition will depend
on the type of environmental impact, the magnitude of
the impact, and the ecogeomorphic type of mangrove
wetland that is impacted. The success of any mangrove
restoration project depends on the establishment of
proper site conditions (geophysical processes and geo
morphic features) along with ecological processes of the
site such as the availability of propagules and the recruit
ment of these individuals to sapling stage of development.
Some of the key parameters of a restoration project
include the elevation of the landscape to provide the
proper hydrology of the site, recognizing the significance
of natural processes to sustaining the restored condition,
and proper planting techniques to enhance recruitment.
Several models of different properties of mangroves have
been developed during the last decade to help facilitate
planning and design of mangrove restoration projects and
improve our management of these critical features of
coastal landscape.
See also: Lagoons; Mediterranean.
Further Reading
Alexander TR (1967) Effect of Hurricane Betsy on the southeastern
Everglades. Quarterly Journal of the Florida Academy of Sciences
30: 10 24.
Allen JA, Ewel KC, Keeland BD, Tara T, and Smith TJ (2000) Downed
wood in Micronesian mangrove forests. Wetlands 20: 169 176.
Alongi DM, Christoffersen P, Tirendi F, and Robertson AI (1992) The
influence of freshwater and material export on sedimentary facies and
benthic processes within the Fly Delta and adjacent Gulf of Papua
(Papua New Guinea). Continental Shelf Research 12: 287 326.
Bacon PR (1990) The ecology and management of swamp forests in the
Guianas and Caribbean region. In: Lugo AE, Brinson M, and Brown S
(eds.) Ecosystems of the World 15: Forested Wetlands, pp. 213 250.
Amsterdam: Elsevier Press.
Bacon PR (1994) Template for evaluation of impacts of sea level rise on
Caribbean coastal wetlands. Ecological Engineering 3: 171 186.
315
Baldwin A, Egnotovich M, Ford M, and Platt W (2001) Regeneration in
fringe mangrove forests damaged by Hurricane Andrew. Plant
Ecology 157: 151 164.
Ball MC (1980) Patterns of secondary succession in a mangrove forest
of southern Florida. Oecologia 44: 226 235.
Ball MC (1988) Ecophysiology of mangroves. Trees 2: 129 142.
Berger U and Hildenbrandt H (2000) A new approach to spatially
explicit modelling of forest dynamics: Spacing, ageing and
neighborhood competition of mangrove trees. Ecological
Modelling 132: 287 302.
Blasco F (1984) Climatic factors and the biology of mangrove plants.
In: Snedaker SC and Snedaker JG (eds.) The Mangrove Ecosystem:
Research Methods, pp. 18 35. Paris: UNESCO.
Botero L (1990) Massive mangrove mortality on the Caribbean coast of
Colombia. Vida Silvestre Neotropical 2: 77 78.
Boto KG, Saffingna P, and Clough B (1985) Role of nitrate in nitrogen
nutrition of the mangrove Avicennia Marina. Marine Ecology Progress
Series 21: 259 265.
Boto KG and Wellington JT (1988) Seasonal variations in concentrations
and fluxes of dissolved organic and inorganic materials in a tropical,
tidally dominated, mangrove waterway. Marine Ecology Progress
Series 50: 151 160.
Brown S and Lugo AE (1994) Rehabilitation of tropical lands: A key to
sustaining development. Restoration Ecology 2: 97 111.
Camilleri JC (1992) Leaf litter processing by invertebrates in a mangrove
forest in Queensland. Marine Biology 114: 139 145.
Carlton JM (1974) Land building and stabilization by mangroves.
Environmental Conservation 1: 285.
Chapman VJ (1976) Mangrove Vegetation. Vaduz, Germany: J. Cramer.
Chen R and Twilley RR (1998) A gap dynamic model of mangrove forest
development along gradients of soil salinity and nutrient resources.
Journal of Ecology 86: 1 12.
Chen R and Twilley RR (1998) A simulation model of organic matter and
nutrient accumulation in mangrove wetland soils. Biogeochemistry
44: 93 118.
Chen R and Twilley RR (1999) Patterns of mangrove forest structure and
soil nutrient dynamics along the Shark River Estuary, Florida.
Estuaries 22: 955 970.
Cintrón G (1990) Restoration of mangrove systems. Symposium on
Habitat Restoration. Washington, DC: National Oceanic and
Atmospheric Administration.
Cintrón G, Lugo AE, Martinez R, Cintrón BB, and Encarnacion L (1981)
Impact of oil in the tropical marine environment, pp. 18 27. Technical
Publication. Division of Marine Resources, Department of Natural
Resources of Puerto Rico.
Corredor JE and Morell MJ (1994) Nitrate depuration of secondary
sewage effluents in mangrove sediments. Estuaries 17: 295 300.
Craighead FC and Gilbert VC (1962) The effects of Hurricane Donna on
the vegetation of southern Florida. Quarterly Journal of the Florida
Academy of Sciences 25: 1 28.
Davis JH (1940) The ecology and geologic role of mangroves in Florida.
In: Carnegie Institution, Publication No. 517, pp. 303 412.
Washington, DC: Carneigie Institution.
Davis S, Childers DL, Day JWJ, Rudnick D, and Sklar F (2001)
Wetland water column exchanges of carbon, nitrogen, and
phosphorus in a southern everglades dwarf mangrove. Estuaries
24: 610 622.
Davis S, Childers DL, Day JW, Rudnick D, and Sklar F (2003) Factors
affecting the concentration and flux of materials in two southern
everglades mangrove wetlands. Marine Ecology Progress Series
253: 85 96.
Duke NC (2001) Gap creation and regenerative process driving diversity
and structure of mangrove ecosystems. Wetlands Ecology and
Management 9: 257 269.
Duke NC and Pinzon Z (1993) Mangrove forests. In: Keller BD and
Jackson JBC (eds.) Long Term Assessment of the Oil Spill at Bahia
las Minas, Panama, Synthesis Report, Volumn II, Technical Report,
pp. 447 553. New Orleans, LA: US Dept of the Interior, Minerals
Management Service, Gulf of Mexico OCS Regional Office.
Ellison JC (1993) Mangrove retreat with rising sea level Bermuda.
Estuarine, Coastal and Shelf Science 37: 75 87.
Ellison AM (2000) Mangrove restoration: Do we know enough?
Restoration Ecology 8: 219 229.
316
Mangrove Wetlands
Ellison AM and Farnsworth EJ (1992) The ecology of Belizean
mangrove root fouling communities: Patterns of epibiont
distribution and abundance, and effects on root growth.
Hydrobiologia 20: 1 12.
Ellison JC and Stoddart DR (1991) Mangrove ecosystem collapse during
predicted sea level rise: Holocene analogues and implications.
Journal of Coastal Research 7: 151 165.
Ewe SML, Gaiser EE, Childers DL, et al. (2006) Spatial and temporal
patterns of aboveground net primary productivity (ANPP) in the
Florida Coastal Everglades. Hydrobiologia 569: 459 474.
Ewel KC, Ong JE, and Twilley R (1998) Different kinds of mangrove
swamps provide different goods and services. Global Ecology and
Biogeography Letters 7: 83 94.
Ewel KC, Zheng SF, Pinzon ZS, and Bourgeois JA (1998) Environmental
effects of canopy gap formation in high rainfall mangrove forests.
Biotropica 30: 510 518.
Farnsworth EJ and Ellison AM (1991) Patterns of herbivory in Belizean
mangrove swamps. Biotropica 23: 555 567.
Farnsworth EJ and Ellison AM (1993) Dynamics of herbivory in Belizean
mangal. Journal of Tropical Ecology 9: 435 453.
Farquhar GD, Ball MC, von Caemmerer S, and Roksandic Z (1982)
Effect of salinity and humidity on 13C values of halophytes
evidence for diffusional isotope fractionation determined by the ratios
of intercellular/atmospheric CO2 under different environmental
conditions. Oecologia (Berlin) 52: 121 137.
Fell JW and Master IM (1973) Fungi associated with the degradation of
mangrove (Rhizophora mangle L.) leaves in south Florida.
In: Stevenson LH and Colwell RR (eds.) Estuarine Microbial Ecology,
pp. 455 465. Columbia, SC: University of South Carolina Press.
Feller IC (1993) Effects of Nutrient Enrichment on Growth and
Herbivory of Dwarf Red mangrove. PhD Dissertation, Georgetown
University.
Feller IC (1995) Effects of nutrient enrichment on growth and herbivory of
dwarf red mangrove (Rhizophora mangle). Ecological Monographs
65: 477 505.
Feller IC and McKee KL (1999) Small gap creation in Belizean mangrove
forests by a wood boring insect. Biotropica 31: 607 617.
Feller IC, Whigham DF, McKee KL, and Lovelock CE (2003) Nitrogen
limitation of growth and nutrient dynamics in a disturbed mangrove
forest, Indian River Lagoon, Florida. Oecologia 134: 405 414.
Feller IC, Whigham DF, O9Neill JP, and McKee KL (1999) Effects of
nutrient enrichment on within stand cycling in a mangrove forest.
Ecology 80: 2193 2205.
Field CD (1996) Restoration of mangrove ecosystems. In: International
Society for Mangrove Ecosystems. Hong Kong: South China
Printing.
Fry B, Bern AL, Ross MS, and Meeder JF (2000) 15N studies of nitrogen
use by the red mangrove, Rhizophora mangle L., in south Florida.
Estuarine, Coastal and Shelf Science 50: 723 735.
Fry B and Smith TJ, III (2002) Stable isotope studies of red mangroves
and filter feeders from the Shark River estuary, Florida. Bulletin of
Marine Sciences 70: 871 890.
Garrity SD, Levings SC, and Burns KA (1994) The Galeta oil spill. I. Long
term effects on the physical structure of the mangrove fringe.
Estuarine, Coastal and Shelf Science 38: 327 348.
Getter CD, Scott GI, and Michel J (1981) The effects of oil spills on
mangrove forests: A comparison of five oil spill sites in the Gulf of
Mexico and the Caribbean Sea. Proceedings of the 1981 Oil Spill
Conference, pp. 65 11. Washington, DC: API/EPA/USCG.
Gilmore RG, Jr. and Snedaker SC (1993) Mangrove forests.
In: Martin WH, Boyce SG, and Echternacht AC (eds.) Biodiversity of
the Southeastern United States/Lowland Terrestrial Communities,
pp. 165 198. New York: Wiley.
Glynn PW, Almodovar LR, and Gonzalez JG (1964) Effects of hurricane
Edith on marine life in La Parguera, Pueto Rico. Caribbean Journal of
Science 4: 335 345.
Gosselink JG and Turner RE (1978) The role of hydrology in freshwater
wetland ecosystems. In: Good DFWRE and Simpson RL (eds.)
Freshwater Wetlands: Ecological Processes and Management
Potential, pp. 633 678. New York: Academic Press.
Hedgpeth JW (1957) Classification of marine environments. Geological
Society of America, Memoir 67(1): 17 28.
Huston MA (1994) Biological Diversity. Cambridge: Cambridge
University Press.
Iizumi H (1986) Soil nutrient dynamics. In: Cragg S and Polunin N (eds.)
Workshop on Mangrove Ecosystem Dynamics, p.171. New Delhi:
UNDP/UNESCO Regional Project (RAS/79/002).
Jones DA (1984) Crabs of the mangal ecosystem. In: Por FD and Dor I
(eds.) Hydrobiology of the Mangal, pp. 89 109. The Hague: Dr. W.
Junk Publishers.
Koch MS and Snedaker SC (1997) Factors influencing Rhizophora
mangle L. seedlings development into the sapling stage across
resource and stress gradients in subtropical Florida. Biotropica
29: 427 439.
Krauss KW, Allen JA, and Cahoon DR (2003) Differential rates of vertical
accretion and elevation change among aerial root types in Micronesian
mangrove forests. Estuarine Coastal and Shelf Science 56: 251 259.
Krauss KW, Doyle TW, Twilley RR, Smith TJ, Whelan KRT, and
Sullivan JK (2005) Woody debris in the mangrove forests of south
Florida. Biotropica 37: 9 15.
Kristensen E, Andersen FØ, and Kofoed LH (1988) Preliminary
assessment of benthic community metabolism in a Southeast Asian
mangrove swamp. Marine Ecology Progress Series 48: 137 145.
Lee SY (1989) Litter production and turnover of the mangrove Kandelia
candel (L.) Druce in a Hong Kong tidal shrimp pond. Estuarine,
Coastal and Shelf Science 29: 75 87.
Leh CMU and Sasekumar A (1985) The food of sesarmid crabs in
Malaysian mangrove forests. Malay Naturalist Journal
39: 135 145.
Lewis RR (1982) Mangrove forests. In: Lewis RR (ed.) Creation and
Restoration of Coastal Plant Communities, pp. 153 171. Boca
Raton, FL: CRC Press.
Lewis RR (1990) Creation and restoration of coastal plain wetlands in
Florida. In: Kusler JA and Kentula ME (eds.) Wetland Creation and
Restoration, pp. 73 101. Washington, DC: Island Press.
Lewis RR (1990) Creation and restoration of coastal wetlands in Puerto
Rico and the US Virgin Islands. In: Kusler JA and Kentula ME (eds.)
Wetland Creation and Restoration, pp. 103 123. Washington, DC:
Island Press.
Lin G and Sternberg LSL (1992) Differences in morphology, carbon
isotope ratios, and photosynthesis between scrub and fringe
mangroves in Florida, USA. Aquatic Botany 42: 303 313.
Lin G and Sternberg LSL (1992) Effect of growth form, salinity, nutrient
and sulfide on photosynthesis, carbon isotope discrimination and
growth of red mangrove (Rhizophora mangle L.). Australian Journal
of Plant Physiology 19: 509 517.
Lovelock CE, Feller IC, Mckee KL, Engelbrecht BMJ, and Ball MC (2004)
The effect of nutrient enrichment on growth, photosynthesis and
hydraulic conductance of dwarf mangroves in Panama. Functional
Ecology 18: 25 33.
Lugo AE (1980) Mangrove ecosystems: Successional or steady state?
Biotropica 12: 65 72.
Lugo AE (1998) Mangrove forests: A tough system to invade but an easy
one to rehabilitate. Marine Pollution Bulletin 37: 427 430.
Lugo AE and Snedaker SC (1974) The ecology of mangroves. Annual
Review of Ecology and Systematics 5: 39 64.
Lynch JC, Meriwether JR, McKee BA, Vera Herrera F, and Twilley RR
(1989) Recent accretion in mangrove ecosystems based on 137Cs
and 210Pb. Estuaries 12: 284 299.
Macnae W (1968) A general account of the fauna and flora of mangrove
swamps and forests in the Indo West Pacific region. Advances in
Marine Biology 6: 73 270.
Malley DF (1978) Degradation of mangrove leaf litter by the tropical
sesarmid crab Chiromanthes onychophorum. Marine Biology
49: 377 386.
McKee KL (1993) Soil physicochemical patterns and mangrove
species distribution Reciprocal effects? Journal of Ecology
81: 477 487.
McKee KL, Feller IC, Popp M, and Wanek W (2002) Mangrove isotopic
(15N and 13C) fractionation across a nitrogen vs. phosphorus
limitation gradient. Ecology 83: 1065 1075.
Medina E and Francisco M (1997) Osmolality and 13C of leaf tissues of
mangrove species from environments of contrasting rainfall and
salinity. Estuarine, Coastal and Shelf Science 45: 337 344.
Mangrove Wetlands
Naidoo G (1985) Effects of waterlogging and salinity on plant water
relations and on the accumulation of solutes in three mangrove
species. Aquatic Botany 22: 133 143.
Nedwell DB (1975) Inorganic nitrogen metabolism in a eutrophicated
tropical mangrove estuary. Water Research 9: 221 231.
Nixon SW (1980) Between coastal marshes and coastal waters A
review of twenty years of speculation and research on the role of
salt marshes in estuarine productivity and water chemistry.
In: Hamilton P and MacDonald KB (eds.) Estuarine and Wetland
Processes with Emphasis on Modeling, pp. 437 525. New York:
Plenum Press.
Odum WE and Heald EJ (1972) Trophic analysis of an estuarine
mangrove community. Bulletin Marine Science 22: 671 738.
Odum WE and McIvor CC (1990) Mangroves. In: Myers RL and Ewel JJ
(eds.) Ecosystems of Florida, pp. 517 548. Orlando, FL: University of
Central Florida Press.
Odum WE, McIvor CC, and Smith TJ (1982) The Ecology of the
Mangroves of South Florida: A Community Profile. FWS/OBS 81/24.
Washington, DC: US Fish and Wildlife Service, Office of Biological
Resources.
Parkinson RW, DeLaune RD, and White JR (1994) Holocene sea level
rise and the fate of mangrove forests within the wider Caribbean
region. Journal of Coastal Research 10: 1077 1086.
Pinzon ZS, Ewel KC, and Putz FE (2003) Gap formation and forest
regeneration in a Micronesian mangrove forest. Journal of Tropical
Ecology 19: 143 153.
Ponnamperuma FN (1984) Mangrove swamps in south and Southeast
Asia as potential rice lands. In: Soepadmo E, Rao AN, and
McIntosh DJ (eds.) Proceedings Asian Mangrove Symposium,
pp. 672 683. Kuala Lumpur: University of Malaya.
Pool DJ, Lugo AE, and Snedaker SC (1975) Litter production in
mangrove forests of southern Florida and Puerto Rico. In: Walsh G,
Snedaker S, and Teas H (eds.) Proceedings of the International
Symposium on the Biology and Management of Mangroves,
pp. 213 237. Gainesville, FL: Institute of Food and Agricultural
Sciences. University of Florida.
Rabinowitz D (1978) Early growth of mangrove seedlings in Panama,
and an hypothesis concerning the relationship of dispersal and
zonation. Journal of Biogeography 5: 113 133.
Rivera Monroy VH, Day JW, Twilley RR, Vera Herrera F, and Coronado
Molina C (1995) Flux of nitrogen and sediment in a fringe mangrove
forest in Terminos Lagoon, Mexico. Estuarine, Coastal and Shelf
Science 40: 139 160.
Rivera Monroy VH and Twilley RR (1996) The relative role of
denitrification and immobilization in the fate of inorganic nitrogen in
mangrove sediments. Limnology and Oceanography 41: 284 296.
Rivera Monroy VH, Twilley RR, Boustany RG, Day JW, Vera Herrera F,
and Ramirez MdC (1995) Direct denitrification in mangrove
sediments in Terminos Lagoon, Mexico. Marine Ecology Progress
Series 97: 97 109.
Rivera Monroy VH, Twilley RR, Bone D, et al. (2004) A conceptual
framework to develop long term ecological research and
management objectives in the wider Caribbean region. Bioscience
54: 843 856.
Robertson AI (1986) Leaf burying crabs: Their influence on energy flow
and export from mixed mangrove forests (Rhizophora spp.) in
northeastern Australia. Journal of Experimental Marine Biology and
Ecology 102: 237 248.
Robertson AI and Alongi DM (1992) Tropical Mangrove Ecosystems,
vol. 41. Washington, DC: American Geophysical Union.
Robertson AI, Alongi DM, and Boto KG (1992) Food chains and carbon
fluxes. In: Robertson AI and Alongi DM (eds.) Tropical Mangrove
Ecosystems, pp. 293 326. Washington, DC: American Geophysical
Union.
Robertson AI and Blaber SJM (1992) Plankton, epibenthos and fish
communities. In: Robertson AI and Alongi DM (eds.) Tropical
Mangrove Ecosystems, pp. 173 224. Washington, DC: American
Geophysical Union.
Robertson AI and Daniel PA (1989) The influence of crabs on litter
processing in high intertidal mangrove forests in tropical Australia.
Oecologia 78: 191 198.
Robertson AI and Duke NC (1990) Mangrove fish communities in
tropical Queensland, Australia: Spatial and temporal patterns in
317
densities, biomass and community structure. Marine Biology
104: 369 379.
Rodelli MR, Gearing JN, Gearing PJ, Marshall N, and Sasekumar A
(1984) Stable isotope ratio as a tracer of mangrove carbon in
Malaysian ecosystems. Oecologia 61: 326 333.
Rojas Galaviz JL, Yánez Arancibia A, Day JW, Jr., and Vera Herrera FR
(1992) Estuarine primary producers: Laguna de Terminos a study
case. In: Seeliger U (ed.) Coastal Plant Communities of Latin
America, pp. 141 154. San Diego, CA: Academic Press.
Romero LM, Smith TJ, and Fourqurean JW (2005) Changes in mass and
nutrient content of wood during decomposition in a south Florida
mangrove forest. Journal of Ecology 93: 618 631.
Ross MS, Meeder JF, Sah JP, Ruiz LP, and Telesnicki GJ (2000) The
Southeast saline Everglades revisited: 50 Years of coastal vegetation
change. Journal of Vegetation Science 11: 101 112.
Roth LC (1992) Hurricanes and mangrove regeneration: Effects of
Hurricane Juan, October 1988, on the vegetation of Isla del Venado,
Bluefields, Nicaragua. Biotropica 24: 375 384.
Rutzler K and Feller C (1988) Mangrove swamp communities. Oceanus
30: 16 24.
Rutzler K and Feller C (1996) Carribbean mangrove swamps. Scientific
American 274: 94 99.
Saenger P, Hegerl EJ, and Davie JDS (1983) Global Status of Mangrove
Ecosystems. Commission on Ecology Paper No. 3, pp. 83.
International Union for the Conservation of Nature (IUCN).
Saenger P and Snedaker SC (1993) Pantropical trends in mangrove
above ground biomass and annual litterfall. Oecologia
96: 293 299.
Sauer JD (1962) Effects of recent tropical cyclones on the coastal
vegetation of Mauritius. Journal of Ecology 50: 275 290.
Scholander PF, Hammel HT, Hemmingsen E, and Garay W (1962) Salt
balance in mangroves. Plant Physiology 37: 722 729.
Sherman RE, Fahey TJ, and Battles JJ (2000) Small scale disturbance
and regeneration dynamics in a neotropical mangrove forest. Journal
of Ecology 88: 165 178.
Simberloff DS and Wilson EO (1969) Experimental zoogeography of
islands: The colonization of empty islands. Ecology 50: 278 289.
Smith TJ, III (1987) Seed predation in relation to tree dominance and
distribution in mangrove forests. Ecology 68: 266 273.
Smith TJ, III (1992) Forest structure. In: Robertson AI and Alongi DM
(eds.) Tropical Mangrove Ecosystems, pp. 101 136. Washington,
DC: American Geophysical Union.
Smith TJ, Boto KG, Frusher SD, and Giddins RL (1991) Keystone
species and mangrove forest dynamics: The influence of burrowing
by crabs on soil nutrient status and forest productivity. Estuarine
Coastal and Shelf Science 33: 419 432.
Smith TJ, III, Robblee MB, Wanless HR, and Doyle TW (1994)
Mangroves, hurricanes, and lightning strikes. BioScience
44: 256 262.
Snedaker S (1982) Mangrove species zonation: Why? In: Sen DN and
Rajpurohit KS (eds.) Tasks for Vegetation Science, vol. 2,
pp. 111 125. The Hague: Junk.
Snedaker SC (1986) Traditional uses of South American mangrove
resources and the socio economic effect of ecosystem changes.
In: Kunstadter P, Bird ECF, and Sabhasri S (eds.) Proceedings,
Workshop on Man in the Mangroves, pp. 104 112. Tokyo: United
Nations University.
Snedaker SC (1989) Overview of ecology of mangroves and information
needs for Florida Bay. Bulletin of Marine Science 44: 341 347.
Snedaker SC, Meeder JF, Ross MS, and Ford RG (1994) Discussion of
Ellison, JC and Stoddart, DR 1991. Mangrove ecosystem collapse
during predicted sea level rise: Holocene analogues and
implications. Journal of Coastal Research 7: 151 165, Journal of
Coastal Research 10: 497 498.
Snedaker SC and Snedaker JG (1984) The Mangrove Ecosystem:
Research Methods. London: UNESCO.
Sousa WP, Quek SP, and Mitchell BJ (2003) Regeneration of
Rhizophora mangle in a Caribbean mangrove forest: Interacting
effects of canopy disturbance and a stem boring beetle. Oecologia
137: 436 445.
Sutherland JP (1980) Dynamics of the epibenthic community on roots of
the mangrove Rhizophora mangle, at Bahia de Buche, Venezuela.
Marine Biology 58: 75 84.
318
Mangrove Wetlands
Teas HJ (1981) Restoration of mangrove ecosystems. In: Carey RC,
Markovits PS, and Kirkwood JB (eds.) Proceedings of Workshop on
Coastal Ecosystems of the Southeastern United States, pp. 95 103.
Reno, NV: US Fish and Wildlife Service, Office of Biological Services,
FWS/OBS 80/59.
Thayer GW, Colby DR, and Hettler WF, Jr. (1987) Utilization of the red
mangrove prop root habitat by fishes in south Florida. Marine
Ecology Progress Series 35: 25 38.
Thom B (1967) Mangrove ecology and deltaic morphology: Tabasco,
Mexico. Journal of Ecology 55: 301 343.
Thom BG (1982) Mangrove ecology A geomorphological perspective.
In: Clough BF (ed.) Mangrove Ecosystems in Australia, pp. 3 17.
Canberra: Australian National University Press.
Thom BG (1984) Coastal landforms and geomorphic processes.
In: Sneadeker SC and Sneadeker JG (eds.) The Mangrove
Ecosystem: Research Methods, pp. 3 17. Paris: UNESCO.
Tilman D (1982) Resource Competition. Princeton, NJ: Princeton
University Press.
Tomlinson PB (1995) The Botany of Mangroves. New York: Cambridge
University Press.
Twilley RR (1988) Coupling of mangroves to the productivity of estuarine
and coastal waters. In: Jansson BO (ed.) Coastal Offshore
Ecosystems: Interactions, pp. 155 180. Berlin: Springer.
Twilley RR (1995) Properties of mangroves ecosystems and their
relation to the energy signature of coastal environments.
In: Hall CAS (ed.) Maximum Power, pp. 43 62. Denver, CO:
Colorado Press.
Twilley RR (1997) Mangrove wetlands. In: Messina M and Connor W
(eds.) Southern Forested Wetlands: Ecology and Management,
pp. 445 473. Boca Raton, FL: CRC Press.
Twilley RR, Cárdenas W, Rivera Monroy VH, et al. (2000) Ecology of the
Gulf of Guayaquil and the Guayas River Estuary. In: Seeliger U and
Kjerve BJ (eds.) Coastal Marine Ecosystems of Latin America,
pp. 245 263. New York: Springer.
Twilley RR and Chen RH (1998) A water budget and hydrology model of
a basin mangrove forest in Rookery Bay, Florida. Marine and
Freshwater Research 49: 309 323.
Twilley RR, Chen RH, and Hargis T (1992) Carbon sinks in mangroves
and their implications to carbon budget of tropical coastal
ecosystems. Water, Air and Soil Pollution 64: 265 288.
Twilley RR, Gottfried RR, Rivera Monroy VH, Armijos MM, and
Bodero A (1998) An approach and preliminary model of integrating
ecological and economic constraints of environmental quality in the
Guayas River estuary, Ecuador. Environmental Science and Policy
1: 271 288.
Twilley RR, Lugo AE, and Patterson Zucca C (1986) Production,
standing crop, and decomposition of litter in basin mangrove forests
in southwest Florida. Ecology 67: 670 683.
Twilley RR, Pozo M, Garcia VH, Rivera Monroy VH, Zambrano R, and
Bodero A (1997) Litter dynamics in riverine mangrove forests in the
Guayas River estuary, Ecuador. Oecologia 111: 109 122.
Twilley RR and Rivera Monroy VH (2005) Developing performance
measures of mangrove wetlands using simulation models of
hydrology, nutrient biogeochemistry, and community dynamics.
Journal of Coastal Research 40: 79 93.
Twilley RR, Rivera Monroy VH, Chen R, and Botero L (1998) Adapting
and ecological mangrove model to simulate trajectories in restoration
ecology. Marine Pollution Bulletin 37: 404 419.
Twilley RR, Snedaker SC, Yanez Arancibia A, and Medina E (1996)
Biodiversity and ecosystem processes in tropical estuaries:
Perspectives from mangrove ecosystems. In: Mooney H,
Cushman H, and Medina E (eds.) Biodiversity and Ecosystem
Functions: A Global Perspective, pp. 327 370. New York:
Wiley.
Vermeer DE (1963) Effects of Hurricane Hattie, 1961, on the cays of
British Honduras. Zeitschrift fur Geomorphologie 7: 332 354.
Wadsworth FH (1959) Growth and regeneration of white mangrove in
Puerto Rico. Caribbean Forester 20: 59 69.
Waisel Y (1972) Biology of Halophytes, 395pp. New York: Academic
Press.
Walsh GE (1974) Mangroves: A review. In: Reimold R and Queen W
(eds.) Ecology of Halophytes, pp. 51 174. New York: Academic
Press.
Wanless HR, Parkinson RW, and Tedesco LP (1994) Sea level control
on stability of Everglades wetlands. In: Davis S and Ogden J (eds.)
Everglades: The Ecosystem and Its Restoration, pp. 199 223. Delray
Beach, FL: St. Lucie Press.
Watson J (1928) Mangrove Forests of the Malay Peninsula. Singapore:
Fraser & Neave.
Woodroffe CD (1990) The impact of sea level rise on mangrove
shoreline. Progress in Physical Geography 14: 483 520.
Woodroffe C (1992) Mangrove sediments and geomorphology.
In: Robertson AI and Alongi DM (eds.) Tropical Mangrove Ecosystems,
pp. 7 42. Washington, DC: American Geophysical Union.
Woodroffe CD, Chappell J, Thom BG, and Wallensky E (1986)
Geomorphological dynamics and evolution of the South Alligator tidal
river and plains. In: ANU, North Australia Research Unit Monograph
3. Darwin: North Australian Research Unit.
Yanez Arancibia A (1985) Fish Community Ecology in Estuaries and
Coastal Lagoons: Towards an Ecosystem Integration. Mexico City,
UNAM Press.
Yanez Arancibia A and Day JW, Jr. (1982) Ecological characterization of
Terminos Lagoon, a tropical lagoon estuarine system in the
Southern Gulf of Mexico. Oceanologica Acta SP: 431 440.
Yanez Arancibia A and Day JW, Jr. (1988) Ecology of Coastal
Ecosystems in the Southern Gulf of Mexico: The Terminos Lagoon
Region. Mexico City: Universidad Nacional Autonoma de Mexico,
Ciudad Universitaria, Mexico.
Yánez Arancibia A, Lara Domı́nguez AL, and Day JW (1993)
Interactions between mangrove and seagrass habitats mediated by
estuarine nekton assemblages: Coupling of primary and secondary
production. Hydrobiologia 264: 1 12.
Yánez Arancibia A, Lara Domı́nguez AL, Rojas Galaviz JL, et al. (1988)
Seasonal biomass and diversity of estuarine fishes coupled with
tropical habitat heterogeneity (southern Gulf of Mexico). Journal of
Fish Biology 33(supplement A): 191 200.
Mediterranean
319
Mediterranean
F Médail, IMEP Aix-Marseille University, Aix-en-Provence, France
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Main Environmental Characteristics of
the Mediterranean Ecoregions
Patterns and Determinants of Mediterranean
Biodiversity
Historical Biogeography and Evolution of
Mediterranean Biodiversity
Convergence versus Nonconvergence of
Mediterranean Ecosystems
Ecosystem Characteristics and Processes
Disturbances and Ecosystem Dynamics
Conclusion: Current Evolution of Mediterranean
Ecosystems under Global Changes
Further Reading
Introduction
the existence of a combined dry and hot summer period of
variable length, which imprints a strong water stress on
species and ecosystems during summer. A high unpredict
ability characterizes these Mediterranean climates, with
high yearly variation in timing as well as amount of rainfall
or occurrence of extreme temperatures.
Rainfall is extremely variable, with mean annual values
ranging from 100 to 2000 mm. The lowest values are found
at desert margins, especially in North Africa and the
Near East. The isohyet of 100 mm yr 1 represents the
borderline between the Mediterranean and the Saharan
climates. Rainfalls higher than 1500 mm are mostly found
at medium altitudes of some coastal mountain ranges. But
Mediterranean type climates differ markedly between and
within the Mediterranean regions, in terms of total rainfall
and seasonality. For example, annual rainfall of parts of
southern California and central Chile are comprised
between 250 and 350 mm, whereas some Mediterranean
montane sites of SW Africa receive as much as 3000 mm
by year, and the summer rainfall is similar here to annual
totals for California or Chile.
Mean minimum temperatures of the coldest month (m)
are often used to define climatic subdivisions in the
Mediterranean Basin (Table 2). These values are corre
lated to elevation and to a lesser extent to increased
latitude and continentality. In most places, m is between 0
and þ3 C although extremes can reach þ8 to þ9 C in
desert margins and 8 to 10 C on the highest mountains.
Aridity and temperature play an essential role in the
structure and composition of Mediterranean ecosystems.
The Emberger pluviothermic quotient (Q2) constitutes the
most utilized index for classifying Mediterranean climates:
Mediterranean type ecosystems occur in areas character
ized by winter rainfall and summer drought. Five
ecoregions of the world possess a Mediterranean climate
and form the Mediterranean biome: the Mediterranean
Basin, California (see Chaparral), central Chile, the
southern and southwestern Cape Province of South
Africa (SW Cape), the southwestern and parts of southern
Australia (SW Australia) (Table 1). These Mediterranean
ecoregions are all centered between 30 and 40 north or
south of the equator, and are exposed to similar atmo
spheric and oceanic circulation patterns with cool ocean
currents. Mediterranean ecoregions occur only along the
western sides of continents, and occupy limited areas
between deserts and temperate regions.
The most typical characteristics of Mediterranean
ecosystems, compared to temperate or boreal biomes,
are their spatial and temporal complexity inducing strong
heterogeneities, in terms of physical factors (geography,
geology, geomorphology, pedology, bioclimate) and of
their biological components and species life history
traits. Paleogeographical and historical episodes, current
geographical and climatic contrasts have molded both an
unusually high biodiversity and ecological complexity, and
favored the emergence of a functional uniqueness for several
ecosystems. High species richness and endemism due to
contrasted biogeographical origins, and original functional
dynamics at local and landscape levels linked to stress effects,
represent indeed key components of these ecosystems.
Main Environmental Characteristics of
the Mediterranean Ecoregions
Climate
The Mediterranean ecoregions are usually defined by their
particular climates, which are transitional between tempe
rate and dry tropical climates. The main characteristic is
Q2 ¼
2000P
M 2 m2
where P is the annual rainfall (in mm), M is the mean
maximum temperatures of the warmest month of the
year, m is the mean minimum temperatures of the coldest
month of the year.
Table 1 Main environmental characteristics and major ecosystem-types of the Mediterranean ecoregions
North Hemisphere
South Hemisphere
Mediterranean Basin
California
Central Chile
SW Australia
Cape Region
Surface (km2)
Topographic
heterogeneity
Climatic
heterogeneity
Rainfall reliability
Main lithological
substrates
2 300 000
High
324 000
High
140 000
Very high
310 000
Low
90 000
Moderate
Very high
Very high
High
Moderate
High
Moderate
Calcareous rocks, occasional
siliceous rocks
Argillaceous and mafic igneous
rocks
Very high
Siliceous rocks (sandstones,
quartzites)
High
Siliceous, argillaceous and mafic
igneous rocks
Soil fertility
Natural fire
frequency
(year)
Forests and
woodlands
High moderate
25 50
Low
Argillaceous and mafic igneous
rocks, occasional ultramafic
rocks
Moderate
40 60
High
Fire free
Very low low
10 15
Very low moderate
10 20
Diverse forests with thermophilous
conifers (Pinus attenuata, P.
sabiniana, Cupressus
macrocarpa) and oaks (Quercus
douglasii, Q. agrifolia, Q. lobata),
and mesophilous conifers (Abies,
Pinus . . .) at higher altitudes;
coast redwood (Sequoia
sempervirens)
Chaparrals with Adenostoma
(chamisal), Arctostaphylos
(manzanita chaparral),
Ceanothus, scrub oaks (Quercus
dumosa); coastal scrubs with
Artemisia, Baccharis, Salvia
Very diverse, with semiarid Acacia
caven and Prosopis chilensis
forests in the north; subtropical
broad leaved and sclerophyllous
forests with Peumus boldus and
Cryptocarya alba in the central
region; deciduous Nothofagus
forests farther south, with
Araucaria araucana
Open shrubland with Acacia caven
(espinal); matorrals with Lithraea
caustica, Quillaja saponaria;
coastal matorrals with cacti
(Trichocereus) and bromeliads
(Puya)
Patchy and open woodlands
dominated by Eucalyptus (E.
diversicolor, E. marginata); low
woodlands with Banksia; thickets
with Acacia, Melaleuca, and
Allocasuarina
Very patchy and scarce forests;
composed of cool and humid
Afromontane plants, with warm
subtropical elements;
sclerophyllous trees and conifers
(Afrocarpus, Podocarpus)
Native perennial bunchgrasses with
Stipa, Poa and Koeleria, replaced
by annual grasslands with forbs
(Avena, Bromus, Lolium,
Erodium)
Anthropogenic prairies with
numerous European herbs and
grasses; wet grasslands with
native Juncus procerus
Kwongan and scrub heaths with
Proteaceae (Banksia, Grevillea,
Hakea) and ericoids plants
(Epacridaceae); mallee
dominated by shrubby
Eucalyptus (E. incrassata, E.
oleosa, E. socialis)
Very scarce and patchy; grasslands
on granite outcrops with annual
everlastings (Helichrysum;
Helipterum) or perennial
Lechenaultia
Fynbos with major plant types:
restioids (Restionaceae),
ericoids, proteoids (Proteaceae)
and geophytes; renosterveld
dominated by ericoids
(renosterbos: Elytropappus);
succulent karoo with Aizoaceae
Very scarce, fire prone grasslands
and grassy shrublands
dominated by geophytes
Shrublands
Grasslands
Very diverse and heterogeneous;
with many sclerophyllous
(Quercus ilex, Q. suber) and
broad leaved oaks (Quercus
pubescens, Q. faginea, Q.
ithaburensis), and conifers (Pinus
halepensis, P. brutia, Cedrus
atlantica, C. libani, Abies,
Juniperus)
Maquis with Erica, Arbutus on
siliceous soils; garrigues with
Quercus coccifera, Cistus, Ulex,
on calcareous soils; phryganas
with spiny shrubs
(Sarcopoterium, Astragalus,
Genista)
Very diverse grasslands with
numerous annuals and
perennials herbs (Poaceae,
Fabaceae, Asteraceae); steppes
with Stipa tenacissima and
Lygeum spartum in North Africa
Data from Davis GW and Richardson DM (1995) Ecological Studies, Vol. 109: Biodiversity and Ecosystem Function in Mediterranean Type Ecosystems. Berlin and Heidelberg: Springer; Cowling RM, Rundel PW, Lamont BB,
Arroyo MK, and Arianoutsou M (1996) Plant diversity in Mediterranean climate region. Trends in Ecology and Evolution 11: 362 366; Cowling RM, Ojeda F, Lamont BB, Rundel PW, and Lechmere Oertel R (2005) Rainfall
reliability: A neglected factor in explaining convergence and divergence of plant traits in fire prone Mediterranean climate ecosystems. Global Ecology and Biogeography 14: 509 519; Dalmann PR (1998) Plant Life in the
World’s Mediterranenan Climates. Oxford: Oxford University Press. Médail, ined.
Mediterranean
321
Table 2 Vegetation levels showing the correspondence between thermal variants and dominant woody types of the Mediterranean
Basin
Vegetation level
Thermal
variant
Infra-Mediterranean
Thermo-Mediterranean
m ( C)
T ( C)
Dominant woody species
Very hot
Hot
> þ7 C
þ3 to þ7 C
> þ17 C
> þ17 C
Meso-Mediterranean
Supra-Mediterranean
Temperate
Cool
0 to þ3 C
3 to 0 C
þ13 to þ17 C
þ8 to þ13 C
Mountain-Mediterranean
Oro-Mediterranean
Cold
Very cold
7 to 3 C
< 7 C
þ4 to þ 8 C
< þ4 C
Argania, Acacia gummifera
Olea, Ceratonia, Pinus halepensis and P. brutia,
Tetraclinis, (Quercus)
Sclerophyllous Quercus, Pinus halepensis and P. brutia
Deciduous Quercus, Ostrya, Carpinus orientalis (Pinus
brutia)
Pinus nigra, Cedrus, Abies, Fagus Juniperus,
prostrate spiny xerophytes
m, mean minimum temperatures of the coldest month; T, mean annual temperature.
Modified from Quézel P and Médail F (2003) Ecologie et biogéographie des forêts du bassin méditerranéen. Paris: Elsevier.
Table 3 Main types of bioclimates and their theoretical correspondence with the dominant vegetation types of the Mediterranean
Basin
Bioclimate
Mean annual rainfall
(for m 0 C)
Number of months
without rainfall
Per-Arid
Arid
< 100 mm
100–400 mm
11–12
7–10
Semi-Arid
Sub-Humid
400–600 mm
600–800 mm
5–7
3–5
Humid
800–1000 mm
1–3
Per-Humid
>1000 mm
<1
Main vegetation type
Saharan
Steppe and pre-steppe (Juniperus turbinata, Pinus halepensis,
Pistacia atlantica)
Pre-forest (Pinus halepensis, P. brutia, Juniperus spp., Quercus)
Forest (Mostly sclerophyllous Quercus, Pinus halepensis, P. brutia,
P. pinaster, P. pinea, P. nigra, Cedrus)
Forest (Mostly deciduous Quercus, Pinus brutia, P. pinaster, P.
nigra, Cedrus, Abies, Fagus)
Forest (Deciduous Quercus, Cedrus, Abies, Fagus)
m: mean minimum temperatures of the coldest month.
Modified from Quézel P and Médail F (2003) Ecologie et biogéographie des forêts du bassin méditerranéen. Paris: Elsevier.
According to the levels of humidity and the winter
severity, several bioclimatic zones and thermal variants
are respectively defined, and they can be included in the
climagram of Emberger; their combination permits to
define six main bioclimatic types (Table 3).
Soils and Nutrients Availability
The five Mediterranean ecoregions are characterized by
different geologies and soils characteristics due to their
contrasted physiographic histories (Table 1). Landscapes
of SW Australia and SW Africa consist of inland mass of
geologically older origins than the three other ecoregions
where mountain building events occurred as recently as
the Tertiary and the Quaternary.
In the two South Hemisphere ecoregions, soils on older
substrata of uplands are generally highly leached lithosols in
South Africa and by laterites and the process of podzoliza
tion in southern Australia; they have been exposed to
weathering since the Paleozoic or even the Precambrian.
Coastal deposits are younger and determine calcareous
sands, decalcified humus podzols, and bleached sands.
Calcareous soils of limestone origin are scarce and occur
only in few places of the South African coasts and in the
central–southern region of Mediterranean Australia.
In California and Chile, the violent tectonic activity
down the west coasts during the Late Tertiary and
Early Quaternary has given rise to rugged landscapes of
the Cordilleran mountain chain. Upland areas possess
generally coarse textured lithosols, whereas the major
inland valleys (Great Valley of California and Central
Valley of Chile) have more fertile soils linked to alluvial
deposits.
In the Mediterranean Basin, the diverse tectonic and
orogenic activities, and also the consequences of Pleisto
cene glaciations, induced a complex patchwork of
landscapes and a mosaic of soil types. The predominantly
limestone rocks have given rise to the terra rossa soil, a
clay rich and relatively fertile soil of the lowland areas.
Soils of the humid uplands are often leached podzols or
brown forest soils occurring within forested landscapes.
In spite of these sometimes large differences in sub
stratum geology, there are several similarities between
soils of the Mediterranean ecoregions, due to similar
322
Mediterranean
pedogenic processes linked to water driven erosion and
leaching. These seasonally droughted and moderately to
strongly leached soils are indeed characterized by a low
availability of several nutrients, especially phosphorous,
and nitrogen which is greatly affected by fire.
Patterns and Determinants of
Mediterranean Biodiversity
To compare the biodiversity of Mediterranean type eco
systems, it is useful to partition species richness and
diversity into three main types of spatial scales, and to
consider successively regional diversity, differentiation
diversity, and local diversity (Table 4). Regional diver
sity is the product of local richness and turnover along
habitat and geographic gradients.
The five Mediterranean climate regions harbor a
remarkable and huge regional biodiversity, among the high
est in the world. With only 2% of the world’s terrestrial
surface, the Mediterranean biome contains nearly 20% of
the Earth’s total plant diversity, making very significant
biodiversity hot spots, second only after tropical ones. The
Mediterranean Basin exhibits the greatest diversity of plant
species, both general and endemic, but with the much
greater surface area (84% of the total of Mediterranean
ecoregions). This ecoregion possess a higher tree richness
(290 indigenous trees with 201 endemics) than the
California Floristic Province (173 trees with 77 endemics),
although its surface is seven times larger. In fact, the latter
ecoregion has more or less the same surface area and plant
biodiversity as Morocco, but California is four times richer
in strictly local endemics. The case of the Cape Floristic
Region is even more remarkable, since the endemic plant
richness reaches close to 70% and the total plant species
reaches 9090 taxa, which makes it one of the world’s richest
areas. The interplay between diverse processes of historical
biogeography and heterogeneous environmental conditions
has promoted these considerable species richness and ende
mism levels in the different Mediterranean ecoregions.
Regional diversity peaks generally in areas with high topo
graphical and climatic heterogeneity. However, the two
highest plant species rich regions of the SW Cape and
SW Australia are characterized by global topographically
and climatically uniform lowlands. Edaphic complexity, and
more recently rainfall reliability (measured as interannual
variation in seasonal and monthly rainfall and as the fre
quency of different sized rainfall events), have been invoked
as main determinants of this exceptionally high biodiversity.
The regional scale plant richness is indeed twofold higher
in the western Cape Region, which receives reliable winter
rainfall than the less reliable and nonseasonal zone in the
eastern Cape. Reliable rainfall regimes are argued to pro
mote higher and rapid speciation and lower extinction rates,
and this pattern could partly explain the overall highest
plant diversity of the SW Cape and SW Australia which
have significantly more reliable regime than the other
three Mediterranean ecoregions. If we consider diverse
groups of vertebrates (Table 4), biodiversity patterns are
more contrasted, and species richness and endemism are
often attenuated compared to plants, notably for birds.
Nevertheless, for reptiles, amphibians, and freshwater fishes,
the uniqueness of Mediterranean biotas is again noteworthy
since the endemism rate is generally comprised between
30% and 60% (Table 4).
The differentiation diversity refers to changes of species
composition along habitat gradients (beta diversity) or geo
graphical gradients (gamma diversity). Highest levels of
differentiation diversity are recorded for plant species in
Table 4 Main biodiversity components of the five Mediterranean ecoregions
North Hemisphere
Biodiversity components
Local diversity
Differentiation diversity
Regional diversity
Plant richness/endemism
Mammal richness/
endemism
Bird richness/endemism
Reptile richness/endemism
Amphibian richness/
endemism
Freshwater fish richness/
endemism
South Hemisphere
Mediterranean
Basin
California
Central Chile
SW Australia
Cape Region
Low–very high
Moderate
Moderate
c. 25 000/12 500
(50%)
224/25 (11%)
Low–moderate
Moderate
Moderate
3488/2128
(61%)
151/18 (12%)
Low–?high
Low–moderate?
Low
3539/1769
(50%)
65/14 (22%)
Low–high
High
High
5710/3000 (52.5%)
Moderate–high
High
High
9086/6226 (68.5%)
57/12 (21%)
90/4 (4%)
497/32 (6%)
228/77 (34%)
86/27 (31%)
341/8 (2%)
69/4 (6%)
54/25 (46%)
226/12 (5%)
41/27 (66%)
43/29 (67%)
285/10 (4%)
177/27 (15%)
33/19 (58%)
324/6 (2%)
100/22 (22%)
51/16 (31%)
216/63 (29%)
73/15 (21%)
43/24 (56%)
20/10 (50%)
34/14 (41%)
Data from Cowling RM, Rundel PW, Lamont BB, Arroyo MK, and Arianoutsou M (1996) Plant diversity in Mediterranean climate region. Trends in
Ecology and Evolution 11: 362 366; and Mittermeier RA, Robles Gil, Hoffmann M, et al. (2004) Hotspots Revisited: Earth’s Biologically Richest and
Most Endangered Terrestrial Ecoregions. Monterrey: CEMEX, Washington: Conservation International, Mexico: Agrupación Sierra Madre.
Mediterranean
the winter rainfall zones of SW Cape and SW Australia,
with a high turnover for fire killed shrub lineages.
Disproportionate radiation of several shrub genera (e.g.,
667 species of Erica in the Cape Floristic Region with
96.5% of endemic heaths) explains the strong dissimilarities
in species composition between morphologically close com
munities. A similar, but attenuated, pattern is found in
California and in the Mediterranean Basin for relatively
recent shrub lineages and annual herbs, which are respec
tively the keystone species of matorrals and grasslands.
At the local scale, that is, less than 0.1 ha (alpha diversity),
Mediterranean biodiversity is two times lower than that of
tropical regions. However, a great variation exists within
each ecoregion and between different habitats. Open and
frequently burned shrublands and heaths on nutrient poor
soils, in particular, fynbos in SW Cape and kwongan in SW
Australia, xerophytic rocky grasslands, and temporary pools
encompass the highest plant diversity. Postfires communities
in dense shrublands, notably chaparral in California (see
Chaparral), and maquis in the Mediterranean Basin, are
also characterized by a rich fire ephemeral flora with
numerous annual plants. Mean local plant richness of
Mediterranean forests is comprised between 10 species
per m2 and 25–110 species per 1000 m2. At this spatial
scale, woody plant communities of the Mediterranean
Basin are both very heterogeneous, but also among the
richest types, ahead of the alpha diversity found in the SW
Cape. Several nonexclusive determinants have been invoked
to explain this high local diversity and species coexistence:
an important regional species pool linked to complex
historical biogeography, differentiation, and character
displacement along structural niche axes, spatiotemporal
variations in resource availability, recurrent disturbances
(fire, grazing), neighborhood effects, and lottery processes.
Finally, few generalizations are available from the numerous
studies on local plant diversity in the Mediterranean vegeta
tion, and the strongest evidence is that diversity represents an
unimodal function of productivity or nutrient supply of soils.
Species–area relationships fit a power function model for the
majority of Mediterranean plant communities, but commu
nities with a preponderance of perennials and paucity of
annuals (e.g., Australian heathlands, mature Californian
matorrals) are fitted by the exponential species–area model.
Historical Biogeography and Evolution of
Mediterranean Biodiversity
Mediterranean floras and faunas constitute highly com
plex assemblages of species of different biogeographical
origins. These huge species diversities and levels of ende
micity are in great part related to the contrasted historical
biogeography of each Mediterranean ecoregion, with dif
ferent evolutionary patterns and processes before and
after the Pleistocene period.
323
Pre-Pleistocene History: The Diversification
and Mixing of Different Lineages
A general picture suggests that humid and mesothermic
climates predominated throughout most of the Late
Cretaceous and Tertiary in Mediterranean ecoregions
where subtropical forests were dominating. The onset of
Mediterranean climates is relatively young, and pale
oclimatic reconstructions demonstrate that cooling
and drying, with the combination of summer drought
and mild winter temperatures rose quite rapidly in
Late Miocene or Early Pliocene (c. 5–3 Ma). In the
Mediterranean Basin, a gradual but deep climatic change
occurs during the Pliocene (3.5–2.4 Ma), with a significant
drop in temperature and a marked seasonality in thermal
and rainfall regimes; the stabilization of the summer
drought arises here at c. 2.6 Ma. Concomitant with the
appearance of Mediterranean climate conditions, a
massive ecological radiation was initiated in the
Late Miocene–Pliocene. However, there are also phylo
genetic evidences of earlier radiation events during
the Oligocene involving geographic speciation; this is
the case of the shrub genus Protea (Proteaceae) in South
African fynbos and the geophytic genus Androcymbium
(Colchicaceae) in xerophytic grasslands of South Africa
and the Mediterranean Basin. The deterioration of
Tertiary warm climates resulted in the extinction of
several subtropical and warm temperate species during
Plio Pleistocene, but historical discrepancies exist
between the Mediterranean ecoregions, from the point
of view of the impact of environmental changes and the
macroscale process of species recolonization after drastic
climatic events.
Mediterranean ecoregions closely situated to subtro
pical regions and without major geographical barriers
between these two biomes have significantly experienced
moderate cases of species extinctions, because of possible
latitudinal shifts in species range or a higher climatic
stability. This is the case of the eastern lowlands of the
Cape Floristic Region, where patches of subtropical
thickets and warm temperate forests are still dominated
by trees and shrubs of paleotropical and Gondwanian
affinities. These highly species and genus rich forma
tions are extremely ancient, with many elements
phylogenetically basal to the western Cape species. The
precise timing of this lineage diversification is still
unknown but geographic speciation was probably the
key process.
In south–central Chile, the actual coastal forests at
mid latitudes remained fairly stable despite major cli
matic and tectonic changes in southern South America
during the Pleistocene. These forests are indeed charac
terized by a notable evolutionary stability, favoring the
conservation of ancient species assemblages. The nearest
ancestor of the Mediterranean sclerophyllous vegetation
324
Mediterranean
of Chile corresponds probably to the Neogene subtropical
paleoflora that occured in the Proto Andean foothills of
this area during the lower to mid Miocene (c. 20–15 Ma),
and the pre Pleistocene paleofloras were developed
under a more humid and warmer paleoclimate induced
by an Andean rain shadow effect.
In California, shrublands (chaparrals) are composed of
temperate, subtropical, and desert elements. Many of
these Californian chaparral genera appeared in the Early
Tertiary, following the trends since the Eocene toward
cooler and drier climates. This sclerophyllous flora seems
to have arisen as an understory component of evergreen
woodlands and form only the chaparral in response to the
latter appearance of the Mediterranean climate and
related increased disturbance by fire. During unfavorable
periods, southward migration of species in Baja California
or central Mexico, or their location in southern coastal
refugia and subsequent recolonization explain the actual
persistence in California of several outstanding trees
(Sequoia, Sequoiadendron, Tsuga, Umbellularia, etc.), whereas
these genera have disappeared around the Mediterranean
sea because of the existence of several east–west oro
graphic and maritime barriers which prevent putative
latitudinal migrations of species.
Moreover, in the Mediterranean ecoregion, climatic
coolings during the Neogene have provoked severe extinc
tions, for example, of 45 genera of megathermous and
warm temperate ligneous species distributed in the north
western part of the Mediterranean sea. The main
diversification of the Mediterranean Basin species took
place during the Miocene, notably in conjunction with the
collision of the African and Eurasian platforms through the
Arabian Plate. Recurrent episodes of species dispersal and
vicariance have occurred between the eastern and western
Mediterranean regions due to several marine regressions/
transgressions and the ongoing rise of Alpine and Atlas
orogenesis. Another major biogeographic event of the end
of the Tertiary is the closure of the Mediterranean–Atlantic
gateways (now the Gibraltar Strait). This episode known as
the Messinian salinity crisis (5.77–5.33 Ma) induced a con
siderable evaporation of the Mediterranean Sea and the
formation of several land bridges suitable for dispersal–
vicariance events and species radiations. Thus, the
Mediterranean species pool arises from diverse biogeo
graphic origins and if it includes still some subtropical
species, mostly originated from African and Asian lineages,
the extratropical species of autochthonous and northern
lineages predominate.
Influence of Pleistocene Glacial Refugia to
Current Patterns of Mediterranean Biodiversity
Pleistocene climatic cycles have profoundly affected the
biogeographical footprint of several Mediterranean spe
cies, notably in the Mediterranean Basin, California, and
Chile. Recent researches combining genetics and biogeo
graphy (i.e., phylogeography) and paleocology underline
that glacial refugia represent crucial areas for the long
term persistence and dynamics of modern biodiversity in
temperate regions. Glacial refugia constitute territories
sheltered from the strong climatic deteriorations during
Ice Ages, and where species survived the drastic conse
quences of severe cold and aridity. A major and
noteworthy glacial event is the Last Glacial Maximum
(LGM) that occurred c. 20 000 years BP. First, there exists
a clear influence of Pleistocene climatic cycles on patterns
of species richness and endemism. Second, full glacial
refugia have also had a powerful influence in shaping
current patterns of genetic diversity in several temperate
and Mediterranean ecoregions. Finally, these refugia
played an important role on vegetation dynamics
during previous interglacial periods of the Pleistocene;
these areas contributed to the forest recolonization
process that started approximately 13 000 years ago in
the Mediterranean Basin and lasted throughout the
Holocene. Once climatic conditions became truly favor
able, the expansion of a highly diversified deciduous
forest could happen rapidly over large territories, such
as in the Pindos mountains in northwestern Greece where
over 20 deciduous woody plants could already be found
in around 10 000 yr BP.
There is growing evidence, both at a global scale and
for Mediterranean type ecosystems, indicating that bio
diversity hot spots coincide generally with areas that were
buffered against climatic extremes. Reduced impacts of
Milankovitch climate oscillations and smaller amplitude
climatic changes during the LGM constitute the best
descriptors to explain both the survival of paleoendemics
and the speciation of neoendemics. Therefore, the climate
stability–diversity pattern accounts for the location of
most of the endemic rich Mediterranean areas, and this
pattern appears clearly when we compare two ecoregions
with distinct historical biogeography, the Mediterranean
Basin and SW Africa.
Around the Mediterranean, the alternation of humid and
hyperarid phases in North Africa or interglacial and glacial
episodes in Europe have induced profound shifts in the
evolution and geographical distribution of species lineages,
resulting in, respectively, expansion from Mediterranean
refugia or extinction–reduction of populations during unfa
vorable periods. The major Mediterranean areas where
temperate and thermophilous species survived are the
three Iberian, Italian, and Balkan peninsulas, but also the
largest Mediterranean islands and the submontane and
mountain margins of North Africa, Turkey, and
Catalonia Provence. The glacial events induced the extinc
tion of several subtropical lineages and paleoendemics, and
speciation by radiation was quite reduced.
In South Africa, glacial climates of Pleistocene were
moister in western Cape and drier in eastern Cape than
Mediterranean
present. It seems that the western Cape elements have
extended northward into the present succulent karoo and
Namib desert during glacial episodes, whereas drier con
ditions in the east have induced the restriction of Cape
vegetation and species to some mesic refugia. Due to the
peninsula configuration surrounded by ocean, southwes
tern Africa was one of the mildest continental landmasses
of the world, and Pleistocene climates there were excep
tionally stable favoring higher rates of speciation and
lower rates of extinction compared to eastern Cape or to
northern Mediterranean ecoregions.
Convergence versus Nonconvergence of
Mediterranean Ecosystems
Mediterranean climate ecosystems have been often cited
as classic examples of convergence in ecosystem structure
and function owing to their similar environments. But
comparative studies during the 1980s have demonstrated
that this convergence pattern is too simplifying, and sev
eral divergences exist also.
Most discussions of the Mediterranean convergence
hypothesis have focused on similarities among communities
and on ecological similarity of distantly related taxa.
Several cases of ecological convergence deserve attention.
SW Australia and SW Cape are characterized by larger
amounts of summer rain, lower soil fertility, and more
frequent fires, compared to the three other Mediterranean
regions. These two Southern Hemisphere Mediterranean
ecosystems show remarkable convergence of plant traits
and community structure on climatically and edaphically
matched sites. Patterns of plant biodiversity are also similar
on nutrient poor soils in South African fynbos and
Mediterranean heathlands of Spain. Striking convergences
exist in the two Northern Hemisphere Mediterranean eco
regions, and several forest types of the Mediterranean
Basin are relatively similar from a biogeographic and eco
logical point of view, to those of the California Floristic
Province. Ancient landmass connections which lasted until
the Eocene (Madro Tethysian flora) explain the existence
of some common tree genera (Pinus, Quercus, Arbutus,
Cupressus, Juniperus, Platanus) between these regions. At
low and medium altitudes, physiognomical similarities
exist for sclerophyll oak forests. The existence of thermo
philous coniferous forests at the thermo and meso
Mediterranean (Pinus halepensis, P. brutia, P. pinaster) and
the thermo and meso Californian (Pinus attenuata, P. sabini
ana, Cupressus macrocarpa) levels, but also the presence of
several mesophilous coniferous (Abies, Pinus, etc.) at higher
altitude constitutes another major resemblance.
Life history traits shared by many sclerophyllous
woody plants of Mediterranean ecosystems represent an
outstanding example of convergence, with evergreen,
small, or even needle like (ericoid) leaves, low specific
325
leaf area (SLA: ratio of fresh leaf area to dry mass), strongly
seasonal photosynthesis and growth, vigorous resprouting
following fire or cutting. Sclerophylly (evergreen and thick
leathery leaves) is the most common and widespread strat
egy of plants in these Mediterranean ecosystems.
Convergence can be explained by similar abiotic fac
tors (climate and notably the intensity of summer
drought, soil nutrient status fertility, fire regimes), and
phylogenetic composition of lineages induced by com
mon historical biogeography. Rainfall reliability was also
recently suggested as another important factor; the two
more ecologically similar regions (SW Australia and the
SW Cape) have indeed significantly more reliable
regimes than California and the Mediterranean Basin.
If we consider plant traits, and notably leaf evolution
and resprouting capacity of evergreen woody plants, two
hypotheses were formulated to explain the observed con
vergences: (1) evolutionary adaptation, that is, the
production of new phenotypes by the action of natural
selection; (2) niche conservatism, with a relative stasis in
trait evolution, induced by the ecological match between
organisms and their environment caused by the spatial and
temporal sorting of existing lineages. Methods of phyloge
netic comparative biology have recently demonstrated that
similarity between traits of Mediterranean woody plants
could be due more to phylogenetical inertia than to com
mon adaptive strategies under Mediterranean climate. The
absence of deep morphological changes in leaf size and
SLA suggests that most of the ancestors of shrubland taxa
had already acquired plant life history traits that contrib
uted to their success under Mediterranean climates. Thus,
these subtropical ‘phantoms’ predate the onset of the med
iterraneity during the mid Pliocene.
More precisely, there exists a clear co variation of life
history traits with regards to the lineage age, and two
groups with distinctive characters associations can be
defined for Mediterranean angiosperms:
1. A pre Pliocene group, consisting of mostly sclerophyl
lous, vertebrate dispersed, fleshy fruited, and large seeded
plants which resprout from stump after disturbance (fire,
clearing) and are often late colonizers in successional stages
of ecosystem dynamics (e.g., Arbutus unedo, Olea spp., Quercus
spp.). This resprouting ‘strategy’ often considered as typi
cally ‘Mediterranean’ represents in fact an ancient trait that
emerged under a subtropical climatic regime, well before
the advent of the Mediterranean climates.
2. A post Pliocene group, including nonsclerophyllous
plants which are obligate seeders after disturbance (e.g.,
Cistus spp., Lavandula spp.). These taxa successfully diver
sified and competed with taxa of the pre Pliocene group
due to their short life cycle, high seed production with
small and dry fruited anemochorous seeds. Seeders grow
significantly faster and allocate more leaf biomass than
resprouters. These set traits account for the important
326
Mediterranean
ecological plasticity of these plants which are associated
with earlier successional stages.
Therefore, historical biogeography could largely explain
the physiognomic similarities of some ecosystems or spe
cies found also in non Mediterranean type climate
regions. This is the case of several modern matorral
type communities present in the American Southwest,
Mexico, or eastern Australia, whose sclerophyllous vege
tation is roughly similar from a structural and functional
point of view to the true Mediterranean shrublands. Even
in southwestern China, some sclerophyllous species pos
sess striking morphological convergence with
Mediterranean trees; for example, Quercus phillyroides in
the Danxiashan escarpments (Guangdong Province) is
astonishingly similar to the evergreen Mediterranean
oaks (Q. ilex and Q. coccifera), whereas the matorrals with
Olea and Pistacia recently discovered in some gorges of the
Yunnan Province are globally alike to the Mediterranean
ones. These similarities support the view that sclero
phylly represents more a general response of semiarid
conditions with a severe dry season, few intense frost
episodes, and low nutrient soils, rather than a specific
adaptation to the summer drought pattern per se.
Evolutionary convergence among bird communities was
also tested between California, Chile, and the
Mediterranean region, along matched habitat gradients of
increasingly complex vegetation structure (from shrublands
to forests), and compared with non Mediterranean commu
nities. Results of these ecomorphological comparisons
indicate that the Mediterranean bird communities do not
resemble each other any more than the non Mediterranean
ones. To summarize, convergence depends on the ecologi
cal compartments considered; Mediterranean convergence
may exist for some community attributes (e.g., species rich
ness, community structure), and is more likely to occur
among organisms such as plants, invertebrates, or reptiles
that depend on seasonal patterns of climate and nutrient
cycling, whereas homeotherm vertebrates are more deeply
influenced by the structure of ecosystems.
Ecosystem Characteristics and
Processes
Main Ecosystem Types
The vegetation types usually considered as ‘typically
Mediterranean’ are the evergreen and sclerophyllous
shrublands or heathlands, named maquis and garrigue in
the Mediterranean Basin, chaparral in California, mator
ral in Chile, fynbos in SW Africa, kwongan and mallee in
SW Australia. But there exists considerable differences in
the composition and structure of these shrublands
(Table 1). Depending on the type of shrubland chosen,
it is possible to find strong similarities or dissimilarities,
between and within the diverse Mediterranean ecore
gions. The similarity most frequently cited is that
between kwongan and fynbos which show a striking eco
logical convergence with an open shrub cover and a high
shrub diversity with the quasi absence of annuals, a domi
nance of postfire seeders, and serotinous shrubs, and the
frequent occurrence of seed dispersal by ants (myrme
cochory). Shrublands occur generally in a mosaic with
xerophytic grasslands, steppes, woodlands, or forests.
True Mediterranean forests are rare and they repre
sent 1.8% of world’s forest area. Northern Hemisphere
Mediterranean forests show a higher structural and spe
cies diversity than those of the Southern Hemisphere,
because the latter cover less extensive areas and, as in
South Africa, may be outside the range of Mediterranean
bioclimate. The Southern Cape forests are very patchy
with mainly subtropical sclerophyllous trees and conifers
(Afrocarpus, Podocarpus). Forests of Mediterranean Chile
are more diverse due to the strong latitudinal gradient
and the increase in rainfall from north to south; semiarid
Acacia caven and Prosopis chilensis forests in the north are
succeeded by subtropical broad leaved and sclerophyl
lous forests in central Chile, and by deciduous Nothofagus
forests farther south. Together with species rich sclero
phyllous shrublands (kwongan and mallee), the forests
and woodlands of SW Australia are dominated by
Eucalyptus, Acacia, and Casuarina on poor sandy soils,
where mean annual rainfall exceeds 400 mm; several
types of Australian woody vegetation are distinguished
according to the foliage cover of tallest stratum and the
high of the trees.
In the Mediterranean Basin, the current diversity of
forest structures can be organized into three major struc
tural types based on bioclimatic and/or human impact
criteria (Tables 2 and 3).
True forest vegetation types are related to metastable
equilibrium of vegetation structures. They represent the
potential structures at the end of a dynamic ecological
cycle, which can be achieved where soil and climate
conditions are favorable and where the impact of man is
not too strong. Dominant species are sclerophyllous oaks
in semiarid bioclimates and deciduous oaks in more
humid conditions.
Preforest types can be divided into two categories.
Under perhumid, humid, and subhumid bioclimates,
they consist of vegetation structures that have undergone
severe human impact, although their soil is still relatively
well preserved. They are transitory structures from true
forests to more open systems. Under semiarid bioclimatic
conditions, or under particularly stressful conditions (e.g.,
ultramafic substrates) in any bioclimate, preforests are
comprised of shrub dominated vegetation structures
with scattered trees (matorrals). Conifer species (Pinus,
Tetraclinis) play an important role in these structures.
Mediterranean
Presteppic forest types, very frequent in southern and
eastern Mediterranean, consist of open vegetation struc
tures dominated by nonforest plant species under
scattered trees. Nonforest species are steppe type peren
nial species that can eventually be replaced by ruderal
annual species when grazing occurs. Presteppes are most
frequent under warm and hot temperature variants of arid
(and sometimes semiarid) bioclimates. They gradually
merge into steppes under hotter and drier conditions.
On mountains, presteppes are a transitional vegetation
structure from forests (or preforests) to high elevation
steppes dominated by low and scattered cushion like
spiny xerophytes.
Annual grasslands represent also key ecosystems in the
two northern Mediterranean ecoregions. The composi
tion and structure of grassland communities are strongly
controlled by disturbances, which create a complex pat
tern of microsites and canopy gaps. Therefore, the heavily
grazed grasslands of the Mediterranean Basin have prob
ably the greatest alpha diversity of any temperate plant
community, and annuals represent half of the total species
found in this region. Nearly one fifth of California is
covered by grasslands, but most of them are dominated
by non native annuals (Bromus, Avena, Erodium, etc.) origi
nated from the Mediterranean Basin.
Strategies of Resistance to Climatic Stress
Mediterranean climates induce severe and contrasted
stresses to habitats and species. These stresses are com
pounded by the unpredictable nature of weather patterns,
and organisms have to cope with this temporal variation
in climate and resource availability. Ecological and eco
physiological studies indicate that Mediterranean species
demonstrate similar strategies to resist climatic and
edaphic stress. Drought stress proves to be the essential
climatic factor responsible for the restriction of produc
tivity, growth, and survival of several groups of
Mediterranean plant species, notably the evergreen
woody plants. Sclerophyllous leaves also exhibit other
water conservation features such as sunken stomata and
low cuticular conductance. Other strategies to cope with
water stress are related to complex root systems, cellular
tolerance to low water potentials or high secondary com
pound production (e.g., terpenes, tannins).
The drought summer season also induces original
physiological strategies for the two main groups of the
Mediterranean soil microfauna; oribatid mites are more
resistant to dryness and only migrate into the deep soil
layers when the soil water content becomes too low;
collembolas cross the summer in an egg stage, and several
species can surmount the dryness by a deshydratation
process similar to anhydrobiosis. There exists a balance
of the collembola composition between, on one hand, the
winter populations which are composed of common
327
species but highly diversified qualitatively and quantita
tively, and the other hand, the ‘reserve populations’
present in a latent state in the soil which are expressed
only when exceptional summer rainfalls occur or during
the onset of the wet season.
On the community level, a striking example is repre
sented by the biotic assemblages of vernal pools,
precipitation filled seasonal wetlands found mainly in
the Mediterranean climate regions. Inundation during
the growing season largely eliminates colonization by
upland species in the pools whereas the terrestrial phase
is sufficiently desiccating to prevent establishment of
typical wetland species. Several cosmopolitan aquatic
plant genera are shared between the five Mediterranean
ecoregions, such as ferns (Pilularia, Marsilea, Isoetes) or
dicots (Callitriche, Elatine, Ranunculus), whereas vernal
pool specialists are often derived from genera of terres
trial origin. The essential ecological characteristic of these
temporary ponds is the stochastic alternation of flooded
and dry ecophases within and between years. This
induces strong year to year differences not only in pool
hydrology, but also in the composition and dynamics of
vernal pool communities, with a strong temporal and
spatial segregation which limits competition as it is the
case of the larvae of anuran amphibians. The succession of
contrasted phases favors the emergence of varied and
highly specialized plant and animal communities, parti
cularly adapted to this high habitat instability. These
environmental factors also played a significant selective
force in shaping life strategies of temporary pool species.
Dormancy provides a determinant means of enduring
prolonged unfavorable dry periods, and several typical
vernal pool species possess mechanisms that keep them
from emerging under unsuitable conditions: drought
resistant reproductive organs such as seeds or oospores
for the well represented annual plants (c. 80% of the
whole vernal pool specialists in California and the
Mediterranean Basin), and eggs or cysts for crustaceans
(e.g., cladoceran and anostracan branchiopods). A great
adaptability of the life cycle often exists, and when water
levels are shortened and water temperature increases,
invertebrates and amphibians can present an advanced
metamorphosis and several annual plants known as ephe
merophytes can complete their entire life cycle within
only a few weeks.
If we consider the main ecological processes linked to
climatic stress, several studies have demonstrated that
competition increases with aridity. Summer drought and
poor soil nutrients, coupled with frequent disturbances,
explain the reduced rates of competitive displacement
observed in most of the Mediterranean communities.
Changes in species interactions along water gradients
were also demonstrated in arid Mediterranean environ
ments. The importance of positive interactions through
facilitation in arid plant communities represents a
328
Mediterranean
complex and unrecognized process. Nevertheless, some
recent studies suggest that facilitation decreases with ele
vation in dry Mediterranean mountains (Sierra Nevada in
Spain, central Andes of Chile) because of the prevalence
of water stress over temperature stress compared to mesic
alpine communities.
In addition to summer drought, low winter tempera
tures and episodic frost exert a strong influence in limiting
distribution range and causing alteration in species compo
sition and productivity, notably for ecosystems situated in
the northern limits of the Californian and Mediterranean
Basin ecoregions. From freezing tests, three groups of
plants can be distinguished with regard to frost suscept
ibility in the Mediterranean Basin: (1) the most sensitive
species with 50% frost injury to the leaves at 6 to 8 C
and to the shoots at 9 to 15 C (Ceratonia siliqua, Myrtus
communis); (2) the medium sensitive species with serious
damage to the foliage at 12 to 14 C, and to the stems
at 15 to 20 C (Olea europaea, Quercus coccifera, Pinus
halepensis); (3) the resistant species, not seriously damaged
until 15 to 25 C (Quercus ilex, Cupressus sempervirens).
The distribution of Mediterranean species of the Northern
Hemisphere is indeed shaped by frost events such as late
spring below freezing temperatures and absolute minimum
temperatures. For example, the extremely cold winters of
1956 and 1985 in the northern Mediterranean contributed
to the determination of the distribution of some keystone
trees such as olive tree, Aleppo pine, and holm oak.
probability to include highly productive species in spe
cies rich communities; (2) niche complementarity
between species, leading to more complete utilization of
resource in intact ecosystems relative to depauperate
ones: space filling, for example, by herbaceous plants
occupying the gaps in shrub canopies, favors the exploita
tion of different soil layers and resource supply by
different root systems and this phenomenon increases
productivity; (3) the increase of positive interactions
between species in complex assemblages.
Disturbances and Ecosystem Dynamics
Natural disturbances represent a determinant component
of the Mediterranean ecosystem dynamics and for the
maintenance of their high biodiversity. Indeed, the long
established mature forests which constitute the typical final
stage of ecological successions in the Mediterranean Basin,
harbor few typical Mediterranean species and the more
architecturally complex vegetation exhibit a high degree of
convergence in species richness and composition, with a
domination of boreal and eurasiatic forest species. On the
contrary, nonmature ecosystems with a more reduced
architectural complexity and a higher structural heteroge
neity comprise a dominance of Mediterranean species.
Thus, natural disturbances such as fire or herbivory can
play a major role in the expression of the mediterraneity by
rejuvenating mature ecosystems.
Links between Biodiversity and Ecosystem
Function
Ecosystem Responses to Fire
The relationship between diversity and productivity of
ecosystems remains a controversial issue, and there are
still few experimental evidences for Mediterranean eco
systems. Nevertheless, some results related to grasslands
and shrublands of the Mediterranean Basin suggest that
plant diversity is positively correlated with ecosystem
function, notably with primary production. However, in
mixed grassland communities, where several growth
forms coexist, one of a few dominant species (keystone
species) may hide or reverse the observed diversity–
biomass production relationship: the inclusion of a
dominant grass in experimental low diversity plots can
produce a constant level of productivity along the gradi
ent of species diversity. Thus, the species composition
appears to be the main determinant of the productivity
performance at each diversity level. Shrubs, grasses, and
geophytes have different attributes in these communities
and they can significantly affect the value of the ecosys
tem functioning. Each growth form or functional type
contributes significantly to the overall productivity per
formance of the studied Mediterranean ecosystems.
Several mechanisms were proposed to explain these
patterns: (1) the ‘sampling effect’, related to the higher
Fire determines major vegetation structure and composi
tion of several world biomes, as well as temperature,
precipitation, and water balance. The fire prone vegetation
cover constitutes at present 40% of the world’s land surface,
including especially the Mediterranean regions, except
central Chile where fires are less frequent (Table 1). In
Mediterranean type ecosystems, most of the areas occupied
by shrublands have the climate potential to be forests and
these shrublands are generally fire maintained. But in con
trast with current opinion that often attributes the existence
of these ecosystems only to anthropogenic burning,
Mediterranean shrublands seem to have naturally expanded
in the Late Tertiary, with flammable C4 grassy biomes. In
the Mediterranean Basin, natural development of pine for
ests has also played a determinant role in increasing fire
frequency since the Miocene (c. 9 Ma).
Forest fires are generally considered as catastrophic
events in the Mediterranean Basin and California where
big fires occur, but the evolutionary and ecological influ
ences of fire also represent key parameters for driving
landscape diversity, ecosystem heterogeneity, vegetation
dynamics, and species differentiation. Fire determines
indirect environmental changes with greater fluctuations
Mediterranean
in temperature and increasing oxygen concentration
in soils, increases light and water availability, reduces
aboveground competition and determines a proper
regeneration niche for fire adapted plants. In burned eco
systems, a higher abundance and species richness is
quickly observed several months, or even weeks, after
fire and important changes in species occurrence and
diversity occur. These immediate and profound ecologi
cal modifications are considerably attenuated during the
second postfire season because Mediterranean ecosystems
are particularly resilient and fire adapted. Nevertheless,
these brief structural changes are of particular importance
for the regeneration window of fire adapted species.
Fire represents a major selective force that shapes the
evolution of plant reproductive traits in fire prone
Mediterranean environments, and thus ecosystem
dynamics. Strong serotiny, fire stimulated flowering or
germination, smoke or charred wood induced germina
tion, resprouting ability through lignotubers, stumps or
burls, and seed dormancy favor plant persistence and
these traits are more common in Mediterranean ecore
gions with a longer fire history or a higher fire frequency.
Several lines of evidence corroborated by recent phylo
genetic analysis indicate that there has been stronger
selective pressure for fire persistence traits in California
than in the Mediterranean Basin. Because of a more
drastic fire occurrence in California, plant species have
evolved here more frequently toward the association of
resprouting capacity and propagule persistence than
Mediterranean Basin plants. Furthermore, another pecu
liarity is the specialization of Californian annual plants
which encompass numerous fire endemics that can persist
as dormant seed banks for many decades between fires
and that occur only 1 or 2 years after a fire.
Influence of Herbivory and Grazing
Herbivore pressures also constitute a crucial feature in
some Mediterranean ecosystems, and grazing affects often
seriously the structure and diversity of Mediterranean com
munities, notably grasslands. In the Mediterranean Basin,
grasslands and shrublands have experienced frequent man
induced grazing for c. 9000 years and ungulate browsing.
These ancient selective pressures explain the widespread
existence of distinct plant species life attributes, which can
contribute efficiently to reduce the cumulative damage
induced by herbivores and linked to the long leaf life span
of evergreen plants. Morphological changes following her
bivory include the reduction of leaf size, low SLA, physical
toughness, modification of branch density, developments of
thorns and hairs, and increase of leaf chemical defenses
by phenolic or tannin compounds. Variations of phenolic
levels in relation to leaf age and season suggest an evolu
tionary adaptation of Mediterranean evergreen plants to
high densities of large herbivores.
329
Without any disturbance for a few years, grasslands
become dominated by a few species of perennial grasses,
forbs, or tall large seeded annuals, which form closed
swards with high biomass, cover, and height. Thus, spe
cies regeneration will be only successful in gaps created
by light grazing, and the observed peaks of species
richness with moderated grazing are consistent with the
classic intermediate disturbance hypothesis. The barren
areas between mature shrublands or grasslands are also
possibly driven by rodent and rabbit activities, which
control not only plant community boundaries and struc
ture but also dynamics by influencing nutrient linkages
between communities. The complex mosaic of microsites
and the constant grazing pressure shift advantage to smal
ler plants that can occupy the microheterogeneity finely,
and the diversities observed can be considerable with
more than 50 plant species on 1 m2 plots in moderately
grazed grasslands or steppes of the Mediterranean Basin.
Fire and grazing must be regarded as two disturbances
with not always additive consequences, but often with
distinct and interactive effects on structure and dynamics
of communities. Moreover, the ability of Mediterranean
woody species to resprout after fire seems not to originate
from an adaptation to recurrent fires, but rather from an
older adaptation to losses of aboveground biomass mainly
induced by herbivory.
Conclusion: Current Evolution of
Mediterranean Ecosystems under
Global Changes
The originality of Mediterranean climate ecosystems can
be explained by complex interaction between historical
biogeography patterns and unique ecological processes.
Insights of paleoecology and phylogeography indicate the
importance of paleogeographical and paleoclimatic
events in shaping this massive and unique Mediterranean
biodiversity and ecosystem types. Recent phylogenetic
methods of comparative biology have demonstrated that
the evolution of evergreen sclerophylls, formerly desig
nated as ‘typically Mediterranean’, may have occurred
long before the origin of Mediterranean climates. The pre
dominance of sclerophylly and related conserved life
history traits in numerous keystone ligneous plants of
Mediterranean ecosystems reflect mainly the ecological
success of these taxa under climatic stress and fires, fol
lowing by some evolutionary refinements in physiology,
rather than a real origin of these traits under Medi
terranean climates. But if these species and ecosystems
were able, to a great extent, to surmount past environmental
changes, their immediate future seems now quite worrying.
At present, these Mediterranean ecosystems are faced
with rapid and previously unknown global environmental
changes with important repercussions in structure and func
tion. Several models predict that the greatest changes on a
330
Peatlands
world scale are expected in Mediterranean regions, with the
highest vulnerability in mountain areas. Climate projections
for Mediterranean type ecosystems suggest drier and war
mer conditions, which have probably already triggered
species distribution shifts, ecophysiology, phenology, and
species interactions. The elevation of atmospheric CO2
linked to anthropogenic causes could induce an increase of
the productivity of many Mediterranean trees, as has already
been measured for Quercus ilex and Quercus pubescens in the
northern Mediterranean. On the other side, experimental
studies have demonstrated that a decrease in water availabil
ity and an increase in temperature might affect the growth
pattern and annual productivity of dominant shrubs, with
alteration of competitive abilities. These modifications could
change, in turn, species composition and structure of several
Mediterranean habitats. Furthermore, the generalized and
deep magnitude of human impact in the Mediterranean
regions accentuates the effects of climatic change, and could
obliterate the efficient capacity of ecological resilience of
Mediterranean type ecosystems, even if they have been sub
mitted in the past to other drastic and rapid changes.
See also: Chaparral.
Further Reading
Ackerly DD (2004) Adaptation, niche conservatism, and convergence:
Comparative studies of leaf evolution in the California chaparral.
American Naturalist 163: 654 671.
Arianoutsou M and Groves RH (1994) Plant Animal Interactions in
Mediterranean Type Ecosystems. Dordrecht: Kluwer Academic
Publishers.
Arroyo MTK, Zedler PH, and Fox MD (1995) Ecological Studies, Vol.
108: Ecology and Biogeography of Mediterranean Ecosystems in
Chile, California and Australia. New York: Springer.
Blondel J and Aronson J (1999) Biology and Wildlife of the
Mediterranean Region. Oxford: Oxford University Press.
Cowling RM (1992) Fynbos, Nutrients, Fire and Diversity. Cape Town,
South Africa: Oxford University Press.
Cowling RM, Rundel PW, Lamont BB, Arroyo MK, and Arianoutsou M
(1996) Plant diversity in Mediterranean climate region. Trends in
Ecology and Evolution 11: 362 366.
Cowling RM, Ojeda F, Lamont BB, Rundel PW, and Lechmere Oertel R
(2005) Rainfall reliability: A neglected factor in explaining
convergence and divergence of plant traits in fire prone
Mediterranean climate ecosystems. Global Ecology and
Biogeography 14: 509 519.
Dalmann PR (1998) Plant Life in the World’s Mediterranean Climates.
Oxford: Oxford University Press.
Davis GW and Richardson DM (1995) Ecological Studies, Vol. 109:
Biodiversity and Ecosystem Function in Mediterranean Type
Ecosystems. Berlin and Heidelberg: Springer.
Di Castri F and Mooney HA (1973) Ecological Studies, Vol. 7:
Mediterranean Type Ecosystems. Origin and Structure. Berlin,
Heidelberg, and New York: Springer.
Di Castri F, Goodall DW, and Specht RL (1981) Ecosystems of the
World, Vol. 11: Mediterranean Type Shrublands. Amsterdam:
Elsevier.
Hinojosa LF, Armesto JJ, and Villagrán C (2006) Are Chilean coastal
forests pre Pleistocene relicts? Evidence from foliar physiognomy,
palaeoclimate, and phytogeography. Journal of Biogeography
33: 331 341.
Keeley JE and Fotheringham CJ (2003) Species area relationships in
Mediterranean climate plant communities. Journal of Biogeography
30: 1629 1657.
Mazzoleni S, Di Pascale G, Di Martino P, Rego F, and Mulligan M (2004)
Recent Dynamics of Mediterranean Vegetation and Landscape.
London: Wiley.
Mittermeier RA, Robles Gil, Hoffmann M, et al. (2004) Hotspots
Revisited: Earth’s Biologically Richest and Most Endangered
Terrestrial Ecoregions. Monterrey: CEMEX, Washington:
Conservation International, Mexico: Agrupación Sierra Madre.
Moreno JM and Oechel WC (1994) Ecological Studies, Vol. 107: The
Role of Fire in Mediterranean Type Ecosystems. New York:
Springer.
Ornduff R, Faber PM, and Keeler Wolf T (2003) California Natural History
Guides, Vol. 69: Introduction to California Plant Life. Berkeley and
Los Angeles: University of California Press.
Quézel P and Médail F (2003) Ecologie et biogéographie des forêts du
bassin méditerranéen. Paris: Elsevier.
Smith Ramı́rez C, Armesto JJ, and Valdovinos C (2005) Historia,
biodiversidad y ecologı́a de los bosques costeros de Chile. Santiago
de Chile: Editorial Universitaria.
Peatlands
D H Vitt, Southern Illinois University, Carbondale, IL, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Occurrence
Environmental Limiting Factors
Peatland Types
Important Processes in Peatlands
Initiation and Development of Peatlands
Peatlands as Carbon Sinks
Further Reading
Introduction
when carbon that is sequestered in plant biomass through
the process of photosynthesis exceeds the long term loss of
this carbon to the atmosphere via decomposition plus losses
of carbon dissolved in water removed from the peatland
through hydrological flow. Globally, peatlands contain
Peatlands, or mires as they are sometimes called, are char
acterized by often deep accumulations of incompletely
decomposed organic material, or peat. Peat accumulates
Peatlands
about 30% of the world’s terrestrial soil carbon, while
covering only about 3–4% of the Earth’s surface, and as
such their carbon storage is considerably greater than their
land surface area might indicate. Peatlands, in general, are
relatively species poor when compared to upland
communities in the same geographic region. However,
due to the specialized environmental conditions often asso
ciated with peatlands, plants, and animals found only in
these ecosystems are sometimes present. Peatlands are espe
cially known for the presence of carnivorous plants such as
Sarracenia and Drosera and for the occurrence of a large
number of species of peat mosses (the genus Sphagnum).
331
Mire located between the Ob and Irtysh Rivers at about
58 N and 75 W). Two other large peatland complexes
are the Hudson Bay Lowland in eastern Canada and
the Mackenzie River Basin in northwestern Canada.
Although peatlands have long been associated with cool,
oceanic climatic regimes such as those in Britain and
Ireland and indeed peatlands are common in these areas
(in fact peatlands are most abundant in areas where the
regional climate is continental with short cool summers
and long cold winters), the vegetation is coniferous and
evergreen, and the upland soils are podzolic.
Environmental Limiting Factors
Occurrence
Globally, peatlands occupy about 4 million km2, with the
boreal and subarctic peatland area estimated to be approxi
mately 3 460 000 km2, or about 87% of the world’s
peatlands. Six countries have greater than 50 000 km2 of
peatland and these account for 93% of the world’s
peatlands – five of these countries are predominantly boreal.
Russia contains 1.42 million km2, Canada 1.235 million km2,
the US 625 000 km2, Finland 96 000 km2, and Sweden
70 000 km2; Indonesia has an estimated 270 000 km2 as
well. Although peat forming plant communities occur in
most of the world’s nine zonobiomes, they are most pre
valent in zonobiome VIII (cold temperate), or more
commonly termed the boreal forest or taiga (Figure 1).
The world’s largest peatland complex is located in west
ern Siberia (especially noteworthy is the Great Vasyugan
The initiation, development, and succession of peatland
ecosystems are influenced by a number of regional, exter
nal factors. Especially important are hydrological and
landscape position, climate, and substrate chemistry.
These regional allogenic factors determine a number of
site specific factors that influence individual peatland
sites. These local factors include rate of water flow, quan
tity of nutrient inputs, the overall chemistry of the water
in contact with the peatland, and the amount of water
level fluctuation. Additionally, there are a number of
internal, or autogenic, processes that help regulate peat
land form and function (Figure 2). These allogenic and
autogenic factors operate in an everchanging world of
disturbance that includes natural disturbances, especially
wildfire, as well as anthropogenic disturbances such as
mining, forestry, and agriculture.
180° W 140° W 100° W 60° W 20° W 20° E 60° E 100° E 140° E 180° E
70° N
70° N
50° N
50° N
30° N
30° N
10° N
10° N
10° S
10° S
30° S
30° S
50° S
50° S
70° S
70° S
180° W 140° W 100° W 60° W 20° W 20° E 60° E 100° E 140° E 180° E
Figure 1 Estimated global distribution of peatlands. Areas colored in light green are those having >10% peat cover. The orange
areas in North America and Siberia are the world’s largest peatland complexes. The dot in western Siberia is the location for the
Vasyugan peatland.
332
Peatlands
Position
Substrate
Climate
Autogenic
change
Water
quantity
Water
quality
Vegetation, flora, fauna
Bog
development
Decomposition
Succession
Function
Form
Patterns
Carbon
sequestration
Landforms
Nutrient
inputs
Disturbance
Production
Water level
fluctuation
Figure 2 Substrate, position, and climate are regional factors that influence six local factors that are shown in boxes in the diagram.
These local drivers direct both the form and function of bogs and fens. Adapted from Vitt DH (2006) Peatlands: Canada’s past and
future carbon legacy. In: Bhatti J, Lal R, Price M, and Apps MJ (eds.) Climate Change and Carbon in Managed Forests,
pp. 201–216. Boca Raton, FL: CRC Press.
Peatland form and function are dependent on the
process of peat accumulation and the pattern of loss or
gain of carbon from habitats. Peat accumulation is depen
dent on the input of organic matter produced by
photosynthesis. This organic matter is first accumulated
in the upper, aerobic (or acrotelm) peat column wherein
relatively rapid rates of decomposition occur. The rate at
which this partially decomposed organic matter is depos
ited into the water saturated, anaerobic peat column (the
catotelm), wherein the rate of decomposition is extremely
slow, largely determines the amount of carbon that will
accumulate at a given site. Thus, the amount of carbon,
and hence the quantity of peat, that is deposited at a
peatland site is dependent on photosynthesis, aerobic
decomposition within the acrotelm, and subsequent anaero
bic processes in the catotelm, including methenogenesis
and sulfate reduction.
Peatland Types
Peat forming wetlands are in general ecosystems that
have accumulated sufficient organic matter over time to
have a well developed layer of peat. In many soil classi
fications, this is defined as soils having greater than 30%
organic matter that forms deposits greater than 30–40 cm
in depth. Non peat forming wetlands such as marshes
(wetlands without trees) and swamps (wetlands domi
nated by a tree layer) mostly have less than 30–40 cm of
accumulated organic material and over time have not
been able to sustain continued accumulation of a car
bon rich peat deposit. Numerous classifications have
been proposed that distinguish between various peatland
types. For example, peatlands have been classified based
on the source of water that has the primary influence on
the peatland. Thus, peatlands that are influenced by water
that has been in contact with soil or lake waters are
termed geogenous and are divided into three types.
Peatlands may be topogenous (influenced by stagnant
water, mostly soil water, but also nonflowing water bodies
as well), limnogenous (influenced by flood water from
water courses resulting in lateral flow away from the
direction of stream flow), or soligenous (influenced by
flowing water, especially sheet flow on gentle slopes,
including seepages and springs). Contrasted to these geo
genous types of peatlands, others may be ombrogenous
(influenced only by rain water and snow).
Peatlands are extremely variable in vegetation
structure; they may be forested (closed canopy), wooded
(open canopy), shrub dominated, or sedge dominated.
Ground layers may be moss dominated, lichen domi
nated, or bare. Finally, peatlands vary as to where they
occur on the landscape: in association with streams, lakes,
springs, and seeps or isolated at higher elevations in the
watershed. Peatlands often occur on the landscape as
‘complex peatlands’, wherein several distinctive peatland
types occur together (Figure 3). Finally, and perhaps
most universally utilized, is a classification that combines
aspects of hydrology, vegetation, and chemistry into a
functional classification of peat forming wetlands. In
Peatlands
333
early field studies in Sweden provided an overarching
view of how hydrology, water chemistry, and flora are
associated, and more recent studies delineate how these
combined attributes together form a functional classifica
tion of northern peatlands that provides an ecosystem
perspective.
Bogs
Figure 3 Peatland complex in northern Alberta, Canada.
Patterned fen in left foreground, bog island with localized
permafrost (large trees) and melted internal lawns to left, and
curved treed bog island to right background. Small treed, oval
island in center is upland.
general, this view of peatlands would consider hydrology
as fundamental to peatland function and recognize two
peatland types – fens and bogs.
Fens are peatlands that develop under the influence of
geogenous waters (or waters that influence the peatland
after being in contact with surrounding mineral, or
upland, substrates). Waters contacting individual peat
lands have variable amounts of dissolved minerals
(especially base cations (Naþ, Kþ, Ca2þ, Mg2þ) and asso
ciated anions (HCO3, SO2 , Cl )), and may also vary in
the amount of nutrients (N and P) as well as the number
of hydrogen ions. Further complicating this minerotrophy
is variation in the flow of water, including amount of flow
and as well as source of the water (surface, ground, lake, or
stream). Peatlands receiving water only from the atmos
phere via precipitation are hydrologically isolated from
the surrounding landscape. These ombrogenous peat
lands, or bogs, are ombrotrophic ecosystems receiving
nutrients and minerals only from atmospherically depos
ited sources.
In summary and from a hydrological perspective, in
fens water flows into and through the peatland after it has
been in contact with surrounding materials, whereas in
bogs water is deposited directly on the peatland surface
and then flows through and out of the bog directly onto
the surrounding landscape. Thus, fens are always lower in
elevation than the surrounding landscape, while bogs are
slightly raised about the connecting upland areas.
The recognition that hydrology is the prime factor for
dividing peatlands into fens and bogs dates back to the
1800s. However, in the 1940s, Einar DuReitz recognized
that vegetation composition and floristic indicators could
be used to further characterize bogs and fens. Somewhat
later, Hugo Sjörs associated these floristic indicators with
variation in pH and electrical conductivity (as a surrogate
for total ionic content of the water). The results of these
Bogs are functionally ombrotrophic. At least in the
Northern Hemisphere, they have ground layers domi
nated by the bryophyte genus Sphagnum (Figure 4).
Sedges (Carex spp.) are absent or nearly so. The shrub
layer is well developed and trees may or may not be
present. Nearly, all of the vascular plants have associa
tions with mycorrhizal fungi. Microrelief of raised
mounds (hummocks) and depressions (hollows) is gener
ally well developed. The peat column consists of a deep
anaerobic layer (the catotelm), wherein decompositional
processes are extremely slow and a surficial layer of
1–10 dm of the peat column that occupies an aerobic
zone (the acrotelm). The acrotelm extends upward from
the anaerobic catotelm and is mostly made up of living
and dead components of Sphagnum plants, wherein vascu
lar plant roots and fallen vascular plant aboveground litter
occur. Well developed acrotelms are unique to ombro
trophic bogs and provide opportunities to study
atmospheric deposition and ecosystem response to such
deposition.
Bogs are acidic ecosystems that have pH’s of around
3.5–4.5. Base cations are limited owing to the ombrogen
ous source of water and to the cation exchange abilities of
Sphagnum (see below). Bicarbonate is lacking in bogs and
carbon is dissolved in the water column only as CO2. The
lack of geogenous waters limits nutrient inputs to those
derived only from atmospheric deposition, and thus nitro
gen and phosphorus are in short supply.
Figure 4 Mixed lawn of the peatmosses: Sphagnum
angustifolium, mostly to the left, and S. magellanicum (red),
mostly to right.
334
Peatlands
Bogs appear to be limited in distribution to areas
where precipitation exceeds potential evapotranspiration.
In many oceanic regions of the Northern Hemisphere
(especially Britain, Ireland, Fennoscandia, and coastal
eastern Canada), bogs form large treeless expanses. In
Europe, the Ericaceous shrub, Calluna vulgaris, forms a
characteristic component of these treeless landscapes.
Many of these oceanic bogs are patterned, with a series
of pools of waters separated by raised linear ridges. This
sometimes spectacular pool/ridge topography forms
either concentric or eccentric patterns (Figure 5), with
water flowing from the highest raised center of the bog to
the lower surrounding edges. Runoff from the surround
ing upland (and from the raised bog itself) is concentrated
at the margins of these raised bogs and due to increased
nutrients, decomposition processes are greater and peat
accumulation somewhat less. Thus, the central, open,
raised ‘mire expanse’ part of a bog is surrounded by a
wetter, often shaded lagg, or moat, and this ‘mire margin’
zone may be dominated by plants indicative of fens. Some
oceanic bogs have a rather flat mire expanse, with occa
sional pools of water. Whereas the mire expanse surface of
these raised bogs is flat, the dome of water contained
within the bog peat is convex and thus the driest part of
the bog is at the edges just before contact with the fen
lagg. This marginal, relatively dry upslope to the mire
expanse is usually treed and is termed the ‘rand’.
In continental areas, bogs have a very different appear
ance (Figure 6). These continental bogs have a conspicuous
tree layer and abundant shrubs (mostly Ledum spp. or
Chamaedaphne calyculata) while pools of water are not pre
sent. In North America, the endemic tree species, Picea
mariana, dominates these continental bogs, while in Russia
bogs have scattered individuals of Pinus sylvestris. Farther
north in the subarctic and northern boreal zones, peat soils
contain permafrost. When entire bog landforms are frozen,
the bog becomes drier and dominated by lichens (especially
species of the reindeer lichen, Cladina). Unfrozen or melted
areas contained within these peat plateaus are easily
Figure 5 An oceanic eccentric bog. Maine, USA. Highest
elevation of bog is to center left, with elongate axis sloping to
distant right. Photo is courtesy of Ronald B. Davis.
Figure 6 A continental ombrotrophic bog from western
Canada. Tree species is Picea mariana (black spruce).
recognized features termed collapse scars (Figure 7). Peat
plateaus form extensive landscapes across the subarctic
zone of both North America and Siberia. Farther south in
the boreal zone, bog landforms may contain only scattered
pockets of permafrost (frost mounds), that over the past
several decades have been actively melting. Recent melting
of the raised frost mounds results in collapse of the mound
and active revegetation by fen vegetation to form wet,
internal lawns with associated dead and leaning trees
(Figure 8).
Fens
Fens are peatlands that are minerotrophic that when com
pared to bogs have higher amounts of base cations and
associated anions. All fens have an abundance of Carex and
Eriophorum spp. and water levels at or near the surface of
the peat (thus acrotelms are poorly developed). Unlike
bogs that are characterized by high microrelief of hum
mocks and hollows, fens feature a more level topography
Figure 7 Extensive peat plateaus with permafrost (whitish
areas dominated by the reindeer lichens in the genus Cladina),
with isolated collapse scars (without permafrost – greenish
circular to oblong areas), and with lush growth of Sphagnum
species and sedges.
Peatlands
335
Poor fens
These Sphagnum dominated peatlands are associated with
acidic waters (pH 3.5–5.5) that contain the least amount of
base cations and little or no bicarbonate alkalinity.
Rich fens
Figure 8 Bog dominated by Picea mariana in background, with
dead snags in foreground, indicating recent permafrost collapse
and the formation of an internal lawn and dominated by carpet
and lawn species of Sphagnum.
True mosses dominate the ground layer of rich fens,
especially a series of species that are red brown in color
and often termed ‘brown mosses’. Examples of important
species would be Drepanocladus, Hamatocaulis, Warnstorfia,
Meesia, Campylium, Calliergon, and Scorpidium. Waters have
pH varying from 5.5 to more than 8.0 and base cations are
relatively abundant, especially calcium. Alkalinity varies
from very little to extremely high amounts of bicarbonate.
Rich fens occur as two types centered on the chemistry of
the pore waters. ‘Moderate rich fens’ have pH values
between 5.5 and 7.0, with little alkalinity. Both brown
mosses and some mesotrophic species of Sphagnum (e.g.,
S. subsecundum, S. teres, and S. warnstorfii) dominate the
ground layer. ‘Extreme rich fens’ are bicarbonate rich
peatlands, often with deposits of marl (precipitated
CaCO3) and pH ranging from around neutral to over
8.0. Species of Scorpidium, Campylium, and Hamatocaulis
dominate the ground layer.
Whereas water quality (¼ chemistry) is the main factor
controlling fen type and flora, water quantity (¼ flow) con
trols vegetation structure and surface topography. Fens,
whether poor or rich, are vegetationally extremely variable,
ranging from sites having abundant trees (dominated by
Larix laricina in North America), to sites dominated by
shrubs (mostly Betula, Alnus, and Salix), to sites having only
sedges and mosses. Topographically, fens may be homo
geneous and dominated by lawns and carpets. However, as
water flowing through the fen increases, the surface vegeta
tion develops a reticulation of wet pools and carpets
separated by slightly raised ridges. Further increase in
flow of water directs the patterns into linear pools (some
filled with floating vegetation ¼ carpets), sometimes termed
flarks, alternating with linear ridges (termed strings;
Figure 10). These pool/string complexes are oriented per
pendicular to water flow, with smaller pools always
upstream from the larger ones. Especially prevalent in
Scandinavia and Russia, these patterned fens and associated
bog islands form extensive peatlands termed aapamires.
Important Processes in Peatlands
Figure 9 A carpet of the brown moss, Scorpidium scorpioides,
a characteristic species of rich fens.
of extensive carpets and lawns dominated by species of
mosses (Figure 9). Depending on the characteristics of the
surrounding water, fens can by divided into three types.
Acidification
Sphagnum species have cell walls rich in uronic acids that
in aqueous solution readily exchange a hydrogen ion for a
base cation. The base cations that are in solution in bogs
and poor fens are received by the peatland from atmo
spheric deposition or inflowing water and are always
associated with an inorganic anion (HCO3, SO24 , Cl ).
336
Peatlands
If Sphagnum species establish, then cation exchange
proceeds, acidity increases while alkalinity decreases,
and rich fen plant species are replaced by poor fen species
tolerating acidic conditions. This acidification of rich fens
has been documented in the paleorecord wherein the
change from rich fen to poor fen vegetation takes place
extremely rapidly, perhaps in the order of 100–300 years.
As a result, these transitional rich fen–poor fen commu
nities are short lived on the landscape and among the
most rare of peatland types.
Water Retention
Figure 10 A patterned fen in western Canada characterized by
elongate pools (flarks) separated by raised ridges (strings),
oriented perpendicular to water flow.
When the base cation is exchanged for the organically
produced Hþ, acidity of the peatland waters is produced.
This acidity thus originated through the exchange of an
inorganic base cation for an Hþ produced by Sphagnum
growth – hence this is termed inorganic acidity. Inorganic
acidity relies on the presence of base cations and can only
produce acidity when base cations are present in the pore
water to exchange. Inorganic acidity is an extremely
powerful process when abundant base cations are present
such as in rich fens transitional to poor fens and in poor
fens. In bogs, with limited supplies of base cations due to
their ombrogenous water supply, inorganic acidity is less
important.
Organic material produced by plants is decomposed
and carbon mineralized through bacterial and fungal
respiration. Under aerobic conditions, bacteria break
down long cellulose chains and in doing so eventually
produce short chained molecules that are small enough to
be dissolved in the pore waters. This dissolved organic
carbon (DOC) may be lost to the peatlands via runoff or
may remain suspended in the pore waters for some length
of time. These decompositional processes produce acidity
through dissociation of humic acids, acidity that is com
pletely produced via organic processes; hence, peatland
acidity produced via decompositional processes, and
extremely important in ombrotrophic bogs, is termed
organic acidity.
Rich fens, with pH above 7.0, also accumulate deep
deposits of peat and are well buffered by large inputs of
bicarbonate alkalinity. With continued inputs of bicarbo
nate, rich fens may remain stable for millennia, dominated
by brown mosses that have little capacity for inorganic
acidification, but strong tolerance for the alkaline peat
land waters. However, as rich fens accumulate peat to
depths of several meters, there is the possibility that the
active surface layer will become more isolated from
the bicarbonate inputs and alkalinity may decrease to
the point that some tolerant species of Sphagnum may invade.
The surface of a peatland lies on a column of water
contained within the peat column. The peatland surface
consists of a nearly complete cover of mosses (either peat
mosses (Sphagnum) or true mosses (brown mosses)) that
are continually pushed upward by the accumulating peat.
This upward growth is limited only by the abilities of the
peat and living moss layer to maintain a continuous water
column that allows the living moss layer to grow. The
vascular plants that grow in this water soaked peat col
umn produce roots that are largely contained in the small
upper aerobic part of the peat. The mosses, however, alive
and growing only from their uppermost stem apices, must
maintain contact with the water column; thus, wicking
and retaining of water above the saturated water column
is paramount for maintenance of the moss layer. Peatland
mosses have special modifications that help in this regard.
Although some brown mosses have adaptations for water
retention, such as the development of a tomentum of
rhizoids along the stems, numerous branches along the
stem that provide small spaces for capillarity, and leaves
that have enlarged bases that retain water, it is in species
of Sphagnum where water retention (up to 20 times dry
plant weight) is greatly enhanced through a number of
morphological modifications. Sphagnum has unistratose
(one celled thick) leaves consisting of alternating, large,
dead, hyaline cells and small, partially enclosed, living,
green cells. The walls of the hyaline cells are perforated
with pores and are strengthened by the presence of cross
fibrils. Stems and branches are often encased in an outer
layer of one or more rows of dead, enlarged cells. All of
these hyaline cells have lost their living cell contents very
early on in development and as a result the ratio of carbon
to nitrogen is high. In addition to the features that allow
the plants to hold water internally, the entire Sphagnum
plant is a series of tiny spaces that serve as reservoirs for
capillarity. The branches are surrounded by numerous,
overlapping, very concave branch leaves (one cell thick).
The branches are attached to the stem in fascicles of three
to five branches, half of which hang along the stem and
half extend outward at more or less 90 . The fascicles of
branches originate at the stem apex, and slowly develop
while still close together at the apex of the stem. This
Peatlands
Figure 11 A longitudinal view of the canopy of Sphagnum;
each stem is terminated by a capitulum of young branches. The
branches along the stem are covered with numerous overlapping
leaves and organized into fascicles that have branches that hang
down along the stem as well as branches that spread outward
from the stem allowing the individual stems to be evenly spaced
from one another.
group of maturing branches, the capitulum, along with the
top 1–5 cm of mature stem and associated branches form a
dense canopy. In total, this canopy (Figure 11) consists of
numerous small spaces of different sizes and, along with
the dead hyaline cells of the leaves and branches, provides
the mechanism for wicking and retention of capillary
water far above the actual water table, which in turn
provides the framework for the aerobic peat column that
is so characteristic of bogs.
Nutrient Sequestration (Oligotrophification)
Peat forms due to slow decompositional processes that
allow organic materials to be deposited as peat. As organic
material is deposited, it contains within its carbon matrix
nutrients, especially nitrogen and phosphorus, which
were originally incorporated in the cell structure of the
living plants, especially those of Sphagnum and brown
mosses. Relatively rapid decomposition in the acrotelm
mineralizes only a portion of the total nutrients tied up in
the plant material, making these available for further
plant growth as well as fungal and bacterial processing.
However, upon entry to the catotelm, almost all decom
positional activity stops and the nutrients become tied to
organic materials in unavailable forms. Thus, rather than
being recycled and remaining available for new plant
growth, nitrogen and phosphorus become part of long
term unavailable nutrient pools. The lack of ability to
utilize this unavailable pool of nutrients causes peatlands
over time to become more oligotrophic at their surface
yet also having large amounts of stored nitrogen and
phosphorus. For example, Sphagnum peat is generally
about 1% nitrogen; however, almost all of this catotelmic
337
nitrogen is unavailable for plant and microorganism use
while in place in the peat deposit. When exposed to the
atmosphere (e.g., as a garden amendment), the carbon is
oxidized to CO2 and the nitrogen is mineralized to NO3
and NHþ
4 and available for plant uptake. Although the
actual percent of nitrogen, and other nutrients, may not
be as high as that in inorganic soils, the total amount in the
soil within any one square meter surface area of the peat
land is greater in peat soils due to the depth of the peat
present. This oligotrophification, and consequently nutri
ent storage, is autogenetically enhanced through the
buildup of the peat column, placing the peat surface
farther from the source of the nutrient inputs. The long
term result of oligotrophication is the regional storage of
large pools of both carbon as well as important nutrients,
especially nitrogen and phosphorus.
Methane Production
Methane is a highly potent greenhouse gas that originates
from both natural and anthropogenic origins. On a weight
basis, methane is 21 times more efficient at trapping heat
and warming the planet than carbon dioxide. Methane
emissions from wetlands account for more than 75% of
the global emissions from all natural sources. Methane is a
highly reduced compound produced as the end product of
anaerobic decomposition by a group of microorganisms
called methanogens, which phylogenetically belong to
Archaea. These strict anaerobes can utilize only a limited
variety of substrates with H2–CO2 and acetate being
the most important too. The H2–CO2 dependent
methanogenesis is considered the dominant pathway of
methane production in boreal peatlands. However,
acetate dependent methanogenesis sometimes dominates
in fens. In rich fens, higher nutrient availability promotes
the growth of vascular plants (primarily sedges). Roots of
these vascular plants penetrate deep into the peat column
and therefore transport potential carbon rich substrates,
such as acetate, into the anaerobic layer. Rapid decom
position of organic matter also provides abundant
substrates for methanogens. Poor fens, with lower vascu
lar plant cover than that of rich fens, generally have
lower potentials for CH4, and a higher portion of the
produced CH4 comes from H2–CO2. Similar to poor
fens, Sphagnum dominated bogs also have a higher pro
portion of CH4 produced from H2–CO2, and it may be
that the dominance of mosses (without roots) and mycor
rhizal vascular plants (without deep carbon rich roots),
along with the reduced abundance of sedges with well
developed deep roots, prohibit movement of labile carbon
substrates to the anaerobic peat layer. Low decomposition
rates in acidic bogs also limit the amount of acetate that
can be produced during peat decomposition, which in
turn limits the acetoclastic pathway. Methanogen diver
sity in bogs is very low and the composition of the
338
Peatlands
methanogen community in bogs also differs greatly from
that characteristic of fens. In general, higher CH4 produc
tion is found in peatlands with higher vascular plant
cover, and higher water tables are found in rich fens.
Sulfate Reduction
In peatlands, sulfur occurs in several different redox states
(S valences ranging from þ6 in SO24 to –2 in hydrogen
sulfide (H2S), S containing amino acids, and other com
pounds), and conversions between these states are the
direct result of microbially mediated transformations. In
bogs, the sole sulfur input is via atmospheric deposition,
while in fens atmospheric deposition can be augmented by
surface and/or groundwater inputs, which may contain
sulfur derived from weathering of minerals in rock and
soil. Regardless of the sulfur source, when sulfur enters a
peatland, there are a variety of pathways through which it
can cycle. In the aerobic zone, sulfate can be adsorbed onto
soil particles, or assimilated by both plants and microbes. In
the anaerobic zone, sulfate can also be adsorbed onto soil
particles, assimilated by plants or microbes, or reduced by
sulfate reducing bacteria through the process of dissimila
tory sulfate reduction. Dissimilatory sulfate reduction is a
chemoheterotrophic process whereby bacteria in at least 19
different genera oxidize organic matter to meet their
energy requirements using sulfate as the terminal electron
acceptor. Thus, this process is one way in which carbon is
lost from the catotelm. If the sulfate is reduced by sulfate
reducing bacteria, the end product (S2 ) can have several
different fates. In the catotelm, where S2 is formed, it can
react with hydrogen, to produce H2S gas, which can diffuse
upwardly into or through the acrotelm where it can be
either oxidized to sulfate, or lost to the atmosphere.
Alternatively, H2S can react by nucleophilic attack with
organic matter to form organic or C bonded sulfur (CBS).
If Fe is present, S2 can react with Fe to form FeS and FeS2
(pyrite), which is referred to as reduced inorganic sulfur
(RIS). The RIS pool tends to be unstable in peat and can be
reoxidized aerobically with oxygen if the water table falls,
or anaerobically probably using Feþ
3 as an anaerobic elec
tron acceptor. If Hg is present, and combines with S2 to
form neutrally charged HgS, then Hg sulfide is capable of
passive diffusion across cell membranes of bacteria that
methylate Hg. Alternatively, bacteria can transfer the
methoxy groups of naturally occurring compounds, such
as syringic acid, to S2 , and form methyl sulfide (MeSH) or
dimethyl sulfide (DMS), although the exact mechanisms
by which this occurs are still unknown.
Initiation and Development of Peatlands
Peatlands initiate in one of four ways. The first, the most
common, appears to through paludification (or swamping),
wherein peat forms on previously drier, vegetated habitats
on inorganic soils and in the absence of a body of water,
generally due to regional water table rise and associated
climatic moderation. Additionally, local site factors also
have strong influences on paludification. Second, peat
may form directly on fresh, moist, nonvegetated mineral
soils. This primary peat formation occurs directly after
glacial retreat or on former inundated land that has risen
due to isostatic rebound. Third, shallow bodies of water
may gradually be filled in by vegetation that develops
floating and grounded mats – thus terrestrializing the for
mer aquatic habitat. Both lake chemistry and morphometry
as well as species of plants in the local area influence the
rates and vegetative succession. Fourth, peat may form and
be deposited on shallow basins once occupied by extinct
Early Holocene lakes. These former lake basins, lined with
vegetated impervious lake clays, provide hydrologically
suitable sites for subsequent peat development.
Across the boreal zone, peatland initiation appears to
be extremely sensitive to climatic controls. For example,
in oceanic areas, peatlands often initiated soon after gla
cial retreat some 10 000–12 000 years ago. Many of these
oceanic peatlands began as bogs and have maintained bog
vegetation throughout their entire development. In more
continental conditions, most peatlands were largely
initiated through paludification. In areas where the bed
rock is acidic, most of these early peatlands were poor
fens, whereas in areas where soils are base rich and alka
line, rich fens dominated the early stages. Like oceanic
peatlands, subcontinental peatlands initiated soon after
glacial retreat; however, throughout most of the large
expanses of boreal Canada and Siberia, peatland initiation
was delayed until after the Early Holocene dry period,
initiating 6000–7000 years ago. Many of these peatlands
initiated as rich or poor fens and have remained as fens for
their entire existence, whereas others have undergone
succession and today are truly ombrotrophic bogs.
A recent study in western Canada correlated peak times
of peatland initiation to Holocene climatic events that are
evident in US Midwest lakes, North Atlantic cold cycles,
and differing rates of peat accumulation in the one rich
fen studied in western Canada.
Peatlands as Carbon Sinks
Peat is about 51% carbon and peatlands hold about
270–370 Pg (petagram) of carbon or about one third of
the world’s soil carbon. For example in Alberta
(Canada), where peatlands cover about 21% of the
provincial landscape, the carbon in peatlands amounts
to 13.5 Pg compared to 0.8 Pg in agricultural soils,
2.3 Pg in lake sediments, and 2.7 Pg in the province’s
forests. Estimates for apparent long term carbon accu
mulation in oceanic, boreal, and subarctic peatlands
Polar Terrestrial Ecology
range from around 19 to 25 g C m 2 yr 2. However,
disturbances can have a dramatic effect on carbon
accumulation. Wildfire, peat extraction, dams and asso
ciated flooding, mining, oil and gas extraction, and
other disturbances all reduce the potential for peatlands
to sequester carbon, while only permafrost melting of
frost mounds in boreal peatlands has been documented
to have a positive effect on carbon sequestration. One
recent study has suggested that effects from disturbance
in Canada’s western boreal region have reduced the
regional carbon flux (amount of carbon sequestered in
the regional peatlands) from about 8940 Gg (gigagram)
C yr 1 under undisturbed conditions to 1319 Gg carbon
sequestered per year under the present disturbance
regime, yet only 13% of the peatlands have been
affected by recent disturbance. These data suggest
that although for the long term peatlands in the boreal
forest region have been a carbon sink and have been
removing carbon from the atmosphere, at the present
time, due to disturbance, this capacity is greatly dimin
ished. Furthermore, when disturbance is examined in
more detail, it is wildfire that is the single greatest
contributor to loss of carbon sequestration, both from
a direct loss as a result of the fire itself as well as from
a loss of carbon accumulation due to post fire recovery
losses. If wildfire greatly increases as is predicted by
climate change models, then the effectiveness of peat
lands to sequester carbon may be greatly reduced and it
has been proposed that an increase of only 17% in the
area burned annually could convert these peatlands to a
regional net source of carbon to the atmosphere. If
boreal peatlands become a source for atmospheric car
bon, then the carbon contained within the current
boreal peatland pool, in total, is approximately two
thirds of all the carbon in the atmosphere.
339
See also: Boreal Forest; Botanical Gardens; Chaparral.
Further Reading
Bauerochse A and Haßmann H (eds.) (2003) Peatlands: archaeological
sites archives of nature nature conservation wise use. Proceedings
of the Peatland Conference 2002 in Hanover, Germany, Hanover:
Verlag Marie Leidorf GmbH (Rahden/Westf.).
Davis RB and Anderson DS (1991) The Eccentric Bogs of Maine: A Rare
Wetland Type in the United States, Technical Bulletin 146. Orono:
Maine Agricultural Experiment Station.
Feehan J (1996) The Bogs of Ireland: An Introduction to the Natural,
Cultural and Industrial Heritage of Irish Peatlands. Dublin: Dublin
Environmental Institute.
Fraser LH and Kelly PA (eds.) (2005) The World’s Largest Wetlands:
Their Ecology and Conservation. Cambridge: Cambridge University
Press.
Gore AJP (1983) Ecosystems of the World. Mires Swamp, Bog, Fen
and Moor, 2 vols. Amsterdam: Elsevier Scientific.
Joosten H and Clarke D (2002) Wise Use of Mires and Peatlands
Background and Principles Including a Framework for Decision
Making. Jyvaskyla, Finland: International Mire Conservation Group
andInternational Peat Society (http://www.mirewiseuse.com).
Larsen JA (1982) The Ecology of the Northern Lowland Bogs and
Conifer Forests. New York: Academic Press.
Moore PD (ed.) (1984) European Mires. New York: Academic Press.
Moore PD and Bellamy DJ (1974) Peatlands. London: Elek Scientific.
National Wetlands Working Group (1988) Wetlands of Canada.
Ecological Land Classification Series, No. 24. Ottawa: Sustainable
Development Branch, Environment Canada, and Montreal:
Polyscience Publications.
Parkyn L, Stoneman RE, and Ingram HAP (1997) Conserving Peatlands.
NewYork: CAB International.
Vitt DH (2000) Peatlands: Ecosystems dominated by bryophytes.
In: Shaw AJ and Goffinet B (eds.) Bryophyte Biology, pp. 312 343.
Cambridge: Cambridge University Press.
Vitt DH (2006) Peatlands: Canada’s past and future carbon legacy.
In: Bhatti J, Lal R, Price M, and Apps MJ (eds.) Climate Change and
Carbon in Managed Forests, pp. 201 216. Boca Raton, FL: CRC
Press.
Wieder RK and Vitt DH (eds.) (2006) Boreal Peatland Ecosystems.
Berlin, Heidelburg, New York: Springer.
Wright HE, Jr., Coffin BA, and Aaseng NE (1992) The Patterned
Peatlands of Minnesota. Minneapolis: University of Minnesota Press.
Polar Terrestrial Ecology
T V Callaghan, Royal Swedish Academy of Sciences Abisko Scientific Research Station, Abisko, Sweden
ª 2008 Elsevier B.V. All rights reserved.
The Future: Polar Regions and Climate Change
Further Reading
The polar regions are situated at latitudes beyond which
the Earth’s angle to the Sun is shallow and the input of
thermal radiation is low. During the winter period, the Sun
is below the horizon and there are prolonged periods of
darkness. The resulting low temperature regimes domi
nate ecological processes, either directly by affecting
plant growth, microbial activity, animal behavior, organism
reproduction and survival, or indirectly by controlling the
length of the snow and ice free periods in which most
primary production and dependent biological activity
occurs, the availability of water in liquid form, and the
expansion and contraction, and other active layer proper
ties in generally primitive soils underlain by permafrost.
Feedback mechanisms from polar regions and their
340
Polar Terrestrial Ecology
ecological systems to the climate system affect local, regio
nal, and global climate. The balance between greenhouse
gas emissions from decomposition, particularly soil micro
bial respiration, and photosynthesis has resulted in a large
net accumulation of carbon in arctic soils while ice and
snow that cover low, tundra vegetation reflect incoming
radiation. Both mechanisms lead to cooling. In contrast,
global ocean circulation leads to the redistribution of the
Earth heat by cooling the tropics and warming the high
latitudes.
Both the Arctic and the Antarctic are characterized
by vast wilderness areas that are generally young, as
most land areas, with some extensive exceptions in the
Arctic, were glaciated in the Pleistocene. Polar regions
host some of the Earth’s most extreme environments
and organisms such as snow algae, lichens that inhabit
the crevices within crystalline rocks, and the simple
communities of soil fauna in the dry valleys of
Antarctica.
Polar environments vary between the Arctic and the
Antarctic and also within each region (Figures 1 and 2).
The Arctic is dominated by a polar ocean surrounded
by continental land masses and islands, whereas the
Antarctic is dominated by a polar, largely ice covered,
land mass surrounded by oceans. Terrestrial ecosystems
are extensive (7.5 million km2) and varied and stretch
from the closed canopy northern boreal forests in the
south, through the latitudinal treeline ecotone and tundra
wetlands to the polar desert in the north. Along this
latitudinal gradient, mean July temperature varies from
about 12 C in the south to 2 C in the north, total annual
precipitation varies from about 250 to 75 mm (mainly as
snow), and net primary production varies from about 1000
to 1 g m 2 yr 1. Approximately 6000 animal and 5800
plant species inhabit arctic lands (3 and 5% of global
biodiversity, respectively). Biodiversity decreases geome
trically along this gradient. Although plant biodiversity is
low in comparison with many biomes, it is surprisingly
high per square meter because of the small scale of plants,
and over 6000 species of animals and plants have been
cataloged in and around Svalbard at about 79 N. There is
also large environmental variation associated with the
climatic effects of northern ocean currents: in arctic
Norway, Sweden, and Finland, forests grow north of the
Arctic Circle (66.7 N) because of the warming effect of
the northward flowing Gulf Stream, whereas polar bears
and tundra vegetation are found at about 51 N in eastern
Canada because of the cooling influence of southward
flowing cold ocean currents. In the Arctic, indigenous and
other arctic peoples have been part of the ecosystem for
millennia.
The large land masses of the Arctic have great
connectivity with land masses further south: great
rivers flow from low latitudes to the Arctic Ocean,
and mammals and hundreds of millions of birds
migrate between the summer breeding grounds in
the Arctic and overwintering areas in boreal or tem
perate regions. Food chains in the Arctic are more
complex than those in the Antarctic and at the top
of the chain are mammalian carnivores such as the
polar bears, wolves, and arctic foxes. Population cycles
characteristic of arctic animals together with relatively
Polar
Polar desert,
desert, Cornwallis
Cornwallis Island,
Island, Canada
Canada
Boreal
Boreal forest,
forest, Sweden
Sweden
Polar
Polar semidesert,
semidesert, Svalbard
Svalbard
Shrub tundra and graminoid tundra, Alaska
Figure 1 Ecosystems of the Arctic.
Polar Terrestrial Ecology
341
The Future: Polar Regions and Climate
Change
Figure 2 Coastal ecosystem, sub-Antarctic South Georgia.
few species in each trophic level can result in ecolo
gical instability and ecological cascades: increasing
numbers of snow geese in arctic Canada have denuded
vegetation resulting in habitat hypersalinity.
The Antarctic land mass covers some 12.4 million
km2 but less than 1% is seasonally ice free. In the
Antarctic, the major environmental variation is asso
ciated with the relatively moist and ‘warm’ maritime
climate of the west coast of the Antarctic Peninsula
(temperatures are between 0 and 2 C for 2–4 months
in summer) contrasted with the cold, dry polar desert
climate of the continental land mass. Consequently,
most biological activity and most species are found
on the west coast of the Antarctic Peninsula.
Vegetation is dominated by relatively simple plant
communities of lichens, mosses, and liverworts that
support simple soil invertebrate communities. Only
two species of higher plant and higher insects occur.
Terrestrial mammals are absent and this short trophic
structure, together with the isolation of the land mass,
has enabled the establishment of a highly specialized,
and commonly endemic fauna of ground nesting birds
(e.g., penguins) and seals that depend on the coastal
land areas for breeding and moulting, and the sea, for
food. Nutrients for plant growth in these areas are
mainly derived from the sea and are deposited on
land by wind or birds. In contrast, over much of the
tundra, low nutrient availability to plants limits pri
mary production. There are no indigenous peoples in
the Antarctic and human activities there have been
restricted to the past 200 years.
Human activity has, until recently, influenced both
Arctic and Antarctic ecosystems less than most biomes
on Earth. However, the polar amplification of global
climate change together with the inherent sensitivity
of polar ecological systems to invasion by species from
warmer latitudes has resulted in the vulnerability of polar
ecosystems which are now under threat of rapid change.
The polar regions are undergoing rapid climate change.
There is a general amplification of global warming in the
Arctic: surface air temperatures have warmed at approxi
mately twice the global rate, although there are local
variations. The average warming north of 60 N has
been 1–2 C since a temperature minimum in the 1960s
and 1970s with the largest increase (c. 1 C per decade) in
winter and spring. Continental arctic land masses together
with the Antarctic Peninsula are the most rapidly warm
ing areas of the globe. Precipitation in the Arctic shows
trends of a small increase over the past century (about 1%
per decade), but the trends vary greatly from place to
place and measurements are very uncertain. There are
reductions in Arctic sea ice, river and lake ice in much of
the sub Arctic, and Arctic glaciers. Reduction in Arctic
sea ice has occurred at a rate of 8.9% per decade for
September relative to the 1979 values and there was an
un predicted extreme reduction in 2007. Permafrost has
warmed. Although changes in the active layer depth have
no general trend, in some sub Arctic locations, discontin
uous permafrost is rapidly disappearing and changes in
permafrost are driving changes in hydrology and ecosys
tems. In Arctic Russia, ponds are drying in the continuous
permafrost zone and waterlogging is occurring where
there is discontinuous permafrost.
In Antarctica, temperature trends show considerable
spatial variability: the Antarctic Peninsula shows signifi
cant warming over the last 50 years, whereas cooling has
occurred around the Amundsen Scott Station at the
South Pole and in the Dry Valleys. Consequently, there
is no continent wide polar amplification of global change
in Antarctica.
Current polar warming is leading to changes in spe
cies’ ranges and abundance and a northward and upward
extension of the sub Arctic treelines. Forest is projected
to displace considerable areas of tundra in some places.
Species tend to relocate, as they have in the past, rather
than adapt to new climate regimes. However, this process
is likely to lead to the loss of some species: polar bears and
other ice dependent organisms are particularly at threat.
In other areas, where rates of species relocation are slower
than climate change, the incidence of pests, disease, and
fire is likely to increase. Changes in vegetation, particu
larly a transition from grasses to shrubs, have been
reported in the North American Arctic, and satellite
imagery has indicated an increase in the ‘normalized
difference vegetation index’ (a measure of photosyntheti
cally active biomass) over much of the Arctic. This index
has increased by an average of about 10% for all tundra
regions of North America, probably because of a longer
growing season. However, such increases in productivity
342
Riparian Wetlands
and changes in plant functional types have been shown
experimentally to displace mosses and lichens that are
now major components of Arctic vegetation.
In Antarctica, warming has caused major regional
changes in terrestrial and marine ecosystems. The abun
dances of krill, Adelie, and Emperor penguins and
Weddell seals have declined but the abundances of the
only two native higher plants has increased. On continen
tal Antarctica, climate change is affecting the vegetation
composed of algae, lichens, and mosses. Introductions of
alien species, facilitated by increased warming and
increased human activity, are particular threats to south
ern ecosystems. Recent studies on sub Antarctic islands
have shown increases in the abundance of alien species
and negative impacts on the local biota. In contrast,
cooling has caused clear local impacts in the Dry
Valleys where a 6–9% reduction in lake primary produc
tion and a 10% per year decline in soil invertebrates has
occurred.
The responses of polar environments to climatic
warming include feedbacks to the global climate system
and other global impacts. Increased runoff from arctic
rivers could affect the thermohaline circulation that redis
tributes the Earth’s heat, thereby causing cooling in the
North Atlantic and further warming in the tropics.
Reductions in sea ice extent and snow cover together
with a shift in vegetation from tundra to shrubs or forests
are likely to reduce albedo (reflectivity of the surface) and
lead to further warming despite the increased uptake of
carbon dioxide by a more productive vegetation.
Thawing permafrost is likely to release methane, a
particularly powerful greenhouse gas, and evidence of
this is already available from various arctic areas.
Not all impacts of climate warming in polar regions are
disadvantageous to society: the reduction of sea ice in the
Arctic is likely to lead to increased marine access to
resources and new fisheries and reduced length of sea
routes, while warming on land will probably lead to
increased productivity and increased potential for for
estry and agriculture.
See also: Alpine Ecosystems and the High-Elevation
Treeline; Biological Wastewater Treatment Systems.
Further Reading
Anisimor OA, Vaughan DG, Callaghan TV, et al. (2007) Polar regions
(Arctic and Antarctic). In: Parry ML, Canziani OF, Palutikof JP,
Hanson CE, and Van der Linder PJ (eds.) Climate Change 2007:
Impacts, Adaptation and Vulnerability. Contribution of Working
Group II to the Fourth Assessment Report of the Intergovernmental
Panel on Climate Change, pp. 655 685. Cambridge: Cambridge
University Press.
Callaghan TV, Bjorn LO, Chapin FS, III, et al. (2005) Tundra and polar
desert ecosystems. In: ACIA. Arctic Climate Impacts Assessment,
pp. 243 352. Cambridge: Cambridge University Press.
Chapin FS, III, Berman M, Callaghan TV, et al. (2005) Polar ecosystems.
In: Hassan R, Scholes R, and Ash N (eds.) Ecosystems and Human
Well Being: Current State and Trends, vol. 1, pp. 719 743.
Washington, DC: Island Press.
Convey P (2001) Antarctic ecosystems. In: Levin SA (ed.) Encyclopaedia
of Biodiversity, vol. 1, pp. 171 184. San Diego: Academic Press.
Nutall M and Callaghan TV (2000) The Arctic: Environment, People,
Policy, 647pp. Reading: Harwood Academic Publishers.
Richter Menge J, Overland J, Hanna E, et al. (2007) State of the Arctic
Report.
Walther GR, Post E, Convey P, et al. (2002) Ecological responses to
recent climate change. Nature 416(6879): 389 395.
Riparian Wetlands
K M Wantzen, University of Konstanz, Konstanz, Germany
W J Junk, Max Planck Institute for Limnology, Plön, Germany
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Definitions and Concepts
Environmental Conditions Determining Riparian
Wetlands
Types of Riparian Wetlands
Typical Biota and Biodiversity in Riparian Wetlands
Ecological Services of Riparian Wetlands
Conservation
Further Reading
Introduction
occur in this zone exchange water with the aquifer and
with the main channel during flood events (Figure 1).
Riparian wetlands are buffer zones for the water budget of
the landscape: they take up excess water from flood events
and release it gradually afterwards.
The riparian zone of running water systems is a site of
intensive ecological interactions between the aquatic and
the terrestrial parts of the stream valley. Wetlands that
Riparian Wetlands
343
Figure 1 Inputs, turnover, and exchange of organic matter in the stream channel (left) and a riparian wetland water body (center) at
low and high water levels. Black arrows indicate organic matter inputs, white arrows indicate water exchange pathways, spirals indicate
nutrient spiralling or downriver transport, and circular arrows indicate sites of organic matter turnover in situ. Curly brace indicates
water-level fluctuations during flood events. Modified from Wantzen KM, Yule C, Tockner K, and Junk WJ (2006) Riparian wetlands.
In: Dudgeon D (ed.) Tropical Stream Ecology, pp. 199–217. Amsterdam: Elsevier.
Modern ecological theory recognizes the important role
riparian wetlands play for biodiversity and for the energy and
matter budgets along the whole range of river courses. The
carbon and nutrient budgets are influenced by dissolved and
particulate substances from the bordering terrestrial ecosys
tems, by the autochthonous production from the wetland
plants, and by allochthonous organic matter delivered by
the floodwater. The proportions between these sources are
defined by the hydrological patterns, landscape morphology,
and climatic conditions. (see Rivers and Streams: Physical
Setting and Adapted Biota and Rivers and Streams:
Ecosystem Dynamics and Integrating Paradigms).
The crossover between humid and dry conditions
creates habitats for organisms coming from either aquatic
or terrestrial ecosystems, and for those biota that are
specialized on wetland conditions. As the transversal
dimension of streamside wetlands is generally small,
their overall importance for landscape ecology, biogeo
chemistry, and biodiversity is often overlooked. However,
the total size of these wetlands can be considerable in
areas with dense stream networks. Moreover, the corri
dor shaped extension of riparian wetlands makes them
perfect pathways for the gene flow between remote popu
lations of aquatic and terrestrial biota. Many ecological
services are uniquely provided by riparian wetlands,
including erosion control, filtering of nutrients and pesti
cides from adjacent cropland, mitigation of floods, and
recreation, which increases their conservation value in a
socioeconomic context.
There is a large array of environmental conditions that
vary between the different types of riparian wetlands, espe
cially climatic region and prevailing vegetation type, and
landscape morphology and hydrologic patterns. This article
deals with the different types of riparian wetlands, their
deterministic environmental conditions, prevailing ecologi
cal processes, typical biota, and aspects of conservation.
Definitions and Concepts
There are many definitions of riparian wetlands. A hydro
logical definition defines riparian wetlands as
lowland terrestrial ecotones which derive their high water
tables and alluvial soils from drainage and erosion of
adjacent uplands on the one side or from periodic flood
ing from aquatic ecosystems on the other (McCormick,
1979)
A functional definition states that
riparian areas are three dimensional ecotones of interac
tion that include terrestrial and aquatic ecosystems, that
extend down to the groundwater, up above the canopy,
outward across the floodplain, up the near slopes that
drain to the water, laterally into the terrestrial ecosystem,
and along the water course at a variable with (Ilhard et al.,
2000).
Both definitions point to the ecotonal character of
riparian wetlands between water bodies on one side and
the upland on the other. Riparian wetlands can be, at the
smallest scale, the immediate water’s edge where some
aquatic plants and animals form a distinct community,
and pass to periodically flooded areas of a few tens of
meters width. At medium scale they form bands of vege
tation, and at the largest scale they form extended
floodplains of tens of kilometers width along large rivers.
344
Riparian Wetlands
In this case, complexity of the riparian wetlands increases
so much that many scientists give them the status of
specific ecosystems (see Floodplains).
There are several concepts that deal with different aspects
of stream and river ecology but two of them are of specific
interest to rivers and riparian zones (see Rivers and Streams:
Ecosystem Dynamics and Integrating Paradigms). The ‘river
continuum concept’ (RCC) of Vannote et al., describes the
longitudinal processes in the river channel and the impact of
the riparian vegetation on the physical and chemical condi
tions and as carbon source to the aquatic communities in the
channel. The ‘flood pulse concept’ (FPC) of Junk et al. stresses
the lateral interaction between the floodplain and the river
channel and describes the specific physical, chemical, and
biological processes and plant and animal communities inside
the floodplain. The predictions of the RCC fit well for rivers
with narrow riparian zones but with increasing lateral extent
and complexity of the riparian zone the FPC becomes more
important. Here, we restrict our discussion to riparian wet
lands along streams and low order rivers. Since lateral extent
of the riparian zone along low order rivers can vary consid
erably in different parts of the same river or between different
rivers of the same river order, the applicability of the con
cepts may also vary.
Environmental Conditions Determining
Riparian Wetlands
Riparian habitats are integral parts of a larger landscape
and therefore influenced by factors operating at various
special and temporal scales. The physical setting that
determines rivers and streams basically defines the ripar
ian wetlands (see Rivers and Streams: Ecosystem
Dynamics and Integrating Paradigms); however, some
environmental features have specific importance on the
wetlands that will be dealt with in the following.
Spatial and Temporal Scales
At the regional scale, geomorphology, climate, and vege
tation affect channel morphology, sediment input, stream
hydrology, and nutrient inputs. At the local scale, land use
and related alteration to stream habitats, but also the
activity of bioengineers such as beavers, can be of signifi
cant influence. At short timescales, individual heavy
rainfall events affect the riparian systems; at an annual
basis climate induced changes in light, temperature, and
precipitation trigger important cyclic biological events,
such as autochthonous primary and secondary produc
tion, litterfall, decomposition, and spawning and hatching
of animals. On multiannual timescales, extreme flood and
drought events, debris torrents, landslides, heavy storms
or fire can have dramatic effects on the riparian zone and
its biota.
Climatic Region
Climate controls the availability of the water in the wet
lands and the activity period of the organisms. If the
flooding and activity periods match, the floodborne
resources can be used by the adapted floodplain biota
(e.g., during summer floods). On the other hand, winter
floods are generally less deleterious for little flood
adapted tree species.
In the boreal and temperate regions, freezing and
drought in winter and snowmelt floods in spring are
predictable drivers of the interplay between surface
water and groundwater in riparian wetland hydrology.
Ice jams may cause stochastic flood events in winter.
Normally, stream runoff is reduced during winter, and
groundwater fed riparian wetlands discharge into the
stream channel as long as possible. In wetlands with
organic sediments, this water is often loaded with large
amounts of dissolved organic carbon. In shallow streams
that freeze completely during winter, riparian wetlands
may serve as refuges for the aquatic fauna, for example,
for amphibians and turtles. Spring snowmelt events
generally provoke prolonged flood events that exceed
the duration of rain driven floods. These long floods
can connect the riparian wetland water bodies to the
stream, so that organic matter and biota become
exchanged. At the same time, there is often an infiltra
tion (downwelling) of surface water into the riparian
groundwater body.
In
seasonal
wet and dry
climates
(both
Mediterranean and tropical savanna climates) water
supply by rainfall is limited to a period of several
months during which very strong rainstorms may
occur. These events, albeit short, are of great impor
tance for the release of dissolved substances and for
the exchange of organic substances and biota between
wetland and main water course. Moreover, energy
rich organic matter (e.g., fruits) may become flushed
from the terrestrial parts of the catchment into ripar
ian wetlands. On the other hand, flash floods can
cause scouring and erosion of fine sediments (includ
ing organic matter). During the dry season,
groundwater levels are lower and may cause a seaso
nal drought in the riparian wetlands. In these periods,
the aquatic biota either estivate or migrate into the
permanent water bodies, and large parts of the stocked
organic matter become mineralized. However, even in
strongly seasonal zones, like the Brazilian Cerrado,
groundwater supply may be large enough to support
permanent deposition of undecomposed organic
matter.
The distribution of water conductive (coarse) and
impermeable substrate (bedrock and loam) of the valley
bottom influences the thickness of the stagnant water
body in the riparian zone and thus the extension of
Riparian Wetlands
organic matter layers. Permanently humid conditions
are found in many riparian wetlands of the boreal
zone and in the humid tropics. These permanent ripar
ian wetlands can accumulate large amounts of organic
carbon. In tropical Southeast Asia (Malayan Peninsula
and parts of Borneo), a special case of riparian wetland
occurs, the peat swamps. These swamps develop when
mangrove forests proceed seawards, and the hinterland
soils lose their salt content. Here, large amounts of
organic matter from the trees become deposited and
the streams flow within these accumulations (see
Peatlands).
Valley Size, Morphology, and Connectivity
The common textbook pattern of steep valleys in the
upper sections of the streams and open, shallow floodplains
in the lower river sections holds true only for very few
cases in nature. Rather, we find these two valley types
interspersed in an alternating pattern like ‘beads on a
string’. Shallow areas are more likely to bear extended
riparian wetlands; however, if groundwater levels are
high enough, even steep valleys may be covered with
wetlands. The morphology of riparian wetlands can be
described by the entrenchment ratio (i.e., the ratio of valley
width at 50 years flood level to stream width at bankfull
level) or by the belt width ratio, that is the distance
between opposing meander bends over a stream section
to stream width at bankfull level. Fifty years flood often
intersect the terrace slope.
Riparian wetlands of different catchments may be
linked with each other through swamp areas (e.g., in old
eroded landscapes of the Brazilian and Guyana Shields in
South America) so that biogeographical barriers can be
overcome by aquatic biota even without a permanent
connection between the water courses. The term connec
tivity describes the degree by which a floodplain water
body is linked to the main channel. Riparian wetlands
may also be connected to the stream, either in a direct
connection by a short channel, or indirectly by a longer
channel which may be intercepted by a pond. In some
cases, these channels can be cryptic/hidden when they
are formed by macropores in the organic soils. Alluvial
riparian wetlands may be connected to the stream via the
hyporheic interstitial zone provided that the sediments
are coarse enough to conduct water. Wetlands without
any of these pathways exchange water, biota, and organic
matter with the main channel during overbank flow of the
stream. Purely aquatic organisms depend on the existence
of connection channels to migrate between wetland and
main water body. For example, amphibia are especially
sensitive to fish predation, so that the highest biodiversity
of amphibia is found at riparian wetland habitats with the
lowest accessibility for fish.
345
Hydrology and Substrate Type
The slope of the landscape and the rock characteristics of
the catchment define the physical habitat characteristics
of the stream–wetland system. Riparian wetlands provide
habitats with different hydraulic and substrate conditions
than the stream channel. Although flooding in streams is
generally shorter, less predictable and ‘spikier’ than in
large rivers, there is a large number of exchange processes
between the main channel and the riparian zone during
these flood events. Major flood events, albeit rare, act as
‘reset mechanism’ in the floodplain that rejuvenates the
sediment structure and the successional stage of the vege
tation. Between these rare events, riparian wetlands act as
sinks for fine particles and organic sediments that were
washed out of the stream channel, the terrestrial zone of
the catchment, or derive from an autochthonous biomass
production.
Vegetation
Vegetation bordering to and growing within riparian wet
lands fulfils many functions: it delivers both substrate for
colonization and food resources for aquatic animals, it
strips nutrients from the incoming water, and it provides
raw material for the organic soils. It retards nutrient loss,
filters nutrient input from the upland, reduces runoff by
evapotranspiration, and buffers water level fluctuations.
Shading by tree canopies reduces light conditions for
algal and macrophyte primary production and it equili
brates soil temperatures. Therefore, riparian wetlands
differ completely according to their vegetation cover.
Unvegetated riparian wetlands occur at sites where
establishment of higher plants is hampered by strong
sediment movement (e.g., high gradient and braided riv
ers), low temperatures (high elevation and polar zones),
rocky surfaces, or periodical drought (desert rivers). The
lack of shading and nutrient competition by higher plants
favors growth of algae on the inorganic sediments, and
productivity may be high, at least periodically.
High altitudes and/or elevated groundwater levels
may preclude tree growth but allow the development of
grass or herbal vegetation on riparian wetlands. Hillside
swamp springs (helokrenes) can coalesce and form exten
sive marshes far above the flood level of the stream
channel, so that the distinction between ‘riparian’ and
‘common’ wetland is difficult.
The tree species of forested riparian wetland are
adapted to periodical or permanent waterlogging of the
soils. They contribute an important input of organic car
bon to the stream system. Large tree logs shape habitat
structure by controlling flow and routing of water and
sediment between stream channel and wetland. Tree
roots increase sediment stability, sequester nutrients,
and form habitats.
346
Riparian Wetlands
Types of Riparian Wetlands
Rockpools
Riparian wetlands are very variable in size and environ
mental characteristics. In the following, we list the most
common types according to their hydrological and sub
strate characteristics (Figures 2 and 3).
Hygropetric Zone
At sites where groundwater outflows run over rocky
surfaces, hygropetric zones develop. In the thin water
film, there is a vivid algal production and a diverse,
however, less studied fauna of invertebrates (mostly
aquatic moths, chironomids, and other dipterans). Biota
of the hygropetric zone need to be adapted to harsh
environmental conditions such as periodical freezing
and drying of the surfaces.
Many streams run through bedrock or large boulders
which have slots that fill with flood or rain water. Biota
colonizing these pools have to be adapted to relatively
short filling periods, high water temperatures, and
solar radiation. High algal production and low predator
pressure (at least at the beginning of the filling period)
attract many invertebrate grazers.
Parafluvial and Orthofluvial Ponds
In alluvial stream floodplains, permanent or temporary
ponds develop from riverine dynamics either within
the active channel (parafluvial pond) or in the riparian
zone (orthofluvial pond). They are fed by both surface
water and groundwater. In coarse grained sediments,
Hygropetric zone
Rockpools
Bedrock
pool
Boulder
pool
Stream
Stream
Riparian flood zone
Active alluvium
Old alluvium
Orthofluvial pond
Parafluvial pond
High water level
Stream
Swamps and hillside wetlands
Hillside
wetlands
Valley bottom
macrophyte
swamps
Logjam and beaver ponds
(aerial view)
Valley
bottom
swamp
forest
Beaver pond
Stream
Logjam or
beaver dam
Water
Figure 2 Types of riparian wetlands.
Organic matter
accumulation
Permeable soil
Aquiclude
Riparian Wetlands
(a)
347
at either side of the stream channel. In fine grained
sediments (including organic soils), the contribution of
groundwater is much more important, and these ponds
are often brownish from dissolved organic matter (humic
acids and yellow substances). Para and orthofluvial
ponds contribute disproportionately to total species
richness along riparian corridors.
Riparian Flood Zones
(b)
Even if no basin like structures are present, flooding
events create wetted zones on either side of the stream,
independent of sediment type. Extension and perma
nence of the wetted zone depends on the valley shape,
the porosity of the sediments, and eventual backflooding
from tributary streams. In temporarily flooded forests
with thick organic layers and in stranded debris dams,
the moisture conditions may be long enough to bridge
the gap between two flood events, so that many aquatic
biota such as chironomids and other midges can com
plete their larval development in these semiaquatic
habitats.
Riparian Valley Swamps
(c)
Swamps occur on soils that are waterlogged for most
of the year. The lack of oxygen in the sediments allows
accumulation of organic matter and selects for tree
or herb species that have specific adaptations to these
conditions, for example, pressure ventilation in the roots.
The vegetation consists of either macrophytes or trees.
Due to the shading and oxygen consumption during
decomposition of organic matter, some of these riparian
wetlands are hostile environments for aquatic metazoa
that depend on dissolved oxygen. Some trees such as the
Australian gum (Melaleuca sp.) shed bark which release
secondary compounds that influence biota.
Hillside Wetlands
Figure 3 Photographs of riparian wetlands (Tenente Amaral
Stream, Mato Grosso, Brazil): (a) Stream channel with
hygropetric zone (foreground) and floodplain forest
(background), (b) Rockpool carved into the sandstone bedrock,
(c) moist organic soil colonized by many aquatic invertebrate
taxa. Leaf litter was removed. All photographs by K. M. Wantzen.
these ponds are connected to the main channel by the
hyporheic interstitial zone, that is, an ecotone between
groundwater and surface water that extends below and
In areas where the aquiclude extends laterally from the
stream, the riparian swamps can merge into hillside
wetlands far above the flood level. Given that waterlogged
ness is permanently provided, these ecosystems tend to
develop black organic soil layers from undecomposed
plant material. The anoxic conditions in these soils favor
denitrification and nitrogen may become a limiting factor
for plant growth. Carnivorous plants (Droseraceae,
Lentibulariacea, Sarraceniaceae) that replenish their nitro
gen budget with animal protein are commonly found in
these habitats. At sites where drainage is better, woody
plants invade these natural meadows. The soft texture of
the soils and their position in hill slope gradients makes
these ecosystems highly vulnerable to gully erosion.
348
Riparian Wetlands
Logjam Ponds and Beaver Ponds
Falling riparian trees are stochastic events which may have
dramatic consequences for the hydraulics of a stream sys
tem. Many tree species are soft wooded, and tree dynamics
are generally high in riparian wetlands. A fallen log blocks
the current and creates a dam that accumulates fine parti
cles. These natural reservoirs often extend far into the
riparian zone.
Dams built by beaver (Castor sp.) can significantly alter
the hydrological and biogeochemical characteristics of
entire headwater drainage networks in Northern
America and Eurasia. Fur trade led to the regional extinc
tion of beavers. Few decades after reintroduction of
beavers on a peninsula in Minnesota, they converted a
large part of the area into wetlands, which led to a mani
fold increase in the soil nutrient concentrations. The
activity of beavers considerably enhances the biodiversity
of wetland depending species. Beavers increase regional
habitat heterogeneity because they regularly abandon
impounded areas when the food supply is exhausted and
colonize new ones, thereby creating a shifting mosaic of
patches in variable stages of plant succession.
Typical Biota and Biodiversity in Riparian
Wetlands
The importance of riparian wetland habitats for the con
servation of biodiversity is well documented for several
watersheds. Riparian areas generally have more water
available to plants and animals than adjacent uplands.
This is of specific importance in regions with a pro
nounced dry season, where lack of water affects plant
growth. Abundance and richness of plant and animal
species tend to be greater than in adjacent uplands
because they share characteristics with the adjacent
upland and aquatic ecosystems and harbor a set of specific
riparian species. Because of their richness and their spatial
distribution, the relative contribution of riparian ecosys
tems to total compositional diversity far exceeds the
proportion of the landscape they occupy.
Apart from beavers, several other biota act as ‘ecologi
cal engineers’ that create and modify riparian wetlands.
African hippopotamus deepen pools and form trails that
increase the ponding of the water. Several crocodilians
maintain open water channels. Digging mammals, fresh
water crabs, and insects like mole crickets increase
the pore space in riparian soils and enhance the water
exchange between wetland and stream channel. Similar
macropores develop from fouling tree roots. Plants
also strongly modify the habitat characteristics in riparian
wetlands, either actively, by influencing soil, moisture,
and light conditions or, passively, by changing the
hydraulic conditions through tree fall or organic debris
dams.
Typical wetland species are adapted to the amphibious
characteristics of the habitats. They are either permanent
wetland dwellers that cope with aquatic and dry conditions
or they temporarily colonize the wetlands during either the
dry or the wet phase. There are many animal species that
permanently colonize riparian wetlands, especially anurans,
snakes, turtles, racoons, otters, and many smaller mammals,
like muskrats, voles, and shrews. Aquatic insects have devel
oped special adaptations to survive periodical droughts, for
example, by having short larval periods or drought resis
tance. Many birds profit by the rich food offered from the
aquatic habitats like dippers, kingfishers, jacamars, warblers,
and rails. Periodical colonizers from terrestrial ecosystems
are bats, elks, moose, and several carnivorous mammals and
birds. Many aquatic species like fish and aquatic inverte
brates periodically colonize riparian wetlands. Riparian
wetland biota belong to the most threatened species as
they suffer from both the impacts on the terrestrial and
aquatic systems, and many riparian species are threatened
with extinction. The effects of extinction of a species
are especially high if it is an ecological engineer or a key
stone species, for example, a top predator. Extinction of
wolves in the Yellowstone National Park in the US led to
overbrowsing of broad leaved riparian trees by increased
elk populations.
Ecological Services of Riparian Wetlands
Riparian wetlands are intrinsically linked to both the
stream and the surrounding terrestrial ecosystems of
the catchment. In many places of the world, however,
riparian zones have remained the only remnants of
both wetland and woody habitats available for wildlife.
They are surrounded by intensively used areas for either
agriculture or urban colonization. The performance of
riparian wetlands to provide ecological services becomes
reduced by the same degree as these bordering ecosys
tems become degraded. However, even in degraded
landscapes, the beneficial effects of the riparian wetland
ecosystems are astonishingly high. For humans, healthy
riparian wetlands are vital as filters and nutrient attenua
tors to protect water quality for drinking, fisheries, and
recreation.
Nutrient Buffering
Riparian wetlands are natural traps for fine sediments and
for organic matter, but they may vary from a nutrient sink
to a nutrient source at different times of a year depending
on high or low water levels. Particle bound nutrients, such
as orthophosphate ions, become deposited in the riparian
Riparian Wetlands
wetlands during spates and may accumulate there. This
may increase the amount of phosphate that becomes
released during the following flood event. Therefore, tech
nical plans for phosphorus retention in artificial wetlands in
agricultural landscapes include a hydraulic design which
hampers the release of particles from the wetland, for
example, by providing continuous, and sufficiently broad
wetland buffer strips along the streams.
For the removal of nitrogen inputs from floodwater and
from lateral groundwater inputs, riparian wetlands are very
efficient. Generally it can be taken for granted that the
slower the water flow (both ground and surface water) the
higher is the nitrate uptake rate; however, the precise
flow pathways in the sediments have to be considered. In
anoxic soils, reduction and denitrification processes trans
form inorganic nitrogen forms into nitrogen gas which is
then released into the atmosphere. Once the nitrate has
been completely reduced, sulphate is also reduced in the
anoxic sediments. Nitrogen also becomes immobilized by
bacterial growth and/or condensation of cleaved phenolics
during the aerobic decay of organic matter. Aquatic macro
phytes and trees growing in the riparian wetlands are very
efficient in nitrogen stripping by incorporating mineral
nitrogen forms into their biomass. They can represent the
most important nitrogen sinks in riparian systems. Some
riparian wetland plants (e.g., alder, Alnus sp., and several
leguminous trees) have symbiotic bacteria associated to
their roots that can fix atmospheric nitrogen when this
nutrient is scarce in the soils. Thus, not all riparian wet
lands exclusively remove nitrogen.
Carbon Cycle
Like other wetlands, riparian wetlands are important
players in the carbon cycle of the watershed. They accu
mulate large amounts of coarse particulate organic matter
(CPOM) and they release dissolved organic matter into
the stream and gaseous carbon compounds into the atmo
sphere (Figure 1).
In the boreal zone, the spring snowmelt runoff contri
butes to more than half of the annual total organic carbon
(TOC) export. The larger the riparian wetland zone, the
bigger the amount of exported TOC. On the other hand,
riparian wetlands receive large amounts of dissolved car
bon from litter leachates from the surrounding forests,
especially during the leaf fall period. These leachates
can be an important source for phosphorous and other
nutrients, as well as for labile carbon compounds. These
substances enhance heterotrophic microbial (bacterial and
fungal) activity.
Spring snowmelt also carries large amounts of
fine particulate organic matter (POM). Riparian wet
lands often provide surface structures that act like a
comb to accumulate these particles (e.g., macrophytes),
349
and enhance the production of detritivores. Additional
POM is produced by riparian trees. The general
trend for litter production to increase with decreasing
latitude (valid in forests) is overlain by species specific
productivity and physiological constraints due to the
waterloggedness in riparian wetlands. Here, the litter
production is generally higher in periodically flooded,
than in permanently flooded, wetlands. Depending on
the oxygen content of the soils, the chemical composi
tion of the leaves, and the activity of detritivores, more
or less dense layers of ‘leaf peat’ can accumulate in the
sediments. This organic matter stock can be increased
by undecomposed tree logs and bark. A reduction of the
water level in the riparian wetlands leads to an
increased mineralization of the carbon stocks and
enhances the release of carbon dioxide.
Hydrological Buffering and Local Climate
Riparian wetlands have an equilibrating effect on hydro
logical budgets. Riparian vegetation dissipates the kinetic
energy of surface flows during spates. Riparian wetlands
store stormwater and release it gradually to the stream
channel or to the aquifer between rainstorm events.
Moreover, they are important recharge areas for aquifers.
Several current restoration programmes try to increase
this recharge function of riparian wetlands in order to
stabilize the groundwater stocks for drinking water
purposes.
Riparian wetland trees and macrophytes contribute
considerably to evapotranspiration and to local and regio
nal climate conditions. The rate of vapor release depends
on the plant functional group which needs to be consid
ered for basin scale water budgets.
Corridor Function for Migrating Species
Riverine wetlands represent a web of ecological corridors
and stepstones. In intense agricultural areas they can be
considered as ‘green veins’ that maintain contact and gene
flow between isolated forested patches. Providing shadow,
balanced air temperatures and moisture, shelter, resting
places, food and water supply, they cover the requirements
of a great deal of amphibian, reptile, bird, and mammal
species. These not only use the longitudinal connection but
also migrate laterally and thus reach the next corridor
aside. Moreover, long range migrating birds use the
green corridors of riparian zones in general as landmarks
for migration. Networks of riparian corridors also facilitate
the movement of non native species. In some US riparian
zones, their richness was about one third greater in riparian
zones than on uplands and the mean number and the cover
of non native plant species were more than 50% greater
than in uplands.
350
Riparian Wetlands
Refugia and Feeding Ground for Riverine Biota
During flood, drought, and freezing events, but also dur
ing pollution accidents in the stream channel, connected
riparian wetland habitats represent refugia for riverine
animals. In extreme cases, residual populations from the
wetlands may contribute to the recolonization of defau
nated stream reaches. Riparian wetlands also act as traps
and storage sites for seeds both from the upstream and
from the uphill areas. The seed banks contain propagules
from plants that represent a large range of moisture tol
erances, life spans, and growth forms. These seeds may
also become mobilized and transported during spate
events.
Riparian wetlands offer a large variety of food sources.
Connected wetland water bodies ‘comb out’ fine organic
particles including drifting algae from the stream water,
they receive aerial and lateral inputs of the vegetation, and
they have a proper primary productivity which profits by
the increased nutrient input and storage from the sur
roundings. Many riverine fish and invertebrate species
are known to migrate actively into the riparian wetlands
in order to profit by the terrestrial resources that are
available during flood periods. In analogy to the ‘floodpulse
advantage’ of fish in large river floodplains, stream biota
that temporarily colonize riparian wetlands have better
growth conditions than those that remain permanently in
the stream channel. For example, the macroinvertebrate
community of riparian sedge meadows in Maine (USA) is
dominated by detritivorous mayfly larvae (over 80% of
the invertebrate biomass) during a 2 month period in
spring. The larvae use the stream channel as a refuge and
use the riparian wetland as feeding ground where they
perform over 80% of their growth.
Reciprocal Subsidies between Aquatic and
Terrestrial Ecosystems
Many aquatic species profit by the terrestrial production
and vice versa. Apart from leaf litter, large quantities of
fruits, flowers, seeds, as well as insects and feces fall from
the tree canopies into the streams where they represent
important energy and nutrient sources for the biota. In
Amazonian low order rainforest streams, terrestrial
invertebrates make up a major portion of the gut content
of most fish species. Fruits and seeds are preferred food
items for larger fish species that colonize medium and
high order rivers. Riparian wetlands increase the area of
this active exchange zone, and they retain these energy
rich resources for a longer period than a stream bank
alone would do.
Aquatic organisms also contribute to the terrestrial
food webs. For example, bats are known to forage on the
secondary production of emerging insects in riparian wet
lands, and the shoreline harbors a large number of
terrestrial predators, such as spiders, tiger beetles, and
riparian lizards. Experimental interruption of these lin
kages (e.g., by covering whole streams with greenhouses)
has shown that the alteration of riparian habitats may
reduce the energy transfer between the channel and the
riparian zone.
Recreation
The sound of the nearby stream, the equilibrated cli
mate, and the occurrence of attractive animal and
plant species render riparian wetlands highly attractive
for recreation purposes such as hiking, bird watching,
or meditation. These can be combined with ‘in channel’
recreation activities such as canoeing, rafting, or fishing,
and represent an economically valuable ecosystem
service, that should be considered in management and
conservation plans.
Conservation
Water is becoming scarce in many areas worldwide. Water
mining reduces water levels, but high and stable ground
water tables are a prerequisite for the existence of riparian
wetlands. In addition to direct water withdrawal, predic
tions about climatic changes include other threats.
Increased stochasticity of the runoff patterns and reduced
snowmelt floods are severe threats to the existence of
riparian wetlands. The riparian zones of streams and rivers
have been sought after by humans since early days. High
productivity, reliable water supply, and climatic stability
make these ecosystems suitable for a range of human use
types, such as wood extraction, hunting, aquaculture, and
agriculture. In areas of intensive agriculture, riparian zones
including their wetlands have shrunk to narrow strips or
have completely vanished. On the other hand, the ecosys
tem services are good socioeconomical arguments to
restore and enlarge riparian wetlands.
For conservation planning, it is very important to
bear in mind that riparian wetlands are very diverse
and have typical regional characteristics. Secondly, the
whole riparian zone is very dynamic. Many tree spe
cies are relatively short lived and well adapted to
changes in the floodplain morphology or in the
hydrology of the wetland. The existence of variable
hydrological patterns is a prerequisite for the coexis
tence of annually varying plant and animal
communities. Often, large scale projects restore ripar
ian zones including wetlands according to a single
pattern that does not consider these dynamic changes
in habitat and species diversity. If large flood events
are precluded by dam constructions in the upstream
region, the natural habitat dynamics are blocked and
the vegetation will develop towards a late successional
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
stage without pioneer vegetation, and with a reduced
range of moisture tolerance. Several studies could
prove that once the hydrological fluctuations become
reduced by water level regulation, exotic species can
invade river valleys more efficiently.
While many animal species depend exclusively on
the specific habitat conditions of wetlands, most riparian
amphibians and reptiles migrate into the drier zones
of the aquatic–terrestrial ecotones for a part of their life
cycle. This makes them vulnerable to increased mortality
in the neighboring ecosystems, especially if these have
been converted into agricultural or urban use. Therefore,
a buffer zone considering the home range of these species
is needed to fully protect these species.
See also: Floodplains; Rivers and Streams: Ecosystem
Dynamics and Integrating Paradigms; Rivers and
Streams: Physical Setting and Adapted Biota.
Further Reading
Ilhardt BL, Verry ES, and Palik BJ (2000) Defining riparian areas.
In: Verry ES, Hornbeck JW, and Dolloff CA (eds.) Riparian
Management in Forests of the Continental Eastern United States,
351
pp. 23 42. Boca Raton, London, New York, Washington, DC:
Lewis Publishers.Junk WJ and Wantzen KM (2004) The flood pulse
concept: New aspects, approaches, and applications An update.
In: Welcomme RL and Petr T (eds.) Proceedings of the Second
International Symposium on the Management of Large Rivers for
Fisheries, vol. 2, pp. 117 149. Bangkok: FAO Regional Office for Asia
and the Pacific.
Lachavanne J B and Juge R (eds.) (1997) Man and the Biosphere
Series, Vol. 18: Biodiversity in Land Inland Water Ecotones. Paris:
UNESCO and The Parthenon Publishing Group.
McCormick JF (1979) A summary of the national riparian symposium.
In: U.S. Department of Agriculture, Forest Service (ed.) General
Technical Report WO 12 Strategies for Protection and Management
of Floodplain Wetlands and Other Riparian Ecosystems,
pp. 362 363pp. Washington, DC: US Department of Agriculture,
Forest Service.
Mitsch WJ and Gosselink JG (2000) Wetlands, 3rd edn. New York:
Chichester, Weinheim, Brisbane, Singapore Toronto: Wiley.
Naiman RJ, Décamps H, and McClain ME (2005) Riparia Ecology,
Conservation, and Management of Streamside Communities.
Amsterdam: Elsevier.
Peterjohn WT and Correll DL (1984) Nutrient dynamics in an agricultural
watershed: Observations on the role of a riparian watershed. Ecology
65: 1466 1475.
Verry ES, Hornbeck JW, and Dolloff CA (eds.) (2000) Riparian
Management in Forests of the Continental Eastern United States.
Boca Raton, London, New York, Washington, DC: Lewis
Publishers.
Wantzen KM, Yule C, Tockner K, and Junk WJ (2006) Riparian
wetlands. In: Dudgeon D (ed.) Tropical Stream Ecology,
pp. 199 217. Amsterdam: Elsevier.
Rivers and Streams: Ecosystem Dynamics and Integrating
Paradigms
K W Cummins and M A Wilzbach, Humboldt State University, Arcata, CA, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Energy Flux
Flux of Matter
Integrative Paradigms in Lotic Ecology
Conservation and Human Alterations of Streams
and Rivers
Further Reading
Introduction
Energy Flux
Research scientists, watershed managers, and conserva
tionists alike agree that following an ecosystem
perspective is the most productive way to examine
streams and rivers. The integration of physical–chemical
with biological processes, which is the study of ecosys
tems, has largely replaced single physical factor or single
species approaches to management and rehabilitation of
running waters. In the discussion that follows, fluxes of
energy and matter into, through, and out of lotic ecosys
tems are used as basic processes embraced by the
integrating paradigms (conceptual models) that presently
underlie inquiry into the structure and function of
streams and rivers.
Energy Sources
Streams and rivers are driven almost entirely by two
alternate energy sources: (1) sunlight that fuels the in
stream growth of aquatic plants (primary production), and
(2) plant litter from stream side (riparian) vegetation. The
relationship between these two energy drivers is essen
tially inverse. The heavier the riparian cover over the
stream/river channel, the greater the plant litter inputs
and the greater the limitation of light reaching the water
and therefore in stream algae and vascular plant growth.
In contrast to nonfilamentous algae, very few stream/
river consumers utilize macrophytes, filamentous algae,
and rooted vascular plants. Rather, the macrophytes enter
352
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
the energy transfer to consumers after they die as detritus.
Stream and river systems in which the majority of the
energy transfer is from in stream plant growth to consu
mers are termed autotrophic. Those systems dominated
by the detrital pathways of energy transfer are hetero
trophic. As discussed in the ‘river continuum concept’
(RCC), the relative importance of these two energy
sources changes with stream size. Smaller streams in
forested catchments are usually dominated by litter
energy sources and wider, mid sized stream segments
are dominated by plant growth in the water. Larger rivers
are dependent upon organic matter (OM) delivered from
the upstream tributary network.
A model of energy flux, that is, the transfer of energy
between trophic levels of plants and animals, was pro
duced by Lindeman in the early 1940s and various forms
of this model have more or less been the basis for the
investigation of energy flux in running waters ever since.
These studies most frequently take the form of energy
budgets; an accounting of the energy in and the energy
out of a given ecosystem (Figure 1) or biological popula
tion (Figure 2) or community within the system. OM
budgets are useful in identifying the sources, magnitude,
and fates of energy and provide insight into internal
dynamics of a river system. At the system level, inputs
include autotrophic production plus energy originating
from the surrounding terrestrial environment (allochtho
nous) that is brought in by various physical vectors.
CPOM
111.9
Terrestrial inputs
FPOM
436.0
CPOM
export 19.1
CO2
14.2
92.8
CPOM
92.9
CPOM
microbes
24.5
CPOM
microbe
processing
29.3
CO2
12.2
Colonization
< 0.01
FPOM
export 190.5
245.5
Outputs include community respiration and losses by
downstream transport. Energy retained within a stream
reach over a given time interval is referred to as storage.
Comparisons of energy flux among and between trophic
levels commonly express biomass as caloric equivalents.
Animal ingestion, egestion, and growth (increase in mass)
are all measured as biomass. Respiration (metabolism)
is readily converted to calories consumed using an oxy
calorific equivalent. Tables are available that provide
conversions of mass to calories for freshwater organisms.
Feeding Roles and Food Webs
Feeding studies of benthic macroinvertebrates have
shown that, based on food ingested, most taxa are omni
vorous. For example, invertebrates that chew riparian
derived leaf litter in streams, termed ‘shredders’, ingest
not only the leaf tissue and associated microbiota, (e.g.,
fungi, bacteria, protozoans, and microarthropods), but
also diatoms and other algae that may be attached to the
leaf surface, as well as very small macroinvertebrates (e.g.,
first instar midge larvae). For this reason, trophic level
analysis does not lend itself well to simple trophic cat
egorization of stream macroinvertebrates.
An alternate classification technique, originally
described by Cummins in the early 1970s, involves the
functional analysis of stream/river invertebrate feeding.
The method is based on the combined morphological and
Light
DOM
438.6
DOM
248.7
export
0.1?
Leaching
Autumnal mortality
13.9
Flocculation
CPOM
38.0
Abrasion
71.0
Leaching +
7.9
ex-cell release
0.5?
CO
2
Ingestion
FPOM
Conversion
+
378.7
41.8
microbes
to
418.6
particulate
Feces
cells
Shredders
Feces
25.1
60.8
4.5
Ingestion 23.6
39.4
Collectors
Ingestion CO2
4.2
Feces
0.5
0.6
11.6
Ingestion
0.5
Predators
0.1
Producer
carbon fixation
CO2
0.1?
189.9
Ex-cell release 0.1?
Ex-cell release 0.1?
DOM
204.5
DOM
microbe
processing
166.5
CO2
105.7
DOM
microbes
60.8
CO2
0.3
Macroproducers Microproducers
12.7
0.3?
Feces
2.5
Ingestion
0.05
CO2
8.4
Ingestion
4.2
Scrapers
0.45
CO2
1.2
Community respiration:
field data 518.7
Component summation
531.1
Figure 1 Example of an energy budget for a small woodland stream ecosystem (Augusta Creek watershed, Michigan, USA). All
values are in grams ash-free dry mass m 2 yr 1. Squares represent pools of organic matter in various states; arrows represent transfers
and circles represent respiratory consumption of organic matter. From Saunders GW et al. (1980) In: LeCren ED and McConnell RH
(eds.) The Functioning of Freshwater Ecosystems. Great Britain: Cambridge University Press.
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
Ingestion
Winter
1.21
cal ind–1 d–1
Summer
353
Egestion
1
44.97
2
57.26
1
54.63
2
37.57
Winter
66%
0.91
Summer
cal ind–1 d–1
1
35.68
2
49.45
1
36.96
2
27.18
AVG kcal m–2 generation–1 =
32.06
AVG kcal m–2 generation–1 =
48.61
34%
Assimilation
Winter
0.30 cal ind–1 d–1
Summer
Respiration
1
9.29
2
7.81
1
17.66
2
11.85
AVG kcal m–2 generation–1 =
11.653
Winter
74%
Summer
0.24 cal ind–1 d–1
1
5.45
2
4.8
1
14.32
2
9.93
kcal m–2 generation–1 =
AVG
8.623
26%
Production
Winter
0.06 cal ind–1 d–1
Summer
AVG
3.03
1
3.85
2
3.01
1
3.35
2
1.92
kcal m–2 generation–1 =
Figure 2 Example of an energy budget constructed over 2 years of study for a population of a stream invertebrate (Glossosoma
nigrior, Trichoptera) from Augusta Creek, Michigan, USA The budget is based on independent measurements of ingestion, production,
and respiration. Modified from Cummins KW (1975) Macroinvertebrates. In: Whitton BA (ed.) River Ecology. Berkeley: University of
California Press.
behavioral mechanisms of food acquisition used by the
invertebrates and four fundamental categories of their
food found in running waters (Figure 3). There is a direct
correspondence between the availability of categories of
nutritional resources and the relative abundance of inver
tebrate populations that are adapted to efficiently harvest
a given food resource. Five invertebrate functional feed
ing groups (FFG) have been designated. These include
shredders, filtering collectors, gathering collectors, scra
pers, and predators. These partition four food resource
categories in running waters that are defined on the basis
of particle size and type: (1) coarse particulate organic
matter (CPOM), which is primarily riparian litter that
has been conditioned, that is, microbially colonized,
within the stream; (2) fine particulate organic matter
(FPOM), which are particles generally smaller than
1 mm in diameter that are largely derived from the bio
logical and physical breakdown of CPOM and whose
surfaces are colonized by bacteria; (3) periphyton, that
is, tightly accreted algae and associated organic material;
and (4) prey, that is, invertebrate species or larval/
nymphal stages small enough to be captured and con
sumed by invertebrate predators. As the relative
availability of the basic food resources changes, there is
a concomitant change in the corresponding ratios of the
FFGs of freshwater invertebrates adapted to specific
resource categories.
Obligate and facultative members occur within each
FFG. These can be different species or different stages in
the growth period of the life cycle of a given species. For
example, it is likely that most aquatic insects, including
predators, are facultative gathering collectors as first instars
newly hatched from the egg. It is with obligate forms that
linkages between invertebrates with their food resource
categories are most reliable. The distinction between obli
gate and facultative status is best described by the efficiency
with which a given invertebrate converts the resource
acquired to growth; that is, obligate forms are more efficient
consumers of a given resource, such as conditioned leaf
litter, than are facultative forms. For example, shredders
feeding on litter consume the fungal rich leaf matrix,
whereas scrapers only abrade the much less nutritious leaf
354
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
Coarse particulate organic matter
Dissolved organic matter
Light
Macroproducers
Microbes
Microproducers
Dissolved
organic matter
Flocculation
Microbes
Fine particulate
organic matter
Shredders
Scrapers
Collectors
Predators
Predators
Figure 3 Conceptual model of invertebrate functional feeding groups and their food resources in a small, forested stream ecosystem.
Modified from Cummins KW (1974) Structure and function of stream ecosystems. Bioscience 24: 631–641.
cuticle. The high efficiency of obligate forms feeding on a
particular resource category is in contrast with the wider
array of food types consumed by facultative forms, but with
lower efficiency. The same morpho behavioral mechanisms
can result in the ingestion of a wide range of food items, the
intake of which constitutes herbivory (consumption of liv
ing plants), detritivory (consumption of dead OM), or
carnivory (consumption of live animal prey).
Although intake of food types changes from season
to season, habitat to habitat, and with growth stage,
limitations in food acquisition mechanisms have
been shaped over evolutionary time and these are
relatively more fixed than the food items ingested.
Morphological structures that enable aquatic
insects to harvest a given food resource category
exhibit significant similarities across diverse taxa.
This convergent or parallel evolution lies at the
heart of the FFG classification method. For example,
larvae of the 26 North American caddisfly
(Trichoptera) families are spread among the four
major nonpredaceous FFGs. The less highly evolved
mayflies (Ephemeroptera) and stoneflies (Plecoptera)
are adapted to acquire fewer food resource categories
(Table 1).
An advantage of the FFG procedure is that it does
not require detailed taxonomic separations of the
invertebrates. Broad, easily distinguished characteris
tics allow FFG classification, preferably in the field
with live specimens. Separations usually involve sys
tematic distinctions at the level of family or higher,
and cut across taxonomic lines. As an example, two
groups of case bearing larval caddisflies (Trichoptera)
are sufficient to separate FFG categories at better
than 90% efficiency. All families, or genera within
Table 1 Numbers of families and functional group assignment of some representative orders of benthic macroinvertebrates in
running-water ecosystems
Number of families by dominant functional feeding groups
Order
Total number of
families
Shredders
Ephemeroptera
Plecoptera
Trichoptera
21
9
26
6
5
Scrapers
Filtering
collectors
Gathering
collectors
2
5
10
8
6
4
Predators
Filamentous algal
piercers
4
3
2
1
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
355
families, of Trichoptera that construct mineral cases
are scrapers. Those that construct organic cases are
shredders.
Given the coupling of FFGs and food resource cat
egories, ratios of the different groups can serve as
surrogates for ecosystem parameters. For example, the
ratio of the functional groups linked to in stream primary
production (scrapers plus those shredders that may harvest
live plant tissue) to those groups dependent upon the
CPOM and FPOM heterotrophic food resources (shredders
of detrital material plus gathering and filtering collectors)
provides an index of the ratio of autotrophy to heterotrophy
at the lotic ecosystem level. When measured directly, an
ecosystem ratio of autotrophy/heterotrophy 1 indicates an
autotrophic system. A surrogate FFG ratio of 0.75 has been
measured in such autotrophic stream/river systems.
modifying this view of closed cycles in lakes to an open
cycle model; that is, the open nutrient cycles in streams and
rivers follow a spiraling pattern in which nutrients gener
ated (or delivered) at one point along a stream or river
complete the recycling to their initial state at a displaced
location downstream (Figure 4). Total spiral length repre
sents the sum of the distance traveled by an element as an
inorganic solute until its uptake by the biota, plus the
distance traveled within the biota until its release back
into the water column. If nutrients such as nitrogen or
phosphorous are cycled rapidly, the spirals are ‘tight’, that
is, the downstream completion of the cycle is short. If
cycling is slow, the closing of the loop is displaced a longer
distance downstream and the spirals are more open. The
tighter the spiraling cycling loops, the more retentive (con
servative) is the stream or river reach.
Flux of Matter
Transport and Storage of OM
Nutrient Cycles and Spiraling
The limnological study of standing waters has always been
dominated by a conceptual model of closed ecosystems, in
which nutrients recycle seasonally, totally within the sys
tem. The unidirectional flow of running waters necessitated
Mechanism
Retention
Biological
activity
Effect on nutrient cycling
Rate of
recycling
High
High
Cycling
rate–1
Fast
(a)
The transport and storage of OM in running water eco
systems involves complex interactions between (1) the
state of the OM, (2) the source of the OM, and (3) the
physical, chemical, and biological retention potential for
any given reach of stream or river.
Distance between
spiral loops
Ecosystem
response to
nutrient addition
Ecosystem
stability
Short
Stream
flow
Conservative
(I > E )
High
Storing
(I > E )
High
Intermediately
conservative
< A but > D
Low
Exporting
(I = E )
Low
Distance between loops
Slow
(b)
(c)
(d)
High
Low
Low
Short
Low
Fast
Long
Slow
Long
High
Low
Figure 4 Nutrient spiraling depicted as the effects of different interactions between the distance of downstream movement
(velocity time) and measures of biological activity such as metabolism by benthic microbes. Modified from Minshall GW, Petersen RC,
Cummins KW, et al. (1983) Interbiome comparison of stream ecosystem dynamics. Ecological Monographs 53: 1–25.
356
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
State of the OM
Three broad categories of OM are dissolved (DOM,
size range <0.45 mm), fine particles (FPOM, size
range >0.45 mm to 1 mm), and coarse particles (CPOM,
size range >1 mm). While FPOM particles are colonized
primarily on the surface by bacteria, CPOM is colonized
by fungus, bacteria, and microzoans that penetrate the
matrix of the material. Aquatic hyphomycete fungi usually
penetrate the CPOM leaf and needle litter first. Bacteria
and microzoans follow the fungal hyphal tracks into the
matrix of the CPOM. The OM in solution (DOM) includes
a full range of molecules from simple very labile ones
such as sugars and amino acids to complex recalcitrant
compounds such as phenolic compounds.
Sources of the OM
A major source of OM in streams (orders 0–5) is the
riparian zone. This border of stream side vegetation pro
duces litter (e.g., leaves, needles, bud and flower scales,
seeds and fruits, small wood and bark) that enters on a
seasonal schedule depending upon the relative propor
tions of deciduous and evergreen species. Other sources
of OM are solutions and particles from bank erosion,
DOM leachates from litter, exudates, and leachates from
periphytic algae and vascular aquatic plants together with
their physical fragmentation and mortality.
Physical, chemical, and biological retention
potential
The retention of DOM involves physical flocculation of
the OM in solution with divalent cations, such at Caþþ,
and biological uptake by resident bacteria and fungi.
Chemical reactions between the smaller molecular weight
organic compounds may precede the physical complexing
with cations. The rate and extent of biological uptake of
DOM depends upon factors such as the lability or recal
citrance of the compounds, density and composition of
the microbial flora, and water temperature. These
mechanisms that convert DOM to FPOM, flocculation
and microbial uptake, are quite important ecosystem pro
cesses. The conversion of DOM in solution to particles
significantly increases the retention of the OM. The dif
ference in the efficiency of retention of OM between soft,
stained water streams and hard, clear water streams
accounts in part for the greater productivity of the latter.
The POM that results from the conversion of DOM is
more likely to remain in a given reach of stream or river
and enter into trophic pathways.
Retention of POM depends upon channel geomor
phology. Large wood debris (LWD), branches and
exposed bank roots, coarse sediments, backwaters, side
channels, and settling pools are all important retention
features. For any given reach of stream or river, a major
source of OM is transport from upstream. In addition,
OM is retained when bankfull flow is exceeded and
material is deposited on the upper banks or on the
floodplain. OM is returned to the channel when water
levels recede. Whether these off channel areas serve as
sources or sinks for OM over an annual cycle depends
upon the configuration of the upper banks and flood
plains and the patterns of the flood flows. The general
fertility of floodplains suggests that they are largely
sinks.
Integrative Paradigms in Lotic Ecology
Paradigms, or conceptual models, have continued to be
developed, modified, and integrated since the 1980s. The
RCC, arguably the most encompassing of these, has
guided a large portion of the research on lotic ecosystems
in the interim. However, a number of other models have
served to elucidate specific components of running water
structure and function or have proposed alternative broad
integrating principles.
The RCC
The major goal of the architects of the RCC was to
examine the patterns of biological adaptation that overlay
the physical setting (template) of stream/river channels in
a watershed.
The RCC views entire fluvial systems, from head
waters to their mouths, as continuously integrated series
of physical gradients together with the linked adjustments
in the associated biota. The RCC was founded on many
antecedent studies and many correlates have been incor
porated into the general paradigm. Subsequent views and
critiques of portions of the RCC also have had significant
impact on the present form of the RCC as a general model
of lotic ecosystem structure and function. This model
focuses on the gradient of geomorphological–hydrologi
cal characteristics as the fundamental template along
intact catchments upon which biological communities
become and remain adapted. This physical template,
and biological communities adapted to it, are viewed as
changing in a predictable fashion from stream headwaters
to river mouth (Figures 5–7). Major generalizations of
the RCC involve seasonal spatial variations in OM supply
(e.g., algal/detrital biomass), structure of the invertebrate
community, and resource partitioning along drainage net
works (Figure 8).
The RCC predicts that recognizable patterns in the
structure of biological communities and the input, utiliza
tion, and storage of OM will be observable along the
continuum. Light limitation inhibits primary production
in the headwaters (orders 1–3) due to shading of the
channel by riparian vegetation and in the larger rivers
(orders greater than 7 or 8) because of light attenuation
through the turbid water column that is typical of the
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
357
Figure 7 The Smith River in coastal northern California USA.
This high order river with a canyon-controlled channel is
dependent upon organic matter delivered from the upstream
tributary network.
Figure 5 A headwater stream in the Cascade Mountains of
Oregon, USA at winter base flow. The coniferous riparian zone
provides partial canopy closure and supplies large woody debris
to the channel.
waters (orders 4–6). High biotic diversity is supported in
rivers in this size range both because of the variety of
habitats and food resources for consumer organisms, and
because of overlapping ranges of organisms with evolu
tionary terrestrial origins (such as the insects) that are
dominant in the headwaters with those of marine origins
(e.g., annelids and mollusks) that are more prevalent
downstream (Figure 9).
The RCC has been widely utilized as an organizing
principle and has been the subject of many studies result
ing in various tests of the concept. As would be expected,
a degree of unpredictability in the physical template
leads to correspondingly less predictability in the overlay
of biological communities. This lack of predictability is
often a function of the spatial and temporal scales
of reference employed and it can also be induced by
human interference. For example, systems may appear
more or less variable over time spans of less than a decade,
the time period of observation, but long term variability,
at the scale of centuries or greater, is usually obscured by
short term variations.
Other Paradigms
Figure 6 The Firehole River, a mid-sized woodland-meadow
stream in Yellowstone National Park, USA. Reduced riparian
shading enables abundant growth of in-stream algae.
At least eight paradigms, other than the RCC, continue to
guide the development of running water ecosystem the
ory (Table 2). These include serial discontinuity,
hierarchical scales, riparian zone influences, flood pulse,
hyporheic dynamics, hydraulic stream ecology, patch
dynamics, and network dynamics.
Serial discontinuity
lower portions of stream/river networks. The cumulative
effect of drainage networks tends to increase nutrient
levels in the down stream/river direction. Overall per
iphyton and rooted vascular plant biomass, and insect and
fish diversity are all maximized in the mid sized running
Interruptions in the longitudinal continuum, as proposed
by the RCC, are caused by engineered impoundments
which serve to reset the general patterns of biotic organi
zation. Above the dam, the system exhibits characteristics
of a higher order than the impounded stream. Below
358
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
Shredders
Microbes
Grazers
Predators
1
CPOM
Collectors
2
Collectors
P/ R < 1
Periphyton
Microbes
Shredders
Grazers
Vascular
hydrophytes
FPOM
Stream size (order)
3
Predators
4
P/ R
>
5
tary
1
u
Trib
Periphyton
CPOM
OM
FP
P/ R < 1
FPOM
6
7
P/ R < 1
8
Phytoplankton
9
Collectors
Microbes
Predators
10
11
12
Zooplankton
Relative channel width
Figure 8 The ‘river continuum concept’ (RCC). A proposed relationship between stream size (order) and the progressive shift in structural
and functional attributes of lotic biotic communities. The heterotrophic headwaters and the large rivers are both characterized by an
autotrophic index, or P/R (ratio of gross primary production to total community respiration) of less than 1 (P/R <1). The largely unshaded
mid-sized rivers are generally classified as autotrophic with a P/R >1.The invertebrate communities of the headwaters are dominated by
shredders and collectors, the mid-sized rivers by grazers ( scrapers) and collectors. The large rivers are dominated by FPOM-feeding
collectors. Fish community structure grades from invertivores in the headwaters to invertivores and piscivores in the mid-sized rivers to
planktivores and bottom-feeding detritivores and invertivores in the largest rivers. From Vannote RL, Minshall GW, Cummins KW, Sedell JR,
and Cushing CE (1980) The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37: 130–137.
the dam, the regulated flows often completely alter the
seasonal hydrological patterns of the receiving channel.
For example, the normal pattern imposed by the discon
tinuity resulting from a dam is to decrease the flows
during natural high water periods (reservoir storage
phase) and release the water during natural low flow
periods (reservoir release phase). The storage phase
retains water to prevent flooding and to provide a later
water supply during dry periods. During the release
phase, water is delivered for irrigation, drinking,
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
359
Diversity and/or abundance
Light Periphyton Major algal
limitation biomass growth form
Diatoms
Greens
Vascular
plants
Bluegreens
Mollusks
annelids
Insects
1
Benthic
3
Epiphytic
6
9
Planktonic
Figure 9 Patterns in categories of biotic diversity, from small streams to large rivers, compared on a relative scale for each parameter, as
predicted by the ‘river continuum concept’. Numbers at the right are general stream/river order ranges. Modified from Cummins KW (1997)
Stream ecosystem paradigms. In: CNR – Instituto di Ricerca Sulle Acque. Prospettive di recerca in ecologia delle acque. Roma, Italia.
Table 2 Comparison of most appropriate scales of application for eight commonly used paradigms (conceptual models) for
running-water ecosystem analysis
Basin or reach scale
Basin
Reach
Macro
Meso
Micro
Macro
Meso
Micro
Stream orders or reach length
RCC
HS
RZI
FPC
HD
HSE
PD
ND
0/1 Order to estuary
0/1 Order to order 6
0/1 Order to order 2–5
>1000 m
100–1000 m
<10 m
a
a
a
d
d
d
a
a
a
a
a
a
b
b
a,e
b
a
a
a
a
d
d
c
c
c
c
c
a
a
a
c
c
c
c
a
a
c
c
c,a
a
a
a
b
b
b
c
a
d
RCC, River Continuum Concept; HS, hierarchical scales; RZI, riparian zone influences; FPC, flood pulse concept; HD, hyporheic dynamics;
HSE, hydraulic stream ecology; PD, patch dynamics; ND, network dynamics.
a, most direct influence on stream biota and ecosystem processes; b, if channels are braided, ranking moves down in c\scale to a lower order;
c, beyond the scale to detect specific (local) differences; d, influence too local to detect general, large scale patterns; e, may be of less direct
importance in naturally deranged (lake interrupted) or beaver influenced stream systems.
recreation, and, in some cases, to improve fish habitat. In
some basins, interruptions in the longitudinal profile of
river networks occur in the form of natural lakes or
impoundments. Although these may change the sequen
cing of stream orders, they do not change the annual
hydrograph.
reach scale and over short time periods do not adequately
represent the patterns viewed over greater spatial and
temporal scales. Therefore, descriptions of stream/river
ecosystem structure and function must be placed in the
appropriate context of space and time.
Hierarchical scales
Ecotones
The hierarchical scales paradigm addresses a weakness
of the RCC. The relative significance of the factors driv
ing the physical, chemical, and biological components of
running waters changes with scale. The data on which the
RCC is based were all collected at the reach scale and
during only several seasons. The hierarchical approach
recognizes that ecosystem processes operating at the
Several paradigms focus on the ecotones that bridge
between the stream/river channel and its surroundings
and underpinnings. These include the riparian border
primarily along small streams, the aquatic terrestrial
interface of large rivers with floodplains, and the subsur
face region of the sediments beneath running waters (the
hyporheic zone).
360
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
Riparian zone influences
of the inputs of litter to streams varies among ecoregions
and with the species composition of the riparian vegetation.
Roots of riparian vegetation stabilize banks at the edge of
the channel and influence the chemistry of subsurface flow
into the channel. The width of the riparian zone that
encompasses these root functions also varies. Thus, a com
plete definition of the riparian zone that encompasses all
these functions would need to be of sufficient width to
accommodate all of them. Just as zones of influence of
these riparian functions vary, so do the zones of associated
management. For example, if a goal is to manage for the
long term input of large woody debris to the channel from
the riparian zone to provide habitat structure for fish and
invertebrates, all trees tall enough to reach the channel
when they fall and large enough to provide habitat structure
should be left in place. However, to accommodate varia
tions in channel and bank morphology and the composition
of the riparian vegetation, this management would need to
be implemented at the reach scale.
The riparian zone paradigm attempts to integrate the phy
sical processes that shape the valley floor of streams and
rivers with the coupled succession of terrestrial plant com
munities in the riparian zone along the channel and the role
they play in the formation of stream habitat and the pro
duction of nutritional resources for organisms that reside in
running waters. The ecotone between wetted channel and
terrestrial bank vegetation that constitutes the riparian zone
is a critical coupling, especially in small streams. The con
fined or unconstrainted nature of the channel system, for
example, exert a major influence on the nature of the
riparian vegetation that develops along the banks.
Definition of the lateral boundaries of the riparian zone,
or buffer strip width in the parlance of timber managers, is a
continuing debate. From the perspective of the stream/
river ecosystem per se, the functional roles of the riparian
corridor encompass differing areas along the stream bank
(Figure 10). Shading of the channel, which along with
nutrient levels regulate in stream primary production,
which in turn depends upon the height and foliage density
of the vegetation, steepness of the side slopes, and aspect
(compass direction) of the channel. The width of the ripar
ian zone that yields litter inputs and large woody debris to
the channel can also vary with height and species composi
tion of the stream side vegetation. Seasonal timing of litter
drop and its introduction into the stream produces patterns
around which the life cycles of many steam invertebrates
have become adapted. This coupling between riparian litter
inputs and stream invertebrates is most direct for inverte
brate shredders that feed on conditioned litter. The timing
Flood pulse concept
The flood pulse concept addresses the ecotones between
rivers and their floodplains. Unlike the lateral riparian influ
ence on stream ecosystem processes where the impact is
largely from the landscape to the stream, the flood pulse
concept emphasizes the reciprocal exchange between the
major river channel and its floodplain. A consequence of this
distinction is that the overwhelming bulk of riverine animal
biomass derives directly or indirectly from production on
the floodplain and not from downstream transport of OM
produced higher in the watershed. Although the importance
Riparian influences
Shading
Channel form
Wood debris
Scour and fill
Litter inputs
Roots
Bank
stabilization
Fish habitat
Microbial
populations
In-stream
primary
production
Woody debris
Shade
Invertebrate
populations
Shade
Litter inputs
Fill
Riparian
Stream
Riparian
Figure 10 Influences of the riparian zone on streams. Redrawn from Cummins KW (1988) The study of stream ecosystems:
A functional view. In: Pomeroy LR and Alberts JJ (eds.). New York: Springer.
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
of this aquatic/terrestrial transition zone, the floodplain
ecotones, is widely acknowledged, there are few hard data
to indicate whether over annual cycles, or longer periods,
the primary movement of nutrients and biomass is onto or
off the floodplain, or in balance. The general perception of
‘fertile floodplains’ suggests that the periodically inundated
floodplains are sinks relative to the river channel. However,
the high productivity of adult fish in many floodplain rivers
and the concentration of reproductive activity on the flood
plain supports the notion that floodplains are sources and the
river is a sink, gradually exporting to the sea. At any rate, the
seasonal pulsing of river discharge, the flood pulse, is the
major force controlling existence, productivity, and interac
tions of biota in river–floodplain ecosystems.
For any given storm or series of storms, the movement
of material and organisms on to the floodplain follows the
rising limbs of the hydrographs, and the return to the river
channel follows the falling limbs. Unfortunately, applica
tion of the flood pulse concept is restricted because of the
wide scale engineering modifications that have isolated
rivers from their floodplains. Natural exceptions to the
flood pulse concept are rivers flowing through deeply
incised canyons.
Hyporheic dynamics
The hyporheos is the subsurface region beneath and
adjacent to stream/river channels that exchanges water
with the surface water. This surface water–ground water
ecotone is spatially and temporally very dynamic. The
conceptual framework of the hyporheic dynamics para
digm has resulted in the incorporation of channel aquifer
dynamics into the general model of the RCC. The hypor
heic zone occurs, at least to some extent, beneath and
lateral to the active channel from the headwaters to river
mouths, except in bedrock channels. Water, solutes, inor
ganic and organic complexes, and uniquely adapted biota
move through interstitial pathways into and out of
the sediments. These flowpaths are determined by the
bedform of the channel. Where the bedform is convex,
there is groundwater recharge from the channel into the
sediments. These locations where oxygenated water is
driven into the sediments are often spawning sites of
salmonid fishes. Concave bedforms are sites of ground
water upwelling into the channel. Invertebrates found in
the hyporheos include the small early instars of a wide
range of taxa and larger forms at times of extremes inflow,
either high or low. In addition, there are some microbe
nthic forms that are specifically adapted to a groundwater
existence. The presence of hyporheic invertebrates is
determined largely by siltation and the availability of
oxygen. If interstitial sediment spaces are filled with fine
sediments and/or conditions are anaerobic, the fauna will
be excluded.
361
Hydraulic stream ecology
The hydraulic stream ecology model emphasizes that the
local responses of stream organisms to flow conditions can
serve as an organizing principle for running waters. Lotic
animals are not well adapted to hydraulic stress and can
sustain exposure to such stress for only short periods of
time. Common patterns of diurnal and seasonal drift of
stream invertebrates along vertical gradients of sediment
and current velocity are a manifestation of this stress
response. The model identifies mean water velocity and
depth as more critical than characteristics of substrates in
determining the distributions of stream animals. The
model has yet to adequately incorporate the extensive
data that implicates the diurnal light cycle as the major
control parameter of invertebrate stream drift.
Patch dynamics
The patch dynamics concept emphasizes the patchy dis
tribution of riverine habitats in space and time, and argues
that an ever shifting mosaic of patches enables a greater
number of species to co occur than would be the case
under greater environmental constancy. Environmental
conditions are predictable in aggregate, but not within a
particular patch, and these aggregate conditions confer
some regularity in species composition. In this model,
particular patch types can be found at any point along
the general longitudinal gradient proposed by the
RCC. However, there are clear examples of invertebrate
‘patches’ that at least change in abundance from head
waters along the continuum to large rivers. For example,
small headwater streams (orders 1–3) are generally better
shaded than higher orders (>3) and sustain less suitable
algal periphyton to support scrapers. Further, the domi
nance of the CPOM–detrital shredder linkage correlates
with stream width and the close availability of riparian
tree and/or shrub litter, and this generally matches with
stream orders 1–3. The extension of the shading of per
iphyton growth and the riparian CPOM–shredder linkage
to larger rivers can occur along braided channels, but
these ‘patches’ will always be more abundant in the head
waters than in mid sized or larger rivers.
Network dynamics
The network dynamics hypothesis, which combines the
hierarchical scales and patch dynamics models, is based
on the observation that there are abrupt changes that
occur at the confluences of tributaries with the receiving
channel. Changes in water and sediment flux at these
locations result in changes in the morphology of the
receiving channel and its floodplain. In this view, the
branching nature of river channel network, together
with infrequent natural disturbances, such as fire, storms,
and floods, are the formative elements of the spatial and
temporal organization of the nonuniform distribution of
riverine habitats. Further, the tributary junctions are
362
Rivers and Streams: Ecosystem Dynamics and Integrating Paradigms
proposed hot spots of biological activity. Some data show
increased fish diversity and abundance at these junctions,
but the influence on other components of the biota has
yet to be investigated. The ‘network dynamics hypothesis’
does not address ‘patches’ represented by braided
channels.
Whether hydraulic characteristics, tributary junctions,
or other patch phenomena, represent local conditions that
need to be integrated along river continua to account for
whole profile trends that are clearly apparent, or whether
such phenomena are localized specific modifiers that
differentially affect stream orders along profiles has yet
to be demonstrated clearly.
Conservation and Human Alterations of
Streams and Rivers
A great challenge for stream and river ecology in the twenty
first century will be the restoration of degraded running
water ecosystems while preserving those systems that still
remain in good condition. Restoration will dominate in more
developed regions where modifications of running waters
and their watersheds have been more extensive. In less
developed regions, preservation of many running waters
may still be possible, but the distinction between pristine
and degraded systems is disappearing rapidly. The historical
scientific databases for running waters are generally poor,
with largely anecdotal or very incomplete information avail
able. The lotic ecosystem paradigms described above can
serve as tools for evaluating present conditions of running
waters, surmising their likely antecedent condition, and
developing targets and strategies for restoration. Because
the majority of degraded streams and rivers have changed
beyond our ability to return them to their historical state, it is
more logical to use the term rehabilitation. Often the actions
will take the form of returning certain organisms or processes
to a condition that addresses societal objectives.
In the context of preserving and rehabilitating
streams and rivers, it will be important to enlist the best
scientific understanding of the structure and function of
running water ecosystems. For example, regulations gov
erning the protection and width of riparian buffer strips,
designed to protect stream organisms (usually fish) vary
from one area to another, wider in some areas, narrower
in others. However, managers and environmentalists
should not limit their view of riparian buffers as only a
matter of vegetative composition and buffer width with
the sole aim of providing shading to reduce water tem
peratures, a source of large woody debris, or stream bank
stabilization. This view of riparian buffers ignores the
often completely different in stream trophic role played
by the coupled riparian ecosystem. The buffer width
required to produce shade, litter, large wood, nutrients,
and bank stabilization are often quite different. Thus, the
management and rehabilitation of a given reach of
running water requires an integrated approach that
acknowledges all the riparian functions and places the
actions within the context of the larger watershed.
See also: Desert Streams; Estuaries; Floodplains;
Freshwater Lakes; Riparian Wetlands.
Further Reading
Benda L, Poff NL, Miller D, et al. (2004) The network dynamics
hypothesis: How channel networks structure riverine habitats.
Bioscience 54: 413 427.
Cummins KW (1974) Structure and function of stream ecosystems.
Bioscience 24: 631 641.
Cummins KW (1975) Macroinvertebrates. In: Whitton BA (ed.) River
Ecology. Berkeley: University of California Press.
Cummins KW (1988) The study of stream ecosystem: A functional view.
In: Pomeroy LR and Alberts JJ (eds.). New York: Springer.
Cummins KW (1997) Stream ecosystem paradigms. In: CNR Instituto
di Ricerca Sulle Acque. Prospettive di recerca in ecologia delle
acque. Rome, Italy.
Frissell CA, Liss WJ, Warren CE, and Hurley MD (1986) A hierarchical
framework for stream classification: Viewing streams in a watershed
context. Environmental Management 10: 199 214.
Gregory SV, Swanson FJ, McKee WA, and Cummins KW (1991) An
ecosystem perspective of riparian zones. Bioscience
41(8): 540 551.
Junk WJ, Bayley PB, and Sparks RE (1989) The flood pulse concept in
river floodplain systems. Canadian Journal of Fisheries and Aquatic
Sciences, Special Publication 106: 110 127.
Minshall GW, Petersen RC, Cummins KW, et al. (1983) Interbiome
comparison of stream ecosystem dynamics. Ecological Monographs
53: 1 25.
Stanford JA and Ward JV (1993) An ecosystem perspective of alluvial
rivers: connectivity and the hyporheic corridor. Journal of The North
American Benthological Society 12: 48 60.
Statzner B and Higler B (1986) Stream hydraulics as a major
determinant of benthic invertebrate zonation patterns. Freshwater
Biology 16: 127 139.
Saunders GW, et al. (1980) In: LeCren ED and McConnell RH (eds.) The
Functioning of Freshwater Ecosystems. Great Britain: Cambridge
University Press
Townsend CR (1989) The patch dynamics concept of stream
community ecology. Journal of the North American Benthological
Society 8: 36 50.
Vannote RL, Minshall GW, Cummins KW, Sedell JR, and Cushing CE
(1980) The river continuum concept. Canadian Journal of Fisheries
and Aquatic Sciences 37: 130 137.
Ward JV and Stanford JA (1983) The serial discontinuity concept of river
ecosystems. In: Fontaine TD and Bartell SM (eds.) Dynamics of Lotic
Ecosystems, pp. 29 42. Ann Arbor, MI, USA: Ann Arbor Science
Publications.
Rivers and Streams: Physical Setting and Adapted Biota
363
Rivers and Streams: Physical Setting and Adapted Biota
M A Wilzbach and K W Cummins, Humboldt State University, Arcata, CA, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
History of the Discipline of Stream and River Ecology
The Physical and Chemical Setting
The Adapted Biota
Further Reading
Introduction
through the 1950s and this period also marked the begin
ning of a focus by lotic ecologists on human impacts.
Descriptive studies detailed the taxonomic composition
and density of the benthic invertebrate fauna found in
reaches of streams and rivers variously affected by human
impacts. Beginning in the 1960s and 1970s there was a
shift to more holistic views of flowing water ecosystems,
with research concentrated on a synthetic view of lotic
ecosystems, on energy flow, and on organic matter bud
gets for first order catchments. In 1970, Noel Hynes,
father of modern stream ecology, published his landmark
book The Ecology of Running Waters which summarized
concepts and literature to that point. With the 1980s,
came the realization that running water dynamics could
be fully understood only through an integrated spatial
and temporal perspective, and that whole catchments
were the basic units of stream/river ecology. For exam
ple, holistic organic budget analyses of running water
ecosystems cannot be constructed unless both spatial
and temporal scales are applied.
The hallmark of lotic research during the 1980s
and 1990s was its interdisciplinary nature. Interactions
involved stream ecologists, fishery biologists, aquatic ento
mologists, algologists, hydrologists, geomorphologists,
microbiologists, and terrestrial plant ecologists. It was
these interactions between the disciplines that focused the
attention of stream biologists on physical processes and
greater spatial and temporal scales. This perspective of
stream ecosystems continues to direct the science in the
twenty first century, aided immensely by the incorpora
tion of geographic information systems (GIS) analysis.
Although there is general acceptance that the logical
basic unit for the study of streams and rivers is the
watershed or catchment, most measurements of lotic eco
system structure and function are still made at the reach
or microscale level. Recently, there has been strong
impetus to extend the scope of understanding to the
watershed mesoscale and beyond because ecosystem pro
cesses exhibit effects of differing importance at different
spatial and temporal scales and these processes interact
across scales. The concern for issues of global climate
change in regard to streams has provided additional moti
vation to analyze entire basins or all the basins in
Streams and rivers are enormously important ecologically,
economically, recreationally, and esthetically. This impor
tance far outweighs their proportional significance on the
landscape. Running waters constitute less than 1/1000th of
the land surface and of freshwater resources of Earth and
contribute only 2/10 000th of annual global freshwater bud
gets. Streams and rivers are significant agents of erosion and
serve a range of human needs, including transportation, waste
disposal, recreation, and water for drinking, irrigation, hydro
power, cooling, and cleaning. At the same time, flooding of
streams and rivers pose potential natural hazards to human
populations. Irrespective of their impact on man, streams and
rivers are rich, complex ecosystems that are diagnostic of the
integrity of the watersheds through which they course.
There has always been a general anecdotal notion of what
constitutes a stream and what constitutes a river; that is,
streams are small, narrow, and shallow while rivers are
large, wide, and deep. However, the difference between
them is without clear distinction in the literature of the last
100 years. For the purposes of this article, streams refer to
channels in drainage networks of orders 0–5 and rivers as
orders 6–12 and above (see definition of stream order under
the section titled ‘Channel morphology’). In this article, the
history of stream ecology is discussed followed by a treatment
of the physical and chemical setting and biological features of
major groups of lotic organisms. In a companion article,
ecosystem dynamics and integrating paradigms in stream
and river ecology are covered.
History of the Discipline of Stream and
River Ecology
The formal published beginning of the study of
flowing waters (lotic ecology) dates to the early twentieth
century in Europe, where initial work focused on the
distribution, abundance, and taxonomic composition of
lotic organisms. In North America, the ecological stream
studies began shortly after. In the 1930s, North American
stream ecology was dominated by fishery biology. Stream
and River studies worldwide remained descriptive
364
Rivers and Streams: Physical Setting and Adapted Biota
continental regions. Thus, a challenge for lotic ecologists
in the twenty first century remains the integration of
data rich studies at the reach level to entire watersheds
and finally the coarse resolution of regional basin analysis
relying on satellite imagery. The ‘river continuum con
cept’ and other stream/river conceptual models described
below should continue to aid in the integration of
knowledge about lotic ecosystems along whole catch
ments, from micro to macroscale levels.
The Physical and Chemical Setting
Stream and river biota evolved in response to, and in
concert with, the physical and chemical setting.
Although traditionally the domain of hydrologists, geo
morphologists, and chemists, study of processes driving
the physical and chemical templates have been
embraced by stream ecologists for interpreting patterns
in organismic distributions and lotic ecosystem struc
ture and function. From a purely physical perspective,
the primary function of rivers is to transfer runoff and
move weathering products away from terrestrial por
tions of the Earth for delivery to the oceans. Despite
tremendous variability in the morphology and behavior
of rivers, each results from the interaction between
geomorphic and hydrologic processes. These processes
and their effect on river morphology are summarized,
followed by a discussion of major physical (current,
substrate, and temperature) and chemical factors that
affect the functioning of river ecosystems and the adap
tations of stream organisms.
Hydrologic Processes
The total amount of the Earth’s water does not change,
and is continuously recycled among various storage com
partments within the biosphere in a process referred to
as the hydrologic cycle (Figure 1). The cycle involves
evaporation from land and evapotranspiration from
terrestrial vegetation driven by solar energy, cloud forma
tion, and precipitation.
Annual global precipitation averages about 100 cm,
but the majority evaporates and little falls directly
into streams. The remainder either infiltrates into the
soil or becomes surface runoff. The relative contributions
of different pathways by which water enters streams and
rivers varies with climate, geology, watershed physiogra
phy, soils, vegetation, and land use.
Water that infiltrates becomes groundwater, which
makes up the largest supply of unfrozen freshwater.
Groundwater discharges gradually to stream channels
through springs or direct seepage when a channel inter
sects the groundwater table. Baseflow describes the
proportion of total stream flow contributed from ground
water, and sustains streams during periods of little or no
precipitation. Running waters may be categorized by
the balance and timing of stormflow versus baseflow.
Ephemeral streams carry water only in the wettest
years and never intersect the water table. Intermittent
streams flow predictably every year only when they
receive surface runoff (Figure 2). Perennial streams flow
continuously during wet and dry periods, receiving both
stormflow and baseflow. The duration, timing, and pre
dictability of flow greatly affect the composition and life
history attributes of stream communities.
Stream and river discharge, the most fundamental of
hydrological measurements, describes the volume of
water passing a channel cross section per unit time.
Any increase in discharge must result in an increase in
channel width, depth, velocity, or some combination of
these. Discharge increases in a downstream direction
through tributary inputs and groundwater addition and
is accompanied by increases in channel width, depth, and
velocity. An estimated 35 000 km3 of water is discharged
annually by rivers to the world’s oceans, with the Amazon
River alone accounting for nearly 15% of the total.
Hydrographs depict changes in discharge over time.
Individual storm events display a steep rising limb
from direct runoff, a peak, and a gradually falling reces
sion limb as the stream returns to baseflow conditions
(Figure 3). Variability in the shapes of hydrographs
among streams reflects differences in the climatic, geo
morphic, and geologic attributes of their watersheds and
differences in the distribution of runoff sources.
Discharge records of sufficient duration allow predic
tion of the magnitude and frequency of flood events for a
given river and year. Recurrence interval (T, in years) for
an individual flood may be estimated as
T ¼ ðn þ 1Þ=m
where n is the number of years of record, and m is ranked
magnitude of the flood over the period of record, with the
largest event scored as m ¼ 1.
The reciprocal of T is the exceedance probability,
which describes that statistical likelihood that a certain
discharge will be equaled or exceeded in any given
year. Thus a 1 in 100 year flood has a probability of
1% of occurring in any given year. The probability
that a 100 year flood will occur in a river is the same
every year, regardless of how long it has been since
the last 100 year flood. Recurrence interval informa
tion provides an extremely important context for
studies of lotic organisms.
Geomorphic Processes
Discharge and sediment supply represent the physical
energy and matter that move through river systems,
365
Rivers and Streams: Physical Setting and Adapted Biota
Cloud formation
Rain clouds
Sur
face
n
Transpiratio
From soil
From ocean
Fro
m
Precipitation
Fro
ms
veg
eta
t
ion
trea
ms
Evaporation
Lake
storage
runo
ff
Infiltration
Soil
Rock
Percolation
Ocean
Deep percolation
Groun
dwate
r
Figure 1 The hydrologic cycle. From Stream Corridor Restoration: Principles, Processes, and Practices, 10/98, by the Federal
Interagency Stream Restoration Working Group (FISRWG).
Lag time
St
Rainfall intensity
Stream discharge (cfs)
(inches h–1)
Rising
limb
w
flo
m
or
Baseflo
w
0
Figure 2 An intermittent stream at 3.4 km elevation in the
Andes Mountains in Chile, bordered by riparian vegetation of
herbs and grasses. Intermittent streams are often important in
exporting invertebrates and organic detritus to downstream
fish-bearing reaches.
Recession
limb
1
Time
of rise
2
3
Time (days)
4
Figure 3 Stream hydrograph from a rainstorm event. From
Stream Corridor Restoration: Principles, Processes, and
Practices, 10/98, by the Federal Interagency Stream Restoration
Working Group (FISRWG).
366
Rivers and Streams: Physical Setting and Adapted Biota
and channel form and profile change over time to accom
modate the energy and matter delivered to it. Three
primary geomorphic processes, including erosion, transport,
and deposition, supply sediment to streams and rivers.
Physical/chemical weathering of bedrock and soils, together
with channel, bank, and floodplain erosion account for short
and long term lotic sediment supply. Initiation of sediment
movement in the channel is a function of drag and lift forces
exerted on sedimentary particles. The greater the velocity
and shear stress exerted on the streambed, the greater the
grain size that can be entrained. Stream competence and
stream capacity refer to the largest grain size moved by a
given set of flow and the total amount of sediment that can be
transported, respectively.
Coarse sediment moves along the stream/river bottom
as bedload, and fine sediment moves downstream in the
water column as suspended load. The suspended load, or
turbidity, screens out light and scours off organisms
attached to the bottom while the organic fraction serves
as the food resource for invertebrate filtering collectors.
Whereas sediments may be temporarily deposited within
mid channel or point bars, longer term storage occurs on
floodplains and elevated alluvial terraces.
Channel Morphology
Within a reach, channel cross sections reflect the interaction
between bank materials and flow and vary from symmetri
cal in riffles to asymmetrical in pools as flow meanders.
Bankfull discharge, when discharge just fills the entire
channel cross section, occurs every 1.5–2 years on average
in unregulated systems. Erodible banks lead to wide shallow
rivers dominated by bedload, while resistant banks produce
narrow, deep channels transporting high suspended loads.
Channel pattern is described by its sinuosity (amount
of curvature) and thread (multiple channel braiding).
Sinuosity index is measured as channel length along the
thalweg (deepest portion of the channel), divided by valley
length. If the index exceeds 1.5, the stream/river is classi
fied as meandering. Erosion of the channel bank carves
the river bends, with the fastest current at the outside of
the bend where the bank erodes. The greater the curve, the
faster the water flows around the bend, deflecting to
the other bank and forming the next curve. This pattern
repeats downstream, creating regular swings in the river
with a meander wavelength approximately 11 times the
channel width.
Riffles are topographic high spots along the channel
composed of the coarsest bedload sediments transported
by the river, and with a water surface slope that is steeper
than the mean stream gradient at low flow (Figure 4).
They are typically spaced every five to seven channel
widths. Pools are topographic depressions with fine sedi
ments and reduced velocity.
Straight
Riffle
Pool
Thalweg
line
Sinuous
Pool
Riffle
or cross over
Figure 4 Riffle and pool sequences in straight and sinuous
streams. From Stream Corridor Restoration: Principles,
Processes, and Practices, 10/98, by the Federal Interagency
Stream Restoration Working Group (FISRWG).
The longitudinal profile of a river is relatively stable
over time, adjusting slowly to discharge and sediment
supply. The profile is generally concave, with a steep
gradient in its headwaters, and a gentle gradient at its
mouth. The concavity reflects the adjustment between
climate and tectonic setting (land relief and base level)
and geology, which controls sediment supply and resis
tance to erosion. Base level describes the limit to which a
river cannot erode its channel. For streams emptying into
the ocean, this is sea level.
Within a drainage basin, stream channels and their net
works grow in size and complexity in a downstream direction
as described by stream order (Figure 5). A first order stream
1
1
1
1
1
1
1
1
2
1
2
1
3
3
3
1
4
1
2
1
1
2
2
2
1
1
1
4
Figure 5 Ordering of stream segments within a drainage
network. From Stream Corridor Restoration: Principles,
Processes, and Practices, 10/98, by the Federal Interagency
Stream Restoration Working Group (FISRWG).
Rivers and Streams: Physical Setting and Adapted Biota
lacks permanently flowing upstream tributaries and order
number increases only where two stream of equal order join
together. Employing this system, the Mississippi and the
Nile Rivers at their mouths are order 10. There are usually
3–4 times as many streams of order n – 1 as of order n, each of
which is roughly half as long, and drains a little more than
one fifth of the land area. In the United States, nearly half of
the approximately 5 200 000 km total river length are first
order. As discussed later, many features of stream ecosystem
structure and function are correlated with stream order.
Drainage basins, or watersheds, are the total area of
land draining water, sediment, and dissolved materials to a
common outlet. Watersheds occur at multiple scales, ran
ging from the largest river basins to first order watersheds
measuring only a few hectares in size. Larger watersheds
are comprised of smaller watersheds and stream segments
in a nested hierarchy of ecosystem units. The size and
shape of the watershed, and the pattern of the drainage
network within the watershed, exerts a strong influence on
the flux of energy, matter, and organisms in river systems.
Because some movement of energy, matter, and organisms
move across and through landscapes independently of
drainage basins, a more complete perspective of stream
ecology requires consideration of landscape ecology.
Physical Factors
Current
Current (m s 1 of flow) is the central defining physical
variable in running water systems. Velocity and asso
ciated flow forces exert major effects on stream
organisms. Current shapes the nature of the substrate,
delivers dissolved oxygen, nutrients, and food, removes
waste materials, and exerts direct physical forces on
organisms on streambed and in the water column, result
ing, for example, in the dislodgement and displacement of
organisms downstream. Current velocity, which rarely
exceeds 3 m s 1 in running waters, is influenced by the
river slope, average flow depth, and resistance of bed and
bank materials.
Flow in running waters is complex and highly variable
in space and time. At a given velocity, flow may be
laminar, moving in parallel layers which slide past each
other at differing speeds with little mixing, or turbulent,
where flow is chaotic and vertically mixed. The dimen
sionless Reynolds number, the ratio of inertial to viscous
forces, predicts the occurrence of laminar versus turbu
lent flow. High inertia promotes turbulence. Viscosity is
the resistance of water to deformation, due to coherence
of molecules. At Reynolds numbers <500, flow is laminar;
at >2000 flow is turbulent with intermediate values tran
sitional. Although laminar flow is rare in running waters,
microenvironments may contain laminar flow environ
ments, even within turbulent, high flow settings.
367
In cross section, a vertical velocity gradient decreases
exponentially with depth. Highest velocities are at the sur
face where friction is least, and zero at the deepest point of
the bottom where friction is the greatest. Mean current
velocity is at about 60% of the depth from the surface to
bottom. A boundary layer extends from the streambed to a
depth where velocity is no longer reduced by friction and a
thin viscous sublayer of laminar flow exists at its base.
Microorganisms and small benthic macroinvertebrates
experience shelter from fluid forces within the sublayer.
However, most stream organisms must contend with
complex, turbulent flow where they exhibit a variety of
morphological and behavioral adaptations for reducing
drag and lift. Adaptations of macroinvertebrates and fishes
may include small size, dorsoventral flattening to reduce
exposure to current, streamlining to reduce current drag,
the development of silk, claws, hooks, suckers, and fric
tion pads as holdfasts, and behavioral movement away
from high velocity areas.
Substrate
In running waters, substrate provides food or a surface
where food accumulates, a refuge from flow and preda
tors, a location for carrying out activities such as resting,
reproduction, and movement, and material for construc
tion of cases and tubes. Algal growth, invertebrate growth
and development, and fish egg incubation largely occur
on or within the substrate. Substrate includes both inor
ganic and organic materials, often in a heterogeneous
mixture. Mineral composition of the substrate is deter
mined by parent geology, modified by the current.
Organic materials include aquatic plants and terrestrial
inputs from the surrounding catchment ranging from
minute fragments and leaves to fallen trees (Figure 6).
Figure 6 Small headwater stream in old-growth Douglas-fir
forest in Oregon, showing large woody debris spanning the
channel. This spanner log forms a retention structure for organic
detritus and sediment as well as refugia and habitat when the
channel is inundated by high flows.
368
Rivers and Streams: Physical Setting and Adapted Biota
Table 1 Size categories of inorganic substrates in streams and
rivers
Size category
Particle diameter (range in mm)
Boulder
>256
Cobble
Large
Small
128–256
64–128
Pebble
Large
Small
32–64
16–32
Gravel
Coarse
Medium
Fine
8–16
4–8
2–4
Sand
Very coarse
Coarse
Medium
Fine
Very Fine
1–2
0.5–1
0.25–0.5
0.125–0.25
0.063–0.125
Silt
stability, are also determinants of benthic community
structure, but these are less easily quantified. In general,
larger, more stable rocks support greater diversity and
numbers of individual organisms than smaller rocks, but
smaller rocks with a higher ratio of surface area to volume
support higher densities.
Evaluation of the ecological role of substrate is difficult
because of its heterogeneity and covariance with velocity
and oxygen supply. Heterogeneity is expressed along
the length of a river as decreasing particle size in a down
stream direction and at a reach scale as pool and riffle
sequences, meandering, and point bar development.
Substrate embeddedness describes the degree to which
larger sediments, such as cobbles, are surrounded or cov
ered by fine sand and silt. Significant embeddedness
reduces streambed surface area and organic matter stor
age, the flow of oxygen and nutrients to incubating fish
eggs and aquatic invertebrates, and entrance to and move
ment within the streambed by invertebrates.
<0.063
Modified from Cummins KW (1962) An evaluation of some techniques
for the collection and analysis of benthic samples with special emphasis
on lotic waters. American Midland Naturalist 67: 477 504.
Inorganic and organic materials are often classified by
size according the Wenthworth scale (Table 1). A broad
classification of organic materials is discussed in Rivers and
Streams: Ecosystem Dynamics and Integrating Paradigms.
Organic particles <1 mm in diameter and >0.45 mm (fine
particulate organic matter or FPOM) often function as
food rather than substrate, and larger organic materials
(CPOM) serve as substrate or food, for example, for
litter feeding invertebrates (Figure 7). Other substrate
attributes, including shape, surface texture, sorting, and
Figure 7 Accumulation of leaf litter in a second-order stream in
Oregon (USA) flowing through a second-growth forest with a red
alder riparian zone. The litter that is retained at the leading edge
of the cobbles provides the major food resource for stream
invertebrate shredders and habitat for other invertebrates.
Temperature
Temperature affects all life processes, including those in
running waters. For example, decomposition, primary pro
duction and community respiration, and nutrient cycling
are all temperature dependent. Most stream organisms are
ectothermic, and their metabolism, growth rates, life cycles,
and overall productivities are all temperature dependent.
Annual temperature changes often serve as environmental
cues for and/or regulate life history events of invertebrates
and fishes, especially emergence or spawning. The tem
perature regime sets limits to where species can live, and
many species are adapted to certain thermal regimes.
Increasing water temperature decreases dissolved oxygen
solubility at the same time that it increases metabolic
demand. Thus preferences of such organisms as salmonid
fishes for cold water may have as much to do with tem
perature effects on oxygen availability as with effects of
temperature per se.
Stream temperature is the net result of heat exchange
via (1) net solar radiation, which reflects direct beam
solar radiation, modified by cloud cover, day length, sun
angle, vegetation, and topographic shading; (2) evaporation
and convection, which are affected by vapor pressure
and air temperature differentials as well as wind speed;
(3) conduction, or heat exchange with streambed; and
(4) advection from upstreamwater inputs, including ground
water and tributaries. On a diel basis, stream temperature
varies less than air temperature because of the high specific
heat of water. The greatest daily fluxes occur in summer
in temperate regions, with a minimum before dawn.
These fluxes are greatly affected by canopy cover and con
tributions of groundwater, which usually enters the channel
at a temperature within 1 C of mean annual air tempera
ture. At the catchment level, daily temperature flux
increases with distance from the headwaters, with a
Rivers and Streams: Physical Setting and Adapted Biota
maximum in mid order segments. Thermal stratification is
rare except in large rivers and at tributary junctions.
Seasonal variations in temperature mirror trends in mean
monthly air temperature. The timing of the summer max
imum often lags the timing of maximum solar radiation.
Year to year variation in monthly temperatures is low, typi
cally less than 2 C. The annual temperature range of
temperate streams is generally 0–25 C, and 0–40 C in
intermittent desert streams. The lower Amazon River is
always within one or two degrees of 29 C. Extremes in
temperature occur in hot water springs, which can exceed
80 C, and in subarctic and arctic streams that may comple
tely freeze in winter. Surface freezing is usually prevented
by snow and ice bridging, but underwater ice may form on
streambeds as anchor ice or in the water column as slush or
frazil.
Stream ecologists often evaluate temperature effects
on stream organisms and ecosystem processes on the
basis of degree day accumulation rather than temperature
maxima or minima. Degree days, which are calculated by
summing daily mean temperatures above 0 C, can differ
among streams with similar maximum or minimum
average daily temperatures. Such differences can affect
voltinism (number of annual generations) of some species
of aquatic insects.
Water Chemistry
Constituents of river water can be divided into five cate
gories, which include dissolved gases, dissolved inorganic
ions and compounds, particulate inorganic material, parti
culate organic material, and dissolved organic ions and
compounds. Dissolved gases include oxygen, carbon diox
ide, and nitrogen. Dissolved inorganic ions and compounds
include major and minor ion groups and trace elements,
such as copper, zinc, iron, and aluminum which occur in
minute quantities. Nitrogen and phosphorus are minor ions,
which function as nutrients essential to plant and animal
growth. Major ion groups include cations of calcium, mag
nesium, sodium, and potassium and the anions bicarbonate,
sulfate, and chloride.
The pH, which measures hydrogen ion activity, is
affected by concentrations of dissolved gases and major
ions and determines the solubility and biological availabil
ity of nutrients and heavy metals. Hardness is a measure of
calcium and magnesium concentrations normally used to
assess the quality of water supplies. Hardness is associated
with, but not identical to alkalinity, which measures
the ability of streamwater to absorb hydrogen ions, thus
buffering changes in pH. Alkalinity is primarily due to
bicarbonate and carbonate ions. Total dissolved solids,
the sum of the concentrations of major cations and anions,
are often estimated as specific conductance. Hardness,
alkalinity, and ionic concentrations are frequently posi
tively correlated with stream productivity and taxonomic
369
richness. Particulate inorganic and organic materials
together make up the suspended load in lotic systems,
and contribute to turbidity.
Carbon dioxide and oxygen are the most biologically
important dissolved gases. Diffusion from the atmos
phere maintains concentrations of both oxygen (O2) and
carbon dioxide (CO2) in streams at close to equilibrium.
However, CO2 is more soluble in water than is O2, which
is 30 times less available in water than air. Groundwater
and sites of organic matter decomposition are low in O2
and enriched in CO2. Photosynthesis and respiration can
alter diel concentrations of oxygen and carbon dioxide in
productive systems, with O2 elevated and CO2 reduced
during day, and the reverse occurring at night. If produc
tion is high relative to diffusion, diel changes in O2 are
used to estimate photosynthesis and respiration. Because
current and turbulence continually renew O2 supply, its
concentrations are problematic for stream organisms only
in sites severely contaminated with organic pollutants or
through a combination of high temperatures, drought, and
dense populations of aquatic plants. Low O2 concentra
tions are better tolerated by stream animals at faster
current speeds.
Typical rivers have been described as essentially dilute
calcium bicarbonate solutions dominated by a few cations
and anions. The considerable natural spatial variability in
lotic chemistry largely reflects the type of rocks available for
weathering and the amount, chemical composition, and
distribution of precipitation. For example, total dissolved
solids are approximately twice as great in rivers draining
sedimentary terrain compared with igneous and meta
morphic rock. Most rivers contain 0.01–0.02% dissolved
minerals, about 1/20–1/40th the salt concentration of the
oceans, with an average concentration of 100 mg l 1.
Generally 50% of this is bicarbonate and 10–30% is
chloride and sulfate. River water contains more dissolved
solids than does rainwater, because of evaporation, weath
ering, and anthropogenic inputs. Rainwater, although nearly
pure, contains dissolved minerals from dust particles and
droplets of ocean spray.
Rainwater is also naturally acidic due to atmospheric
carbon dioxide dissolving in the water droplets, forming a
weak carbonic acid (H2CO3). In catchments with hard
rocks resistant to weathering, little buffering capacity, or
where decaying plant matter is abundant, streamwater
can be acidic even in absence of pollution. Water perco
lating through the soil enters the stream and is enriched
with CO2 from plant and microbial respiration and forms
carbonic acid. The carbonic acid dissolves the calcium
carbonate in rocks, producing calcium bicarbonate, which
is soluble in water and the source of carbon atoms
for aquatic photosynthesis. The dissolution of calcium
carbonate increases the amount of stream calcium and
bicarbonate ions and the latter dissociates to carbonate
ions. At equilibrium, bicarbonate and carbonate ions
370
Rivers and Streams: Physical Setting and Adapted Biota
dissociate, forming hydroxyl ions and resulting in weak
alkaline waters, with a pH > 7. At equilibrium, water
resists changes in pH because the addition of hydrogen
ions is neutralized by the hydroxyl ions formed by
dissociation of bicarbonate and carbonate, and added
hydroxyl ions react with bicarbonate to form carbonate
and water. Thus the buffering capacity of a stream is
largely determined by its calcium bicarbonate content.
The pH of most natural running waters ranges between
6.5 and 8.5, with values below 5 or above 9 being harmful
to most stream organisms. Industrially derived sulfuric
and nitric acids have seriously lowered pH in surface
waters of large areas of Europe and North America,
resulting in reduced species diversity and density.
The Adapted Biota
Many taxonomic groupings inhabit running waters. Key
biological attributes, life histories, and distribution pat
terns of organisms that play a central role in energy flux
within lotic ecosystems or that are of significant human
interest – namely algae, macrophytes, benthic macroin
vertebrates, and fishes – are summarized below.
Algae
Algae are the most important primary producers in run
ning water ecosystems and because of their sessile nature
and short life cycles, their assemblages are used to eval
uate stream ecosystem health. Algae are thalloid
organisms, bearing chlorophyll a and lacking multicellu
lar gametangia. Algal evolution radiated from a common
ancestry to several diverse kingdoms. For example, blue
green algae are classified as bacteria, and dinoflagellate
algae as protozoans. Algal taxonomy is based on pigmen
tation, the chemistry and structure of internal storage
products and cell walls, and number and type of flagellae.
Five major divisions of algae are common in streams,
including the Bacillariophyta (diatoms), Chlorophyta
(green algae), Cyanobacteria (blue green algae),
Chrysophyta (yellow green algae), and Rhodophyta (red
algae). Of these, the diatoms, green algae, and cyanobac
teria are most prevalent. Assemblages of algae attached to
the substrate are referred to as periphyton or aufwuchs.
Periphyton attached to submerged substrates is a complex
assemblage of algae, bacteria, fungi, and meiofauna bound
together with a polysaccharide matrix referred to as bio
film. Algae of the water column are phytoplankton,
occurring chiefly in slowly moving lowland rivers as
sloughed benthic cells or exports from connected stand
ing waters within the watershed.
Diatoms are extremely abundant in freshwater as well
as in saltwater, and typically comprise of majority of
species within the periphyton. Generally microscopic,
diatoms are brownish colored single celled algae con
structed of two overlapping siliceous cell walls, or
valves, fit together like the halves of a petri dish. Valves
are connected to each other by one or more ‘girdle’ bands.
The two valves form the frustule, which is uniquely
decorated with pores (punctae), lines (striae), or ribs
(costae). The symmetry of these decorations defines
two groups: radially symmetrical centric diatoms and
bilaterally symmetrical pennate diatoms. Diatoms may
occur individually, in chains, or in colonies, and those
with a divided cell wall (raphe) are able to move.
In temperate streams, diatoms exhibit two growth blooms:
in spring prior to shading by deciduous canopies as water
temperatures rise and nutrients are plentiful; and in fall
following leaf abscission, when nutrients released from
decaying green algae and deciduous litter are available.
Diatoms constitute a high quality, rapid turnover food
resource for macroinvertebrate scrapers and collectors.
Representative diatoms common in stream periphyton
are shown in Figure 8.
Diatoma
x500
Melosira x500
Meridion x350
Cymbella x200
Gomphonema
x150
Cocconeis
x500
Navicula x500
Achnanthes x500
Nitzschia x700
Synedra x250
Figure 8 Representative diatoms common in stream periphyton. From Hynes HBN (1970) The Ecology of Running Waters. Liverpool:
Liverpool University Press.
Rivers and Streams: Physical Setting and Adapted Biota
Green algae occur in a variety of habitats, and are dis
tinguished by the number and arrangement of flagella, their
method of cell division, and their habitat. In streams, dis
tinctions are made between micro and macroforms.
Macroalgae occurs as a thallus or as filaments. Filamentous
forms may be branched or unbranched. Green algae provide
attachment sites for diatoms, and are a source of FPOM
and photosynthetic oxygen, but are fed upon by few
invertebrates.
Blue green algae, or cyanobacteria, are prokaryotic
organisms of ancient lineage which contain the photosyn
thetic pigment phycocyanin, used to capture light for
photosynthesis. They occur in a variety of habitats and are
one of very few groups of organisms that can convert inert
atmospheric nitrogen into an organic form. Bluegreen algae
may be filamentous or nonfilamentous, and only filamentous
forms with heterocysts are capable of nitrogen fixation in
aerobic settings. Several of the heterocyst containing fila
mentous taxa, (e.g., Anabaena, Aphanizomenon, and Microcystis)
can form dense blooms and produce toxins in warm, nutri
ent rich waters. Nitrogen fixing Nostoc, common in small
streams, forms a unique commensal association with the
chironomid midge Cricotopus.
Macrophytes
Macrophytes include vascular flowering plants, mosses and
liverworts, some encrusting lichens, and a few large algal
forms such as the Charales and the filamentous green alga
Cladophora. Light and current are among the most important
factors limiting the occurrence of macrophytes in running
waters. Major plant nutrients, particularly phosphorus, can
be limiting in nutrient poor waters but are likely to be
present in excess in eutrophic lowland rivers. Three ecolo
gical categories include those that are attached to the
substrate, those that are rooted into the substrate, and free
floating plants. Attached plants include the mosses and liver
worts, certain lichens, and some flowering plants of the
tropics. These are all largely found in cool, headwater
streams. The mosses are unusual in their requirement
for free CO2, rather than bicarbonate, as their carbon source.
In shaded, turbulent streams, their contribution to primary
production may override that of the periphyton. Mosses
also support very high densities of macroinvertebrates.
Rooted plants include submerged (e.g., Hydrocharitaceae,
Ceratophyllaceae, and Halorgidaceae) and emergent (e.g.,
Potamogetonaceae, Ranunculaceae, and Cruciferae) forms
and require slow currents, moderate depth, low turbidity,
and fine sediments for rooting. They are most common in
mid sized rivers and along the margins of larger rivers where
they reduce current velocity, increase sedimentation, and
provide substrate for epiphytic microflora. Tough, flexible
stems and leaves, attachment by adventitious roots, rhizomes
or stolons, and vegetative reproduction are important
adaptations. Free floating plants (e.g., Lemnaceae and
371
Pontederiaceae) are of minor importance in running waters
at temperate latitudes as they depend largely on lacustrine
conditions. They may accumulate significant biomass in
subtropical and tropical settings. Macrophytes in lotic eco
systems contribute to energy flow predominantly through
decomposer food chains, as few macroinvertebrates feed on
the living plants.
Benthic Macroinvertebrates
The major groups of invertebrates in running waters
include three phyla: Annelida (worms) and Mollusca
(snails, clams, and mussels) of marine evolutionary origin
that are most abundant and diverse in larger rivers, and
Arthropoda (crustaceans and insects) that dominate the
headwaters, but are abundant all along drainage networks.
Representative taxa are illustrated in Figure 9.
The Oligochaeta is the most abundant and diverse
group of annelids, and are notable for their ability
to inhabit low oxygen environments. Oligochaetes
inhabit the sediments, some in tubes, and are almost
all gathering collector detritivores. The worms are
segmented with two pairs of stout, lateral chetae on
each segment. Annelid leeches (Hirudinea), a minor
group occurring in small streams to mid sized rivers,
are gathering collectors or predators.
Gastropod (limpets and snails) and bivalve (clams and
mussels mollusks) are restricted in their occurrence in
streams and rivers by their calcium requirement for
shell formation. Limpets, such as Ferrissia (Ancylidae),
frequent small, fast flowing streams where their hydro
dynamic shape and sucker formed by the mantle allow
them to move over rocks in the current and scrape loose
attached algal food with a rasping radula. Snails, such as
Physa, are abundant scrapers in river macrophyte beds
where they employ their radulas to rasp vascular plant
surfaces, removing periphyton and epidermal plant tissue.
Clams and mussels (Bivalvia ¼ Pelecypoda) are filtering
collectors that burrow in the sediments with their incur
rent and excurrent siphons exposed. They pump water in
to extract dissolved oxygen and FPOM, and out to elim
inate wastes. Because bivalve mollusks are sensitive to
water quality, they have been used worldwide as indica
tors of lotic ecosystem health. However the small,
ubiquitous fingernail clams (Sphaeridae) are more toler
ant, inhabiting a wide range of stream and rivers.
The common Crustacea of running waters include
Amphipoda (scuds), Isopoda (aquatic pill bugs), benthic
Copepoda (Harpactacoida), and Decapoda (crayfish
and freshwater shrimps). Most isopods and amphipods
(except Hyallela) are detrital shredders feeding on
stream conditioned riparian litter in headwater streams.
Although decapod shrimps and crayfish have species
found in all sizes of running waters, the former tend to
be more abundant in streams, the latter in mid sized
372
Rivers and Streams: Physical Setting and Adapted Biota
(a)
(b)
2
1
4
5
3
3
1
2
5
6
4
6
7
8
9
8
7
10
12
10
11
13
9
Figure 9 (a) Examples of lotic benthic invertebrates. 1, Annelida, Oligochaeta (Tubificidae); 2, Mollusca, Gastropoda (left, Ancylidae;
right, Physidae); 3, Mollusca, Bivalvia (Spaeridae: left, lateral view; right, dorsal view); 4, Crustacea, Amphipoda; 5, Crustacea, Isopoda;
6, Insecta, Ephemeroptera; 7, Insecta, Plecoptera; 8, Insecta, Megaloptera (Sialidae); 9, Insecta, Odonata, Anisoptera (left, nymph; right
upper, lateral view of head with extended labium; 10, Insecta, Odonata, Zygoptera (right, nymph; left lower, lateral view of head with
extended labium). (b) Examples of lotic benthic invertebrates. 1, Insecta, Trichoptera (mineral case bearers); 2, Insecta, Trichoptera
(organic case bearers); 3, Insecta, Trichoptera (net spinner, fixed retreat above); 4, Insecta, Coleoptera (Elmidae adult); 5, Insecta,
Coleoptera (Elmidae larvae); 6, Insecta, Coleoptera, Psphenidae larvae (left, ventral; right, dorsal); 7, Insecta, Diptera, Tipulidae;
8, Insecta, Diptera, Athericidae; 9, Insecta, Diptera, Simuliidae (left, dorsal; right, lateral view); 10, Insecta, Diptera, Chironomidae
(left, Chironominae; right, Tanypodinae); 11, Insecta, Diptera, Chironomidae (filtering tube of Rheotanytarsus); 12, Insecta, Hemiptera,
Corixidae; 13, Insecta, Hemiptera, Belastomatidae.
rivers. Decapods are scavengers, but are usually classified
as facultative shredders of plant litter. These crustaceans
have always been of interest because of their large size,
commercial food and bait value, and importance as food
for large game fish. The minute harpactacoid copepods
are poorly known, but are often in small streams to large
rivers where they are gathering collectors inhabiting
accumulations of benthic FPOM.
Aquatic insects (Arthropoda) are the most conspicuous
and best studied invertebrates of running waters. They
can be subdivided into the more primitive hemimetabo
lous orders, in which immature nymphs gradually
metamorphose into mature winged adults, and the more
evolved holometabolus orders that have a larval and
pupal stage. Insect growth is accomplished by the nymphs
or larvae and lasts for weeks to years, while the adults feed
little and are short lived (a day to weeks). Terrestrial
insects are much more abundant and diverse than lotic
forms but there are 13 orders of aquatic or semiaquatic
(occurring at lotic margins) taxa. The orders in which
all larvae are aquatic are as follows: the hemimetabolous
mayflies (Ephemeroptera), stoneflies (Plecoptera), dra
gon and damselflies (Odonata), and the holometabolous
caddisflies (Trichoptera), and dobson and alderflies
(Megaloptera). These are signature taxa represented
in almost all unpolluted lotic ecosystems. Mayflies,
which are the only insects that molt as winged subadults
(subimagos) to sexually mature adults (imagos), are of
immense importance to sport flyfishing. All the odonate
and about half of the plecopteran nymphs are predaceous.
Rivers and Streams: Physical Setting and Adapted Biota
The dragonflies and damselflies occur in small streams to
large rivers, with many species associated with aquatic
vascular plants. The nonpredaceous stonefly nymphs are
shredders feeding upon conditioned riparian litter.
Caddisflies are a large aquatic order in which a major
ity of species construct portable cases made of plant
pieces (the shredders) or mineral particles (the scrapers)
held together with silk extruded from glands in the
head. All the cases are lined with silk into which hooks
on the hind prolegs are hooked to maintain the larvae in
the case. Larvae circulate water through the case by
undulating the abdomen to irrigate the gills and inte
gument and facilitate respiration. Five families of
Trichoptera larvae, and all families in the pupal stage,
construct nonportable, fixed retreats of organic and
mineral material. Most larvae of the five families spin
silk nets with which they filter out FPOM food from the
flowing water. Species of the family Rhyacophilidae are
free ranging without cases and almost exclusively predac
eous. Some of the predaceous Megaloptera are among the
largest of the lotic aquatic insects, and they are typical of
slow flowing areas and often associated with submerged
woody debris.
The holometabolous Coleoptera (beetles), Diptera (true
flies), Lepidoptera (aquatic moths), and Hymenoptera
(aquatic wasps) constitute the largest insect orders and
have some aquatic or semiaquatic representatives, as do
the spongeflies of the Neuroptera. The beetles are the
only aquatic insects with representatives in which both the
larvae and adults live in the water. One family of Diptera,
the midges (Chironomidae), is usually more abundant and
diverse in running waters than all other aquatic insects
combined. Chironomid species are represented in all lotic
habitats and all functional feeding groups. Their use in
ecological studies has been hampered by the difficulty of
identifying the larvae. Very few aquatic moths are found in
running waters. A few are scrapers inhabiting fast flowing
streams, but the majority live and feed on the leaves of
aquatic macrophytes. Hymenoptera, a large terrestrial
order containing many social species, has some parasitic
forms in which the females enter the water to oviposit in
the immatures of aquatic and semiaquatic orders. The larvae
of spongeflies inhabit freshwater sponges where they are
either predators or feed directly on sponge tissue.
The hemimetabolous Hemiptera (true bugs), Orthoptera
(grasshoppers, etc.), and Collembola (springtails) have
aquatic or semiaquatic species. All the widely distributed
hemipterans are active predators, occupying the full range
of slow water and marginal habitats where they capture prey
and imbibe their body fluids using piercing mouth parts.
All the Orthoptera and Collembola of running waters are
semiaquatic and function as detrital gathering collectors.
Functional feeding roles are explained in greater detail in
Rivers and Streams: Ecosystem Dynamics and Integrating
Paradigms.
373
Fishes
Fishes, the principal group of vertebrates found in
running waters, are of great human interest because of
their commercial and recreational value. Approximately
41% (about 8500 species) of the world’s fishes live in
freshwater. Of these, almost all have representatives
that occur in running waters, although with varying
degrees of river dependency and saltwater tolerance.
Groups with little or no tolerance for saltwater (e.g.,
Cyprinidae, Centrarchidae, and Characidae) are consid
ered to be primary freshwater fishes, and have dispersed
through freshwater routes or evolved in place from dis
tant marine ancestors. Secondary freshwater fishes
(e.g., Cichlidae and Poeciliidae) are usually restricted
to freshwater but have some tolerance to saltwater.
Diadromous fishes migrate between freshwater and salt
water. Anadromous fishes, including many salmonids,
lampreys, shad, and sturgeon, spend most of their lives
in the sea and migrate to freshwater to reproduce.
American and European eels are catadromous fishes,
which spend most of their lives in freshwater and
migrate to the sea to reproduce. Catadromy appears to
be more prevalent in the tropics, and anadromy more
common at higher latitudes.
Longitudinal gradients of fish assemblages are com
mon within river systems, and have resulted in several
attempts to classify stream zones by the dominant fish
species or assemblage found. Because fish faunas vary
considerably among geographic and climatic regions,
zonation schemes can usually be applied only locally
except in Europe. Longitudinal gradients arise as the
result of species addition and/or replacement, and
reflect adaptations to the type and volume of habitat
and available food along the river continuum. Upstream
fishes, typified by salmonids and sculpins, have high
metabolic rates and consequent high demands for oxy
gen. Salmonids are active, streamlined fishes with strong
powers of locomotion that can maintain position in swift
water to feed upon drifting invertebrates. Sculpins, with
depressed heads and large pectoral fins, hold close to the
streambed and forage for invertebrates among stones on
the bottom. Upstream fishes are usually solitary in habit
and may exhibit territoriality associated with both breed
ing and spatial resources. They may extend downstream
where oxygen and temperatures are suitable, to join
deeper bodied fishes more tolerant of warmer tempera
tures and reduced oxygen. Species richness is usually
greatest in the mid order segments, in association with
increased pool development and overall habitat hetero
geneity. The Cyprinidae, one of the largest and most
widespread of primary fish families, is characteristic of
moderate gradient streams. Shoaling behavior is common
within this group. In high order reaches, fish assemblages
include larger, deep bodied fishes such as suckers and
374
Rocky Intertidal Zone
catfishes that feed on bottom deposits, invertivorous sun
fishes, and predatory pike.
See also: Desert Streams; Rivers and Streams:
Ecosystem Dynamics and Integrating Paradigms.
Further Reading
Allan JD (1995) Stream Ecology: Structure and Function of Running
Waters. London: Chapman and Hall.
Cummins KW (1962) An evaluation of some techniques for the collection
and analysis of benthic samples with special emphasis on lotic
waters. American Midland Naturalist 67: 477 504.
Giller PS and Malmqvist B (1998) The Biology of Streams and Rivers.
Oxford: Oxford University Press.
Hauer FR and Lamberti GA (1996) Methods in Stream Ecology. San
Diego: Academic Press.
Hynes HBN (1970) The Ecology of Running Waters. Liverpool: Liverpool
University Press.
Knighton D (1998) Fluvial Forms and Processes: A New Perspective.
London: Arnold Publishers.
Leopold LB (1994) A View of the River. Cambridge, MA: Harvard
University Press.
Rocky Intertidal Zone
P S Petraitis and J A D Fisher, University of Pennsylvania, Philadelphia, PA, USA
S Dudgeon, California State University, Northridge, CA, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Physical Aspects of the Shore
Attached Organisms
Mobile Organisms
Zonation
Rocky Intertidal Shores as an Important System in
Development of Ecology
Unresolved Problems and Future Directions
Further Reading
Introduction
frequency of the tides are altered by the phases of the
Moon, the Earth’s orbit and declination, latitude, and the
configurations of the shoreline and the seafloor. The tidal
range tends to be smaller toward the equator and can vary
from several meters in high latitudes to less than tens of
centimeters near the equator. Configuration of the coast
and the ocean basin can cause harmonic resonances and
create tides that vary dramatically in amplitude and fre
quency. In extreme cases, the reinforcing and canceling
effects can produce a single high and low tide per day or
almost no change over the course of a day.
The timing of low tides can have a profound effect by
exposing organisms to extreme conditions. For example,
the lowest tides in the Gulf of Maine, USA tend to occur
near dusk or dawn, and so organisms are rarely exposed to
mid day sun in the summer but are often exposed to
below freezing temperatures on winter mornings. In con
trast, the lowest summer tides in southeastern Australia
occur mid day and expose organisms to extraordinarily
high temperatures.
The British ecologist A. J. Southward described the
intertidal zone as ‘‘the region of the shore between
the highest level washed by the waves and the lowest
level uncovered by the tide,’’ and thus communities on
rocky intertidal shores are primarily defined by the
tides and the presence of hard surfaces. The types of
organisms, the number of species, and the distribution
and abundance of individual species found in a parti
cular rocky intertidal community also depend on the
physical aspects of the shore, the supply of resources,
food and larvae from overlying water, the biological
interactions among the species present, and the regio
nal pool of species. Although rocky intertidal shores
cover only a small fraction of the Earth’s surface, they
contain a large diversity of organisms – ranging from
highly productive microalgae to transient vertebrate pre
dators (Figure 1).
Physical Aspects of the Shore
Tides
Tides are caused by the gravitational effects of the Moon
and Sun, which ideally produce a cycle of two high tides
and two low tides per day. However, the amplitude and
Characteristics of the Shore
Any firm stable surface in the intertidal zone has the
potential to support the organisms that commonly occur
in rocky intertidal communities, and at low tide, intertidal
habitats can range from dry rock to filled tide pools. Rock
Rocky Intertidal Zone
375
production from suspension feeders, such as barnacles
and mussels, which link the ocean’s productivity to the
shore.
Algae
Figure 1 Closeup of predatory snails, mussels, barnacles, and
brown algae in Maine, USA. Photo by P. S. Petraitis.
surfaces can vary from very hard to relatively soft rock
such as from granite to sandstone and can range from
smooth platforms to irregular fields of stone cobbles and
boulders. Topography, inclination, color, and texture of
the rock affect rate of drying and surface temperature,
which can limit the distribution and abundance of species.
Man made surfaces such as rock jetties and wooden pier
pilings and biogenic surfaces such as mangrove roots can
also support communities that are indistinguishable from
the communities found on nearby rocky shores.
Tide pools can be very different than the surrounding
shore because of thermal variability, changes in salinity
from evaporation and runoff, and changes in pH, nutri
ents, and oxygen levels caused by algae. Pools often
support residents such as sea urchins, snails, and fish
that would otherwise be restricted to subtidal areas.
The amount of wave surge affects the types of organ
isms found on the shore and their distribution. Wave
surge and breaking waves tend to expand the extent of
the intertidal zone and distribution of species by continu
ally wetting the shore and allowing species to extend
farther up the shore. Wave surge can also cause mobile
animals to seek refuge and can limit the distribution of
slow moving species, and the force of breaking waves can
damage and sweep away organisms. Sand and debris such
as logs swept up by the waves can scour organisms off the
surface. In areas of low wave surge, sedimentation of sand
and silt may bury organisms or clog gills and other filter
feeding structures.
The term ‘algae’ refers to an extraordinarily diverse and
heterogeneous group comprising about seven major
lineages, or roughly 41% of the kingdom level branches
in the Eukarya domain. Most lineages consist of unicellular
microalgae, but the multicellular macroalgae that dominate
many rocky shores worldwide occur in only three groups
(Rhodophyta, Chlorophyta, and Phaeophyta) (Figure 2).
Microalgae are ubiquitous and although inconspicu
ous, they are important members of rocky intertidal
communities. For example, diatoms are the primary
food source of many grazing gastropods and form bio
films, which facilitate settlement of invertebrate larvae
and stabilize meiofaunal assemblages.
Benthic macroalgae (i.e., seaweeds) dominate many
rocky shores, especially the low and mid intertidal
zones of temperate regions, and many exhibit morpholo
gies adaptive for life on wave swept shores. The idealized
body plan of a seaweed consists of a holdfast, a stipe, and
one or more blades. The holdfast usually attaches the alga
either by thin encrusting layers of cells tightly appressed
to the rock surface or by a massive, thick proliferation of
tissue that often produce mucilaginous ‘glues’ to adhere
the tissue to the rock. The stipes are analogous to plant
stems and display remarkable material properties that
enable seaweeds to withstand the tremendous hydrody
namic forces imposed by breaking waves. The blade is the
principal structure for the exchange of gases and nutri
ents, and the capture of light for photosynthesis. Blades
also contain reproductive tissue, either within a vegeta
tive blade, or in sporophylls (i.e., special blades for
reproduction). Some larger brown seaweeds, such as
Attached Organisms
Unlike terrestrial habitats, which depend largely on local
plant material to support resident animal populations,
rocky intertidal assemblages are supported not only by
algal primary production but also by secondary
Figure 2 Extensive brown algal beds in Maine, USA. Photo by
P. S. Petraitis.
376
Rocky Intertidal Zone
fucoids and kelps, have gas filled floats called pneumato
cysts that buoy the blade so that it remains closer to the
surface where light intensity is greater.
The diversity and complexity of the life cycles of most
seaweeds contributes to their great abundance on rocky
shores. The life cycle of most seaweeds consists of an
alternation of separate gametophyte and sporophyte
generations. The two generations can either look the
same (i.e., isomorphic) or different (heteromorphic). In
some species, the heteromorphic generations are so differ
ent that they were originally described as different species.
Heteromorphic life histories are hypothesized to represent
an adaptation to grazing pressure, and heteromorphic gen
erations clearly show tradeoffs with respect to competitive
ability, resistance to disturbance and longevity associated
with upright foliose and flat encrusting morphologies.
Sessile Invertebrates
Adults of many invertebrate species are attached perma
nently to the rock or other organisms (epibiota). These
include members of the phyla Porifera (sponges), Cnidaria
(hydroids and sea anemones), Annelida (tube building
polychates), Arthropoda (barnacles), Mollusca (mussels
and clams), Bryozoa (moss animals), and Chordata (tuni
cates). Suspension feeding – either by pumping water
through a sieve structure or trapping particles carried on
induced or external currents – is a common feature of
sessile animals and serves to transfer inputs of energy and
nutrients produced in the water column into the intertidal
zone via the ingestion of plankton. Additionally, by feeding
on locally derived detritus, suspension feeders capture some
of the nutrients that are produced by neighboring
inhabitants.
Sessile intertidal animals are often physically or che
mically defended against predation and display plastic
phenotypes in response to changing environmental con
ditions because they are fixed in place and cannot move to
avoid predators. For example, the presence of the preda
tory gastropod Acanthina angelica induces change in the
shell shape of its barnacle prey Chthamalus anisopoma, and
the barnacle forms a curved shell making it more difficult
for the predator to attack.
algae, or in tide pools, while other species attach to
exposed rock surfaces just ahead of the incoming tide.
Transient species are those that spend only a small part
of their life cycles in the intertidal zone (e.g., as juveniles)
or are those that enter and leave the intertidal zone during
low or high tide.
Invertebrates
Large, mobile invertebrate consumers are ecologically the
most intensively studied guild on rocky shores and
include species from Turbellaria (flatworms), Crustacea
(e.g., crabs, shrimp, amphipods, and isopods), Annelida
(e.g., polychaetes), Gastropoda (e.g., snails, nudibranchs,
and chitons), and Echinodermata (sea urchins, brittle
stars, and sea stars). Herbivores range from grazers of
diatom films to browsers of macroalgae, and predators
exploit a variety of methods (crushing, stinging, drilling,
and partial consumption) to overcome the defenses of
their prey.
Small mobile metazoans (roughly 0.1–1 mm and col
lectively termed meiofauna) thrive on and among the
algae, animals, and the trapped sediments on rocky shores.
Meiofauna include consumers from many invertebrate
phyla, that – due to their small sizes, extremely high
abundances, and high turnover rates – are an important
guild of consumers whose effects have largely been
neglected in comparison to studies of larger invertebrates.
Vertebrates
Vertebrates tend to be transient species that use the
intertidal zone to feed or hide and include fish and marine
mammals that enter at high tide and birds and terrestrial
mammals that enter at low tide (Figure 3). For instance,
marine iguanas (Amblyrhynchus cristatus) of the Galápagos
Islands, Ecuador forage extensively on intertidal algae on
Mobile Organisms
Mobile invertebrates and vertebrates that are found on
rocky intertidal shores are typically divided into two
categories based on the amount of time spent between
tidemarks. Resident species remain in the intertidal zone
throughout most of their life and face a large range of
local physical conditions that they mitigate by a variety of
behavioral and physiological adaptations. Many residents
find shelter during low tides, either between rocks, under
Figure 3 Rocky shore in Central California, USA with elephant
seals on the beach. Photo by S. Dudgeon.
Rocky Intertidal Zone
lava reefs during low tides. The major exceptions are
resident intertidal fishes, which are often cryptic and
less than 10 cm in length. Resident and transient fishes
include hundreds of species from dozens of families,
though members of the families Blenniidae, Gobiidae,
and Labridae are the most common.
Birds and mammals, characterized by high endother
mic metabolic rates and large body sizes, have significant
impacts on intertidal communities even at low densities.
Birds include locally nesting and migratory species and
can remove millions of invertebrates during a season. In
addition, birds in some communities provide major inputs
of nutrients via guano and prey remains. More than two
dozen terrestrial mammals, mostly carnivores, rodents,
and artiodactyls, have been reported as consumers or
scavengers of rocky intertidal organisms on every conti
nent except Antarctica. Most recorded prey species are
mollusks, crabs, or fish. Probably one of the most unusual
cases is a population of feral rabbits on a small island off
the coast of South Africa that forage on seaweeds in the
intertidal zone. Given the mobility of vertebrates, their
impact on rocky intertidal shores has been difficult to
assess and intertidal activity is often discovered by finding
exclusively intertidal animals or algae in the gut contents
of otherwise pelagic or terrestrial species.
Little is known about the effects of harvesting by
humans in the rocky intertidal zone. Results from a few
large scale studies in Australia, Chile, and South Africa,
however, have demonstrated that harvesting has had sig
nificant effects on intertidal assemblages.
Zonation
Patterns
Rocky intertidal shores often display a vertical zonation
of fauna and flora associated with the strong environmen
tal gradient produced by the rise and fall of the tides. For
example, most moderately exposed rocky shores of the
northern hemisphere have kelps at the littoral sublittoral
interface, followed by rhodophyte algae dominating the
low intertidal zone, by fucoid algae, mussels, and barna
cles dominating the mid intertidal zone, and by
cyanobacteria, lichens, and a variety of small tufted,
encrusting, or filamentous ephemeral seaweeds occurring
in the high intertidal zone. While species from many
phyla may be found together, often a single species or
group is so common; vertical zones are named according
to the dominant group (e.g., the intertidal balanoid zone
named after barnacles in the family Balanidae).
Combinations of various physical factors acting upon
different inhabitants in intertidal zones that vary in their
exposure to waves can lead to complex patterns of dis
tribution and abundance along shorelines in a particular
region. Nevertheless, some general patterns are evident at
377
a regional scale. Geographically, vertical zonation pat
terns are most pronounced on temperate rocky shores
where species diversity is high and tidal amplitudes tend
to be greatest. On rocky shores in the tropics, biotic zones
are compressed into narrow vertical bands because of
small tidal amplitudes. In polar regions, annual ice scour
and low species diversity tend to obscure any conspicuous
vertical zonation.
Causes
It is often stated that the upper limits of organisms are set
by physical factors, whereas the lower limits are set by
biological interactions but there are many exceptions to
this rule. The specific causes of the zonation seen on most
rocky shorelines vary with geographic location, but zona
tion results primarily from behavior of larvae and adults,
tolerance to physiological stress, the effects of consumers,
and the interplay between production and the presence of
neighbors.
Adult movements and larval behavior during settlement
from the plankton onto rocky shores have major effects on
the distribution of animals. For example, studies of barnacles
have shown that vertical zonation of larvae in the water
column contributes to corresponding vertical zonations of
both larval settlement and adults on the shore, a pattern
previously ascribed solely to interspecific competition. For
seaweeds, behavior is a relatively unimportant cause of their
zonation since adult seaweeds are sessile and settling spores
are mostly passively transported.
Marine organisms living higher on the shore are
faced with more frequent and extreme physiological
challenges than their lower shore counterparts, and
the upper limits of intertidal distributions for most
species are set by cellular dehydration. Dehydration
can occur either from freezing during winter or simply
desiccation associated with long emersion times. High
temperatures and wind, which accelerate the rate of
water loss from tissues, exacerbate the effects of
desiccation.
Primary and secondary production by sessile organisms
can be limited at higher tidal elevations because nutrients
and other resources can be acquired only when immersed.
Respiration rates of seaweeds and invertebrates are tem
perature dependent and thus can be greater when an
organism is exposed at low tide. For seaweeds, prolonged
exposure to dehydration also reduces photosynthesis.
The reduced productivity associated with increased
exposure at higher tidal elevations modifies intra and
interspecific interactions. For instance, competition
between seaweeds, which may be intense lower on the
shore, is reduced at higher tidal elevations and enables
coexistence. Competition among intertidal seaweeds is
hierarchical with lower shore species dominating those
of the higher shore. Thus, fucoid species of the mid
378
Rocky Intertidal Zone
intertidal zone are outcompeted for space in the low zone
by foliose red seaweeds that pre empt space with an
encrusting perennial holdfast. There is also a competitive
hierarchy among mid intertidal zone fucoids with those
typically occurring lower on the shore competitively
dominant to those higher up. This is most apparent on
European rocky shores where the diversity of intertidal
fucoids is greatest.
Grazing rates tend to be greater lower on the shore,
although there are cases of herbivory by insects setting
the upper limits of ephemeral green algae. Grazing by
sea urchins at the interface with the sub littoral zone can
limit the lower distributions of macroalgae, but there is
little evidence for grazing on perennial seaweeds setting
the lower limits of those taxa within the intertidal
zone. Grazing of perennial seaweeds is most intense at
the sporeling stage soon after settlement. Grazing by
gastropods and small crustaceans certainly contributes
to losses of biomass of established individuals, but does
not affect distributions within the intertidal zone. In
contrast, the grazing of established ephemeral species
both on emergent rock and tidepools is intense during
spring and summer in many regions eventually eliminat
ing those algae from their respective habitats. There are
also many examples of consumers using seaweeds as
habitat as well as food.
Rocky Intertidal Shores as an Important
System in Development of Ecology
The rocky intertidal zone has been a stronghold for eco
logical research, and the success of intertidal experiments
stems in part from the fact that intertidal assemblages are
often comprised of the few species that are able to survive
the environmental variation associated with the cycling of
tides. In addition, many resident intertidal species are
small, common, and slow moving or fixed in one place.
Thus rocky intertidal shores historically appeared as sim
ple, well defined habitats in which easily observed and
manipulated local interactions control the dynamics of
the assemblages. Such initial appearances, however, have
been deceiving, and variation in recruitment of offspring
from the plankton, a characteristic of many marine species,
has stimulated an increased appreciation of the role of
oceanographic conditions.
the West Indies; South and Central America; the coasts
of Africa; the Mediterranean; the Black Sea; Indian Ocean
Islands; Singapore; Pacific Islands, Australia, and Tazmania.
These early accounts of the rocky intertidal remain a poten
tially valuable source for comparison to contemporary
patterns of species distributions due to local species extinc
tions and introductions.
The Rise of Experimental Studies: 1960–80
Direct experimental manipulation of intertidal organisms
accelerated in the 1960s with the groundbreaking work of
J. H. Connell and R. T. Paine. Connell manipulated the
presence of two species of barnacles in Scotland by selec
tively removing individuals from small tiles fashioned
from the sandstone rock from the shore. He showed that
the lower limit of the high intertidal species Chthamalus
stellatus was set by competition with the mid zone species
Balanus (now Semibalanus) balanoides and that the upper
limit of S. balanoides was set by physical factors. Paine
removed the predatory seastar Pisaster ochraceus from an
area of the intertidal shore in Washington and showed
that Pisaster was responsible for controlling mussels,
which are successful competitors for space and dominate
the intertidal shore in the absence of Pisaster. These early
investigations provided a framework for the rapid growth
of experimental studies that characterized the field in
recent decades (Figure 4).
In general, the observation and experimental manip
ulations of mobile consumers and their prey has often
revealed predation by mobile consumers as an impor
tant factor that contributes to the structure of rocky
intertidal assemblages. Consumers have been repeatedly
shown to be prey species and prey size selective,
while algal grazing consumers can inadvertently
Descriptive Studies: Research Prior to 1960
Descriptions of rocky shores and speculation about the
causes of vertical zonation go back more than 195 years.
Before the 1960s, ecologists had published descriptions of
intertidal areas from more than a dozen large geographical
regions that spanned much of the globe and included both
sides of the North Pacific and North Atlantic; Greenland;
Figure 4 Grindstone Neck in Maine, USA with Mount Desert
Island in the background. This site was used by Menge and
Lubchenco in their groundbreaking work in the 1970s. Photo by
P. S. Petraitis.
Rocky Intertidal Zone
379
remove newly settled animals and algae as well as their
intended prey.
Supply-Side Ecology and External Drivers:
1980–2005
Marine ecologists have known for a long time that success
of many intertidal species depend on the supply of pro
pagules (larvae, zygotes, and spores) from the plankton,
but it was not until the 1980s that experiments were
executed to assess how the supply of propagules influ
enced the patterns of distribution and abundance of adults
in benthic assemblages.
Propagule supply and early post settlement mortality
markedly influence both the strength of interactions among
established individuals and overall patterns of distribution
and abundance on rocky shores. Abundance of established
individuals is often directly proportional to the density of
settlement and consequently, and strength of adult inter
actions depends on variation of settlement. In contrast, if
settlement is high enough to consistently saturate the sys
tem, then local populations tend to be driven by strong
interactions among adults regardless of settlement varia
tion. In some cases, heavy early postsettlement mortality
can lead to low densities of adults despite an abundance of
settlers, and this has been shown for several seaweeds and
many invertebrate species. The causes of variation in pro
pagule supply can be classified into two broad categories –
oceanographic transport or regional offshore production.
Although invertebrate larvae and some macroalgal spores
are motile, their movements are most directly important
at small spatial scales near the substrate just prior to settle
ment. By and large, propagules of benthic species are
transported at the mercy of currents and other oceanic
transport phenomena. For instance, coastal upwelling
results in a net offshore transport of propagules and leads
to a reduction in settlement along a shoreline. This com
monly occurs with invertebrate species that have long
residence times in the plankton. In contrast, seaweeds,
which have very short planktonic stages, often dominate
intertidal sites within regions characterized by seasonal or
permanent upwelling (Figure 5).
Regional offshore production influences the supply
of larvae to a coastal habitat in two ways. First, phyto
plankton production in nearby waters offshore affects
the abundance of planktotrophic larvae that feed for
several weeks in the plankton potentially leading to
greater larval supply in areas with greater phytoplank
ton production. Second and in opposition, increased
production in offshore can generate increased resources
and habitat for the associated pelagic community that
preys upon larvae and thus leads to a reduced larval
supply.
Figure 5 The intertidal zone near Antofagasta in northern Chile,
a region with upwelling and abundant seaweeds. Photo by
P. S. Petraitis.
Unresolved Problems and Future
Directions
Marine ecologists have been remarkably successful in
advancing our knowledge of how strong local interactions
affect the composition of communities, yet it is not yet
clear how the results of small scale experiments can be
scaled up into broad scale generalizations. This is one
of the major challenges of rocky intertidal ecology
since practical, everyday concerns of management, com
mercial harvesting, biodiversity, and restoration demand
answers on the scale of square kilometers of habitat,
not square meters of experimental site. One current
approach has been to use teams of researchers undertake
identical small scale experiments over a broad geographi
cal region (e.g., EuroRock in Great Britain and Europe) or
over similar oceanographic conditions (e.g., the ongoing
studies of rocky shore in upwelling systems on the Pacific
Rim by PISCO). Another approach has been the integra
tion of ‘real time’ physical, chemical, biological data from
in situ and remote sensors (e.g., satellites that can reveal
near shore temperature and primary productivity) with
experimental studies on community dynamics.
Neither approach solves the difficulties of working
with large mobile consumers such as mammals, whose
importance is under appreciated because of the difficul
ties inherent with studying mammals. Even the rat (Rattus
norvegicus) – the most widely recorded introduced inter
tidal mammal with the broadest documented intertidal
diet – likely remains underreported as a rocky intertidal
consumer from many coastal locations where it is known
to be established. It is likely that rocky intertidal organ
isms supply terrestrial consumers significant amounts of
energy, yet there are few data on intertidal–terrestrial
380
Saline and Soda Lakes
linkages and how intertidal shores serve as important
subsidies for terrestrial habitats.
It is also unclear if detailed information from one
area can be informative about another area. For exam
ple, rocky intertidal shores on both sides of the Atlantic
Ocean look surprisingly alike with not only the same
species of plants and animals present but also similarities
in their abundances and distributions. The similarity is
so striking that a good marine ecologist, knowing little
more than the direction of the prevailing swells, can list
the 20 most common species on any 100 m stretch of
shoreline. The average beachcomber could not tell if he
or she were in Brittany, Ireland, Nova Scotia, or Maine.
The causes of this similarity are not well understood.
Rocky shores in Europe and North America may look
similar because of strong biological interactions maintain
species in balance or because of historical accident,
and these opposing views are endpoints on a continuum
but represent one of the major intellectual debates in
ecology today.
Finally ecosystems are not static, and rocky intertidal
systems, which lie at a land–sea boundary, will be doubly
affected by climate change as both oceanic conditions
such as storm frequency and surge extent, and terrestrial
conditions, such as air temperatures, are altered. Such
changes could affect local communities by altering the
disturbance dynamics and changing the geographic limits
of intertidal species.
See also: Saline and Soda Lakes; Salt Marshes.
Further Reading
Connell JH (1961) The influence of interspecific competition and other
factors on the distribution of the barnacle Chthamalus stellatus.
Ecology 42: 710 723.
Denny MW (1988) Biology and Mechanics of the Wave Swept
Environment. Princeton, NJ: Princeton University Press.
Graham LE and Wilcox LW (2000) Algae. Upper Saddle River, NJ:
Prentice Hall.
Horn MH, Martin KLM, and Chotkowski MA (eds.) (1999) Intertidal
Fishes: Life in Two Worlds. San Diego, CA: Academic Press.
Koehl MAR and Rosenfeld AW (2006) Wave Swept Shore: The Rigors
of Life on a Rocky Coast. Berkeley, CA: University of California
Press.
Levinton JS (2001) Marine Biology. New York: Oxford University Press.
Lewis JR (1964) The Ecology of Rocky Shores. London: English
Universities Press.
Little C and Kitching JA (1996) The Biology of Rocky Shores. New York:
Oxford University Press.
Moore PG and Seed R (eds.) (1986) The Ecology of Rocky Coasts.
New York: Columbia University Press.
Ricketts EF, Calvin J, and Hedgpeth JW (1992) Between Pacific Tides,
5th edn., revised by Phillips DW. Stanford, CA: Stanford University
Press.
Southward AJ (1958) The zonation of plants and animals on rocky sea
shores. Biological Reviews of the Cambridge Philosophical Society
33: 137 177.
Stephenson TA and Stephenson A (1972) Life between Tidemarks on
Rocky Shores. San Fransisco, CA: W. H. Freeman.
Underwood AJ (1979) The ecology of intertidal gastropods. Advances in
Marine Biology 16: 111 210.
Underwood AJ and Chapman MG (eds.) (1996) Coastal Marine Ecology
of Temperate Australia. Sydney: University of New South Wales
Press.
Underwood AJ and Keough MJ (2001) Supply side ecology: The nature
and consequences of variations in recruitment of intertidal
organisms. In: Bertness MD, Gaines SD, and Hay ME (eds.) Marine
Community Ecology, pp. 183 200. Sunderland, MA: Sinauer
Associates.
Saline and Soda Lakes
J M Melack, University of California, Santa Barbara, Santa Barbara, CA, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Geographic Aspects
Environmental and Biological Characteristics
Examples of Ecological Processes
Economic Aspects
Further Reading
Introduction
lakes, many are in remote locations and require explora
tory sampling as a first step, often with surprising findings.
For example, a trans Saharan expedition discovered iso
lated villagers eating cakes of an alga called Spirulina that
has led to an aquaculture industry.
Since 1979 a series of eight international symposia on
inland saline lakes have served to strengthen and expand
the scope of scientific understanding and foster a world
wide cadre of researchers. While distinctive because of
their chemical conditions and biota, all ecological
Saline lakes occur on all continents. Lying in hydrologi
cally closed basins where evaporation exceeds local
precipitation, their size and salinity varies markedly and
they are particularly susceptible to climatic variations and
water diversions. Aquatic biota from microbes to inverte
brates to fish and birds frequent these environments and
can attain spectacular numbers. While modern scientific
techniques are increasingly being applied to a few saline
Saline and Soda Lakes 381
processes occur in saline lakes and they provide an excel
lent system in which to observationally and
experimentally examine these processes. A treatise by
Ted Hammer and synthetic reviews by several others
offer comprehensive information about these diverse
and fascinating environments. This is especially impor
tant because inland saline waters are threatened in many
regions by diversion of their inflows and economic
development.
Geographic Aspects
Saline lakes are widespread globally and occur pre
dominately in dry areas, regions that occupy about
30% of the world’s landmass. The volume of water
in saline lakes is about 80% as large as that in fresh
water lakes. Though about 70% of the total volume of
saline water is held in the Caspian Sea, it is worth
noting that about 40% of the freshwater in lakes is
held in Lake Baikal and the Laurentian Great Lakes
(see Freshwater Lakes). Further, many of the world’s
largest lakes are saline and include Great Salt Lake
(USA), Lake Shala (Ethiopia), Lake Van (Turkey), the
Dead Sea, Qinghai and Lop Nor (China), Nan Tso
(Tibetan Plateau, China), Balkhash (Russia), Urmia
(Iran), Issyk kul (Kyrgystan), the Aral Sea, Mar
Chiquita (Argentina), and Lake Eyre (Australia) and
Salar of Uyuni (Bolivia) (these two lakes vary greatly
in size, as is typical of many shallow playas).
Environmental and Biological
Characteristics
Lakes with salinities above 3 g l 1 are usually consid
ered saline, though this value is somewhat arbitrary.
Salinity is defined as the sum of total ions by weight
and usually includes the major cations (sodium, potas
sium, calcium, and magnesium) and anions
(bicarbonate plus carbonate, chloride, and sulfate).
Natural waters can attain salinities of several hundred
grams per liter and vary considerably in their chemi
cal composition. The ionic composition of saline lakes
depends on the ionic ratios in the inflows and extent
of evaporative concentration. As the saturation of spe
cific salts is exceeded, they precipitate and can lead to
the formation of large evaporite deposits. Typically,
calcium and magnesium carbonates are the first
minerals to precipitate. If sufficient calcium remains
in solution, calcium sulfate often precipitates next. In
the most concentrated waters, chloride is the dominant
anion and sodium is usually the dominant cation;
Great Salt Lake in Utah (USA) is such an example.
In rare cases, other combinations of ions can occur in
Figure 1 Lake Mahega, Uganda.
highly concentrated waters such as the sodium–mag
nesium–chloride waters of the Dead Sea, the sodium–
chloride–sulfate brine in Lake Mahega, Uganda
(Figure 1), or the exceptional calcium chloride brine
in Don Juan Pond (Antarctica). Lakes of intermediate
salinities include the sodium carbonate or soda lakes
of eastern Africa and the triple salt waters (sodium
carbonate–chloride–sulfate) of Mono Lake, California,
USA.
A considerable diversity of halophilic microorganisms
with representatives from the three domains of life, the
Archaea, Bacteria, and Eukarya, inhabit saline lakes. Only
recently have modern molecular techniques, such as gene
sequencing, been applied to natural communities of
microbes and much remains to be learned. At especially
high salt concentrations, microbes lack grazers and can
attain very high abundances that can color saline lakes
bright red or orange. Only a very few metabolic processes
have not been observed at high salinities and these
include halophilic methanogenic bacteria able to use
acetate or hydrogen plus carbon dioxide and halophilic
nitrifying bacteria.
As salinity increases in inland waters, biodiversity
tends to decrease, but in the mid range of salinities
other factors cause considerable variation in species
diversity. The strongest relationship between species
richness of plants, algae, and animals occurs, generally,
below a salinity of about 10 g l 1. An investigation by
William D. Williams, an Australian professor who pio
neered studies of saline lakes, found that species richness
of macroinvertebrates in Australian lakes highly corre
lated with salinity over a salinity from 0.3 to 343 g l 1 but
nonsignificant over intermediate ranges of salinity. Many
taxa had broad tolerances of salinity at the intermediate
values. Instead, a variety of other factors, including dis
solved oxygen concentrations, ionic composition, pH, and
biological interactions, appear to influence species rich
ness and composition.
382
Saline and Soda Lakes
Examples of Ecological Processes
The very wide range of environmental conditions and
geographic distribution of saline lakes results in a large
variety of biological communities with differing species
diversity and ecological interactions. Moreover, few saline
lakes have been examined sufficiently with a combination
of field observations and measurements of important pro
cesses, experiments, and models. Hence, three examples of
ecological processes in saline lakes that are reasonably well
studied and that span a wide range of physicochemical and
biological conditions are presented in this article.
Eastern African Soda Lakes
Figure 3 Lake Elmenteita, Kenya.
Saline lakes rich in bicarbonate and carbonate, usually
called soda lakes, are widespread in eastern Africa and are
among the world’s most productive, natural ecosystems. A
conspicuous feature of these lakes is often the presence of
enormous numbers of lesser flamingos (Phoeniconaias minor)
(Figure 2) grazing on thick suspensions of the phytoplank
ter, Arthrospira fusiformis (previously called Spirulina
platensis), but species diversity is low. Heterotrophic bacteria
attain very high numbers, but have not been characterized
with molecular methods. Phytoplankton and benthic algae
include several species of green algae, diatoms, and cyano
bacteria. Of the few species of aquatic invertebrates,
protozoa are the most diverse with 21 species reported
from Lake Nakuru (Kenya). Consumers in Lake Nakuru
at salinities of around 20 g l 1 include one species of fish,
Sarotherodon alcalicus grahami (introduced from springs near
Lake Magadi, a neighboring salt pan), one copepod
(Paradiaptomus africanus), and two rotifers (Brachionus dimi
datus and B. plicatilis), and several aquatic insects including
corixids, a notonectid, and chironomids. Modest changes in
the salinity and in the vertical distribution of salinity can
have major impacts on trophic structure and nutritional
status of these lakes.
Biological communities in shallow, tropical saline
lakes are susceptible to slight variations in water balances
and salinities. For example, intensive sampling during a
period of low rainfall and abrupt increase in salinity in
Lake Elmenteita (Kenya) (Figure 3) and Lake Nakuru
(Kenya) revealed a precipitous drop in the abundance of
phytoplankton and major shift in the zooplankton. As
species of phytoplanktons, such as Arthrospira fusiformis,
were replaced by much smaller phytoplanktons, the
abundance of lesser flamingos decreased markedly.
Scattered across eastern Africa are numerous saline lakes
inside volcanic craters. Several of these lakes have been
studied in Ethiopia, Kenya, Uganda, and Tanzania. One
common feature in the saline, crater lakes of eastern Africa
is persistent chemical stratification, that is, they are mero
mictic, which has significant biological consequences. For
example, Lake Sonachi (Kenya), a meromictic crater lake,
had much lower algal biomass and rates of photosynthesis
than the neighboring soda lakes that mixed more often.
Moreover, studies of phosphorus uptake indicated that the
lake was deficient in phosphorus, although a large reservoir
of phosphorus was trapped below the chemocline.
Mono Lake
Figure 2 Flamingos (Lake Bogoria, Kenya).
One of the most thoroughly studied saline lakes is Mono
Lake, which lies on the western edge of the North
American Great Basin just east of the Sierra Nevada.
With recent salinities in the range from 70 to 90 g l 1, a
pH of about 10 and very high concentrations of bicarbo
nate and carbonate, it is an alkaline, saline lake. As is often
typical of saline lakes, Mono Lake is productive: rapidly
growing algae support a simple food web that includes
very abundant brine shrimp, Artemia monica, and an alkali
fly, Ephydra hians, which in turn feed thousands of birds.
No fish occur in the lake. The lake is a major breeding site
for the California gull (Larus californicus), and is a critical
stop over for migrating phalaropes (Phalaropus spp.) and
Saline and Soda Lakes 383
eared grebes (Podiceps nigricollis). The streams that flow
into Mono Lake from the Sierra Nevada are a plentiful
source of freshwater that were tapped by the City of Los
Angeles by a complex diversion scheme initially imple
mented in 1941. Largely as a consequence of this
interbasin transfer of water, the lake’s level had fallen
about 14 m and its salinity doubled from 1941 to 1982.
Laboratory experiments indicated that further increases
in salinity were likely to have profound impacts on the
ecology as photosynthesis was found to decline about
10% for each 10% increase in salinity, and survival and
reproduction of the brine shrimp was found to be increas
ingly impaired to the point where cyst hatching would
cease if salinities were to increase by about 50% from
their 1980 values. If diversions by Los Angeles were to
have continued unabated, this salinity would have been
reached within several decades. The end result in the
mid 1990s of almost two decades of litigation and envir
onmental assessment was modifications to the water
rights of the City of Los Angeles, which led to higher
lake levels. In contrast to the dismal conditions at a
number of saline lakes, such as the Aral Sea, and continu
ing declines in level at other lakes, such as Walker Lake
(Nevada), the resolution of the contest at Mono Lake is a
good example of how scientific expertise can contribute
in a positive way to solutions of environmental problems.
As was observed in eastern African soda lakes, cli
matic variations as well as diversions have significant
influences on Mono Lake. In the early 1980s, California
experienced substantially above average snow and rain
fall resulting in a large rise in lake level and chemical
stratification that blocked the complete vertical mixing
that usually occurred during the winter. Ammonium,
which would have been replenished in the upper lake,
accumulated in the deep water, but remained very low
in the euphotic zone. Since Mono Lake is a nitrogen
limited lake, phytoplankton abundance and productivity
declined. The combination of resumed diversions and
drought conditions led to sufficient evaporative concen
tration to weaken the chemical stratification and permit
wind and cooling to turn over the lake in the late 1980s,
entrain ammonium rich water, and restore higher algal
biomass and productivity. After a series of years with
winter mixing and average productivity in the early
1990s, diversions were curtailed in the mid 1990s, as
ordered by the revised water rights agreement, and
California experienced above average precipitation.
Mono Lake became meromictic again with subsequent
reductions in productivity. Multiyear records of annual
primary productivity by phytoplankton have conspicu
ous differences as a function of meromictic or
monomictic conditions. During meromixis, the develop
ment of persistent anoxia below the chemocline alters
other chemical conditions with biological consequences.
Methane and dissolved sulfide accumulate, and bacterial
communities adapted to metabolize reduced forms of
elements become active.
Artemia monica is the only macrozooplankter in Mono
Lake. Each year a first generation hatches from overwin
tering cysts, matures, and produces a second generation
via release of live nauplii. A small third generation some
times occurs, but very few animals survive through the
winter. Besides exerting strong grazing pressure on the
phytoplankton, Artemia regenerate ammonium that sup
ports algal growth. Artemia are an important food for large
numbers of gulls breeding at the lake in the spring and for
as many as one million grebes in the autumn. Some life
history characteristics of Artemia are indicative of differ
ences in algal abundance and primary productivity.
Although large numbers of eggs are produced in all
years, on average, fewer cysts and live nauplii are pro
duced during meromictic years, and maturation of the
first generation can be slowed and fecundity and body
size reduced as compared to nonmeromictic years.
Changes in the Artemia populations translate to influences
on the birds feeding at the lake. The fledging rate per pair of
California gulls reflects their clutch size and prefledging
survival, both of which should be influenced by the adult
food supply. In fact, fledging success was low immediately
following the onset of meromixis and remained low during
the subsequent 3 years of meromixis in the 1990s.
Dead Sea
Lying about 400 m below sea level in a rift valley along the
Israel–Jordan border, the surface of the Dead Sea is the
lowest of any lake, and it is one of the saltiest with a current
salinity of around 340 g l 1. Diversions of the Jordan River,
the main inflow, resulted in a 20 m decline in lake level
over the last century and an increase in salinity. One
consequence of the evaporative concentration of the
upper waters was the termination of meromixis that had
persisted for several hundred years. With the exception of a
few years, the lake now mixes completely each year.
At the time of the pioneering microbiological studies
by Benjamin Elazari Volcani in the 1930s and 1940s, the
lake’s salinity was about 260 g l 1. Using enrichments and
microscopy he was able to describe a variety of halophilic
and halotolerant microbes as well as the phytoflagellate,
Dunleilla viridis, several cyanobacteria, diatoms, green
algae, and a ciliate. Subsequent application of modern
molecular techniques has considerably expanded the
number of microbes, but the higher salinities have elimi
nated some organisms noted earlier.
During times when the whole lake reaches salinities of
around 340 g l 1, bacterial densities are low and algae are
absent. However, in response to periods with large
amounts of rainfall and runoff, the upper waters can be
diluted to as low as 250 g l 1, and blooms of Dunaliella and
red Archaea develop. The abrupt decline of the bacterial
384
Salt Marshes
blooms cannot be attributed to protozoan grazing, since
these organisms no longer occur, and may be caused by
bacteriophages, as viruses have been identified in the lake.
Economic Aspects
The salts precipitated from saline waters are a rich source
of chemicals used in a variety of industrial processes and
are mined from salt lakes. In coastal areas with high
evaporation rates, a series of salterns allow progressive
concentration of solutes and the production of useful salts.
In a few saline lakes with strong chemical stratification,
transparent surface waters and a turbid layer within the
chemocline, high temperatures have been recorded in the
turbid layer. These features have guided the construction
of artificial, so called solar ponds, with similar character
istics, for power production and heating purposes.
A common feature of tropical African soda lakes is high
concentrations of nearly unialgal populations of the cyano
bacteria, Arthrospira fusiformis, which support huge numbers
of lesser flamingos and are used as a protein rich food by
people in Chad. These observations, laboratory studies, and
development of mass culture methods have led to
Arthrospira, often marketed as Spirulina, becoming a widely
used food supplement. Other species of algae found in saline
waters are commercially exploited because of their high
glycerol or carotene content (e.g., Dunaliella). Additional
applications include the production of salt resistant enzymes
and the use of organic osmolytes to protect enzymes.
Artemia are an important food for aquaculture of some
fish and other organisms. Typically, cysts are harvested
from lakeshores and maintained dry until needed, when
they are readily hatched by submerging in saline water.
Occasionally, such as at Mono Lake, adult Artemia are
collected, frozen, and shipped to aquaculture facilities.
The impressive numbers of birds that frequent saline
waters and the striking scenery has led to tourism as an
increasingly important aspect of their economic value.
World famous examples include Lake Nakuru, with it
shoreline fringed by pink flamingos, Mono Lake with its
peculiar tufa towers and thousands of waterfowl, and the
Dead Sea with its historical significance and highly buoy
ant water. Some less saline lakes, such as Pyramid Lakes,
harbor fish (e.g., Lahonton cutthroat trout, Oncorhynchus
clarki henshawi) that support recreational fishery.
See also: Freshwater Lakes.
Further Reading
Eugster HP and Hardie LA (1978) Saline lakes. In: Lerman A (ed.) Lakes:
Chemistry, Geology, Physics, pp. 237 293. New York: Springer.
Hammer UT (1986) Saline Lake Ecosystems of the World. Dordrecht:
Dr. W. Junk Publishers.
Melack JM (1983) Large, deep salt lakes: A comparative limnological
analysis. Hydrobiologia 105: 223 230.
Melack JM (2002) Ecological dynamics in saline lakes. Verhandlungen
Internationale Vereinigung Limnologie 28: 29 40.
Melack JM, Jellison R, and Herbst D (eds.) Developments in
Hydrobiology 162: Saline Lakes. Dordrecht: Kluwer.
Oren A (ed.) (1999) Microbiology and Biogeochemistry of Hypersaline
Environments. New York: CRC Press.
Vareschi E and Jacobs J (1985) The ecology of Lake Nakuru. VI.
Synopsis of production and energy flow. Oecologia
65: 412 424.
Williams WD (1996) The largest, highest and lowest lakes in the world:
Saline lakes. Verhandlungen Internationale Vereinigung Limnologie
26: 61 79.
Salt Marshes
J B Zedler, C L Bonin, D J Larkin, and A Varty, University of Wisconsin, Madison, WI, USA
ª 2008 Elsevier B.V. All rights reserved.
Physiography
Extent
Habitat Diversity
Salt Marsh Plants
Salt Marsh Animals
Ecology
Ecosystem Services
Challenges for Salt Marsh Conservation
Research Value
Restoration
Further Reading
Physiography
relatively flat topography), herbaceous vegetation, and
diverse invertebrates and birds. They occur along shores
in estuaries, lagoons, forelands (open areas), and barrier
islands in marine environments, and in shallow inland
sinks where salts accumulate. They are not found where
Salt marshes are saline (typically at or above seawater,
>34 g l 1) ecosystems with characteristic geomorphology
(sedimentary environments, fine soil texture, and
Salt Marshes
waves, currents, or streamflow create strong erosive
forces. Salt (which stresses most species) severely limits
the pool of plant species that can colonize saline sedi
ments, and wetness typically confines the vegetation to
herbaceous species, although some species are long lived
‘subshrubs’. Given a near surface water table, most
shrubs and trees cannot establish their extensive root
systems.
Plants of tidal marshes are usually able to colonize
sediment above mean high water during neap tides
(MHWN ¼ average higher high tide level during
lower amplitude neap tides, which alternate with the
broader amplitude spring tides). Sediment stabilization
by halophytes initiates salt marsh formation. Plants not
only slow water flow and allow sediments to settle out,
but also their roots help hold sediments in place.
Gradual accretion around plant shoots can further ele
vate the shoreline, allowing development of a marsh
plain and transition to upland. This process can reverse,
with tides eroding accumulated sediments. When sedi
mentation is outweighed by erosion, salt marshes
retreat.
The overriding physiochemical influence is salt,
which comes from marine waters, from exposed or
uplifted marine sediments, or from evaporation of
low salinity water in arid region sinks. Salt marshes
along coasts typically have tidal influence (Figure 1),
although many nontidal lagoons have saline shores
that support salt marsh vegetation. Salt marshes in
inland settings occur in shallow sinks (e.g., around
the Great Salt Lake, Utah, USA). The salts that con
tribute to salinity are primarily those of four cations
(sodium, potassium, magnesium, calcium) and three
anions (carbonates, sulfates, and chlorides); the relative
proportions differ widely among soils of inland salt
marshes, but sodium chloride is the predominant salt
of seawater.
385
Tidal regimes differ around the globe, but most
tidal marshes experience two daily high tides of
slightly different magnitude, while some have the
same high and low tides from day to day. Levels
alternate weekly as neap and spring amplitudes, with
the amplitudes readily predicted given gravitational
forces between the Earth, the Moon, and the Sun
(astronomic tides). Forces vary in relation to global
position and coastal morphology; in southern
California, mean astronomic tidal range is 3 m, while
in the Bay of Fundy it is 16 m. The influence of
seasonal low and high pressure systems on water
level oscillations (atmospheric tides) also vary greatly.
For example, in Western Australia’s Swan River
Estuary, atmospheric tides outweigh astronomic tides.
In the Gulf of Mexico, astronomic tides are minimal
because of limited seawater connection with the
Atlantic Ocean. Water levels within the Gulf vary
only a few centimeters except during storms and
seiches.
In tidal systems, marsh vegetation generally ranges
from MHWN to the highest astronomic tide.
Depending on tidal amplitude and the slope of the
shore, salt marshes can be very narrow or kilometers
wide. Strong wave action limits the lower salt marsh
boundary, but a sheltered area can extend the lower
boundary below MHWN.
Animal diversity is high, especially among the
benthic and epibenthic invertebrates and the arthro
pods in the soil or plant canopies. Species that
complete their life cycles within salt marshes either
tolerate changing salinity and inundation regimes or
avoid them by moving elsewhere or reducing contact.
Globally, salt marshes are known to support large
populations of migratory birds in addition to resident
birds, insects, spiders, snails, crabs, and fin and shell
fish. Indeed, foraging is the most visible activity in salt
marshes.
Extent
Figure 1 A tidal marsh in San Quintin Bay seen from the air.
Image by the Pacific Estuarine Research Lab.
Salt marsh area is not well inventoried. The global extent
of pan, brackish, and saline wetlands is approximately
435 000 km2, or 0.3% of the total surface area and 5% of
total wetland area. In USA, the 48 conterminous states
have about 1.7 Mha of salt marshes, out of a total of
42 Mha of wetlands.
While broadly distributed, salt marshes are most
common in temperate and higher latitudes where the
temperature of the warmest month is >0 C. Closer to
the equator, where the mean temperatures of the
coldest months are >20 C, salt marshes are generally
replaced by mangroves. Salt marshes sometimes occur
386
Salt Marshes
inland of mangroves or instead of mangroves where
woody plants have been removed.
Habitat Diversity
Habitats within the salt marsh vary with elevation,
microtopography, and proximity to land or deeper
water. In southern California, the high marsh, marsh
plain, and cordgrass (Spartina foliosa) habitat tend to
follow elevation contours, although cordgrass is often
restricted to low elevations adjacent to bay and chan
nel margins. Other habitats are related to minor
variations in topography, which impound fresh or
tidal water. For example, back levee depressions,
tidal pools, and salt pans occur where drainage is
somewhat impaired. Salt marshes along the Atlantic
Coast of USA are very extensive, with S. alterniflora
creating a monotype except for a narrow transition at
the inland boundary where succulent halophytes or
salt pans are found.
Tidal creeks provide diverse habitats for plants and
animals. Banks are often full of crab burrows, and creek
bottoms harbor burrowing invertebrates and fishes. They
also serve as conduits for fish, fish larve, phyto and
zooplankton, plant propagules, sediments, and dissolved
materials, which move between the salt marsh and sub
tidal channels.
Adjacent habitats can include small, unvegetated
salt pans that dry and develop a salt crust, especially
during neap tides. Salt pans occur where salt concen
trations exceed tolerance of halophytes. During heavy
rains or high tides, water fills the pan, creating tem
porary habitat for aquatic algae and animals and
permanent habitat for the species that survive the
dry spells in situ as resting stages. More extensive
salt pans are sometimes called salt flats. Other nearby
habitats usually include mudflats (where inundation
levels exceed tolerance of halophytes), brackish
marsh (where salinities are low enough for brackish
plants to outcompete halophytes), sandy or cobble
beaches (where wave force excludes herbaceous vege
tation), sand dunes (where soils are too coarse and dry
for salt marsh plants), and river channels (where fresh
water enters the estuary and is not sufficiently saline).
Salt Marsh Plants
Salt tolerant plants (halophytes) include herbaceous
forbs, graminoids, and dwarf or subshrubs. Many of
the forbs are succulent (e.g., Sarcocornia and Salicornia
spp.). Graminoids often dominate Arctic salt marshes,
while subshrubs dominate salt marshes in
Mediterranean and subtropical climates. Many salt
marshes support monotypic stands of cordgrass
(Spartina spp.) (Table 1).
Floristic diversity of salt marshes is low because few
species are adapted to saline soil. Members of the family
Chenopodiaceae comprise a large proportion of the flora
(e.g., species of Arthrocnemum, Atriplex, Chenopodium,
Salicornia, Sarcocornia, and Suaeda). In contrast to the flow
ering plants, salt marsh algae are diverse in both species
and functional groups (green macroalgae, cyanobacteria,
diatoms, and flagellates).
NaCl is a dual stressor, as it challenges osmotic reg
ulation and sodium is toxic to enzyme systems. Salt
marsh halophytes cope with salt by excluding entry
into roots, sequestering salts intracellularly (leading to
succulence), and excreting salt via glands, usually on leaf
Table 1 Representative species of global salt marshes based on a summary by Paul Adam
Arctic Puccinellia phryganodes dominates the lower elevations
Boreal Triglochin maritima and Salicornia europea are widespread. Brackish conditions have extensive cover of Carex spp.
Temperate
Europe: Puccinellia maritima dominated lower elevations historically (but Spartina anglica often replaces it). Juncus maritimus
dominates the upper marsh; Atriplex portulacoides is widespread
USA:
Atlantic Coast: Spartina alterniflora is extensive across seaward marsh plain; S. patens occurs more inland
Gulf of Mexico: Spartina alterniflora and Juncus roemerianus dominate large areas
Pacific Northwest: Distichlis spicata in more saline areas, Carex lyngbei in less saline areas
California: Spartina foliosa along bays, Sarcocornia pacifica inland
Japan: Zoysia sinica dominates the mid-marsh
Australasia: Sarcocornia quinqueflora dominates the lower marsh, Juncus kraussii the upper marsh
South Africa: Sarcocornia spp. are abundant in the lower marsh, Juncus kraussii in the upper marsh. Spartina maritima is sometimes
present
Dry coasts vegetation tends toward subshrubs, such as Sarcocornia, Suaeda, Limoniastrum, and Frankenia species
Tropical Sporobolus virginicus and Paspalum vaginatum form extensive grasslands. Batis maritima, Sesuvium portulacastrum, and
Cressa cretica are also found
Salt Marshes
surfaces. One succulent, Batis maritima, continually drops
its older salt laden leaves, which are then washed away
by the tide. I. Mendelssohn has attributed moisture
uptake from seawater to the ability of some species to
synthesize prolines.
Prolonged inundation reduces the supply of oxygen to
soils, causing anoxia and stressing vascular plants. In addi
tion, abundant sulfate in seawater is reduced to sulfide in salt
marsh soil, with high sulfide concentrations, which are toxic
to roots.
Salt marsh vascular plants withstand brief inundation
but do not tolerate prolonged submergence, as occurs
when a lagoon mouth closes to tidal flushing and water
levels rise after rainfall. Salt marshes in lagoons thus
experience irregular episodes of dieback and regeneration
in relation to ocean inlet condition.
Regular inundation benefits halophytes by importing
nutrients and washing away salts. Salts that accumulate on
the soil surface during daytime low tides and salts
excreted by halophytes are removed by tidal efflux.
Thus, soil salinities are relatively stable where tidal inun
dation and drainage occur frequently. Inland salt marshes,
however, experience infrequent reductions in salinity
during rainfall, and soils can become extremely hypersa
line (e.g., >10% salt). In between irregular inundation
events, halophytes and resident animals endure hypersa
line drought.
Salt Marsh Animals
The salt marsh fauna includes a broad taxonomic
spectrum of invertebrates, fishes, birds, and mammals,
but few amphibians and reptiles. Resident fauna are
adapted to the land–sea interface, while transient users
benefit from the foraging, nursery, and reproductive
support functions.
Salt marsh animals cope with inundation regimes
that differ seasonally, monthly, daily, and hourly.
Vertebrates accomplish this largely through mobility.
For example, fishes exploit marsh surface foraging
opportunities during high tides and then retreat to
subtidal waters. Birds time their use to take advantage
of either low or high tide. Residents, such as the light
footed clapper rail (Rallus longirostris levipes), nest dur
ing the minimum tidal amplitude. Migrants, such as
curlews, move upslope at a high tide and feed during
low tide during their seasonal visits. Many inverte
brates move away from adverse conditions. Some
beetles climb tall plants to escape rising tides.
A springtail, Anurida maritime, has a circatidal rhythm
of 12.4 h that enables it to emerge for feeding shortly
after tides ebb and retreat underground prior to the
next inundation. For less mobile fauna, physiological
adaptations are essential. Gastropods avoid desiccation
387
during low tides by sealing their shells. Some arthro
pods avert drowning by trapping air bubbles in their
epidermal hairs during high tides.
Another challenge is fluctuating salinities, which
salt marsh residents handle with exceptional osmore
gulatory ability. The southern California intertidal
crab species Hemigrapsus oregonensis and Pachygrapsus
crassipes are able to hypo and hyperosmoregulate
when exposed to salt concentrations ranging from
50% to 150% of seawater (brackish to hypersaline).
Tidal marsh fishes also have wide salinity tolerances.
Cyprinodontiform tidal marsh fishes can tolerate sali
nities as high as 80–90 ppt. One species, Fundulus
majalis, hatched at salinities up to 72–73 ppt. Lower
salinity limits for mussels can be as low as 3 g l 1 and
they can tolerate high salinities as well, with mussels
able to tolerate losing up to 38% of their water con
tent. Even birds have adaptations for dealing with salt
water and saline foods; for example, the Savannah
sparrow (Passerculus sandwichensis beldingi) has specia
lized glands that excrete salt through the nares.
Because salt marshes have continuously changing
hydrology, small differences in elevation and topogra
phy (e.g., shallow, low order tidal creeks) influence
foraging activities of fishes and birds by regulating
inundation and exposure times, enhancing marsh
access for fishes, and increasing edge habitat.
Ephemeral pools of just centimeters in depth provide
valuable bird habitat, enhance macroinvertebrate
abundance and diversity, and support reproductive,
nursery, and feeding support functions for fishes.
Ecology
Salt marshes are well studied relative to their limited
global area. Knowledge of salt marsh ecology is strongest
for vegetation, soil processes, and food webs. Conservation
is an emerging issue, given threats of sea level rise in
concert with global warming.
Vegetation and Soils
In Europe, salt marsh ecology developed around floristics
and phytosociology. In USA, research on the Atlantic
and Gulf Coasts characterized salt marsh ecosystem func
tioning, especially productivity, microbial activities,
outwelling of organic matter, food webs, and support of
commercial fisheries, while on the Pacific Coast, studies
concern the impacts of invasive species of Spartina and
effects of extreme events on vegetation dynamics. In
Canada, effects of geese damaging vegetation are a
research focus. Studies of USA’s inland salt marshes
388
Salt Marshes
have contributed knowledge of waterfowl support func
tions and halophyte salt tolerance. In South Africa’s small
estuaries, Spartina productivity and shifts of vegetation in
response to altered freshwater inflows have been
explored. In Asia, widespread plantings of S. alterniflora
have been undertaken in order to extend coastal land
area, provide forage, and produce grass for human use.
In general, salt marshes of Asia, Central America, and
South America are poorly known.
Salt marshes develop primarily on fine sediments, but
salt marsh plants can grow on sand and sometimes gravel.
Older salt marshes have peaty soils, especially in cooler
latitudes where decomposition is slow.
Both roots and burrowing invertebrates affect soil struc
ture by creating macropores in soil. Invertebrates also
cause bioturbation, a process whereby sediments are re
suspended and potentially eroded away. This activity can
be countered by algae and other microorganisms, which
form biofilms on the soil surface. Biofilms cement soil
particles and reduce erosion; they also add organic matter,
and those that contain cyanobacteria fix nitrogen.
Salt marsh soils are often anoxic just below the surface
due to high organic matter content and abundant moist
ure for microorganisms. This is especially so in lower
intertidal areas and in impounded marshes. Tidal marsh
soils are typically high in sulfur, which forms sulfides that
blacken the soil, emit a distinctive rotten egg smell, and
stress many plants. Across intertidal elevation ranges, soil
microorganisms, sulfides, and inundation regimes reduce
species richness where inundation is most prolonged,
often to a single, tolerant species.
Food Webs
Studies of salt marshes have made important advances
in food web theory. Early papers focused on primary
productivity measurements and attempts to explain
differences in rates within and among salt marshes.
The energy subsidy model described S. alterniflora’s
high productivity at low elevations as a function of
increased rates of nutrient delivery and waste removal,
due to frequent tidal inundation. It also explained how
salt marshes with decreasing tidal energy across Long
Island, New York, had a corresponding decrease in S.
alterniflora productivity. R. E. Turner added the role of
climate by relating higher productivity of S. alterniflora
to warmer latitudes.
In the 1960s, E. Odum’s interest in energy flow
led several investigators at the University of Georgia to
quantify productivity, consumption, and decomposition of
various components of Sapelo Island salt marshes. J. Teal’s
energy flow diagram depicted Georgia’s S. alterniflora
marsh as exporting organic matter. Although estimated
by subtraction rather than measurement, detrital export
became a textbook example of how ecosystems channel
and dissipate energy.
Later, advances were made in exploring the quantity
and fate of detritus derived from salt marsh primary
producers. J. Teal’s suggestion that substantial organic
matter is transported to estuarine waters supported
E. Odum’s ‘outwelling hypothesis’, that estuarine derived
foods drive coastal food webs and benefit commercial
fisheries. A number of ecosystem scale tests of outwelling
ensued, and although outwelling did not prove to be
universal, the research demonstrated connectivity
between riverine, salt marsh, and open water ecosystems.
Also, the copious detrital organic matter provided by salt
marshes was shown to be high in nutritional value once
detrital particles were enriched by microorganisms, but
microalgae were also shown to be an important food
source. Even though their standing crop is low, high turn
over rates lead to high primary productivity. In salt
marshes with ample light penetration through the vascu
lar plant canopy, microalgae can be as productive as
macrophytes, and some species (notably cyanobacteria)
are much richer in proteins and lipids. Algae also hold
much of the labile nitrogen in salt marshes, widely
thought to be the limiting factor for growth of inverte
brate grazers.
Food webs are driven by both ‘bottom up’ or ‘top
down’ processes. Evidence for bottom up control of
trophic interactions comes from experimental addition
of nitrogen. Nitrogen has been shown to limit algae,
vascular plants, grazers, and predatory invertebrates in
nearly every salt marsh field experiment. Recently, how
ever, P. V. Sundareshwar and colleagues showed that
phosphorus can limit microbial communities in coastal
salt marshes.
Despite widespread evidence for bottom up effects,
there is expanded recognition of the top down role of
consumers in regulating salt marsh food webs. Populations
of lesser snow geese have increased due to agricultural
grains that are left in the fields after harvest, and large flocks
now cause large scale destruction of vegetation in Arctic
salt marshes due to rampant herbivory. In Atlantic salt
marshes of southern USA, snail herbivory accompanies
drought induced die back of S. alterniflora.
Ecosystem Services
Several ecosystem services provided by salt marshes are
appreciated by society, and some protective measures are
in place. The regular rise and fall of water in salt marshes,
either daily with tides or seasonally with rainfall,
enhances at least six valued functions:
Denitrification improves water quality. The sediments of
tidal marshes are well suited to denitrification, which
occurs most rapidly at oxic–anoxic interfaces. The first
Salt Marshes
step, nitrification, occurs near soil–water or root–soil
interfaces or along pores where oxygen enters the soil at
low tide. The second step requires anoxic conditions and
proceeds rapidly where moisture is sufficient for bacteria
to respire and remove oxygen. In this step, nitrate is
reduced to nitrogen gas in a series of microbially
mediated steps. The rise and fall of tide waters ensures
that oxic and anoxic conditions coexist.
Carbon sequestration slows greenhouse warming. The high
net primary productivity of salt marshes creates high
potential for carbon storage and the anoxic soils slow
decomposition, so carbon can accumulate as peat.
Large standing crops of roots, rhizomes, and litter are
fractionated by a diversity of invertebrates and microor
ganisms and incorporated into soil. Rates are potentially
highest at cooler latitudes, where decomposition is slowed
by low temperatures. Sea level rise is also a key factor; as
coastal water levels become deeper, decomposition slows.
Sedimentation also buries organic matter, making it less
likely to decompose. With sea level rising a millimeter or
more per year, on average, salt marsh vegetation can build
new rooting zones above dead roots and rhizomes of past
decades. Along the USA Gulf of Mexico, the ability of salt
marshes to keep up with rising sea level is attributed to
root and rhizome accumulation, not just sedimentation. If
decomposition proceeds anaerobically to states that pro
duce methane, however, not only is carbon storage
reversed, but carbon is also released in a form that con
tributes more to global warming than carbon dioxide.
Fin and shellfish have commercial value. Tidal marshes
are valued for their nursery function, meaning that the
young of many fishes, crabs, and shrimp make use of
estuarine waters as ‘rearing grounds’. In the USA, it is
estimated that some 60% of commercial species spend at
least part of their life cycle in estuaries. Several attributes
of salt marshes contribute to the food web support func
tion, including high productivity of both algae and
vascular plants, detritus production and export to shallow
water feeding areas, refuge from deepwater predators,
plant canopy cover as a refuge from predatory birds,
warmer temperatures that can accelerate growth, and
potential to escape disease causing organisms and para
sites that might have narrower salinity tolerance.
Forage is used to feed livestock. In Europe and Asia, gra
ziers move cattle, horses, sheep, or goats onto the marsh
plain during low tides. It is common to see ponies teth
ered to stakes in Puccinellia dominated salt marshes of UK.
The temporary availability (between tides) allows recov
ery between use and, potentially, high quality forage and
salt for livestock.
Recreational opportunities and esthetics are appreciated by
people who live near or visit coastal areas. By virtue of
their low growing vegetation and locations between
open water and urban areas, salt marshes attract both
wildlife and people. The combination provides high
389
value for birdwatchers, hikers, joggers, and artists.
Where there is flat topography above and near the
salt marsh, the needs of elderly and disabled visitors
can be accommodated along with hikers, school chil
dren, and those seeking a refuge from city life. Of
particular interest is the ever changing view, as tides
rise and fall along marine coasts, and as water levels
change with season in inland systems. Visitor centers
have been constructed near many urban salt marshes.
Ecotourism then adds economic value to the local
municipality as well as the larger region.
Shorelines are anchored by salt marsh vegetation. Recent
damages from hurricanes and tsunamis have called atten
tion to the protection that wetland vegetation provides to
coastal lands, and especially high cost real estate. Water
flow is slowed by stems and leaves of salt marsh plants, and
their roots and rhizomes bind inflowing sediments.
Mucilage produced by biofilms (algae, fungi, and bacteria)
can then cement particles until new plant growth anchors
the substrate. The stems of vascular plants are often coated
with biofilms, particularly those of tuft forming cyanobac
teria, such that the total surface area available for sediment
trapping and anchoring is greatly enhanced. Floating mats
of green macroalgae (Ulva, Enteromorpha) also collect sedi
ments and, when they move to the wrack line and join
other debris, add to accretion at the upper marsh plain
boundary.
Challenges for Salt Marsh Conservation
Habitat Loss
Estuaries, where rivers meet the sea, are not only suitable
for salt marsh development but also ideal places for
human habitation. The ocean–river connection is a navi
gational link, flat land is easy to build upon, the river
provides drinking water, the salt marsh and coastal fish
eries provide food, outgoing tides facilitate wastewater
disposal, and seawater provides an essential preservative
and universal seasoning, NaCl. Thus, many cities, such as
Venice, Boston, Amsterdam, London, Buenos Aires,
Washington, DC, and Los Angeles, were built on or
rapidly grew to displace salt marsh ecosystems. Major
ports within smaller natural bays, such as San Diego,
have displaced nearly all the natural salt marsh, while
others, such as San Francisco, sustain large salt marshes
despite extensive conversion.
The process of converting salt marsh into nontidal
land was historically called reclamation. The practice of
building embankments to exclude tidal flows eliminated
thousands of hectares of European salt marshes. In the
Netherlands, embankments reclaimed substantial land as
polders for agriculture. In USA, reclamation reduced salt
marsh area by 25% between 1932 and 1954. While the
trend is to halt or reverse this practice, estuaries are being
390
Salt Marshes
dammed in Korea to create tillable fields from mudflats.
In Vietnam, Mexico, and other coastal nations, salt
marshes are yielding to fish and shrimp impoundments.
In such cases, people who use mudflats for fishing and
crabbing are displaced by farmers.
Although salt marshes are highly valued, they are
increasingly threatened by human population growth. It
is estimated that 75% of the global population will live
within 60 km of the coast. Thus, coastal ecosystems are
particularly at risk.
Eutrophication
Salt marshes are enriched when phosphorus and/or nitro
gen flow into waters that ultimately flood the salt marsh.
Agricultural fertilizers applied to fields throughout
coastal watersheds move downslope into waters that
flow toward salt marshes. Because many salt marshes are
nitrogen limited, the effect is to increase the productivity
of both algae and vascular plants. Increased nitrogen
loading stimulates algal growth, especially of green
macroalgae, which form large mats that can smother
vascular plants and benthic invertebrates. Indirect degra
dation occurs when microbial decomposition increases
oxygen demand, causing soil hypoxia or anoxia and sul
fide toxicity.
I. Valiela’s long term eutrophication experiment in a
New England salt marsh indicates that nitrogen addition
shifts S. alterniflora to S. patens and increases competition
for light. Such altered competitive relationships are likely
widespread, especially where considerable nitrogen is
deposited from the air (e.g., from dairy operations in the
Netherlands).
Sediment Supply
Both reduced and enhanced sediment supplies can threa
ten the persistence of salt marsh ecosystems. Sediment
supplies are reduced when water is removed from rivers
for irrigation, human consumption, and industrial use, or
when overbank flooding is prevented by engineering
works. Reduced sediment supply from the Mississippi
River is one factor contributing to salt marsh loss in
Louisiana.
Excessive sediments flow into salt marshes where
the catchment has lost vegetative cover as a result of
logging, farming, or development. Inflows also occur
where mining operations discharge materials directly
to streams. Wastes from California’s gold rush are still
making their way to San Francisco Bay. At a much
smaller estuary, the marsh plain of Tijuana Estuary in
southern California has elevated 25–35 cm since 1963
due to erosion from rapidly urbanizing canyons in
nearby Tijuana, Mexico. The impacts have been losses
of microtopographic variation and local species
richness.
Global Change
Increases in global mean temperature will have sub
stantial impacts on the world’s salt marshes. Sea levels
rise when high elevation glaciers and polar ice caps
melt and when seawater warms and expands. The
impacts of more rapidly rising sea level depend on
rates of sedimentation and uplift. If sediment accretion
is equal to sea level rise, the salt marsh remains in
place, but when sea level rise exceeds sediment accre
tion, the salt marsh moves inland – unless bluffs or
development limits salt marsh migration. As sea level
rises relative to the land, salt marsh communities will
experience increased inundation, such that plant and
animal species should shift upslope. However, not all
species will be able to disperse or migrate as rapidly
as tidal conditions change. In a few cases, for example,
Scandinavia, the coast is still rebounding from the
pressure of former glaciers, and land is rising faster
than sea level. Salt marsh is then lost at the upper end
and slowly gained near the water.
Globally, mean sea level has risen 10–25 cm during
the last century. Current models predict an additional
5.6–30 cm rise in sea level by 2040. In areas of rapid
shifts in sedimentation or high erosion due to wind and
waves, salt marshes are destabilized and threatened
with compositional changes and/or loss of marsh area.
Salt marshes are also threatened by subsidence; if the
land settles faster than sediment or roots and rhizomes
can accumulate, vegetated areas convert to open water.
USA’s largest area of salt marsh loss is along the
Louisiana coastal plain, where subsidence, decreased
sedimentation, canal dredging, levee construction, and
other human disturbances eliminate more than
4300 ha yr 1.
Coastal watersheds that experience increased stormi
ness as a result of climate change will discharge water,
sediments, nutrients, and contaminants more erratically
than at present, with resulting impacts on salt marshes
downstream.
Soil salinity might also rise with higher tempera
tures,
increased
evaporation,
and
increased
evapotranspiration. With more rainfall and freshwater
flooding, however, soil salinity might decrease. The net
effect of warming on salt marsh soil salinity is difficult
to predict. Increased storminess could translate into
more or stronger dune washover events during high
tides, and stronger ocean swells would transport
seawater further inland. The toxic effect of salt on
upland vegetation, coupled with persistent salt in
the soil, would favor halophytes over glycophytes in
an increasingly broader wetland–upland transition
Salt Marshes
Figure 2 Saltmarsh vegetation from the upland–wetland
interface (foreground) to San Quintin Bay, Baja California
Peninsula, Mexico. Photo by J. Zedler.
areas (Figure 2). This prediction is most likely for
areas of low annual rainfall, such as Mediterranean
type climates.
Climate change is likely to affect species differen
tly, potentially altering competitive relationships.
Photosynthesis, transpiration, nutrient cycling, phenol
ogy, and decomposition are influenced by temperature.
Salt marshes with a mixture of C3 and C4 plants might
shift toward C4 plants as mean temperature climbs;
however, elevated CO2 might favor C3 species. In sub
tropical regions, a warming trend and sea level rise
would likely allow mangroves to move northward and
displace salt marshes.
Impacts of climate change to plants and animals are
difficult to estimate. European ecologists, however, have
detailed information on bird use of salt marshes and can
predict shifts in invertebrate foods and shorebirds given
various scenarios of sea level rise.
Invasive Species
Plant and animal species are inadvertently moved around
the globe when ships take on ballast water in one port and
discharge it in another; seeds of alien plants and either
live animals or dormant stages are then available to colo
nize salt marshes. When the USA resumed trade with
China, new invaders gained access to San Francisco Bay.
Fred Nichols traced the arrival of a small clam,
Potamocorbula amurensis, to 1876. Now it coats some
benthos with thousands of clams/m2.
Other alien species have been intentionally intro
duced. In the 1950s, the US Army Corps of Engineers
experimentally introduced S. alterniflora onto several
dredge spoil islands to stabilize the material and provide
wildlife habitat. A region wide invasion of the Pacific
Northwestern USA followed several decades of ‘benign’
behavior. Today, the species is dominant along the lower
391
edge of salt marsh shorelines, where it displaces oysters
and eliminates shorebird feeding habitat.
Once a species has taken up residence, it might hybri
dize with native species and become more aggressive,
either as the hybrid or subsequent genetic variants. Such
is the case for S. alterniflora, which has been widely planted
in Europe, China, Great Britain Australia, and New
Zealand. In Great Britain, it hybridized with the native
S. maritima to form S. townsendii, which then underwent
chromosomal doubling to form S. anglica. S. anglica can
grow at lower elevations than native species and vigor
ously colonizes mudflats. Dense clones of S. anglica reduce
habitat for wading birds and displace native salt marsh
plants.
Non native strains of Phragmites australis were
introduced to the USA 200 years ago, and they have
since spread throughout much of North America.
Today, the alien strain dominates the less saline por
tions of salt marshes in the northeastern USA, where it
displaces native plant species, alters soil conditions,
and decreases waterfowl use. Disturbances such as
ditching or dredging open salt marsh canopies and
allow invasion of P. australis, while eutrophication,
altered hydrologic regimes, and increased sedimenta
tion favor its spread.
Invasive plant species have been linked to reduced
diversity, shifts in trophic structure, habitat alteration,
and changes in nutrient cycling. Invasive alien animals
are equally problematic. In San Francisco Bay wetlands,
alien mudsnails outcompete native ones and the
Australasian isopod, Sphaeroma quoyanum, burrows into
and destabilizes creek banks of tidal marshes, causing
erosion. Marsh edge losses exceeding 100 cm yr 1 have
been reported in heavily infested areas. Another invader,
the green crab, Carcinus maenus, has altered the food web
of Bodega Bay, California, by reducing densities of a
native crab, two native mussels, and other invertebrates.
As the green crab moves north, it will likely reduce food
availability for shorebirds.
In the southeastern USA, fur farmers introduced nutria
(Myocastor coypus) from South America in the 1930s. These
rodents feed on roots of salt marsh plants. When fur
clothing went out of style, nutria populations expanded
and began converting large areas of marsh to mudflat and
open water.
Chemical Contamination
Chemical contaminants accumulate in salt marshes
that receive surface water runoff and/or direct dis
charges of waste materials. Among the most toxic are
halogenated hydrocarbons, which include many insec
ticides, herbicides, and industrial chemicals. When
accumulated in the tissues of salt marsh animals a
392
Salt Marshes
wide range of disorders can result, for example, immu
nosuppression, reproductive abnormalities, and cancer.
Petroleum hydrocarbons pollute harbors and remnant
salt marshes following oil spills, urban runoff, and influxes
of industrial effluent and municipal waste. Once they
move into anoxic sediments, they can persist for decades,
reducing primary production, altering benthic food webs,
and accumulating in bird tissues. Polycyclic aromatic
hydrocarbons have additional carcinogenic and muta
genic potential for aquatic organisms.
Heavy metals are also toxic to aquatic organisms and
can impair feeding, respiration, physiological and neuro
logical function, and reproduction, as well as promote
tissue degeneration and increase rates of genetic muta
tion. Mercury is especially problematic because it is
methylated in the anoxic soils of salt marshes and is
then able to bioaccumulate in food chains.
Salt marsh plants in urban areas take up, accumulate,
and release heavy metals. Judith Weis and others have
found lowered benthic diversity and impaired fish behav
ior in contaminated sites. Fish are slower to catch prey
and less able to avoid predators where heavy metals
contaminate their habitat.
Research Value
Tidal marshes include an impressive array of environ
mental conditions within about a meter of elevational
range. The compressed environmental gradient invites
studies of species abiotic factors, and over time, their
contributions proceeded from community ecology to eco
system science and, finally, integration of the two.
Community Ecology
The limited number of vascular plant species has
made salt marshes very suitable for both descriptive
and manipulative studies. Early researchers attributed
plant species distributions to their physiological tolerance
for the abiotic environment, without regard to species
interactions. J. A. Silander and J. Antonovics used pertur
bation response methods to determine that biotic forces
also affected species distributions. Others effectively used
reciprocal transplanting to examine the relative impor
tance of abiotic conditions and interspecific competition
to species distributions. For example, S. Pennings and
R. Callaway revealed interspecific interactions among
southern California halophytes, and S. Hacker and
M. Bertness reported interspecific interactions among
New England halophytes. Manipulative transplantation
has shown that species distributions respond to abiotic
conditions, facilitation, and competition.
The wide latitudinal range of salt marshes allowed
study of community structure and function in relation to
sea level variations, for example, James Morris documen
ted and modeled interannual variations in salinity and its
effect on S. alterniflora growth. Such studies led
to predictions of changes in response to global climate
change.
Ecosystem Functioning
The monotypic nature of USA Atlantic Coast salt
marshes aided early studies of vascular plant produc
tivity and considerable literature developed around
the rates of productivity and alternative methods of
calculating gross and net productivity – work that
transferred to grasslands and other nonwoody vegeta
tion. Nitrogen dynamics were a later focus. The first
marine system to have a nitrogen budget was Great
Sippewisset Marsh in Massachusetts. The budget
quantified nitrogen inputs from groundwater, precipi
tation, nitrogen fixation, and tidal flow, and nitrogen
outputs from tidal exchange, denitrification, and bur
ied sediments.
Integrating Structure and Function
A long controversy over the causes of height variation
in Spartina spp. has involved USA researchers on both
the Atlantic and Pacific Coasts and has linked plant
and ecosystem ecology. The most convincing evidence
for a genetic (‘nature’) component is that of D.
Seliskar and J. Gallagher, who grew genotypes from
Massachusetts, Georgia, and Delaware for 11 years in
a common garden and documented persistent pheno
typic differences. A series of papers on soil
biogeochemistry explained the role of ‘nurture’.
Nitrogen was shown to be a key limiting factor for
S. alterniflora plant growth because nitrate is quickly
reduced to ammonia by bacteria in poorly drained
areas away from creeks, where soils have lower soil
redox potential. Sulfate reducing bacteria were also
implicated, because they reduce sulfate to sulfide,
which impairs the growth of sensitive plant species.
Increased soil redox potential and greater pore water
turnover in creek side habitat contributes to taller
height forms of S. alterniflora. Thus, both genetics
and environment influence height forms of
S. alterniflora, an outcome of both community and
ecosystem research.
Restoration
With recognition of lost ecosystem services, interest in
restoring salt marshes is growing in Europe and the
USA. One way that the British are combating rising
Salt Marshes
Figure 3 Tidal marsh vegetation is typically dominated by
salt-tolerant grasses and succulent forbs, easily distinguished in
this restored marsh at Tijuana Estuary, near San Diego,
California. Photo by J. Zedler.
393
across a tidal floodplain affected salmon use (Oregon),
and J. Callaway, G. Sullivan, J. Zedler, and others
showed that planting diverse assemblages and incising
tidal creeks jumpstarted ecosystem functioning in
salt marsh restoration sites (Tijuana Estuary,
California) (Figure 3). In Spain, restoration of tidal
ponds is being accomplished in replicate excavations
that test the effect of size and depth on use by salt
marsh animals (Doñana Marshlands).
In conclusion, salt marshes perform highly valued
ecosystem services that are lost when habitats are
developed and/or degraded. Further innovations will
likely take place in both the practice and science of
restoration, because salt marshes are highly amenable to
experimentation.
Further Reading
sea levels is via ‘managed retreat’, which involves
breaching of embankments to restore tidal flushing to
lands that were once salt marshes. In the Netherlands,
tidal influence is being reinstated to various
polders along the southwestern coast to restore natural
processes and diverse estuarine biota to former
polders.
Some of the earliest salt marsh restoration in USA has
been accomplished as mitigation for damages to other
sites as required by federal regulatory agencies. In
North Carolina, S. alterniflora was being replanted in the
1970s, and the practice has expanded widely to mitigate
damages due to development.
Some of the most innovative research on wetland
restoration has been accomplished in salt marshes by
replicating variables in restoration sites; for example,
D. Seliskar and J. Gallagher showed that genotypic
variation in S. alterniflora has implications for nearly
every component of the food web (in Delaware),
T. Minello and R. Zimmerman showed that channels
in replanted salt marshes enhanced fish support
(Galveston Bay, Texas), I. Mendelssohn and
N. Kuhn showed that dredge spoil addition acceler
ated S. alterniflora recovery in subsiding wetlands
(Louisiana), Cornu showed that topographic variation
Adam P (1990) Saltmarsh Ecology. Cambridge, UK: Cambridge
University Press.
Adam P (2002) Saltmarshes in a time of change. Environmental
Conservation 29: 39 61.
Allen JRL and Pye K (1992) Saltmarshes: Morphodynamics,
Conservation and Engineering Significance. Cambridge, UK:
Cambridge University Press.
Chapman VJ (1960) Salt Marshes and Salt Deserts of the World. Plant
Science Monographs. London: Leonard Hill [Books] Limited.
Daiber FC (1982) Animals of the Tidal Marsh. New York, NY: Van
Nostrand Reinhold Co.
Long SP and Mason CF (1983) Saltmarsh Ecology. Glasgow: Blackie &
Sons Ltd.
Pennings SC and Bertness MD (2000) Salt marsh communities.
In: Bertness MD, Gaines SD, and Hay ME (eds.) Marine Community
Ecology, pp. 289 316. Sunderland, MD: Sinauer Associates Inc.
Pomeroy LR and Weigert RG (1981) The Ecology of a Salt Marsh. New
York: Springer.
Reimold RJ and Queen WH (eds.) (1974) The Ecology of Halophytes.
New York, NY: Academic Press Incorporated.
Seliskar DM, Gallagher JL, Burdick DM, and Mutz LA (2002) The
regulation of ecosystem functions by ecotypic variation in the
dominant plant: A Spartina alterniflora salt marsh case study. Journal
of Ecology 90: 1 11.
Threlkeld S (ed.) Estuaries and Coasts: Journal of the Estuarine
Research Foundation. Lawrence, KS: Estuarine Research
Federation.
Weinstein MP and Kreeger DA (eds.) (2000) Concepts and Controversies
in Tidal Marsh Ecology. Boston, MA: Kluwer Academic Publishers.
Zedler JB (ed.) (2001) Handbook for Restoring Tidal Wetlands. New
York, NY: CRC Press.
Zedler JB and Adam P (2002) Saltmarshes. In: Perrow MR and Davy AJ
(eds.) Handbook of Ecological Restoration vol. 2: pp. 238 266. Ress,
Cambridge, UK: Cambridge University Press.
394
Savanna
Savanna
L B Hutley and S A Setterfield, Charles Darwin University, Darwin, NT, Australia
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Definition and Occurrence
Adaptive Traits of Savanna Vegetation
Environmental Factors Determining Savanna Structure
Conceptual Models of Tree and Grass Coexistence
Savanna Biomass and Productivity
Threats to Long-Term Sustainability
Further Reading
Introduction
Definition and Occurrence
This article examines the ecological features of one
of most important tropical ecosystems, the savannas.
Savannas feature the coexistence of both trees and herbac
eous plants and are distinct from grasslands (absence of
woody plants) and closed forests (tree dominant). Savanna
ecosystems occur in over 20 countries, largely in the sea
sonal tropics. Much of the world’s livestock occurs in
savanna, underlining their social and economic importance.
Approximately 20% of the world’s land surface is covered
with savanna vegetation, which produces almost 30% of
global net primary production (NPP). With tree and her
baceous components, savanna biodiversity is high, often
higher than associated dry deciduous forests. Globally,
tenure of savanna lands incorporates pastoral, private use,
indigenous and national parks, with the disparate manage
ment aims of grazing, mining, tourism, subsistence
livelihoods, and conservation. Given their size, savannas
affect global carbon, nutrient and water cycles, and, with
their frequent fires, significantly influence atmospheric
chemistry. Savanna ecosystems have existed for millions
of years in many regions, although paradoxically, many
ecologists regard savannas as an ecologically unstable mix
ture of trees and grasses. Savanna boundaries are dynamic
in space and time and their occurrence and structure are
determined by a combination of environmental factors,
such as available water, nutrients, the frequency of distur
bances (e.g., fire and herbivory), and stochastic weather
events. This range of factors results in significant structural
variation and providing an overarching and strict definition
of what constitutes a savanna has been problematic. This
article provides a commonly used definition, describes
savanna distribution, and examines factors that influence
their structure and function. Understanding the determi
nants of savanna functioning, resilience and stability are
vital ingredients for improved management. Management
of savannas is especially important, as they are under
increasing development pressure, especially in tropical
regions, and threats to their long term sustainability are
examined.
Savanna ecosystems predominantly occur in the seasonal
tropics and are a unique mix of coexisting trees, shrubs,
and grasses (Figure 1). Debate surrounds the use and
definition of the term savanna, reflecting the range of
tree:grass ratios found in these ecosystems. Savanna eco
systems feature a range of structures, from near treeless
grasslands to woody dominant open forest/woodlands of
up to 80% woody cover. A widely used definition
describes a savanna ecosystem as one consisting of a
continuous or near continuous C4 grass dominated
understorey, with a discontinuous woody overstorey.
Woody components can be a mix of trees and shrubs of
evergreen or deciduous phenology, broad or needle
leafed. The grass dominated understory can consist of a
mix of species with either annual or perennial habit (often
>1 m in height). Ecosystems that fit this definition have
ambiguously been termed woodlands, rangelands, grass
lands, wooded grasslands, shrublands, open forests, or
parklands.
Savanna formations occur on all continents of the
world (Figure 2), with the largest extent found in the
wet–dry tropical regions of Africa, South America, and
Australia. Smaller areas occur in Asia, including Sri
Lanka, Thailand, Vietnam, and Papua New Guinea.
Savanna also occurs in India, although these tree and
grass systems tend to be derived from dry deciduous
forest and subhumid deciduous forest due to land use
changes and population pressure. Tropical savanna occu
pies an area of approximately 27.6 million km2 including
the Asian savanna regions. Tree–grass mixtures also occur
in temperate regions, in North America (Florida, Texas),
Mediterranean Europe, and Russia, although these tem
perate savannas are far smaller in extent at approximately
5 million km2. In total, the savanna biome occupies one
fifth of the global land area and supports a large and
growing population.
The existence of a dry season is a defining feature of
savannas; rainfall is seasonal and ranges from 300 to
2000 mm, with a dry season lasting between 2 and 9
Savanna 395
(a)
(b)
(c)
Figure 1 Savanna ecosystems of the world, featuring the coexistence of a discontinuous woody overstorey with a continuous
herbaceous understorey. Plates (a) and (b) are of a north Australian savanna site that receives approximately 1100 mm rainfall and is
dominated by evergreen trees (Eucalyptus sp.) and tall C4 tropical grasses (Sarga spp.). Canopy fullness and grass growth are
significantly differently in the wet (a) and dry (b) seasons. Tower-mounted instrumentation in plate (a) is monitoring ecosystem
productivity and water use over wet and dry seasons. Plate (c) African savanna of the Kalahari Gemsbok National Park, Botswana.
(a, b) Photo courtesy of Joerg Melzheimer.
months of the year. There can be a single, extended dry
season or several shorter dry periods. Inter annual varia
tion of rainfall is typically high, as is the commencement
and cessation of the wet season and growing season
length, making cropping in savanna lands difficult.
Indeed, historical rainfall plays an important role in
determining the vegetation structure of a savanna.
Seasonally available moisture dramatically influences
plant productivity, which in turn determines the timing
of available resources for savanna animals.
Given their wide biogeographic range, savannas
occur on a number of soils types, typically oxisols,
ultisols, entisols, and alfisols (using US soil taxonomy).
In general, these soils are ancient and highly weath
ered, low in organic matter and cation exchange
capacity (CEC). Oxisols occur in tropical savanna
regions of South America and central and eastern
African savanna and consist of highly weathered, trans
ported, and deposited material occurring on fluvial
terraces. Extensive weathering of primary minerals has
occurred and they are dominated by clay minerals
such as kaolinite and gibbsite which have low CEC.
Also present in the soil are acidic Fe and Al sesqui
oxides, which limits nutrient availability, especially
phosphorus. Savanna soils tend to be sands to sandy
loams, deep and well drained but with low soil
396
Savanna
San Fernando De Apure
73 m
1432 mm
250
40
200
30
150
20
100
10
50
°C 27.4
0
J FM A M J J A S O N D
0
Kano
40
30
20
10
0
472 m
°C 26.2
Nyala
866 mm
250
200
150
100
50
J FM A M J J A S O N D
0
40
°C 26.9
675 m
495 mm
30
20
10
0
J FM A M J J A S O N D
Gulbarga
250
200
150
100
50
0
40
458 m
°C 27.2
753 mm
30
20
10
0
J FM A M J J A S O N D
Darwin
250
200
150
100
50
0
40
30
20
10
0
30 m
°C 27.6 1714 mm
250
200
150
100
50
J FM A M J J A S O N D
0
30
30
15
15
0
0
15
15
30
30
Barra
408 m
°C 26.3
684 mm
40
250
200
30
150
20
100
10
50
0
J FM A M J J A S O N D
0
Goiania
729 m
40 °C 21.9 1487 mm
30
20
10
0
J FM A M J J A S O N D
250
200
150
100
50
0
40
30
20
10
0
J FM A M J J A S O N D
Daly Waters
Bulawayo
Ndola
1269 m
°C 19.7 1168 mm
250
200
150
100
50
0
40
1343 m
°C 19.2
581 mm
30
20
10
0
J FM A M J J A S O N D
250
200
150
100
50
0
40
°C 26.9
210 m
628 mm
30
20
10
0
J FM A M J J A S O N D
250
200
150
100
50
0
Figure 2 The distribution of the world’s savannas. Temperature and monthly rainfall data for a range of savannas are also given, with
highly seasonal rainfall clearly evident.
moisture holding capacity. Entisols that occur in
Australian savanna also feature the occurrence of fer
ruginous gravels, further reducing water and nutrient
holding capacity. Bioturbation by earthworms and
termites are critical in the cycling of nutrients through
the poor soil systems. Termites essentially act as primary
consumers and in savannas that lack a significant herbi
vore biomass (e.g., Australian and some South American
savannas), they have an ecological function similar to that
of herbivorous mammals.
Savannas of Australia, Africa, and South
America
Tropical savanna is the predominant vegetation type across
the northern quarter of Australia where rainfall is above
600 mm yr 1, an area of 2 million km2 (Figures 1a, 1b,
and 2). These savannas are open woodlands and open for
ests, with tree cover declining as rainfall decreases
with distance from the northern coast. The overstorey
flora is typically dominated by Eucalyptus spp., particularly
Eucalyptus tetrodonta, E. dichromophloia, and E. miniata.
Melaleuca viridiflora, M. nervosa, and E. pruinosa assemblages
occur in the drier regions of this biome where annual
rainfall <1000 mm yr 1. The ground layer is dominated by
annual and perennial grasses from the Sarga, Heteropogon,
and Schizachrium genera. A variety of other tall grasses
(>1m height) dominate the ground layer of the monsoonal
savannas, which extend from Western Australia to the
Cape York Peninsula in Queensland. Heteropogon contortus
(black speargrass) dominates the tropical savanna unders
tory in eastern Queensland, with Themeda triandra, Aristida,
Bothriochloa, and Chrysopogon bladhii becoming more domi
nant as rainfall declines. Acacia dominated savanna
communities include extensive areas of brigalow (A. harpo
phylla), lancewood (A. shirleyi), and gidgee (A. cambegei and
A. georginae).
The neotropical savannas of South America cover
more than 2 million km2. The Brazilian cerrado and the
Colombian and Venezuelan llanos are a continuous forma
tion, interrupted by narrow gallery forests. The cerradão
includes a range of vegetation formations from the pure or
almost pure grassland of camp limpo, to open woodland with
scattered tree cover of campo cerrado. These savanna can
grade into denser woodland or open forests, the cerradão,
where tree cover is greater than 50%. The dominant
grasses are Andropogon, Aristida, Paspalum, and Trachypogon.
The Orinoco llanos comprise grasslands or grasslands with
scattered trees which are typically <8 m tall. Common
trees include Byrsonima spp. Curatella americana, Bowdichia
virgioides, and grasses include Trachypogon and Andropogon.
Hyper seasonally flooded savannas and esteros (savanna
wetland) occur in Brazil and Bolivia. Other savanna types,
such as savanna parkland and mixed woodland, occur
through tropical America.
The African savannas occur across a range of soil types
within a rainfall range of 200–1800 mm. One of the most
extensive savanna areas is the miombo which covers about
Savanna 397
2.7 million km2 across central and southern Africa.
The miombo is characterized by a discontinuous canopy of
10–12 m tall deciduous species of Brachystegia, Isoberlinia,
and Julbernardia, with an herbaceous layer of tall grasses
including mainly Andropogon species. In southern Africa,
fine leaved savannas, dominated by Acacia species, occur
over fertile soils in low lying, semiarid (250–650 mm yr 1)
areas. Broadleaved savannas, including Burkea africana,
Combretum spp., and Brachystegia, occur on weathered, infer
tile soils. The northern Sudanian savannas have scattered
deciduous trees, typically Isoberlinia doka, over xerophytic
grasslands. These are bordered on the north by the drier
Sahelian savannas and on the south by the wetter Guinea
type savannas. The arid and semiarid east African savannas
are grasslands dominated by Aristida spp. and Brachiara spp.,
with scattered shrubs or trees (including Acacia, Grewia, and
Commiphora), for example, the Serengeti.
Savannas occur throughout Asia, although many of these
are derived from human disturbance. Savanna is fairly
extensive in the Indian subcontinent, although tree clearing
has increased their extent, and many areas have been con
verted to agriculture. The most significant and widespread
savanna type in Southeast Asia is the dry dipterocarp forest,
which occurs in Vietnam, Laos, Cambodia, Thailand,
Burma, and a small area in India. The region receives
1000–1500 mm rain per year and is dominated by the
deciduous Dipterocarpus species, which can grow to 20 m,
over a dense grass and herb layer including Imperata cylind
rical, Apluda mutica, and Arundinaria spp. (pygmy bamboo).
Adaptive Traits of Savanna Vegetation
Savanna plants display a suite of traits to cope with seasonal
drought, low water and nutrient availability, and the impacts
of regular fire and herbivory. Adaptive traits which aid in
the survival of fire for woody plants include thick insulating
bark, high wood moisture content, and significant resprout
ing capacity. Resprouting can occur via lignotubers and
from other underground and stem basal tissues following
the death of aerial stems. This enables recovery with mini
mal developmental costs. Vegetative reproduction from
roots, rhizomes, or stolons is dominant in much of the
savanna biome. Adaptations to low nutrient availability
include root mycorrhizal associations, particularly of ecto
mychorrhizae. Savanna trees can rapidly translocate
sequestered nutrients from the leaves to other tissues (e.g.,
bark) prior to leaf fall. Woody savanna plants often have
thorns that restrict grazing, as well as chemical features, such
as tannins making leaves less palatable. Savanna grasses also
display morphological features, such as serrated edges, and
chemical features, including tannins and silica bodies, to
restrict grazing.
The herbaceous grass layer is dominated by grasses
with a C4 photosynthetic pathway. This pathway enables
high photosynthetic rates at high temperatures and irra
diance and low water availability. Most savanna trees and
shrubs have the C3 photosynthetic pathway that has a
higher efficiency under low light when compared to the
C4 pathway, a characteristic which facilitates recruitment
and establishment under shaded tree canopies. The
growth of savanna plants tends to occur mostly during
the wet season with senescence or dormancy in the fire
prone dry season, a trait that facilitates persistence in
unfavorable conditions. Annual herbaceous species persist
via a soil seed bank, whereas aboveground parts of per
ennial herbaceous species die during dry periods, with
dormant, regenerative buds protected within below
ground rhizomes or by cataphylls. Some annual grass
species use hygroscopically active awns and pointed cal
luses on their seed to work them into the soil, also
protecting them from fire. Perennial herbaceous species
require wet season rains to produce their first green
shoots as carbohydrate storage from the previous wet
season is limited. Rainfall stimulates germination of
annual herbaceous and grass species and the early wet
season is a period of rapid growth. Most herbaceous
species flower in the wet season, although in contrast,
many woody species flower in the dry season.
Woody species have evolved physiological and mor
phological mechanisms to either tolerate (evergreen
habit) or avoid (deciduous habit) prolonged periods of
water stress. Deep rooting woody plants (usually ever
green) are able to access water resources throughout the
year and provides them with their full photosynthetic
capacity when favorable conditions occur. Deciduous
species rehydrate stems prior to onset of wet season
rains, which is then followed by leaf expansion to max
imize photosynthetic activity during the wet season.
Deciduousness and evergreeness represent extremes of
physiological adaptations to survive the seasonal savanna
climate. Evergreen species invest more resources in
longer lived leaves, whereas deciduous species tend to
support shorter lived leaves with high leaf photosynthetic
capacity. Deciduous species need to acquire enough
nutrient and photosynthate to ensure persistence and
reproduction during the wet season, whereas evergreen
species tend to have slower growth rates but persist
throughout the seasonal cycle. Evergreeness also allows
opportunistic acquisition of resources when soil nutrients
are severely limiting and the cost of producing new leaves
to respond to change in soil moisture is prohibitive.
Although this section has described broad seasonal
growth patterns, it is important to note that the world’s
savanna plants include a high diversity of species and
life forms, with many distinct phenological patterns. All
periods of the climatic cycle is favorable to certain
vegetative or flowering phenophases in at least one
group of species.
398
Savanna
Environmental Factors Determining
Savanna Structure
The adaptive traits described above enable individual
plant survival in seasonally variable climates, but what
environmental factors operate at a landscape or regional
scale that determine savanna structure? Evidence sug
gests that four key environmental factors are responsible:
(1) plant available moisture (PAM); (2) plant available
nutrients (PANs); (3) fire regime; and (4) herbivory.
Herbivores include vertebrates and invertebrates and
consist of both browsers consuming woody biomass and
grazers consuming grasses and herbs. The overarching
determinants of savanna physiognomy (relative abun
dance of the tree and grass layer) are climate and soil
type (PAM and PAN), which determines the potential
growth and survival of trees and grasses at a given site.
Growth potential is moderated by disturbance agents,
fire, herbivory, and stochastic events (such as cyclones).
These factors act in concert to influence both competi
tive interactions and facilitation of tree and grass growth
and determine savanna structure, floristics, and
productivity (Figure 3). The interaction of these factors
is poorly understood and their variation in space and
time makes experimental testing and isolation of any
single determinant difficult. Spatial heterogeneity of
vegetation due to local site histories (determined by
antecedent rainfall, fire history, and herbivore numbers)
and an inability to quantify these factors exacerbates this
difficulty.
Available Moisture and Nutrients
Savannas tend to occur in warm climates with an annual
drought, with soils typically low in nutrient capital and
poor water holding capacity. Interactions between trees
and grasses is dominated by competition for water and
nutrients, rather than light or growing space. At a broad,
continental scale, PAM is the most significant of the four
ecological determinants, with increasing rainfall corre
lated with increased tree cover and in general, a
decreased grass biomass. PAM can be quantified via a
range of parameters, from simple measures such as annual
rainfall or via water balance parameters (rainfall as a
fraction of potential or actual evapotranspiration) or soil
characteristics (water release characteristics, soil storage
capacity). At fine spatial scales, soil physiochemical prop
erties (PAN) have a more significant influence and the
interaction with PAM is often termed the PAM/AN plane.
Nutrient availability is largely a function of soil moisture
and dry season nutrient uptake, and nitrogen mineraliza
tion, in particular, is limited by low levels of PAM.
Significant plant growth is only possible during periods
of high PAM that releases available nutrient via miner
alization. Soils of semiarid savanna can have a higher
intrinsic fertility when compared to highly leached soils
of mesic sites, but this nutrient capital is only available for
uptake during moist periods. Savanna vegetation receiv
ing similar rainfall can exhibit contrasting structure and
floristics, simply due to fine scale changes in soil type.
A good example of this interaction comes from the long
Figure 3 Interactions between the environmental determinants of savanna structure. The relative tree and grass biomass and
productivity is determined by available water, nutrient, and disturbance regime (fire and herbivory). These determinants are in turn
characterized by climate and soil type for any given location. Reproduced from House JI, Archer S, Breshears DD, and Scholes R (2003)
Conundrums in mixed woody-herbaceous plant systems. Journal of Biogeography 30: 1763–1777, with permission from Blackwell
Publishing.
Savanna 399
(a)
(b)
Figure 4 Impacts of overgrazing and fire on savanna structure. Plate (a) is an overgrazed native grass paddock in semiarid savanna in
north Australia (Kidman Springs Station, Victoria Rivers District, NT) at the end of the wet season of 1973. This site would be subjected
to wind and water erosion, resulting in further decline in health and productivity of such sites. Exclusion of grazing and fire (Plate (b) has
resulted in a complete recovery of structure and function, with return of trees and grasses stabilizing soil surfaces, increased water
capture and a recovery in nutrient availability and cycling. Photos courtesy of John Ludwig, CSIRO.
term savanna research site of Nylsvley in South Africa;
here, nutrient poor, broad leafed savanna, dominated by
Burkea africana surround patches of nutrient rich soil that
support a very different savanna type, a fine leafed
savanna dominated by Acacia tortilis. Both savanna types
experience the same climate, but differences in soil parent
material result in higher levels of soil available N and P in
the fine leafed patches. Productivity of the fine leafed
savanna is approximately double that of the broad leafed
system and attracts a larger grazing and browsing fauna.
Similarly in South American savanna, soil acidity and
aluminum levels significantly affect structure and floris
tics independent of rainfall.
Fire
Fire is an important landscape scale determinant that
impacts all of the world’s savanna. Fire is an inevitable
consequence of the annual cycle of profuse herbaceous
production during the wet season followed by curing of
this material in the dry season, when climatic conditions are
ideal for burning. Savanna fires are virtually all surface fires,
consuming the highly flammable herbaceous layer. Crown
fires rarely occur, as the foliage of savanna trees and shrubs
are of low flammability. Human ignitions largely control
fire behavior and extent in savanna, with some fires started
by dry lightning strikes. Savanna fires spread rapidly
through the surface fuels and high soil temperatures do
not persist for longer than a few seconds to minutes.
While these fires have a significant impact on aboveground
plant parts, there is limited impact on savanna seed banks or
belowground regenerative plant parts.
Fire has a major role in restricting tree establishment
and growth, as evident from long term fire exclusion plots
(>25 years) in southern African and north Australian
savanna (Figure 4), which have resulted in a woody thick
ening. Frequent fire events can reduce tree seedling
establishment and the ability of saplings to escape the
flame zone via height growth. This limitation on tree
establishment enables grass persistence and growth, main
taining the fuel load. The aerial stems of small seedlings
and suckers are often killed during fire but the individuals
are able to resprout from lignotubers or from other under
ground and stem basal tissues. Seedlings less than 6 months
old have been observed to resprout in some species
(e.g., Eucalyptus miniata) and frequent fire in the savannas
will kill or maintain tree seedlings as a suppressed woody
sprout layer until there is a sufficient fire free period for
them to escape the fire damage zone. Species can survive
for at least 40 years as suppressed sprouts, during which
time they develop significant lignotubers which aid in
rapid growth during fire free periods.
The timing of fires in relation to reproductive phenol
ogy can constrain or promote plant reproduction. Studies
on the woody species in the Brazilian cerrado and mesic
Australian savannas have indicated that frequent fire can
reduce seed production and sexual recruitment and could
cause a shift in species composition, favoring vegetatively
reproducing species. However, fire is also important for
the sexual regeneration of some species, as burning
induces flowering and fruit dehiscence in many cerrado
species and facilitates pollination in others. Most peren
nial grass species are generally less affected by burning
and regenerate from basal leaf sheaths protected under
ground. Some perennial (e.g., Trachypogon plumosus) and
annual (e.g., Andropogon brevifolius) grass species decrease
in abundance after a long term absence of fire.
Prior to human occupation and use of fire in savannas,
lightning would have been the dominant source of igni
tion and it is likely that extensive but infrequent fires
400
Savanna
would have occurred. In Australia, humans have inten
tionally used fire for at least 40 000 years and in Africa for
potentially 1 million years or more. Large proportions of
savanna regions are burnt each year for a variety of
reasons: land clearing, livestock management, property
protection, conservation management, and cultural pur
poses. In African savannas, fires burn between 25–50% of
the arid ‘Sudan Zone’ and 60–80% of the humid ‘Guinea
Zone’ each year. Approximately 65% of Eucalyptus domi
nated savanna woodland and 50% of savanna open forest
in Kakadu National Park, northern Australia was burnt
annually between 1980 and 1994. With the progression of
the dry season, fire intensity increases due to fuel accu
mulation from curing litterfall and grass senescence
resulting in an increased combustibility of fuels plus
more severe fire weather (i.e., higher temperatures, stron
ger winds, and lower humidities). Early dry season fires
(when fuel accumulation is low and curing incomplete)
tend to be low intensity, patchy, and limited in extent.
Fires later in the season are of higher intensity and pro
duce more extensive and homogeneous burning. Impacts
on vegetation depend on fire intensity, distribution, and
timing (fire regime) in relation to the vegetative and
phenological cycles. Determining direct effects of fire on
savannas is often difficult due to confounding effects of
herbivory. Nevertheless, long term burning experiments
have shown that the higher intensity, late dry season fires
are the most damaging to woody species.
Herbivory
Two common images of savannas are herbivory by large,
native ungulates, particularly in Africa and the wide
spread grazing by domestic herds, particularly cattle.
A more neglected group of savanna herbivores are the
invertebrates, particularly grasshoppers, caterpillars, ants,
and termites. Mammal herbivores are typically categor
ized as grazers, browsers, or mixed feeders, who can vary
their diet depending on food availability. Mammal and
insect herbivores impact on savanna structure and func
tion via consumption of biomass, seed predation,
trampling of understory, and the pushing over and killing
of trees and shrubs. The importance of herbivory as a
determinant varies between savanna regions, and appears
to largely reflect the abundance of large herbivores pre
sent. Large herbivore diversity and abundance are much
higher in Africa than in Australia, Asia, or South America.
More than 40 large wild herbivore species have been
described in African savanna. In contrast, only six species
of megapod marsupial have been considered as large
herbivorous mammals in the Australian savannas, and
only three species of ungulates are regarded as native
South American savanna inhabitants. Domestic animals,
particularly cattle, buffalos, sheep, and goats, are now the
dominant, large herbivores in most savannas.
Large herbivores can lead to changes in species compo
sition, woody vegetation density, and soil structure. For
example, grazing pressure in Africa and Australia has led
to a decrease in palatable, perennial, grazing sensitive tus
sock grasses, and an increase in less palatable perennial and
annual grass and forb species. Changes to the soil surface
can occur, including loss of crusts (important in nutrient
cycling), development of scalds, compaction, increased run
off, soil erosion, and nutrient loss. In parts of Africa, woody
vegetation density has sometimes been reduced by large
herbivores, for example, uprooting of trees by elephants
when browsing. Browsers such as giraffes can reduce
woody seedling and sapling growth, thereby keeping them
within a fire sensitive heights for decades. By contrast, in
many of the world’s savannas the density of woody vegeta
tion has increased at the expense of herbaceous vegetation;
one of the major causes has been high rates of herbivory.
A decrease in grass biomass following grazing leads to a
reduction fuel and thus fire frequency and intensity, enhan
cing the survival of saplings and adult tress. Fire also affects
herbivory as herbivores may favor postfire vegetation
regrowth. Clearly, fire and herbivory have an interactive
effect on savanna structure and function.
While less spectacular than large browsers and grazers,
insects are often the dominant group of herbivores in savan
nas, especially on infertile soils supporting low mammal
biomass. There is a paucity of data describing their abun
dance or role in these ecosystems. In a broad leaved, low
fertility savanna of southern Africa, a grasshopper biomass of
0.73 kg ha 1 can consume almost 100 kg ha 1 of plant mate
rial and damage an additional 36 kg ha 1. This represents a
loss of 16% of aboveground grass production. Grasshoppers
and caterpillars can account for up to half the grass herbiv
ory, although the rate and proportion varies substantially
between years. Fertile, fine leaved savannas are able to
support a larger mammal biomass, and the proportion of
herbivory resulting from insect consumption is lower when
compared to infertile African sites. The impact of insect
herbivores on physiognomy has not been established but
they are clearly important herbivores in savannas through
their impact on productivity and ecosystem properties.
Conceptual Models of Tree and Grass
Coexistence
Interactions between the coexisting lifeforms in savanna
communities are complex and over the last 40 years, a
range of conceptual or theoretical models has been pro
posed to explain tree and grass mixtures. Contrasting
models have all been supported by empirical evidence
for particular sites, but no single model has emerged that
provides a generic mechanism explaining coexistence.
Models can be classified into several categories.
Competition based models feature spatial and temporal
Savanna 401
Table 1 Conceptual models explaining the coexistence of trees and grasses in savanna ecosystems in equilibrium (tree:grass ratio
relatively stable at a given site), nonequilibrium (tree : grass ratio variable) or disequilibrium (disturbance agents essential for the
maintenance of tree:grass coexistence)
Competition-based
Demographic-based
Mechanisms of coexistence
Spatial and temporal niche separation of resource usage enables
both life forms to coexist
Root-niche separation
Tree and grasses exploit deep and shallow soil horizons
Mechanisms of coexistence
Climatic variation and disturbance impacts on tree demography
Extremes of climate and disturbance influence tree germination
and/or establishment and/or transition to mature size classes
enabling coexistence
At low rainfall sites, tree establishment and growth occurs only in
above average rainfall periods
At high rainfall sites, high fuel production maintains frequent fire
to limit tree dominance
Phenological separation
Temporal differences in leaf expansion and growth, trees have
exclusive access to resources at beginning and end of growing
season, grasses competitive during growing season
Balanced competition
Trees are the superior competitor but become self-limiting for a
given rainfall and unable to exclude grasses
Competition–colonization
Rainfall variability results in a tradeoff between tree and grass
competition and colonization potential. Higher than mean
rainfall
favours tree growth, lower than mean favours grasses
Primary determinants
PAM variability, PAN, fire regime, herbivory
Primary determinants
PAM, PAN
Secondary determinants
Fire regime, herbivory
separation of resource usage by trees and grasses that
minimizes competition and enables the persistence of
both lifeforms. Alternatively, demographic based models
have been described, where mixtures are maintained by
disturbance, resulting in bottlenecks in tree recruitment
and/or limitations to tree growth and grasses can persist.
Table 1 provides a summary of these models. Root niche
separation models suggest that there is a spatial separation
of tree and grass root systems, with grasses exploiting
upper soil horizons and trees developing deeper root
systems. Trees rely on excess moisture (and nutrient)
draining from surface horizons to deeper soil layers.
Phenological separation models invoke differences in
the timing of growth between trees and grasses. Leaf
canopy development and growth in many savanna trees
occurs prior to the onset of the wet season, often before
grasses have germinated or initiated leaf development. As
a result, trees can have exclusive access to resources at the
beginning of the growing season, with grasses more com
petitive during the growing season proper. Given their
deeper root systems, tree growth persists longer into the
dry season, providing an additional period of resource
acquisition at a time when grasses may be senescing.
This spatial and temporal separation of resource usage is
thought to minimize competition, enabling coexistence.
Other competition models suggest that density of trees
becomes self limiting at a threshold of PAM and PAN
and is thus unable to completely exclude grasses. These
models assume that high rainfall years favor tree growth
and recruitment, with poor years favoring grasses, and
high interannual variability of rainfall maintaining a rela
tively stable equilibrium of trees and grasses over time.
Alternatively, savannas can be viewed as meta stable
ecosystems (narrow range of stabile states) with a
dynamic structure over time. Demographic based models
suggest that determinants of tree demographics and
recruitment processes ultimately set the tree:grass ratios
(Table 1). Fire, herbivory, and climatic variability are
fundamental drivers of tree recruitment and growth,
with high levels of disturbance resulting in demographic
bottlenecks that constrain recruitment and/or growth of
woody components and grass persistence results. At high
rainfall sites, in the absence of disturbance, the ecosystem
tends toward forest. High levels of disturbance, particu
larly fire, can push the ecosystem toward a more open
canopy or grassland; this ecosystem trajectory is more
likely at low rainfall sites.
There is observational and experimental data to sup
port all of the above models and it is highly likely that
savanna structure and function results from the interac
tion of all processes. In many savannas, root distribution is
spatially separated with mature trees exploiting deeper
soil horizons as the competitive root niche separation
model predicts. Root partitioning favors tree growth in
semiarid systems where rainfall occurs during periods
when grass growth is dormant; rainfall can drain to deep
402
Savanna
layers supporting tree components. By contrast, in semi
arid savanna where rainfall and growing seasons coincide,
investment in deep root systems could result in tree water
stress, as rainfall events tend to be sporadic and small in
nature, with little deep drainage. In this case, surface roots
are more effective at exploiting moisture and mineralized
nutrients following these discrete events. In these savan
nas, tree and grass competition for water and nutrients
would be intense. In mesic savanna sites, root competition
between both trees and grass roots in upper soil layers is
apparent, contrary to predictions of niche separation
models. Mesic savannas of north Australia (rainfall
>1000 mm) are dominated by evergreen Eucalyptus tree
species, and during the wet season these trees compete
with high growth rate annual grasses for water and nutri
ents in upper soil layers (0–30 cm). However, by the late
dry season, tree root activity has shifted to subsoil layers
(up to 5 m depth) and herbaceous species have either
senesced or are physiologically dormant. These root
dynamics suggest that grasses are essentially drought
avoiders but are able to compete with trees during the
wet season. This system serves as an example where both
root niche and phenological separation are occurring.
Tree to tree competition is also significant, as suggested
by the strong relationship observed in most savanna regions
between annual rainfall and indices of tree abundance, be it
tree cover (Figure 5), tree basal area (area occupied by tree
stems), or tree density. As PAM decreases, tree abundance
declines. Competition models also fail to consider impacts
of savanna determinants on different demographics of a
population, such as recruitment, seedling establishment,
and tree sapling growth. Root niche or phenological
separation models largely consider impacts acting on
mature individuals, whereas demographic models include
impacts of climate variability and disturbance on critical
life history stages (e.g., seedling establishment and accession
to fire tolerant size classes). Demographic models assume
that savanna tree dynamics are central to savanna ecosys
tem functioning and that savanna trees are the superior
competitors under most conditions; grass persistence only
occurs when determinants act to limit tree abundance. It is
clear that competition, both within and between savanna life
forms, occurs and that tree abundance is moderated by
climate variability and disturbance. A more comprehensive
model would integrate both competition and demographic
theories to yield a model in which competitive effects are
considered for each life history stage.
The complexity inherent in these models is evident
when savanna structure is correlated with any of the
environmental determinants. Figure 5 describes the rela
tionship between tree cover and mean annual rainfall, in
this case a surrogate for PAM. Tree cover data are shown
for African and Australian savanna sites. The figure shows
a large scatter of tree cover possible at any given rainfall,
especially for the African sites. For African savanna, rain
fall sets an upper limit on tree cover, with the relationship
linear until approximately 650 mm rainfall with little
increase in tree cover observed above this threshold
(Figure 5). Points below the line represent savanna sites
with a tree cover determined by PAM plus the interaction
of other determinants to reduce tree cover below the
maximum possible for a given rainfall. At semiarid
savanna sites (<650 mm rainfall), it is likely that rainfall
limits tree cover and canopy closure, permitting grass
100
Africa (Sankaran et al. 2005)
Australia (Williams, unpublished)
Woody cover (%)
80
60
40
20
0
0
200
400
600
800
1000
MAP (mm)
1200
1400
1600
1800
Figure 5 Relationship between mean annual rainfall (MAP) and tree cover for African and Australian savannas, with rainfall setting
a maximal climate-determined woody cover. Other factors such as available nutrient, fire frequency and herbivory determine woody
cover at any given site. Modified from Sankaran M, Hanan NP, Scholes RJ, et al. (2005) Determinants of woody cover in African
savanna. Nature 438: 846–849 (Macmillan Publishers Ltd), with Australian tree cover data from R. J. Williams, unpublished data.
Savanna 403
coexistence. At rainfalls >650 mm, tree canopy closure
may be possible, with disturbance limiting woody dom
inance. For Australian savanna, there is a simpler
relationship evident, with a linear increase in tree cover
with annual rainfall and less scatter. Australian savannas
also have a reduced tree cover (and biomass) for a given
rainfall when compared to African systems (Figure 5).
This suggests that while PAM is determining tree cover,
other factors such as fire frequency or PAN are also
playing a role. Australian savanna soils (PAN) may be
systematically poorer than African soils or fire frequency
higher, limiting tree cover and productivity.
Savanna Biomass and Productivity
Global NPP, the net production of plant biomass, is
approximately 67.6 Gt C yr 1 of which almost 30%
occurs in savanna ecosystems (19.9 Gt yr 1). This produc
tion occurs on 18% of the global land surface,
demonstrating that savannas are relatively productive
ecosystems. Mean savanna NPP has been estimated at
7.2 t C ha 1 yr 1 (Table 2), lower than typical values for
the other major tropical ecosystem, rainforest, which
ranges from 10 to 15 t C ha 1. Savanna NPP and biomass
varies by an order of magnitude (Table 2), as would be
expected given their geographic range and structural
variation. The relative production of trees versus grasses
is also highly variable, but in general, NPP of the C4 grass
layer is 2–3 times that of tree NPP. Biomass stored in
above and belowground pools determines the root:shoot
ratio and these data from a range of savanna sites around
the world give a global mean of approximately 2
(Table 2). This reflects the investment in root systems
and belowground storage organs, such as lignotubers, to
maintain uptake of moisture and nutrient from sandy,
nutrient poor savanna soils and to survive disturbance.
Table 2 Savanna biomass, soil carbon stocks and productivity
Parameter
Mean (sd)
Range
Biomass and soil stocks (t C ha 1)
Aboveground biomass
Belowground biomass
Total biomass
Root : shoot ratio
Soil organic carbon
Savanna area (M km 2)
Total carbon pool (Gt C)
10.6 (9.0)
19.5 (14.9)
33.0 (22.9)
2.1 (2.0)
174.2 (126.0)
27.6
326
1.8–34
4.9–52
9.4–84
0.6–7.6
18–373
7.2 (5.1)
0.14
1.4–22.8
Productivity (t C ha
NPP
NEP
1
Savanna photosynthesis and growth is highly seasonal
and interannual variability high. Mesic savanna may receive
annual rainfall associated with rainforest ecosystems, yet
productivity is significantly lower, due largely to annual
drought, poor soils, and impacts of disturbance. Long term
(as opposed to annual) estimates of savanna productivity
need to include loss of biomass due to fire and herbivory.
Including fire and herbivory impacts on productivity esti
mates gives the carbon sequestration rate, which represents
the net gain (sink) or loss of carbon from the ecosystem to
the atmosphere. While wet season productivity can be very
high in savannas, much of a wet season’s herbaceous pro
ductivity can be lost via fire or grazing. Woody biomass
tends to be a less dynamic, longer term carbon storage pool
than the herbaceous components of savanna. Savanna fire
results in a significant release of greenhouse gases, including
CO2, CO, methane, nonmethane hydrocarbons, nitrous
oxide, particulate matter and aerosols, equivalent to 0.5–
4.2 Gt C yr 1. Fire reduces net savanna sequestration rate
by about 50% and protection of savannas from fire and
grazing results in an increase in woody biomass which can
result in a long term increase in stored soil carbon.
Savanna sink strength in mesic Orinoco savannas in
South America (1500 mm annual rainfall) has been mea
sured at 1 t C ha 1 yr 1, with this sink maintained over a
25 year period in plots with fire and grazing excluded.
Similarly, the carbon sink strength of north Australian,
Eucalyptus dominated savannas receiving approximately
the same rainfall has also been estimated at approximately
1 t C ha 1 yr 1, with this sink measured at sites burnt but
not grazed. This carbon is likely being stored in woody
biomass and soil organic carbon pools, with a small frac
tion being stored as black carbon (charcoal), a resilient
carbon pool. Savanna soil carbon storage is by far the
largest pool of carbon (Table 2) and soil carbon repre
sents a longer term storage of carbon when compared to
the more dynamic vegetation components. Burning also
influences nutrient dynamics via losses due to volatiliza
tion (vaporization) of lighter elements such as nitrogen
and sulfur. At a global scale, savannas and tropical
seasonally dry forests represent a significant source of
N2O to the atmosphere (4.4 Tg N2O yr 1). Shifts to a
more frequent fire regime may result in a significant net
loss of nitrogen, as savannas are in general nitrogen poor.
Many grass species are able to recover quickly after fire,
with re growth attractive to grazing animals, due to the
relatively high nutrient content of the foliage.
Threats to Long-Term Sustainability
y 1)
Data from Grace J, San JJ, Meir P, Miranda HS, and Montes RA (2006)
Productivity and carbon fluxes of tropical savannas. Journal of
Biogeography. 33: 387 400.
Savannas are ancient ecosystems. They are the location
of human evolution, and humans are an integral compo
nent of these ecosystems. Humans have influenced the
determinants of savannas for thousands of years via
404
Savanna
modification to nutrient availability from fire and clearing
for agriculture. Human cultures have used fire as a vegeta
tion management tool and introduced animal husbandry
systems, changing grazing and browsing pressures and
modified tree–grass competitive balances (e.g., Figure 4).
A contemporary impact is now being experienced via
climate change and its influence on rainfall distribution,
temperature increases, and climate conditions conducive to
fire and increased atmospheric CO2 concentration. Human
usage of the savanna biome is increasing, which can lead to
degradation of vegetation and soil resources, resulting in
nutrient losses and shifts in water balance and availability.
Brazilian cerradão contains over 800 species of trees and
shrubs alone; approximately 40% of the cerradão and llanos
have now been cleared or altered for agricultural uses with
crops such as coffee, soybeans, rice, corn, and beans. Soil
management is critical given their low nutrient status,
acidity and friability. Alterations in grazing pressure and
fire suppression in managed savannas have also resulted
in woody dominance, which ultimately reduces grazing
production, severely impacting communities relying on
cattle derived incomes and reducing local biodiversity.
This thickening or woody encroachment is being observed
in areas subjected to extensive grazing activities in both
African and Australian savannas.
Clearing for alternative land uses can also result in
exotic species invasions, a problem for much of the
world’s savannas. African savanna, especially in South
Africa, are being invaded by woody species, often Acacia
or Eucalyptus species from Australia, introduced for fuel
wood or timber production. Low herbivory of these spe
cies results in high growth rates and water use. The
development of thickets reduces deep drainage, ground
water recharge, and streamflow, consequently affecting
water supplies. In an attempt to increase the grazing
potential of north Australian and South American
savanna, fast growing African grasses such as Andropogon
gayanus have been introduced. They are more productive
than native species; however, they develop far larger and
more flammable fuel loads. At infested sites in north
Australia, resultant fire intensity is 5 times that observed
from native grass savanna and impacts on tree mortality
and recruitment. This in turn will result in a demographic
bottleneck, long term loss in tree cover, and the instiga
tion of a grass fire cycle. Introductions of African grasses
such as Brachiaria, Melinis, and Andropogon species have
occurred in the llanos of Colombia and Venezuela and
the cerrado of Brazil. These grasses are used as fodder for
cattle and are displacing native species, causing a loss in
biodiversity of these savannas.
Climate change will alter the distribution of rainfall,
thus influencing PAM and PAN. Shifts in temperature
regimes and atmospheric CO2 concentration may also
alter the relative growth rates of trees and grasses, mod
ifying competitive balances. Trees (C3 photosynthetic
pathway) can potentially utilize high CO2 concentrations
more efficiently than grasses (C4 photosynthetic path
way) due to increased carbon allocation to roots and
lignotubers plus greater water use and nutrient use effi
ciency apparent at high atmospheric CO2 concentrations.
As CO2 concentrations increase, physiological differences
between trees (carbon rich lifeforms) may be favored
over grasses (carbon poor) and trees may gain a compe
titive edge. Tree saplings may grow to fire tolerant sizes
faster, limiting the impact of fires that maintain grasses in
savanna.
All of the above examples involve human impacts
acting on one or more of the determinants of savanna
structure and function. Clearly, increased knowledge of
their interactions will provide improved understanding of
savanna processes and enable better management in a
rapidly changing world. Savannas may be ideal ecosys
tems for agro forestry applications, rather than traditional
cropping systems. Small shifts in fire regime may drama
tically increase productivity; thus, savanna systems could
be used for carbon sequestration and greenhouse gas
mitigation schemes, providing alternative livelihoods
and aiding in the maintenance of biodiversity.
See also: Mediterranean; Swamps.
Further Reading
Andersen AN, Cook GD, and Williams RJ (2003) Fire in Tropical
Savannas: The Kapalga Experiment. New York: Springer.
Baruch Z (2005) Vegetation environment relationships and classification
of the seasonal savannas in Venezuela. Flora 200: 49 64.
Bond WJ, Midgley GF, and Woodward FI (2003) The importance of low
atmospheric CO2 and fire in promoting the spread of grasslands and
savannas. Global Change Biology 9: 973 982.
du Toit JT, Rogers KH, and Bigg HC (eds.) (2003) The Kruger
Experience: Ecology and Management of Savanna Heterogeneity.
Washington, DC: Island Press.
Furley PA (1999) The nature and diversity of neotropical savanna
vegetation with particular reference to the Brazilian cerrados. Global
Ecology and Biogeography 8: 223 241.
Grace J, San JJ, Meir P, Miranda HS, and Montes RA (2006)
Productivity and carbon fluxes of tropical savannas. Journal of
Biogeography 33: 387 400.
Higgins SI, Bond WJ, and Trollope WSW (2000) Fire, resprouting and
variability: A recipe for grass tree coexistence in savanna. Journal
of Ecology 88: 213 229.
House JI, Archer S, Breshears DD, and Scholes R (2003) Conundrums
in mixed woody herbaceous plant systems. Journal of Biogeography
30: 1763 1777.
Mistry J (2000) World Savanna: Ecology and Human Use. Harlow:
Prentice Hall.
Rossiter NA, Setterfield SA, Douglas MM, and Hutley LB (2003)
Testing the grass fire cycle: Exotic grass invasion in the tropical
savannas of northern Australia. Diversity and Distributions
9: 169 176.
Sankaran M, Hanan NP, Scholes RJ, et al. (2005) Determinants of
woody cover in African savanna. Nature 438: 846 849.
Scholes RJ and Archer SR (1997) Tree and grass interactions in
savanna. Annual Review of Ecology and Systematics
28: 517 544.
Steppes and Prairies
Scholes RJ and Walker BH (eds.) (1993) An African Savanna:
Synthesis of the Nylsvley Study. Cambridge: Cambridge University
Press.
Solbrig OT and Young MD (eds.) (1993) The World’s Savannas:
Economic Driving Forces, Ecological Constraints, and Policy
Options for Sustainable Land Use. New York: Parthenon
Publishing Group.
405
van Langevelde F, van de Vijver CADM, Kumar L, et al. (2003) Effects of
fire and herbivory on the stability of savanna ecosystems. Ecology
84: 337 350.
Williams RJ, Myers BA, Muller WJ, Duff GA, and Eamus D (1997) Leaf
phenology of woody species in a north Australian tropical savanna.
Ecology 78: 2542 2558.
Steppes and Prairies
J M Briggs, Arizona State University, Tempe, AZ, USA
A K Knapp, Colorado State University, Fort Collins, CO, USA
S L Collins, University of New Mexico, Albuquerque, NM, USA
ª 2008 Elsevier B.V. All rights reserved.
Grasslands
Grassland Types
The Grassland Environment
Fire in Grasslands
Steppes and prairies (grasslands) are ecosystems that
are dominated by grasses and to help understand grass
lands, it is important to know something about grass
morphology and growth forms. The remarkable ability
of grasses to thrive in so many ecological settings
and their resilience to disturbance is largely attributa
ble to their growth form. Grasses are characterized by
streamlined reduction and simplicity with tillers being
the key adaptive structural element of the plant
(Figure 1). Tillers originate from growing parts (meri
stems) typically just near, at, or below the surface of
the soil. The meristems that produce tillers are gener
ally well protected by their location near or beneath
the soil surface. It is the location of the meristem that
explains much of the resilience of grasses and thus
grasslands to disturbance.
Grass leaves are narrow and generally well supplied
with fibrous supporting tissue that has thick walled cells.
These features, along with a capacity to fold or roll the
leaves along the vertical plane, permit the plant to endure
periods of water stress without collapse. Another feature of
grass leaves is the presence of siliceous deposits and sili
cified cells (phytoliths). Although silica is present in many
plant families, phytoliths are characteristic of grasses.
Phytoliths often have distinctive forms within taxonomic
groups and since they persist in soil profiles for a very long
time, they can be used by paleobotanists to determine
shifts in dominance from one grass form to another.
Silica also makes grass forage very abrasive and it is now
generally accepted that the evolution of abrasion resistant
teeth present in many modern grazing animals was an
evolutionary response to tooth wearing effects of a diet
Grazing in Grasslands
Threats to Grasslands and Restoration of Grasslands
Further Reading
Node
Spikelets
Blade
Blade
Sheath
Ligule
Node
Culm
Panicle
Figure 1 Common oat, Avena sativa, ½. From Hubbard
(1984).
406
Steppes and Prairies
high in grass. This also suggests that the grasses and their
megaherbivore grazers are highly coevolved. But recent
discovery of grass phytoliths in Late Cretaceous dinosaur
coprolites in India suggest that grasses were already sub
stantially differentiated and that abrasive phytoliths were
present in many grasses before the explosion of grazers in
the Oligocene and Miocene time periods.
Grasses show a very large variation in the way tillers are
aggregated as they expand from their origin, but two general
forms of grasses are recognized: bunch forming (caespitose)
and sod forming (rhizomatous). This description captures
the major features of the dominant grass species but there
are some species and groups that deviate from this general
pattern. The most obvious include the woody bamboos
(some of which can reach tree size and for the most part
are restricted to forest habitats in the tropics and subtropics).
In addition to growth form, grasses can also be roughly
divided into two categories based upon their photosynthetic
pathways: cool season (C3) and warm season (C4). C4 photo
synthesis is a variation on the typical C3 pathway and is
thought to have an advantage in high light and temperature
environments typical of many grassland regions worldwide.
Throughout the world today, tropical, subtropical, arid,
semiarid, and mesic grasslands are typically dominated
by C4 grasses while in cooler high elevation or northern
climates, C3 grasses are more common.
as much as 25–40% of the Earth’s land surface although
much of the original extent of native grassland has been
plowed and converted to other grass production (corn and
wheat) or other row crops such as soybeans. Indeed, grass
lands are important from both agronomic and ecological
perspectives. Grasslands are the basis of an extensive
livestock production industry in North America and else
where. In addition, grasslands sequester and retain large
amounts of soil carbon and thus, they are an important
component of the global carbon cycle.
Indeed, because grasslands store a significant amount
of carbon in their soils and they contain relatively high
biodiversity, then now play a prominent role in the dis
cussion about biofuel production. Biofuels may offer a
mechanism to generate energy that releases less carbon
into the atmosphere. Some energy producers recommend
intensive agricultural production of corn, or other grasses
such as switchgrass or elephant grass for biofuel produc
tion. However, agricultural practices have significant
energy costs that may reduce the value of these fuel
sources. A recent study has suggested, however, that
diverse prairie communities on marginal lands are poten
tially ‘carbon negative’ because they provide significant
biomass for fuel and store carbon belowground. Much
additional research is needed to assess the sustainability
of grasslands for biofuel production, but the prospects are
certainly tantalizing to energy producers and conserva
tionists alike.
Grasslands
Grassland Types
As mentioned above, ecosystems in which grasses and
grass like plants (including sedges and rushes and collec
tively known as graminoids) dominate the vegetation are
termed grasslands. In its narrow sense, ‘grassland’ may be
defined as ground covered by vegetation dominated by
grasses, with little or no tree cover. UNESCO defines
grassland as ‘‘land covered with herbaceous plants with
less than 10 percent tree and shrub cover’’ and wooded
grassland as 10–40% tree and shrub cover. Grassland
ecosystems are notable for two characteristics: they have
properties that readily allow for agricultural exploitation
through the management of domesticated plants or her
bivores, and a climate that is quite variable both spatially
and temporally. They are found in regions where drought
is fairly common but where precipitation is sufficient for
their growth. In addition, they can also dominate wetlands
in both freshwater and coastal regions. They also occur in
sites where more predictable rainfall occurs and soils are
shallow or poorly drained, or in areas with topography too
steep for woody plants. To put it simply, grasslands
usually occupy that area between wetter areas dominated
by woody plants and arid desert vegetation.
Grassland biomes occur on every continent except
Antarctica. It is estimated that grasslands once covered
It is estimated that prior to the European settlement of
North America, the largest continuous grasslands in the
United States stretched across the Great Plains from the
Rocky Mountains and deserts of the Southwestern states
to the Mississippi river. Other extensive grasslands are, or
were, found in Europe, South America, Asia, and Africa
(Figure 2). Grasslands can be broadly categorized as
temperate or tropical. Temperate grasslands have cold
winters and warm to hot summers and often have deep
fertile soils. Surprisingly, plant growth in temperate grass
lands is often nutrient limited because much of the soil
nitrogen is stored in forms unavailable for plant uptake.
These nutrients, however, are made available to plants
when plowing disrupts the structure of the soil. The
combination of high soil fertility and relatively gentle
topography made grasslands ideal candidates for conver
sion to crop production and thus have led to the demise of
much of the grasslands across the world.
Grasslands in the Midwestern United States that receive
the most rainfall (75–90 cm) are the most productive and are
termed tallgrass prairies. Historically, these were most abun
dant in Iowa, Illinois, Minnesota, Missouri, and Kansas. The
driest grasslands (25–35 cm of rainfall) and least productive
Steppes and Prairies
Percent grasslands
<25%
25–50%
50–75%
>75%
No data
407
C 2003 World Resources Institute
Figure 2 Map of the grasslands of the world. World Resources Institute – PAGE, 2000. Sources: GLCCD, 1998. Loveland TR, Reed
BC, Brown JF, et al. (1998) Development of a Global Land Cover Characteristics Database and IGBP DISCover from 1 km AVHRR Data.
International Journal of Remote Sensing 21(6–7): 1303–1330. Available online at http://edcaac.usgs.gov/glcc/glcc.html. Global Land
Cover Characteristics Database, Version 1. Olson JS (1994) Global Ecosystem Framework – Definitions, 39pp. Sioux Falls, SD: USGS
EDC.
are termed shortgrass prairie or steppe. These grasslands are
common in Texas, Colorado, Wyoming, and New Mexico.
Grasslands intermediate between these extremes are termed
mid or mixed grass prairies. In tallgrass prairie, the grasses
may grow to 3 m tall in wet years. In shortgrass prairie,
grasses seldom grow beyond 25 cm in height. In all tempe
rate grasslands, production of root biomass belowground
exceeds foliage production aboveground. Worldwide, other
names for temperate grasslands include steppes throughout
most of Europe and Asia, veld in Africa, puszta in Hungary,
and the pampas in South America.
Tropical grasslands are warm throughout the year but
have pronounced wet and dry seasons. Tropical grassland
soils are often less fertile than temperate grassland soils,
perhaps due to the high amount of rainfall (50–130 cm) that
occurs during the wet season and washes (or leaches) nutri
ents out of the soil. Most tropical grasslands have a greater
density of woody shrubs and trees than temperate grass
lands. Some tropical grasslands can be more productive
than temperate grasslands. However, other tropical grass
lands grow on soils that are quite infertile or these
grasslands are periodically stressed by seasonal flooding.
As a result, their productivity is reduced and may be similar
to that of temperate grasslands. As noted for temperate
grasslands, root production belowground far exceeds foliage
production in all tropical grasslands. Other names for
tropical grasslands include velds in Africa, and the compos
and llanos in South America.
Although temperate and tropical grasslands encompass
the most extensive grass dominated ecosystems, grasses are
present in most types of vegetation and regions of the world.
Where grasses are locally dominant they may form desert
(see Deserts) grassland, Mediterranean (see Mediterranean)
grassland, subalpine and alpine grasslands (sometimes
referred to as meadows or parks), and even coastal grassland.
Most grasslands are dominated by perennial (long lived)
plants, but there are some annual grasslands in which the
dominant species must reestablish each year by seed.
Intensively managed, human planted, and maintained grass
lands (e.g., pastures, lawns) occur worldwide as well.
The Grassland Environment
Grassland climates can be described as wet or dry, hot or
cold (typically in the same season), but on average are
intermediate between the climates of deserts and forests.
The climate of grasslands is best described as one of
extremes. Average temperatures and yearly amounts of
rainfall may not be much different from desert or forested
areas, but dry periods during which the plants suffer from
water stress occur in most years in both temperate and
408
Steppes and Prairies
Raceme
Panicle
Spike
Rachis
Spikelet
Lemma
Caryopsis (seed)
Palea
Leaf blade
Ligule
Glumes
Culm (stem)
One floret
per spikelet
Internode
Collar
Auricle
Sheath
Veins
Rachilla
Node
Glumes
Several florets
Sterile
shoot
Leaf blade
Sheath
Stolon
Soil surface
Crown
Rhizome
Figure 3 Structure and architecture of the grass plant. From Ohlenbusch et al. (1983).
tropical grasslands. An excellent example of this comes
from North America, where in the area around
Washington, DC (dominated by eastern deciduous forest),
the annual precipitation is 102 cm whereas at Lawrence,
KS (dominated historically by tallgrass prairie), the annual
precipitation is 100 cm. But the way the rainfall is dis
tributed is notably different. At Lawrence, KS, over 60% of
the rainfall occurs in the growing season (April–
September), whereas at Washington, DC, the precipitation
is uniformly distributed throughout the year. The open
nature of grasslands is accompanied by the presence of
sustained high wind speeds. Windy conditions increase
the evaporation of water from grasslands and this increases
water stress in the plants and animals. Another factor that
increases water stress is the high input of solar radiation in
these open ecosystems. This leads to the convective uplift
of moist air and results in intense summer thunderstorms.
Rain falling in these intense storms may not be effectively
captured by the soil and the subsequent runoff of this water
into streams reduces the moisture available to grassland
plants and animals. In addition to periods of water stress
within the growing season, consecutive years of extreme
drought are more common in grassland than in adjacent
forested areas. Such droughts may kill even mature trees,
but the grasses and other grassland plants have extensive
root systems and belowground buds that help them survive
and grow after drought periods (Figure 3).
Fire in Grasslands
It is generally recognized that climate, fire, and grazing
are three primary factors that are responsible for the
origin, maintenance, and structure of the most extensive
natural grasslands. These factors are not always indepen
dent (i.e., grazing reduces standing crop biomass which
can be viewed simply as a fuel for fire, and biomass is also
highly dependent upon the amount of precipitation).
Historically, fires were a frequent occurrence in most
large grasslands. Most grasslands are not harmed by fire,
409
Steppes and Prairies
Figure 4 Photograph of a spring fire at the Konza Prairie
Biological Field Station. The fire in the background is occurring
2 weeks after the area in the foreground was burned.
Photograph by Alan K. Knapp.
Unburned
Annually burned
700
ANPP (g m–2)
600
500
400
300
1500
200
1000
100
0
500
0
1975
1980
1985
1990
Year
1995
Precipitation (mm)
many benefit from fire, and some depend on fire for their
existence. When grasses are dormant, the moisture con
tent of the senesced foliage is low and this fine textured
fuel ignites easily and burns rapidly. The characteristic
high wind speeds and lack of natural fire breaks in grass
lands allow fire to cover large areas quickly. Because fire
moves rapidly and much of the fuel is above the ground,
temperatures peak rapidly and soil heating into the range
that is biological damaging (>60 C) occurs for only a
short period of time and only at the surface or maybe a
few centimeters into the soil. Thus, the important parts of
the grasses (roots and buds) have excellent protection
against even the most intense grass fires. Fires have been
documented to be started by lightning and set intention
ally by humans in both tropical and temperate grasslands.
Fires are most common in grasslands with high levels of
plant productivity, such as tallgrass prairies, and in these
grasslands fire is important for keeping trees and adjacent
forests from encroaching into grasslands. Many tree spe
cies are killed by fire, or if they are not killed, they are
damaged severely because their active growing points are
aboveground. Grassland plants survive and even thrive
after fire because their buds are belowground where they
are protected from lethal temperatures (Figure 4).
The response of grassland species to fire mostly
depends upon the production potential of the grassland.
In the more highly productive grasslands (e.g., tallgrass
prairie), fire in the dormant season (usually right before
the growing season) results in an increase in growth of the
grasses and thus greater plant production or total biomass.
This occurs because the buildup of dead biomass (detri
tus) from previous years inhibits growth; fire removes this
layer. However, in drier grasslands, or even in years in
productive grasslands when the precipitation is low, the
burning of this dead plant material may cause the soil
to become excessively dry due to high evaporation losses.
As a result, plants become water stressed and growth is
2000
Figure 5 Long-term record (26 years) of aboveground net
primary production (ANPP) at Konza Prairie Biological Field
Station from unburned sites (clear triangles) and annually burned
sites (solid circles). The growing season precipitation (April–
September; solid bars) and annual precipitation (clear bars) is
also shown.
reduced after fire, thus resulting in lower productivity. It
is only with long term data that the true impact of fires on
grasslands can be determined (Figure 5).
So what are the mechanism(s) behind the increase in
production in mesic grasslands after a fire? One of the
most common misconceptions is that fire in grasslands
increases productivity by increasing (releasing) the
amount of nitrogen (N), a key limiting nutrient in terres
trial ecosystems. Actually, soil N decreases with burning.
However, as mentioned above, the primary mechanism
by which fire increases production in tallgrass prairie is
through the removal of the accumulation of detritus pro
duced in previous years. Standing dead biomass has been
reported to accumulate to levels of up to 1000 g m 2 in
tallgrass prairie and a steady state is achieved c. 3–5 years
after a fire. The specific effects of this blanket of dead
biomass on production are numerous and manifest on
individual through the ecosystem levels. This detritus
may accumulate to >30 cm deep, and this nonphotosyn
thetic biomass shades the soil surface and emerging
shoots. This reduction in light available to shoots in sites
without fire occurs for up to 2 months and because soil
moisture is usually high in the spring, loss of energy at this
time is especially critical for primary production. In con
cert with reductions in light available to the grasses, the
early spring temperature environment is much different
between burned and unburned sites, with burned sites
having a higher temperature favoring the dominant C4
grasses. All of these factors result in less production in
unburned tallgrass compared to annually burned prairie
(Figure 5). Other evidence that fire does not increase N
availability in mesic grasslands comes from N fertilization
experiments. Within tallgrass prairie, in annually burned
sites, N fertilizer had a strong impact on production, but
in sites that have not been burned for several years,
additional N did not enhance production and sites with
410
Steppes and Prairies
intermediate fire histories had intermediate responses to
N fertilization. The results of many studies suggest that
one generality regarding grasses and fire is that grasses
tolerate fire extremely well and in most cases reach their
maximum production in the immediate post fire years.
One qualification to this statement is that the beneficial
effect of fire is not uniform across all precipitation gradi
ents. In addition, the growth form type of the dominant
grass is also very important. Highly productive grasslands
on the high end of precipitation gradients show moderate
to high positive response to burning whereas more arid
grasslands and some bunchgrass grasslands show reduced
productivity in the first few years after fire.
Most grasslands have an active growing season as well
as a dormant season. Although fire can occur year round
in many grasslands, fire is most likely to occur during the
dormant season and it is most rare in the middle of the
growing season during normal (non drought) years. Given
the fact that so many aspects of a grassland change during
the yearly cycle, it seems fair to expect that a fire in
different seasons would have dramatically different
impacts. However, in spite of the many studies that have
examined the impact of fires at different times of the year,
there does not seem to be a general consensus on fire
seasonality. Rather, it is probably best to say that grass
lands seem somewhat sensitive to ‘season of burn’. In one
long term study, it was found that the dominant grass in
the tallgrass prairie (Andropogon gerardii ) increased with
burning in autumn, winter, or spring (dormant season),
whereas burning in summer (growing season) resulted in
an increase in many of the subdominant grasses with a
reduction in A. gerardii.
Research indicates that community structure and eco
system functioning in grasslands are impacted strongly by
fire frequency. Plant species composition, in particular,
differs dramatically between annually burned and less
frequently burned sites in mesic grasslands. In tallgrass
prairie, annually burned sites are dominated strongly by
C4 perennial grasses. Although C4 grasses retain domi
nance at infrequently burned sites, C3 grasses, forbs, and
woody species are considerably more abundant resulting
in greater diversity and heterogeneity in unburned
prairie. In fact, the flora on annually burned sites is a
nested subset of that found on less frequently burned
areas. Thus, the differences reflect shifts in dominance
between frequently and infrequently burned sites, rather
than difference in composition per se. Again as with
response of production to fire, there appears to be a
gradient of response in community structure to grassland
fires. In more northern prairies of North America, burning
has not been shown to strongly affect community struc
ture. However, these northern grasslands are dominated
by C3 grasses, which tend to decrease with burning, unlike
the C4 grasses that dominate prairies in warmer climates.
Thus, the role of competition and fire in structuring
grassland plant communities may increase along a latitu
dinal gradient throughout the Great Plains.
At a mesic grassland (Konza Prairie Biological Station),
a clear picture of fire effects on plant community structure
has emerged from the long term (>20 years) empirical and
experimental research done at the site. In the absence of
large herbivores, the system is strongly driven by
bottom up forces associated with light, soil resource
availability, and differential ability to compete under
low resource conditions. Although light availability
increases with burning, the abundance of other critical
limiting resources, N and water, declines as fire frequency
increases. This is especially true in upland areas (with
shallow soils) where production is likely limited by
water. These changes in resource availability favor the
growth and dominance of a small number of perennial
C4 grasses and forbs. As dominance by these competitive
species increases, general declines in plant species diver
sity and community heterogeneity occur.
Impact of Fire on Consumers
Direct effects
Most grassland animals are not harmed by fire, particularly
if fires occur during the dormant season. Those animals
living belowground are well protected, and most grassland
birds and mammals are mobile enough to avoid direct
contact with fire. For example, there were few differences
in the kinds and abundances of ground dwelling beetles in
frequently and infrequently burned Kansas tallgrass prairie.
Insects that live in and on the stems and leaves of the plants
are the ones that are most affected by fire. Fire has been
shown to reduce directly the abundance of caterpillars
which means fewer butterflies, which are important polli
nators, in frequently burned prairies. Fortunately, most
natural fires are patchy in that many unburned areas
remain throughout a larger burned area. These patches
serve as refugia for many insect populations. Given that
these animals have short generation times these refugia
often allow insect populations to recover quickly following
a fire.
Indirect effects
Given the distinct effects of fire frequency on plant com
munity structure and dynamics within and among
burning treatments, it seems plausible that consumers
that depend on the primary producers for food and habitat
structure will be indirectly affected because fire alters
food availability and habitat structure. Given that fire
usually homogenizes grassland plant communities, one
would predict that this would hold true for consumers.
However, there does not appear to be tight linkages
between changes in vegetation composition and structure
animal populations. Indeed, work in an Oklahoma prairie
shows that more grassland birds occur in areas with
Steppes and Prairies
Grazing in Grasslands
Grazing is a form of herbivory in which most of the leaves
or other plant parts (small roots and root hairs) are
consumed by herbivores. Grazing, both above and
belowground, is an important process in all grasslands.
The long association of grazers and grasslands has
prompted the hypothesis that grasses and their megaher
bivore grazers are a highly coevolved system, but, as
mentioned above, there is some more recent evidence
that this might not be the case. However, there is no
disagreement that large grazers have been a factor in
grassland ecology since their origin. The herbivory
actions of many other smaller organisms including small
mammals and insects may be equally important. There is
no doubt that the impact of native grazers in grasslands
can be extensive and work on the East African Serengeti
plains estimated that 15% to >90% of the annual above
ground net primary productivity can be consumed by
ungulates. However, data from small mammal exclosures
suggest that small mammals can also impact grasslands as
when small mammals were excluded from plots in Kenya;
biomass was 40–50% higher than in adjacent plots where
small mammals occurred.
Due to the ability of grasses to cope with high rates of
herbivory, many former natural grasslands are now being
managed for the production of domestic livestock, pri
marily cattle in North and South America and Africa, as
well as sheep in Europe, New Zealand, and other parts of
the world. Grasslands present a vast and readily exploited
resource for domestic grazers. However, like many
resources, grasslands can be overexploited (discussed in
more detail below).
Grazing systems can be roughly divided into two main
types – commercial and traditional – with the traditional
type often mainly aimed at subsistence. Commercial graz
ing of natural grasslands is very often at a large scale and
commonly involves a single species, usually beef cattle or
sheep for wool production. Some of the largest areas of
extensive commercial grazing developed in the nine
teenth century on land which had not previously been
heavily grazed by ruminants; these grazing industries
were mainly developed in the Americas and Australia,
and to a much less degree in southern and eastern Africa.
Traditional livestock production systems vary according
to climate and the overall farming systems of the area.
They also use a wider range of livestock, including buffa
loes, asses, goats, yaks, and camels. In traditional farming
systems, livestock are often mainly kept for subsistence
and savings, and are frequently multipurpose, providing
meat, milk, and manure as fuel.
Grazing aboveground by large herbivores alters grass
lands in several ways. Grazers remove fuel and may
lessen the frequency and intensity of fires. Most large
grazers such as cattle or bison primarily consume the
grasses; thus the less abundant forb species (broad leafed,
herbaceous plants) may increase in abundance and new
species may invade the space that is made available.
Thus, fire reduces heterogeneity in mesic grassland
(a few species dominate) while grazers increase hetero
geneity regardless of fire frequency. In other words,
grazing decouples the impact of fire in productive grass
lands (Figure 6). As a result; grazing increases plant
species diversity in mesic grasslands. In xeric grasslands,
on the other hand, grazing may lower species diversity,
particularly by altering the availability of suitable micro
sites for forb species. These effects are strongly
dependent on grazing intensity. Overgrazing may rapidly
degrade grasslands to systems dominated by weedy and
non native plant species.
100
Ungrazed
90
80
70
60
50
40
Cover (%)
patchy burns than in areas that are uniformly burned or
not burned. Much more work on how fire affects habitat
heterogeneity and grassland consumers communities is
needed.
411
30
Grazed
100
90
C4 grass
80
Forbs
70
60
50
40
30
0
2
4
6
8 10 12 14 16 18 20
Number of fires
Figure 6 Aboveground biomass removal by large ungulates
modulates plant community responses to fire in mesic
grasslands. In ungrazed prairie (top), cover of dominant C4
grasses increased with increasing fire frequency, while cover of
forbs decreased, resulting in a loss of diversity. However, in
prairie grazed by bison (bottom), the cover of forbs was positively
correlated with fire frequency and the cover of grasses was
unaffected, resulting in high diversity in spite of frequent fires.
From Collins SL, Knapp AK, Briggs JM, et al. (1998) Modulation
of diversity by grazing and mowing in native tallgrass prairie.
Science 280(5364): 745–747.
412
Steppes and Prairies
Grazers may also accelerate the conversion of plant
nutrients from forms that are unavailable for plant uptake
to forms that can be readily used. Essential plant nutrients,
such as nitrogen, are bound for long periods of time in
unavailable (organic) forms in plant foliage, stems, and
roots. These plant parts are slowly decomposed by
microbes and the nutrients they contain are only gradually
released in available (inorganic) forms. This decomposition
process may take more than a year or two. Grazers con
sume these plant parts and excrete a portion of the
nutrients they contain in plant available forms. This hap
pens very quickly compared to the slow decomposition
process, and nutrients are excreted in high concentrations
in small patches. Thus, grazers may increase the availabil
ity of potentially limiting nutrients to plants as well as alter
the spatial distribution of these resources.
Some grasses and grassland plants can compensate for
aboveground tissue lost to grazers by growing faster after
grazing has occurred. Thus, even though 50% of the grass
foliage may be consumed by bison or wildebeest, when
compared to ungrazed plants at the end of the season, the
grazed grasses may be only slightly smaller, the same size,
or even larger than ungrazed plants. This latter phenom
enon, called ‘overcompensation’ is controversial, yet the
ability of grasses to compensate partially or fully for foliage
lost to grazers is well established. Compensation occurs for
several reasons, including an increase in light available to
growing shoots in grazed areas, greater nutrient availability
to regrowing plants, and increased soil water availability.
The latter occurs after grazing because the large root
system of the grasses is able to supply abundant water to
a relatively small amount of regrowing leaf tissue.
As with fire, the impact of grazing on grasslands
depends upon where in the precipitation gradient the
grassland occurs (usually more mesic grasslands can
recover more quickly than arid grasslands) as well as the
growth form – cespitose (bunch forming grasses) versus
rhizomatous grasses. But another key factor is the evolu
tionary history of the grassland. In general, grasslands with
a long evolutionary history of grazers, as in Africa, are very
resilient to grazing whereas grasslands with a short evolu
tionary history such as desert grasslands in North America
can easily be damaged by even light grazing.
Threats to Grasslands and Restoration of
Grasslands
Grassland environments are key agricultural areas world
wide. In North America and elsewhere, grasslands are
considered to be endangered ecosystems. For example,
in US Great Plains up to 99% of native grassland ecosys
tems in some states have been plowed and converted to
agricultural use or lost due to urbanization. Similar but
less dramatic losses of mixed and shortgrass prairies have
occurred in other areas. While the loss of native grass
lands due to agricultural conversion is still occurring in
some places, dramatic increases in woody shrub and tree
species threatens many remaining tracts of grasslands.
Indeed, across the world, the last remaining native grass
lands are being threatened by an increase in the
abundance of native woody species from expansion of
woody plant cover originating from both within the eco
system and from adjacent ecosystems. Increased cover
and abundance of woody species in grasslands and savan
nas have been observed worldwide with well known
examples from Australia, Africa, and South America.
In North America, this phenomenon has been documented
in mesic tallgrass prairies of the eastern Great Plains, sub
tropical grasslands and savannas of Texas, desert grasslands
of the Southwest, and the upper Great Basin. Purported
drivers of the increase in woody plant abundance are
numerous and include changes in climate, atmospheric
CO2 concentration, nitrogen deposition, grazing pressure,
and changes in disturbance regimes such as the frequency
and intensity of fire. Although the drivers vary, the con
sequences for grassland ecosystems are strikingly
consistent. In many areas, the expansion of woody species
increases net primary production and carbon storage, but
reduces biodiversity. The full impact of shrub encroach
ment on grassland environments remains to be seen.
Another threat to native grasslands is the increase of non
native grass species. For example, in California, it is estimated
that an area of approximately 7 000 000 ha (about 25% of the
area of California) has been converted to grassland domi
nated by non native annuals primarily of Mediterranean
origin. Conversion to non native annual vegetation was so
fast, so extensive, and so complete that the original extent and
species composition of native perennial grasslands is
unknown. In addition, across the western United States,
invasive exotic grasses are now dominant in many areas
and these species have a significant impact on natural dis
turbance regimes. For example, the propensity for annual
grasses to carry and survive fires is now a major element in
the arid and semiarid areas in western North America. In the
Mojave and Sonoran deserts of the American Southwest, in
particular, fires are now much more common than they were
historically, which may reduce the abundance of many native
cactus and shrub species in these areas. This annual grass fire
syndrome is also present in native grasslands of Australia and
managers there and in North America are using growing
season fire to try to reduce the number of annual plants
that set seed and thus reduce the populations of exotics,
usually with very mixed results.
Conservation and Restoration
Because grasslands have tremendous economic value as
grazing lands and also serve as critical habitat for many
plant and animal species, efforts to conserve the
Steppes and Prairies
remaining grasslands and restore grasslands on agricul
tural land are underway in many states and around the
world. The most obvious conservation practice is the
protection and management of existing grasslands. This
includes both private and public lands. Probably the lar
gest private holder of grasslands in the world is The
Nature Conservancy. The Nature Conservancy is a glo
bal organization that works in all 50 states in the United
States of America, and in 27 countries, including Canada,
Mexico, Australia, and countries throughout the Asia
Pacific region, the Caribbean, and the Latin America.
However, as mentioned numerous times, the factors
that led to the establishment of grasslands and, in particu
lar, the organic rich soils derived from the dominant biota
have facilitated the agricultural exploitation of grasslands.
Consequently, many grasslands that were historically per
sistent have been converted to cropland. Thus, restoration
of grasslands is also a very important conservation prac
tice. Grassland restoration is the process of recreating
grassland (including plant and animal communities, and
ecosystem processes) where one existed but now is gone.
Grassland restoration can include planting a new grassland
where one had been broken and farmed, or it can include
improving a degraded grassland (e.g., one that was never
plowed but lost many plant and animal species due to
prior land management practices). Restoration practices
of existing grasslands may include reintroducing fires into
grasslands following extended periods of fire suppression.
On areas that have been moderately to heavily grazed (but
not completely overgrazed), reducing the intensity of
grazing may be required. In addition, mowing is also a
cost effective method of restoring grasslands. Mowing can
be effective on sites that have been invaded by brush and
forest, but the grasses are still present.
In areas where the grasses are completely absent (agri
culture fields) or in a very degraded state, reseeding of
grasses is usually necessary. There are proven techniques,
complete with specialized equipment (seed drills) for
restoration of grasslands, and, for the most part, it is fairly
easy to get the dominant grasses established in an area.
Indeed, some of the earliest examples of restoration ecol
ogy come from efforts to restore native tallgrass prairie in
North America. As a result, the market for restoration of
grasslands (at least in North America) has developed to
the point that obtaining enough grass seed (sometimes
even local native seed) is not a problem. A bigger chal
lenge, however, in restored grasslands is increasing
establishment of the nongrass species which are so critical
for biodiversity. Seeds may be more difficult to obtain
(especially for rarer plants), and then getting the forbs to
survive and reproduce in many grassland restoration pro
jects has been challenging. Further research is needed
regarding what management techniques are important to
their establishment and growth in these restored areas.
413
In addition to the prairie flora that is at risk, grassland
animals (particularly birds and butterflies) suffer when
grassland quality declines. In North America, grassland
birds were historically found in vast numbers across the
prairies of the western Great Plains. Today, the birds of
these and other grasslands around the world have shown
steeper, more consistent, and more geographically wide
spread declines than any other group. These losses are a
direct result of the declining quantity and quality of
habitat due to human activities like conversion of native
prairie to agriculture, urban development, and suppres
sion of naturally occurring fire.
See also: Agriculture Systems; Savanna.
Further Reading
Borchert JR (1950) The climate of the central North American
grassland. Annals of the Association of American Geographers
40: 1 39.
Briggs JM, Knapp AK, Blair JM, et al. (2005) An ecosystem in transition:
Woody plant expansion into mesic grassland. BioScience
55: 243 254.
Collins SL, Knapp AK, Briggs JM, Blair JM, and Steinauer EM (1998)
Modulation of diversity by grazing and mowing in native tallgrass
prairie. Science 280(5364): 745 747.
Collins SL and Wallace LL (1990) Fire in North American Tallgrass
Prairies. Norman, OK: University of Oklahoma Press.
Frank DA and Inouye RS (1994) Temporal variation in actual
evapotranspiration of terrestrial ecosystems: Patterns and ecological
implications. Journal of Biogeography 21: 401 411.
French N (ed.) (1979) Perspectives in Grassland Ecology. Results and
Applications of the United States International Biosphere Programme
Grassland Biome Study. New York: Springer.
Knapp AK, Blair JM, Briggs JM, et al. (1999) The keystone
role of bison in North American tallgrass prairie. BioScience
49: 39 50.
Knapp AK, Briggs JM, Hartnett DC, and Collins SL (1998) Grassland
Dynamics: Long Term Ecological Research in Tallgrass Prairie,
364pp. New York: Oxford University Press.
Loveland TR, Reed BC, Brown JF, et al. (1998) Development of a
Global Land Cover Characteristics Database and IGBP DISCover
from 1 km AVHRR Data. International Journal of Remote Sensing
21(6 7): 1303 1330.
McNaughton SJ (1985) Ecology of a grazing ecosystem: The serengeti.
Ecological Monographs 55: 259 294.
Milchunas DG, Sala OE, and Lauenroth WK (1988) A generalized model
of the effects of grazing by large herbivores on grassland community
structure. American Naturalist 132: 87 106.
Oesterheld M, Loreti J, Semmartin M, and Paruelo JM (1999) Grazing,
fire, and climate effects on primary productivity of grasslands and
savannas. In: Walker LR (ed.) Ecosystems of the World,
pp. 287 306. Amsterdam: Elsevier.
Olson JS (1994) Global Ecosystem Framework Definitions, 39pp.
Sioux Falls, SD: USGS EDC.
Prasad V, Stromberg CAE, Alimohammadian H, and Sahni A (2005)
Dinosaur coprolites and the early evolution of grasses and grazers.
Science 310: 1177 1190.
Sala OE, Parton WJ, Joyce LA, and Lauenroth WK (1988) Primary
production of the central grassland region of the United States.
Ecology 69: 40 45.
Samson F and Knopf F (1994) Prairie conservation in North America.
BioScience 44: 418 421.
Weaver JE (1954) North American Prairie. Lincoln, NE: Johnsen
Publishing Company.
414
Swamps
Swamps
C Trettin, USDA, Forest Service, Charleston, SC, USA
Published by Elsevier B.V.
Introduction
General Properties of a Swamp
Ecological Functions
Restoration
Ecosystem Services and Values
Further Reading
Introduction
General Properties of a Swamp
Swamp is a general term that is defined as ‘‘spongy
land, low ground filled with water, soft wet ground’’
(Webster, 1983), hence its association with a wide vari
ety of terrestrial ecosystems. Typically, a swamp is
considered a forested wetland. A wetland is a type of
terrestrial ecosystem that has a hydrologic regime
where the soil is saturated near the surface during the
growing season; the soil has hydric properties, expres
sing characteristics of anaerobic conditions; and the
dominant vegetation is hydrophytic with adaptations
for living in the wet soils. In the case of a swamp, the
forest species are adapted to the wet soil conditions.
Without geographic context, there is little functional
information conveyed by the term swamp other than
the prevalence of wetland conditions and dense forest
vegetation (Figure 1).
The following discussion is designed to convey the
common hydrologic settings, soil conditions, and vege
tative communities that occur within the common
usage of the term swamp. References focusing on
swamp forests should be consulted for specific geo
graphic regions.
Hydrology
The hydrologic setting controls the form and function of
the wetland because of the dependence on excess water
to mediate biological and geochemical reactions. There
are four general settings that may be used to characterize
the swamp hydrology (Figure 2). The riverine or
floodplain setting is the most commonly associated
hydrogeomorphic setting for swamps. It is characterized
by periodic flooding from the river or stream, and it may
also receive runoff from adjoining uplands. The periodi
city, and flood depth and duration are the key factors
that affect the type of forest communities present in the
swamp. Depressional wetlands occur where there are
surface depressions which receive water from the sur
rounding uplands, directly from precipitation, and in
certain instances, they may also intersect a shallow
water table. Lacustrine and estuarine fringe wetlands
Depression
Estuarine fringe
Figure 1 Bottomland hardwood swamp, characteristic of
floodplains in the Southeastern United States.
Riverine
Lacustrine fringe
Figure 2 Influence of geomorphic position on hydrology. The
arrows show the dominant direction of water flow for the four
dominant types of geomorphic positions that are characteristic
for swamps. After Vasander H (1996) Peatlands in Finland, 64pp.
Helsinki: Finnish Peatland Society.
Swamps
receive their water primarily from an open water body;
runoff from adjoining uplands and precipitation also
contribute to the water balance. The common hydrologic
attribute of swamps in each of those settings is the
presence of water above the soil surface, but the period
of inundation varies widely. While it is common for
swamps to have flooded conditions for periods ranging
from days to months on an annual basis, it is not uncom
mon for there to be multiyear intervals between flood
events. The factors that affect the flooding regime
include timing and amount of precipitation, groundwater
level, land use in the watershed contributing to the
swamp, and evapotranspiration.
415
Vegetation
The term swamp generally implies a forested wetland.
However, due to the wide range of physical settings (see
the previous two sections) and geographic locations ran
ging from the boreal to the tropical climatic zones on each
continent, there are no consistent characteristics or attri
butes beyond the occurrence of hydrophytic vegetation.
Accordingly, swamps may be dominated by either con
ifers or angiosperms, but a common situation would be a
mixture of species and communities reflecting relatively
minor differences in a microsite. For example, while a
floodplain forest may be broadly characterized as a bot
tomland hardwood swamp, it contains a mosaic of
vegetative communities which reflect small differences
in hydrology and soils.
Soils
Swamp soils cover the full range of texture classes and
degrees of organic matter accumulation (Figure 3). The
wet mineral soils are characteristic of riverine and
depressional settings. The histic mineral soils have a
moderately thick accumulation of surface organic matter
(<40 cm) reflecting prolonged periods of saturation and
little scouring action if located in a floodplain, hence
they may be found in any of the four hydrologic settings.
The histosols or peat soils have a thick layer (>40 cm) of
organic matter accumulation, representing the long per
iods of saturation on an annual basis. These soils
typically occur in depressional settings and are not com
mon in floodplains due to the periodic scouring that
occurs during flood events.
Mineral
0–5 cm
All drainage
classes, long
periods of aeration
Histic mineral
5–40 cm
Somewhat poorly
drained, frequent periods of
high water table
Peat
(Histosol)
>40 cm
Poorly drained,
long periods of
saturation
Figure 3 Types of soils common in swamp forests. The three
categories reflect the amount of organic matter that has
accumulated on the soil surface, which is in turn controlled by the
soil drainage and hydrology. After Trettin CC, Jurgensen MF,
Gale MR, and McLaughlin JA (1995) Soil carbon in northern
forested wetlands: Impacts of silvicultural practices. In: McFee
WW and Kelly JM (eds.) Carbon Forms and Functions in Forest
Soils, pp. 437–461. Madison, WI: Soil Science Society of
America.
Ecological Functions
The ecological functions of swamps are significant,
because of their prevalence and the wide range of condi
tions that they occupy. The following overview highlights
some of the major ecosystem functions that are provided
from swamp wetlands; specifics for a particular type of
swamp are available from the regional references.
Hydrology
Hydrologic functions that are mediated by swamp wet
lands depend on the hydrogeomorphic setting. Riverine
swamps provide temporary storage for floodwaters,
thereby reducing the peak flow to downstream areas.
This function is physically based, with little interaction
with the type of forest vegetation. However, changes in
land use, especially conversion to agriculture, in the
floodplain, may reduce the water storage potential, result
ing in enhanced downstream conveyance of flow. The
flood storage function also serves to sustain stream flow,
as the waters slowly drain from the area. Swamps occur
ring in a depressional setting may be a source of
groundwater recharge, where accumulated surface water
slowly infiltrates through the subsurface sediments. In
estuarine and lacustrine settings, swamps occurring at
the land–water margin are important for the stability of
the shoreline.
Water quality
The effects of a swamp on water quality depend on the
hydrogeomorphic setting. The riverine swamp affects
water quality in two primary ways – by physical and
biogeochemical reactions. Sediment removal is an impor
tant function of the riverine swamps; this is a process
where sediment in the floodwaters settles out onto the
416
Swamps
floodplain surface. The deposited sediment provides
nutrients to the swamp vegetation and it represents
the removal of a contaminant from the floodwater.
Floodplains with dense understory vegetation can be
more effective than open forest settings in filtering sedi
ment from the floodwaters.
The floodplain and riparian zone swamps may also
remove chemical constituents from the water, particularly
nitrogen and phosphorus. As a result of the anaerobic soil
conditions, nitrate nitrogen, which is a common pollutant
in surface and shallow subsurface runoff, can be con
verted to nitrogen gas, thereby removing it from the
water. The removal of phosphorus compounds typically
involves reactions associated with the sediments.
restoring the converted wetlands back to swamp forests
include the reestablishment of flood water storage, in
the case of floodplains, and the development of wildlife
habitat. The restoration of swamp forests is complicated
by the myriad of soil and hydrologic conditions that
one may encounter, and the effects of past management
practices which necessitate the restoration may also
exacerbate the situation. However, with proper consid
eration of the hydrologic setting and matching species
to the soil and water regimes, functional restoration is
feasible. The typical sequence of restoring swamp for
ests is to reestablish the wetland hydrology by blocking
drainage ditches, and planting appropriate tree and
understory species.
Habitat
Swamps are important for the diversity of habitat condi
tions that they provide. At the large scale, swamps
comprise part of the mosaic of land types, yielding wet,
vegetative conditions among uplands. At smaller scales,
within a swamp, there are a multitude of habitat condi
tions that are largely dictated by elevation relative to the
mean high water level.
Terrestrial
The terrestrial habitats provided by swamps are diverse
due to variations in vegetative composition and structure,
which are largely regulated by the hydrologic conditions
of the site. The habitat also changes through the devel
opment of the forest. In early successional stages, the
vegetation is typically a dense combination of shrubs
and trees; then, as the trees gain dominance, the shrub
layers die back yielding a less dense understory.
Correspondingly, the habitat conditions for amphibians,
birds, reptiles, and mammals change as the stand evolves.
The swamp forests are particularly important habitat for
birds, especially migratory song birds.
Aquatic
Swamps also provide important aquatic habitat for fish,
birds, and amphibians. Organic matter produced in the
swamp is an important energy source for aquatic organ
isms, including those living in water bodies within the
swamp and also larger receiving bodies such as lakes,
rivers, and oceans. In floodplains, the floating debris and
logs provide physical structures that are an important
component of the aquatic habitat.
Restoration
In many areas, swamps have been converted into agri
cultural use, through the use of drainage systems and
clearing of the forest vegetation. The merits of
Ecosystem Services and Values
Swamps provide both direct and indirect values to society.
Direct values include raw materials, such as timber and
food stocks. Indirect values include floodwater storage,
water supply, water quality, recreation, esthetics, wildlife
diversity, and biodiversity. The valuation will depend on
inherent characteristics of the resource that are largely
constrained by the biogeographic zone and location within
a watershed, societal norms, and economic conditions.
Further Reading
Barton C, Nelson EA, Kolka RK, et al. (2000) Restoration of a severely
impacted riparian wetland system The Pen Branch Project.
Ecological Engineering 15: S3 S15.
Burke MK, Lockaby BG, and Conner WH (1999) Aboveground
production and nutrient circulation along a flooding gradient in a
South Carolina Coastal Plain forest. Canadian Journal of Forest
Research 29: 1402 1418.
Conner WH and Buford MJ (1998) Southern deepwater swamps.
In: Messina MG and Conner H (eds.) Southern Forested Wetlands
Ecology and Management, pp. 261 287. Boca Raton, FL: CRC
Press.
Conner WH, Hill HL, Whitehead EM, et al. (2001) Forested wetlands of
the Southern United States: A bibliography. General Technical
Report SRS 43, 133pp. Asheville, NC: US Department of Agriculture,
Forest Service, Southern Research Station.
Conner RN, Jones SD, and Gretchen D (1994) Snag condition and
woodpecker foraging ecology in a bottomland hardwood forest.
Wilson Bulletin 106(2): 242 257.
Conner WH and McLeod K (2000) Restoration methods for deepwater
swamps. In: Holland MM, Warren ML, and Stanturf JA (eds.)
Proceedings of a Conference on Sustainability of Wetlands and
Water Resources, 23 25 May. Oxford, MS: US Department of
Agriculture, Forest Service, Southern Research Station.
de Groot R, Stuip M, Finlayson M, and Davidson N (2006) Valuing
wetlands: Guidance for valuing the benefits derived from wetland
ecosystem services. Ramsar Technical Report No. 3, CBD Technical
Series No. 27, Convention on Biological Diversity. Gland,
Switzerland: Ramsar Convention Secretariat. http://www.cbd.int/
doc/publications/cbd ts 27.pdf (accessed November 2007).
Messina MG and Conner WH (eds.) (1998) Southern Forested
Wetlands Ecology and Management, 347pp. Boca Raton, FL: CRC
Press.
Mitch WJ and Gosselink JG (2000) Wetlands, 920pp. New York: Wiley.
Temperate Forest 417
National Wetlands Working Group (NWWG) (1988) Wetlands of Canada.
Ecological Land Classification Series, No. 24, 452pp. Ottawa:
Sustainable Development Branch, Environment Canada.
Stanturf JA, Gardiner ES, Outcalt K, Conner WH, and Guldin JM (2004)
Restoration of southern ecosystems. In: General Technical Report
SRS 75, pp. 123 11. Asheville, NC: US Department of Agriculture,
Forest Service, Southern Research Station.
Trettin CC, Jurgensen MF, Gale MR, and McLaughlin JA (1995) Soil
carbon in northern forested wetlands: Impacts of silvicultural
practices. In: McFee WW and Kelly JM (eds.) Carbon Forms and
Functions in Forest Soils, pp. 437 461. Madison, WI: Soil Science
Society of America.
Vasander H (1996) Peatlands in Finland, 64pp. Helsinki: Finnish
Peatland Society.
Webster N (1983) Unabridged Dictionary, 2nd edn. Cleveland, OH:
Dorset and Baber.
Relevant Websites
http://www.aswm.org Association of State Wetland Managers.
http://www.ncl.ac.uk Mangrove Swamps WWW Sites,
Newcastle University.
http://www.ramsar.org Ramsar Convention on Wetlands.
http://www.sws.org Society of Wetland Scientists.
http://www.epa.gov Wetlands at US Environmental
Protection Agency.
http://www.wetlands.org Wetlands International.
http://www.panda.org World Wildlife Fund.
Temperate Forest
W S Currie and K M Bergen, University of Michigan, Ann Arbor, MI, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Physiography, Climate, and the Temperate Forest
Biome
Disturbance and Forest Structure
Ecological Communities and Succession
Water and Energy Flow, Nutrient Cycling, and Carbon
Balance
Temperate Forest Land Cover
Further Reading
Introduction
temperate forest (sometimes called subtropical ever
green), and the temperate rainforest. Major taxa include
pines (Pinus spp.), maples (Acer spp.), beeches (Fagus spp.,
Nothofagus spp.), and oaks (Quercus spp.) in the mixed
deciduous and mixed evergreen temperate forests;
spruces (Picea spp.), Douglas fir (Pseudotsuga menziesii),
and redwoods (Sequoia sempervirens, Sequoiadendron gigan
teum) in the Northern Hemisphere temperate rainforests;
and southern beeches (Nothofagus spp.) and eucalyptus
(Eucalyptus spp.) in Southern Hemisphere temperate
rainforests.
Within a continent, forests in the temperate biome
grade into subdivisions based on latitude, elevation, and
large scale patterns of precipitation. In North America,
for example, the predominant natural vegetation in the
eastern United States and the southern reaches of eastern
Canada is mixed deciduous temperate forest. This forest
grades to the south through broad leaved coniferous
mixtures to the mixed evergreen forest along the
Atlantic coastal plain (Figure 2). Temperate rainforests
in North America are found in the coastal Pacific
Northwest where marine climates together with oro
graphic lifting produce high rainfall. In South America
temperate forests are found in Chile and parts of
Patagonia. In Europe, within the temperate forest biome
the mixed deciduous forests dominate in the western
The temperate forest biome is characterized by a distinct
seasonality that includes a long growing season together
with a cold winter season in which much of the vegetation
may be dormant. The strong seasonality drives physiolo
gical events to occur at regular annual intervals for plant
species. These include bud break, flowering, and foliar
and shoot extension. As the growing season ends, marked
by dropping temperatures and shortening photoperiod
(day length), trees and shrubs undergo seasonal physiolo
gical changes that include the senescence and abscission
of foliage (although in evergreen species some foliage is
also retained) and the setting of buds for the next growing
season. Because of the cold winters, the dominant woody
vegetation is characterized by freeze hardy species.
During the winter season, the air temperature drops
below freezing and soils are frozen or cold and wet,
impeding decomposition of plant litter and promoting
the accumulation of an organic layer on the soil surface.
The temperate forest is distributed over portions of
five regions of the globe: North America, South America,
Europe, Asia, and Australia–New Zealand (Figure 1).
Within this biome, distinct biogeographic units are recog
nized, particularly the mixed deciduous temperate forest
(the largest in terms of area), the mixed evergreen
418
Temperate Forest
N
Temperate forest
0
2 500
5 000
Km
Figure 1 The distribution of temperate forests of the world. Map data: Olson DM, Dinerstein E, Wikramanayake ED et al. (2001)
Terrestrial ecoregions of the world: A new map of life on earth, BioScience 51: 933–938. Map prepared by the Environmental Spatial
Analysis Laboratory, University of Michigan, USA, 2006.
Figure 2 The edge of a mixed coniferous–deciduous forest in
southeastern Maine, USA. Photo by W. S. Currie.
continent, Great Britain, southern Eastern Europe, and
southern European Russia. In Near East Asia, the temper
ate forest occurs in Turkey and Iran and a narrow band is
found in Central Asia as a transition between the boreal
forest to the north (see Boreal Forest) and steppe to the
south. The temperate forests in East Asia occur predomi
nantly in northern and central China, but also over most
of Japan, Korea, and of the southern tip of Siberia.
Temperate forests, including rainforests, are also found
in parts of New Zealand, the southeast coast of Australia,
and Tasmania.
Temperate forests are distinguished from boreal forests
by having a 4–6 month (140–200 days) frost free growing
season with on average at least 4 months at 10 C or above
and mean annual temperatures from 5 C to 20 C. At
higher latitudes, the temperate forest transitions to the
boreal forest (see Boreal Forest), a biome of evergreen
cold tolerant forests with much shorter growing seasons.
The latter are also found in middle latitudes as montane
forests at high elevations and are often closer, floristically
and functionally, to boreal than to temperate forests.
The occurrence of frost (at 0 C or colder) differentiates
the extratropical (including temperate) from tropical
regions (see Tropical Rainforest). Moisture also distin
guishes temperate regions from drier forested regions,
such as chaparral (see Chaparral) and wetter forested
regions such as tropical rainforests (see Tropical
Rainforest). In temperate regions, precipitation exceeds
potential evaporation and water is available at approxi
mately 50–200 cm yr 1. Precipitation in most temperate
regions is fairly evenly distributed throughout the year in
contrast to the tropics where there are typically pronounced
wet and dry seasons.
Physiography, Climate, and
the Temperate Forest Biome
Climatic and Physiographic Controls on
the Distribution of Temperate Forests
The geographic distributions of the different vegetation
biomes of the world are dependent on the physical envi
ronment and climate in the form of light, temperature,
and moisture. In middle latitudes (30 –60 N and S), these
controls result in a temperate forest biome within each
Temperate Forest 419
hemisphere that is discontinuous, separated by the oceans
and the tropics, and by moisture and physiographic bar
riers. The present day distribution of temperate forests
derives not only from present climatic controls but also
from paleoclimates and past connections among the con
tinents. Climates during the Pleistocene (c. 1.8 million to
10 000 years ago) set the stage for the present day dis
tribution. During glacial maxima, ice sheets covered large
parts of Europe and North and South America as well as
isolated areas in East Asia. In North and South America,
plants migrated to unglaciated refugia and re migrated,
as glaciers receded, to their present day distributions.
Evidence suggests that many genera of forest trees that
remain in North America and unglaciated East Asia were
extirpated from Europe because the east–west running
Alpine range blocked migrations to refugia during
Pleistocene glaciations. Similarly important were conti
nental connections between North America (the
Nearctic), East Asia, and Europe (the Palearctic) at dif
ferent points in geologic history. As a result, floristic
differences are relatively small across the Holarctic,
which spans from the west coast of North America to
the east coast of Asia and includes the majority of the
world’s temperate forests.
Temperate forests occur across a wide range of local
physiographic landforms, from rocky slopes to rolling
plains and river floodplains, although generally under
non extreme physiographic conditions. Trees that occupy
slopes or well drained substrates with low organic matter
such as sandy outwash plains (e.g., pines and some oaks)
are adapted for drier (xeric) sites low in nutrients. Trees
adapted for moderate (mesic) sites are found on plains,
glacial moraines, or low hills with greater stocks of soil
organic matter. Nutrient and moisture demanding
broad leaved species, for example, maples and beeches,
thrive in mesic landscapes. Trees occupying river flood
plains, wetlands, or bogs have environments that can be
very moist to wet (wet mesic to hydric). These soils are
relatively rich in organic matter but trees in these land
scapes must be adapted to withstand flooding, including
long periods with wet, anoxic soils with low nutrient
availability.
Climatic and Physiographic Subdivisions
Given the great geographic extent of temperate forests it
is not surprising that regional differences are observed.
Systematic classifications of ecoregions and climates
describe subdivisions within the biome (Table 1). The
extensive temperate mixed deciduous forest occurs pri
marily in Bailey’s warm continental division (210), hot
continental division (220), and marine division (240);
these are Köppen–Trewartha classes Dcb, Dca, Do, and
Cf. The warm continental division has snowy cold win
ters, while the hot continental division has warmer, wetter
summers and milder winters. In the marine division (240)
winters are mild, summers relatively cool, and precipita
tion occurs most of the year.
The temperate mixed evergreen forests occur
primarily in Bailey’s temperate and rainy subtropical
division (230) which is most analogous to the Köppen–
Trewartha mid
and lower latitude Cf (humid
Table 1 Temperate forest biome types and corresponding geographic regions, Bailey ecoregions, and Köppen–Trewartha climate
classes
Temperate forest type
Geographic region
Bailey ecoregion
Köppen–Trewarthaa climate class
Temperate mixed-deciduous
forest
Eastern North America
Asia
Europe
South America
Australia/New Zealand
Humid temperate domain
(200)
Warm continental division
(210)
Hot continental division
(220)
Marine division (240)
Dcb: Temperate continental, cool
summer
Dca: Temperate continental, warm
summer
Do: Temperate oceanic
Cf: Humid subtropical
Temperate mixed-evergreen
forest
Southeast North
America
Asia
South America
Australia/New Zealand
Humid temperate domain
(200)
Subtropical division (230)
Cf: Humid subtropical
Temperate rainforest
Northwestern North
America
South America
Southeast Australia/New
Zealand
Humid temperate domain
(200)
Marine division (240)
Cf: Humid subtropical
Do: Temperate oceanic
a
Dc: Temperate continental: 4 7 months above 10 C, coldest month below 0 C; Cf: Humid subtropical: 8 months 10 C, coldest month below 18 C,
no dry season; Do: Temperate oceanic: 4 7 months above 10 C, coldest month above 0 C.
420
Temperate Forest
subtropical) class (Table 1). These climates have no dry
season, with even the driest months having at least 30 mm
of rain, and have hot summers with the average tempera
ture of warmest month greater than 22 C.
Temperate rainforest conditions largely occur where
ocean moisture is abundant and prevented from moving
inland by mountain ranges. These conditions occur in
particular continental placements within Bailey’s marine
division (240) and Köppen–Trewartha Do class in higher
latitudes and within Bailey’s subtropical division (230)
and Köppen–Trewartha Do and Cf classes in lower lati
tudes (Table 1).
Disturbance and Forest Structure
Major disturbances occur naturally in temperate forests,
although particular locations vary in the types, fre
quencies, and severities of disturbance. Major natural
disturbances include fires, windthrow during severe
storms, ice storms, flooding, disease, and irruptions of
defoliating or wood boring insects. The array of natural
disturbances that occur at a particular location consti
tutes its disturbance regime, a strong force in shaping
forest structure and composition. Smaller scale distur
bances also shape forests over long time periods in the
absence of a major disturbance. These include the
production of forest gaps from the mortality of one to
a few large trees. In some cases, idiosyncratic combina
tions of processes may produce repeated disturbance.
An example is ‘fir waves’ that occur only in Japan and
the northeastern US. In these waves of mortality that
pass through the forest repeatedly, a fungal pathogen
weakens the roots in mature trees while wind gusts
cause the weakened roots to break as they rub against
sharp gravel in the rocky soil. Because of the repetitive
nature of natural disturbances and the long lifetimes of
temperate forest trees, trees are often adapted (through
what is termed ‘vital attributes’) to withstand particular
disturbances or to regenerate following disturbance.
Some examples are trees that re sprout from stumps
following fire or from branches following windthrow,
cones that require fire to open, and seeds that germi
nate best on exposed soil.
Human activities have substantially altered the dis
turbance regimes in many temperate forests. The large
scale harvesting of trees for timber, whether cutting
selected sizes or species of trees or cutting all of the
trees in a stand, are relatively new forms of disturbance
that now affect forest structure and community compo
sition throughout much of the temperate biome.
Human activities also cause large scale chronic distur
bances, including polluted rainfall (e.g., acid rain) that
causes soil acidification and nitrogen enrichment over
large regions of the US, Western Europe, and
increasingly in eastern Asia. Still another category of
human induced disturbance is in the introduction of
invasive species. In the eastern US, the introduction
of a fungal pathogen in the early twentieth century
caused the chestnut blight, essentially eradicating one
of the dominant trees (the American chestnut, Castanea
dentata) from a large region.
Structural Layers of Vegetation
Disturbances in temperate forests vary not only in their
type and frequency but also in their intensity or severity,
the latter gauged by the percentage of vegetation mor
tality. A major disturbance that causes widespread or
near total mortality of trees in a forest stand, followed
by the development of a new (secondary) forest stand, is
known as a stand initiating event. Following such an
event, but mediated by the occurrence and severities of
subsequent disturbances, the vertical structure of a forest
stand tends to grow more complex over time. More
favorable site conditions such as organic rich, fertile
soils and ample moisture also promote structural com
plexity. With full development, the vertical structure
includes a canopy overstory, understory, a shrub layer,
and an herbaceous layer. In achieving such development
the forest passes through several stages. These include a
stand initiation stage in which seedlings and saplings
dominate and new species may continue to arrive; a
stem exclusion stage in which the canopy closes, shading
out shorter individuals; an understory re initiation stage
in which shade tolerant species grow as seedlings and
saplings; and finally an old growth or steady state stage.
In the old growth stage, the overstory typically includes
both canopy dominants and subdominants (the latter
with crowns only partially in sunlight) together with
understory and shrub layers made up of mature, shade
tolerant individuals. Old growth stands can be identified
through a few key characteristics, including a distribu
tion of age and size classes of trees, the absence of
saw cut stumps, and the presence of decaying logs the
size of overstory trees.
The understory in a structurally complex temperate
forest stand comprises trees and shrubs that spend their
entire life cycle there as well as young or suppressed
individuals of potential canopy dominant species.
Understory tolerant species are those that can survive
in, or even require, the shade of a forest canopy (e.g.,
sugar maple, Acer saccharum). In old growth stands or
those not recently disturbed it is common to see shade
tolerant species in both the understory and overstory
because the overstory trees are those that regenerated in
the shade of the canopy. Some temperate forests have a
dense layer of understory shrubs, for example Kalmia spp.,
Rhododendron spp., and Vaccinium spp. (blueberry). The
herbaceous layer of a temperate forest commonly
Temperate Forest 421
Soils and Woody Debris
Soils provide a physical rooting medium, the capacity to
store and release water, and the capacity to store and
release nutrients for growing trees. The soils of the tem
perate forest regions occur in five orders of the system of
soil taxonomy, namely Spodosols, Alfisols, Ultisols,
Entisols, and Inceptisols. They range from somewhat
infertile (Spodosols) to quite fertile (Alfisols). Spodosols
are characterized by a heavily leached surface mineral
horizon and a deeper accumulation of Al and Fe rich
organomineral complexes. Spodosols form under conifer
ous or mixed forests in relatively cool regions with
substantial hydrologic leaching, particularly at the north
ern borders of the biome in the Northern Hemisphere.
Further toward the subtropical in cooler areas of eastern
North America, Europe, and parts of Asia and Australia,
Alfisols form, characterized by organic rich mineral soil
horizons throughout the soil profile, moderate leaching
and high fertility. Ultisols, the oldest and most highly
weathered soils in temperate zones, are located in the
unglaciated and warmer portions of the biome, including
southern North America, Asia, Australia, and New
Zealand. Because of their advanced age and weathering,
these can be deep soils with relatively poor fertility.
Inceptisols and Entisols, the youngest soils characterized
by less weathering and poor horizon development, are
widely distributed in temperate forests. In particular,
these form in areas where glaciers left behind new parent
material either as till or outwash.
A characteristic that distinguishes temperate from tro
pical forest soils is the much larger stores of soil organic
matter typically present in temperate soils. In temperate
regions, litter in various stages of decomposition from
fresh litter to humified matter often accumulates atop
the mineral soil, forming the forest floor. This organic
layer is key in retaining water, retaining and releasing
nutrients, and providing animal habitat. It varies in
thickness from a few centimeter to tens of centimeters,
depending on the age of the stand, the soil pH, the
inherent decomposability of the species of litter, the
amount of rainfall, and the presence or absence of
earthworms.
An additional important category of organic detritus
found in many temperate forests is coarse woody debris.
This includes standing dead trees and downed, decom
posing logs. Rotting woody debris provides a rooting
medium, a habitat for soil fauna, a substrate for the
saprotrophic flow of energy to the food web, and a
means for returning nutrient elements to soils, as well
as important structural material for forest streams. Logs
undergo a wide range of decay rates, from relatively
rapid (a few years) where logs are small and wetting–
drying cycles are rapid, to very slow (lasting to a cen
tury) where logs are large and the environment is wet
and cool. In harvested or managed forests, coarse woody
debris may be absent because logs are removed for
timber. In unmanaged temperate forests, the long time
periods needed for large logs to be produced and decom
posed produces a U shaped curve in the mass of woody
debris over time (Figure 3). After a stand initiating
disturbance, woody debris from the previous stand accu
mulates rapidly and then decays slowly. A lag time of
several decades typically exists before woody debris
from the new stand begins to accumulate. If the new
stand remains even aged, a second peak may occur as
the new stand passes through the stem exclusion stage of
development and widespread mortality occurs in smaller
trees that compete unsuccessfully for light after the
canopy has closed.
Ecological Communities and Succession
Vegetation Communities
Temperate forest vegetation communities span the range
from single species stands to mixed species stands as well as
the range from even aged to all aged stands. Which type of
community is present at any point in space and time
depends on the site physiography, soil, and climate together
with its disturbance history. Species such as pines,
CWD mass
contains mosses, lichens, vines, and forbs. Many shrubs
and herbs are adapted to low light environments or grow
before canopy leaf extension in the spring or after overs
tory leaf abscission in fall; in summer only about 10% of
full sunlight reaches the herbaceous layer, but this figure
can rise to 70% in deciduous stands in winter. Shrubs and
herbs that require more light grow in well lighted gaps or
extend their crowns into openings. Vines grow into forest
canopies to access light and may be plentiful following a
disturbance that kills canopy trees but leaves the dead
trees standing.
Stand-initiating
disturbance
Time
Figure 3 Dynamics in the mass of coarse woody debris (CWD)
before, during, and after a major stand-initiating disturbance in a
temperate forest. The solid line represents a U-shaped curve in
CWD mass over time. The dashed line represents a secondary
peak that may occur if the newly initiated stand remains evenaged and undergoes a self-thinning stage.
422
Temperate Forest
Figure 4 The canopy of an even-aged red pine plantation,
aged about 75 years, in Massachusetts, USA. Photo by W. S.
Currie.
eucalyptus, cottonwoods (Populus spp.), and others may form
natural single species, even aged stands (Figure 4). Pioneer
species such as aspens (Populus spp.) and some pines may
initially form even aged monocultures which eventually
diversify in composition and vertical structure as growth,
self thinning, or succession proceeds. Long lived hardwoods
and other conifers also form stands where, increasingly with
forest age, great diversity exists in tree ages and sizes. An
example of the latter are hemlock northern hardwood for
ests of the Great Lakes region of the United States. If
horizontal structure and heterogeneity are taken into
account, small patch mosaics of even aged forests of varying
ages form larger landscapes of mixed aged stands, known at
the landscape scale as a shifting mosaic steady state.
It is easy to observe apparent associations of forest
trees that occur at certain scales, for example old growth
hemlock–sugar maple (Tsuga canadensis–Acer saccharum)
stands that form at the scale of square kilometers, or the
oak–hickory (Quercus–Carya) associations that form more
loosely in secondary forests over hundreds of thousands of
square kilometers. However, a longstanding debate con
cerns whether forest communities represent organized
associations or simply continuously varying associations
as tree species respond individualistically to environmen
tal gradients.
Temperate forest tree species form apparent associa
tions with one another and with the abiotic environment
not only across space but also over time at a particular
location. A key organizing principle in understanding such
temporal associations is the concept of succession, or the
replacement of one dominant species or set of dominant
species with another, over time, on a particular soil.
Primary succession refers to the replacement of species
over time occurring in the first forest stand to grow on a
newly exposed soil, for example, following the retreat of a
glacier. Secondary succession refers to species replace
ment over time following a major disturbance such as
massive windthrow, mortality, or forest harvest. Early
successional species, termed pioneers, are those that are
able to fix nitrogen from the atmosphere (see the section
titled Nutrient cycling) or those that grow rapidly under
high light conditions but cannot tolerate shade. Late
successional forest species are typically those that can
tolerate low light or low nutrient conditions as understory
trees, while continuing to grow over long time periods,
eventually reaching the overstory. Forest ecologists have
long sought general principles of succession – for example,
the identification of a deterministic sequence leading to a
particular stable endpoint or ‘climax’ vegetation commu
nity in a particular climate and physiographic landform.
Current understanding, however, emphasizes that while
certain successional mechanisms exist, the particular
sequences and possible endpoints of succession at a parti
cular location are typically numerous, ultimately depending
on a complex interplay among competition, species
arrival, regeneration, disturbance regimes, and species’
modification of the environment.
Temperate Forest Fauna
Faunal biodiversity in temperate forests is not as great as
that in tropical forests (see Tropical Rainforest), but is
greater than in boreal forests (see Boreal Forest). Because
temperate forests are highly seasonal in their climate and
cycles of vegetation physiology and production, faunal
life cycles, ecology, and populations are often tied
strongly to the seasons. Animal habitats within temperate
forests are numerous and heterogeneous, including soils,
the forest floor, woody debris, woody stems, and the
layers of vegetation canopies. Although some animals
depend on particular tree species, many are more depen
dent on certain aspects of forest structure.
In the temperate forest, the greatest concentration of
fauna is on and just below the forest floor, in the litter,
humus, and soil. Animals not only inhabit these strata, but
through their activities drive soil carbon and nutrient
cycling. Also within these strata are gradients of moisture,
temperature, gases, and organic matter. Soil microhabitats
are pore spaces, water film on soil particles, plant remains,
the rhizosphere, and tunnels and burrows. Together, soil
Temperate Forest 423
fauna and saprophytic flora contribute to the decomposi
tion of organic matter. While most decomposition and
nutrient release take place in the warm, humid summers,
coinciding with the growing season of the vegetation,
microorganisms and invertebrates can remain active
below the insulating winter snowpack. Some animals
occupy the litter in summer and move to the mineral
soil in winter.
Because of the moist soil conditions, many temperate
forest floors are home to reptiles (turtles and lizards) and
amphibians (toads, frogs, newts, and salamanders). In the
mixed deciduous temperate forests there are over 230 spe
cies of reptiles and amphibians. These animals live on the
forest floor close to streams, depressions, or lakes where
there is available moisture. Lizards are found in moist
woods and also in disturbed areas. Turtles live in or near
bodies of water and toads and frogs are widespread, needing
only shallow water. Temperate forest streams and rivers can
support abundant fish populations, particularly under less
disturbed conditions and in coastal temperate rainforests.
Mammal populations in temperate forests tend to be
comprised of scattered individuals or groups, and their
habitat ranges from the forest floor to the canopy layers.
Examples of small mammals are squirrels, rabbits, mice,
chipmunks, skunks, and bats. Very large mammals are the
exception and in temperate forests may include bear,
mountain lions, deer, and other ungulates such as moose
and elk. These mammals depend on the herb and shrub
layers of the forest in addition to the litter and woody
debris for food and habitat. Edge areas form transition
zone habitats; for example, deer and other large animals
usually live near the edges of forest openings with the
trees providing shelter while edible ground vegetation is
available in the openings all year.
Trunks are also habitats for spiders, beetles, and slugs.
Birds are especially versatile across habitat structures; they
are found on the forest floor and in several of the vegeta
tion layers depending on nesting and foraging preferences.
Types of birds that breed in mixed deciduous forests
include bark foragers (woodpeckers, flickers), canopy glea
ners and pursuers (chickadee, vireo, flycatchers), ground
species (thrushes, ovenbirds), and warblers. Deciduous for
est are also breeding habitat for larger avian species
including turkeys, vultures, owls, and hawks. In addition,
moths, butterflies, and other flying insects feed and repro
duce in the canopy, the understory, and the forest floor.
Water and Energy Flow, Nutrient Cycling,
and Carbon Balance
Water, Evapotranspiration, and Energy
Water enters temperate forests as rainfall, snowfall, fog,
and the direct condensation of water vapor onto plant or
soil surfaces. Some water, amounting to less than 10% of
rainfall under most conditions, is lost immediately to the
atmosphere through evaporation. Depending on the sea
son, water drips from the forest canopy to enter soils or
accumulates as snow until a mid winter thaw or spring
snowmelt. Entering the soil, water is stored, taken up by
plant roots, or moves to groundwater or surface water.
Water taken up by plants moves upward through the
xylem and exits as water vapor through leaf stomates in
the process of transpiration. Typically, through the com
bined processes of evaporation and transpiration, less than
half of the annual precipitation is passed directly back to
the atmosphere as water vapor. Somewhat more than half
of the annual precipitation passes through the rooting
zone of the soil to enter groundwater or surface water
such as streams and lakes.
Evapotranspiration, or evaporation and transpiration
taken together, makes a large contribution to the ecosys
tem energy budget and to the regulation of temperature.
In the conversion of liquid water to gas, evapotranspira
tion carries away large amounts of heat as latent heat.
This cooling effect combines with other terms in the
energy budget of a forest canopy to regulate the tempera
ture of leaves and of the forest as a whole. Other major
terms in the energy budget include the absorption or
reflection of short wave (sunlight) and long wave radia
tion (from sunlight and from the atmosphere), the
emission of long wave radiation, and the gain or loss of
sensible heat from the atmosphere. On a typical summer
day the vegetation canopy absorbs energy in short wave
radiation from the Sun and dissipates the energy as sen
sible and latent heat to the atmosphere, heating the
troposphere from below. On warm days with strong sun
light, the ability of forest tree canopies to dissipate heat
allows the trees to maintain leaf temperatures closer to
the photosynthetic optimum while also minimizing plant
respiration. The opening and closing of stomates, govern
ing transpiration, is under plant physiological control and
is an important aspect of plant adaptation to life in a
particular environment. During prolonged periods of
drought, when trees are less able to use water to cool
the canopy and maintain leaf turgor, foliar wilting and
tissue damage can occur. Some temperate forest trees can
be unexpectedly drought deciduous, dropping their foli
age during a late summer drought.
The photosynthetic conversion of light energy to
stored chemical energy is a minor term in the physical
energy budget of a forest, amounting to no more than 2%
of the energy in sunlight. At the same time, this energy
conversion represents the largest term in the ecological
energy budget of a forest. The energy stored in photo
synthate drives the life processes of all of the plants and
animals in the ecosystem. A large portion of this energy is
consumed by the vegetation itself through plant respira
tion, supplying energy for growth, metabolism, and
reproduction. Another large flow of energy enters the
424
Temperate Forest
food web through herbivory; herbivores eat seeds, fruits,
and living plant tissues. The consumption of living leaves
by insects, while normally minor, can grow during insect
irruptions to encompass virtually the entire forest canopy
over large areas. Similarly, the consumption of living
leaves by forest ungulates including deer and moose are
typically small energy fluxes at the ecosystem scale
(although the browsing of seedlings and saplings can
have a strong impact on forest regeneration and the future
composition of the vegetation community). The chief
means of energy flow to the faunal food web is through
the saprotrophic pathway. Fungi and bacteria (often
called soil flora) decompose dead and senesced plant
material including leaves, roots, and woody debris. The
soil flora is grazed upon by soil microfauna, which are in
turn preyed upon by other fauna including arthropods,
amphibians, and birds.
Nutrient Cycling and Carbon Balance
To achieve the high levels of productivity typical of
temperate forests, trees require ample and reliable sup
plies of nutrient elements. Those required in the largest
supplies include N, P, K, Ca, Mg, S, and Mn. Trees
acquire most of their nutrients through root uptake from
soils, which store nutrients in soil solution, on the surfaces
of mineral grains, on the surfaces of organic matter, and in
decomposing organic matter itself. A forest ecosystem
receives inputs of nutrients from the atmosphere and
from mineral weathering, experiences losses of nutrients
via leaching (the water driven movement of elements out
of the rooting zone, ultimately to streams), and cycles
nutrients internally (Figure 5). A key internal cycle is
the plant–soil cycle in which an element such as calcium
(Ca) is taken up by plant roots, used nutritionally by the
tree, returned to the soil in foliar litterfall, and returned to
the pool of soil available nutrients during decomposition
of the litter. Temperate forests are characterized by the
fact that, for most nutrient elements required for plant
growth, the internal cycling is greater than the ecosystem
inputs and losses of these elements.
While most of the required nutrient elements can be
released through mineral weathering, a notable exception
is nitrogen (N). Temperate forests rely on inputs of N
from the atmosphere. Combined with the fact that trees
have a high demand for N, and with the fact that N is
strongly retained in unavailable forms in soil organic
matter, this makes N the most limiting nutrient for plant
growth in most temperate forests. Trees have a high
demand for N because photosynthesis and plant metabo
lism require enzymes, which are made of N rich amino
acids. Amino acids are also one of the primary needs of
Atmospheric
deposition
Gaseous
loss
Fixation
Translocation
Aboveground
litter
production
Belowground
litter
production
Nutrient uptake
and assimilation
Nutrients
in
soil solution
Mineral
weathering
Organic matter
forest floor and
mineral soil
Decomposition
and
mineralization
Leaching
Figure 5 Schematic diagram of generalized nutrient cycling in temperate forests. Shaded terms represent nutrient cycling fluxes
within the system, while unshaded terms represent ecosystem inputs or losses. From Barnes BV, Zak DR, Denton SR, and Spurr SH
(1998) Forest Ecology, 4th edn. New York, USA: Wiley.
Temperate Forest 425
herbivores that consume plant tissues, including defoliat
ing insects, deer that browse saplings, and beavers that
girdle trees by eating the cambium around the tree base.
Given the high demand for N by forest trees, it is some
what ironic that trees in temperate forests are surrounded
by two large, potential sources of N that are limited in
availability because of the chemical form of the N. The
first is N2 gas, which is the primary constituent of the
atmosphere. Most forest trees cannot access gaseous N2,
although a few exceptions include red alder (Alnus rubra)
and black locust (Robinia pseudoacacia) in the USA, which
access atmospheric N2 through the process of N fixation
(Figure 5). In this process, symbiotic bacteria living
in root nodules fix the N2 into plant available forms.
The second large pool of poorly available N occurs in
humus and soil organic matter, made up of partially
decayed and humified plant and microbial detritus.
Typically, large accumulations of N are bound in this
material in large, polyfunctional macromolecules that
form during litter decomposition. Temperate forests
are characterized by the combined facts that (1) cold,
wet winters impede microbial decomposition and allow
these pools of organic matter to accumulate, and
(2) warm, humid summers promote decomposition by
fungi, causing these soil organic matter pools to turn
over and release nutrients at slow but continuing rates.
Nutrient release during decomposition is termed miner
alization because N is converted from organic to the
inorganic forms of nitrate (NO3) and ammonium (NH4)
which are easily taken up and used by plants (Figure 5).
Carbon is the primary elemental constituent of both
forest vegetation and the organic matter in forest soils.
Carbon (C) is not considered a nutrient element per se
because a C atom passes through a forest once, in a single
direction, closely linked to the flow of energy; unlike
nutrients, carbon does not cycle between plants and soils
repeatedly. Forests are highly open systems with respect
to carbon, exchanging large quantities of CO2 with the
atmosphere. The carbon balance of a temperate forest
arises from the interplay among processes controlling
forest sources and sinks of atmospheric CO2.
Photosynthesis, or primary production, converts atmo
spheric CO2 to reduced organic compounds, storing
energy and C in the forest. Autotrophic respiration, the
conversion of organic compounds to CO2 by plants, pro
vides energy for plant metabolism. Heterotrophic
respiration, the conversion of organics to CO2 by herbi
vores, microorganisms, soil fauna, and other animals in
the food web, releases energy for animal life processes.
Fire, the rapid oxidation of organics, also releases CO2 to
the atmosphere. Depending on the balance among these
processes, temperate forests can either store or release
large quantities of carbon. The primary storage pools
include growing trees (particularly the woody stems),
the forest floor, standing and downed woody debris
Figure 6 An old-growth sugar maple–birch–hemlock forest
showing a large piece of downed woody debris. This forest is
located in northern Michigan, USA. Photo by W. S. Currie.
(Figure 6), and soil organic matter. The transfer of carbon
among these pools is linked to forest disturbance and
stand dynamics including aggradation and succession.
The flows of carbon into and out of the ecosystem are
closely coupled to the availability of water, flows of
energy, and the cycling of nutrients.
Temperate Forest Land Cover
Historical Land Cover and Land-Cover Change
Temperate forests in all regions of the globe have been
significantly altered by human activities for thousands of
years. Their moderate climates, fertile soils, and vegeta
tion productivity have been favorable to human
settlement and clearing for agriculture, as well as direct
use of trees themselves for lumber and fuels. Agricultural
and settlement activities have included development of
urban areas, widespread grain and other crop (e.g., corn,
vegetables) cultivation, livestock grazing, gathering of
mulch, and alteration of natural water drainage. Under
these historical pressures, it is estimated that only 1–2%
of the original temperate forest remains as never
harvested remnants scattered around the globe. The vast
majority of temperate forest land cover is in secondary
forest responding to human harvest or other human
induced disturbance.
The longest histories of substantial forest clearing have
been in Asia and Europe. In China clearing for agriculture
probably began some 5000 years ago, where the Chinese
civilization is believed to have begun around the Huang
He (Yellow River). The primary sociopolitical factor
contributing to deforestation of China over the centuries
has probably been the focus on an agriculture based
economy. At present, there is negligible large scale
426
Temperate Forest
reforestation in temperate China and significant soil ero
sion problems hampering reforestation.
Forest clearing for agriculture in Europe began over
5000 years ago starting in present day Turkey and
Greece and moving northwest through Middle Europe
to Northern Europe. Forests of Britain were substantially
cleared for agriculture and grazing. Woodlands regained
some area in the Middle Ages; however, even remaining
European temperate forests were degraded, being used
for fuelwood, woodland pasture, and later for charcoal.
Coppice practices promoted species that re sprouted
more quickly than beech – including maples and oaks,
and this activity altered the natural floristic composition.
Tall trees in Britain and Western Europe were removed
for shipbuilding. Manorial estates provided some of the
few refuges for natural forests. Reforestation in recent
centuries in Europe began subsequent to reduction in
the use of woodlands for pasture and fuel; reforestation
has also occurred through the introduction of planted
managed forests and scientific forestry. However,
spruce, pine, and larch have been widely planted on
areas previously occupied by once deciduous temperate
forests.
North American indigenous populations cleared or
burned small areas for some agriculture, but land
cover change in North American temperate forests
began at large scales in the late sixteenth century with
the European settlement. Eastern North American was
rapidly cleared as the population moved westward in the
nineteenth century. By the start of the twentieth century
only a small amount of the original North American
temperate forest remained. When the richer soils of the
topographically level Midwest and Great Plains were
found to be more productive for agriculture than those
of eastern North America, eastern farms were abandoned
and natural forests began to re grow. At present, second
ary forests are regrowing in the eastern and central
United States.
In the Near East the temperate forest occurs in a
narrow belt including in Turkey and Iran. This area
probably served as a plant refugium during the Ice
Ages and the floristic composition is more diverse
than that in Europe. Some forests have been exploited
for coppice, timber, or grazing and others transformed
into agriculture and fruit tree plantations. Beech forests
are the most significant of the present day broad leaved
forests in the region. In the small area of temperate
deciduous forest in South America, forests have been
moderately altered since the arrival of the Spaniards in
the sixteenth century; the further south one goes the
more recently the vegetation has been undisturbed and
wooded areas remain. Australia first saw introduction of
European agricultural practices only approximately 150
years ago.
Present-Day Land Cover and Rates of Change
The global temperate forest continues to be changed by a
combination of long term effects of historical land cover
change and by present day change agents. Present day
drivers of land cover change in temperate forests include
accelerated population growth, continued industrializa
tion, and changes in agricultural practices. These are
expressed on the landscape as continued clearing for
settlement and agriculture in some regions, abandonment
of agriculture and reforestation in other regions, and
widespread alteration in landscape spatial structure and
biodiversity.
While rates of tropical deforestation increased
between 50% and 90% in the 1980s, the area of temperate
forests has remained constant or increased in the last 50
years in the form of new second growth forests. In some
areas, in eastern North America and parts of Northern
Europe, farming is less economically viable than in other
parts of the temperate region, leading to reforestation in
these areas. Preservation in the form of parks has
expanded by active conservation efforts worldwide.
Managed forestry has maintained existing temperate for
est lands by re planting after harvest, and sustainable
forestry practices are receiving increasing attention.
While the temperate forests may have stabilized or
increased in terms of total area, most regions continue to
experience other alterations manifested in the landscape
spatial patterns and forest biodiversity. Today the tempe
rate forest biome is a mosaic of settlements, patches of
forest, and agriculture. Large expanses of unbroken for
ests from past centuries have been replaced by
considerable landscape scale heterogeneity and fragmen
tation. Temperate forest communities have changed
compositionally, as disturbance regimes have shifted
from natural to a combination of natural and human
caused, producing different patterns of regeneration and
succession. While some recently established nature
preserves have a natural forest structure, reduced biodi
versity characterizes many temperate managed and
secondary forests. Considerable present day challenges
lie in understanding and addressing the impacts of land
use change and other aspects of global environmental
change in the temperate biome on forest biodiversity
and forest ecology.
See also: Boreal Forest; Chaparral; Tropical Rainforest.
Further Reading
Bailey RG (1998) Ecoregions: The Ecosystem Geography of the Oceans
and Continents. New York: Springer.
Barbour MG and Billings WD (2000) North American Terrestrial
Vegetation. Cambridge, UK: Cambridge University Press.
Barnes BV, Zak DR, Denton SR, and Spurr SH (1998) Forest Ecology.
New York: Wiley.
Temporary Waters 427
Currie WS, Yanai RD, Piatek KB, Prescott CE, and Goodale CL (2003)
Processes affecting carbon storage in the forest floor and in
downed woody debris. In: Kimble JM, Heath LS, Birdsey RA, and
Lal R (eds.) The Potential for U.S. Forests to Sequester Carbon and
Mitigate the Greenhouse Effect, pp. 135 157. Boca Raton, FL:
Lewis Publishers.
Frelich LE (2002) Forest Dynamics and Disturbance Regimes. Studies
from Temperate Evergreen Deciduous Forests. Cambridge Studies
in Ecology. Cambridge, UK: Cambridge University Press.
Lajtha K (2000) Ecosystem nutrient balance and dynamics. In: Sala O,
Jackson RB, Mooney H, and Howarth RW (eds.) Methods in
Ecosystem Science, pp. 249 264. New York: Springer.
Olson DM, Dinerstein E, Wikramanayake ED, et al. (2001) Terrestrial
ecoregions of the world: A new map of life on earth. BioScience
51: 933 938.
Rohrig E and Ulrich B (1991) Temperate Deciduous Forests.
Amsterdam: Elsevier.
Temporary Waters
E A Colburn, Harvard University, Petersham, MA, USA
ª 2008 Elsevier B.V. All rights reserved.
Overview
Introducing Temporary Waters
The Ecology of Temporary Waters
Ecosystem Ecology
Applied Ecology
Further Reading
Overview
What Is Covered in This Article?
What Are Temporary Waters, and Why Are
They of Interest Ecologically?
This article is divided into two sections. The first intro
duces temporary waters – definitions, important variables,
types, geographic distributions, and terminology. The
second section examines the ecology of temporary
waters, with an overview of the biota and their adapta
tions, and summaries of some key questions in organismal
and community ecology, ecosystem ecology, and applied
ecology.
Temporary waters are shallow lakes, ponds, pools, rivers,
streams, seeps, wetlands, depressions, and microhabitats that
contain water for a limited period of time and are otherwise
dry. They occur across the globe, on all continents and oceanic
islands, at all latitudes, and in all biomes, wherever water can
collect long enough for allow aquatic life to develop.
Numerous and widespread, many temporary waters
are small and easily studied. Their communities are diverse,
with much among site variation (i.e., high diversity), and
differ from those in permanent waters, contributing to
regional ( ) biodiversity. Endemic species are often present.
Organisms survive through species specific behavioral,
physiological, and life history adaptations. Community
composition and structure change in response to environ
mental variations. Temporary waters are highly productive
and their food webs are relatively simple. For all of these
reasons, temporary waters lend themselves to surveys and
experimental manipulations designed to test hypotheses
about biological adaptation, population regulation, evolu
tionary processes, community composition and structure,
and ecosystem functioning.
In many parts of the world, most temporary waters
have been lost. The conservation and restoration of
vulnerable temporary waters is a major thrust of applied
ecology. Also important are applications of ecological
understanding to the control of disease vectors, especially
pathogen transmitting mosquitoes, from temporary water
habitats.
Introducing Temporary Waters
Definition
In temporary waters, aquatic habitat is present for non
continuous lengths of time, in contrast to permanent
water bodies, which are always flooded except under
unusual conditions such as extreme droughts. This dis
continuity in the availability of water is the defining
characteristic of temporary waters.
For this article, temporary waters include temporary
inland salt waters, whose chemistry and biota are allied
to fresh waters and not to marine ones, but they do not
include coastal areas flooded by ocean tides. Also excluded
from this discussion are subterranean waters.
Important Variables
Apart from periodic drying, there are no hard and fast
rules about the characteristics of temporary waters.
Classification may be useful, provided it contributes to
understanding. The important considerations governing
428
Temporary Waters
how to classify temporary waters in a given situation
should be: what is the purpose of classification, and what
are the desired outcomes in terms of distinguishing
different types of temporary water bodies? Researchers
have developed many approaches to classifying tempor
ary waters using the descriptive variables listed below.
Water body type
Temporary waters may be lotic (flowing) or lentic (still).
There are several major categories, and many unique
regional names (Tables 1–3). Some categories overlap;
for example, pools formed after thunderstorms on exposed
rocks on coastal Scandinavian islands are both rainpools
and rockpools.
Geography
Regional location (e.g., Ontario, Malay Archipelago), lati
tude (e.g., tropical, Arctic), or climate (e.g., humid, arid)
may contribute to similarities among temporary waters.
Biome
Temporary waters occur in all terrestrial biomes, even
the wettest. Regardless of their location globally, habi
tats within a particular biome, with similar hydrologic
characteristics on similar substrates, often are much
alike.
Substrate
Substrate (e.g., rock, organic debris, sand, clay, limestone,
mud, basalt, wood) influences hydrology, water chemistry,
and temperature and is an important habitat variable in
its own right (e.g., for seed germination or shelter for
burrowing animals).
Size
Some classifications distinguish microhabitats, mesohabi
tats, and macrohabitats.
Table 1 Major types of temporary waters found throughout the world
Rockpools or rock pools – Accumulations of rainwater or floodwater in depressions on exposed bedrock or boulders
Rainpools or rain pools – Accumulations of rainwater on any substrate
Seasonal woodland pools – Fill annually, usually as a result of winter or spring rains, and from melting snow in northern areas, and dry
later in the year
Grassland pools – Temporary ponds in grassland environments
Marsh pools – Temporary ponds that occur within larger grass, sedge, or rush-dominated wetlands and remain flooded after most of the
wetland has drawn down
Swamp pools – Depressions within larger wooded wetlands that remain flooded after the surface of the swamp has dried
Floodplains – Land areas that are inundated seasonally by high waters spilling over the banks of rivers and streams
Floodplain pools – Low areas in floodplains that remain flooded after floodwaters have withdrawn and left most of the floodplain dry
Springs, seeps, and spring seeps – Sources of water derived from groundwater or from subsurface flow reaching the land surface
after heavy rains. Springs are expressions of the groundwater table and tend to be relatively permanent; seeps may be more
transitory. Both vary in output with rainfall over the source area, and both may provide seasonal or continuous sources of water. Flow
from springs and seeps may extend from the source as marshes, pools, or streams that may contain water during cool or wet seasons
and become dry during periods of high temperature and/or low precipitation
Intermittent headwater streams – The smallest tributaries at the head of stream systems, often seasonal in their flow, containing water
during the wet and/or cooler months and becoming dry during the hot/dry months
Arid-land rivers, intermittent rivers, or ephemeral rivers – Flowing waters that occur in regions where the groundwater table is far below
the surface and where annual potential evapotranspiration is greater than precipitation. They typically flow only during the rainy
season, when runoff travels over the land and is carried downstream; some only carry storm runoff, but others may have extended
flow maintained by seasonal groundwater discharges. During the dry season, there may be water below the surface and in isolated
pools within the channel, and there may be brief spates of flow following cloudbursts
Dry lakes or playas – Shallow water bodies in arid regions, especially in closed basins, where water collects from large areas. Due to the
arid conditions, the water usually evaporates rapidly. A long history of flooding and drying leads to accumulations of salts in these
basins, and dry lakes are typicaly saline. Many dry lakes occupy basins that contained large freshwater lakes earlier in geological
history. Deposits of salts and sediments left behind when the lakes dried may be tens or hundreds of meters deep beneath the lake
beds, and they contribute to saline conditions in the playa
Sinkholes or sink holes – Depressions created in calcareous bedrock by the gradual dissolution of the rock by water. They range in
diameter and depth from meters to kilometers. Sinkholes that contain water are fed by groundwater, precipitation, and/or streamflow
and include both permanent and temporary waters
Snowmelt pools, icemelt pools, and meltwater pools – Formed by the seasonal melting of ice and snow in the Arctic and Antarctic, along
the margins of icefields and mountain glaciers, and in areas that receive snowfall
Meltwater streams – Flowing waters that develop seasonally as glaciers, icefields, and winter snows melt; they often flow during the day
and stop flowing at night as low temperatures inhibit melting
Plant-associated microhabitats or natural containers (phytotelmata) – Microhabitats formed where plants produce small depressions in
which water can collect (see Table 2)
Artificial containers – Any human-made concavity where water can collect, including gutters, birdbaths, tires, empty cans, tractor ruts,
canoes, split and discarded coconuts, and other water-holding depressions
Temporary Waters 429
Table 2 Examples of phytotelmata and other natural
containers that provide temporary aquatic habitats for mosquito
larvae and other organisms
Ant nests
Insect-bored bamboo, bamboo stumps
Fungal cap concavities
Log holes
Buttress-root slits
Eggshells
Flower bracts
Fruits
Horns
Leaf axils
Fallen leaves
Nuts
Modified leaves of pitcher plants and analogs
Pods
Reeds
Rockholes, potholes
Mollusk shells
Skulls and other skeletal remains
Stumps and trunk cavities
Treeholes
Derived from Index in Laird M (1988) The Biology of Larval Mosquito
Habitats. Boston: Academic Press.
Hydrology
Hydrologic variables are the most important factors influ
encing aquatic life in temporary waters.
Water sources
Water sources include groundwater, runoff, precipitation,
snowmelt, streamflow, and floodwater.
Flood timing
Flood timing encompasses both season and predictabil
ity. Vernal, estival, autumnal, and hibernal (or brumal)
refer respectively to filling in spring, summer, fall, or
winter. Intermittent systems flood predictably at annual
(seasonal) to multiyear intervals. Waters that flood
unpredictably are ephemeral if they fill several times
a year, and episodic if they fill just once or twice a
decade.
Seasonality and predictability of flooding influence the
biota. Predictable filling of Mediterranean vernal pools by
rainfall during the winter growing season facilitates plant
growth and has contributed to the development of an
Table 3 Some terms used to describe temporary waters around the world
Avens
Baias
Billabongs
Bogs
Buffalo wallows
California vernal
pools
Carolina bays
Corixos
Dambos
Dismals
Doline
Fens
Gator holes
Gnammas
Heaths
Mires
Kettles, kettle
holes
Moors
Mosses
Muskegs
Oshanas
Pakahi
France: depressions hollowed out in limestone
South America: temporary lakes
Australia: pools that are left behind in floodplains as large, seasonal rivers recede after flooding
Worldwide: freshwater peatlands with acidic water chemistry; usually with limited connection to other
surface waters, often fed exclusively by rainfall
North America: created by buffalo (Bison bison) rubbing their bodies on the ground, these shallow
excavations on the prairies fill seasonally with water
Western North America: seasonally flooded pools in Mediterranean scrub of western North America,
especially California, and characterized especially by rich plant communities with large numbers of
endemic species
North America: round or oval depressions of uncertain origin in the coastal plain of the Southeastern
United States, often supporting endemic plant communities and temporary-pond fauna
South America: temporary-water bodies in floodplains, especially in the Pantanal region
Southern Africa: shallow, treeless, seasonally inundated wetlands at heads of drainage networks
North America: swamps or marshes in the Mid-Atlantic region of Virginia, Delaware, and the Carolinas
Western Balkan states/Dinaric Alps: depressions and sinkholes in limestone
Worldwide: freshwater peatlands with alkaline water chemistry
North America: excavations made by alligators (Alligator mississippiensis) in the Florida Everglades;
they
remain flooded when waters recede and serve as refugia for aquatic animals during droughts
Western Australia: temporary waters formed on granitic outcrops
Great Britain: freshwater peatlands with acidic water chemistry
Northern Europe: freshwater peatlands with acidic water chemistry
Worldwide, in areas affected by continental glaciation in the past: largely circular depressions formed
by
the melting of blocks of ice calved off of retreating continental glaciers and buried in morainal debris
Great Britain: freshwater peatlands with acidic water chemistry located on hilltops
Scotland: raised bogs, i.e., freshwater peatlands with acidic water chemistry located on hilltops or
above
the groundwater table
North America: freshwater peatlands with acidic water chemistry (Algonquin)
Namibia and Angola: linearly linked shallow pans that are filled by floodwater and precipitation
New Zealand: shallow, groundwater-flooded areas with acid soils, inappropriate for cultivation (Maori)
(Continued )
430
Temporary Waters
Table 3 (Continued)
Pans, panes, pannes
Phytotelmata
Plunge pools
Pocosins
Potholes, pot
holes
Prairie potholes
Ramblas
Sabkhas, seabkhas
Salinas
Sinkholes
Sinking creeks
Sloughs
Swallow holes
Takyrs
Tenajas
Turloughs
Vasante
Vernal pools
Vleis
Whale wallows
Worldwide: shallow temporary waters that flood periodically from rainfall in arid regions; also refers to
temporary pools that form in salt marshes from monthly flooding by spring tides
Worldwide: a technical term describing temporary waters associated with plants, in axils of leaves or
branches, modified pitchers and similar structures, nuts
Worldwide: deep holes that form in bedrock at the base of waterfalls through the action of water over
time, and that retain water for a period of time after the stream has dried up
North America: upland-coastal floodplains or groundwater-flooded seasonal wetlands in the South
Atlantic United States
Worldwide: rockpools in or along streambanks and streambeds, created by the action of water and
rock
scouring out round depressions into boulders or bedrock. Potholes may be a few centimeters to
more
than a meter in diameter
North America: in the Great Plains, largely circular depressions formed as blocks of ice left by departing
continental glaciers were covered by morainal debris and then melted
Spain: temporary streams that usually flow only after rainstorms
Arabian Gulf: saline lakes
South America: saline lakes
Worldwide: depressions in limestone, formed by the solution of surface rock or by the collapse of
underground caverns or caves collapse where the subsurface has been dissolved by gradual
solution
in water. Sinkholes may be dry on the bottom, intermittently flooded, or contain water continuously
North America: flowing streams that disappear from the surface into one of the many cracks or
sinkholes
in limestone regions, or into the ground in arid areas
Worldwide: the term has a variety of meanings. In Great Britain it refers to muddy and shallow waters.
In North America it is used to refer to prairie potholes, temporary ponds, oxbow wetlands, permanent
ponds, deepwater areas in the Everglades, brackish marshes on the west coast, seasonally flowing
depressions in forests, freshwater wetlands in the Great Plains. Some use the term to refer to areas
where water is not stagnant but rather, flowing slowly; others specifically define sloughs as areas
with
stagnant water
Great Britain: sinkholes in limestone, especially deep holes through which water funnels underground
Turkmenistan: pans in the desert
North America: rockpools, usually in temporary stream channels, that remain flooded for several
months
after the stream dries; some develop plant communities similar to those in vernal pools
Ireland: temporary waters formed in limestone, filled primarily by groundwater although may
sometimes
fill from precipitation; usually fill in fall and dry in spring or early summer
South America: temporary streambeds connecting lakes in the Pantanal region during the rainy season
North America: temporary woodland pools that fill in spring and dry in summer; applied more broadly to
all seasonal woodland pools that reach maximum depth and volume in spring. Worldwide, the term
applies to any temporary pools that fill in spring. The term ‘California vernal pools’ is used to
represent
a class of Mediterranean biome temporary ponds characterized primarily by their endemic plant
communities
Southern Africa: seasonally inundated wetlands in southern Africa, typically flooded by rivers at
high water
Eastern North America: seasonal woodland ponds along the Delaware coast in the United States
endemic flora. When the pools dry, high summer tem
peratures prevent the establishment of terrestrial
vegetation.
Flood duration, or hydroperiod
Across most categories of temporary waters, there is a
continuum of flood duration: days, weeks, months, or
years. Ephemeral waters are flooded for hours, days, or
weeks. Intermittent refers to flood durations of several
months. Semipermanent or near permanent waters dry
only occasionally, during major droughts. Within a water
body, the hydroperiod varies across filling cycles, depend
ing on weather, with some waters being more stable than
others (Figure 1).
Typically, with increasing hydroperiod, the poten
tial aquatic community becomes richer, and the
adaptations of the flora and fauna become less
extreme. Waters with shorter hydroperiods have
fewer total species but more that are unique to tem
porary habitats.
Temporary Waters 431
Seasonal changes in maximum water depth in
ten Cape Cod, MA, vernal pools
in a wet year, 1997
Maximum depth (cm)
140
120
E2
All pools flooded through 1997 and 1998
E3
100
E4
80
E6
60
E7
40
E8
20
E9
0
E10
1/6/1997
4/3/1997
7/1/1997
Date
10/1/1997
E11
Seasonal changes in maximum water depth in
ten Cape Cod, MA, vernal pools
in a drought year, 1999
Maximum depth (cm)
E1
50
45
40
35
30
25
20
15
10
5
0
1/27/1999
E1
E2
E3
E4
E6
E7
E8
E9
E10
4/28/1999
8/2/1999
Date
10/27/1999
E11
Figure 1 Water depths differ within and between years in ten temporary ponds clustered together on Cape Cod, MA, USA.
Chemistry
Important chemical characteristics include salinity (fresh,
<3 g l 1 salts; brackish, 3–35 g l 1; saline, 35 g l 1),
major ions (e.g., sulfate vs. chloride dominated desert
waters), color (e.g., clear vs. stained dark with organic
acids), pH, and dissolved oxygen.
Distribution of Temporary Waters
Most types of temporary waters occur widely across the
world’s biomes, from the poles to the equator. Their
numbers and varieties vary with annual precipitation,
temperatures, and local geology and geography. They
are most common in arid or cold areas where liquid
water is unable to persist for long periods of time.
Figure 2 Bromeliads and other plants serve as natural
containers for rainwater and provide microhabitats for
microorganisms, mosquito larvae, and some tropical amphibians.
Tropical rainforests
Tropical rainforests, although well watered, contain
many temporary waters. Cavities in bromeliads and
other epiphytes retain rainwater (Figure 2) where
decaying organic materials support microorganisms,
insects, and amphibians. Rainpools on the forest floor
fill, dry within days, and support distinct communities.
Lowlands of great tropical river systems, including the
Amazon and the Paraná–Paraguay, are inundated during
432
Temporary Waters
the rainy season, and the retreating floodwaters create a
mosaic of ponds that retain water for varying time
periods.
Boreal and temperate forests
Temporary waters in deciduous and coniferous forests
include rainpools, rockpools, and treeholes. Intermittent
headwater streams dry in summer when forest trees are
transpiring (Figure 3). Floodplain pools fill in spring or
after major storms. Seasonal woodland pools, commonly
called vernal pools, fill from groundwater, snowmelt, and
spring rainfall and dry in summer (Figure 4). Many are
important breeding habitats for amphibians, crustaceans,
and aquatic insects. Carolina Bays in the Southeastern
United States, and other previously unglaciated systems,
support endemic plants.
Figure 4 Temporary woodland ponds, or vernal pools, are
common in temperate and boreal forests. Water levels vary
significantly over time. In this pool, normal high water reaches the
base of encircling maples, and in wet years it is more than a
meter deep.
Tundra and icefields
Where temperatures are cold and the growing season is
short, temporary waters appear when the summer sun
melts glaciers, ice, and snow. For a few months, Antarctic
rockpools, high mountain ponds, and myriad shallow
water bodies perched over permafrost in Arctic tundra
teem with bacteria, protozoans, planktonic crustaceans,
and insect larvae. This broth of aquatic life provides food
for nesting birds which flock in hundreds of thousands to
high latitudes to raise their young.
Mediterranean scrub
The Mediterranean scrub biome occurs along the
Mediterranean Sea; from Baja California to eastern
Washington; and in parts of Chile, southern Africa, and
Australia. Rains during the winter–spring growing season
collect above impervious substrates, forming water bodies
known as vernal pools, vleis, pans, Mediterranean tem
porary pools, and gnammas. They support endemic floras,
including Isoe¨tes spp.; endemic faunas, including fairy
shrimp and other crustaceans; and cosmopolitan tempor
ary pool plants and animals. On other substrates,
temporary pools are less predictable and lack endemics.
Most rivers in this biome flow only during the wet season,
although isolated pools retain water for part of the dry
season. Treeholes and other natural containers provide
microhabitats after rains.
Deserts
Figure 3 Intermittent headwater streams drain up to 80% of
the landscape in temperate forests and support distinctive
communities of aquatic invertebrates and stream salamanders.
In deserts, extreme aridity, high temperatures, salinity, and
isolation of waters are especially stressful for aquatic life.
Brief rainstorms create ephemeral pools on rocks and other
surfaces. Extended rains collect water from large areas to fill
closed basins, forming shallow, usually saline lakes that
leave extensive deposits of encrusting salts upon drying
(Figure 5). Many rivers and streams flow seasonally, espe
cially in wet winters, or flash flood unpredictably after
storms, leaving behind pools of varying permanence
(Figure 6). Permanent springs overflow during winter,
creating seasonal streams, marshes, and thickets. Salts accu
mulate in the soil along the edges of desert waters, and
temporary water bodies are generally brackish or saline.
Temporary Waters 433
lower Mekong river plains in Asia; the southern African
veldt; and the Pampas, Campos, and Pantanal regions of
South America.
The Ecology of Temporary Waters
The Biota
Figure 5 A salt crust left by evaporating water overlies dry
lakes, or playas, in many desert basins.
Grasslands
Temporary waters in grasslands include pools, marshes,
floodplains, and seasonal rivers and streams. Rich assem
blages of plants, invertebrates, and amphibians occur in
these waters and are critical for bird populations in the
prairie pothole region of North America (Figure 7); the
Eurasian steppes; the Indus, Ganges, Assam, Sylhet, and
(a)
All major groups of freshwater organisms occur in tem
porary waters. Many families and genera are found in
similar habitats throughout the world, and there are
some cosmopolitan species.
Hundreds of species of prokaryotes, including photo
bacteria and bacterial decomposers, proctotists, including
green algae and diatoms, and protozoans, including
ciliates, flagellates, and sarcodines, have been identified
from temporary waters. Plants include mosses and liver
worts, ferns, grasses, sedges, rushes, spike rushes, and
other taxa typical of local wetlands. Microinvertebrates
include rotifers, tardigrades, and gastrotrichs. Arthropods,
especially water mites, crustaceans, and insects, dominate
the macroinvertebrates. Many microcrustacean species,
especially ostracodes and copepods, swim in the water,
feed in the sediments, or cling to surfaces. Branchiopod
(b)
Figure 6 (a) Seasonal rivers in arid regions flow seasonally. (b) When flow ceases, the pools that remain persist for varying lengths
of time.
434
Temporary Waters
Water mites selectively feed on insects and crustaceans.
Platyhelminthes (flatworms and flukes), Annelida (seg
mented worms and leeches), Nematoda (roundworms),
Nematomorpha (gordian worms), and Mollusca (snails
and bivalves) are also represented. Annual tropical killi
fishes (Cyprinodontiformes) and African and South
American lungfishes (Lepidosireniformes) survive peri
odic drying, and fully aquatic fishes move seasonally into
floodplains and intermittent headwater streams to feed
and breed. Most anuran amphibian species, including true
frogs, treefrogs, and toads, and some salamanders, prefer
entially breed in temporary habitats. For many of the
world’s birds, temporary waters are critical food sources
during breeding or migration. Reptiles and mammals feed
and hydrate in these seasonal waters.
Figure 7 Prairie potholes dot the landscape of the upper Great
Plains in North America, provide habitat for aquatic life, and support
breeding waterfowl. Reproduced from Sloan CE (1972) GroundWater Hydrology of Prairie Potholes in North Dakota. USGS
Professional Paper 585-C. Reston, VA: US Geological Survey.
Autecology: Organisms and Populations
Temporary waters lend themselves to thousands of
ecological questions about adaptations, population regu
lation, and evolutionary pathways linking sibling species
and cosmopolitan taxa.
Adaptations to drying
crustaceans, particularly Notostraca (tadpole shrimps)
(Figure 8), Anostraca (fairy shrimps), Conchostraca
(clam shrimps), and some Anomola (daphnias and other
water fleas), are largely restricted to temporary waters. All
aquatic insect orders include temporary water species,
with the largest number in the Diptera, or true flies.
Inhabitants of temporary waters are distinguished by
their ability to survive periodic drying. Adaptations
include diapause, quiescence, and active avoidance.
Rapid responses to flooding, fast growth, and flexibility
in initiating the drying response maximize organisms’
habitat use.
Diapause
Figure 8 Notostracan crustaceans, known commonly as
tadpole shrimp (left of center), are temporary-water specialists.
Diapausing eggs lie for months, years, or decades in sediments
of desert playas, rockpools, and woodland ponds. Hatching
upon flooding, the animals are voracious predators and
scavengers and their presence restricts the distributions of other
temporary pool animals.
Diapause involves suspended development. Hormonally
controlled, and initiated and terminated by specific
environmental cues, diapause is the most common and
most effective drought survival mechanism. It can
allow survival over years – even decades – of continuous
drying.
The rapid appearance of living organisms when water
fills formerly dry puddles, containers, and floodplains is
not, as formerly believed, spontaneous generation, or life
miraculously developed from nothing. Instead, much of
the life in newly flooded areas emerges from cysts, spores,
seeds, or eggs diapausing on the dry substrate.
Found from bacteria to fishes in temporary waters,
diapause is common in organisms with limited dispersal.
Typically, the organism is replaced by a small, highly
desiccation resistant structure that awaits rehydration in
the sediment. The substrate reservoir of diapausing
microbes, plants, and animals is termed a seed bank, egg
bank, or propagule bank. Diapause also occurs in larval
and adult stages. Reproductive diapause is seen in some
insects, and certain flatworms and annelids enter diapause
after encysting in mucus.
Temporary Waters 435
Other dormancy
Other responses to drying involve decreased activity and
lowered oxygen consumption. Some bdelloid rotifers,
tardigrades, and nematodes survive complete dehydration
to revive when flooded. Perennial plants may lose their
leaves and die back to subsurface roots, tubers, or rhi
zomes when water levels recede. African and South
American lungfish on drying floodplains encase in mud,
breathe air, and more than halve their metabolism. Many
mollusks burrow into sediments and estivate. Some insect
pupae become dormant, delaying adult emergence.
Dormancy is generally less effective than diapause for
surviving extended drying or unpredictable flooding.
Avoidance
Anatomical, behavioral, or physiological adaptations can
help organisms avoid drying. Some plants extend long
roots deep into groundwater (Figure 9). Crayfish exca
vate burrows that remain flooded after surface drying.
Figure 9 Along the banks of ephemeral rivers and seasonal
waterbodies in arid regions, deep-rooted trees and shrubs such
as these tamarisks (Tamarix spp.) tap the groundwater and can
influence drying of the waters at the surface. A gradient of
increasing salinity tolerance is seen in plants radiating outward
from the water source.
Animals in intermittent streams move downward into
areas of high moisture or subsurface flow. Some insects
and fish migrate between permanent and temporary
waters. Amphibians and some insects have aquatic larvae
and terrestrial adults.
Physiological ecology
Life in temporary waters may require biochemical mod
ifications and major physiological adaptations. Many
endemic plants from temporary waters use C4 or cras
sulacean acid metabolism (CAM) photosynthesis,
biochemical pathways that use water more efficiently
than the C3 photosynthesis of most plants. Species
along salinity gradients show increasing osmoregulatory
specializations.
Temporary waters have large local thermal gradients
and over time may be subfreezing or above 40 C.
Biological processes vary with temperature, typically
doubling with each 10 C increase. Most species grow
within narrow temperature ranges, and thermal cues reg
ulate many life cycle events. Enzymes need to function
over temperature ranges found in temporary water bodies
during organisms’ life cycles. For example, inhabitants of
Antarctic rockpools are active in cold water, and they
diapause or secrete antifreeze substances to avoid conse
quences of subfreezing temperatures; their physiology
differs markedly from relatives in temperate or desert
pools.
Most freshwater plants and animals cannot regulate
internal ionic concentrations in salt water. Inhabitants of
many desert waters have impermeable body surfaces,
salt exporting cells, modified life histories, and well
developed drought resisting adaptations (Figure 10).
Figure 10 The stick-like cases of salt-tolerant caddisfly larvae
(Insecta: Trichoptera: Limnephilidae: Limnephilus assimilis) litter
the bottom of Salt Creek in Death Valley, CA, in winter. The ability
to regulate hemolymph osmotic and ionic concentrations in
brackish waters, rapid growth, adult reproductive diapause
during the hot summer months, and the presence of fewer
predators than in low-salinity waters contribute to the species’
persistence in this temporary desert stream.
436
Temporary Waters
Many are active in winter, when salinities and tempera
tures are low. Energetic costs of osmotic and ionic
regulation must be compensated for by benefits, such as
abundant food or reduced predation. Distributions in
desert waters reflect species’ physiological tolerances,
with no plants and few highly adapted animals in hypersa
line pools, greater diversity at low salinities, and low
richness in highly unpredictable, ephemeral, freshwater
rainpools.
Populations
Bet hedging
Desiccation resistant seeds of annual plants from tempor
ary waters germinate when chemicals in the seed coat are
washed away. Seeds from the same parent have different
levels of resistance, ensuring that not all germinate at
once, and that some remain in the sediment seed bank.
Similarly, crustaceans and some insects form an egg bank
comparable to the seed bank of plants; some of the eggs
hatch upon flooding, others hatch another time. African
and South American annual killifish deposit diapausing
eggs into the egg bank in the sediment of floodplain pools,
and these eggs, too, hatch differentially upon flooding.
This strategy of spreading risk is predicted from game
theory and is termed bet hedging. The new field of
resurrection ecology uses egg and seed banks to establish
new communities in restoration projects, and obtains
insights into evolutionary processes by growing indivi
duals from samples collected a century or more ago and
comparing them with modern individuals.
Life-history strategies
Numerous studies address short term controls on popu
lation growth and survival in temporary waters. What are
appropriate responses to flooding and drying, when they
occur at different times from one year to the next, or to
salinity and temperature? Theories of r and K selection
predict that some species produce many, small progeny,
raising the odds that some will survive. Others produce
fewer, but larger progeny with more reserves to support
them over adverse conditions. Examples tending toward
both extremes can be found in temporary waters.
Environmental conditions and evolutionary history
shape species’ responses, and many cues stimulate the
initiation and termination of life cycles (Tables 4 and 5).
Short, irregular hydroperiods should favor r selected
life histories including rapid hatching/germination upon
flooding, fast growth, and timely entry of many propa
gules into a drought resistant state. Under longer,
predictable hydroperiods, k selection should produce
slower growth, larger sizes, and longer life spans. In rock
pools worldwide, with hydroperiods from hours to weeks,
algae, insects, and crustaceans in the most ephemeral
pools complete development in less than 24 h. Life spans
are longer in longer duration pools, and in those with
Table 4 Some cues stimulating the termination of diapause in
temporary waters
Hydration
Hydration plus temperature
Hydration plus chemical cues
Hydration plus chemical and thermal cues
Hydration after drying (a minimum period of drying may be
required)
Hydration plus chemical cues after drying
Hydration plus chemical and thermal cues after drying
Hydration after drying and low temperatures or freezing
(a minimum period of drying and exposure to cold temperatures
may be required)
Hydration plus chemical cues after drying and low temperatures
or freezing
Hydration plus chemical and thermal cues after drying and low
temperatures or freezing
Photoperiod in combination with one or more of the above
Table 5 Some cues initiating life stages adapted to drying in
temporary watersa
Developmental stage (obligate diapause/dormancy/
transformation once development reaches a critical threshold,
regardless of habitat favorability)
Developmental stage plus other cues (diapause/dormancy/
transformation initiated facultatively after development
reaches a critical threshold, only after habitat becomes
unfavorable)
Water temperature
Photoperiod
Chemical cues (pH, dissolved oxygen, chemical signals from
predators or competitors, salinity, nutrients, other)
Drawdown-associated cues (chemical concentrations,
crowding, depth)
a
Note that if drying occurs before necessary developmental thresholds
have been reached, cues cannot initiate drought resistant stages, and
organisms may die without completing their life cycles.
predictable flood regimes. Temperate Eubranchipus fairy
shrimp and some aedine mosquitoes have one generation
per year; they hatch, grow, mature, mate, deposit diapaus
ing eggs, and die. In some fingernail clams, only young
individuals resist drying, and they enter obligatory dia
pause as soon as they are born. Similar patterns are seen in
many species.
Life history tradeoffs, such as the ability to grow
while water remains, potentially allow production of
more offspring, or development to a larger size, which
may enhance survival and reproductive fitness. A
longer developmental period may also mean higher
intraspecific densities and competition as habitat
shrinks, and it increases the risk of being stranded if
drying occurs rapidly. Haematococcus pluvialis, a photo
synthetic flagellate related to Volvox from rockpools
worldwide, is typical – diapausing spores develop
rapidly when flooded, the organisms grow and repro
duce, and upon pool drying, the motile cells form
Temporary Waters 437
aplanospores that withstand drying and high tempera
tures. The diapausing spores form in less than a day, at
any stage in the life cycle, providing Haematococcus with
flexibility in the face of variable habitat duration.
Widely different taxa grow rapidly to a minimum size
threshold, after which they can reproduce and grow
through multiple generations (e.g., Daphnia spp., snails,
some bivalves), or become larger (e.g., amphibians,
insects), as long as water is present, or until other
cues initiate diapause, dormancy, or transformation.
Complex life histories
Many species’ life cycles are complex. Post hatching popu
lations of the cosmopolitan water flea Daphnia pulex are all
female and reproduce parthenogenetically while conditions
are favorable. When drying threatens, males are produced,
and fertilized eggs develop into diapausing, drought resist
ing ephippia that lie in the substrate until the next flooding
event and temperature cues stimulate hatching. Similar
alternating generations occur in some rotifers.
Diving beetles (Agabus spp.) have a 2 year life cycle.
They hatch from eggs in temporary pools and, when
mature, fly to permanent waters to overwinter, returning
to pools to breed the following spring. The eggs they
leave hatch the following year, before the next wave of
adults arrives. Some water mite larvae that parasitize
Agabus and other migratory insects are transported by
their hosts from temporary waters in fall and back in
spring; they then pass through two predatory life stages
before laying eggs that hatch into new parasitic larvae.
Dispersal, population maintenance, and evolutionary
ecology
How nonmotile organisms disperse has long fascinated
biologists and has implications for community composi
tion and stability. Mechanisms include transport by wind;
bird feet, feathers, and digestive tracts; water; humans; and
insects. Many temporary water populations are units of
metapopulations; they undergo periodic local extinctions
and are recolonized from other waters or provide coloni
zers for other sites. Genetic analysis and modeling help
determine the extent of genetic mixing needed to main
tain populations or allow divergence.
Endemic species, especially in crustaceans and some
plant taxa, are widespread in temporary waters. New species
of copepods, anostracans, and other crustaceans are still
being identified from all over the world and provide exciting
opportunities to understand evolutionary processes.
Community Ecology
Community studies include questions about local and
regional biodiversity; community composition and struc
ture in relation to environmental and biological variables
and disturbances; patterns of colonization and extinction;
predator–prey, host–parasite, and competitive interac
tions between species; and food webs.
Comparable temporary waters differ in their biota.
Distributions of species, and thus community com
position, shift along gradients of size, hydroperiod,
predictability, and salinity, with richness increasing with
decreasing stress. Community composition may change
between years, and it can also vary seasonally, with a
succession of new hatches and migrants entering waters
over time. The presence of potential community mem
bers as unhatched propagules in the sediment complicates
assessments of community composition and structure.
Community theory
The theory of island biogeography postulates that species
richness in isolated habitats is regulated by local extinction
and colonization and should vary with habitat size and
proximity to potential sources of colonizers. The intermedi
ate disturbance hypothesis predicts high richness in
communities subject to a moderate degree of disturbance
or stress; according to this model, high stress leads to
mortality in all but fast growing individuals, and under
low stress, inter and intraspecific interactions such as com
petition and predation determine community structure.
Other models look at resource and habitat partitioning/
niche diversification, temporal offsets in life histories, and
other mechanisms controlling community composition and
structure. Studies of amphibians, plants, invertebrates, and
algae in temperate woodland pools, Mediterranean tempor
ary pools, Negev and Namibian desert pools, Scandinavian
rockpools, Arctic snowmelt pools, and other areas show
complex relationships between community composition
and habitat variables such as size, hydroperiod, frequency
of flooding, hydrologic predictability, distance from other
waters, and salinity. The data suggest that community
richness is related to both degrees of disturbance and the
predictability of disturbance. Isolation is also important,
with greater richness in waters that are connected to larger
bodies (e.g., in floodplains) but also fewer taxa specifically
adapted to temporary habitats. Species pools in individual
water bodies are poor in comparison to the regional set of
species (Table 6), and experimental assemblages comprised
of larger subsets of available species function differently
than the smaller natural communities.
Interspecific interactions
Food web manipulations allow examination of relation
ships among species and show interesting relationships.
For instance, algae grow better when grazed by tadpoles
than alone. Some potential competitors avoid conflict by
preferentially choosing waters with different hydrologic or
other characteristics when the other species are present.
The survival outcomes for some species of amphibians and
insects when they co occur with competitors depend on
which species becomes established first.
438
Temporary Waters
Table 6 Regional species pools ( diversity) are greater than
local species pools ( diversity), as illustrated by numbers of nondipteran macroinvertebrates found in early spring from nine
adjacent temporary pools on Cape Cod, Massachusetts, USA
Water body
Number of taxa
Pool 1
Pool 2
Pool 3
Pool 4
Pool 5
Pool 6
Pool 7
Pool 8
Pool 9
Total species
34
22
24
37
12
38
48
28
22
89
Modified from figure 2 in Colburn EA (2004) Vernal Pools: Natural History
and Conservation. Blacksburg, VA: McDonald and Woodward.
Certain aquatic insects, crustaceans, and vertebrates
can survive in pools with long flood durations, but they
are typically found only at the more ephemeral end of the
flooding continuum. They are excluded from the longer
hydroperiod pools by predators such as amphibian larvae,
tadpole shrimp, and water bugs. The ovipositing females of
some species explicitly avoid pools with predators. For
example, vulnerable species of mosquitoes avoid laying
eggs in pools containing predatory backswimmers, whereas
predation resistant midge larvae do not; American toads
(Bufo americanus) avoid temporary pools with omnivorous
wood frog tadpoles (Rana sylvatica).
Ecosystem Ecology
There are many questions about temporary waters as
ecosystems. How do tiny, intermittently flooded water
bodies produce huge numbers of insects, amphibians,
and other organisms? How do nutrients, carbon, and
energy flow within temporary waters, and between them
and adjacent terrestrial landscapes?
Some temporary waters are among the most produc
tive ecosystems known. In some temporary habitats,
photosynthesis by microscopic producers is the base of
the food web. For many, from microhabitats in plant
leaves to large woodland and floodplain pools, decompos
ing detritus is the primary energy source. Much remains
to be learned about the sources and fluxes of energy and
nutrients in temporary water ecosystems.
Applied Ecology
Vector Control
Mosquito borne diseases including encephalitis, yellow
fever, West Nile fever, and, especially, malaria affect
millions of people and are a major focus of world health
agencies. Most mosquitoes breed in temporary waters.
Their populations have expanded following human
alterations of natural habitats, the creation of flooded
areas by equipment and land use change, and the disper
sal of water retaining containers. The effective long term
control of disease vectors requires understanding of the
ecology of the pest animals and of their habitats.
Ecological Engineering and Conservation
Bird and amphibian populations and unique aquatic
species depend on temporary waters, and the overall
contributions of these systems to biodiversity are still
being explored. Losses of these habitats are severe (e.g.,
loss estimates for California vernal pools exceed 90%),
and remaining sites face draining, filling, excavation, pol
lution, water abstraction, invasive species, and climate
change. In many regions, seasonal rivers and pools are
important water sources; elsewhere, temporary waters
provide the only arable areas. Many have been dammed,
or converted for rice culture and other crops. Hydraulics,
hydrology, surface–groundwater interactions, and
biology affect management of these systems for human
use and conservation, habitat restoration, and habitat
creation.
See also: Freshwater Lakes; Freshwater Marshes; Saline
and Soda Lakes.
Further Reading
Batzer DP, Rader RB, and Wissinger SA (eds.) (1999) Invertebrates in
Freshwater Wetlands of North America: Ecology and Management.
New York: Wiley.
Belk DA and Cole GA (1975) Adaptational biology of desert
temporary pond inhabitants. In: Hadley NF (ed.) Environmental
Physiology of Desert Organisms, pp. 207 226. Stroudsburg, PA:
Dowden, Hutchinson and Ross, Inc.
Caceres CE (1997) Dormancy in invertebrates. Invertebrate Biology
116(4): 371 383.
Calhoun AJK and DeMaynadier P (eds.) (2007) Science and
Conservation of Vernal Pools in Northeastern North America.
New York: CRC Press.
Colburn EA (2004) Vernal Pools: Natural History and Conservation.
Blacksburg, VA: McDonald and Woodward.
Eriksen C and Belk D (1999) Fairy Shrimps of California’s Pools,
Puddles, and Playas. Eureka, CA: Mad River Press.
Fryer G (1996) Diapause, a potent force in the evolution of fresh water
crustaceans. Hydrobiologia 320: 1 14.
Hartland Rowe R (1972) The limnology of temporary waters and the
ecology of Euphyllopoda. In: Clark RB and Wooton EF (eds.)
Essays in Hydrobiology, pp. 15 31. Exeter, UK: University of
Exeter.
Laird M (1988) The Biology of Larval Mosquito Habitats. Boston:
Academic Press.
Simovich M and Hathaway S (1997) Diversified bet hedging as a
reproductive strategy of some ephemeral pool anostracans
(Branchiopoda). Journal of Crustacean Biology 16(3): 448 452.
Sloan CE (1972) Ground Water Hydrology of Prairie Potholes in North
Dakota. USGS Professional Paper 585 C. Reston, VA: US
Geological Survey.
Tropical Rainforest
Wiggins GB, Mackay RJ, and Smith IM (1980) Evolutionary and
ecological strategies of animals in annual temporary pools. Archiv für
Hydrobiologie (Supplement) 38: 97 206.
Williams DD (1987) The Ecology of Temporary Waters. Portland, OR:
Timber Press.
Williams DD (2006) The Biology of Temporary Waters. London: Oxford
University Press.
439
Witham CW, Bauder ET, Belk D, Ferren WR, Jr., and Ornduff R (eds.)
(1998) Ecology, Conservation, and Management of Vernal Pool
Ecosystems Proceedings from a 1996 Conference. Sacramento,
CA: California Native Plant Society.
Zedler PH (1987) The Ecology of Southern California Vernal Pools.
Biological Report 85(7.11). Washington, DC: US Fish and Wildlife
Service.
Tropical Rainforest
R B Waide, University of New Mexico, Albuquerque, NM, USA
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Definitions
Distribution
Climate and Soils
Forest Structure
Biodiversity
Conservation Issues
Further Reading
Introduction
elevation, the boundary between tropical rainforest and
other forest types is by necessity arbitrary.
The classical definition of tropical rainforest focuses
on features of the vegetation: evergreen, hygrophilus, tall,
and rich in lianas and epiphytes. Additional characteris
tics of tropical rainforest include the dominance of woody
plants (Figure 1), principally trees; high species richness;
sparse undergrowth; relatively slender trunks compared
to trees of temperate forests; straight boles without
branches except near the top; buttresses (Figure 2);
large, dark green leaves with entire margins; the
occurrence of flowers on the trunk or branches; and
inconspicuous green or white flowers.
Alternative definitions of tropical rainforest focus on
characteristics of the forest community and its environ
ment, including the proportion of deciduous trees in the
canopy, the elevation of the forest, and the length and
severity of the dry season. Some local classifications of
forest type incorporate floristic information, but such
classifications require detailed knowledge of the flora as
well as trained experts to implement them.
Several schemes attempt to classify vegetation types
based on climatic conditions including temperature, rainfall,
length of dry periods, and evapotranspiration. Such classifi
cation systems avoid the subjectivity inherent in definitions
that depend on relative terms, but by necessity are over
simplifications of the factors controlling the distribution of
tropical rainforests. The Köppen classification uses the aver
age annual precipitation, average monthly precipitation, and
average monthly temperature to divide the globe into six
major climate regions and their subregions. Under this sys
tem, the tropical rain climate has no month whose mean
Of all of the Earth’s ecosystems, tropical rainforests exist
on the extremes of temperature, rainfall, biodiversity, and
structural complexity. Tropical rainforests exist only
where high year round temperatures are found in con
junction with moderate to high rainfall, which limits both
their latitudinal and elevational distribution. Compared
to other ecosystems, tropical rainforests have high num
bers of plant and animal species and often show great
specificity in their biological relationships. This high
taxonomic diversity contributes to high functional diver
sity, which results in a complex forest structure
comprising many different life forms and sizes of plants.
This diversity and structural complexity makes tropical
rainforests one of the most interesting and complex
ecosystems on Earth, and, as such, tropical rainforests
have captivated the imaginations of scientists and the
public alike.
Definitions
The common use of the term ‘tropical rainforest’ varies
among regions of the globe depending on the ecological
context. There is general agreement that tropical rain
forests are tall, dense evergreen forests existing in wet and
warm places, but since there is a degree of subjectivity in
these terms, the name ‘tropical rainforest’ is applied to
forests on different continents that may be quite different
structurally. Moreover, since the characteristics that
define tropical rainforest grade with latitude and
440
Tropical Rainforest
rainforests occurring where the mean annual temperature
exceeds 24 C. Shifts between forests classes are determined
by changes in rainfall and temperature related to elevation
and latitude. These classification systems work well in areas
where plant formations are strongly controlled by climate,
such as Central America and northern South America, but
are less useful where edaphic factors or other environmental
factors are major controlling factors, as in the lower
Amazonian region in Brazil.
The rest of this article focuses on lowland, evergreen
tropical forests occurring in hot, wet conditions. Tropical
forests at higher elevations or in areas with a pronounced
dry season are covered elsewhere.
Distribution
Figure 1 Tall, relatively slender trees with straight boles are
characteristic of tropical forest on Barro Colorado Island,
Panama. Credit: Nicholas V. K. Brokaw.
Figure 2 Broad buttresses are found on many rainforest trees,
such as this specimen from Providencia, Antioquia, Colombia.
Credit: Robert B. Waide.
temperature is less than 18 C and the mean rainfall of the
driest month is >60 mm. In the Holdridge classification,
rainforests are defined as areas where the ratio of potential
evapotranspiration to rainfall is low, with tropical lowland
Tropical rainforests exist wherever conditions are appro
priate, but are mostly confined to a broad belt around
the equator. The latitudinal distribution of tropical rain
forests is limited by the distribution of freezing
temperatures, which tropical plants are unable to with
stand. A circumglobal belt of dry conditions also limits
the distribution of tropical rainforest and, except in rare
cases, prevents a continuous transition between tropical
and temperate forests. Some gaps occur in the equatorial
band of tropical rainforests, such as in eastern Africa,
where prevailing conditions are too dry for tropical rain
forest to develop. Moreover, in some areas on the eastern
margins of continents, conditions suitable for tropical
rainforest exist outside of the tropics. In all areas suitable
for the development of tropical rainforest, human actions
limit the present distribution of forests.
Large areas of tropical rainforest exist on continents
and large islands that straddle the equator. Roughly half of
the tropical rainforests on the planet occur in three areas
in tropical America. The largest of these forest areas
(somewhat over 3 million km2) occupies the drainage
basins of the Amazon and Orinoco Rivers in northern
and central Brazil and surrounding countries. A narrow
strip of tropical rainforest runs along the Atlantic coast of
Brazil from 7 to 28 S (from Recife nearly to São Paulo),
but less than 5% of this forest remains in its original
condition. A third block of forest occupies southern
Mexico, Central America, and the area of northern
South America west of the Andes. Many Caribbean
islands also have small areas of tropical rainforest.
In Africa, another large block of tropical rainforest
occupies the basin of the Congo River in the Democratic
Republic of the Congo, the Republic of the Congo, Gabon,
and Cameroon. Previously, part of this forest extended into
Nigeria. Belts of tropical rainforest also extended along the
coast of West Africa and the eastern part of the island of
Madagascar, but little remains of these forests but isolated
patches.
Tropical Rainforest
A third large area of tropical rainforest existed on the
Malay Peninsula and the islands of Borneo, Sumatra, and
Java. Sulawesi, the Philippines, and many of the smaller
islands in Indonesia also have substantial areas of rainforest,
but the condition of the remaining forest varies widely from
island to island. Rainforests also occupied parts of mainland
Southeast Asia where rainfall was sufficiently high. Isolated
patches of tropical rainforest occur in the area of the Western
Ghats in India and on the island of Sri Lanka. Most of the
island of New Guinea supports tropical rainforest, and there
is also a small area of rainforest in NE Australia. Patches of
tropical rainforest also occur on some of the Pacific Islands
(Solomons, New Hebrides, Fiji, Samoa, New Caledonia).
Climate and Soils
Tropical rainforests are found under a surprisingly wide
range of climatic conditions. Annual rainfall is generally
high in tropical rainforest compared to other ecosystems,
but can range from 1700 to 10 000 mm. Many rainforests
experience 1–4 dry months a year, when the rainfall is less
than the water lost through evaporation and transpiration.
These annual dry periods exert a strong effect on the
phenology of biotic processes such as flowering and fruit
ing. In some tropical rainforests, rainfall is uniformly high
throughout the year, and no annual dry periods exist. In
these forests, dry periods may occur at multiyear intervals
and trigger strongly synchronized biotic responses
including mass flowering, increased animal reproduction,
and migration. In some parts of the world, these multiyear
cycles are related to periodic El Niño–Southern
Oscillation (ENSO) events. While strong ENSO events
result in more severe dry periods in many tropical rain
forests, the strongest biological consequences seem to
occur in forests that do not normally experience a dry
season, especially in areas of Indonesia and Malaysia.
Mean annual temperatures in tropical rainforests gen
erally fall in the range between 24 and 28 C near the
equator, but a consistent characteristic is the absence of a
cool season. In general, diurnal temperature differences
(6–10 C) exceed monthly differences. The amount of
solar radiation is higher in the topics than in temperate
zones, but tropical rainforests generally have lower avail
able solar radiation than drier tropical forests because of
the greater amounts of water vapor and increased cloudi
ness in more humid climates. As a result, plant growth in
closed canopy tropical rainforests is often light limited.
The environments of tropical rainforests are charac
terized by high relative humidity during the daytime and
generally saturated conditions at night. However, because
much of the rainfall in tropical rainforests occurs in
intense events, even months with high rainfall can have
periods of a few days when little or no rain falls, saturation
deficits increase, and plants wilt. The dry periods can be
441
exacerbated by winds; evaporation rates are higher in the
trade wind zone than in equatorial forests where average
wind speeds are less.
Tropical cyclones can have severe effects on tropical
rainforests. In general, areas within 10 latitude of the
equator are not subject to tropical cyclones, but tropical
rainforests in the Caribbean, Madagascar, northeastern
Australia, many oceanic islands, and parts of Central
America and Southeast Asia are affected by these storms.
The strongest tropical cyclones can have severe but, in
most cases, temporary effects on forest structure and
composition. Tree mortality can be high as a result of
one of these storms but the forest recovers quickly
through regeneration, new growth, and refoliation of
damaged trees. In those areas of tropical rainforest subject
to recurrent tropical cyclones, forest structure and the
biological traits of forest species may be affected by the
frequency and intensity of storms.
The soils underlying tropical rainforests can have
important effects on plant distribution and primary produc
tivity. The complex interactions between soil
characteristics (e.g., soil texture, age, drainage characteris
tics, nutrients) and the considerable topographic and
geographic variation in these characteristics make it diffi
cult to determine the importance of specific soil properties.
Most areas supporting tropical rainforests have very old
soils that are highly leached and weathered and as a result
acidic and infertile. Such soils have low levels of the nutri
ents necessary for plant growth and high levels of toxic
aluminum and thus are unsuitable for most forms of perma
nent agriculture. However, these soils can sustain high
diversity, high biomass tropical rainforests because plants
of these forests recycle nutrients efficiently. Some tropical
rainforests occur on relatively fertile volcanic or floodplain
soils and can sustain permanent agriculture.
In areas of the Amazon with low local relief, soil
properties can have a strong effect on plant communities
and therefore overall biodiversity. Small changes in topo
graphy and the depth of sand overlying the clay subsoil
can cause large changes in the plant community.
Forest Structure
Tropical rainforests support a more diverse set of organ
isms than other kinds of forests. The number of different
life forms, or synusiae, is greater in tropical forests that in
temperate forests. A synusia is a group of organisms whose
members are ecologically equivalent. When applied to
plants, the term reflects an aggregation of species with
similar life form and function. Autotrophic plants
(e.g., those that photosynthesize) include those that do
not need mechanical support (i.e., trees, shrubs, and
herbs) and those that do (i.e., climber, epiphytes, and
442
Tropical Rainforest
Figure 3 Bromeliads and other epiphytes cover branches of
trees in La Planada Reserve, Colombia. Reproduced by
permission of Art Wolfe/Photo Researchers, Inc.
hemi epiphytes; Figure 3). Heterotrophic plants include
saprophytes and parasites.
The structure of a tropical rainforest arises from each
synusia’s methods for obtaining resources for survival and
growth: water, nutrients, and sunlight. In some forests,
photosynthetic, self supporting plants seem to form dis
tinct strata depending on their size. Such stratification is
by no means a uniform characteristic of tropical rainfor
ests. Photosynthetic plants that are not self supporting use
other plants as a platform for growth. Climbers (lianes)
have roots in the ground but use other plants to support
their elongated stems. Epiphytes depend on their host
plants for support only, although a specialized group of
this synusia (the mistletoes) obtains both support and
water and dissolved substances from the support tree.
Hemi epiphytes initially live as epiphytes on supporting
plants but eventually send roots down to the ground.
Saprophytic and parasitic plants obtain required energy
and nutrients from other living or dead plants, and there
fore do not require light for growth or reproduction.
Background mortality of individual trees from natural
causes is a major cause of spatial heterogeneity in tropical
rainforests. Gaps caused by dead or fallen trees change the
structure and the environmental characteristics of the forest.
However, tropical rainforests are also dynamic ecosystems
subject to a large number of natural disturbances including
storms, lightning strikes, landslides, and the effects of ani
mals, all of which can produce gaps in the forest canopy.
The canopy is an important structural element of tropi
cal rainforests because the height and degree of closure of
the canopy plays an important role in determining condi
tions in the understory (Figure 4). Moreover, the lack of
easy access to forest canopies means that their importance as
a source of biodiversity and an influence on ecosystem
processes has probably been underestimated. Forest cano
pies have important roles in the regulation of nutrient
cycling and in the storage of carbon. Large pools of nutrients
exist in live and dead components of the canopy, and
Figure 4 Canopy of lowland tropical rainforest of La Selva
Biological Station, Costa Rica. Photographed from a light plane
flying 200 feet above the canopy. Reproduced by permission of
Gregory G. Dimijian, MD/Photo Researchers, Inc.
decomposition of organic matter in the canopy influences
access to these nutrient pools. The forest canopy serves to
filter air and waterborne nutrients and to provide a site for
nitrogen fixation. Canopy dwelling organisms are efficient
at acquiring and storing nutrients, thus providing a buffer for
pulsed nutrient releases. Forest canopies are rich in species
of plants and animals that are independent of the forest floor.
Moreover, canopy trees and their epiphytes provide impor
tant sources of food for birds, mammals, and insects that
occupy other strata.
Biodiversity
Understanding of the biodiversity of tropical rainforests is
still being refined. New species of all taxonomic groups are
found every year, and knowledge of the diversity of some
taxa, especially insects, is rudimentary. Tropical rainforests
are extremely rich in species of all taxa compared to other
terrestrial ecosystems. For example, the tropical rainforests
of the world have an estimated 175 000 species of plants,
which constitutes about two thirds of the global total.
Considerable variation in diversity occurs among tropical
rainforests around the world, with the largest number of tree
species (>250 species per hectare) occurring in Amazonia
and Malaysia, followed by the islands of New Guinea and
Madagascar, and then Africa. The largest areas of tropical
rainforest (Neotropics, Africa) have the greatest number of
primate species. Similar comparisons for other taxa are
difficult because of the lack of data.
Conservation Issues
Because of their global significance with regard to carbon
storage and the maintenance of biodiversity, conservation
of tropical rainforests is an important and hotly debated
Tundra 443
topic. Solution of conservation issues is made more diffi
cult because most areas of tropical rainforest occur in
countries that are trying to increase the standard of living
of their people. Partly because of this controversy, ade
quate data to judge the loss of tropical forests are difficult
to come by. However, it is clear that tropical forests,
including tropical rainforests, are disappearing at an
increasing rate. The percent of the original forest habitat
that has been lost exceeds 90% for some countries
(Ghana, Bangladesh, Philippines). Estimates suggest that
very little tropical rainforest will remain by the year 2050.
The ultimate causes of forest loss include increasing
populations in countries with tropical rainforest, extreme
poverty, and the lack of effective government protection
for forests. Proximate causes of forest loss include logging,
clearcutting for agriculture, loss of ecosystem integrity
because of forest fragmentation, and hunting (Figure 5).
Hunting of large animals may have insidious effects on
forest structure, as the populations of prey species may
explode when released from predation. Increased popula
tions of small mammals, for example, may have severe
effects on other organisms, leading to the breakdown of
whole ecosystems over time.
Because the issues facing tropical rainforests vary con
siderably from one place to another, generic conservation
solutions are not practical. However, the major elements of
a conservation strategy for tropical rainforests will include
the creation of reserves to protect biodiversity, the regula
tion of exploitative use of tropical rainforest products, the
engagement of traditional societies, the development of
sustainable use strategies that will address the issue of
poverty, and an increased effort by developed countries
to form partnerships with developing countries.
Figure 5 Rainforest has been cleared for timber and agriculture
in this subsistence farm in Providencia, Antioquia, Colombia.
Credit: Robert B. Waide.
Further Reading
Denslow JS and Padoch C (eds.) (1988) People of the Tropical Rain
Forest. Berkeley, CA: University of California Press.
Gentry AH (ed.) (1990) Four Neotropical Rainforests. New Haven, CT:
Yale University Press.
Golley FB (ed.) (1989) Tropical Rain Forest Ecosystems. New York,
NY: Elsevier.
Primack R and Corlett R (2005) Tropical Rain Forests: An Ecological and
Biogeographical Comparison. Oxford: Blackwell Science.
Richards PW (1996) The Tropical Rain Forest: An Ecological Study.
Cambridge: Cambridge University Press.
Sutton SL, Whitmore TC, and Chadwick AC (eds.) (1983) Tropical Rain
Forest: Ecology and Management. Oxford, UK: Blackwell Scientific
Publications.
Terborgh J (1992) Diversity and the Tropical Rain Forest. New York:
Scientific American Library.
Whitmore TC (1998) An Introduction to Tropical Rain Forests. New York,
NY: Oxford University Press.
Tundra
R Harmsen, Queen’s University, Kingston, ON, Canada
ª 2008 Elsevier B.V. All rights reserved.
Introduction
The Periglacial Environment
Landscape and Species Diversity
Vegetation and Succession
Ecosystem Structure and Function
Special Adaptations to Tundra Conditions
Global Warming and Other Anthropogenic Effects
Further Reading
Introduction
action, and long periods of shortage of liquid water caused
by freezing and/or drought. These stresses combine to
create what is called the periglacial environment, which
is defined by repeated effects of freezing and thawing on
soils and water bodies. Tundra comes in many different
Tundra ecosystems are widely distributed over all conti
nents. Tundra is characterized by climatic stress consisting
of low temperatures, strong winds, low precipitation, frost
444
Tundra
kinds. The two main categories are arctic and alpine tun
dra. Each of these can be divided into subcategories or can
be seen as gradients from a richly vegetated tundra with tall
shrubbery adjacent to the ‘tree line’ through categories
with less vegetation to barren areas with only a minimum
of vegetation adjacent to ice fields and permanently frozen
polar or alpine areas (see Boreal Forest and Alpine Forest).
Included in the tundra biome are tundra ponds, lakes,
streams, marshes, and other wetlands.
Since tundra is found at the cold limit of life forms on
Earth, climatic changes of the past have had major effects
on tundra ecosystems and the plant and animal species of
these systems. With each Pleistocene ice age, big areas of
arctic tundra were eradicated, while others shifted south
wards, as entirely new areas of forest or prairie became
tundra, only to be reversed with the subsequent intergla
cials. Similar changes would have occurred in mountainous
regions. These major changes resulted in the extinction of
species and in the disruption of coevolved, interactive
plant, and animal assemblages. These changes in tundra
communities persist today, resulting in low species diver
sity and the scarcity of complex food chains. During the
current interglacial, many areas that were pushed down by
the weight of the ice were first flooded by the rise in sea
level, but have subsequently, in part at least, re bounded
and developed into tundra. Some parts of Beringia (eastern
Siberia, northern Alaska, and into the Yukon and Banks
Island) were not glaciated, and retained a far northern
tundra during the last glacial period. Species of plants
and animals that now form arctic tundra communities
survived the ice ages either south of the glaciated land
area or in unglaciated refuges such as Beringia.
The Periglacial Environment
Periglacial conditions are the result of current or geolo
gically recent frost and ice formations on a landscape.
Glaciers affect landscapes in major ways, which can
have long lasting effects on geomorphology, drainage
systems, and soils. But even temperature regimes that
cause frequent freeze–thaw cycles – for example,
annually in the high arctic and daily on high tropical
mountains – affects not only plants and animals directly,
but also have indirect effects on soils and water, which
results in specific types of erosion and the formation of
characteristic landscapes. Furthermore, the usual pre
sence of permafrost under tundra ecosystems is of
critical importance, in that it forms a permanently
impenetrable floor, preventing biological penetration
and vertical movement of water and nutrients.
The freeze–thaw cycle causes expansion and contrac
tion of soils and water, while the gradual freezing of wet
soils will also cause a nonrandom redistribution of water
into ice lenses and ice wedges. These processes can result
in frost heave and long term vertical and horizontal
movements of soils, debris, and even large rocks, creating
typical landscape features (such as polygons), frost
mounds (such as palsas and pingos), slope solifluction,
and others. These land forms in turn affect the vegetation
and all other life forms.
The permafrost is ubiquitous in the arctic tundra, but
less frequently found in alpine tundra sites, as alpine
landscapes are more diverse and the summers are warmer.
It is not found in any but the highest tropical mountains.
During spring, the thawing of the soil starts from the
surface down, gradually releasing the vegetation from
the grip of the frozen soil. There is usually an overlap
between snow melt and the thawing of the soil, especially
in undulating landscapes. All melt water must run off,
accumulate in low areas, or evaporate, as no vertical
movement of water is possible due to the impervious
permafrost. This can cause erosion, affecting plants and
small animals. During the summer, thawing of the perma
frost continues till autumn, when the surface may already
start to refreeze. During later autumn and early winter,
the frost will penetrate deeper into the soil from the
surface, as it also comes up from the main body of the
permafrost. This process can cause considerable expan
sion and result in frost heave and can cause much damage
to root systems and animal burrows. In many areas, tundra
soils are low in nutrients, because the permafrost prevents
vertical movement of soil water.
Some lakes (e.g., kettle lakes and moraine lakes) have
their origins in major ice formations dating from the latest
ice age, while others are recent formations. Tundra lakes
and ponds are severely affected by annual freezing, espe
cially those lakes that freeze each winter right to the
bottom and beyond to the permafrost. Freezing of lake
ice causes expansion and results in the shoreline with its
vegetation being elevated above the surrounding low
lands. Lake sediments are often high in inorganic matter
from spring runoff, but low in organic matter due to low
productivity reflecting low nutrient levels. Frost action
and wind effects on ice tend to disturb lake sediments in
shallow lakes.
Landscape and Species Diversity
Many parts of the arctic tundra are flat, especially in areas
adjacent to the sea. These areas are often covered with
ponds and shallow lakes, separated by marshes and con
nected by meandering streams and rivers. These areas can
accumulate peat and develop into fens. Along the sea
shore these habitats tend to merge into salt marshes,
brackish lagoons, and beach ridges. On higher ground,
with hills and rock outcrops, the landscape diversity is
much greater, especially since north and south facing
slopes have very different microclimates, and hence,
Tundra 445
very different biological communities. Here one can also
find deep lakes and fast flowing rivers. Both erosion and
the underlying rock type will also add to ecosystem
diversity. In mountainous areas the arctic tundra merges
into an alpine version.
The diversity of alpine tundra worldwide is enormous,
as it is found on all continents and in many climatic zones.
Snow accumulation during winter, combined with slope,
wind, and summer climate affect the length of the grow
ing season of alpine tundra ecosystems. Tropical alpine
tundra occurs only at very high altitudes, with unique
climates varying from desert to some of the wettest con
ditions on Earth (Figure 1). It should also be noted that
many of the alpine tundra zones are isolated from other
such zones by hundreds or even thousands of kilometers,
so that they have undergone independent evolution of
their flora and fauna. Especially geologically old high
mountains contain many endemic species derived from
Figure 1 Mount Kenya, tropical Africa. High tropical alpine
tundra. In the foreground a boulder moraine with lichens,
mosses, scattered tussock grasses, and a few rosettes of the
large Seneciodendron keniensis. In the middleground a sparse
stand of the yet larger Seneciodendron keniodendron. The genus
Seneciodendron is endemic to the east-central African
mountains. In the background the Tyndall Glacier. Photo by
W. C. Mahaney.
local forest or savannah species. For instance, the
Southern Alps of New Zealand have over 600 species of
alpine plants, very few of which are found elsewhere on
Earth.
Roughly 5% of the Earth’s surface is covered with arctic
vegetation and 3% with alpine vegetation. The alpine tun
dra worldwide, as well as per hectare for most alpine
systems, has a much higher biodiversity than the arctic
lowland tundra. Species richness declines with altitude on
mountains and with latitude in the arctic, and also is depen
dent on local climatic conditions, nutrient availability, etc.
Vegetation and Succession
Whether one climbs a mountain and crosses the timber
line, or travels northwards in the arctic, and crossing the
tree line, one enters the low tundra, which is character
ized by shrubs. A combination of low temperatures,
shallow soils, and strong winds prevents tree growth, but
a tight shrub cover manages to thrive under such condi
tions. On each mountainous area on Earth shrub tundras
can be found, which are superficially quite similar to
other isolated alpine shrub tundra communities; even
many of the individual species have a remarkably similar
appearance. However, mostly unrelated species form such
shrub communities in different parts of the world. For
instance, most of the species of the shrub vegetation on
East Africa’s Mount Kenya, New Guinea’s Mount
Wilhelm, and Pico Mucuñuque of the Venezuelan
Andes belong to different families. This is a good example
of convergent evolution acting on divergent taxa, causing
adaptation to a specific environment. The shrub zone in
the Canadian arctic has a more impoverished vegetation
than the shrub zones on tropical mountains. It is domi
nated by several species of willow and birch, and a
smattering of other species (Figure 2). Again, the arctic
Figure 2 Hudson Bay Lowlands, Northern Manitoba, Canada
60 N. Low arctic willow (Salix spp.) and graminoid tundra. Note
the radio-collar on the polar bear.
446
Tundra
tundra in Greenland, Scandinavia, or Siberia also looks
very similar, but in this case the species are all close
relatives or even the same circumpolar species on the
different continents. Another difference is that on tropical
mountains there are a lot of shrub species that are not
found below the tree line, whereas in the arctic many of
the shrub species are also found south of the tree line.
These differences are the result of the different effects of
the ice ages, which on mountains merely caused the
vertical movement of more or less entire plant commu
nities up and down alpine valleys and slopes, while in the
arctic, changes in the climate can cause north–south dis
placements of the conditions suitable for shrub tundra of
over hundreds of kilometers.
The more typical graminoid, forb, and moss tundra
found higher up the mountains and further north in the
arctic is adapted to extreme cold, long periods of tem
peratures permanently below freezing (and permanent
darkness in the arctic) and strong winds. It is the strong
winds blowing ice crystals which abrade any vegetation
above snow level, combined with desiccation that makes
tree and tall shrub growth impossible in high arctic and
alpine tundra. Especially in arctic deserts, where snow
cover is low, vegetation remains very low to the ground
(Figure 3). For instance on Banks Island at 70 N, arctic
willow (Salix arctica) grows horizontally along the ground,
forming matted areas of intertwining branches that form
catkins and leaves in summer. One such willow can live
and grow for decades. All grasses, sedges, and forbs die
back in autumn and survive the winter as belowground
root masses, or as ground hugging rosettes.
One advantage of being a plant in a dense, low to the
ground plant community is that on cool, sunny summer
days radiant heat from the 24 h solar radiation is trapped
Figure 3 Banks Island, North West Territories, Canada 70 N.
Upland high arctic tundra, also described as arctic desert. The
vegetation is dominated by mountain avens (Dryas integrifolia)
and various species of arctic vetch (Oxytropis spp. and
Astragalus spp.) and scattered clumps of small graminoids. In the
background is the Thomsen River valley with sedge meadow
tundra and tundra ponds.
within the air between the plants, keeping temperatures
high enough for growth and maturation of seed. There are
very few annual plants in the high arctic tundra, because
the season is not long enough to germinate, grow, and
reproduce. A few very small species, such as Koenigia
islandica and Montia lamprosperma, maintain an annual life
strategy. Uniquely a few species of semiparasitic members
of the Scrophulariaceae, such as Euphrasia arctica, do so as
well. These species have a distinct early season advantage
being able to grow very rapidly by gaining nutrients and
photosynthate from neighboring perennials.
The frequent disturbances due to the freeze–thaw
cycles often lead to local eradication of vegetation. This
creates openings for reinvasion and subsequent succes
sion. One of the most interesting examples of this is the
result of solifluction of soil clumps on south facing slopes
in the arctic. Soil clumps with vegetation surrounded by
clefts get heated by the sun on the downslope side,
causing them to thaw out and slump downwards, burying
the lowest vegetation, while at the same time exposing a
small strip of upslope bare soil (Figure 4). It takes up to 30
years for the clump to make one entire downhill rotation.
On each clump, one can see a successional sequence of
plant maturity, species composition, and diversity, as the
oldest community gets buried and an opening appears at
the top end for reinvasion. Succession on a larger scale
occurs after slope collapses, frost mounting, stream ero
sion, mud deposits after flooding, etc.
Ecosystem Structure and Function
Very few species remain active within arctic tundra eco
systems during the winter. Only most mammals such
as the muskox (Ovibos moschatus), the reindeer (Rangifer
Figure 4 Banks Island, North West Territories, Canada 70 N.
Two types of high alpine tundra. In the foreground a wet
graminoid tundra fed by snowmelt water. On the opposite slope
a sparsely vegetated dry tundra showing solifluction. The
muskoxen feed primarily on the graminoid slope, but will venture
onto drier tundra types to feed on high nitrogen species such
as arctic vetch.
Tundra 447
tarandus), the arctic hare (Lepus arcticus), lemmings, and
the wolf (Canis lupus arctos) remain fully active. A few
birds, for example, raven (Corvus corax) and rock ptarmi
gan (Lagopus mutus), manage as well. During the autumn
and early winter months, soil microbial metabolic activity
continues down to at least 12 C. The vast majority of
organisms that spend the winter on the tundra do so in
some form of dormancy. Alpine tundra, being much more
diverse, and much of it having periods of daylight
throughout the year, varies greatly in the degree of winter
activity of the fauna. The brief summer on the tundra is
enormously productive, and provides food for a wide
variety of organisms. The vegetation starts to bloom and
grow as soon as the snow starts to melt. At that time of
year the sun hardly sets if at all and temperatures rise
quickly. Dormant overwintering insect larvae start to feed
and eggs eclose to add innumerable larvae in snow melt
ponds, in the soil, and on the new vegetation. The eco
system seems to burst into active life. High availability of
edible vegetation, exploding insect, bird and rodent popu
lations, and young birds lasts till just before freeze up in
autumn (Figure 4).
Many bird species migrate annually from more south
erly wintering sites to the tundra to breed, taking
advantage of, and adding to, the burst of summer produc
tivity. Some of these species arrive in extremely large
numbers. Most of these birds are insectivorous or feed
on pond crustaceans, some such as loons and grebes are
pisciverous, falcons and hawks are predators, and geese
are herbivorous. Especially the colonially nesting geese
can have major destructive effects on the vegetation,
which in turn can affect many other species.
In some tundra ecosystems some small mammals, espe
cially two species of lemmings, show extreme oscillations
in population density, making them keystone species in the
tundra ecosystem. For instance, on Banks Island in north
ern Canada both the collared lemming (Dicrostonyx
torquatus) and the brown lemming (Lemmus sibiricus)
undergo sharp population oscillations with a 3–5 year per
iod. At peak populations the lemmings are all over the
place, whereas the year after it is hard to find a single
lemming. During the outbreak phase, several predatory
birds, including snowy owls (Nyctea scandinaca), rough
legged hawks (Buteo lagopus), and jaegers (Stercorarius spp),
migrate long distances and concentrate in the regions with
high lemming populations. They lay large clutches and
raise many young, only to disperse to other areas when
the lemming population collapses (Figure 5). Mammalian
predators are not as able to respond by migration. Arctic
foxes (Alopex lagopus) and ermines (Mustella erminea) are the
main mammalian predators; they also take advantage of
lemming outbreak with large litters. However, this leaves
relatively dense populations of these predators after the
collapse of the lemming population. This has a major feed
forward effect in that the half starved predators exert a
Figure 5 Nest of snowy owl (Nyctea scandiana) with six eggs
and one hatchling. Snowy owls start incubating as soon as their
first egg is laid, so that the young are hatched sequentially. Note
the seven dead lemmings surrounding the nest, intended as food
for the hatchlings. Later that summer, the lemming population
crashed. Only the two eldest hatchlings survived to fledge, the
others were eaten by the older ones.
strong negative effect on other less favored prey species,
mostly birds, from small passerines to ducklings and even
goslings. Only after the predator population has collapsed
can the lemming population start to grow again.
The ultimate cause of the collapse of the lemming
population is not the predation pressure, but the exhaus
tion of quality vegetation and a delay in nutrient cycling.
However, once the lemming population has collapsed, the
subsequently declining predator population can drive the
lemming population further down to its minimum. The
vegetation, litter layer, and soils are strongly affected by
the lemming cycles. This is shown by the enormous dif
ference between the tundra in northern Canada and central
Greenland, as in Greenland there are no lemmings, much
more accumulated litter, differences in relative abundance
of plant species, and far fewer predators. Exclosure experi
ments in Canadian tundra have similar results.
Special Adaptations to Tundra Conditions
Many species have evolved special adaptations to the
rigorous, but often predictable conditions of the tundra.
This article presents four cases of such adaptations as
examples of this phenomenon: the muskox, two species
of arctic bumblebee, an alpine lobelia, and two congeneric
alpine beetle species.
The muskox of Banks Island in Canada’s
Northwest Territories
The muskox (Ovibos moschatus) is a surviving species of
the Pleistocene megafauna; it survived the ice age both in
Beringia and south of the ice sheet in what is now
448
Tundra
southern Canada and the northern United States. It has a
very long adaptive history in arctic conditions, which
shows in a number of very effective adaptations to
extreme cold. Besides the obvious anatomical features
such as the extremely effective insulating wool under
the shaggy guard hair and the front hooves that are
perfectly shaped to scratch the hard arctic snow to
expose vegetation, this animal has a set of integrated
physiological and behavioral traits making up a unique
reproductive strategy. A muskox cow responds to her
nutritional condition in autumn by not going into heat
when in poor condition, and only going into heat early in
the rutting season when in excellent condition. This
means that cows in poor condition, which would not
have been able to survive the winter and produce a calf
the next spring, will live and have another chance at
reproduction the next year. The cows that do get preg
nant, when faced with a bad winter, will either abort their
fetus or abandon the calf after birth. Since most calves are
born well before snow melt and the reappearance of new
fodder, the cows have to be in good shape to not only
carry the calf to birth, but also lactate for several weeks.
However, only calves born early in the year have a good
chance of gaining enough weight and reserve fat to
survive their first winter.
Integrated with this strategy are some significant traits.
At birth, the calf weight over cow weight ratio is one of
the lowest among ungulates, making abortion or abandon
ment a relatively minor cost for the cow, which can then
cut lactation. Once the calf is born and the cow is lactat
ing, she licks the calf when it urinates and swallows the
urine. The urea of the urine is rebuilt into protein by the
cow’s gut flora and will eventually be available for milk
production. This is important because storage of protein
over the winter is difficult, and late winter forage is scarce
and low in protein. As soon as new forage is available
during snow melt, the cows graze selectively on high
protein vegetation, such as willow catkins and sprouting
rosettes of arctic vetch (Oxytropis spp.). In far northern
parts of their range, muskox cows live long lives, but only
reproduce every second or third year and still lose some
of their calves.
Two Species of Bumble Bee from the Canadian
Arctic
The author has a personal recollection of working in
early July on the tundra on northern Banks Island when
in the middle of a snow squall a bumblebee flew by. This
seemingly incongruous event is explained by the fact that
the common large bumblebee (Bombus polaris) has an
unusually well insulated thorax, which allows it to keep
its flight muscles at approximately 30 C even when the
ambient drops to the freezing point. What is even more
special about this species is that the queen keeps her
abdomen also near 30 C, which presumably allows its
eggs to develop faster. However, early in the season the
queen also warms her eggs and larvae in the nest by
inserting her abdomen into the middle of the nest and
producing heat by vibrating her flight muscles and cir
culating the heat to her abdomen. The queen, after over
wintering, builds the nest, often in an abandoned lem
ming burrow, using bits of dead vegetation and muskox
wool. There she raises one brood of workers before
switching to start raising reproductives for the next
year. The other species of bumblebee (B. hyperboreus)
found on Banks Island is an obligatory brood parasite of
B. polaris. The queens of this species lay only eggs for
reproductives, and lay them in the nest of their host
species. This strategy is obviously adapted to the very
short summer season in the high arctic tundra, but it also
depends on the presence of B. polaris. The ratio of the
densities of the two species is stabilized by frequency
dependent selection.
Flightless Beetles of the Genus Parasystatus
on Mount Kenya
In the tussock grass alpine tundra of Mount Kenya
between 3200 and 4000 m, there are six described species
and at least one undescribed species of the genus
Parasystatus. These large beetles must be adapted to the
diurnal extremes of the climate, which has been
described as summer each day and winter each night.
Two of these species, P. elongates and one undescribed
species, have been studied in some detail as to their
adaptation to the nightly frost of that zone. P. elongates
spends its entire larval and pupal development inside a
tussock of the grass Festuca abyssinica, where it is not
affected by the nightly frost. As an adult beetle, it is
active by day, shielded from the intense solar radiation
by inflated elytra and a shiny, reflective outer cuticle. At
night, the beetle hides under vegetation to avoid the
worst of the frost; it has an ineffectively high supercool
ing point, but an effective freeze tolerance. The other
species of the same genus is active well into the night,
and protects itself with a much lower supercooling point,
but is freezing sensitive. (Cooling a liquid to below its
freezing point without phase transition; here pertaining
to the avoidance of ice formation due to the presence of
antifreeze substances and/or the absence of crystalliza
tion nuclei.) These two different physiological
adaptations to nightly frost within one genus indicate
that the two species have independently invaded the
alpine tundra, rather than having arisen through specia
tion in the alpine zone. Being flightless – a typical
adaptation to mountain top ecosystems – also rules out
invasion from another mountain
Tundra 449
The Giant Lobelia and Its Insect Commensals
on Kilimanjaro
Between 3000 and 4000 m on the slopes of Mount
Kilimanjaro, the giant lobelia (Lobelia deckenii) also has to
face the stress of nightly frost, which can be severe due to
parts of the Kilimanjaro alpine tundra being relatively
dry. The plant has evolved into a ball shaped rosette
consisting of a fleshy center surrounded by concave
spiky leaves, which are arranged in such a manner as to
trap rainwater. A single plant can contain, trapped in its
rosette, a compartmentalized mass of several liters of
water. This volume is large enough to prevent it from
freezing right to the middle in any one night. Indeed, the
center of the plant where the growing tip is located
maintains a very even temperature throughout the diur
nal cycle. Not surprisingly, this water mass of the lobelia
plants with its relatively even temperature has become
the breeding environment for a few species of insects with
aquatic larvae, the most abundant of which is a chirono
mid midge. The water in the lobelias also contains
microorganisms, which feed on decomposing debris and
are in turn the food for the insect larvae.
Global Warming and Other Anthropogenic
Effects
Extensive research in arctic and alpine regions including
ice core analysis, paleolimnology, palynology, and geo
morphology has provided a detailed picture of the climatic
history of these regions. This allows us to conclude that, as
well as the major changes at the end of the last ice age,
frequent climate oscillations have subsequently occurred
that caused major changes in tundra ecosystems.
Furthermore, there have been times when tundra types
existed that are no longer extant. The species complexes
that now exist consist of species that have been sufficiently
flexible and/or dispersible to have survived the climatic
and landscape oscillations of the past. However, this does
not necessarily bode well for the future of tundra ecosys
tems and species, as anthropogenic changes are certain to
be increasingly imposed on the Earth. Already, the most
likely reason for the extinction of most of the Pleistocene
arctic megafauna is a combination of climate change and
human hunting. The disappearance of the large herbivores
at that time caused a major switch in plant dominance on
tundra ecosystems from graminoids to mosses, with con
comitant changes in long term soil and peat formation.
We must expect similar major changes in the coming
century, associated with at least some extinctions.
Climate change will be severe and direct human effects
will also increase. Already, several species are declining
due to pollution and over hunting. Some of the most at
risk tundras (and associated endemic tundra species) will
be isolated alpine tundra systems on relatively low moun
tains, where climatic warming will cause the entire system
to be replaced with forest.
See also: Alpine Ecosystems and the High-Elevation
Treeline; Alpine Forest; Boreal Forest; Cycling and
Cycling Indices; Freshwater Lakes; Polar Terrestrial
Ecology; Steppes and Prairies.
Further Reading
Chapin FS and Korner C (eds.) (1995) Arctic and Alpine Biodiversity.
Patterns, Causes and Ecosystem Consequences, 332pp. Berlin:
Springer.
Coe MJ (1967) The Ecology of the Alpine Zone of Mount Kenya, 136pp.
The Hague: Junk.
Craeford RMM (ed.) (1997) Disturbance and Recovery in Arctic Lands,
621pp. Dordrecht: Kluwer Academic.
French HM and Williams P (2007) The Periglacial Environment, 478pp.
Toronto: Wiley.
Goulson D (2003) Bumblebees: Their Behavior and Ecology, 235pp.
Oxford: Oxford University Press.
Jones HG, Pomeroy JW, Walker DA, and Hoham RW (eds.) (2001)
Snow Ecology: An Interdisciplanary Examination of Snow Covered
Ecosystems, 378pp. Cambridge University Press.
Laws RM (ed.) (1984) Antarctic Ecology, vol. 1, 344pp. London:
Academic Press.
Mahaney WC (ed.) (1989) Quaternary and Environmental
Research on East African Mountains, 483pp. Rotterdam:
Balkema.
Pienitz R, Douglas MSV, and Smol JP (eds.) (2004) Long Term
Environmental Change in Arctic and Antarctic Lakes, 562pp.
Dordrecht: Springer.
Rosswall T and Heal OW (eds.) (1975) Ecological Bulletin, Vol. 20:
Structure and Function of Tundra Ecosystems, 450pp. Stockholm:
Swedish Natural Science Research Council.
Wielgolaski FE (ed.) (1997) Ecosystems of the World 3: Polar and Alpine
Tundra, 920pp. Amsterdam: Elsevier.
450
Upwelling Ecosystems
Upwelling Ecosystems
T R Anderson and M I Lucas, National Oceanography Centre, Southampton, UK
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Primary Production and Lower Trophic Levels
Fish and Higher Trophic Levels
Climatic Forcing
Further Reading
Introduction
Hemisphere, and left in the Southern Hemisphere. The
result is horizontal flow at the ocean surface in the so
called Ekman layer, typically tens of meters deep.
Upwelling occurs in areas where this flow diverges, the
Ekman flow or divergence, so that water displaced at the
surface must be replaced by deeper water from beneath.
Depending on the nature of this divergence, two major
types of upwelling systems can be distinguished.
First, coastal upwelling systems occur where the
Ekman layer is directed offshore resulting in flow diver
gence near the coast. Such systems tend to occur on the
eastern boundary of ocean basins, major examples being
the Canary, Benguela, Humboldt (Peru), and California
Current systems (Figure 1). Offshore Ekman flow in east
ern boundary current (EBC) systems is driven by local
equatorward winds associated with the pressure gradient
between the quasi stationary atmospheric high pressure
systems over the subtropical oceans relative to adjacent
continental low pressure atmospheric systems. Seasonal
north–south progressions of these high pressure systems
(poleward in spring, summer) cause increased upwelling
and nutrient supply that, along with increased day length
and light, drive latitudinal shifts in phytoplankton bio
mass and productivity. The other major coastal upwelling
system is the Somali Current, driven by seasonal mon
soon winds of the Arabian Sea. Coastal upwelling is often
enhanced by topographical features such as capes or can
yons where local upwelling cells form.
Throughout the world’s oceans, phytoplankton commu
nity structure and rates of primary production are
determined by the interplay between available light and
nutrient supply (NO3 , Si, PO24 , dissolved Fe) as well as by
grazing. Winds blowing over the ocean create a surface
mixed layer, the depth of which is of great importance for
production by phytoplankton. If mixing is vigorous, as is
often the case at high latitudes, then nutrients are plentiful
but plankton circulating within a mixed layer that may be
hundreds of meters deep are exposed to low average light
intensities. In contrast, mixing is inhibited in warm strati
fied waters such as those of the vast subtropical gyres that
cover 40% of the surface ocean, in which case light is
plentiful and limitation is instead by nutrients. The unique
physical circulation of upwelling systems leads to condi
tions that, to varying degrees, provide both light and
nutrients together in quantities that considerably exceed
rate limiting requirements for sustaining maximal growth
rates of phytoplankton. As a result, upwelling ecosystems
are among the most productive in the ocean.
Upwelling Circulation
The Coriolis effect, whereby the Earth’s rotation causes
moving bodies at its surface to be deflected, means that
wind driven ocean currents turn right in the Northern
CC
Ca
SC
EP
Hu
EA
Be
SO
Figure 1 Global map of major upwelling systems. Be, Benguela; Ca, Canary; CC, California Current; EA, Equatorial Atlantic; EP,
Equatorial Pacific; Hu, Humboldt; SC, Somali Current; SO, Southern Ocean.
Upwelling Ecosystems
Second, upwelling occurs in the open ocean, being
most marked where easterly trade winds give rise to
Ekman divergence north and south of the equator. The
resulting area of equatorial upwelling in the Pacific is
vast, extending westwards from the coast of South
America to beyond the international date line. A smaller
belt of upwelling occurs in the equatorial Atlantic. In the
Southern Ocean, the Antarctic Circumpolar Current con
tains another zonal upwelling region, most vigorous
between 50 and 60 S, driven by northerly Ekman flow
that is generated by the strongest prevailing westerly
winds in the 40–50 latitudes.
General Characteristics
Nutrients are present in high (e.g., NO3 , 35; Si, 30–60;
PO24 , 1–2 mmol l 1) concentrations in subsurface waters
of the global oceans. Upwelling brings them to the surface,
fertilizing the resident phytoplankton assemblage.
Stratification of the surface mixed layer between the
strongest upwelling pulses provides favorable light con
ditions for algae to grow and take up the nutrients at their
disposal. Resulting rates of primary production are often
among the highest seen in marine systems. Coastal upwel
ling systems, for example, occupy just 0.5% of the ocean
surface area, yet contribute to 2% of global marine pri
mary production. Supporting an abundance of higher
trophic orders such as fish, birds, seals, and whales, they
also contain some of the world’s major fin fisheries.
Intermittence is a key feature of upwelling systems.
Upwelling intensity is seasonally episodic in systems such
as the Canary Current, Benguela, and Somali systems,
whereas in others such as the Humboldt and Southern
Ocean, upwelling is semicontinuous all year round. In all
systems, wind strength varies on shorter timescales of
days to weeks, leading to periods of strong and weak
upwelling, or times when upwelling ceases altogether.
Organisms must be able to tolerate these changes in
upwelling intensity and the resulting impact on nutrient
supply and spatiotemporal variation in food resources, as
well as variations that may occur from year to year and on
longer timescales. In addition, they face the prospect of
either themselves, or their reproductive products, being
swept away in the Ekman layer toward less favorable
habitats. A key feature of these organisms, including
phytoplankton, zooplankton, and fish, is that their life
histories and behavior are specifically geared toward
maintaining populations in the regions of the upwelling
centers.
Understanding the ecosystem structure and function
ing of upwelling systems, and in particular how they are
influenced by climate variability, is increasingly recog
nized as essential to the management of sustainable
fishery resources.
451
Primary Production and Lower Trophic
Levels
Primary Production
The EBC systems provide a highly favorable combina
tion of light and nutrient supply for primary production
by virtue of a strongly shoaling pycnocline (the density
gradient that signifies the base of the mixed layer)
toward the coast and a relatively shallow (<500 m)
shelf environment. A near surface pycnocline has the
dual effect of facilitating the injection of nutrients into
surface waters and maintaining phytoplankton in a shal
low (usually <50 m) well lit euphotic layer environment.
Shallow shelf sediments augment the nutrient concentra
tion of deeper upwelling source waters. As the four
major EBC systems lie predominantly in mid latitudes
(40 N/S to 10 N/S), insolation rates are seasonally
high, providing both the light and necessary surface
warming and stratification to optimally drive photosyn
thetic carbon fixation supported by a nutrient replete
environment. Taken together, they have a combined
productivity estimated to be 1 Gt C yr 1. Chlorophyll
concentrations typically exceed 2 mg m 3 but can reach
up to 50 mg m 3 locally where intense dinoflagellate
blooms are present. Highest production rates are consis
tently found in the Humboldt system (2–6 g C m 2 d 1)
due to a higher average irradiance and a less fluctuating
nutrient environment associated with relatively consis
tent upwelling.
A key feature of primary production in upwelling
ecosystems is that it is fuelled by large amounts of NO3
that is ‘new’ (i.e., allochthonous) to the euphotic zone.
Phytoplankton production based on nitrate uptake is
therefore termed ‘new production’. Nitrate arises almost
entirely from remineralization of organic matter below
the pycnocline, notably from dead phytoplankton and
other material that had earlier ‘rained’ down from surface
waters. In contrast, ‘regenerated production’ is based on
nitrogen (NHþ
4 , urea, and dissolved organic nitrogen)
excreted by organisms within (i.e., autochthonous) the
euphotic zone. The relative importance of new produc
tion is often stated by expressing it as a fraction of total
phytoplankton production (i.e., the sum of new and
regenerated production), this fraction being known
as the f ratio. Values for the f ratio are usually high
(0.5–0.7) in most EBC upwelling ecosystems, although
in seasonally pulsed upwelling systems such as the
Southern Benguela, the yearly averaged f ratio is lower
(0.3). High rates of new production make available
carbon and nutrients for transfer to higher trophic levels
and are the fundamental reason why EBC systems can
sustain productive fisheries. They also provide the poten
tial for large downward fluxes of sinking particles from
the euphotic zone in the event that phytoplankton are
452
Upwelling Ecosystems
inefficiently grazed. In contrast, systems based on regen
erated production gradually run downhill unless there are
new inputs of nitrogen because nutrients are never
recycled with 100% efficiency.
0.01
10.0
0.1
1.0
Chlorophyll (mg m–3)
Phytoplankton Community Structure
A characteristic succession is seen in the composition of
phytoplankton communities of coastal upwelling ecosys
tems. This succession is driven by changes in the nutrient
and light environment, linked closely with upwelling
frequency and the three dimensional (3D) circulation of
water as it flows away from the upwelling centers. Large
individual (20–200 mm), colonial, and chain forming dia
toms of up to 500 mm in length proliferate as newly
upwelled water arrives and stabilizes in the sunlit surface
layer. Analogous to weeds, these algae grow quickly
because of intrinsically fast growth rates and an ability
to take up nutrients rapidly, provided that concentrations
remain sufficiently high. Cell division rates of 2–4 d 1
quickly result in population growth that outstrips
zooplankton herbivory, leading to extensive diatom
dominated blooms. The accumulating chlorophyll bio
mass can exceed 6 mg m 3 in just a few days, high
enough to be easily visible in ocean color satellite imagery
(Figure 2).
Most diatom species within EBC systems are adapted
to avoid lateral dispersal in surface currents away from
the upwelling centers. Many species trigger a resting stage
in their life cycle in response to diminishing nutrient
availability. Spores are formed that sink rapidly and
become entrained into deeper shelf edge waters and sur
face sediments. Sinking of vegetative cells, or chain
formation that increases sinking rates, provide alternative
strategies to counteract dispersal. Spores and physiologi
cally inactive diatoms remain within the sediment–water
interface layer and await the next upwelling event that
will entrain them back into near shore and nutrient rich
sunlit surface waters, so initiating another bloom event.
The construction of silica tests and spiny frustules from
dissolved silicate, along with their large size and ability to
form chains and colonies, offer initial protection from
herbivorous zooplankton such as copepods. The efficacy
of grazers is further weakened in pulsed upwelling sys
tems that do not settle into steady state. As initial
exponential growth rates of diatoms are much faster
(hours, days) than those of herbivorous copepod consu
mers (weeks), episodic and short term upwelling events
(days) produce a ‘mismatch’ between phytoplankton and
zooplankton, the former breaking free from top down
grazer control that might otherwise prevent blooms
from occurring. Nevertheless, the grazing that does
occur permits carbon to be efficiently transferred to
pelagic fish in a short two step food chain (diatoms )
mesozooplankton ) pelagic fish).
30°S
South Africa
35°S
15°E
20°E
Figure 2 Satellite image of chlorophyll biomass in the Southern
Benguela region. Note high chlorophyll concentrations inshore
where upwelling of nutrient rich water is strongest, and also
offshore filaments in the chlorophyll signal which weakens
offshore as nutrients become exhausted. Courtesy Stewart
Bernard, Univ. Cape Town.
A shift in phytoplankton community structure occurs as
nutrients become depleted in the well stratified surface
waters downstream of coastal upwelling centers. Diatoms
give way to smaller cells such as nanoplanktonic phytofla
gellates (2–20 mm) as well as other smaller picoautotrophs
(<2 mm) that do not require Si and are better able to sca
venge nutrients at low concentrations because of their high
surface area to volume ratio. As nutrients in the surface
layers are scarce, these small phytoplankton primarily
occupy the thermocline where nutrients diffuse slowly
from below. A deep chlorophyll maximum (DCM) develops
that involves a delicate tradeoff between maximizing nutri
ent availability, but having sufficient light in a near light
limited environment. Microzooplankton grazers keep the
numbers and biomass of these small phytoplankton in
check (<0.5–1 mg chl a l 1). Within this ‘microbial loop’, par
ticulate organic nitrogen (PON) is efficiently recycled via
microzooplankton and bacteria into NHþ
4 and urea to sup
port further phytoplankton growth. As nitrogen
remineralization is usually balanced by the rapid uptake of
such regenerated nitrogen by phytoplankton, concen
trations of these nutrients remain low (<0.5–1 mmol l 1).
Upwelling Ecosystems
Paradoxically therefore, grazing pressure is essential to sup
port further algal growth. Carbon is only inefficiently
transferred along an extended food chain (pico , nanoplankton
) microzooplankton ) mesozooplankton ) fish) because,
at each step, something approaching 90% of the transferred
carbon is lost through respiration.
453
Offshore hypoxia can have profound effects on the
near shore intertidal environment. When gentle upwel
ling begins, low oxygen water is driven into the near
shore and intertidal zone, killing all before it except those
most resistant to hypoxia. In the Benguela system, cray
fish ‘walkouts’ by animals fleeing from low O2 water can
leave thousands of tons of crayfish stranded on the
beaches in sheltered embayments.
Oxygen Depletion
Dead and decaying material is a feature of all ecosystems,
those of upwelling areas being no exception. Detrital
particles are produced in abundance, either as senescent
phytoplankton or as zooplankton fecal material. A ‘rain’ of
sinking particulate organic material (‘marine snow’) is
exported from the euphotic zone and decomposed either
in mid water by heterotrophic bacteria, or by both
benthic organisms and bacteria on the seafloor. Large
quantities of oxygen are consumed, creating an oxygen
minimum zone (OMZ) in the sediments and the overlying
water. The resulting oxygen concentrations of <0.5 ml l 1
are inhospitable to many animals, pelagic and benthic
alike.
Coastal upwelling systems are particularly susceptible
to hypoxic (<0.5 ml l 1) or anoxic (near zero O2) events
because of the high rates of diatom dominated phyto
plankton productivity that quite suddenly become
nutrient limited and therefore senescent. The oxycline
may often extend to near the surface (<50 m). Most zoo
plankton that can actively migrate often do so in order to
maintain their position in the oxygenated waters near the
surface. Others, such as Eucalanus inermis in the Humboldt
system, are able to withstand low oxygen concentrations,
and indeed are known to congregate in the OMZ, perhaps
exploiting it as a refuge from predation. In similar fashion,
juvenile hake (Merlucius capensis and M. paradoxus) in the
northern Benguela system off Namibia are known to
exploit the OMZ as a refuge from their cannabilistic
parents!
Oxygen consumption in upwelling areas leads to
extensive seafloor habitats subject to permanent hypoxia.
Diversity is low, but those animals that tolerate low oxy
gen conditions are abundant. Calcareous foraminiferans,
nematodes, and annelids utilize the influx of organic
material from above, but rely on anaerobic metabolism
to do so. Chemoautotrophic bacteria use NO3 and sulfur
as terminal electron acceptors instead of O2, first stripping
the anoxic water column of NO3 (denitrification) before
specific anaerobic sulfur and sulfate reducing bacteria
release sulfurous and foul smelling H2S into the water
column creating so called ‘black tides’ and ‘sulfur erup
tions’. However, it appears that some nitrogen losses
previously ascribed to denitrification should instead be
linked to the process of ammonium oxidation ‘annamox’,
first described in Dutch sewerage works.
Zooplankton
Great variety is seen in the zooplankton of upwelling
ecosystems. Smallest are the microzooplankton, including
ciliates, heterotrophic dinoflagellates, and flagellates (typical
size 2–5 mm) that efficiently graze the smallest phytoplank
ton. Reproducing by cell division, their high growth rates
(e.g., 1.0 d 1) are similar to those of their algal prey such that
their grazing is sufficient to prevent small phytoplankton
cells from blooming. Microzooplankton do not fit the
general paradigm that organisms eat prey items significantly
smaller than themselves. Instead, they have evolved various
specialized feeding mechanisms including direct engulf
ment, tube feeding in which a feeding tube, the peduncle,
pierces prey which then has its insides sucked out, and
pallium feeding, where a feeding veil envelops and digests
prey in situ. Prey items as large as, or larger than, the
microzooplankter’s own body size can be consumed using
these adaptations. It has been suggested, for example, that
heterotrophic dinoflagellates are able to compete with cope
pods for diatom prey, although the extent to which this
competition operates in marine ecosystems is as yet poorly
known.
It is the larger mesozooplankton (0.2–2 mm), nota
bly copepods and to a lesser extent euphausiids, that are
the major grazers of diatoms and which form the main
trophic link with fish and other higher trophic levels.
Phytoplankton are captured by filter feeding or by selec
tive particle capture (raptorial feeding) based on size
and/or palatability. Calanus is the dominant copepod
genus in upwelling ecosystems. Although not reprodu
cing as fast as microzooplankton, it may achieve as many
as ten generations per year, each with its own life cycle
of eggs, nauplii, copepodites, and adults. When food
conditions are favorable, fecundity is high and egg pro
duction is rapid. Upon hatching, the planktivorous
juvenile stages are swept along in the Ekman layer and
will starve if they do not encounter suitably dense
patches of appropriately sized food particles. For adults,
survival is enhanced by the storage of energy reserves in
the form of lipids. Given their size, copepods are unable
to maintain their position within upwelling systems by
swimming against lateral advection that is often offshore.
This problem is overcome by diel vertical migration.
Offshore surface Ekman flow is balanced by deeper
shoreward flow onto the shelf. By migrating into this
454
Upwelling Ecosystems
deeper layer by day, copepods utilize the natural circu
lation pattern to maintain their inshore position where
food resources are richest.
Euphausiids are also a significant component of the
zooplankton community in upwelling ecosystems, for
example, Euphasia lucens in the Benguela system. They
are much larger (1–2 cm) than copepods and have a
longer life span of about a year. This longevity, along
with an omnivorous diet, means that euphausiids are
better able to cope with the fluctuating food conditions
of upwelling ecosystems than are copepods. Nevertheless,
physical transport away from upwelling centers remains a
problem, and these animals also employ diel vertical
migration into the subsurface countercurrent to maintain
their position in the flow field. Their larger size makes
euphausiids a key prey item for larger zooplankton con
sumers, including baleen whales that are often temporary
residents of upwelling systems.
Open-Ocean Upwelling Systems
The general principles and characteristics that govern pro
ductivity in EBC regions apply also to the major open
ocean upwelling systems (Equatorial Pacific, Equatorial
Atlantic, and Southern Ocean). There are nevertheless key
differences, notably that upwelling strength tends to be
lower and there is no influence of the seabed (e.g., in
supplying iron) on euphotic zone processes, now that it is
3000–4000 m beneath the ocean surface.
The Equatorial Pacific is a vast upwelling system, as
well as being a good example of a so called high nutrient
low chlorophyll (HNLC) ecosystem. Phytoplankton
biomass is generally low and relatively constant
(0.2–0.4 mg chl a m 3) which, along with low productiv
ity of 0.1–0.5 g C m 2 d 1, occurs despite the presence
of sufficient macronutrients (NO3 , Si, PO42 ) and light.
Iron, however, is in short supply. This micronutrient is
needed by phytoplankton to harvest light using their
photosynthetic machinery (photosystems I and II), as
well as by the enzymes nitrate and nitrite reductase to
reduce NO3 within cells to NH4þ. Without a sedimen
tary source, aeolian supply is the primary source of Fe to
the open ocean. However, most aeolian dust supply is
from the Saharan desert, far distant from the Equatorial
Pacific. The resulting shortage of iron impacts most
severely on large cells, notably diatoms, because of their
inability to compete with smaller phytoplankton at low
nutrient concentrations. In the western basin, phytoplank
ton biomass is dominated by small solitary picoplanktonic
cells (0.2–2 mm) within a DCM comprising prochloro
phytes, Synechococcus, and small eukaryotes. These cells
utilize what little iron supply there is from the waters
upwelled from below, starving the surface ocean of this
element. Diatoms are more abundant (6%) to the east of
140 W where deep nutrient rich upwelling outcrops at
the surface but, nevertheless, the overall biomass is still
picoplankton dominated. Grazing by microzooplankton
keeps phytoplankton stocks in check, but small natural
enhancements of iron that occur in the Equatorial Pacific
in response to the passage of tropical instability waves
promote transient increases in primary production.
The upwelling region at the Antarctic Polar Front is
another HNLC system with low Fe concentrations.
Chlorophyll biomass is typically 0.5 mg m 3 in the aus
tral summer with a productivity of 0.5–1 g C m 2 d 1.
Unlike the Equatorial Pacific, however, winter gales drive
deep mixing that entrains nutrients, including Fe, into
surface waters. This is sufficient to initiate short diatom
dominated blooms in the early spring (September,
October) as the light environment improves. Iron limitation
throughout the rest of the year opens the way for a more
typical HNLC community of pico and nanoplanktonic
phytoflagellates that are microzooplankton controlled.
Populations of the prymnesiophyte Phaeocystis antarctica
may also develop, an organism that exists both as solitary
cells and mucilaginous colonies, and which is the main
producer of volatile organic sulfur (dimethyl sulphide,
DMS) in the region.
The Equatorial Atlantic is unique among open ocean
upwelling systems both in terms of its hydrography and
because of its close proximity to aeolian dust sources from
the Sahara meaning that productivity is far less Fe limited
than in other open ocean systems. Between June and
January, water flowing from the Amazon basin floods
eastwards (the North Equatorial Counter Current) across
the northern margin of the equatorial upwelling region.
Stripped of nutrients as it crosses the Amazonian shelf,
this fresher, buoyant water forms a layer 40 m deep that
caps the nutrient rich water below and also limits light
penetration. A DCM forms at the juncture of these two
water types, the low light intensities being particularly
well exploited by the cyanobacterium Prochlorococcus. The
nutrient depleted waters above are home to a separate
community in which nitrogen fixers such as
Trichodesmium utilize atmospheric nitrogen as a nutrient
source. The aeolian flux of dust plays an important role
since nitrogen fixers have a particularly high requirement
for Fe.
Between February and May, the Amazonian outflow
diverts northwards toward the Caribbean. Phytoplankton
now find themselves 40 m closer to the surface in a
higher light environment and productivity increases but,
as the rate of upwelling is weak, the upward flux of
nutrients is insufficient to support diatom blooms except
in the eastern basin near the African coast.
In addition to the major upwelling systems described
above, the upper ocean contains numerous mesoscale
eddies whirling current systems analogous to weather
systems in the atmosphere but only about a tenth of the
size (tens instead of hundreds of kilometers across).
Upwelling Ecosystems
Produced through the conversion of potential energy to
kinetic energy as part of the ocean’s annual energy
cycle, both cyclonic and anticyclonic eddies (depending
on the vertical structure of the water column) can
result in the localized doming of isopycnals (constant
density surfaces) and upwelling of nutrient rich waters
into the euphotic zone as they form. Eddies themselves
then decay as they release their potential energy over
periods of weeks to months, vertical motions both
upward and downward occurring on their periphery
during this time. Primary production is generally sti
mulated through nutrient enrichment. As in other
upwelling systems, regions of higher nutrients and shal
lower mixed layer depth associated with eddies tend to
promote the growth of larger phytoplankton cells such
as diatoms, their concentrations typically being higher
within eddies than in surrounding waters. Ubiquitous in
nature, eddies provide a significant vertical transport
mechanism for nutrients throughout much of the
world’s oceans.
Fish and Higher Trophic Levels
The EBC upwelling ecosystems of the world support
major commercial fisheries based on the shoals of sardine,
anchovy, and mackerel that thrive on the abundance of
phytoplankton and zooplankton food. In the Humboldt
system alone, for example, catches have been around 12
million tons in peak years, although this decreases by
>50% during unfavorable conditions. Indeed, stocks of
different fish species have been highly variable over the
years, suggesting a remarkable responsiveness in ecosys
tem structure to changing conditions. Understanding the
links between fish, lower and higher trophic levels, and
environment is essential to ensuring the sustainable man
agement of these important fish resources.
Small Pelagic Fish
The food chain of upwelling systems embraces phyto
plankton and zooplankton at its base, linking to small
pelagic fish which are in turn consumed by higher pre
dators such as piscivorous fish, birds, and seals (Figure 3).
A curious aspect of this trophic network is that there are
many species at low (phytoplankton, zooplankton) and at
high trophic levels, but only a few species of small pelagic
fish in between. Indeed, the fish biomass of coastal upwel
ling systems is typically dominated by either a single
species of sardine (Sardinops) or a single species of
anchovy (Engraulis) at any one time.
Although food resources are generally favorable, the 3D
circulation makes upwelling systems a hazardous environ
ment for fish. Losses of eggs and juvenile stages may occur
due to offshore transport or because of starvation when
being carried by currents from the spawning to nursery
areas. Spawning grounds are therefore often strategically
Higher predators
Piscivorous fish
Birds
Mammals
Man
Regime shifts
Sardine
Anchovy
Large zooplankton (>200 μm)
Small zooplankton (<200 μm)
Cyclopoid copepods
Microzooplankton
Calanoid copepods
Euphausiids
Phytoplankton
Picoautotrophs
Decreasing
Flagellates
455
Dinoflagellates
Upward nutrient flux + Fe
Diatoms
Increasing
Figure 3 Idealized flow diagram for an upwelling ecosystem food web. Dashed arrows indicate weak flows.
456
Upwelling Ecosystems
100 m
Orange River Mouth
Spawning area
200 m
Transport
30
Losses
Latitude (°S)
Nursery area
Recruitment
32
Port Elizabeth
34
Cape Town
Jet Current
EAB
Agulhas
Current
CAB
WAB
36
Indian
Ocean
Atlantic
Ocean
16
18
20
22
24
26
28
Longitude (°E)
Figure 4 Map of the Southern Benguela off South Africa showing the locations of small pelagic fish spawning and nursery grounds
and transport and loss processes that impact on eggs and larvae. WAB, CAB, and EAB indicate the Western, Central, and Eastern
Agulhas Banks, respectively. Redrawn from Lehodey P, Alheit J, Barange M, et al. (2006) Climate variability, fish and fisheries. Journal of
Climate 19: 5009–5030. ª Copyright 2006 American Meteorological Society (AMS).
positioned in quieter areas surrounding the upwelling
centers such as downstream of capes in sheltered embay
ments. The result is a complex network of spawning
grounds, transport pathways, and migration patterns, a
typical example being the Benguela system (Figure 4).
Anchovy spawn on the Western Agulhas Bank in spring
and summer (with a maximum in November), while sardine
have a longer spawning season in the same area, peaking in
both October and March. Once fertilized, eggs and larvae
drift northwards in the Benguela ‘Jet’. Larvae feed selec
tively on small particles and juvenile fish recruitment occurs
at several locations north of St. Helena Bay on the West
Coast. There, anchovy recruits feed primarily on larger
zooplankton (copepods) as they slowly migrate southwards,
returning as 1 year old adults to the Agulhas Bank to spawn
in the following austral spring/summer.
The survival of small pelagic fish is thus determined to
a large degree by direct physical factors such as circula
tion patterns and the intensity and duration of upwelling
that simultaneously control egg and larval survival,
recruitment success, and food supply. Population control
is therefore neither exclusively ‘bottom up’ via primary
producers nor ‘top down’ by higher predators. Instead it is
from the ‘waist’, both up and down, the so called ‘wasp
waist’ hypothesis. Small pelagic fish provide higher tropic
levels such as birds and seals with food, while at the same
time keeping phytoplankton and zooplankton numbers in
check. As a result, ecosystem functioning as a whole may
be remarkably sensitive to fluctuations in pelagic fish
numbers. Direct environmental forcing or commercial
fishery exploitation of ‘wasp waist’ populations may
cause disruption to these ecosystems by undermining
the stability of the entire food web.
Bottom up and top down controls of small pelagic
fish populations are nevertheless by no means unim
portant. Sardines and anchovy, for example, have
different feeding strategies. Sardines are mostly indis
criminate filter feeders on phytoplankton and smaller
zooplankton, including small cyclopoid copepods,
whereas anchovy use biting behavior to selectively
ingest individual particles such as larger (2 mm) cope
pods and euphausiids. Strong upwelling should
therefore favor anchovy by promoting the diatom
growth that supports larger zooplankton. In contrast,
the more nutrient depleted waters present during per
iods of weaker upwelling favor smaller phytoplankton
and consequently smaller zooplankton that are pre
ferred by filter feeding sardines.
Variability in small pelagic fish populations has con
sequences for their predators. Evidence from the
Benguela region shows that during periods of pelagic
fish abundance populations of piscivorous fish (e.g.,
snoek, hake), seals and birds (gannets, cormorants) gen
erally increase and, in doing so, begin to exert a stronger
top down control on the small pelagic fish. This in turn
relaxes anchovy predation pressure on copepods and so
mesozooplankton numbers recover, in turn exerting a
higher grazing pressure on phytoplankton. Not only
does top down predator control on small pelagics equal
or exceed that by commercial fishermen, it can
Upwelling Ecosystems
substantially shape community structure right down to
the level of primary producers.
457
Optimal
environmental
window
High primary production fuelled by new nutrients
undoubtedly contributes to the prodigious fish produc
tion of coastal upwelling systems. Back in 1969, John
Ryther proposed that high fish yield should be expected
where phytoplankton cells are large, or exist as colonies
or chains, thereby leading to only one or two trophic links
from primary producers to fish. The greater number of
trophic links stemming from smaller phytoplankton cells
should lead instead to greater respiration losses and recy
cling of organic matter.
Fish catches vary markedly between EBC systems,
typical values being 0.05%, 0.09%, and 0.16% of primary
production for the Canary, Benguela, and Humboldt
Current systems respectively. Much of this variation
may be due to differences in upwelling intensity and
frequency that impact on lower trophic levels and fish
recruitment, although intensity of fishing may also play a
part. The strongly seasonal and pulsed nature of upwel
ling in the Canary and Benguela systems, for example,
leads to temporal mismatches between primary producers
and copepods, depressing their fecundity and therefore
population size, so reducing the food supply for fish. In
contrast the Humboldt system experiences less variation
in the intensity of upwelling, leading to tighter coupling
between phytoplankton and copepods, greater zooplank
ton production, and ultimately higher fish production.
The low fish catch per unit primary production of the
Canary system relative to that of the Benguela is due to its
narrow continental shelf (20 km; the Benguela’s is
85 km), such that proportionately more primary pro
duction may be advected offshore away from the main
zones of fish production.
The impact of environment on the reproductive success
of fish in upwelling ecosystems can be thought of in terms
of a fundamental triad of processes: enrichment (nutrient
supply for primary production), concentration processes
(convergence, water column stability), and retention
(within favorable habitat). Acting at the base of the food
chain, nutrient enrichment via upwelled water fertilizes
primary producers, although excessive wind may deepen
the surface mixed layer, leading to light limitation.
Upwelling also stimulates small scale turbulence, increas
ing encounter rates between both zooplankton and fish
larvae and their prey. On a larger scale, convergence pro
motes food particle aggregation, but divergent upwelling
flow will tend to dissipate particles offshore. Based on these
pros and cons, an optimal level of upwelling intensity can
be defined, the ‘optimal environmental window’ (OEW),
that maximizes fish yield (Figure 5). When on the left side
of the OEW (too little wind) upwelling is weak and
Recruitment
Fish Production
Weak
Moderate
Strong
Upwelling intensity
Figure 5 Optimal environmental window for fish recruitment.
Reproduced from Cury P and Roy C (1989) Optimal
environmental window and pelagic fish recruitment success in
upwelling areas. Canadian Journal of Fisheries and Aquatic
Sciences 46: 670–680.
primary production, and hence also food for fish, is
restricted by insufficient nutrient supply. On the other
hand too much upwelling (right side of the OEW) leads
to dispersal of organisms away from upwelling centers and
provokes light limitation in phytoplankton as they are
mixed deeper into the water column because stratification
is not established.
Higher Trophic Levels
The abundance of zooplankton and small pelagic fish in
coastal upwelling systems provides food for a range of
higher trophic levels including piscivorous fish, seabirds,
pinnipeds, and cetaceans. Predatory fish such as horse
mackerel and deep water hake are themselves important
fishery resources, the latter being caught by mid and
deep water trawling. Another economically viable pro
duct of upwelling systems, particularly in the Humboldt
and Benguela, is the production of bird droppings, guano,
which is prized as a fertilizer because of its high N and P
content. The so called ‘guano birds’ such as the guanay
cormorant, Peruvian booby, Chilean pelican, Cape cor
morant, and Cape gannet are part of resident seabird
populations that breed along the coast and on adjacent
islands feeding on small pelagic fish such as anchovies and
sardines. Both the Benguela and Humboldt systems also
support populations of small sized penguins. The African
Penguin (Spheniscus demersus) extends from central
Namibia to Algoa Bay on the south coast of South
Africa. Its population has dwindled from more than one
million in 1900 to about 200 000 now, feeding mainly on a
diet of pelagic schooling fish (anchovy, sardine, redeye).
Reasons for the declining population are ecologically
complex, but include competition with commercial fish
eries for food, habitat degradation because of removal of
Upwelling Ecosystems
Benguela off Namibia, for example, is now thought to
exceed that of commercially important fish stocks. Once
established, this regime shift may be difficult to reverse
because jellyfish are predatory on fish eggs and juveniles.
Climatic Forcing
Changes have occurred in the dominant fish species of
upwelling ecosystems, from year to year, over decades
and indeed centuries. Worldwide, populations of anchovy
and sardine have exhibited ‘flip flops’ in which one spe
cies is replaced by the other. Fishing is one factor that
may influence these changes. Comparison of the fish
catches of different upwelling systems, however, reveals
a remarkable synchronicity in their behavior (Figure 6)
suggesting a climate linkage via global ‘teleconnections’.
Humboldt
California
Benguela
1.0
Sardine
0.8
Normalized catch
guano from islands that they burrow into for nesting,
pollution (oiling), and predation by seals. The
Humboldt Penguin (Spheniscus humboldti) breeds mainly
from 5 to 33 S along the Peruvian and Chilean coast,
with another small colony at 42 S. Like the African
Penguin, it also feeds on small pelagic fish, and its popu
lation has declined to around 30 000 for similar reasons.
Because of the episodic nature of upwelling, higher
predators must endure large seasonal or interannual fluc
tuations in prey availability, for example, in response to El
Niño events (see below). Resident seabird and pinniped
populations are particularly susceptible. In catastrophic
instances, starvation may cause adult seabirds to die,
although more often food scarcity affects breeding success
by decreasing the proportion of adults breeding and the
growth rate of the hatchlings. In similar fashion, there is
often a high incidence of seal pup mortality during food
shortages, the adult females being unable to provide
enough milk for their survival. Seabirds and pinnipeds
are able to employ various strategies to compensate when
food resources are in short supply including increasing
the time and/or distance spent foraging for offspring,
delaying reproduction until such time that food becomes
available, or ceasing breeding efforts altogether. Others
may target alternative food resources, such as squid, or
migrate to other regions within the system where food
resources are more plentiful.
Many migratory species are attracted to the high pro
ductivity of upwelling systems. Blue whales, for example,
feed on dense swarms of euphausiids that exist in the
California Current system. Many birds that nest else
where also benefit from the abundance of food in
upwelling areas. The California Current system, for
example, is visited by sooty shearwaters, which breed off
South America, and red necked phalaropes which nest in
the Arctic. Arctic terns migrate to the Southern
Hemisphere in austral winter, feeding in the Benguela
and Southern Ocean upwelling systems.
The catch of pelagic anchovy and sardine by the pre
dators described above is generally considered to surpass
that by commercial purse seine fishermen, even in the
Humboldt which is heavily exploited. Seals in particular
are unpopular competitors, not least because they cause
damage to nets and generally interfere with fishing opera
tions. Nevertheless, the potential consequences of
overfishing in upwelling ecosystems should not be under
estimated. Over the last few decades the fishing industry
has progressively concentrated on species at relatively
low trophic levels, with emphasis on small pelagic species
such as sardine and anchovy with decreased catches of
predatory fish such as hake and horse mackerel. An appar
ent consequence of this ‘fishing down marine food webs’
has been a decline in pelagic fish and a proliferation of
jellyfish that occupy the vacant niche, the two utilizing
the same food resources. Jellyfish biomass in the northern
0.6
0.4
0.2
0.0
1950
1960
1970
1980
1990
1970
Year
1980
1990
1.0
Anchovy
0.8
Normalized catch
458
0.6
0.4
0.2
0.0
1950
1960
Figure 6 Comparison of sardine and anchovy catches in the
Humboldt, California, and Benguela systems. Data are
normalized to maximum catch (million tons): sardine: 5.62,
Humboldt; 0.29, California; 1.51, Benguela; anchovy: 12.9,
Humboldt; 0.32, California; 0.97, Benguela. Reproduced from
Schwartzlose RA, Alheit J, Bakun A, et al. (1999) Worldwide
large-scale fluctuations of sardine and anchovy populations.
South African Journal of Marine Science 21: 289-347.
Upwelling Ecosystems
The implication is that fish populations are driven pri
marily by natural climate variability and its influence on
ecosystem structure and recruitment success. Short term
events such as El Niño cause calamitous declines in fish
stocks that lead to hardship for wildlife such as bird and
seal populations, and of course fishermen. Superimposed
on this short term variability are longer term trends that
occur in response to factors such as climate change.
Understanding these variations and their causes is crucial
to the maintenance of sustainable fisheries in upwelling
areas.
El Niño
The El Niño Southern Oscillation (ENSO) is the most
important example of a relatively short term impact of
climatic forcing on upwelling ecosystems, with a typical
periodicity of 3–5 years. In normal (La Niña) years, easterly
trade winds blow across the surface of the equatorial Pacific
from Peru/Chile to Indonesia creating the general divergent
open ocean upwelling that occurs across the eastern half of
the equatorial Pacific. This process sets up a surface tem
perature gradient of <20 C in the east to >30 C in the
west, resulting in a shallow thermocline (20 m) in the east,
but a much deeper one (80 m) in the west. At its eastern
end, along shore winds off Peru and Chile drive the coastal
upwelling of the Humboldt Current System. El Niño occurs
when the easterly trade winds lose intensity, allowing warm
water from Indonesia and eastern Australia to flood east
wards across the Pacific, ‘capping’ the deeper nutrient rich
waters that lie below (Figure 7). The coastal winds that
drive upwelling in the Humboldt System possess insufficient
energy to erode and mix this stratified surface layer.
(a)
459
Upwelling continues during El Niño but the water
arriving at the surface is depleted in nutrients with drastic
consequences for marine life. Diatom blooms are sup
pressed with a dramatic shift to a community structure
dominated by the small cells of the microbial loop. Fish
stocks collapse in response to the low availability of food,
the starvation of adults and/or larvae leading to recruit
ment failure. High mortality rates are also seen in top
predators. Major ENSO events occurred in 1972 and
1976, as well as in later years, economic disaster following
in their wake. For example, revenue losses to Chile and
Peru resulting from decimated fish stocks were about $8
billion for the 1997–98 ENSO event.
Further afield, the effects of ENSO events are felt
throughout the Southern Hemisphere, and indeed the
globe. The warm waters of El Niño in turn affect atmos
pheric circulation, the resulting teleconnections instigating
changes in other upwelling systems. For example, so called
‘Benguela Niño’ events occur about 6 months after the
onset of activities in the Pacific. During these events, the
Angola Benguela front moves southwards by several hun
dred kilometres, bringing low oxygen warm water into the
Namibian upwelling region that results in a southward
displacement of pelagic fish stocks.
Long-Term Climate Variability
Having high fecundity, small pelagic fish are able to
recover from events such as El Niño within a year or
two. Yet the observed anchovy–sardine flip flops persist
over many years suggesting that climatic factors operating
on longer timescales play an important role in structuring
coastal upwelling ecosystems and fish stocks (Figure 6).
(b)
Drought in
Australia and
southeast Asia
Surface winds
blow westward
Winds weaken,
causing updrafts
and storms
Equator
Equator
Australia
Warm waters
South
America
pushed westward
Warm water
e
lin
oc
erm
Th
Cold water
Warm water
South
Australia flow stopped
America
or reversed
Warm water deepens off
South America
Warm water
cline
rmo
The
Cold water
Figure 7 Atmospheric and ocean circulation patterns associated with La Niña (a) and El Niño (b). Reproduce from SEPM Photo CD-5,
Oceanography Series (edited by Peter A. Scholle), with permission from Society for Sedimentary Geology (SEPM).
460
Upwelling Ecosystems
Anchovy were the dominant fish species in the
Humboldt Current System until the mid 1970s. Catches
peaked at about 12.9 million tons in 1970, but were
followed by a severe decline that may have been precipi
tated by the major El Niño of 1972. Recovery of the
anchovy stock did not occur until the mid 1980s, sardine
being dominant during the interim period. The variability
of fish catches appears to have followed cycles of around
55–65 years over the last century. Analysis of atmospheric
circulation patterns (e.g., the ‘atmospheric circulation
index’, ACI) reveals that the dominant direction of air
masses has also changed on similar timescales. Fish scales
preserved in anoxic and undisturbed shelf sediments off
California and off Namibia reveal 50–70 year cycles of
anchovy and sardine abundance that are linked to changes
in sea surface temperature over the last 1600 years. In the
Humboldt system, regime shifts appear to correlate with
lasting periods of warm or cold temperature anomalies
related to the approach and retreat of warm subtropical
water toward the coasts of Peru and Chile. Sardine are
favored during periods of warm water intrusion (1970–85)
whereas the anchovy fishery prospers during periods
when temperatures remain relatively cool (1950–70,
1985 to present). Resolving the underlying, probably
basin scale, physical processes that lead to such
patterns, along with teleconnections linking different
upwelling systems, remains a priority for scientific
investigation.
Changes in local atmospheric pressure gradients with
global warming might be expected to increase upwelling
frequency and intensity, with accompanying changes in
the structure and function of ecosystems. Nutrient concen
trations have for example increased in the Benguela region
over recent decades, suggesting an increase in upwelling.
There has at the same time been a shift from sardine prior
to the mid 1960s, to anchovy in subsequent years. The
recent dominance of anchovy has impacted on the zoo
plankton population by selective predation pressure on
larger zooplankton, so that smaller cyclopoid copepods
become dominant.
Variability in the trophic structure of upwelling ecosys
tems, be it short term regime shifts or longer term trends,
occurs as a consequence of a range of processes operating
via both environment and man’s direct intervention by
fishing. Understanding these interactions in order to predict
the response of upwelling ecosystems to climatic forcing
and fishing strategies involves unravelling a multiplicity of
factors that affect primary production, zooplankton and
fish recruitment, a challenging task for the scientific
community.
Further Reading
Alheit P and Niquen M (2004) Regime shifts in the Humboldt Current
ecosystem. Progress in Oceanography 60: 201 222.
Bakun A (1990) Global climate change and intensification of coastal
ocean upwelling. Science 247: 198 201.
Barange M and Harris R, eds. (2003) Marine Ecosystems and Global
Change. IGBP Science no. 5, 32pp. Stockholm: IGBP.
Croll DA, Marinovic B, Benson S, et al. (2004) From wind to whales:
Trophic links in a coastal upwelling system. Marine Ecology Progress
Series 289: 117 130.
Cury P and Roy C (1989) Optimal environmental window and pelagic
fish recruitment success in upwelling areas. Canadian Journal of
Fisheries and Aquatic Sciences 46: 670 680.
Cury P, Bakun A, Crawford RJM, et al. (2000) Small pelagics in upwelling
systems: Patterns of interaction and structural changes in ‘wasp
waist’ ecosystems. ICES Journal of Marine Science 57: 603 618.
Cury P and Shannon L (2004) Regime shifts in upwelling ecosystems:
Observed changes and possible mechanisms in the northern and
southern Benguela. Progress in Oceanography 60: 223 243.
Hare CE, DiTullio GR, Trick CG, et al. (2005) Phytoplankton
community structure changes following simulated upwelled iron inputs
in the Peru upwelling region. Aquatic Microbial Ecology 38: 269 282.
Lehodey P, Alheit J, Barange M, et al. (2006) Climate variability, fish and
fisheries. Journal of Climate 19: 5009 5030.
Lynam CP, Gibbons MJ, Axelsen BE, et al. (2006) Jellyfish overtake fish
in a heavily fished ecosystem. Current Biology 16: R492 R493.
Mann KH and Lazier JRN (2006) Dynamics of Marine Ecosystems.
Biological Physical Interactions in the Ocean. Oxford, UK: Blackwell.
Moloney CL, Jarre A, Arancibia H, et al. (2005) Comparing the
Benguela and Humboldt marine upwelling ecosystems with
indicators derived from inter calibrated models. ICES Journal of
Marine Science 62: 493 502.
Murray JW, Barber RT, Roman MR, Bacon MP, and Feely RA (1994)
Physical and biological controls on carbon cycling in the Equatorial
Pacific. Science 266: 58 65.
Payne AIL, Brink KH, Mann KH, and Hilborn R (1992) Benguela trophic
functioning. South African Journal of Marine Science 12: 1 1108.
Peterson W (1998) Life cycle strategies of copepods in coastal upwelling
zones. Journal of Marine Science 15: 313 326.
Ryther JH (1969) Photosynthesis and fish production in the sea. Science
166: 72 76.
Schwartzlose RA, Alheit J, Bakun A, et al. (1999) Worldwide large scale
fluctuations of sardine and anchovy populations. South African
Journal of Marine Science 21: 289 347.
Summerhayes CP, Emeis K C, Angel MV, Smith RL, and Zeitschel B
(eds) Upwelling in the Ocean: Modern Processes and Ancient
Records, 422pp. New York: Wiley.
Van der Lingen CD, Shannon LJ, Cury P, et al. (2006) Resource and
ecosystem variability, including regime shifts, in the Benguela
Current System. In: Shannon V, Hempel G, Malanotte Rizzoli P,
Moloney C, and Woods J (eds.) Benguela: Predicting a Large Marine
Ecosystem, Large Marine Ecosystem Series, vol. 14, pp. 147 184.
Amsterdam: Elsevier.
Urban Systems 461
Urban Systems
T Elmqvist, Stockholm University, Stockholm, Sweden
C Alfsen, UNESCO, New York, NY, USA
J Colding, Royal Swedish Academy of Sciences, Stockholm, Sweden
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Urbanization and Plant and Animal Communities
Urban Habitats and Gradient Analyses
Urban Systems and Ecosystem Services
Urban Restoration
Urban Landscapes as Arenas for Adaptive Management
Linking Humans and Nature in the Urban Landscape
Further Reading
Introduction
Mismatches between spatial and temporal scales of
ecological process on the one hand, and social scales of
monitoring and decision making on the other have not
only limited our understanding of ecological processes in
urban landscapes, they have also limited the integration of
urban ecological knowledge into urban planning. In ecol
ogy there is now a growing understanding that human
processes and cultures are fundamental for sustainable
management of ecosystems, and in urban planning it is
becoming more and more evident that urban management
needs to operate at an ecosystem scale rather than within
the traditional boundaries of the city.
Although studies of ecological patterns and processes in
urban areas have shown a rapid increase during the last
decade, there are still significant research gaps that con
strain our general understanding of the effects of
urbanization processes. The vast majority of studies so
far have been short term (one to two seasons), conducted
in cities in Northern Europe or the US, have lacked
experimental approaches, focused on either birds or
plants, while other taxa are rarely represented and have
only included portions of a rural–urban gradient. Most
significantly, we nearly completely lack studies in rapidly
growing urban landscapes in tropical developing countries
that are rich in biodiversity and are just beginning to
address the complexity of human settlements in the
tropics.
Of further significance is that urban landscapes pro
vide important large scale probing experiments of the
effects of global change on ecosystems, since, for example,
significant warming and increased nitrogen deposition
already are prevalent and because they provide extreme,
visible, and measurable examples of human domination of
ecosystem processes. Urban landscapes may be viewed as
numerous large scale experiments producing novel
types of plant and animal communities and novel types
of interactions among species, and as such deserve the full
attention of not only evolutionary biologists and ecolo
gists but also of students of social–ecological interactions.
Urbanization is a global multidimensional process that
manifests itself through rapidly changing human popu
lation densities and changing land cover. The growth of
cities is due to a combination of four forces: natural
growth, rural to urban migration, massive migration
due to extreme events, and redefinitions of administra
tive boundaries. Half of the world’s population today
lives in urban areas, a proportion expected to increase
by 2/3 within 50 years. Today, over 300 cities have a
population of more than 106 and 19 megacities exceed
107. As urbanization is accelerating, the growth of cities
forms large urban landscapes, particularly in developing
countries. (Urban landscape is here defined as an area
with human agglomerations with >50% of the surface
built, surrounded by other areas with 30–50% built,
and overall a population density of >10 ind. ha 1.)
For example, during the last 20 years in China, clusters
of cities have emerged forming at least five mega urban
landscapes. These large and densely urbanized regions
have each between 9 and 43 large cities located in close
proximity and a population ranging from 27 to 75 million
people. This rapid urbanization represents both a chal
lenge and an opportunity to ensure basic human welfare
and a viable global environment. The opportunity lies in
that urban landscapes also are the very places where
knowledge, innovations, human and financial resources
for finding solutions to global environmental problems
are likely to be found.
Since urbanization is a process operating at multiple
scales, factors influencing environmental change in urban
landscapes often originate far beyond city, regional, or
even national boundaries. Fluctuation in global trade,
civil unrest in other countries, health pandemics, natural
disasters, and possibly climate change and political
decisions are all factors driving social–ecological trans
formations of the urban landscape.
462
Urban Systems
Urbanization and Plant and Animal
Communities
Urbanization is today viewed to endanger more species
and to be more geographically ubiquitous than any other
human activity (Figure 1). For example, urban sprawl is
rapidly transforming critical habitats of global biodiver
sity value, for example, in the Atlantic Forest Region of
Brazil, the Cape of South Africa, and coastal Central
America. Urbanization is also viewed as a driving force
for increased homogenization of fauna and flora. In the
urban core in Northern Hemisphere cities, a similar set of
species is recorded that is, often cosmopolitan plant and
animal species tolerant of anthropogenic impacts. For
example, the composition of communities of wildlife spe
cies found in cities across the US is remarkably similar
despite large variation in climate and geographical fea
tures. A common pattern among cities is that they often
show a high turnover of species with losses of native
species and gains of nonnatives over time. For example,
it is documented that New York has lost 578 native plant
species while it gained 411, and Adelaide lost 89 while it
gained 613 new plant species over a period of 166 years.
Although cities may be species rich, frequently having
higher species diversity than surrounding natural habitats,
this is often due to a high influx of nonnative species and
formation of new communities of plants and animals.
A trend of increasing nonnative species from the suburbs
to the urban core is well documented for plants, birds,
Figure 1 Cape Town, South Africa with more than 3 million
residents is located in the Cape Floristic Region, an area with the
highest density of plant species in the world with more than 9600
plants species of which 70% are endemic. Through initiatives
like Working for Wetlands and Cape Flats Nature, successful
efforts are taken to address the large challenge of conserving
precious biodiversity in fragmented natural habitats in an urban
setting where poverty is widespread. These initiatives focus on
building bridges between people and nature and demonstrating
benefits from conservation for the surrounding communities,
particularly areas where incomes are low and living
conditions are poor, and encouraging local leadership for
conservation action.
mammals, and insects. For example, in Berlin the propor
tion of novel species increased from 28% in the outer
suburbs to 50% in the built up center of the city. In New
York, the abundance and biomass of earthworms
increased tenfold when comparing rural and urban for
ests, mainly due to increased numbers of introduced
species in urban areas. Over broad geographical scales
urbanization seems to have an effect of convergence in
species composition with loss of native species and inva
sion of exotics. Nevertheless, a remarkable amount of
native species diversity is known to exist in and around
large cities, such as Singapore, Rio de Janeiro, Calcutta,
New Dehli, and Stockholm.
Interestingly, the number of plant species in urban
areas often correlates with the human population size.
Species number often increases with log number of
human inhabitants, and that relationship is stronger than
the correlation with city area. The age of the city also
affects species richness; large, older cities have more plant
species than large, younger cities. Also of interest is that
diversity may correlate with measures of economic
wealth. For example, in Phoenix, USA, measures of
plant and avian diversity in urban neighborhoods and
parks show a significant positive correlation with mea
sures of median family income levels.
In general, urban landscapes present novel ecological
conditions, such as rapid rate of change, chronic distur
bances, and complex interactions between patterns and
processes. Organisms that have survived in urbanized
areas have been able to do so for at least two reasons:
(1) they evolved rapidly and adjusted genetically or
(2) they were largely preadapted to this environment
and required little or no genetic adjustment. There are
several documented cases of rapid evolution in urban
areas, involving, for example, tolerance to toxic sub
stances and heavy metals in plants, such as lead
tolerance in urban roadside Plantago lanceolata. Among
insects there are many cases of rapid evolution in urban
areas, notable example being the famous case of industrial
melanism among Lepidoptera in UK, a phenomenon also
documented from areas in USA, Canada, and elsewhere in
Europe. Also of interest is that specific urban and rural
races have been identified within well studied Drosophila
species.
In Table 1 we have summarized some of the effects of
urbanization including abiotic and biotic changes. Human
activities may cause increased deposition of nutrients
such as nitrogen and phosphorus and emission of toxic
chemicals which influence urban soil processes.
Decomposition rates in urban soils are often negatively
affected by pollution and toxic chemicals, but positively
affected by increased soil temperature. Decomposition
rates may therefore often be higher in urban than in
rural soils. However, urban litter tends to have higher
C:N ratios and therefore also tends to be more recalcitrant
Urban Systems 463
Table 1 Ecological effects of urbanization
Physical and chemical
environment
Air pollution increases
Hydrological changes
Local climate change
Soil changes
Water changes
Population and community characteristics
Ecosystem structure and function
Altered reproductive rates
Genetic drift, changes in selection
Social and behavioral changes
High species turn over, increase of exotic
species
Loss of K-species and gain of r-species
Increased dominance of generalist species
Altered disturbance regimes
Altered succession
Altered decomposition rates
Altered nutrient retention
Habitat fragmentation
Changes in trophic structure, domination of
omnivores
Modified from McDonnell MJ and Pickett STA (1990) Ecosystem structure and function along urban rural gradients: An unexploited opportunity for
ecology. Ecology 71: 1232 1237.
than rural litter. Urbanization affects in complex ways
both directly and indirectly C pools and N transformation
rates and contrasting dynamics in urban and rural soils is
an area where much more research is needed.
Biotic changes influencing ecosystem functioning are
listed in Table 1. There are a number of reasons why new
human imposed scales for ecological processes are found
within urban areas. First, compared with ecosystems in
rural areas, urban systems are highly patchy and the
spatial patch structure is characterized by a high point
to point variation and degree of isolation between
patches. Second, disturbances such as fire and flooding
are suppressed in urban areas, and human induced dis
turbances are more prevalent as well as intense human
management of urban habitats. Third, because of the
‘heat island’ effect, that is, higher mean temperatures in
cities than in the surroundings, cities in temperate cli
mates have significantly longer vegetation growth
periods. Fourth, ecological successions are altered, sup
pressed, or truncated in urban green areas, and the
diversity and structure of communities of plants and
animals may show fundamental differences from those
of nonurban areas. In general, with increasing urbaniza
tion there is a trend toward dominance of generalist
species with high reproductive capacity and short gen
eration times.
Urban Habitats and Gradient Analyses
Urban habitats are extremely diverse and examples
include parks, cemeteries, vacant lots, streams and lakes,
gardens and yards, campus areas, golf courses, bridges, air
ports, and landfills (see Landfills). These habitats are
highly dynamic, influenced by both biophysical and eco
logical drivers on the one hand and social and economic
drivers on the other. Urban landscapes often represent
cases of extreme habitat fragmentation. Habitat patches in
the urban core are more or less strongly isolated from
each other by a matrix of built environment making
dispersal risky and difficult at least for poorly dispersing
organisms. There are numerous studies analyzing effects
of isolation of urban habitats and, for example, in urban
gardens in UK, the best predictor of species richness of
ground arthropods was found to be the proportion of
green areas within a 1 km radius of the sampling site.
Analyses of the distribution of plant species in vegetation
fragments in Birmingham, UK showed a positive correla
tion between the density of patches available to a species
and the proportion of these patches that was occupied.
For many plant species the rate of occupancy increased
with site age, area, and similarity of adjacent
habitats. Similarly, for urban amphibian assemblages in
Melbourne, Australia, an increase in species richness was
associated with pond size and a decrease with increasing
isolation. Habitat quality also influenced species compo
sition. The importance of isolation is likely to increase
over time and, for example, in Boston an isolated urban
park lost 25% of its plant diversity over 100 years. To
what extent greenways and corridors increase connectiv
ity and contribute to maintain viable populations in urban
green areas is poorly understood. But greenways may, in
multiple ways, provide a chain of different habitats per
meating the urban environment and be of benefit for
many organisms. Apart from preventing local extinction
and facilitating re colonization, increased habitat connec
tivity is important to maintaining vital biological
interactions, for example, plant–pollinator interactions
and plant–seed disperser interactions. Although most of
the studies in urban landscapes address the continuous
loss of green areas due to urban growth and expansion,
this is not the case in all cities. For example, in Shanghai
the proportion of green areas has increased in parallel
with urban expansion and the total area expanded from
less than 9 km2 in 1975 to more than 250 km2 in 2005.
Gradient analyses have a long tradition in ecology and
go back to the pioneering work by Whittaker in the late
1960s. Gradient analyses have also been a rather common
way of disentangling the complexity of urban habitats and
have been used to investigate how urbanization changes
464
Urban Systems
Species diversity
ecological patterns and processes across landscapes, for
example, in invertebrate, plant, and bird community com
position, leaf litter decomposition and nutrient cycling,
and the structure of landscape elements. Almost all the
gradients that have been used for urban studies have been
one dimensional in the sense that they only describe
physical features of the gradient such as proportion of
impervious surface, while the characteristics of the human
population occupying a particular portion of the land
scape often have been neglected. Because urbanization is
an exceedingly complex amalgamation of factors, by
using only a single axis the interpretation of the under
lying processes has often been severely constrained. It has
been suggested that a more comprehensive gradient ana
lysis should include not only physical geography,
demography, rates of ecological processes, and energy,
but also history of land use, socioeconomic analyses, and
patterns of management.
Variation in species densities across an urban gradient
suggests that some individual species disappear with
urbanization, whereas other species invade in response
to the environmental changes associated with develop
ment. At least for birds, species richness has often been
found to peak at intermediate levels of urbanization and
decrease with either more or less development. Some
species are classified as urban avoiders with their highest
densities at the most natural sites, whereas many species
seem to be able to adapt to suburban environments, with
densities peaking at intermediate levels of development.
Some species are urban exploiters whose highest densities
are found at the urban core (see Figure 2).
Suburban adaptable
Urban avoiders
Rural
Urban exploiters
Urban
Degree of anthropogenic impact
Figure 2 Plants and animals may respond differently to
increasing human impact. Urban avoiders are large-bodied
species or species linked to late successional stages. These
species might be very sensitive and show a decline already at
moderate human impacts. Suburban adaptable speceis may, to
various degrees, utilize human modifications of the landscape;
the majority of plant and animal species likely belong to this
group. Urban exploiters directly benefit from human presence for
food, reproduction or protection, and may often be
cosmopolitan, generalist species. Terminology after Blair RB
(1996) Land use and avian species diversity along an urban
gradient. Ecological Applications 6(2): 506–519.
A multitude of factors are likely to influence this
pattern of extinction and colonization, of which changes
in predation rates have been suggested to be among the
most important. Predation on artificial nests has often
been found to be higher in urban parks than in neighbor
ing woodlands and the abundance of predators such as
corvids, rats, and house mice are often more in urban
parks compared to the rural end of the gradient.
However, there are also studies showing no correlation
or a declining predation pressure along the urban–rural
gradient. Observed patterns of extinction and invasion in
urban landscapes may also be linked to gaps in the spec
trum of body masses exhibited in the community and
there are documented cases that body mass patterns are
correlated with invasion and extinction in other human
transformed ecosystems.
In cities, ownership and management of urban habitats
is extremely diverse and complex. In addition to land
managed by government, municipalities, churches, and
foundations, there is also land managed by local user
groups that often covers substantial tracts. For example,
domestic gardens cover 23% of the land area of Sheffield,
and as much as 27% of the city of Leicester. Lands
appropriated for allotment areas, domestic gardens, and
golf courses were found to cover nearly 18% of the total
green space of greater Stockholm, Sweden, representing
well over twice the area covered by nature protected
areas. While the numbers of ecological studies of urban
green areas are limited, there is evidence that different
types of urban green areas used for purposes other than
biodiversity conservation play an important role in sus
taining urban biodiversity. For example, in many cities of
Asia, educational institutions sometimes harbor the lar
gest and last remaining green areas in extensively urban
developed settings. These campus areas can be extremely
significant for biodiversity. A good example is the uni
versity campuses of Pune city, India, which harbor up to
half the plant, bird, and butterfly species of the region
despite the fact that these campuses only cover some 5%
of the land area. Another illustrative case is the Musahi
Institute of Technology in Yokohama, Japan, where a
former community managed forest now restored for stu
dent education, has revived interest in reducing the loss of
biodiversity rich forests in semiurban areas of Japan.
Ecological studies also show that domestic gardens some
times hold a rich flora of plants, including rare and
threatened ones. Thompson and colleagues found that
the private gardens in Sheffield, UK, contained twice as
many plant species as any other habitat assessed. These
gardens also supported surprisingly high numbers of
invertebrates and this regardless of whether garden plants
were native or alien. Even such a controversial land use as
golf courses can contribute with important biodiversity
functions in cities when courses are wisely managed and
well designed. For example, golf courses contribute to
Urban Systems 465
sustain urban woodlands in many cities of Japan, and in
some larger cities of Sweden they may harbor significant
populations of species of both amphibians and macro
invertebrates, which are declining in rural areas.
Also, smaller habitat parcels in urban settings can
provide high quality habitats. One illustrative example
of this is allotment areas, common in many city regions
in developed countries. For example, while allotment
areas only cover some 0.3% of the land in greater
Stockholm, they tend to be extremely biodiversity rich.
In Stockholm city one allotment garden was found to
contain 447 different plant species in an area of 400 m2.
Urban Systems and Ecosystem Services
The concept of ecosystem services has proved to be useful
in describing human benefits from urban ecosystems. For
example, urban vegetation may significantly reduce air
pollution, mitigate the urban heat island effect, reduce
noise, and enhance recreational and cultural values, of
importance for urban citizen’s well being (Table 2).
The scale of importance for generation of these ser
vices is often much larger than a city, for example, for
reduction of air pollutants and water regulation, while for
some, such as recreational and educational services, gen
eration often occur within city boundaries. In most cases,
these services tend to be overlooked by urban planners
and decision makers, despite the fact that the potential of
generation of ecosystem services can be quite substantial.
In a study made within Stockholm County, it was assessed
that this region’s ecosystems potentially could accumulate
about 41% of the CO2 generated by traffic and about 17%
of total anthropogenic CO2. In the Chicago region, trees
were found to remove some 5500 t of air pollutants per
Table 2 Examples of services generated by urban ecosystems
Ecosystem services
Supporting services
Soil formation
Nutrient recycling
Provisioning services
Freshwater
Food, fiber, and fuel
Genetic resources
Regulating services
Air quality
Local climate regulation
Water purification and waste treatment
Biological control
Pollination
Cultural services
Esthetic and recreational
Educational
year, providing a substantial improvement in air quality.
It was also found that the present value of long term
benefits from the trees of the Chicago region was more
than twice the present value of costs related to planting.
Moreover, wetlands in urban settings can substantially
lower the amount of money spent on sewage treatment
costs and in many cities large scale experiments are tak
ing place where wetlands are being used to treat sewage
water. It has been estimated that up to as much as 96% of
the nitrogen and 97% of the phosphorus can be retained
in wetlands through the assimilation of wetland plants and
animals. Green spaces of cities also provide ample oppor
tunity for recreation. In a study on the response of persons
put under stress, it was shown that when subjects of the
experiment were exposed to natural environments the
stress level decreased, whereas during exposure to built
up urban environments the stress levels remained high or
even increased.
A major challenge in urban areas is how to sustain the
capacity to generate ecosystem services. This capacity is
mainly but not exclusively related to the diversity of
‘functional groups’ of species in a system, like organisms
that pollinate, graze, predate, fix nitrogen, spread seeds,
decompose, generate soils, modify water flows, open up
patches for reorganization, and contribute to the coloni
zation of such patches. In urban areas, such functional
groups may be substantially reduced in size or show
changes in the composition due to high species turnover,
both of which may increase vulnerability in maintaining
ecosystem services. To what extent exotic species con
tribute to reduce or enhance the flow of ecosystem
services is virtually unknown for any urban area. But,
since introduced species make up a large proportion of
the urban biota, it is important to know not only to what
extent introduced species are detrimental, but also to
what degree some of the introduced species may enhance
local diversity and maintain important functional roles.
For urban ecology to significantly contribute to improv
ing management of urban habitats and the maintenance of
ecosystem services, the following research questions are
particularly urgent to address:
what extent are urban ecosystems sinks for many
• To
animal and plant species and what are the effects of
•
•
•
species loss on ecosystem functions?
To what extent do novel species play important func
tional roles in urban ecosystems by replacing role of
extinct native species and enhancing ecosystem func
tions and services?
What is the importance of source–sink dynamics and
matrix permeability for maintenance of urban biodi
versity and important ecosystem services?
How do we develop management systems that
match the spatial and temporal scales of ecological
processes?
466
Urban Systems
Urban Restoration
Designed, replicated urban restoration experiments
could substantially advance our knowledge and under
standing of processes of importance for generating
urban ecosystem services, for example, through better
understanding of population and community responses
to disturbances, patterns of self organization and suc
cession, assembly rules, and through identifying
components that contribute to resilience or vulnerabil
ity. Urban restoration also represents an interesting
opportunity for ecologists to work in partnership with
landscape architects, urban designers, and architects and
help in designing urban environments based on ecolo
gical knowledge but merged with the functional and
esthetic design of urban space.
The majority of urban restoration projects deal with
transforming brown areas (abandoned industrial lots or
air fields, landfills, etc.) to functioning green areas, such
as Fresh Kills landfill on Staten Island, New York, or
the Olympic Park in Beijing. Other large scale restora
tion projects involve, for example, substantial wetland
restoration such as Kristianstad, Sweden, and New
Orleans, USA. In New Orleans, coastal wetlands have
eroded substantially during the last 50 years and restor
ing wetlands is viewed as one important measure to
reduce vulnerability to hurricanes. Important lessons of
urban restoration projects so far are that restoring eco
logical functions in urban areas is possible but time
consuming and that there are often significant effects
on many ecosystem services even after one or two
seasons. Although the costs are initially high, these
could be offset by increases in property values and
increased investments in development in areas sur
rounding the restoration site.
Urban Landscapes as Arenas for Adaptive
Management
In cities that experience rapid social and environmental
transformation, it is critical to develop a capacity to
respond to potential surprises. One important aspect of
such capacity building is to facilitate for a wider integra
tion of local people and interest groups in the use and
management of urban green areas. There are several
reasons for a wider integration of local people in urban
ecosystem management. First, governments cannot
entirely rest on protected area management to safeguard
the native flora and fauna found in city regions. As cities
expand, there will be an increased lack of natural lands for
the establishment of protected areas. Studies also show
that many urban nature reserves are unable to sustain
native species in the long run. In addition, protected
area management is financially costly to most local gov
ernments. In London, for example, parts of the protected
green belt have been severely degraded due to lack of
money and this has resulted in urban residents avoiding
these areas for various activities. Second, much of the
flora and fauna depend on well functioning habitats pro
vided by privately owned lands. In the US, for example,
almost two thirds of all the endangered and threatened
species depend on private lands for their continued exis
tence. Also, urban homeowners with gardens have been
engaged to support declining pollinator populations in
Great Britain through the deliberate planting of certain
nectar providing plants in their yards. They have also
helped sustain urban frog populations during their period
of main rural declines through a massive establishment of
garden ponds. Homeowners in Britain are also involved in
programs for monitoring trends in the population status of
birds. Third, a number of international treaties that have
been signed by national governments around the world,
including local Agenda 21, the Convention of Biological
Diversity, and the Malawi principles, strive toward a
decentralization of biodiversity management down to
local people. Recently, the Millennium Ecosystem
Assessment (MA 2005) concluded that a wider coopera
tion among people within different sectors in society is
necessary for more efficient land use that contributes to
the support of ecosystem services.
One approach increasingly used to achieve colla
borative partnership in urban ecosystem management
is ‘adaptive co management’. The approach rests on
the notion of the sharing of resource management
responsibility and authority between users of ecosys
tems and government agencies. This typically involves
local people and interest groups, scientists and local
authorities, with the potential to promote information
exchange to effectively deal with and respond to
change and issues that often transcend locality.
Adaptive co management emphasizes ‘learning by
doing’ in ecosystem management, where management
objectives are treated as ‘experiments’ from which
people can learn by testing and evaluating different
management policies. This form of ecosystem manage
ment avoids set prescriptions of management that may
be superimposed on a particular place, situation, or
context. Such designs have the potential to lower
overall costs of management, most notably costs
incurred for describing and monitoring the ecosystem,
designing regulations, coordinating users and enfor
cing regulations, and depend on the self interest of
participants. Co management arenas could, for exam
ple, also serve as platforms for designed experiments
and urban restoration as discussed above and improve
ecological functions and designs in cities.
Urban Systems 467
Linking Humans and Nature in the Urban
Landscape
Urban landscapes are not only ecological experiments but
also long term experiments in social, economic, and cul
tural transformations shaped by cultures, property rights,
and access rights. Since cities are places where knowl
edge, human and financial resources are concentrated,
rapid urban transformations can likely be more readily
monitored and observed than similar processes in more
rural areas. Studies of transformations in urban landscapes
may therefore well provide the ground for a better under
standing of socio economic drivers of changes also in other
ecosystems. After decades of mutual neglect and artificial
divide between nature on the one hand, and cities with their
respective urban processes on the other hand, the conserva
tion community has started to shift its perception to include
cities as a component of natural landscapes. Just as it is now
increasingly recognized that in protected nature reserves,
conservation will not be successful as long as it is at the
expense of human aspirations, urban planners increasingly
acknowledge that functioning natural systems such as
watersheds, mangroves, and wetlands are indispensable for
reducing vulnerabilities to natural disasters and building
long term resilience.
In New Orleans for example, it has been argued that
population growth and urban economic growth is neces
sary for meeting the costs of building a viable defense
against the grave environmental problems of massive
coastal erosion. In the New York Metropolitan region,
sustainable management of the Catskills, the land around
the upland water reservoirs supplying New York City
with drinking water, has been chosen as an important
complement to building water treatment plants.
The urban landscape provides a public space for the
cross fertilization of minds and various disciplines,
enabling a new perspective on man in nature, one that
could place human well being at the core, break the
artificial and largely culturally biased divide between
the pristine and the human dominated ecosystems, and
contribute to the creation of a new language, with signs,
concepts, words, tools, and institutions that would gather
rather than divide, broker conflicts rather than create
them, and establish responsible environmental steward
ship at the heart of public interest.
See also: Landfills; Riparian Wetlands.
Further Reading
Adams CC (1935) The relation of general ecology to human ecology.
Ecology 16: 316 335.
Adams CE, Lindsey KJ, and Ash SJ (2006) Urban Wildlife Management.
Boca Raton: CRC Press, Taylor and Francis.
Alfsen Norodom C (2004) Urban biosphere and society: Partnership of
cities. Annals of New York Academy of Sciences 1023: 1 9.
Blair RB (1996) Land use and avian species diversity along an urban
gradient. Ecological Applications 6(2): 506 519.
Colding J, Lundberg J, and Folke C (2006) Incorporating green area user
groups in urban ecosystem management. Ambio 35(5): 237 244.
Collins JP, Kinzig A, Grimm NB, et al. (2000) A new urban ecology.
American Scientist 88: 416 425.
Felson AJ and Pickett STA (2005) Designed experiments: New
approaches to studying urban ecosystems. Frontiers in Ecology and
the Environment 10: 549 556.
Kinzig AP, Warren P, Martin C, Hope D, and Katti M (2005) The effects of
human socioeconomic status and cultural characteristics on urban
patterns of biodiversity. Ecology and Society 10(1): 23.
http://www.ecologyandsociety.org/vol10/iss1/art23 (accessed
December 2007).
McDonnell MJ and Pickett STA (1990) Ecosystem structure and
function along urban rural gradients: An unexploited opportunity for
ecology. Ecology 71: 1232 1237.
McDonnell MJ and Pickett STA (1993) Humans as Components of
Ecosystems: Subtle Human Effects and the Ecology of Populated
Areas, 363pp. New York: Springer.
McGranahan G, Marcotullio P, Bai X, et al. (2005) Urban systems. In:
Scholes R and Ash N (eds.) Ecosystems and Human Well being:
Current State and Trends, ch. 27, pp. 795 825. Washington, DC:
Island Press. http://www.maweb.org/documents/document.296.
aspx.pdf (accessed December 2007).
Millennium Ecosystem Assessment (2005) Ecosystems and Human
Well being: Synthesis. Washington, DC: Island Press.
Pickett STA, Cadenasso MI, Grove JM, et al. (2001) Urban ecological
systems: Linking terrestrial ecological, physical and socioeconomic
components of metropolitan areas. Annual Review of Ecology and
Systematics 31: 127 157.
Sukopp H, Numata M, and Huber A (1995) Urban Ecology as the Basis
of Urban Planning. The Hague: SPB Academic Publishing.
Turner WR, Nakamura T, and Dinetti M (2004) Global urbanization
and the separation of humans from nature. Bioscience
54: 585 590.
468
Wind Shelterbelts
Wind Shelterbelts
J-J Zhu, Institute of Applied Ecology, CAS, Shenyang, People’s Republic of China
ª 2008 Elsevier B.V. All rights reserved.
Introduction
Interactions between Wind and Trees
Shelterbelt Structures
Determination of Optical Porosity
Wind Profiles near Shelterbelts
Establishment and Management of Shelterbelts for
Wind Protection
Further Reading
Introduction
The quantity of falling solar energy and the propor
tion that is absorbed by either the atmosphere or the
surface varies greatly from one place to another,
which results in regional variations in temperature
both for Earth’s surface and atmosphere. The varia
tions of temperature cause a difference in atmospheric
pressures, which leads to the movement of air from
high pressure area to low pressure area, that is, the
wind. Wind moves horizontally, vertically, and turbu
lently. It is affected by the conditions of the surfaces it
encounters. The surface wind influences the habitats
for wildlife, the growth of crop and livestock, soil
erosion, snow distribution, sand blowing, etc., and
causes extreme damages when it is very strong.
Shelterbelt is always called windbreak (some authors
distinguish the usage between the two terms based on
their objectives; here no distinction is adopted), which
can be defined as the barrier used to reduce wind speed.
It usually consists of trees and shrubs, or even perennial
or annual crops, wooden fences or other materials
(Figures 1 and 2). Shelterbelts when they are reasonably
designed can provide large areas of reduced wind speed
because they increase the roughness. The areas of
reduced wind speed, especially in windy regions, are
generally called sheltered zones, which are very useful
Figure 2 Shelterbelt provides wind prevention in farmland.
for wildlife, agriculture, and for the people suffering from
the severe climates. In fact, there are many ecological
functions such as protecting against erosion, improving
crop production, filtering air and water, ameliorating cli
matic extremes, improving ecological environmental
qualities, reducing and ameliorating potential conflicts
that may arise, fulfilled by the shelterbelts through alter
ing wind behavior.
However, shelterbelt structural factors such as the
height, density, orientation, length, width, continuity,
cross section shape, and the pattern of tree arrangements
in windbreaks all influence shelter effectiveness.
Therefore, how to manipulate shelterbelt structures
through management practices to meet different objec
tives is one of the key scientific problems in windbreak
studies. Obviously, shelterbelts involve comprehensively
complex systems of establishment and management to
satisfy multiple objectives.
Interactions between Wind and Trees
Wind Parameters
Figure 1 Farmland shelterbelt.
There are four parameters of wind that are measured:
(1) wind direction (the direction from which the wind
blows); (2) wind speed (measured from mechanical
Wind Shelterbelts
469
the pressure decreases as the wind passes through the trees,
and reaches a minimum just to the lee side of the trees. The
pressure gradually increases to the lee side, returning to the
original pressure condition beyond a certain distance.
Tree Shelterbelts/Windbreaks
anemometers, m s 1) (Figure 3); (3) wind gusts and squalls
(gust, a sudden significant increase of wind speed, the peak
wind speed must reach at least 8.0 m s 1, and the variation
between peaks and lulls is at least 5.1 m s 1, the duration is
usually less than 20 s; squall, a sudden onset of strong winds
with speeds increasing to at least 8.0 m s 1 and sustained at
11.2 m s 1 or more for at least 1 min); and (4) shifts (the
changes of wind directions).
When the principles of the ecological effects of vegeta
tions on wind are applied in practice, the shelterbelts are
formed. It is reported that the origin of shelterbelts was in
the middle of 1400 s when the Scottish Parliament urged
the planting of tree belts to protect agricultural produc
tion. The primary purpose of a tree shelterbelt is to create
a barrier that reduces the wind speed for preventing soil
particles blowing away from fields. Shelterbelts are com
monly established with one or more closely spaced rows,
containing one or more tree species. Because a shelterbelt
increases the surface roughness and thus reduces the wind
speed when it is properly designed on the farmland or
other areas that need protection, it can provide large
areas with reduced wind speed for both agriculture, and
people and animals. The main effect of a tree shelterbelt is
to provide shelter. The effects of shelterbelts on wind
(benefits) are more dependent on their structures, which
human beings can manage or control.
Ecological Effects of Wind
Effects/Benefits of Shelterbelts
Wind has a wide variety of ecological effects, and plays an
important role in the development of agriculture and
forestry. For example, wind can transport water vapor,
heat energy, pollen grains, spores, and seeds of plants,
generate static electricity, and affect evaporation and
transpiration, etc. In contrast, gale (wind speed from 14.3
to 28.3 m s 1), can erode soil, do damage to farm, live
stock, and to trees or forests.
Effects of shelterbelts on microclimates. Shelterbelt can reduce
soil evaporation rates when leaf area index is low. This
may improve the soil water availability for the crop later
in the season and thus improve seasonal water use effi
ciency. The increase in diffuse radiation near a shelterbelt
can increase light transmission into plant canopies. This
may be a mechanism that could enhance photosynthetic
activity of crops. Shelterbelts have thermal radiation
effects too as a result of the sky view effect, where the
cold night sky is replaced by warmer trees. A shelterbelt
can reduce thermal radiation losses from the surrounding
crop out to a certain distance from the shelterbelt. This
may reduce frost incidence. Shelterbelts can lower soil
and canopy temperatures through shading, which may
limit growth of crops. However, we should also note
that shading may decrease evaporation rates close to a
shelterbelt and, therefore, reduce moisture losses from
bare soil prior to germination and emergence of crops.
Therefore, shelterbelts can provide improved growing
conditions in the sheltered areas.
Effects of shelterbelts on improving crop yield and quality.
Due to higher soil moisture, daytime temperatures,
humidity, and nighttime carbon dioxide levels, as well
as lower evaporation and nighttime air temperature in
sheltered areas, the yield, quality, and maturity of crops
are improved compared with unsheltered crops,
Figure 3 Wind measurement with a three-dimensional
ultrasonic anemometer.
Effects of Trees on Wind
Vegetations, especially trees, exert significant effects on
winds close to the ground through altering wind behaviors.
This is because the roughness of trees can provide surface
friction, which makes the wind speed decrease, and the
turbulence increase. Trees can be considered as a barrier
on the land surface that changes the wind patterns. When
wind approaches the trees, a portion of the airflow passes
through the trees due to the porosity of leaves
and branches. The remaining airflow is forced up and
over the trees. When the wind moves through the trees,
the surface area of leaves, branches, and the bark of stems
provides a very frictional surface, which can reduce
the wind speed effectively. Pressure on the ground
increases as the wind approaches the trees and reaches a
maximum at the windward edge of the trees. Meanwhile,
470
Wind Shelterbelts
particularly, in locations where snow accumulation or
wind damage often occur.
Effects of shelterbelts on fauna and vegetation. Shelterbelts
can provide wildlife habitats. Generally, macrofauna is
lower in the field as compared to the shelterbelt and
ecotones. The soil organic matter content, as well as
microbial and faunal biomass, decrease gradually from
the shelterbelt toward the field center. The shelterbelts
influence the biomass, density, and composition of many
soil and aboveground fauna taxa and individual size of
animals occurring in bordering fields. In addition, the
transport of pests, pollen and pathogens, etc., all rely on
wind. Thus, providing shelter will modify the pathways
for this around the crop environment or farmland.
A well designed tree shelterbelt can increase the bio
logical diversity that potentially may introduce natural
predators to prey on pests and so reduce the need for
pesticides.
Effects of shelterbelts on erosion control and others. Because
of reduced wind velocity, erosion can be controlled by
shelterbelts, especially on exposed sandy or dry soils.
Well designed shelterbelts can provide for uniform
snow deposition/distribution on farmland across fields.
Single row tree shelterbelts are the most effective and
permanent barriers for uniform snow distribution.
Stockyards and around farm buildings can also be
protected through control of wind by shelterbelts.
Additionally, esthetics and values in the landscape, tree
based resources such as timber, nuts, fiber or biomass, etc.,
are the benefits of shelterbelts.
turbulence generation decreases. In contrast, as porosity
decreases, less wind passes through the shelterbelt and
wind speed reduction is greater, but the extent of the
protected distance decreases with more turbulence
generation.
Shelterbelt density ( d) is the ratio of the solid portion
of the shelterbelt to the total volume of the shelterbelt. It
has the same meaning as the term of porosity, that is, the
ratio of the open portion of the shelterbelt to the total
volume. The two terms are complementary.
Optical porosity and optical density. Although shelterbelt
porosity or density is important in describing the shelter
belt structure, unfortunately, it is nearly impossible to
physically measure the aerodynamic porosity of plants
because of the three dimensional nature of the pores
through which the wind flows. Therefore, much effort
has been directed toward finding an alternative measure
ment. Optical porosity ( ), a two dimensional measure of
porosity, which is defined as a simple ratio of perforated
area to total area on the vertical section of a shelterbelt,
has been employed as a descriptor of shelterbelt structure.
It has proved to be a promising alternative to aerodynamic
porosity, especially for narrow shelterbelts. Generally,
optical porosity is not equivalent to aerodynamic porosity
since it does not take into account the three dimensional
nature of the pores, but for a narrow artificial windbreak,
is close to a.
Optical density ( r) has the reciprocal or complemen
tary implication as optical porosity. It is defined as the
ratio of projected solid area to the total side view area of a
shelterbelt.
Shelterbelt Structures
Shelterbelt structure can be considered as the distribution
patterns of tree stems, branches, and leaves (tree elements)
in a shelterbelt stand, which is determined by tree species,
stem density, composition and the arrangement pattern of
the trees in the shelterbelt, DBH (stem diameter at breast
height, 1.3 m from the ground), tree height (H), tree age,
etc. Additionally, the shelterbelt structure is also affected
by its orientation, length, width, cross sectional shape,
continuity, and uniformity.
Internal Structural Characteristics
Porosity and density. The most commonly applied descrip
tor of the internal structure of shelterbelts is porosity.
Shelterbelt porosity is defined as the ratio or percentage
of pore space to the space occupied by tree elements. The
optimum aerodynamic porosity ( a) is usually considered
to be 0.35–0.45. Porosity affects the turbulence level in
and around the shelterbelt. As porosity increases, more
wind passes through the shelterbelt, that is, less wind
speed reduction, but the protected distance increases and
External Structural Characteristics
Height. Shelterbelt height (H) is the most important factor
determining the extent of the protected zone. Both theory
and empirical measurements have shown that the pro
tected distance is proportional to shelterbelt height. The
shelterbelt height varies according to tree species, site
conditions, and management levels, and keeps increasing
before the maturity of the shelterbelt. The shelterbelt
height can be described by the maximum height of indi
vidual trees, the average height to the tops of the taller
trees or as the height averaged over randomly located
points along the length of the shelterbelt. In order to
facilitate comparisons of shelter effectiveness between
shelterbelts, the protected distance from the shelterbelt
is usually expressed in times of the shelterbelt height.
Orientation. Shelterbelt orientation is another impor
tant factor to shelterbelt structure. The maximum shelter
effectiveness is generally obtained when the shelterbelt
is oriented perpendicular to the problem wind. If the
direction of the problem wind becomes oblique to
the windbreak, the size of protected area decreases.
Wind Shelterbelts
When the wind becomes parallel to the windbreak, wind
speed may increase under some conditions.
Length. The length of a shelterbelt varies greatly from
area to area. The influence of the shelterbelt length on
shelter effectiveness lies on the ends of the shelterbelt.
Wind speeds at the end of a windbreak are generally
greater than those at the open area. This is caused by
the flow around the ends of the shelterbelt, where the
wind seeks the path with least resistance (Figure 4).
Width. The width of a shelterbelt (W) affects the shelter
effectiveness by influencing the porosity or optical por
osity of the shelterbelt. Generally, shelterbelt can be
widened by adding rows, but additional width of shelter
belt has minimal influence on shelter response until the
ratio of width to height (W/H) of a shelterbelt is around 5
(Table 1). However, shelter effectiveness decreases with
increasing width when the ratio of W/H is greater than 5
(Figure 5). Numerical simulation results suggest that the
complex interactions between the shelterbelt width and
the internal shelterbelt structure may be more important
than previously believed; and the simulated data indicate
that as the width increases, the location of maximum
reduction of wind speed moves closer to the shelterbelt.
Cross sectional shape. The cross sectional shape of a
shelterbelt is the external profile of the cross section,
which is described by the geometry of its boundary. The
windbreak shapes are formed by the layout of the trees,
shrubs, the composition of tree species and their arrange
ment patterns within the windbreak. The aerodynamics of
cross sectional shape of shelterbelts indicates that the
cross sectional shapes affect the magnitude and extent of
wind speed reduction in the protected zone. There are
many shelterbelt cross sectional shapes; the most com
mon shapes include rectangle, triangle (windward and
leeward), gable roof/pitched roof (symmetry and asym
metry), and notch (Figure 6). The observations in the
field model windbreaks indicated that the windbreaks
with rectangle cross sectional shape exhibit better shelter
Windbreak /shelterbelt
End
End
Figure 4 Flowing pattern when wind passes through and
around the ends of the shelterbelt.
471
effectiveness (Figure 7). It is reported that windbreak
cross sectional shape has more effective influence than
horizontal distribution on shelter effectiveness.
Continuity and uniformity. The continuity and unifor
mity of shelterbelts influence shelter efficiency. Any gaps
or separations in a shelterbelt will concentrate wind flow,
which creates a zone on the leeward side of the gap or
separation where the wind speed exceeds that on the open
field (Figure 4). Therefore, lanes or other openings
through a windbreak should be avoided.
Aerodynamic Parameters of Shelterbelt
Apparent porosity. Apparent porosity ( 0) is defined as eqn
[1], which is an estimate of porosity based on the mini
mum relative wind speed. The minimum relative wind
speed is considered as a function of 0:
0
¼
0:2 þ
p
0:04 þ 3:2Ur
min
½1
where Ur-min is minimum relative wind speed (m s 1).
Based on the measurement of the minimum wind
speed on the leeside of a windbreak, 0 can be obtained.
0 is generally used as a standard for comparison of the
characteristics of diverse shelterbelts rather than a sub
jective verbal characterization.
Penetration coefficient. Penetration coefficient ( 0) is also
called permeability or aerodynamic porosity. The mean value
of 0 is defined under the conditions of an infinitely long
shelterbelt on flat ground with the z axis upward, x axis hor
izontally perpendicular, and y axis parallel to the shelterbelt as
1
H
0
¼
1
H
Z
H
z
Z0 H
z0
U ð0;zÞdz
U0 ðzÞdz
Z
H
¼ Zz0 H
z0
U ð0;zÞdz
U0 ðzÞdz
½2
where z is height above the ground (m), z0 is roughness length
(m), U(0, z) and U0(z) are horizontal wind speeds at the
leeward of the shelterbelt and at the same height in the open
area (m s 1), respectively, and H is the shelterbelt height (m).
Resistance coefficient. The resistance coefficient (Rc) is
defined as the ratio of the difference in pressure across the
shelterbelt to the product of the shelterbelt height and the
dynamic pressure of the approaching flow. The aero
dynamic properties of a porous shelterbelt are more
directly affected by the resistance coefficient.
Drag coefficient. Drag coefficient (Cd) is defined as the
ratio of the drag force per unit area on the shelterbelt to the
dynamic pressure of the approaching flow. In principle, the
shelterbelt exerts a drag force on the wind field, causing a
net loss of momentum in the incompressible airflow and
thus producing the sheltering effect. A common physical
way to express the aerodynamic effect of a shelterbelt is in
terms of its resistance to the flow, or in terms of dimension
less form such as a drag coefficient
472
Wind Shelterbelts
Table 1 Relationships between windbreak width and relative wind velocity (%)a
Relative wind speed at various distances from the shelterbelt (%)
Width
W (m)
Height
H (m)
W/H
Rows
Optical porosity
54
5
12
10
4.5
0.5
25
3
0.320
0.335
5H
51
0H
5H
10 H
15 H
20 H
25 H
Total
44
75
44
33
62
52
72
54
89
64
89
62
62
55
a
Wind direction is perpendicular to the shelterbelt.
usually a simple mathematical exercise. But when dealing
with living shelterbelts, the estimation of is much more
complicated because the openings of the living shelterbelt
are irregular in shape and in distribution. The method of
digital image processing for measuring
has been
applied. The basic principle is to partition the pores and
the tree elements using computer system through digital
images. The steps of the processes for determining are
as follows.
(a)
Relative wind velocity (%)
100
80
600 m
60
Shelterbelt
40
20
Photographing
0
10
0
10
20
10
20
30
Distance (H)
20
10
20
30
Distance (H)
(b)
Relative wind velocity (%)
100
80
20 m
60
Shelterbelt
40
20
0
10
0
10
Figure 5 Comparison of protected distances between two
types of shelterbelts: (a) wide shelterbelt (H 28 m, W 600 m)
and (b) relatively narrow shelterbelt (H 28 m, W 20 m).
Cd ¼
2D
Uc2 H
½3
where D is drag force (N), is air density (kg m 3), Uc is
reference wind velocity (m s 1), which is usually replaced
by UH, the wind speed at H.
Determination of Optical Porosity
Generally, when dealing with artificial windbreaks such
as salt fences, stubble barriers, and other windbreaks con
sisting of nonliving materials, the calculation of
is
In order to get an image with higher resolution, mono
chrome photographs should be taken during calm days
without too strong light to avoid excessive reflection from
the tree elements. Photographs should be taken as close as
possible to the shelterbelt for ensuring that the smallest
pores or tree elements are resolved by the computer
system. In practice, a mark at a certain height (almost
the same as the eye level of the photographer) is fixed on
one tree of the front row before taking the photograph.
Then, the photographer focuses the mark perpendicular
to the shelterbelt and takes one photograph. The photo
grapher moves 20 m left or right from the position and an
other two photographs are taken. The distance between
the photographer and the shelterbelt usually changes
according to the type of camera, for example, in a Ricoh
KR 10, f ¼ 50 mm with a zoom of 52 mm, the distance
between the photographer and the windbreak can be as
long as about 100 m.
Image Processing
In the laboratory, digital photographs (the photographic
negatives should be digitized) of shelterbelts can be divided
into two parts in the computer system, that is, the crown
and the trunk, according to the composition of shelterbelts.
Gray tones produced by light reflected from tree elements
are treated the same as transmitted light of the same inten
sity. The light intensity of each pixel could have a value
ranging between 0 (black) and 255 (white). It was possible to
assess each of the pixels in the photograph by selecting a
threshold value for the intensity of the transmitted light
between 0 and 255. Each pixel above or below the threshold
value was automatically assigned a binary value of false or
Wind Shelterbelts
(a)
(b)
(c)
(d)
(e)
(f)
473
(g)
Figure 6 Sketches of various cross-sectional shapes for shelterbelts: (a) rectangle, (b) windward triangle, (c) leeward triangle,
(d) notch, (e) gable roof/pitched roof (symmetry), and (f, g) gable roof/pitched roof (asymmetry).
number of pixels of the image and the tree elements,
respectively. Then, the optical porosity of a shelterbelt
can be obtained.
(%)
Relative wind speed (%)
100
80
Error Analysis of Optical Porosity
60
40
20
10
5
Rectangle
Triangle (leeward)
Notch
5
10
15
20
25
30
35 (H)
Triangle (windward)
Gable roof (symmetry)
Figure 7 The shelter effectiveness of various cross-sectional
shapes for model windbreaks (optical porosity 0.60, rows 7,
height 1.85 m).
true, respectively. A preliminary threshold value is selected
and the entire image is digitized. The digitized image
formed by pixels being either true or false can be displayed.
If the pores or tree elements are omitted, the process can be
repeated by replacing different threshold values until the
representative image is obtained. Once the representative
image is achieved, the software can provide the total
As the optical porosity of a shelterbelt is estimated from
photographs through image processing, the formation of
shelterbelt image is the centricity of the true shelterbelt.
Particularly, if the shelterbelt is composed of more than
two rows, the projective error and the contractive error
from the camera must be produced. Therefore, some errors
exist between the optical porosity ( p) estimated from
images and the optical porosity ( ) of true shelterbelts.
For these reasons, it is necessary and useful if the errors
can be deleted or limited during the image processing.
The errors of p can be attributed to the following
factors: (1) the characteristics of the formation of shelterbelt
image, depending on whether it can reflect the tree elements
truly or not; (2) the disturbance of the windbreak back
ground; and (3) random errors, including the errors
produced by the camera, situation of the sample shelterbelts,
and the measuring processes. As the shelterbelt image is a
centric projection, both projection and contractive errors
should exist between true tree shelterbelt and the imaged
shelterbelt. The projection error is defined as the negative
error, as it makes the optical porosity larger, and contractive
474
Wind Shelterbelts
error is defined as the positive error, as it makes the optical
porosity smaller. The errors caused from the disturbance of
the windbreak background can be eliminated through mod
ulating the threshold value and using an eraser during image
processing.
Therefore, the errors of optical porosity for shelter
belts can be summarized as
(a)
100
Ulee /Uopen (%)
80
60
40
20
p
¼
þ p þ c þ
½4
where p is projection error, c is contractive error, and
is random error.
Obviously, it is necessary to determine the errors of
p, c, and for estimating from p. The two errors of
projection and contractive can counteract each other to
some extent. Therefore, the problem should be concen
trated on . According to eqn [4], the expected values of
p and can be written as
E
p
¼
þ
½5
E ð Þ ¼ 0
½6
where E( p) is the expected value of optical porosity and
E() is the expected value of random error.
If the number of the shelterbelt samples for estimating
optical porosity is m, p1, p2, . . ., pm, are the values of
optical porosity estimated from image processing with
random errors. The estimation of E( p) and for a shel
terbelt can be obtained as
E
p
¼
¼
p
m
X
¼
m
1X
mi 1
pi
pi
p
i 1
½7
½8
where p is the mean value of optical porosity estimated
from the image without deleting errors.
0
– 10 – 5
0
5
10
15
20
25
30
35
Distance from windbreaks in times of windbreak height (H)
Single row β = 0.20
Single row β = 0.60
Single row β = 0.40
Single row β = 0.80
(b)
100
Ulee /Uopen (%)
80
60
40
20
0
– 10 – 5
0
5
10
15
20
25
30
35
Distance from windbreaks in times of windbreak height (H)
Single row β = 0.20
Single row β = 0.60
Single row β = 0.40
Single row β = 0.80
Figure 8 Wind profiles near the model shelterbelt in relation to
the composition of rows and the optical porosity, Ulee and Uopen
are wind speeds in open field and the leeward area (m s-1). (a) one
row, optical porosity
0.20, 0.40, 0.60, and 0.80; (b) two rows,
0.15, 0.26, 0.47, and 0.69.
Figure 8 shows the examples of wind profiles near
model shelterbelts with different optical porosities.
Wind Profiles near Shelterbelts
Wind profiles near the shelterbelts have been intensively
studied because of their importance related to shelter
effectiveness. The wind profiles near the shelterbelts
are influenced by many factors from both shelterbelt
structures and climatic conditions. Although shelterbelt
length, cross sectional shape, and width may influence
shelter effectiveness at leeward, tree height and shelter
belt porosity are more important. Obviously, shelter
effectiveness is proportional to tree height if other factors
are all the same. Therefore, porosity is considered as one
of the most important key shelterbelt structural indices in
determining the effectiveness in wind reduction.
Generally, the observations of wind reduction were car
ried out in both field and wind tunnel experiments.
Establishment and Management of
Shelterbelts for Wind Protection
Design of Shelterbelts
The designing of shelterbelts is determined by the objectives
of shelterbelt establishment. The goal of any shelterbelt is to
provide favorable microclimate conditions to landowners;
these conditions are obtained by altering wind patterns
directly or indirectly. Therefore, there are general principles
that apply to the majority of situations.
Shelterbelt orientation. Field shelterbelts should be
oriented perpendicular to the prevailing winds in order
to maximize the protected zone leeward. This orientation
minimizes the number of shelterbelts needed to protect a
given area. The orientation can vary within a certain
Wind Shelterbelts
Shelterbelt
90°
18H
30°
45°
21H
25H
Protected distance
Figure 9 Effects of angles of shelterbelt to wind direction on
effectively protected distance (H).
range according to the situations of the mechanical equip
ment and land use because the protected zone is not
reduced greatly when the angle between the shelterbelt
and the direction of prevailing winds is less than a certain
degree (e.g., 30 ) (Figure 9).
Intervals between shelterbelts. The protected zone of a
shelterbelt is limited, but the requirements for protection
usually exceed these limited influences. Therefore, a
system of properly oriented shelterbelts should be estab
lished. The number of shelterbelts that are required to
provide protection for a given area is directly related to
the average height of the tallest trees in the shelterbelt.
Intervals between adjacent shelterbelts are determined
by tree height, structure of the shelterbelt, and problem
wind velocity. Typically, the interval or distance between
adjacent shelterbelts should range from 10H to 25H,
475
depending on the extent of protection desired and the
size of the field.
Tree species selection and composition. Many factors such
as local climate and soil conditions, wind firmness,
features of tree species (height, crown spread, competi
tiveness), compatibility with the crops in the farmland
and pest problem, etc., determine the selection of tree
and shrub species for a shelterbelt. Species adaptability
is the most critical of all the selection factors. The most
desirable tree species for a variety of shelterbelt uses
should be relatively free of diseases and pest problem,
with a narrow crown and deep roots, a certain potential
height and long lived. Generally, native tree species are
usually a good choice.
Species combination. Most shelterbelts consist of one tree
species, but mixed shelterbelts can make use of the site
conditions efficiently, improve the stability and pest and
disease resistance, and provide good shelterbelt structures
if the mixed tree species are combined reasonably.
However, there are few mixed tree shelterbelts in practice
because of the difficulties in planting and managing.
Table 2 lists the sample results of mixed shelterbelts.
The mixed tree species include Ulmus pumila, Populus
xiaozhuanica, Salix matsudana, Pinus sylvestris var. mongolica,
and shrubs Amorpha fruticosa, Lespedeza bicolor. Five mixing
patterns, that is, mixing between trees in one row, row
mixing (dissymmetry and symmetry), mixing among
segments, and mixing among shelterbelts, are formed.
Shelterbelt structure/optical porosity. According to the
results obtained on the basis of wind reduction experi
ments, an optimum optical porosity exists. Shelterbelts
with low or high optical porosity are generally ineffective
for wind protection. There is considerable variation in the
Table 2 Arrangements of mixing patterns for shelterbeltsa
Mixed patterns
Arrangement types (4 rows)
Stem density
Mixing between trees in one row
1. Salix – Populus – Salix – Populus
2. Ulmus – Populus – Ulmus – Populus
Mixing in rows (dissymmetry)
3. Salix – Populus – Salix – Populus
4. Ulmus – Populus – Ulmus – Populus
2.0 m 2.0 m
2.0 m 2.0 m
Mixing in rows (symmetry)
5. Salix – Salix – Pinus – Pinus
6. Pinus – Pinus – Populus – Populus
7. Ulmus – Ulmus – Populus – Populus
8. Ulmus – Populus – Populus – Ulmus
9. Salix – Populus – Populus – Salix
10. Populus – Populus – Salix – Salix
Mixing among segments
Mixing among shelterbelts
11. Populus (500 m) – Pinus (500 m) – Ulmus (500 m)
12. Populus (500 m), Pinus (500 m)
2.0 m 2.0 m
2.0 m 2.0 m
2.0 m 2.0 m
2.0 m 2.0 m
2.0 m 2.0 m
2.0 m 2.0 m
2.0 m 2.0 m
2.0 m 2.0 m
2.5 m 2.5 m
2.0 m 2.0 m
a
All the shelterbelts were composed of 4 rows with shrubs. The shrubs were planted aside the shelterbelts with density of 1 m 1 m. 10 years later
(in 2002), the mixed shelterbelts of type (5) and type (6) failed because of the growth difference between Pinus and the deciduous species. The other
shelterbelts such as type (11) and type (12), which mixed with Pinus, Populus, and Ulmus, succeeded. Among the mixed shelterbelts of deciduous
tree species, type (8) and type (9) exhibited better patterns because the relatively fast growing species (Populus) were planted in the inner rows, and
the relatively slow growing species (Salix and Ulmus) were planted in the outer rows. This arrangement made full use of the edge effect of the
shelterbelts, i.e., the tree species growing more slowly with age should be planted in outside rows and those growing faster should be planted in the
inner rows for mixed shelterbelts.
476
Wind Shelterbelts
results of optimum optical porosity. The variation might
be caused by the differences in shelterbelt structures, the
effects of thermal instability in the field, the type of
instruments used, and the method used to determine the
optical porosity. Despite these differences, most studies
have suggested that better sheltering is obtained by shel
terbelts with porosities ranging from 0.20 to 0.50. The
characteristics of foliages and branches of individual tree
or shrub are important in determining the shelterbelt
porosity or optical porosity. Therefore, porosity or optical
porosity can be modified by changing tree species or the
spacing within and between tree rows; however, their
effects on shelterbelt porosity are not well studied.
Spacing within and between rows. Spacing within and
between tree rows varies with regions, tree species,
desired density or optical porosity, and the number of
rows. Generally, within the row spacing is as follows:
shrubs, 1 m; trees, from 2.0 to 3.0 m according to the row
number of the shelterbelts.
Planting arrangements. The planting arrangements of
trees and shrubs in a shelterbelt determine the shelterbelt
structure, and further influence the shelter efficiency.
Generally, there are three types of planting arrangements
(i.e., rectangle, triangle, and random). For maximizing the
benefits of shelterbelts, the planting arrangements of mul
tirow shelterbelts should be in triangle patterns. This is
because the triangle planting arrangement of trees in a
shelterbelt is favorable to the shelterbelt structure asso
ciated with shelter efficiency.
Length and width. The influence of the shelterbelt length
on shelter effectiveness lies on the ends of the shelterbelt
(Figure 4). For this reason, the length of a shelterbelt
should be at least 10 times as long as its mature height for
minimizing such an influence. The shelterbelt width influ
ences the shelter effectiveness through changing the
porosity of the shelterbelt. In wide or multirow shelterbelts,
however, tree competition may weaken the stability of the
shelterbelt stand; in order to solve this problem, mixed
shelterbelts as described in Table 2 (i.e., type 8 and type
9) are recommended. The problem of the single or two
row shelterbelt is the potential damage of continuity when
the shelterbelt loses some of trees. Therefore, the shelter
belt width should be as narrow as possible under the
condition of high preserved rate of trees, that is, the smal
lest proportion of land devoted to tree shelterbelts and the
largest complete protection for the remaining land.
Competition zone. The most common negative com
ments concerning tree shelterbelts are related to the
impact of competition between trees and the adjacent
crops, especially under conditions of limited moisture.
The extent of competition varies greatly with crop spe
cies, tree species, geographic location, and weather
conditions. Some types of competition can be reduced
by root pruning, that is, cutting of the lateral tree roots
extending into crop field.
Management of Shelterbelts
Management of tree shelterbelt may be similar to gen
eral forest management. But the substaintial difference
between shelterbelt and forest management is the man
agement purpose. The major goal of shelterbelt
management is to obtain shelter effectiveness. The
care of a shelterbelt is a continuous responsibility, for
example, intensive shelterbelt management begins soon
after planting in China and Russia. The first thinning is
generally carried out within 4–10 years followed by a
release cutting designed to keep the best trees from
undesirable competition. Shelterbelts in the American
Great Plains are generally located in areas where cli
matic and soil conditions are not conducive to the
natural occurrence and regeneration of trees and shrubs.
The management of shelterbelt is to keep all grass and
weeds out of shelterbelt until crown closure (by the time
when the shade is too dense for weed to grow). The
success of a shelterbelt establishment depends not only
on the initial design, species selection, and site prepara
tion, but also on the subsequent care and the
management level. There are numerous silvicultural
techniques that can be used to maintain a shelterbelt
well beyond the life expectancy of the original tree
planting. Generally, shelterbelt systems can provide
more sheltering benefits than an individual shelterbelt.
The spatial patterns and the future development of
shelterbelt systems can be more easily and clearly
exhibited in the landscape level. Thus, landscape con
sideration can provide more important references for
the management of shelterbelts system.
Further Reading
Brandle JR, Hodges L, and Wight B (2000) Windbreak practices.
In: Garrett HE, Rietveld WJ, and Fisher RF (eds.) North American
Agroforestry: An Integrated Science and Practice, pp. 79 118.
Madison: American Society of Agronomy.
Caborn JM (1965) Shelterbelts and Microclimate. London: Faber and
Faber.
Cao XS (1983) Shelterbelt for Farmland. Beijing: Chinese Forestry Press,
(in Chinese).
Ennos AR (1997) Wind as an ecological factor. Trends in Ecology and
Evolution 12: 108 111.
Everham EM (1995) A comparison of methods for quantifying
catastrophic wind damage to forest. In: Coutts MP and Grace J
(eds.) Wind and Trees, pp. 340 357. Cambridge: Cambridge
University Press.
Heisler GM and DeWalle DR (1988) Effects of windbreak structure on
wind flow. Agricultural Ecosystems and Environment 22/23: 41 69.
Jiang FQ, Zhu JJ, Zeng DH, et al. (2003) Management for Protective
Plantation Forests. Beijing: China Forestry Publishing House.
Kenney WA (1987) A method for estimating windbreak porosity using
digitized photographic silhouettes. Agricultural and Forest
Meteorology 39: 91 94.
Loeffler AE, Gordon AM, and Gillespie TJ (1992) Optical porosity and
windspeed reduction by coniferous windbreaks in Southern Ontario.
Agroforestry Systems 17: 119 133.
Peltola H, Kellomaki S, Kolstrom T, et al. (2000) Wind and other abiotic
risks to forests. Forest Ecology and Management 135: 1 2.
Wind Shelterbelts
Ruck B, Kottmeier C, Matteck C, Quine C, and Wilhelm G (2003)
Preface. In: Ruck B, Kottmeier C, Matteck C, Quine C, and
Wilhelm G (eds.) Proceedings of the International Conference Wind
Effects on Trees, pp. iii. Karlsruhe, Germany: Lab Building,
Environment Aerodynamics, Institute of Hydrology, University of
Karlsruhe.
Zhou XH (1999) On the Three Dimensional Aerodynamic Structure of
Shelterbelts. PhD dissertation, Graduate College at the University of
Nebraska.
477
Zhou XH, Brandle JR, Takle ES, and Mize CW (2002) Estimation of the
three dimensional aerodynamic structure of a green ash shelterbelt.
Agricultural and Forest Meteorology 111: 93 108.
Zhu JJ, Gonda Y, Matsuzaki T, and Yamamoto M (2002) Salt distribution
in response to optical stratification porosity and relative windspeed in
a coastal forest in Niigata, Japan. Agroforestry Systems
56(1): 73 85.
Zhu JJ, Matsuzaki T, and Jiang FQ (2004) Wind on Tree Windbreaks.
Beijing: China Forestry Publishing House.
This page intentionally left blank
INDEX
NOTES:
Cross reference terms in italics are general cross references, or refer to subentry terms within the main entry (the main entry
is not repeated to save space). Readers are also advised to refer to the end of each article for additional cross references not
all of these cross references have been included in the index cross references.
The index is arranged in set out style with a maximum of three levels of heading. Major discussion of a subject is indicated
by bold page numbers. Page numbers suffixed by t and f refer to Tables and Figures respectively. vs. indicates a comparison.
This index is in letter by letter order, whereby hyphens and spaces within index headings are ignored in the alphabetiza
tion. Prefixes and terms in parentheses are excluded from the initial alphabetization.
A
aapamires, 335
Abies spp.
alpine ecosystems, 151, 159t
boreal forest ecosystems, 181
Mediterranean ecosystems, 320t,
321t, 325
abiotic environments
agroecosystems, 146
coral reefs, 205, 205f
ecosystem dynamics, 21
indirect interactions, 86
landfills, 305
aboriginal populations, 238
aboveground net primary production
(ANPP), 409f
Abronia maritima, 243f
Acacia auriculifornis, 306
Acacia cambegei, 396
Acacia confusa, 306
Acacia georginae, 396
Acacia harpophylla, 396
Acacia mangium, 306
Acacia shirleyi, 396
Acacia spp.
dune ecosystems, 246
floodplain ecosystems, 259
Mediterranean ecosystems, 320t, 321t, 326
postclosure landfills, 306
savanna ecosystems, 396, 398, 404
Acacia tortilis, 398
Acanthamoebidae, 288t
Acanthaster planci, 204
Acanthina angelica, 376
Acanthochitonidae, 288t
Acanthuridae, 203–204, 288t
Acer saccharum, 420–421
Acer spp.
Europe, 261
temperate forest biomes, 417, 420
United States, 262
Achnanthaceae, 288t
Achnanthes spp., 370f
freshwater ecosystems, 369–370
lake ecosystems
freshwater lakes, 273
peatlands, 335
Acinetobacter spp., 170
Acmaeidae, 288t
Acnathophora spicifera, 293f
acorn barnacles, 83
Acremonium spp., 88
Acrochaetiaceae, 288t
Acropora cervicornis, 210
Acropora hyacinthus, 207f
Acropora palmata, 210
Acropora spp., 202
Acroporidae, 288t
acrotelm, 332–333, 337
Actiniidae, 288t
Actinobacteria, 169
Actinodiscidae, 288t
Actinophyridae, 288t
basic concepts, 168f, 176
benefits, 179
biological processes, 177
continuous flow systems, 177, 177f
simulation models, 179
nitrogen removal, 178
nutrient removal capacity, 178
phosphorus removal, 179
acyclic networks, 55
adaptability
orientation theory, 122, 124,
124t, 125f
addax (Addax nasomaculatus), 239
Adelaide, Australia, 462
Adelie penguins, 342
Adelocosa anops, 193, 193f
adenosine diphosphate (ADP)
microbial processes, 169, 169f
adenosine triphosphate (ATP)
exergy, 137, 138t
microbial processes, 169, 169f
Adenostoma fasciculatum, 195, 320t
Adh gene, See alcohol dehydrogenase (ADH)
adominance hierarchies, 103
aeolian dust, 454
estuarine ecosystems, 249
nitrogen cycle
anammox process, 171f, 172, 178
sulfur cycle, 453
Africa
alpine forests, 157, 159t, 160f
botanic gardens, 184t, 186
desert environments
animals, 229
major deserts, 223t
plants, 228
floodplain ecosystems, 257t, 259
grasslands, 406, 411
intertidal area research, 378
saline/soda lakes, 381, 381f, 382, 382f
savanna ecosystems, 394, 395f,
396f, 396–397
temporary water bodies, 429t
tropical ecosystems, 440, 442
wetland ecosystems, 281
African fish eagles, 259–260
African lungfish, 218
African penguin, 457
African springhares, 230
Afrocarpus, 320t, 326
Agabus spp., 437
agamid lizards, 230
Agaricia spp., 202–203
Agavaeae, 230
Agelasidae, 288t
agroecosystems, 145–150
forest plantations, 264–270
anthropogenic impacts, 146f
basic concepts, 145
abiotic constraints, 146
biodiversity, 146f, 148
box diagram, 146f
crop production, 146
migration effects, 147
pest control methods, 147
weeds, 146
climate effects, 146
cropland resources
bee pollination, 28
shelterbelts/windbreaks, 469
decision-making processes, 148, 149f
desert stream ecosystems, 221
ecological effects, 148
United States, 261
grazing effects, 147
historical background, 237
hunter-gatherer societies, 238
desert environments, 237
historical background, 237
mangrove forests, 313–314
natural systems, 14
Neolithic era, 238
nutrient resources, 146
479
480
Index
agroecosystems (continued )
pest control
crop production, 147
postclosure landfills, 304
temperate forest biomes, 425
forest plantations, 264–270
basic concepts, 264
conservation strategies, 267
ecological effects, 267
economic factors, 265, 265f
exotic species, 267
global distribution, 264, 264f, 265f
global production, 265f
influential factors
government policies, 266
production costs, 266
timber prices, 266
management strategies, 269–270
timber production, 265
Agropyron junceiforme, 242
Agropyrum junceum, 243f
Agulhas Current, 455, 456f
Aiptasia pallida, 294f
Aiptasiidae, 288t
air quality
urban environments, 465, 465t
Aizoaceae, 235, 320t
Alabama (USA), 266
Alaria marginata, 18, 19f
Alaska (USA), 181
albedo
desert environments, 216
Albizia lebbeck, 306
Albuquerque, New Mexico (USA), 112
Alces alces, 182
See also moose
alcohol dehydrogenase (ADH)
floodplain ecosystems, 257
Alcyoniidae, 288t
alderflies, 372–373
alder trees, 181, 424f
Alexander the Great, 183
Alfisols, 269, 395–396, 421
algae
agroecosystems, 148
algal blooms
lagoon ecosystems, 299
cell structure, 166
coral-algal mutualism, 201, 205
coral reef microcosm, 288t
desert environments
desert crusts, 228
poikilohydry, 230
stream ecosystems
colonization, 219f, 219f
energetics, 218
life-history patterns, 216
nutrient dynamics, 219
successional patterns, 215
eastern African soda lakes, 382
estuarine ecosystems, 250
exergy calculations, 134t
Florida Everglades mesocosm project, 293f
food webs, 383
intertidal environments
characteristics, 375
photographic views, 375f
wave surge effects, 375
lake ecosystems
food webs, 383
large-scale whole ecosystem experiments, 21
saline/soda lakes
characteristics, 381
Dead Sea, 384
eastern African soda lakes, 382
Mono Lake (USA), 383
polar ecosystems, 340, 342
saline/soda lakes
characteristics, 381
Dead Sea, 384
eastern African soda lakes, 382
Mono Lake (USA), 383
salt marsh ecosystems, 387, 388, 389
standing crop biomass-production rate
relationship, 18, 19f
stream ecosystems
biotic diversity patterns, 359f
characteristics, 370, 370f
energy sources, 351
taxonomy, 370
temporary water bodies, 433
upwelling ecosystems, 452
See also cyanobacteria
Algeria, 223t
algorithms
animat algorithms, 126, 127f
Alice Springs Desert Park, Australia, 189
alkaline saline lakes, 383
alkalinity, 369
allelopathy
chaparral ecosystems, 197
coral reefs, 202
allies, 227–228
Alligator mississippiensis, 429t
alligators
marshes, 275, 278, 279f, 280
temporary water bodies, 429t
Allocasuarina spp., 320t
Allogromiidae, 288t
alluvial fans, 223
alluvium, 214
Alnus glutinosa, 306
Alnus rubra, 424f
Alnus spp., 151, 181, 335
Alopex lagopus, 447
Alpheidae, 288t
alpine ecosystems, 150–156
adaptive processes, 151, 152f
alpine forests, 156–165
abiotic environments
aboveground, 159, 165
belowground, 162
altitude effects, 162
biogeography, 157, 159t, 160f
characteristics, 156, 157f, 161f
facilitation-competition comparisons,
161f, 165
influential factors, 164t
microclimates, 162
treeline ecotone, 162
treeline formation mechanisms, 163,
164t, 165
anthropogenic impacts, 155
biodiversity, 153, 154f
carbon dioxide (CO2) concentrations, 155
climate change effects, 155
climatic boundaries, 150, 151f
ecosystem processes, 153
flowering plants, 153, 154f
high-elevation treeline, 151, 152f, 156, 157f,
161f
land-use change, 155
mismanagement concerns, 155
nutrient cycling, 153
semidesert environments, 153, 153f
tundra ecosystems, 17f, 444, 446f
water consumption, 153, 153f
Alps, 157, 159t, 160f
Altai Mountains, 159t, 160f
self-organization, 103
aluminum (Al)
elemental concentrations, 10t
stream ecosystems, 369
Amazon River/Amazon Basin
blackwater systems, 256f
floodplain ecosystems, 262
temperature, 368–369
temporary water bodies, 431
tropical rainforests, 440, 442
upwelling ecosystems, 454
wetland ecosystems, 275, 281
Amazon water lilies, 277f, 277–278
Ambrosia dumosa, 224f, 233, 236f, 237f
Ambunguipedidae, 288t
American bulrush, 276f
American chestnut tree, 420
amino acids
exergy calculations, 132
temperate forest biomes, 424–425
ammensalism, 81, 84
ammonia (NH3)
assimilation processes, 171, 171f
ammonidscidae, 288t
ammonium (NH4)
anaerobic ammonium oxidation, 171f, 172
phytoplankton, 452–453
saline/soda lakes, 383
temperate forest biomes, 424–425
upwelling ecosystems, 451
Ammophila arenaria, 242, 243f, 246
Ammophila breviligulata, 243f
Ammophila spp., 246
amoebae, 148
Amorpha fruticosa, 475
amphibians (Amphibia)
boreal forest ecosystems, 182
desert environments, 229–230
exergy calculations, 134t
floodplain ecosystems, 276
marshes, 280
Mediterranean ecosystems, 322, 322t
postclosure landfills, 307
riparian systems, 348
salt marsh ecosystems, 387
temperate forest biomes, 423
temporary water bodies, 431, 431f, 432, 433,
436–437
urban environments, 464–465
amphicarpic plants, 233
Amphileptidae, 288t
Amphinomidae, 288t
amphipods (Amphipoda)
cave habitats, 193
Florida Everglades mesocosm project, 294f
intertidal environments, 376
stream ecosystems, 371, 372f
amphisbaenians, 230
Amphithoidae, 288t
Ampithoe spp., 294f
amplification of ribosomal DNA restriction
analysis (ARDRA), 169
Anabaena spp.
stream ecosystems, 371
anadromous organisms, 373
analysis of variance (ANOVA), 85
Anamixidae, 288t
anammox process
activated sludge systems, 178
microbial processes, 171f, 172
upwelling ecosystems, 453
Anaperidae, 288t
Anastatica hierochuntica, 233, 233f
anchovies
climate change effects, 458, 458f
upwelling ecosystems, 455, 455f
Ancylidae, 371, 372f
Andean flamingoes, 225f
Anderson, Ray, 14
Andes Mountains, 157, 159t, 160f, 225f, 365f
Androcymbium, 323
Andropogon brevifolius, 399
Andropogon gayanus, 404
Andropogon gerardii, 410
Andropogon spp., 396
Index
anemones
exergy calculations, 134t
Angola, 429t
animals
African floodplain ecosystems, 259–260
alpine ecosystems, 153, 157
animal husbandry., See agroecosystems;
farming systems; livestock operations
Australian floodplain ecosystems, 261
biomass-population density relationships, 137
boreal forest ecosystems, 182
breeding behaviors, 101
cell structure, 166
chaparral ecosystems, 196
desert environments
characteristics, 229
drought tolerance, 230
endangered species, 239, 240f
excretion, 230
hunter-gatherer societies, 238
water uptake, 231
dune ecosystems, 246
ecological network analysis (ENA), 73t
desert environments, 239, 240f
grasses, 240
invasive species, 240
predators, 239
estuarine ecosystems, 248
exergy calculations, 134t
floodplain ecosystems
Africa, 259–260
Australia, 261
marshes, 276
South America, 262–263
United States, 261
grasslands, 410
human-environment interactions, 462,
463t, 464f
indirect interactions, 85
invasive species, 391
landfills, 305, 307
mangrove forests, 310
marshes, 278, 279f
Mediterranean ecosystems, 322, 322t, 329
polar ecosystems, 340
riparian systems, 348
salinity
saline/soda lakes, 381
salinity tolerances, 382, 387
salt marsh ecosystems, 386, 387, 391
savanna ecosystems, 400
shelterbelts/windbreaks, 470
South American floodplain ecosystems,
262–263
temperate forest biomes, 422
temporary water bodies
boreal/temperate forests, 432
drying adaptations, 434
Mediterranean ecosystems, 432
tropical rainforests, 431
tundra ecosystems, 432
tundra ecosystems, 446
urban environments, 462, 463,
463t, 464f
See also invertebrates; vertebrates
animat algorithms, 126, 127f
Anisolabis howarthi, 192, 192f
Anisoptera, 372f
annelids (Annelida)
exergy calculations, 134t
Florida Everglades mesocosm project, 294f
intertidal environments, 376
stream ecosystems, 356–357, 359f, 371, 372f
temporary water bodies, 433
upwelling ecosystems, 453
Anomola, 433
Anostraca, 433
Antarctica
desert environments, 223t
desert stream ecosystems, 216
polar ecosystems, 339–342
temporary water bodies, 432
Antarctic Circumpolar Current, 450
Antarctic Polar Front, 454
antelope, 259–260
Anthothelidae, 288t
anthropogenic activities
agroecosystems, 146f
alpine ecosystems, 155
coral reefs, 205, 205f, 210f
desert environments, 238
desert stream ecosystems, 221
ecosystem component interactions, 10,
11t, 25
indirect interactions, 87
intertidal environments, 377
land-use change
alpine ecosystems, 155
savanna ecosystems, 403
marshes
dam construction, 279
draining effects, 279
food webs, 280
general discussion, 279
nutrient resources, 279–280
road networks, 280
salt marsh ecosystems, 390
savanna ecosystems, 398f, 403
stream ecosystems
general discussion, 362
temperate forest biomes, 420, 425
tundra ecosystems, 449
urban environments, 462, 463t
wildfires, 399
anthropoid apes, 134t
ant nests, 429t
Antofagasta, Chile, 379f
ants
desert environments, 229, 230
exergy calculations, 134t
group behaviors, 103
organizational structure, 109
savanna ecosystems, 400
stigmergy, 101, 102f
task allocation, 109, 110f
anurans, 348
temporary water bodies, 433
Anurida maritime, 387
Aphanizomenon spp., 371
Aphelocoma californica, 199
aphids
agroecosystems, 148
chaparral ecosystems, 196
Apis mellifera, 28
Apluda mutica, 397
Appalachian mountain region (USA), 159t, 160f
apparent competition, 82, 83f
apple trees, 28
usability of ecology, 12–16
Apseudidae, 288t
mangrove forests, 313–314
saline/soda lakes, 384
aquatic environments
aquatic-terrestrial interactions, 350
body size-abundance distributions, 47
caves, 190
desert environments, 214
indirect interactions, 85
swamps, 416
trophic transfer efficiency, 19
wetland ecosystems, 275
See also intertidal environments; marine
ecosystems; oceans; riparian systems
aquatic moths, 373
aquatic organisms
estuarine ecosystems, 248
intertidal environments
algae, 375, 375f
mobile organisms
general discussion, 376
invertebrates, 376
vertebrates, 376f, 377
sessile invertebrates, 376
mangrove forests, 310
riparian systems, 348
temporary water bodies, 432
Arabian Desert, 223t, 224f
arable land
definition, 145
See also agroecosystems
arachnids (Arachnida)
agroecosystems, 148
cave habitats, 191, 193, 193f
salt marsh ecosystems, 386
temperate forest biomes, 423
Aral Lake, 273
Aral Sea, 381
Araucaria araucana, 320t
Arbutus, 320t
Arbutus spp., 325
Archaea
cell structure, 166, 167t
exergy calculations, 134t
methanogenesis, 337
saline/soda lakes, 381–382
Arcidae, 288t
arctic environments
desert environments, 222
polar ecosystems, 339–342
Antarctica, 341f
Arctic region, 340f
characteristics, 339–340
climate change effects, 341
desert environments, 222
temporary water bodies, 429t, 432
tundra ecosystems, 444, 446f
arctic foxes, 340–341, 447
arctic hare, 446–447
Arctic Ocean, 340–341
Arctic region, 339–342
Arctic terns, 458
arctic vetch, 446f, 448
arctic willow, 445f, 446
Arctostaphylos spp., 195, 320t
Argania, 321t
Argentina
desert environments, 223t
Argestidae, 288t
Argid soils, 226
argon (Ar), 10t
Argonemertes dendyi, 193
arid environments
intermittent streams, 428t, 433f
Mediterranean ecosystems, 319, 321t
physical characteristics, 214
semiarid environments, 153
stream ecosystems, 214–221
temporary water bodies, 428t, 429t,
432, 433f
vegetation, 218
Aridisols, 226
aridopassive plants, 233–234
Aristida spp., 396
Aristotle, 183
Aristotle’s lantern, 203
Arizona (USA), 184t
Army Corps of Engineers (US), 391
arroyos, 225
Artemia monica, 383
Artemia spp.
saline/soda lakes, 383, 384
Artemisia spp., 320t
481
482
Index
Artemisia tridentata, 233
Arthrocnemum spp., 387
arthropods (Arthropoda)
cave habitats, 191, 193
desert environments, 229, 230, 232
Florida Everglades mesocosm project, 294f
intertidal environments, 376
landfills, 305
salt marsh ecosystems, 386, 387
stream ecosystems, 371
temperate forest biomes, 423–424
temporary water bodies, 433
urban environments, 463
Arthrospira fusiformis, 382, 384
Arthrospira spp., See Spirulina spp.
artificial containers, 428t
artificial life, 102
artiodactyls, 377
Arundinaria spp., 397
ascendancy, 4, 57–64
autocatalysis, 57–58
average mutual information (AMI)
uncertainty analysis, 59, 59f
background information, 57
basic concepts, 57
biomass inclusive ascendancy, 62
calculation techniques, 60
complex systems, 57
conditional probabilities, 57, 58t
calculation techniques, 60
ecosystem applications, 63
limiting factors, 62
overhead calculations
basic concepts, 61
calculation techniques, 61
exports, 61
imports, 61
redundancy analysis, 62
respiration, 61
total system throughput (TST)
matrix models, 58
Ascidiacea, 288t
Ascomycetes spp., 227–228
asexual reproduction
See also sexual reproduction
Ashby, WR, 99
Asia
alpine forests, 157, 159t, 160f
botanic gardens, 184t
desert environments
animals, 229
major deserts, 223t
plants, 228
floodplain ecosystems, 260
grasslands, 406
saline/soda lakes, 381
savanna ecosystems, 394, 396f, 397
temperate forest biomes, 425–426
tropical ecosystems, 441
wetland ecosystems, 281
aspen, 182, 421
Aspidiscidae, 288t
Aspidosiphonidae, 288t
Assam River, 433
assemblages, 104
asses, 239, 411
temperate forest biomes, 421
Asteraceae
chaparral ecosystems, 199
desert environments, 229
Mediterranean ecosystems, 320t
Astragalus spp., 320t, 446f
Astrorhizidae, 288t
Atacama Desert, 223t, 224f
atelechory, 233
Athericidae, 372f
Athrotaxis spp., 159t
Atlantic Ocean
ocean currents, 450
atmosphere
carbon dioxide (CO2) concentrations
alpine ecosystems, 155
circulation patterns, 459, 459f
clouds
hydrologic cycle, 365f
elemental concentrations, 10t
indirect interactions, 87
Atriplex leucophylla, 243f
Atriplex spp., 387
Audouinella violacea, 293f
aufwuchs, 370
Augusta Creek, Michigan (USA), 352f, 353f
Australia
alpine forests, 157, 159t, 160f
botanic gardens, 184t
coral reefs, 206, 207f
desert environments
aboriginal populations, 238
animals, 229
major deserts, 223t
plants, 229
floodplain ecosystems, 257t, 261
intertidal area research, 378
lagoon ecosystems, 296–297, 297f, 300f
mangrove forests, 311, 313
Mediterranean ecosystems
biodiversity patterns, 322, 322t
biogeographical evolution, 323
characteristics, 319, 320t
climatic stress, 327
convergence studies, 325
soils, 321
wildfires, 328
salt marsh vegetation, 386t
savanna ecosystems, 394, 395f, 396, 396f, 399f
temporary water bodies, 429t
tropical ecosystems, 441
urban environments, 462
Australian gum trees, 347
Austria
definition, 6
community ecology
general discussion, 437
interspecific interactions, 437
richness theories, 437, 438t
drying adaptations
avoidance strategies, 435, 435f
diapause, 434, 436, 436t
dormancy, 435
general discussion, 434
physiological ecology, 435, 435f
population dynamics
bet hedging strategies, 436
complex life histories, 437
dispersal mechanisms, 437
life-history strategies, 436
autocatalysis, 4, 41–43
centripetal actions, 42, 42f, 43f
competition, 42–43, 43f
definition, 41
food webs, 41, 42f
schematic diagram, 42f
selection processes, 41
systems ecology, 41, 57–58
automata theory, 101
characteristics, 17, 167t
desert stream ecosystems, 219
tropical rainforests, 441
upwelling ecosystems, 452–453, 455f
Avena sativa, 405f
Avena spp., 320t, 327
avens, 429t
average mutual information (AMI)
ascendancy, 59, 59f
uncertainty analysis, 59, 59f
Avicennia marina, 313
avocados, 28
Axinellidae, 288t
Azorella compacta, 154f
B
Baccharis halimifolia, 320t
Bacillariophyta, See diatoms
bacteria
cell structure, 166, 167t
community structure, 168f
dune ecosystems, 246
eastern African soda lakes, 382
electron-donor/electron-acceptor regimes, 167t
estuarine ecosystems, 249, 250, 251
exergy calculations, 134t
filamentous bacteria, 168f, 177
polyphosphate-accumulating bacteria,
170, 170f
saline/soda lakes, 381–382
stream ecosystems, 356
sulfur cycle
sulfate-reducing bacteria
peatlands, 338
temperate forest biomes, 423–424
temporary water bodies, 433, 434
upwelling ecosystems, 453
Bacteroidetes spp., 169
Bactrian camels, 239
Badain Jaran Desert, China, 241
badlands, 225
Bafut Botanic Garden (Cameroon), 183, 189
Bahamas
coral reefs, 212f, 213
baias, 429t
Bailey ecoregion, 419t
Bairdiidae, 288t
bajadas, 223, 224f
Balanidae, 377
Balanus balanoides, 378–379
Balkhash (Russia), 381
Baltic Sea
cycle analysis, 51, 52f
bamboo stumps, 429t
Bangladesh
floodplain ecosystems, 260
mangrove forests, 313
tropical rainforests, 442
Banksia, 320t
Banks Island, Northwest Territories, Canada,
446f, 447
barchan dunes, 241
bark beetles
boreal forest ecosystems, 182
forest plantations, 269
bark (floodplain ecosystems), 257
barkhan dunes, 224–225
barnacles
competition studies, 83
intertidal environments
characteristics, 376
photographic views, 375f
vertical zonation patterns, 377
stable isotope analysis, 21
barracuda, 207–208
barrier islands, 247
Barro Colorado Island, Panama, 440f
basal swelling, 258
Bassin d’Arcachon, France, 241
Batillipedidae, 288t
Batis maritima, 386t, 387
bats
cave habitats, 191
pollen vectors, 24, 24f
postclosure landfills, 307
riparian systems, 348
temperate forest biomes, 423
bdelloid rotifers, 435
Index
bears
boreal forest ecosystems, 182
seed dispersal, 199
temperate forest biomes, 423
beavers
aquatic wetlands, 275
riparian systems, 348
Bedouin culture, 238
bedrock flats, 224, 224f
beech trees
alpine forests, 158
temperate forest biomes, 417
bee-eaters, 259–260
bees
exergy calculations, 134t
pollen vectors, 24, 24f, 28
beetles
boreal forest ecosystems, 182
cave habitats, 193
desert environments, 229, 235
exergy calculations, 134t
flightless beetles, 448
forest plantations, 268
grasslands, 410
pollen vectors, 24, 188
salt marsh ecosystems, 387
stream ecosystems, 373
temperate forest biomes, 423
temporary water bodies, 437
tundra ecosystems, 448
Belastomatidae, 372f
Belize, 184t
Benguela Current, 450, 450f, 455,
456f, 458f
benthic macroinvertebrates, 371, 372f
Benyus, Janine, 14
Beringia, 444
See also tundra ecosystems
Berlin, Germany, 462
Bertalanffy, L von, 99
best mangement practices (BMPs), 269
bet hedging strategies, 436
Betula spp., 151, 159t, 181, 335
bicarbonate
saline/soda lakes
eastern African soda lakes, 382
ionic composition, 381
Mono Lake (USA), 383
billabongs, 261, 429t
billfish
trophic transfer efficiency, 19
indirect interactions, 89
biochemical oxygen demand (BOD5)
activated sludge systems, 178
constructed wetlands
horizontal subsurface flow (HSSF)
treatment efficiency, 176
lagoon ecosystems, 297
wastewater treatment, 176
biocoenosis
agroecosystems, 146f, 148
alpine ecosystems, 153, 154f
body size-species distributions, 48
boreal forests, 182
botanic gardens, 188
desert environments
animals, 229
convergent evolution, 230
general discussion, 227
microorganisms, 227
plants, 228, 229f
stream ecosystems, 218t
ecosystem function-biodiversity relationship,
22, 23f
ecosystem-level diversity, 28, 28f
emergence/emergent properties, 95t
forest plantations, 267
exergy, 131, 131f, 135
insects (Insecta), 148
intertidal environments, 374, 375f
mangrove forests, 310
marshes, 278, 279f
Mediterranean ecosystems, 320t,
322, 322t, 328
orientation theory, 124t
plants, 148
polar ecosystems, 339–342
richness measurements
alpine ecosystems, 153, 154f
boreal forests, 182
forest plantations, 267
temporary-water populations, 437, 438t
riparian systems, 348
saline/soda lakes, 381–382
shelterbelts/windbreaks, 470
soils, 148
species diversity
biotopes
Shannon index, 60
tundra ecosystems, 444
urban environments, 463, 464f
tropical ecosystems, 442
tundra ecosystems, 444
biodynamic farming, 147
biofilms
community characteristics, 168
community structure, 168f, 177
intertidal environments, 375
nutrient cycling, 220
salt marsh ecosystems, 388, 389
wastewater treatment, 175
indirect interactions, 85
biogeochemical cycles
biological cycles
alpine ecosystems, 153, 155
boreal forests182
peatlands, 330, 332f, 338
savanna ecosystems, 394
mangrove forests, 311, 312f
riparian systems, 349
salt marsh ecosystems, 389
savanna ecosystems, 394, 403, 403t
temperate forest biomes, 425
total organic carbon (TOC), 349
ecosystem ecology, 9, 10t
indirect interactions, 86
iron cycle
anammox process
activated sludge systems, 178
microbial processes, 171f, 172
upwelling ecosystems, 453
denitrification
desert stream ecosystems, 220, 220f
grasslands, 409–410
mangrove forests, 311, 312f
marshes, 278
temperate forest biomes, 424–425
urban environments, 462–463
mangrove forests, 311, 312f
urban environments, 462–463
microbial processes, 453
sulfate-reducing bacteria
peatlands, 338
bioindicators
See also organic farming
Biological Diversity Convention, See Convention
on Biological Diversity (CBD)
biological systems
desert stream ecosystems, 216, 218t
emergence/emergent properties, 93, 95t
riparian systems, 348
biological wastewater treatment systems,
166–180
basic concepts, 168f, 176
benefits, 179
biological processes, 177
483
continuous flow systems, 177, 177f
simulation models, 179
nitrogen removal, 178
nutrient removal capacity, 178
phosphorus removal, 179
background information, 166
biological processes
analytical techniques, 169
cellular organisms, 166, 167t
chemotrophs, 167t
kinetic mechanisms
enzymatic reactions, 173
flow models, 172, 173f
potential denitification activity (PDA)
measurements, 174, 175f
microbial carbon processes, 169, 169f
microbial communities, 168, 168f, 177
microbial phosphorus processes, 170, 170f
nitrogen transformation processes
anaerobic ammonium oxidation, 171f, 172
denitrification, 171, 171f, 175f
general discussion, 170
immobilization processes, 171, 171f
mineralization processes, 171, 171f
nitrification, 171, 171f
respiration, 169, 169f
water circulation, 172, 173f
constructed wetlands
classification, 175
free water surface flow (FSW), 175
subsurface water flow, 175,
176, 176f
future perspectives, 180
biomarkers
alpine ecosystems, 153
biomass-population density relationships, 137
body size relationships, 44–50
background information, 44
community level patterns
body size-abundance distributions, 45
body size-energy use distributions, 48
body-size ratios, 49
body size-species distributions, 48
global-scale size-abundance
distributions, 46
local-scale size-abundance distributions,
46–47, 47f
decoding mechanisms, 49
measuring complexities, 44
range-size patterns, 45
species-level patterns, 45
chlorophyll biomass, 452, 452f
estuarine ecosystems, 249
floodplain ecosystems, 258, 258f
quantity estimates
boreal forests, 182
fundamental ecosystem theory, 34, 36t
grasslands, 408
orientation theory, 124t
net primary production (NPP)
energy flow, 9t
secondary productivity, 17
savanna ecosystems, 398f, 403, 403t
standing crop biomass-production rate
relationship, 18, 18f, 19f
upwelling ecosystems, 452, 452f
biomimicry, 14
biosphere
indirect interactions, 85
elemental concentrations, 10t
fungi
exergy, 131, 131f, 135
self-organization, 105
Biosphere II project, Arizona (USA), 281, 292
bioturbation
salt marsh ecosystems, 388
savanna ecosystems, 395–396
484
Index
birch trees
boreal forest ecosystems, 181
temperate forest biomes, 425f
tundra ecosystems, 445
birdbaths, 428t
birds (Aves)
African floodplain ecosystems, 259–260
alpine forests, 157
Australian floodplain ecosystems, 261
boreal forest ecosystems, 182
breeding behaviors, 101
cave habitats, 191
competition studies, 82–83
coral reefs, 211f
desert environments
characteristics, 230
drought tolerance, 230
endangered species, 239
predation, 236
water uptake, 231
estuarine ecosystems, 249
exergy calculations, 134t
floodplain ecosystems, 276
forest plantations, 268
grasslands, 410
intertidal environments, 377
mangrove forests, 310
marshes, 277, 279f
Mediterranean ecosystems, 322, 322t, 326
pollen vectors, 24, 24f
postclosure landfills, 307
riparian systems, 348
saline/soda lakes, 381, 381f, 382, 384
salt marsh ecosystems, 386, 387
seabirds
upwelling ecosystems, 457
temperate forest biomes, 423
temporary water bodies, 433
tundra ecosystems, 446, 447f
upwelling ecosystems, 451, 455f, 457
urban environments, 463
waterfowl
aquatic wetlands, 275, 277
Asian floodplain ecosystems, 260
Australian floodplain ecosystems, 261
saline/soda lakes, 384
salt marsh ecosystems, 388
upwelling ecosystems, 457
Birge, Edward A, 7
Birmingham, England, 463
bison, 411, 411f, 429t
bivalves (Bivalvia)
estuarine ecosystems, 250, 251f
exergy calculations, 134t
Florida Everglades mesocosm project, 294f
stream ecosystems, 371, 372f
temporary water bodies, 433, 436–437
black band disease, 202–203
blackbody radiation, See solar radiation/solar
energy
Blackburnia spp., 193
black locust, 186, 424f
black mangrove, 291f, 293f
Black Sea ecosystem
body size-abundance distributions, 47f
intertidal area research, 378
black-tailed deer, 196
black tides, 453
blackwater systems, 255, 256f, 262
bladderworts, 41, 42f, 278
bleaching
coral reefs
anthropogenic impacts, 205, 210f
coral-algal mutualism, 201
significance, 209
Blenniidae, 377
Blepharismidae, 288t
blowouts, 242, 244
blueberries, 420–421
blue crabs, 280
bluefin killifish, 295f
blue-green algae, See cyanobacteria
bluestriped grunt, 207–208
Blutaparon portulacoides, 243f
Bodensee, See Lake Constance
body size relationships, 44–50
background information, 44
body size-abundance distributions
general discussion, 45
global-scale size-abundance
distributions, 46
local-scale size-abundance distributions,
46–47, 47f
nontaxonomically based size-abundance
distributions, 47
taxonomically based size-abundance
distributions, 45
body size-energy use distributions, 48
body-size ratios, 49
body size-species distributions, 48
competition, 49
decoding mechanisms, 49
measuring complexities, 44
metabolic rate (MR)
decoding mechanisms, 49
ecological efficiency, 48
range-size patterns, 45
species-level patterns, 45
bogs
characteristics, 175, 275, 333, 429t
continental bogs, 334, 335f
development factors, 332f
ecological network analysis (ENA), 71,
72f, 73t
largest wetlands, 281
nutrient cycling, 278
oceanic bogs, 334, 334f, 338
Russia, 71, 72f, 73t
water availability, 276–277
boids model, 103
Bolivia
alpine ecosystems, 150, 153f
savanna ecosystems, 396
Bolivinitidae, 288t
Boltzmann free energy constant, 129–130
Bombus hyperboreus, 448
Bombus polaris, 448
Bondonidae, 288t
bonitos, 19
boobies, 457
Boodieopsis pusilla, 293f
boojum trees, 229, 229f
Borassus spp., 259
boreal forests, 181–183
animals, 182
biodiversity, 182
biomass estimates, 182
bogs, 334
characteristics, 181
climate, 181
climate change effects, 182
ecosystem dynamics, 182
peatlands, 331
primary production, 234t
soils, 181
structural characteristics, 181
temperate forest biomes, 417–418
temporary water bodies, 428t, 429t,
432, 432f
wildfires, 182
See also taiga
Borneo, 441
Boston, Massachusetts (USA), 463
Bostrychia montagnei, 293f
botanic gardens, 183–189
ancient gardens, 183
biodiversity impact, 188
Canada, 189, 189f
colonial period, 186
educational programs, 189
France, 184t, 185, 185f
geographic locations, 184t
historical background, 185
medicinal gardens, 185
research approaches, 183
Singapore, 184t, 187, 187f
United Kingdom, 183, 184f
United States, 184t, 187, 187f
Botanic Gardens Conservation International,
183, 189
Bothriochloa spp., 396
Botswana, 223t, 395f
bottlenecks
ascendancy applications, 63–64
Bowdichia virgioides, 396
Brachiara spp., 396–397, 404
Brachionus dimidatus, 382
Brachionus plicatilis, 382
Brachionus spp.
saline/soda lakes, 382
brachiopods (Brachiopoda)
exergy calculations, 134t
Brachystegia spp., 396–397
Brahmaputra River
floodplain ecosystems, 260
branchiopods, 433
Branta canadensis, 278
Brassicaceae, 233f
Braunton Burrows, Great Britain, 244
Brazil
botanic gardens, 184t
dune ecosystems, 244
forest plantations, 265f
riparian systems, 347f
savanna ecosystems, 396
tropical ecosystems, 440
urban environments, 462
breadfruit, 186
Briareidae, 288t
brigalow, 396
brine shrimp, 383
bristlecone pines, 158
British Honduras, 313
brittlestars
intertidal environments, 376
bromeliads, 431, 431f, 442f
Bromus spp., 238
Bromus tectorum, 320t, 327
brown algae
intertidal environments, 375, 375f
brown bears, 182
brown lemmings, 447
brush rabbits, 196
bryophytes
alpine ecosystems, 153
bryozoans (Bryozoa)
intertidal environments, 376
buckbrush, 195
Buddleia davidii, 186
buffalo, 259–260, 261, 400, 411, 429t
buffalo wallows, 429t
Bufo americanus, 437
bugs
cave habitats, 193
exergy calculations, 134t
stream ecosystems, 373
bullfrogs, 279
Bullidae, 288t
bulltongue, 276f
bulrush
marshes, 275f, 276f
bumblebees, 28–29, 448
buried seeds, 278
Burkea africana, 396–397, 398
Index
Burma, 397
Buteo lagopus, 447
butterflies
exergy calculations, 134t
grasslands, 410
pollen vectors, 24, 24f
postclosure landfills, 307
temperate forest biomes, 423
butterfly bush, 186
butterworts, 278
buttonwood, 285f
buttressing processes, 258
buttress-root slits, 429t
Byrsonima spp., 396
C
C3 pathway
desert environments, 230
exergy, 137
grasses, 406
salt marsh ecosystems, 391
savanna ecosystems, 397, 404
C4 pathway
desert environments, 230
exergy, 137
grasses, 406
salt marsh ecosystems, 391
savanna ecosystems, 394, 395f, 397
temporary water bodies, 435
Caconemobius spp., 193
Cactaceae, 229, 230
cactus
alpine ecosystems, 153
desert environments, 229, 230, 238
caddisflies
salinity tolerances, 435f
stream ecosystems
general discussion, 372–373
predator-prey dynamics, 354, 354t
Cakile maritima, 243
Calamagrostis epigejos, 246
Calanus spp., 453–454
calcium (Ca)
carbonates (CaCO3)
bicarbonate
eastern African soda lakes, 382
ionic composition, 381
Mono Lake (USA), 383
stream ecosystems, 369
coral reef microcosm, 287f
saline/soda lakes
eastern African soda lakes, 382
ionic composition, 381
Mono Lake (USA), 383
elemental concentrations, 10t
saline/soda lakes, 381
stream ecosystems, 369
temperate forest biomes, 424
Calcutta, India, 462
California Current, 450, 450f, 458
California Floristic Province, 320t, 322
California gull, 383
California lilac, 195, 320t
Californian palm, 233
California (USA)
botanic gardens, 184t
fish/fisheries, 458f, 459
grasslands, 412
intertidal environments, 376f
invasive species, 412
Mediterranean ecosystems
biodiversity patterns, 322, 322t
biogeographical evolution, 323
characteristics, 320t
climate, 319
climatic stress, 327
convergence studies, 325
Pleistocene glaciation, 324
soils, 321
vernal pool ecosystems, 429t, 432
wildfires, 328
river system dynamics, 357f
saline/soda lakes, 383
California vernal pools, 429t
Calliergon spp., 335
Callithamnion spp., 293f
Callitriche spp., 327
Calluna vulgaris, 334
Calvin-Benson cycle, 137
Calystegia soldanella, 243f
Cambodia, 397
camels, 230, 231, 238, 239, 411
Camelus bactrianus, 240
Cameroon, 183, 184t, 186, 440
camouflage, 236
Campylium spp., 335
Canada
alpine forests, 157, 159t, 160f
boreal forests, 181
botanic gardens, 184t, 189, 189f
forest plantations, 267
peatlands, 276–277, 331, 331f
tundra ecosystems
bumblebees, 448
musk oxen, 447
vegetation, 445, 445f, 446f
Canada geese, 278, 280
Canary Current, 450, 450f, 457
Canavalia rosea, 242
cane toads
floodplain ecosystems, 261
Canis latrans, 199
Canis lupus, 182
Canis lupus arctos, 446–447
canoes, 428t
CANON removal process, 179
Cape Cod, Massachusetts (USA), 431f, 438t
Cape cormorant, 457
Cape Floristic Region, South Africa,
320t, 322, 462f
Cape gannet, 457
capercaillie, 182
Cape Town, South Africa, 462f
Capitellidae, 288t
Capparis spp., 193
Carangidae, 204
carbon (C)
carbonates (CaCO3)
bicarbonate
eastern African soda lakes, 382
ionic composition, 381
Mono Lake (USA), 383
stream ecosystems, 369
coral reef microcosm, 287f
saline/soda lakes
eastern African soda lakes, 382
ionic composition, 381
Mono Lake (USA), 383
alpine ecosystems, 153
carbon sinks
boreal forests, 182
peatlands, 330, 332f, 338
global carbon cycle
boreal forests, 182
savanna ecosystems, 394
mangrove forests, 311, 312f
riparian systems, 349
salt marsh ecosystems, 389
savanna ecosystems, 403, 403t
temperate forest biomes, 425
total organic carbon (TOC), 349
carbon dioxide (CO2)
alpine ecosystems, 155
alpine forests, 159
eco-exergy losses, 136
mangrove forests, 312f
485
microbial processes, 169, 169f
regime shifts, 19
stream ecosystems, 369
temperate forest biomes, 425
carbonic acid (H2CO3), 369–370
elemental concentrations, 10t
microbial processes, 169, 169f
peatlands, 330, 331, 332f, 338
stable isotope analysis, 20
carbon cycle, 140
Carcinus maenus, 391–392
Carditidae, 288t
Carex curvula, 154f
Carex lyngbei, 386t
Carex nigra, 244
Carex spp., 333, 334
Caribbean
coral reefs, 210
floodplain ecosystems, 257t
mangrove forests, 313
tropical ecosystems, 440
carnivores (Carnivora)
desert environments, 231, 236
intertidal environments, 377
plants, 41, 42f
polar ecosystems, 340–341
stream ecosystems, 353–354
trophic interactions, 17
carnivorous plants
autocatalysis, 41, 42f
wetland ecosystems
marshes, 278
peatlands, 330
riparian systems, 347
wet meadows, 275
Carolina bays, 429t, 432
Carpinus orientalis, 321t
Carya spp.
temperate forest biomes, 421
United States, 262
Caryophylliidae, 288t
Cascade Mountains, Oregon (USA), 157, 357f
Caspian Sea, 381
Castanea dentata, 420
Castor spp., 348
Casuarina equisetifolia, 306
Casuarina spp., 326
Cataglyphis bicolor, 230
catastrophic events
mangrove forests, 314
caterpillars
grasslands, 410
savanna ecosystems, 400
catotelm, 332, 333, 337
cattails
constructed wetlands, 176
marshes, 275, 278
cattle
grasslands, 411
grazing impact, 147
savanna ecosystems, 400
Caucasus Mountains, 159t, 160f
Caulerpaceae, 288t
Caulerpa fastigiata, 293f
Caulerpa verticillata, 293f
Cavaticovelia aaa, 193
caves, 190–194
cave-like habitats, 192, 192f
characteristics, 190
deep cave communities, 191
environmental adaptations, 192
guano communities, 191
habitats
aquatic environments, 190
food resources, 191
terrestrial environments, 190, 191f
Hawaii
food webs, 193
486
Index
caves (continued )
nonindigenous species, 193
succession studies, 193
research environment, 194
Ceanothus spp., 195, 320t
Cedar Bog Lake, Wisconsin (USA), 7
Cedrus spp., 159t, 320t, 321t, 321t
biological/life system models, 102
emergence/emergent properties, 94
Game of Life (Conway)
basic concepts, 102
self-organization, 102
Celtis spp., 262
centipedes
cave habitats, 191, 193
landfills, 305
Central America
floodplain ecosystems, 257t
intertidal area research, 378
tropical ecosystems, 440
Centrarchidae, 373
centripetal actions, 42, 42f, 43f
Ceramiaceae, 288t
Ceratonia spp., 321t, 328
Ceratophyllaceae, 371
Ceratophyllum demersum, 176
cereals
See also grasses
Cerianthidae, 288t
Cerithium lutosum, 294f
cerrado landscape, 396
cetaceans
upwelling ecosystems, 457
Chad, 223t, 384
Chaetodontidae, 288t
Chaetomorpha gracilis, 293f
Chaetomorpha minima, 293f
Chaetonotidae, 288t
Chaetopteridae, 288t
Chaetospiridae, 288t
Chamaecrista chamaecristoides, 243f, 243–244
Chamaecyparis spp., 159t
Chamaedaphne calyculata, 334
Chamidae, 288t
chamise, 195, 320t
chamomile, 185
Champiaceae, 288t
channelization
floodplain ecosystems, 259
stream ecosystems
desert environments, 221
ecological applications, 34
self-organization, 101
chaparral cherry, 195
chaparral ecosystems, 195–200
allelopathy, 197
bare zone, 197f
biogeographical evolution, 324
characteristics, 195, 196f, 326
community succession, 196
desert stream ecosystems, 218t
fire-dependent regeneration, 195–196
invasive species, 200
management strategies, 200
regional variations, 199
seed dispersal, 199
seed germination, 199
vegetation communities, 195
wildfires
characteristics, 197
dominance-diversity curves, 198f
management strategies, 200
postfire recovery rates, 197, 198f
regional variations, 199
seed germination, 199
wind effects, 197, 199
chaparral holly, 195
chaparral honeysuckle, 196
Characidae, 373
Charales, 371
cheatgrass
Mediterranean ecosystems, 320t, 327
chemical contamination, 392
chemical oxygen demand (COD)
activated sludge systems, 178
chemical warfare, See allelopathy
chemoautotrophs, 453
chemotrophs, 167t
Chenopodiaceae, 387
Chenopodium spp., 387
chenopods, 229
Chesapeake Bay (USA)
ascendancy applications, 63–64
estuarine ecosystems
complex systems, 248
riverine estuaries, 251
lagoon ecosystems, 296–297
riverine estuaries, 248
chickadees, 423
Chihuahuan Desert, 223t, 227
Chile
desert environments, 223t
fish/fisheries, 21, 459
forest plantations, 265f
intertidal environments, 379f
Mediterranean ecosystems
biodiversity patterns, 322, 322t
biogeographical evolution, 323
characteristics, 320t
climate, 319
climatic stress, 327
convergence studies, 325
Pleistocene glaciation, 324
soils, 321
wildfires, 328
Chilean pelican, 457
China, 140
botanic gardens, 183, 184t
desert environments, 223t
floodplain ecosystems, 260–261
forest plantations, 265, 265f
temperate forest biomes, 425–426
chipmunks, 182, 423
chironomids/Chironomidae
saline/soda lakes, 382
stream ecosystems, 372f, 373
Chirotidae, 288t
chitons
intertidal environments, 376
Chlamydotis spp., 240
chlorine (Cl)
elemental concentrations, 10t
saline/soda lakes, 381
stream ecosystems, 369
Chlorophycota, 288t
chlorophyll
algae, 370
deep chlorophyll maximum
(DCM), 452–453
high-nutrient low-chlorophyll (HNLC)
ecosystems, 454
primary production, 451
upwelling ecosystems, 452, 452f
Chlorophyta/Chlorophyceae
desert stream ecosystems, 216
Florida Everglades mesocosm project, 293f
intertidal environments, 375
river system dynamics, 359f
saline/soda lakes, 384
salt marsh ecosystems, 387, 389
stream ecosystems, 370
temporary water bodies, 433
Choanozoa, 288t
Chondrosiidae, 288t
Chordata
exergy calculations, 134t
intertidal environments, 376
Chorizanthe rigida, 233
chotts, 225
chromium (Cr)
eco-exergy losses, 136t
Chromophycota, 288t
Chroococcaceae, 288t
Chrysanthemum alpinum, 154f
Chrysobalanus icaco, 244
Chrysophyta
stream ecosystems, 370
Chrysopogon bladhii, 396
Chthamalus anisopoma, 376
Chthamalus stellatus, 378–379
chubs, 203–204
Cibicidiidae, 288t
Cichlidae, 373
Cidaroidae, 288t
ciliates
saline/soda lakes, 384
upwelling ecosystems, 453
Cinetochilidae, 288t
Cirratulidae, 288t
Cirriformia filigera, 294f
Cistus spp., 320t, 325–326
citric acid cycle, 169, 169f
Cixiidae, 192
Cladina spp., 334, 334f
cladocera
See also Daphnia spp.
Cladophora, 219f
Cladophoraceae, 288t
Cladophora crispata, 293f
Cladophora repens, 293f
Cladophora spp., 371
clams
intertidal environments, 376
invasive species, 391
lagoon ecosystems, 299–300
stream ecosystems, 371
temporary water bodies, 436
Clark’s nutcracker, 158
Clathrinidae, 288t
clay
desert soils, 226
cleaner fishes, 206
agroecosystems, 146
alpine forests, 159, 165
boreal forests, 181
alpine ecosystems, 155
boreal forests, 182
estuarine ecosystems, 252
fish/fisheries, 458, 458f
indirect interactions, 87
polar ecosystems, 341
salt marsh ecosystems, 390
savanna ecosystems, 403
tundra ecosystems, 443, 449
upwelling ecosystems, 458, 458f
forest ecosystems, 30
glaciation
desert stream ecosystems, 218t
Last Glacial Maximum (LGM), 324
Pleistocene, 321, 324
grasslands, 407
Mediterranean ecosystems
characteristics, 319, 320t
rainfall, 321t
stressors, 327
temperature, 321t
alpine forests, 162
shelterbelts/windbreaks, 469
Milankovitch climate oscillations, 324
peatlands, 331, 332f
riparian systems, 344, 349
savanna ecosystems, 398, 398f, 401t
temperate forest biomes, 418, 419t
tropical ecosystems
Index
precipitation, 439–440, 441
rainforests, 441
temperature, 439–440, 441
See also global warming
climate, 140, 141t
climate change, 140, 141t
climbing plants, 442
Clionidae, 288t
closed landfills, 304
closed systems
emergence/emergent properties, 95t
clouds
hydrologic cycle, 365f
Cnidaria
exergy calculations, 134t
intertidal environments, 376
coarse particulate organic matter (CPOM), 349,
352f, 354f, 356, 358f, 368
coastal dune ecosystems, 241–246
abiotic factors
gradient structure, 242
habitats, 244, 245f
nutrient resources, 242
salinity, 243
sand movement, 242, 245f
temperature variations, 244
water sources, 243
dune profile characteristics, 245f
formation processes, 241
general discussion, 241
photographic views, 17f
vegetation communities
characteristics, 241
competition, 245
diseases, 246
general discussion, 244
pioneer zone, 245
predators, 246
sand movement, 242, 243f
succession and facilitation, 245
symbiotic relationships, 246
coastal zones
beaches, 241–246
abiotic factors
gradient structure, 242
habitats, 244, 245f
nutrient resources, 242
salinity, 243
sand movement, 242, 245f
temperature variations, 244
water sources, 243
formation processes, 241
general discussion, 241
vegetation communities
competition, 245
diseases, 246
general discussion, 244
predators, 246
succession and facilitation, 245
symbiotic relationships, 246
desert environments, 222
lagoon ecosystems, 296–303
catchment loads, 297, 297f, 300f
characteristics, 296
land-use change, 297, 297f
modeling studies
empirical models, 298
multifractal biomass/species
distributions, 301
nonequilibrium dynamics, 300
NPZD (nitrogen-phytoplanktonzooplankton-detritus) model, 299
physical processes, 297
simulation models, 299
Coccoloba uvifera, 246
Cocconeis spp., 370f
Cocculus orbiculatus, 193
cockroaches
cave habitats, 193
coconut palms, 246
coconuts, 428t
Codakia orbiculata, 294f
Codiaceae, 288t
Codosigidae, 288t
Coelenterata, 294f
body-size ratios, 49
indirect interactions, 88
orientation theory, 122, 124t
savanna ecosystems, 401t
coffee
bee pollination, 29
botanic gardens, 186
coffeeberry, 195
Colchicaceae, 323
Coleodiea, 134t
Coleoptera
cave habitats, 193
desert stream ecosystems, 217
stream ecosystems, 372f, 373
Colepidae, 288t
collared lemmings, 447
Collembola
indirect interactions, 85
stream ecosystems, 373
Collins, Terry, 14
Colochaetaceae, 288t
Colombia
rainforests, 440f, 442f, 443f
savanna ecosystems, 396
desert stream ecosystems, 218, 218t,
219f
invasive species, 391
self-organization, 104
salt marsh ecosystems, 391
savanna ecosystems, 401t
urban environments, 464
Colorado (USA), 406–407
Columbiformes
desert environments, 231
Combretum spp., 396–397
indirect interactions, 81,
83, 83f
Commiphora spp., 396–397
common pine sawfly, 182
common reed, See Phragmites australis
community assembly, 77
community ecology
communities versus assemblages, 104
dimensional units, 7t
emergence/emergent properties, 95t
hysteresis
community assembly, 77
salt marsh ecosystems, 392
self-organization, 104
temporary water bodies
general discussion, 437
interspecific interactions, 437
richness theories, 437, 438t
comparative theoretical ecosystem analysis
(CTEA), 90
apparent competition, 82, 83f
autocatalysis, 42–43, 43f
body-size factors, 49
coexistence models, 49
coral reefs, 202, 202f
dune ecosystems, 245
exploitative competition
network environ analysis, 80t
research areas, 83, 83f
grazing populations, 83
indirect interactions, 81
interference competition
research areas, 83, 83f
interspecific competition
indirect interactions, 82, 83f
487
intertidal environments, 378
mangrove forests, 311
intraspecific competition
intertidal environments, 378
network environ analysis, 79, 80t
hierarchy theory, 114
resource competition
coral reefs, 202, 202f
savanna ecosystems, 401t
self-organization, 98
complementarity theory, 34
complex peatlands, 332–333, 333f
complex systems, 106–114
ascendancy, 57
basic concepts
decision-making processes, 108
feedback loops, 107–108
general discussion, 106
predictable behaviors, 107
simple systems, 107
weak interactions, 108
characteristics, 111t
complex adaptive systems, 102
components
intelligence, 113
local information, 113
number of agents, 113
ecological complexity, 110
ecosystems, 110
emergence, 109, 110t
estuarine ecosystems, 248
hierarchy theory, 118
indirect interactions, 89
modeling studies
Sugarscape, 112, 112f, 113f
Tierra, 102, 111
TRANSIMS, 112
self-organization, 99, 102, 109, 110, 110t, 111t
surprise-generating mechanisms
connectivity, 109, 110t
emergence, 109, 110f, 110t
incompatible behaviors, 108–109, 110t
paradoxes, 108, 108f, 110t
unstable systems, 108, 110t
compos, 407
carbon:nitrogen (C:N) ratio, 462–463
computational emergence, 92, 96f
CompuTerrarium (computer simulation),
112, 113f
Condylostomatidae, 288t
confocal laser scanning microscopy (CLSM), 169
Congo River
tropical ecosystems, 440
wetland ecosystems, 281
conifers/coniferous forests
dune ecosystems, 246
mixed coniferous-deciduous forests,
417–418, 418f
physiographic regions, 419t
complex systems, 109, 110t
coral reefs, 208f
corridors, 349, 463
riparian systems, 345
self-organization, 100
Connell, JH, 378
conservation strategies
ecosystem-based management, 26
ecosystem services, 24, 24f
forest plantations, 267
grasslands, 412
Millennium Ecosystem Assessment (MA), 25
riparian systems, 350
river systems, 362
social-ecological systems, 25
temporary-water populations, 438
tropical rainforests, 442, 443f
constructed wetlands
classification, 175
488
Index
constructed wetlands (continued )
functionality, 175
subsurface water flow, 175, 176, 176f
wastewater treatment, 175
contamination studies
chemical contamination, 392
salt marsh ecosystems, 392
continental bogs, 334, 334f, 334f, 335f
continental interior deserts, 222
Convention on Biological Diversity (CBD), 11,
11t, 466
convergence studies, 230
Conway’s Game of Life
basic concepts, 102
coontail, 176
self-organization, 103
coral reefs, 206
self-organization, 103
copepods (Copepoda)
saline/soda lakes, 382
stream ecosystems, 371
temporary water bodies, 433
upwelling ecosystems, 452, 453–454, 455f
eco-exergy losses, 136t
stream ecosystems, 369
Corallimorphidae, 288t
Corallinaceae, 288t
coral reefs, 201–213
allelopathy, 202
Caribbean, 210
coral-algal mutualism, 201, 205
ecological interactions
competition, 202, 202f
disturbances, 205, 205f, 210f
herbivores, 203, 203f
mutualism, 201, 205
positive interactions, 205
predators, 204, 204f, 212f, 213
exergy calculations, 134t
geographic distribution, 208, 209f
greenhouse ecosystems
carbonate cycle, 287f
general discussion, 286
gross primary production (GPP), 287f
organism families, 288t
oxygen (O) concentrations, 286f
physico-chemical parameters, 287t
schematic diagram, 286f
importance, 201, 202f
landscape ecology, 207, 208f
recruitment effects, 206, 207f
reproduction, 206
stressors
coral bleaching
anthropogenic impacts, 205, 210f
coral-algal mutualism, 201
significance, 209
diseases, 202–203, 205, 205f, 210, 210f
general discussion, 209
overfishing, 205, 205f, 210f,
211, 211f
protective marine reserves, 212, 212f
shifting baseline syndrome, 211
cordgrass, 249, 249t, 386
See also Spartina alterniflora
Coriolis effect, 450
corixids/Corixidae, 372f, 382
corixos, 429t
cormorants
upwelling ecosystems, 457
water level changes, 277
corn
desert environments, 238
See also maize
Corophiidae, 288t
Corophium spp., 294f
corridors
habitat connectivity, 463
riparian systems, 349
Cortez, Hernando, 183
Corvus corax, 446–447
Costa Rica, 29, 159t, 160f, 442f
cottonwood trees, 268, 421
counterintuitive behaviors, 108
coyotes
desert environments, 235
seed dispersal, 199
crabs
estuarine ecosystems, 249, 252
intertidal environments, 376, 377
invasive species, 391–392
lagoon ecosystems, 299–300
mangrove forests, 310, 311
marshes, 280
riparian systems, 348
salt marsh ecosystems, 386, 387, 391–392
craneflies, 193
Crassostrea virginica, 250
Crassulaceae, 153, 230
Crassulacean acid metabolism
(CAM) pathway
desert environments, 230
exergy, 137
temporary water bodies, 435
Crater Lake, Oregon (USA), 271f
crayfish
stream ecosystems, 371
temporary water bodies, 435
upwelling ecosystems, 453
creaon, 118–119
creosote bush, 233–234
crescent-shaped dunes, 224–225
Cressa cretica, 386t
Cretaceous
desert flora, 228
grass phytoliths, 405–406
Mediterranean ecosystems, 323
crickets
cave habitats, 193
riparian systems, 348
Cricotopus spp., 371
critical phenomenon, 100
crocodiles
Australian floodplain ecosystems, 261
coral reefs, 211, 211f
desert environments, 230
marshes, 278
overfishing, 211, 211f
riparian systems, 348
Cronartium spp., 269
crop production
bee pollination, 28
biodiversity, 148
crop characteristics, 146
pest control methods
characteristics, 147
shelterbelts/windbreaks, 469
temperate forest biomes, 425
See also agroecosystems
crossbills, 182
crown-of-thorns starfish, 204
crows, 182
Cruciferae, 371
crustaceans (Crustacea)
cave habitats, 191
desert stream ecosystems, 216
estuarine ecosystems, 250
exergy calculations, 134t
intertidal environments, 376, 378
mangrove forests, 310
stream ecosystems, 371, 372f
temporary water bodies, 432, 433,
436, 437
crypsis, 236
Cryptocarya alba, 320t
Cryptomonadaceae, 288t
Cryptosporidium spp., 274
Ctemodrillidae, 288t
cucumbers, 28
Cupressus macrocarpa, 320t, 325
Cupressus sempervirens, 328
Curatella americana, 396
curlews, 387
Cuscuta spp., 236f
cushion plants, 153, 154f
cutthroat trout, 384
cyanobacteria
agroecosystems, 148
coral reef microcosm, 288t
desert environments, 228
desert stream ecosystems, 215, 216,
219, 219f
lake ecosystems
indirect interactions, 86
river system dynamics, 359f
saline/soda lakes, 384, 384
salt marsh ecosystems, 387
stream ecosystems, 370
upwelling ecosystems, 454
Cyanophota, 288t
Cyathura polita, 294f
cycle analysis, 50–57
adjacency matrices, 50, 52, 53f
definition, 50
ecological applications, 56
ecological networks
acyclic networks, 55
cycle removal, 55
cycle searches, 55
matrix limitations, 54
matrix models, 52, 53f
strongly connected components, 51, 52f
Finn’s cycling index, 52
food webs
direct interactions, 50, 51f
feeding cycles/nonfeeding cycles, 51
indirect interactions, 51, 51f
number of simple cycles, 54, 55t
general discussion, 50
cycles/cycling indices, 4
Cyclidiidae, 288t
Cycling, 4
Cyclocephala hardyi, 188
cyclones, 313, 441
Cymbaloporidae, 288t
Cymbellaceae, 288t
Cymbella spp., 370f
Cynipidae, 196
Cyperus articulatus, 244
Cyprididae, 288t
Cypridina spp., 294f
Cypridopsis vidua, 294f
Cyprinidae, 373
Cyprinodon spp., 218, 229
Cyprinodontiformes, 433
Cyprinodon variegatus, 295f
D
Daisyworld model, 103, 105
dambos, 429t
coral reefs
importance, 203–204
positive interactions, 206
damselflies, 372–373
Danube River, 261
Daphnia pulex, 437
Daphnia spp.
temporary water bodies, 433, 436–437
darkling beetles, 235
Darwin, Charles
coral reefs, 201
date palms, 233
Dead Sea, 235f, 381, 384
See also mortality studies
Index
Death Valley, California (USA), 226, 435f
decapods (Decapoda)
stream ecosystems, 371
Dechloromonas spp., 170
deciduous forests
alpine ecosystems, 159t
mixed-deciduous temperate forest biomes, 417,
418f, 419t
primary production, 234t
savanna ecosystems, 397
decision-making processes
agroecosystems, 148, 149f
simple systems, 108
decision-tree models
boreal forests, 181
carbon:nitrogen (C:N) ratio, 462–463
desert environments, 235
ecological network analysis (ENA), 73t
estuarine ecosystems, 249
hierarchy theory, 118–119
mangrove forests, 311
peatlands, 331
urban environments, 462–463, 463t
deep cave communities, 191
deep chlorophyll maximum (DCM), 452–453
deer
chaparral ecosystems, 196
temperate forest biomes, 423
deer mice, 196
defense strategies
desert environments, 235
savanna ecosystems, 397
mangrove forests, 313–314
temperate forest biomes, 426
Delesseriaceae, 288t
Deltracia bullaoides, 294f
Democratic Republic of the Congo, 440
demographic analysis
community assembly, 77
savanna ecosystems, 401t, 402
See also population dynamics
denaturing gradient gel electrophoresis
(DGGE), 169
activated sludge systems, 178
desert stream ecosystems, 220, 220f
lagoon ecosystems, 298
mangrove forests, 311, 312f
microbial processes
basic concepts, 171f
nitrogen transformation processes, 171
potential denitification activity (PDA)
measurements, 175f
salt marsh ecosystems, 389
sewage treatment systems, 178
density-dependent effects
indirect interactions, 81
deoxyribonucleic acid (DNA), See DNA
(deoxyribonucleic acid)
Deppea splendens, 188
Derbesiaceae, 288t
Derbesia spp., 293f
desert crusts, 228
desert environments, 222–240
anthropogenic impacts, 238
arctic environments, 222
biodiversity
animals, 229
convergent evolution, 230
general discussion, 227
microorganisms, 227
plants, 228, 229f
stream ecosystems, 218t
coastal zones, 222
continental interior deserts, 222
definition, 222
ecophysiological strategies
drought, 230
unpredictable water resources, 233, 233f
water uptake
animals, 231
microorganisms, 232
plants, 232
flash floods, 225, 226f
geographic distribution, 222, 223f
human ecology
anthropogenic impacts, 238
endangered species, 239, 240f
historical background, 237, 237f
hunter-gatherer societies, 238
nomadic cultures, 238f
research areas, 240
invasive species, 238
landforms
badlands, 225
desert flats/basins, 224, 224f
desert mountains, 223, 224f
ephemeral streams, 225, 225f,
428t, 435f
general discussion, 222
piedmont bajada formations, 223, 224f
playas, 225, 225f, 428t, 433f
sand dunes, 224f, 224–225
limiting factors, 225
major deserts, 223t
physical characteristics, 214
polar deserts, 222
precipitation, 225, 226f
rain shadow deserts, 222
research areas, 240
semidesert environments, 153, 153f
soils, 226
stream ecosystems, 214–221
anthropogenic impacts, 221
biota, 216, 218t, 229
boundary zones, 214–215
energetics, 218
gaining/losing reaches, 215
nutrient dynamics, 219, 220f
physical characteristics, 214
riparian zones, 214, 214f
temporal dynamics
disturbances, 215
drying disturbances, 215–216, 217f
flash floods, 215–216, 217f, 218t, 219f,
225, 226f
flow seasonality, 216
general discussion, 215
interannual/decadal variability,
216, 217f
spatial scales, 215f
successional patterns, 215
temperature variations, 216
subtropical deserts, 222
systems ecology
general discussion, 234
nontrophic interactions, 236, 237f
primary production, 234, 234t, 235f
trophic interactions
decomposition processes, 235
general discussion, 235
herbivory, 235
parasites, 236, 236f
predation, 236
temporary water bodies, 428t, 429t,
432, 433f
tundra ecosystems, 446, 446f
vegetation, 226, 227f
See also dune ecosystems
desert hedgehogs, 236
desertification
anthropogenic impacts, 238
desert pupfish, 218, 229
desert sand rats, 232f
desert shaggy mane, 227–228
desert tortoises, 230, 236, 240
desert truffles, 227–228
489
desert varnish, 228
detritivores
desert environments, 229
stream ecosystems, 353–354, 371
trophic interactions, 17
composition
carbon:nitrogen (C:N) ratio,
462–463
ecological network analysis (ENA), 73t
ecological significance
vegetal detritus, 255–256
exergy calculations, 132, 134t
floodplain ecosystems, 260
indirect interactions, 86
mangrove forests, 312
NPZD (nitrogen-phytoplankton-zooplanktondetritus) model, 299
postclosure landfills, 305
river system dynamics, 351
temperate forest biomes, 421, 421f
Devonian
desert flora, 228
Dexmoxyidae, 288t
Diablo winds, 197, 199
Diadema antillarum, 203, 205, 210
diadromous organisms, 373
Diamon, Jared, 13
diapause
temporary water bodies, 434, 436,
436t
Diatomaceae, 288t
Diatoma spp., 370f
diatoms
desert stream ecosystems, 216, 219f
estuarine ecosystems, 248
indirect interactions, 86
intertidal environments, 375
lake ecosystems, 86
nutrient resources, 454
river system dynamics, 359f
saline/soda lakes, 384
salt marsh ecosystems, 387
stream ecosystems, 370, 370f
temporary water bodies, 433
upwelling ecosystems, 452, 453, 455f
dicotyledons
Mediterranean ecosystems, 327
Dicranomyia spp., 193
Dicrostonyx torquatus, 447
diffusion, See molecular diffusion
dimethyl sulfide ((CH3)2S2)
upwelling ecosystems, 454
dinoflagellates
coral-algal mutualism, 201
estuarine ecosystems, 248
salt marsh ecosystems, 387
upwelling ecosystems, 451,
453, 455f
Dinophilidae, 288t
Diogenidae, 288t
Diosaccidae, 288t
Dipodidae, 230
Dipodomys spp., 232f
dippers, 348
Diprion pini, 182
Diptera
exergy calculations, 134t
stream ecosystems, 372f, 373
temporary water bodies, 433
Dipterocarpus spp., 397
direct interactions, See indirect interactions
Discocephalidae, 288t
Discorbidae, 288t
Discovery Bay, Jamaica, 205f
diseases
agroecosystems, 146f
coral reefs, 202–203, 205, 205f, 210f
dune ecosystems, 246
490
Index
diseases (continued )
forest plantations, 267
vector control, 438
dismals, 429t
modeling studies
chaparral ecosystems, 199
mangrove forests, 311
postclosure landfills, 307
spatial pattern formation, 104, 104f
temporary-water populations, 437
dissipative structure
ecological systems, 34
exergy, 132
thermodynamic systems, 34, 99
dissolved organic matter (DOM)
estuarine ecosystems, 249, 249f
stream ecosystems, 356
Distichlis spicata, 386t
disturbances
coral reefs, 205, 205f, 210f
desert environments, 238
desert stream ecosystems, 215
estuarine ecosystems, 252
grasslands, 408, 409f
mangrove forests, 313
marshes, 278
Mediterranean ecosystems
general discussion, 328
herbivores, 329
wildfires, 320t, 328
savanna ecosystems, 398, 398f, 401t
temperate forest biomes
characteristics, 420
detritus, 421, 421f
soils, 421
structural layers, 420
tropical ecosystems, 441
urban environments, 463t
diversity indices
Shannon index, 60
diving beetles, 437
DNA (deoxyribonucleic acid)
microbial ecology, 169
dobsonflies, 372–373
Dodonaea viscosa, 193
doline, 429t
domesticated animals, 238
domesticated plants, 238
dominance-diversity curves
plants
chaparral ecosystems198f
self-organization, 103
urban environments, 463t
Don Juan Pond (Antarctica), 381
donkeys, 230, 238
chaparral ecosystems, 197
drying adaptations, 435
Mediterranean ecosystems, 327
tundra ecosystems, 446
Dorvilleidae, 288t
Douglas fir trees, 267, 417
Douz, Tunisia, 238f
doves, 231
Draconematidae, 288t
dragonflies
stream ecosystems, 372–373
Drepanocladus spp., 335
Droseraceae, 347
Drosera spp., 330
Drosophila melanogaster
urban environments, 462
drought
desert environments
ecophysiological strategies, 230
stream ecosystems, 215–216, 217f
El Niño Southern Oscillation (ENSO), 459f
grasslands, 407
marshes, 278
Mediterranean ecosystems, 327
temperate forest biomes, 423
Dryas integrifolia, 446f
dry lakes, 428t, 433f
ducks
water level changes, 277
duckweed
marshes, 277–278
Dunaliella viridis, 384
Dune du Pyla, France, 241
dune ecosystems, 241–246
abiotic factors
gradient structure, 242
habitats, 244, 245f
nutrient resources, 242
salinity, 243
sand movement, 242, 245f
temperature variations, 244
water sources, 243
dune profile characteristics, 245f
formation processes, 241
general discussion, 241
photographic views, 17f
vegetation communities
characteristics, 241
competition, 245
diseases, 246
general discussion, 244
pioneer zone, 245
predators, 246
sand movement, 242, 243f
succession and facilitation, 245
symbiotic relationships, 246
dust
aeolian dust, 454
Dysdera spp., 193
E
eared grebes, 383
earthworms
agroecosystems, 148
landfills, 305
savanna ecosystems, 395–396
urban environments, 462
earwigs, 192, 192f
East Africa Highlands, 159t, 160f
eastern boundary current (EBC), 450, 451
eastern gem clam, 391–392
Eastern Scheldt, 248
Echeveria spp., 154
Echinodermata
Florida Everglades mesocosm project, 294f
intertidal environments, 376
Echinometra mathaei, 204
Ecklonia radiata, 21, 22f
ecoclines, 247
eco-exergy
basic concepts, 130, 130f
ecological systems, 34
exergy calculations, 132, 134t
fundamental ecosystem theory, 34, 36f
living organisms, 132, 134t
ecohydrology
mangrove forests, 314
body size-abundance distributions, 48
classification, 175
emergent macrophytes
functionality, 175
biochemical oxygen demand
(BOD5), 176
subsurface water flow, 175, 176, 176f
temporary water bodies, 438
Ecological Footprint analysis
industrial ecology, 87–88
ecological indicators
average mutual information (AMI)
uncertainty analysis5959f
biomass inclusive ascendancy, 62
calculation techniques, 60
calculation techniques, 60
ecosystem applications, 63
limiting factors, 62
overhead calculations
calculation techniques, 61
exports, 61
imports, 61
redundancy analysis, 62
respiration, 61
overhead calculations
calculation techniques, 61
exports, 61
imports, 61
redundancy analysis, 62
respiration, 61
total system throughput (TST), 58
diversity indices
Shannon index, 60
marine ecosystems
biological/life system models, 102
emergence/emergent properties, 94
Game of Life (Conway)
basic concepts, 102
self-organization, 102
ecological network analysis (ENA), See environ
analysis
ecological networks, 4, 141t
cycle analysis
acyclic networks, 55
cycle removal, 55
cycle searches, 55
matrix limitations, 54
matrix models, 52, 53f
strongly connected components, 51, 52f
ecological scaling
body size-abundance distributions, 48
lagoon ecosystems, 298
Ecosystem Approach Principles, 11t
ecosystems, 16–26
basic concepts, 16, 17f
indirect interactions, 85
complex systems, 110
components, 17
conservation strategies
anthropogenic impacts, 25
ecosystem-based management, 26
ecosystem services, 24, 24f
social-ecological systems, 25
eco-exergy losses, 136
ecosystem dynamics
boreal forests, 182
energy and material flow, 18, 18f
large-scale whole ecosystem experiments, 21
management policies as experiments, 21,
22f, 23f
stable isotope analysis, 20
standing crop biomass-production rate
relationship, 18, 18f, 19f
ecosystem ecology, 6–12
anthropogenic impacts, 10, 11t
basic concepts, 6
biogeochemical cycles, 9, 10t
definition, 7
dimensional units, 7t
Ecosystem Approach Principles, 11t
energy flow, 8, 9t
general discussion, 11
historical background, 6, 7t
international agreements and conventions,
10, 11t
research approaches, 10
schematic diagram, 8f
Index
ecosystem function-biodiversity relationship,
22, 23f
emergence/emergent properties, 4, 94, 95t
energy transfer efficiency, 19
exergy, 130, 130f, 132, 134t
fundamental laws, 33–38
basic properties, 37
ecosystem theory, 34, 36f, 36t
general discussion, 37
general scientific theories, 34
solar radiation, 36f
theory development, 33
Gaia hypothesis, 105
hierarchy theory, 114–120
community conception model, 115, 115f
connection strengths, 117, 117f
historical research, 117
holons, 117, 117f
hypothesis testing, 115
importance, 114
levels and sets, 116, 116f
logical types, 115
nested hierarchies, 117f, 118
plant competition, 114
process-functional conception model,
115, 115f
scale factors, 114–115, 119
spatiotemporal variability, 116f, 116–117
typology, 119
high-nutrient low-chlorophyll (HNLC)
ecosystems, 454
historical research, 16
indirect interactions, 81–91
background information, 81
basic concepts, 81
classification, 84
comparative theoretical ecosystem analysis
(CTEA), 90
definitions, 84
future research areas, 90
measurement techniques, 88
occurrences
abiotic environments, 86
aquatic environments, 85
evolutionary role, 88
global relevance, 87
industrial ecology, 87
terrestrial ecosystems, 85
research areas
apparent competition, 82, 83f
exploitative competition, 83, 83f
indirect mutualism/commensalism, 83, 83f
interaction modification, 83, 83f, 84f
interference competition, 83, 83f
interspecific competition, 82, 83f
trophic cascades, 82, 83f, 85
research issues
complex systems, 89
environmental management, 90
modeling studies, 85, 90
spatial separation, 89
system boundaries, 90
time lags, 89
uncertainty, 89
large-scale shifts, 19, 20f
regime shifts, 19, 20f
river system dynamics, 351–362
conservation strategies, 362
energy flow
energy sources, 351, 352f, 353f
functional feeding groups (FFGs), 352,
354f, 354t
matter fluxes
nutrient cycling, 355, 355f
organic matter, 351, 352f, 355
modeling studies
biotic diversity patterns, 359f
ecotones, 359
flood pulse concept, 344, 359t, 360
hierarchical scales, 359, 359t
hydraulic stream ecology, 359t, 361
hyporheic dynamics, 359t, 361
network dynamics, 359t, 361
patch dynamics, 359t, 361
riparian zone influences, 359t, 360, 360f
river continuum concept (RCC), 344, 356,
357f, 358f, 359t
serial discontinuity, 357
organic matter
characteristics, 356
energy sources, 351, 352f, 354f
retention potential, 356
sources, 356
transport/storage mechanisms, 355
salinity/salinization
saline/soda lakes
Dead Sea, 384
eastern African soda lakes, 381f, 382, 382f
general discussion, 382
Mono Lake (USA), 383
salt marsh ecosystems, 392
self-organization
basic concepts, 98
biosphere, 105
communities versus assemblages, 104
evolutionary development, 105
food webs, 104
fragmented populations, 105f
persistence, 104
social groups, 103
spatial pattern formation, 104, 104f
stability analysis, 104
system concepts, 120–128
evolutionary development, 121
general discussion, 120
orientation theory
animat modeling studies, 126
basic orientors, 122, 123f, 124t
complex environments, 121
environment properties, 122, 123f
goal functions, 124t
implicit attractors, 123
influential factors, 124, 125f
orientor properties, 123
stimulus-response relationships, 124, 125f
system organization, 120, 121f
temporary water bodies, 438
thermodynamic hypothesis
basic concepts, 137
Le Chatelier’s Principle, 137
trophic transfer efficiency, 19
See also biodiversity
ecosystems cycles, 4
ecosystem services, 26–33
categories, 27, 27t
cost-benefit analysis, 31, 32t
forest ecosystems, 30, 31f
general discussion, 26, 32
Millennium Ecosystem Assessment (MA), 10
pollination, 24, 24f, 28
salt marsh ecosystems
carbon sequestration, 389
commercial fishing, 389
denitrification processes, 389
forage resources, 389
recreational opportunities, 389
shoreline protection, 389
swamps, 416
tradeoffs, 27, 28f, 31f
urban environments, 465, 465t
valuation criteria, 31, 32t
wetland ecosystems, 29
ecotones
alpine ecosystems
491
facilitation-competition comparisons, 161f, 165
snow cover, 155
treeline vegetation zone, 156, 157f,
161f, 162
vegetation zones, 151, 151f
definition, 16
estuarine ecosystems, 247
hierarchy theory, 116
stream ecosystems, 359
lead (Pb)
salt marsh ecosystems, 390, 392
Ectocarpaceae, 288t
ectomycorrhizal fungi, 397
ectotherms
body size-abundance distributions, 46
Ecuador, 152f, 311
eddy systems
upwelling ecosystems, 454–455
Eden Project (United Kingdom), 189
self-organization, 99, 101
eels
stream ecosystems, 373
effectiveness theory, 122, 124t
eggs
eggshell containers, 429t
egrets
African floodplain ecosystems, 259–260
water level changes, 277
Egypt, 223t
Eidmanella spp., 193
Ekman layer, 450, 453–454
Elatine, 327
See also solar radiation/solar energy
elephants
savanna ecosystems, 400
elephant seals, 376f
elk
boreal forest ecosystems, 182
riparian systems, 348
temperate forest biomes, 423
Elmidae, 372f
El Niño Southern Oscillation (ENSO)
desert stream ecosystems, 216
estuarine ecosystems, 252
tropical ecosystems, 441
upwelling ecosystems, 459, 459f
Elodea nuttallii, 176
Elton, Charles, 16, 44
Elymus arenarius, 242
Elytrigia spp., 246
Elytropappus, 320t
Emberger pluviothermic quotient (Q2), 319
emergence/emergent properties, 91–98
classification, 96, 96f
complex systems, 109, 110f, 110t
general discussion, 91
hierarchical structure
basic concepts, 92
biological systems, 93, 95t
classification, 96f
physico-chemical processes, 93
protobiological processes, 93
thermodynamic analysis, 118
historical concepts, 92
quantification, 97, 97f
self-organization, 94, 96f, 98, 100
emergence relative to a model, 92, 96f
ecological systems, 34
process development, 99
Emmenanthe penduliflora, 199
Emperor penguins, 342
lagoon ecosystems, 298
empty cans, 428t
ENA, 4
Enchelyidae, 288t
Enchytraeidae, 148
desert environments, 239, 240f
492
Index
Enchytraeidae (continued )
grasses, 240
invasive species, 240
predators, 239
endorheic environments
desert stream ecosystems, 218t
body size-abundance distributions, 46
intertidal environments, 377
energetic equivalence rule (EER), 48
energetics
desert stream ecosystems, 218
energy flow, 8, 9t
photosynthesis, 219
energy, See exergy
energy balance, See global energy balance
Engelmann spruce, 158, 161f
Engraulis spp., 455
Enteromorpha spp., 389
Entisols, 269, 395–396, 421
Entomoneidaceae, 288t
Entoprocta
exergy calculations, 134t
ecological systems, 34
emergence/emergent properties, 95t
fundamental ecosystem theory, 34, 36t
irreversible processes, 34
orientation theory, 124t
See also exergy
environ analysis, 4, 76–81
ascendancy, 57–64
autocatalysis, 57–58
average mutual information (AMI)
uncertainty analysis, 59, 59f
background information, 57
basic concepts, 57
biomass inclusive ascendancy, 62
calculation techniques, 60
complex systems, 57
conditional probabilities, 57, 58t
calculation techniques, 60
ecosystem applications, 63
limiting factors, 62
overhead calculations
basic concepts, 61
exports, 61
imports, 61
redundancy analysis, 62
respiration, 61
total system throughput (TST), 58
background information, 76
basic concepts
boundary zones, 76
input-output models, 76
partitioning processes, 76–77
community assembly rules, 77
data requirements, 77
energy analysis, 64–75
analysis levels, 65
applications
consumer goods and services/cost of living,
72, 73t, 74f, 75f
energy taxes, 74, 74f, 75f, 75t
food webs, 71, 72f, 73t
dynamic system indicators
calculation techniques, 70, 70f
modeling studies, 71, 71f, 72f
energy intensities
basic concepts, 65
consumer goods and services/cost of living,
72, 73t, 74f, 75f
dynamic system indicators, 72f
embodied energy flows, 66f, 67f
energy taxes, 74, 74f, 75f, 75t
explicit energy flows, 67f
feedback function, 67f, 68f
Russian bog food web, 73t
two-compartment steady-state analysis,
65, 65f, 66f
indirect interactions, 64
two-compartment steady-state analysis
balance equations, 69t
energy intensities, 65, 65f, 66f, 67f, 68t
input-output models, 65, 65f, 66f, 68t
nutrient intensities, 67, 67f, 68f,
68t, 69t
path length, 68t, 69, 69t, 72f, 73t
residence time, 68t, 69, 69f, 69t, 72f, 73t
terminology, 66t
trophic position, 68, 68t, 69f, 69t,
72f, 73t
hierarchy theory, 118–119
methodology, 77, 77f, 78t
network properties
direct interactions, 79, 80t
general discussion, 79
homogenization effects, 79
indirect interactions, 79, 80t
mutualism, 79, 80t
quantitative environ, 80
purpose, 80
sample network, 77, 77f
environmental forest plantations, 264
indirect interactions, 90
life cycle assessment (LCA)
indirect interactions, 87
environmentally sensitive business, 14
Environmental Protection Agency (EPA)
kinetic mechanisms, 173
Michaelis-Menton equation
wastewater treatment, 173
Monod equation
wastewater treatment, 174
wastewater treatment, 173
Eocene
Mediterranean ecosystems, 325
Epacridaceae, 320t
ephemeral streams, 225, 225f, 428t, 435f
Ephemeroptera
stream ecosystems, 354, 354t, 372f, 372–373
Ephydra hians, 383
epidemiology
estuarine ecosystems, 249t, 250
temporary water bodies, 431, 431f
tropical rainforests, 441, 442f
Epinephelus striatus, 213
epiphytes, See epifauna/epiflora
Epithemiaceae, 288t
Equatorial Atlantic Current, 450f, 454
Equatorial Pacific Current, 450f, 454
Equus hemionus, 240
ergs, 224f, 224–225
See also sand dunes
Ericaceae, 244
Erica spp., 151, 159t, 320t
Erica tetralix, 244
Erigone spp., 193
Eriophorum spp., 334
ermines, 447
Erodium spp., 320t, 327
alpine ecosystems, 155
freshwater lakes, 273
savanna ecosystems, 399f
shelterbelts/windbreaks, 470
tundra ecosystems, 444
Espeletia spp., 151, 152f
estuarine ecosystems, 247–252
complex systems, 248
definitions, 247, 247f
food webs, 249, 249f, 250, 252
freshwater-seawater interface, 247
general discussion, 247
geomorphology
bar-built/lagoonal estuaries, 247
riverine estuaries, 248, 248f
mangrove forests, 308
material fluxes
global climate effects, 252
organismic transport, 251
residence time, 250, 251f
water fluxes, 250
mature systems, 248
microbial-viral loop food web, 249, 249f
nitrogen toxicology
resilience, 252
subsystems
habitats, 248
intertidal environments, 249
invertebrate reefs and beds, 250
mudflats/sand flats, 250
seagrasses, 250
water column, 248
swamps, 414, 414f
See also salt marsh ecosystems;
tidal wetlands
Eubranchipus spp., 436
Eucalanus inermis, 453
Eucalyptus camaldulensis, 268
Eucalyptus dichromophloia, 396
Eucalyptus horistes, 268
Eucalyptus kochii, 268
Eucalyptus miniata, 396, 399
Eucalyptus pruinosa, 396
Eucalyptus robusta, 268
Eucalyptus spp.
alpine ecosystems, 151, 159t
floodplain ecosystems, 261
invasive species, 267
Mediterranean ecosystems, 320t, 326
savanna ecosystems, 395f, 396, 404
temperate forest biomes, 417
Eucalyptus tereticornis, 268
Eucalyptus tetrodonta, 396
Eucinostomus gula, 295f
Eudendridae, 288t
Euglenozoa, 288t
eukaryotes
cell structure, 166, 167t
estuarine ecosystems, 248
saline/soda lakes, 381–382
upwelling ecosystems, 454
Eunicidae, 288t
Euphasia lucens, 454
euphausiids
upwelling ecosystems, 453–454, 455f
Euphorbiaceae, 230
Euphorbia spp., 230
Euphrasia arctica, 446
Euphrates River, 238, 276
Euplotidae, 288t
Europe, 140
alpine ecosystems, 150, 152f, 157, 159t, 160f
botanic gardens, 184t, 185
dune ecosystems, 241, 244
floodplain ecosystems, 257t, 261
grasslands, 406
salt marsh vegetation, 386t, 391, 393
savanna ecosystems, 394
temperate forest biomes, 426
temporary water bodies, 429t
European beech, 158
European larch, 158
European spruce, 158
Eurythor complanata, 294f
eusociality, 101
eutrophication
coral reefs, 205f
lake ecosystems
freshwater lakes, 272, 272f
large-scale whole ecosystem experiments, 21
salt marsh ecosystems, 390
evaporation, See evapotranspiration; water (H2O)
evaporative water loss (EWL)
desert environments, 214, 216, 227, 227f
emergence/emergent properties, 95t
Index
hydrologic cycle
schematic diagram, 365f
landfills, 305
salt marsh ecosystems, 391
savanna ecosystems, 398f
temperate forest biomes, 423
tropical ecosystems, 439–440
United States, 261
Evechinus chloroticus, 21, 22f
Everglades, Florida (USA)
biodiversity, 279f
Florida Everglades mesocosm project
air/water movement, 284f
algal biomass, 293f
ecosystem subunits, 285f
fish distribution, 295f
general discussion, 290
invertebrate abundance, 294f
physico-chemical parameters, 292t
plan diagram, 291f
plant communities, 291f
structural characteristics, 283f
vertical/longitudinal section, 292f
water supply, 284f
lagoon ecosystems, 296–297
water availability, 277
evergreen forests
alpine ecosystems, 159t
physiographic regions, 417, 419t
savanna ecosystems, 397
evolution
self-organization, 105
desert environments, 230
exergy, 4–5, 128–139, 5t
anthropogenic effects, 136, 136t
calculation techniques, 132
definition, 128, 128f
dissipative structure, 132
eco-exergy
basic concepts, 130, 130f
ecological systems, 34
exergy calculations, 132, 134t
fundamental ecosystem
theory, 34, 36f
living organisms, 132, 134t
ecological systems
irreversible processes, 34
emergence/emergent properties, 95t
fundamental ecosystem theory, 34, 36f
information theory, 131, 131f, 135
loss calculations, 136, 136t
nitrogen-phosphorus ratio plot, 139f
organic matter, 132, 134t, 137, 138t
thermodynamic hypothesis
basic concepts, 137
Le Chatelier’s Principle, 137
existence, 122, 124t
exorheic environments, 218t
exotic species
forest plantations, 267
freshwater lakes, 273
savanna ecosystems, 404
See also invasive species
exploitative competition
network environ analysis, 80t
research areas, 83, 83f
extinction models
desert environments, 239, 240f
grasses, 240
invasive species, 240
predators, 239
extinction rates
urban environments, 464
Mediterranean ecosystems, 323
tundra ecosystems, 449
extreme environments
coral reefs, 205, 205f
desert environments, 226, 230
desert stream ecosystems, 216, 218t
tundra ecosystems, 444, 447
Exuma Cays, Bahamas, 212f
F
Fabaceae
Mediterranean ecosystems, 320t
facilitation models, 245
Fagus spp.
Mediterranean ecosystems, 321t
temperate forest biomes, 417
Fagus sylvatica, 158
fairness, See cooperation
fairy shrimp, 433, 436
falcons
tundra ecosystems, 447
farm forest plantations, 264
farming systems
bee pollination, 28
desert environments, 237
historical background, 237
shelterbelts/windbreaks, 469
fauna, See animals
Faviidae, 288t
feces
coral reefs, 206
feedback
emergence/emergent properties, 95t
food webs, 101, 102f
fundamental ecosystem theory, 34, 36t
orientation theory, 124, 124t, 125f
self-organization, 101, 102f, 104
simple systems, 107–108
Fennoscandia, 181
fens
characteristics, 175, 275, 333,
334, 429t
development factors, 332f
largest wetlands, 281
patterned fens, 335, 336f
poor fens, 335
rich fens, 335, 335f
water availability, 277
ferns
Mediterranean ecosystems, 327
temporary water bodies, 433
Ferrissia spp., 371
fertilization
forest plantations, 269
indirect interactions, 87
Festuca abyssinica, 448
Festuca rubra, 243f
Festuca spp., 88
Ficus spp., 259
Fiji, 441
filamentous organisms, 168f, 177
Fimbrisitylis bahiensis, 244
fine particulate organic matter (FPOM), 349,
352f, 354f, 356, 358f, 368
fingernail clams, 371, 372f, 436
Finland
boreal forests, 181
peatlands, 331
polar ecosystems, 339–342
Finn, JT, 53–54
Finn’s cycling index, 52
fire-dependent regeneration, 195–196
Firehole River, Yellowstone National Park
(USA), 357f
first law of thermodynamics, See exergy;
thermodynamic laws
fir trees
boreal forest ecosystems, 181
fir waves, 420
fish/fisheries
African floodplain ecosystems, 259–260
mangrove forests, 313–314
saline/soda lakes, 384
493
Australian floodplain ecosystems, 261
competition studies, 83
coral reefs
competition, 203
importance, 203, 203f, 207
overfishing, 205, 205f, 210f,
211, 211f
positive interactions, 205
predators, 204
recruitment effects, 206
desert stream ecosystems, 217–218, 229
eco-exergy losses, 136
ecosystem dynamics, 21, 22f, 23f
estuarine ecosystems, 249, 252
exergy calculations, 134t
floodplain ecosystems
Africa, 259–260
Australia, 261
marshes, 276
South America, 262–263
Florida Everglades mesocosm
project, 295f
food webs
upwelling ecosystems, 455, 455f
freshwater ecosystems, 373
indirect interactions, 85
intertidal environments, 375, 377
lake ecosystems
saline/soda lakes, 381, 382, 384
mangrove forests, 310, 312
marshes, 278
Mediterranean ecosystems, 322, 322t
overfishing
coral reefs, 205, 205f, 210f,
211, 211f
freshwater lakes, 274
indirect interactions, 85
recreational fishing, 384
recruitment
optimal environmental window (OEW),
457, 457f
upwelling ecosystems, 457, 457f
riparian systems, 344
salinity
saline/soda lakes, 381, 382, 384
salinity tolerances, 373
salt marsh ecosystems, 386, 387, 389
South American floodplain ecosystems,
262–263
stable isotope analysis, 21
stream ecosystems
characteristics, 373
desert environments, 217–218, 229
temporary water bodies, 433, 434
trophic transfer efficiency, 19
upwelling ecosystems, 451
climate change effects, 458, 458f
fish production, 457, 457f
food webs, 455, 455f
general discussion, 455
small pelagic fish, 455, 456f
trophic levels, 457
Fissurellidae, 288t
fitness variance
implications
fragmented populations, 105, 105f
flagellates
agroecosystems, 148
estuarine ecosystems, 249
indirect interactions, 86
temporary water bodies, 436–437
upwelling ecosystems, 453, 455f
flagged branching effects, 156, 157f,
161f, 162
flamingoes
desert environments, 225f
marshes, 278, 279f
saline/soda lakes, 381f, 382, 384
494
Index
flamingoes (continued )
water level changes, 277
flarks, 335, 336f
flash floods, 215–216, 217f, 218t, 219f, 225,
226f, 344
flatworms
exergy calculations, 134t
intertidal environments, 376
temporary water bodies, 433
flickers
temperate forest biomes, 423
flies
exergy calculations, 134t
pollen vectors, 24
saline/soda lakes, 383
stream ecosystems, 373
temporary water bodies, 433
flightless beetles, 448
Flint River, Georgia (USA), 256f
flocs, 168, 168f, 177
floodplain ecosystems, 253–263
Africa, 257t, 259
anthropogenic impacts, 259
Asia, 257t, 260
Australia, 257t, 261
biogeochemical processes, 255
biomass productivity, 258, 258f
blackwater systems, 255, 256f, 262
cross-section diagram, 254f
Europe, 257t, 261
general discussion, 263
geomorphological processes, 253, 254f
hydrologic processes, 254
importance, 253, 276
marshes, 275, 276, 276f, 281
North America, 257t, 261
nutrient cycling, 255, 256f
panoramic view, 253f
river-floodplain interactions, 254, 359t, 360
seasonal pools, 428t, 429t, 432, 433
sedimentation processes, 253, 256f
South America, 257t, 262
swamps, 414, 415
topographic diagram, 254f
vegetation
nutrient cycling, 255
physiological adaptations, 257
structural characteristics, 257, 257t
whitewater systems, 262
wildlife populations, 276
floods
Asian floodplain ecosystems, 260
desert environments, 214, 215–216, 217f, 218t,
219f, 225, 226f
river-floodplain interactions, 254, 359t, 360
vegetation adaptations, 257
Florida Everglades mesocosm project
air/water movement, 284f
algal biomass, 293f
ecosystem subunits, 285f
fish distribution, 295f
general discussion, 290
invertebrate abundance, 294f
physico-chemical parameters, 292t
plan diagram, 291f
plant communities, 291f
structural characteristics, 283f
vertical/longitudinal section, 292f
water supply, 284f
Florida (USA)
botanic gardens, 184t
Everglades
biodiversity, 279f
lagoon ecosystems, 296–297
water availability, 277
Florida Everglades mesocosm project
air/water movement, 284f
algal biomass, 293f
ecosystem subunits, 285f
fish distribution, 295f
general discussion, 290
invertebrate abundance, 294f
physico-chemical parameters, 292t
plan diagram, 291f
plant communities, 291f
structural characteristics, 283f
vertical/longitudinal section, 292f
water supply, 284f
Kissimmee River, 259
mangrove forests, 310, 313
savanna ecosystems, 394
Floridichthys carpio, 295f
flowers, See plants
flukes, 433
fluorescent in situ hybridization (FISH), 169
flycatchers, 423
fog
desert environments, 226
dune ecosystems, 243
temperate forest biomes, 423
foliovores, 310
Folliculinidae, 288t
Folsomia candida, 85
Folsomia inoculata, 85
food industry
ecosystem services, 27t
urban environments, 465, 465t
algae, 383
autocatalysis, 41, 42f
cave habitats, 193
coral reefs
herbivores, 203
overfishing, 211, 211f
predators, 204
cycle analysis
direct interactions, 50, 51f
feeding cycles/nonfeeding cycles, 51
indirect interactions, 51, 51f
number of simple cycles, 54, 55t
desert stream ecosystems, 218
ecological network analysis (ENA), 71,
72f, 73t
estuarine ecosystems, 249, 249f, 250, 252
feedback, 101, 102f
fish/fisheries, 455, 455f
herbivores, 280
indirect interactions, 85
lake ecosystems, 383
landfills, 305
mangrove forests, 310, 312
marshes, 280
matrix models, 50
ecosystem dynamics, 21, 22f
riparian systems, 344
river system dynamics, 352
salt marsh ecosystems, 388
self-organization, 104
stable isotope analysis, 21
temperate forest biomes, 423–424
upwelling ecosystems, 455, 455f
footprint analysis, See Ecological Footprint
analysis
foraminifera
upwelling ecosystems, 453
forbs
grasslands, 411, 411f
marshes, 276f
temperate forest biomes, 420
tundra ecosystems, 446
Forcipomyia pholeter, 193
forests
acidic deposition
forest plantations, 264–270
alpine forests, 156–165
abiotic environments
aboveground, 159, 165
belowground, 162
altitude effects, 162
biogeography, 157, 159t, 160f
characteristics, 156, 157f, 161f
facilitation-competition comparisons,
161f, 165
influential factors, 164t
microclimates, 162
treeline ecotone, 162
treeline formation mechanisms, 163,
164t, 165
boreal forests, 181–183
animals, 182
biodiversity, 182
biomass estimates, 182
bogs, 334
characteristics, 181
climate, 181
climate change effects, 182
ecosystem dynamics, 182
peatlands, 331
primary production, 234t
soils, 181
structural characteristics, 181
temperate forest biomes, 417–418
temporary water bodies, 428t, 429t, 432,
432f
wildfires, 182
conifers/coniferous forests
dune ecosystems, 246
mixed coniferous-deciduous forests,
417–418, 418f
physiographic regions, 419t
deciduous forests
alpine ecosystems, 159t
mixed-deciduous temperate forest biomes,
417, 418f, 419t
primary production, 234t
savanna ecosystems, 397
mangrove forests, 313–314
temperate forest biomes, 426
desert stream ecosystems, 218
ecosystem services, 30, 31f
evergreen forests
alpine ecosystems, 159t
physiographic regions, 417, 419t
savanna ecosystems, 397
floodplain ecosystems
Australia, 261
Europe, 261
litterfall, 255–256
physiological adaptations, 257
South America, 262
structural characteristics, 257, 257t
United States, 261
forested wetlands
mangrove forests, 308–318
biodiversity, 310
characteristics, 308
coral reefs, 207, 208f
ecogeomorphology, 308, 309f, 310f
environmental impacts, 313
estuarine ecosystems, 250
food webs, 310, 312
gradients, 308, 309f, 310f
greenhouse ecosystems, 284f
hydroperiod (flood duration), 308,
309f, 310f
management strategies, 314, 314f
nutrient resources, 311, 312f
photographic views, 17f
productivity, 311
restoration methods, 314
succession, 311
swamps, 414–417
Africa, 259
aquatic environments, 416
characteristics, 175, 275
Index
definition, 414
ecological functions, 415
ecosystem services, 416
habitat conditions, 416
hardwood forests, 414f
hydrologic characteristics, 414, 414f
hydrologic functions, 415
largest wetlands, 281
peat formation, 338
restoration methods, 416
riparian systems, 346f, 347
seasonal pools, 428t
soils, 415, 415f
terrestrial ecosystems, 416
United States, 262
vegetation, 415
water quality, 415
forest plantations, 264–270
basic concepts, 264
conservation strategies, 267
ecological effects, 267
economic factors, 265, 265f
exotic species, 267
global distribution, 264, 264f, 265f
global production, 265f
influential factors
government policies, 266
production costs, 266
timber prices, 266
management strategies, 269–270
timber production, 265
hardwood forests, 181, 281, 414f
hierarchy theory, 115, 115f
hydropower dam construction and operation,
30, 31f
Mediterranean ecosystems
biogeographical evolution, 323
characteristics, 320t, 326
climate, 321t
wildfires, 328
modified natural forests, 264f, 265f
postclosure landfills, 304, 306
primary forests, 264f, 265f
reforestation, 266
riparian zones, 218
riverine forests, 281
seminatural forests, 264f, 265f
temperate forests, 417–427
carbon balance, 425
characteristics, 417
climate, 418, 419t
disturbances
characteristics, 420
detritus, 421, 421f
soils, 421
structural layers, 420
ecological communities
faunal communities, 422
succession, 421
vegetation communities, 421, 422f
energy flow, 423
evapotranspiration, 423
global distribution, 417, 418f
land cover
historical land cover, 425
present-day land cover, 426
mean annual temperatures, 418
mixed coniferous-deciduous forests,
417–418, 418f, 419t
nutrient cycling, 424, 424f, 425f
physiographic regions, 418, 419t
precipitation, 418
temporary water bodies, 428t, 429t, 432,
432f
water cycle, 423
primary production, 234t
temporary water bodies, 429t, 431
See also leaves; plants; temperate forests
fossil fuels
eco-exergy losses, 136, 136t
indirect interactions, 87
Fouquieria columnaris, 229, 229f
foxes
boreal forest ecosystems, 182
competition studies, 83–84, 84f
polar ecosystems, 340–341
tundra ecosystems, 447
fractal landscapes
hierarchy theory, 119
France
botanic gardens
colonial period, 186
Jardin des Plantes, 185, 185f
medicinal gardens, 185
selected gardens, 184t
forest plantations, 265f
temporary water bodies, 429t
Frankenia spp., 386t
Franseria pinnatifida, 243f
Fraxinus spp.
Europe, 261
floodplain ecosystems, 258
United States, 262
freedom of action, 122, 124t
freezing behaviors, 236
Fremont Valley, California (USA), 224f
Fresh Kills Landfill, New York (USA), 304, 466
freshwater ecosystems
acidic deposition
ionic composition, 369–370
aquatic wetlands, 275
estuarine ecosystems, 249
fish, 373
freshwater-seawater interface, 247
indirect interactions, 85
lake ecosystems, 270–274
characteristics, 270
Crater Lake, Oregon (USA), 271f
importance, 271
Lake Baikal, 271f, 271t
major lakes, 271, 271t
water quality problems
acidification, 273
eutrophication, 272, 272f
exotic species introductions, 273
overfishing, 274
pathogenic contamination, 274
salinization, 273
siltation, 273
toxic substances, 273
water-level changes, 273
marshes, 274–281
anthropogenic impacts
dam construction, 279
draining effects, 279
food webs, 280
general discussion, 279
nutrient resources, 279–280
road networks, 280
biodiversity, 278, 279f
characteristics, 275, 275f, 280
environmental factors
disturbances, 278
nutrient cycling, 278
general discussion, 274
geographic distribution, 278
restoration methods, 280
vegetation, 275f, 277, 277f
water availability
floodplains, 276, 276f
general discussion, 275
peatlands, 276–277
seepage wetlands, 277
temporary wetlands, 277
water level changes, 277, 277f
West Siberian Lowland, 276f
urban environments, 465, 465t
wet meadows, 275
See also lagoon ecosystems
frogs
marshes, 278
temperate forest biomes, 423
temporary water bodies, 433, 437
temporary wetlands, 277
frugivores, 310
Fsciolariidae, 288t
fucoid seaweeds, 375–376, 377
Fundulus chrysotus, 295f
Fundulus confluentus, 295f
Fundulus grandis, 295f
Fundulus majalis, 387
Fundulus similis, 295f
fungi
desert environments, 227
exergy calculations, 134t
forest plantations, 268
fungal cap concavities, 429t
indirect interactions, 85, 88
mycorrhizas
bogs, 333
desert environments, 227
ectomycorrhizal fungi, 397
symbiotic relationships
dune ecosystems, 246
root systems, 397
vesicular arbuscular mycorrhizae
(VAM), 246
stream ecosystems, 356
temperate forest biomes, 420, 423–424
fungicides
agroecosystems, 147
fynbos ecosystems, 320t, 325, 326
G
Gabon, 440
Daisyworld model, 103, 105
global energy balance, 105
Galápagos Islands
intertidal environments, 377
gall-forming insects, 236
gall wasps, 196
Gambusia affinis, 295f
Game of Life (Conway)
basic concepts, 102
Gammaridae, 288t
Gammarus spp., 294f
Ganges River
floodplain ecosystems, 260
mangrove forests, 313
temporary water bodies, 433
gannets, 457
garrigue ecosystems, 326
gasoline
eco-exergy losses, 136
Gasteromycetes spp., 227–228
Gastrophysa viridula, 88
gastropods (Gastropoda)
exergy calculations, 134t
grazing populations, 378
intertidal environments, 375, 376, 378
salt marsh ecosystems, 387
stream ecosystems, 371, 372f
gastrotrichs, 433
gator holes, 429t
gazelles
desert environments, 230
geese
marshes, 278, 280
salt marsh ecosystems, 388
tundra ecosystems, 447
water level changes, 277
Gelidiaceae, 288t
Gelisols, 181
495
496
Index
genetics
genetic drift
urban environments, 463t
Genista, 320t
genon, 118–119
Gentiana puncata, 154f
geochemical cycles, See biogeochemical cycles
Geodiidae, 288t
geogenous peatlands, 332
geophytes, 195–196, 233–234
Georgia (USA)
estuarine ecosystems, 250
salt marsh ecosystems, 388
gerbils, 230
Germany
botanic gardens, 184t, 186
urban environments, 462
Gexhouba Hydroelectric Power Plant,
China, 30
Ghana
forest plantations, 267
tropical rainforests, 442
giant lobelias, 449
giant sequoias, 44, 417
Gibb’s free energy, 130
gibbsite, 395–396
Gibson Desert, 223t
gidgee, 396
gila monsters, 236
giraffes, 400
glaciation/glaciers
desert stream ecosystems, 218t
Last Glacial Maximum (LGM), 324
Pleistocene, 321, 324
temperate forest biomes, 418
temporary water bodies, 429t
tundra ecosystems, 444
global ecology, 87
global emergence, 96f
ecosystems
Gaia hypothesis, 105
See also solar radiation/solar energy
global warming
alpine ecosystems, 155
indirect interactions, 87
polar ecosystems, 341
salt marsh ecosystems, 389
tundra ecosystems, 449
Glossosoma nigrior, 353f
Glyceria maxima, 176, 244
glycogen, 170, 170f
Glycymerididae, 288t
gnammas, 429t, 432
gnathostomes
exergy calculations, 134t
goal functions, 120–128
goats
desert environments, 230, 238
grasslands, 411
savanna ecosystems, 400
Gobi Desert, 223t, 241
gobies, 206
Gobiidae, 206, 377
Gödel’s theorem, 34
goldspotted killifish, 295f
Golfingiidae, 288t
Gomphonema spp., 370f
Goniotrichaceae, 288t
Gonyaulacaceae, 288t
Gopherus agassizii, 240
gordian worms, 433
gradients
estuarine ecosystems, 247, 247f
estuarine ecosystems, 247, 247f
mangrove forests, 308, 309f, 310f
stream ecosystems, 373
urban environments, 463
graminoid grasses, 153, 154f, 446, 446f
Grammidae, 288t
granivores, 229, 236
grasses
Africa, 259
alpine ecosystems, 153, 153f, 154f
aquatic wetlands, 275, 276f
defense strategies, 397
desert environments, 229, 238
desert stream ecosystems, 218
dune ecosystems, 242, 245
endangered species, 240
grazing rates, 83–84, 84f
growth forms, 405, 405f, 408f
invasive species, 238
marshes, 276f, 278
morphology, 405, 405f, 408f
photosynthetic pathways, 406
postclosure landfills, 305
riparian zones, 218
savanna ecosystems, 394
temporary water bodies, 433
tundra ecosystems, 445f, 446, 446f
grasshoppers
savanna ecosystems, 400
stream ecosystems, 373
grasslands, 405–413
as agroecosystems, 145
alpine ecosystems, 151, 152f
animals, 410
climate, 407
conservation strategies, 412
disturbances, 408, 409f
geographic distribution, 406, 407f
grass morphology, 405
grazing systems, 329, 411, 411f
indirect interactions, 88
Mediterranean ecosystems
biogeographical evolution, 323
characteristics, 320t, 327
grazing, 329
primary production
aboveground net primary production
(ANPP), 409f
occurrences, 406
postclosure landfills
faunal communities, 307
use patterns, 304
vegetation, 305
prairie ecosystems, 405–413
shortgrass prairies, 406
tallgrass prairies, 406
primary production, 234t
restoration methods, 412
seasonal pools, 428t, 429t, 433
shortgrass prairies, 406
steppe environments, 405–413
Mediterranean ecosystems, 321t, 327
shortgrass prairies, 406
tallgrass prairies, 406
threats, 412
wetland ecosystems, 281
wildfires, 408, 409f
gray jay, 158
gray squirrel, 268
gray tree frogs, 279
agroecosystems, 147
alpine ecosystems, 155
competition studies, 83
dune ecosystems, 246
grasslands, 329, 411, 411f
intertidal environments, 376, 378
marshes, 278
Mediterranean ecosystems, 329
overgrazing effects, 155, 278, 399f, 411
savanna ecosystems, 398f
temperate forest biomes, 425
upwelling ecosystems
phytoplankton, 452
zooplankton, 453
Great Altas Mountains, 159t, 160f
Great Barrier Reef, Australia, 206, 207f
Great Basin Desert, 223t, 228
great crested newts, 307
Great Lakes (USA)
characteristics, 271t
freshwater ecosystems, 381
marshes, 276f
temperate forest biomes, 421
Great Plains (USA), 406, 412, 434f
Great Salt Lake, Utah (USA), 381, 385
Great Sand Dunes National Park Preserve,
Colorado (USA), 241
Great Vasyugan Mire, Siberia, 331, 331f
Great Victoria Desert, 223t
grebes, 383, 447
Greece, 183
green algae
desert stream ecosystems, 216
Florida Everglades mesocosm project, 293f
river system dynamics, 359f
saline/soda lakes, 384
salt marsh ecosystems, 387, 389
stream ecosystems, 370
temporary water bodies, 433
green buildings, 14
greenhouse ecosystems, 281–295
basic concepts, 281
Biosphere II project, Arizona (USA), 281, 292
coral reef microcosm
carbonate cycle, 287f
general discussion, 286
gross primary production (GPP), 287f
organism families, 288t
oxygen (O) concentrations, 286f
physico-chemical parameters, 287t
schematic diagram, 286f
Florida Everglades mesocosm
algal biomass, 293f
fish distribution, 295f
general discussion, 290
invertebrate abundance, 294f
physico-chemical parameters, 292t
plan diagram, 291f
plant communities, 291f
vertical/longitudinal section, 292f
Florida Everglades mesocosm project
air/water movement, 284f
ecosystem subunits, 285f
structural characteristics, 283f
water supply, 284f
operational requirements, 285
physico-chemical control parameters
air/water movement, 283, 284f
ecosystem interchange, 285
ecosystem structure, 284
ecosystem subunits, 284, 285f
light requirements, 283
physico-chemical factors, 283
structural characteristics, 282, 283f
water supply, 283, 284f
Greenland
intertidal area research, 378
tundra ecosystems, 445
green turtles, 211–212
Grevillea, 320t
Grewia spp., 396–397
grey herons, 259–260
Grindstone Neck, Maine (USA), 378f
gross primary production (GPP)
coral reef microcosm, 287f
energy flow, 9t, 65–66
groundwater
desert stream ecosystems, 221
Europe, 261
hyporheic zone, 214f, 214–215, 359t, 361
riparian systems, 344
Index
river-floodplain interactions, 255, 359t, 361
water cycle, 364, 365f
group behaviors, 103
groupers
coral reefs
ecological interactions, 204, 204f
marine reserves, 212f, 213
overfishing, 211, 211f
positive interactions, 206
overfishing, 211, 211f
grouses, 182
grunts, 207, 208f, 295f
guanay cormorant, 457
guano communities, 191, 377, 457
Guayas River, Ecuador, 311
gulf killifish, 295f
Gulf of Mexico, 278
gulf toadfish, 295f
gulls, See seagulls
guppies, 295f
gutters, 428t
Gymnodiniaceae, 288t
gymnosperms
exergy calculations, 134t
H
alpine ecosystems, 153
body size-species distributions, 48
corridors, 349, 463
dune ecosystems, 244, 245f
environmental impact assessments
general discussion, 248
intertidal environments, 249
invertebrate reefs and beds, 250
mudflats/sand flats, 250
seagrasses, 250
water column, 248
fractal dimensions, 48
indirect interactions, 85
invasive species, 391
loss calculations
salt marsh ecosystems, 390
mangrove forests, 308, 309f
marshes, 278, 279f, 279f
Mediterranean ecosystems, 327
orientation mechanisms (taxis)
self-organization, 104
postclosure landfills
faunal communities, 307
use patterns, 304
vegetation, 305
riparian systems, 348
salt marsh ecosystems, 386
implications
fragmented populations, 105, 105f
shelterbelts/windbreaks, 470
implications
fragmented populations, 105, 105f
swamps
aquatic environments, 416
habitat conditions, 416
terrestrial ecosystems, 416
temperate forest biomes, 422
two-dimensional (2D)/three-dimensional (3D)
habitats, 48
urban environments, 463, 463t
See also riparian systems
Haeckel, Ernst Heinrich, 12
Haematococcus spp., 436–437
Haemulidae, 207
Haemulon macrostomum, 295f
Haemulon sciurus, 207–208, 208f
hake, 453, 457
Hakea, 320t
Halacaridae, 288t
Halichondridae, 288t
Haliclona spp., 294f
Haliclonidae, 288t
Halimeda opuntia, 202–203
Halimeda spp., 287
salt marsh ecosystems, 392
halophilic microorganisms, 381–382, 384
halophytes, 232, 386–387
Halorgidaceae, 371
Hamatocaulis spp., 335
Hamiltonian cycles, 54
hammadas, 224
hammock trees, 291f
Hammurabi (King), 237–238
hanta-virus, 196
hardwood forests, 181,
281, 414f
hares, 446–447
Harpactacoida, 371
Harpacticidae, 288t
Hartmannellidae, 288t
Hawaiioscia spp., 193
Hawaii (USA)
botanic gardens, 184t
cave habitats
cave-like habitats, 192
food webs, 193
nonindigenous species, 193
succession studies, 193
coral reefs, 204
Hawinella spp., 193
hawks
temperate forest biomes, 423
tundra ecosystems, 447
heathlands
temporary water bodies, 429t
Hedophyllum sessile, 18, 19f
Helianthemum spp., 227–228
Helichrysum, 320t
Helipterum, 320t
Hemidiscaceae, 288t
hemi-epiphytes, 441
Hemigrapsus oregonensis, 387
Hemiptera
desert stream ecosystems, 217
stream ecosystems, 372f, 373
hemlock trees, 421, 425f
herbicides
agroecosystems, 147
forest plantations, 268
salt marsh ecosystems, 392
herbivores/herbivory
competition studies, 82–83, 88
coral reefs, 203, 203f
desert environments, 231, 235
food webs, 280
grasslands, 411, 411f
indirect interactions, 88
intertidal environments, 376, 378
marshes, 280
Mediterranean ecosystems, 329
savanna ecosystems, 398f, 400, 401t
stream ecosystems, 353–354
trophic interactions, 17
herbs
alpine ecosystems, 153, 154f
temperate forest biomes, 420
herons
African floodplain ecosystems, 259–260
water level changes, 277
Heterandria formosa, 295f
Heteromeles arbutifolia, 195
Heteromyidae, 230
Heteropogon contortus, 396
Heteropogon spp., 396
heterotrophs
characteristics, 17, 167t
desert stream ecosystems, 219
electron-donor/electron-acceptor
regimes, 167t
saline/soda lakes, 382
497
tropical rainforests, 441
Hevea brasiliensis, 186–187
hickory trees, 421
hierarchical structures
social groups, 103
complex networks, 100
emergence/emergent properties
basic concepts, 92
biological systems, 93, 95t
classification, 96f
physico-chemical processes, 93
protobiological processes, 93
hierarchy theory, 114–120
community conception model, 115, 115f
connection strengths, 117, 117f
historical research, 117
holons, 117, 117f
hypothesis testing, 115
importance, 114
levels and sets, 116, 116f
logical types, 115
nested hierarchies, 117f, 118
plant competition, 114
process-functional conception model,
115, 115f
scale factors, 114–115, 119
spatiotemporal variability, 116f, 116–117
typology, 119
soils, 226
hierarchy theory, 4–5
high-nutrient low-chlorophyll (HNLC)
ecosystems, 454
Himalaya Mountains, 150, 157,
159t, 160f
Hippolytidae, 288t
Hippophae rhamnoides, 244, 246
hippopotamus, 259–260, 280, 348
Hirudinea, 371
Histosols, 181, 415, 415f
hogs, See pigs
holarchy, 118
Holdridge vegetation classification system,
439–440
holistic versus reductionistic research
community assembly, 77
ecosystem ecology, 6
fundamental ecosystem theory, 33
philosophy of ecology, 92
Holland, See Netherlands
Holland, John, 99
holons, 117, 117f
Holothuriidae, 288t
home-based forest plantations, 264
homegardens, 463
self-organization, 101
Homo sapiens
desert environments, 237
exergy calculations, 134t
indirect interactions, 87
Homotremidae, 288t
honeybees
group behaviors, 103
pollen vectors, 28
horned lizards, 230, 236
horse mackerel, 457
horses
desert environments, 230, 238
horticulture, 145
houbara, 239
house flies
See also Diptera; flies
Hudson Bay Lowlands, Manitoba, Canada, 278,
281, 331, 331f, 445f
desert environments
anthropogenic impacts, 238
endangered species, 239, 240f
historical background, 237, 237f
hunter-gatherer societies, 238
498
Index
Hudson Bay Lowlands, Manitoba,
Canada (continued )
nomadic cultures, 238f
research areas, 240
usability of ecology, 12–16
See also environmental space
human-environment interactions
urban environments, 462, 463t, 467
social groups, 103
upwelling ecosystems, 455f
urban environments, 467
anthropogenic activities
Humboldt Current, 450, 450f, 457, 458f
Humboldt penguin, 457
humidity
Mediterranean ecosystems, 321t
tropical ecosystems, 441
Humiria balsmifera, 244
hummock dunes, 241
Hungary
desert environments, 239
tropical ecosystems, 442
hurricanes
coral reefs, 205, 205f
estuarine ecosystems, 252
mangrove forests, 313
salt marsh ecosystems, 389
tropical ecosystems, 441
Hutchinson ratio, 44, 49
Hwang Ho River, 276
Hyalella azteca, 294f
Hyallela spp., 371
Hyalodiscidae, 288t
hybridization
invasive species, 391
Hydra spp., 294f
hydraulic stream ecology, 359t, 361
hydrocarbons
salt marsh ecosystems, 392
Hydrocharitaceae, 288t, 371
Hydrocotyle bonariensis, 244
hydrogen (H)
elemental concentrations, 10t
stable isotope analysis, 20
hydrographs, 364, 365f
hydroids
intertidal environments, 376
characteristics, 364
evapotranspiration
schematic diagram, 365f
hydrographs, 364, 365f
intermittent streams, 364, 365f, 428t, 432f
mangrove forests, 308, 309f
meltwater pools/streams, 428t
runoff, 364, 365f
savanna ecosystems, 394
schematic diagram, 365f
temperate forest biomes, 423
elemental concentrations, 10t
hyenas
desert environments, 235
hygropetric zones, 346, 346f, 347f
Hymenoptera
stream ecosystems, 373
Hynes, Noel, 363
Hyphanene spp., 259
Hyphomicrobiom spp., 178–179
Hypneaceae, 288t
hypolimnion
lake ecosystems
water quality problems, 272
Hypolytus spp., 294f
hyporheic zone
desert stream ecosystems, 220
groundwater, 361
stream ecosystems, 214f, 214–215, 359t
hypoxia, 453
I
ibex, 230
ibis, 277
ice
global warming effects, 342
scouring effects, 278
seasonal pools, 428t
igapo floodplains, 262
iguanas, 377
Illinois (USA), 184t, 406–407
Imperata cylindrical, 397
impossible staircase, 108f
Inceptisols, 421
incipient dunes, 242, 245f
incised streams, 254f
incompatible behaviors, 108–109, 110t
India
botanic gardens, 184t, 186
desert environments, 223t
rainforests, 441
savanna ecosystems, 394, 396f, 397
urban environments, 462
Indian Ocean, 378
indices
diversity indices
Shannon index, 60
Finn’s cycling index, 52
ratio-based indicators
ecological indicators, 60
indirect interactions, 81–91
background information, 81
basic concepts, 81
classification, 84
comparative theoretical ecosystem analysis
(CTEA), 90
definitions, 84
energy analysis, 64
food web cycles, 50, 51f
future research areas, 90
grassland wildfires, 410
life cycle assessment (LCA), 87
measurement techniques, 88
mutualism, 57–58, 60, 81, 83, 83f
network environ analysis, 79, 80t
occurrences
abiotic environments, 86
aquatic environments, 85
evolutionary role, 88
global relevance, 87
industrial ecology, 87
terrestrial ecosystems, 85
research areas
apparent competition, 82, 83f
exploitative competition, 83, 83f
indirect mutualism/commensalism, 83, 83f
interaction modification, 83, 83f, 84f
interference competition, 83, 83f
interspecific competition, 82, 83f
trophic cascades, 82, 83f, 85
research issues
complex systems, 89
environmental management, 90
modeling studies, 85, 90
spatial separation, 89
system boundaries, 90
time lags, 89
uncertainty, 89
Indonesia
botanic gardens, 184t, 186
forest plantations, 265f
mangrove forests, 313–314
peatlands, 331
tropical ecosystems, 441
Indus River, 260, 276, 313, 433
Ecological Footprint analysis, 87–88
indirect interactions, 87
industrial forest plantations, 264
infectious diseases
exergy, 131, 131f, 135
probability calculations, 58
Inner Niger Delta, Mali, 259
input-output models
ecological networks, 52, 53f
environ analysis, 76
two-compartment steady-state energy analysis,
65, 65f, 66f, 68t
insecticides
salt marsh ecosystems, 392
biodiversity, 148
boreal forest ecosystems, 182
cave habitats, 191, 193
competition studies, 82–83, 88
desert environments, 229, 231
desert stream ecosystems, 216, 218
exergy calculations, 134t
forest plantations, 267
grasslands, 410
indirect interactions, 88
intertidal environments, 378
mangrove forests, 310
pest control
crop production, 147
postclosure landfills, 307
riparian systems, 348
saline/soda lakes, 381, 382
salinity tolerances, 435f
salt marsh ecosystems, 386
savanna ecosystems, 400
stream ecosystems, 356–357, 359f, 371, 372f
temperate forest biomes, 423
temporary water bodies
bet hedging strategies, 436
boreal/temperate forests, 432
characteristics, 433
interspecific interactions, 437
life-history strategies, 436–437
tropical rainforests, 431
tundra ecosystems, 432
tundra ecosystems, 446, 448
urban environments, 462
integrity
system concepts, 124, 125f
interaction modification, 83, 83f, 84f
interference competition
research areas, 83, 83f
intermittent streams, 364, 365f, 428t
international agreements and conventions
urbanization process, 466
International Biological Program (IBP), 10,
118–119, 240
interspecific competition
indirect interactions, 82, 83f
intertidal environments, 378
mangrove forests, 311
intertidal environments, 374–380
anthropogenic impacts, 377
aquatic organisms
algae, 375, 375f
mobile organisms
general discussion, 376
invertebrates, 376
vertebrates, 376f, 377
sessile invertebrates, 376
biodiversity, 374, 375f
ecological research
general discussion, 378
historical research, 378
late-twentieth century/early twenty-first
century, 379
mid-twentieth century, 378, 378f
estuarine ecosystems, 249
future research areas, 379
mangrove forests, 308
physical characteristics
rock surfaces, 375
Index
tidepools, 375
tides, 374
wave surge effects, 375
standing crop biomass-production rate
relationship, 18, 19f
causal effects, 377
vertical zonation patterns, 377
intraspecific competition
intertidal environments, 378
intuition, 108
invasion fitness
botanic gardens, 189
desert environments, 238
endangered species, 240
forest plantations, 267
grasslands, 412
natural communities
chaparral ecosystems, 200
desert environments, 238
marshes, 280
salt marsh ecosystems, 391
tundra ecosystems, 446
urban environments, 462
polar ecosystems, 342
salt marsh ecosystems, 388, 391
savanna ecosystems, 404
urban environments, 462
invertebrates
body size-abundance distributions, 45, 47f
cave habitats, 191
desert stream ecosystems, 215, 216, 218t
eastern African soda lakes, 382
estuarine ecosystems, 250, 251
intertidal environments, 376, 376
landfills, 305
mangrove forests, 310
riparian systems, 344
river system dynamics, 351–362
energy flow
energy sources, 351, 352f, 353f
functional feeding groups (FFGs), 352,
354f, 354t
matter fluxes
nutrient cycling, 355, 355f
organic matter, 351, 352f, 355
modeling studies
biotic diversity patterns, 359f
ecotones, 359
flood pulse concept, 344, 359t, 360
hierarchical scales, 359, 359t
hydraulic stream ecology, 359t, 361
hyporheic dynamics, 359t, 361
network dynamics, 359t, 361
patch dynamics, 359t, 361
riparian zone influences, 359t, 360, 360f
river continuum concept (RCC), 344, 356,
357f, 358f, 359t
serial discontinuity, 357
organic matter
characteristics, 356
energy sources, 351, 352f, 354f
retention potential, 356
sources, 356
transport/storage mechanisms, 355
salinity
saline/soda lakes, 381, 382
salinity tolerances, 382, 387
salt marsh ecosystems, 386, 387, 388
savanna ecosystems, 400
sessile invertebrates, 376
stream ecosystems, 371, 372f
temporary water bodies
boreal/temperate forests, 432
characteristics, 433
intermittent streams, 432f
phytotelmata, 428t, 429t
tropical rainforests, 431
tundra ecosystems, 432
urban environments, 463
Iowa (USA), 406–407
Ipomoea pes-caprae, 242
Ipomoea spp., 242
Iran, 223t, 426
Iraq, 223t
Ireland
peatlands, 276–277
temporary water bodies, 429t
iron (Fe)
elemental concentrations, 10t
open-ocean upwelling systems, 454
stream ecosystems, 369
irrigation
Asian floodplain ecosystems, 260
desert environments, 237
desert stream ecosystems, 221
historical background, 237
cave-like habitats, 192
Isoberlinia doka, 396–397
Isoberlinia spp., 396–397
isoclines, See zero net growth isocline (ZNGI)
Isoëtes spp., 327, 432
Isognomonidae, 288t
isopods (Isopoda)
cave habitats, 193
desert environments, 235
Florida Everglades mesocosm project, 294f
intertidal environments, 376
landfills, 305
salt marsh ecosystems, 391–392
stream ecosystems, 371, 372f
isotopic analyses, 20
Israel
desert environments, 223t
Issyk-kul (Kyrgystan), 381
Italy
botanic gardens, 184t, 185
lagoon ecosystems, 296–297
Ixeris repens, 243f
Ixodes pacificus, 196
J
jacamars, 348
jacanas, 259–260
jackals, 235
jacks, 204
Jackson, Wes, 14
jaegers, 447
Jamaica, 205, 205f
Japan
alpine forests, 159t, 160f
botanic gardens, 184t
urban environments, 464–465
Jardim Botanico, University of Coimbra,
Portugal, 185
Jardı́n Botánico de Madrid, Spain, 185
Jardin botanique, Montreal, Canada, 184t,
189, 189f
Jardin des Plantes, France, 184t, 185, 185f
Jasus edwardsii, 21, 22f
Java, 441
jellyfish
exergy calculations, 134t
upwelling ecosystems, 458
jerboa, 230
jirds, 230
Jordan, 223t
Jordan River, 384
Joshua trees, 229
Juday, Chauncey, 7
Judean Desert, Palestine, 227f
Julbernardia spp., 396–397
Juncus effusus, 244
Juncus kraussii, 386t
Juncus procerus, 320t
Juncus roemerianus, 386t
499
Juniperus spp.
alpine ecosystems, 151, 159t
Mediterranean ecosystems, 320t,
321t, 325
Juniridae, 288t
Jurong Bird Park, Singapore, 279f
K
Kalahari Desert, 223t, 237
Kalahari Gemsbok National Park,
Botswana, 395f
Kalmia spp., 420–421
Kalyptorychidae, 288t
kangaroo rats
desert environments, 230, 231
kangaroos, 230
Kansas (USA), 406–407
kaolinite (Al2Si2O5(OH)4), 395–396
Karakum Desert, 223t
Kauffman, S, 99
Kavir Desert, 223t
Kay, James, 99
Kazakhstan, 223t
kelp forests
intertidal environments, 375–376
photographic views, 17f
predator-prey dynamics, 21, 22f
stable isotope analysis, 21
standing crop biomass-production rate
relationship, 18, 19f
Kentrophoridae, 288t
Kenya
grasslands, 411
Keronidae, 288t
kettle lakes, 429t, 444
Kew Gardens, See Royal Botanical Gardens at
Kew (United Kingdom)
Kidman Springs Station, Victoria Rivers District,
Northwest Territories, Australia, 399f
Kigelia spp., 259
killifish, 295f, 433, 436
kingfishers, 259–260, 348
Kinorhyncha, 134t
Kissimmee River, Florida (USA), 259
kissing bugs, 196
knees (floodplains), 257
Koeleria spp., 320t
Koenigia islandica, 446
Konza Prairie Biological Field Station, Kansas
(USA), 409f, 410
Köppen-Trewartha climate class, 419t
Köppen vegetation classification
system, 439–440
krill, 342
Kristianstad, Sweden, 466
krummholz mats, 156, 157f, 161f, 162
emergence/emergent properties, 95t
fundamental ecosystem theory, 36t
postclosure landfills, 306
temporary water bodies, 436
urban environments, 463t
Kullback information measure
emergence/emergent properties, 97, 97f
exergy
information theory, 131
fundamental ecosystem theory, 34
!Kung culture, 238
Kuwait
desert environments, 223t
kwongan ecosystems, 326
Kyzyl Kum, 223t
L
Labridae, 206, 377
Lacrymariidae, 288t
Lagodon rhomboides, 295f
500
Index
lagoon ecosystems, 296–303
barrier islands, 247
catchment loads, 297, 297f, 300f
characteristics, 296
land-use change, 297, 297f
mangrove forests, 308, 309f
modeling studies
ecological impacts
empirical models, 298
general discussion, 298
NPZD (nitrogen-phytoplanktonzooplankton-detritus) model, 299
simulation models, 299
multifractal biomass/species
distributions, 301
nonequilibrium dynamics, 300
physical processes, 297
tundra ecosystems, 444–445
Lagopus mutus, 446–447
Lahontan cutthroat trout, 384
Lake Aral, 273
Lake Baikal, 271f, 271t, 381
Lake Biwa, Japan, 271t, 272, 272f
Lake Bled, 272f
Lake Bogoria (Kenya), 381f
Lake Chad, Africa, 281
Lake Constance, 271t, 272
lake ecosystems
algae
food webs, 383
large-scale whole ecosystem experiments, 21
saline/soda lakes
characteristics, 381
Dead Sea, 384
eastern African soda lakes, 382
Mono Lake (USA), 383
alpine habitats, 17f
disturbances, 278
freshwater ecosystems, 270–274
characteristics, 270
Crater Lake, Oregon (USA), 271f
importance, 271
Lake Baikal, 271f, 271t
major lakes, 271, 271t
water quality problems
acidification, 273
eutrophication, 272, 272f
exotic species introductions, 273
overfishing, 274
pathogenic contamination, 274
salinization, 273
siltation, 273
toxic substances, 273
water-level changes, 273
indirect interactions, 86
peat formation, 338
radiation exposure effects
eutrophication control methods
large-scale whole ecosystem
experiments, 21
salinity
saline/soda lakes, 380–384
biodiversity, 381–382
characteristics, 381, 382f
Dead Sea, 381, 384
eastern African soda lakes, 381f, 382, 382f
ecological processes, 382
economic factors, 384
geographic distribution, 381
ionic composition, 381
Mono Lake (USA), 381, 383
temporary water bodies, 429t, 432,
433f
stable isotope analysis, 21
swamps, 414, 414f
temperate forest biomes, 423
tundra ecosystems, 444
water cycle, 365f
Lake Elmenteita (Kenya), 382, 382f
Lake Erie, 271t
Lake Eyre (Australia), 381
Lake Huron, 271t
Lake Maggiore, 271t
Lake Mahega (Uganda), 381, 382f
Lake Malawi, 271t
Lake Michigan
freshwater ecosystems, 271t, 274
marshes, 276f
Lake Nakuru, 384
Lake Okeechobee, Florida (USA), 277
Lake Shala (Ethiopia), 381
Lake Sonachi (Kenya), 382–383
Lake Superior, 271t
Lake Tanganyika, 271t
Lake Titicaca, 271t
Lake Van (Turkey), 381
Lake Victoria, 271t, 273
La Mancha, Mexico, 244
Laminariales, 21
lampreys, 373
lancewood, 396
landfills, 303–307
biota, 305
characteristics, 303
faunal communities, 307
postclosure uses, 304
restoration methods, 466
soil cover, 304
successional development, 307
vegetation, 305
Land Institute, 14
Land-Ocean Interactions in the Coastal Zone
(LOICZ) project, 299
indirect interactions, 85
forest plantations, 268
fractal landscapes
hierarchy theory, 119
hierarchy theory, 116f, 116–117
land cover
temperate forests
historical land cover, 425
present-day land cover, 426
coral reefs, 207, 208f
land-use change
alpine ecosystems, 155
lagoon ecosystems, 297, 297f
savanna ecosystems, 403
self-organization, 105, 105f
shelterbelts/windbreaks, 470
spatial pattern formation, 104f
treeless landscapes, 334
tundra ecosystems
diversity, 444
freeze-thaw cycles, 444
urbanization process
adaptive management, 466
basic concepts, 461
ecological effects, 462, 462f, 463t, 464f
ecosystem services, 465, 465t
gradients, 463–464
habitat quality, 463
human-environment interactions, 462,
463t, 467
international agreements and conventions,
466
restoration methods, 466
La Niña, 459f
See also El Niño Southern Oscillation (ENSO)
Laos, 397
La Planada Reserve, Colombia, 442f
larch trees, 158, 181
Larix deciduas, 158
Larix laricina, 335
Larix spp., 151, 159t, 181
larks, 230, 231
Larrea divaricata, 228
Larrea tridentata
characteristics, 228
drought tolerance, 233–234
parasites, 236
piedmont bajada environment, 224f
Larus californicus, 383
La Selva Biological Station, Costa Rica, 442f
Last Glacial Maximum (LGM), 324
Lavandula spp., 325–326
lava tubes, 190
ascendancy, 62
Lawrence, Kansas (USA), 407
leachate seepage, See landfills
eco-exergy losses, 136, 136t
least killifish, 295f
leaves
defense strategies, 397
exergy calculations, 137
leaf litter
floodplain ecosystems, 255–256
stream ecosystems, 368, 368f
urban environments, 463
Le Chatelier’s Principle, 137
Lechenaultia, 320t
lechwe, 259–260
leeches
exergy calculations, 134t
stream ecosystems, 371
temporary water bodies, 433
Leguminosae, 238
Leicester, England, 464–465
Leigh Marine Reserve, New Zealand, 21
lemmings, 446–447
Lemmus sibiricus, 447
Lemnaceae, 371
Lentibulariacea, 347
lenticel hypertrophy, 258
Leontief input-output model, 52, 53f
Leontief matrix, 52, 53f
leopards, 239
Lepidoptera
exergy calculations, 134t
melanism, 462
stream ecosystems, 373
Lepidosireniformes, 433
Lepomis spp., 295f
Leptochelia savignyi, 294f
Lepus arcticus, 446–447
Lespedeza bicolor, 475
lesser flamingoes, 382, 384
lethal yellowing disease, 246
Leucaena leucocephala, 306
Leucettidae, 288t
Leucothoe revoluta, 244
Leucothoidae, 288t
Lewis, GH, 92
Leymus arenarius, 243f
lianes, 442
Libya, 223t
lichen
alpine ecosystems, 153
bogs, 334, 334f
desert environments, 228, 230
polar ecosystems, 340, 342
stream ecosystems, 371
temperate forest biomes, 420
tundra ecosystems, 445f
ascendancy, 62
life
fundamental ecosystem theory, 34, 36t
mathematical approaches, 102
indirect interactions, 87
desert stream ecosystems, 216, 218t
Mediterranean ecosystems, 325
seaweed, 376
temporary water bodies, 436
light
alpine forests, 159, 165
Index
greenhouse ecosystems, 283
river system dynamics, 351, 356–357, 359f
upwelling ecosystems, 450–460
characteristics, 451
chlorophyll biomass, 452, 452f
climate
climate change effects, 458, 458f
El Niño Southern Oscillation (ENSO),
459, 459f
long-term trends, 459
eddy systems, 454–455
fish/fisheries
climate change effects, 458, 458f
fish production, 457, 457f
food webs, 455, 455f
general discussion, 455
small pelagic fish, 455, 456f
trophic levels, 457
general discussion, 450
intermittence, 451
occurrences, 450, 450f
open-ocean upwelling systems, 454
oxygen depletion, 453
phytoplankton community structure, 452
primary production, 451
zooplankton, 451, 453
See also solar radiation/solar energy
light-footed clapper rails, 387
lightning
wildfires, 182, 399
limber pine, 158
limestone (CaCO3)
cave formation processes, 190
Limidae, 288t
limiting factors
coral reefs, 202
Limnephilidae, 435f
Limnephilus assimilis, 435f
limnogenous peatlands, 332
Limoniastrum spp., 386t
limpets, 371
Lindeman efficiency, 7
Lindeman, Raymond L, 7, 16
linear dunes, 224–225
Lineidae, 288t
Linyphiidae, 192
lions
social groups, 103
Lippia nodiflora, 244
Liquidambar spp., 262
lithoautotrophs, 167t
Lithobius spp., 193
Lithops spp., 235
lithosphere
elemental concentrations, 10t
Lithraea caustica, 320t
litterfall
boreal forest ecosystems, 182
floodplain ecosystems, 255–256
mangrove forests, 311, 312f
postclosure landfills, 305
riparian systems, 349
river system dynamics, 351
temperate forest biomes, 422–423
Littorophiloscia spp., 193
liverworts, 371, 433
livestock operations
alpine ecosystems, 155
forest ecosystems, 31f
grassland models, 411
living stones, 235
lizards
desert environments, 230
temperate forest biomes, 423
llanos landscape, 396, 407
Lobelia deckenii, 449
lobelias, 449
Loblolly pines, 269
lobsters
predator-prey dynamics, 21, 22f
locust trees, 424f
log holes, 429t
Loiseleuria procumbens, 154f
Lolium spp., 320t
longnose killifish, 295f
Long-Term Ecological Research
(LTER) Sites, 10
Long Term Ecological Research (LTRT)
program, 240
Lonicera spp., 195
loons, 277, 447
Lop Nor (China), 381
lotic research, 363
Louisiana (USA), 276f
Louriniidae, 288t
Lovelock, James
climate change effects, 13
Gaia hypothesis
global energy balance, 105
population dynamics, 103
Loxahatchee National Wildlife Refuge, Florida
(USA), 279f
Loxosceles spp., 193
Lucania goodei, 295f
Lucinidae, 288t
Lumbrineridae, 288t
lungfishes
desert stream ecosystems, 218
temporary water bodies, 433, 435
Lupinus arboreus, 246
Lupinus spp., 243
Lutjanidae, 207–208
Lycosa howarthi, 193
Lygeum spartum, 320t
Lyme disease, 196
Lysianassidae, 288t
Lythrum salicaria
aquatic wetlands, 275f
M
Mackenzie River Basin, Canada, 281, 331, 331f
mackerels
upwelling ecosystems, 457
macroalgae
competition studies, 83
coral reefs, 211f
food webs, 21, 22f
intertidal environments, 375, 377
life cycles, 376
predator-prey dynamics, 21, 22f
salt marsh ecosystems, 387, 389
stable isotope analysis, 21
standing crop biomass-production rate
relationship, 18, 19f
trace elements
See also seaweed; fungi
macrophytes
autocatalysis, 41, 42f
desert stream ecosystems, 220, 220f
estuarine ecosystems, 249t
exergy calculations, 139
stream ecosystems, 351, 371
Macropodidae, 230
Madagascar
savanna ecosystems, 396f
tropical ecosystems, 440, 442
Magellanic moorland, 281
magnesium (Mg)
elemental concentrations, 10t
saline/soda lakes, 381
stream ecosystems, 369
Magnoliophyta, 288t
Maharashtra (Mahim) Nature Park, Mumbai,
India, 189
maidencane, 276f
maize,
See also corn
malachite kingfishers, 259–260
Malaysia
botanic gardens, 184t
mangrove forests, 311
tropical ecosystems, 441, 442
mallee ecosystems, 326
mammals (Mammalia)
boreal forest ecosystems, 182
desert environments
characteristics, 230
decomposition processes, 235
drought tolerance, 230
predation, 236
water uptake, 231
estuarine ecosystems, 249
exergy calculations, 134t
forest plantations, 268
grasslands, 410
intertidal environments, 377, 380
Mediterranean ecosystems, 322, 322t
riparian systems, 348
salt marsh ecosystems, 387
savanna ecosystems, 400
temperate forest biomes, 423
tundra ecosystems, 446
upwelling ecosystems, 455f
urban environments, 462
manatees
overfishing, 211, 211f
mangrove forests, 308–318
biodiversity, 310
characteristics, 308
coral reefs, 207, 208f
ecogeomorphology, 308, 309f, 310f
environmental impacts, 313
estuarine ecosystems, 250
food webs, 310, 312
gradients, 308, 309f, 310f
greenhouse ecosystems, 284f
hydroperiod (flood duration), 308,
309f, 310f
management strategies, 314, 314f
nutrient resources, 311, 312f
photographic views, 17f
productivity, 311
restoration methods, 314
salinity, 311
succession, 311
manroot, 196
manzanita, 195
Maoke Mountains, 159t, 160f
maple trees
temperate forest biomes, 417
maquis ecosystems, 326
Marah macrocarpus, 196
Marcetia taxifolia, 244
Mar Chiquita (Argentina), 381
Marginellidae, 288t
marine ecosystems
body size-abundance distributions, 47
indirect interactions, 85
iron cycle
pelagic ecosystem models
NPZD (nitrogen-phytoplanktonzooplankton-detritus) model, 299
See also aquatic environments; marine
ecosystems
marine reserves
coral reefs, 212, 212f
ecosystem dynamics, 21,
22f, 23f
Marmota bobac, 83–84, 84f
marram grass, 246
marsh ecosystems, 274–281
anthropogenic impacts
dam construction, 279
draining effects, 279
501
502
Index
marsh ecosystems (continued )
food webs, 280
general discussion, 279
nutrient resources, 279–280
road networks, 280
biodiversity, 278, 279f
characteristics, 175, 275, 275f, 280
environmental factors
disturbances, 278
nutrient cycling, 278
estuarine ecosystems, 249
general discussion, 274
geographic distribution, 278
restoration methods, 280
seasonal pools, 428t
vegetation, 275f, 277, 277f
water availability
floodplains, 276, 276f
general discussion, 275
peatlands, 276–277
seepage wetlands, 277
temporary wetlands, 277
water level changes, 277, 277f
West Siberian Lowland, 276f
See also salt marsh ecosystems; tidal wetlands
marsh killifish, 295f
Marsilea spp., 327
marsupials
desert environments, 230
savanna ecosystems, 400
Martinique, 186
mass extinctions, 449
Mastogloiaceae, 288t
materials science research, 14
See also modeling studies
matoral ecosystems, 326
adjacency matrices, 50, 52, 53f
ecological networks, 52, 53f
environ analysis
community assembly rules, 77
methodology, 77, 77f, 78t
network properties
direct interactions, 79, 80t
general discussion, 79
homogenization effects, 79
indirect interactions, 79, 80t
mutualism, 79, 80t
purpose, 80
quantitative environ, 80
sample network, 77, 77f
food webs, 50
Leontief matrix, 52, 53f
total system throughput (TST), 58
Mauritania, 223t
Mauritius, 184t, 186, 313
See also fish/fisheries
mayflies
stream ecosystems
general discussion372–373
predator-prey dynamics, 354, 354t
Mayorellidae, 288t
McKinley, Daniel, 13–14
meadows, 275, 281
mean annual surface temperature (MAST), 190
Mediterranean Basin
biodiversity patterns, 322, 322t
biogeographical evolution, 323
characteristics, 319, 320t
climatic stress, 327
convergence studies, 325
Pleistocene glaciation, 324
soils, 321
tectonic activity, 321
wildfires, 328
Mediterranean ecosystems, 319–330
biodiversity patterns, 322, 322t
biogeographical evolution, 323
body size-abundance distributions, 47f
chaparral, 195
characteristics
bioclimates, 321t
climate, 319, 320t, 321t
soils, 320t, 321, 327
vegetation, 320t, 321t
convergence studies, 325
ecosystem characteristics
biodiversity connections, 328
climatic stress, 327
disturbances
general discussion, 328
herbivores, 329
wildfires, 320t, 328
vegetation types, 326
global change effects, 329
occurrences, 319
Pleistocene glaciation, 324
temporary water bodies, 427, 429t
wildfires, 320t, 328
Mediterranean Sea
intertidal area research, 378
Meesia spp., 335
Megadrili, See earthworms
megafauna, 449
Megaloptera, 372f, 372–373
Megaselia spp., 193
Meghna River, 260
meiofauna, 376
Meioneta spp., 193
Mekong River, 433
Melaleuca nervosa, 396
Melaleuca spp., 258, 261,
320t, 347, 396
Melaleuca viridiflora, 396
melanism, 462
Melanoides tuberculata, 294f
Melanophila spp., 198–199
Melbourne, Australia, 463
Melinis spp., 404
Melinna maculata, 294f
Melosira spp., 86, 370f
meltwater pools/streams, 428t
mercury (Hg)
eco-exergy losses, 136t
trace element concentrations
salt marsh ecosystems, 392
Meridion spp., 370f
Merlucius capensis, 453
Merlucius paradoxus, 453
meromixis, 383
meroplankton, 249
Mesembryanthemum aequilaterale, 243f
mesic environments
chaparral ecosystems, 195
desert stream ecosystems, 218t
grassland models, 409–410, 411, 411f
savanna ecosystems, 401–402
water uptake, 231
See also desert environments
mesocosms
Florida Everglades mesocosm project
air/water movement, 284f
algal biomass, 293f
ecosystem subunits, 285f
fish distribution, 295f
general discussion, 290
invertebrate abundance, 294f
physico-chemical parameters, 292t
plan diagram, 291f
plant communities, 291f
structural characteristics, 283f
vertical/longitudinal section, 292f
water supply, 284f
greenhouse ecosystems, 281–295, 283f
Mesopotamia, 238
Mesozoa
exergy calculations, 134t
mesozooplankton, 453–454
mesquite trees, 232
Messerschmitia sibirica, 243f
Messinian salinity crisis, 324
body size relationships
decoding mechanisms, 49
ecological efficiency, 48
metabolic rate (MR)
intertidal environments, 377
Metacystidae, 288t
metals/metallurgy
salt marsh ecosystems, 392
Metazoa
intertidal environments, 376
methane (CH4)
mangrove forests, 312f
methanogenesis, 337
peatlands, 337
Metrosideros polymorpha, 193
Mexico
alpine forests, 157, 159t, 160f
botanic gardens, 183, 184t
desert environments, 223t, 238
dune ecosystems, 244, 244
mangrove forests, 311, 313
tropical ecosystems, 440
Michaelis-Menton equation
wastewater treatment, 173
microalgae
estuarine ecosystems, 251
intertidal environments, 375
microarray technology, 169
bioenergetic processes
nitrogen cycle
anammox process, 171f, 172, 178
sulfur cycle, 453
sulfur cycle, 453
denitrification
exergy calculations, 132, 134t
comparative analyses, 169
electron-donor/electron-acceptor regimes,
167t
alpine forests, 162
shelterbelts/windbreaks, 469
Microcoleus calcicola, 293f
greenhouse ecosystems, 281–295
basic concepts, 281
Biosphere II project, Arizona (USA),
281, 292
biotic parameters
ecosystem interchange, 285
ecosystem structure, 284
ecosystem subunits, 284, 285f
coral reef microcosm
carbonate cycle, 287f
general discussion, 286
gross primary production (GPP), 287f
organism families, 288t
oxygen (O) concentrations, 286f
physico-chemical parameters, 287t
schematic diagram, 286f
Florida Everglades mesocosm
algal biomass, 293f
fish distribution, 295f
general discussion, 290
invertebrate abundance, 294f
physico-chemical parameters, 292t
plan diagram, 291f
plant communities, 291f
vertical/longitudinal section, 292f
Florida Everglades mesocosm project
air/water movement, 284f
ecosystem subunits, 285f
structural characteristics, 283f
water supply, 284f
operational requirements, 285
physico-chemical control parameters
air/water movement, 283, 284f
Index
light requirements, 283
physico-chemical factors, 283
structural characteristics, 282, 283f
water supply, 283, 284f
Microcystis spp.
lake ecosystems, 86
stream ecosystems, 371
microorganisms
community characteristics, 168
community structure, 168f, 177
desert environments, 227, 232
estuarine ecosystems, 249, 249f
halophilic microorganisms, 381–382, 384
indirect interactions, 85
landfills, 305
comparative analyses, 169
electron-donor/electron-acceptor
regimes, 167t
saline/soda lakes, 381
salt marsh ecosystems, 388
stream ecosystems, 356, 358f
sulfate-reducing bacteria
peatlands, 338
temporary water bodies, 431
microphytobenthos, 249t
Microthrix parvicella, 177
microzooplankton, 451, 452–453, 455f
Micruridae, 288t
midges
cave habitats, 193
stream ecosystems, 373
temporary water bodies, 437
migration
agroecosystems, 146f, 147
climate change effects, 182
riparian systems, 349
upwelling ecosystems, 458
Milankovitch climate oscillations, 324
Miliolidae, 288t
milkweeds, 230
Millennium Ecosystem Assessment (MA)
conservation strategies, 25
ecosystem services, 10, 27
trends, 11t
urban environments, 466
Millennium Seed Bank, 188
millipedes
agroecosystems, 148
cave habitats, 193
landfills, 305
mimicry
biomimicry, 14
mimosa
floodplain ecosystems, 261
minerals
mineral resource depletion
eco-exergy losses, 136
rock size categories, 368, 368f, 368t
See also soils
minerotrophic ecosystems, 334
Minnesota (USA), 406–407
mint, 185
Miocene
desert flora, 228
Mediterranean ecosystems, 323
miombo landscape, 396–397
mires, See peat/peatlands
Mississippi River
wetland ecosystems, 281
Missouri Botanical Garden, 184t, 189
Missouri (USA), 406–407
mistletoes, 442
mites
agroecosystems, 148
desert environments, 229
landfills, 305
Mediterranean ecosystems, 327
Mitridae, 288t
modeling studies
simulation models, 179
biological/life systems, 102
complex systems
Sugarscape, 112, 112f, 113f
Tierra, 102, 111
TRANSIMS, 112
savanna ecosystems, 400, 401t
dynamic system indicators, 71, 71f, 72f
ecosystems
lagoon ecosystems, 298
exergy calculations, 137
geographic information systems (GIS)
primary production
aboveground net primary production
(ANPP), 409f
hierarchy theory, 115, 115f
indirect interactions, 85, 90
input-output models
ecological networks, 52, 53f
environ analysis, 76
two-compartment steady-state energy
analysis, 65, 65f, 66f, 68t
ecological impacts
empirical models, 298
general discussion, 298
logistic growth models
NPZD (nitrogen-phytoplanktonzooplankton-detritus) model, 299
simulation models, 299
multifractal biomass/species
distributions, 301
nonequilibrium dynamics, 300
physical processes, 297
pelagic ecosystem models
NPZD (nitrogen-phytoplanktonzooplankton-detritus) model, 299
matrix models
adjacency matrices, 50, 52, 53f
ecological networks, 52, 53f
environ analysis
community assembly rules, 77
direct interactions, 79, 80t
homogenization effects, 79
indirect interactions, 79, 80t
methodology, 77, 77f, 78t
mutualism, 79, 80t
network properties, 79
purpose, 80
quantitative environ, 80
sample network, 77, 77f
food webs, 50
Leontief matrix, 52, 53f
total system throughput (TST), 58
orientation theory
animat algorithms, 126
lifestyle development, 126, 127f
river system dynamics
biotic diversity patterns, 359f
ecotones, 359
flood pulse concept, 344, 359t, 360
hierarchical scales, 359, 359t
hydraulic stream ecology, 359t, 361
hyporheic dynamics, 359t, 361
network dynamics, 359t, 361
patch dynamics, 359t, 361
riparian zone influences, 359t, 360, 360f
river continuum concept (RCC), 344, 356,
357f, 358f, 359t
serial discontinuity, 357
savanna ecosystems, 400, 401t
sewage treatment systems, 179
two-dimensional (2D) system dynamics models
uncertainty analysis
average mutual information (AMI),
59, 59f
wastewater treatment, 179
modified natural forests, 264f, 265f
503
Mojave Desert (USA), 223t, 224f
molds
exergy calculations, 134t
mole crickets, 348
mollusks (Mollusca)
desert environments, 232
exergy calculations, 134t
Florida Everglades mesocosm project, 294f
intertidal environments, 376, 377
stream ecosystems, 356–357, 359f,
371, 372f
temporary water bodies, 433, 435
Moloch horridus, 230
Mongolia Plateau, China, 223t
monkeys
exergy calculations, 134t
monk seals, 211, 211f
monocotyledons
wastewater treatment, 174
Mono Lake (USA), 381, 383, 384
Mono winds, 197, 199
monsoonal weather patterns
upwelling ecosystems, 450
montane forest vegetation zone, 151, 151f
Montastrea spp., 202–203, 208f
Monte Desert, 223t
Montezuma, 183
Montia lamprosperma, 446
moors, 429t
moose
boreal forest ecosystems, 182
marshes, 280
riparian systems, 348
temperate forest biomes, 423
moraine lakes, 444
moray eels, 206
Morocco, 223t
mortality studies
marine ecosystems, 458
temperate forest biomes, 420
tropical rainforests, 442
mosquitoes
exergy calculations, 134t
temporary water bodies, 429t, 436, 437
vector control, 438
mosquito fish
Florida Everglades mesocosm project, 295f
moss animals, 376
mosses
desert environments, 228, 230
exergy calculations, 134t
fens, 277, 334, 335f
peat moss, 330, 333f, 336
polar ecosystems, 342
stream ecosystems, 371
temperate forest biomes, 420
temporary water bodies, 429t, 433
tundra ecosystems, 445f
water retention, 336
moths
cave habitats, 193
chaparral ecosystems, 196
pollen vectors, 24
stream ecosystems, 373
temperate forest biomes, 423
mountain avens, 446f
mountain lions
temperate forest biomes, 423
Mount Kenya, 445, 445f, 448
Mount Kilimanjaro, 155, 449
Mount Wilhelm, New Guinea, 445
mouse/mice
chaparral ecosystems, 196
temperate forest biomes, 423
See also rats; rodents (Rodentia)
mudflats
estuarine ecosystems, 250
invasive species, 391
504
Index
mudflats (continued )
salt marsh ecosystems, 386
mudsnails, 391–392
Muldanidae, 288t
Murreyella periclados, 293f
muskegs, 429t
musk oxen
tundra ecosystems, 446–447, 447
muskrats
marshes, 280
riparian systems, 348
mussels
competition studies, 83
estuarine ecosystems, 250
intertidal environments
characteristics, 376
photographic views, 375f
vertical zonation patterns, 377
stream ecosystems, 371
Mussidae, 288t
Mustella erminea, 447
autocatalysis, 41, 57–58
coral-algal mutualism, 201, 205
indirect interactions, 57–58, 60,
81, 83, 83f
network environ analysis, 79, 80t
Myanmar, 397
Mycalidae, 288t
mycorrhizas
bogs, 333
desert environments, 227
ectomycorrhizal fungi, 397
symbiotic relationships
dune ecosystems, 246
root systems, 397
vesicular arbuscular mycorrhizae (VAM), 246
Myocastor coypus, 278, 392
See also nutria
Myriophyllum spicatum, 176
Myrtus communis, 328
Mysidae, 288t
Mytillidae, 288t
Mytilus spp., 250
myxomatosis, 246
N
Namib Desert, 223t
Namibia
desert environments, 223t
fish/fisheries, 459
temporary water bodies, 429t
Nannolene spp., 193
nanoplanktonic phytoflagellates, 452–453, 454
Nan Tso (China), 381
Nassau grouper, 213
natrium, See sodium (Na)
Natufian culture, 238
natural containers, 428t, 429t
natural forests, 264f, 265f, 267
Nature Conservancy, 412
Naviculaceae, 288t
Navicula spp., 370f
Neanura spp., 193
nebkhas, 224–225
negative feedback
predator-prey dynamics, 101
self-organization, 101, 102f
Negev Desert, Israel, 225f, 226, 232f, 238
nekton, 249, 252
nematodes (Nematoda)
desert environments, 229
exergy calculations, 134t
temporary water bodies, 433, 435
upwelling ecosystems, 453
Nematomorpha, 433
Nemertinea (nemertine worm)
cave habitats, 193
exergy calculations, 134t
Nemertodermatidae, 288t
Neogene
Mediterranean ecosystems, 324
Neolithic era, 238
neon (Ne), 10t
Neotoma fuscipes, 196
Neotropics, 442
Nephropidae, 288t
Nephtys bucera, 294f
Neritidae, 288t
Nesomedon spp., 193
nested hierarchies, 117f, 118
Nesticella spp., 193
Netherlands
botanic gardens, 184t, 186
net primary production (NPP)
aboveground net primary production
(ANPP), 409f
desert environments, 234, 234t, 235f
energy flow, 9t
floodplain ecosystems
biomass productivity, 258, 258f
general discussion, 253
nutrient cycling, 256–257
structural characteristics, 257t
savanna ecosystems, 394, 403, 403t
network, 4, 5, 5t
network methods
environ analysis, 76–81
background information, 76
basic concepts
boundary zones, 76
input-output models, 76
partitioning processes, 76–77
community assembly rules, 77
data requirements, 77
methodology, 77, 77f, 78t
network properties
direct interactions, 79, 80t
general discussion, 79
homogenization effects, 79
indirect interactions, 79, 80t
mutualism, 79, 80t
quantitative environ, 80
purpose, 80
sample network, 77, 77f
neural systems
emergence/emergent properties, 93
Neuroptera, 373
New Caledonia
tropical ecosystems, 441
New Delhi, India, 462
New Guinea
alpine forests, 157
savanna ecosystems, 394
tropical ecosystems, 441, 442
New Hebrides, 441
New Mexico (USA), 406–407
New Orleans, Louisiana (USA), 466
newts
temperate forest biomes, 423
New York Botanical Gardens (USA),
183, 187, 187f
New York, New York (USA), 184t, 462
New Zealand
alpine forests, 157, 159t, 160f
dune ecosystems, 242
fish/fisheries, 21, 22f, 23f
temporary water bodies, 429t
tundra ecosystems, 445
Nezahat Gokyigit Memorial Park (Turkey), 183
emergence/emergent properties, 95t
fundamental ecosystem theory, 36t
savanna ecosystems, 401t, 401–402
nickel (Ni)
eco-exergy losses, 136t
Nigeria, 440
Nile River
floodplain ecosystems, 276
irrigation systems, 238
wetland ecosystems, 281
Nitrobacter spp., 169, 171
nitrogen (N)
ammonia (NH3)
assimilation processes, 171, 171f
ammonium (NH4)
anaerobic ammonium oxidation,
171f, 172
saline/soda lakes, 383
temperate forest biomes, 424–425
upwelling ecosystems, 451
activated sludge systems, 178
desert stream ecosystems, 220, 220f
lagoon ecosystems, 298
mangrove forests, 311, 312f
microbial processes
basic concepts, 171f
nitrogen transformation processes, 171
potential denitification activity (PDA)
measurements, 175f
sewage treatment systems, 178
desert stream ecosystems, 219, 220f
salt marsh ecosystems, 390, 392
elemental concentrations, 10t
desert environments, 237
lagoon ecosystems, 298
postclosure landfills, 306
temperate forest biomes, 424–425
upwelling ecosystems, 454
forest plantations, 269
lagoon ecosystems, 297, 297f, 300f
microbial processes
anaerobic ammonium oxidation, 171f, 172
denitrification
basic concepts, 171f
nitrogen transformation processes, 171
potential denitification activity (PDA)
measurements, 175f
general discussion, 170
immobilization processes, 171, 171f
mineralization processes, 171, 171f
nitrification, 171, 171f
molecular nitrogen (N2), 312f
nitrate (NO3)
desert stream ecosystems, 220, 220f
lake ecosystems, 21
phytoplankton, 452–453
temperate forest biomes, 424–425
upwelling ecosystems, 451
nitric oxide (NO)
microbial processes, 171f
ecological roles
oxygen minimum zones (OMZs), 453
lagoon ecosystems, 298
microbial processes
basic concepts, 171, 171f
nitrite (NO2)
anammox process
activated sludge systems, 178
microbial processes, 171f, 172
upwelling ecosystems, 453
grasslands, 409–410
mangrove forests, 311, 312f
marshes, 278
temperate forest biomes, 424–425
urban environments, 462–463
nitrogen-phosphorus ratio plot, 139f
nitrous oxide (N2O)
microbial processes, 171f
NPZD (nitrogen-phytoplankton-zooplanktondetritus) model, 299
particulate organic nitrogen (PON), 452–453
stable isotope analysis, 20
stream ecosystems, 369
Nitrosomonas europaea, 171
Index
Nitrosomonas spp., 169
Nitrospira spp., 169, 171
Nitzchiaceae, 288t
Nitzschia spp., 370f
nival vegetation zone, 150, 151f
nomadic cultures, 238, 238f
nonincised streams, 254f
nonlinear biological processes, 94
North America
alpine forests, 157, 159t, 160f
desert environments
animals, 229
major deserts, 223t
plants, 228
floodplain ecosystems, 257t, 261
intertidal area research, 378
marshes, 278
savanna ecosystems, 394
temperate forest biomes, 426
temporary water bodies, 429t, 433, 434f
wetland ecosystems, 278, 281
North Equatorial Counter Current, 454
Northern Shoveler, 277
Norway
alpine forests, 158
polar ecosystems, 339–342
Nostoc spp., 371
Nothofagus spp.
alpine ecosystems, 159t
Mediterranean ecosystems, 320t, 326
temperate forest biomes, 417
notonectids, 382
Notostraca, 433, 434f
NPZD (nitrogen-phytoplankton-zooplanktondetritus) model, 299
nucleotides, See DNA (deoxyribonucleic acid)
nudibranchs, 376
nuthatches, 182
nutria
marshes, 278, 280
salt marsh ecosystems, 392
nutrient cycling
agroecosystems, 146
alpine ecosystems, 153
desert stream ecosystems, 218, 219, 220f
dune ecosystems, 242
embodied nutrient flows, 67, 67f,
68f, 68t, 69t
emergence/emergent properties, 95t
estuarine ecosystems, 249
floodplain ecosystems, 255, 256f
forest plantations, 267
grasslands, 406
indirect interactions, 85
lagoon ecosystems, 299
mangrove forests, 311, 312f
marshes, 278, 279–280
nutrient sequestration, 337
orientation theory, 124t
peatlands, 331, 332f, 337
postclosure landfills, 305
riparian systems, 348
salt marsh ecosystems, 391
savanna ecosystems, 394, 395–396,
398, 398f, 401t
stable isotope analysis, 21
stream ecosystems
matter fluxes, 355, 355f
surface waters, 220
temperate forest biomes, 424, 424f, 425f
tropical rainforests, 442
upwelling ecosystems, 450–460
characteristics, 451
chlorophyll biomass, 452, 452f
climate
climate change effects, 458, 458f
El Niño Southern Oscillation (ENSO),
459, 459f
long-term trends, 459
eddy systems, 454–455
fish/fisheries
climate change effects, 458, 458f
fish production, 457, 457f
food webs, 455, 455f
general discussion, 455
small pelagic fish, 455, 456f
trophic levels, 457
general discussion, 450
intermittence, 451
occurrences, 450, 450f
open-ocean upwelling systems, 454
oxygen depletion, 453
phytoplankton community structure, 452
primary production, 451
zooplankton, 451, 453
urban environments, 463, 463t, 465, 465t
Nyctea scandiana, 447, 447f
Nyssa aquatica, 257–258
Nyssa spp., 255–256, 262
O
oak trees
temperate forest biomes, 417, 421
oats, 405f
Ob River, Siberia, 260
Oceanapiidae, 288t
Oceania, See Pacific Island region
oceanic bogs, 334, 334f, 338
oceans
carbon cycle
Coriolis effect, 450
Ekman layer, 450
high-nutrient low-chlorophyll (HNLC)
ecosystems, 454
nitrification processes
oxygen minimum zones (OMZs), 453
climate
climate change effects, 459, 459f
regime shifts, 19, 20f
upwelling ecosystems, 450
overfishing
coral reefs, 205, 205f, 210f, 211, 211f
freshwater lakes, 274
indirect interactions, 85
peat formation, 338
upwelling ecosystems, 450–460
characteristics, 451
chlorophyll biomass, 452, 452f
climate
climate change effects, 458, 458f
El Niño Southern Oscillation (ENSO),
459, 459f
long-term trends, 459
eddy systems, 454–455
fish/fisheries
climate change effects, 458, 458f
fish production, 457, 457f
food webs, 455, 455f
general discussion, 455
small pelagic fish, 455, 456f
trophic levels, 457
general discussion, 450
intermittence, 451
occurrences, 450, 450f
open-ocean upwelling
systems, 454
oxygen depletion, 453
phytoplankton community
structure, 452
primary production, 451
zooplankton, 451, 453
505
water cycle, 365f
See also aquatic environments; marine
ecosystems
Odocoileus hemionus, 196
Odonata, 372f, 372–373
Odum, Eugene Plesants
ecosystem growth, 16
emergence/emergent properties, 91
salt marsh ecosystems, 13, 388
synecology, 76
Odum, Howard Thomas
fundamental ecosystem theory, 36t
salt marsh ecosystems, 13
self-organization, 99
offspring
environmental fate
eco-exergy losses, 136
oil spills, 314
salt marsh ecosystems, 392
Okavango Delta, 259
old-growth forests, 420, 425f
Olea europaea, 321t, 325, 328
Oliarus spp., 193
oligochaete worms, 371, 372f
oligotrophic ecosystems
peatlands, 337
Olindiidae, 288t
Olividae, 288t
Olympic Park, Beijing, China, 466
Oman, 223t
ombrogenous peatlands, 332
ombrotrophic ecosystems, 333, 334f
omnivores
desert environments, 229
onagers, 239
Oncorhynchus clarki henshawi, 384
Oonops spp., 193
oozes, 193
open systems
emergence/emergent properties, 95t
fundamental ecosystem theory, 34
hierarchy theory, 118
self-organization, 98
Ophiactidae, 288t
Ophiactis spp., 294f
Ophiocomidae, 288t
Opsanus beta, 295f
Opuntia spp., 238
Orbiniidae, 288t
orchids
postclosure landfills, 306
wetland ecosystems, 275
Ord River, Australia, 261
Oregon (USA)
stream ecosystems, 367f, 368f
organelles, 166
organic farming, 140
organic matter
coarse particulate organic matter (CPOM),
349, 352f, 354f, 356, 358f, 368
carbon:nitrogen (C:N) ratio, 462–463
dissolved organic matter (DOM)
estuarine ecosystems, 249, 249f
stream ecosystems, 356
exergy calculations, 132, 134t,
137, 138t
fine particulate organic matter (FPOM), 349,
352f, 354f, 356, 358f, 368
particulate organic matter (POM)
riparian systems, 349
upwelling ecosystems, 453
peatlands, 331
riparian systems, 343f, 344
stream ecosystems
characteristics, 356
dissolved organic matter (DOM), 356
energy sources, 351, 352f, 354f
506
Index
organic matter (continued )
modeling studies
biotic diversity patterns, 359f
ecotones, 359
flood pulse concept, 344, 359t, 360
hierarchical scales, 359, 359t
hydraulic stream ecology, 359t, 361
hyporheic dynamics, 359t, 361
network dynamics, 359t, 361
patch dynamics, 359t, 361
riparian zone influences, 359t, 360, 360f
river continuum concept (RCC), 344, 356,
357f, 358f, 359t
serial discontinuity, 357
retention potential, 356
sources, 356
transport/storage mechanisms, 355
organismal ecology, 7t
organotrophs, 167t
orientation theory
basic orientors, 122, 123f, 124t
complex environments, 121
environment properties, 122, 123f
goal functions, 124t
implicit attractors, 123
influential factors, 124, 125f
modeling studies
animat algorithms, 126
lifestyle development, 126, 127f
orientor properties, 123
stimulus-response relationships, 124, 125f
Orinoco River
floodplain ecosystems, 262
savanna ecosystems, 396
tropical rainforests, 440
Orthid soils, 226
Orthoptera, 373
oryx, 230, 239, 240f
Oryx leucoryx, 240, 240f
Oscillatoriaceae, 288t
oshanas, 429t
ostracods
Florida Everglades mesocosm project, 294f
temporary water bodies, 433
ostriches
desert environments, 239
Ostrya, 321t
otters
riparian systems, 348
ovenbirds, 423
overfishing
coral reefs, 205, 205f, 210f, 211, 211f
freshwater lakes, 274
indirect interactions, 85
overgrazing effects, 155, 278, 399f, 411
Ovibos moschatus
tundra ecosystems, 446–447
Oweniidae, 288t
Owens Valley, California (USA), 237f
owls
boreal forest ecosystems, 182
temperate forest biomes, 423
tundra ecosystems, 447
Oxford Physic Garden, England, 185
Oxisols, 395–396
Oxychilus spp., 193
oxygen (O)
biochemical oxygen demand (BOD5)
activated sludge systems, 178
constructed wetlands
horizontal subsurface flow (HSSF), 176
lagoon ecosystems, 297
wastewater treatment, 176
chemical oxygen demand (COD)
activated sludge systems, 178
elemental concentrations, 10t
oxygen minimum zones (OMZs), 453
stable isotope analysis, 20
stream ecosystems, 369
upwelling ecosystems, 453
Oxytrichidae, 288t
Oxytropis spp., 446f, 448
oysters
competition studies, 83
estuarine ecosystems, 250
Florida Everglades mesocosm project, 293f
mangrove forests, 310
P
P450 gene, See cytochrome P450
Pachastrellidae, 288t
Pachygrapsus crassipes, 387
Pacific Coast Range (USA), 159t, 160f
Pacific Decadal Oscillation (PDO), 216
Pacific Island region
intertidal area research, 378
tropical ecosystems, 441
packrats, 196
Pagrus auratus, 21, 22f
Paine, RT, 378
pakahi, 429t
Pakistan, 223t
Palaemonetes palidosus, 294f
Palafoxia lindenii, 243f
Palm House, Kew Gardens (United Kingdom),
183, 184f
Palo Verde Marsh, Costa Rica, 253f
palsas, 444
paludification, 338
Pamir Mountains, 159t, 160f
pampas, 406–407, 433
Panama
rainforests, 440f
Panamint Valley, California (USA), 236f
Panicum hemitomon, 276f
Panicum maximum, 246
Panicum racemosum, 243f
pans, 429t, 432
Pantanal River Basin, 259, 281
Panthera pardus jarvisi, 240
Papua New Guinea
alpine forests, 157
savanna ecosystems, 394
tropical ecosystems, 441, 442
papyrus, 259–260
parabolic dunes, 241
Paracerceis caudata, 294f
Paracoccus denitrificans, 171
Paradiaptomus africanus, 382
Paradoxostomatidae, 288t
paradox theories
impossible staircase, 108f
surprise-generating mechanisms, 108,
108f, 110t
parafluvial zone, 214f, 214–215, 220
paragrass, 261
Parameciidae, 288t
Páramo grasslands, Ecuador, 151, 152f
Parana River
floodplain ecosystems, 262
temporary water bodies, 431
Paranthuridae, 288t
Paraonidae, 288t
parapatry
bumblebees, 448
cleaner fishes, 206
desert environments, 236, 236f
tropical rainforests, 441
Parasystatus elongates, 448
Paratanaidae, 288t
parrotfishes
coral reefs
ecosystem connectivity, 208f
importance, 203f, 203–204, 207–208
marine reserves, 212f, 213
positive interactions, 206
parrots, 230
particulate organic matter (POM)
riparian systems, 349
upwelling ecosystems, 453
particulate organic nitrogen (PON), 452–453
Paspalum maritimum, 244
Paspalum spp., 396
Paspalum vaginatum, 386t
Passerculus sandwichensis beldingi, 387
pastoralist ecosystems, 238, 238f
Patagonian Desert, 223t
patch dynamics
river system dynamics, 359t, 361
pathogens
coral reefs, 210
forest plantations, 269
freshwater lakes, 274
temperate forest biomes, 420
Patten, Bernard, 76
peacocks, 259–260
peat/peatlands, 330–339
bogs
characteristics, 175, 275, 333, 429t
continental bogs, 334, 334f, 335f
development factors, 332f
ecological network analysis (ENA), 71,
72f, 73t
largest wetlands, 281
nutrient cycling, 278
oceanic bogs, 334, 334f, 338
water availability, 276–277
boreal forests, 181
carbon sinks, 330, 332f, 338
characteristics, 330
classification, 332, 333f
environmental factors, 331, 332f
fens
characteristics, 175, 275, 333,
334, 429t
development factors, 332f
largest wetlands, 281
patterned fens, 335, 336f
poor fens, 335
rich fens, 335, 335f
water availability, 277
formation processes, 338
global distribution, 331, 331f
key processes
acidification, 335
methanogenesis, 337
nutrient sequestration, 337
sulfate reduction, 338
water retention, 336
occurrences, 331
peat moss, 330, 333f, 336
swamps, 338, 415, 415f
temporary water bodies, 429t
tundra ecosystems, 444–445
water availability, 276–277
West Siberian Lowland, 275, 276f
wildfires, 278, 338
Pectinidae, 288t
Pedetes spp., 230
PeeDee limestone, 20
pelagic ecosystems
modeling studies
NPZD (nitrogen-phytoplanktonzooplankton-detritus) model, 299
Pelecypoda, 371
pelicans, 457
Peneroplidae, 288t
penguins
climate change effects, 342
polar ecosystems, 342
upwelling ecosystems, 457
Percolozoa, 288t
perennial plants
savanna ecosystems, 397
Index
periglacial environments, 444
periphyton
food webs, 41, 42f
stream ecosystems
biotic diversity patterns, 356–357, 359f
characteristics, 370
organic matter, 352f, 358f
Periplaneta spp., 193
Peritromidae, 288t
permafrost
boreal forests, 181
continental bogs, 334, 334f, 335f
tundra ecosystems, 444
Peromyscus maniculatus, 196
persistence, 104
Peru
alpine forests, 159t, 160f
desert environments, 223t
fish/fisheries, 459
Peruvian booby, 457
pest control
crop production, 147
forest plantations, 269
pesticides
agroecosystems, 146f, 147
Petra, Jordan, 237f
gasoline
eco-exergy losses, 136
oil spills, 314
salt marsh ecosystems, 392
Peumus boldus, 320t
Peysonneliaceae, 288t
pH
stream ecosystems, 369
Phaeocystis antarctica, 454
Phaeophyceae
intertidal environments, 375, 375f
Phalaris arundinacea, 278
phalaropes, 383
Phalaropus spp., 383
Phascolionidae, 288t
Phascolion strombi, 294f
Phascolosomatidae, 288t
Phasianellidae, 288t
Phasmida, 134t
Philippines, 441, 442
restrictive mode
holistic versus reductionistic research, 77, 92
usability, 12–16
Phoeniconaias minor, 382
Phoenicopterus andinus, 225f
Phoenix, Arizona (USA), 462
Phoenix dactylifera, 233
PHOREDOX removal process, 179
phorid flies, 193
phosphorus (P)
activated sludge systems, 179
desert stream ecosystems, 219
elemental concentrations, 10t
forest plantations, 269
lagoon ecosystems, 297
lake ecosystems
freshwater lakes, 272
large-scale whole ecosystem experiments, 21
marshes, 278
microbial processes, 170, 170f
nitrogen-phosphorus ratio plot,
137–138, 139f
phosphates
microbial processes, 170, 170f
mangrove forests, 311, 312f
urban environments, 462–463
savanna ecosystems, 395–396
sewage treatment systems, 179
photoautotrophs, 167t
photosynthesis
biochemical pathways, 137
carbon cycle
alpine ecosystems, 155
coral-algal mutualism, 201
desert environments, 230
desert stream ecosystems, 219
energetics, 219
grasses, 406
peatlands, 331, 332f
photosynthetically active radiation (PAR)
desert stream ecosystems, 218
primary production, 18
salt marsh ecosystems, 391
savanna ecosystems, 397, 403
solar energy availability, 18
temperate forest biomes, 423–424
temporary water bodies, 435
tropical rainforests, 442
photosynthetically active radiation (PAR)
desert stream ecosystems, 218
phototrophs
characteristics, 167t
Phragmites australis
invasive species, 391
marshes, 275
wastewater treatment
characteristics, 175
vegetated beds, 176
Phragmites spp., 259
Phrynosoma spp., 230
Phyramidellidae, 288t
Physa spp., 371
Physella cubensis, 294f
Physidae, 372f
autecology, 435, 435f
temporary water bodies, 435, 435f
physiology
autecology, 435, 435f
temporary water bodies, 435, 435f
phytoflagellates, 384, 452–453, 454
phytoliths, 405–406
phytophagous insects, 182
phytoplankton
community structure, 452
eastern African soda lakes, 382
estuarine ecosystems, 248, 249t,
250, 251
exergy calculations, 130–131
indirect interactions, 85
lake ecosystems
freshwater lakes, 272
saline/soda lakes
eastern African soda lakes, 382
Mono Lake (USA), 383
mangrove forests, 312
nitrification processes, 452–453
NPZD (nitrogen-phytoplankton-zooplanktondetritus) model, 299
nutrient resources, 454
river system dynamics, 358f
saline/soda lakes, 382, 383
stream ecosystems, 370
trophic transfer efficiency, 19
upwelling ecosystems, 450–460
characteristics, 451
chlorophyll biomass, 452, 452f
climate
climate change effects, 458, 458f
El Niño Southern Oscillation (ENSO),
459, 459f
long-term trends, 459
community structure, 452
eddy systems, 454–455
fish/fisheries
climate change effects, 458, 458f
fish production, 457, 457f
food webs, 455, 455f
general discussion, 455
small pelagic fish, 455, 456f
trophic levels, 457
general discussion, 450
intermittence, 451
occurrences, 450, 450f
open-ocean upwelling systems, 454
oxygen depletion, 453
primary production, 451
zooplankton, 451, 453
phytotelmata, 428t, 429t, 429t
Picea engelmannii, 161f
Picea mariana, 334, 334f, 335f
Picea spp.
alpine ecosystems, 151, 159t
bogs, 334, 334f, 335f
boreal forest ecosystems, 181
temperate forest biomes, 417
picoautotrophs, 452–453, 455f
Pico Mucuauque, Venezuela, 445
picoplankton, 454
piedmont bajada formations, 223, 224f
pigs
floodplain ecosystems, 261
pill bugs, 371
Pilularia, 327
pine bark beetles, 269
pine sawflies, 182
pine trees
alpine ecosystems, 158
boreal forest ecosystems, 181
forest plantations, 266, 269
temperate forest biomes, 417
wetland ecosystems, 275
pinfish, 295f
pingos, 444
Pinguicula spp., 278
Pinkerton, Richard, 99
pink flamingoes, 384
pinnipeds
upwelling ecosystems, 457
Pinus cembra, 151f, 158
Pinus elliottii, 269
Pinus mugo, 158
Pinus spp.
alpine ecosystems, 151, 151f, 159t
bogs, 334
boreal forest ecosystems, 181
floodplain ecosystems, 258
forest plantations, 269
Mediterranean ecosystems
characteristics, 320t, 326
climatic stress, 328
convergence studies, 325
rainfall, 321t
temperature, 321t
shelterbelts/windbreaks, 475t
temperate forest biomes, 417, 422f
United States, 262
wetland ecosystems, 275
Pinus sylvestris, 334, 475
Pinus taeda, 269
self-organization, 104
Pisaster ochraceous, 378–379
Pistacia, 326
Pistacia atlantica, 321t
pitcher plants, 277, 278
placozoa, 134t
Plakinidae, 288t
plankton
estuarine ecosystems, 248
intertidal area research, 379
Planorbulinidae, 288t
Plantago lanceolata, 462
Plantanus spp., 262
planthoppers, 192
plant lice, See aphids
507
508
Index
plants
African floodplain ecosystems, 259–260
agroecosystems, 146f
alpine ecosystems
adaptive processes, 151, 152f
biodiversity, 153, 154f
climatic boundaries, 150, 151f
ecosystem processes, 153
flowering plants, 153, 154f
high-elevation treeline, 151, 152f
nutrient cycling, 153
semidesert environments, 153, 153f
water consumption, 153, 153f
Australian floodplain ecosystems, 261
biodiversity, 148
breeding behaviors, 101
carnivorous plants
autocatalysis, 41, 42f
wetland ecosystems
marshes, 278
peatlands, 330
riparian systems, 347
wet meadows, 275
cell structure, 166
chaparral
biogeographical evolution, 324
characteristics, 326
chaparral ecosystems, 195–200
allelopathy, 197
bare zone, 197f
characteristics, 195, 196f
fire-dependent regeneration, 195–196
invasive species, 200
management strategies, 200
regional variations, 199
seed dispersal, 199
seed germination, 199
succession, 196
vegetation communities, 195
wildfires
characteristics, 197
dominance-diversity curves, 198f
management strategies, 200
postfire recovery rates, 197, 198f
regional variations, 199
seed germination, 199
wind effects, 197, 199
commensalism
hierarchy theory, 114
coral reef microcosm, 288t
desert environments, 235
savanna ecosystems, 397
desert environments, 228, 229f, 230, 232
detritus
composition
carbon:nitrogen (C:N) ratio, 462–463
ecological significance
vegetal detritus, 255–256
floodplain ecosystems, 255–256
postclosure landfills, 305
dominance-diversity curves
chaparral ecosystems, 198f
dune ecosystems
characteristics, 241
competition, 245
diseases, 246
dune profile characteristics, 245f
general discussion, 244
pioneer zone, 245
predators, 246
sand movement, 242, 243f
succession and facilitation, 245
symbiotic relationships, 246
ecological network analysis (ENA), 73t
environmental tolerance
United States, 261
exergy calculations, 134t
floodplain ecosystems
Africa, 259–260
Australia, 261
South America, 262–263
grazing systems, 329
human-environment interactions, 462,
463t, 464f
indirect interactions, 85
botanic gardens, 189
desert environments, 238
marshes, 280
salt marsh ecosystems, 391
urban environments, 462
exergy calculations, 137
marshes, 277, 278, 280
Mediterranean ecosystems
biodiversity patterns, 322, 322t
biogeographical evolution, 323
characteristics, 320t, 326
convergence studies, 325
grazing, 329
rainfall, 321t
temperature, 321t
wildfires, 328
natural containers, 428t, 429t
nitrogen cycle
photosynthetic pathways
desert environments, 230
exergy calculations, 137
phytotelmata, 428t, 429t
self-organization, 104
alpine ecosystems, 153
ecosystem services, 24, 24f, 28
pollen vectors, 28
urban environments, 465, 465t
demographic traits
seed germination, 161f, 164t
seedling establishment, 161f, 164t
postclosure landfills, 304, 305
riparian systems, 348
river system dynamics, 351
root systems
cave habitats, 193
estuarine ecosystems, 249–250
floodplain ecosystems, 257
postclosure landfills, 305
savanna ecosystems, 397
symbiotic relationships, 397
salinity tolerances, 435f
salt marsh ecosystems
characteristics, 385, 387
ecological systems, 388
global change effects, 390
global distribution, 386t
invasive species, 391
research value, 392
upland-wetland interface, 391f
savanna ecosystems, 397
seasonality
chaparral ecosystems, 199
postclosure landfills, 307
spatial pattern formation, 104, 104f
seeds
agroecosystems, 146f
buried seeds, 278
chaparral ecosystems, 199
desert environments, 233, 233f
fire-dependent regeneration, 195–196, 197
floodplain ecosystems, 257
postclosure landfills, 306
chaparral ecosystems, 199
postclosure landfills, 307
spatial pattern formation, 104, 104f
self-organization, 98, 98f
shelterbelts/windbreaks, 470
South American floodplain ecosystems,
262–263
succession studies, 182
tropical ecosystems, 439–440, 441
urban environments, 462, 463, 463t, 464f
See also photosynthesis; vascular plants
Platanus spp., 325
Platyhelminthes
exergy calculations, 134t
temporary water bodies, 433
playas, 225, 225f, 428t, 433f, 434f
Plecoptera
stream ecosystems, 354, 354t, 372f,
372–373
Pleistocene
desert environments, 228, 237
glaciation, 321, 324
Mediterranean ecosystems, 323, 324
temperate forest biomes, 418
tundra ecosystems, 443
Pleurocapsaceae, 288t
Pleuronematidae, 288t
Plexauridae, 288t
Pliocene
desert flora, 228
Mediterranean ecosystems, 323
Pluchea odorata, 244
plunge pools, 429t
pneumatocysts, 375–376
Poaceae
desert environments, 229
Mediterranean ecosystems, 320t
Poa spp., 320t
pocosins, 429t
Podaxis psitillaris, 227–228
Podiceps nigricollis, 383
Podocarpus spp., 159t, 320t, 326
Podsols, 181
Poecilia latipinna, 295f
Poecilia reticulata, 295f
Poeciliidae, 373
poikilohydry, 230
poisons, as defense mechanism, 236
polar bears
polar ecosystems, 340–341
polar ecosystems, 339–342
Antarctica, 341f
Arctic region, 340f
characteristics, 339–340
climate change effects, 341
desert environments, 222
alpine ecosystems, 153
ecosystem services, 24, 24f, 28
pollen vectors, 28
urban environments, 465, 465t
pollutants, 141t
pollution, 140
coral reefs, 205, 205f, 210f
landfills, 303
membrane systems
salt marsh ecosystems, 390, 392
urban environments, 463t
pollution effects, 140
poly -hydroxyalkanoate, 170, 170f
polychaetes (Polychaeta)
Florida Everglades mesocosm project, 294f
intertidal environments, 376
lagoon ecosystems, 299–300
tube-building polychaetes
intertidal environments, 376
polycyclic aromatic hydrocarbons (PAHs)
salt marsh ecosystems, 392
polygon landscape features, 444
Polygonum spp., 259
Polylepis spp., 151, 159t
polymerase chain reaction (PCR) technique, 169
Index
polysaccharides, 168f
Polysiphonia subtilissima, 293f
Pomacentridae, 203–204, 288t
ponds, 444–445
pondweeds, 176, 275
Pontederiaceae, 371
pools and riffles, 366, 366f
poor fens, 335
poplars, 181
Popper, Karl, 57
population dynamics
Daisyworld model, 103, 105
emergence/emergent properties, 94
fragmented populations, 105, 105f
K-strategists, 436
modeling studies
autocatalysis, 41, 57–58
coral-algal mutualism, 201, 205
indirect interactions, 57–58, 60, 81, 83, 83f
network environ analysis, 79, 80t
community assembly, 77
population growth rate
negative feedback, 101
stream ecosystems, 352, 352f, 354f,
354t, 358f
upwelling ecosystems, 455f, 457
desert stream ecosystems, 218t
estuarine ecosystems, 252
r-strategists, 436
self-organization, 98
temporary water bodies
bet hedging strategies, 436
complex life histories, 437
dispersal mechanisms, 437
life-history strategies, 436
population ecology, 7t
Populus deltoides, 256–257, 268
Populus spp.
boreal forest ecosystems, 181
Europe, 261
forest plantations, 268
shelterbelts/windbreaks, 475t
temperate forest biomes, 421
United States, 262
Populus xiaozhuanica, 475
Porcellio spp., 193
Porifera
coral reefs, 205, 211f
exergy calculations, 134t
Florida Everglades mesocosm project, 294f
intertidal environments, 376
mangrove forests, 310
Poritidae, 288t
Portugal
botanic gardens, 184t, 185
invasive species, 267
positive feedback
reproduction, 101
postfire regeneration, 197, 198f
Potamocorbula amurensis, 391
Potamogetonaceae, 371
Potamogeton pectinatus, 176
potassium (K)
elemental concentrations, 10t
forest plantations, 269
saline/soda lakes, 381
stream ecosystems, 369
potholes, 429t, 429t
prairie ecosystems, 405–413
shortgrass prairies, 406
tallgrass prairies, 406
temporary water bodies, 429t, 433, 434f
prairie potholes, 278, 281, 429t, 433, 434f
prawns, 299–300
precipitation
alpine forests, 162
boreal forests, 181
desert environments, 214, 224, 225, 226f
dune ecosystems, 243
grasslands, 407
hydrologic cycle, 364, 365f
Mediterranean ecosystems, 319, 320t, 321t
riparian systems, 344
savanna ecosystems
conceptual models, 401t
environmental factors, 398, 398f
seasonality, 394–395, 395f, 396f
temperate forests, 418
tropical ecosystems
rainforests, 439–440, 441
agroecosystems, 147
cave habitats, 193
competition studies, 82
coral reefs, 204, 204f, 212f, 213
desert environments, 229, 236
dune ecosystems, 246
endangered species, 239
indirect interactions, 81, 85
network environ analysis, 79, 80t
population dynamics
desert environments, 236
food webs, 21, 22f
intertidal environments
invertebrates, 376
vertebrates, 377
negative feedback, 101
stream ecosystems, 352, 352f,
354f, 354t, 358f
upwelling ecosystems, 455f, 457
salt marsh ecosystems, 389
tundra ecosystems, 447
urban environments, 464
predictable behaviors, 107
Prigogine, I
dissipative systems, 99
input-output models, 76
irreversible processes, 34
primary emergence, 92, 96f
primary forests, 264f, 265f
alpine ecosystems, 153
autotrophs, 17
boreal forest ecosystems, 182
desert environments, 234, 234t, 235f
dune ecosystems, 242
estuarine ecosystems, 249, 249t, 251
grassland models, 409f
gross primary production (GPP)
coral reef microcosm, 287f
energy flow, 9t, 65–66
intertidal environments, 377–378
mangrove forests, 311
net primary production (NPP)
aboveground net primary production
(ANPP), 409f
desert environments, 234, 234t, 235f
energy flow, 9t
savanna ecosystems, 394, 403, 403t
photosynthesis, 18
river system dynamics, 351
secondary productivity, 17
upwelling ecosystems, 450–460
basic concepts, 451
characteristics, 451
chlorophyll biomass, 452, 452f
climate
climate change effects, 458, 458f
El Niño Southern Oscillation (ENSO),
459, 459f
long-term trends, 459
eddy systems, 454–455
fish/fisheries
climate change effects, 458, 458f
fish production, 457, 457f
food webs, 455, 455f
509
general discussion, 455
small pelagic fish, 455, 456f
trophic levels, 457
general discussion, 450
intermittence, 451
occurrences, 450, 450f
open-ocean upwelling systems, 454
oxygen depletion, 453
phytoplankton community structure, 452
zooplankton, 451, 453
probability density functions (PDFs), 172
Prochlorococcus, 454
productive forest plantations, 264,
264f, 265f
prokaryotes
cell structure, 166, 167t
temporary water bodies, 433
See also Archaea; bacteria
propagule pressure, 379
Prorocentraceae, 288t
Prorodontidae, 288t
Prosopis chilensis, 320t, 326
Prosopis spp., 232
Protea, 323
Proteaceae, 320t, 323
protective forest plantations, 264,
264f, 265f
Protists
coral reef microcosm, 288t
Protium icicariba, 244
Protocruziidae, 288t
protozoa
cell structure, 166, 167t
community structure, 168f
desert environments, 229
estuarine ecosystems, 249, 251
exergy calculations, 134t
saline/soda lakes, 382
temporary water bodies, 433
Providencia, Antioquia, Colombia, 440f, 443f
Prunus ilicifolia, 195
Psammomys obesus, 232f
Pseudocyclopidae, 288t
pseudoscorpions, 193
Pseudotsuga menziesii, 267, 417
Psilotrichidae, 288t
Psphenidae, 372f
ptarmigans, 446–447
Pteroclidiformes, 231
Ptycocyclidae, 288t
public education programs, 14–15
Puccinellia maritima, 386t
Puccinellia phryganodes, 386t
Puerto Rico, 313
puffballs, 227–228
puku, 259–260
puma
temperate forest biomes, 423
Puma concolor
temperate forest biomes, 423
Pune, India, 464–465
purple loosestrife
aquatic wetlands, 275f
botanic gardens, 185
puszta, 406–407
Puya, 320t
pycnoclines, 451
pygmy bamboo, 397
Pyramid Lake, 384
Q
Qatar, 223t
Qinghai (China), 381
quantum theory, 34
Quaternary
tectonic activity, 321
510
Index
Quercus robur, 261
Quercus spp.
alpine ecosystems, 159t
chaparral ecosystems, 195
floodplain ecosystems, 255–256
Mediterranean ecosystems
characteristics, 320t
climatic stress, 328
convergence studies, 325
global change effects, 329
rainfall, 321t
temperature, 321t
temperate forest biomes, 417, 421
United States, 262
Quillaja saponaria, 320t
R
rabbitfishes, 203–204
rabbits
chaparral ecosystems, 196
desert environments, 238
dune ecosystems, 246
intertidal environments, 377
temperate forest biomes, 423
raccoons, 348
rails, 348
rainbow parrotfish, 207–208
rainfall
acidification processes, 369–370
agroecosystems, 146f
alpine forests, 162
boreal forests, 181
desert environments, 214, 224, 225, 226f
dune ecosystems, 243
grasslands, 407
marshes, 278
Mediterranean ecosystems, 319, 320t, 321t
riparian systems, 344
savanna ecosystems
conceptual models, 401t
environmental factors, 398, 398f
seasonality, 394–395, 395f, 396f
tree cover relationship, 402f, 402–403
temperate forest biomes, 423
tropical ecosystems
rainforests, 439–440, 441
United States, 261
rainforests
photographic views, 17f
temperate forest biomes, 417, 419t
tropical ecosystems, 439–443
biodiversity, 442
canopy characteristics, 442, 442f
characteristics, 439, 440f
climate, 441
conservation strategies, 442, 443f
general discussion, 439
global distribution, 440
soils, 441
structural characteristics, 441, 442f
vegetation classifications, 439–440
rainpools, 428t, 431
rain shadow deserts, 222
Rallus longirostris levipes, 387
ramblas, 429t
Rana sylvatica, 437
Randia laetevirens, 243f, 244
random networks, 100
Rangifer tarandus, 446–447
Ranunculaceae, 371
Ranunculus spp.
Mediterranean ecosystems, 327
rats
cave habitats, 191, 193
chaparral ecosystems, 196
intertidal environments, 380
See also mouse/mice; rodents (Rodentia)
Rattus norvegicus, 380
Rattus rattus, 193
ravens
tundra ecosystems, 446–447
recreational activities
riparian systems, 350
urban environments, 465, 465t
coral reefs, 206, 207f
fish/fisheries
optimal environmental window
(OEW), 457, 457f
upwelling ecosystems, 457, 457f
red alder, 424f
red algae
exergy calculations, 134t
Florida Everglades mesocosm project, 293f
intertidal environments, 375
stream ecosystems, 370
redberry, 195
Redfield ratio
ecological stoichiometry
lagoon ecosystems, 298
red fox, 182
red mangrove, 284f, 291f, 293f
red-necked phalaropes, 458
red pine trees, 422f
red squirrels
forest plantations, 268
reductionistic research, See holistic versus
reductionistic research
Reduviidae, 196
redwood trees, 417
reeds, See Phragmites australis
reed sweet grass, 176
Rees, Martin, 13
reforestation, 266
refuges
riparian systems, 344
regs, 224
regular networks, 100
rehabilitated landfills, 304
reindeer
tundra ecosystems, 446–447
reindeer lichen, 334, 334f
relativity theory, 34
renewable resources
eco-exergy losses, 136, 136t
reproductive systems
coral reefs, 206
desert stream ecosystems, 217
fire-dependent regeneration, 195–196
positive feedback, 101
reptiles (Reptilia)
boreal forest ecosystems, 182
desert environments
characteristics, 230
drought tolerance, 230
predation, 236
water uptake, 231
exergy calculations, 134t
Mediterranean ecosystems, 322, 322t
postclosure landfills, 307
riparian systems, 348
salt marsh ecosystems, 387
temperate forest biomes, 423
Republic of the Congo, 440
reserves, See marine reserves
reserve selection
desert stream ecosystems, 218t
estuarine ecosystems, 252
resource competition
coral reefs, 202, 202f
respiratory systems
desert stream ecosystems, 219
emergence/emergent properties, 95t
microbial processes, 169, 169f
temperate forest biomes, 425
Restionaceae, 320t
Reticulosidae, 288t
Rhamnus californica, 195
Rhamnus crocea, 195
Rheotanytarsus spp., 372f
Rhizobium spp.
dune ecosystems, 246
Rhizoclonium riparium, 293f
Rhizophora spp., 249t, 250
Rhizopoda, 288t
Rhodocyclus spp., 170
Rhododendron spp., 159t, 420–421
Rhodomelaceae, 288t
Rhodophyta/Rhodophyceae
coral reef microcosm, 288t
exergy calculations, 134t
Florida Everglades mesocosm project, 293f
intertidal environments, 375
stream ecosystems, 370
Rhodymeniaceae, 288t
Rhyacophilidae, 373
ribbon kelp, 18, 19f
ribonucleic acid (RNA), See RNA
(ribonucleic acid)
rice
exergy calculations, 134t
rich fens, 335, 335f
richness measurements
biodiversity
alpine ecosystems, 153, 154f
boreal forests, 182
forest plantations, 267
saline/soda lakes, 382
species diversity
alpine ecosystems, 153, 154f
boreal forests, 182
temporary-water populations, 437, 438t
tundra ecosystems, 445
urban environments, 463, 464f
Ridgewayiidae, 288t
Rio de Janeiro, Brazil, 462
Rio Grande, New Mexico (USA), 216
Rio Negro, Brazil, 256f
Rio Puerco, New Mexico (USA), 216
Rio Solimoes, Brazil, 256f
riparian systems, 342–351
basic concepts, 344
biota/biodiversity, 348
characteristics
beaver ponds, 346f, 348
flood zones, 346f, 347
hillside wetlands, 346f, 347
hygropetric zone, 346, 346f, 347f
logjam ponds, 346f, 348
parafluvial/orthofluvial ponds, 346, 346f
rockpools, 346, 346f, 347f
swamps, 346f, 347, 415
conservation strategies, 350
definitions, 343
desert environments, 214, 214f
ecological services
aquatic-terrestrial interactions, 350
carbon cycle, 349
climate effects, 349
corridors, 349
food sources, 344
general discussion, 348
hydrological buffering, 349
migration effects, 349
nutrient buffering, 348
recreational activities, 350
refuges, 344
environmental factors
climate, 344
connectivity, 345
Index
general discussion, 344
hydraulic/substrate conditions, 345
morphology, 345
spatiotemporal scales, 344
vegetation, 345
floodplain ecosystems, 253
flood pulse concept, 344
importance, 342
organic matter inputs and exchange, 343f
river continuum concept (RCC), 344
stream ecosystems
energy sources, 351
modeling studies, 359t, 360, 360f
nutrient cycling, 220–221
physical characteristics, 214
schematic diagram, 214f
spatial scales, 216
successional patterns, 215
vegetation, 218
Rissoellidae, 288t
Rissoidae, 288t
riverine forests, 281
river systems, 363–374
alkalinity, 369
biota
algae, 370, 370f
benthic macroinvertebrates, 371, 372f
fish, 373
general discussion, 370
macrophytes, 371
channel structure
channelization
desert environments, 221
pools and riffles, 366, 366f
stream order, 366f, 367
structural characteristics, 366
watersheds, 367
dissolved organic matter (DOM), 356
ecosystem dynamics, 351–362
conservation strategies, 362
energy flow
energy sources, 351, 352f, 353f
functional feeding groups (FFGs), 352,
354f, 354t
matter fluxes
nutrient cycling, 355, 355f
organic matter, 351, 352f, 355
modeling studies
biotic diversity patterns, 359f
ecotones, 359
flood pulse concept, 344, 359t, 360
hierarchical scales, 359, 359t
hydraulic stream ecology, 359t, 361
hyporheic dynamics, 359t, 361
network dynamics, 359t, 361
patch dynamics, 359t, 361
riparian zone influences, 359t,
360, 360f
river continuum concept (RCC), 344, 356,
357f, 358f, 359t
serial discontinuity, 357
organic matter
characteristics, 356
energy sources, 351, 352f, 354f
retention potential, 356
sources, 356
transport/storage mechanisms, 355
estuarine ecosystems, 248, 248f, 251
floodplain ecosystems, 254, 359t, 360
food webs, 352
hardness measurements, 369
importance, 363
ionic composition, 369
lotic research, 363
mangrove forests, 308, 309f
physico-chemical processes
channel morphology
pools and riffles, 366, 366f
stream order, 366f, 367
watersheds, 367
general discussion, 364
geomorphic processes, 366
hydrologic cycle
characteristics, 364
hydrographs, 364, 365f
intermittent streams, 364, 365f, 428t, 432f
meltwater pools/streams, 428t
schematic diagram, 365f
physcial factors
currents, 367
substrate materials, 367, 367f, 368f, 368t
temperature, 368
water chemistry, 369
sedimentation processes, 366
research background, 363
swamps, 414, 414f
temperate forest biomes, 423
tundra ecosystems, 444–445
See also stream ecosystems
Rivulariaceae, 288t
rivulus, 295f
Rivulus marmoratus, 295f
Robinia pseudoacacia, 186, 424f
rock crickets, 193
rockholes, 429t
rock lobsters, 21, 22f
rockpools, 346, 346f, 347f, 428t, 432, 434f
rock ptarmigans, 446–447
rocks, See minerals
rock size categories, 368, 368f, 368t
Rocky Mountains (USA), 150, 157,
159t, 160f, 406
rodents (Rodentia)
chaparral ecosystems, 196
competition studies, 83–84, 84f
desert environments, 230, 231, 232f
intertidal environments, 377
See also mouse/mice; rats
roots/tubers
temporary water bodies, 435
root systems
cave habitats, 193
estuarine ecosystems, 249–250
floodplain ecosystems, 257
mangrove forests, 312f
postclosure landfills, 305
savanna ecosystems, 397
symbiotic relationships, 397
Rose of Jericho, 233f
rosette plants, 151, 152f
rotifers (Rotifera)
exergy calculations, 134t
saline/soda lakes, 382
temporary water bodies, 433, 435
rough-legged hawks, 447
round worms
exergy calculations, 134t
temporary water bodies, 433
Royal Botanical Gardens at Kew (United
Kingdom), 183, 184f, 185
emergence/emergent properties, 95t
fundamental ecosystem theory, 36t
postclosure landfills, 306
temporary water bodies, 436
urban environments, 463t
rubber, 186
Rumex spp., 88
runoff
desert environments, 226
El Niño Southern Oscillation (ENSO), 216
forest ecosystems, 30
grasslands, 407
mangrove forests, 313
stream ecosystems
hydrologic cycle, 364, 365f
swamps, 414
water cycle, 364, 365f
rushes
marshes, 275f, 276f
temporary water bodies, 433
Russia
bog analysis, 71, 72f, 73t
boreal forests, 181
forest plantations, 265, 265f
peatlands, 276–277, 331
savanna ecosystems, 394
rust, 269
S
Sabella spp., 294f
Sabellidae, 288t
sabkhas, 429t
saccamoebidae, 288t
Sagittaria lancifolia, 276f
sago pondweed, 176
saguaro cactus, 229
Sahara desert, 223t, 224f
salamanders
marshes, 278
temperate forest biomes, 423
temporary water bodies, 433
temporary wetlands, 277, 432f
Salar of Uyuni (Bolivia), 381
Salicornia europea, 386t
Salicornia spp., 387
salinas, 429t
desert stream ecosystems, 221
dune ecosystems, 243
faunal communities, 382, 387
floodplain ecosystems
Asia, 260
Australia, 261
United States, 261
freshwater ecosystems
lakes, 273
mangrove forests, 311
Mediterranean ecosystems, 324
saline/soda lakes, 380–384
biodiversity, 381–382
characteristics, 381, 382f
ecological processes
Dead Sea, 384
eastern African soda lakes, 381f,
382, 382f
general discussion, 382
Mono Lake (USA), 383
economic factors, 384
geographic distribution, 381
ionic composition, 381
temporary water bodies, 429t,
432, 433f
salt marsh ecosystems, 385, 390
soils, 243, 390
Salix arctica, 446
Salix herbacea, 154f
Salix matsudana, 475
Salix spp.
boreal forest ecosystems, 181
fens, 335
shelterbelts/windbreaks, 475t
tundra ecosystems, 445f
salmon
stream ecosystems, 373
Salpingoecidae, 288t
salt
See also salinity/salinization
saltcedar, 233
salt flats, 386
salt marsh ecosystems, 384
characteristics, 385
conservation issues
chemical contamination, 392
eutrophication, 390
global change effects, 390
511
512
Index
salt marsh ecosystems (continued )
habitat loss, 390
invasive species, 391
sediment supply, 390
ecological systems
food webs, 388
soils, 388
vegetation, 388
ecosystem services
carbon sequestration, 389
commercial fishing, 389
denitrification processes, 389
forage resources, 389
recreational opportunities, 389
shoreline protection, 389
estuarine ecosystems, 249
faunal communities, 386, 387, 391
global distribution, 386
greenhouse ecosystems, 285f, 291f
habitat diversity, 386
nitrogen toxicology, 390, 392
research value
community ecology, 392
ecosystem functioning, 392
plant growth, 392
restoration methods, 393, 393f
salinity, 385
tidal systems, 385, 385f
tundra ecosystems, 444–445
vegetation
characteristics, 385, 387
ecological systems, 388
global change effects, 390
global distribution, 386t
invasive species, 391
research value, 392, 392
upland-wetland interface, 391f
See also estuarine ecosystems;
tidal wetlands
salt pans, 386
Salt River, Arizone (USA), 221
Salvia spp., 320t
Salvinia spp., 261
Samoa, 441
sand dunes, 241–246
abiotic factors
gradient structure, 242
habitats, 244, 245f
nutrient resources, 242
salinity, 243
sand movement, 242, 245f
temperature variations, 244
water sources, 243
characteristics, 224f, 224–225
dune profile characteristics, 245f
formation processes, 241
general discussion, 241
photographic views, 17f
vegetation communities
characteristics, 241
competition, 245
diseases, 246
general discussion, 244
pioneer zone, 245
predators, 246
sand movement, 242, 243f
succession and facilitation, 245
symbiotic relationships, 246
sand flats, 250
sand grouse, 230, 231
sand wave dunes, 241
San Francisco Bay, 248
San Francisco Botanical Garden (USA), 183
sanitary landfills, 303
San Pedro de Atacama, Chile, 225f
San Quintin Bay, Mexico, 385f, 391f
Santa Ana winds, 197, 199
Santa Cruz River, Arizona (USA), 221
saprophytes, 441
Sarcocornia pacifica, 386t
Sarcocornia quinqueflora, 386t
Sarcocornia spp., 387
Sarcopoterium, 320t
sardines
climate change effects, 458, 458f
upwelling ecosystems, 455, 455f
Sardinops spp., 455
Sarga spp., 395f, 396
Sarotherodon alcalicus grahami, 382
Sarraceniaceae, 347
Sarracenia spp., 278, 330
Saudi Arabia, 223t
savanna ecosystems, 394–405
Africa, 394, 395f, 396f, 396–397
anthropogenic impacts, 403
Asia, 394, 396f, 397
Australia, 394, 395f, 396, 396f
biomass concentrations, 398f, 403, 403t
characteristics, 394, 395f
competition-colonization tradeoffs, 401t
conceptual models, 400, 401t
environmental factors
general discussion, 398
herbivory, 398f, 400, 401t
interactive factors, 398f
moisture availability, 398, 398f, 401t
nutrient cycling, 398, 398f, 401t
wildfires, 398f, 399, 399f, 401t
general discussion, 394
global distribution, 394, 396f
India, 394, 396f, 397
invasive species, 404
net primary production (NPP), 234t,
394, 403, 403t
South America, 394, 396, 396f
temperature data, 396f
tree-grass ratios, 394
vegetation adaptations, 397
wetland ecosystems, 275, 281
wildfires, 394, 398f, 399f, 401t
Savannah sparrows, 387
sawflies, 182
Scaevola plumieri, 243f
ecological scaling
body size-abundance distributions, 48
scale-free networks, 100
Scandinavia
alpine forests, 157, 159t, 160f
boreal forests, 181
tundra ecosystems, 445
Scaridae, 203f, 203–204, 207–208
Scarus guacamaia, 207–208, 208f
Schizachrium spp., 396
Schizachyrium scoparium, 246
Schoenoplectus americanus, 276f
Schoenoplectus spp., 275f
Schrankia spp., 193
sclerophyllous plants, 325
See also bark beetles
scope for change in ascendency (SfCA), 63
Scorpidium scorpioides, 335f
Scorpidium spp., 335
scorpions
desert environments, 229, 236
Scotland
alpine forests, 159t, 160f
botanic gardens, 184t
intertidal area research, 378
temporary water bodies, 429t
Scots pine, 158
Scrophulariaceae, 446
scrub jay, 199
scrub oak, 195
scuds, 371
sculpin, 373
Scytonema hofmanni, 293f
Scytonemataceae, 288t
sea anemones
intertidal environments, 376
seabirds
upwelling ecosystems, 457
See also birds (Aves); waterfowl
sea cabbage, 18, 19f
seagrasses
coral reefs, 207, 208f, 211f
estuarine ecosystems, 250
seagulls
saline/soda lakes, 383
sea ice
climate change effects, 342
sea-level changes
estuarine ecosystems, 252
mangrove forests, 311, 313
salt marsh ecosystems, 389, 390
seals
upwelling ecosystems, 451, 455f, 457
sea otters, 21
sea rocket, 243
Sears, Paul, 13–14
seasonal forests
desert environments, 226
desert stream ecosystems, 215, 217f
Mediterranean ecosystems, 319
savanna ecosystems, 394–395, 395f, 398f
sea squirts
exergy calculations, 134t
seastars
intertidal environments, 376
sea turtles
overfishing, 211, 211f
coral reefs
competition, 203
disease outbreaks, 210
food webs, 203
predators, 204
food webs, 21, 22f
intertidal environments, 375, 376, 378
predator-prey dynamics, 21, 22f
stable isotope analysis, 21
seawater,
See also oceans
seaweed
coral reefs, 202, 205f
intertidal environments, 375, 377, 379, 379f
life cycles, 376
sebkas, 225
secondary emergence, 92, 96f
secondary production, 17, 311, 377–378
second law of thermodynamics, See exergy;
thermodynamic laws
secure landfills, 303
security, 122, 124t
sedges
alpine ecosystems, 153
aquatic wetlands, 275, 276f
bogs, 333
temporary water bodies, 433
tundra ecosystems, 446
sediment capping
freshwater lakes, 273
removal mechanisms, 415
salt marsh ecosystems, 390
swamps, 415
Wentworth scale, 368, 368t
Sedum spp., 154
seeds
agroecosystems, 146f
buried seeds, 278
chaparral ecosystems, 199
desert environments, 233, 233f
fire-dependent regeneration, 195–196, 197
floodplain ecosystems, 257
postclosure landfills, 306
riparian systems, 344
Index
chaparral ecosystems, 199
mangrove forests, 311
postclosure landfills, 307
spatial pattern formation, 104, 104f
seepage wetlands, 277
seeps, 428t
segmented worms, 433
Selaginella spp., 230
selection processes
autocatalysis, 41
self-organization, 4, 98–106
altruism, 103
basic concepts, 98, 98f
complex systems, 102, 109, 110, 110t, 111t
cooperation, 103
ecosystem dynamics
biosphere, 105
communities versus assemblages, 104
dissipative systems, 34
evolutionary development, 105
food webs, 104
fragmented populations, 105f
persistence, 104
social groups, 103
spatial pattern formation, 104, 104f
stability analysis, 104
emergence/emergent properties, 94,
96f, 98, 100
fundamental ecosystem theory, 34
hierarchy theory, 118
historical research, 99
homeostasis, 101
orientation theory, 124, 125f
practical considerations, 105
succession, 104
theoretical models
artificial life, 102
complex adaptive systems, 102
complex networks, 99
connectivity/criticality, 100
encapsulation/inheritance, 100
feedback, 101, 102f, 104
self-organized criticality (SOC), 101
stigmergy, 101, 102f
synchronization, 101
thermodynamic laws, 99
self-organizing holarchic open systems
(SOHOs), 118
self-regulation, 4
semiarid environments, 153, 238
Semibalanus balanoides, 378–379
semidesert environments, 153, 153f
seminatural forests, 264f, 265f
Seneciodendron keniensis, 445f
Seneciodendron keniodendron, 445f
sensory systems, 93
Sequoiadendron, 324
Sequoiadendron giganteum, 44, 417
sequoias
body size relationships, 44
temperate forest biomes, 417
Sequoia sempervirens, 320t, 324, 417
Serengeti ecosystem
grazing systems, 411
savanna ecosystems, 396–397
Serpulidae, 288t
Serranidae, 204
sessile epifauna, 376
Sesuvium portulacastrum, 386t
sewage/sewage treatment
simulation models, 179
constructed wetlands
classification, 175
emergent macrophytes
functionality, 175
biochemical oxygen demand (BOD5), 176
subsurface water flow, 175, 176, 176f
as nutrient source, 146
sewage sludge,
See also sewage/sewage treatment
sexual reproduction
pheromones
alpine ecosystems, 153
ecosystem services, 24, 24f
shad, 373
Shanghai, China, 463
ecological indicators, 60
sharks
coral reefs, 204, 204f, 211, 211f
overfishing, 211, 211f
SHARON removal process, 179
shearwaters, 458
sheep
desert environments, 230, 238
grasslands, 411
savanna ecosystems, 400
sheepshead minnows, 295f
Sheffield, England, 464–465
shellfish, See fish/fisheries
shelterbelts/windbreaks, 468–477
characteristics, 468
design principles, 474, 475f, 475t
farmland shelterbelts, 468f
management strategies, 476
mixing patterns, 475t
optical porosity determination
error analysis, 473
general discussion, 472
image processing, 472
photographic techniques, 472
structural characteristics
aerodynamic parameters, 471
cross-sectional shape, 471, 473f
external structural characteristics, 470
general discussion, 470
internal structural characteristics, 470
length effects, 471f
width effects, 472f, 472t
wind profiles, 474, 474f
wind-tree interactions
benefits, 469
ecological effects, 469
tree effects, 469
wind effects, 469
wind parameters, 468, 469f
Shepard, Paul, 13–14
shifting baseline syndrome, 211
shortgrass prairies, 406
shrews
riparian systems, 348
shrimp
estuarine ecosystems, 249, 252
intertidal environments, 376
mangrove forests, 312, 313–314
stream ecosystems, 371
temporary water bodies, 433, 436, 437
shrub ecosystems
alpine ecosystems, 152f, 153, 153f, 154f
chaparral ecosystems, 195–200
allelopathy, 197
bare zone, 197f
characteristics, 195, 196f
fire-dependent regeneration, 195–196
invasive species, 200
management strategies, 200
regional variations, 199
seed dispersal, 199
seed germination, 199
succession, 196
vegetation communities, 195
wildfires
characteristics, 197
dominance-diversity curves, 198f
management strategies, 200
postfire recovery rates, 197, 198f
regional variations, 199
513
seed germination, 199
wind effects, 197, 199
desert environments, 224f, 229,
233, 236f, 237f
desert stream ecosystems, 218
grazing systems, 329
Mediterranean ecosystems
biogeographical evolution, 323
characteristics, 320t, 326
grazing, 329
wildfires, 328
riparian zones, 218
temperate forest biomes, 420
tundra ecosystems, 445
wetland ecosystems, 281
Sialidae, 372f
forests
alpine forests, 157
boreal forests, 181
peatlands, 331, 331f
tundra ecosystems, 445
wetland ecosystems, 275, 276f, 281
Sierra Madre Mountains (Mexico), 159t, 160f
Sierra Nevada Mountains (Spain), 159t, 160f
Sierra Nevada Mountains (USA), 157,
159t, 160f
Siganidae, 203–204
Silene exscapa, 154f
silicon (Si)
elemental concentrations, 10t
grass phytoliths, 405–406
siltation (lakes), 273
silver jenny, 295f
silviculture
forest plantations, 268
simple systems, 107
Simpson Desert, 223t
Simuliidae, 372f
Sinella spp., 193
Singapore, 184t, 186, 378, 462
Singapore Botanic Gardens, 184t,
187, 187f
sinkholes, 428t, 429t
sinking creeks, 429t
Siphonidae, 288t
Sipunculids, 294f
sitatunga, 259–260
Skanderma, 159t, 160f
skeletal material, 429t
skunks
temperate forest biomes, 423
slacks, 242, 244, 245f
slash pines, 269
slender waterweed, 176
slimes
cave habitats, 193
sloughs, 429t
sludge,
See also sewage/sewage treatment
slugs, 148, 423
Smith River, California (USA), 357f
smoke beetles, 198–199
Smuts, Jan, 6
snails
agroecosystems, 148
cave habitats, 193
competition studies, 83
desert environments, 232
Florida Everglades mesocosm project, 294f
intertidal environments, 375, 375f, 376
marshes, 280
salt marsh ecosystems, 386
stream ecosystems, 371
temporary water bodies, 433, 436–437
snakes
desert environments, 230, 236
marshes, 280
riparian systems, 348
514
Index
snappers, 21, 22f, 207–208, 208f
snow
polar ecosystems, 340
snow cover
alpine ecosystems, 155
boreal forests, 181
snowfall/snowmelt
alpine forests, 162
desert environments, 214, 226
seasonal pools, 428t
temperate forest biomes, 423
snowshoe hares, 182
snowy owls, 447
Snowy Range, Medicine Bow Mountains,
Wyoming (USA), 157f
urban environments, 463t
social contagion, 102
conservation strategies, 25
social groups, 103
soda lakes, See lake ecosystems; salinity/
salinization
sodium (Na)
elemental concentrations, 10t
saline/soda lakes, 381
stream ecosystems, 369
See also salinity/salinization
agroecosystems, 146f
alpine forests, 162
biota
biodiversity, 148
boreal forests, 181
cave habitats, 191
desert environments, 226
dune ecosystems, 242
floodplain ecosystems, 253, 260
forest ecosystems
forest plantations, 269
freeze-thaw cycles, 444
hierarchical structure, 226
hydrologic cycle, 365f
indirect interactions, 85
landfills, 304
mangrove forests, 308, 309f
Mediterranean ecosystems, 320t, 321, 327
peat/peatlands, 330–339
bogs
characteristics, 333
continental bogs, 334, 334f, 335f
development factors, 332f
oceanic bogs, 334, 334f, 338
carbon sinks, 330, 332f, 338
characteristics, 330
classification, 332, 333f
environmental factors, 331, 332f
fens
characteristics, 333, 334
development factors, 332f
patterned fens, 335, 336f
poor fens, 335
rich fens, 335, 335f
formation processes, 338
global distribution, 331, 331f
key processes
acidification, 335
methanogenesis, 337
nutrient sequestration, 337
sulfate reduction, 338
water retention, 336
occurrences, 331
peat moss, 330, 333f, 336
tundra ecosystems, 444–445
polar ecosystems, 340
salinity, 243, 390
salt marsh ecosystems, 388, 390
savanna ecosystems, 395–396, 398f, 401t
swamps, 415, 415f
temperate forest biomes, 419, 421
tropical ecosystems, 441
United States, 261
urban environments, 462–463, 465, 465t
solar energy, See exergy; solar radiation/solar
energy
solar ponds, 384
solar radiation/solar energy
agroecosystems, 146f
alpine forests, 159, 165
desert environments, 227, 227f
eco-exergy storage, 34, 36f
grasslands, 407
photosynthesis, 18
savanna ecosystems, 398f
stream ecosystems, 368
temperate forest biomes, 423
tundra ecosystems, 446
See also light
solifluction, 444, 446f
soligenous peatlands, 332
Solomon Islands, 441
Somali Current, 450, 450f
Sonoran Desert (USA), 223t, 229
sooty shearwaters, 458
Sorbus spp., 151
Sossuvlei Dunes, Namib Desert, 241
South Africa
botanic gardens, 184t
desert environments, 223t
intertidal environments, 377
Mediterranean ecosystems
biodiversity patterns, 322, 322t
biogeographical evolution, 323
characteristics, 320t
climate, 319
climatic stress, 327
convergence studies, 325
Pleistocene glaciation, 324–325
soils, 321
wildfires, 328
salt marsh vegetation, 386t
savanna ecosystems, 398
upwelling ecosystems, 455, 456f
urban environments, 462f
South America
alpine forests, 157, 159t, 160f
botanic gardens, 184t
desert environments
animals, 229
major deserts, 223t
floodplain ecosystems, 257t, 262
grasslands, 406
intertidal area research, 378
saline/soda lakes, 381
savanna ecosystems, 394, 396, 396f, 398
temporary water bodies, 429t, 433
tropical ecosystems, 440
wetland ecosystems, 281
South Carolina (USA)
estuarine ecosystems, 250, 252
Southeast Asia
floodplain ecosystems, 257t, 260
savanna ecosystems, 394, 396f
tropical ecosystems, 441
Southern Alps, New Zealand, 159t, 160f, 445
Southern Ocean, 450f, 451
Spain
botanic gardens, 184t, 185
invasive species, 267
temporary water bodies, 429t
spanish grunt, 295f
Sparidae, 203–204
sparrows
desert environments, 231
Spartina alterniflora
estuarine ecosystems, 249, 249t
salt marsh ecosystems
food webs, 388
global distribution, 386t
habitats, 386
invasive species, 388, 391
plant growth research, 392
resource competition, 390
restoration methods, 393
Spartina anglica, 391
Spartina ciliata, 243f
Spartina foliosa, 386, 386t
Spartina maritima, 386t, 391
Spartina patens, 390
Spartina spp., 387, 388
Spartina townsendii, 391
indirect interactions, 85
biological impacts
group behaviors, 103
self-organization, 104, 104f
spatial pattern formation, 104, 104f
species/speciation
body size-species distributions, 48
desert stream ecosystems, 218t
ecological indicators
desert environments, 239, 240f
grasses, 240
invasive species, 240
predators, 239
forest plantations, 267
Mediterranean ecosystems, 322, 322t
self-organization, 104
richness measurements
biodiversity
alpine ecosystems, 153, 154f
boreal forests, 182
forest plantations, 267
saline/soda lakes, 382
temporary-water populations, 437, 438t
tundra ecosystems, 445
urban environments, 463, 464f
species density, 463, 464f
species diversity
Shannon index 60
alpine ecosystems, 153, 154f
boreal forests, 182
temporary-water populations, 437, 438t
tundra ecosystems, 444
urban environments, 463, 464f
tundra ecosystems, 444
urban environments, 463, 464f
speed of light theory, 34
Spelaeorchestia spp., 193
Sphaeridae, 371, 372f
Sphaeroma quoyanum, 391–392
Sphaeromatidae, 288t
Sphagnum angustifolium, 333f
Sphagnum magellanicum, 333f
Sphagnum spp.
acidification, 335
bogs, 333, 333f
canopy characteristics, 337f
methanogenesis, 337
nutrient cycling, 278
nutrient sequestration, 337
peatlands, 330
poor fens, 335
rich fens, 335
water retention, 336
Sphagnum subsecundum, 335
Sphagnum teres, 335
Sphagnum warnstorfii, 335
Spheniscus demersus, 457
Spheniscus humboldti, 457
Sphyraena barracuda, 207–208
spiders
agroecosystems, 148
cave habitats, 191, 193, 193f
desert environments, 229, 236
Index
salt marsh ecosystems, 386
temperate forest biomes, 423
spiked watermilfoil, 176
spike rushes, 433
spines, 235
Spinifex hirsutum, 242, 243f
spiny ceanothus, 195
Spionidae, 288t
Spirastrellidae, 288t
Spirogyra spp., 293f
Spirorbidae, 288t
Spirulina platensis, 382
Spirulina spp.
saline/soda lakes, 381, 384
Spodosols, 181, 269, 421
spongeflies, 373
sponges (Porifera)
coral reefs, 205, 211f
exergy calculations, 134t
Florida Everglades mesocosm project, 294f
intertidal environments, 376
mangrove forests, 310
Sporobolus virginicus, 386t
springs, 428t
springtails
agroecosystems, 148
cave habitats, 193
landfills, 305
salt marsh ecosystems, 387
stream ecosystems, 373
Spriofilidae, 288t
Spriorbus spp., 294f
spruce budworm, 182, 269
spruce trees
alpine forests, 158
boreal forest ecosystems, 181
boreal forests, 417
squash, 28
squirrels
boreal forest ecosystems, 182
forest plantations, 268
temperate forest biomes, 423
Sri Lanka
botanic gardens, 186
savanna ecosystems, 394
tropical ecosystems, 441
self-organization, 104
stable isotope analysis, 20
stand-initiating events, 420, 421f
starfishes
competition studies, 83
coral reefs, 204
Staten Island, New York (USA), 466
static models
indirect interactions, 85, 88
See also regression models
St. Croix coral reef
carbonate cycle, 287f
general discussion, 286
gross primary production (GPP), 287f
oxygen (O) concentrations, 286f
physico-chemical parameters, 287t
stem hypertrophy, 258
Stenetriidae, 288t
Stephanopogonidae, 288t
steppe environments, 405–413
Mediterranean ecosystems, 321t, 327
shortgrass prairies, 406
temporary water bodies, 433
Stercorarius spp., 447
Stichodactylidae, 288t
stigmergy, 101, 102f
Stipa tenacissima, 320t
St. John’s wort, 185
St. Louis Declaration on Invasive Plant Species,
189
Stockholm Conference (1972), 14–15
Stockholm, Sweden, 462, 464–465
lagoon ecosystems, 298
stoneflies
stream ecosystems
general discussion, 372–373
predator-prey dynamics, 354, 354t
stone pines, 158
storks, 278
storm surges, 252
stream ecosystems, 363–374
algae
biotic diversity patterns, 359f
characteristics, 370, 370f
energy sources, 351
alkalinity, 369
biota
algae, 370, 370f
benthic macroinvertebrates,
371, 372f
fish, 373
general discussion, 370
macrophytes, 371
channel structure
channelization
desert environments, 221
pools and riffles, 366, 366f
stream order, 366f, 367
structural characteristics, 366
watersheds, 367
chemical components
water chemistry, 369
desert environments, 214–221
anthropogenic impacts, 221
biota, 216, 218t, 229
boundary zones, 214–215
energetics, 218
ephemeral streams, 225, 225f,
428t, 435f
gaining/losing reaches, 215
nutrient dynamics, 219, 220f
physical characteristics, 214
riparian zones, 214, 214f
temporal dynamics
disturbances, 215
drying disturbances, 215–216, 217f
flash floods, 215–216, 217f, 218t, 219f,
225, 226f
flow seasonality, 216
general discussion, 215
interannual/decadal variability,
216, 217f
spatial scales, 215f
successional patterns, 215
temperature variations, 216
dissolved organic matter (DOM), 356
drying disturbances, 215–216, 217f
ecosystem dynamics, 351–362
conservation strategies, 362
energy flow
energy sources, 351, 352f, 353f
functional feeding groups (FFGs), 352, 354f,
354t
matter fluxes
nutrient cycling, 355, 355f
organic matter, 351, 352f, 355
modeling studies
biotic diversity patterns, 359f
ecotones, 359
flood pulse concept, 344, 359t, 360
hierarchical scales, 359, 359t
hydraulic stream ecology, 359t, 361
hyporheic dynamics, 359t, 361
network dynamics, 359t, 361
patch dynamics, 359t, 361
riparian zone influences, 359t, 360, 360f
river continuum concept (RCC), 344, 356,
357f, 358f, 359t
serial discontinuity, 357
organic matter
515
characteristics, 356
energy sources, 351, 352f, 354f
retention potential, 356
sources, 356
transport/storage mechanisms, 355
flash floods, 215–216, 217f, 225, 226f
food webs, 352
hardness measurements, 369
importance, 363
incised streams, 254f
intermittent streams, 364, 365f, 428t, 432f
ionic composition, 369
lotic research, 363
meltwater pools/streams, 428t
nonincised streams, 254f
physico-chemical processes
channel morphology
pools and riffles, 366, 366f
stream order, 366f, 367
watersheds, 367
general discussion, 364
geomorphic processes, 366
hydrologic cycle
characteristics, 364
hydrographs, 364, 365f
intermittent streams, 364, 365f,
428t, 432f
meltwater pools/streams, 428t
schematic diagram, 365f
physcial factors
currents, 367
substrate materials, 367, 367f, 368f, 368t
temperature, 368
water chemistry, 369
sedimentation processes, 366
research background, 363
swamps, 414
temperate forest biomes, 423
tundra ecosystems, 444–445
See also floodplain ecosystems; riparian
systems; river systems
stressors
coral reefs
coral bleaching
anthropogenic impacts, 205, 210f
coral-algal mutualism, 201
significance, 209
diseases, 210
general discussion, 209
overfishing, 205, 205f, 210f, 211, 211f
protective marine reserves, 212, 212f
shifting baseline syndrome, 211
Mediterranean ecosystems, 327
strings, 335, 336f
Strombidiidae, 288t
Struthio camelus syriacus, 240
sturgeon, 373
Suaeda spp., 386t, 387
subalpine habitats
forests, 158
See also riparian systems
Suberitidae, 288t
subsurface water flow, 175, 176, 176f
subtidal environments, See intertidal
environments
subtropical deserts, 222
subtropical forests
chaparral ecosystems, 196
desert stream ecosystems, 215
dune ecosystems, 245
landfills, 307
mangrove forests, 311
natural examples, 193
self-organization, 104
plants, 182
self-organization, 104
temperate forest biomes, 421
tundra ecosystems, 446, 446f
516
Index
subtropical forests (continued )
urban environments, 463t
succulent plants, 229
Sudan, 265f
Sudd, Upper Nile, 259
sugar maples, 420–421, 425f
sugars, See carbohydrates
Sugarscape (computer simulation), 112, 112f,
113f
suitable habitats
implications
fragmented populations, 105, 105f
Sulawesi, 441
sulfur (S)
dimethyl sulfide ((CH3)2S2)
upwelling ecosystems, 454
eco-exergy losses, 136
elemental concentrations, 10t
reduced inorganic sulfur (RIS), 338
stable isotope analysis, 20
sulfates (SO4)
ionic composition, 381
stream ecosystems, 369
microbial processes, 453
sulfate-reducing bacteria
peatlands, 338
upwelling ecosystems, 453, 454
Sumatra, 441
sunflowers, 199
See also riparian systems
surface temperature
caves, 190
surface waters
hyporheic zone, 214–215
nutrient cycling, 220
surgeonfishes, 203–204, 206
See also mortality studies
educational programs, 14–15
technological approaches
green buildings, 14
swallow holes, 429t
swamps, 414–417
Africa, 259
characteristics, 175, 275
definition, 414
ecological functions
general discussion, 415
habitats
aquatic environments, 416
habitat conditions, 416
terrestrial ecosystems, 416
hydrologic functions, 415
water quality, 415
ecosystem services, 416
hardwood forests, 414f
hydrologic characteristics, 414, 414f
largest wetlands, 281
peat formation, 338, 415, 415f
restoration methods, 416
riparian systems, 346f, 347, 415
seasonal pools, 414, 415, 428t
soils, 415, 415f
United States, 262
vegetation, 415
swans, 277
Sweden
alpine forests, 158
botanic gardens, 184t
peatlands, 331
polar ecosystems, 339–342
urban environments, 462
Swiss mountain pine, 158
Switzerland
alpine ecosystems, 152f, 158
Sycamore Creek, Arizona (USA), 216,
217f, 219f
Sylhet River, 433
Syllidae, 288t
Sylvilagus bachmani, 196
Symbiodinium spp., 201
symbiosis
coral-algal mutualism, 201, 205
dune ecosystems, 246
emergence/emergent properties, 95t
fungal interactions
mycorrhizas
dune ecosystems, 246
orientation theory, 124t
yucca plants, 196
synchronization, 101
Synechococcus, 454
definition, 6, 76
Synedra spp., 370f
synusial classification approaches, 441
Syria, 223t
Syrian Desert, 223t
system concepts, 120–128
complex systems, 106–114
basic concepts
decision-making processes, 108
feedback loops, 107–108
general discussion, 106
predictable behaviors, 107
simple systems, 107
weak interactions, 108
characteristics, 111t
components
intelligence, 113
local information, 113
number of agents, 113
ecological complexity, 110
emergence, 109, 110t
modeling studies
Sugarscape, 112, 112f, 113f
Tierra, 102, 111
TRANSIMS, 112
surprise-generating mechanisms
connectivity, 109, 110t
emergence, 109, 110f, 110t
incompatible behaviors,
108–109, 110t
paradoxes, 108, 108f, 110t
unstable systems, 108, 110t
evolutionary development, 121
general discussion, 120
orientation theory
basic orientors, 122, 123f, 124t
complex environments, 121
environment properties, 122, 123f
goal functions, 124t
implicit attractors, 123
influential factors, 124, 125f
modeling studies
animat algorithms, 126
lifestyle development, 126, 127f
orientor properties, 123
stimulus-response relationships, 124, 125f
system organization, 120, 121f
system dynamics models
autocatalysis, 41, 57–58
desert environments
general discussion, 234
nontrophic interactions, 236, 237f
primary production, 234, 234t, 235f
trophic interactions
decomposition processes, 235
general discussion, 235
herbivory, 235
parasites, 236, 236f
predation, 236
environ analysis, 76–81
ascendancy, 57–64
background information, 57
basic concepts, 57
background information, 76
basic concepts
boundary zones, 76
input-output models, 76
partitioning processes, 76–77
community assembly rules, 77
data requirements, 77
energy analysis, 64–75
analysis levels, 65
applications, 71
consumer goods and services/cost of living,
72, 73t, 74f, 75f
dynamic system indicators, 70
energy intensities, 65
energy taxes, 74, 74f, 75f, 75t
food webs, 71, 72f, 73t
indirect interactions, 64
input-output models, 65, 65f, 66f, 68t
nutrient intensities, 67, 67f,
68f, 68t, 69t
path length, 68t, 69, 69t, 72f, 73t
residence time, 68t, 69, 69f, 69t, 72f, 73t
terminology, 66t
trophic position, 68, 68t, 69f, 69t,
72f, 73t
two-compartment steady-state analysis,
65, 65f, 66f
energy intensities
balance equations, 69t
basic concepts, 65
consumer goods and services/cost of living,
72, 73t, 74f, 75f
dynamic system indicators, 72f
embodied energy flows, 66f, 67f
energy taxes, 74, 74f, 75f, 75t
explicit energy flows, 67f
feedback function, 67f, 68f
input-output models, 68t
Russian bog food web, 73t
two-compartment steady-state
analysis, 65f
hierarchy theory, 118–119
methodology, 77, 77f, 78t
network properties
direct interactions, 79, 80t
general discussion, 79
homogenization effects, 79
indirect interactions, 79, 80t
mutualism, 79, 80t
quantitative environ, 80
purpose, 80
sample network, 77, 77f
fundamental laws, 33–38
basic properties, 37
ecosystem theory, 34, 36f, 36t
general discussion, 37
general scientific theories, 34
solar radiation, 36f
theory development, 33
conservation strategies, 25
T
Tachys spp., 193
tadpole shrimp, 433, 434f, 437
taiga
peatlands, 331
See also boreal forests
Tainter, Joseph, 13
Taklamakan, 223t
takyrs, 429t
Talamanca Mountains, 159t, 160f
tallgrass prairies, 406
Tamarix spp., 233, 435f
Tanaidae, 288t
tanaids, 294f
Tanalis cavolini, 294f
tannins, 397
Tansley, Arthur, 7, 16, 17f
Tanypodinae, 372f
Tanzania, 184t, 186
Index
tardigrades, 433, 435
Tasmania
alpine forests, 157, 159t, 160f
intertidal area research, 378
Tawharanui Marine Park, New Zealand, 21
Taxodium spp., 257–258, 262
taxonomy
algae, 370
Tbilisi Conference (1978), 14–15
Teal, J, 16, 388
Tegiticula maculata, 196
Tehaucan Valley, Mexico, 238
Tellinidae, 288t
temperate forests, 417–427
carbon balance, 425
characteristics, 417
climate, 418, 419t
deforestation consequences, 426
disturbances
characteristics, 420
detritus, 421, 421f
soils, 421
structural layers, 420
ecological communities
faunal communities, 422
succession, 421
vegetation communities, 421, 422f
energy flow, 423
evapotranspiration, 423
global distribution, 417, 418f
land cover
historical land cover, 425
present-day land cover, 426
mean annual temperatures, 418
mixed coniferous-deciduous forests, 417–418,
418f, 419t
nutrient cycling, 424, 424f, 425f
physiographic regions, 418, 419t
precipitation, 418
temporary water bodies, 428t, 429t, 432, 432f
water cycle, 423
temperature
alpine forests, 159, 165
desert environments, 214, 218t, 226
dune ecosystems, 244
Mediterranean ecosystems, 319, 321t
salt marsh ecosystems, 391
savanna ecosystems, 396f, 398f
seed germination, 199
stream ecosystems, 368
temperate forests, 418
tropical ecosystems, 439–440, 441
temporary water bodies, 427–439
biota
autecology
community ecology, 437
drying adaptations, 434
physiological ecology, 435, 435f
population dynamics, 436
characteristics, 433, 434f
community ecology
general discussion, 437
interspecific interactions, 437
richness theories, 437, 438t
drying adaptations
avoidance strategies, 435, 435f
diapause, 434, 436, 436t
dormancy, 435
general discussion, 434
population dynamics
bet hedging strategies, 436
complex life histories, 437
dispersal mechanisms, 437
life-history strategies, 436
classification approaches
biomes, 428
general discussion, 427
geographic distribution
boreal/temperate forests, 432
desert environments, 432
general discussion, 431
grasslands, 433
Mediterranean ecosystems, 432
tropical rainforests, 431
tundra ecosystems, 432
geographic location, 428
hydrologic variables
chemical characteristics, 431
flood timing, 429
hydroperiod (flood duration), 430, 431f,
436, 436t
water sources, 429
size, 428
substrate materials, 428
water body types, 428, 428t, 429t
definition, 427
ecological applications
conservation strategies, 438
ecological engineering, 438
vector control, 438
ecosystem ecology, 438
marshes, 277
occurrences, 427
swamps, 414, 415, 428t
tenajas, 429t
Tenente Amaral Stream, Mato Grosso,
Brazil, 347f
teosinte, 238
Terebellidae, 288t
Terfezia spp., 227–228
terminal restriction fragment polymorphism
(T-RFLP) analysis, 169
termites
desert environments, 229, 235
group behaviors, 103
savanna ecosystems, 395–396, 400
terns, 458
terrestrial ecosystems
aquatic-terrestrial interactions, 350
caves, 190, 191f
indirect interactions, 85
landfills, 303–307
biota, 305
characteristics, 303
faunal communities, 307
postclosure uses, 304
restoration methods, 466
soil cover, 304
successional development, 307
vegetation, 305
polar ecosystems, 339–342
swamps, 416
See also plants; riparian systems
territoriality
self-organization, 103
Tertiary
desert flora, 228
Mediterranean ecosystems, 323
tectonic activity, 321
Tethyidae, 288t
Tetillidae, 288t
Tetraclinis, 326
Tetrasphaera spp., 170
Texas (USA)
grasslands, 406–407
savanna ecosystems, 394
Textulariidae, 288t
Thailand
forest plantations, 265f
savanna ecosystems, 394, 397
Thalassia spp., 250
Thalestridae, 288t
517
thalweg line, 366f
Thar Desert, 223t
Thaumatogryllus spp., 193
Thecamoebidae, 288t
Themeda triandra, 396
Theophrastus, 183
Theridion spp., 193
thermodynamic emergence, 92, 96f
thermodynamic laws
emergence/emergent properties, 95t, 118
fundamental ecosystem theory, 34
hierarchy theory, 118
probability calculations, 58
irreversible processes, 34
self-organization, 99
thermodynamic processes, See exergy
Thienemann, August, 6
thorns, 397
thorny devils, 230
Three Gorges, Hubei Province, China, 30
thrushes
boreal forest ecosystems, 182
temperate forest biomes, 423
Tibetan Plateau, 150
ticks, 196
tidal cycles
lagoon ecosystems, 297
tidal wetlands
characteristics, 175
salt marsh ecosystems, 385, 385f
See also estuarine ecosystems; salt marsh
ecosystems
tidepools
physical characteristics, 375
tides, 374
Tien Shan Mountains, 159t, 160f
Tierra (computer simulation), 102, 111
Tigris River, 238
Tijuana Estuary, San Diego, California (USA),
393, 393f
Tilia spp., 262
timberline vegetation zone, 151, 151f, 156, 161f
timber production
forest plantations, 30, 31f, 265
time-lag models, 89
Timpisque River, Costa Rica, 253f
Tipulidae, 372f
tires, 428t
Tirmania spp., 227–228
Tisbidae, 288t
tits
boreal forest ecosystems, 182
toads
temperate forest biomes, 423
temporary water bodies, 433, 437
tolerance
salinity tolerances
desert environments, 435f
faunal communities, 382, 387
mangrove forests, 311
tundra ecosystems, 446, 447
tomatoes
pollination, 28–29
topogenous peatlands, 332
topography
mangrove forests, 308, 309f
tortoises
desert environments, 230, 236, 240
total system throughput (TST)
matrix models, 58
toxicants
freshwater ecosystems, 273
lead (Pb)
salt marsh ecosystems, 390, 392
salt marsh ecosystems, 392
See also allelopathy
518
Index
Toxopneustidae, 288t
toyon, 195
TPBP (antifouling chemical)
salt marsh ecosystems, 392
stream ecosystems, 369
Trachypogon plumosus, 399
Trachypogon spp., 396
tractor ruts, 428t
tradeoff-based theory
desert environments, 237
ecosystem services, 27, 28f, 31f
temporary water bodies, 436
TRANSIMS (computer simulation), 112
transverse dunes, 241
tree frogs
temporary water bodies, 433
treeholes, 429t, 432, 432
treeless landscapes, 334
treeline vegetation zone, 151, 151f, 155, 156,
157f, 161f
trees, See forests
Trichocereus, 320t
Trichocereus, 320t
Trichodesmium spp., 454
Trichoptera
salinity tolerances, 435f
stream ecosystems, 353f, 354, 354t, 372f,
372–373
Trichosphaeridae, 288t
Tridacna spp., 287
Tridacnidae, 288t
Trifolium spp., 246
triggerfishes, 204
Triglochin maritima, 386t
Trinidad, 184t
Trochidae, 288t
trophic cascades
indirect interactions, 82, 83f, 85
trophic levels
desert environments
decomposition processes, 235
general discussion, 235
herbivory, 235
parasites, 236, 236f
predation, 236
ecosystem component interactions, 17, 18f
stream ecosystems
conservation strategies, 362
energy flow
energy sources, 351, 352f, 353f
functional feeding groups (FFGs), 352,
354f, 354t
matter fluxes
nutrient cycling, 355, 355f
organic matter, 351, 355
modeling studies
biotic diversity patterns, 359f
ecotones, 359
flood pulse concept, 344, 359t, 360
hierarchical scales, 359, 359t
hydraulic stream ecology, 359t, 361
hyporheic dynamics, 359t, 361
network dynamics, 359t, 361
patch dynamics, 359t, 361
riparian zone influences, 359t, 360, 360f
river continuum concept (RCC), 344, 356,
357f, 358f, 359t
serial discontinuity, 357
organic matter
characteristics, 356
energy sources, 351, 352f, 354f
retention potential, 356
sources, 356
transport/storage mechanisms, 355
upwelling ecosystems
fish/fisheries, 457
zooplankton, 453–454
two-compartment steady-state energy analysis,
68, 68t, 69f, 69t, 72f, 73t
urban environments, 463t
climate
precipitation, 439–440, 441
temperature, 439–440, 441
forests
primary production, 234t
rainforests, 439–443
biodiversity, 442
canopy characteristics, 442, 442f
characteristics, 439, 440f
climate, 441
conservation strategies, 442, 443f
general discussion, 439
global distribution, 440
soils, 441
structural characteristics, 441, 442f
temporary water bodies, 429t, 431
vegetation classifications, 439–440
temporary water bodies, 429t, 431
grasslands, 407
photographic views, 17f
tundra ecosystems, 445, 445f
trout
saline/soda lakes, 384
Truncatella pulchella, 294f
Tsuga canadensis, 421
Tsuga spp., 159t, 324
tsunamis
estuarine ecosystems, 252
salt marsh ecosystems, 389
tube-building polychaetes
Florida Everglades mesocosm project, 294f
intertidal environments, 376
lagoon ecosystems, 299–300
stream ecosystems, 371
tubers/roots
temporary water bodies, 435
Tubificidae, 372f
tufa towers, 384
tunas
trophic transfer efficiency, 19
tundra ecosystems, 443–449
alpine habitats, 17f
animals, 446
anthropogenic impacts, 449
bumblebees, 448
characteristics, 443
flightless beetles, 448
freeze-thaw cycles, 444
giant lobelias, 449
global warming effects, 449
invasive species, 446
landscapes, 444
musk oxen, 447
primary production, 234t
species diversity, 444
succession, 446, 446f
temporary water bodies, 429t, 432
tropical environments, 445, 445f
vegetation, 445, 445f, 446f
tunicates
intertidal environments, 376
mangrove forests, 310
Tunisia, 223t
Turbellaria spp., 376
Turbinidae, 288t
Turkey
botanic gardens, 183
forest plantations, 265f
temperate forest biomes, 426
turkeys, 423
Turkmenistan, 223t, 429t
turloughs, 429t
Turner, RE, 388
turtles
coral reefs, 211, 211f
marshes, 278
overfishing, 211, 211f
riparian systems, 348
temperate forest biomes, 423
tussock grasses, 153, 153f, 154f
Tyndall Glacier, Mount Kenya, 445f
Typha latifolia, 176
Typha spp.
floodplain ecosystems, 259
marshes, 278
wastewater treatment, 176
Tyrannochthonius spp., 193
U
Ucides occidentalis, 311
Ulex europaeous, 246
Ulex spp., 320t
Ulmus pumila, 475
Ulmus spp.
Europe, 261
shelterbelts/windbreaks, 475t
United States, 262
Ultisols, 269, 395–396, 421
Ulvaceae, 288t
Ulva spp.
exergy calculations, 137
salt marsh ecosystems, 389
Umbellularia, 324
uncertainty
indirect interactions, 89
uncertainty analysis
average mutual information (AMI), 59, 59f
ungulates, 230, 400, 411, 411f
Uniola paniculata, 243f
Uniola spp., 246
United Arab Emirates, 223t
United Kingdom
alpine forests, 159t, 160f
botanic gardens
colonial period, 186
medicinal gardens, 185
Royal Botanical Gardens at Kew,
183, 184f
selected gardens, 184t
dune ecosystems, 244
temporary water bodies, 429t
urban environments, 463
United Nations Convention to Combat
Desertification, 238
United States
alpine ecosystems, 150, 157, 159t, 160f
botanic gardens, 184t
desert environments, 223t
floodplain ecosystems, 257t, 261
forest plantations
economic factors, 265, 265f
influential factors, 266
grasslands, 406
salt marsh vegetation, 386t, 391, 393
temporary water bodies, 429t
urban environments, 462
unstable systems, 108, 110t
upwelling ecosystems, 450–460
characteristics, 451
chlorophyll biomass, 452, 452f
climate
climate change effects, 458, 458f
El Niño Southern Oscillation (ENSO),
459, 459f
long-term trends, 459
eddy systems, 454–455
fish/fisheries
climate change effects, 458, 458f
fish production, 457, 457f
Index
food webs, 455, 455f
general discussion, 455
small pelagic fish, 455, 456f
trophic levels, 457
general discussion, 450
intermittence, 451
occurrences, 450, 450f
open-ocean upwelling systems, 454
oxygen depletion, 453
phytoplankton community structure, 452
primary production, 451
zooplankton, 451, 453
Ural Mountains, 157, 160f
urban environments, 461–467
adaptive management, 466
desert stream ecosystems, 221
ecosystem services, 465, 465t
floodplain ecosystems, 259
restoration methods, 466
urbanization process
basic concepts, 461
ecological effects, 462, 462f, 463t, 464f
gradients, 463–464
habitat quality, 463
human-environment interactions, 462,
463t, 467
international agreements and
conventions, 466
urea
upwelling ecosystems, 451
uric acid
desert environments, 230
Urmia (Iran), 381
Uromyces rumicus, 88
Uronematidae, 288t
Uronychiidae, 288t
Urostylidae, 288t
Ursus americanus, 199
Ursus arctos, 182
Ursus horribilis, 199
Utricularia spp., 41, 42f, 278
Uzbekistan, 223t
V
Vaccinium spp., 420–421
Vaginicolidae, 288t
Vahlkampfiidae, 288t
Valoniaceae, 288t
Vampyrellidae, 288t
varzea floodplains, 262
vasante, 429t
vascular plants
bogs, 333
desert environments, 232, 237
estuarine ecosystems, 249
salt marsh ecosystems, 387, 389, 392
stream ecosystems, 351, 356–357, 359f, 371
vegetal detritus, 255–256
vegetation
alpine ecosystems
adaptive processes, 151, 152f
alpine forests, 156–165
biodiversity, 153, 154f
climatic boundaries, 150, 151f
ecosystem processes, 153
flowering plants, 153, 154f
high-elevation treeline, 151, 152f, 156
nutrient cycling, 153
semidesert environments, 153, 153f
water consumption, 153, 153f
chaparral
biogeographical evolution, 324
characteristics, 326
chaparral ecosystems, 195–200
allelopathy, 197
bare zone, 197f
characteristics, 195, 196f
fire-dependent regeneration, 195–196
invasive species, 200
management strategies, 200
regional variations, 199
seed dispersal, 199
seed germination, 199
succession, 196
vegetation communities, 195
wildfires
characteristics, 197
dominance-diversity curves, 198f
management strategies, 200
postfire recovery rates, 197, 198f
regional variations, 199
seed germination, 199
wind effects, 197, 199
classification systems, 439–440
desert environments, 226, 227f, 228, 229f, 232
desert stream ecosystems, 218
dune ecosystems
characteristics, 241
competition, 245
diseases, 246
dune profile characteristics, 245f
general discussion, 244
pioneer zone, 245
predators, 246
sand movement, 242, 243f
succession and facilitation, 245
symbiotic relationships, 246
environmental impact assessments
United States, 261
floodplain ecosystems
nutrient cycling, 255
physiological adaptations, 257
structural characteristics, 257, 257t
fragmented populations, 105, 105f
grazing systems, 329
indirect interactions, 85
exergy calculations, 137
mangrove forests, 308, 309f
marshes, 275f, 277, 277f
Mediterranean ecosystems
biodiversity patterns, 322, 322t
biogeographical evolution, 323
characteristics, 320t, 326
convergence studies, 325
grazing, 329
rainfall, 321t
temperature, 321t
wildfires, 328
peatlands, 332–333
polar ecosystems, 340
postclosure landfills, 304, 305
riparian zones, 218, 345
salt marsh ecosystems
characteristics, 385, 387
ecological systems, 388
global change effects, 390
global distribution, 386t
invasive species, 391
research value, 392
upland-wetland interface, 391f
savanna ecosystems, 397
shelterbelts/windbreaks, 470
swamps, 415
temperate forest biomes, 421, 422f
tropical ecosystems, 439–440, 441
tundra ecosystems, 445
See also photosynthesis
veldts, 406–407, 433
Venezuela, 396
Venice, Italy, 296–297
Vermetidae, 288t
vernal pool ecosystems
boreal/temperate forests, 432, 432f
characteristics, 429t
marshes, 277, 278
519
Mediterranean ecosystems, 327, 429t, 432
vertebrates
intertidal environments, 376f, 377
mangrove forests, 310
Mediterranean ecosystems, 322, 322t
salinity tolerances, 387
salt marsh ecosystems, 387
stream ecosystems, 373
temporary water bodies, 437
See also animals
Vertisols, 226, 269
vesicular arbuscular mycorrhizae (VAM), 246
vetch, 446f
Viatrix globulifera, 294f
Victoria amazonica, 188, 277f, 277–278
Vietnam
savanna ecosystems, 394, 397
vines, 420
vireos, 423
viruses
estuarine ecosystems, 249, 249f, 251
exergy calculations, 134t
Vitrinellidae, 288t
vleis, 429t, 432
volcanic eruptions
cave formation processes, 190
voles
boreal forest ecosystems, 182
riparian systems, 348
Volvox spp., 436–437
Vorticellidae, 288t
Vulpes vulpes, 83–84, 84f, 182
vultures
temperate forest biomes, 423
W
Wadden Sea, 251
Wadi Ram, Jordan, 224f
wadis, 225
warblers
boreal forest ecosystems, 182
riparian systems, 348
temperate forest biomes, 423
Warnstorfia spp., 335
washes, 225
Washington, DC (USA), 407
Washingtonia filifera, 233
wasps
chaparral ecosystems, 196
exergy calculations, 134t
stream ecosystems, 373
waste products
eco-exergy losses, 136, 136
salt marsh ecosystems, 392
wastewater/wastewater treatment
biological wastewater treatment systems,
166–180
basic concepts, 168f, 176
benefits, 179
biological processes, 177
continuous flow systems, 177, 177f
modeling studies, 179
nitrogen removal, 178
nutrient removal capacity, 178
phosphorus removal, 179
background information, 166
biological processes
analytical techniques, 169
cellular organisms, 166, 167t
chemotrophs, 167t
enzymatic reactions, 173
flow models, 173f
kinetic mechanisms, 172, 173f
microbial carbon processes, 169, 169f
microbial communities, 168, 168f, 177
microbial phosphorus processes, 170, 170f
nitrogen transformation processes,
170, 171f
520
Index
wastewater/wastewater treatment (continued )
potential denitification activity (PDA)
measurements, 174, 175f
respiration, 169, 169f
water circulation, 172, 173f
constructed wetlands
classification, 175
free water surface flow (FSW), 175
subsurface water flow, 175, 176, 176f
future perspectives, 180
constructed wetlands
classification, 175
emergent macrophytes
functionality, 175
biochemical oxygen demand (BOD5), 176
subsurface water flow, 175, 176, 176f
freshwater lakes, 272
urban environments, 465t
See also sewage/sewage treatment
Everglades, Florida (USA), 277
marsh ecosystems
floodplains, 276, 276f
general discussion, 275
peatlands, 276–277
seepage wetlands, 277
temporary wetlands, 277
water level changes, 277, 277f
peat/peatlands, 276–277
waterbucks, 259–260
water bugs, 437
waterfleas
temporary water bodies, 433, 436–437
See also Daphnia spp.
waterfowl
aquatic wetlands, 275, 277
Asian floodplain ecosystems, 260
Australian floodplain ecosystems, 261
saline/soda lakes, 384
salt marsh ecosystems, 388
upwelling ecosystems, 457
water (H2O)
dune ecosystems, 243
estuarine ecosystems, 248
United States, 261
forest plantations, 267
freeze-thaw cycles, 444
greenhouse ecosystems, 283, 284f
hardness measurements, 369
desert environments, 237
historical background, 237
marshes
floodplains, 276, 276f
general discussion, 275
peatlands, 276–277
seepage wetlands, 277
temporary wetlands, 277
water level changes, 277, 277f
peatlands, 331, 332f
savanna ecosystems, 398, 398f
turnover time
characteristics, 364
evapotranspiration
schematic diagram, 365f
hydrographs, 364, 365f
intermittent streams, 364, 365f, 428t, 432f
mangrove forests, 308, 309f
meltwater pools/streams, 428t
runoff, 364, 365f
savanna ecosystems, 394
schematic diagram, 365f
temperate forest biomes, 423
water quality
forest plantations, 268
swamps, 415
urban environments, 465, 465t
water-holding depressions, 428t, 429t
water hyacinths
freshwater lakes, 273
water-level changes in freshwater lakes, 273
water lilies, 188, 259–260, 275,
277–278
waterlogged systems
boreal forests, 181
water mites
temporary water bodies, 433, 437
watersheds
forest ecosystems, 30
stream ecosystems
characteristics, 367
hydrologic cycle, 364, 365f
water treaders, 193
weasels, 182
savanna ecosystems, 395–396
Wedelia prostrata, 243f
weeds
characteristics, 146
Wells, HG, 13
Welwitchia of the Namib, 229
Welwitschia mirabilis, 229
Wentworth scale, 368t
West Africa, 313–314, 440
West Siberian Lowland, 275, 276f, 281
Africa, 259
aquatic wetlands, 275
Asia, 260–261
classification, 274
constructed wetlands
emergent macrophytes
functionality, 175
biochemical oxygen demand (BOD5), 176
subsurface water flow, 175
wastewater treatment, 175
desert stream ecosystems, 218
ecosystem services, 29
estuarine ecosystems, 249
forested wetlands
mangrove forests, 308–318
biodiversity, 310
characteristics, 308
coral reefs, 207, 208f
ecogeomorphology, 308, 309f, 310f
environmental impacts, 313
estuarine ecosystems, 250
food webs, 310, 312
gradients, 308, 309f, 310f
greenhouse ecosystems, 284f
hydroperiod (flood duration), 308,
309f, 310f
management strategies, 314, 314f
nutrient resources, 311, 312f
photographic views, 17f
productivity, 311
restoration methods, 314
succession, 311
swamps, 414–417
Africa, 259
aquatic environments, 416
characteristics, 175, 275
definition, 414
ecological functions, 415
ecosystem services, 416
habitat conditions, 416
hardwood forests, 414f
hydrologic characteristics, 414, 414f
hydrologic functions, 415
largest wetlands, 281
peat formation, 338
restoration methods, 416
riparian systems, 346f, 347
seasonal pools, 428t
soils, 415, 415f
terrestrial ecosystems, 416
United States, 262
vegetation, 415
water quality, 415
largest wetlands, 281
marshes, 274–281
anthropogenic impacts
dam construction, 279
draining effects, 279
food webs, 280
general discussion, 279
nutrient resources, 279–280
road networks, 280
biodiversity, 278, 279f
characteristics, 275, 275f, 280
environmental factors
disturbances, 278
nutrient cycling, 278
general discussion, 274
geographic distribution, 278
restoration methods, 280
seasonal pools, 428t
vegetation, 275f, 277, 277f
water availability
floodplains, 276, 276f
general discussion, 275
peatlands, 276–277
seepage wetlands, 277
temporary wetlands, 277
water level changes, 277, 277f
West Siberian Lowland, 276f
peat/peatlands, 330–339
bogs
characteristics, 333
continental bogs, 334, 334f, 335f
development factors, 332f
oceanic bogs, 334, 334f, 338
carbon sinks, 330, 332f, 338
characteristics, 330
classification, 332, 333f
environmental factors, 331, 332f
fens
characteristics, 333, 334
development factors, 332f
patterned fens, 335, 336f
poor fens, 335
rich fens, 335, 335f
formation processes, 338
global distribution, 331, 331f
key processes
acidification, 335
methanogenesis, 337
nutrient sequestration, 337
sulfate reduction, 338
water retention, 336
occurrences, 331
peat moss, 330, 333f, 336
tundra ecosystems, 444–445
postclosure landfills, 304, 307
riparian zones, 218
seasonal pools, 428t
wet meadows, 275
See also estuarine ecosystems; floodplain
ecosystems; freshwater ecosystems;
riparian systems; salt marsh ecosystems;
swamps
wet meadows, 275
whales
upwelling ecosystems, 451, 457
whale wallows, 429t
whispering bells, 199
white band disease, 210
whitebark pine, 158
white mangrove, 285f
whitewater rivers, 262
Wiener, Norbert, 99
wild ass, 239
wild dogs
anthropogenic impacts, 399
boreal forest ecosystems, 182
chaparral ecosystems
characteristics, 197
dominance-diversity curves, 198f
Index
management strategies, 200
postfire recovery rates, 197, 198f
regional variations, 199
seed germination, 199
grasslands, 408, 409f
lightning, 182, 399
marshes, 278
Mediterranean ecosystems, 320t, 328
peatlands, 278, 338
savanna ecosystems, 394, 398f,
399f, 401t
boreal forest ecosystems, 182
floodplain ecosystems, 276
marshes, 278, 279f
tundra ecosystems, 446
See also animals
Williams, WD, 382
willows, 181, 445, 445f
wind
alpine forests, 159, 165
dune ecosystems, 242, 245f
El Niño Southern Oscillation (ENSO), 459f
grasslands, 407
ocean currents, 450
shelterbelts/windbreaks, 468–477
characteristics, 468
design principles, 474, 475f, 475t
farmland shelterbelts, 468f
management strategies, 476
mixing patterns, 475t
optical porosity determination
error analysis, 473
general discussion, 472
image processing, 472
photographic techniques, 472
structural characteristics
aerodynamic parameters, 471
cross-sectional shape, 471, 473f
external structural characteristics, 470
general discussion, 470
internal structural characteristics, 470
length effects, 471f
width effects, 472f, 472t
wind profiles, 474, 474f
wind-tree interactions
benefits, 469
ecological effects, 469
tree effects, 469
wind effects, 469
wind parameters, 468, 469f
tundra ecosystems, 445
upwelling ecosystems, 450
chaparral ecosystems, 197, 199
ecological effects, 469
lagoon ecosystems, 297
windbreaks, See shelterbelts/windbreaks; wind
wolf spiders, 193
wolves
boreal forest ecosystems, 182
polar ecosystems, 340–341
tundra ecosystems, 446–447
wood
forest plantations, 266
timber production, 265
See also forests
woodlands
Mediterranean ecosystems, 320t, 321t, 323
postclosure landfills, 306, 307
seasonal pools, 428t, 429t, 432, 432f, 434f
woodpeckers
boreal forest ecosystems, 182
temperate forest biomes, 423
woodrats, 196
woody detritus, 421, 421f
World Conservation Union, 188
worm lizards, 230
worms
estuarine ecosystems, 250
Florida Everglades mesocosm project, 294f
stream ecosystems, 371, 372f
temporary water bodies, 433
See also polychaetes (Polychaeta); tubebuilding polychaetes
wrasses, 204, 206
Wurdemanniaceae, 288t
Wyoming (USA), 157f, 406–407
X
Xanthidae, 288t
xeric environments, 411
xylem, 423
521
Y
yaks, 411
Yangtze River
floodplain ecosystems, 260–261
runoff, 30
yeasts
exergy calculations, 134t
yellow-green algae, 370
Yellow River, 260–261
Yellowstone National Park (USA), 357f
Yemen, 223t
Yokohama, Japan, 464–465
Yucca brevifolia, 229
yucca moths, 196
yucca plants
chaparral ecosystems, 196
desert environments, 230
Yucca whipplei, 196
Z
Zaire peacocks, 259–260
Zaire Swamps, 259
Zea mays, 238
See also corn; maize
zebra mussels, 273
zebras, 259–260
zinc (Zn)
eco-exergy losses, 136t
stream ecosystems, 369
Zoanthidae, 288t
zooplankton
competition studies, 83
estuarine ecosystems, 248
indirect interactions, 85
NPZD (nitrogen-phytoplankton-zooplanktondetritus) model, 299
river system dynamics, 358f
saline/soda lakes, 382
stable isotope analysis, 21
upwelling ecosystems, 451,
453, 455f
Zooxanthellaceae, 288t
zooxanthellae, 201
Zostera spp., 137, 250
Zoysia sinica, 386t
Zygoptera, 372f
This page intentionally left blank
This page intentionally left blank
This page intentionally left blank