Principles of Ecosystem Stewardship

F. Stuart Chapin, IIIGary P. KofinasCarl Folke

Principles ofEcosystemStewardshipResilience-Based NaturalResource Managementin a Changing World

Illustrated by Melissa C. Chapin

123

EditorsF. Stuart Chapin, III Gary P. KofinasDepartment of Biology and Wildlife Institute of Arctic BiologyInstitute of Arctic Biology University of Alaska, FairbanksUniversity of Alaska Fairbanks AK 99775Fairbanks AK 99775 311 Irving Bldg.USA [email protected] [email protected]

Carl FolkeCtr. Research on Natural

Resources and the EnvironmentStockholm UniversitySE-106 91 [email protected]

ISBN 978-0-387-73032-5 e-ISBN 978-0-387-73033-2DOI 10.1007/978-0-387-73033-2

Library of Congress Control Number: 2008942724

c© Springer Science+Business Media, LLC 2009All rights reserved. This work may not be translated or copied in whole or in partwithout the written permission of the publisher (Springer Science+Business Media,LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts inconnection with reviews or scholarly analysis. Use in connection with any form ofinformation storage and retrieval, electronic adaptation, computer software, or bysimilar or dissimilar methodology now known or hereafter developed is forbidden.The use in this publication of trade names, trademarks, service marks, and similarterms, even if they are not identified as such, is not to be taken as an expression ofopinion as to whether or not they are subject to proprietary rights.

Printed on acid-free paper

springer.com

Preface

The world is undergoing unprecedented changes in many of thefactors that determine its fundamental properties and their influ-ence on society. These changes include climate; the chemical com-position of the atmosphere; the demands of a growing humanpopulation for food and fiber; and the mobility of organisms, indus-trial products, cultural perspectives, and information flows. Themagnitude and widespread nature of these changes pose seriouschallenges in managing the ecosystem services on which societydepends. Moreover, many of these changes are strongly influencedby human activities, so future patterns of change will continue to beinfluenced by society’s choices and governance.

The purpose of this book is to provide a new framework for nat-ural resource management—a framework based on stewardship ofecosystems for human well-being in a world dominated by uncer-tainty and change. The goal of ecosystem stewardship is to respondto and shape change in social-ecological systems in order to sus-tain the supply and opportunities for use of ecosystem services bysociety. The book links recent advances in the theory of resilience,sustainability, and vulnerability with practical issues of ecosystemmanagement and governance. The book is aimed at advancedundergraduates and beginning graduate students of naturalresource management as well as professional managers, communityleaders, and policy makers with backgrounds in a wide array of dis-ciplines, including ecology, policy studies, economics, sociology, andanthropology.

The first part of the book presents a conceptual frameworkfor understanding the fundamental interactions and processes insocial—ecological systems—systems in which people interact withtheir physical and biological environment. We explain how thesesystems respond to variability and change and discuss many ofthe ecological, economic, cultural, and institutional processes thatcontribute to these dynamics, enabling society to respond to andshape change. In the second section we apply this theory to specifictypes of social—ecological systems, showing how people adaptively

v

vi Preface

manage resources and ecosystem services throughout the world.Finally we synthesize the lessons learned about resilience—basedecosystem stewardship as a strategy for responding to and shap-ing change in a rapidly changing world. Change brings both chal-lenges and opportunities for managers, resource users, and policymakers to make informed decisions that enhance sustainability ofour planet.

We owe a huge debt of gratitude to Buzz Holling who origi-nated many of the central concepts that link resilience to ecosys-tem stewardship, as well as to several national and interna-tional programs that have developed these ideas and appliedthem to education and to the real-world issues faced by arapidly changing planet. These include the Resilience Network,the Resilience Alliance, the International Geosphere-BiosphereProgramme, the Millennium Ecosystem Assessment, the Stock-holm Resilience Centre, the Beijer Institute, and the Resilienceand Adaptation Program of the University of Alaska Fairbanks.Primary funding for the book came from the US National Sci-ence Foundation and the Swedish Research Council FORMASprogram. In addition, many individuals contributed to the devel-opment of this book. We particularly thank our families, whosepatience made the book possible, and our students, from whomwe learned many of the concepts and applications presented inthis book. In addition, we thank the following people for theirconstructively critical review of chapters in this book: MartyAnderies, Erik Anderson, Archana Bali, David Battisti, HarryBiggs, Oonsie Biggs, Steve Carpenter, Melissa Chapin, JohannColding, Graeme Cumming, Bill Dietrich, Logan Egan, ThomasElmqvist, Walter Falcon, Victor Galaz, Ted Gragson, NancyGrimm, Lance Gunderson, Susan Herman, Buzz Holling, JordanLewis, Chanda Meek, Joanna Nelson, Evelyn Pinkerton, CiaraRaudsepp-Hearne, Marten Scheffer, Emily Springer, SamanthaStaley, Will Steffen, Fred Swanson, Brian Walker, Karen Wang,and Oran Young. We particularly thank Steve Carpenter for histhoughtful comments on most of the chapters in this book.

Fairbanks, AK, USA F. Stuart Chapin, IIIFairbanks, AK, USA Gary P. KofinasStockholm, Sweden Carl Folke

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Part I Conceptual Framework

Chapter 1A Framework for Understanding Change . . . . . . . . . . . 3F. Stuart Chapin, III, Carl Folke, and Gary P. Kofinas

Chapter 2Managing Ecosystems Sustainably: The Key Roleof Resilience . . . . . . . . . . . . . . . . . . . . . . . . . . . 29F. Stuart Chapin, III

Chapter 3Sustaining Livelihoods and Human Well-Beingduring Social–Ecological Change . . . . . . . . . . . . . . . 55Gary P. Kofinas and F. Stuart Chapin, III

Chapter 4Adaptive Co-management in Social–EcologicalGovernance . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Gary P. Kofinas

Chapter 5Transformations in Ecosystem Stewardship . . . . . . . . . . 103Carl Folke, F. Stuart Chapin, III, and Per Olsson

Part II Stewarding Ecosystems for Society

Chapter 6Conservation, Community, and Livelihoods:Sustaining, Renewing, and Adapting CulturalConnections to the Land . . . . . . . . . . . . . . . . . . . . 129Fikret Berkes, Gary P. Kofinas,and F. Stuart Chapin, III

vii

viii Contents

Chapter 7Forest Systems: Living with Long-Term Change . . . . . . . 149Frederick J. Swanson and F. Stuart Chapin, III

Chapter 8Drylands: Coping with Uncertainty, Thresholds,and Changes in State . . . . . . . . . . . . . . . . . . . . . . 171D. Mark Stafford Smith, Nick Abel, Brian Walker,and F. Stuart Chapin, III

Chapter 9Freshwaters: Managing Across Scales in Space and Time . . 197Stephen R. Carpenter and Reinette Biggs

Chapter 10Oceans and Estuaries: Managing the Commons . . . . . . . 221Carl Walters and Robert Ahrens

Chapter 11Coastal Marine Systems: Conserving Fishand Sustaining Community Livelihoodswith Co-management . . . . . . . . . . . . . . . . . . . . . . 241Evelyn Pinkerton

Chapter 12Managing Food Production Systems for Resilience . . . . . 259Rosamond L. Naylor

Chapter 13Cities: Managing Densely Settled Social–EcologicalSystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281J. Morgan Grove

Chapter 14The Earth System: Sustaining Planetary Life-SupportSystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295Oran R. Young and Will Steffen

Part III Integration and Synthesis

Chapter 15Resilience-Based Stewardship: Strategies forNavigating Sustainable Pathways in a Changing World . . . 319F. Stuart Chapin, III, Gary P. Kofinas, Carl Folke,Stephen R. Carpenter, Per Olsson, Nick Abel,Reinette Biggs, Rosamond L. Naylor, EvelynPinkerton, D. Mark Stafford Smith, Will Steffen,Brian Walker, and Oran R. Young

Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . 339

Contents ix

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389

Contributors

Nick Abel CSIRO Sustainable Ecosystems, GPO Box 284,Canberra ACT 2602, Australia, [email protected]

Robert Ahrens Fisheries Centre, University of British Columbia,Vancouver, BC, Canada V6T 1Z4, [email protected]

Fikret Berkes Natural Resources Institute, University ofManitoba, Winnipeg, Manitoba, Canada R3T 2N2,[email protected]

Reinette Biggs Center for Limnology, University of Wisconsin,Madison, WI 53706, USA, [email protected]

Stephen R. Carpenter Center for Limnology, University ofWisconsin, Madison, WI 53706, USA, [email protected]

F. Stuart Chapin, III Institute of Arctic Biology, Universityof Alaska Fairbanks, Fairbanks, AK 99775, USA,[email protected]

Carl Folke Stockholm Resilience Centre, Stockholm University,SE-106 91 Stockholm, Sweden; Beijer Institute of EcologicalEconomics, Royal Swedish, Academy of Sciences, PO Box 50005,SE 104 05 Stockholm, Sweden, [email protected]

J. Morgan Grove Northern Research Station, USDA ForestService, Burlington, VT 05403, USA, [email protected]

Gary P. Kofinas Institute of Arctic Biology, School of NaturalResources and Agricultural Sciences, University of AlaskaFairbanks, Fairbanks, AK 99775, USA, [email protected]

Rosamond L. Naylor Woods Institute for the Environment,Freeman-Spogli Institute for International Studies StanfordUniversity, Stanford, CA 94305, USA, [email protected]

Per Olsson Stockholm Resilience Centre, Stockholm University,SE-106 91 Stockholm, Sweden,[email protected]

xi

xii Contributors

Evelyn Pinkerton School of Resource and EnvironmentalManagement, Simon Fraser University, Burnaby, BC V5A 1S6,Canada, [email protected]

D. Mark Stafford Smith CSIRO Sustainable Ecosystems, PO Box284, Canberra ACT 2602, Australia, [email protected]

Will Steffen Fenner School of Environment and Society,Australian National University, Canberra, ACT 0200, Australia,[email protected]

Frederick J. Swanson Pacific Northwest Research Station, USDAForest Service, Corvallis, OR 97331, USA, [email protected]

Brian Walker CSIRO Sustainable Ecosystems, GPO Box 284,Canberra ACT 2602, Australia, [email protected]

Carl Walters Fisheries Centre, University of British Columbia,Vancouver, BC, Canada V6T 1Z4, [email protected]

Oran R. Young Bren School of Environmental Science andManagement, University of California, Santa Barbara, CA93106-5131, USA, [email protected]

Part IConceptual Framework

1A Framework for UnderstandingChange

F. Stuart Chapin, III, Carl Folke, and Gary P. Kofinas

Introduction

The world is undergoing unprecedentedchanges in many of the factors that deter-mine both its fundamental properties andtheir influence on society. Throughout humanhistory, people have interacted with andshaped ecosystems for social and economicdevelopment (Turner et al. 1990, Redman1999, Jackson 2001, Diamond 2005). Duringthe last 50 years, however, human activitieshave changed ecosystems more rapidly andextensively than at any comparable period ofhuman history (Steffen et al. 2004, Foley et al.2005, MEA 2005d; Plate 1). Earth’s climate,for example, is now warmer than at any timein the last 500 (and probably the last 1,300)years (IPCC 2007a), in part because of atmo-spheric accumulation of carbon dioxide (CO2)released by the burning of fossil fuels (Fig. 1.1).Agricultural development largely accountsfor the accumulation of other trace gases that

F.S. Chapin, III (�)Institute of Arctic Biology, University of AlaskaFairbanks, Fairbanks, AK 99775, USAe-mail: [email protected]

contribute to climate warming (see Chapter 12).As human population increases, in part dueto improved disease prevention, the increaseddemand for food and natural resources hasled to an expansion of agriculture, forestry,and other human activities, causing large-scaleland-cover change and loss of habitats andbiological diversity. About half the world’spopulation now lives in cities and depends onconnections with rural areas worldwide forfood, water, and waste processing (see Chap-ter 13; Plate 2). In addition, increased humanmobility is spreading plants, animals, diseases,industrial products, and cultural perspectivesmore rapidly than ever before. This increasein global mobility, coupled with increasedconnectivity through global markets and newforms of communication, links the world’seconomies and cultures, so decisions in oneplace often have international consequences.

This globalization of economy, culture, andecology is important because it modifies thelife-support system of the planet (Odum 1989),i.e., the capacity of the planet to meet the needsof all organisms, including people. The dramaticincrease in the extinction rate of species (100-to 1,000-fold in the last two centuries) indicatesthat global changes have been catastrophicfor many species, although some species,

3F.S. Chapin et al. (eds.), Principles of Ecosystem Stewardship,DOI 10.1007/978-0-387-73033-2 1, © Springer Science+Business Media, LLC 2009

4 F.S. Chapin et al.

Globalpopulation

Peo

ple

(bill

ion)

0

2

4

6

0

Water use

(10

00 k

m)3

yr –1

2

4

6

1800

1750

1850

1900

1950

2000

Northern hemisphere temperature

Tem

p. a

nom

aly

( o C

)

–0.5

0

0.5

1.0

Land conversion

% o

f lan

d ar

ea

0

25

50

1800

1750

1850

1900

1950

2000

% o

f fis

herie

s

0

20

40

60

80

100

Fisheries exploited

Year

Year

Year

Year

1800

1750

1850

1900

1950

2000

1800

1750

1850

1900

1950

2000

Species extinctions

0

# (1

03 )

10

20

30

Human well-being ?

Ecosystemintegrity

? Challenges to ecosystemstewardship

Human drivers

Human impacts

Feedbacksto people

Ecosystemconsequences

Figure 1.1. Challenges to research stewardship.Changes in human population and resource con-sumption alter climate and land cover, which haveimportant ecosystem consequences such as speciesextinctions and overexploitation of fisheries. These

changes reduce ecosystem integrity and have region-ally variable effects on human well-being, whichfeeds back to further changes in human drivers. Panelinserts redrawn from Steffen et al. (2004).

especially invasive species and some diseaseorganisms, have benefited and expanded theirranges. Human society has both benefited andsuffered from global changes, with increased

food production, increased income and livingstandards (in parts of the world), improvedtreatment of many diseases, and longer lifeexpectancy being offset by deterioration in

1 A Framework for Understanding Change 5

ecosystem services, the benefits that societyreceives from ecosystems. More than halfof the ecosystem services on which societydepends for survival and a good life havebeen degraded—not deliberately, but inadver-tently as people seek to meet their materialdesires and needs (MEA 2005d). Change cre-ates both challenges and opportunities. Peoplehave amply demonstrated their capacity to alterthe life-support system of the planet. In thisbook we argue that, with appropriate steward-ship, this human capacity can be mobilized tonot only repair but also enhance the capacity ofEarth’s life-support system to support societaldevelopment.

The unique feature of the changes describedabove is that they are directional. In otherwords, they show a persistent trend over time(Fig. 1.1). Many of these trends have becomemore pronounced since the mid-twentieth cen-tury and will probably continue or acceler-ate in the coming decades, even if societytakes concerted actions to reduce some ratesof change. This situation creates a dilemmain planning for the future because we cannotassume that the future world will behave aswe have known it in the past or that our pastexperience provides an adequate basis to plan

for the future. This issue is especially acute forsustainable management of natural resources.It is no longer possible to manage systems sothey will remain the same as in the recentpast, which has traditionally been the referencepoint for resource managers and conservation-ists. We must adopt a more flexible approach tomanaging resources—management to sustainthe functional properties of systems that areimportant to society under conditions wherethe system itself is constantly changing. Man-aging resources to foster resilience—to respondto and shape change in ways that both sustainand develop the same fundamental function,structure, identity, and feedbacks—seems cru-cial to the future of humanity and the Earth Sys-tem. Resilience-based ecosystem stewardship isa fundamental shift from steady-state resourcemanagement, which attempted to reduce vari-ability and prevent change, rather than torespond to and shape change in ways that bene-fit society (Table 1.1). We emphasize resilience,a concept that embraces change as a basic fea-ture of the way the world works and devel-ops, and therefore is especially appropriate attimes when changes are a prominent featureof the system. We address ecosystems that pro-vide a suite of ecosystem services rather than a

Table 1.1. Contrasts between steady-state resource management, ecosystem management, and resilience-based ecosystem stewardship.

Resilience-based ecosystemSteady-state resource management Ecosystem management stewardship

Reference state: historic condition Historic condition Trajectory of changeManage for a single resource or

speciesManage for multiple ecosystem

servicesManage for fundamental

social–ecological propertiesSingle equilibrium state whose

properties can be sustainedMultiple potential states Multiple potential states

Reduce variability Accept historical range of variability Foster variability and diversityPrevent natural disturbances Accept natural disturbances Foster disturbances that sustain

social–ecological propertiesPeople use ecosystems People are part of the

social–ecological systemPeople have responsibility to sustain

future optionsManagers define the primary use of

the managed systemMultiple stakeholders work with

managers to define goalsMultiple stakeholders work with

managers to define goalsMaximize sustained yield and

economic efficiencyManage for multiple uses despite

reduced efficiencyMaximize flexibility of future options

Management structure protectscurrent management goals

Management goals respond tochanging human values

Management responds to and shapeshuman values


single resource such as fish or trees. We focus onstewardship, which recognizes managers as anintegral component of the system that theymanage. Stewardship also implies a sense ofresponsibility for the state of the system ofwhich we are a part (Leopold 1949). The chal-lenge is to anticipate change and shape it forsustainability in a manner that does not leadto loss of future options (Folke et al. 2003).Ecosystem stewardship recognizes that soci-ety’s use of resources must be compatible withthe capacity of ecosystems to provide services,which, in turn, is constrained by the life-supportsystem of the planet (Fig. 1.2).

This chapter introduces a framework forunderstanding and managing resources in aworld where persistent directional changes arebecoming more pronounced. We first presenta framework for studying change—one thatintegrates the physical, ecological, and socialdimensions of change and their interactions. Wethen describe the general properties of systemsthat magnify or resist change. Finally we discussgeneral approaches to sustaining desirable sys-tem properties in a directionally changing worldand present a road map to the remaining chap-ters, which address these issues in greater depth.

Earth’s life support system

Human societies

EconomiesSust

ain

abili

ty

Figure 1.2. Social–ecological sustainability requiresthat society’s economy and other human activitiesnot exceed the capacity of ecosystems to provide ser-vices, which, in turn, is constrained by the planet’slife-support system. Redrawn from Fischer et al.(2007).

An Integrated Social–EcologicalFramework

Linking Physical, Ecological,and Social Processes

Changes in the Earth System are highlyinterconnected. None of the changes men-tioned above is purely physical, ecological, orsocial. Therefore understanding current andfuture change requires a broad interdisciplinaryframework that draws on the concepts andapproaches of many natural and social sciences.We must understand the world, region, orcommunity as a social–ecological system (alsotermed a coupled human–environment system)in which people depend on resources and ser-vices provided by ecosystems, and ecosystemdynamics are influenced, to varying degrees, byhuman activities (Berkes et al. 2003, Turner etal. 2003, Steffen et al. 2004). Although the rel-ative importance of social and ecological pro-cesses may vary from forests to farms to cities,the functioning of each of these systems, andof the larger regional system in which they areembedded, is strongly influenced by physical,ecological, economic, and cultural factors. Theyare, therefore, best viewed, not as ecological orsocial systems, but as social–ecological systemsthat reflect the interactions of physical, ecologi-cal, and social processes.

Forests, for example, are sometimes man-aged as ecological systems in which the nitrogeninputs from acid rain or the economic influenceson timber demand are considered exogenousfactors (i.e., factors external to the system beingmanaged) and therefore are not incorporatedinto management planning. Production of lum-ber or paper, on the other hand, is often man-aged as an economic system that must balancethe supply and costs of timber inputs against thedemand for and profits from products withoutconsidering ecological influences on forest pro-duction. Finally, local planners make decisionsabout school budgets and the zoning for devel-opment and recreation, based on assumptionsabout regional water supply, which depends onforest cover, and economic projections, whichare influenced by the economic activity of forestindustries. The system and its components are


more vulnerable to unexpected changes (sur-prises) when each subsystem is managed in iso-lation. These surprises might include harvestrestrictions to protect an endangered species,development of inexpensive lumber supplies onanother continent, or expansion of recreationaldemand for forest use by nearby urban resi-dents. More informed decisions are likely toemerge from integrated approaches that rec-ognize the interdependencies of regional com-

ponents and account for uncertainty in futureconditions (Ludwig et al. 2001). Resource stew-ardship policies must therefore be ecologically,economically, and culturally viable, if they areto provide sustainable solutions.

In studying the response of social–ecologicalsystems to directional change, we pay par-ticular attention to the processes that linkecological and social components (Fig. 1.3).The environment affects people through both

Ecological properties

Climate,regional

biota,etc.

Exogenous controls

Exogenous controls

Social properties

Slowvariables

Fast variables Fast variables

Tem

pora

l sca

le

Ecosystem services Human actorsEnvironmental impacts

Socialimpacts

Inst

itu

tio

nal

resp

on

ses

Social-ecologicalSystem

Regionalgovernance

systems,regional economy,

etc.

Soil resources,functional types,

disturbanceregime,

etc.

Soil nitrate,deer density,

fire event,etc.

Communityincome,

population density,

access to resources,

etc

Wealth andinfrastructure,

cultural tiesto the land,

etc.

Slowvariables

Spatial scale

Globe

Figure 1.3. Diagram of a social–ecological system(the rectangle) that is affected by ecological (left-hand side) and social properties (right-hand side). Inboth subsystems there is a spectrum of controls thatoperate across a range of temporal and spatial scales.At the regional scale exogenous controls respond toglobal trends and affect slow variables at the scaleof management, which, in turn, influence fast vari-ables that change more quickly. When changes in fastvariables persist over long time periods and large

areas, these effects cumulatively propagate upwardto affect slow variables, regional controls, and even-tually the entire globe. Changes in both slow and fastvariables influence environmental impacts, ecosys-tem services, and social impacts, which, together,are the factors that directly affect the well-being ofhuman actors, who modify both ecological and socialsystems through a variety of institutions. Modifiedfrom Chapin et al. (2006a).


direct environmental events such as floodsand droughts and ecosystem services suchas food and water quality (see Chapter 2).Many economic, political, and cultural pro-cesses also shape human responses to the physi-cal and biological environment (see Chapter 3).Human actors (both individuals and groups)in turn affect their ecological environmentthrough a complex web of social processes (seeChapter 4). Together these linkages betweensocial and ecological processes structure thedynamics of social–ecological systems (seeChapter 5).

The concept that society and nature dependon one another is not new. It was wellrecognized by ancient Greek philosophers(Boudouris and Kalimtzis 1999); economistsconcerned with the environmental constraintson human population growth (Malthus 1798);geographers and anthropologists seeking tounderstand global patterns of land use andculture (Rappaport 1967, Butzer 1980); andecologists and conservationists concerned withhuman impacts on the environment (Leopold1949, Carson 1962, Odum 1989). The complex-ity and importance of social–ecological inter-actions has led many natural and social sci-ence disciplines to address components of theinteraction to both improve understanding andsolve problems. For example, resource man-

agement considers the actions that agencies orindividuals take to sustain natural resources,but typically pays less attention to the inter-actions among interest groups that influencehow management policies develop or how thepublic will respond to management. Similarly,environmental policy analysis addresses thepotential interactions of environmental policiesdeveloped by different organizations, but typ-ically pays less attention to potential social orecological thresholds (critical levels of driversor state variables that, when crossed, triggerabrupt changes or regime shifts) that determinethe long-term effectiveness of these policies.The breadth of approaches provides a wealth oftools for studying integrated social–ecologicalsystems. Disciplinary differences in vocabu-lary, methodology, and standards of what con-stitutes academic rigor can, however, createbarriers to communication (Box 1.1; Wilson1998). The increasing recognition that humanactions are threatening Earth’s life-support sys-tem has recently generated a sense of urgency inaddressing social–ecological systems in a moreintegrated fashion (Berkes et al. 2003, Clarkand Dickson 2003, MEA 2005d). This requiresa system perspective that integrates social andecological processes and is flexible enough toaccommodate the breadth of potential humanactions and responses.

Box 1.1. Challenges to Navigating Social–Ecological Barriers and Bridges.

The heading of this box combines the titlesof two seminal books on integrated social–ecological systems (“Barriers and Bridges”and “Navigating Social Ecological Systems”;Gunderson et al. 1995, Berkes et al. 2003).These titles capture the essence of the chal-lenges in integrating natural and social sci-ences. In this book we adopt the follow-ing conventions in addressing two importantchallenges in this transdisciplinary integra-tion (i.e., integration that transcends tradi-tional disciplines to formulate problems innew ways).

The same word often means differentthings.

1. To a sociologist, adaptation means thebehavioral adjustment by individuals totheir environment. To an ecologist itmeans the genetic changes in a pop-ulation to adjust to their environment(in contrast to acclimation, which entailsphysiological or behavioral adjustment byindividuals). To an anthropologist adap-tation means the cultural adjustmentto environment, without specifying itsgenetic or behavioral basis. In this bookwe use adaptation in its most generalsense (adjustment to change in environ-ment).


2. To an engineer or ecologist describing sys-tems with a single equilibrium, resilienceis the time required for a system to returnto equilibrium after a perturbation. Tosomeone describing systems with mul-tiple stable states, resilience is capacityof the system to absorb a spectrum ofshocks or perturbations and still retainand further develop the same funda-mental structure, functioning, and feed-backs. We use resilience in the lattersense.

3. Natural scientists describe feedbacks asbeing positive or negative to denotewhether they are amplifying or sta-bilizing, respectively. These words areoften used in the social sciences (andin common usage) to mean good orbad. The terminology is especially con-fusing for social–ecological systems,because negative feedbacks are oftensocially desirable (= “good”) and pos-itive feedbacks socially undesirable (=“bad”). We therefore avoid these termsand talk about amplifying or stabilizingfeedbacks.

4. Words that represent important conceptsin one discipline may be meaningless orviewed as jargon in another (e.g., post-modern, state factor). We define each tech-nical word the first time it is used and use

only those technical terms that are essentialto convey ideas effectively.

Approaches that are viewed as “good sci-ence” in one discipline may be viewed withskepticism in another.

1. Some natural scientists use systemsmodels to describe (either quantitativelyor qualitatively) the interactions amongcomponents of a system (such as a social–ecological system). Some social scien-tists view this as an inappropriate tool tostudy systems with a strong human ele-ment because it seems too deterministicto describe human actions. We use com-plex adaptive systems as a framework tostudy social–ecological systems because itenables us to study the integrated nature ofthe system but recognizes legacies of pastevents and the path dependence of humanagency as fundamental properties of themodel.

2. Some natural scientists rely largely onquantitative data as evidence to test ahypothesis, whereas some social scien-tists make extensive use of qualitativedescriptions of patterns that are lessamenable to quantification. We considerboth approaches essential to understandingthe complex dynamics of social–ecologicalsystems.

A Systems Perspective

Systems theory provides a conceptual frame-work to understand the dynamics of integratedsystems. A social–ecological system consistsof physical components, including soil, water,and rocks; organisms (plants, microbes, andanimals—including people); and the productsof human activities, such as food, money, credit,computers, buildings, and pollution. A social–ecological system is like a box or a board game,with explicit boundaries and rules, enabling usto quantify the amount of materials (for exam-ple, carbon, people, or money) in the systemand the factors that influence their flows into,through, and out of the system.

Social–ecological systems can be defined atmany scales, ranging from a single householdor community garden to the entire planet. Sys-tems are defined to include those componentsand interactions that a person most wants tounderstand. The size, shape, and boundariesof a social–ecological system therefore dependentirely on the problem addressed and theobjectives of study. A watershed that includesall the land draining into a lake, for example,is an appropriate system for studying the con-trols over pollution of the lake. A farm, city,water-management district, state, or countrymight be a logical unit for studying the effectsof government policies. A community, nation,


or the globe might be an appropriate unitfor studying barter and commerce. A neigh-borhood, community, or multinational regionmight be a logical unit for studying culturalchange. Defining the most appropriate unit ofanalysis is challenging because key ecologicaland social processes often differ in scale andlogical boundaries (for example, watershedsand water-management districts; Ostrom 1990,Young 1994). Most social–ecological systemsare open systems, in the sense that there areflows of materials, organisms, and informationinto and out of the system. We therefore cannotignore processes occurring outside our definedsystem of analysis, for example, the movementof food and wastes across city boundaries.

Social–ecological processes are the intercon-nections among components of a system. Thesemay be primarily ecological (for example, plantproduction, decomposition, wildlife migration),socioeconomic (manufacturing, education,fostering of trust among social groups), or amix of ecological and social processes (plowing,hunting, polluting). The interactions amongmultiple processes govern the dynamics ofsocial–ecological systems. Two types of inter-actions among components (amplifying andstabilizing feedbacks) are especially impor-tant in defining the internal dynamics ofthe system because they lead to predictableoutcomes (DeAngelis and Post 1991, Chapinet al. 1996). Amplifying feedbacks (termedpositive feedbacks in the systems litera-ture) augment changes in process rates andtend to destabilize the system (Box 1.2).They occur when two interacting componentscause one another to change in the samedirection (both components increase or bothdecrease; Fig. 1.4). A disease epidemic occurs,for example, when a disease infects susceptiblehosts, which produce more disease organisms,which infect more hosts, etc., until some otherset of interactions constrains this spiral ofdisease increase. Overfishing can also lead toan amplifying feedback, when the decline infish stocks gives rise to price supports thatenable fishermen to maintain or increase fish-ing pressure despite smaller catches, leadingto a downward spiral of fish abundance. Otherexamples of amplifying feedbacks include

–

Cattle

Grass Shrubs

Soil moisture

A D

EB

C+ +

+

++

+

–

Resource uptakeCompetitionHerbivoryResource exploitationDisturbance cycle

ProcessNature of feedback

A or DA+D

BCE

–

–

Livelihoods

Fire

StabilizingStabilizingStabilizing

StabilizingAmplifying

Figure 1.4. Examples of linked amplifying and stabi-lizing feedbacks in social–ecological systems. Arrowsshow whether one species, resource, or condition hasa positive or a negative effect on another. The feed-back between two species is stabilizing when thearrows have opposite sign (for example, species 1 hasa positive effect on species 2, but species 2 has a neg-ative effect on species 1). The feedback is amplify-ing, when both species affect one another in the samedirection (for example, more cattle providing moreprofit, which motivates people to raise more cattle;feedback loop C in the diagram).

population growth, erosion of cultural integrityin developing nations, and proliferation ofnuclear weapons.

Stabilizing feedbacks (termed negative feed-backs in the systems literature) tend toreduce fluctuations in process rates, although,if extreme, they can induce chaotic fluctuations.Stabilizing feedbacks occur when two interact-ing components cause one another to changein opposite directions (Fig. 1.4). For example,grazing by cattle reduces the biomass of foragegrasses, whereas the grass has a positive effecton cattle production. Any increase in densityof cattle reduces grass biomass, which then con-strains the food available to cattle, thereby sta-bilizing the sustainable densities of both grass


and cattle at intermediate levels. Other exam-ples of stabilizing feedbacks include prices ofgoods in a competitive market and nutrientsupply to plants in a forest. One of the keysto sustainability is to foster stabilizing feed-backs and constrain amplifying feedbacks that

might otherwise push the system toward somenew state. Conversely, if the current state issocially undesirable, for example, at an aban-doned mine site, carefully selected amplifyingfeedbacks may shift the system to a preferrednew state.

Box 1.2. Dynamics of Temporal Change

The stability and dynamics of a systemdepend on the balance of amplifying and sta-bilizing feedbacks and types and frequenciesof perturbations. The strength and natureof feedbacks largely govern the way a sys-tem responds to change. A system with-out strong feedbacks shows chaotic behav-ior in response to a random perturba-tion. Chaotic behavior is unpredictable anddepends entirely on the nature of the pertur-bation. The behavior of a ball on a surfaceprovides a useful analogy (Fig. 1.5; Hollingand Gunderson 2002, Folke et al. 2004). The

location of the ball represents the state of asystem as a function of some variable such aswater availability. In a chaotic system with-out feedbacks, the surface is flat, and we can-not predict changes in the state (i.e., location)of the system in response to a random per-turbation (Fig. 1.5a). This system structureis analogous to theories that important deci-sions can be described in terms of the poten-tial solutions and actors that happen to bepresent at key moments (garbage-can poli-tics; Cohen et al. 1972, Olsen 2001).

a. b.

c. d.

e. f.

Figure 1.5. The location of the ball represents thestate of a system in relationship to some ecological orsocial variable (e.g., water availability, as representedby the position along the horizontal axis). Changesin the state of the system in response to a perturba-tion depend on the nature of system feedbacks (illus-trated as the shape of the surface). The likelihood

that the system will change its state (location alongthe line) differs if there are (a) no feedbacks, (b)stabilizing feedbacks, (c) amplifying feedbacks, (d)alternative stable states, (d–e) changes in the internalfeedback structure (complex adaptive system), and(e–f) response of a complex adaptive system to per-sistent directional changes in a control variable.

A system dominated by stabilizing feed-backs tends to be stable because the interac-tions occurring within the system minimizethe changes in the system in response toperturbations. Using our analogy, stabiliz-ing feedbacks create a bowl-like depressionin the surface so the ball tends to return tothe same location after a random perturba-

tion (Fig. 1.5b). The resilience of the sys-tem, in this cartoon, is the likelihood that itwill remain in the same state despite per-turbations. This analogy characterizes theperspective of a balanced view of nature, inwhich there is a carrying capacity (maximumquantity) of fish, game, or trees that the envi-ronment can support, allowing managers to


regulate harvest to achieve a maximum sus-tained yield. This view is often based on con-siderable depth of biological understandingbut is incomplete (Holling and Gunderson2002).

A system dominated by amplifying feed-backs tends to be unstable because the ini-tial change is amplified by interactions occur-ring within the system. Amplifying feedbackstend to push the system toward some newstate by making the depressions less deepor creating elevated areas on the surface(Fig. 1.5c). This analogy characterizes theview that small is beautiful and that any tech-nology is bad because it causes change. Thereare certainly many examples where technol-ogy has led to unfavorable outcomes, butthis worldview, like the others, is incomplete(Holling and Gunderson 2002).

Many systems can be characterized byalternative stable states, each of which isplausible in a given environment. Neighbor-hoods in US cities, for example, are likely tobe either residential or industrial but unlikelyto be an even mix of the two. In the sur-face analogy, alternative stable states rep-resent multiple depressions in the surface(Fig. 1.5d). A system is likely to return toits original state (=depression) after a smallperturbation, but a larger disturbance mightincrease the likelihood that it will shift tosome alternative state. In other words, thesystem exhibits a nonlinear response to theperturbation and shifts to a new state if somethreshold is exceeded. There may also bepathways of system development, such as thestages of forest succession, in which the inter-nal dynamics of the system cause it to movereadily from one state to another. Some of

these depressions may be deep and representirreversible traps. Others may be shallow, sothe system readily shifts from one state toanother through time. This worldview incor-porates components of all the previous per-spectives but is still incomplete.

The previous cartoons of nature implythat the stability landscape is static. How-ever, each transition influences the internaldynamics of a complex adaptive system andtherefore the probability of subsequent tran-sitions, so the shape of the surface is con-stantly changing (Fig. 1.5e). Reductions inAtlantic cod populations due to overfishing,for example, increased pressures for estab-lishment of aquaculture and charter fishingbusinesses, which then made it less likelythat industrial-scale cod fishing would returnto the North Atlantic. This analogy of astability landscape that is constantly evolv-ing suggests that precise predictions of thefuture state of the system are impossibleand focuses attention on understanding thedynamics of change as a basis for stewardship(Gunderson and Holling 2002).

Now imagine that rather than havinga random perturbation in some importantstate variable like water availability, thisparameter changes directionally. This ele-ment of directionality increases the likeli-hood that the system will change in a spe-cific direction after perturbation (Fig. 1.5f).The stronger and more persistent the direc-tional changes in exogenous control vari-ables, the more likely it is that new stateswill differ from those that we have knownin the past. This represents our concept ofsystem response to a directionally changingenvironment.

Issues of Scale: Exogenous, Slow,and Fast Variables

Changes in the state of a system depend onvariables that change slowly but strongly influ-ence internal dynamics. Social–ecological sys-tems respond to a spectrum of controls thatoperate across a range of temporal and spatial

scales. These can be roughly grouped as exoge-nous controls, slow variables, and fast variables(Fig. 1.3). We describe these first for ecologicalsubsystems, then consider their social counter-parts.

Exogenous controls are factors such asregional climate or biota that strongly shapethe properties of continents and nations. They


remain relatively constant over long time peri-ods (e.g., a century) and across broad regionsand are not strongly influenced by short-term,small-scale dynamics of a single forest stand orlake. At the scale of an ecosystem or watershed,there are a few critical slow variables, i.e., vari-ables that strongly influence social–ecologicalsystems but remain relatively constant overyears to decades despite interannual variationin weather, grazing, and other factors, becausethey are buffered by stabilizing feedbacks thatprevent rapid change (Chapin et al. 1996,Carpenter and Turner 2000). Soil organic mat-ter, for example, retains pulses of nutrients fromautumn leaf fall, crop residues, or windstorms;retains water and nutrients; and releases theseresources which are then absorbed by plants;the quantity of soil organic matter is bufferedby feedbacks related to plant growth and lit-ter production. Critical slow variables includepresence of particular functional types of plantsand animals (e.g., evergreen trees or herbivo-rous mammals); disturbance regime (propertiessuch as frequency, severity, and size that char-acterize typical disturbances); and the capacityof soils or sediments to supply water and nutri-ents. Slow variables in ecosystems, in turn, gov-ern fast variables at the same spatial scale (e.g.,deer or aphid density, individual fire events)that respond sensitively to daily, seasonal, andinterannual variation in weather and other fac-tors. When aggregated to regional or globalscales, changes that occur in ecosystems, forexample, those mediated by human activities,can modify the environment to such an extentthat even regional controls such as climateand regional biota that were once consideredconstant parameters are now directionallychanging at decade-to-century time scales(Foley et al. 2005). Regardless of the causes,persistent directional changes in broad regionalcontrols, such as climate and biodiversity,inevitably cause directional changes in crit-ical slow variables and therefore the struc-ture and dynamics of ecosystems, including thefast variables. The exogenous and slow vari-ables are critical to long-term sustainability,although most management and public atten-tion focus on fast variables, whose dynamics aremore visible.

Analogous to the ecological subsystem, thesocial subsystem can be viewed as composed ofexogenous controls, critical slow variables, andfast variables (Straussfogel 1997). These consistof vertically nested relationships, ranging fromglobal to local, and linked by cross-scale inter-actions (Ostrom 1999a, Young 2002b, Adgeret al. 2005). At the sub-global scale a predomi-nant history, culture, economy, and governancesystem often characterize broad regions ornation states such as Europe or sub-SaharanAfrica (Chase-Dunn 2000). These exogenoussocial controls tend to be less sensitive tointerannual variation in stock-market pricesand technological change than are the internaldynamics of local social–ecological systems;the exogenous controls constrain local options.This asymmetry between regional and localcontrols occurs in part because of asymmetricpower relationships between national and localentities and in part because changes in a smalllocality must be very strong to substantiallymodify the dynamics of large regions. Regionalcontrols sometimes persist for a long time andchange primarily in response to changes thatare global in extent (e.g., globalization of mar-kets and finance institutions), but at other timeschange can occur quickly, as with the collapseof the Soviet Union in the 1990s or the global-ization of markets and information (Young etal. 2006). As in the biophysical system, a fewslow variables (e.g., wealth and infrastructure;property-and-use rights; and cultural ties to theland) are constrained by regional controls andinteract with one another to shape fast variableslike community income or population density.Both slow and fast social variables can havemajor effects on ecological processes (Costanzaand Folke 1996, Holling and Sanderson1996).

Systems differ in their sensitivity to differ-ent types of changes or the range of conditionsover which the change occurs. The !Kung Sanof the Kalahari Desert will be much more sen-sitive than people of a rainforest to a 10-cmincrease in annual rainfall because it repre-sents a doubling of rainfall rather than a 5%increase. Regions also differ in their sensi-tivity to introduction of new biota (sprucebark beetle, zebra mussel, or West Nile virus),


new economic pressures (development of aqua-culture, shifting of car manufacture to Asia,collapse of the stock market), or new culturalvalues. There are typically relatively few (oftenonly three to five) slow variables that are crit-ical in understanding the current dynamics ofa specific system (Carpenter et al. 2002), somanagement designed to reduce sensitivity todirectional changes in slow variables is not animpossible task. The identity of critical con-trol variables may change over time, however,requiring continual reassessment of our under-standing of the social–ecological system. Thekey challenge, requiring collaborative researchby managers and natural and social scientists, isto identify the critical slow variables and theirlikely changes over time.

Incorporating Scale, Human Agency,and Uncertainty into Dynamic Systems

Cross-scale linkages are processes that con-nect the dynamics of a system to events occur-ring at other times or places (see Chapter 5).Changes in the human population of a region,for example, may be influenced by the wealthand labor needs of individual families (finescale), by national policies related to birth con-trol (focal scale), and by global inequalitiesin living standards that influence immigration(large scale). Events that occur at each scaletypically influence events at other scales. Theuniversal importance of cross-scale linkages insocial–ecological systems makes it important tostudy them at multiple temporal and spatialscales, because different insights and answersemerge at each scale (Berkes et al. 2003).

Legacies are past events that have largeeffects on subsequent dynamics of social–ecological systems. This generates a pathdependence that links current dynamics topast events and lays the foundation for futurechanges (North 1990). Legacies include theimpact of plowing on soils of a regeneratingforest, the impact of the Depression in the1930s on economic decisions made by house-holds 40 years later, and the continuation ofsubsistence activities by indigenous people whomove from villages to cities. Because of path

dependence, the current dynamics of a systemalways depend on both current conditions andthe history of prior events. Consequently, dif-ferent trajectories can occur at different timesor places, even if the initial conditions werethe same. Path dependence is absolutely crit-ical to management, because it implies thathuman actions taken today, whether construc-tive or destructive, can influence the future stateof the system. Good management can make adifference!

Human agency (the capacity of humans tomake choices that affect the system) is oneof the most important sources of path depen-dence. Human decisions depend on both pastevents (legacy effects) and the plans that peoplemake for the future (reflexive behavior). Thestrong path dependence of social–ecologicalsystems is typical of a general class of systemsknown as complex adaptive systems. These aresystems whose components interact in waysthat cause the system to adjust (i.e., “adapt”)in response to changes in conditions. This isnot black magic, but a consequence of inter-actions and feedbacks. Some of the most fre-quent failures in resource management occurbecause managers and resource users fail tounderstand the principles by which complexadaptive systems function. It is therefore impor-tant to understand their dynamics. Understand-ing these dynamics also provides insights intoways that managers can achieve desirable out-comes in a system that is responding simultane-ously to management actions and to persistentdirectional changes in exogenous controls.

Whenever system components with differ-ent properties interact spontaneously with oneanother, some components persist and oth-ers disappear (i.e., the system adapts; Levin1999; Box 1.2). In social–ecological systems,for example, organisms compete or eat oneanother, causing some species to become morecommon and others to disappear. Similarly,purchasing or competitive relationships amongbusinesses cause some firms to persist and oth-ers to fail. Those components that interactthrough stabilizing feedbacks are most likelyto persist. This self-organization of compo-nents linked by stabilizing feedbacks occursspontaneously without any grand design. It


causes complex adaptive systems to be rela-tively stable (tend to maintain their proper-ties over time; DeAngelis and Post 1991, Levin1999). This self-regulation simplifies manage-ment challenges in many respects. A complexadaptive system like a forest, for example, tendsto “take care of itself.” This differs from adesigned structure like a car, whose compo-nents do not interact spontaneously and wheremaintenance must be continually applied just tokeep the car in the same condition (Levin 1999).

If conditions change enough to alter theinteractions among system components, the sys-tem adapts to the new conditions, hence theterm complex adaptive system (Levin 1998).The new balance of system components, in turn,alters the way in which the system respondsto perturbations (path dependence), creatingalternative stable states, each of which couldexist in a given environment (see Chapter 5).Given that exogenous variables are alwayschanging on all time scales, social–ecologicalsystems are constantly adjusting and chang-ing. Consequently, it is virtually impossible tomanage a complex adaptive system to attainconstant performance, such as the constant pro-duction of a given timber species. System prop-erties are most likely to change if there aredirectional changes in exogenous controls. Thestronger and more persistent the directionalchanges in control variables, the more likely itis that a threshold will be exceeded, leading toa new state.

If a threshold is exceeded, and the systemchanges radically, new interactions and feed-backs assume greater importance, and somecomponents of the previous system may dis-appear. If a region shifts from a mining to atourist economy, for example, the communitymay become more concerned about fundingfor education and regulations that assure cleanwater. The regime shifts that occur as the sys-tem changes state also depend on the past stateof the system (path dependence). The presenceof a charismatic leader or nongovernmentalorganization (NGO), for example, can be crit-ical in determining whether large cattle ranchesare converted to conservation easements orsubdivisions when rising land values and taxesmake ranching unprofitable.

These simple generalizations about complexadaptive systems have profound implicationsfor resource stewardship: (1) Social and ecolog-ical components of a social–ecological systemalways interact and cannot be managed in iso-lation from one another. (2) Changes in socialor ecological controls inevitably alter social–ecological systems regardless of managementefforts to prevent change. (3) Historical eventsand human actions, including management, canstrongly influence the pathway of change. (4)The thresholds and nonlinear dynamics asso-ciated with path dependence, compounded bylack of information and human volition, con-strain our capacity to predict future change.Resource management and policy decisionsmust, therefore, always be made in an envi-ronment of uncertainty (Ludwig et al. 1993,Carpenter et al. 2006a).

Adaptive Cycles

The long-term stability of systems depends onchanges that occur during critical phases ofcycles of long-term change. All systems expe-rience disturbances such as fire, war, reces-sion, change in leadership philosophy, or clo-sure of manufacturing plants that cause largerapid changes in key system properties. Suchdisturbances have qualitatively different effectson social–ecological systems than do short-term variability and gradual change. Adap-tive cycles provide a framework for describ-ing the role of disturbance in social–ecologicalsystems (Holling 1986). They are cycles of sys-tem disruption, reorganization, and renewal. Inan adaptive cycle, a system can be disruptedby disturbance and either regenerate to a sim-ilar state or be transformed to some new state(Fig. 1.6a; Holling 1986, Walker et al. 2004).Adaptive cycles exhibit several recognizablephases. The cycle may be initiated by a distur-bance such as a stand-replacing wildfire thatcauses a rapid change in most properties ofthe system. Trees die, productivity decreases,runoff to streams increases, and public faithin fire management is shattered. This releasephase occurs in hours to days and radicallyreduces the structural complexity of the system.Other factors that might trigger release include


Connectednessweak strong

Conservation

Release

Renewal

Growth

Cap

ital

littl

elo

ts

a. Adaptive cycle

b. Panarchy

Intermediatesize and speed

Large andslow

Small andfast

Revo

ltRe

volt

Remember

Figure 1.6. (a) Adaptive cycle and (b) cross-scalelinkages among adaptive cycles (panarchy) in asocial–ecological system. At any given scale, a systemoften goes through adaptive cycles of release (col-lapse), renewal (reorganization), growth, and conser-vation (steady state). These adaptive cycles of changecan occur at multiple levels of organizations, suchas individuals, communities, watersheds, and regions.These adaptive cycles interact forming a panarchy.For example, dynamics at larger scales (e.g., migra-tion dynamics or wealth) provide legacies, context,and constraints that shape patterns of renewal (sys-tem memory). Dynamics at finer scales (e.g., insectpopulation dynamics, household structure) may trig-ger release (revolt; e.g., insect outbreak). Redrawnfrom Holling and Gunderson (2002) and Hollinget al. (2002b).

threshold response to phosphorus loading ofa lake, collapse of the local or regional econ-omy, or a transition from traditional to inten-sive agriculture. Following release, there is a rel-atively brief (months to years) renewal phase.For example, after forest disturbance, seedlingsestablish and new policies for managing theforest may be adopted. Many things can hap-

pen during renewal: The species and policiesthat establish might be similar to those presentbefore the fire. It is also a time, however, whenthere is relatively little resistance to the estab-lishment of a new suite of species or poli-cies that emerge from the surrounding land-scape (see Fig. 2.4). These innovations may leadto a system that is quite different from theprefire system, i.e., a regime shift. After thisbrief window of opportunity for change, theforest goes through a growth phase over sev-eral decades, when environmental resources areincorporated into living organisms, and policiesbecome regularized. The nature of the regener-ating forest system is largely determined by thespecies and regulations that established duringrenewal. During the growth phase, the forest isrelatively insensitive to potential agents of dis-turbance. The high moisture content and lowbiomass of early successional trees, for exam-ple, make regenerating forests relatively non-flammable. Constant changes in the nature ofthe forest cause both managers and the publicto accept changing conditions and regulationsas a reasonable pattern. As the forest developsinto the steady-state conservation phase, theinteractions among components of the systembecome more specialized and complex. Lightand nutrients decline in availability, for exam-ple, leading to specialization among plants touse different light environments and differentfungal associations (mycorrhizae) to acquirenutrients. Similarly, in the policy realm, therelatively constant state of the forest leads tomanagement rules that are aimed at main-taining this constancy to provide predictablepatterns of recreation, hunting, and forest har-vest. Due to the increased interconnectednessamong these social and ecological variables, theforest becomes more vulnerable to any factorthat might disrupt this balance, including fire,drought, changes in management goals, or ashift in the local economy. Large changes in anyof these factors could trigger a new release inthe adaptive cycle.

Many human organizations also exhibitcyclic patterns of change. A business or NGO,for example, may be founded in response toa perceived opportunity for profit or socialreform. If successful, it grows amidst constant


adjustment to changes in personnel and activ-ities. Eventually it reaches a relatively sta-ble size, at which time the internal structureand operating procedures are regularized, mak-ing it less flexible to respond to changes inthe economic or social climate. When condi-tions change, the business or NGO may eitherenter a new period of adjustment (growth) ordecline (release), followed by potential renewalor collapse.

Perhaps the most surprising thing aboutadaptive cycles is that the sequence of phases(release, renewal, growth, and conservation)can be used as a way of thinking aboutmany types of social-ecological systems, includ-ing lakes, businesses, governments, nationaleconomies, and cultures, although the sequenceof phases is not always the same (Gundersonet al. 1995). Clearly the specific mechanismsunderlying cycles in these different systemsmust be quite different. One of the unsolvedchallenges in understanding social–ecologicalsystems is to determine the general systemproperties and mechanisms that underlie theapparent similarities in cyclic patterns of dif-ferent types of systems and to clarify the dif-ferences. The specific mechanisms of adap-tive cycles in different types of systems aredescribed in many of the following chapters.

One of the most important managementlessons to emerge from studies of adaptivecycles is that social–ecological systems aretypically most vulnerable (likely to change toa new state in response to a stress or distur-bance) and create their own vulnerabilitiesin the conservation phase, where they typi-cally spend most of their time. In this stage,managers frequently seek to reduce fluctu-ations in ecological processes and preventsmall disturbances in order to increase theefficiency of achieving management goals(e.g., the amount of timber to be harvested;number of houses that can be built; the budgetto pay salaries of personnel), increasing thelikelihood that even larger disturbances willoccur (Holling and Meffe 1996, Walker andSalt 2006). Flood control, for example, reducesflood frequency, which encourages infrastruc-ture development in floodplains where it isvulnerable to the large flood that will eventu-

ally occur. Prevention of small insect outbreaksincreases the likelihood of larger outbreaks.Management that encourages small-scaledisturbances and innovation during the conser-vation phase reduces the vulnerability to largerdisruptions (Holling et al. 1998, Carpenter andGunderson 2001, Holling et al. 2002a). Thespecific mechanisms that link stability in theconservation phase to triggers for disruptionare described in later chapters.

Release and crisis provide important oppor-tunities for change (Gunderson and Holling2002, Berkes et al. 2003; Fig. 1.5b). Someof these changes may be undesirable (inva-sion of an exotic species, dramatic shift inpolitical regimes that decrease social equity),whereas others may be desirable (implemen-tation of innovative policies that are moreresponsive to change). Recognition of thesechanging properties of a system through thelens of an adaptive cycle suggests that effec-tive long-term management and policy-makingmust be highly flexible and adaptive, looking forwindows of opportunity for constructive policyshifts.

Most social–ecological systems are spatiallyheterogeneous and consist of mosaics of subsys-tems that are at different stages of their adap-tive cycles. Interactions and feedbacks amongthese adaptive cycles operating at differenttemporal and spatial scales account for theoverall dynamics of the system (termed panar-chy; Fig. 1.6b; Holling et al. 2002b). A forest, forexample, may consist of different-aged standsat different stages of regeneration from loggingor wildfire. In this case, the system as a wholemay be at steady state (a steady-state mosaic)even though individual stands are at differentstages in their cycles (Turner et al. 2001). In gen-eral, there are different benefits to be gainedat different phases of the cycle, so policies thatpermit or foster certain disturbances may beappropriate. Many families contain individualsat various stages of birth, maturation, and deathand benefit from the resulting diversity of skills,perspectives, and opportunities. Similarly, in ahealthy economy new firms may establish atthe same time that other less-efficient firms goout of business. Maintenance of natural cyclesof fire or insect outbreak produces wildlife


habitat in the early growth phase and preventsexcessive fuel accumulation that might other-wise trigger more catastrophic fires. Perhaps themost dangerous management strategy would beto prevent disturbance uniformly throughout aregion until all subunits reach a similar state ofmaturity, making it more likely that the entiresystem will change synchronously.

Sustainability in a DirectionallyChanging World

Conceptual Frameworkfor Sustainability Science

A systems perspective provides a logical frame-work for managing changes in social–ecologicalsystems. To summarize briefly the previous sec-tions, the dynamic interactions of ecologicaland social processes that characterize most oftoday’s urgent problems necessitate a social–ecological framework for planning and stew-ardship. Any sustainable solution to a resourceissue must be compatible with current socialand ecological conditions and their likely futurechanges. A resource policy that is not eco-logically, economically, and culturally sustain-able is unlikely to be successful. Sustainableresource stewardship must therefore be mul-tifaceted, recognizing the interactions amongecological, economic, and cultural variables andthe important roles that past history and futureevents play in determining outcomes in specificsituations. In addition, systems undergo cyclicchanges in their sensitivity to external perturba-tions, so management solutions that may havebeen successful at one time and place may ormay not work under other circumstances.

The complexity of these dynamics helpsframe the types of stewardship approaches thatare most likely to be successful. It is unlikelythat a rigid set of rules will lead to success-ful stewardship because key decisions mustfrequently be made under conditions of nov-elty and uncertainty. Moreover, under currentrapid rates of global environmental and socialchanges, the current environment for decision-making is increasingly different from past

conditions that may be familiar to managers orthe future conditions that must be accommo-dated. The more rapidly the world changes, theless likely that rigid management approacheswill be successful. By considering the systemproperties presented above, however, we candevelop resilience-based approaches that sub-stantially reduce the risk of undesirable social–ecological outcomes and increase the likelihoodof making good use of unforeseen opportuni-ties. This requires managing for general sys-tem properties rather than for narrowly definedproduction goals. In this section, we present aframework for this approach that is describedin detail in subsequent chapters.

Sustaining the desirable features of our cur-rent world for future generations is an impor-tant societal goal. The challenge of doing soin the face of persistent directional trends inunderlying controls has led to an emerging sci-ence of sustainability (Clark and Dickson 2003).Sustainability has been adopted as a centralgoal of many local, national, and internationalplanning efforts, but it is often unclear exactlywhat it is or how to achieve it. In this bookwe use the United Nations Environment Pro-gramme (UNEP) definition of sustainability:the use of the environment and resources tomeet the needs of the present without com-promising the ability of future generations tomeet their own needs (WCED 1987). Accord-ing to this definition, sustainability requiresthat people be able to meet their own needs,i.e., to sustain human well-being (that is, thebasic material needs for a good life, freedomand choice, good social relations, and personalsecurity) now and in the future (Dasgupta2001; see Chapter 3). Since sustainability andwell-being are value-based concepts, there areoften conflicting visions about what shouldbe sustained and how sustainability should beachieved. Thus the assessment of sustainabil-ity is as much a political as a scientific pro-cess and requires careful attention to whosevisions of sustainability are being addressed(Shindler and Cramer 1999). Nonetheless, anyvision of sustainability ultimately depends onthe life-support capacity of the environmentand the generation of ecosystem services(see Chapter 2).


Types and Substitutability of Capital

Sustainability requires that the productive baserequired to support well-being be maintained orincreased over time. Well-being can be definedin economic terms as the present value offuture utility, i.e., the capacity of individualsor society to meet their own needs (Dasguptaand Maler 2000, Dasgupta 2001). Well-beingalso has important social and cultural dimen-sions (see Chapter 3), but the economic def-inition enables us to frame sustainability ina systems context. Sustainability requires thatthe total capital, or productive base (assets) ofthe system, be sustained. This capital has nat-ural, built (manufactured), human, and socialcomponents (Arrow et al. 2004). Natural cap-ital consists of both nonrenewable resources(e.g., oil reserves) and renewable ecosystemresources (e.g., plants, animals, and water) thatsupport the production of goods and serviceson which society depends. Built capital consistsof the physical means of production beyondthat which occurs in nature (e.g., tools, clothing,shelter, dams, and factories). Human capital isthe capacity of people to accomplish their goals;it can be increased through various forms oflearning. Together, these forms of capital con-stitute the inclusive wealth of the system, i.e.,the productive base (assets) available to soci-ety. Although not included in the formal defini-tion of inclusive wealth, social capital is anotherkey societal asset. It is the capacity of groupsof people to act collectively to solve problems(Coleman 1990). Components of each of theseforms of capital change over time. Naturalcapital, for example, can increase throughimproved management of ecosystems, includ-ing restoration or renewal of degraded ecosys-tems or establishment of networks of marine-protected areas; built capital through invest-ment in bridges or schools; human capitalthrough education and training; and social cap-ital through development of new partnershipsto solve problems. Increases in this productivebase constitute genuine investment. Investmentis the increase in the quantity of an asset timesits value. Sustainability requires that genuineinvestment be positive, i.e., that the productivebase (genuine wealth) not decline over time

(Arrow et al. 2004). This provides an objectivecriterion for assessing whether management issustainable.

To some extent, different forms of capital cansubstitute for one another, for example, naturalwetlands can serve water purification functionsthat might otherwise require the constructionof expensive water treatment facilities. Well-informed leadership may be able to implementmore cost-effective solutions to a given prob-lem (a substitution of human for economic cap-ital). However, there are limits to the extentto which different forms of capital can be sub-stituted (Folke et al. 1994). Water and food,for example, are essential for survival, and noother forms of capital can completely substi-tute for them (see Chapter 12). They there-fore have extremely high value to society whenthey become scarce. Declines in the trust thatsociety has in its leadership; sense of culturalidentity; the capacity of agricultural soils toretain sufficient water to support production;or the presence of species that pollinate criti-cal crops, for example, cannot be readily com-pensated by substituting other forms of capital.Losses of many forms of human, social, and nat-ural capital are especially problematic becauseof the impossibility or extremely high costsof providing appropriate substitutes (Folkeet al. 1994, Daily 1997). We therefore focusparticular attention on ways to sustain thesecomponents of capital, without which futuregenerations cannot meet their needs (Arrowet al. 2004).

Well-informed managers often have guide-lines for sustainably managing the componentsof inclusive wealth. For example, harvestingrates of renewable natural resources shouldnot exceed regeneration rates; waste emissionsshould not exceed the assimilative capacityof the environment; nonrenewable resourcesshould not be exploited at a rate that exceedsthe creation of renewable substitutes; edu-cation and training should provide opportu-nities for disadvantaged segments of society(Barbier 1987, Costanza and Daly 1992, Folkeet al. 1994).

The concept of maintaining positive genuineinvestment as a basis for sustainability is impor-tant because it recognizes that the capital assets


of social–ecological systems inevitably changeover time and that people differ through timeand across space in the value that they placeon different forms of capital. If the productivebase of a system is sustained, future generationscan make their own choices about how best tomeet their needs. This defines criteria for decid-ing whether certain practices are sustainable ina changing world. There are substantial chal-lenges in measuring changes in various forms ofcapital, in terms of both their quantity and theirvalue to society (see Chapter 3). Nonetheless,the best current estimates suggest that manu-factured and human capital have increased inthe last 50 years in most countries but that nat-ural capital has declined as a result of deple-tion of renewable and nonrenewable resourcesand through pollution and loss of the functionalbenefits of biodiversity (Arrow et al. 2004). Insome countries, especially some of the poorerdeveloping nations, the loss of natural capitalis larger than increases in manufactured andhuman capital, indicating a clearly unsustain-able pathway of development (MEA 2005d).Some argue that there have also been substan-tial decreases in social capital as a result ofmodernization and urban life (Putnam 2000).

Managing Change in Ways that FosterSustainability

Managing for sustainability requires atten-tion to changes typical of complex adaptivesystems. In the previous section we definedcriteria to assess sustainability. These crite-ria are of little use if the system to whichthey are applied changes radically. Now wemust place sustainability in the context of the

directional changes in factors that govern theproperties of most social–ecological systems.Three broad categories of outcome are possi-ble: (1) persistence of the fundamental prop-erties of the current system through adapta-tion, (2) transformation of the system to afundamentally different, potentially more desir-able state, or (3) passive changes (often degra-dation to a less-favorable state) of the sys-tem as a result of failure of the system toadapt or transform. Intermediate outcomesare also possible, if some components (e.g.,ecological subsystems, institutions, or socialunits) of the system persist, others transform,and others degrade (Turner et al. 2003). Sus-tainability implies the persistence of the fun-damental properties of the system or of activetransformation through deliberate substitutionof different forms of capital to meet society’sneeds in new ways. In contrast, degradationimplies the loss of inclusive wealth and there-fore the potential to achieve sustainability.

How can we manage the dynamics of changeto improve the chances for persistence ortransformation? Four general approaches havebeen identified as ways to foster sustainabil-ity under conditions of directional change:(1) reduced vulnerability, (2) enhanced adap-tive capacity, (3) increased resilience, and(4) enhanced transformability. Each of theseapproaches emphasizes a different set ofprocesses by which sustainability is fostered(Table 1.2, Fig. 1.7). Vulnerability addresses thenature of stresses that cause change, the sensi-tivity of the system to these changes, and theadaptive capacity to adjust to change. Adap-tive capacity addresses the capacity of actorsor groups of actors to adjust so as to minimizethe negative impacts of changes. Resilience

Table 1.2. Assumptions of frameworks addressing long-term human well-being. Modified from Chapin et al.(2006a).

Assumed change in Nature of mechanisms Other approachesFramework exogenous controls emphasized often incorporated

Vulnerability Known System exposure and sensitivity todrivers; equity

Adaptive capacity, resilience

Adaptive capacity Known or unknown Learning and innovation NoneResilience Known or unknown Within-system feedbacks and

adaptive governanceAdaptive capacity,

transformabilityTransformability Directional Learn from crisis Adaptive capacity, resilience


Learning, coping,

innovating, adapting

Drivers System dynamics Outcomes

Externaldrivers

Imp

acts

Inte

ract

ion

s

Sensitivity

Natural &human capital

Biotic &social

interactions

Vulnerability

Adaptability

Resilience

Transformability

Persistence

Activelynavigated

transformation

Unintendedtransformation

Exp

osu

re

Figure 1.7. Conceptual framework linking humanadaptive capacity, vulnerability, resilience, and trans-formability. See text for definition of terms. Thesystem (e.g., household, community, nation, etc.)responds to a suite of interacting drivers (stresses,events, shocks) to produce one of three potentialoutcomes: (1) persistence of the existing systemthrough resilience; (2) actively navigated transfor-mation to a new, potentially more beneficial statethrough transformability; or (3) unintended trans-formation to a new state (often degraded) due tovulnerability and the failure to adapt or transform.These three outcomes are not mutually exclusive,because some components (e.g., ecological subsys-tems, institutions, or social units) of the system maypersist, others transform, and others degrade. Thesensitivity of the system to perturbations dependson its exposure (intensity, frequency, and duration)to each perturbation, the interactions among dis-tinct perturbations, and critical properties of the sys-tem. The system response to the resulting impacts

depends on its adaptive capacity (i.e., its capacity tolearn, cope, innovate, and adapt). Adaptive capac-ity, in turn, depends on the amount and diversityof social, economic, physical, and natural capitaland on the social networks, institutions, and entitle-ments that influence how this capital is distributedand used. System response also depends on effec-tiveness of cross-scale linkages to changes occurringat other temporal and spatial scales. Those compo-nents of the system characterized by strong stabi-lizing feedbacks and adaptive capacity are likely tobe resilient and persist. Alternatively, if the existingconditions are viewed as untenable, a high adaptivecapacity can contribute to actively navigated trans-formation, the capacity to change to a new, poten-tially more beneficial state of the system or sub-system. If adaptive capacity of some componentsis insufficient to cope with the impacts of stresses,they are vulnerable to unintended transformationto a new state that often reflects degradation inconditions.

incorporates adaptive capacity but also entailsadditional system-level attributes of social–ecological systems that provide flexibility toadjust to change. Transformability addressesactive steps that might be taken to change thesystem to a different, potentially more desirablestate. Although anthropologists, ecologists, and

geographers developed these approaches some-what independently (Janssen et al. 2006), theyare becoming increasingly integrated (Berkeset al. 2003, Turner et al. 2003, Young et al.2006). This integration of ideas provides pol-icy makers and managers with an increasinglysophisticated and flexible tool kit to address


the challenges of sustainability in a directionallychanging world. We apply the term resilience-based ecosystem stewardship to this entiresuite of approaches to sustainability, becauseof its emphasis on sustaining functional proper-ties of social–ecological systems over the longterm despite perturbation and change. Theseissues represent the core challenges of man-aging social–ecological systems sustainably. Wenow briefly outline this suite of approaches.

Vulnerability

Vulnerability is the degree to which a sys-tem is likely to experience harm due to expo-sure to a specified hazard or stress (Turneret al. 2003, Adger 2006). Vulnerability theoryis rooted in socioeconomic studies of impactsof events (e.g., floods or wars) or stresses (e.g.,chronic food insecurity) on social systems buthas been broadened to address responses ofentire social–ecological systems. Vulnerabilityanalysis deliberately addresses human valuessuch as equity and well-being. Vulnerability toa given stress can be reduced by (1) reducingexposure to the stress (mitigation); (2) reduc-ing sensitivity of the system to stress by sustain-ing natural capital and the components of well-being, especially for the disadvantaged; and/or(3) increasing adaptive capacity and resilience(see below) to cope with stress (Table 1.3;Turner et al. 2003). The incorporation ofadaptive capacity and resilience as integralcomponents of the vulnerability framework(Turner et al. 2003, Ford and Smit 2004) illus-trates the integration of different approaches tosustainability science.

Exposure to a stress can be reduced byminimizing its intensity, frequency, duration,or extent. Prevention of pollution or banningof toxic pesticides, for example, reduces thevulnerability of people who would otherwisebe exposed to these hazards. Mitigation(reduced exposure) is especially challengingwhen the stress is the cumulative effect of pro-cesses occurring at scales that are larger thanthe system being managed. Anthropogeniccontributions to climate warming through theburning of fossil fuels, for example, is globally

Table 1.3. Principal sustainability approaches andmechanisms. Adapted from Levin (1999), Folke et al.(2003), Turner et al. (2003), Chapin et al. (2006a),Walker et al. (2006).

VulnerabilityReduce exposure to hazards or stressesReduce sensitivity to stresses

Sustain natural capitalMaintain components of well-beingPay particular attention to vulnerability of the

disadvantagedEnhance adaptive capacity and resilience (see below)

Adaptive capacityFoster biological, economic, and cultural diversityFoster social learningExperiment and innovate to test understandingSelect, communicate, and implement appropriate

solutions.Resilience

Enhance adaptive capacity (see above)Sustain legacies that provide seeds for renewalFoster a balance between stabilizing feedbacks and

creative renewalAdapt governance to changing conditions

TransformabilityEnhance diversity, adaptation, and resilienceIdentify potential future options and pathways to

get thereEnhance capacity to learn from crisisCreate and navigate thresholds for transformation

dispersed, so it cannot be reversed by actionstaken solely by those regions that experiencegreatest impacts of climatic change (McCarthyet al. 2005). Other globally or regionally dis-persed stresses include inadequate suppliesof clean water and uncertain availability ofnutritious food (Steffen et al. 2004, Kaspersonet al. 2005).

Sensitivity to a stress can be reduced in atleast three ways: (1) sustaining the slow eco-logical variables that determine natural capital;(2) maintaining key components of well-being;and (3) paying particular attention to theneeds of the disadvantaged segments of soci-ety, who are generally most vulnerable. Thepoor or disadvantaged, for example, are espe-cially vulnerable to food shortages or eco-nomic downturns, and people living in flood-plains or the wildland–urban interface areespecially vulnerable to flooding or wild-fire, respectively. An understanding of the


causes of differential vulnerability can lead tostrategies for targeted interventions to reduceoverall vulnerability of the social–ecologicalsystem.

The causes of differential vulnerability areoften deeply rooted in the slow variables thatgovern the internal dynamics of society, suchas power relationships or distribution of land-use rights among segments of society (seeChapter 3). Conventional vulnerability anal-ysis assumes that the stresses are known orpredictable (i.e., either steady state or chang-ing in a predictable fashion). However, long-term reductions in vulnerability often requireattention to adaptive capacity and resilience atmultiple scales in addition to targeted effortsto reduce exposure and sensitivity to knownstresses.

Adaptive Capacity

Adaptive capacity (or adaptability) is thecapacity of actors, both individuals and groups,to respond to, create, and shape variabilityand change in the state of the system (Folkeet al. 2003, Walker et al. 2004, Adger et al.2005). Although the actors in social–ecologicalsystems include all organisms, we focus par-ticularly on people in addressing the role ofadaptive capacity in social–ecological change,because human actors base their actions notonly on their past experience but also on theircapacity to plan for the future (reflexive action).This contrasts with evolution, which shapes theproperties of organisms based entirely on theirgenetic responses to past events. Evolutionhas no forward-looking component. Adaptivecapacity depends on (1) biological, economic,and cultural diversity that provides the buildingblocks for adjusting to change; (2) the capacityof individuals and groups to learn how theirsystem works and how and why it is changing;(3) experimentation and innovation to testthat understanding; and (4) capacity to governeffectively by selecting, communicating,and implementing appropriate solutions(Table 1.3) We discuss the social and culturalbases of adaptive capacity in Chapters 3 and

4 and here focus on its relationship to systemproperties.

Sources of biological, economic, and cul-tural diversity provide the raw material onwhich adaptation can act (Elmqvist et al. 2003,Norberg et al. 2008). In this way it defines theoptions available for adaptation. People canaugment this range of options through learning,experimentation, and innovation. This capac-ity to create new options is strongly influencedby people’s access to built, natural, human,and social capital. Societies with little accessto capital are constrained in their capacity toadapt. People threatened with starvation, forexample, may degrade natural capital by over-grazing to meet their immediate food needs,thereby reducing their potential to cope withdrought or future food shortage. Rich coun-tries, on the other hand, have greater capac-ity to engineer solutions to cope with floods,droughts, and disease outbreaks. Natural cap-ital also contributes in important ways toadaptive capacity, although its role is oftenunrecognized until it has been degraded. Sys-tems that have experienced severe soil erosion,for example, have fewer options with whichto experiment and innovate during times ofdrought, and highly engineered systems thathave lost their capacity to store floodwatershave fewer options to adapt in response tofloods. The role of human capital in adaptivecapacity is especially important. It is much morethan formal education. It depends on an under-standing of how the system responds to change,which often comes from experience and localknowledge of past responses to extreme eventsor stresses. As the world changes, and new haz-ards and stresses emerge, this understandingmay be insufficient. Willingness to innovate andexperiment to test what has been learned and toexplore new approaches is crucial to adaptivecapacity.

Social capital through networking to select,communicate, and implement potential solu-tions is another key component of adaptivecapacity. Leadership, for example, is often crit-ical in building trust, making sense of complexsituations, managing conflict, linking actors, ini-tiating partnerships among groups, compilingand generating knowledge, mobilizing broad


support for change, and developing and com-municating visions for change (Folke et al. 2005;see Chapter 5). It takes more than leaders,however, for society to adapt to change. Socialnetworks are critical in effectively mobilizingresources at times of crisis (e.g., war or floods)and in providing a safety net for vulnerable seg-ments of society (see Chapters 4 and 5).

In the context of sustainability, adaptivecapacity represents the capacity of a social–ecological system to make appropriate substi-tutions among forms of capital to maintain orenhance inclusive wealth. In this way the sys-tem retains the potential for future generationsto meet their needs.

Resilience

Resilience is the capacity of a social–ecologicalsystem to absorb a spectrum of shocks orperturbations and to sustain and develop itsfundamental function, structure, identity, andfeedbacks through either recovery or reor-ganization in a new context (Holling 1973,Gunderson and Holling 2002, Walker et al.2004, Folke 2006). The unique contribution ofresilience theory is the recognition and identifi-cation of several possible system properties thatfoster renewal and reorganization after pertur-bations (Holling 1973). Resilience depends on(1) adaptive capacity (see above); (2) biophysi-cal and social legacies that contribute to diver-sity and provide proven pathways for rebuild-ing; (3) the capacity of people to plan for thelong term within the context of uncertainty andchange; (4) a balance between stabilizing feed-backs that buffer the system against stressesand disturbance and innovation that createsopportunities for change; and (5) the capacityto adjust governance structures to meet chang-ing needs (Holling and Gunderson 2002, Folkeet al. 2003, Walker et al. 2006; Table 1.3). Lossof resilience pushes a system closer to its lim-its. When resilience has been eroded, a distur-bance, like a disease, storm, or stock marketfluctuation, that previously shook and revital-ized the resilient system, might now push thefragile system over a threshold into an alter-native state (a regime shift) with a new trajec-

tory of change. Such system changes radicallyalter the flow of ecosystem services (Chapter 2)and associated livelihoods and well-being ofpeople and societies. Clearly, resilience is anessential feature of resource stewardship underconditions of uncertainty and change, so thisapproach to resource management is even moreimportant today than it has been in the past.

We have already discussed the role of sta-bilizing feedbacks in buffering systems fromchange and the role of adaptive capacity incoping with the impacts of those changes thatoccur. Sources of diversity, which is essen-tial for adaptation, are especially important inthe focal system and surrounding landscape attimes of crisis, i.e., during the renewal phaseof adaptive cycles, when there is less resis-tance to establishment of new entities. Fosteringsmall-scale variability and change logicallycontributes to resilience because it maintainswithin the system those components that arewell adapted to each phase of the adaptivecycle—ranging from the renewal to the con-servation phase. This reduces the likelihoodthat the inevitable disturbances will have catas-trophic effects. Conversely, preventing small-scale disturbances such as insect outbreaks orfires tends to eliminate disturbance-adaptedcomponents, thereby reducing the capacity ofthe system to cope with disturbance.

Biophysical and social legacies contribute toresilience through their contribution to diver-sity. Legacies provide species, conditions, andperspectives that may not be widely repre-sented in the current system. A buried seedpool or stems that resprout after fire, for exam-ple, give rise to a suite of early successionalspecies that are well adapted to postdisturbanceconditions but may be uncommon in the matureforest. Similarly, the stories and memories ofelders and the written history of past eventsoften provide insight into ways in which peoplecoped with past crises as well as ideas for futureoptions that might not otherwise be considered.This often occurs by drawing on social memory,the social legacies of knowing how to do thingsunder different circumstances. A key challengeis how to foster and maintain social memory attimes of gradual change, so it is available whena crisis occurs.


One of the key contributions of resiliencetheory to resource stewardship is the recogni-tion that complex adaptive systems are con-stantly changing in ways that cannot be fullypredicted or controlled, so decisions mustalways be made in an environment of uncer-tainty. Research and awareness of processesoccurring at a wide range of scales (e.g.,the dynamics of potential pest populations orbehavior of global markets) can reduce uncer-tainty (Adger et al. 2005, Berkes et al. 2005),but managing for flexibility to respond to unan-ticipated changes is essential. This contrastswith steady-state management approaches thatseek to reduce variability and change as a wayto facilitate efficient harvest of a given resourcesuch as fish or trees (Table 1.1).

Transformability and Regime Shifts

Transformability is the capacity to reconcep-tualize and create a fundamentally new sys-tem with different characteristics (Walker et al.2004; see Chapter 5). There will always be acreative tension between resilience (fixing thecurrent system) and transformation (seeking anew, potentially more desirable state) becauseactors in the system usually disagree aboutwhen to fix things and when to cut losses andmove to a new alternative structure (Walkeret al. 2004). Actively navigated transformationsrequire a paradigm shift that reconceptualizesthe nature of the system. During transforma-tion, people recognize (or hypothesize) a fun-damentally different set of critical slow vari-ables, internal feedbacks, and societal goals.Unintended transformations can also occur insituations where management efforts have pre-vented adjustment of the system to changingconditions, resulting in a fundamentally differ-ent system (often degraded) characterized bydifferent critical slow variables and feedbacks.The dividing line between persistence of a givensystem and transformation to a new state issometimes fuzzy. Total system collapse seldomoccurs (Turner and McCandless 2004, Diamond2005). Nonetheless, actively navigated trans-formations of important components of a sys-tem are frequent (e.g., from an extractive to

a tourism-based economy). In general, diver-sity, adaptive capacity, and other componentsof resilience enhance transformability becausethey provide the seeds for a new beginning andthe adaptive capacity to take advantage of theseseeds.

Transformations are often triggered by crisis,so the capacity to plan for and recognize oppor-tunities associated with crisis contributes totransformability (Gunderson and Holling 2002,Berkes et al. 2003). Crisis is a time when soci-ety, by definition, agrees that some componentsof the present system are dysfunctional. Duringcrisis, society is more likely to consider novelalternatives. It is also a time when, if novel solu-tions are not seized, the system can becomeentrenched in the very policies that led to crisis,increasing the likelihood of unintended trans-formations. Climate-induced increases in wild-fires in the western USA, for example, threatenhomes that have been built in the wildland–urban interface. One potential transformationwould be policies that cease assuming publicresponsibility for private homes built in remotefire-prone areas and instead encouraged moredense development of areas that could be pro-tected from fire and served by public trans-portation. This would reduce the need and costof wildfire suppression, increase the economicefficiency of public transportation, and reducethe use of fossil fuels. Alternatively, currentpolicies of fire suppression and dispersed res-idential development in forested lands mightpersist and magnify the risk of catastrophicloss of life and property as climate warmingincreases wildfire risk and fire suppression leadsto further fuel accumulation.

Sometimes systems exhibit abrupt transi-tions (regime shifts) to alternate states becauseof threshold responses to persistent changes inone or more slow variables. Continued phos-phorus inputs to clearwater lakes, for example,may lead to abrupt transitions to a turbid-wateralgal-dominated regime (Carpenter 2003). Sim-ilarly, persistent overgrazing can cause shrubencroachment and transition from grasslandto shrubland (Walker et al. 2004). Regimeshifts are large changes in ecosystems thatinclude both changes in stability domains ofa given system (e.g., clearwater–turbid-water


transitions; Fig. 1.7d) and system transforma-tions (Carpenter 2003, Groffman et al. 2006).

Challenges to Sustainability

The major challenges to sustainability varytemporally and regionally. Issues of sustainabil-ity are often prominent in developing nations,especially where substantial poverty, inade-quate educational opportunities, and insuffi-cient health care limit well-being (Kaspersonet al. 2005). These situations sometimes coin-cide with a high potential for environmentaldegradation, for example, soil erosion and con-tamination of water supplies, as people try tomeet their immediate survival needs under cir-cumstances of inadequate social and economicinfrastructure. Sustainable development seeksto improve well-being, while at the same timeprotecting the natural resources on which soci-ety depends (WCED 1987). In other words,it seeks directional changes in some under-lying controls, but not others. Questions areoften raised about whether sustainable devel-opment can indeed be achieved, given its twingoals of actively promoting economic devel-opment while sustaining natural capital. Thefeasibility of sustainable development dependson the multiple effects of development on sys-tem properties and the extent to which thesenew system properties can be sustained overthe long term. In other words, how does devel-opment influence the slow variables that gov-ern the properties of social–ecological systemsand how can they be redirected or transformedfor improving the options of well-being withoutdegrading inclusive wealth? Finding sustain-able solutions usually requires active engage-ment of stakeholders (groups of people affectedby policy decisions) who must live with, andparticipate in, the implementation of potentialsolutions.

Enhancing the sustainability of nations withgreater wealth is equally challenging. Coun-tries such as the USA, for example, consumefossil fuels at per-capita rates that are fivefoldgreater than the world average and frequentlyuse renewable resources more rapidly than theycan be replenished. Here the challenge is toavoid degradation of the ecological and cultural

bases of well-being over the long term so thatpeople in other places and in future generationscan meet their own needs (Plate 3).

In summary, virtually all social–ecologicalsystems are undergoing persistent directionalchanges, as a result of both unplanned changesin climate, economic systems, and culture anddeliberate planning to improve well-being.Efforts to promote sustainability must there-fore recognize that many of the attributes ofsocial–ecological systems will inevitably changeover the long term and seek ways to guide thesechanges along sustainable pathways.

Roadmap to SubsequentChapters

The first section of the book presents the gen-eral principles needed for sustainable stew-ardship in a changing world (Table 1.4).Chapter 1 provides a framework for under-standing change and the factors that influ-ence sustainability under conditions of change.A clear message from this chapter is thatsocial–ecological systems are complex andrequire an understanding of the interactionsamong ecological, economic, political, and cul-tural processes. Consequently, key resource-management issues cannot be solved by dis-ciplinary experts but require an integratedunderstanding of many disciplines. Chapter 2describes the principles of ecosystem manage-ment to sustain the delivery of ecosystem ser-vices to society. Chapter 3 describes the rangeof economic, cultural, and political factors thatshape well-being and use of ecosystem ser-vices. Chapter 4 then describes the institutionaldimensions of human interactions with ecosys-tems. Chapter 5 explores the processes by whichsocial–ecological systems transform to a fun-damentally different system with different con-trols and feedbacks.

The second section of the book applies thegeneral principles developed in the first sec-tion to specific types of social–ecological sys-tems and their prominent resource–stewardshipchallenges (Table 1.4), including conservation(see Chapter 6), forests (see Chapter 7),drylands (see Chapter 8), lakes and rivers


Table 1.4. Resource–stewardship challenges and the chapters in which each isemphasized.

Issue Chapter where emphasized

Social-ecological interactions All chapters (2–15)Global change Concepts (2–5), Global (14), Systems (6–13)Ecological sustainability Ecosystems (2), System chapters (6–14)Ecosystem restoration Ecosystems (2), Drylands (8)Biodiversity conservation Ecosystems (2), Conservation (6), Forests (7)Invasive species Ecosystems (2), Freshwaters (9)Landscape management Ecosystems (2), Drylands (8), Freshwaters (9)Range management Ecosystems (2), Drylands (8)Wildlife management Ecosystems (2), Conservation (6), Drylands (8)Fisheries management Freshwaters (9), Oceans (10), Coastal (11)Water management Ecosystems (2), Drylands (8), Freshwaters (9)Disturbance management Ecosystems (2), Forests (7), Freshwaters (9)Pollution Ecosystems (2), Agriculture (12), Cities (13)Urban development Livelihoods (3), Forests (7), Cities (13)Sustaining human livelihoods Livelihoods (3), Conservation (6), Coastal (11)Social and environmental justice Livelihoods (3), Coastal (11), Cities (13), Global (14)Sustainable development Livelihoods (3), Agriculture (12)Local and traditional knowledge Institutions (4), Conservation (6), Drylands (8)Property rights and the commons Institutions (4), Oceans (10), Coastal (11)Natural resource policy Institutions (4), System chapters (6–14)Subsistence harvest Institutions (4), Conservation (6)Resource co-management Institutions (4), Conservation (6), Coastal (11)Adaptive management Institutions (4), Drylands (8), Oceans (10)Long-term planning Transformation (5), Forests (7), Global (14)Managing thresholds Transformation (5), Drylands (8), Oceans (10)Adaptive governance Transformation (5), Forests (7), Global (14)Thresholds and regime shifts Transformation (5), Drylands (8), Freshwaters (9)

(see Chapter 9), oceans and estuaries (seeChapters 10 and 11), food production sys-tems (see Chapter 12), cities and suburbs (seeChapter 13), and the entire Earth (see Chap-ter 14). Each of these chapters describes thesystem properties and dynamics that are espe-cially important in that system, key manage-ment issues, and potential social–ecologicalthresholds. Each chapter then describes a fewcase studies that illustrate resilient or non-resilient management and outcomes and howthe unique properties of each system shapehuman–environment interactions and sustain-ability constraints and opportunities. Eachsystem chapter emphasizes selected generalprinciples that were described in the firstsection of the book.

The final chapter (see Chapter 15) summa-rizes some of the major strategies that haveproven valuable for managing social–ecological

systems and the lessons learned from previouschapters about the role of resilience and adap-tation in sustainable stewardship.

Review Questions

1. What is resilience-based resource steward-ship? How does it differ from steady-stateresource management, and why are thesedifferences important in the current world?

2. How do different types of feedbacks influ-ence the stability and resilience of a system?

3. What are the mechanisms by which com-plex adaptive systems respond to changes?Do they always respond in the same way toa given perturbation? Why or why not? Insocial–ecological systems, why does a givenpolicy sometimes have different effects whenimplemented at different times or places?


4. Why does the sensitivity of social–ecologicalsystems to perturbations depend on the timesince the previous perturbation? What arethe advantages and disadvantages of man-aging systems to prevent disturbances fromoccurring?

5. What are the processes by which vulnerabil-ity, adaptive capacity, resilience, and trans-formability influence sustainability?

Additional Readings

Berkes, F., J. Colding, and C. Folke, editors. 2003.Navigating Social-Ecological Systems: BuildingResilience for Complexity and Change. Cam-bridge University Press, Cambridge.

Carpenter, S.R., and M.G. Turner. 2000. Hares andtortoises: Interactions of fast and slow variablesin ecosystems. Ecosystems 3:495–497.

Chapin, F.S., III, A.L. Lovecraft, E.S. Zavaleta, J.Nelson, M.D. Robards, et al. 2006. Policy strate-gies to address sustainability of Alaskan borealforests in response to a directionally changingclimate. Proceedings of the National Academy ofSciences 103:16637–16643.

Folke, C. 2006. Resilience: The emergence of aperspective for social-ecological systems analysis.Global Environmental Change 16:253–267.

Gunderson, L.H., and C.S. Holling, editors. 2002.Panarchy: Understanding Transformations inHuman and Natural Systems. Island Press,Washington.

Levin, S.A. 1999. Fragile Dominion: Complexity andthe Commons. Perseus Books, Reading, MA.

MEA (Millennium Ecosystem Assessment). 2005d.Ecosystems and Human Well-being: Synthesis.Island Press, Washington.

Steffen, W.L., A. Sanderson, P.D. Tyson, J. Jager, andP.A. Matson, editors. 2004. Global Change and theEarth System: A Planet Under Pressure. Springer-Verlag, New York.

Turner, B.L., II, R.E. Kasperson, P.A. Matson, J.J.McCarthy, R.W. Corell, et al. 2003. A frameworkfor vulnerability analysis in sustainability science.Proceedings of the National Academy of Sciences100:8074–8079.

Walker, B.H., C.S. Holling, S.R. Carpenter, andA.P. Kinzig. 2004. Resilience, adaptability andtransformability in social–ecological systems.Ecology and Society 9(2):5 [online] URL:http://www.ecologyandsociety.org/vol9/iss2/art5/

Walker, B.H., and D. Salt. 2006. Resilience Thinking:Sustaining Ecosystems and People in a ChangingWorld. Island Press, Washington.

2Managing Ecosystems Sustainably:The Key Role of Resilience

F. Stuart Chapin, III

Introduction

The goal of ecosystem management is to pro-vide a sustainable flow of multiple ecosystemservices to society today and in the future. Asan integral component of natural resource stew-ardship, ecosystem management recognizes theintegrated nature of social–ecological systems,their inherent complexity and dynamics at mul-tiple temporal and spatial scales, and the impor-tance of managing to maintain future optionsin the face of uncertainty (Christensen et al.1996; Table 2.1)—i.e., many of the factorsgoverning the resilience and vulnerability ofsocial—ecological systems. As a society, wehave a poor track record of managing ecosys-tems sustainably in part because the short-term use of natural resources often receiveshigher priority than their long-term sustainabil-ity. Environmental degradation contributed tothe collapse of many advanced human soci-eties, including Babylon, the Roman Empire,and the Mayan Civilization (Turner et al. 1990,

F.S. Chapin (�)Institute of Arctic Biology, University of AlaskaFairbanks, Fairbanks, AK 99775, USAe-mail: [email protected]

Diamond 2005). More than half of the servicesprovided by ecosystems have declined globallyin the last half-century (MEA 2005a, d), rais-ing questions about the capacity of human soci-eties to manage ecosystems sustainably. Rapidrates of social and environmental change havemagnified the challenges of sustainable man-agement. We advocate broadening the conceptof ecosystem management to resilience-basedecosystem stewardship. Its goals are to respondto and shape change in social–ecological sys-tems in order to sustain the supply and oppor-tunities for use of ecosystem services by society.Resilience-based ecosystem stewardship buildson ecosystem management by emphasizing (1)the key role of resilience in fostering adapta-tion and renewal in a rapidly changing world;(2) the dynamics of social change in alter-ing human interactions with ecosystems; and(3) the social–ecological role of resource man-agers as stewards who respond to and shapesocial–ecological change. In this chapter weaddress key components of ecosystem steward-ship, emphasizing the ecological consequencesof those human actions that can tip the balancebetween sustainable and nonsustainable flow ofecosystem services to society. In Chapter 3, webroaden this perspective to integrate social pro-cesses that motivate human actions.


30 F.S. Chapin

Table 2.1. Attributes of ecosystem management. Information from Christensen et al. (1996).

Sustainability Intergenerational sustainability is the primary objectiveGoals Measurable goals are defined that assess sustainability of outcomesEcological understanding Ecological research at all levels of organization informs managementEcological complexity Ecological diversity and connectedness reduces risks of unforeseen changeDynamic change Evolution and change are inherent in ecological sustainabilityContext and scale Key ecological processes occur at many scales, linking ecosystems to their

matrixHumans as ecosystem components People actively participate in determining sustainable management goalsAdaptability Management approaches will change in response to changes in scientific

knowledge and human values

An ecosystem consists of organisms (plants,microbes, and animals—including people) andthe physical components (atmosphere, soil,water, etc.) with which they interact. All ecosys-tems are influenced, to a greater or lesser

degree, by social processes (i.e., are social–ecological systems), although ecosystem stud-ies tend to focus on biological interactions.Using the ecosystem-service framework devel-oped by the Millennium Ecosystem Assessment

Ecosystem stewardship

Supporting services

Ecosystem processesDiversity maintenance

Disturbance cycles

Human well-being

Freedom &choice

Climate regulationWater quality & quantity

Disease control

Regulating services

Cultural services

Cultural identityRecreation & tourism

Aesthetic &spiritual benefits

Food

Fuelwood

Water

Fiber

Biochemicals

Provisioning services

Ecosystem services Well-being

Figure 2.1. Linkages among ecosystem services,well-being of society, and ecosystem stewardship, aframework developed by the Millennium EcosystemAssessment (MEA, 2005c). Supporting services arethe foundation for the other categories of ecosystemservices that are directly used by society. In addi-

tion, the goods harvested by people are influencedby landscape processes, which include regulatory ser-vices, and, in turn, influence people’s connection tothe land and sea (cultural services). Adapted fromMEA (2005d).

2 Managing Ecosystems Sustainably 31

Table 2.2. General categories of ecosystem services and examples of the societal ben-efits that are most directly affected.

Ecosystem services Direct benefits to society

Supporting servicesMaintenance of soil resources Nutrition, shelterWater cycling Health, waste managementCarbon and nutrient cycling Nutrition, shelterMaintenance of disturbance regime Safety, nutrition, healthMaintenance of biological diversity Nutrition, health, cultural integrity

Provisioning servicesFresh water Health, waste managementFood and fiber Nutrition, shelterFuelwood Warmth, healthBiochemicals HealthGenetic resources Nutrition, health, cultural integrity

Regulating servicesClimate regulation Safety, nutrition, healthErosion, water quantity/quality, pollution Health, waste managementDisturbance propagation SafetyControl of pests, invasions, and diseases HealthPollination Nutrition

Cultural servicesCultural identity and cultural heritage Cultural integrity, valuesSpiritual, inspirational, and aesthetic benefits ValuesRecreation and ecotourism Health, values

(MEA 2005d), we first provide an overviewof supporting services, which are the funda-mental ecological processes that sustain ecosys-tem functioning (Fig. 2.1). We show how thedegradation of certain key supporting serviceserodes resilience, leading to loss of other ser-vices that are used more directly by society.These services include (1) provisioning services(or ecosystem goods), which are products ofecosystems that are directly harvested by soci-ety; (2) regulating services that influence soci-ety through interactions among ecosystems ina landscape; and (3) cultural services, whichare nonmaterial benefits that are important tosociety’s well-being (Table 2.2). There is broadoverlap among these categories of ecosystemservices, and different authors have thereforeclassified them in different ways. Traditionalfoods, for example, function as both provi-sioning services that provide nutritional ben-efits and cultural services that sustain cul-tural relationships to the land or sea. Theimportant point, however, is that the func-tioning of ecosystems benefits society in somany ways that human well-being cannot

be sustained without the effective function-ing of the ecosystems of which people area part.

Supporting Services: SustainingEcosystem Functioning

Supporting services are the fundamental eco-logical processes that control the structure andfunctioning of ecosystems. Managers and thepublic often overlook these services becausethey are not the products directly valued bysociety. Moreover, they are frequently con-trolled by variables that change relativelyslowly (i.e., slow variables) and are thereforetaken for granted by agencies tasked with amanaging a particular ecosystem good such astrees or fish. However, because of the funda-mental dependence of all ecosystem services onsupporting services, integrity of these servicesgenerally sustains many services that are val-ued more directly by society. In this section wefocus on the slow variables that most frequently

32 F.S. Chapin

control ecosystem processes and therefore abroad suite of ecosystem services.

Maintenance of Soil Resources

Soils and sediments are key slow variablesthat regulate ecosystem processes by providingresources required by organisms. The controlsover the formation, degradation, and resource-supplying potential of soils and sediments aretherefore central to sound ecosystem manage-ment and to sustaining the natural capital onwhich society depends (Birkeland 1999, Chapinet al. 2002). The quantity of soil in an ecosystemdepends largely on the balance between inputsfrom weathering (the breakdown of rocks toform soil) or deposition and losses from ero-sion. In addition, organisms, especially plants,add organic matter to soils through death of tis-sues and individuals, which is offset by lossesthrough decomposition. If an ecosystem wereat steady state, i.e., when inputs approximatelyequal outputs, the quantity of soil would remainrelatively constant, providing a stable capacityto supply vegetation with water and nutrients.Natural imbalances between inputs and outputslead to deeper soils in floodplains and at thebase of hills than on hilltops. When averagedover large regions, however, changes in soilcapital due to imbalances between inputs andoutputs usually occur slowly—about 0.1–10 mmper century (Selby 1993). In general, the pres-ence of a plant canopy and litter layer (thelayer of dead leaves on the soil surface) reduceserosion. Human activities that reduce vegeta-tion cover can increase erosion rates by severalorders of magnitude, just as occurs naturallywhen glaciers, volcanoes, or landslides reducevegetation cover. Under these circumstances,ecosystems can lose soils in years to decadesthat may have required thousands of years toaccumulate, causing an essentially permanentloss of the productive capacity of ecosystems.Similarly, human modification of river chan-nels can alter sediment inputs. In the southernUSA, for example, loss of sediment inputs andsubsequent soil subsidence led to the disappear-ance of barrier islands that had previously pro-tected New Orleans from hurricanes. The loss

of soil resources substantially reduces resilienceby reducing the natural capital by which social–ecological systems can respond to change; thisincreases the likelihood of a regime shift to amore degraded state.

The physical and chemical properties of soilsare just as important as total quantity of soilin determining the productive potential of ter-restrial ecosystems. Fine particles of mineralsoils (clay) and organic matter are particu-larly important in retaining water and nutri-ents (Brady and Weil 2001). Clay and organicmatter are typically concentrated near the soilsurface, where they are vulnerable to loss byerosion. Wind and overland flow (the move-ment of water across the soil surface) transportsmall particles more readily than large ones,tending to remove those soil components thatare particularly important in water and nutri-ent retention. Human activities that foster windand water erosion, such as deforestation, over-grazing, plowing, or fallowing of agriculturalfields, therefore erode the water- and nutrient-retaining capacity of soils much faster than thetotal loss of soil volume might suggest. Prevent-ing even modest rates of erosion is thereforecritical to sustaining the productive capacity ofterrestrial ecosystems.

Accelerated soil erosion is one of the mostserious causes of global declines in ecosys-tem services and resilience. The erosional lossof fine soil particles is the direct cause ofdesertification, soil degradation that occursin drylands (see Chapter 8). Desertificationcan be triggered by drought, reduced veg-etation cover, overgrazing, or their interac-tions (Reynolds and Stafford Smith 2002, Foleyet al. 2003a). When drought reduces vegeta-tion cover, for example, goats and other live-stock graze more intensively on the remain-ing vegetation. Extreme poverty and lack ofa secure food supply often prevent peoplefrom reducing grazing pressure at times ofdrought, because short-term food needs takeprecedence over practices that might preventerosion. Wetter regions can also experiencesevere erosional loss of soil, especially wherevegetation loss exposes soils to overland flow.The Yellow River in China, for example, trans-ports 1.6 billion tons of sediment annually from


agricultural areas in the loess plateau at itsheadwaters. Similar erosional losses occurredwhen grasslands were plowed for agriculture inthe USA during droughts of the 1930s, creatingthe dustbowl. Changes in social processes con-tribute substantially to regime shifts involvingsevere soil erosion.

Soil erosion from land represents a sedi-ment input to lakes and estuaries. At a globalscale the increased sediment input to oceansfrom accelerated erosion is partially offset bythe increased sediment capture by reservoirs.Therefore lakes, including reservoirs, and estu-aries are the aquatic ecosystems most stronglyaffected by terrestrial erosion. Especially inagricultural areas, these sediments representa large influx of nutrients (eutrophication) toaquatic ecosystems that can be just as problem-atic as the loss of productive potential on land(see Chapter 9).

Water Cycling

Water is the soil resource that is used in largestquantities by plants and which most frequentlylimits the productivity of terrestrial ecosys-tems. Water enters ecosystems as precipitation.Two of the major pathways of water loss aretranspiration, the “green water” that sup-ports terrestrial production, and runoff, whichreplenishes groundwater and aquatic ecosys-tems, the “blue water” sources that are oftentapped by people for domestic and indus-trial uses, irrigation, and hydropower (seeChapter 9). Consequently, there are inherenttradeoffs among the ecosystem services pro-vided by water cycling, and some of the biggestchallenges in water management result fromthese tradeoffs.

Climatic controls over water inputs in pre-cipitation place an ultimate constraint on quan-tities of water cycled through ecosystems.Within this constraint the partitioning of waterbetween transpiration and run off depends on(1) the degree of compaction of the soil surface,which influences water infiltration into the soil,(2) soil water-holding capacity, which dependson the quantity of soil and its particle-size dis-tribution (see Maintenance of Soil Resources),

and (3) the capacity of vegetation to transpirewater (Rockstrom et al. 1999).

Vegetation fosters infiltration and storagethrough several mechanisms. The plant canopyand litter reduce compaction by raindrops thatotherwise tend to seal soil pores. In addition,roots and soil animals associated with veg-etation create channels for water movementthrough the soil profile. By facilitating waterinfiltration (due to reduced compaction), water-holding capacity (due to production of soilorganic matter), and reduced soil erosion (dueto reduced overland flow), vegetation generatesstabilizing feedbacks that sustain the productivepotential of soils and reduce ecosystem vul-nerability to drought. Human activities oftendisrupt these stabilizing feedbacks, therebyreducing resilience. For example, high densi-ties of livestock or movement of heavy farmmachinery at times (spring) or places (ripar-ian corridors) where soils are wet can compactsoils and reduce infiltration. Plowing reducessoil organic content substantially—often by50% within a few years—thereby reducing soilwater-holding capacity and the capacity of soilsto support crop growth with natural rainfall(Matson et al. 1997). Alternative agriculturalpractices that conserve or rebuild soil organiccontent (e.g., no-till agriculture) or reducecompaction by livestock or equipment underwet conditions therefore increase the capac-ity of soils to supply water to crops or othervegetation.

Transpiration is tightly linked to the capac-ity of plants to fix carbon and therefore totheir productive potential. This explains whyproductive agricultural systems are such prodi-gious consumers of water and why stream-flow increases after logging. At a more subtlelevel, any factor that increases the productivepotential of vegetation (e.g., nutrient additionsfrom fertilizers, introduction of exotic nitrogen-fixing species; atmospheric deposition of nitro-gen; replacement of shrublands by forests)will increase transpiration (green water flows)and reduce water movement to groundwaterand runoff (blue water flows). The tradeoffsbetween transpiration and runoff have implica-tions for the role of ecosystems in regulatingwater flow, as discussed later.

34 F.S. Chapin

Carbon and Nutrient Cycling

Within a climate zone, the availabilities ofbelowground resources (water and nutrients)are the main factors that constrain terrestrialcarbon cycling and ecosystem productivity. Thecarbon, nutrient, and water cycles of terrestrialecosystems are tightly linked (Fig. 2.2; Chapinet al. 2002, 2008). The major controls are (1) cli-mate, which governs water inputs, rates of soildevelopment, and cycling rates of carbon, nutri-ents, and water; (2) the water- and nutrient-holding capacity of soils (see Maintenance of

Climate Feedbacks

Carbon balance Energy balance Water balance

Photo-synthesis

Respiration

Solar radiation

EvapotranspirationSensible

heatRunoff

Albedo

Longwaveradiation

Figure 2.2. Three major categories of climate feed-backs (each shown by the arrows beneath thebracket) between ecosystems and the climate sys-tem. Carbon balance is the difference between CO2

uptake by ecosystems (photosynthesis) and CO2 lossto the atmosphere by respiration and disturbance.Energy balance is the balance between incomingsolar radiation, the proportion of this incoming solarradiation that is reflected (albedo), and the transferof the absorbed radiation to the atmosphere as sen-sible heat (warming the surface) or evapotranspira-tion (cooling the surface). Longwave radiation fromthe ecosystem or clouds depends on the tempera-ture of these surfaces. Water balance between theecosystem and atmosphere is the difference betweenprecipitation inputs and water return in evapotran-spiration; the remaining water leaves the ecosys-tem as runoff. Each of these ecosystem–atmosphereexchanges influences climate. Cooling effects on cli-mate are shown by black arrows; warming effects bygray arrows. Arrows show the direction of the trans-fer; the magnitude of each transfer differs amongecosystems. Redrawn from Chapin et al. (2008).

Soil Resources); and (3) the productive capac-ity of vegetation. In addition, carbon and nutri-ent cycles are physically linked because carbonforms the skeleton of organic compounds thatcarry nitrogen and phosphorus among plants,animals, and soils.

Because both water and nutrient availabil-ity depend on the fine particles in soils, thefactors that sustain water cycling (i.e., mainte-nance of vegetation and prevention of erosion)also sustain nutrient cycles. This is typical ofmany of the synergies among ecosystem ser-vices: Management practices that protect thebasic integrity of ecosystem structure fostersustainability of multiple ecosystem services.This simplifies the task of ecosystem manage-ment, because most services “take care of them-selves” if ecosystem structure and functioningare not seriously disrupted.

Plants are an important control point in thecycling of carbon through ecosystems, becausethey are the entry point for carbon and deter-mine the chemistry of dead organic mat-ter that eventually becomes food for decom-posers. However, plant production in mostintact ecosystems is limited by water and/ornutrient availability, so the productive capacityof vegetation typically adjusts to the availabil-ity of water and nutrients that a given climateand soil type provide. Consequently, acrossa broad range of ecosystem types, vegetationabsorbs most of the nutrients that are releasedby the decomposition (chemical breakdownof dead organic matter by soil organisms).Consequently, groundwater and runoff leavingthese systems have relatively low concentra-tions of nutrients. If, however, plant productionis reduced below levels that the climate andsoils can support, as for example in a fallow fieldor overgrazed pasture, or if nutrients are addedto the system at rates that exceed the absorptivecapacity of the vegetation, the excess nutrientsleave the system in groundwater and runoff oras trace gases to the atmosphere (e.g., N2O, apotent greenhouse gas that contributes to cli-mate warming). Certain nitrogen-fixing plantspecies form mutualistic relationships with soilmicroorganisms that convert atmospheric nitro-gen to plant-available nitrogen. The expansionof soybean and other nitrogen-fixing species has


substantially increased nitrogen inputs to manyagricultural regions (see Chapter 12). Indus-trial fixation of nitrogen, primarily to producefertilizers, is an even larger source of nitro-gen inputs to managed ecosystems. Ammoniavolatilizes from fertilizers and cattle urine andenters downwind ecosystems in precipitation.In addition, fossil-fuel combustion producesnitrogen oxides (NOx) that are a major compo-nent of acid rain. Together these anthropogenicsources of nitrogen have doubled the naturallyoccurring rates of nitrogen inputs to terrestrialecosystems (Fig. 2.3; Schlesinger 1997, Vitouseket al. 1997). This massive change in global bio-geochemistry weakens the internal stabilizingfeedbacks that confer resilience to ecosystemprocesses (see Chapter 1). Reducing nitrogeninputs to levels consistent with the productivecapacity of vegetation, for example, by reducingfertilizer applications, reducing air pollution, orpreventing the spread of nitrogen-fixing exoticspecies, reduces the leakage of nitrogen fromterrestrial to aquatic ecosystems.

Carbon and nutrient cycles in aquatic ecosys-tems run on the leftovers of terrestrial nutrientcycles. In intact landscapes with tight terrestrialnutrient cycles, the small quantity of nutrientsdelivered to streams spiral slowly downstream,moving through decomposers, stream inverte-brates, algae, and fish (Vannote et al. 1980).Lakes are typically more nutrient-impoverishedthan streams, because they receive relatively

150

100

50

1960 1970 1980 1990

Time

Terr

estr

ial N

fixa

tio

n (

Tg y

r–1) Total anthropogenic

N fixation

Range of estimates ofnatural N fixation N fertilizer

Fossil fuel

Legume crops

Figure 2.3. Anthropogenic fixation of nitrogen interrestrial ecosystems over time compared with therange of estimates of natural biological nitrogen fix-ation on land. Redrawn from Vitousek et al. (1997).

little leaf litter and groundwater per unit ofwater surface, and sediments chemically fixmuch of the phosphorus that enters the lake(see Chapter 9). Aquatic organisms are welladapted to these low-nutrient conditions. Algaeefficiently absorb nutrients from the water col-umn, are eaten by invertebrate grazers thatin turn are eaten by fish. When phosphorusinputs to lakes exceed the chemical fixationcapacity of sediments, algae grow and repro-duce more rapidly than grazers can consumethem, reducing water clarity and increasing therain of dead organic matter to depth. Herebottom-dwelling decomposers break down thedead organic matter, depleting oxygen belowlevels required by fish, which causes fish todie. The high phosphorus-fixation capacity oflake sediments makes most lakes quite resilientto individual eutrophication events. Once thisphosphorus-sequestration mechanism is satu-rated, however, the sediments become a sourcerather than a sink of phosphorus, causing thelake to shift to a eutrophic state (Carpenteret al. 1999, Carpenter 2003; see Chapter 9).

Estuaries and the coastal zone of oceans,which are the final dumping ground of terres-trially derived nutrients, are typically quite pro-ductive. The rapid decomposition and nutrientrelease from sediments supports a productivebottom fishery and a rich nutrient source forthe overlying water column. Chesapeake Bay,for example, was historically an extremely richfishery that supported dense populations ofNative Americans and later of European set-tlers. Just as in lakes, however, excessive nutri-ent and organic matter inputs to estuaries canbe too much of a good thing, depleting oxygenand killing the organisms that would other-wise decompose and recycle the accumulatingorganic matter. The Mississippi River Delta, forexample, has undergone a regime shift froma productive shrimp fishery to a dead zonewith insufficient oxygen to support much bio-logical activity (Rabalais et al. 2002). The bio-logical mechanisms that limit the resilience oflakes and estuaries are well understood, but thefailure of social–ecological systems to preventregime shifts demonstrates the need to incorpo-rate social processes into management planning(see Chapters 3 and 4).

36 F.S. Chapin

Carbon and nutrient cycling in the water col-umn of deep ocean basins is similar to thatdescribed for lakes, except that it is often evenmore nutrient-impoverished, because of thelarge vertical separation of ocean sedimentsfrom the surface where algal production occurs(Valiela 1995). Productive fisheries are oftenconcentrated in zones of upwelling where deepnutrient-rich water moves to the surface nearthe edges of continental shelves or at high lat-itudes, where wind-driven mixing is most pro-nounced.

Maintenance of Biological Diversity

The known effects of biodiversity on ecosys-tem functioning relate most strongly to traitsthat govern the effect of species on ecosystemprocesses and traits that govern the responseof species to environmental variation. Theseknown effects of biodiversity change (bothloss of key species or invasion of species withlarge impacts) can be fostered by maintainingspecies that span the spectrum of effect diver-sity and response diversity present in the system(Elmqvist et al. 2003, Suding et al. 2008).

Keystone species are species that have dis-proportionately large effects on ecosystems,typically because they alter critical slow vari-ables. Ecosystems are most likely to sustaintheir current properties if current keystonespecies (or their functional equivalents) aremaintained, and new ones are not introduced.Species that influence the supply of growth-limiting resources generally have large effectson the functioning of ecosystems. Myrica faya,for example, is a nitrogen-fixing tree intro-duced to nitrogen-poor ecosystems of Hawaiiby the Polynesians. The resulting increasesin nitrogen inputs, productivity, and canopyshading eliminated many plant species fromthe formerly diverse understory of this for-est (Vitousek 2004). Highly mobile animals,such as salmon and sheep, act as keystonespecies governing nutrient supply by feedingin one place and dying or defecating some-where else. Similarly, species that modify dis-turbance regime exert strong effects on ecosys-

tem functioning through their effects on theadaptive cycle of release, renewal, and growth.The introduction of flammable grasses intoa tropical forest, for example, can increasefire frequency and trigger a regime shift fromforest to savanna (D’Antonio and Vitousek1992). One of the main ways that animalsaffect ecosystem processes is through physi-cal disturbance (Jones et al. 1994). Gophers,pigs, and ants, for example, disturb the soil,creating sites for seedling establishment andfavoring early successional species. Elephantstrample vegetation and remove portions oftree canopies, altering the competitive balancebetween trees and grasses in tropical savan-nas. Many keystone species exert their effect bymodifying species interactions, for example, byeating other species (e.g., forest pests) or com-peting or facilitating the growth of other species(Chapin et al. 2000). Conserving key functionaltypes (i.e., a group of species that have similareffects on ecosystem processes) and preventingthe invasion of novel functional types reducesthe likelihood of large changes in ecosystem ser-vices.

Diversity in the environmental responseof species can stabilize ecosystem processes.Many species in a community appear func-tionally similar, for example, algal species ina lake or canopy trees in a tropical forest(Scheffer and van Nes 2004). What are theecosystem consequences of changes in diver-sity within a functional type (i.e., functionalredundancy)? Differences in environmentalresponses among functionally similar speciesprovide resilience by stabilizing rates of ecosys-tem processes (McNaughton 1977, Chapin andShaver 1985). In midlatitude grasslands, forexample, cool-season grasses are particularlyproductive under cool, moist conditions, andwarm-season grasses under hot, dry conditions.As environmental conditions fluctuate withinand among years, different species attain a com-petitive advantage over other functionally sim-ilar species (i.e., other grasses), thus stabilizingrates of ecosystem processes by the entire com-munity (Ives et al. 1999).

The functional redundancy associated withspecies diversity also provides insurance against


more drastic changes in environment, such asthose that may occur in the event of humanmismanagement of ecosystems or change inclimate. Radical changes in environment areunlikely to eliminate all species of a given func-tional type in a diverse ecosystem, allowing the

surviving species to increase in abundance andmaintain functions that might otherwise be seri-ously compromised. Overgrazing in Australiangrasslands, for example, eliminated the domi-nant species of palatable grass, severely reduc-ing the quantity of cattle that the ecosystem

Table 2.3. Examples of biodiversity effects on ecosystem services. We separate the diversity effects into thosedue to functional composition, numbers of species, genetic diversity within species, and landscape structureand diversity. Modified from Dıaz et al. (2006).

Ecosystem service Diversity component and mechanism

1. Production by societallyimportant plants

Functional composition: (a) fast-growing species produce more biomass;(b) species differ in timing and spatial pattern of resource use(complementarity allows more resources to be used)

Species number: large species pool is more likely to contain productive species2. Stability of crop production Genetic diversity: buffers production against losses to pests and environmental

variabilitySpecies number: Cultivation of multiple species in the same plot maintains high

production over a broader range of conditionsFunctional composition: species differ in their response to environment and

disturbance, stabilizing production3. Maintenance of soil

resourcesFunctional composition: (a) fast-growing species enhance soil fertility; (b) dense

root systems prevent soil erosion4. Regulation of water quantity

and qualityLandscape diversity: Intact riparian corridors reduce erosion.Functional composition: Fast-growing plants have high transpiration rates,

reducing stream flow5. Pollination for food

production and speciessurvival

Functional composition: Loss of specialized pollinators reduces fruit set anddiversity of plants that reproduce successfully

Species number: Loss of pollinator species reduces the diversity of plants thatsuccessfully reproduce (genetic impoverishment)

Landscape diversity: Large, well-connected landscape units enable pollinatorsto facilitate gene flow among habitat patches

6. Resistance to invasivespecies with negativeecological/cultural effects

Functional composition: Some competitive species resist the invasion of exoticspecies

Landscape structure: Roads can serve as corridors for spread of invasivespecies; natural habitat patches can resist spread

Species number: Species-rich communities are likely to have less unusedresources and more competitive species to resist invaders

7. Pest and disease control Genetic diversity or species number: Reduces density of suitable hosts forspecialized pests and diseases

Landscape diversity: Provides habitat for natural enemies of pests8. Biophysical climate

regulationFunctional composition: Determines water and energy exchange, thus

influencing local air temperature and circulation patternsLandscape structure: Influences convective movement of air masses and

therefore local temperature and precipitation9. Climate regulation by

carbon sequestrationsLandscape structure: Fragmented landscapes have greater edge-to-area ratio;

edges have greater carbon lossFunctional composition: Small, short-lived plants store less carbonSpecies number: High species number reduces pest outbreaks that cause

carbon loss10. Protection against natural

hazards (e.g., floods,hurricanes, fires)

Landscape structure: Influences disturbance spread and/or protection againstnatural hazards

Functional composition: (a) extensive root systems prevent erosion anduprooting; (b) deciduous species are less flammable than evergreens

38 F.S. Chapin

could support and exposing the soil to winderosion. Fortunately, a previously rare grassspecies that was less palatable, survived theovergrazing and increased in abundance, whenthe dominant grass declined, thus maintaininggrass cover and reducing potential degradationfrom erosion (Walker et al. 1999). These exam-ples of diversity effects on resilience providehints of the general importance of diversity instabilizing ecosystem processes and associatedecosystem services (Table 2.3) and suggest thatmanagement that sustains diversity is critical tolong-term sustainability.

Maintenance of Disturbance Regime

Disturbance shapes the long-term fluctuationsin the structure and functioning of ecosystemsand therefore their resilience and vulnerabil-ity to change. Disturbances are relatively dis-crete events that alter ecosystem structure andcause changes in resource availability or physi-cal environment (Pickett and White 1985). Dis-turbance is not something that “happens” toecosystems but is an integral part of their func-tioning. Species are typically adapted to the dis-turbance regime (i.e., the characteristic sever-ity, frequency, type, size, timing, and intensityof disturbance) that shaped their evolution-ary histories. Grassland and boreal species, forexample, resprout rapidly after fire or havereproductive strategies that enable them to col-onize recent burns (Johnson 1992). In contrast,many tropical tree species produce a pool ofyoung seedlings that grow slowly in the under-story, “waiting” until a hurricane or other eventcreates gaps in the canopy. The tropical treestrategy is poorly adapted to fire, which wouldkill the understory seedlings, and the borealtrees are poorly adapted to wind, which wouldleave an organic seedbed unfavorable for post-disturbance germination. Naturally occurringdisturbances such as fires and hurricanes aretherefore not “bad”; they are normal proper-ties of ecosystems and indeed are essential forthe long-term resilience of species and com-munity dynamics that characterize a particularecosystem.

The adaptive cycle that is triggered by dis-turbance both generates and depends uponlandscape patterns of biodiversity. After dis-turbance (release phase) and colonization(renewal phase), ecosystems undergo succes-sion (growth phase), a directional change inecosystem properties resulting from biologi-cally driven changes in resource supply. Succes-sion is accompanied by changes in the sizes andtypes of plants, leading to a diversity of foodand habitat for animals and soil microbes. Thesechanges in species composition and diversityboth cause and respond to the changing avail-ability of light, water, and nutrients as succes-sion proceeds, leading to characteristic changesin the cycling of carbon, water, and nutrientsand the associated supply of ecosystem services.The scale of this successional dynamic rangesfrom individual plants (e.g., gap-phase succes-sion in moist temperate and tropical forests)to extensive stands (e.g., flood plains or coniferforests characterized by large stand-replacingcrown fires) and from years (e.g., grasslands)to centuries (e.g., many forests). Subsequentrenewal of a disturbed patch draws on bothon-site legacies (e.g., buried seeds and surviv-ing individuals) and colonization from the sur-rounding matrix (Fig. 2.4; Nystrom and Folke2001, Folke et al. 2004). The resilience of theintegrated disturbance-renewal system dependson both a diversity of functional types capa-ble of sustaining the characteristic spectrumof ecosystem functions (effect diversity) andfunctional redundancy within functional types(response diversity). In coral reefs, for exam-ple, storms cause local extinctions that arerepopulated by dispersal from the surroundingmatrix. If overfishing or eutrophication elimi-nates some grazer species, such as parrot fishthat remove invading algae to produce space forrecolonizing coral larvae, the grazer-functionalgroup is less likely to provide the conditions forsuccessful coral recruitment. In the absence ofparrot fish, algae overgrow the corals, leadingto a regime shift that supports substantially lessbiodiversity and ecosystem services (Bellwoodet al. 2004).

As climate-driven stresses become morepronounced, and local extinctions occur morefrequently, the functional redundancy and


Legacy system

Renewing system

Disturbance(opens space)

Matrix systems

Input from

mobile links

Current system

Distant systems

Loss

LossPrior system

Future system

??

Initial landscape

Figure 2.4. Roles of biodiversity in ecosystemrenewal after disturbance (Folke et al. 2004). A dis-turbance such as a fire, hurricane, volcanic erup-tion, or war opens space in a social–ecological sys-tem. In this diagram, each shape represents a dif-ferent functional group such as algal-grazing herbi-vores in a coral reef, and the different patterns ofshading represent species within a functional group.After disturbance, some species are lost, but an on-site legacy of surviving species serves as the start-ing point for ecosystem renewal. For example, afterboreal fire, about half of the vascular plant speciesare lost. The larger the species diversity of the pre-disturbance ecosystem, the more species and func-tional groups are likely to survive the disturbance.(In this figure all functional groups except “squares”survived the disturbance.) Landscape diversity of thematrix surrounding the patch is also important to

ecosystem renewal because it provides a reservoirof diversity that can recolonize the disturbed patchthrough the actions of mobile links (biological orphysical processes that link patches on a landscape).In this figure the “square” function was renewed bycolonization from the matrix surrounding the ecosys-tem. Through time some additional species may begained or lost. In addition, new functional groups(inverted triangles in this diagram) may invade froma distance. The greater the species diversity of thepatch, the less likely is this invasion (and associatedfunctional change). In summary, diversity in boththe patch and the surrounding matrix are essentialto maintaining ecosystem functioning over the longterm. Although we have described the importanceof diversity from an ecological perspective, the samelogic holds true for economic and institutional diver-sity (see Chapters 3 and 4).

biodiversity of the matrix become increasinglyimportant to landscape resilience (Elmqvistet al. 2003). Postage-stamp reserves in a matrix

of agricultural monoculture, for example, areless likely to sustain their functional diversitythan in a diverse landscape, especially during

40 F.S. Chapin

times of rapid environmental change (Fig. 2.4).A patchwork of hedge rows, forest patches, andriparian corridors in an agricultural landscapeor of urban gardens, cemeteries, and seminat-ural landscaping of lawns in cities can substan-tially increase landscape diversity and resilienceof human-dominated landscapes, even thoughthey may occupy only a tiny fraction of theland area (Colding et al. 2006; see Chapters12 and 13). In a rapidly changing, intensivelymanaged world, assisted migration of specieswith low migration potential can supplementlandscape biodiversity as a source of renewal(McLachlan et al. 2007). Intensively managedFennoscandian forests, for example, have lost70% of the insect and bird biodiversity in theirwood-decomposer food webs. Climate warm-ing now creates conditions that are conduciveto migration of more southerly wood decom-posers. This wood-dependent biodiversity couldbe regenerated if the niche becomes available(through protection of older stands and reten-tion of green and dead trees and coarse woodydebris after logging) and if the species canarrive [management of the matrix to fostermigration of late successional species supple-mented by assisted migration (introduction) ofsoutherly taxa; Chapin et al. 2007].

Society typically derives benefits from mostphases of a disturbance cycle. Disturbance itselfreduces population densities of certain pestsand diseases. Early successional stages are char-acterized by rapidly growing species that havetissues that are nutritious to herbivores, suchas deer, and fleshy fruits that are dispersed bybirds and harvested by people. Later succes-sional stages are dominated by species that pro-vide other types of goods, such as timber ormedicines. Swidden (slash-and-burn) agricul-ture is a cultural system that can be an integralsustainable component of some tropical forests,as long as the traditional disturbance regime ismaintained (see Chapter 8). Human activitiesthat alter disturbance regimes, however, mod-ify the suite of goods and services providedby the landscape. Prevention of small insectoutbreaks, for example, increases the continu-ity of susceptible individuals and therefore theprobability of larger outbreaks (Holling 1986).A command-and-control approach to resource

management that prevents disturbances charac-teristic of an ecosystem often reduces resilienceat regional scales by producing new conditionsat all phases of the successional cycle to whichlocal organisms are less well adapted (Hollingand Meffe 1996; see Chapter 4). Decisions thatalter disturbance regimes tend to address onlythe short-term benefits to society, such as floodcontrol, pest control, and fire prevention, andignore the broader context of changes thatpropagate through all phases of the disturbancecycle.

Ecosystem “Restoration”:The Reconstruction of DegradedEcosystems

In degraded ecosystems such as abandonedmines and degraded wetlands, active transfor-mation to a more desirable state may be acentral management goal. In this case, intro-duction of new functional types can fostertransformation for ecosystem reconstructionor renewal (Bradshaw 1983). Introduction ofnitrogen-fixing trees to abandoned mine sitesin England, for example, greatly enhanced theaccumulation of soil nitrogen and soil organicmatter, providing the soil resources necessaryto support forest succession. Planting of metal-tolerant grasses on metal-contaminated sitescan play a similar role. Planting of beachgrasses in coastal developments can stabilizesand dunes that might otherwise be eroded bywinds and storms. Similarly, planting of saltmarsh plants with different salinity tolerancesalong a salinity gradient can renew coastal saltmarshes that were eliminated by human distur-bance or by sediment deposition from upstreamland-use change.

At larger scales, the introduction of aPleistocene-like megafauna has been suggestedas a strategy to convert to moss-dominatedunproductive Siberian tundra into a moreproductive steppe-like ecosystem. This regimeshift could support greater animal produc-tion and partially compensate for the loss ofeconomic subsidies to indigenous communi-ties in the post-Soviet Russian North (Zimovet al. 1995). Parks in cities are savanna-like


environments that provide important culturalservices for all residents, and city gardens pro-vide valuable nutritional and cultural benefitsto people who have moved to cities from ruralagricultural areas (Colding et al. 2006). In aworld dominated by rapid human populationgrowth and directional environmental change,the deliberate introduction of new functionaltypes creates path dependence for ecosystemtransformation to a new, potentially more desir-able state (Choi 2007). However, deliberatespecies introductions have a history of creat-ing unintended undesirable side effects. Exoticgrasses used to stabilize roadsides may expandinto adjacent ecosystems, or biological con-trol agents may expand their diet to nontargetspecies. Consequently, the introduction of novelfunctional types to trigger ecosystem transfor-mation is a tool that requires caution and canoften be avoided through use of locally adaptedspecies and genotypes.

Provisioning Services: Providingthe Goods Used by Society

Provisioning services are the goods producedby ecosystems that are consumed by society.They are the most direct link between ecosys-tems and social systems and are therefore theecosystem properties that receive most directattention from managers and the public. Theyare fast variables that depend on supporting ser-vices in ecosystems and often exhibit rapid non-linear responses to fluctuations in environment.Large changes may be difficult to reverse ifthresholds in supporting services are exceeded.In this section we identify the major provision-ing services and discuss ways to sustain theirsupply.

Fresh Water

Water is the ecosystem service that is mostlikely to directly limit well-being in the twenty-first century. Although water is the most abun-dant compound on Earth, only a small fractionof it (0.1%) is available to people, primar-ily in lakes, rivers, and shallow groundwater

(see Chapter 9). People currently use 40–50%of available freshwater, with use projected toincrease to 70% by 2050 (Postel et al. 1996). Theshortage of clean water is particularly severein developing nations, where future populationgrowth and water requirements are likely tobe greatest. The projected increases in humandemands for fresh water will strongly impactaquatic ecosystems through eutrophication andpollution, diversion of fresh water for irrigation,and modification of flow regimes by dams andreservoirs (see Chapter 9).

Maintaining the essential role of intactecosystems in the hydrological cycle is the sin-gle most effective way to sustain the supplyand quality of fresh water for use by soci-ety. Intact ecosystems that surround reservoirsminimize sediment input, serve as a chemi-cal and biological filter that removes pollutantsand pathogenic bacteria, and buffer seasonalfluctuations in river flows, as described later.There are tradeoffs between the quantity andthe quality of water provided by ecosystems.Forest clearing is sometimes suggested as a wayto increase runoff and therefore the blue-waterflows that can be withdrawn for human use. Theclear-cutting of an experimental watershed atHubbard Brook in the northeastern USA didindeed increase runoff fourfold (Likens et al.1977). However, it also increased stream nitratefluxes 16-fold because of a fourfold increasein nitrate concentration—to levels exceedinghealth standards for drinking water and ledto loss of a spectrum of ecosystem servicesprovided by the intact forest. Understandingthe tradeoffs among water-related ecosystemservices derived from green-water and blue-water flows is critical to ecosystem stewardship(Rockstrom et al. 1999, Gordon et al. 2008).

Expansion of human populations into aridregions is often subsidized by tapping ground-water supplies that would otherwise be unavail-able to surface organisms. In dry regions80–90% of this water is used to support irri-gated agriculture, which can be highly produc-tive once the natural constraints of water limi-tation are removed. The substantial cost of irri-gated agriculture in turn creates incentives forintensive management with fertilizers and pes-ticides, leading to a cascade of associated social

42 F.S. Chapin

and economic consequences. The sustainabilityof irrigated agriculture depends on the rate ofwater use relative to resupply to the groundwa-ter and the downstream consequences of irri-gation (see Chapters 9 and 12). Many irrigatedareas are supported by fossil groundwater thataccumulated in a different climatic regime andis being removed much more rapidly than it isreplenished, a practice that clearly cannot besustained.

More than half of the water diverted forhuman consumption, industrial use, or agricul-ture is wasted. Most irrigation water, for exam-ple, evaporates rather than being absorbedby plants to support production. Managementactions that increase the efficiency of water useand/or reuse water for multiple purposes canincrease the effective water supply for humanuse without additional fresh-water diversionfrom ecosystems (see Chapter 9).

Food, Fiber, and Fuelwood

Management of ecosystems for the produc-tion of food, fiber, and fuelwood cause greaterchanges in ecosystem services and the globalenvironment than any other human activity.Humans have transformed 40–50% of the ice-free terrestrial surface to produce food, fiber,and fuelwood (Vitousek et al. 1986, Imhoffet al. 2004). We dominate (directly or indi-rectly) about a third of primary productivity onland and harvest fish that use 8% of ocean pro-duction (Myers and Worm 2003). Most of thenitrogen that people add to the environment isto support agriculture, either as fertilizer or asnitrogen-fixing crops. The global human popu-lation increased fourfold during the twentiethcentury to 6.1 billion people, with correspond-ing increases in the harvest of ecosystem goodsto feed, clothe, and house these people.

Two general categories of ecosystem changehave enabled food and fiber production tokeep pace with the growing human population.There has been intensification in use of existingagricultural areas through inputs of fertilizers,pesticides, irrigation, and energy-intensive tech-nology and extensification through land-useconversion or modification of existing ecosys-

tems to provide goods for human use. Meetingthe needs for food and fiber of the projected60–70% increase in human population by 2050will further increase the demands for agricul-tural and forestry production. Recent increasesin food production have come primarily fromintensification of agriculture, with much of theexpected future increase expected to comefrom extensification in marginal environmentswhere largest population increases may occur.There are critical tradeoffs between intensifi-cation, which often creates pollution problems,and extensification, which eliminates many ofthe services associated with natural ecosys-tems. Appropriate management can reduce theimpacts of intensive agriculture (Matson et al.1997). For example, no-till agriculture reducessoil disturbance and therefore the decompo-sition of soil organic matter, enhancing thewater- and nutrient-retaining capacity of soils(see Chapter 12). Careful addition of waterand nutrients to match the amounts and tim-ing of crop growth can substantially reducelosses to the environment. There are also waysin which the extensification of agriculture canminimize impacts on ecosystem services by con-sidering the landscape framework in which itoccurs. Swidden (slash-and-burn) agriculture intropical forests, for example, can provide foodin newly cleared lands and forest products inregenerating forests. With appropriate rotationlength, this practice has been sustained forthousands of years (Ramakrishnan 1992). How-ever, in areas where rotation length declinedfrom the traditional 30-year periodicity to10 years or less in response to recent humanpopulation growth, soils had insufficient timeto regain fertility, forest species with long lifecycles disappeared, and the system underwenta regime shift to intensive agriculture of cashcrops with radically different social and eco-logical properties (Ramakrishnan 1992). Just aswith water, some of the greatest opportunitiesto minimize tradeoffs associated with enhanc-ing agricultural production are to explore prac-tices that maximize the effectiveness of landsand resources to support food productionin ways that are consistent with ecologicalsustainability and local cultural norms andvalues.


About 70% of marine fisheries are overex-ploited (see Chapter 10). Much of this fish-ing pressure results from the globalization ofmarkets for fish and from perverse subsidiesthat enable fishermen to continue fishing evenfor stocks that would otherwise no longer beprofitable to harvest. This illustrates the impor-tance of social and economic factors in drivingincreased harvest of many provisioning servicesand a need for improved resource stewardshipof marine ecosystems to sustain provisioningservices.

Other Products

Ecosystems provide a diverse array of otherproducts that are specific to individual ecosys-tems and societies. These include aestheticallyand culturally valuable items such as flowers,animal skins, and shells. In addition, ecosys-tems constitute a vast storehouse of geneticpotential to deal with current and future condi-tions. This includes genes from traditional cul-tivars or wild relatives of crops and other wildspecies that produce products that benefit soci-ety (see Chapter 6). For example, about 25% ofcurrently prescribed medicines originate fromplant compounds that evolved as defensesagainst herbivores (Dirzo and Raven 2003) andhave substantial potential for bioprospectingin regions of high biodiversity (Kursar et al.2006).

Regulating Services: Sustainingthe Social–Ecological System

Regulating services influence processes beyondborders of ecosystems where they originate.They constitute some of the key cross-scalelinkages that connect ecosystems on a land-scape and integrate processes across temporalscales. They are, however, largely invisible tosociety and generally ignored by managers, sofailures to sustain regulating services often havedevastating consequences.

Climate Regulation

The cycling of water, carbon, and nutrientshas important climatic consequences. About25–40% of the precipitation in the Amazonbasin, for example, comes from water that isrecycled by evapotranspiration from terrestrialecosystems (Costa and Foley 1999). If tropi-cal forests were extensively cut and replacedby pastures with lower transpiration rates, thiscould lead to a warmer, drier climate moretypical of savanna, making forest regenera-tion more difficult (Foley et al. 2003b, Balaet al. 2007). At high latitudes, tree-coveredlandscapes absorb more radiation and trans-fer it to the atmosphere than does adjacentsnow-covered tundra. The northward move-ment of treeline 6,000 years ago is estimatedto have contributed half of the climate warm-ing that occurred at that time (Foley et al.1994). Extensive human impacts on ecosys-tems can have similar large effects. In WesternAustralia the replacement of native heath vege-tation by wheatlands increased regional albedo(reflectance). As a result, the dark heathlandsabsorbed more radiation than the cropland,causing air to warm and rise over the heath-land and drawing moist air from the adja-cent wheatlands. The net effect was a 10%increase in precipitation over heathlands and a30% decrease in precipitation over croplands(Chambers 1998). Many vegetation changes,if they are extensive, generate a climate thatfavors the new vegetation, making it diffi-cult to return vegetation to its original state(Chapin et al. 2008). This suggests that ecosys-tem integrity is critical to resilience of the cli-mate system at regional scales.

Ecosystems are also important sources andsinks of greenhouse gases that determinethe heat-trapping capacity of the atmosphereand therefore the temperature of our planet.Approximately half of the CO2 released byburning fossil fuels is captured and stored byecosystems—half on land and half in the ocean.The capacity of ecosystems to remove and storethis carbon therefore exerts a strong influenceon patterns and rates of climate change. Forestsand peatlands are particularly effective in stor-ing large quantities of carbon, in trees and soils,

44 F.S. Chapin

respectively. Maintaining the integrity of theseecosystem types or restoring them on degradedlands enhances the capacity of the terrestrialbiosphere to store carbon. Increased recogni-tion of the value of this climate regulatory ser-vice has led to a market in carbon credits foractivities that enhance the capacity of ecosys-tems to store carbon (see Chapter 7).

Regulation of Erosion, WaterQuantity, Water Quality, and Pollution

As discussed earlier (see Water), intact ecosys-tems regulate many water-related services bybuffering stream flows to prevent floods andsoil erosion and by filtering ground water toreduce pollutant concentrations (Rockstromet al. 1999). Many of the compounds in agri-cultural and urban runoff are identical or sim-ilar to compounds that naturally cycle throughecosystems and are therefore used by organ-isms to support their growth and reproduction.Ecosystems therefore have a natural capacityto absorb these pollutants, cleansing the air orwater in the process (see Chapter 9). Ecosys-tems also process some novel chemicals, withpotentially positive and negative environmen-tal consequences. Oil spills, for example, selectfor oil-degrading bacteria that use oil as anenergy source, although their capacity to doso is generally limited by nutrient availability.Polychlorinated-hydrocarbon pesticides (PCBs)are a potential energy source for those organ-isms that evolve resistance to their toxicity. Therapid evolution that typifies microbial popula-tions in soils and sediments sometimes selects forpopulations capable of degrading or convertingthese compounds to other products. Their activ-ity reduces pollutant concentration. Sometimes,however, thebreakdownproductsareevenmoretoxic to other organisms than was the origi-nal compound, as in the conversion of insec-ticide DDT to DDE, or are environmentallystable and accumulate in ecosystems, as inthe fat-soluble PCBs that accumulate in foodchains and have caused reproductive failure inmany marine birds (Carson 1962). Society there-fore cannot count on ecosystems to providea “quick fix” that solves pollution problems.

Those pollutants that are processed byecosystems frequently alter their structure,diversity, and functioning. The processing ofnitrogen derived from acid rain or agriculturalpollution, for example, increases productivity,altering the competitive balance and relativeabundance of species. The dominant plantsoften increase in size and abundance and out-compete smaller organisms, leading to a lossof species diversity. Moreover, as ecosystemscycle more nitrogen, soil nitrate concentrationsincrease, leading to emissions of more nitrousoxide (a potent greenhouse gas) to the atmo-sphere and leaching of nitrate to groundwater.As nitrate (an anion) leaches, it carries with it acation to maintain charge balance, reducing theavailability of cations such as calcium and mag-nesium, which can be replaced only by slow soilweathering. This is representative of many eco-logical responses to change, in which apparentlybeneficial effects of ecosystems (e.g., removal ofnitrogen-based pollutants or PCBs) can initiatea cascade of unanticipated consequences. Eco-logical research that recognizes the complexadaptive nature of ecosystems (see Chapter 1)can increase the likelihood of anticipating someof these effects as a basis for informed policydecisions.

Natural-Hazard Reduction

Adaptations of organisms to the characteris-tic disturbance regime of their environmentoften reduce the societal impacts of distur-bance. As discussed earlier, every ecosystemhas a particular disturbance regime to whichorganisms are adapted. However, these samedisturbances, such as hurricanes, wildfires, andfloods, often have negative societal impactson the built environment that people create.Incorporation of organisms adapted to a par-ticular disturbance regime into the built envi-ronment sometimes reduces the impact of dis-turbances when they occur. Kelp forests in theintertidal zone, for example, dissipate energyfrom storm surges and protect beaches fromcoastal erosion, just as beach grasses protectsands from wind erosion. Floodplain trees andshrubs reduce the speed of flood waters, leading


to deposition of sediments and reducing thewater energy that would otherwise cause ero-sional changes in channel morphology. Someof the greatest challenges in managing dis-turbance are to identify and separate thoselocations where disturbances have large nega-tive effects on human-dominated environments(e.g., towns and cities) from areas where dis-turbances have greater societal benefits and/orare most likely to occur. For example, concen-trating suburban development in areas that areunlikely to experience fire or flooding reducesrisks to the built environment. Similarly, allow-ing regular small disturbances (e.g., floods, pre-scribed fires, and insect outbreaks) to occurperiodically reduces the likelihood of largerdisturbances that are more difficult to control(Holling and Meffe 1996).

Regulation of Pests, Invasions, andDiseases

Biodiversity often enhances pest resistance inagricultural systems through both ecologicaland evolutionary processes (Dıaz et al. 2006).An increasing tendency in intensive agricul-ture is to reduce weeds, pests, and pathogensusing agrochemical pesticides. Due to their highpopulation densities and short life cycles, how-ever, insects and weeds typically evolve resis-tance to synthetic biocides within 10–20 years,necessitating continuing costly investments todevelop and synthesize new biocides as cur-rent products become less effective. An alter-native more resilient approach is to makegreater use of natural processes that regu-late pests. Increased genetic diversity of cropsnearly always decreases pathogen-related yieldlosses (Table 2.3). Recently, a major and costlyfungal pathogen of rice, rice blast, was con-trolled in a large region of China by plant-ing alternating rows of two rice varieties (Zhuet al. 2000). Similarly, a high diversity of cropspecies reduces the incidence or severity ofimpact of herbivores, pathogens, and weeds(Andow 1991, Liebman and Staver 2001, Dıazet al. 2006). Sometimes these diversity effectshave multiple benefits. Crop diversity treat-

ments that reduce the abundance of insectherbivores also suppress the spread of viralinfection, because plant-feeding insects trans-mit most plant viruses. This leads to lowerviral densities in polycultures than monocul-tures (Power and Flecker 1996).

The species richness of natural enemies(pathogens, predators, and parasitoids) of peststends to be higher in species-rich agroecosys-tems than in monocultures and higher in nat-ural vegetation buffers than in fields, leadingto higher ratios of natural enemies to herbi-vores and therefore lower pest densities. Thespraying of biocides can increase vulnerabilitybecause it reduces the abundance of naturalenemies more than the pests that are targeted.This allows pest populations to rebound rapidly,sometimes causing more damage than if no pes-ticides had been used (Naylor and Ehrlich 1997;see Chapter 12).

Invasive exotic species cost tens of millionsof dollars annually in the USA, primarily incrop losses and pesticide applications (Pimentelet al. 2000). Ecosystems have mechanisms, asyet poorly understood, that reduce invasibility(Dıaz et al. 2006). These include disturbanceadaptations that enable certain native speciesto colonize and grow rapidly after disturbance,a time when invasive species might otherwiseencounter little competition from other plants.In a given environment, more diverse ecosys-tems are less readily invaded, perhaps becauselocal species already fill most of the potentialbiological roles and utilize most of the resourcesthat might be available (Dıaz et al. 2006). Inva-sive species most frequently colonize environ-ments that naturally support high levels ofdiversity. Together these observations suggestthat (1) hot spots for diversity are particu-larly at risk of invasion by introduced species,and (2) the loss of native species may increaseinvasibility.

Natural enemies also reduce ecosystem inva-sibility. One reason that exotic species are oftenso successful is that they escape their special-ized natural enemies that constrain success intheir region of origin. Natural enemies in thenew environment often prevent these exoticspecies from becoming noxious pests (Mitchelland Power 2003).

46 F.S. Chapin

Maintaining the integrity of natural ecosys-tems may reduce disease risk to people. Lymedisease, for example, is caused by a pathogenthat is transmitted by ticks from mice anddeer to people (Ostfeld and Keesing 2000).Although forest mice are the largest reservoirof the disease, deer move the disease from oneforest patch to another. Exploding deer densi-ties have increased the incidence of lyme dis-ease in the northeastern USA in response toagricultural abandonment, expansion of sub-urban habitat, and elimination of the naturalpredators of deer. In Sweden, climate warm-ing has increased overwinter survival of ticks,further increasing disease incidence (Lindgrenet al. 2000, Lindgren and Gustafson 2001).

Currently 75% of emerging human dis-eases are naturally transmitted from animals tohumans (Taylor et al. 2001). Clearly, efforts tocontrol these diseases require improved under-standing of social–ecological dynamics (Patzet al. 2005).

Pollination Services

Much of the world’s food production dependson animal pollination, particularly for fruits andvegetables that provide a considerable portionof the vitamins and minerals in the human diet.The value of these pollination services is likelyto be billions of dollars annually (Costanza etal. 1997). Pollination by animals is obviouslyessential for the success of plants that are notwind- or self-pollinated. These pollination ser-vices often extend well beyond a given standand are often important in pollinating adjacentcrops. Temperate orchards and tropical cof-fee plantations adjacent to uncultivated landsor riparian corridors often have more pollina-tors and are more productive than are largerorchards (Ricketts et al. 2004). Large mono-cultures reduce pollination services by reduc-ing local floral diversity and nesting sites and byusing insecticides that reduce pollinator abun-dances. Pollination webs often connect the suc-cess of a wide range of species in multipleecosystem types (Memmott 1997).

Cultural Services: SustainingSociety’s Connections to Landand Sea

Cultural Identity and Cultural Heritage

Cultural connections to the environment arepowerful social forces that can foster Stew-ardship and social–ecological sustainability (seeChapter 6). Because people have evolved asintegral components of social–ecological sys-tems, this human–nature relationship is oftenan important component of cultural identity,i.e., the current cultural connection betweenpeople and their environment (Plate 4). Cul-tural identity in turn links to the past throughcultural heritage, i.e., the stories, legends, andmemories of past cultural ties to the environ-ment (de Groot et al. 2005). Cultural iden-tity and heritage are ecosystem services thatstrongly influence people’s sense of steward-ship of social–ecological systems (Ramakrish-nan 1992, Berkes 2008) and therefore offeran excellent opportunity for natural resourcemanagers to both learn from and contribute tostakeholder efforts to sustain their livelihoodsand environment (see Chapter 6). Many indige-nous peoples, for example, have traditional eco-logical knowledge based on oral transmissionof their cultural heritage in ways that informcurrent interactions with the environment (i.e.,cultural identity). This cultural heritage pro-vides information about how people coped withpast environmental and social–ecological chal-lenges and about important values that influ-ence likely future responses to changes in boththe environment and the resource managementpolicies (Berkes 1998; see Chapter 4). Integra-tion of traditional knowledge with the formalknowledge (i.e., “scientific knowledge”) thatoften informs resource management decisionsis not easy, because the “facts” (e.g., the natureof the human–environment relationship) some-times differ between the two knowledge sys-tems (Berkes 2008). Both knowledge systemsare important if they influence the ways inwhich stakeholders perceive and interact withtheir environment. The linkage between knowl-edge systems (as informed by cultural heritage),perceptions, and actions is at least as important


to understanding and predicting human actionsas are the biophysical mechanisms that arebelieved to underlie scientific knowledge. Sociallearning that builds new frameworks to sus-tain social–ecological systems is most likely tooccur if both traditional and formal knowl-edge are treated with respect rather thansubjugating traditional knowledge to westernscience.

Local knowledge held by farmers, ranch-ers, fishermen, resource managers, engineers,and city dwellers is also valuable. As in thecase of indigenous knowledge, local knowl-edge consists of “facts” based on cultural her-itage and observations that determine howpeople respond to and affect their environ-ment. Many local residents, whether indige-nous or not, spend more time interactingwith their environment than do policy mak-ers and therefore have a different suite ofobservations and perceptions. Co-managementof natural resources by resource managersand local stakeholders provides one mecha-nism to integrate these knowledge systems andperspectives in ways that increase the likeli-hood of effective policy implementation (seeChapter 4).

Traditional knowledge systems are beingeroded by social and technological changes.Many indigenous traditional knowledge sys-tems are maintained orally and are thereforetightly linked to language. There are about5,000 indigenous languages, half of them intropical and subtropical forests (de Groot et al.2005). Many of these languages are threat-ened by national efforts to assimilate peopleinto one or a few national languages. Lan-guage loss and cultural assimilation generallyerode traditional knowledge, so efforts to sus-tain local languages and cultures can be crit-ical to sustaining the knowledge and prac-tices by which people traditionally interactedwith ecosystems. Similarly, sustaining opportu-nities for locally adapted approaches to farm-ing, ranching, and fishing preserves practicesthat may sustain local use of natural resources(Olsson et al. 2004b). Obviously, many localpractices, whether indigenous or otherwise, donot contribute to sustainable management ina modern world, but they nonetheless pro-

vide information about perceptions, tradeoffs,and institutions that are a source of resiliencefor developing new frameworks to sustainsocial–ecological systems in a rapidly changingworld.

Spiritual, Inspirational, and AestheticServices

The spiritual, inspirational, and aesthetic ser-vices provided by ecosystems are importantmotivations for conservation and long-term sus-tainability. “The most common element of allreligions throughout history has been the inspi-ration they have drawn from nature, leadingto a belief in non-physical (usually supernatu-ral) beings” (de Groot et al. 2005). These spir-itual services provided by ecosystems for bothpersonal reflection and more organized experi-ences have proven to be a powerful force forconservation. Sacred groves in northeast Indiaand Madagascar, for example, are major reser-voirs of biodiversity and places where peoplemaintain their sense of connection with the landin landscapes that are increasingly converted toagriculture to meet the food needs of local peo-ple (Ramakrishnan 1992, Elmqvist et al. 2007).Often these sacred groves are maintained andprotected by local institutions (see Chapters 6and 7). Well-meaning national and interna-tional efforts to protect these few remainingsites of high conservation value by placing themunder national control sometimes underminelocal institutions and lead to degradation ratherthan protection. This underscores the impor-tance of understanding the social context ofcultural services, when addressing conserva-tion and sustainability goals (Elmqvist et al.2007).

The inspirational qualities of landscapesthat motivated ancient Greek philosophers,Thoreau’s writings, French Impressionist paint-ings, and Beethoven’s symphonies continue toinspire people everywhere through both directexperience and increasingly television, films,and the Internet. This provides natural resourcemanagers with a diverse set of media that cansupplement personal experience in reinforcing

48 F.S. Chapin

the human–environment connection (Swansonet al. 2008).

The aesthetic and inspirational properties oflandscapes are closely linked. Research sug-gests that aesthetic preferences are surpris-ingly similar among people from very differ-ent cultural and ecological backgrounds (Ulrich1983, Kaplan and Kaplan 1989, de Grootet al. 2005). For example, when asked whichis more aesthetically pleasing, people gener-ally prefer natural over built environments andpark/savanna-like environments over arid orforest environments, regardless of their back-ground. People do differ in aesthetic pref-erences, of course. Farmers and low-incomegroups generally prefer human-modified overnatural landscapes, whereas city dwellers andhigh-income groups have an aesthetic pref-erence for natural landscapes. In addition,the view of western Euro-Americans towardwilderness has changed through history from aperception of wilderness as a hostile land untilthe late seventeenth century toward a moreromantic view of wilderness in the eighteenthand nineteenth centuries. Even today, wilder-ness views are changing as people move tocities and change their patterns of use of remotelands (de Groot et al. 2005). At a time of rapidglobal change, it seems important to explorepotential changes in the spiritual, inspira-tional, and aesthetic benefits that people derivefrom ecosystems and the resulting humandecisions and actions that influence theirenvironment.

Iconic species are species that symbolizeimportant nature-based societal values. Pro-tection of these species can mobilize publicsupport for protection of values that mightotherwise be difficult to quantify and defendas management goals. Polar bears, wolves,Siberian tigers, eagles, and whales, for exam-ple, are top predators whose population dynam-ics are sensitive to habitat fragmentationand to factors that might affect their preyspecies. Panda bears and spotted owls requireecosystems with structural properties typicalof old-growth ecosystems. Ecosystem man-agement that protects the habitat of thesespecies often sustains a multitude of otherservices.

Recreation and Ecotourism

Recreation and tourism have always beenimportant benefits that people gain fromecosystems, sometimes as rituals and pilgrim-ages, sometimes just for pleasure and enjoy-ment. For example, about a billion people (15%of the global population) visit the Ganges Riverannually. Nature tourism accounts for about20% of international travel and is increasing20–30% annually. At a more local scale, peopleuse parks and other ecosystems as importantcomponents of daily life. Cultural and nature-based tourism constitutes 3–10% of GDP (grossdomestic product) in advanced economies andup to 40% in developing economies. It is themain source of foreign currency for at least38% of countries (de Groot et al. 2005). Clearlythere are both personal and economic incen-tives to manage the recreational opportunitiesprovided by ecosystems in ways that do notdegrade over time.

Synergies and Tradeoffs amongEcosystem Services

Ecosystems that maintain their characteristicsupporting services provide a broad spectrumof ecosystem services with minimal manage-ment effort. At a finer level of resolution, bun-dles of services can be identified that have par-ticularly tight linkages. This creates synergiesin which management practices that sustain afew key services also sustain other synergisticservices. For example, management of fire andgrazing in drylands to maintain grass cover min-imizes soil erosion (sustaining most support-ing services), sustains the capacity to supportgrazers, and reduces vulnerability to invasionby exotic shrubs (see Chapter 8). Managementof fisheries to sustain bottom habitat throughrestrictions on trawling or to maintain popula-tions of top predators reduces the likelihoodof fishery collapse. In general, managementthat sustains slow variables (soil resource sup-ply, disturbance regime, and functional types ofspecies) sustains a broad suite of ecosystem ser-vices. Managers are often tasked with managingone or a few fast variables such as the supply


of corn, deer, timber, or water, each of whichmight be augmented in the short term by poli-cies that reduce the flow of other ecosystem ser-vices (tradeoffs). However, even these fast vari-ables that are the immediate responsibility ofmanagers are best sustained over the long termthrough attention to slow variables that governthe flow of these and a broader suite of services.

Tradeoffs most frequently emerge whenpeople seek to enhance the flow of one ora few services (Table 2.4). For example, agri-cultural production of food typically requiresthe replacement of some naturally occurringecosystem by a crop with the corresponding lossof some regulatory and cultural services. Man-agement of forests to produce timber as a crop(short rotations of a single species) involvessimilar tradeoffs between the efficient produc-tion of a single species and the cultural andregulatory services provided by more diverseforests (see Chapter 7). Many managementchoices involve temporal tradeoffs betweenshort-term benefits and long-term capacity ofecosystems to provide services to future gener-ations. Management of lands to provide mul-tiple services (i.e., multiple-use management)requires identification of tradeoffs among ser-vices and decisions that reflect societal choicesamong the costs and benefits associated withparticular options.

Given the large number of essential ser-vices provided by ecosystems (e.g., about 40identified by the MEA), which services should

receive highest management priority? Oneapproach is to sustain supporting services thatunderpin most other ecosystem services, asdescribed earlier. During times of rapid socialor environmental change, however, inevitabletradeoffs arise that make objective decisionsdifficult. Under these circumstances, it mayprove valuable to identify critical ecosystemservices, i.e., those services that (1) societydepends on or values; (2) are undergoing (orare vulnerable to) rapid change; and/or (3) haveno technological or off-site substitutes (AnnKinzig, pers.com.). Stakeholders often disagreeabout which ecosystem services are most crit-ical, so identification of these services benefitsfrom broad stakeholder participation (Fischer1993, Shindler and Cramer 1999).

Facing the Realities of EcosystemManagement

Sustainability: Balancing Short-Termand Long-Term Needs

Sustainability requires the use of the environ-ment and resources to meet the needs of thepresent without compromising the ability offuture generations to meet their own needs(WCED 1987; see Chapter 1). Balancing thetemporal tradeoff between short-term desiresand long-term opportunities is a fundamentalchallenge for ecosystem stewards. The demandsby current stakeholders are always more certain

Table 2.4. Examples of synergies and tradeoffs among ecosystem services.

SynergiesSupporting services: maintenance of soil resources, biodiversity, carbon, water, and nutrient cyclingWater resources: water provisioning, maintenance of soil resources, regulation of water quantity and quality by

maintaining intact ecosystems, flood preventionFood/timber production capacity: food/timber provisioning, maintenance of soil resources, genetic diversity of

crops/forestClimate regulation: maintenance of soil resources, regulation of water quantity by maintaining ecosystem structureCultural services: maintenance of supporting services (including biodiversity), suite of cultural services

TradeoffsEfficiency vs sustainability:Short-term vs long-term supply of servicesFood production vs services provided by intact natural ecosystemsIntensive vs extensive management to provide food or fiberRecreation vs traditional cultural services

50 F.S. Chapin

and outspoken than those of future genera-tions, creating pressures to manage resourcesfor short-term benefits. Ecosystem stewardshipimplies, however, a responsibility to sustainecosystems so that future generations can meettheir needs. How do we do this, if we do notknow what future generations will want andneed? The simplest approach is to sustain theinclusive wealth of the system, i.e., the total cap-ital (natural, manufactured, human, and social)that constitutes the productive base available tosociety (see Chapter 1). Since natural and socialcapital are the most difficult components of cap-ital to renew, once they are degraded, these arethe most critical components of inclusive wealthto sustain. Social capital is discussed in Chapter3; here we focus on natural capital.

Future generations depend most critically onthose components of natural capital that can-not be regenerated or created over time scalesof years to decades. These always include (1)soil resources that govern the productive poten-tial of the land; (2) biodiversity that consti-tutes the biological reservoir of future options;(3) regulation of the climate system that gov-erns future environment; and (4) cultural iden-tity and inspirational services that provide aconnection between people and the land orsea. Earlier we described strategies for sustain-ing each of these classes of ecosystem services.Other more specific needs of future genera-tions, such as specific types of food or recre-ation, are less certain and therefore less criticalto sustain in precisely their current form.

At intermediate time scales (e.g., years todecades), it is more plausible to assume that theneeds of people in the future will be similar tothose of today, leading to a more constrained(and therefore more precisely defined) set oftradeoff decisions. Depletion of fossil ground-water to meet irrigation needs today reducesthe water available in the future—for exam-ple, for domestic water consumption. Forests orfish stocks that are harvested more rapidly thanthey can regenerate reduce the services in com-ing years to decades. Maximum sustained yieldwas a policy that sought to maximize the har-vest of forests, fish, and wildlife to meet cur-rent needs, while sustaining the potential tocontinue these yields in the future. Although

intended to prevent overharvest, these policiesoften proved unsustainable because of overlyoptimistic assumptions about the current statusand recovery potential of managed populations(see Chapters 7 and 10).

Economists often discount (i.e., reduce) esti-mates of the future value of manufacturedgoods and services because of opportunity costs(i.e., the opportunities foregone to spend themoney on current goods). Discounting thefuture is not appropriate, however, for tempo-ral tradeoffs involving those ecosystem servicesthat cannot be restored once they are lost, forexample, fossil ground water, biodiversity, orsacred groves in a highly modified landscape(Heal 2000). These services will be at least asvaluable, and perhaps much more valuable, inthe future than they are today (Heal 2000).

Multiple Use: Negotiating ConflictingDesires of Current Stakeholders

Tradeoffs among ecosystem services valued bysociety generate frequent conflicts about man-agement of ecosystem services. Clearing foreststo provide new agricultural land, for exam-ple, represents a tradeoff between the bene-fits of harvested timber and increased poten-tial for food production and the loss of otherservices previously provided by the forest suchas regulation of water quality and climate.Recreational use of an area by snow machinesand wilderness skiers creates tradeoffs betweenthe services sought by each group. There arealso tradeoffs in allocation of fresh wateramong natural desert landscapes, agricultural-ists, ranchers, and urban residents. These arejust a few of the many tradeoffs faced by ecosys-tem stewards. How does a manager balancethese tradeoffs or choose among them? Thereis no simple answer to this question, but thefollowing questions raise issues that warrantconsideration. What are the gains and losses inbundles of ecosystem services associated withalternative management options? How muchweight do we give to gains and losses thatare uncertain but potentially large? Are therenew options that might provide many of thesame benefits but reduce potential losses in ser-


vices? Who wins and who loses? Can conflictsbe reduced by landscape approaches that sep-arate in time or space the services that dif-ferent stakeholders seek to sustain? The socialdimensions of tradeoffs among ecosystem ser-vices are at least as important as the ecolog-ical ones, so we address these questions in asocial–ecological context in Chapter 4. We thenprovide numerous examples of challenges andstrategies for managing synergies and tradeoffsin Chapters 6–14 and summarize the lessonslearned in Chapter 15.

Adjusting to Change

Managing ecosystems in the context of dynamicand uncertain change is an integral compo-nent of resilience-based ecosystem steward-ship. Many changes, such as those associ-ated with interannual variability in weather,cycles of disturbance and succession, and eco-nomic fluctuations are widely accepted as nor-mal components of the internal variability ofsocial–ecological systems. Facilitating changeis sometimes an explicit management objec-tive, for example, in sustainable developmentprojects, where the goal is to enhance well-being, while sustaining the capacity of ecosys-tems to provide services (WCED 1987). Inreclamation and restoration projects, the goalis to enhance ecological integrity, while sustain-ing the social and economic benefits accruedfrom development. In still other cases, regionalchange may be driven by exogenous changes inclimate, introduction of exotic species, humanmigration, and/or globalization of culture andeconomy. Many of these latter changes are per-sistent trends that will inevitably cause changesin social–ecological systems (see Chapters 1and 5). The critical implications of managingchange include (1) identifying the persistentchanges that are occurring or might occur andassessing their plausible trajectories; (2) under-standing that attempts to prevent change orto maintain indefinitely the current flow ofecosystem goods (e.g., production of cattle orpulpwood) may be unrealistic; and (3) contin-uously reassessing the social–ecological long-and short-term goals of ecosystem stewardship

in light of the challenges and opportunitiesassociated with change.

Identifying persistent (directional) trends inimportant controls such as climate and humandemography and projecting plausible trendsinto the future is often a more realistic basisfor planning than assuming that the future willbe like the past. Although the future can neverbe predicted with certainty, planning in the con-text of expected changes can be helpful. Someof these changes may be difficult for individ-ual managers to alter (e.g., climate change),but the trajectory of other changes, for exam-ple, in sustainable development or ecologicalrestoration projects, can be strongly influencedby management decisions because these deci-sions create legacies and path dependenciesthat influence future outcomes. That, after all, isthe purpose of management. In general, thosechanges that sustain or enhance natural andsocial capital will likely benefit society [see Sus-tainability (this chapter), Chapters 4 and 5].

One limitation of maximum sustained yieldas a management paradigm is that it usuallyfails to account for variability and change (par-ticularly persistent directional changes) in thoseyield-influencing processes that managers can-not control (see Chapter 7). Recognition ofthis problem was one of the primary motiva-tions for a move toward ecosystem manage-ment as a more comprehensive approach to nat-ural resource management.

Ecosystems are not the only parts of social–ecological systems to undergo persistent direc-tional changes. Changes in societal goals areessential to consider. This requires active com-munication and engagement with a varietyof stakeholders to enable them to participatemeaningfully in the management process (seeChapters 4 and 5).

Regulations and Politics: Practicingthe Art of the Possible

Many of greatest challenges faced by resourcestewards reflect the social and institutionalenvironment in which they work. Differ-ences among stakeholders in goals and norms,power relationships, regulatory and financial

52 F.S. Chapin

constraints, personalities, and other social andinstitutional factors often dominate the day-to-day challenges faced by resource man-agers. Social processes are therefore an inte-gral component of ecosystem stewardship (seeChapters 3 and 4). At times of rapid social orenvironmental change, frameworks for man-aging natural resources may become dysfunc-tional, requiring communication with a broaderset of stakeholders and openness to new waysof doing things (see Chapter 5).

Summary

Resilience-based ecosystem stewardshipinvolves responding to and shaping change insocial–ecological systems to sustain the supplyand opportunities for use of ecosystem servicesby society. The capacity of ecosystems to supplythese services depends on underlying support-ing services, such as the supply of soil resources;cycling of water, carbon, and nutrients; and themaintenance of biological diversity at standand landscape scales. The resilience of thesesupporting services depends on maintaining adisturbance regime to which local organismsare adapted. Directional changes in thesesupporting services inevitably alter the capacityof ecosystems to provide services to society.An understanding of these linkages provides abasis for not only sustaining the services pro-vided by intact ecosystems but also enhancingthe capacity of degraded ecosystems to providethese services by manipulating pathways forecosystem renewal.

Both managers and the public generally rec-ognize the value of provisioning services suchas water, food, fiber, and fuelwood. It is there-fore not surprising that the links between sup-porting and provisioning services are generallywell understood by ecosystem managers andlocal resource users. This provides a strong localand scientific basis to manage ecosystems sus-tainably for these goods. In contrast, regula-tory services (e.g., the regulation of climate andair quality, water quantity and quality, pollina-tion services, and risks of disease and of nat-ural hazards), although generally recognizedas important by society, are often overlooked

when discussions focus on the short-term sup-ply of provisioning services. Some of the great-est opportunities and challenges in ecosystemmanagement involve the stewardship of ecosys-tems to provide bundles of services that bothmeet the short-term needs of society and sus-tain regulatory services that are essential fortheir secure supply at larger temporal and spa-tial scales.

The long-term well-being of society dependssubstantially on the cultural services providedby ecosystems, including aesthetic, spiritual,and recreational values. These values oftenmotivate support for sustainable stewardshippractices, if the basic material needs of societyare met. Understanding this hierarchy of soci-etal needs, which provides the social context forecosystem stewardship is the central topic ofChapter 3.

Review Questions

1. What slow variables are usually most impor-tant in sustaining ecosystem services? Howare these likely to be altered by climatechange in the ecosystem where you live?What are the likely societal consequences ofthese changes?

2. Describe a strategy for enhancing the supplyof ecosystem services for a degraded ecosys-tem such as an abandoned mine or an over-grazed pasture.

3. What are the advantages and disadvantagesof managing an ecosystem to maximize theproduction of a specific resource such astrees or fish? How would your managementstrategy differ if your goal were to maximizeharvest over the short term (e.g., 5 years) vsthe long term (several centuries). How mightdirectional social or environmental changeinfluence your long-term strategy?

4. If the future is uncertain and you do notknow what future generations will want orneed, how can you manage ecosystems sus-tainably to meet these needs?

5. Explain the mechanisms by which biodi-versity influences supporting services, provi-sioning services, and cultural services.


Additional Readings

Carpenter, S.R. 2003. Regime Shifts in Lake Ecosys-tems: Pattern and Variation. International EcologyInstitute, Lodendorf/Luhe, Germany.

Chapin, F.S., III, P.A. Matson, and H.A. Mooney.2002. Principles of Terrestrial Ecosystem Ecology.Springer-Verlag, New York.

Christensen, N.L., A.M. Bartuska, J.H. Brown, S.R.Carpenter, C. D′Antonio, et al. 1996. The reportof the Ecological Society of America committeeon the scientific basis for ecosystem management.Ecological Applications 6:665–691.

Daily, G.C. 1997. Nature′s Services: Societal Depen-dence on Natural Ecosystems. Island Press,Washington.

Dıaz, S., J. Fargione, F.S. Chapin, III, and D.Tilman. 2006. Biodiversity loss threatens humanwell-being. Plant Library of Science (PLoS)4:1300–1305.

Heal G. 2000. Nature and the Marketplace: Captur-ing the Value of Ecosystem Services. Washington:Island Press.

Holling, C.S., and G.K. Meffe. 1996. Command andcontrol and the pathology of natural resourcemanagement. Conservation Biology 10:328–337.

Olsson P., C. Folke, and T. Hahn. 2004. Social-ecological transformation for ecosystem man-agement: The development of adaptive co-management of a wetland landscape in southernSweden. Ecology and Society 9(4):2. [online]URL: www.ecologyandsociety.org/vol9/iss4/art2/.

Schlesinger, W.H. 1997. Biogeochemistry: An Anal-ysis of Global Change, 2nd edition. AcademicPress, San Diego.

Szaro, R.C., N.C. Johnson, W.T. Sexton, and A.J.Malk, editors. 1999. Ecological Stewardship: ACommon Reference for Ecosystem Management.Elsevier Science Ltd, Oxford.

3Sustaining Livelihoods and HumanWell-Being during Social–EcologicalChange

Gary P. Kofinas and F. Stuart Chapin, III

Introduction

Social processes strongly influence the dynam-ics of social–ecological responses to change. InChapter 2, we described the ecological pro-cesses that govern the flow of ecosystem goodsand services to society. Sustainability of theseflows depends not only on ecosystems, but alsoon human actions that are motivated, in part,by desires and needs for these services. Manyof the social and ecological slow variables thatdetermine the long-term dynamics of social–ecological systems act primarily through theireffects on human well-being (see Chapter 1).Today humans are the dominant force driv-ing changes in the Earth System, with the bio-physical changes during the last 250 years sofundamental that they define a new geologicepoch—the Anthropocene (Crutzen 2002; seeChapter 14). Social process play a key rolein driving these changes, so it is essential tounderstand them as critical determinants of

G.P. Kofinas (�)Institute of Arctic Biology, School of NaturalResources and Agricultural Sciences, University ofAlaska Fairbanks, Fairbanks, AK 99775, USAe-mail: [email protected]

sustainability from local to planetary scales.This chapter focuses on those social processesthat affect well-being, society’s vulnerabilityand resilience to recent and projected changes,and strategies for sustainable development thatseek to enhance well-being, while sustaining thecapacity of ecosystems to meet human needs.Institutional dimensions of well-being, whichare also critical to sustainability, are addressedin Chapter 4.

Well-being, or quality of life, is more thanhuman health and wealth. In the context ofecosystem stewardship and sustainability, well-being also includes happiness, a sense of fatecontrol, and community capacity. Livelihoodsof individuals and households include theircapabilities, tangible assets, and means of liv-ing (Chambers and Conway 1991). Well-beingand livelihoods are therefore key elements thatset the stage for sustainability, resilience, andadaptability of people to change.

Livelihoods are both complex and dynamic(Allison and Ellis 2001). People both respondto and cause many ecological changes as a resultof resource consumption by a growing pop-ulation and efforts to meet their desires andneeds in new ways. Perceptions of well-beingare shaped by material conditions, history, andculture. For these reasons, the relationship


56 G.P. Kofinas and F.S. Chapin

between well-being, livelihoods, and naturaland social capital can define the prospects forlong-term sustainability (Janssen and Scheffer2004). Vulnerability, the degree to which a sys-tem is likely to experience harm due to expo-sure to a specified hazard or stress (Turneret al. 2003, Adger 2006), and resilience, thecapacity of a system to respond to and shapechange and continue to develop, are impor-tant concepts in understanding these relation-ships (see Chapter 1). Vulnerability analysis hastypically focused on the potential threats (e.g.,climate change) and endowments (e.g., assets,social capital) that affect livelihoods and well-being (Turner et al. 2003), whereas resiliencethinking has emphasized the capacity of groupsto cope with ecological change and avoid dra-matic and undesirable social–ecological shifts(Walker and Salt 2006). We integrate the vul-nerability and resilience approaches in thischapter to provide a framework for under-standing livelihoods and well-being in the con-text of ecosystem stewardship and argue thatwell-being based on adequate livelihoods isessential to reduce vulnerabilities to the pointthat people can engage in long-term planningfor ecosystem stewardship. We also note that,because stakeholders differ in their values, per-ceptions of needs, and capacity to fulfill them,ecosystem stewardship requires a participatoryplace-based approach to improve well-being,plan for change, and cope with inevitable sur-prises (Maskrey 1989, Smit and Wandel 2006).

From Human Ecologyto Sustainable Developmentto Social–Ecological Resilience

The context for understanding social–ecological interactions is shifting from studiesof ecological effects on society (and vice versa)to development planning that presumes pre-dictable effects on society to resilience-basedstewardship that seeks to respond to and shapea rapidly changing world. The anthropologicalsubdisciplines of human ecology, culturalecology, and political ecology have studied thecoevolution of human–environment relations

and their effects on the well-being of humangroups. These examinations have helped usto understand how ecosystems shape socialorganization, economies, and the exploitationof ecosystems and how human exploitationaffects ecological conditions (Rappaport 1967,Turner et al. 1990). Anthropologists point outthat, although each case is unique, there arecommon patterns that characterize types oflivelihood strategies, such as sharing amonghunting and gathering societies, specializeddivision of labor among family members inpastoralism, and specialization of technicalskills in urban-based capitalism. The studyof changes in these livelihood strategies hashighlighted the current rapid changes in socialcomplexity and emerging issues of disparityboth within and across groups, particularlybetween those of developing and developedsocieties.

In the 1960s and 1970s, developed countriesexperienced a transformation in their thinkingabout resources use and environmental quality(Repetto 2006), in part because of the aware-ness raised by the books Silent Spring (Carson1962) and the Club of Rome’s Limits to Growth(Meadows et al. 1972). These books and severallarge environmental catastrophes, such as oilspills, captured the public’s attention throughmass media (Bernstein 2002). The global scopeand consequences of environmental degrada-tion from pollution and the central role of tech-nology and population growth led to the UNConference on Human Development in Stock-holm in 1972, and later the UN Commissionon the Environment and the Economy, whichproduced the historic document, Our CommonFuture (WCED 1987). Calling for the need tobalance economic development with environ-mental quality, the commission popularized theterm sustainable development, defining it as themeans by which society improves conditionswithout sacrificing opportunities for future gen-erations (see Chapter 1). The dilemmas of thesepotentially conflicting objectives have servedboth as a catalyst for discussions on sustainabil-ity, and the basis for fierce debate about theextent to which economic growth can providefor quality of life, while ensuring that ecosys-tem services are sustained to meet current and

3 Sustaining Livelihoods and Human Well-Being during Social–Ecological Change 57

future needs. These issues underscore the soci-etal dilemmas represented by rising materialconsumption and current positive trends ofpopulation growth.

A dual focus on social–ecological resilienceand well-being puts the debates on sustain-able development into a dynamic context, rais-ing questions about the sources of both socialand ecological resilience available to groupsseeking to shape change and navigate criti-cal thresholds that may affect well-being (seeChapter 5). They also highlight the behavioraltraps that may emerge and ultimately impedehuman development. The focus on vulnerabilityand resilience adds important insight to thesediscussions by directing attention to exposureto risks, potentials for shocks and pulses ofchange, and the capacity of the system to absorband shape those forces. Although much of thesustainable development debate has focusedon issues of well-being and livelihoods in thecontext of poverty, these issues have equallyimportant implications for all societies and theirlife support systems. In the section below, wepresent information on current global trends inquality of life and describe some elements ofwell-being with relevance to ecosystem stew-ardship in a changing world. We also present ananalytical framework for assessing vulnerabil-ity and resilience in the context of sustainablelivelihoods, well-being, and social–ecologicalchange.

Well-Being in a Changing World

Well-being integrates many dimensions of thequality of life. Disciplinary scholars, such aseconomists, anthropologists, and political sci-entists, typically focus on specific indices ofwell-being such as levels of consumption andexpenditures, availability of employment forcash, sense of cultural continuity, and choice,respectively. Although each is important, so isa broader array, including one’s sense of secu-rity, freedom, fate control, and good social rela-tions (Levy et al. 2005, MEA 2005a). Peoplediffer in the ways that they define a sense ofwell-being, depending on culture, age, gender,household structure, and other factors. How-

ever, there is surprising agreement through-out the world that this broader array of ele-ments is essential for a good life (Narayanet al. 2000, Dasgupta 2001). The social psy-chologist Abraham Maslow (Maslow 1943)framed well-being as a hierarchy of motiva-tions, stating that basic physiological needs suchas food and water are the most fundamental,followed by perceptions of safety and secu-rity, then love and belonging through socialconnections with family and community, thenthe need for self-esteem and the respect ofothers, and finally, self-actualization throughpursuits such as creative action and reflec-tive morality. Although Maslow’s focus wasmostly on individual motivation and well-being,there is an extensive literature linking individ-ual and community resilience (e.g., Luthar andCicchette 2000) that has logical consequencesfor social–ecological sustainability. Opportu-nities for sustainable stewardship increase asmore of Maslow’s components of well-being aremet. In the following sections, we summarizethe connections between these components ofwell-being and ecosystem stewardship.

Material Basis of Well-Being

The material basis of well-being includes peo-ple’s essential needs, such as clean air and waterand adequate food supply. Ecosystem services,generated at regional and local levels, con-tribute substantially to the material bases ofwell-being. Failure to meet the basic materialneeds of life generally has negative implicationsfor ecosystems because people tend to priori-tize the fulfillment of basic needs above long-term stewardship goals. In the absence of basicmaterial needs, people are more likely to betrapped in perpetuating conditions of poverty,poor health, and social instability.

In the global aggregate, the material basis ofwell-being has increased substantially in termsof income and food supplies. Gross domesticproduct (GDP—a standard economic measureof income), for example, has doubled in lessthan 50 years and continues to increase in manyregions. Much of the increase in income andassociated consumption, however, has occurred


in developed nations and affluent segments ofsociety, where consumption reflects desires andpreferences rather than the fulfillment of essen-tial material needs. However, in some areasthere has been a genuine reduction in poverty.In East Asia income poverty declined by 50%between 1990 and 2005 (Levy et al. 2005). Yet,in about 25% of countries, particularly in sub-

Saharan Africa, vulnerability as estimated bythe Human Development Index is increasing(Fig. 3.1; UNDP 2003).

Economic disparity between rich and poornations has increased since the beginning ofthe Industrial Revolution, a trend that contin-ues today (Levy et al. 2005). In other words,poor countries are becoming poorer, relative to

Figure 3.1. Vulnerability index for African countries,composed from 12 indicators (food import depen-dency ratio, water scarcity, energy imports as per-centage of consumption, access to safe water, expen-ditures on defense vs. health and education, human

freedoms, urban population growth, child mortal-ity, maternal mortality, income per capita, degree ofdemocratization, and fertility rate. Reprinted fromKasperson et al. (2005).


more developed nations. Within poor countriesand some rich countries, economic disparityis also increasing. For example, as popula-tion increases, income has declined in low-productivity systems such as drylands, rela-tive to more productive coastal areas (seeChapters 8 and 11). In extremely impoverishedareas, people lack the resources to move tocities, which has been the predominant demo-graphic trend in most of the world. Conse-quently, the greatest human pressure on theenvironment is occurring in those regions thatare particularly prone to degradation of ecosys-tem services. Of the 20% of the global popula-tion that lives in poverty (here defined as livingon less than $1 per day), 70% live in rural areas,where they depend directly on ecosystem ser-vices for daily survival (MEA 2005a). Large dis-parities between rich and poor can also createpolitical instability and low resilience to shocks.These same regions of poverty and economicdisparity are commonly subject to violent con-flict, severe hardship, and chronic rates of highmortality.

Poverty and its implications for the degra-dation of ecosystems services can also havecascading and long-term effects on most othercomponents of well-being. As people spendmore time meeting their basic survival needsfor food, for example, they have less energyto pursue education and other opportunities.Inadequate nutrition and education, in turn,can impact health and social relations. In about25% of countries, the decline in natural capitalhas been larger than increases in manufacturedplus human capital, indicating a clearly unsus-tainable trajectory associated with declines inassets for production (see Chapter 1). Manycritical ecosystem services (e.g., pollinationservices, pest regulation, maintenance of soilresources) appear to be declining (Foley et al.2005), although data gaps make it difficult toassess overall trends. This widespread declinein ecosystem services worldwide clearly threat-ens the future well-being of society, particularlythe disadvantaged, and widens the gap betweenhaves and have-nots.

The material basis for well-being is impor-tant but to some extent subjective. Below aper capita income of about $12,000, there is

a strong correlation between the wealth ofnations and the average happiness of their cit-izens (Diener and Seligman 2004). Similar cor-relations are observed within countries. Thereis, however, no significant increase in happi-ness once an individual’s basic material needsare satisfied (Easterlin 2001). As people acquiregreater wealth, they aspire to achieve evengreater wealth, which, in turn, seems to reducetheir happiness and overall satisfaction. In addi-tion to the pursuit of consumption for its ownsake, other factors become important oncebasic needs are met, such as the quality of socialrelations or one’s sense of fate control. Thesefindings suggest two basic approaches to achiev-ing a better match between the flow of ecosys-tem services and the material needs of society:(1) assure that the basic material needs of poorpeople are met and (2) reduce the upward spi-ral of consumption by people whose materialneeds are already met.

Safety and Security

Perceptions of insecurity and risk vis-a-visecosystem services affect decisions that gov-ern well-being. High risks of starvation, disease,armed conflict, economic crises, or natural dis-asters, for example, have psycho-social implica-tions for all aspects of life and inevitably under-mine society’s commitment to ecosystem stew-ardship. Some hazards and risks have declined,and others have increased.

Stable livelihoods are one of the strongestdeterminants of safety and security. Lackof food security, in particular, is a criticalcomponent of well-being (IFPRI 2002; seeChapter 12). Families with insufficient incometo buy food and no suitable land to grow cropsare more vulnerable to environmental hazardsthat influence food supply than are peoplewith greater access to these assets. Althoughthe early literature on risk and vulnerabil-ity focused primarily on exposure to environ-mental hazards, there is increasing recognitionthat livelihood security is generally a strongerdeterminant of vulnerabilities and risks suchas famine than are climatic events such asfloods and droughts (Sen 1981, Adger 2006).


Clearly, basic material resources, security, andwell-being are tightly linked.

Disease risk interacts with other compo-nents of well-being. In aggregate, humanhealth has improved substantially. Reductionsin infant and child mortality have contributedto a doubling of average life expectancyduring the twentieth century. The associatedincreases in human population, however, aug-ment demands for ecosystem services and non-renewable resources and therefore rates of eco-logical change (Fig. 3.2). Changes in land use,irrigation, climate, and human settlement pat-terns in turn contribute to spread of infec-tious diseases that now constitute 25% of theglobal burden of disease (Patz et al. 2005).Most of these diseases are treatable, but eco-logical changes frequently increase the abun-dance or proximity of disease vectors, leadingto high infection rates that overpower the med-ical capacity for treatment. Deforestation, forexample, often brings forest animals in con-tact with livestock, increasing the risk of trans-fer of wildlife diseases to livestock and thento people; about 75% of human diseases havelinks to wildlife or domestic animals (Patz et al.

Supportingservices

(ecosystem processes)

Regulatingservices

(landscapeprocesses)

Provisioningservices

(harvested goods)

Culturalservices

(non-materialbenefits)

Materialbenefits

Self-esteem & good social

relations

Health

Safety and security

Free

do

m a

nd

ch

oic

e

Ecosystem stewardship

Ecosystem services Well-being

Figure 3.2. Relationship between ecosystem ser-vices and well-being. Adapted from the frameworkdeveloped by the Millennium Ecosystem Assess-ment (MEA 2005d).

2004). Malaria, schistosomiasis, and tuberculo-sis are widespread despite active control pro-grams. Other diseases such as trypanosomiasisand dengue in the tropics and Lyme diseaseand West Nile Virus in temperate countries areemerging health risks that appear related to cli-matic and ecological change. Current trends ofclimate warming create significant uncertaintyabout future trajectories of disease.

Several current global trends place increas-ing pressure on natural resources, adding riskto livelihoods and well-being: (1) an increasinghuman population, (2) a decline in basic envi-ronmental resources such as clean water, and(3) increasing demand for energy. The result-ing increase in competition for renewable andnonrenewable resources substantially increasesthe likelihood of local and global conflicts. Oil-producing states, for example, now host a thirdof the world’s civil wars (up from 20% 15 yearsago), and half of the oil-exporting countries(OPEC) are poorer than they were 30 yearsago (Ross 2008). Although there are interna-tional organizations (e.g., the UN), nationaland non-governmental organizations (NGOs),and treaties that could reduce the likelihoodof conflicts, these are not always effective inthe absence of basic social order. In summary,the risks of environmentally induced conflictare increasing, necessitating the adaptation andtransformation of governance processes to alle-viate these risks (see Chapters 4 and 5).

Many factors contribute to overall risk andinsecurity, leading to social–ecological vulnera-bility. The impacts of climate change, for exam-ple, depend strongly on poverty and well-being.Sub-Saharan Africa will likely be much morevulnerable to climate change than will the dry-lands of Australia or the USA, even where theclimatic changes and local ecology might be rel-atively similar. These differences arise out ofthe vastly different endowments that are avail-able to people to deal with these challenges,for example, government services, better infras-tructure, and more personal assets. The effectsof drought on food supply may also be dramat-ically increased by civil strife. Wars accountedfor about half of the major famines in the twen-tieth century (Hewitt 1997). Because vulnera-bility is always a multifactor social–ecological


problem, potential solutions should be place-based and pay careful attention to local con-text and slow variables such as climate, prop-erty rights, and livelihoods (Turner et al. 2003,Adger 2006).

Despite the complex nature of vulnerability,repeatable syndromes emerge (Petschel-Heldet al. 1999). The Sahel and many other aridzones, for example, are characterized by a set ofprocesses that result in the overuse of agricul-turally marginal land and make people vulnera-ble to various hazards, including drought, war,and epidemics of AIDS/HIV. When patternsof correlation among 80 social and environ-mental symptoms of global change were exam-ined, about 16 syndromes emerged that rep-resented repeatable sets of interactions amongthese symptoms. These syndromes were catego-rized based on how humans were using nature:as a source of resources, a medium for eco-nomic development, and environmental sinksfor human pollutants (Table 3.1). They providea basis for identifying intervention strategiesthat might apply to broad categories of social–ecological situations. A combination of soil-conserving agricultural techniques and povertyreduction actions, for example, were thought to

be most promising in reducing vulnerability inthe Sahel of Africa and other similar social–ecological situations (Petschel-Held et al. 1999).

Good Social Relations

Good social relations involve social cohesion,mutual respect, equitable gender relations,strong family associations, the ability to helpothers and provide for children, and the abil-ity to express and experience aesthetic, spir-itual, and cultural values. These higher-ordermotivations (good social relations, self-esteem,need for mutual respect, and actualization)in Maslow’s hierarchy are critical to ecosys-tem stewardship. They are the basis for socialcapital—the capacity of a group to work collec-tively to address and solve problems (Coleman1990). One tangible measure of social capitalis the extent to which social relations provideaccess directly to resources or to those withskills not held by individuals or members of ahousehold. Social relationships and social capi-tal are typically slow to build, but can deterio-rate quickly, especially in times of rapid socialor environmental change. Individuals and orga-nizations interact through social networks, the

Table 3.1. Major patterns of social–ecological inter-actions grouped by the nature of human use ofnature: as a source for production, as a medium forsocio-economic development, or as a sink for outputs

of human activities. Names are derived from eitherprototypical regions or catchwords for characteristicfeatures. Adapted from Petschel-Held et al. (1999).

Nature as a source for productionSahel syndrome Overuse of marginal landOverexploitation syndrome Overexploitation of natural ecosystemsRural exodus syndrome Degradation through abandonment of traditional agricultural practicesDust bowl syndrome Non-sustainable agro-industrial use of soils and bodies of waterKatanga syndrome Degradation through depletion of nonrenewable resourcesMass tourism syndrome Development and destruction of nature for recreational endsScorched earth syndrome Environmental destruction through war and military actionNature as a medium for socio-economic developmentAral Sea syndrome Damage of landscapes as a result of large-scale projectsGreen revolution syndrome Degradation through the transfer and introduction of inappropriate farming methodsAsian tiger syndrome Disregard for environmental standards in the course of rapid economic growthFavela syndrome Socio-ecological degradation through uncontrolled urban growthUrban sprawl syndrome Destruction of landscapes through planned expansion of infrastructureDisaster syndrome Singular anthropogenic environmental disasters with long-term impactsNature as a sink for outputs of human activitiesSmokestack syndrome Environmental degradation through large-scale diffusion of long-lived substancesWaste-dumping syndrome Environmental degradation through controlled and uncontrolled disposal of wastesContaminated land syndrome Local contamination of environmental assets at industrial locations


linkages that establish relations among indi-viduals, households and organizations acrosstime and space (see Chapter 4). Networks canalso serve important functions at the landscapescale, as for example in the regulation of wateruse for rice production in the Balinese watertemple networks (Lansing 2006). It is thereforeimportant to avoid evaluations of well-beingand livelihoods that focus solely on individualconditions. The social processes that form thebasis of social relations and their underlyingsocial networks (See Chapters 4 and 12) aresome of the critical endowments that define thesensitivity of a household, community, and soci-ety to risk and vulnerability.

Self-Esteem and Actualization

Fostering self-esteem and providing outlets foractualization of creative abilities and actionsmotivated by a person’s sense of place providesome of the greatest opportunities to enhanceecosystem stewardship. The highest-order ele-ments in Maslow’s hierarchy of human moti-vations, self-esteem and self-actualization, aremost fully expressed when people have mettheir immediate material needs, feel secure intheir capacity to continue meeting these needs,and have strong social relations. Wealth, how-ever, is not a prerequisite for these higher-ordermotivations. Many preindustrial cultures foundtime and ways to express their relationship withthe environment (e.g., Coastal Indians of theNorthwest Coast of North America) throughart and the celebration of stories that are partof oral traditions. When basic needs are met,there is a sense of fate control, and people feel asense of group and space to practice and appre-ciate music, literature, theater, and others artsthat embody the importance of people’s rela-tionship with ecosystems. These activities helpto embed stewardship as a part of culture byproviding the social space for reflection on thevalue of ecosystems and people’s relationshipwith their environment. People’s appreciationfor their relationship to nature (i.e., a land ethic;Leopold 1949) is often a strong motivation forecosystem stewardship and sustainability and

can be a powerful stabilizing feedback for man-aging rates of change.

As noted above, well-being and ecosys-tem stewardship are defined primarily aroundissues of social conditions, yet Maslow’s workwas to a great extent focused on the indi-vidual. Indeed, ecological stewardship dependssubstantially upon individual resilience—thecapacity of individuals to cope with changewhile remaining healthy. Although the con-nections of individual resilience and ecosystemstewardship may seem distant for professionalresource managers, they are nonetheless worthconsidering when implementing and evaluat-ing specific programs and policies. Where com-munities face epidemic rates of suicide orsubstance abuse, addressing issues of individ-ual resilience under conditions of rapid socio-economic change may prove to be the key tocommunity-scale resource stewardship.

Assessing Vulnerabilitiesto Well-Being

Vulnerability is the degree to which a systemis likely to experience harm due to exposure toa specified hazard or stress (Turner et al. 2003,Adger 2006; see Chapter 1). Social dimensionsof vulnerability are intimately tied to elementsof well-being, particularly to livelihoods; safetyand security; and strong social relations. In thissection, we link the concepts of livelihoods andwell-being to social–ecological vulnerability in arapidly changing world (Fig. 3.3). Vulnerabilityanalysis includes three evaluative steps: identi-fying (1) the type and level of exposure of a sys-tem to hazards or stresses, (2) the system’s sen-sitivity to these hazards based on the social andeconomic endowments available to them, and(3) the adaptive capacity of the system to makeadjustments to minimize the impacts of hazards(Turner et al. 2003, Adger 2006).

An important strength of vulnerability analy-sis is its focus on specific hazards and their inter-actions and on the differential impacts of theseinteracting hazards on specific social groups ata specific place. Vulnerability analysis of poorcoastal communities of Bangladesh to climate


System dynamics

Well-being,social networks:

poverty traps,dependency

cycles

Manufactured,natural, &

human capital

Biotic & socialinteractions

Coping,adaptation,

Interactions

Inte

rnal

stre

sses

Externalhazards

Exposureregime

Immediateimpact

Adaptability& resilience

Adapted resilient impact

Perceptions of risk Sensitivity

resilienceprocesses

Figure 3.3. Components of vulnerability. Vulnera-bility depends on exposure to external hazards orinternal stresses; sensitivity of the system to these

interacting stresses; and the adaptive adjustment inresponse to the resulting impacts.

change, for example, would consider the inten-sity and frequency of storm surges (exposure);the capability of local infrastructure to with-stand storms and flooding (infrastructure sen-sitivity); and the capacity of the communityto mobilize resources in anticipation of thestorm and capacity of aid workers to arriveat the scene after the incident has resulted indamage (social sensitivities). The goal of vul-nerability analysis is to provide the type ofinformation necessary for targeted interventionstrategies that foster adaptation to changingconditions.

Exposure to Hazards and Stresses

Social–ecological systems experience hazardsthat depend on both external events and inter-nal dynamics. Exposure is the nature anddegree to which the system experiences envi-ronmental or socio-political hazards (Adger2006). Mitigation (reduction in exposure to ahazard) is therefore an important strategy forreducing vulnerability. Hazards can be char-acterized in terms of their magnitude, fre-quency, duration, and spatial extent (Burtonet al. 1993). Exposure can reflect exoge-

nous hazards that a system experiences (e.g.,heat waves, earthquakes, tsunamis, coloniza-tion, wars, novel market forces) or endogenousstresses (e.g., food or water shortage, lack offinancial resources, high levels of internal con-flict; Turner et al. 2003, Kasperson et al. 2005),or both. Internal stresses often reflect scarcityof one or more forms of capital or other compo-nents of well-being (Fig. 3.3). Hazards, whetherthey be exogenous or endogenous, can rangefrom events or pulses that shock the system,such as an earthquake or collapse of a financialmarket, to chronic stresses that undermine well-being and social–ecological integrity of the sys-tem, such as rising sea level from climate changeor increased energy cost.

Exogenous Hazards

The increased frequency of extreme eventsadds to the vulnerability of social–ecologicalsystems. Model simulations suggest thathuman-induced climate warming contributesstrongly to many environmental componentsof this trend (IPCC 2007a, b; see Chapter 2).Extreme heat killed 35,000 people in Europein 2003, for example, and extreme droughtsare occurring more frequently. Flooding is also


occurring more frequently in low-lying coun-tries like Bangladesh. Increased sea-surfacetemperatures may intensify tropical storms,perhaps contributing to their fourfold increasein frequency. Melting of sea ice makes huntingof marine mammals more hazardous for arcticsubsistence hunters (Krupnik and Jolly 2002).Each of these changes exposes people to moresevere environmental hazards.

Climatically induced disturbances are inte-gral components of social–ecological systems,so exposure to these disturbances is shapedby both exogenous climate and endogenoussocial–ecological interactions (see Chapter 2).The frequency and extent of wildland fire,flooding, and pest outbreaks, for example, areincreasing in response to climatic warming, butthe extent to which people are exposed tothese disturbances also depends on manage-ment decisions. Frequently, there is public pres-sure to prevent these changes by increasingeffort to suppress wildland fires, building largerflood control structures, and using pesticides toreduce populations of pests and diseases (seeChapter 1). In the short term, these policiesmay reduce exposure, but in the long term, theyincrease the risk of larger, more severe dis-turbances (see Chapter 2; Holling and Meffe1996). An important arena of policy debate istherefore how best to minimize vulnerabilitythat results from climatically induced increasesin exposure to disturbances. In some cases thereare potentially simple solutions such as poli-cies that minimize development in floodplainsor fire-prone wildlands. In other cases, poli-cies designed to reduce exposure simply shiftthe vulnerability to certain segments of society(typically the disadvantaged) that occupy thelands with greatest exposure to hazards. Activepublic discourse on these issues can provide thebasis for evaluating vulnerabilities and inform-ing those potentially affected.

Air and water pollution also increase thehazards to which people are exposed. As withclimatically sensitive disturbances, disadvan-taged segments of society tend to live and workin places where they experience greater expo-sure to these environmental hazards. In somecases regulatory and market approaches canreduce the release of pollutants into the envi-

ronment and therefore the exposure of soci-ety and ecosystems to their effects. For exam-ple, international treaties such as the 1987Montreal Protocol, which banned the use ofozone-destroying chlorofluorocarbons (CFCs),led to a decline in the production and use ofthese compounds. As with all long-lived pol-lutants, there has been a long lag-time in thedecline in atmospheric concentrations of CFCsin response to these “cleanup” policies. Therate of ozone destruction by these compoundsis only now beginning to decline, 20 yearsafter policy implementation. This indicates theimportance of acting early to reduce pollu-tion, as soon as environmental consequencesare identified. There is currently active debateand experimentation with market mechanismsand regulatory limits to the quantities of CO2

that a country or firm can emit. In summary,there are a variety of regulatory and marketmechanisms, as well as the necessary technol-ogy, to reduce vulnerability to pollution. Unfor-tunately, there are often strong economic incen-tives at the levels of nations, regions, and firmsnot to pay the cost of pollution mitigation orto continue polluting illegally (see Chapter 4).Reducing vulnerability to pollution is thus moreof a political and social issue than a scientificchallenge.

Economic globalization increasingly linkseconomic stresses in vulnerable nations toglobal economic fluctuations. Although theselinks may improve living conditions in the shortrun, they may also create vulnerabilities as com-munities become dependent on services sup-plied by central agencies. Food aid, for example,can eliminate markets for local farmers, leadingto a decline in the in-country capacity to pro-duce food (see Chapter 12).

Endogenous Stresses

Endogenous stresses are affected by both eco-logical and social processes (Kasperson et al.2005). Declines in ecosystem services (e.g.,the productive capacity of soils, the filtrationof pollutants by wetlands, and pollination offood crops) exposes people to shortages offood and clean water. Such shortages interact


with socioeconomic factors to contribute tothe vulnerability of some segments of soci-ety, as described in general in Chapter 2, andin greater detail in Chapters 6–14. Similarly,chronic poverty and or the lack of freedomto make lifestyle choices are stresses that con-tribute substantially to vulnerability of dis-advantaged segments of society. Exposure toendogenous stresses is difficult to separate fromsensitivity to these stresses, so we treat bothexposure to endogenous stresses and sensitivityas sources of vulnerability in the next section.

Sensitivity to Hazards and Stresses

Sensitivity to hazards and stresses dependson ecological (see Chapter 2) and social sen-sitivity (previous section on livelihoods andwell-being), as well as the degree of couplingbetween social and ecological components ofthe system. A group with a high dependenceof locally harvested crops and limited inter-action with cash markets, for example, mayhave a high sensitively to drought. Sensitivi-ties can also depend on economic and socialconditions (e.g., wealth stratification, strengthof social capital, infrastructure availability), orpower relationships (e.g., caste systems).

Group members with greater authority andinfluence generally command more resourcesand are therefore less sensitive to hazards andstresses. Within a given location, people of dif-ferent race, gender, and age generally differ intheir power to access resources and/or to influ-ence decision making. Women, children, anddisempowered ethnic groups, for example, usu-ally have less access to resources, and this dif-ference in livelihood makes them more vulner-able to hazards and stresses. For example, tiesto children and social norms associated withtravel may limit the mobility of women in someregions of Asia, whereas men have greatermobility to pursue jobs in urban areas. Liveli-hoods and power relationships are deeply inter-twined, and together determine entitlements,which are the sets of alternative resourcesthat people can access by legal and customarymeans, depending on their rights and oppor-tunities (Sen 1981, Adger 2006). Entitlements

depend in part on the availability of resourcesin terms of built, human, and natural capitaland in part on institutions and power relation-ships that allow people to access resources (seeChapter 4).

Power relationships can vary geographically.People living in remote areas generally haveless influence to affect change than those inurban or government centers. Similarly, thosein inner city ghettos are marginal to those inmore wealthy suburbs (see Chapters 8 and 13).These power relationships can translate intohigher vulnerabilities to environmental shocksof people at periphery (e.g., high dependenceon central government support for crisis relief)than those at core areas (Cutter 1996; seeChapter 8). Nonetheless, those at the marginmay also have options for subsistence harvest-ing in conditions of scarcity that are unavailableto the rich and poor of the inner cities. The dis-tinctions between exogenous and endogenousstresses are difficult to tease apart in an increas-ingly connected world, where local groupswith sufficient resources can garner regional-to-international support for their needs and inter-est through use of the media, policy communi-ties, and strong leadership (Young et al. 2006).While a greater connectivity and stronger voicecan improve conditions for those at the margin,the power imbalances remain.

Cycles of Dependency

Endogenous stresses can be exacerbated whenrepeated patterns of dysfunctional social behav-ior become reinforced through time. Cyclesof dependency are often precipitated by wel-fare or land-use policies that can lure indi-viduals, households, and entire communitiesaway from traditional livelihood practices intosystems with disincentives for self-sufficiency,instead of the safety net or development thatis intended. In many regions of the underde-veloped world, cycles of dependency have beenlinked to the philanthropic activities of NGOs,which result in unanticipated consequences.The resulting conditions can entrap individualsand in some cases entire social groups, increas-ing their vulnerabilities to change. Breaking


patterns of dependency is not easy and needsto consider the broader social–ecological con-text. For example, in the 1880’s loss of tradi-tional livelihoods by several Native Americanstribes of the US West and the resulting cyclesof dependency resulted from both the tribes’exposure to foreign pathogens (e.g., small pox)and the Manifest Destiny policies of the USgovernment. Because it is impossible to restorefully any social–ecological system, finding away to escape from these patterns for NativeAmericans of the US West requires an inte-gration of both traditional and novel solutions.Development of credible and responsive insti-tutions that support self-determination of thoseaffected is an important ingredient in escapingdependency (see Chapter 4).

Social and Environmental Justice

The impacts of social and environmentalchanges are unevenly distributed both locallyand globally, making some regions and seg-ments of society more vulnerable than oth-ers. Environmental injustice—the uneven bur-den of environmental hazards among differentsocial groups (Pellow 2000)—results from dif-ferences in both exposure and sensitivity. Forexample, the arctic, mountains, and drylands,where climatic change is occurring most rapidlyand climatic extremes shape local culture andadaptation, are more exposed to the risks ofclimate change than are some other regions(McCarthy et al. 2005). Similarly, there is anet transport of persistent organic pollutants tohigh latitudes, where they are further concen-trated through food chains in animals harvestedfor food by indigenous residents (AMAP 2003).Local variations in exposure also occur, espe-cially for point sources of contaminants andpollution and for risks such as flooding thatare locally heterogeneous. Disadvantaged seg-ments of society are often most exposed toenvironmental hazards because they lack theresources to avoid exposure to existing hazardsor lack the political influence to prevent hazardsfrom increasing.

Those groups that are most exposed to envi-ronmental hazards are also disproportionately

sensitive to these hazards, because they lackthe material resources and safety from risks tocope and adapt effectively. These same individ-uals are also frequently disengaged from and inmany cases disillusioned with the political pro-cess, or lack the power to make their voicesheard. Consequently, addressing the issues ofsocial and environmental injustices generallyrequires a concerted and personalized effortto engage stakeholders in the decision-makingprocess to solve problems.

Sources and Strategiesof Resilience and AdaptiveCapacity

Sources of resilience in social systems providepeople with the means to buffer against change.Sources of resilience can be material, social,cultural, ecological, and intellectual. Resiliencecan also follow from a group’s capacity to inno-vate in the face of new or rapidly changingsocial–ecological conditions. The central issuehere is to avoid innovations and adaptationssocially and economically that provide short-term benefits at the cost of the longer-termcapacity of ecosystems to sustain societal devel-opment.

Spectrum of Adaptive Responses

Social–ecological systems exhibit a broad spec-trum of responses to hazards and stresses.These responses range from the immediateimpacts of hazards and stresses without adap-tation, to short-term coping mechanisms, toadaptation or transformation, in which adaptiveresponses foster favorable long-term social–ecological adjustments. As described in the pre-vious section, the immediate impacts of haz-ards and stresses are a product of interactionsbetween the exposure and sensitivity of thesocial–ecological system. Although, by defini-tion, immediate impacts involve no adaptation,they differ regionally and among segments ofsociety largely due to differences in poverty andwell-being.


Coping is the short-term adjustment by indi-viduals or groups to minimize the impacts ofhazards or stresses. Coping mechanisms enablesociety to deal with fluctuations that fall withinthe normal range of experience (Fig. 3.4; Smitand Pilifosova 2003). Coping sometimes occursby drawing on savings or social relationships,for example, by harvesting animals at times ofdrought or famine, borrowing from neighborsor family after a flood, or relying on extendedsocial networks during a regional crisis. How-ever, stresses that occur over a sustained timeperiod can lead to a decline in the availabil-ity of these resources. Alternatively, an increasein the population that is dependent on theseresources constrains a system’s coping capac-ity and the range of conditions with which itcan cope (Smit and Wandel 2006). Other fac-tors that reduce the coping range include exter-nal events and political factors (e.g., wars or lossof a leader) and increased frequency of stresses(e.g., drought).

Adaptation is a change in a social–ecologicalsystem that reduces adverse impacts of hazardsor stresses or takes advantage of new oppor-tunities (Adger et al. 2005). In this way, adap-tation acts as a stabilizing feedback that con-fers resilience and reduces vulnerability. Adap-tation occurs through social learning, experi-menting, innovating, and networking to com-municate and implement potential solutions(Fig. 3.5). These processes depend critically onpatterns of social interactions and group behav-ior, which are the main subjects of Chapter 4.Adaptive capacity refers to preconditions that

Mo

istu

re d

efic

it

Time

Coping range with increased adaptive capacityCoping range

Figure 3.4. Coping range and extreme events.Redrawn from Smit and Wandel (2006).

enable adaptation. Adaptations can differ inmany ways. For example, they can be reac-tive, based on past experience, or anticipatory,based on expected future conditions. Adapta-tion can also be local or widespread; techno-logical, economic, behavioral, or institutional(Smit and Wandel 2006). There is a continuumin the degree of adjustment and complexity ofadaptation ranging from simple coping strate-gies by individual actors to complex system-level changes. Some authors prefer to treatthe entire spectrum of adjustment as “adapta-tion” (Gallopin 2006). Substantive adaptationoften requires a substitution among forms ofcapital or a reorganization of the institutionalframeworks that influence the patterns of useof manufactured, human, and natural capital.Hence, adaptation is a continuous stream ofactivities, actions, decisions, and attitudes thatinform decisions about all aspects of life andthat reflect existing social norms and processes.Adaptation for resilience requires directing asocial–ecological system in a way that pro-vides flexibility and responsiveness in dealingwith disturbance and that allows a way to takeadvantage of the latent diversity within the sys-tem and the range of opportunities available(Nelson et al. 2007).

Transformation is the conversion to a new,potentially more beneficial, state with new feed-backs and controls when existing ecological,economic, or social structures become unten-able (Fig. 3.5; see Chapter 5). Transformation inthe social context may include radical changesin governance, such as the American Revolu-tion or the fall of the Soviet Union or shiftsfrom an extractive to a tourism-based economy.There is often a tension between efforts to fixthe current system and a decision to cut lossesand move to a qualitatively different structure(see Chapter 1; Walker et al. 2004). This is dis-cussed in Chapter 5 in the context of social–ecological transformation.

Fostering Diversity: Seedsfor Adaptation and Renewal

Diversity provides the raw material or buildingblocks on which adaptation can act (Fig. 3.5).


Well-being,sense of place,

knowledgesystems

Sociallearning

Immediateimpact

Adapted impact

Transformation

Experimentation, innovation,social networks

Hazards &stresses

New options

Social &biological diversity,

(redundancy)

Original options

System dynamics

Resi

lien

ce

Para

dig

m s

hift

Shared decision-making

Ad

apta

tio

n

Co

pin

g

Figure 3.5. Components of adaptive capacity andresilience. Adaptive capacity depends on capacity tocope with the normal range of variation in hazards

and stresses and the capacity to adapt through sociallearning, experimentation, and innovation.

It increases the range of options, at least someof which are likely to be successful under what-ever new conditions arise, thereby reducing thelikelihood of radical degradation of the system.It also provides redundancy, that is, multiplemeans of accomplishing the same ends, whichaugments the likelihood that valuable functionswill be retained during times of rapid change.Chapter 2 describes the role of biological diver-sity in fostering resilience and adaptation. Eco-nomic and social components of diversity canalso be important. The role of diversity insocial–ecological resilience is well-illustrated ineconomics. Economic diversity reduces vulner-ability and facilitates adaptation to economicor environmental change (Chapin et al. 2006a).For example, a nation whose economy is overlydependent on a single product is vulnerable tochanges in the demand for, or value of, thatproduct. Economic diversity can be enhancedby incentives that encourage entrepreneurshipand initiation of new types of businesses. Thisrole of subsidies contrasts strikingly with their

more typical role in supporting production insectors that have been adversely affected byclimatic, ecological, or economic change (seeChapter 10). Agricultural subsidies in Europeand the USA, for example, typically support thecontinued production of crops that would oth-erwise not be able to compete on the globalmarket with foods produced by countries withlower labor and production costs. In some cases,subsidies for non-competitive sectors of theeconomy are maintained simply because of thepolitical power of groups that benefit fromthem. In other cases, there may be good rea-sons to subsidize traditional sectors. Scandina-vian countries, for example, support agriculturein part to maintain an agricultural capacity asinsurance against global or regional conflictsthat might interrupt food imports. Agricultureor other traditional activities may also be subsi-dized for their cultural value to society, as in thecase of agriculture in northern Norway.

Once decisions have been made to provideincentives to foster economic innovation and


diversity, there are many potential mechanismsto accomplish this, including “carrot and stick”combinations, voluntary agreements, and taxincentive schemes (Baliga and Maskin 2003).Investment in education and infrastructure canincrease the capacity of local residents todiversify their economy and explore options.Economic diversification potentially enhancesresilience because it is unlikely that all eco-nomic activities would be equally sensitive toany given change in the natural environment orin economic and political conditions.

Diversification of livelihoods is a commonstrategy among groups in anticipation of stres-sors and shocks to the system (Ellis 1998). Mostrural peoples have considerable experiencemanaging exposure to risk that occurs within“normal” levels (Fig. 3.4). The development ofdiverse portfolios of livelihood strategies amongmembers of a household or kinship group is com-mon. Diversification may include, for example,planting a variety of crops because of uncertaintyin crop yields or markets. Another form of diver-sification is the cash employment by one or moremembers of a household in an off-site enterprisewhile remaining members work family-farmoperations. A similar role diversification is foundamong some contemporary hunting cultureswhere some family members work full-time jobswhile others are full-time subsistence harvesters,sharing their take with the extended family.

Diversification can include strategies otherthan increasing sources of cash. Householdsmay build their resilience by intentionallyextending their social networks to new groupsthat can provide additional resources or poten-tial support in times of hardship. It is note-worthy that diversification can be used by boththe wealthy and the poor and can contributeto equities and inequities. The elimination ofconstraints that allow for choice in diversifica-tion can enhance adaptive capacity and reducevulnerabilities (Ellis 1998). Diversification oflivelihoods may broaden the capacity of a groupto cope with change (Fig. 3.4), but its effective-ness may be limited if the shocks to the sys-tem are extreme. The capabilities of rich versuspoor households to diversify are also categor-ically different (BurnSilver In press). Well-offfamilies usually have the benefits of larger asset

portfolios on which to diversify their activities,and this can lead to greater wealth stratificationbetween wealthy and poor over time.

The role of cultural diversity in social–ecological resilience is complex. As witheconomic and biological diversity, culturaldiversity provides a range of perspectivesand approaches to addressing problems (seeChapter 4). For example, industrial fishing typ-ically uses nets with a single mesh size capa-ble of catching all large fish, while allowingsmall fish to escape. In contrast, Cree Indians inCanada use mixed-mesh nets that allow escape-ment of a range of size classes. The single-mesh nets are much more efficient at catchingvaluable fish. However, large fish are also dis-proportionately important in reproduction so,as fishing pressure increases, the loss of largefish can decimate the fishery (see Chapter 10).The diversity of fishing approaches providedby two cultures proved important in conservingthe fish on which the Cree depended (Berkes1995). Sometimes, however, cultural perspec-tives are so radically different or are so stronglyrooted in historical conflict that cultural diver-sity creates more challenges than opportunities,especially where institutions for shared decisionmaking are lacking. The major point here is thatcultural diversity broadens the range of per-spectives and experiences with which to addresschange.

Institutional diversity provides a range ofmechanisms for fostering innovation and imple-menting favorable outcomes and will be dis-cussed in Chapter 4.

Legacies and Social Memory: LatentSources of Diversity

Social legacies and social memory contributeto resilience by providing insight into thealternatives for responding to disturbance andchange. Social legacies are the lasting effectsof past events affecting current social condi-tions. Social memory is the collective memoryof past experiences that is retained by groups(Folke et al. 2003; see Chapter 4). Because itendures longer than the memory of individualpeople, it contributes strongly to resilience and


sustainability. Social memory includes both thewritten record (e.g., books, articles, reports, andregulations) and oral traditions and stories ofevents, strategies for coping with those events,and their lasting lessons. Thus, social mem-ory provides a wealth of ideas on how social–ecological systems and their components haveresponded and adapted to changes in the past.This memory extends the range of potentialfuture options beyond those that dominate thecurrent system (Fig. 3.5). For example, the sto-ries and memories of elders of many culturesprovide historical accounts of coping strategiesas well as insights that inspire innovation inproblem solving. Social memory can be partic-ularly important in times of crisis, when cur-rent options, by definition, are insufficient, anda broader range of possibilities is essential todiscover pathways to favorable outcomes (seeChapter 4). Social memory constantly evolvesin response to system change. Social memoryis important because it provides a context inwhich to frame novelty and change (Taylor2000).

People who are negotiating social–ecologicalsystems that are subject to disturbance andchange may draw on their internal social mem-ory as sources of resilience for renewal andreorganization. Others may use diverse memo-ries and recombine them in ways that lead tosocial–ecological transformations through, forexample, new inventions or leadership with newvisions or worldviews.

Individual actors, communities, and wholesocieties may successfully adapt to chang-ing social and economic circumstances in theshorter term, but such an adaptation may beat the expense of the essential capacity ofecosystems to provide support in the longerterm. For example, the unprecedented expan-sion of human activities since the Second WorldWar made possible through the exploitationof fossil energy has in many respects been animpressive human adaptation, with numerouslocal and regional transformations (see Chap-ter 14). However, the cumulative effects ofthese changes have contributed to global cli-mate change in the present.

To what extent can social memory contributeto such adaptations? People’s knowledge of

resource uses and ecosystem dynamics is oftentacit or implicit, linked to norms and rules,embedded in rituals and social capital, andframed by worldviews and belief systems. Thisis true for both traditional knowledge andthe local knowledge of contemporary societies.It is also true for the more formal science-based approaches to knowledge production.Having the capacity to tap into a group’s localand traditional knowledge, while cultivatingits development, is fundamental for integrat-ing livelihoods and well-being into resilience-based ecosystem stewardship. In many localcommunities there are management practiceswith tacit ecological knowledge that respond toecosystem feedbacks in ways that can enhancesocial–ecological resilience (Berkes and Folke1998a, Berkes et al. 2000). Yet as sources ofresilience, legacy and memory are subject tointerpretation. Indentifying key individuals ina group who hold a rich understanding of theevents of history and facilitating group pro-cesses that lead to reflection on past expe-riences can provide access to these sources.Through social learning, actors, networks, andinstitutions can integrate traditional knowledgewith current practices to find new ways to col-laborate (Schultz et al. 2007).

Innovation and Social Learning:Creating and Sharing New Options

Learning to cope with uncertainty and surpriseis a critical component of adaptive capacity.Individuals and groups are generally familiarwith short-term variability in the hazards andstresses to which they are exposed and havecoping mechanisms and adaptations that do notrequire learning new things. However, whenstresses or conditions change directionally (orvariability increases) and move beyond pastexperience, people may lack the social mem-ories and legacies to formulate an effectiveresponse. Learning is therefore essential to dealwith surprises—unanticipated outcomes or con-ditions that are outside the normal range ofvariability. There are at least three potentialresponses to surprise (Gunderson 2003): (1)No effective response occurs when inertia or


ineffectiveness in resource management pre-vents a response or when vested interestsblock changes that are attempted. This pre-serves the status quo and maintains or aug-ments vulnerability. (2) Response without expe-rience occurs if actions are taken in responseto novel circumstances or if institutions failto incorporate understanding gained from pre-vious experience. The outcomes from inexpe-rienced responses can be good, bad, or neu-tral. (3) Response with experience occurs whensocial learning has retained insights gainedfrom previous experiences (Fig. 3.5; Lee 1993,Berkes and Folke 2002). It occurs through theretention and sharing by groups and organi-zations of knowledge gained from individuallearning (see Chapter 4). Social learning istherefore important in increasing the likelihoodthat the system will respond based on expe-rience and understanding of system dynamics.Social learning can occur in many ways. Con-siderable learning occurs through the cumula-tive experience gained as local knowledge byobserving, managing, and coping with uncer-tainty and surprise. It can also occur throughscientific research or by reading or hearingabout the experiences of others. Local knowl-edge is critical for effective implementationof new solutions because it provides informa-tion about local conditions or “context” thatdetermine how to implement new actions effec-tively. Local knowledge also incorporates peo-ple’s sense of place reflecting how people relateto the social–ecological system in which theyare embedded and therefore how they are likelyto respond to, or participate in, potential newapproaches to ecosystem stewardship.

Social learning that contributes effectivelyto social–ecological adaptive capacity almostalways requires the integration of multipleperspectives and knowledge systems that rep-resent different views about how a systemfunctions (Fig. 3.5). Examples of comple-mentary perspectives and knowledge systemsinclude the natural and social sciences; aca-demics, practitioners, and local residents; andWestern, Eastern, and indigenous knowledgesystems.

Resilience learning is a form of social learn-ing that fosters society’s capacity to be prepared

for the long term by enhancing its capacityto adapt to change while maintaining andsustainability. Like other forms of social learn-ing, resilience learning benefits from cumu-lative experience, learning from others, andintegration of diverse perspectives and knowl-edge systems. However, resilience learning isparticularly challenging because it requires thatindividuals collectively consider the interac-tions of different components of the entiresocial–ecological system, rather than a singleaspect such as forest productivity or poverty,which are complex suites of issues in theirown right. The core of resilience learning isdeveloping the social mechanisms to makedecisions that enhance long-term sustainabil-ity and resilience under conditions of uncer-tainty and surprise. This is radically differ-ent than a paradigm of postponing decisionsuntil the group believes that it knows enoughnot to make mistakes. The latter approach isnever sufficient for addressing long-term com-plex issues and is often used as an excuse foravoiding short-term risks and maintaining thestatus quo. The decision not to take action is justas explicit a decision as acting decisively.

There are at least two important componentsof resilience learning: The first is to reduceuncertainty and improve understanding of thedynamics of social–ecological systems underconditions of variability and change so as toexplore new options. This can occur throughobservation, experimentation, and modeling(see Chapter 4). The second is the process ofsocial learning that requires framing the issuesof sustainability and resilience in a context thatconveys its value to the public and to decisionmakers (Taylor 2000).

Research on complex adaptive systems ischallenging because everything changes simul-taneously, and you can only guess what futurehazards and stresses may be. There are, how-ever, logical approaches to improved under-standing of resilience by (1) identifying criticalslow variables, (2) determining their responsesto, and effects on, other variables (often non-linear and abrupt); and (3) identifying pro-cesses occurring at other scales that mightalter the identity or dynamics of critical slowvariables.


There are typically only a few (about 3–5)slow variables that are critical in understandinglong-term changes in ecosystems (Gundersonand Pritchard 2002). Critical slow variables arefrequently ignored, assumed to be constant, ortaken for granted by policy-makers, who tend tofocus on the fast variables that fluctuate morevisibly and are often of more immediate con-cern to the public. The dynamics of ecologicalslow variables were discussed in Chapter 2. Theidentity of critical social slow variables is cur-rently less clear but probably includes variablesthat constrain well-being and effective gover-nance. Critical social slow variables are oftenstable or change slowly over long periods oftime, but may change abruptly if thresholds areexceeded. For example, property rights of manypastoral societies in Africa changed abruptly asa result of European colonization and policydebates over how to promote pastoral develop-ment, causing changes in land use, social disrup-tion and declines in well-being (Mwangi 2006).Oil development or building fences to creategame parks for ecotourism can today cause sim-ilar disruption to pastoral livelihoods. Many ofthe factors that disrupt well-being were dis-cussed earlier in this chapter, and factors influ-encing governance are addressed in Chapter 4.

Interactions among key variables operatingat different scales are also important causesof abrupt changes in social–ecological systems(Gunderson 2003). For example, interactionsamong changes in global climate and local pop-ulation density may cause soil loss in drylandsto exceed the changes that would occur basedon either factor acting alone. Research andawareness of processes occurring at a widerange of scales (e.g., the dynamics of potentialpest populations to behavior of global markets)can reduce uncertainty and enhance resilience(Adger et al. 2005, Berkes et al. 2005). Oftenthese cross-scale effects are difficult to antic-ipate, so managing to maintain or enhanceresilience of desired social–ecological statesmay be the most pragmatic approach to mini-mizing undesirable effects of cross-scale inter-actions.

Planning for the long term requires an under-standing of the tradeoffs between short-termand long-term costs and benefits. Actions that

contribute to long-term sustainability oftenincur short-term costs. For example, managingforests or economies for diversity usually incursshort-term costs in terms of reduced efficiencyand profits. Monospecific single-aged forestplantations, for example, can be harvested effi-ciently, and the harvested material can be trans-ported and processed to produce one or afew products, with substantial economies ofscale. These plantation forests are, however,more prone to losses from windthrow, disease,and changes in global markets than are morediverse multi-species stands and therefore lackresilience to long-term changes in these hazardsand stresses (Chapin et al. 2007; see Chapter 7).Substantial social learning must occur beforepeople or groups are willing to look beyond theshort-term costs and benefits to address long-term resilience and sustainability. In economicterms people tend to discount future value tosuch a degree that many decisions are basedprimarily on short-term rather than long-termbenefits. Politicians may be interested in max-imizing benefits over their term of office (afew years), and community planners seldomthink beyond 5–20 years. Resilience learningrequires a reassessment of these discount ratesand exploration of new options that extendthe time horizon over which benefits are val-ued by society. This is most likely to occur ifstrategies are developed that maximize syner-gies (win-win situations) and reduce the magni-tude of tradeoffs between short-term and long-term benefits (see Chapter 1). Now may bethe best time in history to convince society ofthe importance of long-term thinking becausethe magnitudes of directional changes that pre-viously occurred over many generations (cen-turies and millennia) are now evident withina single human lifetime due to the acceleratedrates of change and increased human longevity(see Fig. 1.1).

Incorporating Noveltyin Social–Ecological Systems

Experimentation and innovation allow peopleto test what they have learned from observa-tions and social learning. Managers are often


reluctant to engage in social–ecological exper-iments under conditions of uncertainty becauseof the risk that outcomes will be unfavorableto some stakeholders (see Chapter 4). An alter-native approach is to learn from the diversityof approaches that inevitably occur as a resultof temporal and spatial variation in land own-ership, culture, access to power and resources,and individual ingenuity. Multiple suburbs ofa city or school districts within a region, forexample, represent different efforts to address acommon set of problems. In general, this diver-sity in approaches leads to improved levels ofperformance, despite the inefficiency associatedwith redundancy (Low et al. 2003). Similarly,social–ecological differences between produc-tion forests, multiple-use forests, and wilder-ness conservation areas provide opportunitiesto learn about resilience to climatic, ecological,and economic shocks and openness to innova-tion and renewal (see Chapter 7). As expectedin complex adaptive systems, those experi-ments that are successful persist, and the fail-ures disappear. Social learning further increasesthe likelihood that successful solutions willbe adopted by others. Providing opportunitiesfor a diversity of social–ecological approachesspreads the risk of failure and the motivationsfor success.

Networking increases the efficiency of sociallearning. Two essential features of adaptationin any physical, biological, or social–ecologicalsystem are the presence of a diversity of compo-nents and the selection and persistence of thosecomponents that function effectively. In social–ecological systems, diversity is created throughexperimentation and innovation, and selectionoccurs through the process of governance—the pattern of interaction among actors thatsteer social and environmental processes withina particular policy arena. System-level adap-tation requires that successful innovation tocope with surprise be communicated broadly,so other people and groups can avoid the mis-takes of unsuccessful experiments and build onthe successes. These decisions and innovationstake place at the level of livelihood choicesof groups and individuals. Identifying success-ful options requires leadership and appropri-ate participation to select adaptation strate-

gies and an appropriate governance structure toshare information and resources and implementsolutions. These topics are central features ofChapters 4 and 5.

Participatory VulnerabilityAnalysis and Resilience Building

A key challenge of applying vulnerability andresilience frameworks to ecosystem steward-ship is developing qualitative and quantitativemeasures of the framework elements. It is alsoimportant to move beyond an examination ofthe vulnerability of only one aspect of the sys-tem (e.g., a fishing community’s fishing activi-ties) to a more holistic assessment (Allison andEllis 2001). The scale of vulnerability analysismust also be appropriate to inform those whoare vulnerable. National or global assessments,for example, might show that Bangladesh ismore vulnerable than Nepal, which is morevulnerable than Greenland. These top-downapproaches to vulnerability analysis providelittle direct help to those at the local level.Given the importance of informal social net-works for coping with shocks to a system andthe tendency of bureaucracies to capture thelion’s share of financial resources associatedwith disaster relief, engaging local communitiesin the assessment links theory and research withaction.

Participatory vulnerability analysis is a sys-tematic process of involving potentially affectedcommunities and other stakeholders in an in-depth examination of conditions. Participa-tory vulnerability analysis provides a basisfor formulating actions that support resilience-building. It can also be the basis for mit-igation. The steps for the analysis includeassessments of (1) exposure and sensitivitiesto determine past and current levels of risk,(2) historic strategies and constraints for cop-ing with risks, (3) identification of possiblefuture risks, and (4) development of plansfor mitigating and or adapting to future risks(Smit and Wandel 2006). To be effective,the analysis should use multiple sources ofknowledge (local and traditional knowledge,


modeling techniques, science-based analysis,remote sensing imagery) and address the multi-ple levels at which local, regional, and nationaldecision makers might be engaged to reducevulnerability. The objective is not to arrive ata single vulnerability score, but to build theresilience of the system to cope with anticipatedand potentially surprising change. Address-ing issues that reduce vulnerability and buildresilience is usually incremental, so the pro-cess must track emerging conditions and makethe necessary adjustments. The participatoryapproach can be helpful in energizing a commu-nity into action, but it must be approached witha grounded sense of realism that recognizesthe day-to-day demands of local residents (seeChapter 4).

Summary

Well-being based on adequate livelihoods isessential to reduce vulnerabilities to the pointthat people can engage in long-term planningfor ecosystem stewardship. Well-being dependson the acquisition of basic material needsincluding ecosystems services, as well as othersocial factors such as freedom of choice, equity,strong social relations, and pursuit of liveli-hoods. Once these basic human needs are met,people have greater flexibility to think cre-atively about options for ecosystem steward-ship to meet the needs of future generationsin a rapidly changing world. There has been atendency however, for people to seek greaterlevels of consumption, once their basic needsare met, even though this leads to no mea-surable increase in happiness. Two importantstrategies to enhance sustainability are there-fore to meet basic human needs and to shiftfocus from further increases in consumptionto social–ecological stewardship, once the basicneeds are met.

Social dimensions of vulnerability are inti-mately tied to elements of well-being, par-ticularly to livelihoods; safety and security;and good social relations. Vulnerability can beassessed by identifying (1) the type and levelof exposure of a system to hazards or stresses,(2) the system’s sensitivity to these hazards,

and (3) adaptive capacity and resilience of thesystem to make adjustments to minimize theimpacts of hazards. Although many systemsare experiencing increased stress from exoge-nous drivers like climate change and globaliza-tion of markets, the impacts of these stressesare strongly shaped by levels of well-being andassociated endogenous stresses, such as povertyand conflict. Both exposure and sensitivity tostresses vary tremendously with time and placeand among social groups. Reducing vulnerabil-ity therefore requires an understanding of spa-tial context and engagement of local stakehold-ers, particularly the disadvantaged, who mightotherwise lack the power and resources to copewith change.

Resilience thinking adds to resource stew-ardship by addressing the capacity of socialgroups to cope with change, learn, adapt, andpossibly to transform challenging circumstancesinto improved social–ecological situations.

Review Questions

1. What factors determine quality of life for anindividual, household, or social group? Whyis it important to understand people’s qualityof life and livelihood in a region if your goalis to foster ecosystem stewardship?

2. Describe how well-being influences expo-sure and sensitivity of a group to externalhazards and stresses. Why is it often insuf-ficient to simply provide food and shelter topeople who have experienced an event suchas a hurricane or drought if the goal of theaid is to reduce vulnerability?

3. What economic policies would be most likelyto increase regional resilience to uncertainfuture changes in climate and a globalizedeconomy? How might this be fostered?

4. How do conditions of rapid change enhanceand or confound a household and a commu-nity’s capacity to adapt?

5. What are some of the potential problems andpractical applications of vulnerability andresilience frameworks?


Additional Readings

Adger, W.N. 2006. Vulnerability. Global Environ-mental Change 16:268–281.

Berkes, F., and C. Folke. 1998. Linking social andecological systems for resilience and sustainabil-ity. Pages 1–25 in F. Berkes and C. Folke, edi-tors. Linking Social and Ecological Systems: Man-agement Practices and Social Mechanisms forBuilding Resilience. Cambridge University Press,Cambridge.

Kasperson, R.E., K. Dow, E.R.M. Archer, D.Caceres, T.E. Downing, et al. 2005. Vulnerablepeoples and places. Pages 143–164 in R. Has-san, R. Scholes, and N. Ash, editors. Ecosystemsand Human Well-Being: Current State and Trends.Millennium Ecosystem Assessment. Island Press,Washington.

Nelson, D.R., W.N. Adger, and K. Brown. 2007.Adaptation to environmental change: Contri-butions of a resilience framework. AnnualReview of Environment and Resources 32:doi10.1146/annurev.energy.1132.051807.090348.

Petschel-Held, G., A. Block, M. Cassel-Gintz, J.Kropp, M.K.B. Ludeke, et al. 1999. Syndromes ofglobal change: A qualitative modeling approachto assist global environmental management. Envi-ronmental Modeling and Assessment 4:295–314.

Smit, B. and J. Wandel. 2006. Adaptation, adaptivecapacity and vulnerability. Global EnvironmentalChange 16:282–292.


4Adaptive Co-managementin Social–Ecological Governance

Gary P. Kofinas

Introduction

Directional changes in factors that controlsocial–ecological systems require a flexibleapproach to social–ecological governance thatpromotes collaboration among stakeholders atvarious scales and facilitates social learning.Previous chapters showed that environmentaland social changes are rapidly degrading manyecosystem services on which human livelihoodsdepend. However, simply knowing that degra-dation is occurring seldom leads to solutions.People have tremendous capacity to modifytheir environment by changing the rules thatshape human behavior, yet much of the conven-tional thinking on resource management offerslimited insights into how to steward sustainabil-ity in conditions of rapid change. Consequently,there is a critical need to understand therole of people and their social institutions asmechanisms for negotiating social–ecologicalchange. Designing and implementing appro-

G.P. Kofinas (�)School of Natural Resources and Agricultural Sciences,Institute of Arctic Biology, University of AlaskaFairbanks, Fairbanks, AK 99775, USAe-mail: [email protected]

priate resource management in conditions ofchange requires an understanding of both theprocesses by which groups make decisionsand the mechanisms by which these decision-making processes adjust to change. It alsorequires moving beyond notions of resourcemanagement as control of resources and peo-ple, toward an approach of adaptive social–ecological governance.

Social–ecological governance is the collectivecoordination of efforts to define and achievesocietal goals related to human–environmentinteractions (Young et al. 2008). It is a processby which self-organized citizen groups, NGOs,government agencies, businesses, local commu-nities, and partnerships of individuals and orga-nizations are part of a stewardship process. Thisprocess may or may not involve government. Inthe context of sustainability, social–ecologicalgovernance addresses problems of maintain-ing social and natural assets while sustainingecosystem services (Folke et al. 2005). Adap-tive co-management emerges as an approachfor meeting this challenge through the inten-tional efforts at and across multiple scales tounderstand emergent conditions, learn fromexperience, and act in ways that maintain thedesirable properties of social–ecological sys-tems (Folke et al. 2002, Olsson et al. 2004a,Armitage et al. 2007). The ideas of adaptive


78 G.P. Kofinas

co-management build, in part, on the concept ofco-management, which is the sharing of powerand responsibility among local resource usercommunities and resource management agen-cies (Pinkerton 1989b, 2003) for more collab-orative and coordinated actions, and adaptivemanagement, which is an approach to resourcemanagement based on the science of learn-ing by doing (Holling 1978, Walters 1986, Lee1993). A successful adaptive co-managementprocess is (1) flexible and responsive to change,(2) focused holistically on social–ecologicalinteractions, (3) well informed by a diversityof perspectives, (4) reflective in decision mak-ing, and (5) innovative in problem solving. It isachieved through networks of decision-makingarrangements that are guided by good leader-ship to effectively link communities of resourceusers with regional-, national-, and global-scalegovernance (Olsson et al. 2004b, Armitage et al.2007). In practice, adaptive co-managementis implemented through a range of manage-ment activities, such as social and ecologi-cal monitoring and data collection; research,model building, and data analysis; habitatprotection and impacts assessment; enforce-ment; resource allocation; education; and pol-icy making. To build resilience, adaptive co-managers must integrate these activities withinand across organizational, spatial, and tempo-ral scales to facilitate learning, adaptation, and,where needed, social–ecological transformation(see Chapter 5). In these ways adaptive co-management is particularly suited to addressingconditions of dynamic change.

Realizing successful adaptive social–ecological governance is not easy becauseit requires collaboration and social learningacross multiple scales. Although there is nosingle governance arrangement appropriate forall resource situations (Ostrom et al. 2007),there are basic concepts, frameworks, andprinciples that can guide the design, imple-mentation, and evaluation of such a process.This chapter begins by exploring some of theshortcomings of conventional approaches toresource management, then presents six keychallenges of adaptive social–ecological gov-ernance. These challenges include (1) buildingresponsive institutions that provide for collec-

tive action; (2) finding a good fit of institutionswith social–ecological systems; (3) building andmaintaining strong community-based resourcegovernance; (4) linking scales of governance forcommunication, responsiveness, and account-ability; (5) facilitating adaptive learning; and(6) generating innovation.

Why Adaptive Social–EcologicalGovernance?

Decades of research and practical experienceprovide resource managers with many of thetools necessary to address critical resource chal-lenges, but are insufficient to manage social–ecological systems under conditions of rapidchange. Conventional resource management(sometimes called “scientific management”)often seeks to maintain ecosystems in a steadystate by perpetuating current desired ecolog-ical conditions. This approach is based onwell-accepted ecological principals and adjustspractices to fit local conditions. When appliedappropriately, steady-state resource manage-ment often organizes information in ways thatefficiently address specific problems, such ascompaction and erosion of overgrazed range-lands, site conditions that favor establishmentof preferred forest trees after logging, or themaintenance of desired stocking levels of fishor deer (see Chapter 2). These approachesto management have been somewhat effectivein addressing short-term variability during thegrowth and conservation phases of the adap-tive cycle, when conditions are either relativelyconstant or change in a predictable fashion(see Chapter 1). However, under conditionsof rapid directional social–ecological change(the release and renewal phases of the adap-tive cycle), new conditions emerge for whichsteady-state resource management is no longeradequate. Under these conditions, adaptive co-management provides mechanisms to adjust tochange (Brunner et al. 2005, Armitage et al.2007). Dietz et al. (2003) liken the challenge offinding appropriate management systems to aco-evolutionary race in which governance sys-tems must keep pace with social, economic and

4 Adaptive Co-management in Social–Ecological Governance 79

technological changes that increase the poten-tial for human impact on the biosphere, includ-ing the ingenious efforts by people to evaderules for conservation.

In many cases, steady-state resource man-agement occurs through top-down controlof people and resources by bureaucraticallystructured government agencies that operatethrough a set of well-defined rules and reg-ulations (Gunderson et al. 1995, Gundersonand Holling 2002). A command-and-controlapproach to management tends to focus onenvironmental and ecological controls andviews human behavior as exogenous to ecosys-tems (see Chapter 6). In a top-down approach,agency personnel typically assume responsibil-ity for enforcement of rules, with controver-sial decisions sometimes leading to complianceproblems, limited trust relations, and protractedconflict. Consequently, the top-down approachcan be insensitive and unresponsive to localconditions, human livelihoods, and communityconcerns.

Conventional approaches to resource man-agement can also become overly focused on“science of the parts,” while ignoring more inte-grated approaches to knowledge production.An integration of the parts requires recognitionof the range of social-ecological interactionsthat can occur and the importance of uncer-tainty in limiting our capacity to predict out-comes of these interactions (Holling 1998). Italso requires a culture of respect for differentdisciplinary and cultural approaches to knowl-edge production. In the absence of these con-siderations, resource managers and resourceusers frequently face surprises, such as signif-icant loss of resource productivity and shiftsin social conditions that previously supportedresource conservation (Holling 1986).

In some cases, the strategies of resource man-agement also include a narrow set of indica-tors of social welfare, employ worldviews thatseparate people from ecosystems, and assumethat it is possible to find substitutes for theloss of ecosystems and the services they gen-erate (Costanza et al. 1997, Folke et al. 1998).Such policies have been inadequate in address-ing social incentives, especially at the locallevel. These approaches have also assumed

a view that technological fixes would resolvefuture problems. Five “recurring nightmares”can emerge from such inadequate approaches(Yaffee 1997): (1) a process in which short-term interests outcompete long-term visionsand concerns; (2) conditions in which competi-tion supplants cooperation because of the con-flicts that emerge in management issues; (3) thefragmentation of interest and values; (4) thefragmentation of responsibilities and authori-ties (sometimes called functional silos or stovepipes); and (5) the fragmentation of informa-tion and knowledge, which leads to inferiorsolutions.

These nightmares, including the widespreadassumption in resource management that sys-tems are approximately at equilibrium andcan be maintained in their current condition,are exacerbated in conditions of directionalchange. Together they suggest that implement-ing an adaptive co-management strategy is animportant way to adjust to change, but onethat will be challenging to actors at all lev-els of the system. Implementation of adaptiveco-management therefore requires more thansimple public participation in decision mak-ing. It also requires collaborative processesof knowledge co-production and problem-solving at multiple scales, with a diversityof parties and incorporating knowledge ofecosystem dynamics. For local resource-basedcommunities, directional changes require anevaluation and readjustment of long-standingrules for interacting with ecological compo-nents of the system while at the same timeinnovating to sustain traditional ways of life.For subnational to national government agen-cies, directional change requires expanding thescope of resource management to consider adiversity of knowledge and to engage a broadset of public and private actors in makingpolicies at various levels. At the internationallevel, these changes necessitate new forms ofgovernance that assess global-scale phenomenawhile linking those assessments with regional-to-local processes in social–ecological systems(see Chapter 14). Working toward the objec-tives of adaptive co-management of ecosystemsalso requires careful consideration of the matchbetween institutions for decision making and

80 G.P. Kofinas

their specific social–ecological conditions. Forthese reasons, adaptive co-management mustbe evaluated not simply on its structural fea-tures (e.g., centralized to decentralized organi-zation) or by a set of management outcomes,but also by the extent to which its processprovides effective feedbacks that contribute torobust human responses in relation to changesin ecosystem dynamics. Achieving these endsrequires effective communication among keyactors, some level of intergroup cooperation, aswell as political know-how and maneuvering.This cooperation, in turn, requires an under-standing of the functions of institutions, humanorganizations, and their social networks in rela-tion to ecosystem feedbacks. Many of theseideas have been developed through the inter-national synthesis efforts, such as the Interna-tional Association for the Study of the Com-mons (NRC 2002, Ostrom et al. 2007) and theInstitutional Dimensions of Global Environ-mental Change program (Young et al. 2008).

Building Responsive Institutionsthat Provide for CollectiveAction

Institutions, organizations, and social networksaffect the feedbacks of social–ecological sys-tems and human well-being. Institutions aresocially constructed constraints that shapehuman choice (North 1991, Young et al. 2008)and are among the most important elements ofgovernance systems. Institutions are “the rulesof the game” that assign the roles assumedby individuals and organizations in society,direct the allocation of resources to indi-viduals and organizations, and affect humaninteractions (Ostrom 1990, Young 2002b). Inthe language of institutionalists, the rules ofthe game (i.e., institutions) affect the playof the game (i.e., human behavior, includ-ing human–environment interactions). Insti-tutions give rise to repeatable patterns ofhuman behavior that become the political, eco-nomic, and social interactions of society (NRC2002). For example, property rights, one ofthe most important types of institutions, are

those rules that govern access to and use ofresources (Bromley 1989, 1992) and thus reflectpower relations that affect social–ecologicaldynamics and resource sustainability. In manyrespects, institutions define a group’s human–environment relations, which in turn affectsocial relations among individuals and groups(Berkes 1989). Institutions also function asproblem-solving devices, providing road mapsof established processes by which emergingissues and conflicts are confronted, evaluated,and potentially resolved. The regularity of theiruse can produce a level of predictability amongactors that can result in trust among groups toengage in a range of social practices. To a largeextent, institutions dictate the rigidity and/orflexibility of a group’s decision making and thusdetermine the responsiveness of a group tochange. Institutions are also historical artifactsof human experience, evolving through timeand in a particular place or time. The coevo-lution of social–ecological conditions in a par-ticular context establishes a path dependenceof events that shape the direction and transfor-mation of institutional legacies (see Chapter 1;North 1990).

Institutions can be formal or informal, withthe interplay (interactions among institutions)between the two affecting social–ecologicalgovernance. Formal institutions, such as consti-tutions, laws, contracts, and legally based con-ventions, are typically written rules enforced bygovernments, such as the US National Envi-ronmental Policy Act, the World Trade Orga-nization Free Trade Agreement, and a localcounty’s ordinance on pollution prevention.Informal institutions are the unwritten rules ofsocial life, such as sanctions, taboos, customs,traditions, and codes of conduct that are partof the fabric of all groups. Examples of infor-mal institutions include the customary rulesfor managing sacred forests and the unwrit-ten sanctions for responding to social miscon-duct (e.g., shunning). Informal rules are alwayspresent and in some cases may be inconsistentwith formal rules. Because formal and infor-mal rules may emerge both as espoused rulesand as rules-in-use (Argyris and Schoen 1978),understanding their respective roles and inter-play may be critical in assessing the effective-


ness of a social–ecological governance system.For example, formal rules or policies such asa government regulation for land managementcan be imposed on a group, while informal rulesmay or may not change as a result of thoseimposed formal rules. Inconsistency betweenformal and informal rules can lead to dysfunc-tional behavior in a governance process, suchas conflicts over regulations and enforcementand covert activities, such as poaching. Effec-tive stakeholder engagement, e.g., through co-management, is critical to resolving problemsthat arise from inconsistency between formaland informal rules.

Institutions play out in the context oforganizations—social collectives with member-ship and resources, which thus differ from rulesof the game (Scott 2000). Human organizations,whether they are a small community, a cor-poration, a club, or citizens of a state, havestructural and behavioral characteristics thataffect their performance in addressing social–ecological governance problems. The organiza-tional environment is the context in which peo-ple assume roles of status and authority, amassresources and power, network to form coali-tions, and exercise property rights. Opennessof membership can vary in ways that affectuse of and relationship to resources. In somecases there is an ebb and flow of membership,such as a rural community experiencing boom-bust economic cycles. In other cases member-ship can be restricted, as in a closed familyclan, private club, or nation that limits accessto resources. As already noted, some organiza-tions are hierarchical, such as government agen-cies, others decentralized and uncoordinatedsuch as some community groups, and others aremultidimensional, highly complex, interorgani-zationally linked, and dynamically responsive.

Typically, large organizations with systemsof centralized authority function with predeter-mined procedures for decision making and areless responsive to change than smaller, moredecentralized organizations (Table 4.1). Theyalso tend to have greater disparity between for-mally stated rules and rules-in-use. Bureaucra-cies are generally predictable in their behav-ior, and, because of their centralized authority,they can be highly effective in mobilizing abun-

Table 4.1. Frequently observed differences betweencentralized and decentralized organizations.

Centralized organizations Decentralizedorganizations

Hierarchicaldecision-making

Democraticdecision-making

Tends to becomebureaucratic

Tends to remain flexible

Decision-making is oftenefficient

Decision-making istime-consuming

Common in largeorganizations

Common in smallorganizations

Prone to functional silos Effective integration ofinformation

Susceptible to agencycapture

Relatively predictable inoutcomes

Relatively unpredictable inoutcomes

Resistant to change Potentially responsive tochange

dant resources to address a problem, as with amilitary response. Bureaucracies are also sub-ject to goal displacement—a condition in whichthe survival of the organization assumes greaterpriority than efforts to meet its stated mission.Where a resource management bureaucracy isfocused on the needs of a single group, thatagency is subject to agency capture, a condi-tion in which a special interest group estab-lishes a controlling relationship, and the agencyworks almost exclusively on the interest group’sbehalf. Goal displacement and agency capturecan reduce the resilience of the system bylimiting the responsiveness of governance tochange. Conversely, highly democratic organi-zations generally require small membership, ahigh diffusion of knowledge among members,and considerable time to deliberate and makedecisions. All organizations are potentially sub-ject to the forces of bureaucratization—a pro-cess by which, over time, an organization hasan increased dependence upon formal rulesand procedures. This evolution frequently leadsto functional silos and related communicationproblems and in some cases increased rigid-ity. The collapse of a social–ecological sys-tem can have its roots in these institutionaland organizational dysfunctions, as noted inthe adaptive cycle described in Chapter 1. Ingeneral, centralized organizations work mosteffectively under relatively constant predictable

82 G.P. Kofinas

conditions, where there is broad agreementon management goals, whereas decentralizedorganizations are more effective in addressinguncertainty and change and therefore in confer-ring social–ecological resilience.

Individuals and organizations interactthrough social networks, the links that establishrelations among individuals and organizations(and their institutions) across time and space.Social networks and their supporting formal andinformal institutions distribute information andresources and can be important to the resilienceof a community. For example, institutions for thesharing of harvested animals among the !KungSan hunters of the Kalahari ensure survivalof the community when only some huntersare successful (Lee 1979). Market networksthat distribute goods and services globallycan function in a similar manner but can alsocreate vulnerabilities where there is overdepen-dence on external sources without adequatealternatives. The principle of “six degrees ofseparation,” which states that everyone onEarth is separated from any other person by amaximum of six linkages (Watts 2003), showsmathematically how interactions among largegroups and individuals can be closely linked.The development and disappearance of link-ages generates constantly shifting patterns ofchange in social networks, as would be expectedin any complex adaptive system. Linkagesof social networks can vary in strength. Thetheory of weak ties (Granovetter 2004) statesthat critical information is most commonlyreceived from those outside one’s strongersocial network, suggesting that resilience ismaintained through both bonding networks oflong-standing familiar relations and bridgingnetworks that connect individuals and groupsto sustain more generalized forms of reciprocity(Putnam 2000). Weak ties developed throughbridging networks can therefore generatenovelty and resilience and shape change.

The interplay of institutions, organizations,and networks constitutes the critical socialdynamics of a social–ecological governancesystem. Said another way, human organiza-tions define and modify institutions. Socialnetworks provide critical pathways for infor-mation exchange, stimulation for innovation,

and resource sharing; and institutions shapeindividual, organizational, and interorganiza-tional behavior. Social–ecological governanceunfolds in highly complex clusters of interac-tions involving institutions, individuals, organi-zations, networks, and ecosystems within andacross multiple scales. Infrastructure is anothercritical element in the interactions of social–ecological governance by enabling communi-cation among parties and affecting ecologicalprocess and the capacity of groups to respond(Anderies et al. 2004). When these interactionsinvolve a particular area of interest or a partic-ular resource, they are referred to as resourceregimes (Young 1982; e.g., resource regime forgovernance of Atlantic cod; see Chapter 11).Although it is possible and sometimes helpfulto examine the amplifying and stabilizing feed-backs of a resource regime, it is important toremember that all interactions of a particularinstitutional arrangement play out as open sys-tems affected by formal and informal as well asinternal and external forces.

A focus on the interplay of formal andinformal institutions, organizations, infrastruc-ture, and social networks shifts the discussionbeyond social–ecological system processes as amechanistic set of components and feedbacksand places human agency, the capacity of peo-ple to make choices and to impose their choiceson the world, as a central driver of the system.Humans are unique in their ability to transformtheir social–ecological environments. Throughhuman agency and collective action (i.e., thecooperation of individuals to pursue a goal),people hold considerable capacity to change therules by which they live, mitigate the causes ofchange, and/or adapt to change. By understand-ing better the relationship among institutions,organizations, networks, and human choice, webegin to explore how people can intervene tosustain the desired features of their world andhow they can learn from experience and workto shape emergent conditions.

No One Model, Discipline, or Solution

There is no one optimal institutional arrange-ment for sustaining resources and building


social–ecological resilience (Ostrom et al. 2007,Young et al. 2008). There has been substan-tial debate about the role of institutions inresource management and which institutionalarrangements are best suited to achieve the sus-tainable use of resources. Much of this debatehas centered on understanding how to avoida tragedy of the commons, an outcome inwhich the self-interest of resource consumersand poorly defined property rights result insignificant degradation of collectively sharedor common-pool resources (termed common-property resources in the early literature; NRC2002, Young et al. 2008). These include freshwater, marine fish, migratory species, the globalatmosphere, biodiversity, and public lands—resources whose integrity is critical to the life-support system of the planet and thereforeto human well-being. Common-pool resourcesconstitute some of the most important natu-ral resources but are highly vulnerable to over-exploitation because of an array of problemsassociated with securing their sustainability. Inthe extreme case these problems can lead toa tragedy of the commons (Hardin 1968) if(1) potential users cannot be excluded fromresource use—a situation called open access,(2) users have the capacity to use the resourcemore rapidly than its rate of renewal, and (3) allusers seek to maximize their self-interest tothe detriment of society as a whole (Acheson1989). Hardin considered the case of individ-ual herders who raise cattle on a village com-mons (i.e., grazing land shared by all villagers).In his account of the commons, economic incen-tives and the absence of institutions support-ing resource conservation motivate each herderto add additional cattle to his/her herd, eventhough each addition would lead to a degra-dation of the village commons at a cost to all.Hardin argued that because of the danger ofover-exploitation, all common-pool resourcesshould either be privatized or be managed withfull government control.

Although Hardin’s model for managing thecommons was initially widely accepted, it ulti-mately generated many questions. What is therole of small-scale institutions in managementof a commons? Should common-pool resourcesbe managed by privatizing resource owner-

ship or primarily by state government agen-cies, or should small communities with a depen-dence on those resources have the lead author-ity? Are there other alternatives? What arethe prospects for public–private partnershipsor hybrid management arrangements in pro-viding for robust social–ecological governance?Underlying these questions are others on howto define the nature of human rationality—the logic by which people make decisionsin resource management institutions. To whatextent are people means-ends oriented whenmaking decisions? When do they pursue indi-vidual benefits versus electing to self-sacrificeand act in the interests of their social collective?How does bounded rationality (deciding withincomplete information) affect decision mak-ing? To what extent are people cost-avoidant?And to what extent do culture and social condi-tions shape human choice situations?

Finding the appropriate solution for social–ecological governance is anything but simple.Extensive research shows that, although thereare many models of social–ecological gover-nance and many forms of institutional designthat can be applied to a resource problem, thereis no single institutional arrangement (or orga-nizational structure) that will guarantee the sus-tainability of a social–ecological system in allcases (Ostrom et al. 2007). There are simplytoo may unique variables in each situation thatare together too dynamic to allow for a gener-alized prescriptive solution. This “no one solu-tion fits all” principle has important implica-tions in the assessment of any social–ecologicalgovernance problem. First, it demands an in-depth understanding of the particular contextto assess conditions (Young et al. 2008). Whoare the players? What are their vulnerabilities?What are the values for the resources in ques-tion? What is the relationship between the useof the resource and the greater social order?What are the power relationships among play-ers? To what extent do ecological conditionscreate bottlenecks or opportunities for alterna-tive actions? What are the possible future statesthat may challenge the system? Simply trans-ferring the “answers” and solutions of resourcemanagement from one situation to another mayresult in a host of unintended consequences.

84 G.P. Kofinas

Table 4.2. Lens for assessment of institutional performance.

Lens Economic Cultural Political

Assumptions Rational choice; utility seeking Group values and norms guidebehavior

Conflicting interests; winnersand losers

Key concepts Incentives; free riding;transaction costs

Membership; identity;worldview

Property rights;interdependence

Moreover, the complexity of social–ecologicalgovernance and the limitations of any oneframe of analysis underscore the need for aninterdisciplinary approach when assessing aresource-governance problem. To address thesequestions, there is a growing body of literaturethat integrates multiple approaches to under-standing commons management.

The three sections below present economic,cultural, and political lenses for assessing thefunctionality and performance of resourceregimes (Table 4.2). Considered together, theyprovide an integrated framework for assess-ing the capacity of a governance system tobe responsive to change and ensure impor-tant ecosystem services and allow for theintegrated understanding of resilient social–ecological governance. In practice they requirethat analysts see problems from a diversity ofperspectives and embrace a variety of disci-plinary approaches to understanding problems.

The Economic Lens: Incentives,Transaction Costs, and SocialDilemmas

Institutions set up incentives that affect humanbehavior, functioning as reward systems thatmotivate certain behaviors and/or dissuade andpunish other behaviors (Bromley 1989). Incen-tive systems can reinforce individualistic behav-ior or encourage cooperation and collectiveaction among actors. Institutional incentivescan be monetary in nature, work through moralsuasion, and/or subject violators of rules to sig-nificant penalties.

Institutional arrangements that place theresponsibilities for sustainability on individu-als without making them sufficiently account-able to the group can lead to the shirking ofgroup responsibilities. The free-rider problem

occurs when individuals or groups of actorsdo not assume their fair share of responsibil-ities while consuming more than their shareof the resources. A free rider, for example, isan able-bodied village resident who gleans thebenefits of a communal irrigation system butshirks his or her responsibilities to help main-tain the system. Rent seeking is the seekingof profit or gain through manipulation of aneconomic or legal system rather than throughtrade or production. For example, profiteeringor monopolization of natural resources as aresult of land trades is a form of rent seek-ing by providing benefits to the individual with-out contributing to the productivity of soci-ety. At modest levels both free riding and rentseeking occur frequently and may have littleeffect on group behavior or the ecological sus-tainability of the system. At high levels, theycan have dramatic impacts and be indicatorsof institutional failure. Institutional arrange-ments can also set up incentives that encour-age the discounting of future resource ben-efits for the pursuit of current resource val-ues. For example, a policy that provides incen-tives to encourage economic development, suchas tax incentives for clear-cutting of a forest,may generate immediate financial benefits toa firm, increase the cash flow, and transfernatural capital into financial capital, but notprovide rewards for maintaining the value ofthose resources for future generations. Suchinstitutions do not account for the opportunitycosts of their actions—i.e., the potential ben-efits that might have been gained from somealternative action. Conversely, institutions thatprovide strong systems of accountability arebased on an adequate level of social–ecologicalmonitoring and are oriented toward a multi-generational view of human actions. Design-ing and transforming institutions to addressthe social dilemmas between individual and/or


group gain versus social benefits are among thegreatest challenges in achieving a resilient formof social–ecological governance.

Decision making involves transaction costs,such as search costs, information costs, nego-tiation costs, monitoring costs, and enforce-ment costs. High transaction costs can be abarrier to the emergence of collective actionbecause they can create difficulties in obtain-ing information or in monitoring the behaviorof resource users or ecological conditions. Theregularity and predictability provided by insti-tutions make the work of society more effi-cient by lowering transaction costs of decisionmaking. Attention to transaction costs reorientsrational choice assumptions of human behav-ior away from an exclusive focus on benefits byrecognizing that costliness of information gath-ering is a key impediment in decision making(Williamson 1993).

Conditions that commonly lower transactioncosts in social–ecological governance (and thusimprove the effectiveness of a resource regime)include homogeneity of the values and interestsamong actors and the longevity and multiplic-ity of relationships among interacting individ-uals and groups (Taylor and Singleton 1993).Collective action theorists argue that up-frontinvestments in communication and negotiationwill lower the transaction costs of enforcement,leading to higher compliance with rules (Hannaet al. 1996). In addition, groups that make up-front investments in detecting and understand-ing the dynamics of social–ecological changewill face fewer transaction costs when con-fronting the challenges of responding to thosechanges. Over time, groups that continue tointeract can improve their institutions for gov-ernance and further develop their capacity forproblem solving. The capacity to solve prob-lems and act collectively is termed social cap-ital, which provides the basis for respondingto unanticipated events (e.g., the emergenceof social–ecological surprise). Social capital isan important but unusual slow variable ofsocial–ecological systems. It is typically slowto develop but can erode quickly if there isa breakdown in institutions. It may developquickly when social actors perceive a commonthreat. For example, public response to the pro-

posed nuclear waste dump in New England ledan unlikely coalition of local residents with lit-tle mutual understanding or respect to speak inone voice and successfully confront the projectproponents.

The Cultural Lens: Identity, MentalMaps, and Ideology

Institutions represent cognitive, symbolic, andideological conditions embedded within a spe-cific social–ecological context. They are morethan simply a set of rules. From a culturalperspective, institutions shape the identity ofactors, generate common discourses, and drawindividuals into routinized activities that donot involve the calculations (e.g., cost–benefitevaluations) of decisions on a day-to-day basis(Douglas 1986, McCay 2002). Accounting forthe cultural dimensions of institutions there-fore highlights a suite of considerations thatwould otherwise be overlooked in a purely eco-nomic analysis. For example, language and therules of interpersonal and intergroup communi-cation shape patterns of communication, whichin turn determine how issues are discussed (e.g.,by what terms); which aspects of those discus-sions are ultimately deemed legitimate in deci-sion making; and which are marginalized. Forexample, while some Asian cultures have tra-ditions of indirect communication and defer-ential behavior in conflict situations, westerncultures often reward for directness, competi-tion, and confrontation. The cognitive or men-tal maps of a culture provide blueprints forgroup thinking that are typically implicit ingroup transactions. Rules for group member-ship establish social and organizational bound-aries in the transmission of ideas and informa-tion and can limit the extent of social networks.These institutional attributes are found in allgroups, from the family to the professionalorganization (e.g., Association of ProfessionalForesters), to the nation state, and beyond. Forexample, in resource management, organiza-tional culture in government agencies, such as anational forest service, is typically pronouncedand entrenched, and has a significant effect onthe interpretation of policies. These norms (i.e.,rules for social behavior) translate into how

86 G.P. Kofinas

staff members act on organizational policies,such as the appropriateness of public participa-tion in decision making or the role of traditionalknowledge in research (Mahler 1997). Embod-ied in the institutions of each culture groupis its worldview, the framework by which thegroup interprets events and interacts with itssocial–ecological system. For example, indige-nous hunters and gatherers commonly viewanimals as sentient beings whose actions caninclude emotion, revenge, and respect, analo-gous to human behavior. Resource managerswho rely solely on the methods of western sci-ence may reject these perspectives as “irra-tional” because of the problem of validat-ing such ideas with scientific test. Worldviewembodies whether human–environment rela-tions convey a sense of dominion over natureor one of deference and risk avoidance.

Negotiating the differences in worldviewbetween groups sufficiently to achieve effectivesocial–ecological governance may require thedevelopment of common vocabulary and mutu-ally agreed upon protocols to establish sharedvisions of problems. A diversity of perspectivescan contribute to problem solving. In othercases, these differences are irreconcilable. Prob-lems not withstanding, institutions that allowfor the insights of multiple perspectives, whilealso managing for conflict, potentially add tothe resilience of a social–ecological system.

The Political Lens: Power and PowerSharing

Power relations are an unavoidable aspect ofsocial–ecological governance that can eitherbuild or undermine social–ecological resilience.Power takes many forms such as the impositionof one’s will on others, the use of direct and/orindirect influence, the threat or use of harm,the power of personal characteristics such ascharisma, the power of argument, the powerof information and resource hoarding, and thepower of rewards (Lukes 2002). Power can beintentionally imposed or achieved through pas-sive or disruptive behavior. Questions of powerin social–ecological governance include whocontrols the process, who ultimately decides,

and who are the winners and losers. Politicalprocess may also involve issues of contestedmeanings. How a group perceives power, eitheras a finite resource or something that is unlim-ited and to be shared, often shapes its use.Power generally influences resource manage-ment decisions where players depend on oneanother to some degree, differ in goals andobjectives or in values about technologies or thedecision-making process, and consider the issueto be important (Pfeffer 1981).

Institutions affect power relations eitherdirectly or indirectly. For example, institutionscan impose ideological agendas, such as throughthe implementation of policies that foster theintroduction of market economies to third-world nations for economic development. Pow-erful interest groups can advance policies thatprivatize collectively held fishing rights andmake them transferable, which can indirectlyresult in a greater ownership of a fishery bythe industrial fleet and a crisis of sustainabilityfor local coastal communities (see Chapters 10and 11). Although power relations are alwayspresent, power can also be shared, contributingto collective action and cooperation.

Because sustainability, resilience, and vulner-ability are normative concepts (i.e., they assumea values orientation), they require a process fordefining their meaning and thus raise the politi-cal question “Sustainability for whom?”. Manyof the questions in defining sustainability aresocial value choices that lack a single “correct”answer. For example, determining a commu-nity’s critical level of approval for a particularissue (e.g., having a 51 vs 66% majority) is bestdetermined through a collective-choice processrather than scientific analysis. That being said,equity in power relations in social–ecologicalgovernance generally contributes to social–ecological resilience. Equity fosters resilienceprimarily by maintaining diversity of opin-ion, cultural orientation, socio–economic class,and empowering disadvantaged people to con-tribute to ecosystem stewardship. Pluralism insocial–ecological governance is an operatingprinciple that affirms and accepts a diversityof perspectives (Norgaard 1989). Institutionaldiversity can add to resilience by providing anarray of approaches to problem solving, which


may be critical in unanticipated future con-ditions. Practices of conflict resolution, suchas principle-based negotiation and third-partymediation or arbitration, are helpful tools inavoiding political gridlock when institutional-ized as part of the resource-management pro-cess (Wondolleck 2000).

In summary, economic, cultural, and politi-cal dimensions of institutions, when applied inisolation, provide an incomplete understandingof the system and can lead to different pol-icy outcomes (Young 2002b). Focusing exclu-sively on economic dimensions may result ina set of unintended consequences that do notaccount for cultural factors that drive groupbehavior. Conversely, images of social choicedictated solely by cultural considerations, suchas group identity or values, can overlook eco-nomic incentives that may lead to unsustainablebehaviors. Given that power relations permeateall human decision-making processes and deter-mining goals of sustainability and resilience areat some levels questions of values, attention tothe distribution of power highlights the issuesof social–ecological justice and the maintenanceof diversity. Striving to assemble a multidimen-sional or pluralist view can bridge these models,leading to more robust forms of governance.

Finding a Good Institutional Fitwith the Social–EcologicalSystem

The effectiveness of institutions in governanceof natural resources depends to a great extenton how governance fits with physical, ecologi-cal, and social conditions. Is the scope of thearrangement a good match for the problemsbeing addressed? Does the geographic scale ofa resource regime match with the distributionand movement of the resource? Are institutionsadequately sustaining the supporting ecosystemservices for important resources? Can the insti-tutional arrangement be responsive in a waythat matches expected rates of social–ecologicalchange?

The “problem of fit” can confound effortsto maintain essential ecosystem services,

lead to high transaction costs, rent seeking,and unpleasant surprises (Folke et al. 2007).Achieving a good fit between institutions,social-ecological systems, and their dynamiccharacteristics requires careful attention tothe specific conditions in question about thesystem, its exogenous drivers, and internalinteractions (Young 2002a). Because social–ecological systems are complex adaptivesystems with nonlinear dynamics, even themost careful considerations may lead toinstitutional misfits (Galaz et al. 2008) andsurprise.

Policy analysts have traditionally used threeterms to evaluate the appropriateness of par-ticular institutional arrangements to ecologicalconditions—subtractability, exclusion, and con-gestion (Fig. 4.1). Subtractable resources arethose that, when used, become less abundantor are degraded. Fish and trees, for example,are subtractable because, when used by one per-son, they are no longer available for use by oth-ers. Sunsets and radio programs, on the otherhand, are nonsubtractable because they are stillfully available to others, no matter how manypeople experience them. Many resources areintermediate in their subtractability, depend-ing on how they are used. Water, for exam-ple, can be used for hydropower generation andstill be available for other functions. Institu-tions for management of subtractable resources

Sub

trac

tab

ility

Common-poolresources

(e.g., migratoryducks)

Cost of exclusion

Private goods(e.g., car)

Club goods(e.g., golf course)

Pure public goods(e.g., military defense)

LowLow

High

High

Figure 4.1. Institutional arrangements that differdepending on subtractability and cost of exclusion ofgoods.

88 G.P. Kofinas

should include design characteristics that allowdecision makers to be sensitive to the natureof its subtractability, such as potential rates ofrenewal and nonreversible tipping points.

The degree of subtractability depends incomplex ways on many social and ecologicalvariables. The productivity of wildlife availablefor human consumption, for example, habitatconditions, life cycle bottlenecks in energy bud-gets, population cycling, and interspecific com-petition all influence production, so produc-tion and consumption are both highly vari-able and difficult to determine from their cur-rent values. To some extent, all resources arecoupled to their ecological context, and insome cases they are tightly coupled with spe-cific ecosystem properties (Berkes and Folke1998b). For example, the cyclical predator–preyrelations of hare and lynx constitute an inter-relationship between two resources, requiringa holistic approach to understanding ecosys-tems. Multiple forces for change, such as humanharvesting with climate change, add to thechallenge. The complex, nonlinear, and chaoticbehavior of ecosystems adds to the difficultly incalculating subtractability in natural resources(Wilson 2002). Assessments of subtractabilitycan also be confounded by the limitations ofhuman constructs used to understand biophysi-cal and social conditions (Young 2007). The useof ecological concepts such as carrying capacityor maximum sustained yield, for example, maylead to policy solutions that do not accuratelycapture ecological dynamics (see Chapters 2,6, and 11). The resulting misfit of institutionswith ecological dynamics suggests the need ofa more adaptive approach to social–ecologicalgovernance.

Excludable resources are those from whichpotential users can be excluded at low cost.Neighbors, for example, can be excluded fromharvesting potatoes from a private garden. Onthe other hand, it is more difficult to excludesomeone from watching a sunset or harvest-ing fish from the open ocean. Excludabilitydepends on the transaction costs of the exclu-sion process. It is much less costly to excludeneighbors from a potato patch than to pre-vent an industry in another country from pol-

luting the atmosphere, although both types ofusers could conceivably be excluded. Such costsdepend to a great extent on a group’s or indi-vidual’s access to power (e.g., money, influence,authority) over others. Governance that doesnot account for transaction costs associatedwith exclusion may result in failures becauseof unenforceable rules. Fugitive resources, suchas anadromous fish, migratory waterfowl, andrivers, move across a range of jurisdictions andcan have many user groups. Fugitive resourcesentail high costs of exclusion. In general, insti-tutional arrangements for resources with hightransaction costs, such as fugitive resources,require a multilateral approach and a high levelof collective action among groups.

Congestible resources decrease in qualitywhen the number of resource users reachesa threshold, for reasons other than ecologicalproductivity. Congestion can be a special man-agement problem for nonconsumptive resourceusers. For example, one’s wilderness experiencemay be unaffected by other visitors at low vis-itation levels, but increased visitation and thelack of appropriate institutions can cause con-gestion, which subtracts from the experienceof all visitors. An unacceptable level of con-gestion is, of course, a societal value judgment.Catch-and-release sport fishing may be ecolog-ically sustainable but not desirable when highnumbers of fishermen (combat fishing) appearin a prized fishing location. Nonlocal huntersat low levels may not be problematic to resi-dent harvesters, but at higher levels their pres-ence may subtract from the traditional hunters’ability to transmit knowledge and values aboutthe need to respect and care for land and ani-mals. Congestion problems can be addressedwith a policy for limited entry, such as requiringpermits in zones of use. Limited entry systemscan be effective but may entail high administra-tive costs. Alternatively, reducing (or simply notimproving) physical access to use areas, such aseliminating the maintenance of roads, can pro-duce the same outcome without the cost of per-mits and enforcement.

More recent analyses of institutional fit haveexpanded on these ideas to address a broaderand more complex set of social–ecological


Table 4.3. Misfits in Governance Adapted from Galaz et al. (2008).

Type of misfit Definition Examples Possible solutions

Spatial Does not match the spatialscales of ecosystemprocesses

• Administrative boundariesmismatch with resourcedistribution creatingcollective-action problems

• Local institutions unable tocope with roving banditsengaged in global market

• Central managers applyone-size-fits-all solutionsthat are inappropriate forlocal users

• Extend or create broaderjurisdictional authority

• Establish multiple-scalerestraining institutions

• Develop plan forco-management that allowsfor power sharing andmultiscale governance

Temporal Does not match the temporalscales of ecosystemprocesses

• Speed of invasive speciesfaster than responsivenessof policy-making process

• Refocus policy to includeconsideration of slowervariables

• Improve monitoring programThreshold behavior Does not recognize or is

unable to avoidsocial-ecological regimeshifts

• Sustained yield harvestingpolicies lead to collapse ofkeystone species

• Runoff from nitrogenfertilizer results in suddenchange in aquatic ecosystem

• Engage in integratedassessments with scenarioanalysis

• Implement adaptivemanagement to identifysolutions throughexperimentation

Cascading effects Unable to buffer or amplifiescascading effects betweendomains

• Reduction in polar ice capopens northern sea route toshipping, creating new formsof land use change andimpacts on local harvesters

• Engage in SES modeling tounderstand multidimensionaldynamics of system

conditions (Folke et al. 1998, Young 2002b,Galaz et al. 2008). Young’s (2002b) frameworkexamined the relationships between ecosys-tems, institutions, and human system charac-teristics, as well as their interactions, point-ing out how efforts to achieve a good fit areconfounded by the many dynamics of the sys-tem. Galaz et al. (2008) noted that mismatchesbetween institutions and ecosystems can be(1) spatial, where social–ecological governancedoes not match the spatial scales of ecosystemprocesses (e.g., too small to capture a full water-shed); (2) temporal, where social–ecologicalgovernance does not match the temporal scalesof ecosystem processes (e.g., the speed of inva-sive species encroachment outpaces the respon-siveness of society); (3) threshold behavior,where the social–ecological governance systemdoes not recognize or cannot avoid abrupt shiftsor social–ecological transformations (e.g., over-

harvesting of keystone species); or (4) cascad-ing, in which governance has limited capacityto buffer against or amplifies cascading effectsthrough the system (Table 4.3). In short, find-ing a good fit is difficult in all cases and,if social–ecological conditions change, requiresthe coevolution of institutional arrangementswith changing governance challenges.

Building and Maintaining StrongCommunity-Based Social–Ecological Governance

Local communities are a key component of anysocial–ecological governance system. Commu-nities represent the social space in which peopleinteract regularly, where people have access toand make use of resources for livelihoods, and

90 G.P. Kofinas

where the use of ecosystem services translatesinto social well-being. Strong local-scale sys-tems of social–ecological governance increasethe likelihood that governance at other scaleswill be successful (Dietz et al. 2003).

Major insights have emerged from recentresearch on the diversity of ways in which com-mons are managed at a local scale (e.g., McCayand Acheson 1987, Berkes 1989, Ostrom 1990).This research has shown that small-scale com-mons have been managed, in many cases sus-tainably for decades to centuries, primarilybecause local institutions evolved to overcomethe problems that Hardin had hypothesized inthe tragedy of the commons. An important ele-ment of success in community-based manage-ment of commons is that resource users foundways to exclude other individuals, preventingopen access to all users, while at the same timehaving strong internal institutions for resourcestewardship. There are cases, however, in whichcommon-pool resources have been degraded,frequently at times of rapid social or economicchange when long-standing self-regulating localinstitutions were undermined. These examplesillustrate ways in which people can manage (or

fail to manage) trade-offs between individualand collective well-being.

As mentioned, there is no universal solu-tion for sustainable management of a commons.When a commons is successfully managed,however, local-level institutions often proveto be important, especially when local resi-dents depend on the resources for their well-being. These local institutions often involvethe prevention of open access; the diffusion ofinformation in ways that facilitate shared deci-sion making; the development of cooperationand social capital and mechanisms to preventfree riding; and the minimizing of transactioncosts associated with communication, negoti-ation, monitoring, coordination, and enforce-ment. A high dependence on the resource,the total number of resource users, their het-erogeneity, leadership, and systems of reci-procity are also important to the success ofthese systems. Through comparative case stud-ies, Ostrom (1990) and others (e.g., Agrawal2002) have generated design principles for suc-cessful management of commons (Box 4.1).Figure 4.2 relates those principles to activitiesassociated in social–ecological governance.

Box 4.1. Design Principles for Successful Management of a Commons (Ostrom 1990, Dietzet al. 2003)

1. Clearly Defined Boundaries. The bound-aries of the resource system (e.g., irriga-tion system or fishery) or households withrights to harvest resource units are clearlydefined. Resource users understand theborders clearly, with respect to both geo-graphical extent of use and boundaries ofacceptable behavior.

2. Proportional Equivalence between Ben-efits and Costs. Rules specifying theamount of resource products that a useris allocated are related to local condi-tions and to rules requiring labor, mate-rials, and/or money inputs. This principleaddresses the need for a sufficient level ofequity in allocation and uses of resources.

3. Collective-Choice Arrangements. Manyof the individuals affected by harvest-

ing and protection rules are included inthe group that can modify these rules.Although the level of participation in rulemaking is situational, there is an effortto avoid decision making by arbitrary orcapricious means.

4. Monitoring. Monitors who actively auditbiophysical conditions and user behaviorare at least partially accountable to theusers and/or are the users themselves. Thissharing of roles has implications for theeffectiveness of social control mechanisms.

5. Graduated Sanctions. Users who violaterules-in-use are likely to receive gradu-ated sanctions (depending on the seri-ousness and context of the offense) fromother users, from officials accountable tothese users, or from both. Graduated


sanctions provide opportunities for indi-viduals and the social system as a whole torespond to changes and learn from experi-ence.

6. Conflict-Resolution Mechanisms. Usersand their officials have rapid access to low-cost, local arenas to resolve conflict amongusers or between users and officials.

7. Minimal Recognition of Rights to Orga-nize. The rights of users to devise their

own institutions are not challenged byexternal governmental authorities, andusers have long-term tenure rights to theresource.

8. Nested Enterprises. Appropriation, pro-vision, monitoring, enforcement, conflictresolution, and governance activities areorganized in multiple layers of nestedenterprises.

Several crosscutting themes emerge fromthese principles (Box 4.1): the need for resourceusers to comply with rules; the need forresource users to have a role the creation anddevelopment of rules; and the problem of exter-nally imposed rules that may be perceived asillegitimate by resource users. These designprinciples constitute a stabilizing feedback loopthat supports social learning—the process bywhich groups assess social–ecological condi-tions and respond in ways that support theirwell-being. Monitoring of resource use by usersmay result in detection of overexploitation orinappropriate uses, which, when appropriatelypunished, reduces the likelihood that theseusers or their neighbors will again overex-ploit the resource. Similarly, graduated sanc-tions and conflict resolution mechanisms pro-vide opportunities for individual and grouplearning through experience by allowing peo-ple to change their behavior over time. Changesthat frequently threaten common-pool resourcemanagement include changes in slow vari-ables that are more rapid than the systemcan adapt to; failure to transmit knowledgeacross generations, which may limit the capac-ity of communities to understand the histori-cal basis of ecosystems and governance; andthe lack of large-scale and well-linked institu-tions that relate community actions to regional-,national- and global-scale interactions. As pre-viously noted, cultural factors also shape theoutcomes of these processes, such as the extentto which a group perceives a threat to overex-ploitation or resource degradation, the degreeto which the group shares a common identity,

and the depth and richness of local knowledgein the governance process.

Community social–ecological governance,like all forms of social–ecological governance,is not a panacea (Ostrom et al. 2007). Localleaders can be subject to corruption, resourceusers can discount future values of resourcesif presented with the wrong incentives, andlocal knowledge holders can be wrong aboutthe causes of social–ecological dynamics. Nev-ertheless, strong local institutions for gover-nance of common-pool resources serve as animportant foundation in a world in which cross-scale interactions (down- and up-scale effects)increasingly affect social–ecological processes.During the past several decades there has beena global trend to decentralize decision makingfrom national-level governance to regional- andlocal-level social–ecological governance, as wellas an emergence of a more global governanceprocess (see Chapter 14). These trends sug-gest new opportunities to develop local-scalesocial–ecological governance while raising chal-lenges in understanding the role of commu-nity governance in a global context. Althoughthe lessons learned from local social–ecologicalgovernance are, to some extent, unique to thelocal scale, there are interesting similaritiesto global-scale governance. These similaritiesinclude a substantial dependence on consensusforms of decision making and informal institu-tions (Koehane and Ostrom 1995). In contrast,national and subnational governance are gener-ally more centralized and top-down in authorityand hierarchal in structure.

While the role of communities in social–ecological governance is important, the forces

92 G.P. Kofinas

Natural resourcegovernance principles

Natural resourcegovernance principles

Requirementsfor social-ecological

governance

Devise rules that arecongruent with

ecological conditions

Clearly define theboundaries ofresources and

user groups

Devise accountabilitymechanisms for

monitors

Apply graduatedsanctions for

violations

Establish/use low-costmechanisms for

conflict resolution

Provide necessaryinformation

Deal with conflict

Induce compliance with rules

Provide physical. technical andinstitutional

infrastructure

Encourageadaptation and

change

Involve interestedparties in informeddiscussion of rules

(analytic deliberation)

Allocate authority to allow for adaptive

convenience atmultiple levels from

local to global(nesting)

Employ mixtures of institutional types

(institutional variety)

Figure 4.2. The general principles for robust gov-ernance of resource regimes (boxes on right andleft) interact with requirements for governance ofsocial–ecological systems. The likely connections

between principles and requirements are presentedwith arrows, which are relevant to local to regional-and global-scale problems. Redrawn from Dietz et al.(2003).

of globalization are making the conditions thatsupport self-governing small-scale resource sys-tems increasingly rare (Dietz et al. 2003). Forexample, roving bandits in the form of piratefishers that range widely to exploit resourceswithout regard to established institutions, rep-resent a significant challenge to local sustain-ability (Berkes et al. 2006; see Chapter 10).Concurrently, there is interest in forms of globalgovernance that address global-scale issues,such as climate change (Biermann 2007; seeChapter 14). Together the potential contribu-tion of strong local-scale systems of governanceand the need for global-scale governance points

to the importance of coordination and account-ability between local, regional, national, andinternational institutions.

Linking Scales of Governance forCommunication, Responsiveness,and Accountability

The increase in geographic connectivity associ-ated with globalization and the complexity ofsocial–ecological governance necessitates effec-


tive cross-scale interactions among institutionsand organizations. In such an environment,local communities, regional and national gov-ernment agencies, research institutes, and inter-national NGOs increasingly interact, formingsocial networks with ties of different strength.Linkages between institutions, organizations,and social networks (similar to what Ostromrefers to as “nested systems”), facilitate inter-actions, providing opportunities for those at onescale to deliberate and solve problems with oth-ers at another scale. In a directionally changingworld, neither top-down nor bottom-up interac-tions are the preferred direction of interaction,but instead two-way transactions are neededto account for observations, understandings,and human needs as perceived at the variouslevels.

There are two general categories of cross-scale linkages (Young 1994, Berkes 2002,Young 2002a). Horizontal linkages occur at thesame level of social organization and acrossspatial scales, for example, in treaties amongcountries or regional compacts. Vertical link-ages occur as part of a hierarchy of interor-ganizational relationships at the same location,such as in the transactions between local- andnational-level management systems that oper-ate in a given region. Linkages that include richinformation exchanges and shared processesfor defining problems and seeking solutionsto problems contribute to the effectivenessof social–ecological governance as do thosethat address accountability across scales. Thelinkages are typically more complex than local–national or lateral relationships and are ori-ented around specific issues, creating polycen-tric governance, a complex array of interactinginstitutions and organizations with overlappingand varying objectives, levels of authority, andstrengths of linkages (Ostrom 1999b).

Deciphering cross-scale institutional inter-play helps to identify inconsistencies in rules,asymmetries or gaps in information flow, politi-cal inequities, and their respective implicationsfor sustainability. Once these inconsistenciesare identified, there are often high transactioncosts required to resolve differences in juris-diction of authority, address power inequities,and develop trust and effective communica-

tion. Sometimes these challenges associatedwith effective linkages are overwhelming, espe-cially when time is limited. For example, theinstitutional failures in the post-Katrina hurri-cane events in New Orleans in 2005 are largelyattributable to inadequate coordination amongscales of governance ranging from neighbor-hoods, enforcement agencies, and emergencyrelief organizations to the policies of state- andfederal-elected officials. However, overlaps inauthority can also provide resilience throughredundancy, because when one of these insti-tutions is ineffective, other institutions maystill be able to accomplish many of the tasksrequired for effective social–ecological man-agement (Berkes et al. 2003).

Strong linkages can also have negative con-sequences that cascade downward to local-scale implementation of governance. Thesemay involve the shifting of management toexclude local knowledge; a colonial perspec-tive that subjugates local authority; the nation-alization of resources that were formally gov-erned successfully through local institutions;and the expansion of commodity-based marketsfor development, with development definedby central (nonlocal) authorities. To coun-terbalance these problems, there are severalstrategies that can allow communities to affectactions at other scales, strengthen steward-ship by keeping local systems accountableto broader principles of sustainability, revital-ize cultural and political empowerment, buildcapacity of local organizations to engage inactivities (political and otherwise) at higherlevels, and help in the establishment anddevelopment of institutions (Berkes 2002,Table 4.4).

Co-management is the sharing of power andresponsibility between resource users’ commu-nities and government agencies in managementof natural resources (Pinkerton 2003). Powersharing through co-management arrangementscan take many forms. It can be establishedthrough formal contractual agreements thatoutline the authority of the arrangement invarious functions of management [e.g., indige-nous land-claim settlements (see Chapter 6) orwater management districts (see Chapter 9)].Co-management can also be informal, based

94 G.P. Kofinas

Table 4.4 Strategies for cross-scale linkages.

OrganizationalOrganization Goal relationship Typical type Short- or long-term

Co-management Sharing of power Agency-community Long-termcollaboration

Short and long

Special interestpressure groups

Influencingmanagement

User organization Issue-based Short

Policycommunities

Influencingmanagement

Multiorganization Issue-basednetwork

Short and long

Social movements Achieve a socialgoal

Multiorganization Issue-basednetwork

Short

Multistakeholderbodies

Scope issues Agency-initiated; Short-termsolutions sought

Generally short,depending onmandate

Publicconsultationprocesses

Hear fromcommunity

Agency-initiated Short-termconcerns; openparticipation

Short

Development andempowerment

Capacity building Vertical NGO orgovernment aid

Short and long

Citizen science Improveunderstanding

Vertical Task-focused Short and long

on personal relationships between agency per-sonnel and local community members. “For-mal co-management agreements” do not alwaysachieve power-sharing, so care must be taken toclarify the terms of working relationships anddevelop the social capital necessary for effec-tive joint problem-solving. Effective involve-ment of co-management partners in policymaking, especially with respect to policies onresource allocation and in the interpretationof research findings, is critical to achieving“complete co-management” (Pinkerton 2003).Many formal co-management arrangements areimplemented with bridging or boundary orga-nizations (boards and councils of representa-tives) that link local communities to policyprocesses at a regional scale. Co-managementoften requires the modification or developmentof new pathways of communication as well asstrong leaders who hold bicultural perspectivesand broker cross-scale solutions. Commonlyaccompanying these and other types of link-ages are shadow networks, groups that are indi-rectly involved in decision making, but support-ive to the process, such as academic researcherswhose work supplements the process (Olssonet al. 2006).

Cultural differences, communication prob-lems, or differing perceptions of appropriateforms of decision making (e.g., consensus vsbureaucratic) can confound efforts to achieveeffective representation in these processes.Even with these challenges, co-managementhas proven in many cases to build commonunderstanding of problems across scales andcan be important where a commons requires ahigh degree of coordinated action among geo-graphically dispersed parties.

Other strategies serve different functions inlinking groups with decision making. Multi-stakeholder bodies, for example, provide policymakers a way to scope issues, seek solutionsto problems, achieve broader public partici-pation, and foster public consensus. They areoften short-lived bodies that lack managementauthority and therefore tend to function pri-marily as forums for discussion, planning, andconflict resolution. However, in cases wherethey have a mandate to negotiate and someauthority to affect policy outcomes, they havethe potential to resolve conflicts, set direc-tion for future actions, and achieve “win–win”solutions (Wondolleck and Yaffee 2000). Pub-lic consultation processes, another strategy for


linking across scales, can be used by govern-ment management agencies to inform decisionmaking about the interests and concerns oflocal communities. For example, environmentalimpact assessments typically provide for pub-lic review and commenting, allowing a formaldocumentation of local users’ perceptions ofpotential impacts. In some cases, public issuesraised in these reviews must be included inthe public record and require that agenciesrespond to them in writing as a part of thefinal impact assessment. Focus groups, publicmeetings, mail-out, and other forms of surveyresearch can also be used for accessing pub-lic input. The strength of the linkage betweenthe public voice in public consultation processesand the final decision depends to a great extenton the receptiveness of government administra-tions and the ability of public interest groupsto organize and articulate their concerns andinterests.

Development and empowerment organiza-tions can work with stakeholder groups tobuild their capacity to address problems, bothcurrent and future, and communicate betterinternally and with other groups. These orga-nizations are typically NGOs, although somefederal programs of developed countries workin developing countries to achieve the sameobjectives (see Chapter 6). The Nature Conser-vancy of the USA, for example, has facilitateda range of “community partnership” programsthat convene key players to address presentand future needs of community developmentand the habitat conservation actions that canbest meet those objectives. Larger-scale orga-nizations with a development and/or empow-erment mission provide important networkingservices to groups in ways that bridge deci-sion making across scales. To be successful, theymust work through a process that avoids depen-dence on those organizations that exacerbateproblems.

Special interest groups, policy communi-ties, and social movements seek managementpolicies and actions that foster their specificinterests. Special interest groups typically self-organize to advocate for policies that servetheir interests at local to regional scales. Hunt-ing organizations and the timber industry, for

example, can organize to seek policies thatfacilitate harvest of specific resources; conser-vation NGOs may seek to reduce these har-vests; and the tourist industry may seek regu-lations that foster infrastructure developmentfor recreation. Policy communities are looselyaffiliated groups that share an interest in one ormore specific issues and collaborate to changepolicy. They are typically interorganizationalnetworks of NGOs, international governments,agencies, and local communities. Collaborativealliances of this kind typically must go throughstages of development, including problem def-inition, direction setting, implementation, andevaluation (Gray 1989, Gray and Wood 1991).Failure to move through these stages can limitthe groups’ collaboration (Kofinas and Griggs1996). Social movements operate over broadergeographic scales than special interest groupsand can be cultivated by organizations and/orstrong leadership to advance a particular ide-ological perspective. Those involved in socialmovements may have little or no history ofworking together and no recognized leaderor organizational center. However, they cantake on a life of their own to become apowerful force for realizing a common vision.For example, many distinct indigenous groupshave created social movements in their effortsto achieve international indigenous empower-ment. The wilderness movement has similarlycreated a push toward wildland preservation.Similarly, many sovereign states have coordi-nated to create a global free-market economythrough the establishment of the World TradeOrganization.

Programs in citizen science provide oppor-tunities for local residents to share observa-tions and, in some cases, local understanding ofenvironmental process with groups operating atother scales. They have been extensive in theirreach and provide information in the assess-ment of environmental change. The ChristmasBird Count, for example, has operated since1900 as a citizen-based network of bird watch-ers in the USA that annually documents thepresence of bird species in their communities.Frog Watch is another geographically exten-sive program of citizen science that has bridgedlocal observations with regional- to global-scale

96 G.P. Kofinas

observations of environmental change. Similarprograms have been established with indige-nous hunters of Africa and the Arctic. Theseprograms can parallel advocacy for specialinterest (e.g., Audubon and its advocacy foravian fauna) or be a complementary part of aco-management arrangement. An example ofone of the most successful areas of citizen sci-ence is watershed stewardship programs (seeChapter 9). Increased connectivity through theinternet is extending the capacity of citizengroups to have influence across scales. In thecontext of citizen science, it is also providingan educational function through classroom pro-grams such as “GLOBE”—an Earth-SystemScience education program that facilitates inter-national interactions among students. Anotherside benefit of citizen science is that citizenslearn about environmental issues, and theirparticipation builds commitment to resourcestewardship.

The effectiveness of each of these strate-gies in linking across scales depends to a greatextent on the sources (money, infrastructure,human capital, organizational capacity) andbalance of power among groups. Formal insti-tutions that recognize the rights of resourceusers may enable local communities to activelyengage regional- and national-level decisionmakers in problem solving, yet the organiza-tion of an active policy community or an agencyculture sympathetic to local interest may pro-vide strong linkages in the absence of formalinstitutions (see Chapters 6 and 11). The ben-efits of cross-scale linkages follow when theinteraction of parties facilitates the building oftrust, mutual respect among parties, and sociallearning.

Facilitating Adaptive Learning

Adaptive learning as a part of resource man-agement provides the means for coping withuncertainty and change in a social–ecologicalenvironment. Adaptive learning in governanceoccurs when one or more groups (1) care-fully and regularly observe social–ecologicalconditions, (2) draw on those observations to

improve understanding of the system’s behav-ior, (3) evaluate the implications of emergentconditions and the various options for actions,and (4) respond in ways that support theresilience of the social–ecological system. Face-to-face deliberation is a vital component of thisprocess, providing for reflection on past experi-ence, revisiting of basic operating assumptions,and careful evaluation of policy alternatives.Adaptive learning in social–ecological gover-nance can occur within and across a number ofgeographic and organizational scales, as notedin the section above. For example, adaptivelearning within a single organization, such as asingle agency may occur through periodic andwell-structured planning processes. Bridgingorganizations, such as management councils,can function as central nodes of cross-scale net-work interactions in these processes. Broadersocietal-level social learning can also unfoldthrough a more diffuse process of communi-ties of practice or learning networks. At thislarge scale the process may be less structuredbut still part of a governance process in whicha society sets its course for the future. Link-ing adaptive learning at the level of the bridg-ing organization with society at large can be aconsiderable challenge where there is limitedsocial capital and no effective communicationsstrategy.

Adaptive learning in social–ecological gover-nance follows from the idea of adaptive man-agement, which is defined as resource manage-ment based on the science of learning by doing(Walters 1986, Gunderson 1999, Lee 1999).Embracing uncertainty is a central tenet ofadaptive management. Uncertainty in resourcemanagement can result from a lack of, or incor-rect, information about environmental factorsthat influence a given situation (observationalerror), a lack of understanding of the underly-ing social and/or ecological processes (processerror), a lack of knowledge about the effectsof incorrect decisions (model error), and/or theinability of the decision maker to assess thelikelihood that a factor will affect the successor failure of the system (implementation error;Francis and Shotton 1997). From a statisticalperspective, uncertainty involves Type 1 errors,which are the claims that something is true


when in actuality it is not (a false positive),and Type 2 errors, which are failures to dis-prove a null hypothesis when another is true(a false negative). Beyond statistics there isalso the error of getting the right answer to awrong question. Uncertainty in resource man-agement can also be cognitive (i.e., mentallyconstructed), such as ambiguity in terminologyof concepts used by groups in the decision-making process.

While it may be compelling to reduce uncer-tainty, an alternative approach is to assume thatuncertainty is the norm and therefore makedecisions based on risk evaluation and reduc-tion rather than predicting a precise futurepath. In such an approach, uncertainty is less ofa central problem to resolve and more a condi-tion to navigate (see Chapters 1 and 5). Evenif assumed, uncertainty presents a number ofchallenges for resource managers, particularlywhen management is undertaken on an ecosys-tem basis. The challenges follow from the com-plex adaptive nature of interacting componentsof a social–ecological system and the limitedability of people to make predictions about thefuture (Wilson 2002). Yet, in the face of uncer-tainty, managers and resource users alike needto act, typically before scientific consensus isachieved (Ludwig et al. 1993); inaction (or nodecision) is in fact a decision that can lead tosurprise (see Chapter 11).

Adaptive management assumes conditionsof uncertainty by viewing policy actions ashypotheses that are tested through policyimplementation and later evaluated to improvethe state of knowledge and practice. Key objec-tives of adaptive management are to anticipatemajor surprises where possible, limit their nega-tive consequences when they do occur, and usechange as an opportunity for adaptive learning.Over time, adaptive learning is a cyclical pro-cess by which observations inform understand-ing, which informs decision making, and so on.Adaptive learning ultimately requires a deliber-ative process by which the state of knowledge isreviewed, evaluated, and, where needed, modi-fied based on improved understanding and thedesired future. The distinguishing feature ofadaptive management is double-loop learning,the feedback process in decision making by

which practitioners reflect on the consequenceof past actions before taking further action(Argyris and Schoen 1978, Senge 1990, Argyris1992). Double-loop learning differs from single-loop learning, which adjusts actions to meetidentified management goals (e.g., modifies har-vest rate to conform to specified catch limits)but does not evaluate basic assumptions andapproaches (Fig. 4.3). It also differs from deci-sion making by “muddling through,” a pro-cess of trial-and-error decision making in whichdecisions are based primarily on politics with-out the benefits of careful reflection (Lindblom1959, 1972).

Ideally, policy actions are viewed as exper-iments that are analyzed in subsequent cyclesof adaptive learning. Policy changes can beoperational, such as a change in a stipula-tion for construction of new infrastructure; orregulatory, such as a change in the harvest-ing period. Building on the idea of policiesas experiments, the process can be undertaken

Sing

le-lo

op le

arning

Do

ub

le-l

oo

p le

arni

ng

Monitoroutcome

Evaluateoutcome

Maintain or adjustdesign to meetthe same target

Implementaction

Consider whether to alter

policies

Re-evaluateassumptions,

key relationships, and mental

models

Re-evaluate and decide on policies

Figure 4.3. Single- and double-loop learning. Single-loop learning involves changing actions to meet iden-tified management goals (e.g., modifying harvest rateto conform to specified catch limits), often throughtrial and error. Double-loop learning includes areflection process of evaluating underlying assump-tions and models that are the basis of defining prob-lems (e.g., revising the indicators and simulationmodels used to calculate allowable catch from stockassessments).

98 G.P. Kofinas

as active adaptive management through theintentional manipulation of the system to testits response or passive adaptive managementthrough an intensive examination of historiccause–effect relationships. Scale is an importantdeterminant in the appropriateness of activeand passive approaches to adaptive manage-ment. For example, experimentation in man-agement of hydrological systems for an oil fieldmay glean good insights into improved designsthat maintain ecological properties, whereasefforts at regional-to-global scale experimenta-tion are inappropriate because of the potentialirreversibility of negative outcomes (see Chap-ter 14).

The ideas of adaptive learning can becomeoperational by integrating the activities of mon-itoring, research, and policy making. Ongoingsocial–ecological monitoring provides the basisfor observing current conditions and the pat-terns and trajectories of change. Monitoringcan assess if there is compliance with policy,if management actions are having the desiredeffect, and if models and their assumptionsare valid (Busch and Trexler 2002). Multiplesources of information (e.g., local knowledge,field-based monitoring, remotely sensed data)enhance the monitoring process by providingmultiscaled understanding of phenomena aswell as interdisciplinary and cross-cultural per-spectives. Programs of citizen science, for exam-ple, can contribute to this process, as do natu-ral science monitoring programs (e.g., Kofinaset al. 2002). Indicators in monitoring charac-terize the structural and compositional parts ofthe system and their processes. In many partsof the world “sustainability indicator programs”are being implemented by municipalities totrack change by integrating social–ecologicalconditions. In some cases, benchmarks of per-formance are identified and reviewed peri-odically to assess change and make adjust-ments as needed to meet desired targets.This is a valuable application of single-looplearning.

Linking observations of change with analysisand evaluation can be improved with the devel-opment and use of models. Models are simpli-fied representations of the real world that pro-vide insight into social–ecological processes and

change. Conceptual models, statistical models,simulation models, and spatial models, wheneffectively used, can serve as decision-supporttools that help assess the state of knowledge,assess management actions, direct future mon-itoring and research, and explore the impli-cations of alternative futures. Effective adap-tive management draws on multiple models totest alternative hypotheses about why givenchanges occurred or are likely to occur in thefuture. As with monitoring, modeling can bebased on multiple knowledge systems or theintegration of expert knowledge with empiri-cally based relationships. Models, developed inconjunction with scenario analysis, help to con-ceptualize emergent conditions and support avisioning process (see Chapter 5). The use ofscenarios can be undertaken through (1) back-casting, which identifies societal objectives anduses models and scenarios to explore how theseobjectives and past events led to the current sit-uation or some desired future condition, and(2) forecasting, which projects future condi-tions and their social–ecological consequencesbased on recent trends (Robinson 1996). Mak-ing models complex enough to capture thenature of the problems while simple enough tounderstand the drivers of change is a significantchallenge in their use (Starfield et al. 1990). Par-ticipation in the development and use of modelsas well as model transparency are also impor-tant in making assumptions and key relation-ships clear for all users.

In practice, the adaptive cycle of social–ecological governance occurs imperfectly indifferent phases and within and across dif-ferent scales. Adaptive management becomesadaptive co-management when groups at vari-ous scales of social–ecological systems, includ-ing local communities, participate in processesof social learning. This broader participationpotentially enhances the learning process, aswell as confounds it. An effective adaptiveco-management process typically requires thatsubgroups have sufficient organizational capac-ity to engage in these activities at any onelevel (e.g., regional) and across scales. Bridgingorganizations, such as co-management boardsor councils, can be critical to facilitating theseprocesses.


Although the concept of adaptive manage-ment has been widely accepted by resourcemanagers as a good idea, it is difficult to imple-ment (see Chapter 10; Walters 1997, Lee 1999).Elected officials and the public at large tend toprefer stability since it conveys a sense of well-being. The idea of “policies as experiments”can be unsettling for policy makers who aretrying to satisfy their myopically oriented elec-torate. Ultimately, the success of adaptive co-management depends to a great extent on thedegree of shared management goals and objec-tives among managers and resource users andtheir support for the long-term implementa-tion of the adaptive management program (Lee1999). This suggests that any adaptive manage-ment process must first seek to identify goalsand objectives, at least among those directlyinvolved with decision making (e.g., membersof a bridging organization). The adaptive man-agement of the Everglades, the Grand Canyon,and the Columbia River Basin in the USA area few cases that illustrate the potential successand challenges of successful adaptive manage-ment. Projects undertaken in Wisconsin pro-vide good examples of participatory scenarioanalysis (Peterson et al. 2003). The monitor-ing of ecosystems and development of modelsfor reflexive learning are less useful if under-taken only for brief periods, given the need tounderstand system dynamics and translate pol-icy experiments into new knowledge. Yet fund-ing of long-term programs such as observationsystems can be problematic, and agreement onmanagement objectives can unravel in the heatof new political controversies, changes in fund-ing priorities, or the recognition of new eco-logical threats. Moving from adaptive manage-ment to adaptive co-management representsan additional challenge since it suggests theneed to develop learning processes while con-currently sharing power and responsibility ingovernance between resource users and otherparties involved in management of ecosystems.While these conditions are challenging, they arenot impossible. In practice passive and activeadaptive management represents endpoints ona spectrum, with passive adaptive management(intentionally learning from past experience)

regarded as responsible resource managementby most professionals.

Generating Innovative Outcomes

The success of social learning through adaptiveco-management depends on more than hav-ing the right process and ultimately rests onarriving at solutions to resolve short- and long-term problems. Given the high rate and nov-elty of global-to-local social–ecological changefaced today, those involved in adaptive co-management will increasingly need to scantheir environment to detect change, under-stand it in ways that are meaningful to deci-sion making, and ultimately achieve robust pol-icy responses that sustain the properties ofsocial–ecological systems deemed desirable bysociety. Facing novel conditions, adaptive co-managers will also need to think beyond politi-cal agendas and reflect more carefully on alter-native solutions to novel problems and theprocesses by which novel solutions are read-ily discovered and tested. Thus, the capac-ity of social–ecological governance systems togenerate innovative solutions to novel prob-lems may be one of the most important out-comes of adaptive co-management (Kofinaset al. 2007). While research on innovationin social–ecological governance is somewhatnew, initial findings suggest that innovative out-comes follow from a number of facilitatingconditions, including the productive friction ofcultural diversity, power sharing, and powerpolitics. Box 4.2 presents a set of conditionsthat facilitate innovation through adaptive co-management. The quest for social learning andinnovation through adaptive governance raisesthe question of whether the production of inno-vation through adaptive co-management canbe sustained or whether it waxes and wanesin response to social need and conditions. Itis probably both: Human responsiveness toaddress emergent problems is alerted throughthe waxing and waning of perceived impendingcrisis, while the quality of our responses towardresilience is enhanced by the development ofour capacity to innovate (Kofinas et al. 2007).

100 G.P. Kofinas

Box 4.2 Conditions that Facilitate Innovation in Adaptive co-management

The novel problems facing society from rapiddirectional change require that managers andresource users work together toward innova-tion in problem solving. Several factors con-tribute to the production of innovative out-comes in adaptive co-management processes.Some are listed below (Kofinas et al. 2007).

1. Interdependence of actors’ needs andinterests and sufficient levels of socialcapital (i.e., trust) provide the basis forcreative engagement in an adaptive co-management process.

2. Appropriate levels of social heterogene-ity and productive conflict provide for thecomparison of perspectives and stimula-tion of novel solutions.

3. A culture of openness to new ideas andthe taking of risk promote an environmentin which innovation can be cultivated.

4. Policy leaders and policy entrepreneurspromote and guide innovative problemsolving and gain the acceptance of inno-vative solutions by the greater public.

5. Reflection and innovation do not just hap-pen, but require the allocation of time andcareful facilitation of process.

6. Decision-support tools, such as the use ofscenario analysis with simulation models,can help in anticipating possible futuresand stimulating creative thinking.

7. Prior experience with successful innova-tion builds confidence to experiment andlearn in the future.

Summary

In spite of the significant challenges of rapidchange, there are conceptual frameworks andpractical tools available to resource managers,resource–user communities, NGOs, and othersfor building resilience and sustaining the flowof ecosystem services to society. This chapterexplored the critical components with whichsocial–ecological governance is implementedand assessed, with a focus on the social mech-anisms by which institutions are linked withecosystems and social learning.

Institutions—the rules of the game—function as incentive structures in humandecision making, shaping the choices of soci-ety and affecting individual and collectiveresponsiveness to change. Institutions alsoconvey cultural identity, shape patterns ofcommunication, and define power relationsthat affect the distribution of costs and benefitsthat follow from policy choices. Institutions canboth constrain and/or enhance the capacity ofgroups to respond to change and innovate foradaptation. Because institutions are artifactsof human invention and creativity, their use

and manipulation provide a significant oppor-tunity for society to increase its sensitivitiesto change and be proactive in anticipation ofsurprise. Institutions, human organizations,and social networks are components of highlydynamic and complex adaptive systems. Therole of human agency and the nonlinearnature of change in these systems add to theunpredictability that groups face in ecosystemstewardship. Because of the heterogeneityand complexity of social–ecological systems,there are no “one size fits all” panaceas forresolving the problems associated with social–ecological governance. Developing institutionalarrangements that are well suited for particularsituations require careful attention to context,the underlying social dilemmas that affectchoice arenas, and the institutional fit andinterplay of social–ecological systems acrossmultiple scales.

The ongoing flow of ecosystem services tosociety depends on institutions that are suffi-ciently robust, that are highly adaptive, and thatlink local communities to other scales. Adap-tive co-management has been suggested as anapproach well suited to address the problems


of rapid change. Adaptive co-management hasthe potential to shift human conflict towardincreased cooperation, facilitate social learn-ing through experience and reflection, andserve as a feedback that shapes the trajecto-ries of social–ecological change. Adaptive co-management is not simply a set of rules ora management structure, but is a process bywhich groups draw on a systems perspectiveto think holistically and learn about social–ecological interactions and gain better insightfrom a diversity of perspectives. When effective,the outcomes of adaptive co-management stim-ulate innovation in problem solving. Realizingthe ideals of adaptive co-management requiresan ongoing reevaluation of conditions, assump-tions, and human values associated with man-agement decisions. It also requires a processof continuous reflection on the performance ofinstitutions to improve and in some cases trans-form them to better suit emergent conditions.

Review Questions

1. How can adaptive co-managementapproaches to social–ecological governanceaddress the shortcomings of steady-stateresource management?

2. How do economic incentives created byinstitutions for management of a com-mons affect the sustainability of importantresources and the social and political systemsthat are part of human communities?

3. What are some of the potential strengths andlimitations of local communities having fullauthority over resource management?

4. How can the misfits in institutions andsocial–ecological systems be avoided in con-ditions of rapid change?

5. Given the multiple ways to achieve cross-scale linkages of institutions in social–ecological governance, what are the condi-tions in which each approach is likely to bemost effective?

6. Provide an example of how the practice ofadaptive co-management, including the useof community-based monitoring, simulationmodeling, and scenario analysis, could beused as part of a broad-scale social learningprocess.

Additional Readings

Armitage, D., F. Berkes, and N. Doubleday, editors.2007. Adaptive Co-Management: Collaboration,Learning, and Multi-Level Governance. Univer-sity of British Columbia Press, Vancouver.

Berkes, F. 2008. Sacred Ecology: Traditional Ecolog-ical Knowledge and Resource Management. 2ndEdition. Taylor and Francis, Philadelphia.

Dietz T., E. Ostrom, and P.C. Stern. 2003. The strug-gle to govern the commons. Science 302:1907–1912.

NRC (National Research Council). 2002. The Dramaof the Commons. Committee on the HumanDimensions of Global Change. E. Ostrom, T.Dietz, N. Dolsak, P.C. Stern, S. Stovich, et al.,editors. National Academy Press, Washington.

Olsson, P., C. Folke, and F. Berkes. 2004. Adaptiveco-management for building resilience in social-ecological systems. Environmental Management34:75–90.

Wilson, D.C., J.R. Nielsen, and P. Degnbol, editors.2003. The Fisheries Co-Management Experience.Kluwer Academic Publishers, Dordrecht.

Young, O.R. 2002. The Institutional Dimensions ofEnvironmental Change: Fit, Interplay, and Scale.MIT Press, Cambridge, MA.

5Transformations in EcosystemStewardship

Carl Folke, F. Stuart Chapin, III, and Per Olsson

Introduction

Changes in governance are needed to dealwith rapid directional change, adapt to it,shape it, and create opportunities for pos-itive transformations of social–ecologicalsystems. Throughout this book we stress thathuman societies and globally interconnectedeconomies are parts of the dynamics of thebiosphere, embedded in its processes, andultimately dependent on the capacity of theenvironment to sustain societal developmentwith essential ecosystem services (Odum 1989,MEA 2005d). This implies that resource man-agement is not just about harvesting resourcesor conserving species but concerns steward-ship of the very foundation of a prosperoussocial and economic development, particularlyunder conditions of rapid and directionalsocial–ecological change (Table 5.1). We firstdiscussed the integration of the ecological (see

C. Folke (�)Stockholm Resilience Centre, Stockholm University,SE-106 91 Stockholm, Sweden; Beijer Institute ofEcological Economics, Royal Swedish Academy ofSciences, PO Box 50005, SE 104 05 Stockholm, Swedene-mail: [email protected]

Chapter 2) and social (see Chapters 3 and 4)aspects of ecosystem stewardship in relationto directional change and resilience in a glob-ally interconnected world (see Chapter 1),emphasizing processes that reduce the likeli-hood of passive degradation that might leadto socially undesirable regime shifts. In thischapter we identify ways to enhance the like-lihood of constructive transformative changetoward stewardship of dynamic landscapesand seascapes and the ecosystem services thatthey generate. Rapid and directional changesprovide major challenges but also opportunitiesfor innovation and prosperous development.Such development requires systems of gov-ernance of social–ecological dynamics thatmaintain and enhance adaptive capacity forsocietal progress, while sustaining ecologicallife-support systems.

An Integrated Social–EcologicalPerspective

Over decades, segregated approaches havedominated policy and the structure of gov-ernmental departments, agencies, and decision-making bodies, with little communicationbetween sectors. This was true in science


104 C. Folke et al.

Table 5.1. Features of shift in perspective from command-and-control to complex systems.Adapted from Folke (2003).

From command-and-control To complex systems

Assume stability, control change Accept change, manage for resiliencePredictability, optimal control Uncertainty, risk spreading, insuranceManaging resources for sustained yield Managing diversity for coping with changeTechnological change solve resource issues Adaptive co-management builds resilienceSociety and nature separated Social–ecological coevolution

as well, with reward systems stimulatingwithin-discipline knowledge generation andlimited collaboration across disciplines (Wilson1998). Resource and environmental manage-ment have been subject to similar divi-sions. Only recently have managers appreci-ated the significance of a broader systems per-spective to deal with rapid and directionalchange. The integration of the human andthe environmental dimensions for ecosystemstewardship is still in its infancy, so analysesof social–ecological systems are not as welldeveloped as those of social or ecological sys-tems alone (Costanza 1991, Ludwig et al. 2001,Westley et al. 2002). A focus that is restrictedto the social dimension of ecosystem stew-ardship without understanding how it is cou-pled to ecosystem dynamics will not be suf-ficient for sustainable outcomes. For exam-ple, the development of fishing cooperativesin Belize was considered a social success bymanagers. However, the local mobilization ofcoastal fishers into socially desirable and eco-nomically effective fishing cooperatives becamea magnet of fishing efforts to capture eco-nomic rent (the income gained relative tothe minimum income necessary to make fish-ing economically viable) and resulted in ashort-term resource-exploitation boom of lob-ster and conch, causing large-scale resource-useproblems and increased vulnerability (Huitric2005).

Similarly, focusing only on the ecologicalaspects as a basis for decision making for sus-tainability leads to conclusions that are toonarrow. Ecosystems can pass a threshold andshift from one state to another, often triggeredby human actions (Folke et al. 2004). Whena lake shifts from a clearwater state attractivefor fishing and recreation to a state of unde-

sired algal blooms and muddy waters, it maylook like an ecologically irreversible transition.However, if there is sufficient adaptive capac-ity in the social system to respond to the shiftand foster social actions that return the laketo a clearwater state, the social–ecological sys-tem is still resilient to such change (Carpenterand Brock 2004, Bodin and Norberg 2005; seeChapter 9).

Hence, in a social–ecological system withhigh adaptive capacity, the actors have the abil-ity to renew and reorganize the system withindesired states in response to changing condi-tions and disturbance events. However, thereare also situations where it would be desirableto move away from the current conditions andtransform the social–ecological system into anew configuration. Transformation is the fun-damental alteration of the nature of a systemonce the current ecological, social, or economicconditions become untenable or are undesir-able (Walker et al. 2004, Nelson et al. 2007;see Chapter 1). The capacity to learn, adaptto and shape change, and even transform arecentral aspects of social–ecological resilienceand require that learning about resource andecosystem dynamics be built into managementpractices (Berkes and Folke 1998b) and sup-ported by flexible governance systems (Folkeet al. 2005). Transformative learning is learningthat reconceptualizes the system through pro-cesses of reflection and engagement. It relatesto triple-loop learning (Fig. 5.1), which directsattention to redefining the norms and protocolsupon which single-loop and double-loop learn-ing (see Chapter 4) are framed and governed.This draws together human agency and indi-vidual and collective learning with processes ofchange, uncertainty, and surprise (Keen et al.2005, Armitage et al. 2007).

5 Transformations in Ecosystem Stewardship 105

Sing

le-lo

op learning

Do

ub

le-l

oo

p le

arni

ng

Tri

ple-

loop

lea

rnin

g

Monitoroutcome

Evaluateoutcome

Maintain or adjustdesign to meetthe same target

Implementaction

Consider whether to alter

policies

Consider whetherto alter rules fordecision-making

Re-evaluateassumptions,

key relationships, and mental

models

Re-evaluate anddecide on

governance

Re-evaluate and decide on policies

Figure 5.1. Triple-looplearning involves the samereevaluation ofassumptions and modelsas double-loop learning(see Fig. 4.2) but considerswhether to alter norms,institutions, andparadigms in ways thatwould require afundamental change ingovernance (e.g., a shiftfrom an agriculturalsystems focused onsupporting farmers to atourist-based economyrequiring a broader, moreinclusive form ofgovernance).

Dealing with Uncertaintyand Surprise

Recognizing and accepting the uncertaintyof future conditions is the primary motiva-tion for incorporating resilience thinking intoecosystem stewardship. We are nowhere closeto a predictive understanding of the complexinteractions and feedbacks that govern trajec-tories of change in social–ecological systemsnor able to anticipate the future human actionsthat will modify these trajectories. Uncertaintyis therefore a central unavoidable conditionfor ecosystem stewardship. There are severalsources of uncertainty, only some of which canbe readily reduced (Carpenter et al. 2006a).Both scientific research and the observationsand experience of managers and other peo-ple provide data that inform our understand-ing. However, there are many uncertaintiesregarding the validity of any dataset and itsrepresentativeness of the real world (Kinzig

et al. 2003). Models, both quantitative com-puter models and conceptual models of howthe world works, also have many uncertain-ties in assumptions and structure. Models aremost useful when based on observations andother data, but there are always important pro-cesses for which data are unavailable and datathat do not fit our current understanding. Sur-rounding these uncertainties in data and modelsare uncertainties in other factors that we knowto be important but for which we have nei-ther data nor models—the “known unknowns”(Fig. 5.2). There are also “unknown unknowns”that we cannot anticipate—the surprises thatinevitably occur (Carpenter et al. 2006a).

There are several types of surprises(Gunderson 2003). Local surprises occurlocally. They may be created by a narrowbreadth of experience with a particular system,either temporally or spatially. Local surpriseshave a statistical distribution, and peoplerespond to these surprises by forming subjec-tive probabilities that are updated when new

106 C. Folke et al.

All possible futures

Data

Models

Uncertainties

Figure 5.2. The full set of possible futures of social–ecological systems is only partially represented inavailable data and models. Together, the data andmodels allow us to project some uncertainties (know-able unknowns). The probability that any model

projection of future conditions will actually occurdepends on the full set of all possible futures, most ofwhich are unknown. Redrawn from Carpenter et al.(2006a).

information becomes available. Based on theseestimates, there is a wide range of adaptationsto risk that are economically rational to indi-viduals, including risk-reducing strategies andrisk-spreading or risk-pooling among indepen-dent individuals. Local surprises are manage-able by individuals and groups of individuals.Their detection requires a comprehensivesystems perspective (e.g., an ecosystem ratherthan a single-species approach). Adaptation-to-risk strategies fail when surprises are notlocal.

Cross-scale surprises occur when there arecross-scale interactions, such as when local vari-ables coalesce to generate an unanticipatedregional or global pattern, or when a pro-cess exhibits contagion (as with fire, insectoutbreak, and disease). Cross-scale surprisesoften occur as the unintended consequencesof the independent actions of many individualagents who are managing at different scales.

Although individual responses are generallyineffective, individuals acting in concert canaddress these surprises, if appropriate cross-scale institutions are available or are readilyformed (see Chapter 4).

True-novelty surprises constitute never-before-experienced phenomena for which strictpreadaptation is impossible. However, systemsthat have developed mechanisms for reorgani-zation, learning, and renewal following suddenchange may be able to cope effectively withtrue-novelty surprises. These are the social–ecological features that nurture resilience todeal with unexpected change.

Directional change in the context of globaland climatic change creates a situation ofincreased likelihood of unknowable surprises.It is within these sources of uncertainty andsurprise that ecosystem stewardship must func-tion and where a resilience approach becomesessential.


Preventing Social–EcologicalCollapse of Degrading Systems

The Interplay Between Gradualand Abrupt Change

Theories, models, and policies used in resourcemanagement have historically been developedfor situations of gradual or incremental changewith implicit assumptions of linear dynamics.These approaches generally disregard interac-tions that extend beyond the temporal and spa-tial scales of management focus. The resilienceapproach to ecosystem stewardship and theadaptive-cycle framework outlined in Chapters1 and 4 indicate that there are times whenchange is incremental and largely predictableand other times when change is abrupt, disorga-nizing, or turbulent with many surprises. It alsodraws attention to ways in which such social–ecological dynamics interact across temporaland spatial scales (the concept of panarchies,see Chapter 1; Gunderson and Holling 2002).This dynamic interaction challenges manage-

ment to learn to live with uncertainty, be adap-tive, prepare for change, and build it intoecosystem stewardship strategies. Periods oflarge abrupt changes in social and ecologicaldrivers, including climatic change and economicglobalization, are occurring more frequently(Steffen et al. 2004; see Chapter 14), increas-ing the likelihood of abrupt social–ecologicalchange. In the absence of resilience-based stew-ardship, these changes are quite likely to trig-ger shifts from one state to another that maybe socially and ecologically less desirable. Afocus on resilience in social–ecological systemsis needed to deal with the challenging newglobal situation of rapid and directional social–ecological change.

Behaviors that reduce the adaptive capac-ity to deal with interactions between gradualand abrupt change may push systems towarda threshold that precipitates regime shifts orcritical transitions. The existence of thresholdsbetween different regimes or domains of attrac-tion has been described for several ecolog-ical systems (Scheffer et al. 2001; Fig. 5.3).

Ecosystem state

Conditions

Perturbation

F1

F2

Figure 5.3.

108 C. Folke et al.

Regimeshift

Clear-water lakes

Phoshorus accumu-lation in agriculturalsoil and lake mud

1 2 3 4

Coral-dominatedreefs

Grassland

Grassland

Kelp forests

Pine forest

Overfishing, coastaleutrophication

Fire prevention

Seagrass beds

Tropical lake with sub-merged vegetation

Hunting of herbivores

Elimination of top predators

Microclimate and soilchanges, loss of pineregeneration

Removal of grazers,lack of hurricanes,salinity moderationspatial homogenization

Nutrient accumulationduring dry spells

Flooding, warming,overfishing,species invasions

Disease, bleaching,hurricane

Good rains, continu-ous heavy grazing

Disease

Thermal event,storm, disease

Decreased firefrequency, increasedfire intensity

Thermal event

Nutrient releasewith water table rise

Turbid-waterlakes

Algae-dominatedreefs

Shrub-bushland

Woodland

Sea urchindominance

Oak forest

Phytoplanktonblooms

Floating-plantdominance

,

Initialstate

Causes ofchange

Alternatestate

Triggers ofchange

Figure 5.3. Regime shifts in ecosystems: Previouspage: Cup-and-ball model illustrating the shift fromone ecosystem regime to another. The bottomplane shows the hysteresis-curve, which highlightsthe nonlinear relation of moving from one regimeto another. Modified from Scheffer et al. (2001).

Above: Alternate states in different ecosystems (1, 4)and some causes (2) and triggers (3) behind lossof resilience and regime shifts. For more exam-ples, see Thresholds Database on the Web sitewww.resalliance.org. Modified from Folke et al.(2004).

Thresholds may also occur in resilient ecolog-ical systems, but fostering resilience reducesthe likelihood of this occurring. Experiencesuggests that critical ecosystem transitions areoccurring increasingly often as a consequenceof human actions and seem to be more commonin human-dominated landscapes and seascapes(Folke et al. 2004). Human actions that are mostlikely to cause loss of resilience of ecosystemsinclude

• introduction or removal of functional groupsof species that reduce effect and response

diversity, including loss of entire trophic lev-els (top-down effects),

• impact on ecosystem resources and toxins viaemissions of waste and pollutants (bottom-up effects), and

• alteration of the magnitude, frequency, andduration of disturbances to which the biotais adapted.

Similarly there are features of social systemsthat impact social–ecological resilience like

• degradation of the components of humanwell-being, including public education andincome levels,


• erosion of social capital and adaptive capac-ity, for example, through corruption, rentseeking, and loss of new opportunities for thefuture, and

• dysfunctional institutions, causing, for exam-ple, weak or insecure property rights andhigh inequality in power and wealth.

The loss of resilience through the com-bined and synergistic effects of such factorsmakes social–ecological systems increasinglyvulnerable to changes that previously could beabsorbed. As a consequence they are morelikely to shift from desired to less desired states.In some cases, these regime shifts or criticaltransitions may be irreversible or too costly toreverse. Irreversibility is a reflection of changesin critical slow variables (e.g., biogeochemical,hydrological, climatic, constitutional, cultural)and loss of diversity and of social–ecologicalinteractions that support renewal and reorga-nization into desired states. A challenge forresilience-based stewardship is to address suchchanges in a more integrated fashion than isusually done today.

In light of the risk for irreversible shiftsand their implication for well-being, the self-repairing capacity of social–ecological systemsin the face of directional change should not betaken for granted. It must be nurtured. Criticalfeatures have been identified (Folke et al. 2003;see Chapter 1) for fostering adaptive capacityand resilience in social–ecological systems:

• learning to live with change and uncertainty,• cultivating diversity for reorganization and

renewal,• combining different types of knowledge sys-

tems for learning,• creating opportunity for self-organization

toward social–ecological sustainability, and• experimenting and innovating to test under-

standing and implement solutions.

Adaptive management (see Chapter 4) andadaptive governance of resilience (discussedlater in this chapter) will be required, for exam-ple, in the context of scenarios of plausiblefutures to prevent social–ecological degrada-tion or to transform systems into more desiredstates.

Multiple Regimes

Shifts between social–ecological states canoccur because of external perturbations, like aclimatic event or a political crisis. They can alsooccur because of complex cross-scale dynamicswithin the social–ecological system, with myr-iad localized interactions among smaller enti-ties serving as a source of adaptation and nov-elty, and larger-scale emergent constructs suchas political systems or climatic conditions serv-ing to frame the behavior and dynamics atsmaller scales.

Critical transitions or regime shifts have beendescribed primarily for either ecological sys-tems (e.g., Folke et al. 2004) or social and eco-nomic ones (e.g., Repetto 2006). However, shiftsalso occur due to interactions and feedbacksbetween social and ecological processes that aretriggered by external events or internal dynam-ics that cause a loss of resilience (Gundersonand Holling 2002, Kinzig et al. 2006). For exam-ple, resource-management institutions that per-form in a socially and economically resilientmanner, with well-developed collective actionand economic incentive structures, may in igno-rance degrade the capacity of ecosystems toprovide ecosystem services. Such behavior maycause a transition to a degraded ecosystem statethat in turn feeds back into the social andeconomic systems, causing unpleasant surprisesand social–ecological regime shifts. As a con-sequence, the social–ecological system may fallinto a rigidity or poverty trap (Gunderson andHolling 2002). In rigidity traps people and insti-tutions try to resist change and persist withtheir current management and governance sys-tem despite a clear recognition that change isessential. The tendency to lock into such a pat-tern comes at the cost of the capacity to adjustto new situations. This behavior constrains theability of people to respond to new problemsand opportunities. A poverty trap, a social–ecological system with persistent poverty, alsoreflects a loss of options to develop or deal withchange (Bowles et al. 2006). It is locked intopersistent degraded conditions and would needexternal support to get out of it. However, sim-ply providing money, technical expertise, infras-tructure, and public education is seldom suffi-

110 C. Folke et al.

cient to move out of a poverty trap. Escape fromrigidity and poverty traps depends on the capac-ity of people within the social–ecological sys-tem to create continuously new opportunities.New opportunities, in turn, are strongly linkedto the existence of sources of resilience andadaptive capacity (see Chapter 3) to help peoplefind ways to move out of traps. Hence, the risksor possibilities of sudden shifts between social–ecological states have profound implicationsfor stewardship of essential ecosystem servicesin a world of rapid and directional change.

Thresholds and Cascading Changes

The movement of a social–ecological systemacross a threshold to a new regime alterssocial–ecological interactions, triggering newsets of feedbacks with cascading social andecological consequences. Once a system hasexceeded a threshold, many changes occur thatcan only be understood or predicted in thelocal context. However, certain repeatable pat-terns emerge that provide a basis for design-ing appropriate management strategies (seeChapters 3 and 4). New interactions frequentlybecome important, and social–ecological pro-cesses become sensitive to different slow vari-ables. These changes require a reassessment ofmanagement goals and priorities and flexibil-ity to seek new solutions through innovation ininstitutions and approaches (double-loop learn-ing; see Fig. 4.3). In the absence of theseresilience-based strategies, the system may con-tinue to degrade. In Western Australia, forexample, extensive areas of native heath veg-etation were converted to wheatlands (Kinziget al. 2006; see Chapter 8), leading to a radi-cally different system, both socially and ecologi-cally, than the shrub savanna occupied by Abo-rigines. Replacement of deep-rooted heath byshallow-rooted cereals altered the hydrologiccycle by reducing transpiration rate, causing thesaline water table to rise close to the surface,creating saline soils that reduced productivity.Declines in production, in turn, caused peopleto leave the region. The combination of declin-ing population and productivity (33% of theland too saline to farm), coupled with changesin national farm policy, led to amalgamationof farms into larger units that could remain

profitable due to economies of scale (a socialtransformation). As population declined, manytowns could no longer support a service sector,causing still more people to leave and towns tobe abandoned. This made it difficult for peo-ple who stopped farming to find other jobs,compounding the levels of social stress. Thisexample of cascading consequences of exceed-ing a social–ecological threshold illustrates sev-eral points (Kinzig et al. 2006).

• New sets of interactions come into play whena social–ecological system crosses a thresh-old, leading to a cascading set of social andecological consequences.

• The reorganization of the system after cross-ing the threshold increases the vulnerabil-ity to further degradation and thresholdchanges.

• Each successive transformation is moreresilient, in the sense that it would be moredifficult to return it to its original state or tosome other more desirable state.

The cascading changes that occur when athreshold is exceeded are typical of the behav-ior of complex adaptive systems (Levin 1999;see Chapter 1), in which any change in the sys-tem triggers additional changes in its funda-mental properties and feedback structure. Newfeedbacks then develop that stabilize the sys-tem in a new state, making it progressively moredifficult to return to the original state.

Directional changes in external drivers suchas climate can trigger similar regime shiftsand passive degradation. Changing sea sur-face temperatures in the Atlantic Ocean, forexample, reduced rainfall in the Sahel regionof sub-Saharan Africa. This reduced precipi-tation, causing declines in vegetation, whichincreased regional albedo, weakening the mon-soon and stabilizing the drought conditions(Foley et al. 2003a; see Chapter 2). Thedeclines in vegetation caused people to con-centrate their herds on the remaining veg-etation, causing further increases in albedoand strengthening the drought. These drought-induced feedbacks probably contributed to thelong (30-year) duration of drought. Fortu-nately, large-scale circulation changes eventu-ally ended the drought. Other potential inter-ventions might have included strategies to


reduce albedo by increasing vegetation cover(e.g., through planting of forests that tapdeep groundwater or introduction of drought-resistant crops that might provide enoughfood that grazing pressure by cattle could bereduced). It is quite likely that current ratesof climate change, if they continue, will triggerregime shifts and cascading social–ecologicalchanges in many parts of the planet. If theseare extensive, they could exceed a global tip-ping point—i.e., in this context a threshold fortransformational change to a new system, lead-ing to novel global changes in climate, economy,and politics (Plate 3).

Endogenous changes in social–ecologicalsystems can also trigger regime shifts with cas-cading effects. Several ancient societies suchas the Roman Empire and Mayan Civilizationappear to have collapsed at least in part becauseof unsustainable practices that caused environ-mental degradation and loss of the produc-tive potential of the ecosystems on which theydepended (Janssen et al. 2003, Diamond 2005).Similarly, collapse of the Soviet Union in 1990was a regime shift with many social, economic,and political consequences.

Institutional Misfits with Ecosystems:A Frequent Cause of Regime Shifts

It is no longer rational to manage systems sothey will remain the same as in the recentpast, which has traditionally been the referencepoint for managers and conservationists (seeChapter 1). We must instead adopt a moreflexible approach to managing resources—management to sustain and enhance the func-tional properties of integrated social–ecologicalsystems that are important to society underconditions where the system itself is constantlychanging. Sustaining and enhancing such prop-erties and recombining them in new waysare the essence of sustaining social–ecologicaldevelopment and the very core of the resilienceapproach to ecosystem stewardship (Folke2006).

The problem of fit between institutions andecosystem dynamics in social–ecological sys-tems (see Chapter 4) is one of the most fre-quent causes of undesirable regime shifts. Thisinterplay takes place across temporal and spa-

tial scales and institutional and organizationallevels in a dynamic manner (see Table 4.4).

Temporal Misfits in Social–EcologicalSystems

The implementation of conventional resourcemanagement tends to lead to governance sys-tems that invest in controlling a few selectedecosystem processes, often successfully in theshort term, in order to fulfill immediate eco-nomic or social goals, such as the produc-tion of wood by forests. But this success tendsto turn into a longer term failure throughthe erosion of social–ecological resilience andkey functions (Holling and Meffe 1996; seeChapter 2). “Science-of-the-parts” perspectives(see Chapter 4) have contributed to resource-management systems that focus on producing anarrow set of resources, often in vast monocul-tures like tree plantations or resource-intensivesystems like chicken farms, or salmon aquacul-ture operations. The widespread approach of“optimal production of single resources” under-lying these production systems (Table 5.1) maybe successful during periods of stable environ-mental and economic conditions (Holling et al.1998). In situations of uncertainty and surprise,however, they become vulnerable because theylack backup systems and sources of reorgani-zation and renewal. These systems are there-fore seriously challenged by rapid directionalchange.

This challenge may also hold true formore diverse management systems with seem-ingly flexible institutional and organizationalarrangements. The Maine lobster fishery, forexample, is a sophisticated collective actionand multilevel governance system that has sus-tained and regulated the economically valu-able lobster fisheries. It has been consideredone of the classic cases of successful people-oriented local management of common-poolresources. However, when the linkage of thesocial domain to the production of lobsters istaken into account, the Maine fishery seems tohave followed the historical pattern of fishing-down food webs (Jackson et al. 2001). Deple-tion of the cod fishery opened up space forthe expansion of species lower down in foodwebs, like lobsters. Currently the coastline is

112 C. Folke et al.

massively dominated by lobsters, like a coastalmonoculture, with the bulk of the lobster pop-ulation artificially fed with herring supplied asbait in lobster pots. The lobster has a high mar-ket price and sustains the social organizationand the fishery. However, such simplificationof marine systems through removal of func-tional diversity (see Chapter 2) has created ahighly vulnerable social–ecological system wait-ing for an accident, like a lobster disease, tohappen. If such a “surprise” occurs, the lob-ster population might be decimated over hugeareas, perhaps triggering a shift into a very dif-ferent social–ecological system in which coastalwaters no longer provide a viable livelihood forlocal fishers (R. Steneck and T. Hughes, pers.comm.). Because lobster fishing is central toregional identity, the potential loss of lobsterfishing could have severe social as well as eco-nomic impacts.

Similar mismatches between short-term suc-cess (a governance system that delivers short-term economic and social benefits) and long-term failure of resource management (lack ofan ecosystem approach and ecosystem stew-ardship leading to erosion of resilience) haveoccurred in forests and lake fisheries (Regierand Baskerville 1986), other coastal andregional fisheries (Finlayson and McCay 1998),crop production (Allison and Hobbs 2004),and a range of other situations (Gundersonet al. 1995; see Chapters 6–14). The ques-tion remains to what extent such patterns ofresource and ecosystem exploitation can fos-ter adaptive capacity to either prevent passivedegradation or actively transform landscapesand seascapes to more beneficial states throughsustainable ecosystem stewardship.

Spatial Interdependenceof Social–Ecological Systems

Human societies are now globally intercon-nected, through technology, financial markets,and systems of governance with decisions inone place influencing people elsewhere. How-ever, the interplay between globally intercon-nected societies and the planet’s ecological life-support systems is not yet fully appreciated.

The seriousness and challenges of the climateissue have begun to mentally reconnect peopleto their dependence on the functioning of thebiosphere. The common policy response to cli-mate change has been to focus on mitigationof greenhouse gases through technical meansor on social and economic adaptation to cli-matic change. The urgent policy response that isbeginning to emerge as a critical complement isthe stewardship of the social–ecological capac-ity to sustain society with ecosystem servicesand its links to adaptation, resilience, and vul-nerability in the face of unprecedented direc-tional changes (see Chapters 2, 3, and 14). Itrequires systems of governance that are adaptiveandthatallowforecosystem-basedmanagementof landscapes and seascapes (see Chapter 4).

Governance to address global issues mustbe aware of, account for, and relate to thedynamic interactions of people and ecosystemsacross local-to-global scales. For example, theefforts by large chains of food stores in devel-oped regions and urban centers to reduce tem-poral fluctuations in the supply of fish, fruits,and other commodities has increased both theextraction and the exploitation of resources inremote areas, creating spatial dependence onother nations’ ecosystems (Folke et al. 1997,Deutsch et al. 2007). People in the cities ofSweden, for example, depend on ecosystem ser-vices over an estimated area about 1000 timeslarger than the actual area of the cities, corre-sponding to about 2–2.5 ha of ecosystem perperson (Jansson et al. 1999; see Chapter 13).In this broader context it becomes clear thatpatterns of production, consumption, and well-being depend not only on locally sustainablepractices but also on managing and enhancingthe capacity of ecosystems throughout the worldto support societal development. For example,salmon and shrimp produced in aquacultureoperations in temperate and tropical regions,respectively, are traded on global markets andconsumed in developed regions and urban cen-ters (Lebel et al. 2002). The feed input toproduce these aquaculture commodities comesfrom coastal ecosystems all across the planet.Shrimp produced in ponds in Thailand, forexample, use meal from fish caught in the NorthSea. Similar globally interconnected patterns,


made possible by fossil-fuel-based technologyand supported by information technology, existin agricultural food and energy production.Demand in one corner of the world shapeslandscapes and seascapes in other parts of theplanet (see Chapters 12 and 14).

Stewardship of ecosystems is continuouslysubject to global drivers (Lambin et al. 2001). Inthis context it becomes important to address theunderlying social causes challenging ecosystemcapacity to generate services. They include thestructure of property rights; macro-economic,trade, and other governmental policies; eco-nomic and legal incentives; the behavior offinancial markets; causes behind populationpressure; transfer of knowledge and technol-ogy; misguided development aid; patterns ofproduction and consumption; power relationsin society; level of democracy; and worldview,lifestyle, religion, ethics, and values.

In the UK in the 1980s, for example, tax con-cessions on afforestation were increased but notfor the purchase of land. Investors thereforeminimized land purchases and located forestplantations on economically low-valued land,such as wetlands, heath, and moorland, therebydepleting ”unpriced” wildlife values (Wibe andJones 1992). In the Brazilian Amazon, onecould only acquire a title to land by living onand ”using” the land, with logging as a proofof the land being occupied and used. A farmcontaining “unused” forests was taxed at higherrates than one containing pastures or cropland.The real interest rate on loans for agriculturewas lower than for other land uses, and agricul-tural income was almost exempt from taxation(Binswanger 1990). Hence, policies and activi-ties that, at first glance, seem to be unrelated tothe capacity of ecosystems to generate servicesmay indirectly counteract ecosystem steward-ships. Such policies serve as subsidies from soci-ety to use living resources and ecosystem capac-ity in unsustainable manners. They need to beredirected into incentives for more sustainableresource use.

Global market drivers sometimes operate soquickly that local governance responses do nothave time to respond or adapt, as illustratedby the “roving bandits” phenomena in coastalfisheries (Berkes et al. 2006), where exploiters

linked to global markets rapidly move from onefish stock to another over wider and wider spa-tial scales. This implies that sustainable adap-tive governance systems for ecosystem stew-ardship need to be prepared to deal in a con-structive manner with sudden external shockslike the rapid development of a new mar-ket demand or sudden shifts in governmentalpolicies.

Fostering DesirableSocial–EcologicalTransformations

Identifying Dysfunctional States

Dysfunctional states occur when society cannotmeet the basic needs of human well-being orwhen environmental, ecological, social, or polit-ical determinants of well-being are degradedto the point that loss of well-being is highlylikely to occur. Some social–ecological sys-tems persist in dysfunctional states, such as dic-tatorships, persistent civil strife, and extremepoverty, for extended periods of time. Otherdysfunctional states result from natural disas-ters such as floods, hurricanes, and tsunamis orfrom social disasters such as wars. When suchsystems experience shocks and surprises theymay lack the adaptive capacity to reorganize, orthey may reorganize in ways that increase thelikelihood of future shocks. Getting out of dys-functional states often requires external insti-tutional, financial, and/or political support, andmany bodies from local nongovernmental orga-nizations (NGOs) to international aid organiza-tions like the Red Cross or economic ones likethe World Bank work actively to support suchtransformations. However, external aid is insuf-ficient. Escape from dysfunctional states alsorequires local development of adaptive capac-ity for innovation.

Systems can also degrade due to gradual lossof ecosystem services and resilience; increaseddemand for ecosystem services because of pop-ulation growth or excessive consumption ofservices; and various social or political trends.As this degradation proceeds, there is often aspectrum of opinion among stakeholders about

114 C. Folke et al.

whether to fix the current system by incremen-tally addressing specific problems or enhanc-ing resilience to deliberately explore transfor-mation to a new social–ecological state. In theGoulburn-Broken Catchment in Australia, theagriculturaldevelopmenttrajectorywasstronglyembedded socially and culturally and econom-ically supported, making it difficult to explorenew ways to manage the land. Local resourcesand institutions were initially focused on main-taining a system that fostered a continual down-ward spiral into a dysfunctional and nonre-silient state. More recently, crisis awareness atthe system level triggered shifts in perceptionand action and transformed whole manage-ment and governance systems toward ecosys-tem stewardship of the social–ecological sys-tem (Walker and Salt 2006). Similar shiftstoward ecosystem stewardship at regional scalesare evident in landscapes of southern Swe-den and the vast Great Barrier Reef seascapeof Australia (Olsson et al. 2006, 2008).

Recognizing Impending Thresholdsfor Degradation

Scenarios of plausible future changes provide astarting point for exploring policy options thatreduce the likelihood of undesirable regimeshifts. Global, national, and local assessmentsoften provide clear evidence of trends in envi-ronmental, ecological, and social conditionsthat are leading in unsustainable directions forsocial–ecological systems at local-to-planetaryscales. The causes of many of these trendsare increasingly well understood, providing abasis for quantitative or conceptual modelsthat project some of these trends into thefuture, assuming that people continue their cur-rent patterns of behavior (“business-as-usual”scenarios). Continuation of current trends infossil-fuel use, for example, will likely cause“dangerous” climatic change within the nextfew decades by altering Earth’s climate systembeyond a tipping point that would have seri-ous ecological and societal consequences andbe difficult to reverse (Stern 2007, IPCC 2007a,b). Similarly, current declines in biodiversityand ecosystem services are degrading liveli-

hoods and well-being of social–ecological sys-tems globally, particularly for underprivilegedsegments of society (MEA 2005a; see Chapters2, 3 and 4).

Scenarios represent plausible futures that arebased on our understanding of past and cur-rent trends. They are not predictions becauseof the considerable uncertainties that surroundfuture trajectories. Scenarios are most usefulfor assessing gradual changes that are con-trolled by processes for which we have bothdata and models. Trends in global climate, forexample, reflect predictable biophysical inter-actions that can be projected with reason-able confidence over decadal time scales. Thesetrends include resource consumption, fossil-fuel emissions, land-use change, and local-vs-global resource dependence. Scenarios can alsobe defined that assume a suite of policies andhuman actions with predictable biophysical,ecological, and social outcomes. These “what-ifgames” allow comparisons of alternative poten-tial future states, depending on policies thatsociety chooses to implement. Two scenarioscommonly accepted by policy makers and thepublic are that (1) there will be no directionalchange in controls over social–ecological pro-cesses (today’s world will remain unchanged) or(2) people will continue their current behavior(business as usual). Scenarios can explore thelogical social–ecological consequences of theseassumptions or alternative policies that mightlead to more desirable outcomes.

The greatest shortcomings of scenarios arethat they do not capture (1) the uncer-tainty associated with processes that are well-understood; (2) the effects of processes that aremissing from assumptions and models; (3) manyof the complexities of social–ecological inter-actions and feedbacks; and (4) the unknow-able surprises that are an increasingly com-mon property of social–ecological dynamics.Given these severe shortcomings in the capac-ity of scenarios to predict the future, whywould anyone want to use them? Clearly, sce-narios should be treated as plausible futuresrather than predictions. Scenarios are most use-ful when used comparatively to explore the log-ical consequence of differences in assumptionsabout how the world works or policy options


that might differ in their social–ecological con-sequences.

World leaders in industry, government,and the environment disagree about howbest to achieve social–ecological sustainabil-ity. The Millennium Ecosystem Assessment(MEA) sought to explore the consequencesof this spectrum of world opinion by describ-ing four general scenarios of policy strategiesintended to enhance social-ecological sustain-ability (Bennett et al. 2003, MEA 2005c).These scenarios differed in the extent towhich policies were global or regional in theirdesign/implementation and whether ecosys-tem management was reactive (respondingto ecosystem degradation after it occurred)or proactive (deliberately seeking to manageecosystem services in sustainable ways (Corket al. 2006; Table 5.2). In Global Orchestration,there is global economic liberalization withstrong policies to reduce poverty and inequal-ity and substantial investment in public goodssuch as education. In Order-from-Strength,economies become more regionalized, andnations emphasize their individual security.

Adapting Mosaic also has more regionalizedeconomies, but there is emphasis on multi-scale, cross–sectoral efforts to sustain ecosys-tem services. In TechnoGarden, the econ-omy is globalized, with substantial invest-ments in sound environmental technology, engi-neered ecosystems, and market-based solu-tions to environmental problems (MEA 2005c,Carpenter et al. 2006a). A combination of quan-titative and qualitative modeling suggested thatthese alternative policy options would lead toquite different ecological and social outcomes.In each of them there are tradeoffs amongecosystem services and among social benefits.The most encouraging result was that all sce-narios except the Order from Strength wouldreduce the current net degradation of ecosys-tem services and improve human well-being,relative to today’s uncoordinated spectrum ofglobal policies. Nonetheless, these net improve-ments result from quite different patterns oftradeoffs and social equity. Some of the keylessons from these scenarios were (Cork et al.2006):

Table 5.2. Defining characteristics of four scenarios1.

Global orchestration Order from strength Adapting mosaic Technogarden

DominantApproach forSustainability

Sustainable development,economic growth,public goods

Reserves, parks,national-levelpolicies,conservation

Local-regionalco-management,common-propertyinstitutions

Green-technology,ecoefficiency, tradableecological propertyrights

EconomicApproach

Fair Trade (reduction oftariff boundaries), withenhancement of globalpublic goods

Regional trade blocs,mercantilism

Integration of localrules regulates trade;local nonmarketrights

Global reduction of tariffboundaries, fairly freemovement of goods,capital, and people,global markets inecological property

Social PolicyFoci

Improve world; globalpublic health; globaleducation

Security andprotection

Local communitieslinked to globalcommunities; localequity important

Technical expertisevalued; followopportunity;competition; openness

Dominant socialorganizations

Transnational companies(Companies thatspread seamlesslyacross many countries):global NGO andmultilateralorganizations

Multinationalcompanies(Companies thatconsist of loosealliances of largelyseparate franchisesin differentcountries)

Local cooperatives,global partnerships,and collaborationsestablished as localgroupings recognizethe need to shareexperiences andsolutions

Transnational,professionalassociations, NGOs

Reprinted from Cork et al. (2006).

116 C. Folke et al.

• No utopian solution is likely to emergebecause of tradeoffs among ecosystem ser-vices and among social benefits.

• Global cooperation to deal with social andenvironmental challenges would lead to bet-ter outcomes than lack of cooperation.

• Proactive environmental policies would leadto lower risks of major environmental prob-lems and loss of well-being than would reac-tive policies.

• Participation by a breadth of stakeholders indesigning the scenarios clarified the variationin visions about how to achieve sustainabil-ity and acceptance of the conclusions of thestudy.

• Comparison of a small number (2–4) scenar-ios allowed a diverse but manageable set ofoptions to be considered.

Creating Thresholdsfor Transformation from UndesirableStates

How can ecosystem stewardship help people,communities, and societies escape rigidity andpoverty traps? In a rigidity trap there is a ten-dency to lock management and governance intotheir existing attitudes or worldviews, makingit difficult to respond to changing conditions.Even if there is a general feeling that somethingneeds to be done, it can be surprisingly diffi-cult to get a group out of such gridlock, and theinvestment in a certain perspective or behaviormay be so strong that it is hard to create incen-tives that are strong enough to change it. Rigid-ity is deeply rooted because it develops as a wayto ensure consistency. In such situations, the“exceptional few” individuals play an importantrole in catalyzing tipping points and shiftingmanagement and governance over a thresholdinto a new direction. Some individuals appearto be able to mobilize groups to remove theinertia and change management behavior andworld views. They may, for example, be particu-larly well connected, have high social capital, beinnovators or early adopters by nature, or havethe charisma to cause emotional contagion. Theabsence of such leaders makes a social group

as a whole rigid and weak when adaptation tochange is required (Scheffer and Westley 2007).

It is more difficult to create tipping points toescape from a poverty trap, because these trapsare generally characterized by very low levelsof social capital and adaptive capacity; initialpoverty is often self-reinforcing; and concen-trated poverty tends to undermine processes ofcommunity organization (Bowles et al. 2006).Even if the group or community can mobilizeinternally and build adaptive capacity to get outof the trap, it may be overwhelmed by broader-scale factors, such as changes in regional andglobal markets, governmental corruption, orlow level of education in a country as whole.There is often a long historical path dependenceof political and economic goals and institutionalstructures that push a social-ecological systeminto a poverty trap (Engerman and Sokoloff2006). The challenges of moving out of povertytraps are huge. Although economic capital andtechnology support are important, they areoften insufficient to help social-ecological sys-tems escape such traps (Bowles et al. 2006).

Moving out of traps requires not just a shiftin the social (including economic) dimensionsbut also active stewardship of ecosystem pro-cesses. A major challenge is to secure, restore,and develop the capacity of ecosystems togenerate ecosystem services because this capac-ity constitutes the very foundation for the socialand economic development needed to escapefrom poverty traps (Enfors and Gordon 2007).Ecological restoration and ecological engineer-ing are subdisciplines of ecology that focus onenhancing the capacity of ecosystems to pro-vide services. These fields tend to emphasize thegrowth and conservation phases of the adap-tive cycle. More recently, research on biolog-ical diversity as sources of resilience is gain-ing momentum (Folke et al. 2004). Biologicaldiversity provides the ingredients for regener-ating an ecosystem within its current state afterdisturbance or the seeds for alternative statesthat might be more viable under new condi-tions (see Chapter 2). Hence, biodiversity playsa central role in the release and renewal phasesof the adaptive cycle. Diverse landscapes andseascapes with resilience have higher capac-ity to regenerate in the face of disturbance


and thereby sustain the supply of ecosystemservices. Management that focuses on usingprotected areas and reserves as reservoirs ofbioidveristy to strengthen resilience is gainingground (e.g., Bengtsson et al. 2003). For exam-ple, to enhance the resilience to climate changeand secure ecosystem services of the Great Bar-rier Reef in Australia, the seascape has beenrezoned into 70 habitats, each of which hasfully protected areas as insurance for ecosystemregeneration after disturbance. A major task ofecosystem stewardship is to identify and man-age the role of functional groups of organisms,their redundancy, and their response diversityin relation to ecosystem services at the land-scape and seascape scale (Walker 1995, Naeem1998, Elmqvist et al. 2003, Nystrom 2006; seeChapter 2).

Resource management for poverty reduc-tion has tended to focus on water use, foodproduction, or management of other crucialresources, but, in our view, these resources needto be managed in the broader social-ecologicalcontext as part of ecosystems and landscapedynamics. Managing for ecosystem resilience isa necessary but insufficient condition for social-ecological transformations from poverty trapsinto improved states.

Recent work on social interactions alsoreveals the significance of diversity in humaninteractions, in institution building, and forcollective action (Ostrom 2005, Page 2007).Diversity is a crucial element of resilience forcoping with extreme events in a world char-acterized by accelerating directional change.Redundancy (backups), functional diversity ofroles (of species in ecosystems, or of peopleand institutions in social systems), and responsediversity (different responses of species, land-scape elements, individuals or institutions tosuites of disturbances) provide options for flexi-ble outcomes and help social-ecological systemsreorganize and develop (see Chapter 2).

These insights illustrate that the search forblueprint solutions, i.e., uniform solutions toa wide variety of problems that are clus-tered under a single name based on one ormore successful exemplars, lead to resource-management failures (Ostrom et al. 2007; seeChapter 4). Yet, diversity seems to be erod-

ing in many dimensions. It is declining sys-tematically in agriculture and most land- andseascapes (MEA 2005a; see Chapter 12). Atthe same time, in human societies, the bene-fits of efficiency, rationality, and standardizationhave resulted in an emphasis on “best prac-tice”, efficiency, and a tendency toward mono-culture and a dominance of the few. All thischallenges resilience, because it leaves us withan impoverished set of sources of novelty forrenewal.

Such erosion of resilience can, for example,be counteracted by increasing the diversity ofproblem solvers in a team, community or soci-ety, thereby stimulating a wide range of men-tal models and also allowing for transparencyregarding conflicting viewpoints (e.g., disci-plinary background, methodology, conflict andlearning styles, age, gender, and cultural back-ground). Complex problems (problems withmany potential solutions that are quite differ-ent in execution and rankable in quality of out-come) may be solved more effectively by adiverse team of competent individuals than bya team composed of the best individual prob-lem solvers (Page 2007). In this sense, socialdiversity contributes to the sources of resiliencethat strengthen social-ecological systems. Com-bining social, ecological, and economic sourcesof resilience in times of directional and oftenunexpected change provides the seeds not onlyfor adaptive capacity but also for transformingsocial-ecological systems into new and poten-tially more desirable states.

New global institutional structures emergeduring rapid globalization from financialmarkets, multinational companies, tradeagreements, IT-developments, and intergovern-mental treaties. Currently, however, we lack orhave primarily weak international institutionsto deal with ecosystem stewardship for sustain-ability (see Chapter 14). Important advanceslike UN declarations and the IPCC are stilllargely disconnected from powerful economicand political institutions. We envision that, inpace with climatic change and associated dis-turbances, new regional and global governancestructures will emerge that will truly merge theecological and social dimensions for improvedstewardship of ecological life-support systems

118 C. Folke et al.

and ecosystem services. For example, struc-tures such as the European Water Directiveand the MEA are already emerging. Insti-tutional scholars talk about these structuresas multilevel governance systems, and somepropose a polycentric governance structure, inwhich citizens are able to organize in multipledemocratic governing bodies at differing scalesin a specified geographical area to deal withcommon pool resources and stewardship ofecosystems. Selforganized resource governancesystems within a polycentric system may beorganized as special districts, nongovernmentalorganizations, or parts of local governments.These are nested in several levels of general-purpose governments that provide civil equityas well as criminal courts. The smallest unitscan be viewed as parallel adaptive systemsthat are nested within ever-larger units thatare themselves parallel adaptive systems. Thestrength of polycentric governance systemsin coping with complex, dynamic biophysicalsystems is that each of the subunits has consid-erable autonomy to experiment with diverserules for using a particular type of resourcesystem and with different response capabilitiesto external shock. In experimenting with rulecombinations within the smaller-scale unitsof a polycentric system, citizens and officialshave access to local knowledge, obtain rapidfeedback from their own policy changes, andcan learn from the experience of other parallelunits. Redundancy builds in considerablecapabilities and small-scale disasters that maybe compensated by the successful reaction ofother units in the system (Ostrom 2005).

Navigating TransformationThrough Adaptive Governance

The capacity to adapt to and shape changeis a central component of resilience of social–ecological systems. When there is high adapt-ability, actors have the capacity to reorganizethe system within desired states in responseto changing conditions and surprises. But highadaptability may also be recombined with inno-vation and novelty to transform a social–

ecological system into a new regime. Adap-tive governance has emerged as a frameworkfor understanding transformations by expand-ing the focus from adaptive management ofecosystems to address the broader social con-texts that enable shifts in governance systemstoward ecosystem-based management (Folkeet al. 2005).

By governance systems we mean the inter-action patterns of actors, their sometimes con-flicting objectives, and the instruments chosento steer social and environmental processeswithin a particular policy area. Institutions area central component in this context, as are theinteractions between actors and the multilevelinstitutional setting, creating complex relation-ships between people and ecosystem dynamics(Galaz et al. 2008, see Chapter 4).

A transformation of governance may includeboth shifts in perceptions or mental modelsand changes in institutions and other essentialsocial features. Adaptive governance researchaddresses transformations of entire governancesystems from one state to another. Transforma-tions that increase the capacity to learn from,respond to, and manage ecosystem feedbacksgenerally require shifts in social features suchas perception and meaning; social network con-figurations and patterns of interactions amongactors; and associated institutional arrange-ments and organizational structures. In thisbook we are concerned with transformationsthat redirect governance into restoring, sustain-ing, and developing the capacity of ecosystemsto generate essential services.

Path Dependence and Windowsof Opportunity

We still know relatively little about how social–ecological transformations can be orchestratedand the enabling social processes that makeit possible for actors to actively push systemsfrom one trajectory to another. Why do cer-tain strategies succeed and suddenly take off,while others utterly fail? There is a need tounderstand transformative capacity, the capac-ity to shift from trajectories of unsustainableresource use to sustainable ones in the face of


increased resource depletion and global change(Chapin et al. 2006b). Path-dependence charac-terizes most institutional development and pub-lic policy-making (Duit and Galaz 2008). Thesepaths often show a punctuated equilibrium, inwhich long periods of stability and incrementalchange are separated by abrupt, nonincremen-tal, large-scale changes (Repetto 2006).

Windows of opportunity often trigger theselarge-scale changes. Sometimes windows opendue to exogenous shocks and crises, includ-ing shifts in underlying economic fundamen-tals like a rapid rise in energy price, a changein the macro-political environment, new scien-tific findings, regime shifts in ecosystems, orrapid loss of ecosystem services. For exam-ple, a window for changing direction openedup in water management for agricultural andurban areas in California. Water managementhad been locked for decades into a highly engi-neered infrastructure that reinforced one pol-icy and excluded others and pushed the social–ecological system into a crisis. As a response, anew awareness emerged among multiple stake-holders in Californian water management thatbusiness-as-usual was no longer a viable option.The window of opportunity opened throughthe awareness of the crisis. As a necessity, pol-icy and management shifted and broadened toincorporate a wider array of state and federalagencies as well as private and public orga-nizations to address the crisis (Repetto 2006).The social–ecological system seems to be goingthrough a transformation.

Leadership, Actor Groups, SocialNetworks, and Bridging Organizations

Leadership in Transformations

The interplay between individual actors, orga-nizations, and institutions at multiple levels iscentral in social-ecological transformations. Aliterature on the role of leadership strategiesin transformations to ecosystem-based manage-ment is emerging (Westley 2002, Olsson et al.2004b, Fabricius et al. 2007; see Chapter 15),with a focus on the relationship between socialstructures and human agency (see Chapters1 and 4). In the governance systems of the

Everglades in Florida in the USA and Kris-tianstad Vattenrike in southern Sweden, suc-cessful transformations occurred because of theability of leaders to

• reconceptualize key issues,• generate and integrate a diversity of ideas,

viewpoints, and solutions,• communicate and engage with key individu-

als in different sectors,• move across levels of governance and poli-

tics, i.e., span scales,• promote and steward experimentation at

smaller scales, and• recognize or create windows of opportunity

and promote novelty by combining differentnetworks, experiences, and social memories.

Leaders who navigate transformations areable to understand and communicate a wide setof technical, social, and political perspectivesregarding the particular ecosystem stewardshipissues at hand. Visionary leaders fabricate newand vital meanings, overcome contradictions,create new syntheses, and forge new alliancesbetween knowledge and action.

Diversity of Actor Groups and SocialNetworks

People with different social functions operat-ing in teams or actor groups play significantroles in mobilizing social networks to deal withchange and unexpected events and to reorga-nize accordingly. Social roles in networks alsointeract to create tipping points and trans-formations. Gladwell (2000) identified tippingpoint roles and labeled them mavens (altruis-tic individuals with social skills who serve asinformation brokers, sharing and trading whatthey know) and connectors (individuals whoknow many people (both numbers and espe-cially types of people). They enhance the infor-mation base of their social network. Mavensare data banks and provide the message. Con-nectors are social glue and spread the mes-sage. There are also salesmen who have thesocial skills to persuade people unconvincedof what they are hearing. Other social rolesof key individuals operating in actor groups

120 C. Folke et al.

include knowledge carriers, knowledge genera-tors, stewards, leaders, people who make senseof available information, knowledge retainers,interpreters, facilitators, visionaries, inspirers,innovators, experimenters, followers, and rein-forcers (Folke et al. 2005).

Social capital (see Chapter 4) focuses onrelationships among groups, i.e., the bridgingand bonding links between people in social net-works (Wasserman and Faust 1994). Applied toadaptive governance, these relationships mustbe fed with relevant knowledge about ecosys-tem dynamics. This is related to the capacityof teams to acquire and process information,to make sense of scientific data and connectit to a social context, to mobilize the socialmemory of experiences from past changes andresponses, and to facilitate adaptive and inno-vative responses. Social roles of actor groupsare all important components of social net-works and essential for creating the conditionsthat we argue are necessary for adaptive gover-nance of ecosystem dynamics during periods ofrapid change and reorganization. Linking dif-ferent societal levels and knowledge systemsrequires an active role of individuals as coordi-nators and facilitators in co-management pro-cesses. Intermediaries, or middlemen can, forexample, play a role in linking local commu-nities to outside markets (Crona 2006). Bring-ing together different actor groups in networksand creating opportunities for new interactionsare important for dealing with uncertainty andchange and critical factors for learning and nur-turing integrated adaptive responses to change.

Bridging Organizations Connect DifferentLevels of Governance

Bridging organizations coordinate collabora-tions among local stakeholders and actors atmultiple organizational levels (Westley 1995,Hahn et al. 2006). Bridging organizations pro-vide arenas for trust-building, vertical and hor-izontal collaboration, learning, sense-making,identification of common interests, and con-flict resolution. As an integral part of adap-tive governance of social–ecological systems,bridging organizations reduce transaction costs

of collaboration and provide social incentivesto participate in ecosystem stewardship. Theinitiative behind a bridging organization maycome from bottom-up, top-down, or from, forexample, NGOs or companies that bridge localactors with other levels of governance to gener-ate legal, political, and financial support. Suchbridging organizations may also filter externalthreats and redirect them into opportunitiesand help transform social–ecological systemstoward resilience-based stewardship (Olssonet al. 2004b, 2008). Their role in resilience andsustainability needs further investigation.

Interplay Between the Microand the Macro

How do new multilevel governance systemsemerge? What are the enabling conditions forthe emergence of innovative initiatives to dealwith ecosystem change, uncertainty and cri-sis, and the social mechanisms that diffuseinnovations across scales? The micro levelinvolves encounters and patterned interactionamong individuals (which include communica-tion, exchange, cooperation, and conflict), andthe macro level refers to structures in soci-ety (groups, organizations, institutions, and cul-tural productions) that are sustained by mech-anisms of social control and that constituteboth opportunities and constraints on indi-vidual behavior and interactions (Munch andSmelser 1987). Could social innovations gen-erated at local/regional scales influence andtransform governance at a global scale? Canmulti-actor experiments be designed that gen-erate new knowledge, network across scales,pressure governance regimes, and ultimatelylead to tipping points and transformations tomore ecosystem-benign management and gov-ernance? These issues are beginning to beaddressed in the context of adaptive gover-nance (Dietz et al. 2003, Folke et al. 2005).

Learning Platforms as Part of AdaptiveGovernance

The adaptive governance framework suggestsa learning approach that includes fostering a


diversity of approaches and creating “learningplatforms” to experiment with social responsesto uncertainty and change. Such a learningapproach has great potential to enhance theresilience of interconnected social–ecologicalsystems and enhance the capacity of ecosystemsto produce services for human well-being. Forexample, initiatives like UNESCO’s Man andthe Biosphere Programme identifies potentiallearning sites and policy laboratories. The cre-ation of transition arenas in the Netherlandsis another example of experimenting with newapproaches for managing and governing waterresources (van der Brugge and van Raak 2007).

Successful large and long-lived companiesthat depend on continuous innovation, suchas Phillips or IBM, have addressed the ten-sion between moving forward in a conven-tional fashion and exploring by “encapsulating”creative or explorative units (Epstein 2008).They often physically separate the research anddevelopment departments or teams from theproduction teams and train special managerswho can champion and shepherd the innovationprocess while buffering it from the demands ofproduction. This allows the company to buildup a bank of new ideas and products to drawupon in future launches, while simultaneouslyproducing and marketing successful initiatives(Kidder 1981, Kanter 1983, Quinn 1985). Oth-ers implement the ideas when they are suc-cessful enough, thereby avoiding the diver-sion of energy from creativity to production(Mintzberg and Westley 1992).

Three Phases of Social–EcologicalTransformation

Transformation can be triggered by perceivedthreats to an area’s cultural and ecological val-ues. In the wetland landscape of Kristianstad insouthern Sweden, for example, people of vari-ous local steward associations and local govern-ment responded to perceived threats by mobi-lizing and moving into a new configuration ofecosystem management within about a decade(Olsson et al. 2006). This self-organizing pro-cess was led by a key individual who pro-vided visionary leadership in directing change

a)

b)

Preparing thesystem for change

Window ofopportunity

Navigating thetransition

Building resilience of the new direction

Figure 5.4. Three phases of social-ecological trans-formation, linked by window of opportunity–preparing the system for change, navigating the tran-sition, and building resilience of the new direction.The transformation is illustrated in two ways: (a) as aregime shift between multiple stable states, passing athreshold or (b) as a tipping point.

and transforming governance. The transforma-tion involved three phases, where phases (a)and (b) are linked by a window-of-opportunity(Fig. 5.4):

(a) preparing the system for change,(b) navigating the transition, and(c) building resilience of the new governance

regime.

Trust-building dialogues, mobilization ofsocial networks with actors and teams acrossscales, coordination of ongoing activities, sensemaking, collaborative learning, and creatingpublic awareness were part of the process.

A comprehensive framework with a sharedvision and goals that presented conservation asdevelopment and turned problems into oppor-tunities was developed and contributed to ashift in values and meaning of the wetland land-scape among key actors. The shift was facili-tated through broader-scale crises, such as sealdeaths and toxic algal blooms in the NorthSea, which raised environmental issues to a topnational political priority at the time that themunicipality was searching for a new identity.This coincidence of events opened a windowof opportunity at the political level, makingit possible to tip and transform the gover-nance system into a trajectory of adaptive co-management of the landscape with extensive

122 C. Folke et al.

social networks of practitioners engaged in mul-tilevel governance (Fig. 5.4).

A broader analysis of five case studies con-firms this pattern of social–ecological transfor-mation (Olsson et al. 2006). The strategies ofpreparing for change are shown in Table 5.3.Key leaders and shadow networks (informalnetworks that are politically independent fromformal organizations) play a key role in prepar-ing a system for change by exploring alternativesystem configurations and developing strate-gies for choosing among possible futures. Keyleaders can recognize and use or create win-dows of opportunity and navigate transitionstoward adaptive governance. Leadership func-tions include the ability to span scales of gov-ernance, orchestrate networks, integrate andcommunicate understanding, and reconcile dif-ferent problem domains. Successful transfor-mations are often preceded by the emergenceof informal networks that help to facilitateinformation flows, identify knowledge gaps, andcreate nodes of expertise in ecosystem manage-ment that can be drawn upon at critical times. Inthe Kristianstads Vattenrike Biosphere Reserveand in the Everglades, these networks werepolitically independent from the fray of reg-ulation and implementation in places whereformal networks and many planning processesfail. These shadow networks serve as incuba-tors for new approaches to governing social–ecological systems (Gunderson 1999). Theseinformal, outside-the-fray shadow groups areplaces where new ideas often arise and flourish.These groups often explore flexible opportuni-ties for resolving resource issues, devise alterna-tive designs and tests of policy, and create waysto foster social learning. Because the membersof these networks are not always under scrutinyor the obligations of their agencies or con-stituencies, they are freer to develop alterna-tive policies, dare to learn from each other, andthink creatively about how to resolve resourceproblems.

In Australia, a flexible organization, theGreat Barrier Reef Marine Park Authority,was crucial in initiating a tipping point of gov-ernance toward ecosystem-based management(Olsson et al. 2008). This agency was also instru-mental in the subsequent transformation of

governance systems, from the level of localfisher organizations to national political pro-cesses, and provided leadership throughout theprocess. The Great Barrier Reef study identifiessocial features and strategies that made it pos-sible to shift the direction of an already exist-ing multilevel governance regime toward large-scale ecosystem-based management. Strategiesinvolved active internal reorganization andmanagement innovation, leading to

• an ability to coordinate the scientific commu-nity,

• increased public awareness of environmentalissues and problems,

• involvement of a broader set of stakeholders,and

• maneuvering the political system for supportat critical times.

The transformation process was driven byincreased pressure on the Great Barrier Reef(from terrestrial runoff, overharvesting, andglobal warming) that triggered a new sense ofurgency to address these challenges. It shiftedthe focus of governance from protection ofselected individual reefs to ecosystem steward-ship of the larger-scale seascape.

The study illustrated the significance of stew-ardship that can change patterns of interac-tions among key actors and allow new formsof management and governance to emerge inresponse to environmental change. The studyalso showed that enabling legislation or otherforms of social bounds were essential, butnot sufficient for shifting governance towardadaptive co-management of complex marineecosystems.

In contrast to the Great Barrier Reef case,marine zoning in the USA has been severelyconstrained due to inflexible institutions, lack ofpublic support, difficulties developing accept-able legislation, and failures to achieve desiredresults even after zoning is established. Under-standing successes and failures of governancesystems is a first step in improving their adap-tive capacity to secure ecosystem services in theface of uncertainty and rapid change.

The case studies discussed here show thattransformation is more complex than sim-ply changing legislation, providing economic

5 Transformations in Ecosystem Stewardship 123Ta

ble

5.3

Key

fact

ors

for

prep

arin

gfiv

edi

ffer

ents

ocia

l–ec

olog

ical

syst

ems

for

tans

form

atio

n1 .

Soci

al–e

colo

gica

lsys

tem

Key

fact

ors

build

ing

know

ledg

eN

etw

orki

ngL

eade

rshi

p

Kri

stia

nsta

dsV

atte

nrik

e(c

reat

ion

ofth

eK

V)

Ane

wpe

rspe

ctiv

eon

ecos

yste

mm

anag

emen

tofi

nteg

rate

dla

ndsc

ape-

leve

lsol

utio

nsgu

ided

the

deve

lopm

ento

fkno

wle

dge.

Itin

clud

edid

enti

fyin

gkn

owle

dge

gaps

for

man

agin

gth

eK

Van

din

itia

ting

stud

ies

tofil

lthe

m.

The

emer

genc

eof

the

netw

ork

inth

em

id-1

980s

conn

ecte

dac

tors

wit

hdi

ffer

enti

nter

ests

.Thi

sin

clud

edve

rtic

allin

ksan

dho

rizo

ntal

links

betw

een

gove

rnm

enta

genc

ies,

NG

Os,

the

mun

icip

alit

y,an

dla

ndow

ners

.

Lea

ders

hip

emer

ged

that

was

impo

rtan

tfo

rco

nnec

ting

peop

le,d

evel

opin

g,an

dco

mm

unic

atin

ga

visi

onof

ecos

yste

mm

anag

emen

t,an

dbu

ildin

gtr

usta

ndbr

oad

supp

ortf

orch

ange

.

The

Eve

rgla

des

(eco

syst

emre

stor

atio

n)A

few

key

scie

ntis

tsw

ere

frus

trat

edby

cont

inui

ngec

osys

tem

degr

adat

ion,

whi

chth

eytr

ied

toad

dres

sin

wor

ksho

ps.T

heec

osys

tem

rest

orat

ion

(res

ilien

ce)

pers

pect

ive

guid

edm

odel

ing

wor

ksho

psin

whi

chin

form

atio

nw

assy

nthe

size

dan

dus

edto

deve

lop

com

posi

tepo

licie

s.

Ane

twor

kof

scie

ntis

tsem

erge

din

the

late

1980

san

dfo

rmed

the

adap

tive

man

agem

entg

roup

.In

1992

,net

wor

king

was

exte

nded

into

the

man

agem

enta

ndpo

litic

alar

eas

tosp

read

the

idea

sof

the

adap

tive

man

agem

entg

roup

,lin

kac

tor

grou

psop

erat

ing

atdi

ffer

ent

orga

niza

tion

alle

vels

,and

repr

esen

tdi

ffer

enti

nter

ests

.

Lea

ders

hip

emer

ged

that

brou

ghti

na

nove

lper

spec

tive

ofec

osys

tem

resi

lienc

e,bu

ilttr

ust,

and

conn

ecte

dpe

ople

.The

lead

ers

wer

ew

eary

ofon

goin

gle

gala

ctio

nsan

dw

ishe

dto

purs

ueal

tern

ativ

ew

ays

ofm

anag

emen

t.T

hey

focu

sed

onen

suri

ngth

een

gage

men

tofa

llgr

oups

,not

just

afe

wsp

ecia

lint

eres

tgro

ups.

The

Nor

ther

nH

ighl

ands

Lak

eD

istr

ict

(sus

tain

able

futu

res)

The

pola

riza

tion

amon

gdi

ffer

enta

ctor

grou

pshi

nder

sth

esh

arin

gof

new

idea

san

din

nova

tion

s.H

owev

er,a

few

brid

ging

effo

rts

are

deve

lopi

ng,a

ndth

ese

coul

dnu

clea

teto

prov

ide

the

nece

ssar

yin

stit

utio

nsfo

rbu

ildin

gan

dsh

arin

gkn

owle

dge.

Net

wor

king

ata

regi

onal

scal

eth

atco

nnec

tsdi

ffer

entg

roup

sof

acto

rsis

poor

lyde

velo

ped.

Lea

ders

hip

for

colle

ctiv

eac

tion

and

ecos

yste

mm

anag

emen

tatt

here

gion

alle

velh

asno

tem

erge

d.In

stea

d,le

ader

ship

has

emer

ged

for

purs

uing

spec

ific

inte

rest

s.

Mae

Nam

Pin

gB

asin

(sus

tain

able

wat

erm

anag

emen

t)

Kno

wle

dge

base

don

the

ecos

yste

map

proa

chha

sbe

enas

sim

ilate

dfr

oma

wid

era

nge

ofso

urce

s,an

din

nova

tive

ecos

yste

ms

appr

oach

esex

istb

utdo

not

guid

ene

twor

king

atth

ere

gion

alle

vel.

Net

wor

king

atth

eba

sin

leve

lis

lack

ing.

Inst

ead,

netw

orks

that

serv

ean

dpr

otec

tsp

ecifi

cin

tere

sts

are

deve

lopi

ng.

Lea

ders

hip

for

colle

ctiv

eac

tion

and

ecos

yste

mm

anag

emen

tatt

heba

sin

leve

lhas

note

mer

ged.

Inst

ead,

lead

ersh

ipha

sem

erge

dfo

rpu

rsui

ngsp

ecifi

cin

tere

sts.

Gou

lbur

n-B

roke

nC

atch

men

t(su

stai

nabl

eag

ricu

ltur

e)

The

rew

asa

lack

ofin

nova

tion

that

mad

eit

impo

ssib

leto

expl

ore

new

confi

gura

tion

sof

the

syst

em,i

npa

rtic

ular

,way

sto

addr

ess

ecos

yste

mpr

oces

ses.

Bui

ldin

gkn

owle

dge

tosu

ppor

tthe

stat

usqu

oap

proa

chto

ecos

yste

mm

anag

emen

twas

emph

asiz

ed.

Net

wor

ksem

erge

dth

atco

nnec

ted

peop

lean

din

tere

sts

atdi

ffer

entl

evel

s.T

hese

netw

orks

wer

ela

ter

form

aliz

edin

tode

cisi

on-m

akin

gan

dim

plem

enti

ngor

gani

zati

ons.

Lea

ders

hip

did

emer

gefo

rco

llect

ive

acti

onat

the

catc

hmen

tlev

el,b

utno

tto

prov

ide

ano

vela

ppro

ach

toec

osys

tem

man

agem

ent.

1.R

epri

nted

from

Ols

son

etal

.(20

06).

124 C. Folke et al.

incentives or introducing new restrictions onresource use. As observed by McCay (1994),the change of perceptions and mental models ofthe significance of ecosystems for human well-being is an important part of ecosystem stew-ardship and can change human behavior on afairly large scale without involving the politicalprocesses of making and changing institutions.

Actions that foster successful transforma-tions of social–ecological systems toward adap-tive governance of landscapes and seascapesoften include:

• Change attitudes among groups to a new,shared vision; differences are good, polariza-tion is bad.

• Check for and develop persistent, embeddedleadership across scales; one person can doit for a time, but several are better locally,regionally, and politically.

• Design resilient processes, e.g., discourse andcollaborations, not fixed structures.

• Evaluate and monitor outcomes of pastinterventions and encourage reflection fol-lowed by changes in practices.

• Change is both bottom-up and top-down.Otherwise, scale conflicts ultimately compro-mise the outcome; globalization is good butcan destroy adaptive capacity both region-ally and locally.

• Develop and maintain a portfolio of projects,waiting for opportunities to open.

• Always check larger scales in different sec-tors for opportunities; this is not science, butpolitics.

• Know which phase of an adaptive cycle thesystem has reached and identify thresholds;talk about it with others.

• Plan actions for surprise and renewal dif-ferently than growth and conservation; effi-ciency is on the last part and resilience on thefirst.

• The time horizon for effect and assessment isat least 30–50 years; restructuring resiliencerequires attention to slow dynamics.

• Create cooperation and transform conflict,but some level of conflict ensures that chan-nels for expressing dissent and disagreementremain open.

• Create novel communication face-to-face,individual-to-individual, group-to-group,and sector-to-sector.

• Encourage small-scale revolts, renewals andreorganizations, not large-scale collapses.

• Try to facilitate adaptive governance byallowing just enough flexibility in institutionsand politics.

These generalizations can help managersnavigate more effectively the periods of uncer-tainty and turbulence that are unavoidablecomponents of any social–ecological transfor-mation.

Summary

This book emphasizes the need for ecosys-tem stewardship to generate a deeper under-standing of integrated social–ecological sys-tems undergoing change. It requires an expan-sion of focus from managing natural resourcesto stewardship of dynamic and evolving land-scapes and seascapes in order to sustain ecosys-tem services. Such stewardship requires gover-nance systems that actively support ecosystem-based management and allow for learningabout resource and ecosystem dynamics. Achallenge is to develop governance systems thatare flexible, adaptive, and have the capacity totransform. It requires dealing with change—not just incremental and predictable change,but uncertain, abrupt, and surprising change.Chapter 5 has identified and discussed featuresof social–ecological systems that create barri-ers and bridges for transformations to moredesired states and presents strategies for build-ing and enhancing their resilience in times ofdirectional change.

The first five chapters of the book pre-sented existing and emerging theory and con-cepts in relation to resilience-based ecosys-tem stewardship and serve as the founda-tion for the remaining chapters. The followingchapters, structured into major types of social–ecological resource systems covering local-to-global scales, illustrate this foundation and pro-vide insights, challenges, and implementationstrategies for improved stewardship of terres-


trial and marine ecosystems and the servicesand fundamental support that they provide tohumanity.

Review Questions

1. How does transformation differ from adap-tation?

2. What types of surprises occur in social–ecological systems? How do managementstrategies differ in preparing for andresponding to each type of surprise?

3. What human actions can change resilience?In what ways might they interact and leadto regime shifts of social–ecological systems?What intervention strategies might addressthese interactions and reduce the likelihoodof undesirable regime shifts?

4. How might appropriate intervention strate-gies differ between poverty traps, rigiditytraps, and cascading effects of regime shifts?

5. In what ways can mismatches of institutionsand ecosystems lead to surprises and regimeshifts?

6. What tools are available to deal with trueuncertainty?

7. What is adaptive governance and how doesit differ from adaptive co-management?

8. Which are the major phases of transfor-mations and their social features? In whatways can management actions foster social–ecological resilience to facilitate activelynavigated transformations?

Additional Readings

Carpenter, S.R., E.M. Bennett, and G.D. Peterson.2006. Scenarios for ecosystem services: Anoverview. Ecology and Society 11(1):29 [online]URL: http://www.ecologyandsociety.org/vol11/iss1/art29/

Folke, C., T. Hahn, P. Olsson, and J. Norberg.2005. Adaptive governance of social–ecologicalsystems. Annual Review of Environment andResources 30:441–473.


Olsson, P., L.H. Gunderson, S.R. Carpenter, P.Ryan, L. Lebel, et al. 2006. Shooting therapids: Navigating transitions to adaptive gov-ernance of social-ecological systems. Ecologyand Society 11(1):18. [online] URL: http://www.ecologyandsociety.org/vol11/iss1/art18/

Repetto, R. 2006. Punctuated Equilibrium and theDynamics of U.S. Environmental Policy. YaleUniversity Press, New Haven.

Scheffer, M., S.R. Carpenter, J.A. Foley, C. Folke, andB.H. Walker. 2001. Catastrophic shifts in ecosys-tems. Nature 413:591–596.

Walker, B.H., C.S. Holling, S.R. Carpenter, andA.P. Kinzig. 2004. Resilience, adaptability andtransformability in social–ecological systems.Ecology and Society 9(2):5 [online] URL:http://www.ecologyandsociety.org/vol9/iss2/art5/

Part IIStewarding Ecosystems

for Society

6Conservation, Community,and Livelihoods: Sustaining,Renewing, and Adapting CulturalConnections to the Land

Fikret Berkes, Gary P. Kofinas, and F. Stuart Chapin, III

Introduction

Since most of the world’s biodiversity is not inprotected areas but on lands used by people,conserving species and ecosystems depends onour understanding of social systems and theirinteractions with ecological systems. Involvingpeople in conservation requires paying atten-tion to livelihoods and creating a local stakefor conservation. It also requires maintainingcultural connections to the land and at timesrestoring and cultivating new connections. Thischapter addresses human–wildlife–land inter-actions across a range of hinterland ecosys-tems, from relatively undisturbed “wildlands”to more intensively manipulated rural agricul-tural areas where local communities are an inte-gral component of the landscape. These regionscommonly comprise unique ecosystems that arein many cases important hotspots of global bio-diversity. Here we examine three case studies –Ojibwa and Cree use of boreal forest biodi-

F. Berkes (�)Natural Resources Institute, University of Manitoba,Winnipeg, Manitoba, Canada R3T 2N2e-mail: [email protected]

versity, the community-based programs for ele-phant conservation in sub-Saharan Africa, andGwich’in engagement in international manage-ment of the Porcupine Caribou Herd in ArcticNorth America. The three cases highlight therelationship between conservation and commu-nity livelihoods to illustrate strategies that com-munities have used and the challenges they faceto sustain land, resources, and their own well-being.

Historically, hinterland regions of low humandensity have been the homelands of people whoare highly dependent for their livelihoods onecosystem services through small-scale agricul-ture, forest resource use, small-scale fisheries,and subsistence economies of hunting and gath-ering. By some estimates, indigenous peoples ofthese hinterland regions inhabit 20–30 percentof the earth’s surface, about four to six timesmore area than is included in all formally des-ignated protected areas of the world (Stevens1997). These areas also provide cultural andsometimes spiritual connections to land, plants,and animals. But in the globalized world oftoday, no place is isolated from external driverssuch as global market forces. Thus, challengesto sustainability are ever-present, even when thesocial–ecological system is distant from denselyinhabited areas. In many cases, these areas


130 F. Berkes et al.

are increasingly encroached upon by nonlocalinterests. In other cases, problems are due tolocal population increase, land-use intensifica-tion, introduction of new markets, persistentsocial inequities at community and regional lev-els, and competition over scarce resources. Andin still others, the challenges are related to theeffects of global-scale processes, such as cli-mate change. A serious challenge in conserva-tion for people in all these regions is how tosustain the connections between people and theland. Approaches to these challenges illustrateseveral principles presented in Chapters 1–5:

1. People and nature are integral components ofall social–ecological systems, from wildernessto cities.

2. Local and traditional knowledge, culturallegacies, social institutions, and social net-works play a critical role in sustaining the useof ecosystem services.

3. Resource uses compatible with conserva-tion objectives can be important sources ofresilience by fostering ecological renewalcycles.

4. Participatory approaches that includeresource users in monitoring, research,and policy-making improve understanding,responsiveness to change, and compliancewith rules.

5. Changes in wildlife habitat often reflectregional and global processes, so sustainabil-ity requires attention to cross-scale interac-tions.

Strategies for Conservationand Biodiversity

Different strategies have been implementedacross the globe to achieve conservation andmaintain biodiversity (Box 6.1). One strategy isto establish protected areas that strictly excludehuman use and provide for the preservation ofwilderness. In other cases, such as BiosphereReserves, human use areas are established adja-cent to core protected areas, with restricted-use areas as an intervening buffer zone. In yetother cases, intensive human use across a regionoccurs with strategies that seek the related goalsof human livelihoods and biodiversity protec-tion. In the face of rapid change in land useand the extinction of species, there is consid-erable debate about which strategies are bestfor achieving conservation objectives. Climatechange complicates this debate, because areasconsidered to be sensitive habitat today mayshift geographically in the future.

Box 6.1. Categories of Protected Areas

Protected areas are defined by World Con-servation Union as “An area of land and/orsea especially dedicated to the protectionand maintenance of biological diversity, andof natural and associated cultural resources,and managed through legal or other effec-tive means.” The World Conservation onProtected areas and the World Conserva-tion Union (IUCN) have provided leader-ship in classifying protected areas into sixcategories (see below), ranging from thosededicated exclusively to preserving ecologi-cally unique areas to those in which peopleare part of important landscapes. This classi-fication system has been modified since firstestablished in 1978, reflecting shifts in think-

ing about ecological change, the recognizedimportance of unique ecosystems providingfor hinterland people’s livelihoods, and theinvolvement of local communities in conser-vation (IUCN 1991, Stevens 1997).

CATEGORY Ia: Strict Nature Reserve:protected area managed mainly for science.

Definition: Area of land and/or sea pos-sessing some outstanding or representativeecosystems, geological or physiological fea-tures, and/or species, available primarilyfor scientific research and/or environmentalmonitoring.

CATEGORY Ib: Wilderness Area: pro-tected area managed mainly for wildernessprotection.

6 Conservation, Community, and Livelihoods 131

Definition: Large area of unmodi-fied or slightly modified land, and/orsea, retaining its natural character andinfluence, without permanent or signifi-cant habitation, which is protected andmanaged so as to preserve its naturalcondition.

CATEGORY II: National Park: protectedarea managed mainly for ecosystem protec-tion and recreation.

Definition: Natural area of land and/orsea, designated to (a) protect the ecologi-cal integrity of one or more ecosystems forpresent and future generations, (b) excludeexploitation or occupation inimical to thepurposes of designation of the area, and (c)provide a foundation for spiritual, scientific,educational, recreational, and visitor oppor-tunities, all of which must be environmentallyand culturally compatible.

CATEGORY III: Natural Monument:protected area managed mainly for conserva-tion of specific natural features.

Definition: Area containing one, or more,specific natural or natural/cultural featurewhich is of outstanding or unique valuebecause of its inherent rarity, representativeor aesthetic qualities or cultural significance.

CATEGORY IV: Habitat/Species Man-agement Area: protected area managed

mainly for conservation through manage-ment intervention.

Definition: Area of land and/or sea sub-ject to active intervention for managementpurposes so as to ensure the maintenance ofhabitats and/or to meet the requirements ofspecific species.

CATEGORY V: Protected Landscape/Seascape: protected area managed mainly forlandscape/seascape conservation and recre-ation.

Definition: Area of land, with coast andsea as appropriate, where the interaction ofpeople and nature over time has producedan area of distinct character with significantaesthetic, ecological and/or cultural value,and often with high biological diversity. Safe-guarding the integrity of this traditional inter-action is vital to the protection, maintenance,and evolution of such an area.

CATEGORY VI: Managed Resource-Protected Area: protected area managedmainly for the sustainable use of naturalecosystems.

Definition: Area containing predomi-nantly unmodified natural systems, managedto ensure long-term protection and mainte-nance of biological diversity, while providinga sustainable flow of natural products andservices to meet community needs.

The American national parks model of pro-tected areas is considered by many as syn-onymous with biodiversity and preservation.Variants of this approach have been appliedthroughout the globe, and in some cases havesignificantly enhanced biodiversity in the faceof global change. Yellowstone National Park,among the first national parks in the world,was established out of concern for the west-ern expansion of settlers and the impacts ofthat expansion on Native American societies,wildlife, and wilderness. In spite of many park-management problems, such as loss of speciesand the suppression of most fires, the protec-tion of these wildlands has supported the per-

sistence of unique species in a wilderness land-scape that is of high cultural value to society.

The national park approach, however, comeswith limitations. First, the vast numbers of visi-tors may impact a region even where no extrac-tive use is allowed (2.8 million visitors to Yel-lowstone in 2006). Second, exclusion of peo-ple and their associated disturbances may notbe sufficient to maintain a particular ecolog-ical state. Ecosystems are naturally dynamic,and sustaining ecosystem resilience requiresthe maintenance of ecological variability andcycles (see Chapter 2). Third, effective pro-tection in some cases may require that pro-fessional resource managers work closely with


local residents, rather than excluding them ortheir livelihood activities from the protectedarea.

The idea of protected areas and notionsof wilderness as free from people have insome regions created serious conflicts with localland and resource use. People-free parks arereferred to by critics as the “fences and fines”approach to conservation.

The American national park model is not theonly way to achieve conservation. National andinternational approaches to conservation haveincreasingly used a mix of strategies to main-tain biodiversity and conservation, includingthose in which community-based managementserves as a foundation (Borgerhoff Mulderand Coppolillo 2005). The appropriateness ofthese strategies depends to a great extenton scale. In the absence of adequate com-munity institutions for conservation, national-level strategies may be critical, especially whereinternal conflict and or national military activ-ities threaten endangered species. But wherelocal communities have a close relationshipwith land and resources, and where institutionsprovide for sufficient local incentives to con-serve, community-based approaches can sup-port biodiversity while addressing issues ofhuman livelihoods and community sustainabil-ity. In those hinterland regions where thereis poverty or marginal economic opportuni-ties, stewardship of nature may be an effec-tive means of addressing the problems of liveli-hoods (WRI 2005).

Critical Social–EcologicalProperties that Support BothConservation and Livelihoods

Many small-scale indigenous and other ruralsocieties throughout the world live in close rela-tionship with lands and resources. These hinter-lands support the rural economy of much of thedeveloping world, as well as the parts of thedeveloped world where people make a liveli-hood using a variety of local resources – forests,wildlife, fisheries, and rangelands, along with

small-scale agriculture. In many regions today,these activities represent mixed economies ofsubsistence harvesting and small-scale cashmarkets.

Human–environment relationships amongmany hinterland peoples are reflected in theirsocial organization, modes of production, andcultural worldview. These lands can includeabandoned agricultural and restored environ-ments, such as rehabilitated wetlands, as wellas the more “pristine.” They can support agro-forestry systems that are low-intensity mixes ofagriculture and tree crops, such as those foundin many parts of Asia, Latin America, andAfrica. The ecosystems discussed in this chapterusually (but not always) have low human pop-ulation densities. These lands often retain a rel-atively high proportion of their native biodiver-sity and a relatively productive set of ecosystemservices. They are the rural areas of the worldthat support a diversity of economic activitiesthat enable the inhabitants to make a livelihoodlargely from their local ecosystems.

Although many systems are losing biologicaldiversity and associated ecosystem services atan accelerating pace (MEA 2005d), the outlookis not entirely negative. There is an enormousreservoir of cultural understanding of ecosys-tems (Posey 1999, Berkes 2000) and an associ-ated diversity of environmental ethics (Taylor2005). Through many generations, many soci-eties have developed sensitivities to ecologicalchange and strategies for responding to themin ways that foster sustainability. On the onehand, we need to study and build on this expe-rience. On the other, we need to seek signif-icant changes in existing policies, institutions,and practices to meet the challenge of revers-ing the degradation of ecosystems in the face ofincreasing demands for ecological services. Asimportant as they are, setting aside protectedareas alone will not be sufficient to meet thischallenge.

There is an increasing interest in conserva-tion on lands used by rural people. Major inter-national organizations have been exploring var-ious options (WRI 2005). One UN agency, theUN Development Programme (UNDP), hasencouraged grassroots projects that combine


biodiversity conservation and poverty allevia-tion goals in tropical countries. This project,called the UNDP Equator Initiative (Timmerand Juma 2005, Berkes 2007, UNDP 2008), isone of many ways in which the internationalcommunity has seriously begun to exploreways to reconcile conservation and livelihoods(Brown 2002). There are active debates on theappropriate role of local communities in thiskind of integrated conservation development.While the track record of community-basedconservation as practiced in the last 20 or soyears is mixed, there is an emerging consensusthat rural peoples of the world and their com-munities need to be involved as major partnersin any conservation effort (Berkes 2004).

Conservation biologists Bawa and colleagues(2004) have been exploring the implicationsof different approaches to conservation. Theyobserved that a number of conservation mod-els have been developed by large internationalconservation organizations. These models seekunifying conservation principles and have beenpartly driven by a need for general application.However, these generalized, monolithic modelstend to view the world through relatively coarsefilters and may be at odds with the emergenceof fine-grained models adapted to local condi-tions (Bawa et al. 2004). The implementation ofsuch monolithic models can lead to the oppo-site effect, ultimately eroding biological diver-sity and ecosystem health.

Conservation efforts at all levels are con-fronted by complexity and heterogeneity ofecosystems. In lands that serve both conserva-tion and livelihood needs, the situation is fur-ther complicated by a myriad of social, eco-nomic, and institutional factors that influencethe prospects for conservation. Monolithic con-servation models that are broadly applied arenot equipped to deal with human issues. How-ever, many protected areas include human uses,and the management of most kinds of pro-tected areas, even national parks, involves deal-ing with people who live in or around them.Conservation science has proceeded to developas a technical field with narrowly trained tech-nical experts, whereas practical conservationrequires interdisciplinary skills, including the

ability to understand social, economic, andinstitutional processes (Berkes 2004).

Before the advent of conservation science,many local groups successfully conserved bio-diversity and protected ecological processesby following a diversity of approaches, rang-ing from the strict preservation of sacredareas (Ramakrishnan et al. 1998) to sustain-able resource use. Obviously, in many areas,human activities have caused the degradationof ecosystems and loss of biodiversity. But thisis not always the case, and it is important to doc-ument and understand how some of the moresuccessful traditional systems work.

The Western Ghats in southern India is oneof the “biodiversity hotspots” of the world.These low-lying beautiful green mountains sup-port high levels of biodiversity as well as a rel-atively high population density (Plate 5). Muchof the landscape is in diverse agroforestry sys-tems (as opposed to monoculture plantations).Annual, perennial, and tree crops are growntogether by landholders in species combina-tions that have evolved over hundreds of years.The patchwork of these agroforestry operationsare interspersed with government-designatedprotected areas and traditional sacred groveswhere people refrain by social custom fromexploitation. Bhagwat et al. (2005) comparedthe biodiversity of several groups of organ-isms in three kinds of areas: forest reserves,sacred groves, and coffee agroforestry planta-tions. They found surprisingly high levels of bio-diversity, comparable to that found in forestreserves, in sacred groves and in coffee planta-tions (Fig. 6.1). Although endemic tree specieswere more abundant in the forest reserve thanin sacred groves, threatened trees were moreabundant in sacred groves than in the for-est reserve. They concluded that sacred grovesmaintained by tradition and the multifunctionallandscapes produced by centuries-old systemsof agroforestry should be considered an integralpart of conservation strategies in this region(Bhagwat et al. 2005).

Based on findings such as these, Bawa et al.(2004) argue that the best hope for conservationin a complex and rapidly changing world is tofocus on locally driven approaches that (1) draw


Coffeeplantations

n = 162species

Coffeeplantations

n = 70species

245

19

9637 14

20

Sacred grovesn = 75

species

Sacred grovesn = 167species

Forest reserven = 39

species

Forest reserven = 134species

10 1 8

434

25

12

Trees Birds

Figure 6.1. Numbers of shared and restricted tree species and bird species found in forest reserves, sacredgroves, and coffee plantations in Kodagu, Western Ghats of India. Adapted from Bhagwat et al. (2004).

on local and traditional practices as well as sci-ence; (2) are locally adaptive; and (3) seek toincrease human and social capital in additionto natural capital. They argue that initiativesthat incorporate participatory approaches, con-flict resolution, partnerships, and mutual learn-ing are likely to garner broad support. Suchapproaches are relevant to many parts of theworld. Empowerment of local communities andstrengthening local institutions are precondi-tions to the success of long-term conserva-tion goals. These goals may be best achievedthrough a focus on building resilience in social–ecological systems to be able to respond toshocks and surprises, and maintaining a recip-rocal relationship between humans and nature.

The sections below present three dimen-sions of linking conservation, communities, andlivelihoods. Each is supported by a case study toillustrate (1) how community-based approachescan support conservation and livelihoods, (2)how community institutions for conservation ofthreatened species can evolve when attentionis paid to the incentive structures that providefor livelihoods, and (3) how co-managementarrangements can serve to link communitiesto the national and international arenas togive local interests a voice in conservation.A fourth section illustrates how conventionalsingle-species approaches to conservation ofwild resources can be counter-productive ifsocial–ecological interactions are neglected.

Addressing Conservation Issues

Traditional Managementof Biodiversity

Many traditional societies have used ecosys-tem services without destroying the ecosystemsof which they have been part for thousandsof years. Otherwise, there would have beenno resources left for us to conserve today. Alandscape produces a variety of goods and ser-vices for livelihood needs, and some humanactions maintain ecological processes and bio-diversity. In practice, people actively disturband manage their landscape through the trig-gering of successional processes. Disturbancehelps maintain spatial and temporal diversityat both landscape and site levels. The spe-cific mechanisms by which biodiversity is con-served in these multifunctional landscapes varywith the ecological and cultural settings. TheCree and Ojibwa (Anishnaabe) people of theboreal forest of central Canada have developedpractices that sustain biodiversity. For example,the use of fire helps maintain ecological func-tions and renewal cycles of the boreal forestecosystem while enabling the Cree and Ojibwato meet their livelihood needs (Box 6.2). Thecase study illustrates the potential of usinglocal cultural practices and values to protectbiodiversity.


Box 6.2. Boreal Case Study of Indigenous People Keeping the Land Healthy

Sources: Berkes and Davidson-Hunt (2006);Davidson-Hunt et al. (2005); O’Flahertyet al. (2008); Lewis and Ferguson (1988).

The central Canadian boreal zone is pop-ulated mainly by the Cree and the Ojibwa(Anishnaabe), two large and related groupsof indigenous people. There is little agricul-ture because of the short growing season andpoor soils in this glaciated landscape. Tradi-tionally, up until 1960s or so, people weremobile, following an annual cycle of wildlifeharvesting activities. The subsistence systemwas based on terrestrial and aquatic ani-mals (moose, caribou, and beaver), water-fowl (geese and ducks), fish, medicinal plants,berries, and other nontimber forest products(NTFPs). The cash economy depended onthe fur trade. Senior hunters or stewards pro-vided leadership, oversaw proper land usepractices, and rules of respect and reciprocity.Social organization was by family groups,loosely organized into local bands, and sincethe 1960s, into permanent villages consistingof several local bands.

In recent decades, traditional family-basedland use practices and institutions havebecome weaker but persisted, with thehunters relying on rapid transport such assnowmobiles and motor canoes to accessresources seasonally. The land-based econ-omy is still important for food, cash (forestry,commercial fishing, hunting and fishing out-fitting, ecotourism), and a sense of culturalidentity. Much of the cash economy is nowbased on service industries in permanenttowns and villages. Store-bought food is gain-ing in importance, even though it providesa poorer diet, high in fat and carbohydrates,than the one based on fresh meat.

The use of many NTFPs is linked to theecological processes of disturbance and suc-cession. Various species of trees and shrubsare distributed in space and time relative todisturbance. For example, in the lands of theShoal Lake Ojibwa people in northwesternOntario, fireweed (Epilobium angustifolium

L.) occurs in the early years following a dis-turbance, ginseng (Panax quinquefolius L.)is found under mature forest canopies, whilehighbush cranberry (Viburnum trilobum L.)often occurs along riverbanks disturbed peri-odically by spring flooding. Fireweed and gin-seng are utilized as medicinal plants whilehighbush cranberry is an edible berry.

Many NTFPs typically depend on succes-sion management, and a huge diversity ofsuccession-management systems have beendocumented from all parts of the world.Ecologically speaking, these succession-management systems all involve renewalcycles initiated by a disturbance event. Thedisturbance could be a natural fire, a pestinfestation, a blowdown, or it could be ahuman-made fire or a forest cut. A typicalrenewal cycle starts with an early succes-sional phase of rapidly growing herbaceousplants (see Chapter 2). Gradually, shrubbyplants take over, shading out the grasses andother pioneer species. Larger trees graduallytake over, approaching a climax phase beforethe cycle repeats.

In the boreal zone, the climax phase isnot biologically productive. To provide freshbrowse for wildlife and NTFPs for people,a disturbance is needed to release nutrientsand start the renewal cycle over again. In thecentral Canadian boreal zone, the typical firefrequency is about 300 years; insect infesta-tions may be more frequent. Traditionally,some groups of boreal indigenous peopleused small-scale fires to clear the land, attractgrazing animals and waterfowl, but this prac-tice was banned in the 1950s by resource-management authorities. As rediscovered inrecent years, and now used by parks author-ities, frequent small disturbance events actu-ally help ecosystem renewal. Conversely, theprevention of small disturbances makes aforest ecosystem increasingly vulnerable tolarge, potentially disastrous disturbances. Aclassical example is Sequoia National Park inthe USA, where a century of fire prevention


allowed white fir trees to form a ladder offuels from the understory to the canopy ofgiant sequoia trees that had resisted under-story fires for thousands of years (Stephens1998).

In resilient ecological systems, small dis-turbances precipitate the release phase thathelps system renewal by leading to a reor-ganization phase in which the memory ofthe system enables ecological cycles to startover (see Chapter 2). The memory can con-sist, for example, of pine cones, anythingthat helps the forest ecosystem to perpetuateitself. It can also consist of the “social mem-ory” of traditional practices such as those ofthe Ojibwa that help renew forest ecosys-tems.

Senior hunters and land experts amongthe Cree and Ojibwa have an understand-ing of renewal cycles that is similar to thatof contemporary ecologists. Although theyrarely practice fire management (it is illegal),they use naturally occurring disturbances andforestry clear-cuts to get what they need fromthe forest. Their ecological understanding ismixed with indigenous spirituality and landethics. For example, in the perspective of theIskatewizaagegan (Shoal Lake) Ojibwa, theCreator provided everything that the peoplewould need for their survival. In return, theOjibwa hold the responsibility to maintainthese gifts that were given to them. Practicesthat harm these gifts can lead to certain unde-sirable consequences for an individual or theindividual’s family. There is a basic duty uponthe Ojibwa not to try to “control” nature. Thepeople believe that the forest changes all thetime but follows a natural pattern. The sameprinciple is followed in Pikangikum, a differ-ent Ojibwa group, and is concisely translatedinto English as: “as was, as is”.

The creation of blueberry patches throughrepeated burning is not seen as a contradic-tion of this principle. Burning or other distur-

bance simply reveals the different combina-tions of plants that naturally occur. “As was,as is” means that all that was on the landbefore should be still be there today and alsotomorrow. While a fire may destroy a forest,it also reveals the natural pattern that blue-berries follow fire on sandy soils. Then otherplants follow the blueberries, and then theforest returns. The cycle can be modified byhumans but should not be disrupted.

Ojibwa elders hold that all plant speciesare equally important. It cannot be said thatsome are more important than others. Whatis important is the protection of the full suiteof plant species. This is because every habi-tat and species has a reason to be there,known or unknown to humans, and for thatreason the full set of plant species shouldbe maintained into the future. There is asecond major argument for maintaining thefull suite of biodiversity. Many species haveknown uses, including survival value in diffi-cult times. But the value of a plant, such asits medicinal value, cannot be known aheadof time. This knowledge is accessed duringthe healing process, as a healer may receivea vision during a dream in which a plant,or other being, offers itself. Hence one musthold an attitude of humility for the mysteriesof nature, as one never knows the real valueof a plant given by the Creator.

The Ojibwa practice of site-specific burn-ing, in combination with landscape-scalenatural fires, would increase the temporaldiversity of the boreal forest. The combinedoutcome is a landscape that is diverse spa-tially. Within that spatial diversity, there arehabitats at different temporal stages thatincrease the overall landscape biodiversity.Small disturbances that create habitat diver-sity may also help fireproof the landscape andincrease resilience by reducing the fuel loadon the forest floor, thus averting large distur-bances.

The first mechanism of conserving andenhancing biodiversity is to maintain all suc-cessional stages. Since each stage in forest suc-cession represents a unique community, having

forest patches at different successional stagesmaintains overall biodiversity. At the sametime, it contributes to the continued renewal ofecosystems by conserving the building blocks


for renewal and reorganization. The second isthe creation of what landscape ecologists mightcall patches, gaps, and mosaics. It is well knownin landscape ecology that low and intermediatelevels of disturbance often increase biodiver-sity as compared to nondisturbed areas. Borealhunters know this mechanism intuitively anduse it effectively, for example, through the cre-ation of moose yards by the use of fire (Lewisand Ferguson. 1988). The third is the creationof edges (ecotones). Edges exist in nature butnew edges can also be created by disturbance.Boundaries between ecological zones are com-monly characterized by high diversity, both eco-logically and culturally (Turner et al. 2003).

Detailed studies with one boreal group, theShoal Lake Ojibwa, show that their principle ofmaintaining the full suite of biodiversity is sim-ilar to the scientific practice of multifunctionallandscape management. But the approach isfrom a different angle. Ojibwa elders do notseek to connect biodiversity to known func-tional properties of a habitat or species, as sci-entists do. Rather, the elders believe that, sincethe Creator has provided everything that theOjibwa need to survive, there must be a rea-son for the existence of every plant, animal, andother beings (Davidson-Hunt et al. 2005). Themaintenance of these species and the health ofthe land on which they grow are part of theirethical responsibility.

Introduction of new values and approachesand global economic drivers have been impact-ing the boreal forest and creating new chal-lenges. Logging is the major management issuefacing the boreal forest of central Canada.Large-scale timber production makes littleallowance for the ecological processes that pro-duce berries and medicinal plants. Loggingoperations have been gradually moving north-ward on Ojibwa traditional lands. What deter-mines a “commercial forest” is the interna-tional price for pulp and paper products. Paperfrom the boreal forest is not used locally butis exported worldwide. Logging is only oneuse of the forest ecosystem but an increas-ingly dominant one. The economic impact offorestry dwarfs any monetary benefits fromwildlife, ecotourism, berries, and medicinalplants.

Can these forests be managed sustainably –in a way that conserves ecosystem processeswhile at the same time providing for timber pro-duction and other ecological services? Learn-ing from traditional management systems, suchas those of the Ojibwa, is important for broad-ening conservation objectives and approaches.The use of local and traditional ecologicalknowledge is an effective mechanism for theempowerment of indigenous communities sothey can participate in making the decisionsthat impact their resources and livelihoods.

Threatened Species andCommunity-Based Conservation

Threats to some of the world’s endangeredspecies are found in social–ecological systemsof the developing world. Elephants of eastAfrica, turtles of Costa Rica, blue sheep ofthe Himalayas, pandas in China, and otherthreatened species are commonly found wherehuman populations face severe poverty; thereare limited options for economic development,and in some cases ecosystem services havebeen dramatically reduced in quality (WRI2005). Historically, nongovernmental organiza-tions (NGOs) have taken a lead role in researchand international advocacy for the conservationof threatened species. Species such as elephantsand whales have become symbolic in fundraising for many of these organizations, withsome advancing the ideals of animal rights andthe prohibition of harvestings (Freeman andKreuter 1994). To meet the conservation chal-lenges of threatened species, protected areas,often in the form of national parks, have beenestablished throughout the developing world,some successful and others failing to achievetheir conservation objectives (Robinson andBennett 2000). In some of these cases, imposedprotected areas have exacerbated problems byrelocating residents from their homelands, cre-ating heightened conflicts among those seekingto conserve resources and potential communityresource stewards (Songorwa et al. 2000). Con-servation issues can be especially problematicadjacent to national parks where wildlife popu-lations are increasing and where local residents


view these species as pests that transform culti-vated lands (Getz et al. 1999).

The paradigm of community-based wildlifemanagement assumes that high levels of com-munity involvement in resource managementand explicit incentives for conservation encour-age stewardship behavior and protect threat-ened species from possible extinction. Duringthe past 20 years the World Bank has shiftedaway from supporting top-down arrangementsfor economic development and adopted thisnew approach. The related shift in fundingfor community-based programs is striking. In1996, US$125 million was loaned by the WorldBank for participatory environmental projects;by 2003, the total of loans had grown to $7 bil-lion (Mansuri and Rao 2004).

Community-based wildlife management isintended to be a bottom-up approach in whichcommunities benefit from the use of wildlifeand sustainable management. In some casesthese programs are implemented adjacent toprotected areas without provisions for con-tributing to livelihoods, but instead are focusedon reducing opposition to the protected areas(Songorwa et al. 2000). In other cases, com-munity livelihoods have been developed hand-in-hand with conservation initiatives and in

partnership with government agencies, safarihunting operators, and environmentally orien-tated NGOs. Ideally, these approaches shouldbe adaptive through ongoing cycles of experi-mentation, regular review, and modification.

The CAMPFIRE program of community-based management in Zimbabwe involving ele-phant conservation (Box 6.3) demonstratesboth the extraordinary opportunities of theseprograms and the significant difficulties in theirsuccessful long-term implementation. The caseof Zimbabwe (and with most other countries ofAfrica) illustrates how traditional resource-usesystems with long-standing biodiversity can beundermined through colonization and unregu-lated markets of animal products. The case ofCAMPFIRE shows how an innovative a pro-gram that is attentive to both conservation andlivelihoods can be initiated first as an exper-iment and later implemented to successfullyreduce conflicts and provide community ben-efits, and how it can be taken up as a modelof resource management by other countries inthe region. Yet the case also illustrates how aproject can face difficulties because of corrup-tion of regional leaders, and ultimately flounderbecause of the collapse of national-level institu-tions that support it.

Box 6.3. CAMPFIRE: Elephant Conservation and Community Livelihoods in SouthernAfrica

Sources: Freeman and Kreuter (1994);Berger (2001); Getz et al. (1999) Peterson(1994); Thomas (1994); Hulme and Mur-phree (2001); WRI (2005)

Because of their huge size, high foodintake, potential damage to farmlands andhuman infrastructure, and occasional aggres-sive behavior, elephants have significantimpacts on their social–ecological systems.Yet these large animals have coexisted forthousands of years with Africans without sig-nificant damage to one another (Freemanand Kreuter 1994). In the precolonial period,human populations were small, elephant har-vest levels were low, traditional institutions

for managing harvests were intact, and thehuman harvesting of elephants was presum-ably sustainable. The export of ivory forEuropean markets, starting in Napoleonictimes and intensifying with the rise of theglobal economy, changed the traditional rela-tionship between humans and elephants andbegan to threaten the future of the animals inmany parts of Africa.

At the time of Kenya’s independence in1963, the elephant population was estimatedat 170,000. However, by 1989 elephant num-bers in Kenya had been reduced to 16,000,with the decline attributed almost entirely topoaching for ivory. During that same time


the value of ivory went from $3 a poundto $300. Similar trends were seen in othercountries as well. For example in Zaire, theelephant population plummeted from about150,000 in 1989 to near extinction during thesame period. In 1977, the Kenyan Govern-ment, with pressure from developed countriesand nongovernmental conservation organiza-tions, put a ban on hunting. Several otherAfrican governments did the same. Colo-nial governments dismissed the knowledge ofindigenous people as related to conservationand neglected the potential benefits from non-preservationists’ forms of management.

In 1989, international concern for ele-phants increased and parties to the Conven-tion on International Trade in EndangeredSpecies (CITES) voted to ban all trade inAfrican elephant products. Elephants tookon a high symbolic role in developed coun-tries like the USA and the UK, embodyingenvironmental organizations’ rationale forpreservation. Animal rights advocates notedthe large brain size and social behavior of ele-phants as evidence of their high intelligence.Attempting to address the problem of ele-phant conservation during the 1970s to 1980s,many African countries established nationalparks and other forms of protected area thatrestricted all local uses of elephants. In somecases villages were relocated outside parks toaid in enforcement of illegal hunting. How-ever, in spite of the CITES designation andprotected areas with enforcement officers, anorganized illegal trade of elephant productsgrew, driven by black market buyers. Thesebuyers sold their contraband internationallyat high rates, little of the profit going toillegal hunters themselves. Although govern-ments spoke of their efforts to implementelephant conservation programs, such effortswere accompanied with high degrees of gov-ernment corruption. Since government offi-cials administered access rights to elephanthabitat, poaching and the ivory trade contin-ued and in many cases thrived.

Zimbabwe was the first country in Africato recognize conservation by utilization,which acknowledged that landowners shouldbenefit from wildlife. The transformation of

policies toward utilization began in the 1960sand was enacted in 1975 with legislation thatgave private commercial ranchers an eco-nomic rationale to conserve resources. Thepopulation of elephants in Zimbabwe wasnot near extinction, and hunting had histor-ically taken place on private reserves, withoperations owned by Africans who weredescendents of colonists and catered to for-eign big game hunters for considerable profit.In the late 1970s, government officials of var-ious departments and wildlife researchers,with local communities began experimentingwith the idea of approaching elephant con-servation by creating incentives that wouldbenefit local villagers and encourage vil-lage participation. In its initial phase it wascalled “Operation Windfall” in which ele-phant culling was linked to participation bycommunities. The rationale for the new pro-cess followed from the acknowledgementthat elephants on communal use areas causedhardship for communities, destroying crops,creating fear among residents, and in somecases killing people by trampling. By the late1970s the Zimbabwe Department of NationalParks and Wildlife Division initiated a 2-yearexperiment on elephant conservation underthe Communal Areas Management Programfor Indigenous Resources (CAMPFIRE).

At the time it started, CAMPFIRE wasunique in Africa in empowering local com-munities by distributing the benefits of tro-phy hunts to villages near game reserves, thusencouraging conservation. One of the imme-diate effects of the program was a dramaticincrease in cited violations for breaking hunt-ing regulations because community residentswere beginning to help enforce regulations tosafeguard their resource.

In one community, the local managementcommittee decided that benefits were to beused for community development projects,including construction of a new school and agrinding mill. As the program became estab-lished and the benefits became clear, it wasexpanded to other regions of the country.As a part of the program, villagers contribu-ted ecological monitoring data for researchsupported by The World Wildlife Fund. In


one region residents decided to relocate tocreate enhanced elephant-hunting opportu-nities that would bring the financial benefitsdirectly to residents.

Although CAMPFIRE was criticized bysome international NGOs for compromisingpure conservation objectives, it was toutedby many as a huge success of community-based conservation. The program was notlimited to elephants but extended to multi-ple wildlife species and safari hunting in gen-eral, distributing back to the communitiesrevenues as well as meat from the hunt. Theprogram did encounter problems associatedwith bureaucratic delays in delivering bene-fits to villages, the interception of funding byregional organizations, and inequities stem-ming from corrupt local leaders. It was alsobecoming clear that CAMPFIRE was run asa top-down program that did not effectivelydevolve the authority to manage wildlifebelow the district level of government. It alsodid not go far in devolving resource tenureon communal lands (Hulme and Murphree2001). Nevertheless, from the 1970s to about2000, CAMPFIRE was a model that inspiredcommunity-based wildlife-management pro-grams in many southern and eastern Africancountries.

Since about 2000, the repressive politicalregime in Zimbabwe has led to a loss of safaritourism and ecotourism and internationalNGO support. These political and economicchanges resulted in the collapse of manyinstitutions of Zimbabwe, and with it, mostof the tourism and safari industry that pro-vided the economic benefits of the CAMP-FIRE program. In other parts of southernand eastern Africa, including Namibia, Zam-bia, and Mozambique, community-based sys-tems originally inspired by CAMPFIRE wereestablished and continue to support con-servation while providing for livelihoods.For example, Namibia’s conservancies, estab-lished through the Nature Conservation Actof 1996, are considered to be among the mostsuccessful African efforts at community-based wildlife management, with significantdevolution of management power to localcommunities, and effective in poverty alle-viation (WRI 2005). Elephant conserva-tion is still a global problem. But in sev-eral of the countries of the sub-SaharanAfrica, elephant numbers have actuallyincreased, and the current problem has iron-ically shifted from a concern about too fewelephants to the management problem oftoo many.

Community-based wildlife management(like co-management described in the nextsection) is neither simple nor automatic. Intheir evaluation of seven community-basedwildlife management programs in Africa,Songorwa et al. (2000) identified insightsabout the potential for success and failureof these programs. These included problemsassociated with implementers not adoptinga bottom-up approach, lack of true partici-pation, and the inability of the program toaddress basic community needs and distributebenefits equitably. In other cases, the authorsfound a mismatch between the goals and theobjectives of NGOs that funded the programsand the livelihood goals of the community.An associated problem was the interruption

of funding from NGOs because of changesin the organization’s budget and/or mission.At the community level, issues can arisearound the difficulties of villagers havingto enforce their own rules on one anotherbut continue to live as neighbors. Othersarose when the job obligations of part-timewildlife enforcement officers competed withthe responsibilities of tending family farmsand caring for the welfare of kin. Economicincentives are also important. For example,Getz et al. (1999) note that programs likeCAMPFIRE are most likely to succeed wherethere are marginal opportunities for agricul-tural development and thus higher incentivesto conserve wildlife that might otherwise be ahardship.


Ecologically and socially, there have beena number of successes with community con-servation. In some countries, community-basedwildlife management has contributed to sus-tainable use and biodiversity conservation. Inmany cases, it has provided a mechanism fordecentralizing resource management, so thatthe people dependent on wildlife participate inmanagement decisions that affect their liveli-hoods. It has also led to the development ofcommunity-based monitoring systems, involv-ing the participation of local harvesters as mon-itors of wildlife, community empowerment, andan increased appreciation of the local and tradi-tional knowledge of people who live in the hin-terlands and interact daily with wildlife (Kofi-nas et al. 2003).

Although the track record of community-based resource management globally is notvery strong, we know that it can work whencommunities have partners who help thembuild their management capabilities and haveco-management arrangement with governmentagencies (Berkes 2004) Where these condi-tions are fulfilled, community-based manage-ment has led to new governance arrangementsthat improve the long-term prospects for bothpeople and biodiversity (WRI 2005).

Resource Co-management

The social systems of small, resource-basedcommunities are embedded within a broadersocial–ecological context. In today’s world, alllocal communities have some level of interac-tion with regional and national governments,NGOs, and neighboring resource users. Theglobal commons has no remote locations. Effec-tive linkages across scales of management thatfoster the evolution of local-level conservationstrategies are therefore critical to sustainabil-ity (Dietz et al. 2003). In most cases, exist-ing indigenous-state relationships constitute thelegacies of past and current colonialism, includ-ing imposed harvesting policies implementedwith little awareness of community livelihoods.These impositions can make the prospects ofcommunity-based conservation efforts difficult

to realize since state agencies typically resistdevolving power to local communities. Top-down imposed policies can also undermine tra-ditional institutions of community managementthat previously supported resource conserva-tion. In several cases around the world, indige-nous peoples of hinterland regions have negoti-ated self-government agreements with nationalgovernments that specify rights for use andself-management of important resources. Buteven in cases of negotiated self-governmentand comprehensive land claims, communitiesare embedded within broader legal institu-tions, creating a need for community rightsthat secures access to resources and systemsthat provide for cross-scale communication andaccountability.

In cases where communities share use ofcommons with other communities and thereare potential issues of resource scarcity, thereis a need to establish vertical (across lev-els of organization) and horizontal (acrossthe same level) linkages of resource gover-nance among resource user communities, stateagencies, and others. These arrangements fallwithin the rubric of co-management, definedas the sharing of power and responsibil-ity in decision-making between state govern-ments and communities in the functions ofresource management. Co-management func-tions may include monitoring, habitat protec-tion, impact assessment, research, enforcement,and policy making. When implemented with afocus on learning-by-doing, these arrangementsare referred to as adaptive co-management(see Chapter 4). Formal and informal co-management has proven critical in the devel-opment of strategies that support livelihoodsand conservation initiatives, contributing tosocial learning and social–ecological resilienceat local and regional scales. We give twoexamples.

In the American Southwest, the US NationalPark Service and Navaho Nation co-manageCanyon de Chelly National Monument, a pro-tected area that is situated entirely on NavajoTribal Trust Land and that celebrates the spiri-tual significance and traditional uses of the areafor the Navajo. The Park’s co-managers increas-ingly recognize that climate changes will alter


the unique ecosystem of the canyon, includingorange groves dating back to the time of Cortez,and therefore the ecosystem services to Navahoresidents.

In the American North, the US–CanadaPolar Bear Agreement established a linkagebetween the two countries and among AlaskanInupiat and Canadian Inuvialuit indigenouspeoples for conservation, helping to coordi-nate the setting of quotas and to facilitatediscussions about habitat issues. With climatechange now threatening the future of the South-west and certain Arctic species, the terms ofthese agreements and discussions among co-managers become increasingly critical in reduc-ing the risk of species loss. With the partner-

ships in place, all parties should be better sit-uated to address these challenges.

Co-management systems are difficult toimplement (Pinkerton 1989a, Wilson et al.2003), even without the stresses of social–ecological change. Because state governmentsare reluctant to give up authority in resourcemanagement to communities and nationalinterests typically take precedence overlocal interests, communities involved in co-management may have to be creative and fullyengaged to be heard in highly charged politicalprocesses. Such a process is illustrated in Box6.4 on caribou co-management involving theGwich’in people of Alaska, Yukon, and theNorthwest Territories.

Box 6.4. The Case of Porcupine Caribou Herd Co-management Across an InternationalBorder

Sources: Kofinas (1998, 2005); Kofinas et al.(2007); Griffith et al. (2002).

The Gwich’in, indigenous people of north-western Canada and northeastern Alaska,have depended on caribou of the US–Canada Borderlands (Alaska, Yukon, andthe Northwest Territories) for millennia. TheGwich’in’s close relationship with the Por-cupine caribou, an internationally migratoryherd of that region, is central to Gwich’inlivelihoods, identity, worldview, and sense ofcultural survival. Like many of the barrenground caribou herds of the CircumpolarNorth, the Porcupine herd is a commons oftens of thousands of animals that overwinterin boreal forests and migrate north to coastalplain tundra for calving and summer insectrelief. Scientific research shows that calvinggrounds are important habitat for caribou,with a strong relationship between tundraforage quality at calving grounds and cariboureproductive success.

The traditional hunting practices of theGwich’in have changed dramatically since1900, although aspects of the relationshipbetween Gwich’in and caribou have per-

sisted. In the precontact period, caribouwere harvested in the fall by family groups,often using an extensive system of cariboufences or corrals into which migrating cari-bou were directed and snared. Caribou werethen butchered, dried, and stored for the win-ter by family groups. Strict rules dictated themanagement of waste to ensure that cap-ture areas were clean and without odorsthat would disturb subsequent caribou drives.Decisions about hunting were directed byleaders and informed by traditional knowl-edge, which provides an intergenerationalunderstanding of annual herd movements,decadal patterns of shifting winter habitat,the effects of severe climatic events, indica-tors of body condition and animal health, andthe implications of human disturbance to ani-mals and hunters. From a Gwich’in perspec-tive, people’s livelihoods and the well-beingof caribou are interrelated and inseparable.

With the introduction of firearms, thesocial organization of harvest was trans-formed, with the eventual abandonmentof caribou fences. Group hunting withleadership remained important until the


introduction of the snowmobile, which madehunting a more individualized activity. Inspite of changes in hunting organization andtechnology, the sharing of harvested meatremains a central element of the Gwich’incontemporary subsistence-cash economy.The “lucky hunter” (i.e., successful) is theone who shares with fellow community mem-bers, is not boastful, and is respectful of landand animals.

Beginning in the 1930s, and more inten-sively from the 1960s to present, national andinternational interest in mineral and oil andgas resources of the region has been a con-cern for the Gwich’in because of its poten-tial impacts on caribou. In the 1950s theCanadian Government constructed a high-way through the winter grounds of the Por-cupine herd, with no formal consultation withlocal communities of the region. US federaland state government efforts to open the“northern frontier” to industrial developmentin the herd’s calving grounds of Alaska in1980 led to reestablishing kinship ties betweenCanada and the US Gwich’in communities toaddress threats to caribou. The early issuesof industrial development also initiated a20-year negotiation process with governmentsthat led to the signing of a Canadian Por-cupine caribou co-management agreement in1985 and a US–Canada international agree-ment for the conservation of Porcupine Cari-bou in 1987. These agreements and theirmanagement boards were intended to pro-vide local-to-international linkages for jointdecision-making about caribou and impor-tant caribou habitat, to ensure the ongo-ing availability of caribou for the Gwich’inand other indigenous peoples of the region.

US–Canada bilateral activities of theseco-management arrangements have beenlargely dominated by the issue of proposedoil and gas development in concentratedcalving areas of the herd, located on thecoastal plain of the Arctic National WildlifeRefuge in Alaska. The international agree-ment for Porcupine Caribou conservationhas been ineffective in addressing oil and gasissues because of political vagaries of US fed-eral administrations, which have included not

selecting an Alaskan Gwich’in user to theinternational caribou conservation board,canceling international board meetings, andsuspending international dialogue on habitatissues.

In response, the Porcupine Caribou Man-agement Board of Canada has worked withits user communities using alternative meth-ods to address the shortcomings of theinternational agreement, by initiating itsown assessments on sensitive caribou habi-tat and distributing the products of thatresearch directly to members of the USCongress. The Canadian caribou board alsodeveloped a relationship with the CanadianConsulate in Washington to monitor polit-ical activities and update local communi-ties on legislative proposals and to com-municate local concerns to the CanadianForeign Affairs Ministry. The Canadian co-management board also raised private fundsto facilitate the grassroots political campaignof Porcupine caribou user community mem-bers, which included training indigenous peo-ple to lobby in Washington and at public pre-sentations across the USA. A central mes-sage of these effective campaigns has notbeen caribou science, but the mutual respectof people and wildlife of the North and therelationship between indigenous livelihoodsand caribou conservation. Building on thisexperience, Gwich’in First Nations formeda political organization, Gwich’in Interna-tional, which was later given observer sta-tus at the Arctic Council, an internationalgovernance body of state officials, indige-nous peoples, and NGOs. Habitat protectionfor the calving grounds in Alaska remainsa critical issue and continues to commandthe attention of the Gwich’in. Through theefforts of the Gwich’in and Canadian Porcu-pine Caribou co-management, the issue of oildevelopment and caribou has also becomewidely recognized across the USA andinternationally.

Co-management of Porcupine Caribouat the regional level has been effective atachieving consensus on a range of diffi-cult issues and resolving them. For exam-ple, barter and trade policies have been


established to provide legal backing for tra-ditional systems of food sharing betweenfamilies in Canada and Alaska to addressexport policies that created barriers for foodexchanges. An offer by Asian buyers to pur-chase antlers for medicinal uses resulted infear among local communities of wastageand disrespect for caribou. Eventually, awell-supported voluntary prohibition on rawantler sales (i.e., not art products) was for-mulated and endorsed by the Canadian Por-cupine Caribou Management Board and thecommunities. Managing and achieving con-sensus on rules for hunting on Canada’sDempster Highway has been more difficult.Repeated efforts to implement rules for man-aging hunting from the highway have beenunsuccessful because of intercommunity con-flict and legal challenges, even when theserules were based on traditional knowledge.More successfully, co-managers, communi-ties, and agencies have together developed a

regional ecological monitoring program thatdraws on local knowledge, as well as science-based indicators. It has also created a strate-gic planning process that provides adaptivelearning through the use of computer simula-tion models that inform policy choices, plan-ning, experimentation, and reflection.

At the community level, local lead-ers have had to confront imposed legalregimes and negotiate new ones, learn tomaneuver through and around bureaucraticinstitutions, while selectively maintainingand adapting traditional institutions in arapidly changing social–ecological setting.Co-management of wildlife involving mul-tiple communities, state agencies, interna-tional transactions is a difficult, often messyprocess. It is not a panacea. It does, how-ever, provide an important mechanism ofmultiscale governance for building resilienceand sustaining community livelihoods andresources.

Among the parties of a co-managementarrangement, power relations are typicallyasymmetrical; that is, there rarely is a level play-ing field. As well, cultural differences amonggroups can create communication problems andlimit understanding of each other’s perspec-tives (Morrow and Hensel 1992, Kofinas 2005).The differences in approaches to knowledgeproduction and legitimacy can also add to thedifficulties (Berkes 2008). Yet many examplesdemonstrate that power sharing can yield bet-ter decisions, enhance adaptation, and con-tribute to social–ecological resilience (Pinker-ton and Weinstein 1995, Armitage et al. 2007).There are many ways in which co-managementcan contribute to community livelihoods andsocial–ecological resilience:

• creating a regional forum for deliberationson resource-management issues;

• maintaining collaborative and systematicsocial and ecological monitoring;

• focusing research that draws on local, tradi-tional, and scientific knowledge;

• evaluating sensitive habitat, protectingimportant habitat, and providing for com-munity participation in impact assessments;

• developing strategic resource-managementplans that are sensitive to community inter-ests in economic development;

• overseeing appropriate policies for barterand trade of subsistence resources;

• guiding effective response polices for ecosys-tem disturbance, such as forest fires;

• building and using decision-support tools;• achieving regional consensus and good com-

pliance with co-management endorsed poli-cies.

Case studies show that improved levels ofsocial capital (i.e., trust) can emerge throughthe security of specified legal managementrights for communities and or the ongoing inter-actions among leaders, resource professionals,and resource harvesters (Wilson and Raakjaer2003). Factors that are particularly critical inensuring the effectiveness of co-managementfor communities are that monitoring, research,and policy making be truly collaborative


enterprises and resources rights be secure(Pinkerton 2003). In conditions of rapid changeand novel problems, effective systems of cross-scale governance may prove critical in arriv-ing at innovative solutions (Kofinas et al. 2007)and ultimately in supporting community andregional adaptation (Armitage et al. 2007).

Single-Species Management of WildResources

Single-species management of wild resources(IUCN Category IV; Box 6.1) often focuses onecological controls rather than institutions thatlink conservation to community livelihoods.The use of local ecosystems as a source offood for personal consumption and as a con-nection to the land is not unique to hinterlands.People living in a broad range of rural to semi-urban environments pick berries, harvest mush-rooms, catch fish, and hunt wildlife as cultur-ally and nutritionally significant activities thatlink their livelihoods to local ecosystems. Theseactivities are often responsive to formal andinformal institutions that foster the conserva-tion of these resources. For those wild resourcessuch as berries and mushrooms in which har-vest has little detectable effect on the futuresupply, local and traditional institutions tend toaddress access and use rights rather than theactual levels of harvest. In some countries, forexample, rights to pick berries may be restrictedto property owners. In other countries, particu-larly on public lands, families that have pickedberries, harvested mushrooms, or caught fishin a particular location may return regularlyto the same places, giving rise to informal, butpredictable patterns of use. In still other cases,these resources may be treated as open-accessresources to which all people have access.

The harvest of fish and wildlife is oftenregulated through additional formal (govern-ment) institutions because human overharvestcan significantly reduce future availability. Thisformal management tends to focus on themaximum sustained production of particularspecies that are compatible with available habi-tat, natural mortality, and harvest rate. Thissingle-species command-and-control approach

to wildlife management (see Chapter 4) restson the dubious assumption that other causes ofvariations in population density are unimpor-tant or can be managed.

Because hunting pressure is usually moresocially than biologically mediated, wildlifemanagers are challenged to manipulate hunt-ing in ways that roughly mimic natural preda-tion pressure, i.e., to increase hunting pressurewhen target populations increase and to reduceit when populations decline. In cases wherepotential harvest pressure may exceed the mor-tality rate that the population can support, man-agement options might include shortened (orclosed) seasons for hunting or fishing, reducedallowable harvest levels (e.g., through rafflesor lottery for the right to hunt a particularspecies), catch-and-release fishing, etc. Alterna-tively, when population densities of the targetspecies exceed the carrying capacity of the habi-tat, management options include increases inallowable harvest, focused harvest on females(which have a disproportionate effect on pop-ulation growth rate), and economic incentivesfor harvest (e.g., bounty on predator species).Efforts to maintain a specific population target(maximum sustained yield) neglect the impor-tance of wildlife harvest for community liveli-hoods. An alternative approach is to managefor resilience (see Chapter 2). Resilience-basedwildlife management accepts a broader range ofvariability in wildlife populations as a normalstate of the social–ecological system and placesgreater attention on community livelihoods asan integral component of the human–wildlifesystem. Many of these practices are similar toindigenous resource management described inprevious sections. For example, habitat man-agement that maintains a habitat mosaic ofmultiple successional stages is a cornerstoneof wildlife management, just as in traditionalresource management of the Ojibwa (Box 6.2).

Management of human harvest is challeng-ing because of both biological and social uncer-tainties and societal pressures to manage popu-lations in ways that deviate substantially fromthose that occur naturally. For example, rarespecies or individuals that are viewed as partic-ularly desirable to maintain in the population(trophy individuals) are subject to more intense


hunting pressure than they would likely expe-rience under natural predation or harvest thatresponded to food needs.

Management is equally challenging when tar-get populations become extremely dense. Deerin suburban areas of the northeastern USA,for example, are abundant because people haveeliminated most potential predators, have notfunctioned as predators themselves, and cre-ated an abundance of favorable early succes-sional habitat. Under these circumstances deerpopulations explode, become prone to wildlifediseases, and are more effective vectors of lymedisease (Ostfeld and Keesing 2000). The result-ing disease risks have the perverse consequenceof disconnecting people from nature by reduc-ing human use of natural areas.

Synthesis and Conclusions

The conservation and maintenance of biodi-versity at various levels are clearly social–ecological, rather than strictly biological, issues.Human actions are the principal causes ofrecent declines in biodiversity, but they mayalso be important pathways to potentially sus-tainable solutions. The establishment of pro-tected areas is an important measure. But pro-tected areas alone are not sufficient to meetthe challenges of biodiversity conservation. Weneed to work with people, encouraging stew-ardship ethics and cultural connections to theland, to protect biodiversity everywhere, andnot create artificial “islands” of conservationthrough protected areas.

The “fence-and-fine” approach to conserva-tion that seeks to separate people from naturetypically creates highly artificial, fragile systemsthat have little historical precedent and areunlikely to be resilient to future environmental,economic, and social changes. The case studiespresented in this chapter show that the integra-tion of livelihoods into a biodiversity conserva-tion ethic has been an integral part of many tra-ditional cultures. We need to support such rela-tionships, build on them, and promote them asa component of innovative solutions to currentconservation challenges.

Developing conservation programs ina social–ecological context is not easybecause the inherent complexities of human–environment interactions, issues of power andpolitics, and uncertainties of future humanbehavior make precise planning impossible.However, these same complexities may beused to generate resilience that enables thesystem to adapt to unknown future changes.Active participation of local resource users inthe design, implementation, and monitoring ofsocial–ecological conservation efforts improvesunderstanding, responsiveness to change, andcompliance with rules. Under these circum-stances, local engagement and innovation canbecome a source of resilience, by seekingsolutions to emerging problems, as opposed toa conservation system that has inflexible rulesand that seeks to exclude people.

Conservation that builds livelihoods for localpeople in ways that are compatible with con-servation objectives seems crucial. Unless localpeople can meet their basic livelihood needs,they may have little choice but to conflictwith conservation objectives. So the solutionsoften involve working with local communitiesin such a way that they can use some of theresources under a conservation scheme, cre-ating a local stake in biodiversity protection.Although engagement of local people in con-servation efforts does not always lead to desir-able outcomes, generalizations are emerging asto the conditions that influence their likelihoodof success or failure (Songorwa et al. 2000,Borgerhoff Mulder and Coppolillo 2005, WRI2005, Berkes 2007).

In developed countries, the management ofharvestable fish and wildlife also benefits froma social–ecological perspective that recognizesthe importance of social determinants of har-vest patterns. Hunting pressure, for example,may tend to increase for species that declineor become uncommon, particularly in areaswhere human densities are highest. This is pre-cisely opposite to the natural predator–preyfeedbacks that convey resilience to wildlifepopulations in less human-dominated land-scapes. Under these circumstances, engage-ment of local communities in understanding theproblems and seeking solutions to conservation


challenges has the greatest likelihood of suc-cessful implementation.

In all cases, formal and informal rules affect-ing conservation of biodiversity and support ofpeople’s livelihoods will need to be evaluatedand adapted, in some cases in subtle ways, andin other cases more radically. These adaptationsmay involve the renewal of traditional ways ofrelating to and managing resources. In othercases they may require a grand transformationof formal institutions to provide for better part-nerships among communities, resource man-agement agencies, and nongovernment organi-zations.

Most current conservation challenges haveroots in processes occurring at many scales,including global climatic and economic changes,national and regional agendas, and local tradi-tions and needs. Resilience-based approachesto conservation recognize the importance ofscale, the inherent uncertainty of future con-ditions, and the need for multilevel manage-ment, as seen in the caribou conservationcase. Resilience-based stewardship thereforerequires the fostering of flexibility and innova-tion in a social–ecological context.

Review Questions

1. In comparing the conditions of the threecase studies (Boxes 6.2, 6.3, and 6.4), identifycommon themes in conservation planning.

2. Can a conservation ethic be created in a soci-ety with little or no tradition of conserva-tion?

3. What are some of the strategies that commu-nity leaders might use to ensure that com-munity needs are seriously considered inthe establishment of a protected area thatincludes community homelands?

4. What are the strengths and weaknesses ofa national-park vs a community-based con-servation approach in conditions of rapidsocial–ecological change? Debate the propo-sition that the priority for our biodiversityconservation efforts should be to establishpeople-free protected areas.

5. In what social–ecological conditions is eachof the six classes of protected areas mostappropriate?

6. What are the key institutions in each of thethree case studies in the boxes? Are theylocal/traditional, governmental, or both?

Additional Readings

Anderson, M.K. 2005. Tending the Wild: NativeAmerican Knowledge and the Management of Cal-ifornia’s Natural Resources. University of Califor-nia Press, Berkeley.

Bawa, K.S., R. Seidler, and P.H. Raven 2004. Recon-ciling conservation paradigms. Conservation Biol-ogy 18:859–860.

Berkes, F. 2008. Sacred Ecology. 2nd Edition.Routledge, New York.

Mulder, M.B. and P. Coppolillo. 2005. Conservation:Linking Ecology, Economics, and Culture. Prince-ton University Press, Princeton.

Brechin, S.R., P.R. Wilshusen, C.L. Fortwangler andP.C. West (eds). 2003. Contested Nature: Promot-ing International Biodiversity with Social Justice inthe Twenty-first Century. State University of NewYork Press, Albany.

Gomez-Pampa, A. and A. Kaus 1992. Taming thewilderness myth. BioScience 42:271–279.

Posey, D.A. (ed). 1999. Cultural and Spiritual Valuesof Biodiversity. UNEP and Intermediate Technol-ogy Publications, Nairobi.

Taylor, B.R. (ed). 2005. Encyclopedia of Religion andNature. Thoemmes Continuum, London.

Turner, N.J. 2005. The Earth’s Blanket. TraditionalTeachings for Sustainable Living. Douglas &McIntyre, Vancouver; and University of Washing-ton Press, Seattle.

7Forest Systems: Livingwith Long-Term Change

Frederick J. Swanson and F. Stuart Chapin, III

Introduction

Human societies depend heavily on forestsfor resources and habitat in many parts ofthe globe, yet the global extent of forests hasdeclined by 40% through human actions sincepeople first began clearing lands for agriculture.Forests still cover about 30% of the terrestrialsurface and support 70% of Earth’s plant andanimal species (Shvidenko et al. 2005). Soci-eties residing in tropical forests alone accountfor half of the world’s indigenous languages.People have lived in and interacted with forestsfor thousands of years, both benefiting fromtheir services and influencing the dynamics ofthe forests in which they live. Forests are there-fore most logically viewed as social–ecologicalsystems of which people are an integral com-ponent. Forests are globally important becauseof their broad geographic extent and the greatwealth and diversity of ecological services they

F.J. Swanson (�)Pacific Northwest Research Station, USDA ForestService, Corvallis, OR 97331, USAe-mail: [email protected]

provide to global and local residents, includingsubstantial biological and cultural diversity.

The character of forests and societal interac-tions with them vary greatly across the globe(Plate 5, Fig. 7.1). Belts of dense tropical rain-forests encircle the earth in the equatorialzone, bordered at higher latitudes of Africaand South America by extensive savannas ofscattered trees. Boreal forests stretch acrosshigh, northern latitudes. Conifer or broadleafand mixed forests are common in intermedi-ate latitudes, especially in the northern hemi-sphere. Individual forest patches may be man-aged by a household, local community, smallor large corporation, or local or national gov-ernment agency. The social settings of forestsrange from urban forests within cities and smallsacred forests adjacent to villages to extensivetracts of forest in remote wilderness areas. Theportfolio of forest uses and the relative extentof human–forest engagement vary greatly fromcountry to country, reflecting the complex his-tories of forestlands and societies.

Rapid global changes in biophysical andsocial factors (see Chapter 1) are particularlychallenging to forest stewardship because manytrees live longer than the professional lifetimesof people who manage them, making it chal-lenging to develop and sustain institutions with


150 F.J. Swanson and F.S. Chapin

Figure 7.1. Patch clear-cutting that leads to plan-tation forests of native Douglas-fir in a matrixof 100- to 500-year native forest in the north-western USA. Photograph by A. Levno, US For-est Service, photograph AEA-002 [online] URL:http://www.fsl.orst.edu/lter/data/cd˙pics/cd˙lists. cfm?topnav=116

time horizons that extend beyond the short-term motivations of individual decision mak-ers. In this chapter we highlight four overlap-ping resource–stewardship challenges faced byforest managers throughout the world that are,in part, logical consequences of the long-livednature of forest trees:

• Sustaining forest productivity. Institutionswith a long-term view of the ecologi-cal conditions necessary to support forestsand forest-dependent peoples are a foun-dation for sustainable harvest from forests.Given harvest intervals that are frequently40–100 years or longer, large areas arerequired for forests containing the spectrumof age classes needed for a continuous flowof wood for human use. The large tempo-ral and spatial scales required for sustainableforestry often clash with the short-term moti-vations and small-scale jurisdiction of forestmanagers. In order to meet these sustainabil-ity challenges it is critical to sustain socialvalues and capacities for dealing with forestsystems over long time periods.

• Sustaining forest ecosystem services. Man-aging forests for multiple ecosystem ser-vices involves strong tradeoffs among costsand benefits to different stakeholders, withchoices having implications for multiplehuman generations. Forests provide a mul-

titude of ecosystem services, including fuel-wood, timber products, water supply, recre-ation, species conservation, and aestheticand spiritual values. Together these servicesgenerate a diverse and challenging mix ofmanagement objectives to meet multiplesocietal needs.

• Sustaining forests in the face of environmen-tal change. Managing forests under condi-tions of rapid change is challenging becausea forest stand is likely to encounter novelenvironmental and socioeconomic condi-tions during the life of the individual trees inthe stand. Forest ecosystems across the globeface myriad threats from both intentionaland inadvertent human impacts, includingair pollution, invasive species, and, perhapsabove all, global climate change.

• Sustaining the forestland base. Forest con-version to new land uses is a state change thatis difficult and time-consuming to reverse,given the long regeneration time of foresttrees and ecosystems. Historic and ongo-ing land use has converted vast areas ofcomplex, native forests to agriculture, builtenvironments, and plantations of a single ornarrowly constrained set of species, oftenexotics. Under other conditions, large-scaleagricultural abandonment or increased eco-nomic value of forests, as for carbon seques-tration, can foster reforestation or afforesta-tion (the regeneration of forest on recentlyharvested sites or planting of new forests onpreviously nonforested sites, respectively).

We next address the importance of social–ecological dynamics in forest planning. Wethen consider each of the above resource–stewardship issues in greater detail, showingboth the challenges and the opportunities forsustainable forest stewardship. We concludewith some recommendations for sustainablestewardship of forests in the future.

Forests as Social–EcologicalSystems

Because of the longevity of forest trees andforest crops, decisions made today for for-est use and management have inevitable con-

7 Forest Systems 151

sequences for future generations operatingin different environmental and social con-texts. Societies often face a broad spectrum offorest-management objectives and associatedapproaches, ranging from maximum wood pro-duction with relatively short (decadal) rotationsto development of old forest attributes on themulticentury time scale to simple retention ofnative forest and the species contained thereinfor conservation, recreation, spiritual, or otherobjectives. Furthermore, shifting legal, regula-tory, and economic contexts of forestry cancause gradual or disruptive, abrupt changes inmanagement, so social disturbances can be asimportant as ecological disturbances in shapingthe future of forest resources.

This complexity of the societal context forforest management is confounded by a webof forces operating at local-to-global scales.Human population growth and sprawl ofrural residential development into forest areasincrease local demand for forest products whileconstraining the range of potential forest man-agement tools, such as prescribed fire, that maybe critical to sustaining the properties of nativeforests. Social forces may shift the emphasisof management objectives within the portfo-lio of ecological services provided by publicor private land, such as from wood productionto protection of species, with various intendedand unintended consequences for society. Thus,social factors can trigger abrupt and profoundchanges in forest policy and management thatcan ripple across scales (see Chapter 5). Global-ization of commerce, for example, may connecta market place in Europe to forestry operationsin a distant part of the globe through forest cer-tification and international agreements againstillegal logging. Forest certification is a pro-cedure for assessing forest-management prac-tices against standards for sustainability so pur-chasers can support sustainable management,even over vast distances.

Changing societal and environmental factorsare colliding and interacting. Global changesin climate, climate variability, and species inva-sions, much of them human-driven, make itincreasingly difficult to ensure a predictableflow of desired quantities and diversity of ser-vices from forests. These global changes haveimpacts at local to regional scales, sometimes

causing landowners to modify forests in expec-tation of climate-change effects not yet real-ized on the land. In some cases these thresh-old changes in system condition and capacitymay be influenced by legacies of past manage-ment that affect the organization (e.g., age classdistribution of forest trees or agency cultureof past management) and vulnerability of theecosystem to disturbance. The resulting uncer-tainties place a premium on social capital andadaptive capacity.

Supply and Use of EcosystemServices

Forests provide many important ecosystem ser-vices to society both globally and locally overa range of time scales (see Chapter 2). At theglobal scale, the human conversion of foreststo other land-cover types in the last two cen-turies has had important effects on the climatesystem (Field et al. 2007). For example, glacial–interglacial changes in forest cover in responseto small changes in solar radiation contributedto massive shifts in global climate (Foley et al.1994, Friedlingstein et al. 1995). As home to70% of Earth’s biodiversity, forests are impor-tant sources of ecosystem services ranging fromfood and medicines to cultural appreciation oftheir spiritual and aesthetic values. In additionto these global services, forests are home tohuman communities, whose local use includesharvest of goods; regulation of water and nat-ural hazards; and recreation and cultural tiesto the land. Different segments of society oftenplace different priorities on these services, rais-ing challenging questions about tradeoffs andsynergies.

Forests are obviously the major source oftimber and its products, including lumber,veneer, and paper. Despite the importanceof lumber and pulp products, more than half(55%) of global wood consumption is for fuel-wood, which is used locally; is not well charac-terized in economic summaries of forest prod-ucts; and is the primary energy source forheating and cooking for 40% of the global pop-ulation, particularly in developing nations andrural areas (Sampson et al. 2005). Nontimber


forest products also have considerable eco-nomic and cultural importance. In the north-eastern USA and Canada, maple syrup, pro-duced from the sap of local trees, providesincome to rural residents and contributes tolocal identity. In Alaska the economic valueof blueberries and moose harvested by localresidents each exceeds the value of harvestedtimber. About 25% of currently prescribedmedicines originate from plant compounds thatevolved as defenses against herbivores andpathogens (Dirzo and Raven 2003). Conse-quently, property rights to the genetic diversityof tropical forests are an issue that is contestedbetween tropical nations, local indigenous peo-ples, and commercial bioprospecting firms(Kursar et al. 2006). The harvest and saleof bush meat to city residents in Africa isalso a controversial topic because it createsboth a source of local income and an extinc-tion threat to many of the species that areharvested.

In some cases the products provided byforests are part of an agricultural rotation(slash-and-burn or swidden agriculture), inwhich small forest patches are cut, the landis cultivated, and forests regenerate, as peo-ple move to clear adjacent forest patches (seeChapter 12). This rotation has been sustainablefor millennia, but increasing human populationdensity is reducing the length of the forest rota-tion; below a rotation length of about 10 years,slash-and-burn agriculture no longer appearssustainable, causing a transformation to contin-uous cultivation of cash crops (Ramakrishnan1992).

Many forestlands are managed for the crit-ical ecosystem services of water supply andwatershed protection (Rockstrom et al. 1999,Vorosmarty et al. 2005). Forested watershedsaccount for more than 75% of the world’s acces-sible freshwater (Shvidenko et al. 2005). Giventhe increasing scarcity of freshwater relative tohuman demands (see Chapters 2 and 9), forestsare likely to become increasingly importantfor their capacity to provide and purify fresh-water. Water quality and quantity also affectother resources, such as fish, so an ecosystemapproach is required for sustainable manage-ment. An important aspect of forest water-

shed management is therefore to meet waterquality objectives by minimizing soil erosion,which also sustains soils as a base for terres-trial productivity, limits sediment accumulationin reservoirs, and prevents damage to down-stream freshwater and marine aquatic habi-tats (see Chapters 2, 9, and 11). Watershed-focused forest management can also reduce thepotential for landslides and snow avalanches,thereby protecting life and property on thehillslopes and valley floor below. Appropri-ate actions include distributing forest harvestand roads to avoid unstable areas, creatingbuffer strips of forest along streams, and adjust-ing the frequency, spatial pattern, and inten-sity of management actions to minimize theirimpact.

Forests and forestry affect the global car-bon budget in ways that can either increaseor decrease CO2 concentration in the atmo-sphere, hence global warming. Forests accountfor about two thirds of terrestrial net primaryproduction and half of the terrestrial carbonstocks (Shvidenko et al. 2005). Any increases inforest extent or a positive growth response toincreasing atmospheric CO2 or nitrogen depo-sition removes CO2 from the atmosphere andreduces the potential climatic impact of fossil-fuel emissions (see Chapter 2). This capacity offorests to sequester carbon appears to be satu-rating (Canadell et al. 2007), indicating that wecannot count on forests to “solve the problem”of climate warming without more concertedefforts to reduce fossil-fuel emissions. Althoughincreased forest extent sequesters more carbon,it also reduces albedo (short wave reflectance),especially in northern forests during the snow-covered seasons. The reduced albedo leads togreater absorption of solar energy and moreheating of the atmosphere. In the tropics, theeffects of carbon storage and cooling of the sur-face by high transpiration rates predominateover energy-exchange effects, so any reductionin deforestation or increase in forest regenera-tion tends to reduce the rate of climate warm-ing. At high latitudes, the tradeoff is less cer-tain; the greater atmospheric heating by forestsdue to their low albedo reduces the benefitsof carbon sequestration, with the net effect onclimate currently uncertain (Field et al. 2007,


Chapin et al. 2008). Therefore efforts to slowthe rates of deforestation, the conversion of for-est to a nonforested ecosystem type such asagriculture, are important to the global climatesystem, especially in the tropics.

Forests also provide important regulatoryservices at more local scales (see Chapter 2).Tropical forest pollinators, for example, are crit-ical to coffee plantations, which show greatestfruit set and productivity adjacent to forests andforest fragments (Ricketts et al. 2004). Simi-larly, forests often harbor insect predators thatreduce the likelihood of insect pest outbreaks(Naylor and Ehrlich 1997). As forests arecleared to support small-scale agriculture, peo-ple and their domestic cattle become exposedto diseases from forest animals. In this way,forest fragmentation reduces the capacity offorests to regulate diseases (Patz et al. 2005).In the northeastern USA forest disturbance andelimination of predators has led to dense deerpopulations that spread lyme disease, whichin turn reduces human use of forests (seeChapter 6).

Cultural and spiritual values of forests havetaken different forms in different parts of theworld. The Druids, for example, held certaintrees to be sacred deities in their Celtic reli-gion. Patches of sacred forest, some only afraction of a hectare, are an important part oflife in many developing nations, including India(Ramakrishnan 1992), Madagascar (Elmqvistet al. 2007), and Benin and Togo (Kokouand Sokpon 2006; see Chapter 6). Forestsalso provide economically important culturalservices such as recreation and tourism (seeChapter 2).

The existence of forests as reservoirs of bio-diversity is important to many people and soci-eties. Tropical forests alone house 50–90% ofEarth’s terrestrial species. Forested mountainlandscapes are especially rich because the com-plex topography and steep environmental gra-dients create great habitat diversity. In someparts of the world large tracts of forestland havebeen reserved specifically for conservation ofbiological diversity, and rules for managementof these reserves may, or may not, allow humanuse of other potential ecosystem services (seeChapter 6).

Sustainable Timber Harvest:A Single-Resource Approachto Forest Management

Evolving Views of People in Forests

Despite the huge ecological and cultural vari-ations among forests as social–ecological sys-tems, changes over time in social–ecologicalforest systems exhibit some striking parallels.Societal engagement with forests in Australia,Canada, and the USA, which share Euro-pean cultural roots, for example, have exhib-ited a similar sequence of stages: “(1) useby hunter-gatherer societies, (2) exploitive col-onization, settlement and commercialization,(3) wood resource protection, (4) multiple usemanagement, (5) sustainable forest manage-ment or ecosystem management” (Lane andMcDonald 2002), and finally (6) sustainableecosystem stewardship (see Fig. 15.1). Transi-tions from one stage to the next have oftenoccurred through a crisis triggered by biologi-cal forces (e.g., extensive insect outbreaks) orsocial events (e.g., law suits and court injunc-tions), or development programs sponsored bynonlocal agencies with interest in moderniz-ing “primitive people” (Gunderson et al. 1995,Lane and McDonald 2002). These transforma-tions can be important opportunities for inno-vation, because the crisis demonstrates that thecurrent system is not working, making man-agers and the public willing to consider newalternatives (see Chapter 5). Although gov-ernment controls on forestland and terminol-ogy for management systems may differ amongcountries, this general trajectory and the punc-tuated pattern of change have been quite simi-lar among countries and biomes that range fromdrylands and freshwaters to cities and agricul-ture (see Chapters 8, 9, 12, 13, and 15).

However, not all societies engage withforests in this sequence. Developing coun-tries, for example, sometimes find a path thatdraws on the ingenuity and knowledge oflocal people to meet pressing demands forboth poverty alleviation and forest steward-ship, providing a potential seedbed for new for-est stewardship approaches that would bene-


fit Western countries. However, indigenous andrural communities often face extremely diffi-cult challenges in their efforts to use local for-est resources because of governmental controlsand corruption (Larson and Ribot 2007). Insome very remote forests, however, popula-tion density may be limited by the capacity offorests to provide food, either through small-scale clearing and recovery of forests or throughsubsistence harvest of foods from the forests.These interactions limit both the human pop-ulation that can be supported and the extentof forest harvest. Forests were harvested pri-marily to meet short-term, local human needsfor resources provided by the forest ratherthan for commercial sale of wood products (seeChapters 6 and 12).

Expansion of agriculture and commercialtrading of forest products led to more exten-sive forest clearing. This began in Europe inthe Middle Ages and in eastern North Amer-ica with European colonization and still charac-terizes many tropical and boreal regions. For-est exploitation is largely driven by demandfor wood and depends on availability oflabor, access, and markets in social contextswith limited local governance. Under condi-tions of illegal and even some corporate- andgovernment-backed forest exploitation, littleattention might be paid to whether forestpractices are sustainable (Burton et al. 2003).For example, in the developing world forestexploitation is often fostered by unclear prop-erty rights, disempowerment of local institu-tions for managing common property, or gov-ernment efforts to generate foreign exchangefrom wood exports (see Chapter 4). Underthese circumstances, any unfavorable effects offorest harvest have relatively few consequencesfor those who make the decisions about theextent and nature of forest harvest.

As exploitation depletes forests, the impor-tance of forest regeneration becomes moreapparent, and there is a gradual transition tomaximum sustained yield (MSY) of timber,rather than short-term profit as a guide for for-est management. MSY sets a harvest level thatdoes not exceed the expected annual growthincrement (see Chapter 2). This leads to a har-vest rotation system, in which forests are man-

aged much like an agricultural crop. In this con-text, the value of the forests reflects both sup-ply and demand. MSY is clearly motivated byefforts to manage sustainably, but with a nar-row focus on wood supply and much less atten-tion to other ecological services or to forestresilience to unexpected changes. Managing forMSY is challenging with long-lived species liketrees and with uncertain variation in climate,pests, other disturbances, markets, and taxa-tion systems. In many cases, estimates of futureyield have been overly optimistic and proba-bilities of environmental consequences and dis-turbance have been underestimated, leadingto harvest schedules that proved unsustainableover the long term. For example, climate warm-ing may increase drought stress and the risksof fire and insect outbreaks to an extent thatwas not anticipated when plans for high lev-els of wood production were developed. Short-rotation production forestry on plantations isstill guided by MSY, which provides a reason-able guide to sustainable timber production, ifchanges in slow variables such as soil fertil-ity, disturbance, and pest populations are takeninto account. Production forestry predominateswhere land is owned by forest companies oron public lands where wood production forshort-term revenues is the primary objective formanagement. Even where wood production isthe explicit management objective, silviculturalpractices have been developed that enhance thedelivery of other ecosystem services to society,as described below.

The shortcomings of MSY and public sen-timent that public forests should serve morethan industrial forestry interests pushed pol-icy for public forestlands to embrace multi-ple use management that explicitly addressesa broad array of ecosystem services. In the1990s growing global interest in species conser-vation, maintaining site productivity, and otherecological services, led to development of anecosystem-management approach that empha-sizes the well-being of the system as a whole,while capitalizing on natural ecological pro-cesses to do the work we wish to accomplish inthe forest (Grumbine 1994). Even in the con-text of ecosystem management, crises may arisefrom influences such as extensive insect damage


to forests or altered public opinion, resultingin environmental policy changes and new envi-ronmental regulations. Over the course of thesechanges in views and approaches to forest man-agement, several trends are evident:

• Broader geographic and temporal scales areconsidered.

• More components of forests, hence moreecosystem services, are included as primarymanagement objectives.

• Therefore, a greater variety of technical dis-ciplines is engaged in planning and imple-menting forest management, and generalprofessional oversight shifts from silvicultur-ists and foresters to broadly interdisciplinaryteams.

• In some cases adaptive management isimplemented to monitor change, adapt plansbased on new information, and, thereby,learn through doing.

• Research-management and science-policyties are broadened and strengthened.

Current conditions of rapid environmen-tal or social change have given rise to addi-tional challenges that require an expandedvision of ecosystem management, which weterm ecosystem stewardship (see Chapters 1and 15). In this context, managers recognizethat change is inevitable, although the natureand rate of change are generally uncertain.Under these circumstances, managers seek torespond to and shape changes and to be veryjudicious in identifying historical properties

of the system to attempt to sustain into thefuture.

Despite these common trends, quite differ-ent forest landscape-management approachesare often evident even within a single forestregion. In the Pacific Northwest (PNW) USA,for example, forestry practices range from anagricultural model (MSY) of 40-year cuttingrotation on industrial lands to an 80- to 100-year rotation on government lands managedfor timber production to wilderness, park, andother lands where no cutting occurs (Box 7.1).In some other countries, by contrast, laws des-ignate narrowly prescriptive forestry manage-ment rules on most public and private lands.In yet other contexts ranging from communi-ties in remote areas of developing countriesto western government forestry, exploration ofalternative future scenarios has been used toset a desirable future course of forest stew-ardship (Wollenberg et al. 2000). These con-trasts suggest that there is no single “right”approach: different governmental jurisdictions,ownerships, and land allocations have differ-ent management objectives, hence manage-ment paradigms, approaches, and supportingscience (see Chapter 4). Policy diversity andadaptive, learning social systems are sourcesof institutional and ecological resilience for anunknown future. Even if a “best policy” couldbe identified for today’s conditions, other poli-cies might prove more favorable as an uncer-tain future unfolds (Bormann and Kiester 2004)(Fig. 7.2).

Box 7.1. Blending Forest Production and Conservation in the Western USA

Development of forest issues and policies inthe PNW of the USA (Fig. 7.2) over thepast few decades presents dramatic exam-ples of the challenge of conducting ecolog-ically and socially sustainable forest man-agement. The conflicting values people holdof PNW forests have been starkly framed—cut majestic, centuries-old forests or sustainlocal economies and families who for severalgenerations have worked in the woods; savea cryptic owl species and iconic salmon or

intensively manage highly productive forest-lands for wood products that benefit a broadcross-section of society. Of course, the realissues are not such simple dichotomies, andsocietal complexities match or exceed eco-logical complexities. We consider this exam-ple of a region’s quest for sustainable forestryin terms of the expanding breadth of multi-ple use objectives, the great diversity of liveli-hoods affected, and successes and failures inadaptive management.


Conflict over these forests grew out ofthree decades of intensive timber productionon Federal forestlands commencing afterWorld War II to supply wood for the postwarhousing boom. Intensive timber harvest dur-ing this period focused on old growth (>200years old) and cumulatively affected about30% of the landscape by the late 1980s. Overthe timber production era local communi-ties grew dependent on livelihoods basedon jobs in the woods and in the mills. Fed-eral forestry agencies flourished by puttinglarge volumes of wood into the market-place, and the applied science communitystudied ways to enhance timber productiv-ity and efficiency of logging systems. Thispattern of natural resource system develop-ment follows the paradox framed by Holling(1995, p. 8): “The very success in manag-ing a target variable for sustained productionof food or fiber apparently leads inevitablyto an ultimate pathology of less resilienceand more vulnerable ecosystems, more rigidand unresponsive management agencies, andmore dependent societies”—until some eco-logical or social disturbance triggers abruptchange.

That disturbance erupted as intense con-troversy in the late 1980s over the fateof old-growth forests, salmon, and north-ern spotted owl. Law suits hinged on pro-tecting the northern spotted owl, “listed”under the Endangered Species Act, stoppedall logging of forests in the 100,000 km2

range of that species, extending from north-ern California to the Canadian border. Tobreak the gridlock created by the resultingcourt injunctions that blocked timber cut-ting, President Clinton convened scientists toconduct a bioregional assessment (FEMAT1993, Szaro et al. 1999) and craft a newplan with objectives of protecting speciesof critical interest and forming an intercon-nected network of old-growth forest reserves,while providing some flow of timber to localcommunities.

The resulting Federal lands policy, theNorthwest Forest Plan (NWFP; USDA andUSDI 1994), was a revolutionary depar-ture from previous forest management in

the region, and set an example of ecosys-tem management with broader impact. TheNWFP greatly expanded the scope of eco-logical considerations under the multiple userubric. For example, hundreds of specieswere designated for special attention underprotocols called “survey and manage.” Thegeography of planning under the NWFPwas based more on ecosystem considera-tions than political jurisdictions, spanningmany Federal agency lands and aligned withwatershed boundaries. The plan placed 80%of the land in reserves for terrestrial andaquatic species and reduced the timber har-vest level by more than 80% of the levelof the 1980s. Harvest of old-growth forestoutside reserves was to provide a signifi-cant share of timber volume in the earlydecades of NWFP implementation. Ecolog-ical considerations strongly influenced theharvesting systems. For example, 15% coverof live trees and substantial amounts of dead-wood were to be retained to meet ecologicalobjectives in harvest units that would havebeen clear-cut in earlier logging systems. TheNWFP incorporated adaptive managementat several scales, including designation of tenAdaptive Management Areas (AMAs) cov-ering 7% of the plan area. The NWFP alsocommissioned a region-wide monitoring pro-gram for change in forest cover, northernspotted owls, socioeconomic conditions, andother attributes, to support adaptations of theplan at the regional scale.

How is the NWFP working? Resultsof a regional monitoring program docu-ment successes and failures over the firstdecade of plan implementation (Hayneset al. 2006). Dire predictions of collapse ofrural communities did not occur, althoughmore isolated communities highly depen-dent of Federal timber did suffer and liveli-hoods dependent on jobs in the woods andmills declined markedly. A long-standingsystem of payment of timber revenues tocounties in lieu of property taxes on Fed-eral lands has collapsed, leaving countieswith extensive Federal timber land with-out funds which they had depended on forroads, libraries, and schools. Federal timber


harvest has not reached even the greatlyreduced level projected by the NWFPbecause of procedural challenges by envi-ronmental organizations, limited funding offorestry agencies, and other issues. The har-vest of old growth was limited because ofstrong public opposition. The regional mon-itoring program revealed that the extent ofold-growth habitat increased because for-est growth exceeded losses to harvest andwildfire. Ironically, although suitable habi-tat increased, spotted owl populations con-tinued to decline, perhaps in part due tocompetition from the more aggressive barredowl expanding its range from the north andeast. Adaptive management proceeded intwo ways: (1) a regional monitoring program(Haynes et al. 2006) that reported findingslikely to influence future plan revision and(2) the AMA program in which scientists,land managers, and in some cases local pub-lic groups undertook studies to test assump-tions in the NWFP and explore other man-agement approaches. However, the regionalmonitoring program faces uncertain fund-ing, and the AMA program faced substan-tial institutional barriers to success (Stankeyet al. 2003). Before the AMAs were a decadeold most funding had disappeared, and thecommitment to this learning process largelyevaporated.

What does the future hold? Federal forestsin the PNW will doubtless present profoundsurprises, as they have in the recent past.Three factors are in play—the forests, theirsocial context, and environmental change. Isthe present NWFP harvest level so low thatit is socially unsustainable? There has beenno great public clamor to increase logging inthe Federal forests of the region, but it maydevelop for social (e.g., more revenue to localcommunities) or ecological (e.g., reductionof vegetation that could fuel fires) reasons.However, changes in the global market placeand other social factors may someday trig-ger increased forest harvest. Will the barredowl displace spotted owls, reducing the legalmotivation to sustain old-growth habitat?This now seems possible, but the spotted owlis no longer critical to protection of nativeforests, especially old growth. The publicseems thoroughly committed to protectionof the remaining old-growth forests. Climatewarming, invasive species, and changing fireregimes are creating a very uncertain futurefor forest management. In short, the forcesof change, both environmental and social,have great potential to trigger new convul-sions. The region remains caught in Holling’sparadox—in part because of limited successin developing sustained learning institutions(Holling 1995).

Forest Exploitation and Illegal Logging

Illegal logging dominates the timber produc-tion of some developing nations and accountsfor up to 15% of timber production glob-ally (Contreras-Hermosilla 2002, Sampson et al.2005). More than half (50–80%) of timber pro-duction in Indonesia, Brazil (in 1998), andCameroon, for example, is estimated to occurthrough illegal logging (Sampson et al. 2005),although precise estimates are seldom avail-able. Illegal logging includes logging in pro-tected areas, without authorization, or morethan authorized; timber theft and smuggling;and fraudulent pricing and accounting prac-tices (Sampson et al. 2005). It can focus on

high-value tropical woods, depleting local diver-sity, or extensive forest clearing, causing landdegradation. Illegal logging deprives govern-ments and local communities of forest rev-enues, often strengthens criminal enterprises,and induces corruption among enforcementand other officials (Contreras-Hermosilla 2002,Sampson et al. 2005). It is analogous (and isa similar proportion of total harvest) to ille-gal and unreported fishing (see Chapter 10). Inthe former Soviet Union (FSU), illegal loggingwas estimated to account for about 20% of tim-ber production, but this practice largely disap-peared, when the collapse of the FSU elimi-nated subsidies for transportation from foreststo processing facilities (Sampson et al. 2005).


Figure 7.2. Shaded relief map of western Washing-ton and Oregon and northwest California showingforested areas of coastal mountains and the CascadeRange and the outline of NWFP area covering therange of the northern spotted owl. Information fromUSDA and USDI (1994).

Just as in the forest exploitation phasesof Europe and North America, illegal log-ging is driven largely by market value, access,and labor, with no regard for regenera-tion or sustainability. It often proliferatesin response to multiple local, national, andglobal circumstances, therefore requiring mul-tipronged responses. Commonly contributingfactors include:

• Disempowerment of local institutions thatprovide for sustainable forest management.These could include a weakening of prop-erty rights or weakening of some of the fac-tors that facilitate effective stewardship ofcommon pool resources (Dietz et al. 2003,Agrawal et al. 2008; see Chapter 4).

• Political corruption that prevents local indi-viduals or communities from benefiting fromthe timber sales or that leads to national poli-cies with other social goals (e.g., forest clear-ing for oil palm plantations in Indonesia ormeat production in Brazil) that subvert for-est sustainability.

• High market prices that increase the bene-fits from illegal logging relative to the socialcosts and political risks.

Actions that might reduce the likelihoodof illegal logging include the building of localsocial capital and livelihood options; bridginginstitutions at local-to-national scales; clarifica-tion of property rights; international aid that istied to changes in the incentive structure forillegal logging; or global forestry certificationprograms that encourage sustainable manage-ment. The engagement of local forest users indecisions about rules for management of pro-tected areas generally results in greater com-pliance and oversight of others than in caseswhere rules are imposed by higher authori-ties (Nagendra 2007, Agrawal et al. 2008). Atthe scale of international trade, on the otherhand, European purchasers of South Americanwood would like to know that it came fromforests managed sustainably. Forest certifica-tion currently covers a negligible proportion oftimber production, primarily in the temperatezone, with variable verification, but ultimatelycould substantially reduce forest exploitation,especially for high-value timber species with


specialized markets. International accords onforest conservation provide potential avenuesto address both illegal logging and forestcertification.

Timber Production from NaturalForests

Natural forests are forests that have regener-ated naturally, in contrast to production forestsin which trees are planted, often in regularlyspaced patterns of a single species (Sampsonet al. 2005). Logging of natural forests accountsfor most (65%) of global timber productionand is a significant source of local employmentand livelihoods. In this chapter we use tim-ber synonymously with industrial roundwood,which includes sawlogs and pulpwood (and theresulting chips, particles, and wood residues).Throughout the world, individuals, families, orlocal enterprises harvest most timber, provid-ing both local income and cultural connec-tions to forestlands. Although timber produc-tion continues to increase globally in responseto global increases in consumption, the rate ofthis increase is slowing and is largely met byincreases in production forestry. Until recently,the growth in timber production was approxi-mately balanced by increased labor productiv-ity (1.45% annually in the USA, for example),as labor-saving technologies and infrastructureexpand, leaving a roughly constant employmentin forest production (Sampson et al. 2005).This apparent stability hides huge national andregional shifts in timber harvest. For exam-ple, harvest declined radically in the FSU dueto economic collapse and in the western USAdue to shifts in forest policy (Box 7.1). Con-versely, harvest from natural forests is increas-ing in many tropical countries, in part throughillegal logging. Given the long-lived nature offorest trees, uncertain environmental change,and shifting markets, how can natural forestsbe managed more sustainably? The answerappears to be context-specific and differs sub-stantially between developing and developednations.

Rural communities in many developingnations depend strongly on the products of

nearby forests, especially in the tropics andsubtropics (Bawa et al. 2004). In some ofthese countries, such as Costa Rica, Mexico,and Nepal, large-scale reforestation is occurringeither coincident with or following deforesta-tion. In other countries, such as China, defor-estation is the predominant trend (Nagendra2007). Even within a country the balancebetween forest cutting and reforestation can beregionally variable. In Nepal, for example, allforests are nationally owned but are manageddifferently, depending on the land-tenure sta-tus of local residents. In national forests andparks that attempt to exclude local residentsto meet conservation goals, deforestation is thepredominant trend as a result of illegal log-ging, whereas areas that are managed by com-munities or by individual leaseholders show netreforestation (Fig. 7.3; Nagendra 2007). Thesetrends toward reforestation are strongest wherelocal residents participate in monitoring thepatterns of forest use. These patterns of com-munity forestry are more sustainable underconditions of clearly specified land-use rightsand active monitoring, consistent with researchon successful management of commons (seeChapter 4). Community engagement does notalways lead to sustainable forestry, however, inpart because several factors create significantmisfits of institutional and ecological condi-tions (Brown 2003). Factors that can underminesustainable forestry include inability of localresidents to exclude nonresidents from forestharvest (e.g., illegal logging) or to participatemeaningfully in defining the rules or resolvingconflicts related to forest use (see Chapter 4).Globally there is a trend toward decentraliza-tion of forest-management authority (Agrawalet al. 2008), providing opportunities for con-certed efforts to strengthen community-basedmanagement.

Sustained timber production from naturalforests involves both the first cutting of nativeforest and the subsequent cuttings of foreststhat regenerate after the initial harvest. If refor-estation is managed to sustain a near-naturalmix of native species; biotic structures, suchas standing and down deadwood; and ecolog-ical processes, such as a natural fire regime,successive forest rotations may provide many


Reforestation

Deforestation

For

est c

hang

e

Nationalforest

Communityforest

Leasehold forest

Tenure regime

0

Figure 7.3. Relative extent of deforestation andreforestation of government-owned forests in Nepal.National forests are managed by the government,with little community engagement or protection,resulting in open-access use by people living nearby.Community forests are managed by local communi-

ties through community forestry user groups. Lease-hold forests are small patches of forest that are gen-erally highly degraded and available to households inregions of high poverty. Leaseholders often replanttheir own forests and harvest from nearby nationalforests. Adapted from Nagendra (2007).

of the same services as the initial forest. Inecological terms, this forestry system contrastssharply with management of plantations ofexotic species. In some parts of the world firstcutting of natural forests remains the domi-nant practice—in Siberia, for example, whereold-growth forests are extensive and forestregrowth is slow, and parts of the tropics, whereecological, social, and economic factors haveconstrained reforestation. Even in places wherenative forests were converted to agricultureand then monocultures of native or exotic treespecies, such as forest plantations on publiclands in the Southeastern USA, some forestsare being managed for more natural biologicalconditions of species and processes combinedwith the intent of maintaining a flow of forestproducts from the land. In these cases forestryhas both wood production and forest restora-tion objectives.

A common theme in forestry in natural sys-tems is the need to understand the ecology ofthe system. This is the gist of ecosystem man-agement (Grumbine 1994). What limits andcapacities does the native ecosystem impose?How can native ecosystem processes help usefficiently work toward our management objec-tives? For example, many native ecosystemscontain plant species with the capacity to fix

nitrogen, which can contribute to soil fertil-ity. Intensive plantation forestry may precludethese nitrogen-fixing species by eliminating thestages of forest development in which theyprosper. More natural and diverse ecosystemsmay help retain the roles of nitrogen-fixingspecies in the forest system, resulting in greater,sustained site productivity.

The practice of sustained production fromnatural forests therefore requires develop-ment of tightly integrated ecological and socialdimensions of forestry practices. Adaptive man-agement (see Chapters 4 and 10; Walters 1986)provides a context for bridging the contrastingworkplace cultures of science and management.Scientists like to ask questions and challengeideas; land managers tend to work together andseek public license to operate by using “bestmanagement practices.” But, if scientists, forestmanagers, and communities can work togetherin an adaptive management framework, it maybe possible to develop management systemsthat are scientifically credible, socially accept-able, and adaptive to changing social and envi-ronmental circumstances (Box 7.2). Some gov-ernment agencies, such as Forestry Tasmaniaand the US Forest Service have both manage-ment and research branches, which have thepotential benefit of cultural proximity, but the


disadvantage of too much familiarity and colle-giality, which may limit objectivity and criticalanalysis. Another research-management part-nership model is the alliance of state forestrywith academia, such as extension programs inthe USA for forestry and agriculture. However,these tend to be simple consultations ratherthan sustained programs of adaptive manage-

ment. In very limited instances the forest indus-try has had in-house research programs thatmay set the stage for a sustained adaptive man-agement program. A critical issue in all suchcases is the institutional capacity to sustainadaptive management programs over the longtime spans appropriate to trees, cropping rota-tions, and institutional change.

Box 7.2. The Many Roles of Scientists in Management and Policy

Scientists involved in forestry stewardshipare among the many people whose liveli-hoods are deeply engaged with forestsystems—such as loggers, heavy equipmentoperators, mill workers, forestry and regula-tory agency staff, and environmentalists, toname but a few (see Chapter 3). Amongthese groups scientists have a unique anddifficult challenge in their need to producenew knowledge, interpret monitoring andresearch findings, and in many cases com-ment on policy alternatives.

To display some of the complexity of typesof livelihoods and other engagements withforests, we briefly examine scientists’ roles inshaping, supporting, and even attacking for-est management and policy (Swanson 2004).It is useful to view these roles in the contextof stages of forestry management and pol-icy paradigms in the panarchy framework ofGunderson and Holling (2002), who envisionnatural resource management as a series ofcross-scale, interacting adaptive cycles punc-tuated by periodic crises (see Chapter 1).However, some roles of scientists are contin-uous, such as conduct of basic science, despiteperiodic crises and regardless of the manage-ment approach of the day. The role of sci-entists can be broadly grouped as operatinginside the system in support of the prevailingmanagement paradigm and those operatingoutside as shadow networks that attempt toinfluence decision making through the pro-duction of knowledge and critique of pro-posed management and policy options.

Recent forestry conflicts reveal the quitevaried roles of scientists over the course of

development of a resource exploitation man-agement paradigm, through the crisis of itsdemise, and into establishment of the next setof policies (Box 7.1, Fig. 7.4). These roles maytake the forms of:

1. Exploitation phase

• “Applied scientists” working in sup-port of the prevailing paradigm study-ing methods to enhance productivityand management efficiency. Informa-tion exchange is termed “technologytransfer,” implying one-way flow ofinformation in the form of technology.Government and industry are likely tofund this science work.

• “Canary in the mine” scientists iden-tify species or other ecosystem compo-nents or processes that are perceived asimperiled by the management paradigmof the day. This is a risky, public, andgenerally unfunded role.

2. Conflict phase

• Scientists may conduct bioregionalassessments of the history and cur-rent conditions of the ecosystems andresources in question to set the stagefor resolution of conflict (Szaro et al.1999).

• Scientists may participate in develop-ment of broad-scale management plansthat set new policy intended to resolvepast conflict. Bioregional assessmentsmay be used as a foundation for the newplan.


Outsideattacker

(canary in the mine)

Bioregional assessment Plan development

C R I I S S

Outside expert

Exploitation management

Adaptive ecosystem management

Maximize productoutput and efficiency

Diverse portfolio ofecosystem services

Applied scientist

Interdisciplinary science team

Predominant in-house scientist roles

Management goals

On-going scientist roles

Basic science, expert opinion (e.g., blue-ribbon committee)

Figure 7.4. Schematic representation of the variedand changing roles of scientists during a transitionfrom a predominantly exploitation phase of forestmanagement through a conflict phase to a postcon-flict ecosystem-management phase. This paradigmshift involves important changes in the roles of in-house (agency or industry) scientists from leadershipby applied scientists, whose goal is to improve pro-ductivity and economic efficiency, to interdisciplinaryteams of basic and applied scientists who may havepivotal roles in bioregional assessments and devel-

opment of new management plans. These scientiststogether with land managers may assume respon-sibility to design and implement adaptive manage-ment. Scientists outside of agencies also play keyroles—most frequently as outside attackers in theexploitation phase or as outside experts in the post-conflict phase. Ongoing roles for scientists at alltimes include conducting basic science and serving assources of expert opinion (e.g., blue-ribbon commit-tees). Adapted from Swanson (2004).

3. Postconflict phase

• “Adaptive management roles” have sci-entists working in support of the newmanagement paradigm. In current ter-minology that may mean participationin adaptive management programs withscientists and land managers working inclose partnership to develop new man-agement approaches that have scien-tific and operational credibility. This isquite different than the old “technol-ogy transfer” model of scientist andmanager relations. The adaptive man-agement relationship between scien-tists and managers can feature two-wayexchange of views and shared learning.A strong base of adaptive managementcan set a stage so that when the pol-icy window opens, innovative, groundedapproaches are ready to be spread via ashift in policy (see Chapter 5).

4. Science roles at any time

• “Long-term basic research” is a contin-uing role of scientists, whose findingsmay be useful in policy and manage-ment at some later point.

• “Outside attackers” criticize the pre-vailing management paradigm and maywork for interest groups or with per-sonal motivations. Attackers may oper-ate at any stage of policy and manage-ment, but are especially significant intriggering crises and bringing down anexisting paradigm. Attackers of a newmanagement approach may come fromthe ranks of those who benefited fromthe earlier management paradigm.

• “Episodic venues” for science input topolicy makers occur via participation inprocesses such as “blue ribbon” com-missions on special topics intending toinform the public and the policies.


• “The isolated individual” is a single sci-entist or small group with limited sup-port operating in areas with little sci-ence infrastructure, including develop-ing countries. In this case an individualmay play many of the above roles to tryto improve stewardship of forests andother natural resources.

Scientists engaging in natural resourceissues, especially leading up to, through,and shortly after a major policy convulsion,can find professional life very challenging.Education programs generally do not trainstudents for the social and political turbu-lence that can come with these roles. Manydilemmas about professional conduct arise inthe context of public policy disputes. Somescientists play multiple roles during theircareers or even simultaneously, such as gov-ernment scientists who may be a dispassion-ate provider of science-based information inthat government role, but work clandestinelyon behalf of an interest group in the offhours. Scientists confront many tough per-sonal choices, such as advocate aggressivelyfor a policy outcome in the immediate cri-sis or hold back to maintain a reputationof objectivity, hoping this to be an asset infuture crises.

Thus, scientists have many decisions tomake about their conduct. Even advocat-ing use of science in natural resource deci-sion making may be a contentious propo-sition in some circumstances. Young scien-

tists in particular have a great deal at stake—credibility with the public and peer groups,continued and future employment prospects,even self-esteem. Attitudes about appropri-ate roles of scientists differ among scien-tists, land managers of science information,interest groups, and the informed public,although these groups do endorse the roleof scientists to help integrate their knowl-edge with other information (Lach et al.2003). Attitudes about appropriate roles ofscientists vary from one societal context toanother (Lach et al. 2003). Decision mak-ing and policy advocacy by scientists is moreaccepted in Europe than in the USA, forexample; and in the USA these actions maybe more accepted for academics than for gov-ernment scientists, who are viewed as sourcesof objective scientific information to deci-sion makers, who may be elected officials ortheir appointees possessing little or no scien-tific background. Therefore, decisions aboutprofessional conduct as a scientist are verydependent on institutional context, potentialimpacts on career, and the environment inboth the near term and the long term. Thisissue has been debated at length (Fischer2000, Weber and Word 2001), especially inthe aftermath of highly disputed cases. Uni-versity programs in natural resources andrelated fields should provide training in val-ues, professional conduct, and environmentalethics, but ultimately the resolution rests withthe scientist or natural resource professional.

Production Forestry

Intensive production forestry on plantationsaccounts for an increasing proportion of theworld’s timber supply, about 35% in 2000, withthe proportion expected to increase to about44% by 2020. As in the case of intensive agricul-ture (see Chapter 12), an important benefit ofproduction forestry is that a substantial propor-tion of the global demand for timber can be meton a relatively small land base, reducing pres-sures for harvest of natural forests that have

high conservation, watershed, and recreationalvalues. Production forests currently account foronly 5% of forestlands, but 35% of timberproduction. Production forests are intensivelymanaged, regularly spaced stands, often mono-cultures of Pinus or Eucalyptus where theyare exotics. They are managed with agricul-tural approaches that maximize the efficiency ofwood production.

Although management practices vary geo-graphically, production forests can providemany important ecosystem services in addition


to the wood for which they are explicitly man-aged. Water and energy exchange, for exam-ple, may not differ substantially between pro-duction and natural forests. Production forestshave high rates of carbon gain, but their rolein carbon sequestration depends on the size ofdead wood carbon pools left on the forest siteand the lifetime of wood products from theseforests. Wood used in construction, for exam-ple, may have as long a lifetime as a solid formof carbon as it would in the forest, whereaspaper products might decompose and returntheir carbon to the atmosphere quite rapidly.Old-growth forests, in contrast, have rates ofproduction that are only slightly higher thanrates of decomposition and therefore sequestercarbon relatively slowly (Schulze et al. 2000).These forests have high standing crops of car-bon, including in the dead wood component,which gets released to the atmosphere uponcutting and conversion to intensively managedyoung forests (Harmon et al. 1990).

Sustainable management practices that max-imize both forest production and other ecosys-tem services generally also sustain slow vari-ables, including the retention of soils and theircapacity to supply water and nutrients, provi-sioning of high-quality water in streams andgroundwater, and many of the aesthetic andcultural benefits that foster public support forforests and forestry (see Chapter 2). Produc-tion forests can provide employment for localresidents, enabling people to remain in ruralareas and sustain their cultural connections tothe land. However, not all production forestsare managed in an environmentally sound way.As with intensive agriculture, some productionforests receive large applications of fertilizersand pesticides, with similar detrimental conse-quences for the environment and local commu-nities (see Chapters 9 and 12).

The most consistent losses of ecosystem ser-vices with production forestry are associatedwith reduced biodiversity. Production forestsare typically single-aged, single-species standsthat maximize the efficiency of planting, thin-ning, and harvesting of trees, but lack the struc-tural and biotic diversity of natural forests. Inregions dominated by production forestry orplantations of trees such as oil palms, much of

the regional diversity has been lost, especiallythose species that require the structural com-plexity, long periods for dispersal, or decayingwood typical of old-growth forests.

Restoring biological diversity within highlymanaged landscapes is one of the greatest chal-lenges faced by production forestry, but in somecases it can be accomplished to a significantextent without greatly sacrificing economic effi-ciency. The most simple and cost-effective wayto restore diversity is to facilitate processesthat naturally generate and maintain ecosys-tem and landscape diversity. These processesinclude periodic disturbance, such as fire—perhaps approximated by harvest patterns thatcreate habitat heterogeneity within and amongstands; maintenance of landscape patterns thatprovide corridors for spread of native speciesand barriers to spread of early successionalexotic species; and protection of rare habitatsto which some species are restricted (Hannahet al. 2002, Chapin et al. 2007, Millar et al.2007). In general, if a diversity of habitatsand environments is created and sustained,the appropriate organisms will find and exploitthem, if they are locally available.

Within-stand heterogeneity and diversity canbe increased by silvicultural practices such asretention of some live trees, dead standingtrees, and decaying wood at the time of har-vest, and protection of key habitats and bufferstrips between managed stands (Larsson andDanell 2001, Raivio et al. 2001). A spectrumof tree ages also fosters diversity, becausewood-decaying fungi, insects, and bird preda-tors found in late-successional forests are quitedifferent than species found in younger stands(Chapin and Danell 2001). A flexible rota-tion schedule can augment stand-age diversity.Allowing some trees to die and produce deadwood is critical to maintaining trophic diversity.As coarse woody debris becomes more avail-able on the landscape, it will likely be colonizedby wood-decaying taxa. In some forestry cul-tures, dead wood was removed to protect croptrees—to remove logging slash to reduce firedanger and ease planting, and to remove habi-tat for pest organisms that might attack livecrop trees (e.g., removal of dead wood underforest hygiene laws in parts of Europe). In a


manner similar to shifting attitudes about fire—from enemy to potential collaborator—forestmanagers have come to see dead wood as inte-gral to ecosystem management rather than athreat to the forest. In summary, a spectrumof forest-management approaches exist fromhighly “agricultural” forest plantations to plan-tations that are managed with substantial struc-tural and biological diversity to natural foreststhat are managed largely for timber productionto natural forests that are never harvested. Thisblurs the distinction between natural and plan-tation forests.

Economic and cultural diversity can be justas important as ecological diversity to sustain-able forestry, even in production forestry oper-ations. Economic subsidies to initiate new typesof businesses, such as the harvest of nontim-ber forest products (e.g., berries, guided hunts,medicinal herbs, edible fungi, greens for flo-ral displays), the production of local crafts,and promotion of recreation can generate awide variety of economic options, increasingthe likelihood of long-term economic vitality,regardless of the social and economic eventsthat may occur (Chapin et al. 2007). Alterna-tively, payment to local forest owners to pro-vide ecosystem services such as berries and wildgame could alter the incentive structure forforestry planning in rural areas. If local usersstrengthen their personal and cultural connec-tions to the land, they will have a stronger long-term commitment to sustaining its importantqualities.

Multiple Use Forestry: ManagingForests for Multiple EcosystemServices

Given the breadth of ecosystem services pro-vided by forests, there is a long history of effortsto balance exploitation of forests for wood pro-duction and their value in providing many otherservices. In some cases this balancing act hasbeen fraught with conflict (Box 7.1); in othercases societies have been very accepting of

tradeoffs to meet multiple objectives. Manage-ment of forests for multiple objectives was wellestablished long before the term “multiple use”gained formal definition in contexts such as theUS Multiple-Use Sustained-Yield Act of 1960.Many indigenous societies traditionally man-aged forests for multiple uses (see Chapter 6)and recognized that the gifts received fromforests (i.e., ecosystem services) entailed aresponsibility to manage the forests in ways thatsustained the capacity of forests to continueproviding these services. Community forestry,which has deep roots in indigenous cultures,has recently experienced a resurgence of inter-est in many societies, leading to establishmentof NGOs and governmental programs to pro-vide support (Gunter 2004). The case of mov-ing from the management of forests for tim-ber to multiple uses requires at least two steps:(1) recognition by forests managers (whetherthey are in a government agency, a company,a forest community, or a family) that forestsprovide multiple services with both synergiesand tradeoffs among the provisioning and useof these services and (2) fostering a sense ofcommunity responsibility to sustain the capac-ity to provide multiple services over the longterm.

During the 1990s public pressures and con-flicts forced forestry from an emphasis onresource extraction to a focus on stewardshipof the ecosystem as a whole. The resultingecosystem-management paradigm emphasizesforest practices for multiple services by capital-izing on and sustaining native ecosystem pro-cesses (Grumbine 1994, Szaro et al. 1999; seeChapter 2). As a consequence, in many jurisdic-tions the extent of forest reserved from cuttinghas been increased; the intensity and frequencyof cutting have declined; and historical distur-bance regimes have been increasingly used toguide future management.

Given the inevitable tradeoffs among ecosys-tem services, multiple use management is chal-lenging and benefits from regular engage-ment and open communication among multiplestakeholders. Sometimes acceptable solutionsare possible by managing different portions ofthe landscape using different but complemen-tary approaches.


Managing Forests UnderConditions of Rapid Change

Global Change

Rapid changes in environmental and socialdrivers of forest dynamics increase the uncer-tainties of how to provide long-term foreststewardship. Environmental changes occurringat local to global scales shape the forest envi-ronment in ways that alter options for sus-tainable forestry. For example, air pollution,such as that resulting in acid rain and nitrogendeposition, affects large regions, such as north-eastern North America and Eastern Europe,with net effects across the gradient of pollu-tion ranging from stimulation of tree growthby nitrogen addition to severe growth reduc-tion due to leaching of essential cations fromsoils (Driscoll et al. 2001). Resulting changes insoil properties, such as organic matter contentand pH, have effects that will likely last for cen-turies. Introductions of exotic species such asearthworms or forest pests have in some casesdirectly and radically altered entire ecosystems.For example, the spread of exotic insects (e.g.,defoliators, bark beetles) and pathogens (e.g.,root-rot fungi) can cause range-wide extirpa-tion or profound suppression of key forest trees,with impacts that ripple through the affectedecosystems (Ellison et al. 2005). Notable exam-ples in the USA include the chestnut blightin the early twentieth century and the hem-lock wooly adelgid and Sudden Oak DeathSyndrome that are current, acute managementconcerns. Perhaps the most daunting driver offorest change will come from global climatechange, which may cause migrations of speciesacross the landscape, reshaping forest commu-nities and making them vulnerable to pests,pathogens, and other disturbance agents atunprecedented scales. Some system responsesto climate change and/or land use history maybe gradual and others abrupt. Threshold sys-tem changes can have critical effects on for-est systems by mechanisms including pine bee-tle outbreaks, permafrost thaw, and altered dis-turbance regimes. Some threshold changes mayshift forests to other ecosystem types (Folke

et al. 2004), thus eliminating forest-specific eco-logical services. Not all forest changes are nega-tive. Abandonment of agricultural lands has ledto forest expansion in portions of Europe andeastern North America. Combinations of risingCO2 and modest nitrogen deposition can stimu-late forest production, which can have both pos-itive (e.g., carbon sequestration) and negativeeffects (e.g., reduced species diversity; Chapinet al. 2002) .

The world is now hyperlinked via near-universal, instantaneous communications andintricate economic networks. Public attentionto the well-being of forests has grown overrecent decades as a result of controversies overdeforestation in the tropics, global loss of bio-diversity, and protection of old-growth forestsin temperate regions. In this context grow-ing attention to sustainable forest stewardshipand combating illegal logging appear poised tomake substantial advances.

Disturbance Management

The paradigm for managing forest disturbanceshas changed dramatically in recent decades.Through much of the twentieth century “dis-turbances” were viewed as bad, so fires andinsect outbreaks were fought with military zeal.Training and practice of this dimension of forestmanagement was termed “forest protection.”When forests are viewed as a crop, disturbancesare a simple loss of capital and investment,so this single-minded management responsewas logical. Mid-twentieth-century practices,such as extensive use of persistent chemicalslike DDT, were gradually revealed to be verydamaging to the environment, so they werereplaced by pesticides more targeted to the“pest” species. But even these pesticides, aswell as native biocontrol agents like Bacillusthuringiensis, a bacterium that attacks the bloodof Lepidoptera, such as gypsy moth, raised chal-lenges because it can kill desirable, nonpest rel-atives of the target species. Thus, as the appre-ciation of ecosystem complexity grew, so toodid the difficulty of combating pests and otherdisturbances.


Furthermore, practical experience and eco-logical studies have shown that disturbances areintegral to the well-being of many types of for-est ecosystems and their provisioning of ecosys-tem services (Gunderson and Holling 2002).Attempts to suppress native disturbance pro-cesses have in some cases resulted in even moreexplosive disturbance events when they escapethat suppression—extremely intense fires mayfeed on an unnatural buildup of fuels; floodsthat breach dikes may create unusual havocwhen the dikes slow drainage of the inundatedarea; insects feeding on stressed, vulnerable for-est stands and landscapes may spread rapidly(Holling and Meffe 1996). Consequently, therehas been an accelerating shift from attempt-ing to exclude fire and other disturbance agentsfrom forests to adopting management systemsthat seek to retain roles of these processeswithin accepted limits. For example, frequent,low-intensity prescribed fire is now an impor-tant management tool in systems where groundfires occurred naturally (Schoennagel et al.2004). In some cases a legacy of fire sup-pression has permitted forest fuels to build tounnatural levels that pose the threat of stand-replacing fires in systems where that was notcharacteristic. In these circumstances mechan-ical reductions of woody biomass may be nec-essary before reintroducing fire to the forest.In other cases managers attempt to control for-est vigor by thinning overcrowded stands, withthe goal of keeping insect outbreaks within lim-its. However, these forestry practices are gen-erally expensive and require extensive imple-mentation to be effective at the landscape scale,so they have been applied primarily in targetedsituations, such as the “wildland–urban inter-face” where high-value homes are threatenedby wildfire (Radeloff et al. 2005).

Historic disturbance regimes are increasinglyused as a reference point to guide managementof future landscapes (Perera et al. 2004). Thisapproach is based on the premise that nativespecies and ecological processes might best bemaintained by retaining ecological structures(including whole landscapes) and disturbanceprocesses in a seminatural range of conditionsto which native organisms are well adapted.

Several provinces in Canada (Perera et al. 2004)and forestry organizations in the USA (e.g.,Cissel et al. 1999), for example, have exploredthis approach to forestland management bystudying historic wildfire patterns over timeand space and then developing managementplans with frequency and severity of harvesttreatments that emulate the wildfire regime.Although this concept has been extensively dis-cussed in the literature (Burton et al. 2003, Per-era et al. 2004), no examples have been imple-mented extensively, in part because of the longtime periods required to produce an imprint onforest landscape structure where cutting rota-tions approach or exceed 100 years.

In addition, many challenging questions havebeen raised concerning the use of the historicrange of variability to guide future manage-ment (see Chapter 2): What period of historydo we use as a guide? What aspects of his-toric disturbance regime do we incorporate inthe management plan (e.g., frequency, sever-ity, spatial pattern of disturbances)? Climatechange presents many challenges that are dif-ficult to anticipate. Millar et al. (2007) presentthe options for employing adaptive strategiesin these cases, but we know of no cases whereactual practices are in place on the ground.

The challenge of using an historic approachbrings into focus the question: What does “con-servation” mean in a world with so muchchange? Given the dynamic state of the envi-ronment, an iterative approach seems prudent:

1. Identify the climatic and other forms of envi-ronmental change that are most likely tooccur and the likely ecological responses,both gradual and abrupt/threshold;

2. examine and implement managementapproaches that may confer social and eco-logical resilience in the face of anticipatedchange;

3. design an assessment process that provides abasis for adaptive change, as learning occurs;and

4. be prepared to adjust objectives from culti-vating resilience to transformation, if envi-ronmental or social change is so great thatan alternative trajectory may provide greater


ecological and social benefits in the face ofinevitable transformation.

This sequence of steps required for adap-tive management (see Chapters 4 and 8) maytake many decades. In many cases, desirableand undesirable outcomes are strongly shapedby social forces, as described in the next section.

Forest Conversion to New LandUses

Conversion of forests to other land uses andpotential for reversion to forest are criticaldeterminants of future forest extent and the ser-vices they provide to humanity. The history offorest clearing is deep, but the process is accel-erating in response to a wide range of socialand economic drivers. Forest clearing for shift-ing agriculture has caused cyclic change in thepast, although increasing population pressuremay lead ultimately to deforestation. Otherforms of forest clearing may be immediate andirrevocable.

Tropical deforestation, both legal and illegal,to make way for agriculture, such as produc-tion of beef and biofuels, has been an issue ofcritical concern because of the profound socialand environmental consequences. All ecologi-cal services that are distinctively provided byforests are lost. Tropical deforestation involvescross-scale interactions in driving local changeand the large-scale feedbacks to the climate sys-tem, as described earlier.

An important form of forest conversion mayoccur due to the intimate mixing of forestand other land uses, even where the landscaperemains largely forested. For example, low-density residential development in forest land-scapes is common in many parts of the USA,creating conflicts when forest processes, suchas wind-toppling of trees, wildfires, and wildanimals (e.g., cougars drinking from children’swading pools) impinge on a suburban life style.In these circumstances, some ecological servicesprovided by forests may be partly maintained(e.g., carbon sequestration, water regulation,habitat for some groups of forest species), butothers are lost (e.g., wood production, fire, and

habitat for species considered threatening topeople).

Not all forest conversion is irrevocable. Forexample, waves of land-use change began inNew England when the “soft” deforestation foragricultural development took place in the lateeighteenth and early nineteenth century (Fos-ter and Aber 2004). Industrialization, access tosuperior farmlands in the Midwest, and otherfactors led to farm abandonment and reforesta-tion largely by natural reseeding. Now, how-ever, the sprawl of urban and rural residentialdevelopment is a new wave of “hard” deforesta-tion; the pavement and dwellings will be moredifficult to return to forest. New conservationstrategies are being put forth to identify the bestof the remaining forestland tracts in a regionaldesign intended to provide large habitatpatches and dispersal corridors between them(e.g., the Wildlands and Woodlands programin Massachusetts and neighboring states—http://harvardforest.fas.harvard.edu/wandw/).

Changes in subsidies and taxation systemscan precipitate shifts in property ownershipand attendant shifts in commitment to sustain-able management. In the USA, forest com-panies that have a long-term stake in forestproductivity are being taken over by Timber-land Investment Management Organizations(TIMOs) and Real Estate Investment Trusts(REITs) that seek very short-term (fraction ofa cutting rotation) financial return in part byselling forestlands for residential development(Fernholz 2007). In the eastern US conserva-tion NGOs are partnering with timber compa-nies to purchase development rights that pro-tect forestlands from development and allowsmall-scale logging to continue (Ginn 2005).Similarly, in some developing nations, debt-for-nature swaps have allowed foreign debts to beforgiven in return for conservation protectionof lands that might otherwise be cleared. Thelong-term outcomes of these emerging trans-actional mechanisms controlling forests are farfrom clear—will they lead to more sustainableforest-management practices or to a seriousdeparture from sustainable forestry? TIMOsand REITs appear to encourage unsustainablepractices, whereas the debt-for-nature programmay protect forests from clearing. The interme-


diate case of conservation easements on privatelands may prove to be a compromise that pro-vides both ecological and social benefits. Man-agement of public lands seems to be on the pen-dulum swing between reserve and rather inten-sive management, as, for example, the NewZealand bifurcation of lands to native forestwithout management and plantations largelyof exotic species (mainly Pinus radiata). Onlytime will tell the outcome of these alternativeapproaches to governance.

Change in national policies to shift energyproduction from fossil fuels to forest-generatedbiofuels will intensify forest management, ascurrently being considered in Sweden, or leadto clearing of native forests for oil palm planta-tions, as is occurring in Indonesia.


These many approaches and challenges toforestry lead us to seek an expanded view of for-est stewardship for tomorrow’s forests, drawingon the large bodies of practical and scientificknowledge of forests and associated communi-ties and also calling for greatly enhanced adap-tive capacities to face the changing societal andenvironmental conditions. Despite their social–ecological diversity, forests face similar chal-lenges throughout the world, suggesting somegeneral strategies for sustainable forest stew-ardship. The implementation of these strategiesmust be context-specific, taking into accountlocal, national, and global drivers of change.

• Institutions governing forest stewardshipmust allow planning for the long term bypromoting practices that maintain the socialand ecological capital required for multi-ple generations of trees and forest users,whether they are in developed or develop-ing countries (see Chapter 1). In the absenceof such institutions, positive social outcomesare most likely to result from attention tocapacity building and adaptive governancethat would foster development of institu-tions and policies that might lead to favor-able social–ecological transformations to analternative state.

• Sustainable timber production requires sus-taining the slow variables that maintain theproductive potential of ecosystems, includingsoil fertility, a disturbance regime to whichlocal organisms are adapted, and both theecological and the socioeconomic diversitiesto maintain future options.

• Effective stewardship to sustain multipleecosystem services provided by forestsrequires a clear understanding of the con-trols over these services and the syner-gies and tradeoffs among them. Althoughit is unlikely that all these services can bemaximized simultaneously, careful planningthat engages stakeholders can lead to well-informed choices that address multiple needsand concerns.

• Given that social and environmental condi-tions are quite likely to change and unknow-able surprises will occur, forest steward-ship should foster flexibility to respond tothese changes through adaptable governanceand the fostering of biological and socioeco-nomic diversity that provides the seeds formultiple future options (see Chapter 5).

• Given the frequent mismatch between therequirement for long-term vision for foreststewardship and the short-term motivationsand path dependence of decision options,forests are quite prone to social–ecologicaltransformations, suggesting the benefit ofplanning that considers multiple alternativestates and their relative social and ecologicalbenefits. Think outside the box.

Although none of these strategies is uniqueto forests, the longevity of forest trees makesthe importance of the long view particularlyapparent and could therefore inform ecosys-tem stewardship in a broad range of social–ecological systems.

Review Questions

1. What are some of the challenges to foreststewardship that result from the long-livednature of trees? What other systems facesimilar stewardship challenges?


2. What are the major stages of developmentof management practices that have char-acterized forestry practices in many partsof the world? Based on this understanding,how might policy makers avoid detrimen-tal phases and foster sustainable forest stew-ardship in regions that are just beginning todevelop forest resources?

3. What common and distinctive challenges docommunities in developed and developingcountries face in exercising local influence onforest management and ecological servicesreceived from forests?

4. How can production forests be managedin ways that provide multiple ecosystemservices within a management frameworkgeared to maximizing wood production?

5. What institutions and social practices aremost likely to foster long-term sustainabilityof forest values that reflect inevitable trade-offs among multiple ecosystem services?

6. What roles do scientists play in naturalresource decision making? What challengesdo they face in each of these roles?

Additional Readings

Brown, K. 2003. Integrating conservation and devel-opment: A case of institutional misfit. Frontiers inEcology and the Environment 1:479–487.

Ellison, A.M., M.S. Bank, B.D. Clinton, E.A.Colburn, K. Elliott, et al. 2005. Loss of founda-tion species: Consequences for the structure and

dynamics of forested ecosystems. Frontiers inEcology and the Environment. 3:479–486.

Folke, C., S. Carpenter, B. Walker, M. Scheffer, T.Elmquist, et al. 2004. Regime shifts, resilience,and biodiversity in ecosystem management.Annual Reviews in Ecology, Evolution, and Sys-tematics 35: 557–581.

Foster, D.R., and J.D. Aber. 2004. Forests in Time:The Environmental Consequences of 1,000 Yearsof Change in New England. Yale University Press,New Haven.

Ginn, W.J. 2005. Investing in Nature: Case Studiesof Land Conservation in Collaboration with Busi-ness. Island Press, Washington.

Lane, M.B. and G. McDonald. 2002. Towards a gen-eral model of forest management through time:Evidence from Australia, USA, and Canada.Land Use Policy 19:193–206.

Millar, C.I., N.L. Stephenson, and S.L. Stephens.2007. Climate change and forests of the future:Managing in the face of uncertainty. EcologicalApplications. 17:2145–2151.

Nagendra, H. 2007. Drivers of reforestation inhuman-dominated forests. Proceedings of theNational Academy of Sciences 104:15218–15223.

Sampson, R.N., N. Bystriakova, S. Brown, P.Gonzalez, L.C. Irland, et al. 2005. Timber, fuel,and fiber. Pages 585-621 in R. Hassan, R. Scholes,and N. Ash, editors. Ecosystems and HumanWell-Being: Current State and Trends. MillenniumEcosystem Assessment. Island Press, Washington.


8Drylands: Coping with Uncertainty,Thresholds, and Changes in State

D. Mark Stafford Smith, Nick Abel, Brian Walker, and F. Stuart Chapin, III

Introduction

Drylands cover 40% of the terrestrial surface(Table 8.1, Plate 6) and are characterized by highecological and cultural diversity. Although theyare, by definition, of low productivity, they havebeen a source of biotic, social, and scientific inno-vation.Athirdof theglobalbiodiversityhotspotsare in drylands, with a diversity of large mam-mals in savannas, high diversity and endemismof vascular plants in shrublands, and a high diver-sity of amphibians, reptiles, birds, and mammalsin deserts. Succulence, the CAM photosyntheticpathway, and camels’ tolerance of changingblood water content are examples of biologi-cal innovations that arose in drylands. Drylandsare culturally diverse and account for 24% ofthe world’s languages (Safriel et al. 2005). Tradi-tionally, many social groups moved both season-ally and in response to prolonged droughts (e.g.,Davidson 2006). The need to cope with harshconditions and repeated episodes of scarcity

D.M. Stafford Smith (�)CSIRO Sustainable Ecosystems, PO Box 284, CanberraACT 2602, Australiae-mail: [email protected]

have given rise to strong cultural traditionssuch as the invasive effectiveness of the Mon-gols, the rule base for several major religions,and traditional ecological knowledge backed bypowerful sanctions as in Aboriginal cultures inAustralia. In ecology, attention to the extremeconditions represented by drylands has helpedcreate paradigm shifts of wider relevance, suchas the development of disequilibrium conceptsand state-and-transition models (Westoby et al.1989, Vetter 2005) that were important driversof the development of resilience theory (Walker1993). Such concepts are especially pertinentas humans prepare for climatic change.

Today, about a third of the world’s pop-ulation (two billion people) live in drylands(Table 8.1), about half of them dependent onrural livelihoods. Water availability stronglyconstrains both biological productivity andhuman development. Drylands receive only8% of the world’s renewable fresh water supply− 30% less per capita than the minimumconsidered essential for human well-being andsustainable development (Safriel et al. 2005).Dryland populations tend to lag behind thosein other parts of the world on a variety of eco-nomic and health indices, even controlling for“ruralness” (MEA 2005a, b), with higher infant


172 D.M. Stafford Smith et al.

Table 8.1. Key features of four dryland subtypes. Information from Safriel et al. (2005).

Hyper-arid Arid Semi-arid Dry subhumid Total

Aridity index1 <0.05 0.05–0.20 0.20–0.50 0.50–0.65Area (million km2) 9.8 15.7 22.6 12.8 60.9

Share of globe (%) 6.6 10.6 15.2 8.7 41.3Total population (million) 101 243 855 910 2109

Share of globe (%) 1.7 4.1 14.4 15.3 35.5

Land-use (% of subtype) % of all drylands2

Rangelands 97 87 54 34 65Cultivated 0.6 7 35 47 25Urban 1 1 2 4 2

1 Ratio of precipitation to potential evapotranspiration2 Does not add to 100% as 8% is taken up in other categories, notably inland waters.

100

80

60

40

20

0

3500

3000

2500

2000

1500

1000

500

0

Land degradation(% of area)

Livestock density(# km–2)

Water use(m3 yr–1 person–1)

Lan

d d

egra

dat

ion

an

d li

vest

ock

den

sity

Wat

er u

se

Po

pu

lati

on

den

sity

, G

NP,

in

fan

t m

ort

alit

y

80

60

40

20

0

Hyper-arid Arid Semi-arid Dry sub-humid

GNP ([$US 100s] person–1)

Population density (persons km–2)

Infant mortality(Deaths [1000 people]–1)

Figure 8.1. Some vital statistics across the dryland aridity gradient: top, GNP per capita (excluding OECDcountries), infant mortality (excluding OECD countries), population density (developing countries only);bottom, proportion of dryland subtype that is degraded (%, 1990 estimates, excluding “low” category), live-stock density (developing countries only), and water supply per capita. Information from Safriel et al. (2005).

8 Drylands 173

mortality, severe shortages of drinking water,and much lower per capita GNP (Fig. 8.1).Dryland populations are among the most eco-logically, socially, and politically marginalizedpopulations on Earth (Khagram et al. 2003).Given that 10−20% (depending on measures)of drylands are “desertified,” their populationsare seen as some of the most vulnerable to theincreased frequency of drought events expectedunder climate change (MEA 2005b, Burke et al.2006). Rapid population increase coupled withlivelihoods at risk creates the potential forfurther water shortages and land degradation,most likely causing movement of environmen-tal refugees and flow-on effects to other biomesof the world.

In many ways, drylands represent the epit-ome of management challenges emphasized inthis book. They are social–ecological systemsthat were traditionally well-adapted to uncer-tainty and variability, and many dryland civi-lizations persisted for millennia. However, inpart through attempts to reduce this variabil-ity, many of the recent ecological and socialchanges have reduced the social–ecologicalresilience of drylands. The social context and

the climate under which these social–ecologicalsystems evolved have changed and will con-tinue to change. What were some of the orig-inal sources of resilience in drylands? Howhave these changed, and how can we restoreor generate new sources of resilience underthe altered social and climate contexts? Insome cases, drylands have undergone dramaticand apparently irreversible degradation. Whatchanges in social–ecological slow variables (seeChapter 1) have caused these thresholds to beexceeded? How can we reduce the likelihoodof this occurring in the future?

Dryland Featuresand Functioning

The key attributes of drylands can be sum-marized as unpredictability, resource scarcity,sparse populations, remoteness, and “distantvoice” – the drylands syndrome (Reynolds et al.2007). Remote regions of Australia clearly illus-trate this syndrome (Fig. 8.2; Stafford Smith2008), although these attributes are impor-tant in most drylands to a greater or lesser

Low, unpredictable, patchy resources

(water and nutrients)

Sparse, mobile,patchy human

population

Few options for livelihood.

wealth creation,and infrastructure Unique cultures

and institutions

Unpredictable markets,

labor, policy,and investments

Distant markets,government and industry

Figure 8.2. Factor interactions and feedbacks in the dryland syndrome. Adapted from Stafford Smith (2008).


Table 8.2. Gross domestic product (GDP) and area of Australian rangelands used bydifferent sectors in 1999. Information from Fargher et al. (2003).

Rangeland contribution to Area of rangelandsSector Australian GDP (% of GDP) occupied (% of area)

Mining 2.4 0.04Tourism 0.4 Not available (small)Pastoral leasehold 0.2 50

extent. Reynolds et al. (2007) applied resilienceand systems theory to drylands with theseproperties to develop a drylands developmentparadigm (DDP), a conceptual framework con-sisting of five principles for analyzing past andfuture policy and management (see Table 2 inReynolds et al. 2007):

P1. Social and ecological systems are coupledin drylands to form complex adaptive sys-tems (see Chapter 1); this is particularlyimportant in drylands because livelihoodsare tightly coupled to the biophysical envi-ronment and vulnerable to market and pol-icy processes that are beyond local control.These systems have no simple target equi-librium point for resource users and pol-icy makers to aim for. Tracking change isthus more difficult and important in dry-lands than elsewhere.

P2. Although drylands are highly variable intime, these dynamics are determined bya few slow variables (see Chapter 1) thatare regulated by feedbacks. Identifying andmonitoring slow social and ecological vari-ables is particularly important in drylandsbecause high variability in fast variablesmasks the fundamental change indicted byslow variables. The limited suite of keyvariables and processes makes the complexsystem tractable.

P3. Thresholds in key slow variables define dif-ferent states of social–ecological systems.Soil loss from a landscape that supportsan arable farming community, for exam-ple, may shift the system to another statethat can only support livestock, when thethreshold of soil loss is exceeded. Thresh-olds are particularly important in drylandsbecause there is usually little capacity to

invest in recovery from an undesirablechange in state. Where calls must be madeon outside agencies, the transaction costs ofdoing so are high in drylands due to relativeremoteness. Such thresholds contribute tothe unpredictability of drylands.

P4. Dryland social–ecological systems are hier-archical, nested, and connected throughsocial networks, communications, andinfrastructure to the broader nation inwhich they are located. This is particu-larly important in drylands because theirtypically low primary productivity supportsonly sparse populations that tend to haveweak political influence in remote seatsof governance. Consequently, they receivelower levels of services and infrastruc-ture than national averages. Cross-scalelinkages are important but usually weakin drylands, thus requiring particularinstitutional attention.

P5. Local ecological knowledge is criticallyimportant for maintaining the functionsand coadaptation of dryland social–ecological systems. This is particularlyimportant in drylands because experi-ential learning and innovation tend tooccur more slowly in these highly variablesystems. In addition, there is often lessformal research than in mesic regions.Consequently, the development of hybridscientific and local knowledge systems isvital for local management and regionalpolicy in drylands.

In this chapter we use the DDP principles,which are derived from resilience theory forenvironments driven by variability and scarceresources, to analyze and seek solutions to dry-lands problems.

8 Drylands 175

Physical Functioning

Drylands span a gradient from arid to semiaridto dry subhumid areas. Precipitation is scarceand variable (Middleton and Thomas 1997)along the entire gradient. Drylands are definedclimatically by their low aridity index (<0.65)– the ratio of precipitation to potential evapo-transpiration (Table 8.1) – resulting from highair temperatures, low humidity, and abundantsolar radiation.

Water is thus the factor that most directlylimits biological productivity and human pop-ulation density. Water’s greatest societal sig-nificance is to support forage production forpastoralists at the drier end of the gradientand to support agriculture at the moister endof the gradient. Not only are rainfall inputsmore-or-less low in drylands, but they are gen-erally less predictable than in other systems.Rainfall variability increases as mean annualrainfall decreases, as latitude decreases, andas the effect of the El Nino-Southern Oscil-lation (ENSO) increases (Nicholls and Wong1990), thus delineating the great midlatitudebelts of deserts in both hemispheres with addedimpacts from ENSO, especially in the southernhemisphere drylands. Even where rainfall itselfis more reliable, other climatic elements (e.g.,extremely cold winters in Kazakhstan or Patag-onia) can create other sources of interannualvariability from the perspective of human use.

Climate is coupled to vegetation cover ata regional scale. Loss of vegetation throughland use increases albedo, which lowers surfacetemperature, reduces convective uplift (and insome cases advection of moist marine air),and reduces monsoon rains (Foley et al. 2003b,Xue et al. 2004). In contrast, controlled graz-ing, afforestation, and irrigated agriculture inthe northern Negev of Israel led to a 10–25%increase in rainfall (Otterman et al. 1990), sug-gesting multiple effects of management on dry-land climate.

Rock type determines parent material andthus affects spatial patterns of plant-availablenutrients (PAN). Parent material affects soiltexture, and thus soil permeability and soilmoisture storage capacity, as well as soil depth.Soil moisture and nutrient levels can be gener-

alized at broad scales. The productivity, struc-ture, composition, and functioning of ecologicalcommunities (Cole 1982) and the resilience ofsoil–vegetation systems can be partly explainedin terms of PAN and plant-available moisture(PAM; Frost et al. 1986, Walker and Langridge1997), both of which act as slow variables, withthresholds that can be crossed if over-croppedor overgrazed. In general (Walker et al. 2002):

• The ratio of woodland to grassland (a slowvariable) increases with increasing rainfall;

• So does the likelihood of increases in unde-sirable shrubs when a site is grazed;

• The productivity of a site is highest wherePAN and PAM are both high, permittingfaster rates of recovery after grazing. How-ever, this says nothing about the propensityof a site to change to a new state.

Biological Functioning

Soil and vegetation characteristics stronglyinfluence water retention and use (d’Odoricoand Porporato 2006). Because natural rates oferosion from wind and water are high, soilstend to be shallow and therefore have low soilmoisture storage capacity, a key slow variable.This determines vegetation productivity as wellas influencing the ratio of water infiltration torunoff. Most dryland soils also have low organicmatter and low aggregate strength, which makethem highly erodible. When tillage or grazingreduces ground cover below a critical thresholdof between 30 and 40%, soil loss from intensewinds and rainfall can reduce soil depth (thussoil moisture-storage capacity) below a criticalthreshold that can preclude recovery of vegeta-tion productivity and consequent ground cover(Fig. 8.3). This negative effect on soil mois-ture is reinforced because a decrease in veg-etation cover reduces the rate of water infil-tration. Losses of mineral soils are accompa-nied by losses of clay and organic matter, socation exchange capacity, a slow variable thatdetermines PAN, is also reduced. Seeds mayalso be transported off the landscape by waterflows, reducing the capacity of vegetation torecover.


Grazing intensityincreases

Rainfall

Runoff increases

Infiltrationdecreases

Ground coverdecreases

Soil faunadeclines

Soil moisturedecreases

Soil depthdecreases

Nutrient, soil, andseed loss increase

Primaryproductivity

decreases

Figure 8.3. Grazing, aspects of landscape function,and feedbacks among these. Changes in groundcover (a fast variable with fluctuations driven mainlyby interannual rainfall variability but affected bygrazing) can cause soil depth (a slow variable) andconsequent soil moisture storage capacity to crossthresholds below which the rest of the cycle moves

into a new state characterized by impoverished soilfauna, reduced nutrients and seed banks, and long-term loss in productivity. There is a negative feed-back to grazing intensity, but management interven-tions generally weaken this control by maintainingstock numbers.

Infiltration and soil moisture storage are alsoinfluenced strongly by soil invertebrates thatcreate macropores and support nutrient cycling.Decreases in soil moisture resulting from graz-ing can cause declines in soil fauna with neg-ative effects on nutrient availability and soilmoisture. This further reduces vegetation pro-ductivity and increases vulnerability to furthersoil loss (Fig. 8.3). Thus a state change can beinduced as thresholds of PAN and PAM arecrossed.

Toward the drier end of the climatic gradi-ent there is insufficient PAM to support morethan very sparse vegetation cover if the landsurface is flat. However, even slight topographicvariations redistribute water (and seeds andnutrients) to produce patches of vegetationat all scales. Soil–water–nutrient–plant–animalprocesses have been integrated in the conceptof landscape function (Ludwig et al. 2005).Landscape function is the capacity of a land-scape to regulate nutrients and water, con-

centrating them in fertile, vegetated patcheswhere soil biota maintain nutrient cycles andwater infiltration and where vegetation coverimpedes surface water flow, retains nutrientsand seeds, maintains infiltration, and protectsagainst erosion (Fig. 8.4). The resulting vegeta-tion structures, such as brousse tigre (bandedlandscape patterns) and perennial grass andshrub “islands” (Ludwig et al. 1999, Rietkerket al. 2004), are self-organizing expressions oflandscape function. These support biota thatcould not occur if the same resources wereevenly distributed across the landscape.

Loss of landscape function, commonly evi-dent from the loss of vegetation patterns andhomogenization of landscape cover, can drivea landscape from a productive state acrossa threshold to a state of low productivity(Fig. 8.3). The key control variable is vegetationcover at ground level. Ground cover on crop-land is affected by crop type, sowing method,fertilizer level, and residues. On rangelands it is

8 Drylands 177

Low infiltration and nutrient recycling rates

Moderate infiltration and nutrient recycling rates

High infiltration and nutrient recycling rates

Runoff

Runoff

Figure 8.4. Functioning of a “brousse tigre” or banded shrub landscape when it selforganizes to capturewater, nutrients, litter, and seeds. Modified from Tongway and Ludwig (1997).

affected by fire and grazing. As cover declines,there is increased loss of water, soil, nutrients,organic matter, and propagules from the system(Ludwig et al. 2005). There is a critical thresh-old level in landscape function below which,even if all livestock are withdrawn or fields areleft fallow, the process continues to an irre-versibly degraded state, or with a slow, hys-teretic return path (Rietkerk et al. 2004). In thiscase we see landscape function as a critical slowvariable with a threshold that separates alter-nate states.

Pringle and Tinley (2003) have extendedthe landscape function of drylands to a catch-ment scale that integrates a geomorphologicalframework. Water erosion caused by grazing orcropping can incise a landscape and lower itsbase level (a slow variable), for example, byincising natural levees and riparian sills. Thiscan increase drainage from, and reduce soilmoisture in, the landscape above the incision.Edaphic wetlands can be transformed to a newstate characterized by scrub, for example. Thedecline in PAM and PAN reduces primary pro-ductivity. The management implication is thatthe problem needs to be addressed at catch-ment scale.

Socioeconomic Functioning

More than in most regions of the world, dry-lands are characterized by a majority of liveli-hoods that still depend on natural and cultural

resources, through grazing, dryland agriculture,tourism, and mining. In this section we focusmainly on grazing and dryland agriculture, witha brief discussion of other uses.

Long-established social systems in dry-lands have coadapted with the unpredictabil-ity, resource scarcity, and large extent of range-lands. Compared to mesic areas, and a fewmajor desert cities notwithstanding, three socialand economic features of dryland systems arepervasive: the human populations of drylandsare usually more sparse, mobile, remote frommarkets, and distant from the centers (and pri-orities) of decision makers. It is consequentlymore costly to deliver services, and institu-tional arrangements devised in other regionsmay be dysfunctional when imposed on dry-lands (Stafford Smith 2008).

Systems of property rights and associatedreciprocal obligations between individuals orsocial groups reduce conflicts over scarce for-age and water by enabling movements of peo-ple and stock. Such arrangements are slowlychanging social variables that have been crit-ical to traditional human adaptation to dry-lands. However, some rangeland regions havebeen privatized, leading to long-term trajecto-ries of changes in systems, driven by increas-ing populations of stock and humans, expandingcommercial opportunities, alienation of land,blockage of traditional migration routes, andsometimes veterinary disease control fences(Fig. 8.5). These processes alter the cost-benefit trade-off of subdivision and enclosure


Consolidation ofresidual rangelands

Permanent excisionof higher-value land

Land tenure policy Commercial arrangements

Mainly driven by

Period of highest degradation risk

Unconstrainednomadicsystems

Privatization of“better” land as

populationincreases

Theorized developmental trajectory over time

Lan

dsc

ape

frag

men

tati

on

Mo

re fr

agm

ente

dM

ore

inta

ct

Figure 8.5. Hypothesized general trajectory of landfragmentation, potential excision for permanentagriculture, and consolidation in drylands. The depthof the curve and the degree to which the dashed orsolid part to the right predominates depends on a

variety of factors including productivity (more exci-sion at the more productive end of the spectrum),population density, and political stability (more exci-sion for higher population and political stability).Modified from Behnke (2008).

of common resources (Behnke 2008). In devel-oping countries, subdividing land and assigningit to individuals, families, or small groups oralienating land for irrigation schemes is oftendriven by development policies devised in dis-tant capitals. It disrupts previously nomadic cul-tures and resource-use strategies. Traditionalstrategies are commonly focused around keyland units where landscape function is robust,and forage and surface water are available dur-ing dry periods (Scoones 1995). Such land iscommonly selected for privatization because itshigher productivity, unlike that of surround-ing land, justifies capital investment (Behnke2008). However, there are thresholds, such that,if too much of the key land resources areremoved from the system, a nomadic strategyon the residual land becomes unsustainable.Cheap and readily available weapons such asassault rifles have further transformed resourceaccess disputes in some dryland regions (Carr2008). Where previously they were resolvedthrough negotiations or with spears, they maynow become much bloodier feuds as in, for

example, Darfur in Sudan, and the Horn ofAfrica among many other regions.

Today pastoralism and rangeland use stilldominate land use in hyperarid and arid(desert) portions of drylands, but agriculturenow covers a greater area than rangelandsat the moist end of the drylands gradient(Table 8.1). Cities are also concentrated inregions with greater water availability, wherethey occupy 4% of the land area (Safriel et al.2005). Consequently, human population den-sity is sevenfold greater at the moist end ofthe dryland moisture gradient (71 persons km-2)than at the dry end (10 persons km-2; Fig. 8.1).Other uses are also important. Mining occupiesa trivial proportion of the area of the drylands,but its economic importance can hugely exceedthat of pastoralism and agriculture. Tourism isalso important, as, for example, in Australia(Table 8.2). In developing countries tourismbased on wildlife has a history at least a centurylong, relying on protected areas from whichlocal people are wholly or partly excluded(see Chapter 6).

8 Drylands 179

Causes of Degradation and TheirImpacts

The central issue in drylands is the same asin other systems – the sustainable produc-tion of outputs valued by humans. In drylandsthese outputs are milk, meat, hides, grains,timber, biodiversity conservation, aesthetic,recreational and tourism experiences, and hunt-ing trophies, as well as minerals. Sustainabil-ity implies that the net production of valuesis maintained indefinitely (see Chapter 1). Thefeatures of the drylands syndrome – unpre-dictability, resource scarcity, sparse populations,remoteness, and distant voice (Reynolds et al.2007) – create particular problems in achievingthis goal compared to other systems. Geist andLambin (2004) provide a comprehensive anal-ysis of 132 cases of major change in differentdryland systems worldwide, identifying six clus-ters of fundamental biophysical or social driversthat underpin the proximate causes of drylanddegradation – climatic, demographic, techno-logical, economic, policy and institutional, andcultural processes.

Climatic – Climate is changing at a varietyof temporal and spatial scales, from the short-and medium-term variability discussed earlier,through regional changes due to deforestationand other land-use changes, to global climaticchange. The past 50 years have already shown atendency toward drying trends in many drylandareas of the world, with these trends expectedto become more pronounced in the next 50years (Burke et al. 2006). Variability is alsolikely to increase. Both drying and increasedvariability can cause degradation if resourceuses and institutions become unsuited to the cli-matic regime.

Demographic – The human population isincreasing more rapidly (18.5% between 1990and 2000; Safriel et al. 2005, p. 645) in drylandsthan in any other major biome. However, dry-lands worldwide can be broadly split betweenthose (mostly in developing nations) wherethere is strong population growth and conse-quent pressures on the resource; and those(mostly in developed nations) where drylandstend to become depopulated (notwithstanding

the development of major cities such as Phoenixand Las Vegas in the USA, and Dubai in theGulf States that are metropolitan islands setin arid hinterlands). Drylands are also a majorsource of migration.

Technological – A whole series of inno-vations continually alter production dynam-ics, so that the balance between managementand environment continually needs updating.These include physical technologies such asmore powerful boring rigs that can dig deeperfor water, new types of fences, new and cheapweapons used in resource-access disputes, andmore accessible four-wheel drives for tourists,as well as biotechnologies such as higher yield-ing or pest-resistant crop cultivars and newbreeds of grazing animals. There are also com-munications and transport technologies thatlink drylands to the opportunities and threatsof globalization.

Policy and institutional – Medium- andlong-term shifts in political structures andgeopolitical allegiances (e.g., loss of the SovietUnion, growing strength of China and India,development of the Convention to Com-bat Desertification, and other multilateralarrangements) alter the pressures on drylands.Changing social processes and institutionalstructures (e.g., methods for delivering aid,increasing roles of NGOs, network develop-ment such as “Desert Knowledge” in Australia[Stafford Smith et al. 2008]) provide opportuni-ties and challenges.

Economic – Changing market needs andaccess, as well as world trade agreements andbilateral arrangements, mostly associated witha broad suite of globalization effects, reachinto drylands profoundly, as exemplified by thehuge surge in mining in deserts worldwide (e.g.,Chile, Australia) as a result of demand fromChina.

Cultural – Positive cultural trends in devel-oped nations such as rising demand for tourism,greater appreciation of cultural and biologicaldiversity, and the opening of premium mar-kets such as Fair Trade deliver opportunities todryland areas, at the same time that advancesin communications technologies increase theimposition of a dominant culture from devel-oped nations into remote areas, causing an


Globalization Technologies

Socialvalues

Climatechange

Marketorientation

Tenuresystems

Political influence

Localknowledge

Landuse

Infrastructure Population

Wateravailability

Income Forage / cropgrowth

Landscapefunction

Figure 8.6. The effect of scale on the role of pro-cesses as external drivers, internal controlling (slow)variables, or fast variables representing immediate

goods and services. Factors outside any ellipse aregenerally slow (but not necessarily controlling) vari-ables, whereas those inside are relatively fast.

unprecedented rate of loss of local languagesand culture.

Whether a factor such as population is adriver or a controlling (slow) variable dependson the scale of analysis (Fig. 8.6). The fol-lowing sections explore some major systemchanges, usually driven by multiple factors incombination, and often creating both risks ofdegradation and new opportunities. The dis-cussion aims to highlight slow variables andthresholds.

Expanding Aridity

The drying trends of the past 50 years andincreased human use of freshwater, made pos-sible in part by improving borehole technology,have combined to cause groundwater tablesto drop in most areas, a widespread prob-lem, for example, on the Indo-Gangetic Plainof India. For a given technology, there is athreshold below which water extraction is notworthwhile or is not feasible, possibly render-ing some areas unsuitable for human occupa-tion. The chronic water shortage that charac-

terizes drylands is expected to become morepronounced, by 11% per capita at the moistend of the gradient to 18% at the dry end ofthe gradient for the period 2000–2010 (Safrielet al. 2005) due to a combination of popula-tion increase, land cover change, and climatechange.

A key climatic feature of drylands is vari-ability of rainfall. Ellis (1994) proposed thatthere is a threshold in the coefficient of vari-ation (CV – standard deviation divided by themean) of annual rainfall of 30% above which itis more useful to think of a system as disequi-librial (that is, perpetually buffeted away fromany equilibrium), than as an equilibrium systemexperiencing occasional disturbance. Increasingvariability resulting from climatic change maycause this threshold to be crossed in regionsthat are currently managed as equilibrial sys-tems. Adaptation to avoid land degradation insuch cases may require resource users to shift toan opportunistic production strategy that is spa-tially more mobile and that can accommodatelarge fluctuations in livestock numbers and out-puts. On the other hand, it may become a ratio-nal alternative to vacate lands that, through

8 Drylands 181

climatic change, have lost productive poten-tial and have become increasingly risky to use.Policies to address the risks of land degra-dation, inequity, and impoverishment couldinclude incentives to leave the region, radicallyreduced stocking densities, or shifts to alterna-tive occupations within or outside the affectedregion.

Biodiversity Loss

Biodiversity is believed to be a key founda-tion of the resilience of social–ecological sys-tems because biodiversity augments responsediversity and the maintenance of functionsthat may become important if circumstanceschange (see Chapter 2; Elmqvist et al. 2003).Biodiversity is also valued for its own sake,its commercial or subsistence utility (forexample, tourism and recreation), future val-ues that are not yet known, or its sup-port of cultural practices, including traditionalones.

Biodiversity conservation policies and man-agement attempt to restore or maintain func-tioning examples of selected ecological commu-nities and the processes and biota they con-tain. Processes include the dynamics and disper-sion of populations, trophic and other interac-tions, and the evolutionary processes that gen-erate diversity. Successful conservation of bio-diversity requires the maintenance of sufficientareas of well-functioning land, representingall the biological units and processes deemedimportant, including the ecological interactionswithin and among the conserved areas (Sarkaret al. 2006). This is an ideal, and policies andmanagement are usually pragmatic rearguardactions against threats from habitat loss, pestsand weeds, increased human use, and othercauses. The best that can be achieved in mostcircumstances is to make and keep each man-aged area:

• large enough (a slow variable with a thresh-old) to maintain the viability of sedentarypopulations;

• sufficiently well connected (slow variablewith a threshold) to other conservation sites

to enable movements of mobile species andrecoveries from local extinctions.

Connectivity depends on distance betweensimilar sites, as well as the suitability of theland between the sites for dispersal by mobilespecies. The majority of drylands, particularlythose that are grazed rather than cropped andnot too fragmented, are often in seminaturalcondition, so the opportunities to combine con-servation with moderate land use are greaterthan in dramatically transformed agriculturallandscapes.

With increasing population and appropria-tion of primary production, humans in dry-lands threaten biodiversity by removing habi-tat, heavy grazing of plant species, block-ing dispersal paths, killing wild species toeat or sell, changing fire regimes, and graz-ing their animals in competition with wildspecies. For example, there is a long historyof conflict between wildlife conservation andindigenous uses of African drylands. Conser-vation areas were established by colonial gov-ernments, and indigenous users were formallyexcluded in most cases, although poachingand other uses continued (Abel and Blaikie1986). Today conservation areas continue tobe supported by postcolonial governmentsbecause of the foreign exchange they earn fromtourism (see Chapter 6). There are thresh-olds in the slow variables of size, representa-tiveness, and connectivity that affect the suc-cess of biodiversity management. Reconsoli-dation of fragmented land, sometimes underlocal initiatives, sometimes promoted by gov-ernment policy, has often been necessaryin order to incorporate animal populationswithin an ecologically viable unit (Abel et al.2006). Where the movements of wildlife val-ued for tourism (e.g., wildebeest, zebra) occurat a very broad scale, incentive schemes arebeing established to encourage landholdersnot to block migration routes with fences orfields.

On the other hand, the establishment ofconservation areas can cause the loss ofkey resource areas or migration routes forhunter-gatherers or agro-pastoralists. As withwild species, the accessible areas of key


resources and the connectivity among themare both slow variables with thresholds that,if exceeded, threaten the viability of partic-ular social–ecological systems. Some of thetraditional uses by hunter-gatherers that aredisallowed in conservation areas may beentirely compatible with conservation and canbe restored through co-management arrange-ments between government and indigenous res-idents, as in some Australian national parks(see Chapter 6).

In sparser rangelands, the number and place-ment of artificial water points are criticallyimportant to the management of forage for pas-toralism, such that more water points enablethe landscape to be grazed more evenly. Graz-ing intensity by domestic stock is highest nearthe water and declines as a power functionwith distance, affecting both landscape func-tion and biodiversity (Ludwig et al. 2004).Improvements in borehole technologies andcheaper water distribution using plastic pipehave caused rapid water point development inAustralia. Consequently, fewer areas are dis-tant from water, leaving less habitat in whichgrazing-sensitive species can persist (Jameset al. 1999, Landsberg et al. 2003). In thisway, slow developments by individual pas-toralists can cause a regional threshold to bereached suddenly, causing widespread biodi-versity losses because no water-remote refugiaremain. Parallel issues occur with respect toborehole provision in the Sahel.

The establishment and management ofregional biodiversity conservation systems arefurther complicated by actual and poten-tial climatic changes. In the absence of cli-matic change a resilience-based approach mightattempt to maintain a conserved area or sys-tem of conserved areas within a particular basinof attraction (see Chapters 1 and 5). Climaticchange causes the basin of attraction to shiftso that the functionality of the conserved areais no longer possible in its current location.When basins of attraction shift along a cli-matic gradient, policies and management mustadapt to these changing conditions or disap-pear entirely. Alternatively, triage may encour-age the reallocation of resources to achieve thefeasible, rather than futile, attempts to main-

tain species and ecological communities thatcannot adapt.

Intensification and Abandonment

Population increase leads to intensificationand fragmentation (with increasing demandsfor services), whereas population declineslead to abandonment or consolidation (cou-pled with withdrawal of services; Fig. 8.5).These different trajectories create distinc-tive interactions with policy issues in dry-lands of developed and developing nation(Campbell et al. 2000).

The size of the human population in a regionstrongly influences the area cropped and thenumbers of stock. One threshold that is crossedas population (a slow variable) increases is thenumber of people who can be supported bypastoralism alone. Once crossed, cropping ormixed farming may become obligatory since itcan support more persons per unit area thancan pastoralism (Ruthenberg et al. 1980). Thuspopulation increase can lead to sedenterizationof human populations that had previously beenmobile (often also promoted by governmentsto enhance the provision of services such aseducation, health and energy, and perhaps con-trol of the population). Drylands of intermedi-ate aridity are the most susceptible to degrada-tion thresholds; this is because they can supportopportunistic cropping, but, being economicallyand ecologically marginal, their low productiv-ity does not justify the establishment of secureproperty rights and investment in sustainablecropping (Safriel et al. 2005; Fig. 8.1). Crop-ping has a much larger negative impact on land-scape function than grazing because it exposessoils to erosion and temperature increase, lossof soil structure, and nutrient loss by runoffand harvesting (Hiernaux and Turner 2002),so landscape function declines. The traditionalmethod of maintaining landscape function, con-trolling weeds, and accumulating soil moisture,is fallowing, which declines when there is exces-sive demand for land under pressure of pop-ulation (Ruthenberg et al. 1980). Mixed farm-ing and manuring offers some remediation, or

8 Drylands 183

fertilizer can be purchased if crops are grownfor sale, but with attendant risks of acidifica-tion and soil structural decline. More inten-sive cropping, as practiced at the wetter end ofthe gradient, tends to have less impact becauseits higher productivity is more likely to justifyhigh labor or capital inputs for soil conservation(Ruthenberg et al. 1980).

In drylands of developed countries, decliningpopulations lead to the withdrawal of key ser-vices, such as health, education, and financialservices. Each service is likely to have a thresh-old of population size below which provisionis deemed by politicians or public servants tobe not cost-effective. Potential policy responsesinclude triage, in which public funds are notinvested in communities that are not viable, butcitizens are assisted in moving to more sociallyand economically viable locations, and consoli-dation of services in selected settlements. Thesetrends parallel patterns observed in remote arc-tic and boreal regions where increasing costsassociated with climate change and/or ruraldepopulation threaten the viability of ruralindigenous communities (Chapin et al. 2008).In developing countries where populations areincreasing, such thresholds for service provisionare likely to be reversed, moving from cost-ineffective to cost-effective as demand grows.The extent to which the services are providedto a region, whether in a developed or a devel-oping country, will depend largely on the polit-ical influence of the region (Sandford 1983,Abel et al. 2006).

Population increase provides more laborfor land management, promotes innovation,and can foster cooperation, so degradation isnot inevitable (Tiffin and Mortimore 1994).Social networks also increase with populationgrowth, perhaps enhancing the potential to sup-port those affected by drought and facilitat-ing the exchange of information about resourceavailability in times of scarcity. Technolog-ical changes such as public electricity sup-plies may have positive impacts on woodlandresources around villages through reducing fire-wood collection. They also allow access to tele-vision and telephones that bring new commer-cial, political, and social influences to remoteplaces.

Loss of Local Community GovernanceStructures

Governance (Ostrom 2005; see Chapter 4)includes:

• the formal and informal social rules (e.g.,property rights) and norms (e.g., equity) thatguide the behavior of individuals and groupstoward one another and their access to anduse of resources and services;

• the political system that makes and changesthe rules and the policies through which theyare implemented; and

• organizations and social networks thatimplement and monitor compliance with therules.

Governance operates across levels from local(e.g., a social network that implements rulesabout reciprocal obligations during drought)to regional, national (e.g., national policy ondrought support for drylands), and global (e.g.,attempts to establish carbon trading to mit-igate climatic change). Interactions of gover-nance rules and organizations across scales arecritically important to the resilience of dry-land regions and the well-being of its occupants(Abel et al. 2006). Cross-scale networks linkindividuals, groups, and organizations in remoteregions to metropolitan governments so thatcalls for infrastructure, services, and droughtrelief are heard. The response depends on thepolitical influence of the remote region, which,because of its remoteness, distant voice, andsparse population, is often weak. The problemcan be exacerbated by differences in the eth-nicity of the members of government and thepublic service, on one hand, and the ethnicity ofpopulations in the dryland region, on the other(Sandford 1983).

Government interventions, coupled with theglobalizing effects of communications andtransport technologies, can disrupt regionaland local governance processes. Early in thetrajectory represented by Fig. 8.5, formaliza-tion and centralization of governance tends toweaken traditional governance that evolved tocope with spatiotemporal variability in rainfall(Sandford 1983). It commonly involves priva-tization of group property rights; provision of


water points, clinics, and schools; and increasedrule of law. The net benefits of these changesdepend on many details, but the process ofchange is often disruptive. For example, col-lectivization was imposed in the USSR onsome nomadic peoples and livestock ownershipappropriated by the state. This policy failedbecause the collectives were sedentary and non-viable in the face of spatiotemporal variability.Sedenterization policies, widely applied in theMiddle East and Africa in the 1970s and 1980s(Sandford 1983), also commonly failed for sim-ilar reasons. In some African rangelands, triballand was excised for small collective groups ofpastoralists, but these “group ranches” wereusually too small to accommodate spatiotem-poral variability. In the Americas, Australia,and Africa, drylands were taken from indige-nous peoples and privatized by invading Euro-peans. In some regions, there has been a suc-cessful effort to rebuild group property rights orat least some regional peer group governanceof resources (e.g., LandCare in Australia; theDesert Community Initiative in Niger).

Privatization (fragmentation) of commu-nal land, commonly following a US ranchingmodel, does not necessarily increase produc-tivity per unit area, measured in physical ormonetary terms, nor the number of peoplethat land can support (Abel 1997). In Australiastate-legislated subdivision of initially exten-sive holdings created units that lacked both thefinancial and the ecological resilience of largerholdings (Young 1985). Policies now promotereconsolidation, but even large units are stillbelow the threshold of property size (a slowvariable) needed to encompass broad-scale spa-tiotemporal variations in forage. In such casesagistment, which is widely practiced in Aus-tralia, has become an important means of redis-tributing livestock at regional and interregionalscales (McAllister et al. 2006). In the practiceof agistment, private properties that have toomany animals for the available forage in a par-ticular year will transport stock to other prop-erties where the imbalance is in reverse, andthe senders pay the receivers. In other yearsand in the long run the imbalance and the pay-ments may be reversed, thus reducing pressureson areas where rain has not fallen. Other enter-

prises have achieved the same effect by owningland in climatically different regions and mov-ing stock among them. Both approaches usecommercial arrangements to mimic traditionalsystems of reciprocal obligations and mobilitythat have been practiced by Aboriginal hunter-gatherers (e.g., Keen 2004) and pastoralists inAfrica and Asia (Sandford 1983, Alimaev andBehnke 2008).

Disruptive Subsidies

At national and international levels a commonresponse to spatiotemporal variation has beenpolicies directed to drought relief. In Australiapastoralists have received support for animalfeed and debt and tax relief during droughts.Many of these measures create perverse incen-tives to retain stock in poor times and causedamage to the land (Stafford Smith 2003).Today some of these measures persist, althoughemergency relief is more directed toward trans-port subsidies to support moving animals, andsome better tax instruments that allow pro-ducers to build up financial reserves in goodyears that are not taxed until they are used.In developing countries famines have triggeredinternational food and medical aid. While thefood has saved lives during the emergency, itslong-term benefits are questionable. In manycases the problem is not lack of food at anational scale that leads to famine, but inabilityof drought-struck people to pay for it. In suchcases it would be better to subsidize the pur-chase of food from farmers within the nationand thus stimulate investment and production(see Chapter 12).

Subsidy of drylands can be ecological as wellas economic. The intensively used parts of thelandscape (often the majority at the moister endof the drylands spectrum) generally depend onimported subsidies of nutrients and chemicalpest and weed control to maintain their produc-tivity. Subsidies can make these areas more pro-ductive when they function well, but highly vul-nerable if these subsidies are withdrawn (as intimes of war or economic recession). Also, fer-tilizer use can lower pH (a slow variable) belowa critical threshold, or irrigation can increase

8 Drylands 185

soil salinity above a threshold so that yields andground cover decline. Because most drylandcrops are annuals, soil is exposed to erosion andthe risk of irreversible loss of landscape func-tion if poorly managed. Pesticides can eliminatepredators and foster resistance in pests, bothchanges reducing the self-organizing capacityof the system and reducing its resilience. Thepolicy implications are clear, but ethically andpolitically difficult, because a commitment tohigh-input development can increase shorterterm welfare substantially, but at the cost oflonger term vulnerability. When populationsare increasing, policy makers may feel theyhave no choice but to support intensification(see Chapter 3). However, the resilience of ahuman–ecological system stems largely from itscapacity to self-organize and recover from adisturbance. Although rebuilding this capacityat times requires access to external resources,excessive subsidy can reduce the incentives andcapacity to self-organize. Cross-scale subsidies,whether of nutrients or through drought relief,increase vulnerability at the local scale and,if feasible, should end when self-organizationbecomes apparent. If many regions are subsi-dized, this can reduce the resilience of the sys-tem as a whole (Abel et al. 2006).

Commercialization and Diversification

All regions are affected by the increasing con-nectedness of the world, but globalization cre-ates particular threats and opportunities fordrylands. Communications and transport tech-nologies provide dryland inhabitants accessto new markets. Thus aboriginal art is soldacross the Internet from remote Australia,animal products are marketed internationally,Botswana beef is sold to the European Union,and nature-based tourism and hunting experi-ences attract international customers. All bringnew money into remote communities. Withsuch changes, of course, comes a market ori-entation, which in developing countries oftenbrings social, cultural, and technological thresh-old changes.

Commercialization can disrupt traditionalcultures by transforming the social relations of

production. For example, a shift from the pro-duction of milk for subsistence, which is tra-ditionally managed by women for their fami-lies, to meat for sale, which is commonly man-aged by men, requires different herd struc-tures and stocking rates. This can alter fam-ily nutrition and vulnerabilities to market fluc-tuations. Alternatively, traditional cultures thatare unwilling or unable to participate in mar-kets may lose political influence and becomemarginalized. Likewise, producing crops forsale often requires the use of fertilizer andpesticides to raise yields above those neededfor subsistence. This increases dependence oninput markets which are themselves vulnerableto fuel price increases and wars. It is evidentthat many thresholds of potential concern canarise from entry into new markets, which wisepolicy makers would consider carefully.

Changes from complex mixes of livestockand other bush products in a subsistence soci-ety to a market orientation toward maximumproduction of one or two saleable products likebeef can have significant flow-on effects to theecosystem as livestock forage needs becomefocused, and complementary mixes of browsers(goats, camels, and browsing wildlife) and graz-ers (cattle, sheep, and grazing wildlife) arereplaced by herds of one commercial species.

New markets can diversify income sourcesand enhance resilience. Income generated on-or off-farm from nonagro-pastoral activitiessuch as a service industry (tourism, for exam-ple) may be more stable or vary asynchronouslywith an agro-pastoral income that is drivenby rainfall. If an alternative or supplementaryincome is from a source not dependent onregional rainfall, such as mining or tourism,it enables farmers and pastoralists to survivedrought and rebuild assets afterwards. Suchdiversification is not new, for many traditionalsocieties in arid lands are accustomed to usingwildlife during times of scarcity.

In some regions cropping and livestock pro-duction by pastoralists are being replaced byemployment in tourism, conservation, special-ist native products, and mining. In easternand southern Africa, for example, commer-cial livestock production has in some casesbeen replaced by or is managed alongside


commercial production of wildlife for sporthunting, meat production, or tourism. In somecases this reverses a colonial process in whichpastoralists and indigenous hunter-gathererswere excluded from the then newly estab-lished national parks. A reduced focus on live-stock production may mean increases in woodyvegetation and other landscape-scale changeswith beneficial implications for landscape func-tion, carbon storage, and energy feedbacks toclimate.

Case Studies

To illustrate the issues raised by an analysisof resilience in different dryland systems, wenow present four case studies (locations shownin Plate 6) that provide examples from devel-oped and developing nations (capturing differ-ing population trajectories, production goals,and connectivity with markets and govern-ments) in different environments, and at a vari-ety of different scales.

A Crop–Livestock System inthe Southern Sahel, Western Niger

Mixed crop–livestock systems in western Nigerillustrate the strong interdependence of thesocial and biophysical subsystems and ways inwhich increasing pressure of use can lead to thecrossing of thresholds in both (Fernandez et al.2002, Hiernaux and Turner 2002). About 15%of this arid region (300–500 mm annual rainfall)is rangeland on nonarable soil, and the remain-der is either active cultivation or fallow land onsandy, infertile soils. The crops are mostly staplecereals for subsistence with some other crops,and livestock are mainly cattle, sheep, and goatswith a few donkeys, camels, and horses used fordraft.

The evolved system of land use is quite com-plex, involving two kinds of households withdifferent livelihoods: a village household (VH,about 70% of the people) and a camp house-hold (CH, about 30%). VHs have primaryrights to arable land. CHs have mostly livestockand undertake wet season transhumance (sea-

sonal movements) to northern areas for live-stock forage. The VH have fewer livestock,because they cannot maintain them year roundon the available forage. The CH livestock usethe fallow and cropped lands in the dry season,providing fertilizer in the form of manure.

Soil fertility is a critical limiting factor forcrop production. It declines under cropping andin the absence of manure from livestock; 3 outof 8 years must be in fallow to maintain long-term fertility. This translates to a minimum of3 out of every 8 hectares in fallow at any onetime. This “3/8” value is a biophysical thresh-old in the system. Below that value, fertilitydeclines; above it the system’s soil fertility isself-sustaining.

Finally, in the rangeland component of thesystem, soils have a higher clay content. Underheavy grazing they are prone to erosion andcompaction, as perennial grass cover gives wayto a sparse cover of small annuals, constrainingthe number of animals and time spent grazingon rangelands. Three indices of sustainabilitycan be defined for this system:

1. Soil fertility is sustained unless land isfarmed more than 5 years out of 8. If manureor other sources of nutrients are added, lessfallow is needed.

2. Grazing is sustained unless there is insuffi-cient palatable herbage to meet intake needsthroughout the season.

3. Household economy is sustained if theaggregate production of crops, dairy prod-ucts, and livestock exceeds the minimumthreshold need for these by a household.

The mixed economy of farmers and camppastoralists is viable as long as none of thesethresholds is exceeded, which requires coopera-tion among the two types of households. If VHsleave too little fallow land in order to increasecrop yields, then soil fertility declines and camppastoralists are forced to spend more timeon rangelands, moving the system toward twothresholds of unsustainability. Alternatively, iflands are left fallow for too long, soil fertil-ity increases but total harvest declines, lead-ing toward a threshold of economic unsustain-ability. Interventions that enhance the system’sadaptive capacity might reverse this trend, for

8 Drylands 187

example, through access to inorganic fertilizers,crop diversification, agro-forestry, and increas-ing farmers’ husbandry skills (human capital).

An important factor in the dynamics of thissystem is the strong linkage between the house-hold and the community scales. Available for-age for livestock depends directly on commu-nity livestock density and reciprocity agree-ments that determine this. In the dry sea-son, the VH depend on the CH for live-stock manure, and the CH depend on VH foraccess to crop and fallow land. The systemas a whole depends strongly on the continuedaccess by CH to grazing resources outside thearea.

Learning from a Centuryof Degradation Episodes in Australia

Eight well-documented degradation episodesassociated with major droughts have occurredin different regions of Australian rangelandsbetween 1880 and 1990 (McKeon et al. 2004). Ametaanalysis of seven of these (Stafford Smithet al. 2007) illustrates the importance of focus-ing on slowly changing state variables in envi-ronments where fast variables are very poorindicators of their slow counterparts. The anal-ysis also enriches our understanding of waysin which local environmental knowledge linkshuman and environmental subsystems at multi-ple scales.

The Australian rangelands constitute threequarters of the Australian land mass, about halfof which is grazed by cattle or sheep underleasehold pastoral systems. Most productivelands are subject to high climatic variability onmultiple time frames, including intra- and inter-annual fluctuations driven in part by ENSOevents with return times of about 4–8 years andby longer term drivers (e.g., the InterdecadalPacific Oscillation) with mean return times ofabout 20 years (White et al. 2003). An analy-sis of social (process and policy) and biophysi-cal (particularly climate and pasture condition)drivers (McKeon et al. 2004; Fig. 8.7) revealedthe following syndrome that was common to allevents:

• [–10 years] Good prices, good rains, animalnumbers increase; human expectations rise.Decline in pasture condition causes reducedresilience in short dry periods.

• [0 year] Onset of severe drought and/ormarket decline; stock retained in hope ofrain/price increase, sometimes encouragedby perverse policy incentives.

• [+2–3 years] Extended dry period revealsdecline in pasture condition more dramati-cally as pasture production collapses; furtherdamage, big stock losses.

• [+5–7 years] Eventual relief through rains,but a more-or-less permanent decline inpasture condition becomes evident in sub-sequent short dry periods. Local learningby individual pastoralists does occur, oftenresulting in reduced stocking rates. Formalgovernment inquiries report on events, butthis occurs too late for intervention in thepresent drought.

• [+20 years] Pastoralists turn over; mem-ory loss in pastoral and policy communitiesoccurs over the next 20 years.

Both industry and government organizationsfailed to recognize key slow variables and theirthresholds. Pastoralists responded to pasturegrowth (and good prices) rather than landcondition (and market trends); governmentsresponded to actual drought events rather thanclimatic cycles or the adaptive capacity of indus-try. Land condition in many cases had crosseda critical threshold before the drought, butthis was only evident once the severe droughtepisodes had begun. By this time it was too late.

There were evident cross-scale influencesaffecting pastoral decision-making, in particu-lar those of markets being driven by nationaland global changes and the influence of policyin creating perverse incentives to retain live-stock in drought times (Stafford Smith 2003).The influence of scale is particularly importantwhen we consider learning. Almost all episodeswere followed by clear evidence of some learn-ing by individual pastoralists. However, therewas little evidence that this learning persistedfrom one drought event to the next, becausethe local knowledge appeared to be too local(Stafford Smith et al. 2007). The long return


External driversand shocks

(e.g. market drop,new policy)

Effects ofdecision-making

(especially stocking rates and tactics in drought)

External driversand shocks

(e.g.. drought,climate change)

Local environmentalknowledge about

environment sub-systemcapabilities

and responses

H - E

E - H

Products of ecosystemservices (forage

production and stability )

Evolving humansub-system (changing

technology, institutionsand human capital)

Evolving environmentsub-system (changing

forage and animalproduction systems)

Human Environ-ment

Figure 8.7. Linking thesocial and ecologicalaspects of an Australianpastoral grazing system.Redrawn from StaffordSmith et al. (2007).

time of major drought events, on the orderof 20 years, was of the same magnitude asthe longevity of a pastoralist’s managing life-time, so the inevitable turnover of managersmeant that their learning was lost by thetime another major drought occurred. ‘Local’learning in this case needed to occur at aregional scale, with a large enough communityinvolved and with the support of policy andresearch so that it would persist over manydecades.

This case study constitutes a “natural exper-iment” of eight regional degradation episodesthat occurred under broadly similar policy andsocial settings over a century in the Australianrangelands. Long-term impacts on pasture pro-duction of significance to household livelihoodswere documented in events that were triggeredby drought but really driven by overoptimistichuman expectations of the environment andmarkets in the years preceding the drought.Both managers and governments were respond-ing to the wrong variables, and governmentssometimes exacerbated the situation with per-verse policy incentives that caused pastoraliststo hold onto stock even longer. Most impor-tantly, this case study illustrates that the crit-

ical scale for learning (DDP P5) may not bevery local, where the controlling variables (inthis case, particularly drought) are operatingat larger scales in time and space. Encourag-ingly, after a century, Australia has been movingtoward a more regional alliance between indus-try, research, and government to deliver peer-supported “safe carrying capacity” estimates topastoralists, monthly drought alerts from thescientists, and hopefully a better policy contextfrom government (Stafford Smith et al. 2007).

Immigration and Off-Site Impacts inInner Mongolia

Between 1949 and 1997 cross-scale interac-tions have caused a pronounced co-evolutionbetween the human and the environmentalsubsystems and influences of Inner Mongo-lia (Jiang 2002). The population of the Uxinbanner on the Mu Us Sandy Land on theOrdos Plateau has tripled, the cash incomeper capita has risen 50-fold, pastureland hasbeen privatized to the household, and irre-versible changes in household consumption andmarket-oriented aspirations have occurred inthe population. During that period, one policy

8 Drylands 189

aimed at making the existing livestock produc-tion hazard-resistant and able to support thegrowing population through forage shrub plan-tations and irrigated cropping. This policy suc-ceeded to the extent that the region is now self-sufficient in winter forage, and stock mortali-ties have dropped from 8–17% to 1.5%. Theirrigated cropping now produces mostly cashcrops but often remains in proximity to plan-tations that help to protect against sand move-ments. Irrigated cropping practices originatedwith in-migrating Chinese Han people but arealso spreading to the indigenous Mongolians.During the same period, a second policy aimedat stabilizing sand dunes, the mobilization ofwhich was the main indicator of desertifica-tion in the region. The North China Revegeta-tion Program sought to plant trees, shrubs, andgrasses on the dunes, with periodic injectionsof effort from the central government in Bei-jing, which is affected by sandstorms from theregion.

Unfortunately, the irrigation appears to bedrawing down the shallow water table in theregion. The landscape is increasingly polarizedand consolidated into irrigated areas with highcover and production, and sandy areas witha declining water table and worsening sandmovement. Intermediate grasslands are declin-ing in productivity and extent. Human aspi-rations for production in the region will notdecline. The issue is now therefore whetherthe local understanding of the water table-mediated connection between higher over-all production and increasing area of mobilesands will develop to the point that actionon the trade-off in values can be taken.Regardless of the social and environmen-tal drivers at the household level, there areemerging regional impacts that must now beaddressed. The latest efforts in revegetationwere driven by Beijing as a result of sand-storms affecting the capital that might affectChina’s reputation during its hosting of the 2008Olympics – international expectations play-ing through a national government to poli-cies that are implemented at regional and locallevels.

As Jiang (2002: 182) notes in her analysisof the effects of cultural change on desertifi-

cation, “Aiming at a moving target, the dry-land ecosystem does not exhibit recovery buttransition toward a new state.” Just as drylandecosystems are disequilibrial, so too are theirembedded social–ecological systems. Uxin hasessentially crossed mixed social and environ-mental thresholds and been transformed intoa new type of system (Walker and Meyers2004), perhaps not yet in balance. The optionof recrossing the thresholds back to the originalregime no longer seems possible or desirable:the new system has reduced livestock deathsand improved household economies (Fig. 8.7).With this change in state have come funda-mental changes in system functioning. At onetime these revolved around grazing and itslocal impacts on grass cover, which at timesbecame too low, causing the mobilization ofsand. Today, the more important functionalityis groundwater for irrigation and the impactof water withdrawal on the water table fortrees and shrubs. With the change in systemhas come a change in the controlling slowvariables. Previously the important variable tomonitor was land condition for pasture produc-tion (and sand mobilization) but today watertable levels are more important in the new sys-tem state. Drivers contributing to the changeinclude population increases, a change from col-lective ownership when pastureland was dis-tributed to households in 1984, and the culturaland technological changes associated with irri-gated agriculture, all of these ultimately drivenby policies from another scale. Survival in thefuture depends on whether the community candevelop new local knowledge and solutions forgroundwater-mediated landscape linkages.

Managing Under Uncertainty in theKruger National Park, South Africa

The Kruger National Park (KNP) illustrates theintegration of all of the issues discussed abovein the multiscaled planning and managementof a major multiple-use dryland region (Biggsand Rogers 2003, du Toit et al. 2003). KNPspans 22,000 km2 of land in northeast SouthAfrica. Across its area, mean rainfall rangesfrom 300 to 700 mm yr–1. At least three extreme


droughts have occurred between 1981 and 2000.The high heterogeneity of its rocks, landforms,and soils is reflected in the diversity of the struc-ture, composition, and functioning of its eco-logical communities (du Toit et al. 2003) andits resulting value for the conservation of bio-diversity. The Park is famous for its large pop-ulations of elephant, buffalo, hippopotamus,rhino, lion, leopard, cheetah, and antelope, butit also conserves an array of other biota. How-ever, this diversity is also a consequence ofits resource scarcity and sparse human popu-lations, because the area is relatively unpro-ductive for farming and supports trypanosome-bearing tsetse fly that make it unsuitable forlivestock. Its ecological communities are con-sequently relatively less simplified by humansthan those in more productive areas. What isnow the KNP was nonetheless established onland and among cultures that have seen wavesof social–ecological changes during which thearea was subject to different burning, resourceuse, and disease regimes that changed the struc-ture, composition, and functioning of its ecosys-tems. The stone-age San hunter-gatherer peo-ples were overrun around 400 AD by iron-agefarming and trading cultures that exported gold,ivory, and other wildlife products. Afrikaanerand British colonizers hunted the area heav-ily from the 1830s until rinderpest, introducedby Europeans to Africa, drastically reduced thenumbers of susceptible large herbivore speciesin 1896. Formal protection of wildlife in whatis now the KNP began in 1902. Protection fromhunting and recovery from rinderpest enabledthe recovery of large herbivore populations.The KNP was established in 1926 (du Toit et al.2003).

Management and use of this relativelyremote area has been strongly influenced bynational policies. In 1948 the National Party waselected to govern South Africa by a white elec-torate and remained in power until 1994. Itsapartheid regime attempted to run the econ-omy and society with races segregated. Itspolicies applied to the KNP too, which wasrun partly to foster the cultural identity ofAfrikaner people. Physical and psychologicalseparation of nonwhite peoples from the KNPremains a problem today, though a de jure mul-

tiracial democratic society was established in1994 (du Toit et al. 2003). The future of the KNPis tied by multiple cross-scale political, cultural,and economic links to the future of the evolvingSouth African nation. It is now surrounded bya variety of land uses, including towns, mining,agriculture, forestry, and wildlife utilization. Itdepends on water from catchments that supportlarge human populations. Its mobile wildlifepopulations, including elephant, move across itspermeable boundaries into neighboring Zim-babwe and Mozambique. It is linked to interna-tional markets through a major tourist industry,and neighboring mines export minerals and usewater.

The KNP has inherited abundant opportu-nities as well as multiple problems from itscultural and ecological past. At the nationalscale, the importance of biodiversity conserva-tion and tourism has to be considered along-side the extremes of poverty, health, and edu-cational opportunities of the population as awhole. Around the KNP, traditional users ofthe land that became the Park press claimto historical rights. Competition for water isalready being addressed by explicit recogni-tion in South African water policy of the tran-spiration of water from each land use, andthe rights to water of the users. This has con-sequences for river flow in the Park and itsaquatic ecosystems. Within the Park the dynam-ics of plants, animals, and fire drive large andoften unpredictable changes in the compo-sition, structure, and functioning of ecologi-cal communities. KNP’s policy makers, man-agers, and users thus face high levels ofuncertainty stemming from the unpredictabil-ity of the social–ecological system at severalscales. This case study describes the adap-tive framework that guides their managementstrategies.

South Africa’s history is consistent withAdaptive Cycle theory (Gunderson andHolling 2002), in which the pre-1994 racistsociety had become increasingly rule-bound,hierarchical, and resistant to new ideas. At aregional scale, the KNP-management systemreflected a similar pattern. The rest of thiscase study is drawn from Biggs and Rogers(2003), who describe how the management

8 Drylands 191

system subsequently adapted. They describethe science-management system of the KNP inthe years shortly before democratic and socialchanges of 1990 as:

• treating nature as linear and predictable;• compartmentalizing knowledge along disci-

plinary grounds and around particular prob-lems;

• insular and autocratic in its decision-making;• neglecting biodiversity and heterogeneity in

its management.

Since then the science-management systemhas moved in directions that acknowledgethe rapid change and high uncertainty at thenational scale, the multiple policy objectives ofa multiracial and economically heterogeneoussociety, and its interaction with a dynamic andheterogeneous ecosystem. Here we describethe current science-management approach –called Strategic Adaptive Management - undersix headings: ecological theory; objectiveshierarchy; thresholds of potential concern;monitoring and adaptive decision making;communities of practice; and outcomes andchallenges.

Ecological Theory. The concept of hierar-chical patch dynamics is the scientific basisfor management interventions in the structure,composition, and processes of the KNP’s eco-logical communities. The ecological system as awhole is seen as a set of nested and interactingpatches. This is consistent with panarchy the-ory (see Chapter 1), with its nested but inter-linked subsystems, each operating at differenttemporal and spatial scales. Management inter-ventions in KNP are made at particular times,particular spatial scales, and in particular waysto maintain or enhance this spatial and tem-poral heterogeneity, both because it is consid-ered fundamental to ecological functioning, andbecause of the intrinsic and market values ofthe biodiversity that it fosters. Gillson and Duf-fin (2007) show how pollen analyses spanning5,000 years can guide thresholds of probableconcern (TPCs – defined below) for vegetationstructure, thus influencing TPCs for the Strate-gic Adaptive Management of elephant densi-ties and fire. That said, the dynamics of thepatches are not well understood, many pro-

cesses are nonlinear and often unpredictable,and the ecological system as a whole is in con-stant flux. Lack of ecological predictability rein-forces the need for adaptive management (seeChapter 4) that would in any case be requiredto address the social, economic, and politicaluncertainties in the broader social–ecologicalsystem (Fig. 8.8).

Vision Statement and Objectives Hierarchy. Avision statement for the KNP, developed withpublic participation, is to “. . .maintain biodi-versity in all its natural facets and fluxes andto provide human benefits in keeping with themission of the South African National Parksin a manner which detracts as little as possi-ble from the wilderness qualities of the KNP”(Biggs and Rogers 2003, p. 60). A hierarchyof objectives has been developed to link theabstract vision to management activities at ahierarchy of scales. Main groups of objectivesare atmospheric, aquatic, terrestrial research,terrestrial management, terrestrial monitor-ing, alien species management, human bene-fits, wilderness, and integrated environmentalmanagement.

Thresholds of Probable Concern. TPCs aremanagement goals that together define currentviews about the spatiotemporal heterogeneityconditions of ecosystems (Biggs and Rogers2003). TPCs are upper and lower levels inselected indicators. When a level is reached, ormodeling predicts it is likely to be reached, rea-sons are investigated, and management actionis applied, if necessary. Alternatively, the levelof the TPC may be reset if it is thought tohave been wrongly set. TPCs are hypothe-ses about the limits of acceptable change inecosystem structure, composition, and function.As hypotheses, they are necessarily modified,replaced, or discarded as new evidence dictates(Fig. 8.8). As a set, they form the multidimen-sional envelope within which variation of theecosystem is acceptable under the vision state-ment. In terms of resilience theory (see Chap-ter 1), it is thus a practical definition of a basinof attraction for the ecological elements of thesocial–ecological system (Table 8.3).

Monitoring and Adaptive Decision-Making.The TPCs indicate what management actionsshould be done, where, and when. The choice


TPC exceeded?

Continuemonitoring

Objectivesachieved?

Adjust TPCvalue

Consider a variety of policy options

Changegovernance

No

No

No

Yes

Yes

Yes

External surprisesor new paradigm

Change values,institutions, and / or

paradigm

Double-loop learning

Triple-loop learning

Single-loop learning

Reassessment: Will current norms and mental models

allow broad objectives to be achieved by taking

corrective action?

Take corrective action, choosing

from a variety of options

Implementmanagement

options

Single-loop learning

(limited opportunity for learning)

Newmanagement

options

Figure 8.8. Strategic approach to river manage-ment that integrates indicators, endpoints, and val-ues in the Kruger National Park. Thresholds of prob-able concern (TPCs) define the acceptable levels ofheterogeneity. If the system remains within theselimits, monitoring continues. If management objec-tives are met despite the TPC being exceeded, thethreshold value is changed, but monitoring contin-ues. If the TPC is exceeded and management failsto meet its objectives, a more fundamental reassess-

ment occurs, leading to a modification or invention ofnew management strategies. Sometimes this reeval-uation leads to an entirely new paradigm or newunderstanding, especially if the process has been trig-gered by unexpected events or is considered in thecontext of a new paradigm. These more fundamentalreevaluations often require a change in governance,involving new sets of actors (Biggs and Rogers 2003).See also Fig. 5.1 on single-, double-, and triple-looplearning.

of action is made, acknowledging uncertaintiesabout outcomes. The TPCs also define whatis monitored and where, so that managers canlearn whether or not their actions maintainedthe ecological system within the TPCs, and

whether the TPC levels, or indeed the TPCsthemselves, were appropriate.

Communities of Practice. Sharing of knowl-edge among managers, stakeholders, andscientists is fundamental to the implementation

Table 8.3. Examples of thresholds of probable concern (TPCs). Adapted from Biggs and Rogers (2003).

Trigger or lead-up TPC exceeded or expected to be exceeded

Silt release episode fromupstream mining operation

Turbidity, resulting in fish kills

Fires started by humans Area burned by fires lit by humans exceeded TPCRiver sedimentation Modeled geomorphic and riparian species indicatorsOccurrence of an alien plant

species along riversNew occurrence of an alien with a high index of potential threat.

8 Drylands 193

of this approach (Biggs and Rogers 2003),supporting the principle of local ecologicalknowledge as critical in maintaining the func-tions and coadaptation of social–ecologicalsystems in drylands. Sharing of views andknowledge in the KNP takes place withinself-organizing communities of practice, whichare networks that link across organizational,disciplinary, and interest-group boundaries, andbetween research, practice, and management(see Chapter 4). In addition to breaking downsuch boundaries, they enhance collective andindividual learning. Formal research processesare thus linked to the local knowledge ofmanagers and stakeholders in a dynamiclearning system that adapts to new problems,opportunities, and knowledge.

Outcomes and Challenges. According toBiggs and Rogers (2003), the application ofStrategic Adaptive Management in the KNPhas resulted in, for example:

• More proactive behavior – for example,actions taken now to forestall sedimentationof rivers, even though these effects are notpredicted to reach unacceptable levels formany years hence;

• Stronger links to external stakeholders – forexample, use of poverty-relief programs toclear alien species and address the externalcause of river sedimentation (mining);

• Acknowledgment and use of conflictingviews about the causes of and remedies forTPCs being exceeded – for example, in themanagement of rare Sable Antelope;

• Revision of fire policy when it was realizedthat the frequency of unintended fires wouldexceed its TPC regardless of managementaction.

Among the challenges of applying StrategicAdaptive Management, Biggs and Rogers iden-tify:

• Need to interpret and sift multiple sourcesof information, of varying quality and oftenconflicting, and synthesize it for monitoringand for setting or resetting TPCs;

• Interactions with programs that operate out-side the Strategic Adaptive Managementframework, such as veterinary control of dis-

eases that affect both wildlife and livestock,with which park managers are legally obligedto comply;

• Keeping the TPCs to a number that achievesthe KNP mission and can be monitored andresponded to with available resources;

• Difficulty of knowing when monitoring dataare signaling an impending crossing of a TPCthreshold.

The KNP’s Strategic Adaptive Managementsystem has continued to learn and develop sincepublication of the du Toit et al. (2003) bookon which this case study was based. It is animportant model for policies and managementfor biodiversity conservation in general, as wellas in drylands, and the concept of TPCs is beingadvocated and applied in Australia at least.

Synthesis

Thanks to their underlying features of anunpredictable climate and resource scarcity,drylands tend to possess sparse populations, beremote from markets, and distant from centersof governance. The analysis of issues in drylandsis therefore most likely to be adequate if car-ried out through the lens of the five principlesof the Dryland Development Paradigm (DDP– Reynolds et al. 2007). Our case studies haveillustrated the importance of each, in particular:

• The coevolutionary nature of social andecological systems, such that system col-lapse principally occurs when this relation-ship becomes dysfunctional, not just becausethere is change (DDP P1);

• The need to focus very carefully on theappropriate slow variables and their thresh-olds in order to determine the state of thiscoevolutionary system as a matter of par-ticular importance in variable environments(DDP P2, P3);

• The massive effect that cross-scale interac-tions can have on dryland systems that areusually particularly poorly equipped to dealwith these because of their distant voice(DDP P4); and,

• The vital importance of the right sharedmental models in the form of local


knowledge at a variety of scales for main-taining the functionality of the coupledsystem – particularly important in drylandswhere variability slows down experientiallearning (DDP P5).

These lessons are true for all levels ofengagement in drylands, from local man-agement, through regional research efforts,to national and international policies. Themarginal nature of productivity in drylands usu-ally means that they also end up being politi-cally marginal. The challenge for dryland inhab-itants is to accept this as inevitable and focus onmanaging it by creating strong links to the out-side world that can be activated readily in timesof need. By contrast, the challenge for (well-meaning but distant) policy makers is to payparticular attention to the form of institutionalarrangements in drylands, to support local gov-ernance where this is possible, to create ade-quate links to other regions, and to put extraeffort into hearing the articulation of needsfrom those regions that otherwise get swampedby demands of more heavily populated areas.

In a number of dryland regions around theworld an unfortunate outcome of the dry-lands syndrome is that cross-scale effects cou-pled with inherently low adaptive capacity hasresulted in changes to undesirable, but resilient,system regimes. The efforts of people to sur-vive and prosper have led to declines in ecosys-tem productivity; further efforts seem to makethings worse, in terms of both ecosystem healthand human well-being. For these regions, theproblem is no longer how to increase resilience,but how to increase transformability (see Chap-ter 5). The need is to facilitate transformationfrom the kinds of systems they now are to someother kind of system. This may entail chang-ing the ways people make a living, develop-ing new “products” (goods and services) andoperating at different scales. There are no clearrules for how to do this, although recognitionof the dryland syndrome is a useful prerequi-site for effective action. It requires strong lead-ership and vision, a high level of social capitalto get agreement on what may be difficult orpainful changes, and most likely financial andother support from higher scales. Transforma-

tion and transformability are emerging as crit-ical areas for research within the broad areaof resilience science, with particular relevanceand poignancy for marginalized peoples indrylands.

Review Questions

1. Why are drylands more likely than manyother systems to undergo changes in state?

2. What social changes in the last century havereduced the resilience of many pastoral soci-eties?

3. What are some of the major causes of deser-tification, and how do these interact? Whatpolicy interventions might reduce the likeli-hood of desertification?

4. Why has it been difficult for landowners inAustralian drylands to learn how to copewith drought?

5. What key attributes of drylands need to beconsidered when analyzing policy or man-agement responses to drylands problems?

6. Which principles of resilience theory aremost important in drylands, and why?

7. Why is an analysis across scales particularlyimportant in drylands?

8. What is the role of local knowledge in man-aging dryland social–ecological systems?

Additional Readings

Biggs, H.C. and K.H. Rogers. 2003. An adaptivesystem to link science, monitoring, and manage-ment in practice. Pages 59–80 in J.T. du Toit, K.H.Rogers, and H.C. Biggs, editors. The Kruger Expe-rience: Ecology and Management of Savanna Het-erogeneity. Island Press, Washington.

Galvin, K.A., R.S. Reid, R.H.J. Behnke, and N.T.Hobbs, editors. 2008. Fragmentation in Semi-Aridand Arid Landscapes: Consequences for Humanand Natural Systems. Springer, Dordrecht.

Geist, H.J., and E.F. Lambin. 2004. Dynamic causalpatterns of desertification. BioScience 53:817–829.

Reynolds, J.F. and D.M. Stafford Smith, editors.2002. Global Desertification: Do Humans CauseDeserts? Dahlem University Press, Berlin.

Reynolds, J.F., D.M. Stafford Smith, E.F.Lambin, B.L. Turner, II, M. Mortimore, et al.

8 Drylands 195

2007. Global desertification: Building a sciencefor dryland development. Science 316: 847–851.

Safriel, U., Z. Adeel, D. Niemeijer, J. Puigdefabre-gas, R. White, et al. 2005. Dryland Systems. Pages623–662 in R. Hassan, R. Scholes, and N. Ash, edi-tors. Ecosystems and Human Well-Being: CurrentState and Trends. Millennium Ecosystem Assess-ment. Island Press, Washington.

Walker B.H. 1993. Rangeland ecology: Understand-ing and managing change. Ambio 22(2–3) 80–87.

Walker, B.H., and M.A. Janssen. 2002. Rangelands,pastoralists and governments: Interlinked sys-tems of people and nature. Philosophical Trans-actions of the Royal Society of London, Series B357:719–725.

9Freshwaters: Managing Across Scalesin Space and Time

Stephen R. Carpenter and Reinette Biggs

Introduction

Freshwaters include groundwater, rivers, lakes,and reservoirs. These ecosystems representabout 7% of earth’s terrestrial surface area.Although aquatic and terrestrial ecosystemsappear clearly separate to the human eye,groundwaters, lakes, and rivers are in factclosely connected to terrestrial systems (Mag-nuson et al. 2006). Climate, soils, and water-usecharacteristics of terrestrial plants affect infil-tration of water to groundwater and runoff tosurface waters. Terrestrial systems contributenutrients and organic matter to freshwater sys-tems. Rivers in flood fertilize their valleys. Ter-restrial organisms are eaten by aquatic ones,and vice versa. A natural unit for consid-ering coupled terrestrial and aquatic ecosys-tems is the watershed. Within a watershed,ecosystems are closely linked through flowsof water, dissolved chemicals, including nutri-ents and organic matter, and movements of

S.R. Carpenter (�)Center for Limnology, University of Wisconsin,Madison, WI 53706, USAe-mail: [email protected]

organisms. Thus watersheds are natural units ofanalysis for freshwater resources.

People are closely connected to freshwatersof the watersheds where they live. Histori-cally, people depended on water for drink-ing, washing, agriculture, and transportation.Rivers and lakes have been important routesfor travel and commerce for thousands ofyears. Today, people also use freshwaters atunprecedented scales for industrial processesas well as irrigation of agriculture for foodproduction. Freshwaters are therefore criti-cal in supporting modern society. Given theclose connections between human and eco-logical aspects, management of freshwaters islikely to be most effective if they are treatedas social–ecological systems. Freshwaters exem-plify several features of complex systems thatmake them fascinating to study and difficultto manage: spatial heterogeneity of flows, mix-tures of fast and slow variables, and thresholds(Box 9.1).


198 S.R. Carpenter and R. Biggs

Box 9.1. Key Principles for Freshwaters

Like other social–ecological systems dis-cussed in this book, freshwaters supporthuman activities and are altered by humanaction. Freshwater social–ecological systemsprovide particularly good examples of threeprinciples that occur frequently in thisbook: spatial heterogeneity and flows acrossboundaries, fast and slow turnover times, andthresholds.

Spatial heterogeneity and flows acrossboundaries. Groundwater, lakes, and riverswithin a watershed are interconnected,exchanging water, nutrients, organic matter,and organisms. Human action alters theseconnections in many ways. People changeland use and land cover, thereby alteringflows from land to freshwater. People changewater movements with channels, levees, anddams. People extract water for irrigation andindustrial and household uses, and peopleharvest aquatic organisms such as rice or fish.Many of the changes in freshwater ecosys-tems and their services are a result of humanalteration of spatial flows of water, solutes,and organisms.

Fast and slow turnover times. Turnovertime is the amount of time that it takesto replace the pool of a material in anecosystem, if the pool size is steady overtime. If a reservoir holds 1 km3 of water,and the annual inflow is 0.5 km3, then theturnover time of water in the reservoir is2 years. Turnover times vary widely amongthe components of freshwater ecosystems.Water turnover times range from millennia ingroundwater to years in lakes to days or min-

utes in streams. Phosphorus turnover timesrange from centuries in soil, to decades insediments, to years in fish, to millisecondsfor phosphate in surface waters. Freshwa-ter food webs also span a wide range ofturnover times, from years for fish biomassto about a day for plankton biomass. Dif-ferences in turnover times sometimes leadto interesting nonlinearities in ecosystemdynamics such as alternate stable states ofaquatic communities, shifts in nutrient lim-itation, alternate states of water quality,and trophic cascades (Carpenter and Turner2000, Carpenter 2003).

Thresholds. Sharp changes in flows, feed-backs, or the state of a system occur at thresh-olds. For example, a light rain on an agricul-tural field will be absorbed by the soil. Asthe intensity of rainfall and water saturationof the soil increases, a threshold is exceeded,causing some of the water to runoff as surfaceflow, carrying soil particles and nutrients intostreams or lakes. Some important thresholdsfor freshwater ecosystems affect dominanceof primary production by rooted plants ver-sus phytoplankton, clear water versus turbidwater, outbreaks of invasive species or water-borne diseases, and large changes in rela-tive abundance of harvested fishes (Schefferet al. 2001, Carpenter 2003). Thresholds alsooccur in human behavior and institutionsfor ecosystem management (Brock 2006).Thresholds are important because systembehavior on one side of a threshold is a poorguide to system behavior on the other side ofthe threshold.

Sustainability of freshwater social–ecologicalsystems involves many issues, of which three areparamount:

1. People use freshwater intensively. In manyregions, the rate of water use by peopleexceeds the rate of supply by the hydro-logic cycle. Extractive use of water com-petes with in-stream uses of water to meethuman needs (pollution dilution, transporta-

tion, fish and game, recreation) and ecosys-tem needs (water for support of terrestrialecosystem services).

2. Freshwater ecosystems are frequentlydegraded by changes in chemical or biologi-cal drivers. Pollution, especially runoff fromagriculture and urban areas and sewagedischarge, causes eutrophication (nutrientenrichment) and sanitation-related disease.In terms of the numbers of people impacted

9 Freshwaters 199

globally, degraded water quality is as severea problem as insufficient water supply.Freshwater ecosystems, like oceanic islands,also tend to have strong internal ecologicalfeedbacks because changes in species traitsstrongly influence ecosystem dynamics. Thusspecies invasions and overfishing can havestrong effects on freshwater ecosystems.

3. Competition for scarce water resources canlead to conflicts and requires managementof tradeoffs. Many kinds of institutions havedeveloped to manage freshwater, dependingon local ecosystem conditions, the social andhistorical context, and the state of knowl-edge about the system. Climate change andrapidly-rising human demand for water areincreasing the pressures on these institutionsas well as ecosystems.

The chapter begins by describing freshwa-ter resources, their use by people, and theecosystem services that freshwaters provide.Next we describe the degradation of freshwaterresources and the drivers of degradation, suchas land use, chemical pollution, habitat loss, andspecies invasion. We then turn to challengesof managing freshwater resources that occurcommonly in case after case – conflict betweenupstream and downstream users, human ver-sus environmental flows, managing for hetero-geneity, and equity across human generations.Such conflicts can be addressed by institutionalprocesses, such as markets, trade in virtualwater, conservation, technological innovation,or rescaling of management institutions. Thechapter closes with a short summary.

Human Use of FreshwaterResources

Water is an essential and nonsubstitutableresource for people and ecosystems. This criti-cal resource is, however, rare and patchily dis-tributed on earth. Only about 2.5% of all wateron the planet is freshwater. Less than 1% ofEarth’s freshwater occurs in lakes, rivers, reser-voirs, or groundwater shallow enough to betapped at affordable cost (Vorosmarty et al.2005). Some freshwater occurs as soil water

available to plants, but most of the remain-ing water occurs as ice or deep aquifers thatare not available to people or other organisms(Fig. 9.1).

Freshwater supports human societiesand ecosystems in two main forms: “bluewater” and “green water” (Falkenmark andRockstrom 2004). Blue water is what wetypically think of when we consider waterresources: liquid water in rivers, lakes, reser-voirs, and groundwater aquifers. Blue wateris important for a host of essential services,including drinking water and sanitation, foodproduction through irrigation, transport, andenergy production. Blue water also supportsa diverse array of aquatic ecosystems. Greenwater is the moisture in the soil that supportsall nonirrigated vegetation, including rainfedcrops, pastures, timber, and terrestrial naturalvegetation. Green water flows via evaporationand transpiration exceed blue water flows inrivers and aquifers (Fig. 9.2).

Available Freshwater Resources

Flows of freshwater are more important thanstorages when considering the freshwater avail-able for use by humans and ecosystems(Fig. 9.1). For example, the total amount ofblue water stored in all the world’s rivers isabout 2,000 km3, much less than the annualwithdrawal of 3,800 km3. A more appropriatemeasure of water availability is the 45,500 km3

that is discharged annually by the Earth’s rivers(Oki and Kanae 2006). Water that is replen-ished by rainfall or snowfall can be regarded asa renewable resource. Renewable water sourcesare available on a sustainable basis provided therate of extraction does not exceed the replen-ishment rate, and adequate water quality ismaintained. Water quality is usually definedby its desired end use. Water for drinking,recreation, and habitat for aquatic organisms,for example, requires higher levels of puritythan water for hydropower. Nonreplenishablestocks of freshwater such as fossil groundwateraquifers are nonrenewable in the same way asoil resources.

Freshwater replenishment rates vary sub-stantially among ecosystems and over time.


Snowfall12.5

Total terrestrialprecipitation

111

Rainfall98.5

Total terrestrialevapotranspiration

65.5

Glaciers and snow

24,064

Net water vaporflux transport

45.5

Precipitationover ocean

391

Groundwater

23,400

River

Reservoirs

7

Wetlands

17

Lakes

175

Soilmoisture

17

Subsurfacerunoff30.2

Perma-frost

300

3

Water vapor over land

Surfacerunoff15.3

2

Biologicalwater

Flux 103 km3 yr –1

Storage, 103 km

10

Evaporationover ocean

436.5

Sea

1,338,000

Domestic

Industry

45.5

0.770.38

Abstraction and return flows

Irrigatedcropland

2.66

1

Water vapor over sea

Figure 9.1. Global hydrological fluxes (1,000 km3

yr–1) and storages (1,000 km3). The direct ground-water discharge, which is estimated to be about

10% of total river discharge globally, is includedin river discharge. Adapted from Oki and Kanae(2006).

The water of a lake in the moist tropics maybe replaced several times per year, whereasaccessible groundwater of a dryland steppemay be replaced only once every few decades.The sustainable rate of extraction for twoequal-size bodies of water will therefore bevery different in these two systems. Addition-ally, many drier parts of the world experi-ence substantial interannual variability in pre-cipitation. What may be a replenishable rateof extraction in a particular year or decademay not be sustainable in another. Such vari-ability poses particular challenges for waterresources policy and management because itdemands high levels of adaptability. Lastly, notall renewable water is available for peopleor ecosystems. Flood waters may flow awaybefore the water can be used, or ground-water may become polluted and unusable.Different watersheds therefore tend to eachhave their own particular set of managementchallenges.

Globally about 3,800 km3 of blue water iswithdrawn annually for human use. Althoughthis represents only 10% of the maximumrenewable freshwater resource (Oki and Kanae2006), high variability in water supplies overspace and time means that extraction inmany regions exceeds renewable supplies (seePlate 7). Overextraction has resulted in dra-matic decreases in the extent of several largeaquatic ecosystems, including the Aral Sea,the Mesopotamian Marshes, and Lake Chad(Fig. 9.3). Approximately 700 km3 of the totalannual withdrawal comes from groundwater,and much of this is extracted from fossilaquifers or at rates substantially greater thanthe rate of recharge. To stabilize the spatial andtemporal variability in water supplies, humanshave substantially altered the flows of fresh-water by constructing dams and reservoirs andthrough interbasin transfers. At any point intime, the world’s man-made reservoirs nowhold three to six times the total amount of

9 Freshwaters 201

Blu

e w

ater

Gre

en w

ater

Eco

syst

em s

ervi

ces

Unused

Uti

li-za

ble

No

n

uti

lizab

le

Used

Ind

irec

t g

reen

wat

erD

irec

t g

reen

wat

er

Food (4%)

Grazing (18%)

Grasslands (11%)

Forest woodlands (17%)

Wetlands (1%)

Arid lands (5%)Lake evaporation (1%)Evaporation reservoirs (0.7%)Others (5%)

Storm runoff (27%)

Instream ecology (5%)

Available for consumptive use (3%)

Food-consumptive use (1.5%)

Domestic and industry (1%)

Food non-consumptive use (0.7%)

Figure 9.2. Blue-water and green-water flows usedto support global ecosystem services, based on datain Rockstrom et al. (1999). Direct green water is

the water directly used by food-production systems;indirect green water benefits society through otherecosystem services.

water in all the world’s rivers (Vorosmarty et al.2005). The total flow of water being divertedwithout return to its stream of origin in Canadaalone was 140 km3 in the 1980s, more than themean annual discharge of the Nile (Vorosmartyet al. 2005).

Society’s management of freshwater islargely organized around the different humanuses of freshwater. While this may be practicalto some degree, it means that different agenciesare commonly responsible for managing differ-ent parts of the hydrological cycle. For example,water utilities are usually responsible for pro-viding drinking water, agricultural departmentsfor providing water for agriculture, and envi-ronmental departments for maintaining aquaticecosystems and water quality. Managementof freshwater systems therefore tends to befragmented.

Freshwater for Drinkingand Sanitation

Access to adequate water for drinking and san-itation is a basic human need, and was declareda human right by the UN in 2002. The min-imum drinking water requirement for humansurvival is 3–5 l per person per day, dependingon climate (Gleick 1996). Drinking water mustbe sufficiently clean to prevent water-relateddiseases such as cholera. The major determi-nant of clean drinking water is access to waterfor sanitation. About 1.7 million people, pri-marily young children, die each year from dis-eases associated with inadequate sanitation andhygiene (Vorosmarty et al. 2005). If water forbasic sanitation, hygiene, and food prepara-tion is taken into account, each person needsat least 20–50 l of clean water each day to


1997

1963 1973 1987

2001

Niger

NigeriaCameroon

Chad

Wetland vegetation

Water

Extent of the lake in 1963

Land

Figure 9.3. Lake Chad was once one of Africa’slargest bodies of fresh water, similar in size to NorthAmerica’s Lake Erie. Lake Chad has decreaseddramatically to about 5% of its original size dueto growing human demand, especially for large-scale irrigation projects, combined with a seriesof droughts. Most of the change in Lake Chadhappened over the 15-year period from 1973 to1987. The drier climate and reduced size of the

lake has compromised the irrigation farmers, fish-ermen who depended on the lake fishery, and pas-toralists whose cattle grazed in the area surround-ing the lake. Some of these people have movedto cities or other regions, where they have oftenjoined the ranks of the jobless or come into con-flict with host communities. Redrawn from http://www.grida.no/climate/vitalafrica/english/14.htm

survive. The minimum requirement varies dueto differences in climate, culture, technology,and lifestyle. Providing basic freshwater needsdepends both on availability of water in theecosystems where people live and on humanknowledge and infrastructure to extract, trans-port, and clean the water.

Much of humanity still lives with inadequatewater and sanitation supplies, a major factoraffecting well-being and limiting development.For each dollar invested in improved water sup-ply and sanitation, an estimated return of $3–34can be expected on account of averted deaths,reduced health care costs, days gained fromreduced illness, and reduced time spent obtain-ing and transporting water (Hutton and Haller2004). Nevertheless, in 2004, approximately 1.1billion people (17% of the world’s population)lacked access to clean water, and 2.6 billion(40% of the global population) lacked access toimproved sanitation (Vorosmarty et al. 2005).Many governments are too poor or dysfunc-tional to make the expensive infrastructuralinvestments required to deliver these basic

services. This is particularly true where localwater supplies have become exhausted or pol-luted, and water supply costs have risen dra-matically. For example, in parts of Bangladeshgroundwater supplies have become contami-nated with arsenic. Arsenic occurs naturally inmany soils, but becomes toxic when exposedto the atmosphere when groundwater tablesare lowered through excessive extraction. Asdiscussed in the section on pollution, devel-oped and developing countries alike also faceincreasing problems with chemical pollution ofdrinking water from industrial and agriculturalwaste (see Chapter 12).

Freshwater for Food Production

Irrigation is by far the largest user of blue water,accounting for 70% of all blue water with-drawals. These withdrawals support 40% of theworld’s crop production on 18% of the world’scropland (Cassman et al. 2005). Advantages ofirrigated agriculture are that crop yields arehigher than for rainfed crops, and production is

9 Freshwaters 203

less susceptible to the vagaries of climate. Theexpansion of irrigated agriculture together withdramatic improvements in transportation hashelped reduce the incidence of severe faminesand malnutrition despite large increases inhuman populations (see Chapter 12). Expand-ing irrigated agriculture in very poor partsof the world such as Africa presents one ofthe largest opportunities for addressing poverty(Shah et al. 2000). However, as discussed later,irrigation is also a major cause of freshwaterand environmental degradation. Furthermore,in some areas irrigation withdrawals far exceedreplenishment rates. The sustainability of irri-gation practices in many parts of the world istherefore of concern (Foster and Chilton 2003,Vorosmarty et al. 2005).

Blue water is also critical in providing food inthe form of fish. Inland fisheries are of particu-lar importance in developing countries becausefish are often the only source of animal pro-tein. However, overharvesting in combinationwith habitat degradation and invasive speciespose substantial threats to freshwater fisheries(Finlayson et al. 2005). Several well-studiedaquatic ecosystems such as the North AmericanGreat Lakes showed dramatic declines in nativefish populations in the twentieth century. Thesedeclines have been related to a combination offactors, including the establishment of invasivespecies such as sea lamprey and alewife, over-harvesting, severed migration routes, and stock-ing of exotic sport fish such as Pacific salmonids.In recent years, the production of inland fish hasbecome dominated by aquaculture, which is thefastest growing food production sector in theworld. Although aquaculture may add a sourceof protein to food supplies, this benefit must bebalanced against adverse environmental effects.Aquaculture facilities are foci for disease, con-centrate pollutants that cause eutrophication,and may increase harvest of wild fishes used asfood (Naylor et al. 2000).

Green water plays a major role in food pro-duction. Almost 40% of the world’s terrestrialsurface is now used for food production: 16 mil-lion km2 for croplands (82% of which is rain-fed), 18 million km2 for managed pastures, and16 million km2 as rangeland. Evapotranspira-tion is estimated to be 7,600 km3 yr–1 from crop-

land and 14,400 km3 yr–1 from permanent graz-ing land used for livestock production (Oki andKanae 2006). Together, cropland and grazingland therefore account for one third of total ter-restrial evapotranspiration. About 40% of theworld’s population depends directly on theseagricultural systems for their livelihoods, withthe proportion increasing to over 60% of thepopulation in poorer parts of the world.

Other Services Supportedby Freshwater

Besides water and food, freshwater ecosystemsprovide a host of other services that sustain mod-ern societies. The most important of these areecological processes that maintain the environ-ment and resources on which people depend.Aquatic ecosystems such as lakes and wetlandsplay an important role in regulating water flow.They help attenuate floods, recharge groundwa-ter, and maintain river flow during dry periodsby releasing water stored during wet periods.Such hydrological regulation reduces the needforexpensiveengineeredfloodcontrolandwaterstorage infrastructure. The vegetation in manyinland waters traps sediments, nutrients, andpollutants such as heavy metals. This helps main-tain water that is of adequate quality for drink-ing and irrigation without the need for expen-sive water treatment. Trapping sediments andbreaking down pollutants also reduces degrada-tion of downstream habitats that are importantfor fish production. Wetlands, and peatlands inparticular, store exceptionally large quantities oforganic carbon per unit area and thereby con-tribute to climate regulation (Finlayson et al.2005, Cole et al. 2005). The value of the regu-lating services provided by functioning naturalaquatic ecosystems is often not recognized untilthey are degraded or destroyed.

By supporting aquatic and terrestrial ecosys-tems, freshwater provides important nonfoodproducts such as fiber, construction timber, andenergy. Nonfood crops such as fiber (e.g., cot-ton for clothes and textiles), biofuels, medicines,pharmaceutical products, dyes, chemicals, tim-ber, and nonfood industrial raw materialsaccount for almost 7% of the world’s harvestedcrop area (Cassman et al. 2005). The area used


for biofuel production could grow substantiallyas petroleum supplies decline or nations seekto decrease their dependence on petroleum-producing nations. In poorer parts of the worldsuch as sub-Saharan Africa, over 80% of thepopulation depends on fuelwood and charcoalfrom natural vegetation for domestic cookingand heating. Peatlands, a type of wetland, havebeen mined extensively for domestic and indus-trial fuel, particularly in Western Europe. Todaypeat mining for use in horticulture is a multimil-lion dollar industry in Europe (Finlayson andMcCay 1998).

Blue water directly supports nonfood sec-tors of modern society by providing water forindustry, electricity generation, and transport.About 21% of blue water withdrawals globallyare used for industrial purposes, which accountfor 32% of global economic activity (CIA2007). Freshwater is also important for elec-tricity production, both through hydropower(17% of global electricity production) and innuclear and coal power plants (16 and 66%of global electricity production, respectively)(EIA 2006). Blue water in rivers and lakeshas long played a major role in transporta-tion. Although cargo traffic has substantiallydeclined due to aviation, railroads, and truck-ing, cargo shipping remains important in areassuch as the North American Great Lakes, theRhine River in Europe, and the Yangtze Riverin China.

Freshwater ecosystems provide manycultural and recreational ecosystem services.People are drawn to water. Lakes, rivers, orwetlands are associated with sacred sites orreligious activities in many cultures and haveinspired artists. Many freshwater ecosystemsare today protected as National Parks, WorldHeritage sites, or Wetlands of InternationalImportance (Ramsar Sites). Income generatedby recreation and tourism associated withfreshwater can be a significant component ofnational and local economies. This includesactivities such as water sports, recreationalriver cruises, and recreational fishing. Theeducational value of freshwater systems, andwetlands in particular, is closely associatedwith recreation. For example, approximately160,000 people visit a 40-ha wetland complex inthe heart of London each year. Similar to many

such centers around the world, this site offersan educational exhibition center and activitieswithin a recreational setting with boardwalks,hides, and pathways (Finlayson et al. 2005).

Degradation of FreshwaterEcosystems

The Millennium Ecosystem Assessment(2005a) concludes that “It is established butincomplete that inland water ecosystems are inworse condition overall than any other broadecosystem type.” Freshwater ecosystem condi-tion has been degraded by draining of wetlands,fragmentation of rivers through constructionof dams and canals, pollution by fertilizers andtoxic chemicals, invasion by harmful exoticspecies, and overfishing (Fig. 9.4).

In preindustrial times, transfers across water-shed boundaries were few, and flows betweenwatersheds were usually small compared toflows within watersheds. Flows between water-sheds occurred in the form of critical nutrients(in dust) and microorganisms moving throughthe atmosphere and mobile animals such asinsects, birds, and mammals moving amongwatersheds. In modern times, people have dra-matically increased the rates of flows betweenwatersheds by moving water, fertilizers, agri-cultural products, and people and by increas-ing the airborne transport of nutrients and con-taminants. These changes have contributed toalterations in the ecological processes withinfreshwater ecosystems and have expanded theimpacts of human activities to far beyond theparticular watershed in which people live.

Draining, Fragmentation,and Habitat Loss

Extensive areas of wetlands have been drainedto make way for agriculture or urban devel-opment. By 1985, losses of wetland area sincethe industrial revolution were 56–65% in NorthAmerica and Europe, 27% in Asia, 6% inSouth America, and 2% in Africa (MEA2005a). Rates of loss have declined steeply inmany rich nations and are rising in some poornations. Degradation of watershed vegetation,

9 Freshwaters 205

Non-point source

pollution

Afforestation and

deforestation

Dams

River channelization

Roads andflood control infrastructure

Urban and industrial

point sourcepollution

Airborne pollution

Groundwaterextraction

Large-scale irrigationand river diversions

Overharvesting of wild

resources

Intoduction of exotic species

Figure 9.4. Ways in which human use affects the water cycle and freshwater ecosystems. Modified fromFinlayson et al. (2005).

shorelands, and wetlands increases the prob-ability of damaging floods and reduces thenatural storage of water for release duringdry periods (MEA 2005a). Floods typicallyaffect the poorest people most severely, thougheven rich countries suffer devastating floodsthat are exacerbated by degradation of wet-lands and terrestrial vegetation (Kundzewiczand Schellnhuber 2004). For example, large-scale clearing of savanna in Brazil was associ-ated with a 28% increase in wet-season flows(Costa et al. 2003).

Impervious land cover, such as buildings,roads, or packed earth, prevents infiltration ofprecipitation into soil water or groundwater.Instead, most of the precipitation runs off theland as surface flow. Consequently, water inputsto streams and lakes are more variable and havehigher concentrations of sediment, phosphorus,and contaminants that are associated with sed-

iment particles. Agriculture and some urbanland uses (such as fertilized lawns) increasethe nutrient content of runoff. Reduced infil-tration contributes to declining groundwatertables, so that groundwater accounts for asmaller proportion of water input to streamsand lakes. The net effect is more variablewater levels, poorer water quality, and lossof springs and small streams that are crucialhabitat for many plants and animals, includingspawning fish.

Dam construction has fragmented about40% of the world’s major river systems.The number of large reservoirs has increasedfrom about 5,000 in 1950 to about 45,000 in2000 (Vorosmarty et al. 2005). This estimateexcludes many small impoundments createdfor local use by farmers or small municipali-ties. While dam construction has contributedto economic development, food security, and


flood control, it has also had substantial envi-ronmental, social, and human health impacts.Dams have had enormous effects on the watercycle and hence on aquatic habitats, suspendedsediment, carbon fluxes, and waste process-ing. In some cases, substantial amounts ofwater are transferred among watersheds, alter-ing the water balance of extensive areas. Largedam projects often involve the displacementand resettlement of people, sometimes causingenormous social disruption. In addition to los-ing their livelihoods and cultural and histori-cal heritage, resettled communities often sufferfrom a marginalized status and cultural and eco-nomic conflicts with the communities in whichthey are settled.

Loss of habitat through draining and frag-mentation is the principal driver of biologi-cal change in freshwater ecosystems. Habitatloss can be caused by changing climate or landuse or by directly engineering waterways forhydropower, transportation, or shoreline stabi-lization. Some habitat losses are obvious. Damsand levees, for example, prevent movement offishes and are an important factor in loss offish species from river networks (Rahel 2000).Other habitat losses are subtle or indirect.Clearing of trees and brush along shorelinesdecreases the numbers of fallen trees in lakes,thereby decreasing habitat for fishes and their

prey and ultimately decreasing fish production(Sass et al. 2006).

Eutrophication, Salinization,and Chemical Pollution

Many freshwater supplies are degraded byeutrophication, chemical pollution, and salin-ization. Eutrophication leads to cloudy water,blooms of cyanobacteria (some of them toxic),oxygen depletion, fish kills, and taste and odorproblems that impair the use of water for drink-ing. In some regions of the world, sewagedischarge is a major cause of eutrophication.In addition to degrading water quality, inade-quate sewage treatment is an important causeof human disease. Most of the world’s wealth-ier countries have adequate facilities for treat-ing sewage before it is discharged to surfacewaters. However, as sewage discharges cameunder control, eutrophication in many placesremained stable or increased because of non-point pollution (Box 9.2). Nonpoint pollutionis nutrients or toxins scoured by runoff fromagricultural or urban lands. Nonpoint pollutionaccounts for more than 80% of the nitrogenand phosphorus inputs to the surface waters ofthe USA and is the major source of pollutantsto surface waters in many countries (Carpenteret al. 1998b).

Box 9.2. Managing Lake Mendota

When Wisconsin’s legislature in 1836selected a location for the capital city, Madi-son, they chose a gorgeous site on an isthmusbetween two lakes (Mollenhoff 2004; Fig.9.5). A reporter visiting from Massachusettsnoted that he could “see the drifts of whitesand far down to the transparent depths.” Bythe 1870s, the fertile savannas of the water-shed had been converted almost entirelyto agriculture, and the formerly white lakebottoms were covered with a thick layer ofblack soil eroded from the uplands (Lath-rop 1992a). By the 1880s, when E.A. Birgearrived to launch America’s first limnol-ogy program at the University of Wisconsin,the lakes were plagued in summer by algaeblooms and fish kills. Carp were introduced

and exacerbated the lakes’ water-qualityproblems. Eutrophication and carp led tochanges in rooted aquatic plants and loss ofsome fish species that depended on theseplants for habitat (Magnuson and Lathrop1992). Other fish species were introducedto the Madison lakes from rivers or lakesthat dried out during the deep drought of the1930s (Magnuson and Lathrop 1992). Start-ing the late 1940s, intensification of agricul-ture and development brought further dete-rioration of water quality (Lathrop 1992a).During the 1950s, most of the deep-waterbenthic invertebrates disappeared from LakeMendota, the largest of the lakes (Lathrop1992b). Thick blooms of cyanobacteria werecommon in summer.

9 Freshwaters 207

Figure 9.5. Lake Mendota. http://mrsec.wisc.edu/Edetc/NSECREU/2007%20N4.jpg.

Management of the Madison lakes hasgone through a series of phases (Carpenteret al. 1998, 2006b, Carpenter and Lathrop1999). During the first half of the 1900s, man-agement focused on treating algal bloomswith direct applications of herbicides. Nui-sance rooted plants were harvested ortreated with herbicides. By the 1950s, publicconcern with water quality evoked proposalsto decrease sewage inflow to the lakes. By1971, all sanitary sewage was diverted fromthe lakes. However, water quality did notchange much. Phosphorus runoff from agri-culture, construction sites, and towns, com-bined with phosphorus recycling from sed-iments, maintained high levels of nuisancealgae in the Madison lakes. The focus ofmanagement shifted to runoff, or nonpointpollution. During the early 1980s, efforts tomanage nonpoint pollution failed because oflow participation by farmers. In 1987, thefood web of Lake Mendota was manipu-lated to increase grazing on phytoplanktonand improve water quality (Kitchell 1992,Lathrop et al. 2002). However, water qual-

ity remained highly variable and was espe-cially poor in years of high runoff. In the late1990s, a massive program was initiated to cutnonpoint pollution of Lake Mendota by half(Carpenter et al. 2006b). At the time of writ-ing, this project was still underway. Thoughthe outcome is unknown, it seems likely thatmany of its goals will be met.

Meanwhile, the Madison area faces ongo-ing changes in water quality and supply.Climate is changing, with unknown butpotentially large consequences for precipita-tion, evapotranspiration, and runoff. Devel-opment has increased the amount of imper-vious surface in the watershed and therebydecreased infiltration of water to soils andincreased runoff. As a consequence, waterlevels in the lakes have become more vari-able. New invasive species, such as zebramussel or silver carp, could become estab-lished and alter water quality. Thus manage-ment of the Madison lakes is always a workin progress. New challenges and (one hopes)new solutions are likely to continue in anongoing cycle.

Overuse of fertilizers and high densitiesof livestock are major causes of nutrientrunoff and eutrophication. Phosphorus and

nitrogen are two of the main plant nutri-ents supplied by fertilizers and have particu-larly large impacts on environmental quality


and ecosystem services. In pre-industrial times,inputs of phosphorus to earth’s terrestrialecosystems were 10–15 Tg yr-1 (Bennett et al.2001). By 2000, mining of phosphate rock andincreased weathering due to land disturbanceby people had increased these inputs to 33–39Tg yr–1 (Fig. 9.6). Most of the mined phos-phorus is used as fertilizer to grow crops. Thephosphorus in crops is fed to domestic ani-mals or people, which recycle the phosphorus inwaste or accumulate it in their bodies. Phospho-rus applied to soils as fertilizer or manure canwash into aquatic systems and cause eutrophi-cation. Globally, the fluvial transport of phos-phorus through freshwater ecosystems to thesea is about 22 Tg yr–1 at present, versus only8 Tg yr–1 before the advent of industrializedagriculture. Human additions to the phospho-rus cycle have therefore almost tripled theamount of phosphorus annually cycled throughthe world’s ecosystems.

About 20% of irrigated croplands worldwidesuffer from salinization, the accumulation ofsalts to the point that soils and vegetation aredegraded. The main cause of salinization isinputs of salt dissolved in irrigation water from

rivers or aquifers. When plants take up irriga-tion water from the soil, the salts are mostlyleft behind. In the absence of sufficient drainageto leach the excess salts deeper into the soilprofile, salts accumulate in the topsoil overtime and result in reduced crop yields. A sec-ondary cause of salinization is waterlogging.Waterlogging occurs when aquifers are unableto take up the excess irrigation water that hasdrained deeper down into the soil profile. Thiscauses the groundwater table (containing saltsleached from the topsoil) to rise to within thecrop rooting zone. Water tables may also riseas a consequence of clearing trees for agricul-ture, especially in arid regions such as Australia(see Chapter 8). The replacement of deep-rooted trees by shallow-rooted crops can dra-matically reduce the amount of water taken upand transpired, resulting in rising groundwatertables.

Chemical pollution of surface and ground-waters is of increasing concern in developedand developing countries alike. Exposure toheavy metals, long-lived synthetic compounds,and toxic substances has been linked to arange of chronic diseases, including cancer,

10 to 15

Weathering

Fluvial drainage

Fluvial drainage

Ancestral phosphorus budget

Change-in-storage =1 to 6 Tg P yr –1

Change-in-storage =10.5 to 15.5 Tg P yr –1

Freshwater and terrestrial ecosystems

of Earth

Freshwater and terrestrial ecosystems

of Earth

Present-day phosphorus budget

18.5

15 to 20

Mining

Weathering

Atmosphericoutput

Atmosphericoutput

22

8

1

1

Figure 9.6. Human impacts on erodable phosphorus. Adapted from Bennett et al. (2001).

9 Freshwaters 209

lung damage, and birth defects. Some pollu-tants may persist in aquatic systems for decadesafter they have been found to be harmful andbanned. Cleaning water of these compoundsis usually very expensive and in some casesimpossible.

Invasive Species and Overharvesting

Habitat loss interacts with other drivers of bio-logical change, including eutrophication (seeabove), species invasions or introductions(Kolar and Lodge 2000), and harvesting (Postet al. 2002). Species invasions often lead tosecondary losses of species that are eaten bythe invader or lose habitat due to expansionof the invading population. Trophic cascades

are changes driven by changes in top preda-tors that affect lower trophic levels, primaryproducers, bacteria, and nutrient cycles (Car-penter and Kitchell 1993). Cascades in aquaticecosystems can be caused by overfishing orby invasions of top predators. Trophic cas-cades change not only food webs, but also pri-mary production, sedimentation rates of nutri-ents and carbon, and rates of release of green-house gases from freshwater ecosystems. Oftenthe effects of species invasions, fish harvest-ing, and nutrient input interact, leading tomassive changes in lakes or rivers (Box 9.3).Such changes are difficult to reverse or can bereversed only slowly. More often, the changeslead to new and different ecosystems as well asnew kinds of relationships between people andfreshwaters.

Box 9.3. Lake Victoria: The Nile Perch “Gold Rush”

Lake Victoria in East Africa is the world’ssecond largest freshwater lake. It is the siteof one of the most rapid and extensive evo-lutionary radiations of vertebrates known,with most of the over 500 endemic speciesof haplochromine cichlid fish having evolvedin the last 15,000 years (Stiassny and Meyer1999, Kocher 2004). It is also the site of oneof the most recent vertebrate mass extinc-tions on the planet. Up to half the endemiccichlids, some of them never described, arebelieved to have gone extinct during the1970s and 1980s. The mass extinction ismainly linked to the population explosion ofthe Nile perch (Lates niloticus) in the mid-1980s, although other factors such as over-fishing and eutrophication also played a role(Kaufman 1992). The predatory Nile Perchwas introduced to Lake Victoria in the 1950sand 1960s by the British colonial administra-tors in an attempt to develop a productivecommercial fishery in the lake and addressthe declining native subsistence tilapia fish-ery (Pringle 2005b).

The establishment of the “perch regime”has had profound impacts on the lake’s

ecology and people. The species commu-nities, food web, and ecological processesin the lake have been radically trans-formed. Increased numbers of people in thewatershed have contributed to large-scaledeforestation and the eutrophication of thelake. These ecological changes may havemade the lake more vulnerable to inva-sion by the South American water hyacinth,which occurred in the 1990s and nega-tively impacted drinking water quality, fish-ing, tourism, and transport (Balirwa et al.2003) – but has subsequently been suc-cessfully controlled. The perch regime alsobrought significant benefits. The large, fleshyperch is highly suited for commercial exportand transformed the regional subsistenceeconomy into a cash-based economy linkedto global markets. This change undoubtedlyprovided valuable employment opportuni-ties, raised the living standards of people liv-ing around the lake, and earned valuableforeign exchange for the bordering coun-tries (Reynolds and Greboval 1988, Pringle2005a).


Today the lake region faces substan-tial challenges in simultaneously address-ing poverty and environmental degradation.There are signs that the Nile perch fish-ery is in decline. Population growth cou-pled with few formal employment opportu-nities and poor environmental enforcementhave resulted in intense fishing pressure.One result of the declining perch popula-tion is that some of the cichlids that werebelieved extinct have recently been recordedagain. On the other hand, globalization ofthe perch fishery means that lakeside com-munities now compete with consumers in theWest, and recent price increases limit localaccess to perch as a food and protein source

(Fig. 9.7). Many people also depend on thejobs centered on the Nile perch export indus-try. Transforming the current regime into onethat is ecologically and economically sus-tainable presents a considerable challenge.No “magic bullet” solution is likely to exist;rather concerted effort will be needed on anumber of fronts, such as developing alter-native livelihood options and improving fish-eries monitoring, regulation, and enforce-ment. Experience elsewhere suggests thatthese are most likely to succeed if local fish-ermen and riparian communities are activelyinvolved in developing and implementingsolutions to address the challenges they face(Dietz et al. 2003).

Figure 9.7. Fish landing on Lake Victoria, Jinja, Uganda. Photograph by Jim Kitchell.

Challenges in WaterManagement

Declines in water supply or water qualityoften lead to conflict among competing users.

These include competition among people orbetween people and ecosystems for scarcewater, as well as political conflict over mit-igation of water quality. Such conflict canlead to nonsustainable use of water supplies,

9 Freshwaters 211

development of costly alternative water sup-plies or purification methods, limitations toeconomic growth and development, pollutionand public health problems, international dis-putes in transboundary river basins, and polit-ical and civil instability (Vorosmarty et al.2005). In this section, we address four chal-lenges that frequently arise in freshwater man-agement: upstream/downstream flows, humanversus environmental flows, managing for het-erogeneity, and cross-generational equity.

Upstream/Downstream Flows

Water extraction and pollutant discharges inupstream areas clearly have direct impactsfor downstream users. They affect both waterquantity and quality, as well as ecosystem ser-vices such as coastal fisheries that are importantin many large river systems. These conflicts aresimilar to those that often exist between multi-ple users in a particular river section or arounda particular lake.

The Watershed Trust Fund of Quito,Ecuador, illustrates a potentially successfulpartnership for maintaining water suppliesand conserving biodiversity in the watershed(Postel 2005). Organized with support fromThe Nature Conservancy and the US Agencyfor International Development, the trust fundpools the demand for watershed conservationamong downstream users, including municipalusers, irrigators, industries, and hydroelectricutilities. In 2004, the trust fund mobilized morethan half a million dollars for conservationof the watershed. An important aspect of thetrust fund is the merger of water supply goalswith conservation of biodiversity. This com-bination expands the pool of participants andfunders.

Upstream/downstream conflicts can be espe-cially difficult to deal with in freshwaters thatare shared across international boundaries.There are 261 international rivers, whose water-sheds cover almost half the total land sur-face of the globe. In addition, many ground-water aquifers and large lakes and reservoirsare shared between two or more countries.Water has been a source of cross-border ten-

sion, for instance in the Middle East, southernAfrica, the US–Mexican border, and parts ofAsia, and there has been much talk of potential“water wars.” However, history suggests thatthe strength of shared interests usually inducescooperation rather than inciting violence. Inthe twentieth century 145 water-related treatieswere signed, while there were only seven minorwater-related skirmishes (Wolf 1998). Such his-torical analyses suggest that war over wateris seldom strategically rational, hydrographi-cally effective, or economically viable. Rather,cooperative water regimes established throughinternational treaties have tended to be impres-sively resilient over time, even between hostilenations that have waged conflicts over otherissues. For example, the 1960 Indus WatersTreaty between India and Pakistan has survivedmajor conflicts between the two nations.

Human versus Environmental Flows

Providing water for domestic, agricultural, andindustrial use, while also meeting the waterrequirements of aquatic ecosystems is a centralchallenge in water resource management. Asdiscussed earlier, maintaining aquatic ecosys-tems is important for the provision of a hostof ecosystem services, such as flood protection,maintaining fisheries, and regulating water-borne diseases. It can also be argued thataquatic organisms and ecosystems have intrin-sic value in and of themselves irrespective oftheir value to human society. Some peoplebelieve that humans have an ethical responsibil-ity to limit their use of freshwater resources sothat other organisms can live and thrive (Singer1993, Agar 2001).

One response to human–environmental con-flicts practiced in parts of Australia, Europe,New Zealand, North America, and SouthAfrica has been to specifically allocate waterfor environmental flows (Box 9.4). Environ-mental flows refer to the quantity, quality, andtiming of water flows considered sufficient toprotect the dependent species and the struc-ture and function of aquatic ecosystems (Kinget al. 2000, Dyson et al. 2007). Flow variabil-ity is important because different benefits are


provided by high and low flows. Low flows,for example, can exclude invasive species, whilehigh flows may provide spawning cues for fishand replenish floodplains. Environmental flowrequirements can range from 20 to 80% ofmean annual flow, depending on the river type,

species composition, river condition objectives(e.g., pristine, moderate modification, minimumflows) (Vorosmarty et al. 2005). The volumes offlow required indicate a high degree of poten-tial conflict between direct human uses and themaintenance of freshwater ecosystems.

Box 9.4. Water Policy in South Africa

The South African National Water Act intro-duced in 1998 is regarded as a landmarkin international water policy (Postel andRichter 2003). The new legislation definesthe freshwater resource as river ecosystemsrather than water, in recognition of ecosys-tem service values. The Act gives priorityto basic human needs and the needs ofaquatic ecosystems through establishment ofa “Reserve.” Instead of private ownership,the legislation is based on the legal prin-ciple of public trust, whereby the govern-ment holds certain rights and entitlementsin trust for the people and is obliged toprotect those rights for the common good.The Act treats the hydrological cycle holis-tically, recognizing that surface and ground-water flow are connected and also linked toland use. The new policy promotes participa-tory decision-making, where allocations arebased on interest-based negotiations ratherthan water rights.

The “Reserve” is one of the most innova-tive aspects of the Water Act and has twoparts: the basic human needs reserve and theecological reserve. The basic human needsreserve provides for 25 l per person per dayfor essential drinking, cooking, and sanita-tion needs. The ecological reserve is definedas the water quantity and quality requiredto protect the structure and functioning ofaquatic ecosystems in order to secure sus-tainable development. This two-part Reservehas priority over all other uses and is theonly water guaranteed as a right. Interna-tional obligations are provided for once theReserve has been met. Only after these needsare satisfied, is the remaining water allocatedto other uses such as irrigation and industry.

Another innovative aspect of the nationalwater legislation is that it is regarded as atool for transforming post-Apartheid SouthAfrica into a socially and environmentallyjust society (Funke et al. 2007). To achievethis, the legislation is based on four key prin-ciples: decentralization, equitable access, effi-ciency, and sustainability. The decentraliza-tion principle makes provision for peopleto participate in decision-making processesthat affect them and for governmental func-tions to be delegated to the lowest appro-priate level. In accordance, Catchment Man-agement Agencies (CMAs) and Water UserAssociations are being established for all themajor watersheds in the country. The CMAgoverning boards must represent all majorstakeholders and are mandated to developdetailed catchment management strategiesfor their watersheds through cooperativeapproaches. The principle of equitable accessis provided through the public trust doc-trine and an administrative licensing systemthat regulates the extraction of water for allnondomestic uses. To ensure efficiency, thesocial, economic, and environmental bene-fits and costs of competing water uses haveto be evaluated. To facilitate this process,provision is made to appeal against licensingdecisions, and economic instruments such aspricing and subsidy programs are used. Inparticular, in order to provide the “lifeline”basic human needs water supply of 6,000 lper household per month free of charge toall households, a sliding-scale pricing schemehas been adopted whereby users pay pro-gressively more per unit of additional wateruse. The two parts of the Reserve are thepillars that provide for the interlinked social,

9 Freshwaters 213

ecological, and environmental sustainabilityof the country’s freshwater resources. TheReserve thereby provides for two constitu-tional rights guaranteed to all South Africans:

the right to enough water to meet basic needsand the right to a safe environment that is suf-ficiently protected to ensure socioeconomicdevelopment and ecological sustainability.

Habitats, Fisheries, and Managingfor Heterogeneity

When relatively undisturbed, freshwaterecosystems have enormous heterogeneity.Sizes and shapes of waterbodies and drainagenetworks are similarly variable (Downinget al. 2006). Time series of discharge fromundisturbed rivers vary across a wide rangeof timescales. Freshwater organisms rangein mass over about 18 orders of magnitude,from bacteria (∼10–12 g) to the largest fresh-water fishes (∼106 g). Thus great variability– in habitat size and shape, in flow, and inorganism size – is characteristic of freshwaterecosystems.

Human intervention tends to reduce struc-tural variety of freshwater ecosystems byengineering shorelines, altering sizes of ecosys-tems using dams or drains, and harvestingspecies in certain size classes (such as largerfishes). While the role of structural variancein the function of freshwater ecosystems is notwell understood, interventions that alter habi-tat shape or the biotic size structure tend to cas-cade, affecting many species as well as ecosys-tem processes such as net ecosystem produc-tion, ecosystem respiration, gas exchange withthe atmosphere, and nutrient supply ratios. Thecapacity to absorb disturbance and maintainecosystem processes seems to be related to thevariability in habitat configuration, flow regime,and biota of freshwater ecosystems.

Maintenance of variability is crucial for man-aging freshwater ecosystems, but most manage-ment practices are designed for stabilization.Paradoxically, uniform application of such prac-tices can create severe instabilities. If fisheriesof a lake district are managed with a “one-size-fits-all” policy that stabilizes management tar-gets for all lakes, there is greater risk that spatialcascades of collapse will spread contagiously

across fisheries of many lakes (Carpenter andBrock 2004). In contrast, policies that allowlocal managers to set incentives or regulationson a lake-by-lake basis, taking account of con-ditions in neighboring lakes, tend to maintaina patchwork of fisheries that are variable yetpersistent. In other words, heterogeneous poli-cies that emerge from local decisions fosterresilience of the fisheries across the lake districtas a whole (see Chapter 1).

Humans often demand stability in freshwa-ter ecosystem services. For example, peoplewant the supply of freshwater, protection fromfloods, the capacity to carry boat traffic, orthe fish production of freshwaters to be pre-dictable and stable. Yet this expectation of sta-bility conflicts directly with the heterogeneitythat creates resilience of freshwater ecosystemservices. Designing institutions that accommo-date the heterogeneity of resilient freshwaterecosystems and the needs of people for fresh-water ecosystem services is one of the greatestchallenges in ecosystem management.

Cross-Generational Equityand Long-Term Change

Human-caused changes to aquatic ecosystemscan have long-lasting consequences. Manyspecies invasions or extirpations are difficultor impossible to reverse. Water bodies mayremain eutrophic for decades or longer, even ifexcess nutrient use ceases, because of the slowturnover of phosphorus in enriched soils andefficient recycling of excess phosphorus withinlakes and reservoirs. Altered flow regimes dueto levees and dams may also last for decades,and species or habitats lost to fragmentationmay be lost permanently. Depleted aquifersmay take decades to replenish, and pollutedgroundwaters may remain unusable for hun-


dreds or thousands of years. Thus, actions takenin the present may affect the condition offreshwater ecosystems for many generations inthe future. There is a conflict between presentuses and future options for use of freshwaterecosystems.

Some economic analyses use discountingto compare the value of natural resources inthe present and the future. Conceptually, thediscount rate is similar to the inflation rate,whereby a given amount of money in the futureis worth less than the same amount of money inthe present. Discounting often seems like a rea-sonable tool for evaluating projects when thestakes and uncertainty are modest, and there isgeneral agreement about the benefits and costsof the project.

In many cases, however, proposed uses ofwater resources involve high stakes, unknownand potentially irreversible outcomes, and con-siderable controversy about the best way toproceed. Such decisions involve judgments ofvalues or ethics that are not well represented bysimple discounting formulations or cost-benefitmethods (Ludwig et al. 2005). In polarizedpolitical contexts, each interest group may haveits own preferred “objective” analysis, and thediversity of analyses adds to the overall uncer-tainty. Interestingly, if there is broad uncer-tainty about the discount rate then on averagediscounting tends toward zero, thereby plac-ing present and future generations on the samefooting (Ludwig et al. 2005). Thus, moral judg-ments about future risk assume a central role.

Institutional Mechanismsfor Solving Conflicts

While water scarcity often generates conflict,it can also provide an opportunity to developinstitutions or technologies that enhance coop-eration and build resilience. Responses to waterscarcity are many and varied and depend onsocial and historical context, local ecosystemconditions, water infrastructure, and the stateof knowledge. Different types of responses tendto be employed depending on the phase ofthe adaptive cycle in which the local social–

ecological freshwater system finds itself (seeChapter 1). For example, markets and virtualwater trade tend to be employed most often inthe growth phase, while conservation responsesfeature prominently in the conservation phase.Technological innovation and decentralizationusually have their biggest impacts in therenewal phase.

Globally, there has been a shift in emphasisfrom increasing water supply (mainly by build-ing dams and reservoirs) to reducing demandthrough conservation or improved technology.At a global scale, this may indicate a shift fromthe growth to the conservation phase in the wayhumanity interacts with freshwater resources.

Markets and Benefit–CostComparisons

Economic studies often show substantial ben-efits from conservation of aquatic ecosystems.New York City avoided $6 billion in capi-tal costs and about $300 million per annumin operating costs for water-purification facili-ties by spending $1.5 billion in watershed pro-tection over a 10-year period (NRC 2000).When the floodplains of northeastern Nigeriawere evaluated for dams and irrigation projects,researchers compared the economic benefitsof the proposed projects with the benefits ofthe intact floodplains for agriculture, fuelwood,and fishing. When used for these purposes,water had a net economic value about 60 timesgreater than its value in the irrigation projects(Barbier and Thompson 1998). Thus economicanalysis favored conservation of the floodplain.

In theory, markets could allocate waterresources efficiently if values of water could beproperly monetized. Yet experiences with watermarkets have been mixed. In sub-SaharanAfrica, markets for drinking water have failedbecause private investments have been lowerthan expected, and in many cases users havebeen unable to meet payments (Bayliss andMcKinley 2007). The solution may lie in pub-lic utilities or in more effective partnershipsof development agencies and private indus-try. Successful water markets involve tiering ofwater rates, so that basic household uses of

9 Freshwaters 215

water are inexpensive while heavy water userssuch as industry or agriculture pay higher rates(Postel 2005).

Nutrient flows that degrade water qualitycan also exhibit remarkable diseconomies. Forexample, farming with a net yield of $4 mil-lion per annum causes losses to society of morethan $50 million per annum due to damagesto water quality of Lake Mendota, Wiscon-sin (Carpenter et al. 1999 and Box 9.2). Mar-ket mechanisms may be useful in solving suchimbalances, though often they are addressed byregulation or litigation instead.

“Cap-and-trade” has been used to createmarkets for nutrient discharge or water with-drawals. A maximum tolerable nutrient dis-charge or water withdrawal (the “cap”) isestablished by a regulatory authority. The mul-tistate commission responsible for the Murray–Darling River basin in Australia set a cap onwithdrawals to mitigate degradation of the riversystem (Postel 2005). To create a market, theregulatory authority issues marketable creditsfor pollutant discharge and water withdrawal(the “trade”). In North Carolina in the USA,state officials set a cap for phosphorus andnitrogen discharges into Pamlico Sound andallowed for trading of nutrient credits amongpolluters. A major problem with cap-and-tradeis that it does not allow for natural variationin the capacity of ecosystems to provide ser-vices. In some years the Murray–Darling Rivermay have abundant flows, and in dry yearsthe flow may be insufficient to meet the cap.In some years, Pamlico Sound may flush fre-

quently and be able to dilute larger nutrientinputs; in other years the estuary may not flushat all, and the cap may lead to dangerous algalblooms and deoxygenation. Making marketsflexible enough to cope with ecological varia-tion is a challenge with cap-and-trade.

Virtual Water and Trade

Virtual water is the volume of freshwaterneeded to produce a specific product or service(Fig. 9.8). While water requirements can be cal-culated for any product, virtual water require-ments have mainly been explored for crop andlivestock production. There is a vast mismatchbetween the weight of agricultural commodi-ties and the virtual water required for their pro-duction. For instance, producing 1 kg of grainrequires 1,000–2,000 kg (liters) of water, and1 kg of beef requires an average of 16,000 kgof water (Hoekstra and Chapagain 2007). Inwater-scarce regions, transporting water overlong distances to meet these large demands isusually too expensive. However, water-scarceregions can offset their water demand for foodproduction by importing products that requirelarge volumes of water for their productionfrom water-rich regions. Such “virtual watertrade” is estimated to be about 1,000 km3 yr–1,although not all of it is to compensate for watershortages (Allan 1998, Oki and Kanae 2006).

Virtual water trade encompasses more thantrade in water. In terms of agricultural produc-tion it involves transfers of crop nutrients. For

Potatoes

Corn

Wheat

Rice

Legumes

Milk

Eggs

Pork

Beef

Poultry

Liters of water per 500 caloriesLiters of waterper g protein

5000 4000 3000 2000 1000 0 1000

Figure 9.8. Virtual water,the liters of water neededto produce 500 calories offood or 10 g of protein.Data from Renault andWallender (2000).


instance, Africa depends on a net import ofmore than 30 million tons of cereal crops annu-ally, accounting for a net inflow of around 1.5million tons of nitrogen, phosphorus, and potas-sium to the African continent each year. Theneed for food imports is partially a result ofthe depletion of local soil nutrients in Africa,which results in low crop yields and inadequatelocal food production. Soil nutrients in muchof sub-Saharan Africa have progressively beendepleted because most farmers cannot affordor do not have access to commercial fertiliz-ers, and production demands have outstrippedthe capacity of traditional soil fertility prac-tices, such as fallow periods or applying animalmanure, to replenish soil nutrients (Cassmanet al. 2005; see Chapter 8).

While food importation may offset localwater and soil limitations, it also displaces theimpacts of crop production. The impacts ofexcess fertilizer runoff and biodiversity impactsdue to habitat loss are borne by the crop-producing country. In other regions, movementof food to feed livestock leads, through manure,to high soil nutrient levels and eutrophica-tion of downstream water bodies (Bennettet al. 1999, 2001). On the other hand, poli-cies aimed at food self-sufficiency are likely tointensify present-day patterns of water scarcityand associated conflicts. Crop production undermarginal or poorly managed conditions canalso have disproportionately large environmen-tal impacts.

Conservation

Although aggregate global withdrawals con-tinue to increase, a major feature of globalwater use is that per capita use rates have beendeclining, from around 700 m3 yr–1 in 1980 toabout 600 m3 yr–1 in 2000 (Vorosmarty et al.2005). Conservation, or increased efficiency ofwater use, offers many ways of mitigating watershortages. Most cities lose 20–50% of theirwater supplies to leaks in the distribution sys-tem or other forms of waste (Postel 2005).Leakage is highly variable among cities; Copen-hagen, Denmark, loses only 3% of its watersupply to leakage. Cities can reduce their waterloss through engineering improvements and

public education. For instance, Boston in theUSA decreased water use by 31% from 1987 to2004 despite substantial economic growth and astable population size (Postel 2005).

Food production offers the greatest oppor-tunities for increased efficiency. Irrigationdiverts 20–30% of the world’s available waterresources, but inefficiencies in distribution andapplication mean that only 40–50% of thatwater is used in crop growth. Irrigation wateris often wasted because governments subsi-dize water supply so that farmers have littleincentive to conserve. Yet technology exists tomake irrigation far more efficient, for instancethrough drip irrigation systems (Postel 1999).Production per unit water can also be improvedon rainfed lands that are not irrigated. Meth-ods vary with local conditions. Changes caninvolve the timing of seed sowing or rootingdepth of crop varieties, intercropping to cre-ate more shade on the soil surface, cultivationto promote infiltration of water, mulching, orweed control (see Chapter 12). Changes in dietscan also affect water usage. Diets high in meatdemand a lot of water, because a unit of animalproduction requires many units of plant pro-duction, as well as the water consumed by theanimal through its lifetime (Fig. 9.8).

Nutrient runoff, the major cause of waterquality degradation, can also be addressed byconservation measures. Nitrogen and phospho-rus that leave a farmer’s field in runoff are a lossto the farmer and to agricultural production andalso degrade water quality. Manure disposal isa major factor in overfertilization of croplands(Bennett et al. 1999, 2001). If manure applica-tions were adjusted to crop demand, waste ofnutrients, and damage to water quality wouldbe avoided. Also, reduced demand for meatwould decrease animal production and therebydecrease nutrient flow through manure.

Terrestrial restoration projects can improveboth water flows and water quality. Forestconservation plays a key role in maintainingthe water supply for New York city (NRC2000). The Working for Water project in SouthAfrica is a very successful land-managementprogram that seeks to simultaneously addresspressing environmental, economic, and socialissues. About 20,000 previously unemployed

9 Freshwaters 217

people are hired per year to manually clearinvasive alien plants; 60% of the jobs arereserved for women, and HIV/AIDS awarenesstraining and child care facilities are included(Magadlela and Mdzeke 2004). The impetusfor the project was the finding that many inva-sive alien plants use substantially more waterthan native species, and that clearing stands ofalien plants could substantially enhance stream-flow in water-stressed regions (Gorgens and vanWilgen 2004).

Technology

Technological advances to improve water avail-ability are likely to have their biggest impactson the largest water user, irrigation. Drip irri-gation is a method that minimizes the use ofwater and fertilizer by allowing water to dripslowly to the roots of plants, either onto the soilsurface or directly onto the root zone. Drip irri-gation methods range from high-tech, comput-erized systems to low-tech and relatively labor-intensive methods. Hydroponics is a methodof growing plants using mineral nutrient solu-tions instead of soil. Water use can be substan-tially less, and soil-borne diseases and weeds arelargely eliminated. However, hydroponic sys-tems require greater technical knowledge andmore careful monitoring than conventional soil-grown crops.

Genetic engineering of crops may haveimportant implications for future use andimpacts on freshwater resources. Geneticallymodified crops have been developed to increaseresistance to pests and harsh environmen-tal conditions such as droughts, as well asto improve shelf life and increase nutritionalvalue. Genetic engineering can therefore playa role in reducing water demand, the need forfertilizer and pesticides, as well as the amountof land needed for agriculture. However, genet-ically engineered crops and other organisms arealso associated with environmental risks, manyof which are as yet unknown.

Traditional supply-side technological mea-sures, such as interbasin transfers and build-ing dams and reservoirs, are today regardedwith some reservation (WCD 2000) but never-theless remain important, especially in devel-

oping countries. An estimated 1,500 dams areunder construction worldwide, and many moreare planned. Interbasin transfers are also likelyto remain an important mechanism for alle-viating regional water shortages (Vorosmartyet al. 2005). Knowledge about the ecologicalimpacts of dams and about ecological function-ing can help design and manage engineeringstructures so that they are less harmful. Forexample, many dams are now built with fish lad-ders that facilitate natural fish migration. Flowsfrom reservoirs can also be managed to mimicnatural flow patterns by allowing for higher andlower releases at different times of the year.

Desalinization, converting saltwater intofreshwater, is currently the most costly meansof supplying freshwater and is highly energy-intensive. Nevertheless by 2002 there were over10,000 desalinization plants in 120 countries,supplying more than 5 km3 yr–1 of freshwa-ter. About 70% of the installed desalinizationcapacity is in the oil-rich states in the Mid-dle East and North Africa (UN 2003). Whileits use may be difficult to justify for high-consumption activities such as irrigation, invest-ments in desalinization technology are likely toimprove efficiency and bring down costs, cre-ating a potentially important source of fresh-water, at least for domestic use (Gleick 2000).Adequately managing brine waste from thedesalinization process to protect nearby coastalecosystems remains an unresolved issue requir-ing special attention (Vorosmarty et al. 2005).

Rainwater harvesting, the collection andstorage of rain water from roofs and othersurfaces can be practiced through traditionalmethods or using modern technology. Tradi-tionally, rainwater has mainly been harvestedin arid and semiarid areas, providing waterfor drinking and domestic use, livestock, small-scale irrigation, and as a way to replenishground water levels. Rainwater harvesting canbe particularly beneficial in urban areas. Itaugments cities’ water supply, increases soilmoisture levels for urban greenery, facilitatesgroundwater recharge, and mitigates urbanflooding. In New Delhi, India, for instance, itis now mandatory for multistoried buildingsto have rooftop rainwater-harvesting systems(Vorosmarty et al. 2005).


Many technologies are available for improv-ing access to adequate sanitation facilities, someof which require very little or no water and lit-tle capital investment (Gleick 1996). Conven-tional high-volume flush toilets use as much as75 liters per person per day, but efficient designsthat require less than 10 liters per person perday are now available. Composting toilets andimproved pit latrines require no water otherthan for hand washing and no connection to acentral sewer.

Freshwater ecosystem restoration is an activearea of research and development (Cooke et al.2005). Although the term “restoration” seemsto imply a return to past conditions, restorationmost often involves the conversion of an ecosys-tem to a more desirable condition that has someelements of the past but is in many respectsa new kind of ecosystem (see Chapter 2).Technologies are well-developed for mitigatingeutrophication by a range of methods, includingchanges in hydrology, in-lake chemical inter-ventions, and manipulation of aquatic foodwebs. Wetland construction, habitat improve-ments, removal of harmful dams or levees,and foodweb management (through managingfisheries) are also practiced in various waysaround the world. Each freshwater restorationproject is in some respects unique, and differentcombinations of methods are appropriate fordifferent circumstances.

Decentralization and Integrationof Management

It is increasingly apparent that the complexityand specificity of issues that characterize fresh-water systems and watersheds require manage-ment approaches that are more collaborativeand integrated than in the past. In contrast toconventional resource management that tendedto be highly centralized, many regions are nowadopting freshwater management structuresthat are far more decentralized and directlyinvolve users (e.g., Box 9.4). Research on nat-ural resource governance has shown that, whenusers are genuinely engaged in deciding on therules that govern their use of the resource, theyare far more likely to follow the rules and

monitor others than when an authority sim-ply imposes the rules (Dietz et al. 2003; seeChapter 4). Such self-enforcement and moni-toring, through formal and informal local insti-tutions, can be particularly important in poorercountries that do not have the resources topolice and enforce regulations.

Users’ direct involvement in designing theresource-management rules also encouragesgreater creativity and sensitivity in the rules,which increases their effectiveness. For exam-ple, treaties around shared international water-ways often show great sensitivity to local con-ditions. In the boundary waters agreementbetween Canada and the USA that allowedfor greater hydropower generation in the Nia-gara region, both states affirmed that protectingthe scenic beauty of the falls was their primaryobligation. The treaty guarantees a minimumflow over the famous Niagara Falls during thesummer daylights hours, when tourism is at itspeak. In the Lesotho Highlands Treaty, SouthAfrica helped finance the hydroelectric/waterdiversion facility in Lesotho, one of the world’spoorest countries. South Africa acquired rightsto drinking water for Johannesburg, whileLesotho receives the power generated in addi-tion to substantial annual water payments fromSouth Africa (Wolf 1998). Although the man-agement rules designed for a particular placemay incorporate elements from other places, itis clear from the great diversity of managementcontexts that there could never be a one-size-fits-all “management recipe.”

Most freshwater systems have clear hier-archical structures: smaller watersheds nestedwithin larger watersheds. Management is oftenmost effective if carried out at multiple lev-els, via hierarchical or polycentric institutions(McGinnis 1999; see Chapter 4). In this waydecision-making processes at different levelspartially overlap, conferring some flexibility inscaling management appropriately to the pat-terns of ecosystems. Polycentric governancesystems therefore provide some autonomy ateach level but also a degree of consistencyacross levels. Polycentric institutions cut acrossscales, involving individuals, local communities,municipalities, and central government in man-aging ecosystem services from the level of a

9 Freshwaters 219

village to a watershed and thereby mediatebetween global and local knowledge. Bridg-ing organizations that link across scales, suchas South Africa’s Water User Associations(Box 9.4), help to balance the roles of localautonomy and cross-scale coordination.

Conflicts in freshwater use are also contribut-ing to the emergence of an alternative modelfor the relationship between science, manage-ment, and society (Poff et al. 2003). This modelemphasizes the need for partnerships betweenscientists and other stakeholders in develop-ing water-management goals. It also empha-sizes the need for new experimental approachesthat advance understanding at scales relevantto whole-river or whole-lake management. Inthis model, existing and planned managementpolicies can be seen as opportunities to con-duct ecosystem-scale experiments, in accor-dance with the ideas of adaptive management(Walters and Holling 1990; see Chapter 4).

Summary

Freshwater resources are embedded in ter-restrial ecosystems and must be understoodas parts of interactive landscapes. Moreover,freshwater resources, including the quantityand quality of the water itself as well as fishand wildlife, interact strongly. Freshwater isa nonsubstitutable resource, in the sense thatpeople have finite needs for drinking, washing,sanitation, and growing food, while both terres-trial and aquatic ecosystem processes dependon adequate supply and quality of freshwater.In a world of expanding human populationsand associated demand, along with a chang-ing climate that alters patterns of precipitationand evaporation, there is increasing pressure onfreshwaters and the many resources that theysupport.

Human interactions with water tend tobe characterized by sector-specific decisionsand unavoidable tradeoffs. The condition offreshwater ecosystems has been compromisedby the conventional sectoral approach towater management and, if continued, will con-strain progress to enhance human well-being(Vorosmarty et al. 2005). For example, flow

stabilization through dam construction canseverely degrade aquatic habitats and lead tolosses of economically important fisheries. Itis clear that substantial inconsistencies willdevelop between major development and sus-tainability strategies, such as the MillenniumDevelopment Goals, the Convention on Biodi-versity, and the Kyoto Protocol, if they do notbecome better integrated.

Problems of water, food, health, and povertyare highly interlinked. This is particularly truewhere freshwater is scarce, and the local econ-omy is too weak to provide the infrastruc-ture for safe domestic water supply or toallow large-scale imports of food. On theother hand, if access to safe drinking watercan be secured through appropriate invest-ments in infrastructure, public health conditionsimprove, the potential for industrial develop-ment increases, and the time devoted to col-lecting water can be spent on more productivework or educational opportunities. Addressingpoverty-related problems in very poor nationsmay require a threshold level of infrastructuralinvestment to enable the economy to developand grow.

Spatial heterogeneity, multiple turnovertimes, and thresholds characterize freshwaterresources. Because of these complexities, fresh-water resources are vulnerable to regime shifts,or large-scale changes maintained by new setsof feedbacks. Some regime shifts are harmful,such as eutrophication, salinization, or loss offish and wildlife stocks. Other regime shifts aredesirable, such as the deliberate destabilizationof eutrophication during lake restoration.Resilience of freshwaters is inversely relatedto the amount of change necessary to causea regime shift. In practice, resilience occursat multiple scales within a watershed. Regimeshifts in a particular subsystem may cascadeto other subsystems through movements ofwater, organisms, or people. Thus managementof watersheds involves the breakdown orenhancement of resilience at multiple scales.

So far, success in managing freshwaterresources has been mixed. Many methodsand technologies exist for managing fresh-waters, and even more are under develop-ment. Freshwater-management problems are


somewhat individualistic; each particular prob-lem seems to require a unique combinationof approaches. Moreover, spatial connectivitymeans that all freshwater resources within awatershed are connected, and these connec-tions must be accounted for. Decentralizedmanagement with institutions nested at severalspatial scales seems to be necessary for success-ful management of freshwater resources. Suchnetworks of institutions seem to be capable ofadapting as circumstances change, and adapta-tion is crucially important because of the direc-tional changes in climate, land use, and humandemand for freshwater resources. It is not easyto create and integrate such networks of institu-tions, yet it is being done in many places aroundthe world. Freshwater management demandsresilience thinking. Therefore management offreshwaters continues to be an incubator fornovel approaches to transform the relationshipsof people and nature.

Review Questions

1. What ecosystem services are connected tofreshwaters, both directly and indirectly?

2. How do changes in terrestrial ecosystemsalter freshwater ecosystems?

3. What are thresholds, and why are theyimportant in ecosystem management?

4. How have people responded to conflictsover freshwater? Which responses build

resilience of regional social–ecological sys-tems, and which responses erode resilience?

5. What are the implications of a singleregional institution, polycentric institutions,and decentralized management for freshwa-ter resources?

Additional Readings

Falkenmark, M., and J. Rockstrom. 2004. Balanc-ing Water for Humans and Nature: The NewApproach in Ecohydrology. Earthscan, London.

Finlayson, C.M., R. D′Cruz, N. Aladin, D.R. Barker,G. Beltram, et al. 2005. Inland water systems.Pages 551–583 in R. Hassan, R. Scholes, andN. Ash, editors. Ecosystems and Human Well-Being: Current State and Trends. MillenniumEcosystem Assessment. Island Press, Washington.

Magnuson, J.J., T.K. Kratz, and B.J. Benson, editors.2006. Long-Term Dynamics of Lakes in the Land-scape. Oxford University Press, London.

Naiman, R.J., H. Decamps and M.E. McClain. 2005.Riparia. Elsevier Academic Press, Amsterdam.

Postel, S.L., and B. Richter. 2003. Rivers for Life:Managing Water for People and Nature. IslandPress, Washington.

Vorosmarty, C.J., C. Leveque, C. Revenga, R. Bos,C. Caudill, et al. 2005. Fresh water. Pages 165–207in R. Hassan, R. Scholes, and N. Ash, editors.Ecosystems and Human Well-Being: Current Stateand Trends. Millennium Ecosystem Assessment.Island Press, Washington.

Walters, C.J., and S.J.D. Martell. 2004. Fisheries Ecol-ogy and Management. Princeton University Press,Princeton.

10Oceans and Estuaries: Managingthe Commons

Carl Walters and Robert Ahrens

Introduction

Most of Earth (71%) is covered by oceans,whose fisheries provide an important pro-tein source for people, both locally and glob-ally. Fisheries also contribute economically tothe well-being of coastal communities andother sectors of society. The demand formarine fish continues to increase because ofhuman population growth, migration to coastalzones, increased income and health concernsthat strengthen a luxury-seafood market, andgrowth of fishmeal-based aquaculture, and live-stock production. About 20% of the most food-deficient countries export fish to provide for-eign exchange and to service their nationaldebt—a further motivation to increase fishharvests (Pauly et al. 2005). Fishing capac-ity has increased substantially to meet thisdemand, primarily through development oflarge industrial-scale fleets that can harvest fishthat were once too remote, deep, or dispersed

C. Walters (�)Fisheries Centre, University of British Columbia,Vancouver, BC V6T 1Z4, Canadae-mail: [email protected]

for efficient commercial harvest. As a result, theannual global fish catch has more than tripledin the last 50 years. Since about 1985, however,global landings appear to have declined, despitecontinued increase in fishing capacity, suggest-ing that past harvest levels are not sustainable.

About half (53%) of marine fish and inver-tebrate harvest comes from coastal (continen-tal shelf) areas that occupy <10% of the globalocean, particularly along steep coastlines thathave high upwelling rates (Peru current offSouth America, Benguela current off Africa).With the exception of upwelling areas, mostfisheries production is from high-latitude (sub-arctic, e.g., Bering Sea and North Atlantic)areas, and from near the mouths of major rivers(like the Mississippi) that provide estuarinerearing-environments for juvenile fish and highnutrient delivery rates to coastal areas. Tropicaloceans tend to have much more diverse, com-plex, and less-productive fish communities thantemperate and subarctic oceans. Most tropi-cal fish harvest comes from very large buthighly dilute pelagic (water column) ecosys-tems, mainly from tuna fisheries.

Severe overfishing has occurred in mostof the major ocean production regions(Christensen et al. 2003, Jackson et al. 2001). Arecent analysis of global fishery catch statistics


222 C. Walters and R. Ahrens

indicate about 25% of global fisheries havecollapsed, i.e., a 90% or greater reduction incatch from peak historic levels (Mullon et al.2005). Notable exceptions to stock depletionsinclude the Bering Sea and Gulf of Alaska,where very conservative harvest policies haverestricted fishing over most of the history ofthe fishery. In some regions like the NorthAtlantic, the overfishing problem has not beensolved and is apparently getting worse. In a fewother high-production regions, like the Peruupwelling system, early (1960–1980) periodsof overfishing have given way to apparentlymore conservative and sustainable harvestmanagement.

In this chapter we explore the interactionsbetween people and fish, focusing primarily onindustrial-scale fisheries that account for mostof the global harvest. Chapter 11 focuses onsmaller scale community-based fisheries thathave tight linkages to the livelihoods of manycoastal communities. Quite different controlsover social–ecological dynamics characterizethe extremes of this continuum of fisheriestypes.

Social–Ecological Structureof a Fishery

Marine fisheries are unusual as a social–ecological system because one process (har-vest) dominates social–ecological interactions.Other processes (e.g., pollution, habitat modi-fication, and diversity effects on ecosystem pro-cesses) vary in importance and future unknownchanges in climate and perhaps ocean circu-lation may have important consequences. Thisprovides an unusual opportunity to use a sys-tems framework focused on a single dom-inant interaction. In this section we brieflydescribe the unique features of fish and fisher-men that are important in understanding theirinteractions.

Most fish species have a basic life history pat-tern that exposes them to a complex set of eco-logical interactions, including human impactsthat alter coastal and estuarine systems. Fishtypically start out life as very small eggs, fol-

lowed by a larval stage in which juveniles arewidely dispersed and depend on small marineorganisms (lower trophic levels, phytoplankton,and zooplankton) for food. This start at the bot-tom of the food web generally occurs in local-ized spawning areas to which older fish havemigrated, for example, areas of high produc-tion (e.g., estuaries) and/or relatively low pre-dation risk. As juveniles grow, they typicallyexhibit complex ontogenetic habitat shifts (i.e.,shifts related to stage of development) to loca-tions where they can utilize larger food organ-isms and where increased body size makes thembetter able to both forage efficiently and moverapidly to avoid predation. These areas includereefs, rocky bottom areas, sea grass beds, andsalt marshes. An important policy implicationof ontogenetic habitat shifts is that marine pro-tected areas (MPAs) cannot easily be designedto cover the full life cycle of the fish.

Since ontogenetic habitat shifts generallyinvolve use of shallow coastal waters and/orestuarine areas, many fish species spend atleast some critical parts of their life cyclein those marine habitats most vulnerable tohuman impacts from habitat modification (e.g.,dredging and filling of mangrove swamps)and eutrophication. But remarkably, long-termstudies of fish (and commercially importantinvertebrates like shrimp) recruitment pat-terns, where recruitment is measured as netnumber of juveniles surviving to harvestableages, have rarely shown temporal patterns ofrecruitment decline that are obviously corre-lated with known historical habitat alterations.We will review one case study (oysters inthe Cheasapeake Bay) where an obvious link-age between recruitment and human coastalhabitat alteration can be demonstrated, butmost such “obvious” linkages have been dif-ficult to demonstrate empirically. There areat least three potential explanations for thegeneral lack of demonstrated linkage betweenhabitat modification and population dynam-ics: (1) overwhelming impact of fishing as acontrol over population dynamics, (2) insuf-ficient observation and/or experimentation todemonstrate habitat effects, and/or (3) habi-tat modification may be less important thansome managers have assumed. One manage-

10 Oceans and Estuaries 223

ment implication is that careful management ofone process (fishing) can have enormous influ-ence on fish stocks and their role in sustainingecosystem services in coastal and marine envi-ronments. We return to the issue of manage-ment experiments later.

The biophysical structure of most produc-tive coastal habitats is maintained througha dynamic balance of structure-building pro-cesses like sedimentation and growth of seden-tary organisms that offer structure (mangroves,sea grasses, reefs), and erosive processes asso-ciated with water movements (currents, waves).These processes result in slow-variable changesin structural patterns, punctuated by rapidchanges during violent physical events likehurricanes. For example, sediment delivery tothe coast has declined globally by about 10%because sediment capture by reservoirs exceedsincreases in erosion (Agardy et al. 2005). Theecological consequences of this slow change canbe quite abrupt (e.g., 1,500 km2 of Louisianawetlands lost during hurricane Katrina in 2005).

Localization of spawning and juvenile nurs-ery opportunities further complicates fisheriesmanagement, because fish populations tend tobe divided into complex metapopulations withmany local subpopulations, where natural selec-tion has led to many adaptations to local habitatconditions (e.g., differentiation in spawn tim-ing, movement patterns of juveniles, growthand maturation schedules). Such complex spa-tial organization inevitably leads to subpopula-tion differences in vulnerability to human har-vest or, alternatively, a compensatory increasein productivity in response to fishing-inducedincreases in mortality. Fishing typically causesat least some loss of spatial population struc-ture (the most vulnerable and sensitive sub-populations are typically overharvested) evenwhen overall population productivity and har-vest appear to be sustainable. For this rea-son harvest statistics often provide inadequatewarning of incipient stock depletion.

There has been much speculation aboutthe possibility of complex ecological dynam-ics and multiple stable states in marine ecosys-tems (Walters and Martell 2004). However,the empirical evidence from long time-series(50–100 years) of harvest data most often indi-

cates relatively simple dynamic change, punc-tuated for some species by dramatic collapsesapparently due to overfishing. Data from trop-ical and pelagic fisheries show no hint of mul-tiple equilibria. But for very productive, rela-tively low-diversity ecosystems like the BeringSea and the northern continental shelves ofthe Atlantic Ocean, there may be at least twostable states, a pelagic (water-column-dwelling)dominance state, where the fish communityis dominated by small planktivores (herring,sprat, sardines, etc., that eat plankton), andlarge piscivores (fish that eat fish) are relativelyrare, and a benthic (bottom) dominance statewhere large benthic piscivores (cod, large flat-fish, pollock) are very abundant, and plankti-vores are reduced through high predation rates.For example, the Bering Sea has been movinginto a benthic dominance state, since initiationof industrial fishing, with growing populationsof Alaska pollock, cod, and arrowtooth floun-ders (NRC 1996, 2003). Over the same period,the Scotian shelf off eastern Canada has beenmoving in the opposite direction, with declinesin large piscivores like cod and increases insmall pelagic species (Choi et al. 2004, Franket al. 2005, Zwanenburg et al. 2002).

The world’s largest fisheries involve specieswith highly dispersive and migratory life histo-ries (like tunas, cods, and small pelagics), butmost of biological diversity of fish communitiesis in coastal areas with complex bottom struc-ture (rocky bottoms, coral reefs). Thus there is abig difference between sustaining fisheries pro-duction in overall biomass terms, versus pro-tecting biological diversity as a conservationobjective. In California, for example, a largenetwork of MPAs (mandated by the Califor-nia Marine Life Protection Act) offers effectiveprotection to less than 20% of the harvestablebiomass, primarily in long-lived benthic specieslike rockfish that have very low sustainable har-vest rates (10% per year or less; Hilborn et al.2006).

Sociocultural systems and their managementinstitutions are closely linked to variation inpatterns of fish production and biodiversity.Where there is high biomass, productivity,and mobile fish stocks (subarctic, temperate,and upwelling systems and the tropical open-


Table 10.1. Comparison between small-scale (artisanal) and large-scale (industrial) fisheries in Norway.Information from Alverson et al. (1994).

Fishery characteristic Large-scale fishery Small-scale fishery

Number of fishermen employed 2 million >12 millionFishermen employed per $1 million invested in fishing vessels 5–30 500–4,000Capital cost per fisherman $30,000–300,000 $25–2,500Annual fuel consumption 14–19 million tons 1–3 million tonsFish caught per ton of fuel 2–5 tons 10–20 tonsAnnual catch for human consumption 29 million tons 24 million tonsAnnual catch for industrial products (fishmeal and oil) 22 million tons almost noneDiscarded bycatch 16–40 million tons almost none

ocean tuna fishery), most fish are harvested byindustrial fleets with efficient gear, relativelylow employment, and well-developed assess-ment and regulatory agencies and commissions(Table 10.1). Conversely, where fish popula-tions are less productive and/or less mobile, forexample, in diverse tropical ecosystems or intemperate invertebrate (e.g., shellfish and lob-sters) fisheries, more localized fishing commu-nities tend to develop (see Chapter 11). In thesesituations fishermen utilize a diverse collectionof simpler technologies, rely more upon localinstitutional arrangements to prevent overfish-ing and generally only cause severe depletionin local areas near communities where fishingactivities are based. We discuss these sociallydiverse types of fisheries later in the chapter,with social dimension discussed in Chapter 11.The cases where local institutions fail to pre-vent overfishing mainly involve industrial fish-ing operations that have moved into tropicalcoastal areas through agreements with localgovernments (often involving corruption andpayoffs to local officials) to fish in their exclu-sive economic zones (EEZs; Bonfil et al. 1998).This has happened, for example, with Europeanfleets using the coast of Africa.

The major institutions that govern industrial-scale fishing are (1) regulations that constrainfishing pressures to protect stocks and governallocation of stocks among competing groups offishermen and (2) subsidies and market forcesthat influence the behavior of fishermen withinthis regulatory framework. Each nation regu-lates fisheries within 200 miles of its coast (itsEEZ). International bodies (e.g., the Interna-tional Whaling Commission) and treaties (e.g.,

Convention on the Conservation of Antarc-tic Marine Living Resources, CCAMLR) pro-vide similar protection to some stocks in inter-national waters. The effectiveness of interna-tional efforts is seriously limited by illegal,unreported, and unregulated (IUU) fishing andby the choice by some nations not to participatein treaties (e.g., Norway with respect to whal-ing). Fishing pressure can be reduced by restric-tions on the number of vessels (e.g., individ-ual transferable fishing quotas (ITQs) that areassigned property rights to a proportion of theannual catch), types of gear, or times and placeswhere fishing is allowed.

Subsidies are a major factor contributing tooverfishing. These subsidies are large (glob-ally about 50–100% of the value of the landedcatch, Christy 1997) and are intended to reduceeconomic hardships to fishermen when stocksdecline or alternatively to increase nationalcompetitive advantages in industrial fishing.About 90% of subsidies go to industrial fish-ermen, who tend to have greatest politicalinfluence (Pauly et al. 2005). The types ofsubsidies vary among countries, for example,unemployment benefits in Canada, tax exemp-tions in the USA, and payment of fees to gainaccess to foreign fishing grounds by the Euro-pean Union. Many food-deficient developingnations sell fishing rights at low prices to othercountries to provide foreign exchange becausethey lack the fishing capacity to exploit off-shore stocks. Other subsidies include cheap fueland vessel buyback programs to reduce fish-ing capacity in overcapitalized fisheries. Vesselbuyback programs often fail to achieve theirobjective because fishermen invest in new boats


with technologies that enable them to exploitdepleted stocks more effectively (Pauly et al.2005). All types of subsidies enable fishermento continue fishing under circumstances whendeclining stocks have reduced profits below lev-els that would permit them to fish in an unsub-sidized fishery.

In order to understand the overfishing issue,it is useful to think of a “fishery” as a complexadaptive system (see Chapter 1) consisting ofthree main subsystems or components:

1) a collection of targeted organisms (stocks)that may range from a single genetic stockof one species to a complex assemblage ofspecies living in an arbitrarily defined area ofinstitutional management responsibility andpursued by

2) a collection of fishermen, which again mayrange from a single fleet of similar fish-ing vessels to a complex assortment of fish-ing gears operated by a multinational setof fleets, motivated by economic and/orrecreational interests and regulated to somedegree by

3) a management authority, which is usually apublic or international agency/commissionwith some legal mandate to monitor thestocks and fishing, and to regulate fishingactivity (input control approach) or catches(output control approach) with the dualaims of assuring ecological sustainabilityand reasonable or fair allocation of benefitsamong competing fishermen. Local informalinstitutions serve this role in some coastalfisheries.

Such systems are essentially a predator–prey interaction, somewhat moderated in itsdynamics by restrictions imposed by regula-tions, incentives, or other institutional arrange-ments. These three subsystems are embed-ded within larger ecosystems that sustain (andsometimes threaten) the target species, largereconomic and social systems that generatedemand for fish, and larger institutional sys-tems that define and constrain the powers ofthe regulatory authority and sometimes cre-ate perverse subsidies for fishermen as part ofsocial programs aimed at maintaining employ-ment and economic well-being. As target stocks

vary over time in response to natural factorsand fishing impacts, fishermen respond throughchanges in investment in new gear and technol-ogy, fishing activity or effort, and spatial choicesabout where to fish. Regulatory authorities arefaced with the complex problems of (1) infer-ring how much of measured changes in out-puts (catches, fisher success rates) and abun-dance are due to human-induced versus naturalchanges and (2) varying fishing restrictions overtime accordingly. Further, in complex multifleetfisheries, regulatory authorities often becomepreoccupied with issues of allocation amongcompeting fleets (or recreational versus com-mercial users) and lose sight of conservation orsustainability objectives.

Regulatory authorities typically exhibit adecision-making pathology known as indecisionas rational choice when faced with evidenceof stock decline (Walters and Martell 2004).In such circumstances, decision makers heartwo types of empirical assertions. On one hand,they hear scientific assessments purporting todemonstrate the decline, generally presented inthe form of complex assessment reports witharcane terminology and much hedging aboutproblems with inadequate data and populationmodels. On the other hand, they hear simple(and hence compelling) assertions from fishingrepresentatives about how there are still plentyof fish and scientists are often wrong. So facedwith a hard choice between causing immediateeconomic and social harm by imposing moresevere regulations (reduced fishing time, lowerquotas, etc.), versus gambling that the scientistsare wrong and that the decline will reverse itselfnaturally if no action is taken, it is quite rationalfor decision makers to choose the gamble andto defer fishery reductions for as long as possi-ble. Such time delays and thresholds in the reg-ulatory subsystem cause dynamic instability inthe fishery system as a whole.

In the following examples we describe howfisheries dynamics vary over a range of situa-tions. We begin with relatively simple cases inwhich the fish–fisherman predator–prey inter-action is not substantially impacted by manage-ment, yet can lead to persistently productivesystems. Then we look at cases in which persis-tence has not occurred, mainly because of par-


ticular ecological spatial dynamics interactingwith an inadequate management subsystem.Finally we examine a few cases in which man-agement has been relatively successful despiteecological and fishing dynamics that could eas-ily produce collapse and review a few key char-acteristics of such success stories.

Why Fisheries Should NotCollapse: Shrimp and TunaFisheries

If fisheries are viewed as dynamic predator–prey systems, there are bioeconomic mech-anisms that could prevent collapse withoutintensive management intervention. It is notself-evident that fishermen in search of profitswill always drive fish stocks to collapse unlessprevented from doing so by effective institu-tional arrangements. If profitability of fishingdecreases as the predator (fleet) populationgrows, the predator recruitment (new invest-ment) rate may decrease and mortality rate mayincrease, leading to a rough balance or bioeco-nomic equilibrium even in the absence of effec-tive regulation (Clark 1985, 2007).

There are now hundreds of time series ofdata showing how fish populations respondto increased fishing pressure. If reproductiveand mortality rates showed no compensatorychange, increased mortality due to fishingshould accelerate declines toward extinction,because fishermen first remove the largest, old-est, most fecund fish. However, most exploitedfish populations do not show this simple rapidpopulation decline but instead show strong,compensatory increases in juvenile survivalrates (from egg to the age of first captureby fishing). Consequently, the net recruit-ment of fish to the harvestable populationremains nearly independent of spawning pop-ulation size until the spawning population isgreatly reduced. Juvenile survival rate typi-cally increases at least threefold as stock sizedeclines, and many fish (like cod and flatfishes)exhibit increases of 50-fold or more (Myerset al. 1999, Goodwin et al. 2006). Increasedjuvenile survival is often accompanied by com-

pensatory improvements in body growth rates.Even when fish populations are also affectedby environmental changes (like ocean regimeshifts), these compensatory responses implythat there can be a rough balance between bio-logical production and fishery removals, if fish-ing mortality rates decline at low fish popula-tion sizes due to loss of economic incentives tokeep fishing.

The open-access character of many fisheries(i.e., where fishermen pursue fish and no oneclaims ownership of the stocks) is often assumedto be the main cause of fishery collapse. Inother words, in the absence of effective regu-latory restraints, individual fishermen have noincentive to restrain their catches, because anyfish so saved can be taken by other fisher-men. However, from a bioeconomic perspec-tive, the real issue is whether there are situationswhere the expected decline in profitability(decreasing income relative to fishing costs) failsto reduce fishing pressure as stocks decline.

There are a few excellent examples of fish-eries that have apparently reached bioeconomicequilibrium and remain highly productive with-out effective regulation of total fishing effortor fleet size. One such example is one of themost valuable fisheries in the USA, the Gulfof Mexico shrimp fishery (Fig. 10.1a). Regula-tions in this fishery are primarily closed seasonsand areas, aimed at protecting shrimp until theyreach marketable body sizes. The total num-ber of vessels allowed to pursue shrimp dur-ing open seasons and in open areas has neverbeen effectively controlled, except through thebioeconomic process of declining profitabilityas more vessels enter the fishery and competefor about the same total biological productioneach year. This apparent bioeconomic equilib-rium does not, however, imply that the Gulf ofMexico ecosystem has reached an equilibrium.There are continuing, slow declines in severalfish (like Atlantic croaker and marine catfishes)that are killed in large numbers by the trawl-ing or discarded as bycatch. Over the long term,disruption of the benthic habitat by trawling ordeclines in other species could impact the sus-tainability of the shrimp fishery itself.

The total fishing effort or catches of themajor pelagic tuna fisheries have also never


Gulf of Mexico shrimp fisheryC

atch

1960 1970 1980 1990 2000

1020

3040

5060

00.

10.

20.

30.

4

Effo

rt

CatchEffort

1950 1960 1970 1980 1990 2000

100

200

300

400

500

Japanese global longline effort

Ho

oks

(mill

ion

s)

1950 1960 1970 1980 1990 2000

500

1000

1500

2000

Total global longline effort

Ho

oks

(mill

ion

s)

1950 1960 1970 1980 1990 2000

108

64

212

14

Japanese global longline catch

Year

Cat

ch (m

illio

ns)

AlbacoreBigeyeYellowfin

SwordfishBluefin Tuna sp.Marlin sp.

1950 1960 1970 1980 1990 2000

1015

2025

5

Total global longline catch

Year

Cat

ch (m

illio

ns)

AlbacoreBigeyeYellowfin

SwordfishBluefin Tuna sp. Marlin sp.

0

Figure 10.1. Long-term changes in two major fish-eries. The Gulf of Mexico shrimp fishery has reacheda “bionomic equilibrium” where stock sizes andcatches have remained at healthy levels despite lackof effective control on the fishing fleet size. The

Japanese pelagic longline fishery for tunas and bill-fish has also reached an apparent bionomic equilib-rium, but global longline effort and catch continueto grow apparently due to economic subsidies that afew nations have provided.


been effectively regulated. For these fisheries,at least some national fishing fleets appear tohave reached bioeconomic equilibrium (e.g.,Japan, Fig. 10.1b), while others continue togrow (Fig. 10.1c) despite slow declines in prof-itability as measured by catch per unit fish-ing effort. The continued growth in effort hasoccurred primarily in fishing fleets of thosenations that provide large financial subsidiesto maintain fishing (e.g., Taiwan and SouthKorea), effectively shifting the bionomic equi-librium to lower or even zero fish abundance.Other subsidies seek to prevent economic hard-ship to fishermen at times when stocks are toolow to make a profit also altering the bioeco-nomic equilibrium.

There are also a few obvious examples offisheries that have been sustained over longperiods of time, most notably the Pacific salmonin British Columbia and Alaska, where thestocks are extremely concentrated at spawning(e.g., at the mouths of salmon rivers) and hencecould be wiped out without fishermen “seeing”the decline in the form of decreasing profitabil-ity of fishing. In these cases, the blatant riskof overfishing led to severe restriction on fish-ing over a century ago, in the form of closingmost of the ocean to fishing for most of theyear. Many salmon fisheries are managed byallowing only brief, intensive fishery openingsat river mouths, which each take limited (andwell-known) percentages of the stocks. Sim-ilar regulatory systems have been developedmore recently for other stocks that form vul-nerable spawning aggregations, most notablyPacific herring and Atlantic herring off easternCanada.

Another reason why the “invisible hand”of declining profitability with stock size can-not completely substitute for an effectiveregulatory subsystem is that declining abun-dance and competition among fishermen cre-ates strong incentives to invest in technolo-gies for increasing search efficiency. This cre-ates a slow-variable change in the predator–prey parameters and a decline over time in thestock size needed for the average fisherman tocatch enough to just meet costs. Technologi-cal change tends to cause fisheries to slowly“ratchet down” (decrease stock sizes) over time

(Ludwig et al. 1993, Hennessey and Healey2000), with episodes of apparent stability inter-spersed with punctuated declines, as new tech-nologies spread through fishing fleets.

Why Some Fisheries CollapseAnyway: Atlantic Codand Herring Stocks

Despite the possibility of bioeconomic balanceor equilibrium even without regulation, reviewsof trends in fishery catch statistics show thata growing percentage of the world’s fisheries(roughly 25% as of 2006) have “collapsed” inthe sense that catches have declined to 10% orless of peak historical levels (Mullon et al. 2005,Caddy and Surette 2005, Worm et al. 2006).Worm et al. (2006) claim that collapses areoccurring at an accelerating rate. Some of theseapparent collapses have led to more stringentregulations aimed at rebuilding stocks that werehistorically overfished, but many have not ledto any apparent regulatory response. Statisticalanalyses of patterns of collapse in 1,519 stocksaround the world (Mullon et al. 2005) showthree general patterns over the decade prior tothe collapse (Fig. 10.2): (1) “plateau” where aperiod of stable catches is followed by rapid,apparently depensatory (accelerating) decline;(2) “smooth” where catch declines more or lesssteadily; (3) “erratic” where there is at least onestrong peak in catches in the decade precedingcollapse. Caddy and Surette (2005) argue thatthese patterns can be largely explained throughdirect impacts of fishing, without invoking morecomplex changes in trophic interactions orother aspects of ecosystem structure.

The most worrisome collapses are the“plateau” pattern cases, where apparent stabil-ity is followed by rapid, accelerating decline.These cases involve species (e.g., cod and her-ring) that show a “range collapse” pattern inwhich the density of fish (fish per unit area)remains stable or even increases as the totalstock declines because of decreases in the areaoccupied by the stock (MacCall 1990). Fish-ermen are generally able to detect this pat-


Plateau and crash

0.0

0.4

0.8

1.2

0.0

0.4

0.8

1.2

Erratic

Years before collapse

0.0

0.4

0.8

1.2

10 8 6 4 2 0

Smooth decline

Rela

tive

cat

chRe

lati

ve c

atch

Rela

tive

cat

ch

Figure 10.2. Selected examples of three patterns oftemporal change in catch over the last decade beforecollapse occurred, for various fish stocks that havecollapsed in the last 50 years as classified by Mullonet al. (2005) from analyses of over 1500 catch time

series. The most worrisome pattern is the “plateau”,where catch statistics (and often other indicators ofstock status and trend) show little or no warningof impending rapid collapse. Modified from Mullonet al. (2005).

tern and concentrate their fishing activity inthe remaining occupied range area. The neteffect of this ability to find the remainingfish is to cause the fishing mortality rate toincrease rapidly even if the total fishing effortdeclines somewhat as the stock collapses. Infisheries terminology, we say that the catcha-bility coefficient (fishing mortality rate gener-

ated by one unit of fishing effort) increases asstock size decreases; this effect has been rec-ognized as a major management risk for manyyears for fisheries where control of fishing effortis seen as the main way to prevent overfishing(Paloheimo and Dickie 1964, Harley et al. 2001,Shertzer and Prager 2007). For cod off New-foundland, the effect was to cause a dramatic


Year

Sto

ck s

ize

1965 1970 1975 1980 1985 1990 19950

500

1000

1500

2000

2500

3000

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Exp

loit

atio

n ra

te

Actual stock

NAFO 1980

NAFO 1983

NAFO 1986

CAFSAC 1991

Exploitation rate

Figure 10.3. The recent history of the Newfoundland“northern” cod stock. Beginning in the 1950s, for-eign fishing fleets caused rapid stock decline. AfterCanada assumed management of the fishery in thelate 1970s (under extended jurisdiction over coastalwaters), stock assessments and projections (trajecto-ries labeled “NAFO” and “CAFSAC”) grossly over-estimated stock size and recovery rate, leading to

development of a Canadian fishery that eventuallydecimated the stock. Note how annual exploitationrates increased greatly during the last few years ofthe collapse, as fishing effort remained high whilethe range area occupied by the stock decreased andfish became more concentrated in remaining fishingareas. Modified from Walters and McGuire (1996).

increase in total fishing mortality as the stockcollapse proceeded (Fig. 10.3; Atkinson et al.1997, Rose et al. 2000, Swain and Sinclair 1994).

Stock assessments often fail to detect rangecollapse for species like cod because popula-tion trends are estimated from trends in rela-tive abundance along with catch and size–agecomposition statistics, without considering thearea occupied by the stock (Walters and Martell2004). The resulting stock assessments overesti-mate stock size when a stock begins to decline,which in turn leads to a delay in implementingneeded catch and effort reductions. The stockestimates and projections of future populationgrowth can be grossly incorrect (Fig. 10.3), rightup to the point where fishermen can no longertake even the allowable catch and the fisherymust be closed completely. This occurred inthe Newfoundland cod case. Management deci-sions based on scientific assessments that did

not consider distortions in relative abundancetrends due to range constriction ultimatelycollapsed the fishery (Walters and McGuire1996).

Even when scientific assessment errors havenot contributed to the problem, two keydecision-making pathologies have character-ized rapid collapses of range-collapsing speciesfor which it remains economical to keep fish-ing until stock size is very low. First, thereis a dynamics of denial, in which fishermenargue that scientific assessments must be wrongbecause they are still catching plenty of fish(and if there is a decline, recovery will occurnaturally due to favorable environmental cir-cumstances). Political decision makers are notblind to such arguments; after all, fishermen arethe ones with firsthand knowledge of what ishappening on the fishing grounds. Second, thereis a kind of ecological brinksmanship or inde-


cision as rational choice (Walters and Martell2004). Decision makers when faced with cer-tain economic pain if they take decisive action,or a chance of natural recovery (or underes-timates of stock size by scientists) if they donot, will make the quite rational choice to gam-ble on the chance of natural recovery ratherthan face the certainty of economic harm. Noone can blame decision makers for taking suchgambles, especially considering how frequentlyscientists have been wrong in claiming “thesky is falling.” A good example of such claimsis the highly publicized assertion by Wormet al. (2006) that there may be global fish-ery collapse within 50 years; such claims haveaimed to foster public support for conserva-tion but may end up having just the oppositeresult by further increasing mistrust in scientificassessments.

The points of the previous paragraph empha-size that decision-making processes in fisheriesharvest management are rarely done by, or evendominated by, scientists in the first place. Thereare rare situations, like the Pacific HalibutCommission and the Pacific Salmon Commis-sion, where independent management author-ities have been empowered to set harvest lim-its supposedly based purely on scientific evi-dence, but even in such cases there is politicalinfluence through the composition of decision-making boards and commissions. In most devel-oped fisheries, decision-making falls ultimatelyon the shoulders of politicians who hear evi-dence from other stakeholders besides scien-tists and are empowered to balance ecologicaland social/economic concerns. There is grossmisunderstanding about this point in the sci-entific community, with most attention beingfocused on ecological changes (such as depen-satory decreases in productivity at very lowstock size) without much thought about whydangerously high harvests were allowed in thefirst place. This disciplinary myopia is evidencedfor example in a review by Longhurst (2006)claiming that the classical “theory of fishing”has failed. In fact that population dynamics the-ory correctly predicted most patterns of fish-ery decline, but its predictions have been widelyignored by decision makers who have engagedin the pathology of indecision as rational choice.

This institutional structure of industrializedfisheries contrasts with informal institutionsoften found in local fisheries (see Chapter 11).

Range collapses and associated stock-assessment errors are not the only causes forfishery collapse, especially in cases of slow col-lapse where no bioeconomic equilibrium (or atleast decline in fishing effort) becomes evidentas the stock becomes very low. Many stockshave continued to collapse simply becausethey have been “caught in the crossfire,” i.e.,are incidental or more sensitive species takenalong with more productive or valuable onesin multispecies fisheries. This has been a prob-lem in both the large pelagic fisheries, wherefishing is targeted on tunas but incidentallygenerates high mortality rates of larger (andless productive) billfishes and dolphins, and themultispecies benthic fisheries where long-lived,slow-growing species like rockfish have beendepleted while fisheries have continued to besupported by more productive cod, flatfish, etc.There are also some cases where environmen-tal change and habitat loss have contributedto apparent collapses, for example, cod inthe Baltic sea where the habitat periodicallybecomes unfavorable for them, and Pacificsalmon species that depend on stream habitatsimpacted by cumulative effects of loggingand urban development. There are cases likethe tuna longline example in the previoussection where fishing effort has been drivento artificially high, eventually nonsustainablelevels by outright government subsidies aimedat maintaining fisheries employment. Finally,there are a few species that bring such highprices that fishing remains economical evenafter abundances have been driven very low(e.g., bluefin tuna and abalone).

Slow-Variable Dynamicsand Shifting Baselines:Chesapeake Oysters

Changes in critical slow variables can causedegradation of a fishery. One of the saddest sto-ries in renewable resource management is the


decline of oysters in the Chesapeake Bay. Thesevaluable creatures were spectacularly abundantwhen the region was first settled, occurring inlarge, inshore reefs where oysters grew atopreefs created from the accumulated shells oftheir ancestors. Early fisheries depleted liveoysters from these accessible reefs, and the shellreefs themselves were “harvested” to providesources of lime and bed material for roads andother construction. By the mid-1800s, the fish-ery had moved offshore, using dredges fromsailboats. From that time onward, the fisheryhas shown a progressive decline (Fig. 10.4;Kennedy and Breisch 1983, Rothschild et al.1994), to the point that it is now almost extinct.An obvious slow variable in this system is theavailability of hard bottom materials, especiallyoyster shells themselves, as settlement sites fornew recruits. This nonliving “biomass” wouldhave declined somewhat over time with reduc-tions in live oyster populations, even if shell hadnot been actively mined for other uses.

The need to maintain and restore shell reefswas recognized even in the 1800s, and attemptsto regulate the fishery led to colorful “oysterwars” by the 1880s. Yet despite such attemptsto regulate the fishery in the face of relativelysimple dynamics of the loss of recruitment habi-tat, there was an apparent lack of learning; com-mons decision-making was defied (e.g., oysterwars). In the 1900s the decline was punctuatedby disease invasions. Recognition of the needto stop dredging and removing shell came early,and efforts to haul shell back into bay thatbegan during the 1920s continue today. Todayin the Cheasapeake there is much public invest-ment in large, complex scientific monitoringprograms and plans for restoration. The stud-ies have shown, for example, that high chloro-phyll (phytoplankton) levels in the Bay are duepartly to eutrophication from surrounding citiesand farmlands, but also to loss of the massivewater-filtering function provided by the oyster

3275 Hand-tong (Ht) boats operating

700dredge

(Dr) vessels

1800 1850 *1884 1900 1950 2000

Year

MSX DermoMSX

11000 hand- and patent-tong

boats (Pt) plus 6000

dredge vessels

Ht,

Dr,

Pt, h

ydra

ulic

pat

ent

ton

g (H

PT)

Ht,

Dr,

Pt,

HPT

, d

iver

har

vest

ing

(Di)

Oys

ter l

and

ing

s (1

05 m

etri

c to

ns)

0

2

4

6

8

Figure 10.4. Development and decline of the Chesa-peake Bay oyster fishery. The long, slow decline ismost likely a slow-variable effect of loss of oystershell bottom habitat (where most successful settle-ment of juvenile oysters occurs), due to mining theshell and loss of “recruitment” to the shell pop-ulation caused by harvesting of live oysters with-

out returning their shells to the beds from whichthey were harvested. During each successive timeperiod (A to E), new harvest methods and additionalharvest pressure were added. MSX arrows refer totimes of exotic disease outbreaks. Modified fromRothschild et al. (1994).


beds, which, prior to exploitation, probably fil-tered the entire Bay’s water mass at least onceper week.

Today, as outside observers, we see a systemin which there has been huge ecological impactof oyster harvesting, combined with massive sci-ence efforts that appear to be leading nowhere.How could this have happened? Given thesedentary oyster stock, why was there neveran institutional move to local governance withincentives for local communities and fishermento protect and restore their resources? Did peo-ple expect the problem to correct itself? Didthey simply ignore the long-term trend, i.e., wasthere a shifting baseline syndrome as defined byPauly (1995)?

Coral reef systems have been lost in somecoastal areas due to slow-variable increasesin nutrient loading and sedimentation fromagricultural and urban activities (Brodie et al.2005, Bellwood et al. 2004) in combination withoverfishing (see Chapter 11), which depletesstocks of parrotfishes and other herbivores thatwould otherwise prevent algae from overgrow-ing the corals (Bellwood et al. 2004, Hugheset al. 2007). Such changes may have con-tributed to development of dynamic instabilityin trophic interactions. For example, the crown-of-thorns starfish has exhibited coral-destroyingoutbreaks since the 1950s on Great BarrierReef of Australia. These outbreaks may havebeen a natural part of this system for at least7000 years (Walbran et al. 1989), but this evi-dence has been disputed (Keesing et al. 1992).Models of the outbreak dynamics look muchlike the fast–slow variable mixed dynamics offorest–insect outbreaks that motivated earlyideas about resilience, with slow coral recov-ery leading under some conditions to increas-ing larval survival of the starfish and eventu-ally to outbreak dynamics where the starfishspreads widely across reef systems. These con-ditions for improved larval survival may involveslow changes in nutrient loading. In othercases, warming climate has been a slow-variablechange that leads to bleaching of corals (lossof their symbiotic alga), death of corals, and, inextreme cases, loss of reef structure that is crit-ical in supporting the fishery, just as in the caseof Chesapeake oysters.

Resilience in BiologicallyVariable Systems: Pacific Salmonand Herring

Appropriate management policies have madesome of the most vulnerable fisheries highlyresilient. As noted above, Pacific salmon andherring fisheries have prospered over most ofthe last century, despite dramatic spatial con-centration and high vulnerability to overfish-ing as fish move into spawning areas. Bla-tant risk of overfishing led as early as the late1800s to management systems involving clo-sure of large areas to fishing for most of eachyear, with fishing concentrated in small areasfor brief periods. This management approachrepresented essentially a reversal of the ideaof MPAs, by creating small exceptional areasfor fishing, instead of closing areas to fishing.Also early in the development of this veryrestrictive approach to management, agenciesbegan to adopt feedback policies for copingwith environmental fluctuations, in the formof rules for varying harvests with changes instock size. The earliest such rules were very sim-ple, so-called fixed exploitation rate rules: catchthe same fixed percentage of the stock eachyear, and leave the rest to spawn. Later, fixedescapement rules became popular, in whichthe management goal is to allow the samefixed number of fish to spawn each year. Opti-mization studies (see review in Walters 1986)showed that yield should be maximized byfixed escapement policies, but less variable andsocially more acceptable patterns of yield overtime should arise from fixed exploitation ratepolicies.

Resilience is the capacity of a system tocope with unexpected changes in its environ-ment. In resource-management contexts, thismeans the capacity of management systems torespond effectively to unplanned, usually sur-prising changes in ecological and resource userdynamics (Walker et al. 2004, Gunderson et al.2006).

A few fisheries management systems havedemonstrated high resilience, coping effec-tively (and continuing to support relatively highbiological production and healthy dependent


communities) on time scales of up to a cen-tury despite wildly variable fish stock sizes, peri-ods of unintentional overfishing, unexpecteddestructive events such as catastrophic die-offsof fish associated with blockage of migrationpaths, and extreme changes in fishing tech-nologies and allocation among license hold-ers and gears. These highly successful manage-ment systems have mainly involved species likePacific salmon and herring that are highly con-centrated in space, relatively easy to monitor(and to overfish), capable of supporting special-ized local communities, and so obviously need-ful of careful management and protection soas to preclude much debate among stakehold-ers and politicians about the basic need formanagement.

Most of the cases of high resilience haveinvolved species with complex spatial popu-lation structures, with many local stocks orgenetic races of fish contributing to overallproduction. This allows fishermen flexibility todampen variation by moving among species andareas. There are successful management sys-tems for stocks with simpler ecological struc-ture, e.g., Pacific halibut and Australian shrimpand rock lobster fisheries, but these cases alsoinvolve spatially complex fish life histories(migration, dispersal) and fishing patterns andspecies that have either been relatively easy tomonitor (shrimp) or have some natural pro-tection from overharvesting due to biologicalcharacteristics (e.g., movement of some fish intodeeper waters that do not attract much fishingeffort, use of juvenile nursery areas where fish-ing is either not practical or is banned deliber-ately to protect the juveniles).

Such cases are at an opposite extreme fromcases like the Chesapeake oyster fishery, whichhas been in decline for over a century andhas shown virtually no capability to respondeffectively to regulatory needs or to unexpecteddisturbances like appearance of exotic diseaseorganisms. Fortunately, such extreme manage-ment failures are also rare.

Between these extremes are a variety offisheries that have shown some capability tocope with “routine” natural variability but havefailed when faced with persistent changes in

productivity. Good examples are the cod andherring fisheries of the north Atlantic, whereperiods of low recruitment led to stock sizedeclines that were too rapid for available mon-itoring data and stock-assessment methods todetect and where the need for reductions inharvest occurred more rapidly than was con-sidered socially acceptable. In these cases, lowresilience results from a complex stew of rapidecological change, inadequate monitoring, andincentives for denial of the need for adaptiveresponse.

Social Sources of Resiliencein Fisheries Systems

Much of the previous discussion has focusedon industrial-scale fisheries. However, there isa continuum of fishery types, including indus-trial fleets that catch and process fish; small-to-large privately owned commercial vesselsthat sell their catch to markets or proces-sors; charter boats and sport fishermen; indi-vidual subsistence fishermen who consume orshare what they catch; and various combina-tions of these categories. Small-scale fishermenare more likely to use inshore fisheries, where itis easier to monitor the status of stocks, fishingeffort, and catch compared to the open oceanwhere at least parts of many stocks are too dis-tant from fishing villages to be worth pursu-ing. In many cases, local institutions (both for-mal and informal) have developed to regulatethe fishery for the long-term benefit of localusers. As in the industrial fishery, some of thesearrangements have been successful and othershave failed to sustain stocks (see Chapter 11).Even in the case of failures, however, theseare usually local in extent, because of the localnature of the fisheries. There has been consid-erable interest in these local fisheries, becausesuccess often occurs without top-down govern-mental regulation. There is no simple formulato success of these local fisheries, but compar-isons of dozens of case studies show that cer-tain features enhance the likelihood of success(see Chapters 4, 9, and 11; Schlager and Ostrom


1993, Pinkerton and Weinstein 1995). Consid-ering the spectrum from small-scale to indus-trial fisheries, common denominators that havedemonstrated high resilience are observed.

In most cases, a degree of localized man-agement control is present. Such control,from government regulatory representatives orlocally evolved institutions, has developed somedegree of cooperation between stakeholdersto design mutually acceptable monitoring pro-grams and rules for responding to change. Suchmonitoring programs are capable of rapidlydetecting change as it occurs and are carriedout in a manner that is easily understood bystakeholders (i.e., do not involve complex scien-tific surveys and arcane stock-assessment mod-eling procedures). Management objectives areclear and relatively simple. Focus is on a fewkey performance measures and managementactions are aimed at improving those measures,often with little regard for collateral damageto other ecological factors like bycatch/discardspecies.

Flexibility in economic opportunity forstakeholders to target different fishing loca-tions, stocks, and jobs without catastrophic lossin income is observed in systems with highresilience. Complex spatial structure in the eco-logical production system, so that disturbancesand irreversible losses in local productivity arenot felt immediately throughout the managedsystem is also a common trait. Finally, whencomponents of the fishery are overfished, thereis a willingness to entertain and implementactively adaptive, experimental managementpolicies. Such policies are aimed at rebuildinghistorically overfished stock components (e.g.,Walters et al. 1993) and evaluating impactsof artificial stock-enhancement (hatchery) pro-grams.

Note that this list does not include fac-tors that ecologists have sometimes incorrectlypointed to as possible requirements for success-ful ecological management, such as high bio-logical productivity (rapid population growthrates after disasters) and predictability or stabil-ity of population sizes over time. Further, notethat not all fisheries with these characteristicshave been managed successfully; for example,

extremely valuable abalone (Haliotis spp.) pop-ulations satisfy most of the criteria, but havebeen decimated along most of the Pacific coastof North America and Australia by illegal fish-ing (poaching).

There are disturbing signs of decreasingresilience in some management systems thatwere historically successful (e.g., Pacific salmonin western Canada). Indicators of an erosion ofresilience in fisheries management include:

1) complication of management objectives,particularly in relation to protection of weakstocks and biodiversity that have resulted insevere, contentious, and economically dis-ruptive regulatory changes (e.g., large areaclosures);

2) diversion of management agency resources(people, funds) resulting in the deterio-ration of necessary monitoring-assessment-enforcement (core adaptive managementactivities) into other activities;

3) adding complexity to stakeholder involve-ment and decision-making processes aimedat meeting the more complex objectiveswhich appear necessary but do nothing toimprove the management system;

4) adopting ecologically trivial or risky projectslike habitat restoration and hatchery pro-duction when such activities detract formcore activities, or no framework is put inplace to determine if such activities are abenefit;

5) complication of scientific research and mod-eling activities aimed at developing moreprecise predictive models (which have beenextremely costly, especially oceanographicresearch, and almost completely ineffective;Myers 1998).

The basic result of increasing institutionalcomplexity has been to generate much verbiageabout planning, but no improvement in essen-tial management activities and in some caseseven an apparent loss of recognition of whatthose essential activities are, i.e., a basic loss ofunderstanding of priorities for effective man-agement.


Actively Adaptive Policiesin Fisheries and Coastal ZoneManagement

Active adaptive management (see Chapter 4)is essential to evaluate the success of ecosys-tem approaches to fisheries management. Theconcept of actively adaptive or experimen-tal management of renewable resources hadits origins in case studies of fisheries, whenwe began to encounter management optionsand questions for which historical data andprocess-based modeling were incapable of pro-viding definitive answers to managers and whenwe started to suggest that the most efficientway to resolve many uncertainties might be totry management alternatives within an experi-mental framework (Walters and Hilborn 1976,Walters 1986). In those early case studies, weexamined simple policy questions with corre-spondingly simple management objectives inmind, for example, “would increases in salmonspawning escapements result in increases inlong-term abundance and harvest, and wouldthose increases be great enough to compen-sate for losses in harvest associated withincreasing spawning escapements in the firstplace?”

Curiously, adaptive management is nowwidely advocated as an approach to environ-mental management in general but has onlyrarely been used in the design of fisheries poli-cies. There have been a few experimental testsof policies involving increased spawning runsof salmon, tests of simple habitat enhancementand restoration measures such as lake fertil-ization, and assessments of impact of fishingon tropical continental shelf and coral reef fishcommunities (Sainsbury 1988, Sainsbury et al.1997, Campbell et al. 2001). There are caseslike the California Marine Life Protection Actwhere an adaptive approach to management ofMPAs has been legally mandated (i.e., requiredby law), but little indication as yet whethersuch mandates will be respected in practice interms of (1) sound experimental design princi-ples used to select experimental areas and (2)adequate investment in monitoring programscapable of detecting differences between pro-

tected and nonprotected areas as these differ-ences change over time.

So the simple fact is that actively adap-tive management is not widely used, eitherin fisheries or in coastal restoration manage-ment in general (Thom 2000). One excel-lent review site for coastal restoration stud-ies (http://www.nemw.org/restoration.htm) onlymentions adaptive management for two casestudies (San Francisco Bay and Florida Ever-glades), where very little is actually being donein the field. For two other cases where verylarge-scale restoration tests are in fact beingconducted (flow and sediment restoration inthe Mississippi River Delta, oyster bed restora-tion in Chesapeake Bay), the tests are not evenlabeled as adaptive management experiments.

Infrequent past application of adaptive man-agement to fisheries does not lessen its rel-evance to current issues. In response towidespread fisheries declines, a wide range ofnew policies is being implemented, often withunknown consequences. For example, fisheriesmanagement agencies are rapidly moving fromtraditional single-species harvest-managementapproaches to broader ecosystem management(Corcoran 2006), largely in response to polit-ical pressure from conservation interests. Theresulting policy initiatives will cause immedi-ate economic hardship for some traditional fish-eries stakeholders, but the ecological outcomesare grossly uncertain, and there is almost no his-torical experience from which to build. Theseinitiatives range from developing networks ofMPAs, to requirements for use of more selec-tive fishing gear, to direct control or cullingprograms for unwanted species, to establish-ment of new legal arrangements of fishing rightsintended to create incentives for fishermen tocooperate in sustainable management. Avail-able ecosystem models warn us that some ofthese initiatives could backfire badly, particu-larly those involving protected areas (see, e.g.,Walters et al. 1999 or Fig. 11.10 in Walters andMartell 2004) and culling of unwanted species(Yodzis 2001). Even the outcomes of “obvious”management improvements aimed at reduc-ing bycatch of nontarget sensitive species (e.g.,bycatch reduction devices for shrimp trawling)are indicated by ecosystem models to be highly


uncertain, due to our inability to predict howthe fish community as a whole will respond tothe changes in mortality rates and competition-predation regimes that they will cause (Walters2007). It is not that ecosystem management is abad idea; rather it is important to treat this shiftin management paradigm as an experimentalopportunity from which managers can learn.

Several authors have reviewed reasons forthe lack of widespread implementation of adap-tive management (Walters 1997, 2007, Allanand Curtis 2005, Schreiber et al. 2004; seeChapters 6–8). Reasons range from the highcost of monitoring programs implied by com-plex spatial experimental designs, to institu-tional barriers created by inappropriate incen-tive systems for government employees, tolack of individual leadership in coordinatingand pushing forward the activities of vari-ous groups (scientists, managers, stakeholders)whose expertise and participation is needed incomplicated management analyses and imple-mentation. But despite much analysis and dis-cussion, the only point on which experiencedpeople can agree is that development andimplementation of an effective adaptive man-agement plan requires a very large personalinvestment of time and effort by at least onekey person in a leadership position, and suchkey people rarely step forward and accept suchan arduous challenge.

Overview of Dynamics andPolicy Options for Sustainability

Careful attention to processes controlling sup-ply and demand for fish provide a frameworkfor assessing fisheries sustainability. A very sim-ple graph can be used to summarize the mainideas presented in this chapter (Fig. 10.5). Overtime, the dynamics of fishery development cre-ate a shrinking domain of stability for sus-tainable management. As harvest grows, stocksizes and spatial stock structure are inevitablyeroded, so the safe catch that can be taken with-out eroding or endangering future options forsustained harvesting decreases over time. Thisshrinkage may be exaggerated by slow changes

Time after fishery development starts

A. Maximum

safe catch

B. Minimum

eco

nom

ic c

atch

C. Critical point

Cat

ch

Figure 10.5. Loss of resilience and adaptability offisheries as they develop is associated with conver-gence of two curves, one representing the maxi-mum harvest that can be safely taken without impair-ing future production (Curve A) and one repre-senting the minimum harvest that must be takento prevent economic disruption (Curve B). If thetwo curves cross (Point C, where minimum eco-nomic catch exceeds maximum safe catch), the fish-ery becomes unsustainable from both ecological andeconomic perspectives, fishery managers are facedwith a pathological situation where they must choosebetween risk of ecological decline or economic hard-ship for fishermen, and there may be either acceler-ating ecological collapse (if the stock shows collapsein its range) or at least partial economic collapse.

in habitat capacities as well. The growth inharvest creates increasing economic commit-ment, in the sense of an increasing minimumcatch that must be taken in order to sustaineconomic and social functioning of the fishery.That minimum tolerable catch increases partic-ularly quickly in technology-intensive fisheries,where participants must typically finance cap-ital development through borrowing, which inturn creates a minimum catch necessary to meetloan repayment (interest) requirements.

Provided stock sizes are maintained highenough, and harvest level is low enough not tocross point C in Fig. 10.5, there is considerableecological and economic flexibility (resilience)to carry out management experiments aimed attesting ecological potential, to cope with unpre-dictable variations in ecological production dueto environmental factors, and to invest in the


development and maintenance of stable andprofitable markets for fish products. But oncea fishery has substantially passed point C, sothat maximum sustainable catch is substantiallyless than minimum required economic catch,decision-making dynamics become dominatedby concerns about economic collapse, risktaking comes to dominate harvest-regulationchoices, and the fishery is bound to col-lapse unless economic factors prevent contin-ued overharvesting as stock sizes decline (i.e.,unless there is a bioeconomic equilibrium atnonzero stock size).

A wide range of strategies have been recom-mended for managing fisheries so as to avoidpoint C in Fig. 10.5. These can be roughlydivided into two categories, production-sideand economic demand-side, corresponding tomanagement of the two curves in Fig. 10.5.

Strategies for production-side managementmainly involve precautionary policies that havebeen developed by highly risk-averse fisheriesscientists and other ecologists, where stockassessments and harvest regulations are delib-erately chosen to be conservative enough toassure that the safe harvest curve remainshigh over time. A second production-side strat-egy, which we might call the engineering pol-icy approach, attempts to shore up the safecatch curve through engineered habitat man-agement and artificial propagation (hatchery)systems. That second strategy has been usedprimarily with Pacific salmon and has beenwidely criticized as a no-win option, wherenatural production systems are likely to bereplaced by expensive and possibly unsus-tainable artificial production. The replacementeffect occurs because of two mechanisms, fail-ure to reduce harvest impact on natural stockswhen enhanced stocks continue to attract fish-ing, and competitive impacts of enhancedstocks on survival and growth rates of natu-rally produced fish (Hilborn 1992, Hilborn andEggers 2000).

Recommended strategies for demand-sidemanagement mainly revolve around assign-ment of property rights or other institutionalconstraints on fishing effort, along with regu-latory approaches that discourage overinvest-ment in fishing capital and create incentives

for fishermen to advocate rather than opposeproduction-side limits on harvesting. Tactically,such systems may be as simple as license lim-itation programs that cap the number of fish-ing vessels and gear deployed by each vessel(to limit fishing effort), or individual transfer-able/vessel fishing quotas (ITQ or IVQs) thatlimit either fishing activity or total catch perowner (see Chapter 11). Once in place, suchrights often lead quickly to demands by fish-ermen for cooperative science and assessmentprograms, and even for management experi-ments, to provide information needed to pro-tect their investments. In some cases where fish-ermen have taken over the financing of researchand assessment (essentially, adaptive manage-ment) programs, spending has increased com-pared to what public agencies were willing tospend. So there is considerable hope for thefuture of fisheries research and managementdespite the rather dismal history implied bymany of the case studies discussed earlier.

Summary

Overfishing is the primary cause of recentunsustainable changes in marine fisheries. Thisresults from both increased demand for fish andincreased capacity to catch fish. About 25% ofthe world’s fish stocks appear to be produc-ing only a fraction (10% or less) of historicalcatches. Subsidies to industrial fleets drive mostof the overfishing that has occurred. These areintended to maintain national capacities to reapprofits from international fisheries and to softenthe economic hardships to individual fishermenwhen a fishery collapse occurs. Subsidies pre-vent the fishery from reaching a bioeconomicequilibrium in which fishing pressure declinesin response to declining stocks. With subsidies,fishermen continue to fish even when stocks aredepressed to levels at which stock recovery isunlikely to occur. In localized fisheries in whichindustrial fishing has not developed, the sta-tus of the fishery depends largely on changesin slow variables, including pollution, habitatstructure, and biological interactions related tobiodiversity (e.g., in coral reefs), and local insti-tutions that govern fishing pressure. Fisheries


management is moving into a new era ofecosystem management involving multispeciesmanagement; establishment of MPAs; regula-tion of bycatch and habitat disruption (e.g., bytrawling); and enhancement of fishery produc-tion (through hatchery and habitat enhance-ment programs). Given the emergent condi-tions, adaptively managed fisheries programsof experimentation are important to evaluatethe currently unknown potential of these pro-grams to enhance fishery sustainability andresilience.

Review Questions

1. Describe the two main ways that spatialstock structure changes when a fish stockis reduced by fishing. When fishermen arecapable of following these changes, whathappens to their perception of available fishabundance (and to fishery-based indices ofabundance used by scientists) as stock sizedeclines?

2. Numerical abundance of most fish is lim-ited to at least some extent by restrictedhabitat use at juvenile life stages, which inturn is driven by selection for habitat choicesand behaviors to reduce predation risk. Suchlimitations on abundance tend to weakentrophic interaction (predator–prey) linkages.Fishing down top predators tends to resultin simple decreases in natural mortality ratesof their prey, and fishing on species inthe middle of the food web tends not tocause large changes in abundances of specieshigher in the web. Does this mean that goodsingle-species assessment and managementis enough to insure sound ecosystem man-agement?

3. The restricted habitats used by most fishspecies are in coastal areas, including shal-low reefs and estuaries. These areas arehighly vulnerable to impacts of human activ-ities ranging from eutrophication to out-right habitat destruction through physicalmodification, freshwater flow modification,etc. What are the key monitoring technolo-gies that can be used to document coastal

changes, and what policy instruments (e.g.,zoning) can be used to prevent them?

4. In a controversial review of impacts offisheries on biodiversity loss and oceanecosystem services, Worm et al. (2006) cor-rectly assert that many multispecies fisheriesinvolve specific targeting behavior by fisher-men, such that fishermen may “switch off”a given species as its abundance declines.They assert further that this switching behav-ior may allow the species some chance torecover. Explain why this second assertionis wrong in situations where overall fishingactivity remains unregulated. Do depletedstocks recover, or are they simply keptdown by reverse switching whenever recov-ery starts to become apparent to fishermen?

5. A fisheries management system (fish +fishermen + management agency) may fail,leading to stock collapse at least for range-collapsing fish species, because scientificassessments are not heeded by the manage-ment agency or because the assessments arecommonly biased upward by the same fac-tors that make it difficult for fishermen to seethat stock size is declining. What does thissay about the need for detailed spatial datain fisheries monitoring programs?

6. Consider a developing fishery for somesedentary species that may play some keyrole in the food web, such as a sea urchinthat may have major impact on develop-ment of kelp communities and the fishes thatuse them. Instead of the usual approach ofallowing fishermen to pursue them over abroad area under overall harvest regulationsand/or individual quotas, develop an alter-native management plan based on territo-rial use rights (TURFs) assigned to individ-ual fishermen, with each right holder fullyresponsible for monitoring and regulation ofharvests within one TURF. What opportu-nities would this create for organizing theallocation of TURFs by the managementagency over time so as to create a large-scaleadaptive management experiment to test forundesirable ecosystem impacts before suchimpacts become widespread?

7. Slow variables that can undermine manage-ment systems based on myopic models and


data range from fishery-induced habitat loss(e.g., oyster shell beds) to erosion of spatialstock structure to technological changes thatreduce the cost of fishing enough to make iteconomical to drive stocks to very low lev-els. Formal monitoring programs (ecologicaland economic) seldom provide quantitativedata over long-enough periods to documentsuch slow-variable changes. Where do scien-tists need to look for evidence of the impactof such variables, e.g., historical accounts,“traditional knowledge” of older fishermen,etc.?

Additional Readings

Bellwood, D., T.P. Hughes, C. Folke, and M. Nystrom.2004. Confronting the coral reef crisis. Nature429:827–833.

Clark, C.W. 2005. Mathematical Bioeconomics: TheOptimal Management of Renewable Resources.2nd Edition. Wiley-Interscience, New York.

Clark, C.W. 2007. The Worldwide Crisis in Fisheries:Economic Models and Human Behavior. Cam-bridge University Press, Cambridge.

Hilborn, R., and C.J. Walters. 1992. QuantitativeFisheries Stock Assessment and Management.Chapman-Hall, New York.

Jackson, J.B.C., M.X. Kirby, W.H. Berger, K.A.Bjorndal, L.V. Botsford, et al. 2001. Historicaloverfishing and the recent collapse of coastalecosystems. Science 293:629–638.

Pinkerton, E.W., and M. Weinstein. 1995. Fisheriesthat Work: Sustainability Through Community-Based Management. The David Suzuki Founda-tion, Vancouver.

Walters, C.J. 1986. Adaptive Management of Renew-able Resources. McGraw-Hill, New York.

Walters, C.J., and S.J.D. Martell. 2004. Fisheries Ecol-ogy and Management. Princeton University Press,Princeton.

11Coastal Marine Systems: ConservingFish and Sustaining CommunityLivelihoods with Co-management

Evelyn Pinkerton

Introduction: What Is a CoastalMarine System and Why Is ItImportant?

Community contributions to fisheries manage-ment are important to the sustainability ofcoastal ecosystems, including the people whomost depend on fisheries. A majority of theworld’s population lives along coastlines. Peo-ple are an integral component of coastal marinesystems because of what they do to eitherconserve or damage marine resources. Justas we think of ecological systems as havingstructure, function, and patterns of interrela-tionship and change, so too are human popu-lations usually organized into communities thathave institutions (laws, rules, customary prac-tices) and measurable effects on coastal marinesystems. Human communities and the coastalmarine zone that they most strongly impact arethus logically viewed as an integrated social–ecological system (see Chapter 1).

E. Pinkerton (�)School of Resource and Environmental Management,Simon Fraser University, Burnaby, BC V5A 1S6,Canadae-mail: [email protected]

Anthropologists have documented theimpact of small-scale preindustrial societieson natural systems in a subfield called cul-tural ecology or human ecology. They thinkof human communities as having adaptedto the natural resources in their immediateenvironment in a manner that has producedsustainable rates of human use. For example,studies of hunter-gatherers concluded thatthese small, mobile groups tended to limittheir populations to about 60% of the carryingcapacity of their adjacent local resources (Leeand DeVore 1968). In other words, given fluctu-ations in resource abundance, human societiesadopted patterns of use that were conservativeenough to allow them to survive during timesof scarcity.

But human communities did far more thanpassively adapt to what was available to them.They also significantly altered the landscapeover time. For example, they burned patchesof grassland and forests to encourage vegeta-tion for their own use and to attract animals forhunting purposes (Lewis 1989; see Chapter 6).In the coastal zone, they burned wetlands,contributing to the raising of land levels andthus reduced flooding, an important functionwhen sea levels were rising (Kirwan 2008).They systematically fished certain species at


242 E. Pinkerton

certain times and places in a manner that clearlyaffected marine populations and food webs. Soit is more appropriate to think of a coastalmarine system as a case of mutual adaptationand social–ecological coevolution. We havemuch to learn from the ways in which tradi-tional human communities coexisted with andactively managed adjacent marine resources.There are few coastal areas without a long his-tory of human use in which these types of inter-actions occurred, so it is inappropriate to thinkof “pristine coastal ecosystems, untouched byhumans,” because these were rare indeed.

Instead, what should impress us is theenormous diversity of species and sustain-able patterns of livelihood that developed incoastal zones before the industrial revolution innow-developed countries and the colonizationof developing countries (Swezey and Heizer1977, Johannes 1978, 1981, McCay and Ache-son 1987, Berkes 1988, Cordell 1989, Ander-son 1996, Walter et al. 2000). The industrialrevolution and colonization introduced out-siders and more intensive fishing methods bygroups that did not always live adjacent tothe marine resources they fished. The intro-duction of industrial processing capacity, along-side the stripping away of local rights to pro-tect local resources and livelihoods, led in manycases to the overfishing and collapse of at leastsome local stocks. Yet in other cases, localsystems were resilient enough to survive, rais-ing fascinating questions. Maritime anthropol-ogy has focused on what made the preindus-trial livelihoods sustainable and what happenswhen these old fishing communities becomepart of the modern world – the industrialeconomy. What changes and what remains thesame? How do they evolve? How do they sur-vive intense external pressures that threaten todestroy local resources? What kind of histo-ries can these local coastal systems tell? Whatcan they teach us about principles for sustain-able use? What is the role and importance ofresource dependence, place orientation, owner-ship, and resilience to resource stewardship?

Increasingly, interdisciplinary scholars havebecome interested in the survival of these tra-ditional mechanisms in contemporary societiesbecause they represent a reservoir of institu-

tional strategies for solving certain problems(Schlager and Ostrom 1993, Wilson et al. 1994).To support the survival of small-scale tradi-tional institutions, there is an interest in co-management arrangements that provide forpower sharing between local communities andagencies in resource management (Pinkerton1989a, 2003), and adaptive co-management asarrangements for building the capacity of con-temporary communities to design institutionscapable of working at multiple levels, learn-ing from experience, and responding to change(Berkes et al.1989, Armitage et al. 2007, Plum-mer and Fennell 2007) (see Chapter 4). Thischapter examines the conditions that facilitatethe success of these arrangements in the contextof coastal marine fishing communities.

Problems of Privatizing CoastalMarine Systems

The rise of neoliberalism (the state acting toadvance capitalist markets) and neoclassicaleconomic paradigms and models for manag-ing resources (focusing on the benefits soughtby individuals and the market as the bestregulatory mechanism) has created problemsfor sustainably managing many coastal marinesystems. In such paradigms, the allocation ofprivate rights to individuals and the free trans-ferability of such rights via the market areconsidered to be the ideal. Such systems areassumed to deliver efficiency and even to cre-ate incentives for conservation of the resource(NRC 1999). Starting with New Zealand andIceland in the 1970s, many countries haveexperimented with privatization of access andwithdrawal rights to marine resources throughindividual transferable quotas (ITQs). Quota-based systems were usually designed to solveproblems in state regulation, but were imple-mented without recognizing the possibility ofcommunity-based regulation. In many casesthis approach led to the transfer of fishing rightsaway from a diverse array of coastal commu-nities (or even out of a country), separatingplace from access, and removing the possibil-ity of capturing the benefits of co-management

11 Coastal Marine Systems 243

or adaptive management as described aboveand below. Above all, ITQ systems can repre-sent a loss in flexibility and adaptability in themanagement system, the hallmarks of resilientsystems (deYoung et al. 1999). Flexibility andadaptability are compromised with these poli-cies because property rights to capture a spe-cific amount or percentage of fish leads to evermore costly investments in quotas, even whileit lowers the cost of fishing through concentra-tion of fewer fishermen on fewer vessels. Espe-cially problematic for the flexible managementof ecosystems is the fact that quota rights areusually based on ownership of single-specieslicences and do not create incentives to attendto habitat linkages or species interactions. Pri-vatization has also created problems with eco-logical sustainability and equity; some analystsfind that quotas do not score well on efficiency– usually their most prominent claim (Tieten-berg 2002, Pinkerton and Edwards 2009). Theconservative Organization for Economic Coop-eration and Development in a study of ITQsworldwide found many cases of stock failures,higher enforcement costs or problems, underre-porting of catch, and data degradation (OECD1997). Agrawal (2002) and Baland and Plat-teau (1996) also found that community institu-tions were not necessarily any less efficient thanprivatization. However, ITQs can be appeal-ing to government agencies because they solvean important problem for fisheries managementagencies: in times of declining agency funding,they allow agencies to use quotas-holders to payfor the cost of stock assessment and monitoringof catch (such as 24-h surveillance cameras onboard vessels, and dockside monitoring at deliv-ery sites on land). The first generation of quotaholders can afford to pay these and other man-agement costs, because they have been giftedwith a valuable public good that they are thenallowed to sell at what the market will bear. Theresults have been mixed, in many cases result-ing in overfishing and struggles over the public’s(and even government’s) right to the catch dataand research of quota holders.

In cases where quotas are unavoidablebecause they are the accepted approach, com-munities and cooperatives have sometimesadapted the model by creating community quo-

tas (Langdon 1999, 2008). In such cases, rightsare usually transferable among individual mem-bers of a specified organization, but not at spec-ulative prices and not out of the organization.Hence they are neither fully individual rightsnor are they fully transferable via the mar-ket. In less-formalized arrangements, fishermenmay develop an arrangement to share theirquota with other community members in timesof resource abundance (Loucks 2005).

The reliance on a privatization model such asquotas to solve problems highlights the morefundamental issue of overreliance on single-species management and a single disciplinaryapproach. Many resource-management situa-tions have shown the need for multidisciplinaryteams. Ever since Peter Larkin’s famous dic-tum that “fisheries management is about man-aging people, not fish,” there has been recog-nition of the need to integrate social sciencesinto problem solving. At least half the prob-lems are caused by difficulties in managingwhat people do, and the failure to understandhow human societies function culturally, soci-ologically, politically, psychologically (Wilson1998). Unfortunately, current rates of social–ecological change confound efforts to under-stand the nature of the problem. Still, it iscollaborative and multiscaled efforts, such asadaptive co-management that offer the bestopportunities for meeting the challenges ofsustainability.

International Recognition ofLocal Fishing Communities

There has been a growing recognition of theimportance of coastal marine systems since theywere acknowledged internationally by the 1992United Nations Conference on the Environ-ment and Development, whose initiative oncoastal communities was followed by the 1994UN convention on the Law of the Sea, and the1995 UN Food and Agriculture Organization’sCode of Conduct on Responsible Fisheries.

Chapter 17 of the Rio Declaration fromthe 1992 UN Conference on Environment andDevelopment (paragraphs 17.79, 17.81, and17.93) stated that governments should “take

244 E. Pinkerton

into account traditional knowledge and inter-ests of local communities, artisanal fisheries[i.e., small-scale mixed commercial and subsis-tence fisheries] and indigenous people in devel-opment and management programs.” The Codeof Conduct went on to state that resourcemanagers should “implement strategies for sus-tainable use of marine living resources, takinginto account the special needs and interests ofsmall-scale artisanal fisheries, local communi-ties, and indigenous people to meet nutritionaland other development needs.” Finally, it notedthat governments should “recognize the rightsof small-scale fish workers and the special situ-ation of indigenous people and local communi-ties, including their rights to utilization and pro-tection of their habitats on a sustainable basis.”The 1995 FAO Fisheries Code of Conduct onSustainable Fishing insisted on “transparencyin data collection and in decision-making prac-tices, to ensure that independent analysis ofdata and assumptions is possible by crediblethird parties, and that the public owners of theresource have access to basic information.” Thedocument also emphasized the importance ofmeasures that prevent excess fishing capacityand ensure that the exploitation of the stocksremains economically viable and that the bio-diversity of aquatic habitats and ecosystems isconserved (FAO 1995).

These international agreements and codesillustrate the growing recognition that the sus-tainability of coastal communities is criticalto the resilience of the costal marine social–ecological system as a whole. Through therecognition of the rights of fishermen, the legit-imacy of their local knowledge, and their mean-ingful role in the data collection, research, anddecision making, resource governance is morelikely to be responsive to change and providefor the livelihoods of those who most dependon the marine coastal system.

These international agreements and codes donot, however, explain the conditions that arenecessary for this vision of community partic-ipation to actually occur. This chapter looksat the key factors that allow the vision tobecome a reality, by asking the following ques-tions: Under what conditions will coastal com-munities play a positive role in fisheries and

coastal management? Under what conditionsmight we expect communities to play a nega-tive role, damaging these same systems? Howcan community involvement in coastal sys-tems governance address the increasing pres-sures from nonsustainable fishing practices bylarge nonlocal industrial fleets and from manyforms of development and pollution in thecoastal zone such as oil and gas development,tourism, and unplanned urban sprawl? Howcan involvement of communities contribute tothe flexibility and responsiveness of coastalmarine systems in conditions of directionalchange?

The pages that follow explore a variety of sit-uations in different parts of the world that illus-trate both the conditions allowing sustainableoutcomes in coastal marine systems and thespecific benefits that are associated with this. Inso doing, critical challenges to sustaining com-munity livelihoods of coastal systems will beexplored.

General Conditions UnderWhich Coastal CommunitiesConserve Marine Resources

If international bodies believe that coastal com-munities can make such a difference in con-serving coastal marine resources, we need toknow more about when we can expect thisto occur and when we should be concernedthat coastal communities might plunder theseresources. It is useful to consider the impact offive conditions discussed below that social sci-entists believe are good predictors of conserva-tion by coastal communities. Because resourcegovernance emerges in each situation throughits own path-dependent set of conditions (seeChapters 1 and 4) and because all situations arecomplex, no specific set of conditions is nec-essary or sufficient to achieving sustainability.Yet identification of key conditions under whichcoastal communities make a positive contribu-tion to fisheries conservation can help groups atall scales to identify the possible actions neededto sustain coastal marine systems.


Condition 1: Communities HaveStrong Access Rights to Local MarineResources

Think of it this way: If you are relatively poorand have a historical dependence on your localmarine resources, but you now have no right tofish, what are you likely to do? Will you poachas a way of taking fish away from the outsiderswho do have the rights? Will you decide todevelop the coastal zone for other purposes thatprovide economic opportunities for you? Willyou support oil and gas development, intensivefinfish aquaculture, wind farming and ferry ter-minals in tidal flats, cruise ship dumping, super-tanker traffic, and many other industrial usesfor the coastal zone that can negatively impactfish habitat, because at least these will providesome economic activity in your community?There must be enough local livelihoods in fish-ing that are perceived as sufficiently importantby locals to counteract possible temptations todevelop coastal zones for other purposes. Sincefishing is often perceived as an occupation oflast resort, or as poverty alleviation in devel-oping countries, a local constituency willing tostand up and speak for protection of the fishis often required to prevent development withno regard for fish habitat. If local communitieshave no economic incentive to fight for thesevalues, there may be little standing in the wayof fish habitat destruction in the estuaries, bays,and river mouths that are often critical nurseryareas for juvenile fish. Fishing rights holders indistant urban centers usually have little knowl-edge of the importance of such areas and donot depend on any one area for their livelihood,so they are not likely to be a strong voice forprotection.

Similarly, it is the rural fishermen catchingfish at a smaller scale than industrial fleetswho are most likely to notice changes in stocksrelated to fishing, as is illustrated in the fol-lowing example. The Catholic priest ThomasKocherry in India organized two successfulnation-wide strikes of fish sellers and the smallfishermen who were their spouses, because thepoorest and smallest fishermen observed thatthe species that were once abundant in theircatches were disappearing due to the growth

of an offshore trawl fleet. As a result, Indiabanned foreign trawlers from its territorialwaters (Thalenberg 1998). In this case, the smallfishermen had no formal legal rights to accessfish that could stand up against the large indus-trial fleets. However, through the organizedaction of Kocherry, they were able to exercisede facto rights to catch their share of the fishand to prevent outsiders from taking it first.Without the action of these small fishermen, itis likely that the trawl fleet would have over-fished or extirpated many species before it wasevident that their actions had this effect andbefore the Indian central government wouldhave been willing to take any action. If thesmall fishermen in the coastal zone had hadno fishing access rights at all, there wouldhave been no nationwide strikes to ban for-eign trawlers. Holding rights to fish is thus thefirst and most fundamental condition that mustexist for coastal communities to play a role inconservation.

Condition 2: Communities HaveStrong Rights to Participate inManagement Decisions

It takes far more than informal access rights,exercised through strikes and protests, to pro-duce conservation, however. To really makea difference, communities must have a bun-dle of rights to participate in the managementsystem at multiple levels, from data collec-tion to policy development (Pinkerton 2003.Table 11.1 illustrates a hierarchy of rights, somefar more powerful than others, that would allowa coastal community to participate meaning-fully in management and assert its vision ofconservation. On a continuum from weakerand smaller-scope rights to stronger and larger-scope rights, these involve rights to data col-lection, data analysis, planning the timing andlocation of the fishery, fishing methods, allo-cation of catch, exclusion of outsiders or non-rights holders, protection of habitat, monitor-ing of harvest and habitat, creation of the rulesby which all these decisions are made, enforce-ment of the rules, coordination of harvests andother competing uses, returning optimum value

246 E. Pinkerton

Table 11.1. Hierarchy ofrights in fisheries-managementdecision making, based onPinkerton and Weinstein(1995).

Type of management right Specific right

Lower order rights Data collectionData analysis

Higher order rights Plan timing and location of fisheryRule-making re fishing methodsAllocation of fishing opportunity among

rights holdersEnforcement of fishing rulesDefining who has fishing rights

Broader rights affecting other Rule-making re fish habitat protectionactors and users of marine Enforcement of habitat-protection rulesspace Coordination of fishing and other

competing uses of marine spaceReturning optimum value to fishermen

Highest level rights Fisheries policy developmentIdentification of key problems, issuesCreating a vision of what fishery is

desired, goals of management

to fishermen, policy development, identificationof key problems, issues, and goals of manage-ment, creating a vision of what kind of fishery isdesired.

Two of the oldest examples of coastal com-munities developing this range of fishing rightsare discussed below: the US western Washing-ton State treaty tribes and the coastal munic-ipalities of the Philippines. Both of these sys-tems evolved from long-term precolonial defacto management by local communities andbecame de jure systems when laws or courtdecisions clarified the nature of the rights inthe 1970s. Both systems continued to evolve,clarifying and extending the rights of coastalcommunities over the next four decades, assummarized in Table 11.2 and briefly discussedbelow.

The 1974 “Boldt Decision” (US v Washing-ton) was a US Federal Supreme Court deci-sion (based on the court’s interpretation of ear-lier treaties) that recognized the right of Indiantribes in Western Washington to up to a 50%share of the salmon that passed their custom-ary fishing sites, mostly located on rivers, bays,and estuaries. Tribes had been unable to exer-cise rights to a share of fish recognized in earliercourt decisions, because the harvest was man-aged by the state government in such a way thatfew salmon remained by the time this migra-tory species reached the marine and riverine

territories in which the tribes could legally fish.The state managed the nontribal commercialand sport fisheries so that all but approximately5% of the salmon were harvested elsewhere.Judge Boldt reasoned that only by recognizingthe tribes’ right to participate in planning andregulating the entire harvest (which he called“concurrent management”) would their alloca-tion right ever be exercised. In other words,the treaty’s promise that the tribes would sharein access to the fish alongside the citizens ofthe territory (Condition 1) was unrealizablewithout a second and higher-level right beinggranted: the right to participate in manage-ment decisions about how the harvest wouldbe conducted. The treaty tribes invented theterm “co-management” to describe this situa-tion and gradually over the next two decadesbegan to assert the full range of rights so thatthey shared management authority with thestate of Washington in all aspects of manage-ment. Thus we can note that, in the hierar-chy of rights, the higher-level more powerfulrights (such as harvest management) do notautomatically accrue simply because a lowerright (such as access to an allocation of fish)exists.

The Washington Department of Fisheries(WDF) originally resisted the exercise of thetribes’ right to co-manage, beginning with tribaldemands for access to WDF’s stock abundance


Table 11.2. Fishery management rights held by Washington tribes and Philippines municipalities and Fisheryand Aquatic Resource Management Councils (FARMCs).

Rights Washington tribes Philippines municipalities and FARMCs

Data collection on fishstock abundance

de jure No

Data collection onfishermen’s gear, vessels

de jure de jure for registration of gears andvessels by municipality (not fish)

Data analysis de jure NoPlan timing and location of

fisheryde jure de jure, local fisheries ordinances,

approved by stateRulemaking re fishing

methodsde jure de jure, local fisheries ordinances,

approved by stateAllocation of fishing

opportunity amongrights holders

de jure de jure only in that local fishingcooperatives have preferential accessand marginal fishermen’sorganizations are exempt from fees

Monitoring andenforcement of fishingrules

de jure de jure, shared with state,municipal-based interagency lawenforcement teams

Rule-making re fish habitatprotection

de jure de jure, shared, municipal involvement indesignation and management ofmarine protected areas

Enforcement of habitatprotection rules

State de jure, by state and/or FARMC in theirown protected areas

Coordination of fishing andother competing uses ofaquatic space

Shared with state; de factoparticipation inRegional ManagementCouncils

de jure, assist in the preparation ofdevelopment plans

Returning optimum valueto fishermen

No No

Fisheries policydevelopment

de facto de jure

Identification of keyproblems, issues

de facto de jure, identification of problems inparticular areas and strategies toaddress them

Creating a vision of whatfishery is desired

de facto de jure, mandated to FARMCs asmethod of rural development

Creation of financialcapacity to manage

de jure (federal funding) de jure, municipalities levy taxes formanagement and partner with NGOsand international developmentagencies to deliver basic services,build capacity, develop livelihoodprojects; FARMCs depend onmunicipalities for budgets and dividefishing violation fines with them

data, catch data, and participation in the verydefinition of conservation. In this case, defin-ing “conservation” meant deciding how muchsalmon should not be fished, but rather allowedto spawn in each stream in order to repro-duce the next generation. (The tribes tendedto advocate lowering the catch and allowingmore salmon to spawn.) The WDF did not want

to reveal the paucity of its stock abundancedata and the level of uncertainty surround-ing its analysis and decision-making about theallowable catch. Furthermore, it did not trustthe tribes more than any other fishermen toreport their catch accurately, especially sincesome tribes had asserted their treaty rights(see Cohen 1986) through illegal fishing for

248 E. Pinkerton

decades. The final negotiation of a formal co-management agreement (see Chapter 4) thatresolved these problems, among others, took 10years and resulted in a complex power-sharingrelationship in which the state and tribes agreedto work jointly on every aspect of data gath-ering, data analysis, and harvest planning, andeventually played complementary and mutu-ally supportive roles. The tribal–state relation-ship was less a delegation or decentralization ofpowers than a complex division of powers anda collegial collaboration in problem-solving aspresumed coequals. The two parties eventuallydeveloped a high level of trust and learned tomake the best use of limited funding by sharing(and sometime agreeing to tradeoff specializa-tions) in virtually every aspect of managementand every stage of planning, from interna-tional negotiations to collecting data on indi-cator streams. The sharing and improvementof data gathering and data analysis throughmutual accountability provided the foundationon which trust was built in harvest planning(Pinkerton 2003).

The court had not envisaged all the complex-ities of co-management; so many of the otherrights were recognized only through negotia-tions after intense political struggles, or con-tinued use of the courts and its extensions.Through the repeated exercise of their har-vest management right, the tribes graduallylearned what other rights were necessary formaking this core co-management function oper-able. As soon as harvest co-management pro-tocols were agreed to in 1984, a new set ofissues around co-management emerged. Theseincluded habitat protection, regional planning,setting broader policies at a higher level, andinternational allocation (interception) agree-ments. The co-management system set thestage for a complex multistakeholder exercisein watershed analysis and eventually for themost challenging exercise in complex collabo-ration ever attempted, involving federal agen-cies regulating endangered species protectionand water quality under federal statutes. Thusharvest regulation emerged as only a smallpart of what would eventually be involved inco-management (Pinkerton 1992, 2003, Ebbin1998, Singleton 1998).

A similar evolution of rights processoccurred in the Philippines, which also had hada system of village jurisdiction over coastalresources before Spanish colonization in theseventeenth century imposed state control offishing and gradually eroded traditional villagerights (Pomeroy 2003). The decreasing catchby small-scale fishermen in the 1970s led thePhilippines government to increase fishingeffort by equipping them with vessels and gearto venture further from shore, a strategy thatultimately resulted in lowered catches. Begin-ning in 1991 with the Local Government Code,which devolved key governance functionsfrom the national to municipal level (coveringareas 15 km from shore), the national govern-ment began to actively promote devolution ofmanagement authority as part of a develop-ment and poverty-reduction initiative. Thenin 1995, the Fishery and Aquatic ResourceManagement Councils (FARMCs) were setup and eventually brought under nationallegislation as government consultative bodiesto allow greater participation of fishermen’sorganizations in management. These advisoryand monitoring bodies work mainly at andacross three levels of governance (munici-pal, regional, and national) and assist in theformulation and implementation of nationalfisheries policy. Advisory and monitoring bod-ies are considered public–private partnershipsbetween government, nongovernmental orga-nizations (NGOs), and private industry. In thiscase “private industry” consist of commercialfishing ventures (6% of fishermen), processors,and aquaculture (26% of fishermen), whileNGOs consist of organizations of fishermenparticipating in municipal fisheries (65%).Since 80% of the dietary protein in ruralareas comes from fish, food security is evenmore important than livelihoods in municipalfisheries. The participation by these groups incivil society (engagement with informal andformal institutions) was deemed to be a criticalelement of poverty alleviation, a key goal ofthe legislation, and hence, fishermen and fishworkers make up one third to two thirds ofall FARMCs. These bridging organizations aremost active in law enforcement and pollutioncontrol (Benjamin et al. 2003).


Initial and subsequent devolution of deci-sion making to communities through the imple-mentation of FARMCs was further supportedby other aspects of the process that (1) simpli-fied procedures for national approval of localfisheries ordinances that had to comply withfederal legislation and policy; (2) strength-ened enforcement of fisheries laws throughmunicipal-based interagency law-enforcementteams; (3) allowed the identification of specificproblems in particular areas and the develop-ment of strategies to address them; (4) man-dated the protection of coastal fish habitat; (5)permitted consolidation and coordination ofthe services and resources of local governingunits for common purposes; (6) allowed munic-ipal governments to levy taxes and fees forresource management that had once been allo-cated to the federal government; (7) allowedmunicipalities to partner with NGOs and inter-national development agencies to deliver basicservices, build capacity, or develop livelihoodprojects; (8) mandated the rights of local fishingcooperatives to have preferential access and,in the case of marginal fishermen’s organiza-tions, exemption from fees; (9) mandated thatthe fisheries management councils at nationaland municipal levels would assist in the prepa-ration of development plans; and (10) man-dated municipal involvement in the designationand management of marine protected areas(Pomeroy 2003, Pomeroy and Viswanathan2003).

In practice, the FARMCs and municipali-ties have been actively involved so far mostlyin the creation of protected areas, the protec-tion of habitat, and in enforcement of regula-tions (Benjamin et al. 2003). One island munic-ipality in the Philippines has used the pro-tected area strategy effectively to rehabilitatecoral reefs and ban use of destructive fish-ing technology (cyanide, explosives, fine meshnets) introduced by new arrivals to this areain the 1970s. By 1988, living coral cover haddeclined to an average 23% and catches hadalso drastically declined. With the help of alocal NGO, in 1989 the municipality created aninner no-take sanctuary and an outer reserveallowing only nondestructive technology, thenproceeded to cooperatively create ordinances

for fishing, provide a boat and equipment topatrol coastal waters, and monitor fishing activ-ities closely. The local daily average catch perfisherman went from 3 kg in 1988 to 10 kgin 1998, while living coral reef cover increasedfrom 23 to 57%, accompanied by increased fishspecies diversity (Pomeroy and Viswanathan2003). This example illustrates a frequent find-ing that when local authorities have a centralrole in the design of marine protected areas(MPAs), and the regulations are well-respectedand well-enforced, local fishermen directly ben-efit from increased fish abundance and stronglysupport the MPA. In this case, the design of theno-take and fishing zones, the banning of intro-duced technology that local fishermen under-stood to be destructive, and effective monitor-ing and enforcement were key to the success ofthe MPA.

Another case demonstrates that flexibilitythrough adaptive co-management can also con-tribute to the success of MPAs. In that case,Cinner et al. (2005) found that locally-maderegulations for a MPA invoked high rates ofcompliance because there were adaptive peri-odic openings and closures practiced explic-itly to meet social goals, including providingfood resources at a time of social significance,rather than to fulfill nonlocal notions of con-servation or resource management. The authorsconcluded that “The reality in many develop-ing countries is that fully closed reserves oftensuffer from low levels of enforcement, moni-toring, and compliance and are thus, in manycases, ineffective at either conserving marineresources or enhancing fisheries. In contrast, wefound that adaptive periodic closures increasedfish biomass and average fish size . . . . When theproper socioeconomic factors are found, andlarge, permanently-closed reserves are unreal-istic or have repeatedly failed, adaptive peri-odic closures may be a more viable conserva-tion strategy” (Cinner et al. 2005).

But sometimes the forces of habitat degrada-tion, global warming, and overfishing threatento overwhelm an MPA’s efforts to the extentthat adaptation at another scale is called for.Australia offers an example of a transformationin MPA governance through reconceptualizingthe need for greater and more interconnected

250 E. Pinkerton

no-take areas after a marine park was longestablished. The Great Barrier Reef MarinePark (GBRMP), the largest coral reef sys-tem in the world, went from 5 to 33% no-take areas in response to a drastic decline inspecies and coral health, combined with greaterunderstanding of the importance of connectiv-ity of larvae and interactions between reef andnonreef habitats for maintaining the resilienceof the entire ecosystem (see Chapter 5). Thefocus of the rezoning was on protecting bio-diversity and maintaining ecosystem functionand services, rather than on maximizing theyield of commercially important fisheries. Keyto the rezoning was internal reorganization ofthe reef-management agency to build supportand understanding for the MPA, to encourageagency scientists to think beyond their individ-ual sample sites or specialized expertise, andfor all groups to collectively reach a biore-gional perspective on the GBR as a whole.Also key was extensive public education, tworounds of extensive public consultation, innova-tive interactive forms of information exchange,and strategic building of public and legislativesupport. As a protected area the size of Califor-nia, this site in Australia set a new standard inmobilizing public support at a scale commensu-rate with the scale at which the agency believedthe park ecosystem should be viewed (Olssonet al. 2008).

Table 11.2 illustrates the similarities in thebundles of management rights held by theco-managing coastal communities in Washing-ton State and the Philippines and also sug-gests some of the differences in the situa-tions of developed and developing countriesin their efforts to enact co-management. Therole of government in providing technical sup-port, credit, marketing assistance, or protec-tive legislation is especially important to co-management in developing countries such asthe Philippines (Chevalier and Buckles 1999,Pomeroy and Berkes 1997), although this isoften the case in developed countries as well.Economic development and the protection ofthe large percentage of poor or marginal fish-ermen is a more central concern of govern-ments in developing countries. It is worth not-ing, however, that external or nonfishing NGOs

such as universities, international bodies, oradvocacy groups may be the chief sourcesof legitimation for the co-management rela-tionship in developing countries such as theDominican Republic (Stoffle et al. 1994, Cheva-lier and Buckles 1999), not necessarily gov-ernment, even though government must even-tually provide protection for co-managementto work. But dependence on legitimation byexternal bodies such as NGOs can be unhelp-ful or even a disadvantage if a developedcountry NGO does not have adequate expe-rience with the community in a developingcountry it is attempting to assist, and if thereis no internally generated effort and strongdesire to develop co-management. If externalNGOs offer funding to communities to exper-iment with and adopt co-management pro-cesses, goal displacement may occur (see Chap-ter 4). A community may discontinue all effortstoward co-management after the external fund-ing is exhausted (Hara and Nielsen 2003). Inthis situation, the community’s goal in the co-management project becomes to secure fund-ing rather than to develop co-management rela-tionships and management rights that wouldcontribute to conservation and livelihoods. Insuch a case, the community may have lackedthe capacity, vision, or will to take on theresponsibilities of co-management (as discussedunder Condition 4), and the mere provisionof funding and legitimation by external NGOswas not sufficient to overcome the absence ofthese conditions and allow the developmentand assertion of management rights. In Chile,most of the existing MPAs are sponsored andadministered by external private organizations(Fernandez and Castilla 2005). Superimposingco-management policy, in the form of territo-rial use rights for fishermen over an existingtraditional community-based natural-resourcemanagement system in Chile, eroded trust inthe community and reduced adaptive capacityand resilience in resource-stewardship (Gelcichet al. 2006).

All of these findings underline the over-whelming importance of management rightsheld by communities as a condition neces-sary for their ability to implement conserva-tion and sustainable fishing rates. As noted


above in the Philippines and in the DominicanRepublic examples, the ability of communitiesto ban destructive technology is a particularlyimportant example of how these rights can beimplemented to the benefit of conservation.Also noted is the fact that institutions for co-management evolve through time, and activeadaptation of these arrangements and the pro-cesses they employ can strengthen their perfor-mance. In the following discussion, these rightsare considered co-management rights (giventhat government will play some role in man-agement, even if communities take the lead onmany issues) and are assumed to be a basicunderlying condition that enables other condi-tions to occur.

Condition 3: The Nature of theResource Must Lend Itself Well toCo-management Arrangements

So far the discussion has focused on the natureof the governance arrangements between thecoastal community and the level of governmentcharged with the legal responsibility to man-age fisheries, usually the national government.But the biophysical and logistical characteris-tics of the fish matter a great deal to sustain-ability because of the difficulty or ease theycreate in management (see Chapters 4 and 10about the fit between institutions and ecosys-tems, and the problems of misfits). For exam-ple, species that are immobile such as shellfishor that remain year-round in nearshore waters,lend themselves well to one of the oldest andmost common forms of local management, Ter-ritorial Use Rights Fisheries or TURFs, as theywere famously named by FAO economist Fran-cis Christy (1982). A TURF is adjacent to acoastal community so that fishing activity canbe monitored and locally made rules can beeasily enforced with low transaction costs. Thecommunity exercises either de facto (informal)or de jure (formal and legal) ownership overits local “territory.” Many of the fish speciesgoverned by municipalities and FARMCs inthe Philippines were immobile or less-mobilespecies, and it was thus possible to exclude

other fishermen from taking them. However,the fact that the western Washington tribescould co-manage salmon, a highly migratoryspecies, shows that mobility does not make localmanagement impossible, merely more difficult.

In addition to resource mobility, Agrawal(2002) identified seven other resource charac-teristics that can affect the ease or difficultyof local management: (2) resource size (affect-ing ease of monitoring and enforcement), (3)boundary clarity (affecting ease of exclusionand governance), (4) riskiness of resource flows(affecting willingness to invest in management),(5) scarcity and value (affecting how impor-tant the resource is economically, which inturn affects willingness to invest), (6) salience(affecting how culturally important and valu-able a resource is considered to be), and (7)resource storability (affecting the logistical andtechnical possibilities for holding or processingafter catch). To this list, Pinkerton and John(2008) added (8) resource visibility (affect-ing ease of monitoring; 9) resource spoilabil-ity (affecting how long a resource can be heldafter being caught without significant loss ofvalue), and (10) resource defendability (thelocation of resource concentrations in relationto possibilities for exclusion, monitoring, andenforcement). A historical or dynamic dimen-sion should be added to this list, because a num-ber of studies suggest that, if a stock is badlyoverfished by outsiders or newcomers, localsmay abandon hope of conserving and partici-pate in finishing it off (Befu 1980, McGoodwin1994). In other words, the resource must be suf-ficiently abundant to not trip off the feedingfrenzy that occurs when everyone decides it isnot possible to conserve the resources and turnsattention instead in getting a share of what willsoon be gone.

Condition 4: The Characteristics of theCommunity Must Lend ThemselvesWell to Co-management

Integrating the work of many scholars, Agrawal(2002) also identified eleven characteristicsof a community that is well-adapted to self-management or co-management. (1) relatively

252 E. Pinkerton

small size (not too large to work together effec-tively), (2) clear membership rules (who is acommunity member and who can fish), (3)shared norms of behavior, (4) trust becauseof experience of past success in workingtogether (social capital), (5) appropriate lead-ership (connected to both traditions and possi-bility of innovation), (6) interdependence (needfor support of other community members tohave a viable livelihood and participation in asocial group), (7) homogeneity of identities andinterests, heterogeneity of endowments (a vari-ety of skills and capacities possessed by commu-nity members), (8) low level of poverty, (9) highlevel of dependence on resource, (10) residentslocated close to resource, (11) low or gradu-ally changing demand, (12) access to conflictresolution.

To this list I add eight other characteristicsrelated to local culture and capacity (Pinker-ton 1992, 1991, 1989): (13) a shared sense ofhistory and cultural continuity (leading indi-viduals to care about more than their immedi-ate self-interest and to identify themselves witha historical tradition and with future genera-tions), (14) shared values that are sufficientlyclear to grant moral legitimacy to a manage-ment system which reflects these (e.g., senseof fairness in access to the resource by users,stewardship), (15) ability of key individuals(leaders) to articulate a broad, holistic visionregarding sustainability, including ecosystemvalues, (16) political will to work consistentlytoward the vision, (17) willingness and abil-ity to develop local skills and capacity, (18) anenergy center, dedicated person, or group whoapplies consistent pressure to advance the pro-cess, (19) political attitudes that allow strate-gic alliances and collaboration, or ability tocontain conflict for the sake of larger goals,(20) local and regional bodies that can coor-dinate planning processes, act as informationdepots and clearinghouses, mediators, regionalconsensus builders, policy developers, and localhuman and financial resource mobilizers andorganizers.

Communities lacking some of these charac-teristics will still be able to co-manage, but themore of these assets they lack, the more difficultwill be their task.

Condition 5: The Nature of theCommunity’s Relationship withOutside Groups and GovernmentMust be Favorable, i.e., Strong Enoughto Promote its Attempt to ClaimPower in Governance

There are also 17 characteristics that enablecommunities to mobilize sufficient politicalsympathy and power to implement power shar-ing with government agencies (Pinkerton 1992,1993). A first set of characteristics emerge fromcommunities’ capacities to express their polit-ical aspirations effectively: (1) ability to articu-late local ecological knowledge and local under-standing of what works in fishing regulations ina manner understandable to natural scientistsand government regulators, (2) ability to iden-tify community interests with the public inter-est in sustainability, (3) ability to demonstratethat radical reform is necessary and not beingaddressed.

A second set of characteristics pertain tocivil society and allow communities to mobi-lize sufficient power to roughly balance thepower of groups that have captured the gov-ernment agency in the past: (4) access to oldand new public forums of debate and dissemi-nation of opinion, including issue networks, as asource of ideas, technical information, alterna-tive models and complementary resources, link-ing community to government agencies, univer-sities, NGOs, and credible organizations, (5) asocial movement in the larger society, in theform of new and expanding organizations, newand expanding forms of political expression,(6) access to sources of power such as legaladvice, strategy advice, legislative bodies, pub-lic boards, (7) access to logistical and financialresources.

A third set of characteristics concern exist-ing institutions that can be accessed by the com-munity to press its case for claiming manage-ment rights: (8) sympathetic courts and sup-portive legal precedents (re attitude towardlocal involvement in decision-making, tribalrights, etc.), (9) protective or permitting legis-lation, (10) an appeal body to assist with localequity questions and principles, (11) an ade-


quate scale of planning to control enough vari-ables, i.e., the political boundary to which rightsapply.

A fourth set of characteristics concerns thenature of the management agencies with whichcommunities have to negotiate power-sharing:(12) agencies have hands-on experienceworking with fishermen, including personalrelationships and common understandings, (13)agencies grasp the importance/value of localknowledge and are willing to work to combineit with natural science, (14) agencies are willingto work with communities to experimentthrough adaptive co-management with whatlevel of fishing impact is sustainable, (15) agen-cies are willing to negotiate with communitiesand together with them experiment and dopilots to get started, (16) pilot agreements endup being formalized, legalized, and multiyear.

Finally (17) market and product form devel-opment is concomitant with the supply (so thatcommunities can successfully sell their fish).

Benefits of InvolvingCommunities inCoastal-Management Systems

The contribution of coastal communities’involvement in fisheries management dependson the specific ecological, economic, and socialobjectives. We saw in the Philippines examplethat some of the national objectives of fisheriesaccess allocation in developing countries maybe primarily social: to provide economic andfood-security opportunity to coastal communi-ties with few alternative livelihoods, and withlong traditions of dependence on fishing, givingpreference to the most needy and dependentgroups. In developed high-latitude countriessuch as Norway and Canada, governmentpolicies for the allocation of fishing opportunityhave historically been a way to encourage andstabilize settlement on the most northern coast-line. Coastal fishing communities serve multiplenational interests by increasing a country’s abil-ity to claim jurisdiction over adjacent waters(thus controlling shipping and oil drilling) andthe ability to support national defense bases

of operation or reconnaissance. Governmentstherefore sometimes allocate fishing opportu-nity in the form of fishing licenses or favorablepolicies toward fishing/farming communitiesfor purposes that have little or nothing todo with conservation or economic efficiency(Jentoft 1993, deYoung et al. 1999).

Whatever multiple purposes may be servedby fisheries management, most federal fisherieslegislation in developed or developing coun-tries is likely to identify the primary goal tobe conserving fish stocks and managing themsustainably. Let us therefore consider what thecontribution of coastal communities might be tothis particular goal.

As human understanding of ecological pro-cesses grows more complex, we increasinglyappreciate the importance of long-term moni-toring, not only for understanding food webs,species-habitat linkages, and trends in abun-dance in particular locations, but also for under-standing and anticipating specific impacts ofregime shifts caused by social or environmentalchange such as global warming (see Chapters 2and 4). Without community involvement in andconcern for monitoring, it would not be possibleto document or grasp in much depth the natureand extent of change. In addition to global pro-cesses that may be reflected in long-term mon-itoring, there are also local processes that havetheir own particular patterns. Documentationof these local processes can help decision mak-ers take advantage of more precise time- andplace-specific knowledge so they can integrateanalyses of different scales (Neis and Felt 2000,Kofinas 2002, Degnbol 2003). Structured andinformal/traditional community-based monitor-ing is also still used by some indigenous com-munities to control local use, and to curtail itin times of overuse (see Chapter 6). Histori-cally, tribal chiefs and heads of families within agiven territory had the authority to allow accessto specific resources, such as shellfish beds, andto call a harvesting moratorium if they deemedthat the local stocks needed to recover (Turneret al. 2000). I was recently told by a BritishColumbia Stolo First Nations chief that hisgrandfather could identify five different stocksof salmon in their territory by examining theirbelly scales. His grandfather told him that his

254 E. Pinkerton

great grandfather would identify all the salmonstocks in their traditional territory (stocks thatwere adapted to spawn in different tributariesor creeks). This kind of detailed local knowl-edge allowed very specific management actionsto be taken in response to detailed observationsof different stock condition.

A major challenge for fish-managementagencies and scientists is how to support andconnect with community willingness to con-tribute to monitoring, often as volunteersbecause of their concern for and connectionto the resource. In Washington State, ProfessorJames Karr at the University of Washington wasable to mobilize hundreds of citizens who liveon reaches (sections) of salmon streams to mon-itor water quality and other conditions. Usingsimple straightforward indices that are easy tomonitor, Karr and his colleagues demonstratedthat citizen science can help a great deal in doc-umenting very specific trends in particular areas(Fore et al. 2001, CCDCD 2004; see Chapter 4).Similarly, fishermen in Nicaragua were willingto monitor local stock condition and report into a research team a great distance away. “Afterthe fieldwork was over they went on recordingthe water level during different months of theyear and mailed self-made charts to the scien-tist. No one had asked them to do this, but theyapparently had the feeling that the study shouldsomehow continue. . .” (Fischer 2000b).

An even more valuable contribution ismade when a community can undertake itsown research and when communities andresearchers work together to link moni-toring at different scales with research toco-produce knowledge on social–ecologicalchange. In several cases these research pro-grams have involved observations by local fish-ermen that were unnoticed by scientists. Forexample, two indigenous communities in theBroughton Archipelago between northern Van-couver Island and mainland British Columbia,Canada, noticed degradation in the conditionof their clam beaches and in the clams them-selves. Because of the timing of these changes,local fishermen suspected that the changesresulted from the presence of salmon farmsthat had proliferated in the area for a cou-ple of decades (Schreiber 2003, 2006). They

noticed an increase in sludge worms (Capitela)and macro algae matting on the beach thatapparently contributed to the beaches becom-ing smelly and unable to drain normally. (Thesephenomena are normally associated with eco-logical stress due to an overabundance of nutri-ents.) They also found blackened clams witha bad taste. The value of multigenerationallocal knowledge in a case like this providesa baseline of what is “normal.” These condi-tions had never been observed before (Heaslip2008). The communities were eventually ableto get the attention of a research scientist atthe Department of Fisheries and Oceans, andengaged the agency in a collaborative study byscientists, communities, and the salmon farm-ers to look at the sediments on particular clambeaches identified by the community as goodcandidates. The study is examining the differ-ent isotopic signatures in sediments on clambeaches to identify different sources of mate-rials. The joint study will test the hypothesis ofthe communities that some of the carbon, met-als, and nutrients so identified will be foundto originate from salmon farm feed (Weinstein2006, 2007). Since salmon farms on the BritishColumbia coast have been highly controver-sial, and the controversy has consumed vastfinancial and time resources, a study that estab-lishes or denies the suspected linkages will bean important contribution to science and publicpolicy. In this case, the co-production of knowl-edge involves systematically testing a hypoth-esis generated by the community that wouldhave been unlikely to gain serious attention byresearchers without community initiative.

A similar and ongoing study in Kitimat onthe northern coast of British Columbia mea-sures the effect of pulp mill effluent on theoolichan run (Thaleichthys pacificus, a majorsubsistence food of the local indigenous com-munity) via sensory evaluation methodologiesin which the panelists determine if the oolichanis tainted or not. The human sensory abili-ties of taste and smell are qualitative measuresthat have been adapted as a standard scientificapproach to detecting contamination in foods.It is widely recognized that the human palateis extraordinarily sensitive to various kindsof “off” tastes and odors, and that properly


established testing protocols can produce con-sistent, replicable results.

In the Kitimat situation, standard analyticalchemistry would not have been helpful becausethere could be up to 2,000 different chemicalsin pulp mill effluent, most of which have notbeen formally identified. Because the taintingpropensity could come from any one of theseor any number acting in combination, it was notpracticable to proceed with analytical chemicalidentification of compounds. Previous studieshave measured the tainting propensity of pulpand paper mill effluent (Shumway and Chawick1971, Brouzes et al. 1978, Gordon et al. 1980) asa means to determine uptake of contaminantsand the potential for biological effects. Further-more, the sensory methods go to the heart ofthe matter by determining whether the foodis “off” and does not concern itself with whatspecific chemical is causing it. Consultants forboth the indigenous community and the pulpmill agreed at the start that sensory testing wasappropriate. Sensory tests can employ expertevaluators, trained panelists, or naıve panelistsdepending on the purpose and objective of theevaluation. In this case, all sides agreed to usea panel made up solely of the local indigenouscommunity, on the grounds that only peoplefrom the local community are sufficiently famil-iar with the normal taste of oolichans to be ableto tell when something is off.

Assessment of tainting with pulp and papermill effluent is a very sensitive indicator of pol-lution effects. And because of their extraor-dinarily high fat content, oolichans are par-ticularly susceptible to becoming tainted fromexposure. Therefore, the ability of the millto eliminate tainting of oolichan would alsogreatly reduce the contaminant loading andpollution effects in the receiving environmentin general. Thus, the Kitamaat indigenous com-munity’s right and ability to protect a tradi-tional food source translates into a broader ben-efit for adjacent resources that do not enjoy thesame protection.

Coastal communities also contribute to con-servation by the very nature of their mode offishing. The small-scale local fishermen who livein these communities are the most adaptablepart of the commercial fleet. They have low

overhead costs and flexibility to shift to otherresources when conservation requires fishingclosures. Unlike the highly invested large indus-trial fleet that exerts political pressure to keepfisheries open because of their high overheadcosts, the small-boat fleet does not require ahuge volume of fish to support it and can adaptflexibly to radical shifts in abundance. Largehighly capitalized vessels are efficient at timesand in places when mass production is calledfor, and risks and fuel costs are relatively low(such as “mop up” fisheries during high abun-dance times or large concentrations of migra-tory fish that must be taken quickly). Smallvessels have a different kind of “efficiency”in that they are well-adapted to the risk anduncertainty prevalent in many modern fisheries(deYoung et al. 1999). These small fishing ves-sels are also “efficient” in that they are usedin coastal communities to procure food fish,firewood, wild game, and supplies and to pro-vide transportation. Loss of commercial fish-ing opportunity means the loss of a boat andthus the loss of the means to survive (Pinker-ton 1987). It is important in viewing the roleof coastal systems in supporting coastal popu-lations to recognize that the food-security rolethey play in the larger economy (providing forthe basic food and livelihood needs of coastalresidents, enabling them to care for their ownneeds and not be dependent on governmentor other forms of social assistance) is enabledby their small-scale commercial fishing activity.In turn, the recognition by coastal communi-ties of their dependence on the local system fortheir livelihood is a driver in their being candi-dates for conservationists. The argument here,as should be clear from the conditions outlinedin the previous sections, is not that coastal com-munities are automatic conservationists, butthat they are predisposed by many conditionsand circumstances to be so, and should be seenas an important form of human and social capi-tal that can aid in management.

The contribution of volunteers in coastalcommunities to the restoration of fish habitatand the enhancement of fish runs is underap-preciated and little understood. Unlike the off-shore marine fisheries described in Chapter 10,nearshore and riverine coastal fisheries are

256 E. Pinkerton

heavily dependent on healthy habitats, andhighly vulnerable to upland uses such as log-ging, farming, and industrial development thatcause soil erosion or release toxins that espe-cially affect nursery and spawning areas. Thereare some 10,000 volunteers in fish habitatrestoration, protection, and stock enhancementin British Columbia. In a study of the rea-sons why people volunteer for this work, Jus-tice (2007) found that the closer the volun-teers lived to the resource, the more likelythey were to perceive the urgency required torehabilitate the resource and also the moretime they had available to participate. The pri-mary reason people volunteered was to “putsomething back,” hoping to improve resourceoutcomes. Secondary reasons for volunteer-ing were to find camaraderie and apprecia-tion among fellow volunteers for doing some-thing worthwhile. Volunteer participation indecision-making forums, where it was perceivedto have an effect, encouraged further volun-teerism (Justice 2007). This research suggeststhat the perception of the importance of theresource, not only to livelihoods but also toidentity, is a factor in producing volunteerism.

This research reminds us that the capac-ity of a resource system to mobilize humanenergy is one of its most important character-istics. Peter Senge (1990), in researching suc-cessful learning organizations, identified threeconditions for effective organizational function.People in the organization must believe in thework; they must feel part of a team; they mustfeel that their individual contribution is valued.Under these conditions, people are likely to “gothe extra mile” to solve problems and facili-tate effective function, creating a working envi-ronment in which extra energy and creativityare produced. Bandura (1982) likewise foundthat when people experience self-efficacy (thesense of having an effect on events throughone’s efforts, having the power to make a dif-ference) in their work, they are able to mobilizemore energy. If we imagine coastal communi-ties as places where people, if they feel empow-ered, are likely to contribute to things that areimportant to them and that they care about,then we can see important possibilities for therole these communities can play in resource

management through their knowledge, skills,vision, and energy. In an ideal world in whichmany of the conditions identified in the last sec-tion were met, coastal communities would forma political constituency with a vision of abun-dant coastal resources that would dedicate itselfto working for this vision.

Conclusions

This chapter has shown how coastal communi-ties have coevolved with coastal ecosystems asparts of linked social–ecological systems. Thediversity of species and the sustainable pat-terns of livelihood that developed in coastalzones both in pre- and postindustrial times areproducts of a number of conditions that per-mit coastal communities to achieve conserva-tion of adjacent marine resources and to remainresilient to change. Social scientists have foundthat five general conditions are good predic-tors of conservation by coastal communities: (1)strong community access rights to local marineresources, (2) strong community rights to par-ticipate in management decisions, (3) resourcecharacteristics that lend themselves well toco-management arrangements, (4) communitycharacteristics that lend themselves well to co-management arrangements, and (5) commu-nity relationships with outside groups and gov-ernment strong enough to support its attemptto claim power in governance. Since there isnot a single formula for success, but rather amultiplicity of possible combinations of favor-able conditions, the listing of conditions canencourage those involved to look for windowsof opportunity. Where these conditions exist,there are strong possibilities for coastal commu-nities to contribute to the ecological, economic,and social objectives of fisheries management,and ultimately, to the resilience of the coastalsocial–ecological system.

Communities can uniquely contribute tolong-term monitoring and research of globaland local trends, model a flexible and adaptableway of fishing, and contribute volunteer energytoward the restoration of fish habitat and theenhancement of fish runs. These activities posi-tion communities to ban destructive gear and


technology that may negatively impact fisheries.Because of their relationship to local aquaticlife, coastal communities are in a constant pro-cess of learning and responding to their envi-ronment. As components of larger, linked, andmultilevel systems of governance, they can actas the tentacles that give a first warning aboutchange and can generate hypotheses about thelinkages between different components of thesystem as these respond to change.

Review Questions

1. What is an appropriate way to conceptualizethe relationship between coastal communi-ties and marine ecosystems? Why?

2. Under what conditions will coastal commu-nities play a positive role in fisheries andcoastal management? Under what condi-tions might we expect communities to playa negative role?

3. What is the role and importance of resourcedependence and place orientation on fish-eries conservation?

4. What can the involvement of coastal com-munities contribute to the ecological, eco-nomic, and social objectives of fisheries man-agement?

5. What role can coastal communities play inmonitoring, research, and policy making?What are some of the potential benefits andhazards of their involvement in these pro-cesses?

6. How might communities resist externalpressures that threaten to destroy localresources?

7. How can co-management contribute toadaptation of small-scale coastal communi-ties that are dependent on fisheries?

8. What lessons can be drawn from the vari-ous histories of community involvement inMPAs?

Additional Readings

Agrawal, A. 2002. Common resources and institu-tional stability. Pages 41–85 in NRC. The Dramaof the Commons. National Academy Press, Wash-ington.

Degnbol, P. 2003. Science and the user perspec-tive: The gap co-management must address.Pages 31–50 in D.C. Wilson, J.R. Nielsen, and P.Degnbol, editors. The Fisheries Co-ManagementExperience: Accomplishments, Challenges andProspects. Kluwer, Dordrecht.

Heaslip, R. 2008. Monitoring salmon aquaculturewaste: The contribution of First Nations’ rights,knowledge, and practices in British Columbia.Marine Policy 32:998–916.

Holm, P., B. Hersoug, and S.A. Ranes. 2000. Revis-iting Lofoten: Co-managing fish stocks or fishingspace? Human Organization 59:353–364.

Neis, B., and L. Felt. 2000. Finding our Sea Legs.Linking: Fishery People and Their Knowledgewith Science and Management. Institute for Socialand Economic Research, St. Johns, Canada.

Pinkerton, E.W. 2003. Toward specificity in com-plexity: Understanding co-management from asocial science perspective. Pages 61–77 in D.C.Wilson, J.R. Nielsen, and P. Degnbol, editors. TheFisheries Co-Management Experience: Accom-plishments, Challenges and Prospects. Kluwer,Dordrecht.

Pinkerton, E.W., and L. John. 2008. Creatinglocal management legitimacy. Marine Policy 32:680–691.

Pinkerton, E.W., and M. Weinstein. 1995. Fisheriesthat Work: Sustainability Through Community-Based Management. The David Suzuki Founda-tion, Vancouver.

Schlager, E., and E. Ostrom. 1993. Property rightsregimes and coastal fisheries: An empiricalanalysis. Pages 13–41 in T.L. Anderson and R.T.Simmons, editors. The Political Economy of Cus-toms and Culture: Informal Solutions to the Com-mons Problem. Rowman and Littlefield Publish-ers, Lantham, MD.

Wilson, J. 2002. Scientific uncertainty, complex sys-tems, and the design of common-pool institutions.Pages 327–360 in NRC. The Drama of the Com-mons. National Academy Press, Washington.

12Managing Food Production Systemsfor Resilience

Rosamond L. Naylor

Introduction

Managing food production systems on asustainable basis is one of the most critical chal-lenges for the future of humanity, for the obvi-ous reason that people cannot survive with-out food. Ecosystem health is both a “means”and an “end” to resilient crop and animal pro-duction. Being fundamentally dependent onthe world’s atmosphere, soils, freshwater, andgenetic resources, these systems generate someof the most essential ecosystem services on theplanet. They are also the largest global con-sumers of land and water, the greatest threats tobiodiversity through habitat change and inva-sive species, significant sources of air and waterpollution in many locations, and major deter-minants of biogeochemical change from localto global scales (Matson et al. 1997, Vitouseket al. 1997, Naylor 2000, Smil 2000). The inher-ent interplay between human welfare, foodproduction, and the state of the world’s nat-

R.L. Naylor (�)Woods Institute for the Environment andFreeman-Spogli Institute for International Studies,Stanford University, Stanford, CA 94305, USAe-mail: [email protected]

ural resources underscores the need to man-age these systems for resilience—to anticipatechange and shape it in ways that lead to thelong-run health of human populations, ecosys-tems, and environmental quality.

The production of crops and animals forhuman consumption epitomizes the social–ecological connection developed throughoutthis volume. This chapter differs from mostother chapters in that food production systemsare, by definition, human-created. The tightcoupling of social and ecological processes forfood production has been complete for at least10,000 years since the early domestication ofcrops and animals (Smith 1998). Few “wild” sys-tems exist today, in which humans hunt and for-age for food. A major exception is the fisheriessector, but even there, the fish species in great-est demand for human consumption have longbeen enhanced through hatchery production,and aquaculture (fish and shellfish cultured inconfined systems) is rapidly becoming the dom-inant source of global fish supplies (Naylor et al.2000, Naylor and Burke 2005, FAO 2006).

Food production systems are also unique interms of their economic, institutional, and cul-tural contexts. Markets link demand and sup-ply of food commodities throughout the world.This linkage is strong within commodity groups


260 R.L. Naylor

(e.g., the wheat market in China is coupled eco-nomically to the wheat markets in the USA,Argentina, and Australia), between commoditygroups (e.g., an increase in the price of maizecauses the quantity demanded for substitutefood crops, such as wheat, to rise), and betweensectors (e.g., the demand for maize as a live-stock feed or feedstock for bio-ethanol affectsthe price and demand for direct consumptionof maize as a food grain; Naylor et al. 2007c). Inalmost all cases, agricultural and food marketstend to be influenced heavily by policy involve-ment within countries. Culture also plays a rolein the design and management of food produc-tion systems. In the words of Anthelme Brillat-Savarin (circ. 1825), “Tell me what you eat, andI will tell you who you are.”

How food production systems are designed,managed, and redesigned throughout the worlddepends on a myriad of social and ecologicalfactors, such as soil type, climate, water avail-ability, pests and pathogens, genetic advances,economic incentives driven by market forcesand policy, and cultural influences includingtastes, traditional practices, and urbanization. Akey question to be addressed before any discus-sion of sustainable management can begin is:What are we trying to sustain? Food suppliesand adequate nutrition per capita over time?The environmental quality of farm systems andecosystems affected by food production prac-tices? The cultural integrity of farming commu-nities? Food quality, diversity, and safety? In adirectionally changing world, with continuallyrising demand for agricultural products drivenby population and income growth pressing on afinite land base, tradeoffs among these sustain-ability goals often occur. These tradeoffs arelikely to become more acute in the future as thedemand for biofuels adds to the already largeand growing global demand for food and ani-mal feed and as climate change limits agricul-tural productivity growth in certain locations.

The Agricultural Enterprisein Perspective

What makes food production systemsinteresting—and challenging—to study istheir wide diversity throughout the world.

Maize is produced with high-yielding, hybridseeds in mechanically dominated systems onthousands of hectares in the USA, Australia,and South Africa. Maize is also producedwith local seeds on small plots of sloping landby farmers in Central America, and by poorfarmers on marginal lands in East Africa usingfew external inputs apart from family labor.Rice is grown in irrigated, lowland fields inChina; it is also grown in deepwater, floodedsystems of Bangladesh and Thailand, on dry-land, hillside plots of Cambodia and Laos, andin integrated rice–fish ponds in Malawi. It isnot uncommon to see rice grown on quarter-hectare plots in rotation with vegetables andcassava in Indonesia and on thousand-hectareplots with no rotation apart from the winterfallow in the USA. Hogs are raised in con-finement in large industrial systems in Mexicoand in small numbers in backyard pens, andeven inside houses, in many Asian countries.Some agricultural regions are dominated bycash crops for export; others are devotedto staple crops for domestic consumption.Large private companies play a major role inthe development of agricultural technology,trade, and policy of most industrial nations,whereas agribusiness involvement directly infarming is minimal in the world’s poorest coun-tries. Some countries like the USA, France,and Japan subsidize agricultural productionheavily, while many developing countriestax it.

These examples simply illustrate the diver-sity in agricultural systems seen throughout theworld and should not be considered as the typ-ical agriculture of any particular region. Forexample, industrial livestock systems are alsofound throughout Asia, and high-productivitylowland rice is grown in Bangladesh, Thailand,Cambodia, and Laos. The USA grows staplecrops both for export and for domestic con-sumption of food, feed, and fuel. Laos exportscassava through Thailand to China for feed andfuel. There is a message to this madness: lessonson resilience and sustainability from one loca-tion do not necessarily apply to other areaswith similar biophysical systems. And lessonsare not easily transferred across scales of pro-duction. The global food production system iscomplex.

12 Managing Food Production Systems for Resilience 261

Intensification

Despite the heterogeneity in food productionsystems worldwide, there are important aggre-gate trends worth noting. The rising globalpopulation—now at 6.5 billion people andheaded toward 8 billion or more by the mid-dle of the twenty-first century (UN 2008)—coupled with steady urbanization and increas-ing demand for animal protein with incomegrowth, has created the need for more high-yielding production systems. Over the past40 years, roughly 80% of the growth in agricul-tural output has resulted from intensification;that is, the move toward high-yielding crops

with adequate water availability, soil quality,and nutrients (including synthetic fertilizers) toachieve significant increases in yields (Conway1997, Evans 1998, MEA 2005a). The pas-toral lifestyles and ecosystem use described inChapter 8 on drylands are examples of extensi-fication, and they stand in contrast to the highlymodified food production systems discussed inthis chapter. Shifting cultivation for subsistencestill represents a large share of global croppedland—particularly in the tropics—but manyof these systems are experiencing decliningproductivity with human population pressure(Box 12.1). The need to intensify all systems islikely to increase over time.

Box 12.1. Shifting Cultivation

Food production systems have experiencedan evolution from extensive to intensive cul-tivation in many parts of the world as a resultof continued population growth on a limitedland base. Yet extensive slash-and-burn pro-duction systems still exist on about 1 millionha of the earth’s land surface, and accountfor 22% of all agricultural land in the trop-ics (Giller and Palm 2004, Palm et al. 2005).In these systems, land is typically clearedby burning and cultivated for several cyclesuntil there is a significant loss in soil fer-tility (see Chapters 2 and 7). The land isthen fallowed in order to recuperate its nat-ural vegetation and nutrients. If the land isgiven sufficient time to recover, it can becleared and cultivated again after a numberof years in a sustainable fashion. However, ifthe land is farmed too long and is not givensufficient time to recover, excessive nutri-ent loss causes yields to decline and forcesfarmers onto more marginal lands. Underthese circumstances, the ecosystem may notrevert back to a mature forest, but insteaddevelop into grassland that is much less pro-ductive biologically (Conway 1997). For thisreason, shifting cultivation is often blamedas a primary cause of detrimental land-use change, most notably of tropical defor-

estation (Amelung and Deihl 1992, Myers1993, Rerkasem 1996, Ranjan and Upadhyay1999).

Shifting cultivation is designed to beresilient through time rather than space (Foxet al. 1995). In other words, when a farmerclears, cultivates, and fallows land for suffi-cient time, the land changes – in fact its eco-logical structure may change dramatically –but as time passes, it returns to alternativestable states of high resilience. Soon afterthe land is cleared, conditions are generallygood for planting; there are abundant nutri-ents from the ashes of burned vegetation, andstressors such as weeds and pests are low.The land can be used to farm productivelyfor a few years before stressors become pro-hibitive, at which point it must be abandonedin order to permit forest regrowth. Forestshave natural stabilizing components such asa diversity of flora and fauna (that preventssudden population swings) and long-livingtrees (that retain nutrients and maintain pro-ductivity). Given sufficient time to recuper-ate, forests generally represent a resilient sys-tem with productive soils, ample water, andresistance to pests, all of which are impor-tant for productive farming activities (Ewel1999).

262 R.L. Naylor

The key to resilience in these systems istiming. It is thus important for managers tounderstand the determinants of reduced fal-low periods or expansion into marginal land.Growth in the human population dependanton slash-and-burn agriculture is the lead-ing culprit of such change; more food isdemanded from the system, and there is sim-ply not time to wait for full forest recuper-ation unless crop yields can be improvedwith some form of productive intensifica-tion. Other factors are also important; forexample, land privatization can disrupt asustainable clear–cultivate–fallow cycle, andaccess to new markets can provide incen-

tives to farm for profit rather than for sub-sistence, perhaps resulting in less resilientmonocultures (Bawa and Dayanandan 1997,Amelung and Diehl 1992, Angelsen andKaimowitz 1999). The introduction of inten-sive and diverse agroforestry systems is oneavenue for resilience-based management inthe face of continued population growth(Palm et al. 2005). The establishment of insti-tutions and policies that support a transitionfrom degraded, extensive systems towardmore productive, intensive systems is essen-tial as population pressures erode the exist-ing land base (Fig. 12.1, taken from Palmet al. 2005).

Time and population density

Eco

syst

em s

tock

s an

d fu

nct

ion

s

Naturalsystems

Slash and burn

agriculture;shortened

fallows

Degradedsystems

Rehabilitationthrough

intensification

Intensivemanagement

c

a

b

Figure 12.1. Land use intensification pathways andchanges in stocks of natural capital such as carbonand nutrient stocks, biodiversity, and other ecosys-tem services, with time and increasing populationdensity in the tropics (Palm et al. 2005). Line a repre-sents the usual pattern of land degradation and even-tual rehabilitation when the proper policies and insti-tutions are in place, line b represents the continuedstate of degradation that can occur in the absence ofappropriate policies and institutions, and line c rep-resents the desired course where there is little degra-dation of the resource base yet improved livelihoodsare achieved. Redrawn from Palm et al. (2005).

Investments in irrigation, improved crop cul-tivars, and animal breeding have resulted inimpressive growth in global food production—around 260% between 1961 and 2003 (FAO-

STAT 2007). Cereal output grew by 2.5-foldduring this period, and poultry, pork, and rumi-nant production rose by 100, 60, and 40%,respectively. However, these changes haveoccurred at the expense of natural ecosystems.During the past half-century, intensive foodproduction systems have contributed substan-tially to anthropogenic emissions of greenhousegases (particularly methane and nitrous oxide)and pollution mainly in the form of nitrates,nitric oxide, and a wide range of pesticides(Tilman et al. 2002). The three largest staplecrops—maize, rice, and wheat, which provide45% of the calories consumed by the humanpopulation—are grown on almost half of thetotal arable cultivated land and account forover half of the synthetic nitrogen fertilizerapplied to agriculture worldwide (FAOSTAT2007, Cassman et al. 2003).

On a global basis, agricultural lands extendover 5 billion ha (34%) of the earth’s terres-trial surface (Fig. 12.2). Agricultural land areaexpansion averaged 7.3 million ha per annumon a net basis between 1990 and 2005, after tak-ing into account urban expansion, conversion ofrural lands to nonagricultural uses, and degra-dation (FAOSTAT 2007). The rate is fallingover time; between 2000 and 2005, net landarea expansion was only 2.6 million ha perannum. Although rates of area expansion havebeen declining over the past quarter century,growth often occurs in natural habitats with


Permanent crops

Permanent meadows andpastures

(Billions of ha)

Arable land

0.14

1.4

3.4

Figure 12.2. Global agricultural land use. Agricul-tural area refers to: arable land—land under tempo-rary crops, temporary meadows for mowing or pas-ture, land under market and kitchen gardens, andland temporarily fallow (less than five years). Theabandoned land resulting from shifting cultivation isnot included in this category. Data for arable landare not meant to indicate the amount of land thatis potentially cultivable; permanent crops—land cul-tivated with crops that occupy the land for longperiods and need not be replanted after each har-vest, such as fruit trees, cocoa, coffee, and rubber;this category excludes land under trees grown forwood or timber; and permanent pastures—land usedpermanently (5 years or more) for herbaceous for-age crops, either cultivated or growing wild (wildprairie or grazing land). Information from FAO-STAT Database 2006. FAO, Rome. 12 Nov 2006(FAOSTAT accessed Nov 27, 2007).

high biodiversity value, such as the Amazonianrainforest (Myers et al. 2000). Given the limitson net land area expansion, maintaining yieldgrowth in food and feed production will beessential in order to meet the expected globaldemand increase of 70–85% between 2000 and2050 as the human population and incomescontinue to rise (MEA 2005a). Even greatergrowth will be needed if significant cultivatedarea is used for biofuels and if climate changereduces productivity in many regions.

The availability of water through rainfalland irrigation is another important determi-nant of yield growth and agricultural produc-tivity throughout the world. Irrigated crop sys-tems account for about 40% of global foodproduction, but less than 20% of the world’scultivated land is irrigated (Gleick 2002). Irri-gated systems are abundant in Asia, whereas

over 90% of Sub-Saharan African agricultureis sown on rainfed lands. Most of the “easy”irrigation investments were made in the sec-ond half of the twentieth century; the annualrate of increase in irrigated area is currentlybelow 1% due to costs and to environmen-tal and social protests (FAO 2004, Khagram2004). Unfortunately, many irrigation systemswere built without proper drainage systems,and now an estimated 20% (45 million ha)of irrigated land suffers from salinization andwaterlogging (Ghassemi et al. 1995, Postel 1999,2001). Roughly 70% of the available surfacewater withdrawn for human activities globallyis used for agriculture (Postel 1999, Gleick 2002;see Chapter 9). The diversion of surface waterfor irrigation often comes at the expense of sur-rounding natural ecosystems.

Hunger

Despite growth in productivity of food sys-tems in many parts of the world during thepast 40 years, over 800 million people still livein chronic hunger (Chen and Ravallion 2007,MDG 2007; see Chapter 3). A set of Mil-lennium Development Goals was adopted bythe United Nations General Assembly in 2000,which included halving the world’s undernour-ished and impoverished by 2015. Today, morethan halfway through this target period, virtu-ally no progress has been made toward achiev-ing the dual goals of global hunger and povertyalleviation. Even worse, most of the gainsmade toward these goals since 2000 have beenerased by the world food crisis that emergedin 2008 (Economist 2008). The world food cri-sis has been characterized by extraordinarilyhigh agricultural commodity prices, low grainstocks, and restrictions on cereal exports by sev-eral major trading countries. The persistenceof global hunger among such a large num-ber of people is particularly disturbing in lightof the widespread economic and technologicalprogress experienced in many parts of the worldand the growing problem of obesity in industrialand middle-income countries, most notably theUSA.

264 R.L. Naylor

The Challenge Ahead

The challenge for the twenty-first century isthus clear: to develop food production sys-tems in a way that will support rural incomes,enhance yield growth, utilize inputs efficiently(particularly water and added nutrients), min-imize environmental impacts, and providehealthy diets for the human population (Con-way 1997, Power 1999, Tilman et al. 2002,Robertson and Swinton 2005). Given the out-look for future global food demand, there isno question that the productivity of food pro-duction systems must continue to increase inorder to achieve any definition of sustainabil-ity. But it is not at all clear that the goalshould be to maintain or augment productiv-ity on existing systems in all cases. Instead, thegoal should be to redesign systems to promotegenetic and crop diversity, stability in the faceof future shocks (e.g., climate change and asso-ciated pest–predator impacts), and food secu-rity as defined by access to affordable foodfor all people at all times (see Chapter 5). Inorder to achieve this goal, efforts should bedirected toward both small-scale farming sys-tems that primarily meet local and regionaldemands, and large-scale surplus systems thatmeet national and global demands (Robertsonand Swinton 2005). A framework of structuraldynamics for the agricultural sector is presentedbelow to help conceptualize the opportunitiesfor—and constraints on—advancing resilience-based management of food production systems.

Structural Dynamics of FoodProduction Systems

Resilience-based management of crop and ani-mal production requires a focus on the dynam-ics of both demand and supply. Humans playa dominant role in these dynamics, funda-mentally shaping the agricultural landscape inan effort to meet local, regional, and globaldemands for food, animal feed, and fuel. Ata broader scale, humans shape the agricul-tural landscape in order to generate incomes,employment, and in some cases ecosystem ser-

vices (e.g., pollination, pest control, watershedmanagement) that have positive impacts on theagricultural system itself and on surroundingareas. Human behavior plays a more intrinsicrole in the dynamics of food production sys-tems than is the case for most other ecosystems.Agriculture is by definition a human artifact—starting with the initial selection of geneticmaterial and encompassing genetic manipula-tion, breeding, the deployment of new cropvarieties and animal breeds, farm management,and food processing to meet consumer demand(Evans 1998, Smil 2000, Pretty 2002, Manning2004).

Determinants of Demand and Supply

Growth in demand is a function of popula-tion increases, per capita income growth, urban-ization, and cultural preferences. Per capitaincome growth and urbanization, in particu-lar, typically lead to diversification of diets.Two empirically based rules in the agriculturaldevelopment field—“Engel’s Law” and “Ben-nett’s Law”—have held up well over spaceand time (Timmer et al. 1983). Engel’s Lawstates that as incomes grow, the share ofhousehold income spent on food in the aggre-gate declines. Although food quality rises withincome growth, there are fundamental lim-its of food intake (the law of the stomach),which lead households to spend an increas-ing share of incremental income on nonfooditems such as education, housing, health, andmaterial goods and services. Engel’s Law holdsover time (the share of household income spenton food declines as countries develop) andover space (households in poor regions spenda greater share of their income on food thando households in wealthier regions). Bennett’sLaw states that the caloric intake of householdsis dominated by starchy staples at low levels ofincome, but is characterized by a diversified dietof fruits, vegetables, and animal products withincome growth.

Urbanization creates another shift indemand; households tend to eat food that iseasier to prepare at home or that is sold atrestaurants and food stalls. The demand for


meat products often rises in urban settings, asdoes the demand for products such as cookingoil used by street vendors. The process ofurbanization and suburbanization also supportsthe rise in national and multinational super-market chains and other large retail operations(e.g., WalMart and Cosco), which favors pro-ducers who are connected to these marketingchains and certain products over others (e.g.,farmed salmon over wild salmon because theformer can be supplied consistently in largevolumes throughout the year; Eagle et al. 2004,Reardon and Timmer 2007). Finally, incomegrowth leads to higher fuel energy demand,including demand for motorized fuels, whichcreates new linkages between the energy andthe agricultural sectors through biofuels whenfossil fuel prices are sufficiently high (Nayloret al. 2007c).

The ability of crop and animal productionto meet these various demands depends crit-ically on factor market conditions in individ-ual locations—that is, on the dynamics of labor,land, and credit markets—and on the state ofinfrastructure (e.g., roads, irrigation networks)and natural capital (water availability, climate,soils, genetic resources). Individual productionactivities are aggregated to create regional ornational supplies, which are either consumeddomestically or traded internationally. Tech-nology also plays an important role region-ally and globally in terms of genetic manipu-lation (e.g., improved crop cultivars and ani-mal breeds), labor-saving mechanization, andtools to maximize input use efficiency (e.g.,nitrogen sensors, machines designed to incor-porate residues in low-till systems, drip irri-gation technology). Producers throughout theworld have a complicated set of decisions tomake each season, particularly given inherentuncertainties in output price, weather, and pestsand pathogens at the beginning of each pro-duction cycle, and the long payoff times formany technological investments. Most farm-ers strive to maximize expected profits givena set of constraints that includes, for exam-ple, factor availability, agro-climatic conditions,technology, and infrastructure (Timmer et al.1983). Other farmers—particularly in very poorregions—strive to minimize variability in pro-

duction systems that are used mainly forsubsistence.

The overall impact of agricultural develop-ment on the environment is determined bythe IPAT Law (Ehrlich and Holdren 1974).This law states that the impact (I) of anyhuman activity such as agriculture is equal tothe product of the human population size (P,mouths to feed), the affluence of the pop-ulation (A, income and related food prefer-ences), and the technology being employed inproduction (T).

Short-Run Adjustments

There are several ways in which food produc-tion systems adjust in the short run to equatedemand and supply at different spatial scales.The most obvious adjustment mechanism is themarket (Fig. 12.3). If demand exceeds supply ina given period, prices rise and provide an incen-tive for producers to increase supply in subse-quent periods. Alternatively, if supply exceedsdemand, prices fall, causing consumption torise, production to drop, and stocks in storage tobe drawn down. Market equilibration does notoccur instantaneously due to the length of cropand animal production cycles, marketing chains,and policy disincentives to change. As a result,market adjustments can have serious impactson consumers when supplies are short, stocksare low, and prices are high. The consequencesof price hikes are particularly serious for theworld’s poorest consumers who typically spend50–75% of their incomes on food (Banerjee andDuflow 2007). Similarly, sharp price declinesdue to excess supplies can have devastatingeffects on producers, especially those who donot have diversified income sources, insurance,or savings.

Substitution plays a key role in producers’and consumers’ responses to price—a conceptthat is often neglected in discussions of carryingcapacity in food production systems. For exam-ple, when the price of rice rises in Asian mar-kets, consumers may switch to wheat or cassavaas the staple in their diets. When the price ofmaize is relatively high, farmers in the MidwestUSA may alter their crop rotation from soy to

266 R.L. Naylor

(1) Maize (2) Maize

(3) Wheat (4) Soy

D1

D0

D1

D0

S1

S0

S1

S0

Pric

e

Quantity

Figure 12.3. Market dynamics of agricultural sup-ply and demand. D = demand curve; S = supplycurve. Panel (1) – rising demand for maize leadsto growth in supply along the curve that includesproduction at higher marginal costs. Panel (2) –longer run shift in supply due to technical change

induced by higher prices. Panel (3) – higher maizeprices increase demand for wheat in livestock mar-kets, causing wheat prices to rise. Panel (4) – greaterarea sown to maize reduces area planted to soy, caus-ing soy prices to rise. Redrawn from Naylor et al.(2007c).

maize. Substitution in production often occurswith a lag, particularly when major assets (e.g.,large farm machinery) cannot be transferredacross crops. It is worth noting that, althoughsubstitutions occur throughout the world in allincome classes, extremely poor consumers andproducers tend to be more limited in theiroptions.

Four other important adjustment processesmay also come into play in response to fluctu-ations in price. The first is through the draw-down of stocks of grain, which are typically car-ried as a reserve from one crop year to the next.Another adjustment mechanism is through live-stock; when cereal and other feed prices rise,animals may be sold or butchered as a directresponse. A third, and much starker, short-runadjustment to high prices is human starvation.Hunger results mainly from the lack of incometo buy food, not from physical food shortages

alone. Finally, natural ecosystems may be theprimary adjusters during high commodity priceperiods, as reflected in agricultural expansioninto pristine rainforests or wetlands, excessivepumping of groundwater resources, and pollu-tion to surrounding ecosystems.

Longer-Run Adjustments

Crop and animal production systems also adjustin fairly predictable ways over the mediumto long run. There is a continuous process ofbottleneck breaking in these systems (Evans1998). For example, when the brown plant hop-per became a major pest in intensive rice sys-tems of Indonesia in the 1970s—as describedin more detail in the Indonesian case study—the government scaled back pesticide subsidies(because the pests had become resistant to the


pesticides), farmers adopted integrated pest-management practices (including crop rota-tions), and plant biologists continued to engi-neer cultivars that were more resistant to thepest. The process of bottleneck breaking canalso occur on the demand side. For instance,when maize surpluses in the international mar-ket caused prices to fall and farm incomesto drop in the USA in the latter half of thetwentieth century, food processors launched themaize-sweetener industry, and technology forthe maize bio-ethanol began to be developed—both of which added new layers to overall maizedemand. The federal government also imple-mented food aid programs to dispose of sur-plus grain, which have created disincentives inmany recipient countries to invest in agricul-ture (Falcon 1991). Although bottleneck break-ing increases the resilience of local food pro-duction systems, its net effect on regional orglobal resilience depends on market and policyresponses.

The decline in maize prices noted abovecame as no surprise and reflects a secondlong-run adjustment in food production sys-tems. Investments in infrastructure and tech-nology, such as irrigation, genetic improve-ments, or planting and harvesting equipment,lead to increased supplies over time. Increasein agricultural productivity typically results ineconomy-wide income growth in agrarian soci-eties. But as Engel’s Law states above, incomegrowth leads to a relative decline in expendi-tures on food over time. High growth in agri-cultural output with declining rates of growthin demand eventually results in excess sup-ply in a closed economy setting. There areseveral ways to break free from this insid-ious feedback, including migration, reducingresource use in agriculture, promoting interna-tional trade (exports), and creating new formsof demand. The ongoing expansion of crop-based biofuels presents an interesting example,because energy demand tends to rise in lock-step with income growth and thus helps “fuel”the demand for agricultural commodities, evenwhen relative expenditures on food are declin-ing (Naylor et al. 2007c). Political support forbiofuels in the USA and the EU is motivatedto a large extent by the goal of revitalizing rural

economies, not just by the goals of expandingand diversifying fuel supplies.

A third adjustment process that occursover the longer run is referred to as inducedinnovation—innovation that arises in responseto price increases for scarce factors of pro-duction (Ruttan and Hyami 1984). For exam-ple, in Asia where arable land is scarce rela-tive to labor, high-yielding seed varieties havebeen introduced as a land-saving technology.In the USA and Canada, where the oppositeholds (agricultural labor is scarce relative toarable land), labor-saving innovations, such asplanting and harvesting machinery and herbi-cides, have been introduced. The same princi-ple can be applied to natural resource inputs forfood production systems. The scarcity of waterin relation to other factors of production hasled to the design of water-saving innovationssuch as drip irrigation or desalinized water sys-tems. In agricultural areas with high soil ero-sion, soil-saving technologies such as conser-vation tillage or cover crops have come intoplay. Like land- and labor-saving technologies,resource-saving innovations tend to be adoptedwhen they are deemed economically profitableover a relevant time frame. Such calculationsoften involve present value accounting with dis-counting of future benefits—a process that canbe controversial depending on the time horizonand which discount rate is used (Kolstad 1999,Portney and Weyant 1999; see Chapter 9). Thebasic point is that investments in new technolo-gies or management practices require some sortof analysis—ranging from sophisticated calcula-tions to more rudimentary weighting schemes—of alternative uses of limited natural and finan-cial capital by current and future generations.

Role of Policy

Market forces shape the typical feedback mech-anisms described above for food consumptionand production. However, government policiesare also important in influencing the dynam-ics of demand and supply and often overridethe market system. The food production sec-tor is exceptional in terms of its heavy policyinvolvement (Naylor and Falcon In press). In

268 R.L. Naylor

most countries, governments have a hand inthe agricultural sector via direct input subsidies,price or income supports, output taxes, creditprograms, or land-reform measures. The agri-cultural sector may also be affected significantlyby government policies implemented for nona-gricultural purposes, such as controls on finan-cial capital or exchange rate adjustments. Tradepolicies, such as tariffs and quotas on agricul-tural commodities or bio-ethanol, can similarlyhave direct and indirect impacts on food pro-duction and consumption. The list goes on. Themain message is that the dynamics of crop andanimal systems throughout the world are influ-enced by policies designed for a wide range ofconstituents and political purposes—includingbut not confined to the food and agriculturalsector. The historical development of one pol-icy overlaid by another, for reasons rangingfrom rural revitalization to food or energy self-sufficiency, is an underlying cause of socialand environmental problems in food produc-tion systems in many countries.

Implications for Resilience-BasedManagement

Understanding the dynamics of food produc-tion systems raises some interesting dilemmasfor resilience-based management. In many situ-ations, market feedbacks help keep these sys-tems on track, either by eliminating crops orproduction practices that are inferior or nolonger valued by society, or by improving theefficiency of input use and resource allocation.Problems arise, however, when nonmarket con-sequences are involved—such as hunger, lossof cultural ties to the land, the destruction ofpristine environments, and damages to humanhealth and ecosystems from pollution—becausethe feedback mechanisms are less direct. Inaddition, policy often overrides market signalsand biophysical responses that might otherwiseenhance resilience. For example, governmentsubsidies for irrigation lead to inefficient wateruse practices, excessive groundwater pump-ing, and salt-water incursion in many loca-tions, obscuring obvious biophysical signals ofscarcity and thus delaying appropriate action

Ag

ricu

ltu

ral p

rod

uct

/ in

com

e

Shock or stress High

Medium

Low

Time

Figure 12.4. Resilience trajectory. Possible produc-tivity or income trajectories of an agro-ecosystemafter being subjected to a stress or shock. Redrawnfrom Conway (1997).

until large or irreversible changes may haveoccurred (see Chapter 9). Similarly, prolongedgovernment support for a particular crop, suchas irrigated wheat in the Sonoran desert oralfalfa in Southern California, leads to contin-ued investments in that crop despite its inap-propriate fit with natural and human resourcebase (Naylor and Falcon In press). Market andpolicy incentives might help a particular foodproduction system cope with external shockssuch as drought, pest infestations, war, and oilprice spikes, as shown by the “high resilience”arrow in Fig. 12.4. But the underlying ques-tion remains: Is the original trajectory of thatsystem designed for resilience, given the fullscope of biophysical, economic, and culturalfactors needed to ensure sustained production,consumption, and environmental quality overtime?

Case Studies on the ResilienceChallenge

Answering the question of whether food pro-duction systems are on a resilient trajectoryrequires a focus on the biophysical dynamicsof the systems as well as the social and eco-nomic determinants of change. Many crop andanimal systems are managed primarily for a sin-gle ecosystem service—the production of a con-sumable or marketable commodity (Robertson


and Swinton 2005). Yet several other ecosystemservices can also be provided by these systems,including soil stability, nutrient balance, pestand weed control, hydrological cycling, biodi-versity protection, climate regulation, clean airand water, cultural value, and human nutri-tion (Daily 1997, Conway 1997). Unfortunately,the goal of managing crop and animal produc-tion for a variety of ecosystem services thatsupport ecological resilience in the long run isoften overridden by social and economic prior-ities in the short run—for example, to supportpolitical constituents, respond to global marketopportunities, reduce poverty, or meet exter-nal demands for food, feed, and fuel. A decou-pling of food production systems from the eco-logical systems on which they fundamentallydepend diminishes their resilience (Robertsonand Swinton 2005), as shown in the case stud-ies that follow. With a continually rising globaldemand for food, feed, and fuel, it will becomeincreasingly important to identify and promotepolicy and institutional mechanisms to cou-ple or recouple agricultural and environmentalsystems.

Case I: Managing for Pro-poor Growthin Indonesia

Indonesia is the world’s fourth most populousnation (∼230 million people) and is extremelydiverse culturally, with more Muslims than anyother country but also with Catholics, Protes-tants, Hindus, Buddhists, and animists. TheIndonesian archipelago has about 9000 inhab-ited islands, 400 language groups, and croppingsystems that vary from Bali’s manicured irri-gated rice systems, to Nusa Tenggara Timor’s(NTT’s) rainfed corn and sorghum-based sys-tems, the Maluka’s rainfed cassava systems,Borneo’s oil palm plantations, and Papua’s sagopalm and sweet potato systems. Rice is themain staple food in most, but not all, parts ofthe country. Monoculture is not typical exceptfor flooded rice during the monsoon season onsome islands. As Wallace’s Line (a zoographicboundary between Asian and Australasian fau-nas) splits the country just east of Bali, thecountry contains wide geographic disparities

not only in soil, rainfall, flora, and fauna, butalso in agriculture, rural incomes, and incomegrowth.

Despite its inherent diversity, Indonesia issecond only to China in terms of its successin poverty reduction during the past 50 years,advancing from about 70% below the povertyline in the late 1960s to 15% in the twenty-first century. Although early land reform and avariety of later fiscal transfers (such as schoolvouchers and health benefits) have been testedfor poverty alleviation over the past half-century, agricultural development focused onpro-poor growth has demonstrated the onlysustained success (Timmer 2005). This develop-ment process was tied primarily to the GreenRevolution in rice production and illustratesa classic case of induced innovation. In mostAsian countries including Indonesia—and par-ticularly the island of Java, one of the world’smost densely populated areas—land is scarcerelative to labor. On Java, it is not uncommonto see a farmer working a quarter-hectare plotof land, or to see thirty unskilled workers showup to help harvest this farmer’s plot in exchangefor a share of their individual harvest output(Naylor 1994). Forty years ago, human pres-sure on this limited land area was so severe thatClifford Geertz (1963), a great anthropologistand agricultural development specialist, coinedthe term “agricultural involution” to depict afragile balance of existence between humansand the rice ecosystems on which they survived.

The scarcity of land in relation to labor—and the extreme poverty it created in manyparts of Asia—induced the Green Revolutionin rice and wheat in the late 1960s and early1970s. In simplest terms, the Green Revolu-tion consisted of the design and dissemina-tion of high-yielding seed varieties for themajor cereal crops; because the seeds are scale-neutral, farmers on small and large plots alikecould adopt the technology (Conway 1997).The new varieties contained dwarf genes, hadshorter maturities, and an absence of photope-riod sensitivity, which allowed a shift in croppatterns from one or two rice crops per yearto three crops (or five crops over 2 years) with-out rotation or fallow. Annual productivity ofrice thus increased substantially. However, a

270 R.L. Naylor

reliable water source and added nutrients werealso required to maximize the potential of thenew crop varieties, and, as a result, the great-est early successes were seen in irrigated areaswhere farmers also had access to credit, afford-able inputs (particularly synthetic fertilizers),and transportation infrastructure.

In Indonesia, the Green Revolution for ricewas promoted as part of a pro-poor growthstrategy. High-yielding seed varieties wereadopted and disseminated in the late 1960sand throughout the 1970s, along with nitrogenfertilizers and improved irrigation infrastruc-ture (mostly “run of the river” systems). Pol-icy incentives to enhance adoption of the newtechnology were created, such as the “farmer’sformula” in 1972 that linked very cheap fer-tilizers to the price of paddy to enhance prof-itability (Mears 1981, Timmer 1975). Improvedmicrocredit programs came 15 years later. Therice sector served as the engine for rural eco-nomic growth throughout the 1970s and 1980s,and led to steady increases in real wagesand incomes for unskilled labor (Naylor 1991,1994). During this period, poverty reductionalso resulted from expanded education (impor-tant especially for girls), better rural health andfamily-planning practices, and a macro policythat did not stifle agriculture or agriculturalexports (Timmer 2005).

The introduction of the Green Revolutionand the implementation of pro-poor growthstrategies in Indonesia caused rice prices to fallfor net consumers and lifted a large number ofpeople out of poverty. But the relevant questionfor this chapter is: Did it lead to a resilient agri-cultural system—one that can withstand shocksand provide multiple ecosystem services forlong-term benefits to the human populationand the environment? The answer is “not com-pletely.” There has been a marked decline inthe diversity of rice varieties used by farmersover the decades (Fox 1991). One of the mostsuccessful early varieties, IR36, was released inIndonesia in 1977. By 1982 it was planted on40% of total rice acreage, and in East Java it wasplanted on 77% of the wet season and 85% ofthe dry season acreage. Between 1975 and 1985,rice output grew by almost 50% in Indonesia.But unfortunately IR36 was susceptible to some

major rice pests such as the brown plant hopperand grassy stunt, and by 1985 the variety beganto fail. Substantial yield losses in IR36 led to therelease of an improved cultivar, IR64, whoseuse grew even more rapidly than IR36. Thisclassic case of bottleneck breaking led to pro-ductivity gains in the short- and medium-term.But the rapid dissemination of IR64 has causedthe country’s field- and landscape-level geneticbase for rice to become even more slender (Fox1991), potentially lowering the resilience of therice sector in the long run.

To add to the loss of genetic diversity seenin farmers’ fields, pest problems associated withIR36 led to the introduction of policies to subsi-dize pesticides (up to 80% of the retail cost) andencourage prophylactic spraying in the 1970sand early 1980s. This pesticide policy was aclear case of bad science, because it soon led toresistance within the brown planthopper pop-ulation and greater pest infestations over largeareas. Moreover, the policy was driven by cor-ruption, as a senior agricultural official was theprincipal owner of the pesticide plant. Oncethe subsidy policy was dismantled, integratedpest management (IPM) became a more popu-lar and successful practice—in conjunction withthe development of new host plant resistanceand limited spraying of chemicals—for stabiliz-ing yields. The IPM practices reestablished croprotations, although rice remains the dominantcrop in the monsoon season in many areas, par-ticularly on Java.

The agricultural sector now faces a new set ofchallenges, such as crop diversification, urban-ization and rising labor prices, new demands,and marketing arrangements with the rise ofsupermarket chains, and global climate change(Timmer 2005, Naylor et al. 2007a). It will beimportant for farmers and policymakers to pre-serve genetic diversity, minimize the use ofharmful chemicals (including particular herbi-cides as chemical weed control replaces handweeding with increased labor costs), and bal-ance incentives for productivity growth andrural poverty alleviation. The Indonesian ricesector has thus far been remarkably adaptableand resilient but will undoubtedly continue tobe tested in biophysical, economic, and culturalterms.


Case II: Managing for Globalizationin the Yaqui Valley, Mexico

The story of the Yaqui Valley in Sonora, Mex-ico, also pertains to the Green Revolution, butin a very different context. The Yaqui Valley,located in Northwest Mexico along the Gulfof California, was actually the home for theGreen Revolution in wheat in the late 1960s.Because the region is agro-climatically rep-resentative of 40% of the developing worldwheat-growing areas, it was selected as an idealplace for the early wheat-improvement pro-gram. It was here where Norman Borlaug, anagricultural scientist who was later awarded aNobel Peace Prize, introduced the first high-yielding dwarf varieties of wheat (Matson et al.2005). Using a combination of irrigation, highfertilizer rates, and modern cultivars, Yaquifarmers produce some of the highest wheatyields in the world (Matson et al. 1998). TheValley consists of 225,000 ha of irrigated wheat-based agriculture and is one the country’s mostproductive breadbaskets (Naylor et al. 2000).However, its agricultural productivity has beenthreatened repeatedly in recent decades bydrought, pests and pathogens, increased salin-ization, price shocks, and policy forces (Naylorand Falcon In press), making it an interestingcase study for resilience.

Although the Yaqui Valley has some sim-ilarities with Indonesia in terms of the high-yielding seed technologies used, the story dif-fers on at least four counts. First, the Valley islocated in the Sonoran desert and depends fun-damentally on irrigation from reservoirs, and toa lesser extent from groundwater pumping, forits intensive agricultural production. Second,while there are a large number of poor farm-ers in the region—mainly in the ejido (collec-tive agriculture) sector (Lewis 2002)—the Val-ley is increasingly characterized by larger, morewealthy farmers who operate over 50 ha apieceand in some cases hundreds of hectares. Third,Yaqui farmers use high rates of chemical inputs;nitrogen applications for wheat, in particular,are among the highest in the world and resultin major losses to the environment through var-ious biogeochemical pathways (Matson et al.1998). Finally, the main economic and policy

determinants of change for Yaqui Valley farm-ers originate at national and international scalesand are often exogenous to production prac-tices in the region (Naylor and Falcon In press).

With a semiarid climate and variable precip-itation rates, the Valley has relied on the devel-opment and maintenance of irrigation reser-voirs for agricultural intensification. By 1963three major dams had been constructed sup-plying irrigation water to 233,000 ha (Nayloret al. 2000). However, the construction of thesereservoirs did not eliminate the region’s sen-sitivity to climatic extremes. For example theprolonged drought during 1994–2002 led todramatic declines in total reservoir volume,increases in well pumping, and reduced waterallocations to farmers, resulting in less than20% of the total area in production in 2003(Matson et al. 2005). By this time, Valley farm-ers had completely drained the 64-km2 reser-voir from the Rio Yaqui. Increased dependenceon groundwater from wells—water that tendsto be more saline than the high-quality freshwater from reservoirs—in turn raised the risk ofsalinization, a problem that affects roughly onethird of the soils in the Valley. Rains returnedto the region by 2005 and provided water tothe reservoir once again, but policy incentivessupporting the production of crops that are notdrought or salt-tolerant continue to weaken theresilience of this desert agro-ecosystem system.

Policy has played an enormous role in agri-cultural development in the Valley, but it hasoften worked at odds with environmental qual-ity and ecosystem health—and on occasioneven with farm profitability (Naylor and FalconIn press). Many of the policies affecting agricul-ture in the region have been macroeconomic,focused on trade, exchange rates, interest rates,and national financial portfolio balances. Theagricultural policies at the microlevel have beenimplemented in Mexico City for the country asa whole and not necessarily for the benefit ofYaqui farmers.

During the 1980s, government involvementin almost all phases of the Mexican food sys-tem was pronounced. Significant price sup-ports for agricultural products, large input sub-sidies on water, credit, and fertilizer, and majorconsumption subsidies on basic food products

272 R.L. Naylor

were justified primarily as poverty-alleviationpolicies. In the early 1990s the governmentunder President Salinas began to withdraw gov-ernment support in agriculture as part of abroader liberalization process that was occur-ring in other sectors of the economy. Newinternational trading arrangements were imple-mented for agriculture, mainly via NAFTA,which reduced trade barriers, motivated largechanges in prices of many agricultural inputsand outputs, and thus dramatically altered rel-ative prices to producers. Producer incentiveswere also altered by replacing price supportsfor wheat and other agricultural products withincome supports—thereby decoupling govern-ment payments from total production, whichhurt farmers in this high-yield region of Mexico.During this period, the government reduced itsinstitutional involvement in agriculture, e.g., byreducing consumer food subsidies, privatizingthe nationalized Mexican Fertilizer Company(FERTIMEX), removing or reducing govern-ment credit subsidies, and largely eliminatingpublic extension services. Finally, the operatingauthority and funding responsibilities for irriga-tion systems were decentralized from federal tolocal water-user groups via the Water Laws of1992 and 1994, and a constitutional change inland rights (Article 27) in 1992 made possiblethe (legal) sale and rental of ejido land (Nayloret al. 2000, Naylor and Falcon In press).

The overall intentions of these numerouspolicy changes were to integrate Mexican agri-culture into the global economy, improve effi-ciency of production, and increase privatesector involvement. But Yaqui farmers wereexposed to markets in unprecedented ways.They were also hit by a series of externalshocks—pest attacks, drought, large fluctua-tions in world commodity markets, and a majordevaluation of the exchange rate—that funda-mentally altered the economic and biophysicalenvironment in which they operated. In prin-ciple, the full suite of policies could have ledto greater input use efficiency (e.g., water andnitrogen), resulting in “win–win” solutions forfarm profitability and the environment. How-ever, despite higher marginal costs, farmersdid not reduce fertilizer and water applicationsuntil forced to do so with the drought (Manning

2002, Addams et al. In press). Moreover, Mex-ico’s main agricultural trading partner—theUSA—continues to subsidize its wheat farmersin many indirect ways, leaving Yaqui farmersat a competitive disadvantage unless they arealso subsidized. Some farmers have attemptedto diversify production into high-valued crops,livestock, or aquaculture, but the economic andbiophysical risks of doing so remain high.

Economic and policy changes affecting agri-cultural decision-making in the Yaqui Valley inthe 1990s were largely exogenous to the farm-ing system. However, farmers in this highlycommercialized region of Mexico are not just“policy-takers.” They also do have a voice inthe development of national agricultural poli-cies. Their declining competitive position rel-ative to the USA, coupled with the fall ininternational commodity prices in the latterpart of the decade, induced an intensive lob-bying effort that ended with the reenactmentof commodity price supports in 2000 for thebulk crops such as wheat, maize, and cotton.Yaqui farm groups were not solely responsi-ble for the renewed protection, but they had—and continue to have—a persuasive influenceon other farm groups in Mexico and a historyof political power in Mexico City. Three Mexi-can presidents came from the Valley’s main city,Ciudad Obregon, and the connection betweenlarge farm organizations in the Yaqui Valley,and politicians in Hermasillo (the state capital)and Mexico City have traditionally been strong.When lobbying has not worked, farm groupshave resorted to other tactics, such as threaten-ing to close down the highway that runs throughthe state of Sonora to the US border.

The policy shift back toward protection atthe turn of the twenty-first century raises impor-tant questions for the future sustainability andresilience of Yaqui Valley agriculture. Willfarmers continue to grow wheat, corn, and cot-ton as a result of the economic security stem-ming from policy support—despite the fact thatthese crops demand high fertilizer, pesticide,and water inputs? If so, will improvements ininput use efficiency be sufficient to reduce thecost-price squeeze that farmers have experi-enced in the past, and to lessen the impacton the environment? Will farmers continue to


band together in the promotion of these cropsin order to preserve their “safety in numbers”?Answers to these questions will depend onagricultural policy dynamics within Mexico—driven in large part by the tension betweennorthern commercial interests versus south-ern social interests—and between Mexico, theUSA, and Europe. Managing for resilience thusrequires the participation of policymakers out-side of the agricultural system in question. Aslong as the USA and the EU persist in subsi-dizing wheat, maize, and cotton through variouspolicy instruments, Mexico will likely be forcedpolitically to subsidize these crops as well. Thefuture course of commodity protection in Mex-ico will depend importantly on progress relatedto agricultural trade negotiations within theWorld Trade Organization, the 2008 US FarmBill, EU farm policy, and trends in internationalcommodity prices. This progress will be influ-enced, in turn, by a set of emerging issues withinthe world food economy, including the indus-trial livestock revolution, the biofuels boom,and global climate change.

Emerging Issues forResilience-Based Management

A key feature of resilience-based managementis to anticipate change and adjust accordinglyin order to preserve long-run ecological func-tioning and human welfare (see Chapters 2and 3). Beyond the specific, anticipated sorts ofchanges in a system, resilience-based manage-ment needs to consider the possibility of com-plete surprises and uncertainties, such as theconfluence of events that led to the world foodcrisis in 2008 (Walker and Salt 2006). Thereare at least three major transitions confrontingglobal food production systems today that areworth considering in this context: growth inindustrial livestock systems, the rising use ofcrops for fuel, and global climate change. Thefirst two cases can be thought of as demand-driven changes, while the third is primarilysupply-driven. In all cases, policies directedtoward both the agriculture and the energy sec-

tors will play a role in determining the resilienceof food production systems.

Industrial Livestock

Domesticated livestock have played a role inhuman societies and evolution for the past10,000 years, providing important sources ofprotein, fertilizer, fuel, traction, and trans-port (Diamond 1997, Smith 1998). Tradition-ally, livestock have been an integral part ofagricultural systems, raised close to their foodsource, and used as an input (soil nutrients andtraction) in crop production. In recent decades,however, livestock have become industrialized,often raised far from its feed source and tradedinternationally (Naylor et al. 2005, Gallowayet al. 2007). At the heart of this transitionis very rapid income-driven growth in meatdemand, particularly in parts of the developingworld such as China, Southeast Asia, and LatinAmerica (Steinfeld et al. 2006). In addition,relatively inexpensive feeds, improved trans-portation, technological innovations in breed-ing and processing, concerns over food safety,and vertical integration of the industry have ledto industrialization and spatial concentration ofintensive livestock systems. Urbanization anddevelopment of large-scale retail chains furthercontribute to intensification of these systems(Steinfeld et al. 2006).

A dominant feature of the geographic con-centration of livestock—and one that has majorimplications for the resilience of the sector—is the de-linking of animal production fromthe supporting natural resource base. Feed issourced on a least-cost basis from internationalmarkets, and the composition of feed is chang-ing from agricultural by-products to grain, oil-meal, and fishmeal products that have highernutritional and commercial value (Naylor et al.2005, Galloway et al. 2007). Synthetic fertiliz-ers as opposed to animal manure are used tofertilize crops, and machines are used insteadof animal traction to plow the land—both con-tributing to higher fossil fuel inputs and greatergreenhouse gas emissions. The pattern of indus-trialization is particularly striking for monogas-tric animals (poultry and hogs), which utilize

274 R.L. Naylor

concentrated feeds more efficiently than rumi-nants (cattle, sheep, and goats) and whichhave short life cycles that favor rapid geneticimprovements (Smil 2002). Pork and poul-try products also tend to be less expensivefor developing-country consumers; as a result,industrial livestock production is expected tomeet most of the income-driven doubling inmeat demand forecast for developing countriesin the coming decades (FAO 2004).

For meat and other livestock products theincome elasticity of demand is high—that is,when incomes grow, expenditures on livestockproducts grow rapidly (Steinfeld et al. 2006).A classic relationship exists between incomesand direct (food) versus indirect (feed) demandfor grains (Fig. 12.5). As incomes rise, moregrain and oilseed crops are needed to feedthe human population via the livestock sec-tor, which in turn has detrimental effects onglobal land and water resources. Through graz-ing and feed crop production, livestock is thelargest global user of land resources, occupyingalmost one third of the ice-free terrestrial sur-face of the earth (Steinfeld et al. 2006, Gallowayet al. 2007; see Chapter 8). With industrializa-tion, land use change associated with livestockis being driven increasingly by feed crop pro-duction as opposed to grazing, although grazingremains a major form of land use (see Chapter 8and Fig. 12.2). A similar pattern holds for water

Direct + indirect

Direct

Annual income (per capita)

Gra

in c

on

sum

pti

on

(kg

per

cap

ita

yr–1

)

800

180

750 10,000

Figure 12.5. Direct and indirect (to support live-stock production) grain consumption as a function ofincome.

resources: agriculture dominates global wateruse, and a growing share of crop production isnow devoted to animal feeds (see Chapter 9).

The de-linking of livestock production fromthe land base, and the increasing intensity ofirrigation and synthetic fertilizer applicationsfor feed crop production, has internal and exter-nal impacts that are often obscured by the lackof appropriate market valuation for agriculturalwater use and pollution. Moreover, with inter-national trade in livestock products now grow-ing faster than production, the link betweenconsumers and producers has weakened—effectively eliminating the accountability thatlivestock consumers might feel in relation to theproducts they eat (Galloway et al. 2007). Poli-cies also play a role in promoting feed grain andindustrial livestock production in many coun-tries, even when factor scarcity or resource con-straints might otherwise lead to a decline inthe livestock sector. Finally, the emerging dom-inance of industrial livestock has curbed themarket access for livestock producers in smallscale or extensive pastoral systems, most ofwhom live in poor rural communities (Steinfeldet al. 2006). The intensification of livestock hasincreased global protein consumption in bothdeveloping and developed countries. However,it is essential that these systems be managedto strengthen economic and ecological feed-backs through improved nutrient cycling, effi-cient water use, and food safety measures. Iflivestock production remains decoupled fromits supporting resource base, the resilience offood production systems at local to global scalesremains in question.

The Biofuels Boom

The integration of the global agricultural andenergy sectors caused by recent and rapidgrowth in the biofuels market raises even moreserious questions than industrial livestock interms of the resilience of food production sys-tems. Investments in crop-based biofuels pro-duction have risen recently around the worldas countries seek substitutes for high-pricedpetroleum products, greenhouse gas-emittingfossil fuels, and energy supplies originating


from politically unstable countries. Some coun-tries such as the USA are also supportingcrop-based biofuels production as a means ofrural revitalization. Growth in the biofuels sec-tor raises two important questions concern-ing the resilience of food production systems.First, can agro-ecological systems be sustainedenvironmentally given the degree of intensifi-cation needed to meet biofuels production tar-gets in various countries over time? And sec-ond, as greater demand pressure is placed onagricultural systems—and as agricultural com-modity prices rise in response—can food secu-rity be maintained for the world’s poorest pop-ulations? In answering these questions, a fewtrends seem clear. Total fuel energy use willcontinue to escalate as incomes rise in bothindustrial and developing countries, and biofu-els will remain a critical energy developmenttarget in many parts of the world if petroleumprices remain high. Even if petroleum pricesdip, policy support for biofuels as a means ofboosting rural incomes in several key countrieswill likely generate continued expansion of bio-fuels production capacity over the next decade(Naylor et al. 2007a).

The rising use of food and feed crops for fuelis altering the fundamental economic dynam-ics that have shaped global agricultural marketsfor the past century. Although both energy andfood demand rise with income growth, the rateof increase is much greater for energy, as shownin Fig. 12.6. Engel’s Law, coupled with impres-sive increases in world food production, has led

to a steady trend decline in real food pricesin international markets for the past severaldecades. But this pattern is changing with thenew linkages between the agriculture and theenergy sectors. As energy markets increasinglydetermine the value of agricultural commodi-ties (Cassman et al. 2006, Schmidhuber 2007),the long-term trend of declining real prices formost agricultural commodities is likely to bereversed and Engel’s Law overridden (Nayloret al. 2007c).

Over the short term this reversal, whilepotentially helping net food producers in poorareas, could have large negative consequencesfor the world’s food insecure, especially thosewho consume staple foods that are director indirect substitutes for biofuel feedstocks.Sugarcane, maize, cassava, palm oil, soy, andsorghum—currently the world’s leading biofuelfeedstocks—comprise about 30% of mean calo-rie consumption by people living in chronichunger around the world (Naylor et al. 2007c).The use of these crops for global fuel consump-tion could thus increase the risk of global foodinsecurity, particularly if rural income growth isnot rising in parallel. Rising commodity pricesfor feedstock crops and their substitutes (e.g.,maize, soy, wheat, and cassava) also have adirect impact on the livestock industry, sincethese crops are an important feed ingredient,particularly for pork and poultry.

The risks of food insecurity associatedwith biofuels development were realized in2008 with the sharp run-up of agricultural

10,000

10,000

1000

1000

100

100

Income ($ per capita yr–1)

Cereal

Energy

Cer

eal a

nd

en

erg

y (o

il eq

uiv

) co

nsu

mp

tio

n (k

g p

er c

apit

a ye

ar –1

)

Figure 12.6. Per capitacereal consumption(circles) and energyconsumption (triangles)by country as a function ofincome in 2003. Redrawnfrom Naylor et al. (2007c).

276 R.L. Naylor

commodity prices worldwide. Increaseddemand for biofuel feedstocks (on top ofthe rising demand for meat in China andother emerging economies)—coupled withsome regional drought- and disease-relatedsupply shocks—led to heightened speculationin financial markets and panic among manygovernments facing food riots and politicalunrest. Although biofuels were not the solecause of the world food crisis as it unfoldedin 2008, the tight linkage between agriculturaland energy markets raises serious questionsfor the future resilience of food productionsystems and the ability of poor people to affordadequate nutrition.

On the supply side, the growth in biofu-els globally and regionally creates risks ofenvironmental decay and resource exhaustion(Naylor et al. 2007c). For example, the USA—now one of the two largest global bio-ethanolproducers along with Brazil—is fundamentallyconstrained in how much maize can be pro-duced for fuel by both land area and yieldpotential. Although maize area has expandedat the cost of other crops such as soy inrecent years, price feedbacks limit the amountof area substitution that will occur over time(Fig. 12.3). Some agricultural lands that werepreviously removed from production for pro-grams like the Conservation Reserve Program(CRP) are now being brought back into maizecultivation (Imhoff 2007). Introducing mono-cultures into CRP lands will likely have adverseeffects on biodiversity and wildlife habitat—two main CRP goals. The yield potential forUS maize is also limited in terms of geneticgains (Cassman and Liska 2007); most yieldgrowth is likely to come from additional inputs(fertilizers, water), creating additional environ-mental stress to supporting and surroundingecosystems unless input use efficiency improvesdramatically (Cassman et al. 2003, 2006).

There are several other examples where cropproduction for biofuels potentially lowers theecological resilience of food production systemsworldwide (Naylor et al. 2007c). For example,oil palm produced in the rainforests of Borneofor biodiesel and food can cause habitat destruc-tion, a loss of biodiversity, degraded water andair quality, the displacement of local commu-

nities, and a change in regional climate (Cur-ran et al. 2004). Land clearing through firecontributes to especially large environmentalthreatsof regionalairpollution,biodiversity loss,and net carbon emissions (Tacconi 2003). Cropproduction in low-productivity, hillside areas ofChina for bio-ethanol can cause soil erosionand flooding. The substitution of maize for soyin the USA can lead to increased soy produc-tion in the Amazonian rainforest, thus poten-tially causing biodiversity loss and a change inregional climate. The wide range of ecological,environmental, and food security effects of bio-fuels production are just starting to be mea-sured and documented. Strategies for managingcrop-based biofuels development for resilienceneed to focus on the implications for land use(soil erosion, biodiversity), water use (quantityand quality), air quality, and net energy andclimate outcomes at local and global scales.

A key to enhancing the resilience of foodproduction systems in an era of biofuels willbe the emergence of economically and techno-logically feasible sources of cellulosic fuels thatcan be grown on degraded lands (Tilman et al.2006). Current cellulosic biomass-to-fuel con-version systems are not yet cost-effective andrequire large amounts of water (Naylor et al.2007c, Wright and Brown 2007). The technol-ogy for large-scale deployment of cellulosic bio-fuels production is probably at least 10 yearsaway (Himmel et al. 2007), although smaller-scale biomass systems using more rudimentarytechnology have long been viable for local fuelproduction. The coupling of the agriculturaland energy sectors at regional to global scalesthrough the development of crop-based biofu-els is thus likely to play an important role inthe resilience of food production systems fordecades to come.

Global Climate Change

A third—and arguably the dominant—emerging influence on resilience in foodproduction systems worldwide is global climatechange. The fourth report of the Intergov-ernmental Panel on Climate Change (IPCC)released in 2007 presented strong scientific


data and consensus once again that the globalclimate is changing, and that humans are bothcausing and will be damaged by this change(IPCC 2007a,b). The agricultural sector is likelyto be affected more directly than any othersector by rising temperatures throughout theworld, sea level rise, changing precipitationpatterns, declining water availability, new pestand pathogen pressures, and declining soilmoisture in many regions. Vulnerability iscreated through biodiversity loss and simpli-fication of landscapes in the face of climatechange (see Chapter 2). The agricultural sectoris also contributing to climate change throughnitrous oxide emissions from fertilizer use(denitrification) and methane release from ricefields and ruminant livestock (e.g., cattle andsheep; Steinfeld and Wassenaar 2007).

Although predicting future climate con-ditions involves many uncertainties (seeChapter 14), there is broad scientific consensuson three points (IPCC 2007a). First, all regionswill become warmer. The marginal change intemperature will be greater at higher latitudes,although tropical ecosystems are likely to beparticularly sensitive to projected temperaturechanges due to the evolution of species undermore limited seasonal temperature variation(Deutsch et al. 2007). Second, soil moisture isexpected to decline with higher temperaturesand evapotranspiration in many areas of thesubtropics, leading to sustained drought condi-tions in some areas and flooding in other areaswhere rainfall intensity increases. In general,wet regions are expected to become wetter,and dry regions are expected to become drier,although changes in precipitation are much lesscertain than those of temperature. Third, sealevel will rise globally with thermal expansionof the oceans and glacial melt.

How the anticipated changes in global cli-mate will affect agricultural productivity aroundthe world and the resilience of food produc-tion systems depends on regional patterns ofchange and the ability of countries within eachregion to adapt over time. Again, there is muchuncertainty in the distribution of future impacts,but some regional predictions can be madewith reasonable confidence—ignoring, for themoment, adaptation. For example, sea level rise

will be most devastating for small island statesand for countries such as Bangladesh that arelow-lying and highly populated. Large areas ofBangladesh already flood on an annual basis andare likely to be submerged completely in thefuture, leading to a substantial loss of agricul-tural land area, even for deep water rice. More-over, therapidmeltingof theHimalayanglaciers,which regulate the perennial flow in large riverssuch as the Indus, Ganges, Brahmaputra, andMekong, is expected to cause these river sys-tems to experience shorter and more intenseseasonal flow and more flooding—thus affect-ing large tracks of agricultural land. Moderatetemperature changes are likely to be more posi-tive for agricultural yields in high latitudes thanin mid- to low-latitudes (IPCC 2007b). In addi-tion, CO2 fertilization will benefit some cropsin the mid-latitudes (provided that temperaturechanges are not extreme and sufficient water isavailable for crop growth) until mid-century—atwhich time the deleterious effects of tempera-ture and precipitation changes are expected tooffset CO2 fertilization benefits. Food insecurepopulations, particularly in southern Africa andSouth Asia, will likely face greater risks of lowcrop productivity and hunger due to increasedtemperature; yield losses could be as high as 30–50% for some crops in these regions if adaptationmeasures are not pursued (Lobell et al. 2008).Africa as a whole is particularly vulnerable toclimate change since over half of the economicactivity in most of the continent’s poorest coun-tries is derived from agriculture, and over 90% ofthe farming is on rainfed lands (Parry et al. 2004,Easterling et al. 2007, World Bank 2007).

Given these large potential impacts, adap-tive capacity is critical to the resilience of anyparticular agricultural system (see Chapter 3).Adaptation measures range in size and expense,with activities such as the shifting of plant-ing dates or substitution among existing cropcultivars at the low end, to the installation ofnew irrigation infrastructure or sea walls atthe high end (Lobell et al. 2008). A commonassumption is that agricultural systems will shiftgeographically over time to regions with suit-able agro-climatic conditions (e.g., crops willmove poleward with warmer temperatures)—resulting in little net impact on global food

278 R.L. Naylor

supplies in the future. However, the necessaryextensive genetic manipulation through breed-ing will require continued collection, evalua-tion, deployment, and conservation of diversecrop genetic material (Naylor et al. 2007b).Because climate change will also affect wild rel-atives of plants and hence the in situ geneticstock on which the agricultural sector funda-mentally depends (Jarvis et al. 2008), there is anurgent need to invest in crop genetic conserva-tion for resilience, particularly in cases where exsitu genetic resources (genetic material storedin gene banks) are scarce (Fowler and Hodgkin2004).

Overall, there are three important points toconsider with respect to the resilience of foodproduction systems under a changing climate.The first is that poor farmers tend to havefewer resources at their disposal than wealthyfarmers—eitherat thehousehold,community,orstate levels—for adaptation. They typically donot have access to irrigation, a wide selectionof seed varieties, knowledge of alternative crop-ping systems, credit, or personal savings (Dercon2004,Burkeetal.2008).Second,becauseadapta-tion strategies that benefit the greatest numbersof producers involve large-scale investments(e.g., irrigation, breeding for new cultivars),there is a “public good” aspect to adaptationthat cannot be ignored. Governments and theinternational donor community have a role toplay in ensuring that adaptation can occur inboth rich and poor countries to protect theresilience of food production systems. Finally,providing information to policymakers, farmers,and agribusiness throughout the world on cli-mate change and its potential impacts is needednow—particularly in poor countries where suchinformation is often lacking—in order to moti-vate private sector solutions to the problem inboth the short and long run. Substantial invest-ments in education, information, infrastructure,and new crop varieties suited to the projected cli-matewillberequiredtosustainallaspectsof foodproduction systems: food security, rural com-munities, agro-ecosystems, crop genetic diver-sity, and agricultural yield growth. Without aforward-looking vision, the resilience of agri-cultural and livestock systems will undoubtedlyerode.

Conclusion

The twenty-first century marks a new era forglobal food production systems along severalaxes. There has never been a time when thehuman population has approached 8 billionpeople, with growing demands for food and ani-mal feeds. Nor has there been a time whenthe agricultural sector promises to be so tightlylinked with the energy sector in terms ofprice level and volatility. Widespread urbaniza-tion is consuming fertile agricultural land andwater and at the same time is creating greaterdemands for food from the arable land thatremains in production (see Chapter 13). Glob-alization is leading to the integration of agricul-tural commodities, inputs, and financial marketsover space and time that increasingly decoupleconsumption from production and leave farm-ers vulnerable to swings in factor and outputprices and to volatility in the cost of financialcapital (see Chapter 14). And perhaps mostdaunting, climate change over the course ofthe twenty-first century will alter modern foodproduction systems in unprecedented ways andthreaten food security in many poor regionsof the world. Crop yields are predicted todecline dramatically in many areas as a resultof climate change—just at a time when ris-ing demands from population growth, incomegrowth, biofuels expansion, and urbanizationrequire significantly higher yield growth than inthe past.

Designing adaptive strategies to ensureresilience in food production systems in thisnew biophysical and socioeconomic era willrequire focused attention by agricultural sci-entists, policymakers, international develop-ment agencies and donors, industry leaders,entrepreneurs, resource managers, crop seedcollectors and conservation experts, NGOs, andfood producers and consumers throughout theworld. How should agricultural systems bedesigned and managed to meet the challengesahead? What incentives should be provided tofarmers to guarantee a wide range of ecosystemservices from agriculture beyond “the globalpile of grain”? How can food security, foodquality, and environmental quality be ensuredfor the generations to come?


Answering these questions is not an easytask, particularly given the diversity of groupsinvolved in the global agricultural enterpriseand their differing priorities with respect toproduction, income, culture, and environmentalgoals now and in the long-run future. Despitesuch complexity, the transition toward resilientfood production systems will require three mainsteps (Lambin 2007). The first step involvesa widespread recognition that the biophysicaland socioeconomic conditions of the twenty-first century are markedly different than thepast and that many crop and animal produc-tion systems will need to be redesigned funda-mentally (see Chapter 5). Some principles couldhelp shape this process of redesign. For exam-ple, given the growing resource constraints onagriculture, it is time to focus more on adaptingcrops and animals to the local resource base andnutritional needs, rather than transforming theresource base to meet specific commodity pro-duction targets (see Chapter 2). Does it makesense to grow water-intensive crops in irrigateddesert ecosystems or erosion-prone crops onhillsides? Or does it make sense to decouplecrops from livestock and use large amountsof fertilizers to produce feed? Should nativelegume-based systems or intercropping be fur-ther encouraged in poor areas of Sub-SaharanAfrica or South Asia where soil fertility is low,fertilizers expensive, and protein deficiencieshigh? The shift toward production systems thatare more compatible with the human and nat-ural resource base leads to another principlefor redesign: the promotion of crop diversityon local to global scales (see Chapter 2). Withthe projected magnitude of climate changesto come, it will be necessary to encourage awide range of cropping systems that are suit-able for the future climate, not just to adaptexisting crops—particularly the major crops ofmaize, rice, and wheat—to increased temper-ature or drought tolerance in existing areas.Creating a resilient path for food productionsystems worldwide begins, in the words of Wen-dell Berry (2005) “in the recognition and accep-tance of limits.” It also begins with a broadunderstanding of the economic, social, and cli-mate forces that are changing the agriculturallandscape (see Chapters 2 and 3).

In order to promote a vision of redesignoperationally, the second step is that govern-ment policies influencing the trajectory of foodproduction systems must be altered in orderto re-couple agriculture with its environmen-tal support systems (Robertson and Swinton2005). For example, distortionary policies thatencourage crop or animal production that isincompatible with resource constraints shouldbe discarded, and incentives to improve theefficiency of input use and eliminate externalimpacts from agriculture should be introduced.The difficulty lies in the many policies createdwithout agriculture or livestock in mind (suchas macro policy), which affect the profitabil-ity and structure of food production systems.Moreover, a wide range of policies initiatedby different agencies with little or no overlaptypically shape food production systems withinany country, a point made clearly by the cur-rent role of energy policy and agricultural pol-icy in the USA. Given the enormous influenceof policies on crop and animal systems through-out the world, aligning economic incentives forresilience is perhaps the most challenging step.

The third and final step toward the resiliencetransition is to promote the scientific and infor-mation tool kit that enables farmers to antic-ipate and respond to change within the localcultural context. The scientific tool kit is broadand includes emerging knowledge from a widerange of disciplines (e.g., genetics, agronomy,biology, hydrology, climate science, economics)to improve management practices, breedingefforts, cultivar development and deployment,local adaptation to climate change, water avail-ability, and options for income and nutri-tional enhancement. Advanced genetics willplay a key role in identifying desired crop traitsand increasing productivity of both major andminor crops under evolving stresses; this fieldextends beyond the use of genetically modifiedorganisms to include marker-assisted breed-ing and bioinformatics (Naylor et al. 2004).Integrated management practices that improveinput use efficiency (water, nutrients) will beequally, if not more important for enhancingcrop productivity, incomes, and environmen-tal services in all areas of the world (Cassmanet al. 2003). Finally, the use of new information

280 R.L. Naylor

technology, such as remote sensing images andgeographic information systems (GIS) to iden-tify yield gaps, input use efficiencies, soil andwater constraints, and climate projections onregional scales, could help transform the abil-ity of farmers in both rich and poor countriesto respond to resource and climate changesover time (Cassman 1999, Lobell and Ortiz-Monasterio 2008).

The main point is that efforts are needed todevelop the full tool kit as an integrated pack-age throughout the world, thus helping to pre-serve options for future adaptation to change(Solow 1991). Such efforts will require the insti-tutional commitment of international agricul-ture and development agencies and donors,as well as the commitment of national gov-ernments, to reach farmers in all regions(Falcon and Naylor 2005). Many farmers arealready redesigning their agricultural systemsto meet local needs and constraints. It is nowtime for the global community to embrace avision of redesign of food production systemsto meet human needs without compromisingfuture options, cultural integrity, and environ-mental quality. The world food crisis that hitagricultural commodity markets in 2008 is astark reminder that all countries need to buildresilient food production systems.

Review Questions

1. What are the major differences betweenagricultural systems and other ecosystemslike drylands or forests? What implicationsdo these differences have for the types ofecosystem stewardship issues that arise inagriculture and the potential challenges andopportunities for solving them?

2. What are the relative costs and benefits ofintensification and extensification of agricul-ture as a way to meet global food needs?How might the costs and benefits be opti-mized in a developing nation like Indonesiaand in a developed country like Sweden orthe USA?

3. How do supply and demand for food changewith income in a developing nation? How isthis affected by eating habits, fuel demand,

and the politics of globalized trade in devel-oped nations?

4. What are important short-term and long-term adjustments to food shortage? Whatpolicies might a developing nation pursueto maximize long-term social–ecological sus-tainability within the context of meetingfuture needs for food?

5. How might food systems be redesigned tomeet the challenges of a rapidly chang-ing planet? What changes in policies mightfoster the stewardship of agricultural sys-tems, and how might these policy changes beachieved?

Additional Readings

Cassman, K.G. 1999. Ecological intensification ofcereal production systems: Yield potential, soilquality, and precision agriculture. Proceedings ofthe National Academy of Sciences 96:5952–5959.

Conway, G. 1997. The Doubly Green Revolution:Food for All in the 21st Century. Cornell Univer-sity Press, Ithaca.

Evans, L.T. 1998. Feeding the Ten Billion: Plants andPopulation Growth. Cambridge University Press.Cambridge.

Matson, P.A., W.J. Parton, A.G. Power, andM. J. Swift. 1997. Agricultural intensificationand ecosystem properties. Science 227:504–509.

Naylor, R.L. 2000. Agriculture and global change.Pages 462–475 in G. Ernst, editor. Earth Systems:Processes and Issues. Cambridge University Press,Cambridge.

Naylor, R.L., A. Liska, M. Burke, W. Falcon,J. Gaskell, et al. 2007. The ripple effect: Biofuels,food security and the environment. Environment49(9):30–43.

Power, A.G. 1999. Linking ecological sustainabilityand world food needs. Environment, Develop-ment, and Sustainability 1:185–196.

Robertson, G.P., and S.M. Swinton. 2005. Reconcil-ing agricultural productivity and environmentalintegrity: A grand challenge for agriculture. Fron-tiers in Ecology and the Environment 3:38–46.

Smil, V. 2000. Feeding the World: A Challenge for the21st Century. MIT Press, Cambridge, MA.

Tilman, D., K. Cassman, P. Matson, R. Naylor,and S. Polasky. 2002. Agricultural sustainabilityand intensive production practices. Nature 418:671–677.

13Cities: Managing Densely SettledSocial–Ecological Systems

J. Morgan Grove

Introduction

Why are Cities and UrbanizationImportant?

The transition from a rural to urban populationrepresents a demographic, economic, cultural,and environmental tipping point. In 1800,about 3% of the world’s human populationlived in urban areas. By 1900, this proportionrose to approximately 14% and now exceeds50% in 2008. Nearly every week 1.3 millionadditional people arrive in the world’s cities(about 70 million a year), with increases due tomigration being largest in developing countries(Brand 2006, Chan 2007). People in developingcountries have relocated from the countrysideto towns and cities of every size during thepast 50 years. The urban population on aglobal basis is projected by the UN to climb to61% by 2030 and eventually reach a dynamicequilibrium of approximately 80% urban to20% rural dwellers that will persist for the

J.M. Grove (�)Northern Research Station, USDA Forest Service,Burlington, VT 05403, USAe-mail: [email protected]

foreseeable future (Brand 2006, Johnson 2006).This change from 3% urban population tothe projected 80% urban is a massive changein the social–ecological dynamics of theplanet.

The spatial extent of urban areas is growingas well. In industrialized nations the conversionof land from wild and agricultural uses to urbanand suburban settlement is growing at a fasterrate than the growth in urban population. Citiesare no longer compact (Pickett et al. 2001); theysprawl in fractal or spider-like configurations(Makse et al. 1995) and increasingly interminglewith wildlands. Even for many rapidly growingmetropolitan areas, suburban zones are grow-ing faster than other zones (Katz and Bradley1999). The resulting new forms of urban devel-opment include edge cities (Garreau 1991) anda wildland–urban interface in which housing isinterspersed in forests, shrublands, and deserthabitats.

Accompanying this spatial change is achange in perspectives and constituencies.Although these habitats were formerly domi-nated by agriculturists, foresters, and conserva-tionists, they are now increasingly dominatedby people possessing resources from urban sys-tems, drawing upon urban experiences, andexpressing urban habits.


282 J.M. Grove

An important consequence of these trendsin urban growth is that cities have become thedominant global human habitat of this centuryin terms of geography, experience, constituency,and influence. This reality has important con-sequences for social and ecological systems atglobal, regional, and local scales, as well asfor natural resource organizations attemptingto integrate ecological function with humandesires, behaviors, and quality of life.

Urbanization is a Dynamic, Social,and Ecological Phenomenon

Urbanization is having significant and unpre-dicted effects on the human global population.According to UN projections, the world’s over-all fertility rate will decline below replacementlevels by 2045, due in large part to decliningfertility in cities. In cities women tend to haveboth more economic opportunities and morereproductive control, and the economic benefitof children depends less on the quantity of chil-dren than on their quality, particularly in termsof education. This trend is amplified by the factthat the social and financial costs of childbear-ing and childrearing continue to rise in cities(Brand 2006).

Birthrates on a national basis have alreadydropped as a result of rapid urbanization inboth developed and developing nations. In thecase of the developing world, the fertility ratehas declined from six children per womanin 1970 to 2.9 currently. In twenty emerging-economy countries—including China, Chile,Thailand, and Iran—the fertility rate hasdeclined below the replacement rate of 2.1 chil-dren per woman (Brand 2006).

What this means in a global, long-termdemographic context is that, although theworld’s population doubled in a single genera-tion for the first time in human history, from 3.3billion in 1962 to 6.5 billion now, this is unlikelyto occur again. The “population momentum” ofour current global population and our childrenwill carry the world population to a peak of 7.5–9 billion around 2050 and then decline (Brand2006).

Urbanization creates both ecological vulner-abilities and efficiencies. For instance, coastalareas, where many of the world’s largest citiesoccur, are home to a wealth of natural resourcesand are rich with diverse species, habitat types,and productive potential. They are also vulner-able to land conversion, changes in hydrologicflows, outflows of waste, and sea level rise (seeChapter 12; Grimm et al. 2008). In the USA,10 of the 15 most populous cities are locatedin coastal counties (NOAA 2004) and 23 of the25 most densely populated US counties are incoastal areas. These areas have already experi-enced ecological disruptions (Couzin 2008).

The link between urbanization and coastalareas is evident on a global basis as well.Because of the coastal locations of many majorcities, urban migration also brings people tocoastlines around the world in one of the great-est human migrations of modern times. Themost dramatic population growth has occurredin giant coastal cities, particularly those in Asiaand Africa. Many experts expect that cities willhave to cope with almost all of the populationgrowth to come in the next two decades, andmuch of this increase will occur in coastal urbancenters (Brand 2006, Johnson 2006).

While ecological vulnerabilities are signifi-cantly associated with urban areas, urbaniza-tion also fosters ecological efficiencies. Theecological footprint of a city, i.e., the landarea required to support it, is quite large(Folke et al. 1997, Johnson 2006, Grimm et al.2008). Cities consume enormous amounts ofnatural resources, while the assimilation oftheir wastes—from sewage to the gases thatcause global warming—also are distributedover large areas. For example, London occupies170,000 ha and has an ecological footprint of21 million hectares—125 times its size (Toepfer2005). In Baltic cities, the area needed from for-est, agriculture, and marine ecosystems corre-sponds to approximately 200 times the area ofthe cities themselves (Folke et al. 1997).

Ecological footprint analysis can be mislead-ing, however, for numerous reasons (Deutschet al. 2000). It ignores the more important ques-tion of efficiency, defined here as persons-to-area: how much land area (occupied area andfootprint area) is needed to support a certain

13 Managing Densely Settled Social–Ecological Systems 283

number of persons? From this perspective, itbecomes clear that urbanization is critical todelivering a more ecologically sustainable andresource-efficient world because the per-personenvironmental impact of city dwellers is gen-erally lower than people in the countryside,and it can be reduced still further (Brand 2006,Johnson 2006, Grimm et al. 2008). For instance,the average New York City resident generatesabout 29% of the carbon dioxide emissions ofthe average American. By attracting 900,000more residents to New York City by 2030, NewYork City can actually save 15.6 million met-ric tons of carbon dioxide a year relative tothe emissions of a more dispersed population(Chan 2007).

The combined effects of urbanization onmigration, fertility, and ecological efficiencymay mean that social–ecological pressures onnatural systems can be dramatically reducedin terms of resources used, wastes produced,and land occupied. This may mean that citiescan provide essential solutions to the long-termsocial–ecological viability of the planet givencurrent population trends for this century.

Are Urban Areas Ecosystems?

Earlier chapters about low-density, social–ecological systems such as drylands, forests,and oceans took pains to point out the per-vasive importance of social processes in gov-erning social–ecological dynamics. Conversely,the fundamental importance of ecological pro-cesses is sometimes overlooked in cities. Anecosystem is an assemblage of organisms inter-acting with the physical environment within aspecified area (see Chapter 1; Tansley 1935,Bormann and Likens 1979). When Tansley(1935) originated the term ecosystem, he care-fully noted that “. . . ecology must be appliedto conditions brought about by human activ-ity. The ‘natural’ entities and the anthropogenicderivatives alike must be analyzed in terms of themost appropriate concepts we can find.” Sincethe 1950s, social scientists have contributed toan expanded view of ecosystems inclusive ofhumans along a continuum from wildernessto urban areas (Hawley 1950, Schnore 1958,

Duncan 1961, 1964, Burch and DeLuca 1984,Machlis et al. 1997). Public health researchersand practitioners have provided supplementalperspectives (Northridge et al. 2003), and trans-disciplinary approaches have been proposedto implement a social–ecological framework(Chapters in this book; Elmqvist et al. 2004,Collins et al. 2007).

An urban ecosystem perspective retains aconcern with ecological structure and func-tion, including biophysical fluxes (Stearns andMontag 1974, Boyden et al. 1981, Burch andDeLuca 1984a, Warren-Rhodes and Koenig2001) and ecological regulation of systemdynamics (Groffman et al. 2003, Pickett et al.2008). At the same time, demographic, social,and economic structures and fluxes clearly exertimportant controls over these dynamics as well(Burch and DeLuca 1984a, Grove and Burch1997, Machlis et al. 1997). Integrating social–ecological structure, function, and regulation ofurban ecosystems is therefore essential to anunderstanding of the ecology of cities (Grimmet al. 2000, Pickett et al. 2001) as open complexadaptive systems that can be characterized interms of vulnerability and resilience. In contrastto rural areas, urban social–ecological systemsare distinguished by a high population density,the built environment, and livelihoods that donot directly depend on the harvest or extractionof natural resources. Finally, ecosystem serviceconcerns are likely to differ between cities andmany rural areas, particularly cultural servicessuch as social identity, knowledge, spirituality,recreation, and aesthetics.

Why Use the Approaches Describedin This Book?

In very broad historical terms we have beguna new paradigm for cities. Since the 1880s, agreat deal of focus has centered on the “Sani-tary City,” with concern for policies, plans, andpractices that promoted public health (Melosi2000). While retaining the fundamental concernfor the Sanitary City, we have begun to enve-lope the Sanitary City paradigm with a con-cern for the “Sustainable City,” which places

284 J.M. Grove

urbanization in a social–ecological context on alocal, regional, and global basis.

Urban ecology has a significant role to playin this context. Already, urban ecology has animportant applied dimension as an approachused in urban planning, especially in Europe.Carried out in city and regional agencies, theapproach combines ecological information withplanning methodologies (Hough 1984, Spirn1984, Schaaf et al. 1995, Thompson and Steiner1997, Pickett et al. 2004, Pickett and Cadenasso2007).

Cities face challenges that are increasinglycomplex and uncertain. Many of these com-plexities are associated with changes in climate,demographics, economy, and energy at multiplescales. Because of these complex, interrelatedchanges, concepts such as resilience, vulnerabil-ity, and ecosystem services may be particularlyuseful for addressing current issues and oppor-tunities as well as preparing for potential futurescenarios requiring long-term, and frequentlycapital-intensive, change.

Cities have already begun to address thesechallenges and opportunities in terms of poli-cies, plans, and management. For example, onJune 5, 2005, mayors from around the globetook the historic step of signing the UrbanEnvironmental Accords—Green City Decla-ration with the intent of building ecologicallysustainable, economically dynamic, and sociallyequitable futures for its urban citizens. TheAccord covered seven environmental cate-gories to enable sustainable urban living andimprove the quality of life for urban dwellers:(1) energy, (2) waste reduction, (3) urbandesign, (4) urban nature, (5) transportation,(6) environmental health, and (7) water(www.urbanaccords.org). International associa-tions such as ICLEI-Local Governments for Sus-tainability (http://www.iclei.org/) are developingand sharing resources to address these issues.

The ability to address these seven categorieswill require numerous, interrelated strategies.New York City’s plaNYC for A Greener,Greater New York (http://www.nyc.gov/html/planyc2030/downloads/pdf/full report.pdf), forexample, includes 127 different but interrelatedstrategies for making the city more sustainable,dynamic, and equitable. However, many citiesare managed in disciplinary and fragmented

ways. In some sense, city agencies and non-governmental organizations (NGOs) too oftenresemble traditional university departmentsseparated by academic disciplines. The essenceof this situation readily maps to Yaffee’s (1997)“recurring nightmares” (Chapter 4): (1) a pro-cess in which short-term interests out-competelong-term visions and concerns; (2) conditionsin which competition supplants cooperationbecause of the conflicts that emerge in manage-ment issues: (3) the fragmentation of interestand values; (4) the fragmentation of responsi-bilities and authorities (sometimes called “func-tional silos” or “stove pipes”); and (5) thefragmentation of information and knowledge,which leads to inferior solutions.

To address these “recurring nightmares,”universities and cities alike have begun to reor-ganize themselves in part by creating Offices ofSustainability (Deutsch 2007). A fundamentalchallenge to these types of offices and the poly-centric networks in which they exist will be tounderstand urban ecosystems as complex adap-tive systems in order to build resilient urbanfutures that are ecologically sustainable, eco-nomically dynamic, and socially equitable.

The following section applies several of theresilience principles described earlier in thisbook for understanding and building moreresilient urban futures: (1) cities are open, andmultiscale systems, (2) cities are heterogeneousand ecosystem composites, and (3) cities arecomplex adaptive systems.

Principles

Cities are Open, and MultiscaleSystems

The recognition that cities are open multi-scale systems has only recently become evi-dent in the ecological study of urban areas(Pickett et al. 1997a, Grimm et al. 2000).Urban ecology began as the study of “ecol-ogy in cities,” which focused historically on eco-logically familiar places and compared urbanand nonurban areas: parks as analogs of ruralforests (e.g. Attorre et al. 1997, Kent et al.1999) and vacant lots as analogs of fields orprairies (Vincent and Bergeron 1985, Cilliersand Bredenkamp 1999). Urban streams, rock


outcrops, and remnant wetlands were theobject of ecological studies similar in scopeand method to those conducted in nonurbanlandscapes. There is a long European tradi-tion of these types of ecological studies incities (Sukopp et al. 1990, Berkowitz et al.2003).

The study of “ecology of cities” buildsupon the focused efforts of “ecology in cities,”while incorporating a more expansive approachto cities that is consistent with the social–ecological approaches described in this book(see Chapters 1–5). In particular, the ecology-of-cities approach developed in response to therecognition of the open and multiscale nature ofcities. Input–output budgets of a city were thefirst type of ecology-of-cities approach address-ing the open nature of cities. This budgetaryapproach relies on a “closed box” approachto ecosystems. Inputs and outputs are mea-sured and the processes within the system areimplicitly assumed to be homogeneous. Thisapproach is similar to the ecosystem ecology ofthe 1960s and 1970s and has been used by ecolo-gists (Bormann and Likens 1967), environmen-tal historians (Cronon 1991), and social scien-tists (Stearns and Montag 1974). The materialand energy budget of Hong Kong (Boyden et al.1981) and the nitrogen budget of New Haven,Connecticut (Burch and DeLuca 1984) areexamples. The lack of interdisciplinary expertsnoted by Boyden et al. (1981) and the appar-ent lack of interest by mainstream ecology con-strained this approach to cities. However, thetwo urban projects of the Long-Term Ecologi-cal Research (LTER) Network, the BaltimoreEcosystem Study (BES) and Central ArizonaPhoenix (CAP) program, have developed nitro-gen budgets in terms of both internal dynamicsand inputs–outputs for their urban ecosystems(Baker et al. 2001, Groffman et al. 2004).

Ecology of cities in its contemporary formincorporates new approaches from ecology ingeneral and from ecosystem ecology in par-ticular. It also benefits from relatively newspecialties such as landscape ecology, whichfocuses on the functional consequences ofspatial heterogeneity. It further benefits fromincreasing interdisciplinary work and training.Together, these developments make the inclu-sive approach to ecology of cities very differ-

ent from the examples from the 1970s and early1980s. There are several reasons for this dif-ference. First, the ecology of cities addressesthe whole range of habitats in metropolitansystems, not just the green spaces that arethe focus of ecology in cities. Second, spa-tial heterogeneity, expressed as gradients ormosaics, is critical for explaining interactionsand changes in the city. Third, the role ofhumans at multiple scales of social organiza-tion, from individuals through households andephemeral associations, to complex and persis-tent agencies, is linked to the biophysical scalesof the metropolis. Finally, humans and theirinstitutions are a part of the ecosystem, not sim-ply external, negative influences. This opens theway toward understanding feedbacks amongthe biophysical and human components of thesystem, toward placing them in their spatialand temporal contexts, and toward examiningtheir effects on ecosystem inputs and outputsat various social scales, including individuals,households, neighborhoods, municipalities, andregions (Grove and Burch 1997).

Cities are Heterogeneousand Ecosystem Composites

Urban ecosystems are notoriously heteroge-neous or patchy (Jacobs 1961, Clay 1973). Bio-physical patches are a conspicuous layer ofheterogeneity in cities. The basic topography,although sometimes highly modified, contin-ues to govern important processes in the city(Spirn 1984). The watershed approach to urbanareas has highlighted the importance of slopes,and of patchiness along slopes, in water flowand quality (Band et al. 2006). Steep areas areoften the sites of remnant or successional for-est and grassland in and around cities. Soil anddrainage differ with the underlying topogra-phy. Vegetation, both volunteer and planted,is an important aspect of biophysical patch-iness. The contrast in microclimate betweenleafy, green neighborhoods versus those lack-ing a tree canopy is a striking example of bioticheterogeneity (Nowak 1994). Additional func-tions that may be influenced by such patchi-ness include carbon storage (Jenkins and Rie-mann 2003), animal biodiversity (Adams 1994,

286 J.M. Grove

Hostetler 1999, Niemela 1999), social cohesion(Grove 1995, Colding et al. 2006), and crime(Dow 2000, Troy and Grove 2008).

Social and economic heterogeneity is alsopronounced in and around cities. Patchinesscan exist in such social phenomena as eco-nomic activity and livelihoods, family structureand size, age distribution of the human popula-tion, wealth, educational level, social status, andlifestyle preferences (Burch and DeLuca 1984,Field et al. 2003).

Temporal dynamics are just as importantas spatial pattern, since none of these socialpatterns are fixed in time. This insight is akey feature of the socio-spatial (Gottdienerand Hutchinson 2001) and patch dynamicsapproaches to urban ecosystems (Pickett et al.1997b, Grimm et al. 2000, Pickett et al. 2001).It is a critical feature for including the builtnature of cities as well. Most people, and indeedmost architects and designers, assume that thebuilt environment is a permanent fixture. How-ever, buildings and infrastructure change, asdo their built and biophysical context. Thiselasticity in the urban system suggests a pow-erful way to reconceptualize urban design asan adaptive, contextualized pursuit (Pickett etal. 2004, Shane 2005, Colding 2007, McGrathet al. 2008). Such dynamism combines withthe growing recognition of the role of urbandesign in improving the ecological efficienciesand processes in cities. Although this appli-cation of patch dynamics is quite new, it hasgreat promise to promote the interdisciplinarymelding of ecology and design and to gener-ate novel designs with enhanced environmentalbenefit (McGrath et al. 2008). Thus, patches inurban systems can be characterized by biophys-ical structures, social structures, built structures,or a combination of the three at multiple scales(Cadenasso et al. 2006).

Not only are urban areas heterogeneous,they are ecological composites, constituted bymany of the ecosystem types described in thisbook: forests, drylands, freshwaters, estuaries,coastal areas, and urban gardens (see Chap-ters 8–12), combined with the social attributesdescribed in Chapters 3–4. Because cities areecological composites, they often include theecological and social characteristics associated

with these individual ecosystem types. In thecase of freshwaters, for example, cities bor-dering lakes and rivers are affected by thespatial heterogeneity of flows; the interac-tion of fast and slow variables; nonlinear-ities and thresholds in system change; theneed to account for blue and green water;and management systems that are fragmentedand require continuous adaptation. Likewise,in the case of estuaries, social sources ofresilience depend upon monitoring programs,spatial complexity, some degree of localizedmanagement control, and a willingness to enter-tain and implement actively adaptive, experi-mental management policies (Felson and Pick-ett 2005). Because of this urban ecological com-position, there is a great deal to learn andadapt from the experiences and knowledgeof these particular ecosystem types to urbansettings.

Cities are Complex Adaptive Systems

The fact that cities are complex adaptive sys-tems is manifest in the definition provided inChapter 1: Systems whose components inter-act in ways that cause the system to adjust or“adapt” in response to changes in conditions.This is a simple consequence of interactions andfeedbacks. These interactions and feedbacks areexpressed in urban areas in several ways.

Cities, like all social–ecological systems, existin a state of nonequilibrium (Pickett and Cade-nasso 2008) as a result of both major disrup-tions (pulses) and chronic stresses (presses).Pulses include disease epidemics, droughts,famines, floods, earthquakes, fires, and warfare.Long-term presses result from demographicchanges caused by immigration, emigration,and/or aging; changes in economy through tran-sitions from agriculture, manufacturing, ship-ping, and service economies (Fig. 13.1); andchanges in transportation systems includingwater, rail, auto, and air. The dynamic results ofthese long-term press and pulse forces are man-ifest in Batty’s (2006) long-term rank clocks forurban areas in the USA (1790–2000) and theplanet (430 BC–AD 2000), which illustrate the


0

200

400

600

800

1000

1880 1900 1920 1940 1960 1980 20000

2

4

6

8

10

12

14

Population

Pop

ula

tio

n (t

ho

usa

nd

s) Manufacturing

Commerce

Services

Year

% o

f to

tal p

op

ula

tio

n e

mp

loye

d b

y se

cto

r

Figure 13.1. Long-term trends in population andeconomy for the city of Baltimore, USA, 1880–2000.Data from the LTER Ecotrends Project: Socioe-

conomic Catalogue (http://coweeta.ecology.uga.edu/trends/catalog˙trends˙base2.php).

long-term dynamics of cities in terms of popu-lation size over time (Figs. 13.2, 13.3).

Social institutions (the rules of the game)play a key role in the adjustments or adap-

tations of cities to changing conditions (seeChapter 4, Burch and DeLuca 1984, Machlis etal. 1997). Institutions, such as property rights,direct the allocation of resources to individuals

1820

1830

1840

1850

1870

1860

188018901900

1910

1920

1930

1940

1950

1960

1970

19801990

17902000

18101800

Atlanta

Richmond,VA

LosAngeles

Boston

Phoenix

Figure 13.2. The trajectory of relative populationrank of four of the 100 most populous US citiesfrom 1790 to 2000. The most populous city is at thecenter of the rank clock, and the hundredth cityis at the periphery. Boston, an important colonialcity in the northeastern USA, has always been oneof the largest US cities. Richmond, another impor-

tant colonial city, declined in importance during theearly twentieth century, as new cities like Los Ange-les in the western USA became important. Phoenix,a desert city attractive to retired persons, becameimportant only in the last 50 years. Very few citieshave remained among the most populous US citiesthroughout their history. Modified from Batty (2006).

288 J.M. Grove

–200

–430

1950

2000

19001850

18001750

1700

1650

1600

1550

1500

1450

1400

1400

1300

1250

1200

1150

1100

1000

900800

622

622

361

100

Rome

Rome

Nanking

Mexico City

Paris

Figure 13.3. The trajectory of relative populationrank of 4 of the 50 most populous world cities from430 BC to 2000 AD. The most populous city is atthe center of the rank clock, and the fiftieth city isat the periphery. Rome (Italy) and Nanking (China)remained important throughout most of this 2500-year history. Most cities, however, have had a highly

volatile population history, with many cities of thedeveloped world (e.g., Paris, France) declining inrelative rank during the industrial revolution andcities from the developing world (e.g., Mexico City)becoming more populous, a trend that is likely tocontinue in the future. Modified from Batty (2006).

and organizations and affect human interac-tions. Social institutions can be thought of asdynamic solutions to universal needs, includ-ing health, justice, faith, commerce, education,leisure, government, and sustenance (Machliset al. 1997). While the structure of these insti-tutions is important, structure should not bemistaken for function: the health of individ-uals and populations, the exchange of goodsand services, the provisioning of food, water,energy, and shelter (Machlis et al. 1997). Inthis context, different forms of social institu-tions might yield identical functions. The abil-ity of social institutions to change in formand yet continue to yield comparable insti-tutional functions is a key element to theadaptive capacity of urban social–ecologicalsystems.

Social institutions are interrelated and oftendepend upon polycentric governance and social

capital within and among cities (see Chapters4 and 5). For instance, social capital is a cru-cial factor differentiating “slums of hope” from“slums of despair” (Box 13.1; Brand 2006).This is where community-based organizations(CBOs) and NGOs that support local empow-erment play critical roles (Colding et al. 2006,Lee and Webster 2006, Andersson et al. 2007).Typical CBOs include, according to a 2003UN report, “community theater and leisuregroups; sports groups; residents associations orsocieties; savings and credit groups; child caregroups; minority support groups; clubs; advo-cacy groups; and more. . . . CBOs as interestassociations have filled an institutional vac-uum, providing basic services such as com-munal kitchens, milk for children, income-earning schemes and cooperatives” (Brand2006). CBOs and NGOs can be diverse, notnecessarily focused on “environmental” issues,


Box 13.1. Slums: Past and Present

It can be argued that cities have rarely beenthe result of grand master plans and thatall of the world’s current major cities beganas disreputable shantytowns (Brand 2006,Neuwirth 2006). Slums remain a prominentissue of concern today and, in many ways, themegacities of the developing world are strug-gling with the same issues of uncharted andpotentially unsustainable growth that indus-trial cities in Europe and the USA facedin the late 1800s and early 1900s (Johnson2006). By 2015 for instance, the five largestcities on the planet will be Tokyo, Mum-bai, Dhaka, Sao Paulo, and New Delhi, withpopulations greater than 20 million each.Most of this growth will occur in shanty-towns: built on illegally occupied land with-out the guidance of civic planning profession-als or traditional infrastructure to supportits growth. By some estimates, 25% of theworld’s population will live in shantytownsby 2030 (Johnson 2006).

Although shantytowns lack the formalplans and infrastructure of urban areas indeveloped countries, they are dynamic placesof social innovation and creativity with eco-nomic activities of ordinary life: shops, banks,and restaurants. All of this has been accom-plished without urban planners, withoutgovernment-created infrastructure, and with-out formal property deeds. While these shan-tytowns might offend some persons’ sense oforder, these shantytowns are not exclusivelyplaces of poverty and crime. In fact, they arewhere the developing world goes to get outof poverty (Brand 2006). They are a reminder

that different social forms might yield identi-cal functions; that the ability of social insti-tutions to change in form yet continues toyield comparable institutional functions is akey element to the adaptive capacity of urbansocial–ecological systems.

It might be negligent to claim shanty-town residents do not need nor want for-mal civic resources in terms of expertise,investments, and opportunities in areas suchas epidemiology, public infrastructure, edu-cation, engineering, waste management, andrecycling (Brand 2006, Johnson 2006). But itwould also be negligent to not recognize theenormous variety of shantytown experiencesamong the thousands of “emerging citieswith different cultures, nations, metropoli-tan areas, and neighborhoods. From this vari-ety is emerging an understanding of bestand worst governmental practices—best, forexample, in Turkey, which offers a standardmethod for new squatter cities to form; worst,for example, in Kenya, which actively pre-vents squatters from improving their homes.Every country provides a different example.Consider the extraordinary accomplishmentof China, which has admitted 300 millionpeople to its cities in the last 50 years withoutshantytowns forming, and expects another300 million to come” (Brand 2006:13). And inthis process it might be important to examinethe ecological footprint of shantytowns, withtheir extreme density, low-energy use, andingenious practices of recycling everything.Maybe there are ideas there that could begeneralized on a global basis (Brand 2006).

Religious groups play significant roles aswell. According to Davis (2006) “PopulistIslam and Pentecostal Christianity (and inBombay, the cult of Shivaji) occupy a socialspace analogous to that of early twentieth-century socialism and anarchism. In Morocco,for instance, where half a million rural emi-grants are absorbed into the teeming cities

every year, and where half the population isunder 25, Islamicist movements like ‘Justice andWelfare,’ founded by Sheik Abdessalam Yassin,have become the real governments of the slums:organizing night schools, providing legal aidto victims of state abuse, buying medicine forthe sick, subsidizing pilgrimages and paying forfunerals.” He adds that “Pentecostalism is. . .the

290 J.M. Grove

first major world religion to have grown upalmost entirely in the soil of the modern urbanslum” and “since 1970, and largely because ofits appeal to slum women and its reputation forbeing colour-blind, [Pentecostalism] has beengrowing into what is arguably the largest self-organized movement of urban poor people onthe planet” (Brand 2006, Johnson 2006).

The networks among cities are importanttoo. As New York City prepared its Greener,Greater New York plan, David Doctoroff,deputy mayor for Economic Development,noted that, “We shamelessly stole congestionpricing from London and Singapore, renewableenergy from Berlin, new transit policies fromHong Kong, pedestrianization and cycling fromCopenhagen, bus rapid transit from Bogota,and water-cleaning mollusks from Stockholm”(Chan 2007).

Many of the social and ecological inter-actions and feedbacks in urban ecosystemsconfer resilience: the capacity of a social–ecological system to absorb a spectrum ofshocks or perturbations without fundamen-tally altering its structure, functioning, andfeedbacks. Resilience depends on (1) adap-tive capacity; (2) biophysical and social lega-cies that contribute to diversity and provideproven pathways for rebuilding; (3) the capac-ity of people to plan for the long term withinthe context of uncertainty and change; (4) a bal-ance between stabilizing feedbacks that bufferthe system against stresses and disturbanceand innovation that creates opportunities forchange; and (5) the capacity to adjust gover-nance structures to meet changing needs (seeChapter 1; Gunderson and Holling 2002, Folke2006, Walker and Salt 2006).

Biophysical and social legacies can signifi-cantly affect the resilience of urban systems.These legacies or dependencies can be tempo-ral or spatial. For instance, historic residentialsegregation and industrial development in Bal-timore, Maryland, created a situation in whichpredominantly white neighborhoods are inproximity to TRI sites (toxic release inventorysites; Boone 2002). Legacies [dependencies] canbe spatial too. For example, as police depart-ments decide how to allocate scarce resources,part of their decision-making process is to

label neighborhoods as green (no crime prob-lems), yellow (some crime problems), and red(severe crime problems). Whether the policedepartment expends resources in a neighbor-hood depends upon their assessment of thatneighborhood as well as adjacent neighbor-hoods. Thus, if a neighborhood is coded yellowand is bordering green neighborhoods, policeresources are dedicated to the yellow neighbor-hood to reduce the risk of spillover or contagioninto green neighborhoods. If a neighborhoodis coded yellow and is bordering red neighbor-hoods, police resources may not be investedbecause the likelihood of decline from yellowto red is too great. Finally, temporal and spa-tial legacies can be prospective. In other words,people take specific actions in specific placestoday because they believe those actions willinfluence the capacity of future residents andgovernance networks to meet short-term andlong-term challenges and opportunities.

Working prospectively to create social-ecological legacies is a profound challenge. AsDoctoroff notes (Chan 2007), “sustainability isan almost sacred obligation to leave this citybetter off for future generations than we whoare here today have found it.” The greatestchallenges are not matters of technology, butrather issues of “will and leadership.” Short-term sacrifices for long-term gains “are notthings that, by its very nature, the political sys-tem is equipped to decide.”

While Doctoroff’s observation is well-founded, there is already evidence for urbanresilience. Cities are the most long-lived ofall human organizations. The oldest survivingcorporations, the Sumitomo Group in Japan,and Stora Enso in Sweden, are about 400and 700 years old, respectively. The oldestuniversities in Bologna and Paris have beenin place more than a 1,000 years. The oldestliving religions, Hinduism and Judaism, haveexisted more than 3,500 years. In contrast tothese corporations and religions, the town ofJericho has been continuously occupied for10,500 years and its neighbor, Jerusalem, hasbeen an important city for 5,000 years, though ithas been conquered or destroyed 36 times andexperienced 11 religious conversions (Brand2006). Many cities die or decline to irrelevance,


but some thrive for millennia (Batty 2006,Brand 2006, Johnson 2006; Fig. 13.3).

Part of a city’s durability is associated withthe fact that it is constantly changing. InEurope, cities replace 2–3% per year of theirmaterial fabric (buildings, roads, and otherconstruction) by demolishing and rebuildingit. In the USA and the developing world,that turnover occurs even faster. Yet withinall of that turnover something about a cityremains deeply constant and self-inspiring(Brand 2006). Some combination of geography,economics, and cultural identity ensures thateven a city destroyed by war (Warsaw, Dres-den, Tokyo) or fire (London, San Francisco)will often be rebuilt (Johnson 2001, Brand 2006,Johnson 2006). Thus, the resilience of urbanecosystems rarely depends upon a single factor,but a diversity of interacting social–ecologicalfeedbacks.

Cities as FunctionalSocial–Ecological Systems

Ecosystem Services and the Dynamicsof Cities

The existence, significance, and dynamics ofecosystem services—supporting, provisioning,regulating, and cultural services—have beenonly partially characterized in urban areas(Bolund and Hunhammar 1999, Colding et al.2006, Farber et al. 2006, Andersson et al.2007). Their existence in urban areas is increas-ingly well-documented, although frequently notidentified as ecosystem services per se. Forinstance, there is growing knowledge and dataabout urban ecosystems describing soil dynam-ics and nutrient flux regulation (supporting);production of freshwater, food, and biodiversity(provisioning); modification of climate, hydrol-ogy, and pollination (regulation); and links tosocial identity and spirituality and importanceto recreation and aesthetics (cultural).

The significance of these ecosystem servicesis often poorly understood. One approach is toestimate the monetary value of ecosystem ser-vices, for example, the capacity of ecosystems topurify water for New York City (see Chapter 9).

Although this approach is valuable in address-ing certain issues, it may miss important aspectsof the dynamics of urban ecosystems: “Do vari-ations in ecosystem services affect the desir-ability of cities?” In other words, do ecosys-tem services affect where households (families)and firms (businesses) choose to locate in orderto avoid some places (push) and seek otherplaces (pull)? Clearly, this may be the case ashouseholds and firms seek places that afford,for instance, clean air and water, recreationand communal opportunities, and efficiencies inenergy and transportation. In contrast to othertypes of ecosystems, provision of and access tothese ecosystem services is regulated by a com-bination of ecological, social, and built systems.

The dynamics or interactions among ecosys-tem services in urban areas are poorly under-stood. Preferences for ecosystem services mayvary over time. For instance, in the early 1900sin Baltimore, Maryland, households preferredto live close to industrial factories where theyworked and were not concerned with air qual-ity. With growing knowledge about the relationbetween air quality and health, households nowvalue clean air more than proximity to theirwork place, and desirability of these neighbor-hoods has declined (Boone 2002).

Preferences for ecosystem services may varyamong social groups. Certain ethnic groups mayprefer different recreation or communal oppor-tunities. Demographics or life-stage is impor-tant too. Young families with children mayprefer houses with large yards and nearby play-parks for their children, while retired couplesmay prefer to live in apartments close to green-ways and waterfront promenades for their dailywalks.

Preferences for ecosystem services may beconditioned by interacting factors. For instance,living close to a park in Baltimore is generallyhighly desirable. Proximity to a park increasesthe value of a home in neighborhoods withlow levels of crime. However, in neighborhoodswith high levels of crime, living close to a parkactually depresses the value of a home (Troyand Grove 2008).

Finally, preferences for ecosystem servicesmay be nonlinear and characterized by thresh-olds. In other words, more is not always bet-

292 J.M. Grove

ter. For example, increases of tree canopycover in Baltimore led to increased environ-mental satisfaction up to a threshold of about˜60%, above which environmental satisfactionno longer increased.

In sum, we increasingly understand andappreciate the existence and importance ofecosystem services in urban areas. However, weare only beginning to understand the dynam-ics of human responses to variation in theseservices. Understanding these dynamics is cru-cial for understanding the push–pull drivers ofurban ecosystems and their resilience over time(Borgstrom et al. 2006).

Management of Complex AdaptiveSystems

Cities rarely are the result of grand mas-ter plans. Rather, they often exhibit emergentproperties that are the result of bottom-up pro-cesses driven by diverse interests, agencies, andevents (Jacobs 1961, Johnson 2001, Shane 2005,Johnson 2006, Batty 2008, Grimm et al. 2008).Given the bottom-up nature of these processesand emergent properties, it is remarkable howsimilar cities tend to be in their functions (Batty2008, Grimm et al. 2008).

Polycentric governance (see Chapter 4)among different types of organizations—government agencies, NGOs, and CBOs—andacross scales often exist. These networks tendto focus on a specific issue or interest: thestovepipes that are part of Yaffee’s recur-ring nightmares. The ability of polycentricgovernance networks to contribute to urbanresilience depends upon their capacity to(1) address the essential interdependence ofdemographic, economic, social, and ecologicalchallenges and solutions that cities face; (2)plan for the long term within the context ofuncertainty and change; and (3) adjust gov-ernance structures to meet changing needs(see Chapter 5). This requires the ability tosense and interpret patterns and processesat multiple scales. At a local scale, there isa growing trend among city governments todevelop GIS-based systems that monitor awide range of indicators in nearly real-time. For

instance, a number of US cities have developedCitiStat-type programs that provide accurateand timely intelligence, develop effective tac-tics and strategies, rapidly deploy resources,and facilitate follow-up and assessments(http://www.baltimorecity.gov/news/citistat/).

New York City has taken the CitiStatapproach to a new level with its 311 system.The 311 system functions in three ways, citizensreporting problems to the city, citizens request-ing information about city services, and as amutually learning system that builds upon thefirst two functions. The novel idea behind theservice is that this information exchange is gen-uinely two-way. The government learns as muchabout the city as the 311 callers do. In a sense,the City’s 311 system functions as an immenseextension of the city’s perceptual systems, har-nessing millions of ordinary “eyes on the street”to detect emerging problems or report unmetneeds (Johnson 2006).

New York City’s 311 approach makes mani-fest two essential principles for how cities cangenerate and transmit good ideas. First, theelegance of technologies like 311 is that theyamplify the voices of local amateurs and “unof-ficial” experts and, in doing so, they make iteasier for “official” authorities to learn fromthem. The second principle is the need for lat-eral, cross-disciplinary flow of ideas that canchallenge the disciplinary stovepipes of knowl-edge, data, interests, and advocacy associatedwith many government agencies, NGOs, andthe training of professionals (see Chapters 4and 5). This second principle is increasinglyrealized by the polycentric and interdisciplinarynature of sensing and interpretation facilitatedby the Web and new forms of amateur cartog-raphy built upon services like Google Earthand Yahoo! Maps. Local knowledge that hadso often remained in the minds of neighbor-hood residents can now be translated into digi-tal form and shared with the rest of the world.These new tools have begun to unleash a rev-olution in the exchange and interpretation ofdata because maps no longer need to be cre-ated by distant professionals. They are mapsof local knowledge created by local residents.And these maps are street-smart and mul-timedia. They can map blocks that are not


safe after dark, playgrounds that need to berenovated, community gardens with availableplots, or local restaurants that have room forstrollers. They can map location of trees, leak-ing sewers, and stream bank erosion. Thesetools enable locals to map their local history aswell—where things were or what things werelike—creating and sharing long-term, localknowledge (Johnson 2006).

The scale of these observations can broadenfrom a neighborhood to an entire planet asthese local data and knowledges become net-worked. Formal examples of these types ofsystems already exist. Public health officialsincreasingly have global networks of healthproviders and government officials report-ing outbreaks to centralized databases, wherethey are automatically mapped and publishedonline. A service called GeoSentinel tracksinfectious diseases among global travelers andthe popular ProMED-mail email list providesdaily updates on all known disease outbreaksaround the world (Johnson 2006). Althoughthese types of systems are intended to be earlywarning systems on specific topics, they arelikely to become interdisciplinary as the inter-dependence of challenges and solutions are rec-ognized and facilitate comparisons and learningamong both official and unofficial experts.

Summary

Cities continue to be tremendous engines ofwealth, innovation, and creativity. They arebecoming something else as well: engines ofhealth that are both public and environmen-tal, Sanitary and Sustainable. Two great threatsloom over this new millennium: global warm-ing and finite supplies of fossil fuel. These twothreats may have massively disruptive effectson existing cities in the coming decades. Theyare not likely to disrupt the macro-pattern ofglobal urbanization over the long term, how-ever (Johnson 2006). The energy efficiencies ofcities can be part of the solution for both globalwarming and energy demands.

The long-term challenges that cities face aresocial, ecological, and interrelated, includingthe growth and aging of urban populations;

aging infrastructure and incorporation of adap-tive technologies; environmental changes andresource limitations; and governance problems,particularly inequality. These challenges andtheir solutions are completely interdependent:sustainability and economic growth can be com-plementary goals (Chan 2007).

However profound the threats are that con-front us today and for the near future, theyare solvable (Johnson 2006). It will requireapproaches that perceive cities as complex,dynamic, and adaptive systems that dependupon interrelated ecosystem services at local,regional, and global scales. Polycentric gov-ernance networks will contribute to urbanresilience depending upon their adaptive capac-ity to (1) address the essential interdependenceof demographic, economic, social, built, andecological challenges and solutions that citiesface; (2) plan for the long term within the con-text of uncertainty and change; and (3) adjustgovernance structures to meet changing needs.This will increasingly involve the ability to senseand interpret patterns and processes at multiplescales. Tools that harness diverse local knowl-edges and “eyeballs on the street” to engage ingenuine exchanges and evaluations of informa-tion will become less novel and more routine.The exchange of knowledge among cities willbe important on a global basis, as we learn thatthere are multiple pathways to similar solutionsfor resilient cities that are ecologically sustain-able, economically dynamic, and socially equi-table.

Review Questions

1. Are the populations of cities growing morerapidly than the global population? Is thislikely to continue indefinitely? Why or whynot?

2. In what ways are cities similar to or differ-ent from social–ecological systems that havelower population density?

3. Are urbanization and the shift from theSanitary to the Sustainable City directionalchanges? Why or why not?

294 J.M. Grove

4. How might demands for ecosystem servicesbe similar and different in urban versus lessdensely settled areas?

5. In what ways is the transition from an“ecology in cities” to an “ecology of cities”important to understanding cities as openand multiscalar?

6. In what ways does expanding urbanizationinfluence ecological vulnerability, resilience,and efficiency of resource use on a local,regional, and global basis?

7. Evaluate the claim that cities can provideessential solutions to the long-term social–ecological viability of the planet given popu-lation trends for this century. How is this trueor not true?

8. What are some of the social challenges andopportunities to developing resilient cities?

Additional Readings

Batty, M. 2008. The size, scale, and shape of cities.Science 319:769–771.

Grimm, N., J.M. Grove, S.T.A. Pickett, andC.L. Redman. 2000. Integrated approaches

to long-term studies of urban ecological systems.BioScience 50:571–584.

Grimm, N.B., S.H. Faeth, N.E. Golubiewski,C.L. Redman, J. Wu, et al. 2008. Global changeand the ecology of cities. Science 319:756–760.

Johnson, S. 2001. Emergence: The Connected Livesof Ants, Brains, Cities, and Software. Scribner,New York.

Northridge, M.E., E.D. Sclar, and P. Biswas. 2003.Sorting out the connection between the builtenvironment and health: A conceptual frame-work for navigating pathways and planninghealthy cities. Journal of Urban Health: Bul-letin of the New York Academy of Medicine 80:556–568.

Pickett, S.T.A., M.L. Cadenasso, and J.M. Grove.2004. Resilient cities: Meaning, models andmetaphor for integrating the ecological, socio-economic, and planning realms. Landscape andUrban Planning 69:369–384.

Pickett, S.T.A., P. Groffman, M.L. Cadenasso, J.M.Grove, L.E. Band, et al. 2008. Beyond urban leg-ends: An emerging framework of urban ecologyas illustrated by the Baltimore Ecosystem Study.BioScience 58:139–150.

Spirn, A.W. 1984. The Granite Garden: Urban Natureand Human Design. Basic Books, Inc., New York.

14The Earth System: SustainingPlanetary Life-Support Systems

Oran R. Young and Will Steffen

What is the Issue?

The Earth as a whole can be viewed as a social–ecological system; in fact, the largest such sys-tem that can exist. The increasing evidencethat human activities are now interacting withthe natural environment of the Earth at thescale of the planet lends credence to this per-spective. The scientific understanding of thehuman imprint on the planet is well recog-nized throughout the policy and managementsectors and is raising severe challenges to gov-ernance structures. Never before has humanityhad to devise and implement governance struc-tures at the planetary scale, crossing nationalboundaries, continents and large biogeographicregions. Responsible stewardship of the globalsocial–ecological system is the ultimate chal-lenge facing humanity, as it entails safeguard-ing our own life-support system. The Earth asa social–ecological system is a very recent phe-nomenon. For nearly all of its existence, Earth

O.R. Young (�)Bren School of Environmental Science andManagement, University of California, Santa Barbara,CA 93106-5131, USAe-mail: [email protected]

has operated as a biophysical system, withoutthe social component, as fully modern Homosapiens arose only about 200,000–250,000 yearsago. This long evolution of Earth as a biophys-ical system provides the canvas on which thehuman enterprise has exploded with exponen-tially growing impact in the last micro-instantof Earth’s existence. A full understanding ofthe implications of this phenomenon requiresan understanding of Earth as a system, and par-ticularly the natural envelope of environmen-tal variability that provides the conditions forhuman life on the planet.

Our planet is about 4.6 billion years old,and life on Earth originated about 4 billionyears ago. The earlier notion that life has beenmaintained over this long period because thegeophysical conditions of the planet happenedto be amenable to life has been modified toacknowledge the role of life itself in maintain-ing its own environment. Biological processesinteract with physical and chemical processes tocreate and maintain the planetary environment,with life playing a much stronger role than pre-viously thought in modifying its own environ-ment (Lovelock 1979, Levin 1999, Steffen et al.2004).

The term Earth System is increasingly usedto describe the complex set of interacting


296 O.R. Young and W. Steffen

physical, chemical, biological, and anthro-pogenic processes that define the planet’s envi-ronment (Oldfield and Steffen 2004). The EarthSystem has the following major characteristics:

• It is a “materially closed system” – i.e., it doesnot exchange significant amounts of matterwith space – and it has a single primaryenergy source, the Sun.

• The major dynamical features of the EarthSystem are (1) the transport and transfor-mation of materials and energy internallythroughout the system and (2) a large arrayof complex feedback processes that give theplanetary environment its characteristic sys-temic nature and, importantly, limit variabil-ity within well-defined bounds.

• Human societies are an integral part of theEarth System but not an outside driver per-turbing an otherwise natural system. Therapid growth of the human enterprise overthe last century or two to become a globalgeophysical force means that the processesof human societies are now influencing thelarge feedback loops that modulate theglobal environment. The Earth System isnow truly a social–ecological system and isnot just a biophysical system.

The most dramatic evidence for the sys-temic nature of the planetary environment dur-ing the period of human existence comes fromthe Antarctic ice cores, such as the Vostokrecord (Fig. 14.1a; Petit et al. 1999). The Vos-tok ice core shows the variation in tempera-ture and atmospheric gas concentrations (car-bon dioxide (CO2) and methane (CH4)) overthe entire period of human existence on Earthand beyond. Three features of the record arestriking. First, the Earth’s environment cyclesthrough cold periods (“ice ages”) and shorter,intervening warm periods at regular – ca.100,000 years – intervals, ultimately paced bythe Earth’s orbit around the Sun. Second, theconcentration of CO2 and CH4, both green-house gases, in the atmosphere and the temper-ature are closely coupled. This is, however, nota simple cause–effect relationship, but rather acomplex coupling involving several global-scalefeedback loops. Third, there are well-definedupper and lower limits to the gas concentrations

and to the temperature as Earth cycles throughglacial and interglacial states. All of these fea-tures are typical of a system with a high degreeof internal self-regulation.

Feedback loops involving physical, chemical,and biological processes link physical climateand the carbon cycle (Fig. 14.2; Ridgwell andWatson 2002). Feedback loops consist of pro-cesses that either amplify or dampen an ini-tial perturbation (see Chapter 1). Combinationsof such processes form linked loops that feedback on themselves, leading to emergent phe-nomena that can either stabilize the global envi-ronment or propel it from one alternative stateto another. Understanding the nature of theseplanetary feedback loops is essential for effec-tive governance of Earth as a social–ecologicalsystem.

For nearly all of human existence, but espe-cially during the last 10,000 years, the dynamicsof the Earth System have provided an excep-tionally accommodating and resilient environ-ment that has allowed our species to growin number and with ability to modify theenvironment around us (Fig. 14.1b). This hasbenefited human society, as we developed agri-culture and then villages, cities, and civiliza-tions. This evolution of the human enterprise,from hunter-gatherers to a global geophysicalforce, is also mirrored in the increase in scaleof social–ecological interactions, from local toglobal.

Humans have been hunter-gatherers living insmall groups and, for the most part, in nomadicconditions throughout most of their existenceas a distinct species. During this period, how-ever, we slowly learned to manipulate the envi-ronment, most notably through the use of fire(Pyne 1997). Other important modifications ofthe environment include the wave of late Pleis-tocene extinctions of megafauna, due at least inpart to human hunting pressures (Martin andKlein 1984), and the domestication of animals,beginning with the dog about 100,000 yearsago. These modifications of the environmentwere largely carried out at local scales, and, forthe most part, were modifications of the natu-ral dynamics of ecosystems rather than whole-sale conversions of ecosystems. The imprint ofthese human activities therefore did not reach

14 The Earth System 297

Time (kyr BP)400 350 300 250 200 150 50100

280260240220200

700600500400

0

CO

2 (p

pm

v)

CH

4 (p

pb

v)

Infe

rred

tem

per

atu

re (o

C)

4

20–2–4–6

–8

CO2

Temp

CH4

a

b

δ180

(per

mil)

First migration offully modern

humans out of Africa

Aboriginesarrive inAustralia

Migrations offully modern humans

from South Asiato Europe

Beginningof agriculture

Great Europeancivilisations:

Greek, Roman

40

–20

0Age (kyr before present)

–38

100

ΔT

0

80 60 20

–34

–42

Figure 14.1. (a) The 420,000-year Vostok (Antarc-tica) ice core record, showing the regular pat-tern of atmospheric CO2 and CH4 concentrationand inferred temperature through four glacial-interglacial cycles (Petit et al. 1999). Redrawn from

Steffen et al. (2004). (b) The last glacial cycle of 18O(an indicator of temperature) and selected eventsin human history. Information from Oppenheimer(2004).

the global scale, nor did it modify the naturaldynamics of the Earth System.

The advent of sedentary agriculture about10,000 years ago was a major turning pointin the development of the human enterprise,with significantly more extensive modificationsof the environment at larger scales. The twomost important in terms of biophysical impactswere the clearing of forests and the tilling ofgrasslands for agricultural crops and the irriga-tion of rice. In some cases, these practices dra-matically altered landscapes at a regional level.However, the rate and geographical scale of thespread of agriculture were still modest enough

to produce no significant impacts on the dynam-ics of the Earth System. There is no strongevidence from global environmental records ofhuman impacts associated with early agricul-ture. Social–ecological systems operated at thelocal and regional scales only.

All of these changed with the beginningand spread of industrialization around 1800(Turner et al. 1990). The critical feature ofindustrialization in terms of the Earth Sys-tem was the use of fossil fuels as an energysource. Coal, and later oil and gas wereaccessed from beneath the Earth’s surface.These carbon-based fuels had been locked away


Climate state(temperature)

AtmosphericCO

2

Dust supply

Biologicalproductivity

+

+

–

–

a b

Ice volume

TemperatureDust supply

Sea level

Atmospheric CO2mixing ratio

Southern oceanproductivity

Hydrologicalcycle

Vegetationcover

Figure 14.2. Feedbacks in the climate system. Dif-ferent components of the Earth system can directlyinteract in three possible ways; a positive influence(i.e., an increase in one component directly resultsin an increase in a second), a negative influence(i.e., an increase in one component directly resultsin a decrease in a second), or no influence at all.An even number (including zero) of negative influ-ences occurring within any given closed loop givesrise to an amplifying (positive) feedback, the oper-ation of which will act to amplify an initial per-turbation. Conversely, an odd number of negativeinfluences give rise to a stabilizing (negative) feed-back, which will tend to dampen any perturbation.In the two schematics, positive influences are shownin grey and negative in black (Ridgwell and Wat-son 2002). Redrawn from Steffen et al. (2004). (a)Schematic of a simplified feedback system, involv-ing dust, the strength of the biological pump, CO2,and climatic state (represented by mean global sur-face temperature). Because there is an even number

of negative influences (=2), this represents an ampli-fying feedback, with the potential to amplify an ini-tial perturbation (in either direction). (b) Schematicof the hypothetical glacial dust-CO2-climate feed-back system, with explicit representation of the var-ious dust mechanisms that we have identified. Pri-mary interactions in the dust-CO2-climate subcycleare indicated by thick solid lines, while additionalinteractions (peripheral to the argument) are showndotted for clarity. For instance, the two-way interac-tion between temperature and ice volume is the ice-albedo feedback. Four main (amplifying) feedbackloops exist in the system: (1) dust supply → pro-ductivity → CO2 → temperature → ice volume →sea level → dust supply (four negative interactions),(2) dust supply → productivity → CO2 → tempera-ture → hydrological cycle → vegetation → dust sup-ply (two negative interactions), (3) dust supply →productivity → CO2 → temperature → hydrologicalcycle → dust volume → dust supply (two negativeinteractions).

from the active biogeochemical cycling of car-bon between land, ocean, and atmosphere formillions of years. The usage of these fuels beganto noticeably perturb the carbon cycle at theglobal scale. The flow-on effects from the useof fossil-fuel energy systems were even more

dramatic. The ability to chemically “fix” nitro-gen from the atmosphere to produce fertilizersled to even more profound changes to the nitro-gen cycle than to the carbon cycle. The ability toclear land and to extract resources from coastaland marine ecosystems accelerated with the use


of fossil-fuel-driven machinery and the trans-formation of the environment became global inscope.

The advent of industrialization as the begin-ning of a new planetary epoch, now calledthe Anthropocene (Crutzen 2002), marks thebeginning of Earth itself as a social–ecologicalsystem. Even more dramatic has been theexplosion of the human enterprise in termsof both population and economic activitiesafter World War II, now labeled the GreatAcceleration (Fig. 14.3; Hibbard et al. 2006,Steffen et al. 2004). Human population dou-bled to over 6 billion people in just 50 years,while the global economy expanded fifteen-fold. Resource use has expanded, while mobil-ity and communication have accelerated to lev-els that could scarcely have been imagined inthe first half of the twentieth century (McNeill2000). The human imprint on the global envi-ronment is now unmistakable (Fig. 14.3; Steffenet al. 2004). Climate change is the most well-known aspect of human-influenced change atthe global scale, but it is now certain that humanactivities have also modified the nitrogen, phos-phorus, and hydrological cycles and have sig-nificantly altered the structure and compositionof the atmosphere, oceans (particularly coastalseas), and land.

Directional Changesat the Planetary Scale

Taken together, the recent changes in humanactions and Earth System responses (Fig. 14.3)demonstrate without doubt that the Earth hasrapidly evolved from its earlier status as abiophysical system to its current status as asocial–ecological system. To understand thenature of this planetary social–ecological sys-tem, we need to examine in more detail therole of human actions in large-scale biophys-ical changes, often called global environmen-tal changes, the parallel changes in large-scalesocial processes, properly interpreted as globalsocial changes, and the interdependencies andinteractions between the two. The basic mes-sage of this brief account is that the Earth

System is a complex and highly dynamic systemwhose future trajectory is difficult to forecast.

Global Environmental Changes(GECs)

It is helpful to divide the human imprint intotwo types – systemic and cumulative. Systemicprocesses are those for which changes any-where on the planet rapidly affect the EarthSystem at the global scale. The classic exam-ple is emissions of greenhouse gases, which,because of the rapid mixing of the atmosphere,affect the climate at the global scale. Cumula-tive processes are those whose initial impact islocal but whose effects occurring in differentparts of the world aggregate to produce globalconsequences. For example, the loss of biodiver-sity is largely a local process with local conse-quences, but over the past two centuries, incre-mental losses of biodiversity on every continenthave aggregated to become a global crisis thatchallenges human well-being (MEA 2005d).

The best-known systemic human imprint onthe Earth System is a suite of changes in thecomposition of the atmosphere. CO2 concentra-tion now stands at 385 ppm (parts per million),100 ppm higher than the preindustrial value(the maximum value observed during previ-ous interglacial periods was 280–300 ppm). Thecurrent rate of increase is 2–3 ppm per year.Concentrations of other important greenhousegases, such as methane and nitrous oxide, havealso risen significantly over the past two cen-turies because of human activities.

The atmosphere has also changed as a resultof human-driven emissions of aerosols, smallparticles of various sizes and composition.Unlike greenhouse gases, which have long life-times in the atmosphere and are thus wellmixed at the global scale, aerosols have life-times on the order of days and sometimesa week. They are therefore largely local andregional phenomena and give rise to con-sequences at that scale, primarily involvinghuman health and the functioning of the terres-trial biosphere.

Climate change is the most obviousconsequence of human-driven changes to


Globalpopulation

Peo

ple

(bill

ion

)

0

0 0

2

4

6Real GDP

1990

Do

llars

(1012

)

0

15

30

45

Atmospheric CO2

CO

2 (p

pm

v)

280

300

320

340

360

380

Northern hemisphere

temperature

Tem

p. a

no

mal

y ( o

C )

–0.5

0

0.5

1.0

Fertilizer use

Nu

trie

nts

(108

ton

s–1)

0

1

2

3

Great floods

Flo

od

s d

ecad

e–1

0.01

0.02

0.03

0.04

0.05

Land conversion

% o

f lan

d a

rea

0

25

50

% o

f fis

her

ies

0

20

40

60

80

100

Fisheries exploited

Species extinctions

0

# (1

03 )

10

20

30

Year

Globalized (foreign)

investment200

400

600Water use

(100

0 km

)3 yr–1

2

4

6

Paperconsumption

100

Ton

s (m

illio

n y

r–1)

200

0

18001750

18501900

19502000

18001750

18501900

19502000

18001750

18501900

19502000

18001750

18501900

19502000

18001750

18501900

19502000

18001750

18501900

19502000

Human Actions

Earth–System Responses

1998

do

llars

(bill

ion

)

Figure 14.3. Increasing rates of change in humanactions and Earth System responses since thebeginning of the Industrial Revolution. Significantincreases in rates of change occur around the 1950s

in each case and illustrate how the past 50 years havebeen a period of dramatic and unprecedented changein human history. Adapted from Steffen et al. (2004).

atmospheric composition. The most recentassessment of the science of climate change(IPCC 2007a) concluded that the warming ofthe global climate system is unequivocal, asshown by the rise in global mean temperature,the widespread melting of snow and ice, rising

sea levels, and other changes in the Earth Sys-tem. In addition to rising temperature, changesin other aspects of climate have been observedat the scale of continents and ocean basins –wind patterns, precipitation, ocean salinity, seaice, ice sheets, and aspects of extreme weather.


The Intergovernmental Panel on ClimateChange (IPCC) has concluded that increases inanthropogenic greenhouse gas emissions havecaused most of the observed increase in globalmean temperature since the mid-twentiethcentury. Although not a focus of attention inthe cautious and consensus-based findings ofthe IPCC reports, many experts now believethat abrupt climate change – taking such formsas the disintegration of ice sheets or the shut-ting down of the thermohaline circulation andoccurring in periods of a few years or at most adecade or two – is a distinct risk.

The projections for climate change for thetwenty-first century present a daunting chal-lenge to the stewardship of Earth as a social–ecological system. The most recent estimatessuggest a rise of global mean temperature by2100, compared to preindustrial temperatures,of at least 1.5◦C and possibly up to 6◦C orslightly more. A global debate has begun onwhat constitutes “dangerous climate change,”with some suggesting that 2.0◦C should be anupper limit, beyond which damage to natural,social, and economic systems will be unaccept-ably high (Stern 2007). Given that the cur-rent rate of increase of greenhouse gas emis-sions is tracking at the upper limit of the pro-jected range (Raupach et al. 2007) and thatboth global mean temperature and sea-levelrise are also tracking at the upper limit of theirprojected ranges (Rahmstorf et al. 2007), thegoal of limiting the temperature rise to 2◦C orless seems unattainable without very large andrapid reductions in greenhouse gas emissions.The ways in which societies around the worldtackle the climate-change challenge over thenext decade will be a turning point for the evo-lution of the Earth System over the next cen-tury and beyond.

Cumulative changes are just as importantand widespread as systemic ones, but perhapsnot as widely publicized. For example, thehuman imprint on Earth’s land surface is stag-gering, with extensive human domination oflarge areas of the planet. It is estimated thatabout 50% of the ice-free land surface has beenconverted or extensively modified for humanuse. As for the remaining 50%, most has beenmodified to some extent for human use; virtu-

ally no land area outside of the most extremeenvironments has been left untouched (Ellisand Ramankutty 2008). Using another mea-sure, between 5 and 50% of the net primaryproduction of the terrestrial biosphere – the netamount of carbon assimilated by the biospherefrom the atmosphere – is used, co-opted, ordiverted from its natural metabolic pathway byhuman activities (Vitousek et al. 1997).

The oceans are now also significantly modi-fied by humans, although not as extensively asthe land surface. The human imprint is mostclearly seen on fisheries, with 47–50% of fish-eries for which sufficient information is avail-able being fully exploited, 15–18% overex-ploited, and 9–10% depleted (see Chapter 10).In a way analogous to human appropriation ofnet primary production on land, it has been esti-mated that humans harvest about 8% of thetotal primary production of the oceans, withmuch higher percentages for the continentalshelves and regions of major upwelling.

The most rapidly changing component ofthe biophysical Earth System is probably thecoastal zone, with about 50% of the humanpopulation now living within 100 km of a coast-line, a percentage that will likely continue torise through this century (see Chapters 11and 13). The human imprint on the coastalenvironment has traditionally been focused onurban areas and port facilities, where the geo-morphology of the coast has been extensivelymodified. The last several decades, however,have seen much more extensive modification ofthe coastline, especially in the tropics, drivenby the demands of a globalizing food system.About 50% of mangrove forests globally havenow been converted to other uses of directbenefit to humans (e.g., for the production ofprawns; see Chapter 2).

Although climate change has focused muchof the public’s attention on global environmen-tal changes, the accelerating loss of biologicaldiversity may be just as serious, or perhaps evenmore so, in terms of the long-term function-ing of the Earth System. The current rate ofextinctions is at least 100 times the backgroundrate, or perhaps even 1000 times greater (MEA2005d). This suggests that the Earth is in themidst of its sixth great extinction event, the first


caused by a biotic force – Homo sapiens. Theestimated future decline in the species richnessof mammals, fish, birds, amphibians, and rep-tiles – 20% of birds are threatened with extinc-tion, 39% of mammals and fish, 26% of reptiles,and 30% of amphibians – is serious enough,but masks the fact that many species that donot become extinct will have greatly reducedranges or will have become ecologically extinct,that is, be so few in number that they cannotcarry out the ecological functions that they pre-viously performed and that are essential in theprovision of ecosystem services for human well-being.

Water resources are central to well-being,and their management has been the definingcharacteristic of many past and present civi-lizations. It is therefore not surprising that thehydrological cycle is one of the most perva-sively modified of the great material cycles ofthe planet (see Chapter 9; Gleick et al. 2006).At present, about 40% of the total global runoffto the ocean is intercepted by large dams anddiverted to human uses. In the northern hemi-sphere, only about 23% of the flow in 139 ofthe largest rivers is unaffected by reservoirs.As the availability of surface water becomesincreasingly limited in many parts of the world,societies are turning toward groundwater as aresource. Although much less is known aboutthe rate of extraction of groundwater, it hasalready become clear in many areas, particu-larly arid and semiarid areas, that groundwateris being extracted much faster than reservoirscan recharge, a situation equivalent to the min-ing of water resources rather than their sustain-able use.

In summary, the advent of humans as a globalgeophysical force has led to dramatic changesin many features of the Earth System’s func-tioning in a remarkably short period of time ascompared with the natural rhythms of this sys-tem. Changes in the physical part of the EarthSystem include the rapid warming of the cli-mate system, the diminution of snow and icecover, and the rising of sea levels. Changes inthe chemical part of the Earth System fromthe impact of the human enterprise include theacidification of the oceans due to increasing dis-solution of atmospheric CO2 and the increas-

ing acidification of soils and landscapes due tothe application and deposition of reactive nitro-gen compounds. Biological aspects of the EarthSystem have also been extensively modified byhuman action, in general trending toward ahomogenized and simplified biosphere with asyet unknown implications for the functioning ofthe Earth System.

Global Social Changes (GSCs)

The preceding paragraphs focus mainly on thebiophysical aspects of the Earth System, withthe human components of the system appear-ing only as drivers and targets of change. Fornearly all of the Anthropocene, there has beena large conceptual disconnect between the bur-geoning human enterprise and its imprint onthe planet. We have typically assumed that theEarth System is large and robust enough to con-tinue indefinitely to provide an accommodatingand pleasant life-support system for humans, nomatter what the nature and size of the humanenterprise. We have also assumed, at leasttacitly, that environmental impacts of humanactions arise only at local and regional scalesand can be ameliorated as societies becomewealthier.

The realization that human activities canimpact significantly the functioning of the EarthSystem as a whole has come only in the lastdecade or two. The Antarctic ozone hole andclimate change have been the main triggers forthis growing realization, with the 1987 Mon-treal Protocol and the 1997 Kyoto Protocolconstituting preliminary attempts at creatingglobal institutions to deal with planetary envi-ronmental problems. Thus, the last decade orso has seen the first real recognition of theEarth as a complex social–ecological system,although the development of this system beganwith the advent of industrialization about twocenturies ago.

Much like GECs, global social changes occurin both systemic and cumulative forms. Undermost circumstances, the changes that occupyhuman attention are social rather than envi-ronmental in nature. A number of these socialchanges have far-reaching consequences for the


Earth System. But for the most part, they areunintended byproducts of social changes; theyare seldom taken into account in efforts to mea-sure social welfare at the national level, muchless at the global level.

A variety of systemic social changes are com-monly lumped together under the heading ofglobalization. But this concept encompasses ahost of distinct processes that are worth differ-entiating in this discussion (Held et al. 1999).Global social change involves the rise of tradeand monetary systems that are planetary inscope. Exports as a percent of gross domesticproduct (GDP) in developed countries (in con-stant prices) rose from 8.3% in 1950 to 23.1% in1985 (Held et al. 1999). Today, this figure for allthe countries of the world has risen to over 40%(World Bank 2004). Among other things, thishas given rise to multinational corporations thatnow rival or exceed many nation states in termsof their global influence and economic size. Thegrowth in foreign direct investment (FDI) hasbeen equally dramatic during this period. FDInow stands at 15–20% of GDP (World Bank2004). Private investment that crosses nationalboundaries has now eclipsed official develop-ment assistance to become a major economicforce at the global level.

Two consequences of these economic devel-opments stand out. These elements of glob-alization affect everyone, even those living insmall communities that are now tied to globalmarkets for energy, wood products, agriculturalcommodities, and so on. At the local level,this leads to asymmetrical dependence in whichglobal economic developments can have dra-matic impacts on communities that are pow-erless to influence global trends. In addition,monetary crises, such as the panic of 1997, canripple through the entire system overnight. Inthe absence of well-developed mechanisms tocounter such phenomena, monetary crises caneasily lead to self-reinforcing processes.

The scope of globalization extends to techno-logical developments as well as matters of tradeand money. Information systems, based initiallyon the fax machine but now mainly on the com-puter and its links to the Internet, have revo-lutionized our ability to communicate rapidlyand extensively with others located anywhere

on the planet. By 2005, over half of the peo-ple living in developed regions were users of theInternet; the number of users is growing rapidlyin developing countries as well. New technolo-gies emerge rapidly and spread immediately atthe global level, despite the efforts of individualinventers and companies to control this processthrough the use of patents enforceable by law(Stiglitz 2006).

Along side these economic and technologi-cal changes are several social and institutionalchanges that are likely to have equally largeimpacts on human–environment interactions ata planetary scale. Although the accuracy oftreating international society as a society ofstates has always been subject to challenge,major changes at this level are now well under-way (Keene 2002). It is not that the state islikely to fade away as a major player at the plan-etary level any time soon. But nonstate actorsof various kinds, including NGOs and multina-tional corporations, are increasingly powerfulplayers at the global level. A suite of devel-opments that deserves the label “global civilsociety” has altered the organizational land-scape in ways that have major implicationsfor Earth System governance (Wapner 1997,Kaldor 2003, Keane 2003). It is no longer suffi-cient to focus on the actions of states in thinkingboth about the causes of stress on planetary life-support systems and about the range of strate-gies available for coming to terms with thesestresses.

Cumulative social changes are also striking.They include population pressures, industrial-ization, changing consumer preferences, urban-ization, and political decentralization. Eco-nomic growth and industrialization have beenoccurring in China since the late 1970s at asustained pace never before seen in the indus-trial age. India is not far behind. Betweenthem, China and India contain more that athird of the world’s human population. Peopleliving in these countries aspire to lead moreaffluent lives and, so far at least, this is tak-ing the form of a rising demand for individ-ual residences, private automobiles, and the fullrange of consumer durables that constitute aprominent feature of contemporary life in theOrganization for Economic Cooperation and


Development (OECD) countries. It is appar-ent that the potential environmental impactsof these cumulative social changes are enor-mous. With respect to natural resources andecosystem services, it would take two to threeplanets to bring everyone on Earth up to cur-rent living standards in Europe and five planetsto emulate living standards in North America(www.footprintnetwork.org)

These developments have fueled both socialprocesses like urbanization and political pro-cesses like decentralization intended to pro-vide enhanced authority to those living outsidedominant urban centers. Lured by the prospectof jobs, people continue to stream into cities,despite urban living conditions that are oftensqualid. City dwellers now constitute approx-imately 50% of the Earth’s human popula-tion, and this figure is destined to continueto rise during the course of this century (seeChapter 13). Chinese planners expect that 75%of China’s population will reside in urban areasby 2050.

Legal and political decentralization is drivenpartly by a legitimate interest in putting author-ity into the hands of those who understandbiophysical systems best but also in part bya desire on the part of central governmentto shed responsibility for environmental prob-lems without providing regional or local gov-ernments with the resources needed to han-dle such tasks effectively. The result is a situa-tion in which it is often easy for unscrupulousindividuals to exploit social–ecological systemsto their own advantage. The remarkable sizeof the informal economy in many parts of theworld is a product, in part, of the actions of cor-rupt officials and private citizens who engagein illegal trade in such products as drugs inColumbia, animal parts in Africa, wood prod-ucts in Indonesia, and so on. (see Chapter 7;Bardhan and Mookerjee 2006).

Environmental–Social Interactions

Interactions between global environmentalchanges and global social changes are ris-ing rapidly and already leading to profoundimpacts on major ecosystems. There is every

reason to believe that this trend will continue inthe absence of a concerted effort at the globallevel to regulate it. A few examples will sufficeto make this proposition more concrete.

In response to the emergence of a globalmarket together with pressures to embraceexport-led growth, many communities in Cen-tral America have cleared sizable areas of for-est and established coffee plantations. But themarket for coffee is both global and notoriouslyvolatile (Bates 1997). Changes in world marketprices, over which local communities have lit-tle control, can affect their welfare dramatically.While it is illogical at the macro-level, it is nothard to understand why individual communi-ties tend to respond by clearing more forest andreplacing it with coffee plantations in an effortto maintain an adequate flow of income and for-eign exchange on which they now depend.

The government of Brazil, motivated largelyby a desire to enhance security along withthe central government’s control over outly-ing regions, built the Trans-Amazonian High-way during the 1960s and 1970s. Whatever itssuccess in terms of security, the construction ofthe highway opened large areas of the AmazonBasin to individual settlers and to corporationsdesiring to clear the forest in order to estab-lish ranches to supply meat to fast-food restau-rants in North America and, to a lesser extent,Europe. The environmental impacts have beenpredictable (see Chapter 7). Settlers graduallyclear larger areas of forest; the ranchers moveon to clear new areas in the forest wheneverthe land initially cleared loses its fertility. Thedestruction of the forest has severe environ-mental impacts both in terms of the releaseof carbon stored in the trees of tropical forestand in terms of the loss of habitat for endemicspecies (see Chapters 6 and 7).

The growth of trade on a global basis can bebeneficial, especially to those seeking to pro-mote export-led economic growth. But it canprove costly in terms of damage to the environ-ment and threats to the human health. Invasivespecies, spread around the world as an unin-tended byproduct of the growth of trade, havenow emerged as one of the two or three mostimportant causes of the loss of biological diver-sity (see Chapter 2). The influx of Eurasian


milfoil, for instance, has brought about dra-matic changes in many North American lakesand ponds. Trade coupled with the rapid growthin long-distance travel on the part of humanshas opened up the prospect of disease vec-tors that are global in scope. Fears regard-ing the prospect of a rapid and uncontrollableglobal spread of AIDS, SARS, and bird flumay be somewhat exaggerated. Yet it is hard todeny that globalization has brought with it thepotential for the development of disease vec-tors whose impacts could dwarf those of the1918–1919 epidemic of influenza (Kolata 2000).

Now that the Earth as a whole has emergedas a social–ecological system, we must directattention to global environmental changes,global social changes, and their interactionsor, to put it in simpler terms, adjust our con-ceptual lens to view the Earth as a cou-pled social–ecological system (Walker and Salt2006). This is especially important to the extentthat we are concerned about sustainable devel-opment as a component of environmental pro-tection (WCED 1987). Large-scale biophysicalchanges, such as substantial shifts in Earth’s cli-mate system, raise profound questions aboutthe future viability of contemporary humanlifestyles. Yet many human beings are still toopreoccupied with efforts to achieve food secu-rity or to cope with chronic shortages of fresh-water to focus on issues like climate change(see Chapter 3). Others who are beginning toexperience the benefits of industrialization wishto emulate the lifestyles of affluent westerners,regardless of the consequences of their actionsfor the Earth System.

Common Threads

The systemic and cumulative directionalchanges we have noted seem quite disparate, atleast on the surface. What is the link betweenforces leading to the destruction of forests inAmazonia and the potential spread of SARSor bird flu from China to the rest of the world?How should we think about the links betweenindustrialization and the loss of biologicaldiversity?

Even in biophysical terms, the links betweenand among these directional changes are oftensubstantial and sometimes surprising. Thereare obvious connections between the climatechange and the progressive loss of biologicaldiversity; both are tightly coupled with theforces leading settlers and ranchers to clear for-est lands in the Amazon Basin. Similar remarksapply to the environmental impacts of the cre-ation of coffee plantations in Central America.

Here, we identify and explore briefly severalkey features of the directional changes identi-fied in the preceding section that pose prob-lems for finding ways to govern these processesthrough national and international measures.Directional changes at the planetary level posechallenges for governance that are unprece-dented in a number of respects and are unlikelyto yield to familiar tools in our governancetoolkit.

To begin with, the impacts of human actionsat a planetary scale now extend far beyond any-thing we have experienced before (Vitouseket al. 1997). As a result, it is essential tothink about coupled social–ecological dynam-ics rather than treating planetary processes asbiophysical systems that are occasionally per-turbed by human actions. In this context, theone thing we can hope to change directly isthe behavior of human beings. We may desireto lower atmospheric CO2 concentrations or toimprove efficiency in the use of water for irri-gation. But all such efforts require changes inhuman behavior.

Uncertainty looms large in any effort tounderstand, much less to forecast, the dynam-ics of the Earth System (Wilson 2002). Thecomplex and dynamic nature of the planet asa social–ecological system can lead to changesthat are nonlinear, abrupt, irreversible, andsometimes nasty, at least from a human per-spective. Simulations with Earth System mod-els show that slight changes in initial conditionscan lead to profound differences in outcomesover time. It is worth differentiating first-orderand second-order uncertainty in this connection(see Chapter 5). We do not know what atmo-spheric concentrations of greenhouse gases arecompatible with limiting temperature increasesto 2◦C. With regard to the loss of biological


diversity, on the other hand, we are uncertainnot only about current extinction rates but alsoabout the total number of species existent onthe planet. Answering questions about the pro-portion of species likely to go extinct duringa given period of time is therefore an orderof magnitude more difficult than addressingthe effects of increases in surface temperatures.It follows that we cannot wait to take actionregarding large-scale changes in the Earth Sys-tem until we are certain about both the changesand the probable consequences of our delib-erate interventions. This would only become arecipe for paralysis in the face of profound chal-lenges to human welfare. Yet our skills at deci-sion making under uncertainty remain limited(Tversky and Kahneman 1974, Kahneman andTversky 1979).

A third common thread has to do with asym-metries in rates of change in social–ecologicalterms and in our capacity to develop effectivegovernance systems, especially at a global scale.Even at the national level, governance systemsare hard to adapt or adjust to major changesin the demand for governance. We generallyassume that problems will evolve slowly enoughto give us time to overcome problems arisingfrom governance systems that are dominated byspecial interests. But we now face a dilemma inthese terms. Global environmental changes andglobal social changes are occurring at an accel-erated pace. As a result, fish stocks collapsebefore we can agree on ways to reduce har-vest levels; habitat destruction occurs before weare able to muster agreement on the need forprotection; uses of the atmosphere as a reposi-tory for various types of waste rise faster thanwe can address the sources of such develop-ments at the level of policy. What is needed,under the circumstances, are governance sys-tems that can produce substantial changes in atimely manner, without falling prey to varioustypes of corruption.

We are also poorly prepared for situationsin which large systems change at a pace thatequals or exceeds rates of change in small sys-tems (Young et al. 2006). We understand thatlocal weather patterns will change rapidly andeven dramatically, but we expect the Earth’sclimate system to remain unchanged. We are

not surprised to experience large swings in theeconomies of local communities, but we expectthe economy of the Earth System to remain sta-ble. We anticipate fluctuations in human popu-lation at the local level, but we are poorly pre-pared to address rapid demographic changes atthe national level, much less at the global level.Global environmental and social changes havenow called these expectations into question.Financial crises that spread like wildfire havealready occurred. Demographic transitions thatwill pose profound questions for public policyare already underway. Abrupt climate changeis a distinct possibility. As a result, we can nolonger assume that large-scale systems are sta-ble enough to justify focusing attention exclu-sively on small-scale reform.

Taken together, these common threadsdemonstrate a need for new systems of gover-nance for sustainable development. Once weshift our paradigmatic perspective to recog-nize that the Earth is a complex and dynamicsocial–ecological system, it is clear that weneed to think in terms of coupled systemsin which change is large-scale, often nonlin-ear, frequently fast, and sometimes irreversible.From a human perspective, the consequences ofthis sort of change may prove not only unfa-miliar but also nasty. Still, introducing newand effective governance systems at a planetaryscale is a daunting task. It will certainly loomlarge as one of the great issues of the twenty-first century (see Chapter 15; Young 2002b,Biermann 2007, Young et al. 2008).

The Challenge of Earth SystemGovernance

Governance, in every setting, centers on thedevelopment and operation of mechanismsdesigned to steer societies away from bad out-comes (e.g., the tragedy of the commons) andtoward good outcomes (e.g., sustainable useof ecosystem services). As we move deeperinto the Anthropocene, an era in which humanactions are major determinants of the con-dition of biophysical systems on a plane-tary scale, environmental governance becomes


increasingly a matter of limiting or constrainingdisruptive impacts of anthropogenic forces onplanetary life-support systems rather than sim-ply a matter of achieving efficiency and equityin human uses of natural resources.

Successful governance in a world featuringboth global environmental changes and globalsocial changes requires the development ofarrangements that are capable of dealing withcross-level interactions and finding effectivemeans of coping with uncertainty.

Local developments often generate effectsthat are significant at a global scale (e.g., thecontributions of deforestation in specific placesto concentrations of carbon dioxide in theEarth’s atmosphere). Conversely, global devel-opments (e.g., climate change) can producesevere impacts on social welfare at a local level.It follows that we need various forms of mul-tilevel governance or, in other words, distinctarrangements operating a different levels ofsocial organization that interact in a mutuallyreinforcing or synergistic manner to provideeffective Earth System governance. The idea ofsubsidiarity, based on the assumption that thereis a proper level at which to address specificproblems, will not suffice to solve Earth Systemproblems.

Because the problems at stake are highlycomplex – featuring nonlinear and often abruptchanges – the governance systems we create toaddress them must be able simultaneously toarrive at decisions in situations under condi-tions of severe uncertainty and to respond inan adaptive manner as new information relat-ing to problems like climate change becomesavailable. This puts a premium not only on thedevelopment of early warning systems but alsoon the capacity to assimilate new informationand to overcome rigidities giving rise to pathdependence when it becomes clear that priordecisions or policies are outmoded.

How can governance systems address spe-cific problems (e.g., climate change, loss of bio-logical diversity) in settings of this sort? Threeapproaches can prove helpful in turning gen-eral objectives like sustainable developmentinto well-defined and operational goals in spe-cific cases. One approach centers on translatinggeneral goals (e.g., avoiding dangerous inter-

ference with the Earth’s climate system) intoclear-cut measures to be used at the operationallevel. Efforts to set specific goals relating to cli-mate change (e.g., no more than 450 ppm ofCO2 in the Earth’s atmosphere or no more thana temperature increase of 2◦C) exemplify thisapproach. So also does the setting of goals call-ing for measurable changes in some dependentvariable (e.g., the effort to halve the numberof people without safe drinking water by 2015as spelled out in the UN Millennium Develop-ment Goals). As these examples suggest, how-ever, this approach has weaknesses as well asstrengths. What reason do we have for setting450 ppm as a cap on the concentration of CO2 inthe Earth’s atmosphere? Is the number of peo-ple lacking safe drinking water a decision vari-able that we can hope to pursue actively andeffectively as a matter of public policy?

A second way of articulating the goals ofgovernance in operational terms focuses onthe development of safeguards to prevent orcontrol runaway processes like abrupt disinte-gration of ice sheets or financial panics. Thisapproach calls for an enhanced interest instrengthening adaptive capacity with regardto climate change and devising countercycli-cal mechanisms to prevent various kinds ofdestructive spirals made possible by globaliza-tion. Some may regard this as an overly pes-simistic strategy. But it is worth noting that weadopt approaches of this sort all the time atthe domestic level by creating mechanisms toprevent escalating financial crises or by empha-sizing prevention and preparedness as well asresponse with regard to extreme events likehurricanes or tsunamis.

Going a step further, there may be a con-vincing case in some situations for adopting aform of worst-case analysis in thinking aboutthe Earth as a social–ecological system. We usethis type of thinking in the realm of nationalsecurity as a matter of course. We focus on thecapabilities of potential opponents without con-sidering their goals or motives, and we toleratethe expenditure of hundreds of billions of dol-lars per year to increase our sense of security,without any serious debate about the wisdomof doing so. Why not take a similar approachto an issue like climate change where most


experts believe an investment of something like$100–200 billion in the near future could offsetmuch larger costs brought on over a period ofyears or decades by the intensification of cli-mate change (Stern 2007)? It seems clear thatthis is more a matter of mindsets or discoursesthan a lack of the capability needed to act vigor-ously to reduce emissions of greenhouse gases,at least among the advanced industrial states inthe world today.

Applications – Climateand the MDGs

Because we are dealing with large, complex sys-tems that are subject to nonlinear and some-times abrupt changes and that are typically sen-sitive to initial conditions, there is little prospectof devising simple recipes for addressing thedemand for governance across the full range ofEarth System issues. What we can say is thatsuccessful governance systems require carefulattention to matching the properties of thesocial–ecological systems in question with theattributes of the governance systems created tomanage them (Young 2002b, Galaz et al. 2008).

We do know that governance systems oper-ating on a global scale must be able to moni-tor the dynamics of planetary processes closely,provide early warning regarding the approachof tipping points or thresholds, and respondto such developments with appropriate adjust-ments in a timely manner. Fulfilling theserequirements with respect to specific problemswill require close attention to the special fea-tures of individual issues. Addressing climatechange, for instance, requires actions that dif-fer from those required to stem the loss of bio-logical diversity. To explore the implications ofthese observations concretely, we turn in thissection to brief case studies of efforts (1) toaddress the problem of climate change and (2)to meet the Millennium Development Goals.Climate change constitutes a systemic problemthat requires an unprecedented response at aglobal scale. Efforts to address this problemhave produced meager results so far. Fulfillingthe Millennium Development Goals, by con-

trast, is a matter of tackling a cumulative prob-lem that requires action in spatially-defined set-tings but that has planetary-scale consequencesin the aggregate. At this stage, the results ofefforts to address this problem are mixed.

Controlling Climate Change

Climate change is arguably the most complexgovernance issue that humanity has faced. Itaffects all countries (but not in the same way);it has a very long timeframe, and it has thepotential to threaten the viability of contem-porary civilization. The first serious attempt atcontrolling climate change was the 1992 UnitedNations Framework Convention on ClimateChange (UNFCCC), which entered into forcein 1994 and which nearly all countries have rat-ified. The critical component of the UNFCCCis Article 2, which states the ultimate objec-tive of the convention as stabilizing “. . . green-house gas concentrations in the atmosphere ata level that prevents dangerous anthropogenicinterference with the climate system.” The 1997Kyoto Protocol, the first attempt to translatethis general goal into an explicit approach toreducing greenhouse gas emissions by assign-ing targets and timetables, became embroiled ininternational politics and has not achieved uni-versal coverage. Although the protocol finallyentered into force in 2005, the USA has refusedto ratify it. The Kyoto Protocol and other effortsoutside it to reduce greenhouse gas emissionshave had no measurable effect, as can be seenin the observed record of greenhouse gas emis-sions through 2006 (Fig. 14.4; Raupach et al.2007). The emissions trajectory is tracking on abusiness-as-usual line, well above all concentra-tion stabilization trajectories.

There are several severe challenges thatmake it difficult to come to terms with climatechange from the perspective of Earth Systemgovernance and that global society has failed sofar to meet:

• Equity issues. The climate change challengeis bedeviled by equity issues of several types.First, most of the increased greenhouse gasesin the atmosphere have been emitted by theindustrialized world, yet the consequences


Historic emissions

Emissionsgap

Projected emissions (A1)

To stabilize at 650 ppm



1850 1900 1950 2000 2050 21000

2

4

6

8

10

12

14

16

18

20

Foss

il fu

el e

mis

sio

n (G

t C

yr–1

)

Year

Figure 14.4. Observed global CO2 emissions compared with emission scenarios and stabilization trajectories.Redrawn from Raupach et al. (2007).

of climate change will be borne dispropor-tionately by the developing world. This isexacerbated by the generally lower adap-tive capacity in developing countries. Sec-ond, although there is much focus now onChina and India as future larger emitters,past emissions have been dominated by theUSA and Europe. There is a tendency tofocus so much on the rapidly developingAsian giants that the historical legacy is over-looked. Finally, countries tend to quantifyemission rates in ways that favor their ownpositions. For example, Australia, with only20 million inhabitants, argues that its emis-sions are a very small fraction of the globaltotal and are already dwarfed by China’semissions. But Australia’s per capita emis-sions vie with those of the US as the high-est on the planet, much higher than China’s.This raises a fundamental question. Shouldemissions permits in any global scheme beallocated to countries or to individuals onthe premise that every human being has anequal right to use the “atmospheric sink”?This question is one of the most contro-versial in the climate change debate. Untilit is resolved, little progress will be madein achieving a globally acceptable emissionsreduction scheme.

• Long lag times. The biophysical componentof the climate system has exceptionally longlag times that challenge our ability to deviseappropriately “fitted” governance systems.The momentum already built into the cli-mate system, due largely to the thermal iner-tia in the oceans, implies that we are com-mitted to another 0.5◦C or 0.6◦C of globalmean temperature rise, regardless of the suc-cess of efforts to reduce future emissions.Thus, the nature of the Earth’s climate in2030 can already be predicted with a highdegree of certainty (in terms of temperature;Plate 8; IPCC 2007a); no matter how strin-gent measures we take in the next decadeor two, it will make little difference by 2030.What government can convince its electorateto make large sacrifices now with no per-ceptible effect for 20 years? There are evenlonger lags to deal with. If temperature risesto 2.5 or 3.0◦C above the preindustrial level,it is highly likely that most of the Green-land ice sheet and parts of the West Antarc-tic ice sheet will be committed to melting,although it may take a few centuries for thisto play out. This would lead to a sea-levelrise of 6–7 m. Once the “tipping point” iscrossed, there is no going back even if we


could reduce greenhouse gas emissions tozero overnight.

• Dangerous climate change. The UNFCCCmakes reference to dangerous interferencein the climate system but gives little guidanceas to what constitutes “dangerous climatechange.” Although scientific research on thenature of climate impacts can inform thedebate, deciding what is really “dangerous”in terms of climate change is an individualand societal value judgment that could varymarkedly around the world. The inhabitantsof low-lying Pacific Island states, for instance,have already decided that the current levelof climate change (a global mean tempera-ture rise of about 0.7◦C) is dangerous; theyare preparing to leave their countries inlarge numbers for New Zealand. On theother hand, Russia may benefit from furtherwarming. Agriculture could move north-wards; the country’s vast boreal forests couldexpand into the present tundra regions, andlarge deposits of oil and gas in the Arc-tic Ocean could become accessible as thesea ice retreats. A large number of inter-pretations of “dangerous climate change” liebetween these two extremes. But the point isthat societal judgments of what is dangerousand what is not will vary widely around theworld, adding another layer of complexity tothe task of building a global governance sys-tem to control climate change.

• Nonlinearities in the climate system. Thedynamics of the climate system are highlynonlinear in many aspects. In that regard,CO2 behaves as a “threshold gas,” a smallincremental change (an increase of a fewparts per million) could tip the climate sys-tem into a cascade of feedbacks that couldpropel the Earth into another climatic andenvironmental state (see Chapter 5). Sucha rapid and irreversible (in human time-frames) transformation could be triggered,for example, by large-scale feedbacks in thecarbon cycle – large outgassing of CO2 fromthe soil, increases in wildfires in the borealand tropical zones, activation of methaneclathrates buried under the coastal seas,and the rapid loss of methane from tundraecosystems as they warm. This raises the

issue of overshoot. The pathway to a givenstabilization target may be as important asthe ultimate target itself. If atmospheric con-centration rises too high during a particu-lar stabilization trajectory, this carbon cyclethreshold may be crossed and the naturalfeedbacks described above may be activated,rendering the original target unattainable. Ina worst-case (but by no means impossible)scenario, such a “runaway” effect could leadto the collapse of modern civilization (i.e., anuncontrollable decline in economic activity,population, social cohesion, and individualand societal well-being). How can currentdecision-making structures deal with sucha momentous issue? What type of gover-nance/institutional system can be devised to“manage” our own life-support system forthe benefit of humanity as a whole?

There is a growing urgency to find workablesolutions to these institutional challenges. Manysocieties have faced momentous environmen-tal challenges of various types. Some could notchange in innovative and adaptive ways and col-lapsed; others found new approaches to dealwith the challenge and transformed themselvesinto flourishing and resilient societies (Tainter1988, Diamond 2005). What does this implyfor contemporary, globalized society and theclimate change challenge? The complexity ofthis particular environmental challenge dwarfsthose of the past and, ultimately, implies that wemust become effective and successful stewardsof our own life-support system, which we havenow transformed into the largest and most com-plex social–ecological system possible. To fail atthis challenge has implications for an increas-ingly connected global society that are hard toimagine.

Because climate change is a systemic issue,avoiding dangerous interference in the climatesystem will require coordinated human actionson a planetary scale. But the fact that themaintenance of a viable climate system alsoexhibits the basic features of a public good (seeChapter 4) means that many actors in the sys-tem will experience strong incentives to becomefree riders, leading to suboptimality or evenoutright failure in efforts to supply this good.


Experience in other realms suggests that thebest prospect for overcoming this problem maybe to start with a coalition of key players and tobuild out from there to draw in free riders step-by-step (Schelling 1978). This is what happenedin the case of ozone depletion. Of course, thetwo problems differ in a number of significantways, but it seems likely that addressing climatechange will require a similar process.

Finally, climate change, although it nowdominates the media and the political arena,is not the only global environmental changethat threatens the well-being of humanity andthe stability of the Earth System. The Mil-lennium Ecosystem Assessment (2005a) high-lighted many other global-scale changes thataffect the ecosystem services on which human-ity ultimately depends. Most of these issues, ifnot all of them, are not as well developed interms of ongoing scientific assessments, inter-national negotiations, and global organizationsas is climate change. Nevertheless, ensuring thatthe Earth System continues to provide the suiteof essential ecosystem services on which wedepend is just as important as stabilizing the cli-mate system in a state within which we can sur-vive and prosper.

Fulfilling the MillenniumDevelopment Goals

The millennium development goals (MDGs)are a set of eight broad policy targets intendedto improve the lives of people in developingcountries (Table 14.1). The problem here iscumulative in the sense that it requires actionin many specific locations but at the same timeglobal because overall success or failure in ful-filling the MDGs will have planetary conse-quences. Adopted initially by all the memberstates of the UN in the Millennium Declara-tion of 2000, the commitment to the MDGs wasreinforced at both the 2002 Monterrey Confer-ence on Financing for Development and the2002 World Summit on Sustainable Develop-ment in Johannesburg. Individual goals cover avariety of issues ranging from the alleviation ofpoverty to the improvement of health and the

adjustment of trade and financial arrangementsto facilitate efforts on the part of developingcountries to improve their lot.

Most of the MDGs call for specific reduc-tions in well-defined conditions (e.g., the num-ber of people living on less than a dollar a day,the incidence of HIV/AIDS, malaria, and otherdiseases) with a target date of 2015. It is pos-sible in a number of cases to track progresstoward meeting these goals in a quantitativemanner. Goal #1, for instance, calls for halv-ing “. . . between 1990 and 2015, the propor-tion of people whose income is less than $1 aday.” Goal #7 envisions halving “. . . by 2015,the proportion of the population without sus-tainable access to safe drinking water and basicsanitation.” Goal #8 speaks to the need to“. . . develop further an open, rule-based, pre-dictable, nondiscriminatory trading, and finan-cial system.”

Have we made progress toward meetingthese goals and, in the process, addressing thesecumulative social–ecological concerns throughinitiatives launched at the global level? Aprogress report released by the UN in July2007 concludes that “[h]alf way to a 2015deadline, there has been clear progress towardimplementing the Millennium DevelopmentGoals . . . But their overall success is still farfrom assured . . .” (UN Press Release, 2 July2007b). Let us take a closer look at progressrelating to Goals #1, 7, and 8 and delve into theissues of governance on a planetary scale thatthey pose. The results are illuminating in termsof the insights they provide regarding issues ofgovernance on a large scale.

There is relatively good news regardingpoverty. The number of people in developing

Table 14.1. UN Millennium Development Goals.Information from United Nations (2007a).

1. Eradicate extreme poverty and hunger2. Achieve universal primary education3. Promote gender equality and empower women4. Reduce child mortality5. Improve maternal health6. Combat HIV/AIDS, malaria, and other diseases7. Ensure environmental sustainability8. Develop a global partnership for development


countries living on less than a $1 a day fellfrom 1.25 billion in 1990 to 980 million in 2004or in other words from nearly a third living inextreme poverty “. . . to 19 percent over thisperiod” (UN 2007a: 6–7). As the UN reportpoints out, however, progress is unequallyshared. Most of the progress occurred in Eastand South Asia, while “. . . poverty rates inWestern Asia more than doubled between 1990and 2005” (UN 2007a: 7). With respect to safedrinking water and basic sanitation, progresshas been made but much more to be done tomeet Goal #7. Thus, “. . . if trends since 1990continue, the world is likely to miss the targetby almost 600 million people” (UN 2007a: 25).Progress toward meeting Goal #8 (fair trade)is harder to track in any quantitative way. Yetthere are reasons for concern in this area. Bythe middle of 2007, the Doha Round of tradenegotiations had failed to reach agreement “. . .

on the overall programme of measures to beadopted” (UN 2007a: 20). Most donor coun-tries have failed to meet the targets for officialdevelopment assistance (ODA) set at the 2002Monterrey Conference. ODA actually fell from2005 to 2006, and “. . . aid to the least developedcountries (LDCs) has essentially stalled since2003” (UN 2007a: 29).

What should we make of this mixed recordregarding fulfillment of the MDGs and espe-cially of the lackluster performance in severalkey areas. The evidence suggests that the maindrivers in this domain include both internal andexternal factors (Collier 2007). Internally, theexistence of good governance appears to be acritical condition for success. This is not, at leastin the first instance, a matter of conforming tosome ideal standard of democracy. More impor-tant with respect to issues like sanitation andpoverty are conditions like the absence of civilwar and limitations on the ability of exploita-tive individuals to find ways to feather their ownnests and, more often than not, manage to holdtheir assets in offshore accounts.

Externally, key factors are the rules of thegame and the willingness of wealthy coun-tries to provide properly targeted and well-managed ODA. It is generally acknowledged,for instance, that the agricultural polices of theEU and the US make it difficult for farm-

ers in developing countries to compete in themarkets of these advanced industrial countries.And there is no reason to expect that these poli-cies will change in any fundamental way, even ifthe Doha Round of trade negotiations producesagreement on some significant issues.

The story regarding ODA is equally trou-bling. Using development assistance to makeprogress toward broader economic, political,and social goals is always tricky; things can andoften do go wrong in the provision of devel-opment assistance. Still, we cannot help beingconcerned by trends regarding the availabilityand use of ODA. The fact that most wealthycountries have failed to make good on the com-mitments they agreed to in 2002 at the Monter-rey Conference is telling. Overall, ODA todayis less than half of the target of 0.70% of GDPfor the wealthy countries as a group. In someof these countries, willingness to provide devel-opment assistance is actually waning. A particu-larly serious concern is the failure to make goodon the pledge to “. . . double aid to Africa by2010 at the summit of the Group of eight indus-trialized nations in Gleneagles in 2005” (UN2007: 29).

What lessons can we draw from this casestudy that are relevant to the challenge of EarthSystem governance? First, and in some respectsforemost, is the limited capacity of a large num-ber of developing countries to participate ina meaningful fashion in addressing planetaryissues like climate change. Leaders of countriespreoccupied with problems of poverty, disease,and a lack of safe drinking water have little timeand energy to consider systemic problems likeclimate change. It is not that they do not careabout climate and fail to grasp the seriousnessof this problem for social welfare in their owncountries. The sad fact is that they do not havethe resources needed to deal with all their prob-lems at the same time and that issues like cli-mate change seem like concerns that can be putoff, at least in the short run (see Chapter 3).

What is to be done under these conditions?Several strategies, implemented simultaneouslyseem promising as answers to this question. Thefirst objective is to provide people right downto the local level with incentives to take large-scale problems seriously. A striking example


involves the rise of tourism in Southern Africaand the incentives associated with this devel-opment to treat wild elephants as a source ofincome rather than as pests. The fact that thisexample relates to biological diversity ratherthan to climate change is not accidental. It maywell be easier to generate interest at the locallevel in taking steps that pertain to issues, likemanaging elephant populations, that seem tan-gible and close to home.

Beyond this, it obviously helps to have exter-nal sources of funding available on favorableterms. The evidence is strong that the Mon-treal Protocol Multilateral Fund played a sig-nificant role in drawing developing countriesinto the effort to protect stratospheric ozone.Even more to the point are the efforts of theGlobal Environment Facility (GEF), a multi-lateral funding mechanism administered jointlyby UNEP, UNDP, and the World Bank anddedicated to providing funds needed to helpdeveloping countries address large-scale envi-ronmental problems like climate change. Tar-geted on issues like climate change, biologicaldiversity, and freshwater, the GEF has emergedas a particularly important source of funds inlight of the failure of most developed countriesto live up to the financial commitments theymade in Monterrey in 2002 and later that sameyear in Johannesburg (Stiglitz 2002).

Addressing cumulative issues, like thosespelled out in the MDGs, differ significantlyfrom efforts relating to systemic issues. It is tobe expected that initiatives aimed at reducingpoverty or providing safe drinking water will befar more effective in some countries or regionsthan in others (Collier 2007). Changes in therules of the game, like adjustments designedto help developing countries to export agricul-tural products, will be more important to somedeveloping countries than others. The capac-ity of individual developing countries to absorband make good use of development assistancevaries dramatically. As a result, any search forgeneral or universal prescriptions in this issuearea is bound to fail (see Chapter 4). This meansthat we are unlikely to find ourselves seekingto build comprehensive regimes, like the globalregime dealing with climate change, when itcomes to battling poverty or improving health

care in the developing countries. Even so, fulfill-ing the MDGs must be treated as a high priorityin any effort to protect the planet’s life-supportsystems.


There is now an inescapable demand for effec-tive governance at the planetary level or, inother words, Earth System governance. Boththe biophysical and the social parts of the EarthSystem are capable of rapid and irreversiblechanges that could be detrimental to the well-being of humans and the viability of mod-ern societies. Interactions between global envi-ronmental changes and global social changesmay trigger events that are nonlinear, abrupt,irreversible, and nasty, at least from a humanperspective. To minimize the risk that humanactions will trigger such events, we need todevelop a capacity to manage major compo-nents of the Earth System in such a way as tokeep close to the bounds of natural variabil-ity. This presents unprecedented challenges tohumanity to conceptualize the Earth as a com-plex social–ecological system and to move frombeing clumsy exploiters to sensitive stewards ofour own life-support system. This is the funda-mental challenge of the twenty-first century.

How can we prepare to meet this challengein an effective manner? We cannot succeed inefforts to address these problems by focusing onbiophysical systems and treating human actionsas perturbations or minor disturbances to beset aside or ignored in most cases. Understand-ing the resultant social–ecological systems willrequire the development of a new generationof models and methods capable of capturing thedynamics of coupled systems in which feedbackloops linking the biophysical components andthe human components are central concerns.Governance systems treated as steering mecha-nisms can play an important role in maintainingthe resilience of these social–ecological systems.But they interact with numerous other driversand must be designed to fit the distinctive prop-erties of the social–ecological systems in ques-tion in order to prove effective.


What recommendations can we offer on thebasis of this analysis that may prove helpful tothose responsible for addressing specific prob-lems of Earth System governance? Here welist and briefly clarify six lessons for policymak-ers who are responsible for addressing bothsystemic and cumulative issues on a planetaryscale:

• Draw on multiple types and sources of knowl-edge. It is important to avoid the pretensethat mainstream, western scientific analy-ses are superior to those arising from otherapproaches to the production of knowledge.The ability to integrate a number of types ofknowledge will emerge as an essential skill(see Chapter 4).

• Pay attention to long-term consequences. Inaddressing issues of the type we have dis-cussed, it is important to avoid the dismissalof long-term consequences arising both fromthe idea of discounting future benefits andcosts and from the dominance of the elec-toral cycle in policymaking. The importanceof considering future needs as well as cur-rent needs is the basic message arising fromthe idea of sustainable development (seeChapters 7 and 9).

• Learn how to cope with uncertainty. In deal-ing with the planetary issues discussed inthis chapter, policymakers will always befaced with the need to make decisions underconditions of uncertainty. This suggests theimportance of thinking in terms of satisfic-ing (i.e., selecting satisfactory options ratherthan holding out indefinitely to find the per-fect option) and making use of heuristicsor rules of thumb. It also provides a strongrationale for erring on the side of cautionin estimating subjective probabilities regard-ing events that could push the Earth Sys-tem across thresholds and into irreversiblechanges (see Chapter 1).

• Create sensitive monitoring systems. In deal-ing with large-scale social–ecological sys-tems, there is a need for sophisticated mon-itoring systems capable of picking up earlysigns of changes and providing informationneeded to respond to changes in a timelymanner. The more dynamic the system and

the higher the level of uncertainty, the moreimportant it is to allocate scarce resources tothe creation and operation of early warningsystems.

• Emphasize social learning as well as adapta-tion management. The idea of adaptive man-agement with its emphasis on approachingpolicymaking with an experimental frame ofmind is a step in the right direction. But indealing with complex and dynamic social–ecological systems, it is important to go a stepfurther to focus on social learning or, in otherwords, to consider adjusting goals or refram-ing cognitive models as well as attending toinstrumental matters like the pros and consof alternative means to pursue existing goals(see Chapters 4 and 5).

• Prepare for crises as periods of opportunity.Crises are, of course, dangerous situations inwhich large systems can cross thresholds andmove into different – and often less desirable– basins of attraction. But they are also peri-ods of opportunity during which it is possi-ble to make important changes in prevailinggovernance systems. Because there is nevertime to engage in extensive analysis duringa crisis, it is highly desirable to think aboutthe relative merits of different approachesto institutional reform before a crisis erupts(see Chapter 5).

Review Questions

1. What developments during the second halfof the twentieth century and the first yearsof the present century have made it appro-priate to treat the Earth System as a wholeas a social–ecological system?

2. What is the significance of introducing theconcept of the Anthropocene to describe thecurrent era in the earth’s history?

3. In what ways do global environmentalchanges and global social changes interactwith one another to form a coupled plane-tary system?

4. How seriously should we take the prospectof nonlinear, abrupt, and irreversiblechanges in the Earth System?


5. What is the difference between systemicchanges and cumulative changes, and whydoes this distinction matter?

6. How should we think about sustainabledevelopment as a criterion for evaluating theresults of Earth System governance?

7. How can we overcome the obstacles thatimpede our efforts to strengthen the climateregime?

8. Why are we making more progress in meet-ing MDG #1 than MDG #7?

9. How do we need to adjust our approach togovernance to address problems arising invery large and highly dynamic systems likethe Earth System?

Additional Readings

Biermann, F. 2007. Earth system governance as acrosscutting theme of global change research.Global Environmental Change 17:326–337.

Costanza, R., L.J. Graumlich, and W. Steffen, editors.2007. Sustainability or Collapse? An IntegratedHistory and Future of People on Earth. MIT Press,Cambridge, MA.

Diamond, J. 2005. Collapse: How Societies Choose toFail or Succeed. Viking, New York.

Gunderson, L.H. and C.S. Holling, editors. 2002.Panarchy: Understanding Transformations inHuman and Natural Systems. Island Press,Washington.

McNeill, J. 2000. Something New under the Sun: AnEnvironmental History of the Twentieth Century.Penguin Press, London.

Schellnhuber, H.-J., P.J. Crutzen, W.C. Clark, andM. Claussen, editors. 2004. Earth System Analysisfor Sustainability. MIT Press, Cambridge, MA.

Steffen, W.L., A. Sanderson, P.D. Tyson, J. Jager,P.A. Matson, et al. 2004. Global Change and theEarth System: A Planet under Pressure. Springer-Verlag, New York.

Tainter, J. 1988. The Collapse of Complex Societies.Cambridge University Press, Cambridge.


Young, O.R. 2002. The Institutional Dimensions ofEnvironmental Change: Fit, Interplay, and Scale.MIT Press, Cambridge, MA.

Young, O.R., L.A. King, and H. Schroeder, editors.2009. Institutions and Environmental Change:Principal Findings, Applications, and ResearchFrontiers. MIT Press, Cambridge.

Part IIIIntegration and

Synthesis

15Resilience-Based Stewardship:Strategies for Navigating SustainablePathways in a Changing World

F. Stuart Chapin, III, Gary P. Kofinas, Carl Folke, Stephen R. Carpenter, PerOlsson, Nick Abel, Reinette Biggs, Rosamond L. Naylor, Evelyn Pinkerton,D. Mark Stafford Smith, Will Steffen, Brian Walker, and Oran R. Young

Introduction

Accelerated global changes in climate, environ-ment, and social–ecological systems demand atransformation in human perceptions of ourplace in nature and patterns of resource use.The biology and culture of Homo sapiensevolved for about 95% of our species’ his-tory in hunting-and-gathering societies beforethe emergence of settled agriculture. We havelived in complex societies for about 3%, and inindustrial societies using fossil fuels for about0.1% of our history. The pace of cultural evo-lution, including governance arrangements andresource-use patterns, appears insufficient toadjust to the rate and magnitude of technolog-ical innovations, human population increases,and environmental impacts that have occurred.Many of these changes are accelerating, caus-ing unsustainable exploitation of ecosystems,including many boreal and tropical forests, dry-lands, and marine fisheries. The net effect hasbeen serious degradation of the planet’s life-

F.S. Chapin (�)Institute of Arctic Biology, University of AlaskaFairbanks, Fairbanks, AK 99775, USAe-mail: [email protected]

support system on which societal developmentultimately depends (see Chapters 2 and 14).

Efforts to redirect exploitation and fostersustainability have led to a gradual shift inresource management paradigms that often fol-low a transition from an intensive resourceexploitation phase to a steady-state resource man-agement paradigm aimed at maximum or opti-mum sustained yield (MSY or OSY) of a sin-gle resource, such as fish or trees, and sub-sequently to ecosystem management to sustaina broader suite of interdependent ecosystemservices in their historic condition (Fig. 15.1).Despite its sustainability goal, management forOSY often results in overexploitation because ofoverly optimistic assumptions about the capac-ity of resource managers to sustain productiv-ity, avoid disturbance and pest outbreaks, regu-late harvesters’ actions, and anticipate surprisessuch as extreme economic or environmentalevents (see Chapters 1, 4, and 5). Actions toaddress emerging problems are often delayeduntil research can provide a more completeunderstanding of likely system response. Evenecosystem management, which is widely viewedas the state-of-the-art management paradigmfor sustainability, is constrained by its focuson historic conditions as a reference pointfor management and conservation planning.



High

Low

Pre -industrial

use

Exploitation

Steady-state resourcemanagement (MSY)

Ecosystem management

Ecosystemstewardship

TimeBoreal Canada (Ch. 6)

Tropical Indonesia (Ch. 7)

Open ocean (Ch. 10)

NW U.S. (Ch. 7)

Sust

ain

abili

ty

S. Sweden(Ch. 5)

Figure 15.1. Evolution ofresource-managementparadigms and their focuson the balance betweenexploitation andsustainability. Arrows atthe bottom show thehistory (up to the present)in representative locationsand the chapter (Ch) inwhich each is described. Amajor opportunity fordeveloping nations is to“leap-frog” frompre-industrial orexploitative phasesdirectly to ecosystemstewardship.

Given the challenges of sustaining social–ecological systems in a rapidly changing world,we advocate a shift to ecosystem stewardship(Table 15.1). The central goal of ecosystemstewardship is to sustain the capacity of ecosys-tems to provide services that benefit soci-ety by sustaining or enhancing the integrityand diversity of ecosystems as well as theadaptive capacity and well-being of society.This requires adaptive governance of coupledsocial–ecological systems to provide flexibilityto respond to extreme events and unexpected

changes. Rather than managing resource stocksand condition, ecosystem stewardship empha-sizes adaptively managing critical slow variablesand feedbacks that determine future trajecto-ries of ecosystem dynamics (see Chapters 2, 4,and 5). Actions that foster a diversity of futureoptions rather than a single presumed optimumprovide resilience in the face of an unknown,but rapidly changing future. In this perspec-tive, uncertainty and change become expectedfeatures of ecosystem stewardship rather thanimpediments to management actions. This shift

Table 15.1. Differences between steady-state resource management and ecosystem stewardship.

Characteristic Steady-state resource management Ecosystem stewardship

Reference point Historic condition Trajectory of changeCentral goal Ecological integrity Social–ecological sustainability benefitsPredominant approach Manage resource stocks and condition Manage stabilizing and amplifying

feedbacksRole of uncertainty Research reduces uncertainty before taking

actionActions maximize flexibility to adapt to an

uncertain futureRole of resource

managerDecision maker who sets course for sustainable

managementFacilitator who engages stakeholder groups

to respond to and shape social-ecologicalchange and nurture resilience

Response to disturbance Minimize disturbance probability and impacts Adapt to changes and sustain optionsResources of primary

concernSpecies composition and ecosystem structure Biodiversity, well-being, and adaptive

capacity

15 Resilience-Based Stewardship 321

in paradigm builds on a wealth of knowledgeand experience of many professionals, includ-ing resource managers and users, policy mak-ers, and business leaders and integrates con-cepts from natural and social sciences and thehumanities.

In Chapters 1–5, we presented a frame-work linking studies in vulnerability analy-sis, resilience theory, and social transforma-tion to address ecosystem stewardship in asocial–ecological context. We emphasized thekey roles of adaptive co-management and adap-tive governance in fostering both ecosystemsustainability and human well-being. We thenillustrated in Chapters 6–14 the applicationof these ideas in specific social–ecological sys-tems. In this chapter, we summarize prac-tical approaches to implementing ecosystemstewardship through the integration of threebroad sustainability strategies: (1) reducing vul-nerability; (2) fostering adaptive capacity andresilience; and (3) navigating transformations toavoid, or allow escape from, undesirable social–ecological states. These three strategies areoverlapping and complementary. They requiremore proactive and flexible approaches to defin-ing the future state of the planet than havecharacterized either exploitative or equilibrialresource management paradigms of the past.

Reducing Vulnerability

Reduce Exposure to Hazardsand Stresses

Vulnerability analysis entails assessing and min-imizing hazards, stresses, and risks and reduc-ing the sensitivity of social–ecological systemsto those threats that cannot be adequatelymitigated (Table 15.2). It is a logical start-ing point for implementing ecosystem steward-ship, because it involves planning in the con-text of known current conditions and expectedchanges. Although vulnerability analysis tra-ditionally emphasizes human vulnerability, wefocus on the vulnerability of social–ecologicalsystems in addressing ecosystem stewardship.

Assessment and mitigation of currently rec-ognized hazards, stresses, and risks have always

Table 15.2. Examples of stewardship strategies toreduce vulnerability.

Reduce exposure to hazards and stresses• Minimize known stresses and avoid or minimize

novel hazards and stresses• Develop institutions that minimize stresses

originating beyond the managed unit• Manage in the context of projected changes rather

than historical range of variabilityReduce social–ecological sensitivities to adverse impacts

• Sustain the capacity of natural capital to providemultiple ecosystem services

• Sustain and enhance critical components of well-being, particularly of vulnerable segments of society

• Engage stakeholders in decision-making to accountfor variation in norms and values in assessingtrade-offs

• Plan sustainable development to address thetrade-offs among costs and benefits for ecosystemsand multiple segments of society

been a fundamental goal of sound resourcemanagement and are routinely applied to man-aging individual stresses such as drought, over-grazing, pollution of freshwaters, and pestoutbreaks (see Chapters 8–10). By monitor-ing trends in indicators of stresses and theirecological impacts, resource users can gaugechanges from some historic reference pointand take appropriate actions to reduce thestress. For example, overgrazing in drylandsreduces the abundance of palatable grasses rel-ative to unpalatable grasses or shrubs, indicat-ing the need to reduce grazing pressure (seeChapter 8). Similarly, a decline in the size ortrophic level of marine fish can be a sensi-tive indicator of overfishing (see Chapter 10).Social–ecological impacts of stresses are some-times masked by ecosystem or social feed-backs, such as phosphorus sequestration in lakesediments (minimizing changes in water col-umn phosphorus concentration; see Chapter 9)or perverse subsidies that motivate fisher-men to increase fishing effort despite stockdeclines (minimizing declines in total catch; seeChapter 10). Such feedbacks make it valuableto identify multiple ecosystem and social indi-cators that are sensitive to initial phases ofdegradation (see Chapters 2 and 3), requiringknowledge and understanding of ecosystemdynamics and social–ecological interactions.


Mitigating novel stresses is an increasingchallenge, because many of these stresses,such as climate change, acid rain, internationalfishing pressure, and global demand for bio-fuels, reflect global-scale economic, informa-tional, and cultural linkages (see Chapters 4,12, and 14). Mitigating global-scale stressesrequires concerted global action, for which cur-rent governance mechanisms are inadequate(see Chapters 4 and 14). Nonetheless, effectivecollaboration among even a few key nationscan accomplish a large proportion of thedesired global mitigation, yielding dispropor-tionately large global benefits (see Chapter 14).Global consensus is not required to make hugeadvances in the stewardship of our planet.

Given that many stresses are likely tocontinue or intensify, trajectories of expectedchange often provide more realistic manage-ment targets than do historical ranges of vari-ability. Projections of climatic and demographicchanges, while uncertain, provide an increas-ingly fine-scale framework for future planning(see Chapters 1 and 14). In cities, for exam-ple, planting trees and protecting green spacesameliorate local impacts of increasingly fre-quent heat waves in urban heat islands, whileproviding other health and social benefits (seeChapter 13). Similarly, countries with trajec-tories of rapid population growth and ruralpoverty benefit from aid that empowers peopleto enhance the capacity for local food produc-tion (see Chapters 3 and 12) and at the sametime avoid poverty traps (see Chapter 5). Thiscontrasts with many current international aidprograms that address immediate food needsin ways that undermine the capacity of localfarmers to sell their crops or otherwise developlivelihoods that are sustainable within the con-straints of locally available ecosystem services(see Chapter 12).

Reduce Sensitivity to Adverse Impactsof Hazards and Stresses

The current dynamics of most social–ecologicalsystems result in part from a long coevolution-ary history that has shaped governance and thecapacity of people and other organisms to cope

with and adapt to a wide range of hazards andstresses. An important starting point for mini-mizing social–ecological sensitivity is thereforeto sustain natural and social capital and liveli-hood opportunities and to understand adapta-tions to both the historical range of stressesand plausible future changes. In this context,resource management has traditionally empha-sized the importance of protecting and sustain-ing the resources (e.g., water and soils) andorganisms (e.g., population sizes of major func-tional groups) to provide multiple ecosystemservices (Table 15.2; see Chapter 2). Similarly,cultural heritage and traditional economies areoften important in sustaining livelihoods inhuman communities, although these economiesmay also reflect lack of access to more favorableopportunities, as in the former Apartheid sys-tem of South Africa (see Chapter 3). These gen-eral strategies of sustaining natural and socialcapital are at the core of steady-state ecosystemmanagement and policy formulation. Nonethe-less, achieving workable solutions is often dif-ficult because of trade-offs among ecosystemservices or societal outcomes and differencesof opinion among stakeholders in evaluat-ing these trade-offs (see Chapter 4). Becauseissues and trade-offs vary temporally and spa-tially, the application of general principles mustbe sensitive to local context. Thus, althoughthe general guidelines are clear, their imple-mentation commonly sparks conflicts that canonly be addressed through stakeholder engage-ment in the decision-making process (seeChapter 4).

When social–ecological conditions change ornew hazards or stresses are imposed, the mech-anisms by which these systems coped with his-torical hazards and stresses may no longer suf-fice, for example, when confronted with newweapons, agricultural technologies, and glob-alized markets. In some places, the break-down of traditional governance and resource-use systems combined with ready access toweapons cause violent local and regional con-flicts that require fundamental solutions (seeChapter 8). These commonly involve renego-tiation of resource access rights and access tonew livelihood opportunities. In other cases,innovation may provide opportunities, as when


chronic food shortages are addressed throughmore productive crop varieties or new agricul-tural practices (see Chapters 8 and 12). Inno-vations that are intended to reduce sensitivityto stress must be implemented cautiously, how-ever, because they can create new vulnerabil-ities, such as susceptibility to crop diseases orincreased dependence on a cash economy tobuy expensive fertilizers or imported food atvariable prices.

Policies that are intended to reduce sensi-tivity to stresses sometimes backfire becauseof an associated loss of resilience. For exam-ple, ecosystem management that is intended toreduce the sensitivity of timber economies topests or fire can reduce the variability in thesestresses—often leading to conditions to whichmany local organisms or social processes, par-ticularly those that are important in postdistur-bance reorganization and recovery, are less welladapted. When patterns of variability are nolonger compatible with current (often chang-ing) conditions, the whole system becomes vul-nerable to large-scale change (see Chapters 1and 5).

Within any social–ecological system, certainsegments of society are particularly vulnerableto specific stresses and hazards. Targeted inter-ventions that reduce the sensitivity of vulnera-ble segments of society are likely to be partic-ularly effective in reducing net social impactsof shocks and stresses. For example, protec-tion of the access rights of coastal fishermencan reduce vulnerability of both the fisher-men and the fish stocks on which they depend(see Chapter 11). Similarly, effective systems ofinformation exchange via the internet or tele-phone hotlines between urban residents, socialnetworks, and city government can reduce therisks and sensitivity of urban slum dwellersto emerging environmental hazards, therebyreducing costs to society when a shock to thesystem occurs (see Chapter 13).

Careful analysis of current and projectedhazards, sensitivities, and their interactions formultiple segments of society provides an excel-lent starting point for developing targets andpathways for sustainable development. Vulner-ability analysis allows an assessment of trade-offs among costs and benefits for ecosystems

and various segments of society. The 1998South African water policy, for example, estab-lished water reserves that guarantee accessto domestic water for all segments of societyand provided flows needed to sustain ecosys-tem integrity at times of drought. Thus domes-tic and ecosystem water rights took prece-dence over irrigated agriculture and industry(see Chapters 8 and 9). Such policies begin toaddress issues of long-term social–ecologicalresilience, discussed in the next section, as wellas current vulnerabilities.

Enhancing Adaptive Capacityfor Social-Ecological Resilience

At times of rapid directional social–ecologicalchange, the historical state of the system maybe poorly suited for adaptation to new condi-tions. Traditional informal property rights, forexample, that may have been locally recognizedand effective for centuries may be transferredfrom local users to newcomers who lack theknowledge or the ethical foundations to useecosystem services sustainably. Key organismsmay decline in response to novel disturbances,landscape structure, and pollutant levels. Peo-ple may be disadvantaged in a more globalizedeconomy or fail to cope with rising populationpressures and shrinking resource supply. Peo-ple and other organisms have always adaptedto new opportunities through changes in theiractivities, ways of life, and locations (see Chap-ter 3). However, the novel character and rapidrate of recent changes challenge the capacity ofsocial–ecological systems to adapt. The EarthSystem is moving into novel terrain.

Resilience-based ecosystem stewardshipshifts the philosophy of resource manage-ment from reactions to observed changes toproactive policies that shape change for sus-tainability, while preparing for the unexpected.This paradigm shift is essential in a worldundergoing rapid and directional change. Acentral approach to both reducing vulnerabilityand enhancing resilience is to enhance theadaptive capacity of social–ecological systemsto both expected changes and unanticipated


surprises. We describe four approaches toenhancing resilience by fostering (1) a diversityof options; (2) a balance between stabilizingfeedbacks and creative renewal; (3) sociallearning and innovation; and (4) the capacityto adapt, communicate, and implement solu-tions (Table 15.3). These approaches provideecosystem stewards with opportunities tothink creatively about ways to sustain sys-tem attributes that society deems importantand potential future pathways to sustain andenhance these attributes.

Foster Biological, Economic,and Cultural Diversity

Diversity, whether it is cultural, biological, eco-nomic, or institutional, is important because itincreases the number of building blocks avail-able to respond to and shape change. Diver-sity also broadens the range of conditions underwhich the system can function effectively (see

Chapter 1). A region whose economy dependsentirely on one extractive industry, for example,is poorly buffered against market fluctuationsor technological innovations that might reducethe value of that product. Similarly, low biologi-cal diversity constrains the capacity of ecosys-tems to adjust to large or rapid changes withsubsequent losses of ecosystem services (seeChapter 2).

Although biodiversity warrants protectioneverywhere because of its functional impor-tance and current high rates of loss (seeChapter 2), certain areas could contributeuniquely to planetary stewardship. For exam-ple, global hotspots of biodiversity or speciesof high cultural or iconic value (e.g., elephants,pandas, and polar bears) are likely to engen-der strong public support; areas with high topo-graphic diversity and intact migration corridorsenable species to migrate readily in responseto rapid environmental change (see Chapter 2).Similarly, retention of corridors such as urbangreen spaces, hedgerows, and riparian corridors

Table 15.3. Examples of stewardship strategies to enhance social-ecological resilience.

Foster biological, economic, and cultural diversity• Prioritize conservation of biodiversity hotspots and locations and pathways that enable species to adjust to rapid

environmental change• Retain genetic and species diversity that are underrepresented in today’s landscapes• Exercise extreme caution when considering assisted migration• Renew the functional diversity of degraded systems• Foster conditions that sustain cultural connections to the land and sea• Foster retention of stories that illustrate past patterns of adaptation to change• Subsidize innovations that foster economic novelty and diversity

Foster a mix of stabilizing feedbacks and creative renewal• Foster stabilizing feedbacks that sustain natural and social capital• Allow disturbances that permit the system to adjust to changes in underlying controls• Exercise extreme caution in experiments that perturb a system larger than the jurisdiction of management

Foster social learning through experimentation and innovation• Broaden the problem definition by learning from multiple cultural and disciplinary perspectives and facilitating

dialogue and knowledge co-production by multiple groups of stakeholders• Use scenarios and simulations to explore consequences of alternative policy options• Test understanding through experimentation and adaptive co-management• Explore system dynamics through synthesis of broad comparisons of multiple management regimes applied in

different environmental and cultural contexts

Adapt governance to changing conditions• Provide an environment for leadership to emerge and trust to develop• Specify rights through formal and informal institutions that recognize needs for communities to pursue livelihoods

and well-being• Foster social networking that bridges communication and accountability among existing organizations• Permit sufficient overlap in responsibility among organizations to allow redundancy in policy implementation


and retention of mobile species that link thesehabitats, such as pollinators, birds, and con-cerned gardeners, sustain diversity in human-dominated landscapes where people are mostlikely to lose their sense of connection tonature (see Chapter 2). Involvement of localresidents in planning and implementing conser-vation efforts in lands that support their ownlivelihoods increases the likelihood that policieswill be respected (see Chapter 6).

Areas that have already lost most of theirbiodiversity (e.g., production forests; mining,stream, or wetland restoration projects; or over-fished coastal systems; see Chapters 2, 7, 9,and 11) are unlikely to return to their formerstate, particularly in the context of rapid envi-ronmental change. Renewal of ecosystem ser-vices in highly modified ecosystems may occurmore readily by developing functional diversitythat is consistent with likely future conditionsand societal goals than by trying to rebuild thehistorical predisturbance species composition.This redefines strategies of restoration ecologyin a context of change.

The last several centuries have substantiallyreduced the cultural diversity of the planet asnation states, both internally and as colonialpowers, sought to eliminate alternate cultures,ideologies, and modes of governance, especiallywhere these involved claims to ownership anduse of resources that differed from the claimsof the state. The rise of nation states in Europeduring the 17th and 18th centuries, for example,involved a deliberate effort to reduce culturaland political complexity, by removing peoplefrom historical relationships to land and water.The modern nation state standardized and thussimplified the formerly diverse and complexproperty rights and boundaries so that land-scapes and seascapes could be mapped, admin-istered, and taxed efficiently. Administrationwas concerned with extracting the maximumvalue from financially valuable resources andremoving traditional uses of these or relatedculturally important resources (see Chapter 3).Similarly, in many colonial situations, indige-nous people were forcibly removed from landsand waters they had previously controlled, sothat the colonizers could lay claim to the vastmajority of resources on these lands. In these

cases, the elimination of the institutions andhuman relationships that were intrinsic to thecultural diversity was a logical part of exploit-ing resources.

More recently, during the mid-20th century,cultural assimilation was advocated globally asa way to rapidly improve opportunities for dis-advantaged groups, for example, through edu-cation using a standardized curriculum in asingle national language. These policies oftenfurther undermined cultural integrity throughloss of language, local institutions, and cul-tural ties to the land and sea (see Chapter 3).Similarly, conservation plans that exclude peo-ple from their traditional homelands or dis-rupt rural economies, such as ranching, canstimulate poaching or land sales to developersthat undermine the original policy intent (seeChapters 6–8, 11). Cultural diversity is valu-able because it provides a diversity of knowl-edge systems, perspectives, and experience onways to meet mutually agreed-upon goals in theface of large uncertain changes (see Chapter 4).On the other hand, cultural differences in pref-erences and belief systems can be sources offriction that are often essential to an equitableevaluation of trade-offs.

Under conditions of rapid change, somecomponents of cultural, biological, and eco-nomic diversity are likely to be altered or lost.It is therefore important to identify, protect,and legitimize latent sources of diversity thatmay be underrepresented in current systemdynamics. This can serve as insurance againstexcessive loss of current diversity components.For example, traditional crop varieties thathave been replaced by higher-yielding hybridor genetically engineered varieties assume anew importance as sources of genetic diver-sity to address the challenges of novel condi-tions and rapid spread and evolution of croppests (see Chapter 12). Elders in most soci-eties remember ways of doing things under arange of circumstances that enrich the optionsthat are potentially available to address cur-rent and future changes. Many of these manage-ment practices reflect ecological understanding,such as indigenous burning in Australia andthe USA, which reduced fuel loads, increasedthe proportion of food plants for both peo-


ple and grazers, and reduced risk of large fires(see Chapter 6). As climate warming and acentury of fire suppression increase risks oflarge fires, traditional indigenous managementstrategies provide alternative options for firemanagers to consider. Other traditions suchas gender inequality in education and votingor rigid property rights can reduce opportu-nities, indicating that some traditional institu-tions may reduce household and communityresilience and ecosystem stewardship as socialcontext changes.

Rapid climate change raises problematicdecisions of whether and how to assist in themigration of long-lived immobile organismslike trees that are unlikely to keep pace withrapid environmental change. Should forestersplant a geographically diverse range of geno-types in forest regeneration projects ratherthan just the locally adapted genotype? Shouldorganisms be transplanted to new, climaticallyfavorable locations (assisted migration) whenthey are likely to go extinct in their cur-rent habitat (see Chapters 2 and 7)? Giventhe checkered history of species introductionsand biocontrol programs, decisions that mod-ify migration corridors or move species to newlocations must be made cautiously and reflecta clear understanding of the trade-offs betweenrisks and opportunities (see Chapter 9).

Like cultural and biological diversity, eco-nomic diversity increases the range of optionsfor adjustment to change. Policies that provideshort-term subsidies and incentives for inno-vation increase the opportunities to adjust tochange (see Chapter 12). Conversely, perversesubsidies that sustain uneconomical practices,such as overfishing of stocks that would other-wise be uneconomical to fish, reduce the capac-ity of social–ecological systems to adjust tochange (see Chapter 10). Economic subsidiesfrequently involve social–ecological trade-offs(e.g., short-term economic hardship to fisher-men vs. long-term viability of a fish stock; via-bility of Scandinavian agriculture in an unfa-vorable economic climate vs. long-term foodsecurity; see Chapters 10–12). Decisions aboutsubsidies often reflect power dynamics amongaffected stakeholders more than their value inecosystem stewardship.

Foster a Mix of Stabilizing Feedbacksand Creative Renewal

In the short term, stabilizing feedbacks tend tosustain the properties of the system and keep itwithin its current state. Under circumstances ofrapid directional change, however, these feed-backs can create a system that is increasinglyout of balance with underlying driving vari-ables. Therefore, a dynamic mix of stabiliz-ing feedbacks and occasional disturbances thatallow adjustments to changing conditions aremost likely to sustain the fundamental proper-ties and dynamics of the system—that is, to con-fer resilience. Managing stabilizing feedbackshas always characterized sound resource man-agement. These include, for example, policiesthat minimize soil erosion and foster vegeta-tion recovery after disturbance (see Chapters 7and 12), market mechanisms that balance sup-ply and demand (see Chapter 12), rangelandproperty rights that encourage pastoralists tocare for the land while reducing resource useconflicts (Chapter 8), and fisheries quota sys-tems that allocate access rights among fisher-men (see Chapters 10 and 11).

Management that allows or fostersdisturbance and renewal is often more con-troversial and may require greater search forcreative solutions. Purchases of conservationeasements that prevent residential devel-opment in rural scenic areas, for example,allow retention of fire as a natural ecologicalprocess in forests (see Chapter 7). Policiesthat allow rural residents to share revenuesfrom big game hunting in African wildlifeparks increase local support for wildlifeconservation and reduce poaching (see Chap-ter 6). Policies that recognize the rights topublic protest may create windows for pol-icy adjustments to changing conditions (seeChapter 13).

Disturbance that creates the opportunity forrenewal along either previous or new trajecto-ries entails both opportunities and risks asso-ciated with new outcomes. Experimentation atsmall scales allows learning to occur so policiescan be adjusted to either favor or reduce thelikelihood of the changes that were observed.Experiments at larger scales than the system


being managed are more dangerous becausethey can release changes that are beyond thecontrol of individual resource managers to pre-vent undesirable outcomes (see Chapter 1).Current “experiments” with the climate, biodi-versity, and cultural diversity of the planet aretherefore of grave concern, including proposedgeoengineering approaches to dealing with cli-mate change, which could lead to unintendedfeedbacks that are equally damaging to theplanetary environment (see Chapter 14).

Foster Innovation and Social Learning

Although diversity provides the raw materialsfor adaptation, innovation and social learningare the core processes that build the adaptivecapacity and resilience of a social–ecologicalsystem. The central roles of innovation andsocial learning in adaptation to changing socialand economic conditions are universally rec-ognized by business. However, they receivesurprisingly little attention by resource man-agers and the public at large. Instead, thereis often an emphasis on reducing variability,preventing change, and maintaining the sta-tus quo. This is no longer a viable manage-ment or policy framework under conditions ofrapid directional social–ecological change. Howcan society shift from a mindset of fearingchange to assessing its value as a way to copewith and realize new opportunities in a rapidlychanging world? Not all changes are construc-tive. As discussed earlier, changes occurringat scales larger than the scale of manage-ment (e.g., the planet) should be approachedcautiously.

An obvious starting point is to broaden theframework of problem definition by integrat-ing a broader range of disciplines, knowledgesystems, and approaches. Resource managers,for example, moved from the management ofsingle species such as tigers, pines, or tuna toecosystem management by acknowledging theimportance of a broader range of ecosystem ser-vices and the key linkages between biophysicaland social processes (see Chapters 1–4). Simi-larly, adaptive co-management of resources byagency managers and resource users providesopportunities to integrate a wealth of scientific

and local understanding to address challeng-ing problems of social–ecological change (seeChapters 4–6).

In a rapidly changing world, however, knowl-edge of how to cope with previous conditions isoften insufficient. Neither is it feasible to post-pone actions until we can observe the perfor-mance of the system in equilibrium with newconditions. Instead, we must learn by doingwithout destroying future options—adaptivelymanaging the global life-support system inwhich society is embedded (see Chapters 4 and8). This requires educational transformationsin both the school system and the workplace,including, as emphasized here, in resource man-agement.

Adaptive management is a critical compo-nent of social learning because it embracesuncertainty and builds social learning into themanagement process. However, it is insufficientin a rapidly changing world because social–ecological systems, like other complex adaptivesystems, are path-dependent, so managementinterventions that proved valuable in one cir-cumstance may have different effects at othertimes and places (see Chapters 1 and 4).

Scenario modeling and analysis providesopportunities to explore those potential futureconditions that cannot be readily predicted,for example, the consequences of new tech-nologies or alternative policy strategies. Thiscan help policy makers, researchers, resourcemanagers, resource users, and the public envi-sion potential futures, assess their fit to societalgoals, and explore potential strategies and path-ways to achieve desired ends (see Chapter 5).For example, scenarios of alternative waterand land management policies in arid areascan broaden the discourse beyond fulfillingcurrent needs to addressing long-term strate-gies that avoid unsustainable development(see Chapter 8). Much can be learned aboutsocial–ecological dynamics and feedbacks fromcomparisons of similar management systemsthat have been applied in different social–ecological settings. Examples include marinereserves established in different marine ecosys-tems, democratic institutions or conservationstrategies employed in different cultural set-tings, and integrated pest management applied


to different agricultural systems. A great dealhas been learned, for example, through com-parative studies about the institutional arrange-ments and circumstances that foster sustainableuse of common pool resources such as water,forests, fish, and rangelands (see Chapters 4 and6–10). Co-management is not always conduciveto flexibility and change. If co-managementarrangements become overly codified and rigid,they constrain opportunities to adjust to change(see Chapter 6). Also, if responsibility is dis-persed and unfocused, the hard choices maynever be made. Leadership is essential in nego-tiating differences, providing vision, and build-ing links between groups and their social net-works at different levels of management andgovernance.

At the global scale, adaptive management,especially “management experiments,” shouldbe applied with extreme caution. There arethresholds or boundaries in the Earth Systemthat would be dangerous to cross or to approachtoo closely. The best-known example is the con-centration of carbon dioxide in the atmosphere.There is a threshold above which reinforcingfeedbacks, such as weakening and then rever-sal of oceanic and terrestrial carbon sinks, couldpush the Earth System into another, muchwarmer state, state that is much less amenablefor human life. At present, the precise locationof this threshold is not well known; estimatesvary between 350 and 600 ppm CO2. Thresholdsalso exist in the chemical and biological compo-nents of the Earth System, but these are gen-erally less studied and much less well knownthan the greenhouse-gas example. Given theexistence of such thresholds and the seriousconsequences for humanity of crossing them,responsible stewardship of the Earth System asa whole requires careful attention to the pre-cautionary principle (see Chapter 14).

Adapt Governance to ChangingConditions

Flexibility in governance structures that candeal with change is critical to long-term social–ecological resilience. Grazing systems in thedrylands (Chapter 8) offer models of how

resource access rules can change according tocircumstances, contrasting with the rigidity ofrules and processes of governments. The con-cept of “rules for changing the rules,” as writ-ten into the constitutions of many countries,is a useful way to bring more flexibility toresource-use rights during this time of height-ening uncertainty. Devolving the powers andresources of government to local scales canalso enhance the responsiveness and adaptabil-ity to change in ways that sustain opportuni-ties instead of constraining options, as long asit comes with the resources needed to navi-gate change and good systems of accountabil-ity. Bridging individuals and organizations suchas NGOs or temporary public advocacy groupsprovide informal communication pathways thatallow dialogue and negotiation to occur out-side the rules and policies of formal institutions.Emergence of Ecuador’s Watershed Trust Fundfrom a constellation of local groups, NGOs,and international aid agencies is an example(Chapter 9).

Decentralized, polycentric governanceresults in some overlap of responsibilities. Thisis analogous to biodiversity in providing redun-dancy in social–ecological functioning. Stateagencies, neighborhood groups, and nationalNGOs, for example, may all support actionsthat protect a certain species or valued habitat.When one of these groups “drops the ball”and fails to provide this governance functionbecause of budget shortfalls or shifting priori-ties, the overlapping activities of other groupscan sustain the basic need (see Chapter 9).Conversely, when national policies usurp thepower of local institutions to protect valuedlocal resources, the polycentric nature of gov-ernance is eroded, and ecosystem stewardshipobjectives are more likely to be threatened (seeChapter 6).

Navigating Transformations

Transformations are fundamental changes insocial-ecological systems that result in dif-ferent control variables defining the state ofthe system, new ways of making a living,and often changes in scales of critical feed-


backs. In the context of ecosystem steward-ship, transformations involve forward-lookingdecisions to convert the system to a funda-mentally different, potentially more beneficialsystem (see Chapter 5). Unintended transfor-mations can also occur in situations where man-agement actions have prevented adjustment ofthe system to changing conditions. Rapid direc-tional changes in the factors that control sys-tem dynamics increase the likelihood that somecritical threshold will be exceeded, so under-standing the actions that increase or decreasethe likelihood of social–ecological transforma-tions is a critical component of resilience-basedecosystem stewardship (Table 15.4).

Preparing for Transformation

The first step in addressing potential transfor-mations (either desirable or not) is to iden-tify plausible alternative states and considerwhether they are more or less desirable than thecurrent state (Table 15.4). Because most trans-formations create both winners and losers, andthe magnitudes of potential gains and lossesare uncertain, stakeholder groups often dis-agree about how serious the problems are andwhether or how to fix them. Important ini-tial steps in preparing for purposeful trans-

formation are mobilizing support for change;engaging stakeholders in identifying and rais-ing awareness of problems (e.g., rigidity trapsor lack of institutional fit); and defining a col-lective vision for the future. Once the vision isdefined, people are more willing to explore andagree on potential pathways to improved situa-tions. This includes identifying knowledge gaps,developing new governance and managementapproaches, identifying barriers to change,and developing strategies to overcome thesebarriers.

Shadow networks often play an impor-tant role in seizing windows of opportunityto make use of abrupt change. They canexplore new approaches and experiment withsocial responses to uncertainty and change andthereby generate innovations that could triggerthe emergence of new forms of governance andmanagement of social–ecological systems (seeChapter 5). An important challenge is to pro-vide space for these networks to form throughenabling legislation and financial, political, andmoral support (see Chapter 5). Such learn-ing platforms can generate a diversity of ideasand solutions that can be drawn upon at crit-ical times. The challenge here is to establishstructures like bridging organizations that allowfor and support ecosystem stewardship fordevelopment.

Table 15.4. Strategies for purposeful navigation of transformations (see Chapter 5 for details).

Preparing for transformation• Engage stakeholders to recognize dysfunctional states and raise awareness of the problem• Identify, recruit, and support potential change agents• Connect nodes of expertise and develop shadow networks of motivated actors• Identify plausible alternative states and pathways• Identify thresholds, potential crises, and windows of opportunity• Identify the barriers to change and prepare strategies to overcome these

Navigating the transition• Use crises or opportunities to initiate change• Maintain flexible strategies for transition• Negotiate the transformation with transparency and active stakeholder participation• Foster structures that facilitate cross-scale and cross-organizational interactions

Building resilience of the new regime• Create incentives and foster values values for stewardship in the new context• Initiate and mobilize social networks of key individuals for problem-solving• Mobilize new knowledge and external funding, when needed• Foster the support of decision makers at other scales• Shape the local context through adaptive co-management


Actively navigated transformations toalternate potentially more desirable statesare frequently observed. These include theestablishment of water management boardsthat guarantee domestic and green water flowsin arid South Africa (see Chapter 9) and ashift from ecosystem management for multipleecosystem services to residential housing in thewildland-urban interface in the northwesternUSA (see Chapter 7). Rigidity traps thatstill exist and require transformational ratherthan incremental solutions include persistentpoverty in sub-Saharan Africa (see Chapter12), failure of global governance to effectivelyaddress climate change (see Chapter 14), andrepeated depletion of the planet’s marine fishstocks by overfishing (see Chapters 10 and 11).Every ecosystem type addressed in this bookshowed the potential to undergo purposefultransformations, although this did not alwaysoccur or sometimes reverted to the originalsystem (see Chapter 6). These observationssuggest that transformations from degradedor dysfunctional states warrant considerationin most social–ecological systems. Indeed,this is the primary motivation of sustainabledevelopment projects undertaken in devel-oping nations and warrants considerationin any ecosystem that confronts persistentsocial–ecological challenges.

Unintended transformations that haveoccurred include shift from slash-and-burnagriculture to intensive agriculture triggeredby population growth and shortened cyclesof forest recovery (see Chapter 12) and shiftfrom production forests to residential housingassociated with rising property values (seeChapter 7). Every ecosystem type addressed inthis book was observed to undergo unintendedtransformations in response to rapid directionalchanges in one or more environmental or socialdrivers (see Chapters 6–14). The nature ofmany of these transformations was highlypredictable from the known sensitivity of thesystem to critical controls, and the movementtoward these state changes could be recognizedfrom observed changes in drivers and knownindicators of the transformation pathway. Whatis often unknown is the time course, abruptness,and sometimes the degree of irreversibility of

the changes—hence the importance of fosteringgeneral resilience and the adaptive capacity toadjust to change. In other cases, the causes andnature of the transformation are unexpected.These observations suggest that unintendedtransformation should be taken as a seriouspossibility in all social–ecological systems andthat resource managers should identify andrespond to likely causes and indicators ofmovement toward transformation.

The properties of thresholds between alter-native social–ecological regimes are poorlyknown and are currently an active area ofresearch. Many thresholds are related to cas-cades through chains of positive feedbacks,such as trophic cascades in aquatic food chains(Chapter 9), spatial cascades of dryland degra-dation (see Chapter 8), fishing down marinefood chains, as upper trophic levels are depleted(see Chapter 10), or agricultural transitions trig-gered by population increases (see Chapter 12).As ecosystems approach important thresholds,they respond more sluggishly to interven-tion, so the ecosystem becomes more diffi-cult to control even as control is more desper-ately necessary. It has lost resilience. Ecosys-tems approaching thresholds may also becomevariable, flickering among alternate states inlocalized patches. Such flickering has beendescribed for drought persistence in drylands.Conceptually, thresholds are expected in sys-tems where human action is causing gradualchange in slowly moving spatially extensivevariables that contribute to strong feedbackcycles with fast-moving variables. For example,gradual loss of habitat may eventually drivelarge predator populations below a thresh-old where they cannot persist, triggering cas-cades of change in lower-level consumers andplants. Preparing a social–ecological system fortransformation therefore entails the capacity todetect early warnings and recognize potentialthresholds.

Navigating the Transition

Preparing for and navigating transformationalchange is challenging because it depends oncircumstances that are specific to each time


and place. Transformational change is mostlikely to occur at times of crisis, when suffi-cient stakeholders agree that the current sys-tem is, by definition, dysfunctional. In such situ-ations, the transformation may take many dif-ferent alternative directions, some desirable,others highly undesirable. It is important thatdiscussions about potential transformations betransparent, objective, and open to all stake-holders so that the process is not co-opted bya particular stakeholder group or agenda. Iden-tifying potential crises that might provide win-dows of opportunity for transformations to pro-mote ecosystem stewardship is a critical step innavigating transformations (see Chapter 5). Forexample, there is increasing recognition thatthe planet is approaching a point of “danger-ous climate change”; this crisis represents anopportunity to implement global governancestructures requiring more aggressive actionsto reduce human impacts on the climate sys-tem (see Chapter 14). Prolonged droughts orcatastrophic wildfires that burn the wildland–urban interface might trigger transformationin fire management and urban development,respectively (see Chapter 7). City infrastruc-ture frequently has a lifetime of 50 years orless, and specific sections of the city may rein-vent themselves even more frequently, provid-ing opportunities to infuse new elements suchas parks, green spaces, or efficient public trans-portation in ways that reshape urban dynam-ics (see Chapter 13). An important lesson fromthese examples is that the nature of crises thatare likely to trigger transformation is oftenobvious ahead of time, providing opportuni-ties to strategize and prepare for windows ofopportunity.

Do we have to experience a crisis beforewe can change? How can we avoid or steeraway from cascading ecological crises, unsus-tainable trajectories, and traps before they hap-pen? There are at least four ways that crisiscan lead to opportunities: (1) Resource stew-ards may actively prepare for change, so tran-sitions happen smoothly; (2) a system may col-lapse locally, which raises awareness of the needfor change; (3) actors may learn from criseshappening in a similar system at other timesor places; and (4) a crisis may happen in other

sectors or at other scales but be seized as anopportunity to make changes. In KristianstadsVattenrike in Sweden, for example, a local eco-nomic crisis coincided with rising national envi-ronmental concern because of pollution-relatedseal deaths along the Swedish coast (a nationalcrisis) to initiate change.

Building Resilience of the New Regime

A successfully navigated transformation is beststabilized by building adaptive capacity andresilience through the processes described ear-lier (Tables 15.3 and 15.4). Because transforma-tions create winners and losers, the resilience ofthe new system may initially be relatively frag-ile. Building resilience in new conditions can bestrengthened by actions that build trust, identifysocial values among players of the new regime,and empower key stakeholders to participatein decisions that legitimize relationships andinteractions of the new regime. Transformationsoften alter the nature of cross-scale interactions,providing both opportunities and challenges.Early attention to these cross-scale interactionsthat ensure good information flows, systems ofaccountability, and sensitivity to differing per-spectives reduces the likelihood of reversionto earlier states or other unfavorable transfor-mations. For example, the rapid transformationfrom rural to urban systems occurring in manyparts of the world may require new systemsof governance and patterns of social–ecologicalstewardship. While these are shaped by histori-cal legacies, the resilience of the new system canbe enhanced by eliminating barriers to coop-eration among agencies and communicating avision of opportunities provided by the trans-formed state throughout the public and privatesectors. This can motivate and entrain actorsand social networks to address new needs andopportunities that inevitably arise with trans-formation. Continuous evaluation and opendiscussion of the associated economic andnoneconomic benefits and costs of change pro-vide a basis for assessing progress in navigat-ing relatively undefined structures and rela-tionships that arise in novel social–ecologicalsituations.


Comparisons of Vulnerability,Adaptive Capacity, andResilience amongSocial-Ecological Systems

Contrasts among Types ofSocial–Ecological Systems

In the following paragraphs, we describekey vulnerabilities and sources of resiliencethat broadly characterize the types of social–ecological systems described in this book, rec-ognizing that regional variation is substantialand that surprises will inevitably occur. Sinceno single type of institutional arrangement isthe best for all situations, this broad social–ecological comparison illustrates the rangeof issues faced by ecosystem stewards andapproaches that have proven useful to addressthe sustainability challenges in particular envi-ronments. Viewed together, the challengesand opportunities across this range of social–ecological systems provide a broader under-standing of ecological stewardship.

Hinterlands (see Chapter 6) are typicallyoccupied by small communities with strong cul-tural ties and informal institutions that link peo-ple to the ecosystems in which they live (Table15.5). Nonresidents, however, often value theseareas for conservation, recreation, or sourcesof resources (e.g., oil and gas) without consid-ering the implications of these uses for hinter-land residents. The integration of livelihoodsinto a biodiversity conservation ethic has beenan integral part of many traditional culturesand can be a component of innovative solutionsto current conservation challenges. Conversely,efforts to isolate parks from local livelihoodstypically create highly artificial, fragile systemsthat have little historical precedent and areunlikely to be resilient to future environmental,economic, and social changes. Consequently,ecosystem stewardship of hinterlands requiresadaptive governance at the local level thatincorporates a range of perspectives on knowl-edge and the implications of change. Efforts tosustain traditional practices, as well as to facili-tate smooth transformations to other forms oflivelihood, must include institutions that pro-

vide strong linkages to regional and global pro-cesses that now shape opportunities in eventhe most remote locations. Resilience of hin-terlands may also benefit from transformationsthat sustain and enhance ecosystem services forurban dwellers, including provisioning serviceslike sustainable foods and cultural services suchrecreational and existence values of wild land-scapes and seascapes.

Forests (see Chapter 7) are dominated byplants that often have life spans lasting mul-tiple human generations. Longevity generatesstability and predictability for human residentsof forests in terms of environment and ecosys-tem services (Table 15.5). However, large, rapidchanges in environment, disturbance regime,and management goals can cause changes thatare irreversible on the timescale of multiplehuman generations. Transformations of particu-lar concern include high rates of tropical defor-estation and temperate suburbanization, whichmay lead to large-scale regional and even EarthSystem feedbacks. These transformations aredriven by global- or regional-scale processesand require intervention at these scales. Thedevelopment and strengthening of institutionswith a long-term view of the ecological andsocial conditions necessary to support forestsand forest-dependent peoples are a foundationfor sustainable forestry.

In contrast to forests, the resilience of dry-lands (see Chapter 8) derives primarily fromeffective adaptation of organisms, traditionalsocieties, and local communities to high envi-ronmental variability. Local knowledge andgovernance arrangements are vitally importantto this resilience at times of rapid environmen-tal and social changes (Table 15.5). Cross-scalelinkages and capacity to engage in diverse mar-kets provide resources at times of crisis. Thesesame characteristics render drylands vulnera-ble to large changes in slow variables, includingprolonged drought and erosion of institutions(e.g., access rights) that allow society to copewith variability, particularly in environmentsof intermediate aridity where both populationpressures and vulnerability to desertificationare high. Natural variability slows down expe-riential learning, making the drylands vulnera-ble to rapid change. Cross-scale interactions can

15 Resilience-Based Stewardship 333Ta

ble

15.5

.Rep

rese

ntat

ive

sour

ceso

fvul

nera

bilit

yan

dre

silie

nce

and

freq

uent

lyob

serv

edtr

ansf

orm

atio

nsof

Ear

th’s

maj

orty

peso

fsoc

ial–

ecol

ogic

alsy

stem

s.W

eco

nsid

eron

lyat

trib

utes

that

char

acte

rize

each

syst

emty

peas

aw

hole

and

igno

rere

gion

alva

riat

ions

wit

hin

syst

ems.

Type

ofso

cial

–eco

logi

cal

Cha

ract

eris

tic

sour

ces

ofC

hara

cter

isti

cso

urce

sof

Pla

usib

leun

desi

rabl

eP

laus

ible

desi

rabl

esy

stem

vuln

erab

ility

resi

lienc

etr

ansf

orm

atio

nstr

ansf

orm

atio

ns

Hin

terl

ands

(Cha

pter

6)W

eak

polit

ical

voic

e;lim

ited

prop

erty

righ

tsto

prot

ect

livel

ihoo

ds

Div

ersi

tyof

know

ledg

esy

stem

san

dec

osys

tem

s;st

rong

soci

al–e

colo

gica

lcou

plin

g

Cul

tura

lass

imila

tion

;los

sof

biod

iver

sity

hots

pots

Em

erge

nce

ofgl

obal

lyco

nnec

ted

loca

lgov

erna

nce

Fore

sts

(Cha

pter

7)R

apid

shif

tsin

fore

stst

ruct

ure;

rapi

dcl

imat

ech

ange

Eco

logi

cals

tabi

lizin

gfe

edba

cks

that

sust

ain

ecos

yste

mse

rvic

es

Def

ores

tati

on;s

prea

dof

hom

esin

tow

ildla

nd–u

rban

inte

rfac

e

Ref

ores

tati

on;f

ocus

edho

me

deve

lopm

ents

that

cons

erve

fore

sts

Dry

land

s(C

hapt

er8)

Ext

ende

ddr

ough

t;ov

ergr

azin

g;di

stan

tvoi

ceA

dapt

atio

nto

tem

pora

land

spat

ialv

aria

bilit

y;se

ason

alm

ovem

ents

;mix

edec

onom

ies

Des

ert;

frag

men

tati

onof

past

oral

land

sth

atpr

even

tse

ason

alm

ovem

ents

Sust

aina

ble

wat

erm

anag

emen

t;no

vel

inst

itut

ions

that

spre

adri

skin

tim

ean

dsp

ace

Fres

hwat

er(C

hapt

er9)

Ove

r-ex

trac

tion

;pol

luti

onw

ith

nutr

ient

sor

toxi

ns;i

nvas

ive

spec

ies

Rip

aria

nve

geta

tion

and

wet

land

s;co

mpl

exaq

uati

cfo

odw

ebs;

effe

ctiv

eco

mm

on-p

rope

rty

inst

itut

ions

No

wat

er;h

eavi

lypo

llute

dsy

stem

s;ex

tirp

atio

nof

long

-liv

edpr

edat

ory

fish;

Nov

elin

stit

utio

nsfo

rw

ater

com

mon

sth

aten

sure

sust

aina

bilit

y

Ope

noc

ean

(Cha

pter

10)

Ove

rfish

ing;

perv

erse

subs

idie

s;op

enac

cess

Com

plex

food

web

s;ex

tens

ive

spat

iall

inka

ges

Fish

erie

sco

llaps

e;no

velf

ood

web

sE

limin

atio

nof

perv

erse

subs

idie

sfo

rin

dust

rial

fleet

sC

oast

al(C

hapt

er11

)L

oss

ofac

cess

righ

ts;p

ollu

tion

;de

clin

ing

stoc

ksfr

omov

erfis

hing

Acc

ess

tobo

thm

arin

ean

dte

rres

tria

lres

ourc

es;s

tron

gtr

adit

iona

lkno

wle

dge

and

ties

toco

ast

Aqu

acul

ture

atth

eex

pens

eof

othe

rec

osys

tem

serv

ices

;ex

trac

tion

ofno

nren

ewab

lere

sour

ces

(e.g

.,oi

land

gas)

atth

eex

pens

eof

fishi

ngliv

elih

oods

;

Cul

tiva

tion

ofne

wm

arke

tsto

sust

ain

trad

itio

nal

livel

ihoo

ds;s

tron

gco

-man

agem

ents

yste

ms

that

sust

ain,

and

whe

rene

eded

tran

sfor

m,l

ocal

com

mun

itie

sA

gric

ultu

re(C

hapt

er12

)L

oss

ofcr

opdi

vers

ity;

depe

nden

ceon

ener

gyse

ctor

;eco

nom

icgl

obal

izat

ion

Hig

hcr

opan

dcu

ltur

aldi

vers

ity;

inte

grat

edpe

stm

anag

emen

t;fle

xibl

efo

odpr

oduc

tion

syst

ems

Col

laps

eof

loca

lagr

icul

ture

inde

velo

ping

nati

ons

asfo

odai

del

imin

ates

loca

lmar

kets

Red

esig

nof

food

prod

ucti

onsy

stem

sto

enha

nce

othe

rec

osys

tem

serv

ices

Cit

ies

(Cha

pter

13)

Sepa

rati

onof

peop

lefr

omfo

odpr

oduc

tion

Hig

hcu

ltur

alan

dst

ruct

ural

dive

rsit

yC

olla

pse

ofec

onom

ies

that

supp

ortu

rban

livel

ihoo

dsan

din

fras

truc

ture

;w

ides

prea

ddi

seas

e

Ren

ewal

ofci

ties

asm

odel

sof

mod

ern

sust

aina

ble

livin

g;ur

ban–

rura

lfun

ctio

nas

com

plem

enta

rysy

stem

sP

lane

tEar

th(C

hapt

er14

)Se

nsit

ivit

yof

clim

ate

syst

emto

surf

ace

prop

erti

es;

inad

equa

tego

vern

ance

ofth

egl

obal

com

mon

s

Stro

ngbi

ophy

sica

lfee

dbac

ks;

dive

rsit

yof

know

ledg

esy

stem

s

Lar

gesc

ale

colla

pse

ofE

arth

asa

life-

supp

orts

yste

m;

glob

al-s

cale

confl

ictd

ueto

geop

olit

ical

issu

esof

clim

ate

chan

gean

den

ergy

dem

ands

Em

erge

nce

ofco

ordi

nate

dsy

stem

sof

glob

algo

vern

ance


have massive effects on drylands that are usu-ally poorly equipped to deal with these externalforces because of their distant voice. Frequenttransformations in drylands include desertifica-tion due to prolonged drought and/or overgraz-ing; conversion of mesic drylands to agriculture,which constrains access rights of pastoralists;and conversion to ranchettes or game ranches.Policies that mitigate these external forces forchange sustain the natural resilience of dry-lands.

Most of the vulnerabilities of freshwaterecosystems are associated with intense bioticinteractions that can drive rapid extirpation orevolution (see Chapter 9; Table 15.5). Freshwa-ters and their living resources are vulnerable tooverextraction, pollution with nutrients or tox-ins, species invasions, and trophic cascades dueto overexploitation of top predators. Resilienceof freshwater ecosystems is supported by bothecological and social factors. On the ecologicalside, riparian vegetation, variable flow regimes,and complex food webs with long-lived toppredators are keys to resilience. Over thou-sands of years, people have developed manyeffective institutions for dealing with watershortages. As these systems change, adaptivegovernance that is grounded in strong linkagesbetween monitoring and decision-making pro-vides an important source of resilience. Thereare both bad and good examples of trans-formation for freshwaters. In recent decades,many freshwater resources have been impairedand some have disappeared altogether. Insome cases, long-standing systems have beeneffective and in others emerging problemshave prompted people to develop new institu-tional arrangements for threatened freshwatercommons.

In open oceans (see Chapter 10), the lackof tight coevolutionary interactions betweenpeople and fisheries has caused people totreat many of these fish stocks as open-accessresources, making them highly vulnerable tooverfishing. Past and current levels of resourceuse in open oceans are the primary cause ofrecent collapses in marine fisheries (Table 15.5).This problem is social in nature, resulting fromboth increased demand for fish and increasedcapacity to catch fish via fossil-fuel-dependent

technologies. Perverse subsidies to industrialfleets drive most of the overfishing that hasoccurred. Subsidies prevent the fishery fromreaching a bioeconomic equilibrium in whichfishing pressure declines in response to declin-ing stocks. The essential feedbacks of thesocial–ecological fisheries systems have beenmasked. Fisheries management is currentlymoving into a new era of ecosystem manage-ment involving multispecies approaches; estab-lishment of marine protected areas; regula-tion of bycatch and habitat disruption (e.g., bytrawling); and enhancement of fishery produc-tion (through hatchery and habitat enhance-ment programs). Given the emergent condi-tions, adaptively managed fisheries programsof experimentation are important in evaluat-ing the currently unknown potential of theseprograms to enhance fishery sustainability andresilience.

In contrast to the open ocean, coastal oceans(see Chapter 11) are tightly coupled social–ecological systems (Table 15.5). The diversityof species and the sustainable patterns of liveli-hood that developed in coastal zones in bothpre- and postindustrial times are products of amultiplicity of possible combinations of favor-able conditions that permit coastal communi-ties to achieve conservation of adjacent marineresources and to remain resilient to change.Sources of resilience are tied to the designand performance of governance that ensuresthe sustainability of local communities, such ascommunity rights to access and to participatein management decisions, especially throughadaptive co-management arrangements and toaccess both marine and terrestrial resourcesat times of scarcity. Depletion of coastal fish-eries and economic globalization have led to thespread of one-species aquaculture, which tieslocal economies to global markets and marinesystems worldwide.

The agricultural production of crops and ani-mals for human consumption epitomizes thetight social–ecological linkages that are lessobvious in some other systems. Agriculturalsystems (see Chapter 12) have adapted to awide variety of environmental and social condi-tions, giving rise to a broad spectrum of genetic,crop, and cultural diversity (Table 15.5). The


resulting resilience can be enhanced by select-ing food production systems that are suitable tolocal social–ecological conditions, rather thanmodifying the environment to suit particularnonnative crops. When this does not occur,agricultural systems become vulnerable due toloss of local crop diversity and local knowl-edge, which could seriously constrain the capac-ity to adapt to directional climatic changes.Globalization of the economy creates addi-tional vulnerabilities by decoupling food andenvironment spatially and temporally and dis-torting agricultural markets. Plausible transfor-mations to address these vulnerabilities couldoccur by redesigning food production systemsin three ways: selection of crops suitable tothe environment; recoupling food productionwith its ecosystem base by removing distor-tionary policy incentives; and investment in abroad tool kit for agricultural improvement,including breeding, sustainable managementpractices and soil improvement techniques,multiple cropping systems, enhanced food qual-ity, advanced genetics, information systems, andconsumer labeling.

Cities (see Chapter 13) are the most rapidlyexpanding social–ecological systems on Earthand therefore provide huge opportunities forinnovation in social–ecological stewardship andpurposeful transformation to systems that max-imize efficiencies of energy and resource useand reduce pressures on the remainder of theplanet. They also represent huge concentrationsof power, with potential to ignore rural issuesand needs for sustainable development. Simi-larly, the tremendous human capital of urbansystems, their mixing of diverse global cultures,and rapid turnover of infrastructure provideboth opportunities for innovation and poten-tials for conflict and social–ecological degra-dation. Local social networks and bridgingorganizations can contribute to urban social–ecological resilience to (1) address new urbanchallenges and opportunities (e.g., rapid rural-to-urban migration, education and job opportu-nities for women that reduce population growthrates, rapid infrastructure turnover); (2) planfor the long-term in ways that reduce resourceextraction from, and impacts on, nonurbanregions; and (3) increase flexibility of gover-

nance structures to meet changing needs. In thiscontext, it becomes essential that city inhabi-tants recognize the significance of life-supportecosystems in rural areas worldwide for theirown well-being and that of the planet.

Given the scale and rate of changes, thereis now an inescapable need for effective gov-ernance at the level of the Earth System (seeChapter 14). Both the biophysical and the socialcomponents of the Earth System are vulner-able to rapid and irreversible changes thatcould be detrimental to human well-being andthe viability of modern societies (Table 15.5).Agreements are needed on issues like trade,security, migration, disease, climate, and otherlarge-scale challenges that take into accountthe significance of resilience-based stewardship.Global social–ecological collaborations and theemergence of multilevel governance systemsare needed to cope with, adapt to, and make useof rapid and directional change to shape trans-formations toward sustainability. Given thatglobal-scale experiments (e.g., climate changeand widespread biodiversity loss) result fromdecisions made at multiple scales, effectiveglobal governance must be linked to actions atmany scales. This is the fundamental challengeof the 21st century.

General Patterns

From our comparative analysis of social–ecological systems of the world, several clearmessages emerge:

• Resilience-based stewardship requiresactions that recognize the coupled, inter-dependent nature of social–ecologicalsystems.

• Every system exhibits critical vulnerabilitiesthat tend to become exacerbated as direc-tional environmental and social changespush these systems beyond the range ofconditions to which organisms and cul-tures are adapted. However, the nature ofthese vulnerabilities differs among social–ecological systems.

• Every system has sources of biological,cultural, and institutional diversity and asubstantial capacity to adapt to change.


This adaptive capacity can be enhancedthrough appropriate ecosystem stewardshipsupported by social–ecological governancefrom local-to-global scales.

• Every system has sources of resilience thatprovide opportunities for transformationto alternative, potentially more desirablesocial–ecological states. Human actions canenhance or degrade this resilience.

These broad conclusions suggest that ecosys-tem stewardship has important contributionsto make in all systems. There is no region soresilient that policy makers and resource man-agers can ignore potential threshold changes,and we doubt that any region is beyond hopeof substantial enhancement of well-being, adap-tive capacity, and resilience. Meeting the goalsof sustaining the important properties affect-ing ecosystem services will, however, requirereconnecting people’s perceptions, values, insti-tutions, actions, and governance systems to theprocesses and dynamics of the biosphere and anactive stewardship of ecosystems.

Critical Opportunities andChallenges for the Future

Resilience-based ecosystem stewardship pro-vides a framework and guidelines that can ame-liorate many problems and increase social–ecological sustainability. Nonetheless, seriouschallenges remain that constitute both majorrisks and opportunities for society. We high-light four social–ecological challenges that, ifaddressed effectively, would greatly enhancethe sustainability of our planet.

• Linking global, national, regional, and localgovernance: The planet appears to be rapidlyapproaching a tipping point of dangerousclimate change requiring new governancesystems to prevent this from occurring. Scat-tered local and national initiatives are begin-ning to address this issue. How can gov-ernance across the full range of scales belinked to respond appropriately to climatechange and avoid dangerous conditions?How can governance systems be designed to

facilitate local adaptation to ecological andsocial changes that are already occurring andreduce the risk of major geopolitical andsocial problems?

• Sustaining cultural and biological diversity ina globalized world: Irreversible losses of cul-tural and biological diversity are reducingresilience at all scales and the options foraddressing an uncertain and rapidly chang-ing future—often as an unintended conse-quence of a globalized market economy thatresponds to increasing demand for renew-able and nonrenewable resources. How canthe costs of globalization be taken intoaccount in ways that avoid the loss of cul-tural and biological diversity? What transfor-mations are needed to retain this diversitywhile fostering the resilience-based steward-ship needed to sustain the Earth System?

• Navigating transitions in human populationand consumption that place increasing pres-sure on the planet’s natural resources: Theabsolute increase in human population andits demand for food, energy, and other nat-ural resources are projected to be greaterin the next four decades than at any timein the history or likely future of our planet.What steps can be initiated now that willreduce this pressure on planetary resourcesand speed the transition to a more sustain-able future path?

• Using urbanization as an opportunity toenhance regional resilience: Rapid move-ment of people from rural areas to urbanareas provides opportunities to reshape pat-terns and pathways of using ecosystem ser-vices. How can rapid urbanization be usedto enhance resilience and the generation ofecosystem services to meet human needs(not desires) and to rebuild rural diversityand support social–ecological sustainability?

We suggest that resilience-based ecosystemstewardship provides a framework to effec-tively address these and other urgent issuesthat face society today and in the future. Likeall complex adaptive systems, resilience-basedecosystem stewardship is always a work inprogress that will inevitably adapt and trans-form as new insights and conditions emerge.


Review Questions

1. Describe, for a social–ecological system ofyour choice, specific changes in policies andinstitutions that would reduce its vulnerabil-ity to economic decline or expected climaticchanges. What practical steps would facili-tate these changes? How can these be initi-ated now?

2. Describe, for a social–ecological system ofyour choice, specific changes in policies andinstitutions that would enhance the adaptivecapacity and resilience to economic declineor expected climatic changes. What practicalsteps would facilitate these changes? Howcan these be initiated now?

3. Describe, for a social–ecological system ofyour choice, the unintended regime shifts or

transformations that might plausibly occurwithin the next 20–40 years. What are thecosts and benefits to different stakeholdersof these large potential changes? What prac-tical steps can be initiated now to reduce thelikelihood of these changes?

4. Describe, for a social–ecological system ofyour choice, plausible transformations thatmight be actively navigated to enhanceopportunities or to avoid major social–ecological problems that currently exist orare likely to occur. Propose a pragmaticset of actions that would make this trans-formation likely to occur, if the appropri-ate window of opportunity presented itself.How might this window of opportunity becreated?

Abbreviations

AD Anno Domini (years after thebirth of Christ)

AIDS Acquired immunodeficiency syn-drome

AMA Adaptive Management AreaB BillionBES Baltimore Ecosystem Study

LTER programCAFSAC Canadian Atlantic Fisheries Sci-

entific Advisory CommitteeCAM Crassulacian acid metabolismCAMPFIRE Communal Areas Manage-

ment Program for IndigenousResources

CAP Central Arizona-Phoenix LTERprogram

CBO Community-based organizationCCAMLR Convention on the Conserva-

tion of Antarctic Marine LivingResources

CFC ChlorofluorocarbonCH Camp householdCH4 MethaneCITES Convention on International

Trade in Endangered SpeciesCMA Catchment management agencyCO2 Carbon dioxideCRP Conservation reserve programCV Coefficient of variationDDE Breakdown product from an

insecticideDDP Dryland development paradigmDDT An insecticideEEZ Exclusive economic zoneENSO El Nino-Southern Oscillation

EU European UnionFAO Food and Agriculture Organiza-

tionFARMCs Fishery and Aquatic Resource

Management CouncilsFERTIMEX Mexican Fertilizer CompanyFDI Foreign direct investmentGBRMP Great Barrier Reef Marine ParkGCM General circulation model;

global climate modelGDP Gross domestic productGECs Global environmental changesGEF Global Environment FacilityGIS Geographic information systemGNP Gross national productGSCs Global social changesHIV Human immunodeficiency virusIBM International Business MachineIPAT Generalization relating human

impact, population, affluence,and technology

IPCC Intergovernmental Panel on Cli-mate Change

IPM Integrated pest managementIT Information technologyITQ Individual transferable quotaIUCN International Union for Conser-

vation of NatureIUU Illegal, unreported, and unregu-

lated (fishing)IVQ Individual vessel quotaKNP Kruger National ParkLDC Least developed countriesLTER Long-Term Ecological Research

program

339

340 Abbreviations

M MillionMDGs UN Millennium Development

GoalsMEA Millennium Ecosystem Assess-

mentMPA Marine protected areaMSY Maximum sustained yieldN2 Di-nitrogen, the predominant

form of nitrogen in the atmo-sphere

NAFO Northwest Atlantic FisheriesOrganization

NAFTA North American Free TradeAgreement

NGO Nongovernmental organizationNTFP Non-timber forest productNTT Nusa Tenggara Timor in Indone-

siaNWFP Northwest Forest PlanNYC New York CityODA Official development assistanceOECD Organization for Economic

Cooperation and DevelopmentPAM Plant-available moisturePAN Plant-available nutrientsPCB Polychlorinated biphenyl (class

of compounds frequently used aspesticides)

ppm Parts per millionREIT Real Estate Investment TrustTg TeragramsTIMO Timber Investment Management

OrganizationTPC Threshold of probable concernTRI Toxic release inventoryTURF Territorial Use Rights FisheryUK United KingdomUN United NationsUNDP United Nations Development

ProgrammeUNEP United Nations Environment

ProgrammeUNESCO United Nations Educational, Sci-

entific and Cultural OrganizationUNFCCC United Nations Framework Con-

vention on Climate ChangeUS United StatesUSDA US Department of AgricultureUSSR Union of Soviet Socialist

RepublicsVH Village householdWCED World Commission on Environ-

ment and DevelopmentWDF Washington Department of Fish-

eriesWRI World Resources Institute

Glossary

Aboriginal. See indigenous. The first group tooccupy a region. In Canada and Australiaused synonymously with indigenous.

Active adaptive management. Intentionalmanipulation of the system to test itsresponse.

Actor group. A group of people often withdifferent skills (e.g. knowledge carriers,entrepreneurs) engaged in an activity suchas ecosystem stewardship or poverty allevi-ation.

Adaptability. See adaptive capacity.Adaptation. Adjustment to a change in envi-

ronment. Defined by biologists and anthro-pologists as a genetic change in a population.Anthropologists also refer to adaptation as asocial, economic or cultural adjustment to achange in the physical or social environment.

Adaptive capacity. Capacity of human actors,both individuals and groups, to respond to,create, and shape variability and change inthe state of the system. Synonymous withadaptability.

Adaptive cycles. Cycles of system disruptionand renewal.

Adaptive co-management. Resource manage-ment that seeks social-ecological sustainabil-ity through a multi-scale and collaborativeprocess of intentionally learning from expe-rience.

Adaptive learning. Process in which oneor more groups (1) carefully and regu-larly observe social-ecological conditions,(2) draw on those observations to improveunderstanding of the system’s behavior, (3)

evaluate the implications of emergent con-ditions and the various options for actions,and (4) respond in ways that support theresilience of the social-ecological system.

Adaptive management. Resource managementapproach based on the science of learningby doing.See also active and passive adaptivemanagement.

Adaptive management. Resource managementapproach based on a systematic process forcontinually adjusting policies and practicesby learning from the outcome of previouslyused policies and practices.See also activeand passive adaptive management.

Adaptive governance. Active experimentationin governance with institutional and politi-cal frameworks designed to adapt to chang-ing relationships between society and ecosys-tems in ways that sustain ecosystem services;expands the focus from adaptive manage-ment of ecosystems to address the broadersocial contexts that enable ecosystem-basedmanagement.

Afforestation. Planting of forests on previouslyunforested lands.

Agency capture. Condition in which a specialinterest group establishes a controlling rela-tionship with an agency, which works almostexclusively on the interest group’s behalf.

Agistment. Practice in which private proper-ties that have too many animals for the avail-able forage in a particular year will transportstock to other properties where the imbal-ance is in reverse, and the senders pay thereceivers.

341

342 Glossary

Albedo. Reflectance of incoming solar (short-wave) radiation.

Alternative stable states. Alternative systemstates, each of which is plausible in a particu-lar environment.

Amplifying feedback. Feedback that augmentschanges in process rates and tend to desta-bilize the system. It occurs when two inter-acting components cause one another tochange in the same direction (both compo-nents increase or both decrease). Synony-mous with positive feedback.

Anion. Negatively charged ion.Anthropocene. New planetary epoch begin-

ning with the advent of industrializationcharacterized by a new situation whereglobal processes are strongly shaped byhumanity.

Aquaculture. Fish and shellfish cultured in con-fined systems.

Arable land. Land under temporary crops,temporary meadows for mowing or pasture,land under market and kitchen gardens andland temporarily fallow (less than five years).It does not include land that is potentiallycultivable.

Aridity index. Ratio of precipitation to poten-tial evapotranspiration.

Artisanal fishery. Small-scale mixed commer-cial and subsistence fishery, predominantly incoastal areas.

Assisted migration. Movement of organisms bypeople to a more favorable climate becausetheir natural migration is too slow to enablethem to adapt to climatic change.

Backcasting. A modeling approach that iden-tifies societal objectives and uses modelsand scenarios to explore how to arrive atthem.

Bennett’s law. Generalization that the caloricintake of households is dominated by starchystaples at low levels of income, but is char-acterized by a diversified diet of fruits, veg-etables, and animal products with incomegrowth.

Benthic. Bottom-dwelling organisms of aquaticecosystems.

Biodiversity. Number and relative abundanceof organisms in an area.

Bioeconomic equilibrium. Equilibrium harvestlevel dictated by harvest effort and prof-itability.

Blue water. Liquid water in rivers, lakes, reser-voirs, and groundwater aquifers.

Bonding network. Network of individualsbased on long-standing familiar relations.

Bottom-up effects. Regulation of consumerpopulations by food supply.

Boundary organizations. Organizations thatfacilitate the transfer of knowledge betweensystems, for example, between science andpolicy.

Bounded rationality. Decisions based onincomplete information.

Bridging network. Network that links individu-als and groups across scales to affect ecosys-tem stewardship process.

Bridging organization. Group, such as a board,council or other organization, that commu-nicates information and coordinates collabo-rations among local stakeholders and actorsacross several organizational levels or cul-tural systems.

Brousse tigre. Banded landscape patterns.Built capital. The physical means of produc-

tion beyond that which occurs in nature(e.g., tools, clothing, shelter, dams, andfactories). Synonymous with manufacturedcapital.

Bundles of services. Groups of ecosystem ser-vices that co-occur because of tight linkagesthrough ecosystem processes.

Bureaucratization. Process by which, overtime, an organization becomes increasinglydependent upon formal rules and proce-dures.

Bycatch. Species that are unintentionallycaught in the process of fishing for otherspecies.

Cap-and-trade. A market-based approach forcontrolling use of a common-pool resource.The cap is the maximum tolerable levelof resource use and is set by regulation.Marketable credits issued by the regulatoryauthority can be bought or sold to allowresource conservation by some users to bal-ance resource use by others.

Glossary 343

Capital. Assets or productive base of a social-ecological system, i.e., its human, manufac-tured, and natural assets.

Carrying capacity. Maximum quantity of anorganism that the environment can support.

Catchability coefficient. Fishing mortality rategenerated by one unit of fishing effort.

Cation. Positively charged ion.Chaotic behavior. Behavior that is unpre-

dictable and depends primarily on initialconditions and/or the nature of the perturba-tion.

CitiStat-type programs. Urban programs thatprovide accurate and timely intelligence,develop effective tactics and strategies,rapidly deploy resources, and facilitatefollow-up and assessments.

Citizen science. A type of knowledge pro-duction in which volunteers perform mon-itoring and research-related tasks, often totrack changes and or answer real-worldproblems.

Civil society. All voluntary civic activities,social organizations, and institutions that arethe basis for a functioning society.

Clay. Fine soil mineral particles.Coefficient of variation. Standard deviation

divided by the mean.Cognitive maps. Method people use to struc-

ture and store spatial knowledge, allowingthe “mind’s eye” to visualize images in orderto reduce cognitive load, and enhance recalland learning of information. Cognition canbe used to refer to the mental models, orbelief systems, that people use to perceive,contextualize, simplify, and make sense ofotherwise complex problems.

Collective action. Cooperation of individuals topursue a common goal.

Co-management. Sharing of power andresponsibility between resource user com-munities and resource management agen-cies.

Command-and-control. Decision making fromthe highest division of the organization thatseeks to reduce variability so as to maintainconstant output of some ecosystem good orservice.

Common-pool resources. Shared resourcesthat are subtractable and from which it

is costly to exclude people’s use. Formerlytermed common-property resources.

Common-property resources. See common-pool resources.

Command-and-control. Decision making fromthe highest division of the organization thatseeks to reduce variability so as to maintainconstant output of some ecosystem good orservice.

Community-based wildlife management pro-gram. Program that assumes that high lev-els of community involvement in resourcemanagement and explicit incentives for con-servation encourage stewardship behaviorand protect threatened species from possibleextinction.

Community quota. A property right held incommon by a community for rights to har-vest, which are issued through legal meansand are usually transferable only among indi-vidual members of a specified organization,but not at speculative prices.

Complex adaptive system. System whose com-ponents interact in ways that cause the sys-tem to adjust (i.e., “adapt”) in response tochanges in conditions.

Complex problem. Problem with many poten-tial solutions that are quite different in exe-cution, and rankable in quality of outcome.

Congestible resources. Resources thatdecrease in quality when the number ofresource users reaches a threshold, for rea-sons other than ecological productivity (e.g.,enjoyment of wilderness).

Connector. Individual who knows lots of peo-ple in terms of both numbers and kinds ofpeople, in particular the diversity of acquain-tances.

Conservation by utilization. Ecosystem man-agement policy, which acknowledges thatlandowners should benefit from ecosystemservices.

Conservation phase. Phase of an adaptive cycleduring which interactions among compo-nents of the system become more specializedand complex.

Context-dependent. Actions framed by anoverall context, for example a legal frame-work, norms and rules, a historical trajectoryor a cultural belief system.

344 Glossary

Coping. Short-term adjustment by individualsor groups to minimize the impacts of hazardsor stresses.

Coupled human-environment system. Seesocial-ecological system.

Crisis. Time when a group of people or asociety agrees that some components ofthe present system are dysfunctional and istherefore more likely to consider novel alter-natives.

Critical ecosystem services. Ecosystem servicesthat (1) society depends on or values; (2)are undergoing (or are vulnerable to) rapidchange; and/or (3) have no technological oroff-site substitutes.

Cross-scale linkages. Processes that connectthe dynamics of a system to events that occurat other times and places.

Cross-scale surprises. Surprises that occurwhen there are cross-scale interactions, suchas when local variables coalesce to generatean unanticipated regional or global pattern,or when a process exhibits contagion (as withfire, insect outbreak, and disease).

Cultural ecology. Study of the relationship ofa given society and its natural environment,with a focus on changing social and ecologi-cal conditions.

Cultural heritage. Stories, legends, and memo-ries of past cultural ties to the environment.

Cultural identity. A sense of membership by anindividual or group to a culture.

Cultural services. Non-material benefits thatsociety receives from ecosystems (e.g., cul-tural identity, recreation, and aesthetic, spir-itual, and religious benefits).

Cumulative processes. Processes that occurlocally in different parts of the world, butover time the effect aggregates to produceregional-to-global consequences.

Cyanobacteria. Nitrogen-fixing bluegreenalgae.

Decision-support tools. Tools that integrateinformation to assess the state of knowledge,assess management actions, direct futuremonitoring and research, and explore theimplications of alternative futures.

Decomposer. Organism that uses dead organicmatter as a source of energy.

Decomposition. Chemical breakdown of deadorganic matter by soil organisms.

De facto. Used in practice, often referred towith respect to property rights.

Deforestation. Conversion of forests to a non-forested ecosystem type, frequently agricul-ture.

Degradation. Deterioration of a system to aless desirable state as a result of failure toactively adapt or transform.

De jure. Formal, based in a legal convention.Denitrification. Conversion of nitrate to

gaseous forms (N2, NO, and N2O).Depensatory decline. Population decline that

accelerates as the stock size declines.Desalinization. Conversion of salt water into

freshwater.Desertification. Soil degradation that occurs

in drylands that is triggered by drought,reduced vegetation cover, over-grazing, ortheir interactions.

Deterministic. Governed by and predictable interms of a set of laws or rules.

Development and empowerment organiza-tions. Organizations that work with com-munities and other stakeholder groups tobuild local capacity to address problems bothcurrent and future, and communicate betterinternally and with other groups. Commonlyassociated with economic development.

Directional change. Change with a persistenttrend over time.

Disciplinary. Belonging to a field of study thatshares a common perspective of the world.

Discount rate. The estimated value of goodsand services that can be obtained in thefuture, relative to the value of the samegoods that are available today.

Discourse. Formal or orderly expression ofthought on a subject, including what is dis-cussed, how it is discussed, and which aspectsare deemed legitimate or are marginalized.

Disequilibrial. Perpetually buffeted away fromany equilibrium.

Disturbance. Relatively discrete event in timeand space that alters ecosystem structure andcauses changes in resource availability orphysical environment.

Disturbance regime. Characteristic severity,frequency, type, size, timing, and intensity ofa set of disturbances in an ecosystem.

Glossary 345

Double-loop learning. Feedback process indecision making by which practitionersreflect on the consequence of past actionsbefore taking further action.

Drip irrigation. Irrigation method that mini-mizes the use of water and fertilizer by allow-ing water to drip slowly to the roots of plants,either onto the soil surface or directly ontothe root zone.

Dryland development paradigm (DDP). SeeDrylands syndrome.

Drylands syndrome. Suite of key attributesthat characterize most drylands of theworld: unpredictability, resource scarcity,sparse populations, remoteness and ‘distantvoice.’ Synonymous with dryland develop-ment paradigm.

Dustbowl. Region of the central US wherewind erosion removed massive amounts ofsoil during the extended drought of the1930s.

Dynamics of denial. Perspective in which agroup of people, for example, fishermenargue that scientific assessments must bewrong because they are still catching plentyof fish (and if there is a decline, recoverywill occur naturally due to favorable environ-mental circumstances).

Earth System. Planet Earth as a social-ecological system.

Economic rent. Income gained relative to theminimum income necessary to make anactivity economically viable. The economicrent in a sales transaction is the differencebetween the payment actually received andthe second-best price the owner could oth-erwise get for using land, labor or capital inanother activity.

Ecosystem. All of the organisms (plants,microbes, and animals—including people)and the physical components (atmosphere,soil, water, etc.) with which they interact.

Ecosystem goods. See provisioning services.Ecosystem management. Management

paradigm that emphasizes practices formultiple ecosystem services by capitalizingon, sustaining, and enhancing ecosystemprocesses.

Ecosystem services. Benefits that peoplereceive from ecosystems, including support-

ing, provisioning, regulating, and culturalservices.

Ecotone. Edge or transition zone between twoecological communities.

Effect diversity. The diversity of organisms withrespect to their effects on ecosystem pro-cesses.

Emergent properties. Property of a complexsystem that emerges from interactions ofsubunits and cannot be predicted by study-ing individual subunits.

Engel’s law. Generalization that, as incomesgrow, the proportion of household incomespent on food in the aggregate declines.

Engineering fishery policies. Increase the safecatch curve through engineered habitat man-agement and artificial propagation (hatch-ery) systems.

Entitlements. Sets of alternative benefits thatpeople can access, depending on their rightsand opportunities, sometimes guaranteedthrough law. It also refers, in a more casualsense to someone’s belief of deserving someparticular reward or benefit.

Environmental event. Physical events occur-ring in ecosystems such as floods anddroughts.

Environmental injustice. Uneven burden ofenvironmental hazards among differentsocial groups.

Equilibrium. Condition of a system thatremains unchanged over time because of abalance among opposing forces.

Equity. Fairness.Erosion. Transport of soil particles by wind or

water from one place to another.Essential resource. Resource for which no sub-

stitute exists.Eutrophication. Nutrient enrichment (typi-

cally nitrogen and phosphorus pollution ofaquatic systems).

Excludable resources. Resources from whichpotential users can be excluded at low cost.

Exclusive economic zone (EEZ). Marinewaters within 200 miles of a nation’scoast.

Exogenous factor. Factor external to the systembeing examined, which therefore is not incor-porated into management practices.

Expert knowledge. Knowledge based onempirical relationships.

346 Glossary

Exposure. Nature and degree to which thesystem experiences environmental or socio-political stress.

Extensification. Increased agricultural produc-tion achieved through ecosystem conversionto agricultural lands.

Evolution. Changes in organisms as a result ofgenetic responses to past events.

Fast variable. Variable that responds sensitivelyto daily, seasonal, and interannual variationin weather and other factors.

Feedback. See Amplifying feedback and stabi-lizing feedback.

Fishery collapse. 90% or greater reduction incatch from peak historic levels.

Fit. Match of characteristics of institutions withthe dimensions of a social-ecological systemthey are intended to govern.

Fixed escapement rule. Allow the same fixednumber of fish to spawn each year.

Fixed exploitation rate. Harvest the same fixedpercentage of the stock each year, leavingthe rest to spawn.

Food security. Ability of all people at all timesto achieve physical and economic access tothe food needed to lead a productive andhealthy life.

Forecasting. Projection of future conditionsand their social-ecological consequences.

Forest certification. Procedure for assessingforest management practices against stan-dards for sustainability so purchasers cansupport sustainable management.

Formal institutions. Formal sets of written rulesenforced by governments such as constitu-tions, laws, and legally based conventions.

Formal knowledge. “Scientific knowledge”learned from books or through formal edu-cational systems.

Fossil groundwater. Groundwater that accumu-lated under a different climate regime andis no longer being renewed at a significantrate.

Framing. Defining a problem or issue in a con-text that conveys its value to the public andto decision makers.

Free-rider problem. Situation that occurs whenindividuals or groups of actors do not assumetheir fair share of responsibilities while

consuming more than their share of theresources.

Fugitive resources. Resources that move acrossa range of jurisdictions and can have manyuser groups.

Functional redundancy. Diversity within afunctional type.

Functional silos. Fragmentation of responsibil-ities and authorities among individuals oragencies. Synonymous with stove pipes.

Functional types. Groups of organisms thatexert similar effects on social-ecological sys-tems (effect functional type) or which showsimilar response (response functional type)to an environmental change (e.g., evergreentrees, algal-eating fish).

Genuine investment. Increase in the produc-tive base (total capital or inclusive wealth) ofa social-ecological system.

Global Environment Facility (GEF). Multilat-eral funding mechanism administered jointlyby UNEP, UNDP, and the World Bankand dedicated to providing funds neededto help developing countries address large-scale environmental problems like climatechange.

Globalization. Global interconnectedness (e.g.,of economy or culture).

Goal displacement. Condition in which the sur-vival of the organization assumes greaterpriority than efforts to meet its statedmission.

Governance. Pattern of interaction amongactors, their sometimes conflicting objec-tives, and the instruments chosen to steersocial and environmental processes within aparticular policy area.

Great acceleration. The explosion of thehuman enterprise in terms of both popula-tion and economic activities after the SecondWorld War.

Green revolution. Design and disseminationof high-yielding seed varieties for the majorcereal crops requiring intensive applicationof fertilizers and pesticides.

Green water. Moisture in the soil that supportsall non-irrigated vegetation, including rain-fed crops, pastures, timber, and terrestrialnatural vegetation.

Glossary 347

Growth phase. Phase of the adaptive cycle dur-ing which environmental resources are incor-porated into living organisms, and policiesbecome regularized.

Habitat/species management area. Protectedarea managed mainly for conservationthrough management intervention.

Hierarchical system. A vertically organizedsystem, often in the shape of a pyramid, witheach row of objects linked to objects directlybeneath it, for example, an army or a com-puter file system.

Horizontal interplay. Interaction among insti-tutions at the same level of social organiza-tion or across space.

Horizontal linkages. Linkages that occur at thesame level of social organization and acrossspatial scales, e.g., in treaties among coun-tries.

Human agency. Capacity of people to makeindividual and collective choices and toimpose their choices on the world.

Human capital. Capacity of people to accom-plish their goals given their skills at hand.

Human ecology. See cultural ecology.Hydroponics. Method of growing plants using

mineral nutrient solutions instead of soil.

Iconic species. Species that symbolize impor-tant nature-based societal values.

Inclusive wealth. Total capital (natural, manu-factured, human, and social) that constitutesthe productive base available to society.

Income elasticity of demand. Growth in thedemand for a good in response to an increasein income of people demanding that good.

Indecision as rational choice. Failure of reg-ulators to make decisions when faced withconvincing evidence from stakeholders ofdetrimental economic impacts and uncer-tain scientific evidence of degradation of theresource being regulated.

Indicator. Parameter monitored to character-ize the unmeasured structural and composi-tional parts of the system and their processes.

Indigenous. People who have lived in a placelonger than the written record, often theoriginal inhabitants. It generally refers to agroup that is ethnically distinct from a group

that colonized a region in historical times. InCanada the preferred term is aboriginal.

Individual transferable quota (ITQ). Quotasissued by government that privatize theaccess and withdrawal rights to a specificmarine resource.

Industrial roundwood. Sawlogs and pulpwood(and the resulting chips, particles, and woodresidues). Synonymous with timber.

Informal institution. Unwritten rules such assanctions, taboos, customs, and traditions.

Institution. Rules in use that create enduringregularities of human action in situationsstructured by rules, norms, and shared strate-gies.

Institutional fit. See fit.Institutional interplay. Interactions among

institutions.Institutional learning. See social learning.Integrated pest management (IPM). Multidi-

mensional approach for managing agricul-tural pests by development of new host plantresistance and limited spraying of chemicals.

Intensification. Increased agricultural produc-tion achieved through increased inputs offertilizers, pesticides, irrigation, and technol-ogy to enhance yield per unit land area.

Interdisciplinary. Integration among disciplinesto define problems, share methods, and cre-ate questions based on a shared perspectiveof how the world works.

Interplay. Interactions among institutions.Investment. Increase in the quantity of an asset

times its value.IPAT law. Rule of thumb stating that the

impact (I) of any human activity is equal tothe product of the human population size(P), affluence of the population (A), and thetechnology employed in production (T).

Keystone species. Species that have dispro-portionately large effects on ecosystems,typically because they alter critical slowvariables.

Knowledge system. Culturally defined way ofknowing.

Landscape function. Capacity of a landscapeto regulate nutrients and water, concen-trating them in fertile, vegetated patches

348 Glossary

where soil biota maintain nutrient cycles andwater infiltration and where vegetation coverimpedes surface water flow, retains nutrientsand seeds, maintains infiltration and protectsagainst erosion.

Learning networks. See communities of prac-tice.

Legacies. Stored past experiences of thedynamics of social-ecological systems.

Life-support system. Supporting ecosystem ser-vices that give rise to the provisioning, reg-ulating, and cultural ecosystem services forsociety.

Limited entry. Policy requiring permits in zonesof use.

Litter. Layer of dead leaves on the soil surface.Livelihood. Strategy undertaken by individu-

als or social groups to create or maintain aliving.

Local knowledge. A dynamic system of place-based observations, interpretations, andlocal preferences that inform people’s use ofand relationship with their environment andwith other people. It may include a mix ofsocial, ecological, and practical knowledgeand involve a belief component. Overlapswith traditional ecological knowledge.

Local surprises. Surprises that occur locally andare created by a narrow breadth of experi-ence with a particular system, either tempo-rally or spatially.

Managed resource protected area. Protectedarea managed mainly for the sustainable useof natural ecosystems.

Manufactured capital. See built capital.Maven. An altruistic individual with social

skills who serve as information brokers, shar-ing and trading what they know.

Maximum sustained yield. Policy that seeks tomaximize the harvest of forests, fish, andwildlife to meet current needs, while notexceeding the current biomass increment,thus sustaining the potential to continuethese yields in the future. Often optimisticabout current and future potential yields.

Mental map. See cognitive map.Metapopulations. Populations of a species that

consist of partially isolated subpopulations.Mitigation. Reduction in the exposure of a sys-

tem to a stress or hazard.

Mobile links. Biological or physical pro-cesses that link patches on a landscape orseascape.

Muddling through. Process of non-strategictrial-and-error policy making in which solu-tions to problems are sought by organiza-tions without the benefits of careful reflec-tion.

Multidisciplinary. Perspectives from multipledisciplines on how the world works, witheach discipline working within its own disci-plinary framework.

Multilevel governance. Governance that takesplace across several levels of institutions.

Multiple-use management. Management thatexplicitly seeks to sustain a broad array ofecosystem services.

Multi-stakeholder body. Group that is oftenconvened by policy makers to scope issues,seek solutions to problems, achieve broaderpublic participation, and foster public con-sensus.

Mycorrhizae. Symbiotic relationship betweenfungi and plant roots leading to an exchangeof fungal nutrients for plant carbohydrates.

National monument. Protected area managedmainly for conservation of specific naturalfeatures.

National park. Protected area managed mainlyfor ecosystem protection and recreation.

Natural capital. Nonrenewable and renewablenatural resources that support the produc-tion of goods and services on which societydepends.

Natural enemies. Pathogens, predators, andparasitoids of agricultural pests.

Negative feedback. See stabilizing feedback.Neo-liberalism. Political movement that

espouses economic liberalism (i.e., minimalgovernmental interference) as a meansof promoting economic development andsecuring political liberty.

Nitrogen fixation. Conversion of gaseous nitro-gen (N2) to ammonium by organisms (bio-logical nitrogen fixation) or industrially usingfossil fuels as an energy source (industrialnitrogen fixation).

Nonpoint pollution. Nutrients or toxinsscoured by runoff from agricultural or urbanlands.

Glossary 349

Norms. Rules for social behavior.Normative concepts. Concepts with a values

orientation.

Ontogenetic habitat shifts. Habitat shifts thatoccur as fish develop from juvenile to adult.

Open access. Situation in which potential usersare not excluded from using a resource.

Open systems. System characterized by flowsof materials, organisms, and information intoand out of the system.

Opportunity costs. Potential benefits that mighthave been gained from some alternativeaction.

Organization. Social collective with member-ship and resources, which functions as a com-ponent of broader social networks.

Overland flow. Movement of water across thesoil surface.

Panarchy. Mosaics of nested subsystems thatare at different stages of their adaptivecycles and operating at different scales withmoments of interaction across scales.

Passive adaptive management. Learningthrough intensive examination of historiccause-effect relationships.

Path dependence. Strong impacts of historicallegacies on the future trajectory of a system,or, more narrowly, the co-evolution of insti-tutions and social-ecological conditions in aparticular historical context.

Pelagic. Water column-dwelling.Piscivore. Organisms that eat fish.Planktivore. Organisms that eat plankton.Permanent crops. Land cultivated with crops

that occupy the land for long periods andneed not be replanted after each har-vest, such as fruit trees, cocoa, coffee, andrubber.

Permanent pastures. Land used permanently(five years or more) for herbaceous foragecrops, either cultivated or growing wild (wildprairie or grazing land).

Pluralism. Operating principle that affirms andaccepts a diversity of perspectives, mentalmodels, or knowledge systems.

Policy community. Group that shares an inter-est in one or more specific issues and collab-orates to change policy.

Polycentric governance. The organization ofsmall, medium, and large-scale democraticunits that each exercise independence tomake and enforce rules within a scope ofauthority for a specified geographical area.

Polycentric institutions. Complex array ofinteracting institutions and organizationswith overlapping and varying objectives, lev-els of authority, and strengths of linkages.

Positive feedback. See amplifying feedback.Poverty. Pronounced deprivation of well-being.Poverty trap. Situation characterized by persis-

tent poverty, reflecting a loss of options todevelop or deal with change.

Power. Having influence over others throughvarious resources, including formal author-ity, threat, charisma, money, and infor-mation; intentionally imposed or achievedthrough passive or disruptive behavior.

Precautionary fishery policies. Stock assess-ments and harvest regulations are deliber-ately chosen to be conservative enough toassure that the safe harvest curve remainshigh over time.

Precautionary principle. If an action or policymight cause severe harm, but the outcome isuncertain, then the burden of proof falls onthose who advocate taking the action.

Pressure group. Group that lobbies to bringabout institutional change.

Property rights. The relationship between aresource user, a community, a society, andthe rights, rules and responsibilities that gov-ern access to and use of resources. The threeclasses of property rights are private, com-munal, and state property.

Prospective actions. Actions taken todaybecause people believe those actions willinfluence the capacity of future residents andgovernance networks to meet short-term andlong-term challenges and opportunities.

Protected landscape/seascape. Protected areamanaged mainly for landscape/seascape con-servation and recreation.

Provisioning services. Products of ecosystemsthat are directly harvested by society (e.g.,fresh water, food, fiber, fuelwood, bio-chemicals, and genetic resources). Synony-mous with ecosystem goods or renewableresources.

350 Glossary

Public consultation processes. Strategy forlinking across scales, commonly used by gov-ernment management agencies to informdecision making about the interests and con-cerns of local communities.

Punctuated equilibrium. Temporal pattern inwhich long periods of stability and incre-mental change are separated by abrupt, non-incremental, large-scale changes.

Pure public goods. Goods that are not sub-tractable and nonexcludable.

Redundancy. Diversity of functionally similarcomponents (e.g., species or institutions) toprovide multiple means of accomplishing thesame ends, in the event that some compo-nents disappear.

Reflexive behavior. Human behavior thatallows planning for the future by taking intoaccount the likely consequences of actions.

Reforestation. Regrowth or planting of forestson previously forested lands.

Regime shift. Abrupt large-scale transition toa new state or stability domain characterizedby very different structure and feedbacks.

Regulating services. Regulation by ecosystemsof processes that extend beyond their bound-aries (e.g., regulation of climate, water quan-tity and quality, disease, and pollination).

Release phase. Phase of an adaptive cycle thatradically and rapidly reduces the structuralcomplexity of a system.

Renewable resource. Resource whose rate ofextraction does not exceed the replenish-ment rate.

Renewal phase. Phase of an adaptive cycle inwhich the system reorganizes through thedevelopment of stabilizing feedbacks thattend to sustain the same properties overtime.

Rent. See economic rent.Rent seeking. Situation that occurs when indi-

viduals or organizations glean the benefits ofa transaction without contributing to, and insome cases subtracting from, the welfare ofsociety.

Resilience. Capacity of a social-ecologicalsystem to absorb a spectrum of shocks or per-turbations and to sustain and develop its fun-damental function, structure, identity, and

feedbacks as a result of recovery or reorga-nization in a new context.

Resilience-based ecosystem stewardship. Asuite of approaches whose goal is to sustainsocial-ecological systems, based on reducingvulnerability and enhancing adaptive capac-ity, resilience, and transformability. Its goalsare to respond to and shape change in social-ecological systems in order to sustain thesupply and opportunities for use of ecosys-tem services by society.

Resilience learning. Form of social learningthat fosters society’s capacity to be pre-pared for the long term by enhancing systemresilience and sustainability.

Resource regime. Cluster of institutions gov-erning management of a particular area ofinterest or resource.

Response diversity. The diversity of responsesto environmental change among organ-isms contributing to the same ecosystemfunction.

Rigidity trap. Situation in which people andinstitutions try to resist change and persistwith their current management and gover-nance system despite a clear recognition thatchange is essential.

Roving bandits. People that range widely toexploit resources without regard to estab-lished institutions.

Rules-in-use. Practiced institutions (vs.espoused rules).

Runoff. Water that moves as overland flow orgroundwater from a terrestrial to an aquaticsystem. Gives rise to blue water.

Salesman. Individual with the social skills topersuade people unconvinced of what theyare hearing.

Salinization. Accumulation of salts in soils orfreshwaters to the point that productivitydeclines.

Salmonid. Fish belonging to the salmon family.Scenario. A plausible, simplified, synthetic

description of how the future of a sys-tem might develop, based on a coherentand internally consistent set of assumptionsabout key driving forces and relationshipsamong key variables.

Glossary 351

Sector. A division of society, often associatedwith an economic activity (e.g., the businesssector).

Self-efficacy. Sense of having an effect onevents through one’s efforts, having thepower to make a difference.

Self-organization. The development of systemstructure as a result of stabilizing feedbacksamong system components.

Sense of place. Self-identification with a partic-ular location or region.

Shadow network. Group that is indirectlyinvolved in decision making, but supportiveto the process.

Shifting baseline syndrome. Depletion of astock that occurs so gradually that currentlevels are accepted as normal, and no poli-cies are implemented to prevent furtherdepletion.

Shifting cultivation. See swidden agriculture.Single-loop learning. Feedback process that

adjusts actions to meet identified manage-ment goals (e.g., modifies harvest rate to con-form to specified catch limits) but does notevaluate basic assumptions and approaches.

Single-species management. Management tomaintain the abundance or productivity of asingle species.

Six degrees of separation. Hypothesis thateveryone on Earth has a maximum of sixlinkages that separate him/her from anyother person.

Slash-and-burn agriculture. See swidden agri-culture.

Slow variables. Variables that strongly influ-ence social-ecological systems but remainrelatively constant over years-to-decades.

Social capital. Capacity of groups of people toact collectively to solve problems.

Social-ecological governance. Collective coor-dination of efforts to define and achievesocietal goals related to human-environmentinteractions.

Social-ecological processes. Interconnectionsamong components of a social-ecologicalsystem.

Social-ecological system. System with inter-acting and interdependent physical, biolog-ical, and social components, emphasizing the‘humans-in-nature’ perspective.

Social learning. Process by which groups assesspast and emergent social-ecological condi-tions and respond in ways to meet objectives.

Social memory. Memory of past experiencesthat is retained by groups, providing a legacyof knowing how to do things under differentcircumstances.

Social movement. Group that operates overbroad geographic scales and seeks toadvance a particular ideological perspective.

Social network. Linkages that establish rela-tions among individuals and organizations(and their institutions) across time andspace.

Soil. Mixture of small mineral and organic par-ticles that retain the water and nutrientsrequired for growth of terrestrial plants.

Special interest group. Organization that advo-cates for policies that serve its interests atlocal to regional scales.

Species diversity. Number of species, adjustedfor relative abundance.

Stability. Tendency of the system to maintainthe same properties over time.

Stabilizing feedback. Feedback that tendsto reduce fluctuations in process rates,although, if extreme, can induce chaoticfluctuations. A stabilizing feedback occurswhen two interacting components cause oneanother to change in opposite directions.Synonymous with negative feedback.

Stakeholders. Individuals and organizations,including government, affected by policydecisions.

Staple. Crops that form the basis of the tradi-tional diet of a region.

Steady state. Condition of a system in whichthere is no net change in system structure orfunctioning over the time scale of study.

Steady-state mosaic. Landscape in which dif-ferent stands are at different successionalstages, but there is no net change in land-scape composition over time.

Stewardship. See resilience-based ecosystemstewardship.

Stove pipes. See functional silos.Strict nature reserve. Protected area managed

mainly for science.Substitution. Replacement of one form of cap-

ital input by another.

352 Glossary

Subtractable. One person’s use of the resourcereduces the availability of that resource foruse by others.

Succession. Directional change in ecosystemproperties resulting from biologically drivenchanges in resource supply.

Supporting services. Fundamental ecologicalprocesses that sustain ecosystem function-ing (maintenance of soil fertility, cycling ofessential elements, biological diversity, andcycles of disturbance and renewal).

Surprise. An unexpected and unimaginedoccurrence.See See cross-scale surprises,local surprises, and true-novelty surprises.

Sustainability. Use of the environment andresources to meet the needs of the presentwithout compromising the ability of futuregenerations to meet their own needs. Main-tenance of the productive base (total capital)over time.

Sustainable development. Development thatseeks to improve human well-being, while atthe same time sustaining the natural resourcebase on which society depends for futuregenerations.

Sustainable management. Management to sus-tain the functional properties of social-ecological systems that are important tosociety.

Sustained yield. Production of a biologicalresource (e.g., timber or fish) under manage-ment procedures that ensure replacement ofthe part harvested by regrowth or reproduc-tion before the next harvest occurs.

Swidden agriculture. Agricultural systeminvolving cycles of forest clearing, growingof crops, and regrowth of forests in smallpatches. Synonymous with slash-and-burnagriculture or shifting cultivation.

Synergy. Ecosystem services or societal bene-fits that co-occur with other services and ben-efits (e.g., aesthetic value and carbon seques-tration provided by natural forests).

Systemic processes. Processes for whichchanges anywhere on the planet rapidlyaffect the Earth System at the global scale.

Temporal tradeoffs. Tradeoffs between short-term benefits and long-term capacity ofecosystems to provide services to future gen-

erations. Synonymous with intergenerationaltradeoffs.

Theory of weak ties. Theory that criticalinformation is most commonly receivedfrom those outside one’s stronger socialnetwork.

Threshold. Critical level of a driver orstate variable that, when crossed, triggersan abrupt change (regime shift) in thesystem.

Threshold of probable concern (TPC). Upperand lower levels in selected indicators ofmanagement goals that together define cur-rent views about acceptable heterogeneity inconditions of ecosystems.

Timber. See industrial roundwood.Tipping point. Threshold for change to a new

system controlled by different critical slowvariables and feedbacks.

Top-down effects. Effects of predators (ecol-ogy) or large-scale organizations and institu-tions on a group of actors.

Tradeoffs. Ecosystem services or societal bene-fits that can be obtained only at the expenseof other services and benefits (e.g., clearcutlogging and old-growth conservation of thesame forest).

Traditional knowledge. A cumulative bodyof knowledge observations, understanding,practice, and belief, evolving by adaptiveprocesses and handed down through genera-tions by cultural transmission, about the rela-tionship of living beings (including humans)with one another and with their environ-ment. Typically viewed as held by indigenouspeople or a group unique to a society. Over-laps with local knowledge.

Tragedy of the commons. Outcome in whichthe self-interest of resource consumers andpoorly defined property rights result in sig-nificant degradation of collectively sharedor common-pool resources, often referred tothe tragedy of open access.

Transaction costs. Costs associated withsearch, communication, negotiation, mon-itoring, coordination and enforcement ofrules.

Transdisciplinary. Integration that transcendstraditional disciplines to formulate problemsin new ways.

Glossary 353

Transformability. Capacity to re-conceptualizeand create a fundamentally new system withdifferent characteristics.

Transformation. Fundamental alteration of thenature of a system when the current ecolog-ical, social, or economic conditions becomeuntenable or are undesirable. Transforma-tion involves a paradigm shift that re-conceptualizes the nature of the system interms of a different set of critical slow vari-ables, feedbacks, and social goals. Transfor-mations can be purposefully navigated orunintended.

Transformative learning. Learning that re-conceptualizes the system through processesof reflection and engagement.

Transhumance. Seasonal movements of pas-toralists.

Transpiration. Evaporation of water from cellsurfaces inside leaves that supports plantproduction. Synonymous with green water.

Triple-loop learning. Learning that redefinesnorms and protocols as a basis for changesin governance.

Trophic cascade. Changes in species composi-tion driven by changes in top predators thataffect lower trophic levels, primary produc-ers, bacteria and nutrient cycles.

True-novelty surprise. Never-before-experienced phenomena for which strictpre-adaptation is impossible.

Turnover time. Amount of time that it takes toreplace the pool of a material in an ecosys-tem, if the pool size is steady over time.

Type 1 error. Acceptance of a proposition thatturns out to be false.

Type 2 error. Failure to reject a false proposi-tion.

Upwelling. Movement of deep nutrient-richwater to the surface.

Utility. Capacity of individuals or society tomeet its own needs.

Vertical interplay. Interaction among institu-tions in multilevel governance systems.

Vertical linkages. Linkages that occur as part ofinter-organizational relationships, such as inthe transactions between local- and national-level management systems operating in agiven region.

Virtual water. Volume of freshwater needed toproduce a specific product or service.

Vulnerability. Degree to which a system islikely to experience harm due to exposure toa specified hazard or stress.

Vulnerable. Likely to change in state inresponse to a stress or stressor.

Wallace’s Line. Zoographic boundary betweenAsian and Australasian faunas.

Watershed. A lake or stream and all the landsthat drain into it.

Weathering. Breakdown of rocks to form soilparticles.

Well-being. Quality of life; basic material needsfor a good life, freedom and choice, goodsocial relations, and personal security. Also,the present value of future utility.

Wicked problem. Problem that is so complexthat each attempted solution creates newproblems for other segments of society orother times and places.

Wilderness area. Protected area managedmainly for wilderness protection.

Window of opportunity. Short periods thatoffer possibilities for large-scale change.Windows may open because of exogenousshocks and crises, including shifts in under-lying economic factors such as a rapid rise inenergy prices, a change in the macropoliticalenvironment, new scientific findings, regimeshifts in ecosystems, or rapid loss of ecosys-tem services.

Worldview. Framework by which a group inter-prets events and interacts with its social-ecological system.

Reference

Abel, N.M., and P. Blaikie. 1986. Elephants, peo-ple, parks and development: The case of theLuangwa Valley, Zambia. Environmental Man-agement 10:735–751.

Abel, N.M., D.H.M. Cumming, and J.M.Anderies. 2006. Collapse and reorgani-zation in social-ecological systems: Ques-tions, some ideas, and policy implica-tions. Ecology and Society 11(1):17. http://www.ecologyandsociety.org/vol11/iss11/art17/.

Abel, N.M. 1997. Mis-measurement of the produc-tivity and sustainability of African communalrangelands: A case study and some principles fromBotswana. Ecological Economics 23: 113–133.

Acheson, J.M. 1989. Management of common-property resources. Pages 351–475 in S. Plattner,editor. Economic Anthropology. Stanford Uni-versity Press, Stanford.

Acheson, J.M., J.A. Wilson, and R.S. Steneck. 1998.AQ1Managing chaotic fisheries. Pages 390–413 in F.Berkes and C. Folke, editors. Linking Social andEcological Systems: Management Practices andSocial Mechanisms for Building Resilience. Cam-bridge University Press, Cambridge.

Adams, L.W. 1994. Urban Wildlife Habitats: ALandscape Perspective. University of MinnesotaPress, Minneapolis.

Addams, L., D. Battisti, E. McCullough, J.L. Min-jares, and G. Schoups. in press. Water resourcemanagement in the Yaqui Valley. in P.A. Matson,W.P. Falcon, editors. Agriculture, Developmentand the Environment in the Yaqui Valley (Mexico).Island Press, Washington.

Adger, W.N. 2006. Vulnerability. Global Environ-mental Change 16:268–281.

Adger, W.N., N.W. Arnell, and E.L. Tompkins. 2005.Successful adaptation to climate change acrossscales. Global Environmental Change 15:77–86.

Agar, N. 2001. Life’s Intrinsic Value: Science,Ethics and Nature. Columbia University Press,New York.

Agardy, T., J. Alder, P. Dayton, S. Curran, A.Kitchingman, et al. 2005. Coastal systems. Pages513–549 in R. Hassan, R. Scholes, and N. Ash, edi-tors. Ecosystems and Human Well-Being: CurrentState and Trends. Millennium Ecosystem Assess-ment. Island Press, Washington.

Agrawal, A. 2002. Commons resources and insti-tutional sustainability. Pages 41–86 in NRC(National Research Council) in E. Ostrom, T.Dietz, N. Dolsak, P.C. Stern, S. Stovich et al.,editors. The Drama of the Commons. Commit-tee on the Human Dimensions of Global Change.National Academy Press, Washington.

Agrawal, A., A. Chhatre, and R. Hardin. 2008.Changing governance of the world’s forests. Sci-ence 320:1460–1462.

Alimaev, I.I., and R.H.J. Behnke. 2008. Ideology,land tenure and livestock mobility in Kazakhstan.Pages 151–178 in K.A. Galvin, R.S. Reid, R.H.J.Behnke, and N.T. Hobbs, editors. Fragmentationin Semi-arid and Arid Landscapes: Consequencesfor Human and Natural Systems. Springer,Dordrecht.

Allan, C., and A. Curtis. 2005. Nipped in the bud:Whyregional-scale adaptive management is not bloom-ing. Environmental Management 36: 414–425.

Allan, J.A. 1998. Virtual water: A strategic resource.Global solutions to regional deficits. GroundWater 36:545–546.

Allen, M.R., and W.J. Ingram. 2002. Constraints AQ2on future changes in climate and the hydrologiccycle. Nature 419:224–232.

Allison, E.H., and F. Ellis. 2001. The livelihoodsapproach and management of small-scale fish-eries. Marine Policy 25:377–388.

355

356 References

Allison, H.E., and R.J. Hobbs. 2004. Resilience,adaptive capacity, and the ‘‘Lock-in Trap’’of the Western Australian agriculturalregion. Ecology and Society 9(1):3. http://www.ecologyandsociety.org/vol9/iss1/art3/.

Alverson, D.L., M.H. Freeberg, S.A. Murawski andJ.G. Pope. 1994. A global assessment of fisheriesbycatch and discards. FAP Fisheries TechnicalPaper 339, 235 pp.

AMAP (Arctic Monitoring and Assessment Pro-gramme). 2003. AMAP Assessment 2002: HumanHealth in the Arctic. AMAP, Oslo.

Amelung, T., and M. Diehl. 1992. Deforestationof Tropical Rain Forests: Economic Causes andImpact on Development. J.C.B. Mohr, Tubingen,Germany.

Anderies, J.M., M.A. Janssen, and E. Ostrom.2004. A framework to analyze the robustnessof social-ecological systems from an institutionalperspective. Ecology and Society 9(1):18. http://www.ecologyandsociety.org/vol19/iss11/art18/.

Anderson, E.N. 1996. Learning from the land otter:Religious representation of traditional resourcemanagement. Pages 54–72 in E.N. Anderson, edi-tor. Ecologies of the Heart: Emotions, Beliefs,and the Environment. Oxford University Press,Oxford.

Andersson, E., S. Barthel, and K. Ahrne. 2007. Mea-suring social-ecological dynamics behind the gen-eration of ecosystem services. Ecological Applica-tions 17:1267–1278.

Andow, D.A. 1991. Vegetational diversity andarthropod population response. Annual Review ofEntomology 36:561–586.

Angelsen, A., and D. Kaimowitz. 1999. Rethinkingthe causes of deforestation: Lessons from eco-nomic models. World Bank Research Observer14:73–98.

Argyris, C. 1992. On Organizational Learning.Blackwell Business, Oxford.

Argyris, C., and D.A. Schoen. 1978. Organiza-tional Learning: A Theory of Action Perspective.Addison-Wesley Publishing Company, Reading.

Armitage, D., F. Berkes, and N. Doubleday, edi-tors. 2007. Adaptive Co-Management: Collabora-tion, Learning, and Multi-Level Governance. Uni-versity of British Columbia Press, Vancouver.

Arrow, K., L. Goulder, P. Dasgupta, G. Daily, P.Ehrlich, et al. 2004. Are we consuming too much?Journal of Economic Perspectives 18:147–172.

Atkinson, B.A., G.A. Rose, E. Murphy, and C.A.Bishop. 1997. Distribution change and abundanceof northern cod (Gadus morhua): 1981–1993.Canadian Journal of Fisheries and Aquatic Sci-ences 54(Suppl. 1):132–138.

Attorre, F., A. Stanisci, and F. Bruno. 1997. Theurban woods of Rome. Plant Biosystems 131:113–135.

Baker, L.A., D. Hope, Y. Xu, J. Edmonds, andL. Lauver. 2001. Nitrogen balance for the Cen-tral Arizona-Phoenix Ecosystem. Ecosystems 4:582–602.

Bala, G., K. Caldeira, M. Wickett, T.J. Phillips,D.B. Lobell, et al. 2007. Combined climate andcarbon-cycle effects of large-scale deforestation.Proceedings of the National Academy of Sciences104:6550–6555.

Baland, J.-M., and J.-P. Platteau. 1996. HaltingDegradation of Natural Resources: Is there a Rolefor Rural Communities? Clarendon Press, Oxford.

Baliga, S., and E. Maskin. 2003. Mechanism designfor the environment. Pages 305–324 in K.-G.Maler and J.R. Vincent, editors. Handbook ofEnvironmental Economics. Elsevier, Amsterdam.

Balirwa, J.S., C.A. Chapman, L.J. Chapman, I.G.Cowx, K. Geheb, et al. 2003. Biodiversityand fishery sustainability in the Lake Victoriabasin: An unexpected marriage? BioScience 53:703–715.

Band, L.E., M.L. Cadenasso, S. Grimmond, and J.M.Grove. 2006. Heterogeneity in urban ecosystems:Pattern and process. Pages 257–278 in G. Lovett,C.G. Jones, M.G. Turner, and K.C. Weathers, edi-tors. Ecosystem Function in Heterogeneous Land-scapes. Springer-Verlag, New York.

Bandura, A. 1982. Self-efficacy mechanisms inhuman agency. American Psychologist 37:122–147.

Banerjee, A.V., and E. Duflo. 2007. The economiclives of the poor. Journal of Economic Perspec-tives 21:141–167.

Barbier, E.B. 1987. The concept of sustainable eco-nomic development. Environmental Conservation14:101–110.

Barbier, E.B., and J.R. Thompson. 1998. The valueof water: Floodplain versus large-scale irriga-tion benefits in northern Nigeria. Ambio 27:434–440.

Bardhan, P., and D. Mookerjee, editors. 2006.Decentralization and Local Governance in Devel-oping Countries. MIT Press, Cambridge, MA.

Bates, R.H. 1997. Open-Economy Politics: The Polit-ical Economy of the World Coffee Trade. Prince-ton University Press, Princeton.

Batty, M. 2006. Rank clocks. Nature 444:592–596.Batty, M. 2008. The size, scale, and shape of cities.

Science 319:769–771.Bawa, K.S., and S. Dayanandan. 1997. Socioeco- AQ3

nomic factors and tropical deforestation. Nature386:562–63.

References 357

Bawa, K.S., R. Seidler, and P.H. Raven. 2004. Rec-onciling conservation paradigms. ConservationBiology 18:859–860.

Bayliss, K., and T. McKinley. 2007. Providingbasic utilities in sub-Saharan Africa. Environment49(3):24–32.

Befu, H. 1980. The political ecology of fishing inJapan: Techno-environmental impact of industri-alization in the Inland Sea. Research in EconomicAnthropology 3:323–392.

Behnke, R.H.J. 2008. The drivers of fragmentationin arid and semi-arid landscapes. Pages 305–340 inK.A. Galvin, R.S. Reid, R.H.J. Behnke, and N.T.Hobbs, editors. Fragmentation in Semi-arid andArid Landscapes: Consequences for Human andNatural Systems. Springer, Dordrecht.

Bellwood, D., T.P. Hughes, C. Folke, and M.Nystrom. 2004. Confronting the coral reef crisis.Nature 429:827–833.

Bengtsson, J., P. Angelstam, T. Elmqvist, U.Emanuelsson, C. Folke, et al. 2003. Reserves,resilience, and dynamic landscapes. Ambio 32:389–396.

Benjamin, C., R. Orozco, R.F.M. Sales, and S.T.Jayme. 2003. Civil Society Participation in theFisheries and Aquatic Resource ManagementCouncil (FARMC) and the National Agricul-ture and Fisheries Council (NAFC): The Caseof Cavite and Bataan. Research conducted bythe Philippine Rural Reconstruction Movement(PRRM) for the NGO for Fishery Reform(NFR).

Bennett, E.M., S.R. Carpenter, and N.F. Caraco.2001. Human impact on erodable phosphorus andeutrophication: A global perspective. BioScience51:227–234.

Bennett, E.M., S.R. Carpenter, G.D. Peterson, G.S.Cumming, M. Zurek, et al. 2003. Why global sce-narios need ecology. Frontiers in Ecology and theEnvironment 1:322–329.

Bennett, E.M., T. Reed-Andersen, J.N. Houser, J.R.Gabriel, and S.R. Carpenter. 1999. A phosphorusbudget for the Lake Mendota watershed. Ecosys-tems 2:69–75.

Berger, J.R. 2001. African elephant, humaneconomies, and international law: Bridging thegreat rift for East and Southern Africa. George-town International Environmental Law Review13:417–462.

Berkes, F. 1988. Environmental philosophy of theCree people of James Bay. Pages 7–21 in M. Free-man and L. Carbyn, editors. Traditional Knowl-edge and Renewable Resource Management inNorthern Regions. Boreal Institute for NorthernStudies, Edmonton.

Berkes, F., editor. 1989. Common PropertyResources: Ecology and Community-BasedSustainable Development. Belhaven Press,London.

Berkes, F. 1995. Indigenous knowledge and resourcemanagement systems: A native Canadian casestudy from James Bay. Pages 99–109 in S. Hannaand M. Munasinghe, editors. Property Rights in aSocial and Ecological Context: Case Studies andDesign Applications. Beijer International Insti-tute of Ecological Economics and the WorldBank, Washington.

Berkes, F. 1998. Indigenous knowledge and resourcemanagement systems in the Canadian subarc-tic. Pages 98–128 in F. Berkes and C. Folke,editors. Linking Social and Ecological Systems:Management Practices and Social Mechanisms forBuilding Resilience. Cambridge University Press,Cambridge.

Berkes, F. 2002. Cross-scale institutional linkages:Perspectives from the bottom up. Pages 293–321in NRC. The Drama of the Commons. NationalAcademy Press, Washington.

Berkes, F. 2004. Rethinking community-based con-servation. Conservation Biology 18:621–630.

Berkes, F. 2007. Community-based conservation ina globalized world. Proceedings of the NationalAcademy of Sciences 104:15188–15193.

Berkes, F. 2008. Sacred Ecology: Traditional Ecolog-ical Knowledge and Resource Management. 2ndEdition. Taylor and Francis, Philadelphia.

Berkes, F., N. Bankes, M. Marschke, D. Armitage,and D. Clark. 2005. Cross-scale institutions:Building resilience in the Canadian North. Pages225–247 in F. Berkes, R. Huebert, H. Fast, M.Manseau, and A. Diduck, editors. Breaking Ice:Renewable Resource and Ocean Management inthe Canadian North. University of Calgary Press,Calgary.

Berkes, F., J. Colding, and C. Folke. 2000. Redis-covery of traditional ecological knowledge asadaptive management. Ecological Applications10:1251–1262.

Berkes, F., J. Colding, and C. Folke, editors. 2003.Navigating Social-Ecological Systems: BuildingResilience for Complexity and Change. Cam-bridge University Press, Cambridge.

Berkes, F., and I.J. Davidson-Hunt. 2006. Biodi-versity, traditional management systems, and cul-tural landscapes: Examples from the boreal for-est of Canada. International Social Science Journal187:35–47.

Berkes, F., D. Feeny, B.J. McCay, and J.M. Ache-son. 1989. The benefits of the commons. Nature340:91–93.

358 References

Berkes, F., and C. Folke. 1998a. Linking socialand ecological systems for resilience and sustain-ability. Pages 1–25 in F. Berkes and C. Folke,editors. Linking Social and Ecological Systems:Management Practices and Social Mechanisms forBuilding Resilience. Cambridge University Press,Cambridge.

Berkes, F., and C. Folke, editors. 1998b. Link-ing Social and Ecological Systems: ManagementPractices and Social Mechanisms for BuildingResilience. Cambridge University Press, Cam-bridge.

Berkes, F., and C. Folke. 2002. Back to the future:Ecosystem dynamics and local knowledge. Pages121–146 in L.H. Gunderson and C.S. Holling, edi-tors. Panarchy: Understanding Transformations inHuman and Natural Systems. Island Press, Wash-ington.

Berkes, F., T.P. Hughes, R.S. Steneck, J.A. Wilson,D.R. Bellwood, et al. 2006. Globalization, rov-ing bandits, and marine resources. Science 311:1557–1558.

Berkowitz, A.R., C.H. Nilon, and K.S. Holweg, edi-tors. 2003. Understanding Urban Ecosystems: ANew Frontier for Science and Education. Springer-Verlag, New York.

Bernstein, S.F. 2002. The Compromise of LiberalEnvironmentalism. Columbia University Press,New York.

Berry, W. 2005. Renewing husbandry: After mecha-nization, can modern agriculture reclaim its soul?Crop Science 45:1103–1106.

Bhagwat, S., C. Kushalappa, P. Williams, and N.Brown. 2005. The role of informal protectedareas in maintaining biodiversity in the West-ern Ghats of India. Ecology and Society 10 (1):8.http://www.ecologyandsociety.org/vol10/iss1/art8/.

Biermann, F. 2007. Earth system governance asa crosscutting theme of global change research.Global Environmental Change 17:326–337.

Biggs, H.C., and K.H. Rogers. 2003. An adaptivesystem to link science, monitoring, and manage-ment in practice. Pages 59–80 in J.T. du Toit,K.H. Rogers, and H.C. Biggs, editors. The KrugerExperience: Ecology and Management of SavannaHeterogeneity. Island Press, Washington.

Binswanger, H. 1990. The policy response in agricul-ture. Pages 231–258 in Proceedings of the WorldBank Annual Conference on Development Eco-nomics 1989. World Bank, Washington.

Birkeland, P.W. 1999. Soils and Geomorphology. 3rdEdition. Oxford University Press, New York.

Bodin, O., and J. Norberg. 2005. Information net-work topologies for enhanced local adaptive

management. Environmental Management 35:175–193.

Bolund, P., and S. Hunhammar. 1999. Ecosystemservices in urban areas. Ecological Economics29:293–301.

Bonfil, R., G. Munro, U.R. Sumaila, H. Valtysson,M. Wright, et al. 1998. Impacts of distant waterfleets: An ecological, economic and social assess-ment. Pages 11–111 in The Footprint of Dis-tant Water Fleets on World Fisheries. EndangeredSeas Campaign, WWF International, Godalming,UK.

Boone, C. 2002. An assessment and explanationof environmental inequity in Baltimore. UrbanGeography 23:581–595.

Borgerhoff Mulder, M., and P. Coppolillo. 2005.Conservation: Linking Ecology, Economics, andCulture. Princeton University Press, Princeton,NJ.

Borgstrom, S.T., T. Elmquist, P. Angelstrom, and C.Alfsen-Norodom. 2006. Scale mismatches in man-agement of urban landscapes. Ecology and Soci-ety 11(2):16. http://www.ecologyandsociety.org/vol11/iss2/art16/.

Bormann, B.T., and A.R. Kiester. 2004. Optionsforestry: Acting on uncertainty. Journal ofForestry 102:22–27.

Bormann, F.H., and G.E. Likens. 1967. Nutrientcycling. Science 155:424–429.

Bormann, F.H., and G.E. Likens. 1979. Patternsand Processes in a Forested Ecosystem. Springer-Verlag, New York.

Boudouris, K., and K. Kalimtzis, editors. 1999. Phi-losophy and Ecology. Iona Publications, Athens.

Bowles, S., S. Durlauf, and K. Hoff, editors.2006. Poverty Traps. Princeton University Press,Princeton.

Boyden, S., S. Millar, K. Newcombe, and B. O’Neill.1981. The Ecology of A City and Its People: TheCase of Hong Kong. Australian National Univer-sity Press, Canberra.

Bradshaw, A.D. 1983. The reconstruction of ecosys-tems. Journal of Ecology 20:1–17.

Brady, N.C., and R.R. Weil. 2001. The Nature andProperties of Soils. 13th Edition. Prentice Hall,Upper Saddle River, NJ.

Branch, T.A., R. Hilborn, A.C. Haynie, G. Fay, L. AQ4Flynn, et al. 2006. Fleet dynamics and fishermenbehavior: Lessons for fisheries managers. Cana-dian Journal of Fisheries and Aquatic Sciences63:1647–1668.

Brand, S. 2006. City Planet. strategy + busi-ness. http://www.strategy-business.com/press/16635507/16606109.

References 359

Brock, W.A. 2006. Tipping points, abrupt opinionchanges, and punctuated policy change. Pages47–77 in R. Repetto, editor. Punctuated Equilib-rium and the Dynamics of U.S. EnvironmentalPolicy. Yale University Press, New Haven.

Brodie, J., K. Fabricius, G. De’ath, and K. Okaji.2005. Are increased nutrients responsible formore outbreaks of crown-of-thorns starfish?Marine Pollution Bulletin 51:266–278.

Bromley, D. 1989. Economic Interests and Institu-tions: The Conceptual Foundations of Public Pol-icy. Basil Blackwell, New York.

Bromley, D.W., editor. 1992. Making the CommonsWork: Theory Practice and Policy. Institute forContemporary Studies, San Francisco.

Brouzes, R.J.P., A.J. Liem, and V.A. Naish. 1978.Protocol for fish tainting bioassays. CPAR Report.No. 775–1. Environmental Protection Service,Ottawa.

Brown, K. 2002. Innovations for conservation anddevelopment. The Geographical Journal 168:6–17.

Brown, K. 2003. Integrating conservation and devel-opment: A case of institutional misfit. Frontiers inEcology and the Environment 1:479–487.

Brunner, R.D., T.A. Steelman, L. Coe-Juell, C.M.Cromley, C.M. Edwards, et al., editors. 2005.Adaptive Governance: Integrating Science, Policy,and Decision Making. Columbia University Press,New York.

Burch, W.R. Jr., and D.R. DeLuca. 1984. Measur-ing the Social Impact of Natural Resource Policies.New Mexico University Press, Albuquerque.

Burke, E.J., S.J. Brown, and N. Christidis. 2006.Modeling the recent evolution of global droughtand projections for the twenty-first century withthe Hadley Centre climate model. Journal ofHydrometeorology 7:1113–1125.

Burke, M., M. Mastrandrea, W.P. Falcon, and R.Naylor. 2008. Ending hunger by assumption. Pro-gram on Food Security and the Environment(under review).

BurnSilver, S.B. In press. Pathways of changeAQ5and continuity: Maasai livelihoods in Amboseli,Kajaido District, Kenya. Pages in K. Home-AQ6wood, P. Kristjanson, and P. Chevevix Trench, edi-tors. Staying Maasai? Livelihoods, Conservationand Development in East African Rangelands.Springer, New York.

Burton, I., R.W. Kates, and G.F. White. 1993. TheEnvironment as Hazard. 2nd Edition. Guilford,New York.

Burton, P.J., C. Messier, G.F. Weetman, E.E. Prepas,W.L. Adamowicz, et al. 2003. The current state of

boreal forestry and the drive for change. Pages1–40 in P.J. Burton, C. Messier, D.W. Smith,and W.L. Adamowicz, editors. Towards Sustain-able Management of the Boreal Forest. NationalResearch Council of Canada, Ottawa.

Busch, D.E., and J.C. Trexler. 2002. Monitor-ing Ecosystems: Interdisciplinary Approaches forEvaluation Ecoregional Initiatives. Island Press,Washington.

Butzer, K.W. 1980. Adaptation to global envi-ronmental change. Professional Geographer 32:269–278.

Caddy, J.F., and Surette, T. 2005. In retrospect theassumption of sustainability for Atlantic fisherieshas proved an illusion. Reviews in Fish Biologyand Fisheries 15:313–337.

Cadenasso, M.L., S.T.A. Pickett, and J.M. Grove.2006. Dimensions of ecosystem complexity:Heterogeneity, connectivity, and history. Ecologi-cal Complexity 3:1–12.

Campbell, R.A., B.D. Mapstone, and A.D.M.Smith. 2001. Evaluating large-scale experimen-tal designs for management of coral trout on theGreat Barrier Reef. Ecological Applications 11:1763–1777.

Campbell, B.D., D.M. Stafford Smith, and GCTEPastures and Rangelands Network members.2000. A synthesis of recent global change researchon pasture and rangeland production: Reduceduncertainties and their management implications.Agriculture, Ecosystems and Environment 82:39–55.

Canadell, J.G., D.E. Pataki, R. Gifford, R.A.Houghton, Y. Luo, et al. 2007. Saturation ofthe terrestrial carbon sink. Pages 59–78 in J.G.Canadell, D. Pataki, and L. Pitelka, editors.Terrestrial Ecosystems in a Changing World.Springer-Verlag, Berlin.

Carpenter, S.R. 2003. Regime Shifts in Lake Ecosys-tems: Pattern and Variation. Ecology Institute,Oldendorf/Luhe, Germany.

Carpenter, S.R., E.M. Bennett, and G.D. Peter-son. 2006a. Scenarios for ecosystem services: Anoverview. Ecology and Society 11(1):29. http://www.ecologyandsociety.org/vol11/iss1/art29/.

Carpenter, S.R., D. Bolgrien, R.C. Lathrop, C.A.Stow, T. Reed, et al. 1998a. Ecological and eco-nomic analysis of lake eutrophication by non-point pollution. Australian Journal of Ecology 23:68–79.

Carpenter, S.R., and W.A. Brock. 2004. Spatial com-plexity, resilience and policy diversity: Fishing onlake-rich landscapes. Ecology and Society 9(1):8.http://www.ecologyandsociety.org/vol9/iss1/art8/.

360 References

Carpenter, S.R., W.A. Brock, and D. Ludwig. 2002.Collapse, learning, and renewal. Pages 173–193 inL.H. Gunderson and C.S. Holling, editors. Panar-chy: Understanding Transformations in Humanand Natural Systems. Island Press, Washington.

Carpenter, S.R., N.F. Caraco, D.L. Correll, R.W.Howarth, A.N. Sharpley, and V.H. Smith. 1998b.Non-point pollution of surface waters with phos-phorus and nitrogen. Ecological Applications8:559–568.

Carpenter, S.R., and L.H. Gunderson. 2001. CopingAQ7with collapse: Ecological and social dynamics inecosystem management. BioScience 51:451–457.

Carpenter, S.R., and J.F. Kitchell. 1993. The TrophicCascade in Lakes. Cambridge University Press,Cambridge.

Carpenter, S.R., and R.C. Lathrop. 1999. Lakerestoration: Capabilities and needs. Hydrobiolo-gia 395/396:19–28.

Carpenter, S.R., R.C. Lathrop, P. Nowak, E.M. Ben-nett, T. Reed, et al. 2006b. The ongoing experi-ment: Restoration of Lake Mendota and its water-shed. Pages 236–256 in J.J. Magnuson, T.K. Kratz,and B.J. Benson, editors. Long-Term Dynamics ofLakes in the Landscape. Oxford University Press,London.

Carpenter, S.R., D. Ludwig, and W.A. Brock. 1999.Management of eutrophication for lakes sub-ject to potentially irreversible change. EcologicalApplications 9:751–771.

Carpenter, S.R., and M.G. Turner. 2000. Hares andtortoises: Interactions of fast and slow variables inecosystems. Ecosystems 3:495–497.

Carr, C. 2008. Kalashnikov Culture: Small ArmsProliferation and Irregular Warfare. Greenwood,Portsmouth, NH.

Carson, R. 1962. Silent Spring. Crest, New York.Cassman, K.G. 1999. Ecological intensification of

cereal production systems: Yield potential, soilquality, and precision agriculture. Proceedings ofthe National Academy of Sciences 96:5952–5959.

Cassman, K.G., A. Dobermann, D.T. Walters, andH. Yang. 2003. Meeting cereal demand while pro-tecting natural resources and improving environ-mental quality. Annual Review of Environmentand Resources 28:316–351.

Cassman, K.G., V. Eidman, and E. Simpson. 2006.Convergence of agriculture and energy: Implica-tions for research and policy. Council for Agricul-tural Science and Technology (CAST) Commen-tary, pp. 12.

Cassman, K.G., and A.J. Liska. 2007. Food and fuelfor all: Realistic or foolish? Biofuels, Bioproducts,and Biorefining 1(1):18–23.

Cassman, K.G., S. Wood, P.S. Choo, H.D. Cooper,C. Devendra, et al. 2005. Cultivated systems.Pages 745–794 in R. Hassan, R.J. Scholes, and N.Ash, editors. Ecosystems and Human Well-Being:Current State and Trends. Millennium EcosystemAssessment. Island Press, Washington.

Chambers, R., and G. Conway. 1991. SustainableRural Livelihoods: Practical Concepts for the 21stCentury. IDS, Brighton.

Chambers, S. 1998. Short- and Long-Term Effects ofClearing Native Vegetation for Agricultural Pur-poses. PhD. Dissertation. Flinders University ofSouth Australia, Adelaide.

Chan, S. 2007. Considering the Urban Planet 2050.New York Times City Blog. Posted December 4,2007, 6:11 pm. http://cityroom.blogs.nytimes.com/2007/12/04/considering-the-urban-planet-of-2050/?scp=1-b&sq=&st=nyt.

Chapin, F.S., III, and K. Danell. 2001. Boreal forest.Pages 101–120 in F.S. Chapin, III, O.E. Sala, andE. Huber-Sannwald, editors. Global Biodiversityin a Changing Environment: Scenarios for the 21stCentury. Springer-Verlag, New York.

Chapin, F.S., III, K. Danell, T. Elmqvist, C. Folke,and N.L. Fresco. 2007. Managing climate changeimpacts to enhance the resilience and sustainabil-ity of Fennoscandian forests. Ambio 36:528–533.

Chapin, F.S., III, M. Hoel, S.R. Carpenter, J.Lubchenco, B.Walker, et al. 2006a. Buildingresilience and adaptation to manage arcticchange. Ambio 35:198–202.

Chapin, F.S., III, A.L. Lovecraft, E.S. Zavaleta, J.Nelson, M.D. Robards, et al. 2006b. Policy strate-gies to address sustainability of Alaskan borealforests in response to a directionally changing cli-mate. Proceedings of the National Academy of Sci-ences 103:16637–16643.

Chapin, F.S., III, P.A. Matson, and H.A. Mooney.2002. Principles of Terrestrial Ecosystem Ecology.Springer-Verlag, New York.

Chapin, F.S., III, J.T. Randerson, A.D. McGuire, J.A.Foley, and C.B. Field. 2008. Changing feedbacksin the earth-climate system. Frontiers in Ecologyand the Environment 6:313–320.

Chapin, F.S., III, and G.R. Shaver. 1985. Individu-alistic growth response of tundra plant species toenvironmental manipulations in the field. Ecology66:564–576.

Chapin, F.S., III, M.S. Torn, and M. Tateno. 1996.Principles of ecosystem sustainability. AmericanNaturalist 148:1016–1037.

Chapin, F.S., III, E.S. Zaveleta, V.T. Eviner, R.L.Naylor, P.M. Vitousek, et al. 2000. Consequencesof changing biotic diversity. Nature 405:234–242.

References 361

Charles, A. 2001. Sustainable Fishery Systems.AQ8Blackwell Science, Oxford.

Chase-Dunn, C. 2000. Guatemala in the global sys-tem. Journal of Interamerican Studies and WorldAffairs 42:109–126.

Chen, S., and M. Ravallion. 2007. Absolute povertymeasures for the developing world, 1981–2004.Proceedings of the National Academy of Sciences104:16757–16762.

Chevalier, J., and D. Buckles. 1999. Conflict manage-ment: A heterocultural perspective. Pages 13–41in D. Buckles, editor. Cultivating Peace: Conflictand Collaboration in Natural Resource Manage-ment. International Development Research Cen-tre, Ottawa.

Choi, J.S., K.T. Frank, W.C. Leggett, and K.Drinkwater. 2004. Transition to an alternate statein a continental shelf ecosystem. Canadian Jour-nal of Fisheries and Aquatic Sciences 61:505–510.

Choi, Y.D. 2007. Restoration ecology to the future:A call for a new paradigm. Restoration Ecology15:351–353.

Chou, C., and J.D. Neelin. 2004. Mechanisms ofAQ9global warming impacts on regional tropical pre-cipitation. Journal of Climate 17:2688–2701.

Christensen, N.L., A.M. Bartuska, J.H. Brown, S.Carpenter, C. D’Antonio, et al. 1996. The reportof the Ecological Society of America committeeon the scientific basis for ecosystem management.Ecological Applications 6:665–691.

Christensen, V., S. Guenette, J.J. Heymans, C.J. Wal-ters, R. Watson, et al. 2003. Hundred-year declineof North Atlantic predatory fishes. Fish and Fish-eries 4(1):1–24.

Christy, F. 1982. Territorial Use Rights in Marine Fish-eries: Definitions and Conditions. Food and Agri-culture Association Fisheries Technical Paper 227.Food and Agriculture Association, Rome.

Christy, F.T. 1997. Economic waste in fisheries:Impediments to change and conditions forimprovement. Pages 28–39 in E.K. Pikitch, D.D.Huppert and M.P. Sissenwine, editors. GlobalTrends: Fisheries Management. American Fish-eries Society, Bethesda, MD.

CIA. 2007. The World Factbook. Central Intelli-gence Agency, Washington.

Cilliers, S.S., and G.J. Bredenkamp. 1999. Analysisof the spontaneous vegetation intensively man-aged open spaces in the Potchefstroom MunicipalArea, North West Province, South Africa. SouthAfrican Journal of Botany 65:59–68.

Cinner, J., M.J. Marnane, T.R. McClanahan,and G.R. Almany. 2005. Periodic closures asadaptive coral reef management in the Indo-

Pacific. Ecology and Society 11(1):31. http://www.ecologyandsociety.org/vol11/iss1/art31/.

Cissel, J.H., F.J. Swanson, and P.J. Weisberg.1999. Landscape management using historical fireregimes: Blue River, Oregon. Ecological Applica-tions 9:1217–1231.

Clallam County Department of Commu-nity Development. 2004. State of theWaters of Clallam County: A Report onthe Health of Our Streams & Water-sheds. http://www.clallam.net/streamkeepers/html/state of the waters.htm.

Clark, C.W. 1985. Bioeconomics Modeling and Fish-ery Management. Wiley Interscience, New York.

Clark, C.W. 2005. Mathematical Bioeconomics: The AQ10Optimal Management of Renewable Resources.2nd Edition. Wiley-Interscience, New York.

Clark, C.W. 2007. The Worldwide Crisis in Fisheries:Economic Models and Human Behavior. Cam-bridge University Press, Cambridge.

Clark, W.C., and N.M. Dickson. 2003. Sustainabil-ity science: The emerging research program. Pro-ceedings of the National Academy of Sciences100:8059–8061.

Clay, G. 1973. Close Up: How to Read the AmericanCity. Praeger Publishers, New York.

Cohen, F. 1986. Treaties on Trial. American FriendsService Committee. University of WashingtonPress, Seattle.

Cohen, M.D., J.G. March, and J.P. Olsen. 1972.A garbage can model of organizational choice.Administrative Science Quarterly 17:1–25.

Colding, J. 2007. Ecological land-use complementa-tion’ for building resilience in urban ecosystems.Landscape and Urban Planning 81: 46–55.

Colding, J., J. Lundberg, and C. Folke. 2006. Incor-porating green-area user groups in urban ecosys-tem management. Ambio 35:237–244.

Cole, J.J., Y.T. Prairie, N.F. Caraco, W.H. McDowell, AQ11L.J. Tranvik, et al. 2007. Plumbing the global car-bon cycle: Integrating inland waters into the ter-restrial carbon budget. Ecosystems 10: 172–175.

Cole, M.M. 1982. The influence of soils, geomorpho-logy and geology on the distribution of plant com-munities in Savanna ecosystems. Pages 145–174 inB.J. Huntley and B.H. Walker, editors. Ecology ofTropical Savannas. Springer-Verlag, Berlin.

Coleman, J. 1990. Foundations of Social Theory.Harvard University Press, Cambridge, MA.

Collier, P. 2007. The Bottom Billion: Why the Poor-est Countries Are Failing and What Can Be DoneAbout It. Oxford University Press, Oxford.

Collins, S., S.M. Swinton, C.W. Anderson, B.J. Ben-son, T.L. Gragson, et al. 2007. Integrated Sci-

362 References

ence for Society and the Environment: A strategicresearch initiative. LTER Network Office, Albu-querque, NM.

Contreras-Hermosilla, A. 2002. Law Compliancein the Forestry Sector: An Overview. The WorldBank, Washington.

Conway, G. 1997. The Doubly Green Revolution:Food for All in the 21st Century. Cornell Univer-sity Press, Ithaca.

Cooke, G.D., E.B. Welch, S. Peterson, and S.A.Nichols. 2005. Restoration and Management ofLakes and Reservoirs. 3rd Edition. CRC Press,Boca Raton, FL.

Copes, P. 1986. A critical review of the individualAQ12quota as a device in fisheries management. LandEconomics 62:278–291.

Corcoran, P.J. 2006. Opposing views of the “Ecosys-tem Approach” to fisheries management. Conser-vation Biology 20:617–619.

Cordell, J. editor. 1989. A Sea of Small Boats. Cul-tural Survival, Inc. Cambridge, MA.

Cork, S.J., G.D. Peterson, E.M. Bennett, G. Petschel-Held, and M. Zurek. 2006. Synthesis of thestorylines. Ecology and Society 11(2):11. http://www.ecologyandsociety.org/vol11/iss2/art11/.

Costa, M.H., A. Botta, and J.A. Cardille. 2003.Effects of large-scale changes in land cover on thedischarge of the Tocantins River, SoutheasternAmazonia. Journal of Hydrology 283:206–217.

Costa, M.H., and J.A. Foley. 1999. Trends in thehydrological cycle of the Amazon basin. Journalof Geophysical Research 104:14189–14198.

Costanza, R., editor. 1991. Ecological Economics:The Science and Management of Sustainability.Columbia University Press, New York.

Costanza, R., and H. Daly. 1992. Natural capitaland sustainable development. Conservation Bio-logy 6:37–46.

Costanza, R., R. d’Arge, R. de Groot, S. Farber,M. Grasso, et al. 1997. The value of the world’secosystem services and natural capital. Nature387:253–260.

Costanza, R., and C. Folke. 1996. The structureand function of ecological systems in relation toproperty-rights regimes. Pages 13–34 in S. Hanna,C. Folke, K.-G. Maler, and A. Jansson, editors.Rights to Nature: Ecological, Economic, Cultural,and Political Principles of Institutions for the Envi-ronment. Island Press, Washington.

Couzin, J. 2008. Living in the danger zone. Science319:748–749.

Crona, B. 2006. Supporting and enhancing devel-opment of heterogeneous ecological knowledgeamong resource users in a Kenyan seascape.

Ecology and Society 11(1):32 [online] URL:http://ecologyandsociety.org/vol11/iss1/art32/.

Cronon, W. 1991. Nature’s Metropolis: Chicago andthe Great West. W.W. Norton & Co., New York.

Crutzen, P. 2002. Geology of Mankind – TheAnthropocene. Nature 415:23.

Cutter, S.L. 1996. Vulnerability to environmen-tal hazards. Progress in Human Geography 20:529–539.

Daily, G.C., editor. 1997. Nature’s Services: SocietalDependence on Natural Ecosystems. Island Press,Washington.

D’Antonio, C.M., and P.M. Vitousek. 1992. Biologi-cal invasions by exotic grasses, the grass-fire cycle,and global change. Annual Review of Ecology andSystematics 23:63–87.

Dasgupta, P. 2001. Human Well-Being and theNatural Environment. Oxford University Press,Oxford.

Dasgupta, P., and K.-G. Maler. 2000. Net nationalproduct, wealth, and social well-being. Environ-ment and Development Economics 5:69–93.

Davidson, R. 2006. No fixed address: Nomadsand the fate of the planet. Quarterly Essay 24:1–53.

Davidson-Hunt, I.J., P. Jack, E. Mandamin, and B.Wapioke. 2005. Iskatewizaagegan (Shoal Lake)plant knowledge: An Anishinaabe (Ojibway) eth-nobotany of northwestern Ontario. Journal ofEthnobiology 25:189–227.

Davis, M. 2006. Planet of Slums. Verso, London.DeAngelis, D.L., and W.M. Post. 1991. Positive feed-

back and ecosystem organization. Pages 155–178in M. Higashi and T.P. Burns, editors. TheoreticalStudies of Ecosystems: The Network Perspective.Cambridge University Press, Cambridge.

Degnbol, P. 2003. Science and the user perspec-tive: The gap co-management must address.Pages 31–50 in D.C. Wilson, J.R. Nielsen, and P.Degnbol, editors. The Fisheries Co-ManagementExperience: Accomplishments, Challenges andProspects. Kluwer, Dordrecht.

de Groot, R., P.S. Ramakrishnan, A. van de Berg,T. Kulenthran, and S. Muller, 2005. Cultural andamenity services. Pages 455–476 in R. Hassan,R. Scholes, and N. Ash, editors. Ecosystems andHuman Well-Being: Current State and Trends.Millennium Ecosystem Assessment. Island Press,Washington.

Dercon, S. 2004. Growth and shocks: Evidencefrom rural Ethiopia. Journal of Development Eco-nomics 74:309–329.

Deutsch, B. 2007. A Threat So Big, Academics Try toCollaborate. The New York Times, New York.

References 363

Deutsch, C., J.J. Tewksbury, R.B. Huey, K. Sheldon,AQ13C. Ghalambor, D.C. Haak, et al. 2008. Cli-mate change will be most detrimental to tropicalecotherms. Nature (in review).

Deutsch, L., S. Graslund, C. Folke, M. Troell, M.Huitric, et al. 2007. Feeding aquaculture growththrough globalization: Exploitation of marineecosystems for fishmeal. Global EnvironmentalChange 17:238–249.

Deutsch, L., A. Jansson, M. Troell, P. Ronnback,C. Folke, et al. 2000. The “ecological footprint”:Communicating human dependence on nature’swork. Ecological Economics 32: 351–355.

de Young, B., R.M. Peterman, A.R. Dobell, E.Pinkerton, Y. Breton, et al. 1999. CanadianMarine Fisheries in a Changing and UncertainWorld. Canadian Special Publication of Fisheriesand Aquatic Sciences 129:199.

Diamond, J. 1997. Guns, Germs and Steel. W.W. Nor-AQ14ton & Co. New York.

Diamond, J. 2005. Collapse: How Societies Chooseor Fail to Succeed. Viking, New York.

Dıaz, S., J. Fargione, F.S. Chapin, III, and D.Tilman. 2006. Biodiversity loss threatens humanwell-being. Plant Library of Science (PLoS) 4:1300–1305.

Diener, E., and M.E.P. Seligman. 2004. Beyondmoney: Toward an economy of well-being. Psy-chological Science in the Public Interest 5:1–31.

Dietz, T., E. Ostrom, and P.C. Stern. 2003. The strug-gle to govern the commons. Science 302:1907–1912.

Dirzo, R., and P.H. Raven. 2003. Global state of bio-diversity and loss. Annual Review of the Environ-ment and Resources 28:137–167.

D’Odorico, P., and A. Porporato, editors. 2006. Dry-land Ecohydrology. Springer, Dordrecht.

Douglas, M. 1986. How Institutions Think. SyracuseUniversity Press, Syracuse.

Dow, K. 2000. Social dimensions of gradientsin urban ecosystems. Urban Ecosystems 4:255–275.

Downing, J.A., Y.T. Prairie, J.J. Cole, C.M. Duarte,L.J. Tranvik, et al. 2006. The global abundanceand size distribution of lakes, ponds and impound-ments. Limnology and Oceanography 51:2388–2397.

Driscoll, C.T., G.B. Lawrence, A.J. Bulger, T J.Butler, C.S. Cronan, et al. 2001. Acidic depositionin the northeastern United States: Sources andinputs, ecosystem effects and management strate-gies. BioScience 51:180–198.

Duit, A., and V. Galaz (2008). Governing com-AQ15plexity: Insights and emerging challenges. Gover-nance.

Duncan, O.D. 1961. From social system to ecosys-tem. Sociological Inquiry 31:140–149.

Duncan, O.D. 1964. Social organization and theecosystem. Pages 37–82 in R.E.L. Faris, editor.Handbook of Modern Sociology. Rand McNally& Co., Chicago.

du Toit, J.T., K.H. Rogers, and H.C. Biggs, editors.2003. The Kruger Experience: Ecology and Man-agement of Savanna Heterogeneity. Island Press,Washington.

Dyson, M., G. Bergkamp, and J. Scanlon, edi-tors. 2007. Flow: The Essentials of EnvironmentalFlows. IUCN, Gland, Switzerland.

Eagle, J., R. Naylor, and W. Smith. 2004. Why farmsalmon out-compete fishery salmon. Marine Pol-icy 28:269–270.

Easterlin, R.A. 2001. Income and happiness:Towards a unified theory. Economic Journal111:465–484.

Easterling, W.E., P.K. Aggarwal, P. Batima, K.M.Brander, L. Erda, et al. 2007. Food, fibre andforest products. Pages 273–313 in M.L. Parry,O.F. Canziani, J.P. Palutikof, P.J. van der Linden,and C.E. Hanson, editors. Climate Change 2007:Impacts, Adaptation and Vulnerability. Contribu-tion of Working Group II to the Fourth AssessmentReport of the Intergovernmental Panel on ClimateChange. Cambridge University Press. Cambridge.

Ebbin, S. 1998. Emerging Cooperative Institutionsfor Fisheries Management: Equity and Empower-ment of Indigenous Peoples of Washington andAlaska. PhD. dissertation. Yale University, NewHaven.

Ehrlich, P.R., and J.P. Holdren. 1974. Impact of pop-ulation growth. Science 171:1212–1217.

EIA. 2006. International Energy Annual 2004. Inter-national Energy Administration, Washington.

Ellis, E.C., and N. Ramankutty. 2008. Putting peo-ple on the map: Anthropogenic biomes of theworld. Frontiers in Ecology and the Environment6:doi:10.1890/070062.

Ellis, F. 1998. Household strategies and rural diver-sification. Journal of Developmental Studies 35:1–38.

Ellis, J. 1994. Climate variability and complexecosystem dynamics: Implications for pastoraldevelopment. Pages 37–46 in I. Scoones, editor.Living with Uncertainty. Intermediate TechnologyPublishers, London.

Ellison, A.M., M.S. Bank, B.D. Clinton, E.A.Colburn, K. Elliott, et al. 2005. Loss of foun-dation species: Consequences for the structureand dynamics of forested ecosystems. Frontiers inEcology and the Environment 3:479–486.

364 References

Elmqvist, T., J. Colding, S. Barthel, S. Borgstrom,A. Duit, et al. 2004. The dynamics of social-ecological systems in urban landscapes: Stock-holm and the National Urban Park, Sweden.Annals of New York Academy of Sciences1023:308–322.

Elmqvist, T., C. Folke, M. Nystrom, G. Peterson, J.Bengtsson, et al. 2003. Response diversity, ecosys-tem change, and resilience. Frontiers in Ecologyand the Environment 1:488–494.

Elmqvist, T., M. Pyykonen, M. Tengo, F.Rakotondrasoa, E. Rabakonandrianina, et al.2007. Patterns of loss and regeneration of tropicaldry forest in Madagascar: The social institu-tional context. Plant Library of Science (PLoS)2:e402:doi:410.1371/journal.pone.0000402.

Enfors, E. and L. Gordon. 2007. Analysingresilience in dryland agro-ecosystems: A casestudy of the Makanya catchment in Tanzaniaover the past 50 years. Land Degradation &Development 18:680–696.

Engerman, S., and K.L. Sokoloff. 2006. The persis-tence of poverty in the Americas: The role of insti-tutions. Pages 43–78 in S. Bowles, S. Durlauf, andK. Hoff, editors. Poverty Traps. Princeton Univer-sity Press, Princeton.

Epstein, M.J. 2008. Making Sustainability Work: BestPractices in Managing and Measuring CorporateSocial, Environmental, and Economic Impacts.Greenleaf Publishing Ltd, Sheffield, UK.

Evans, L.T. 1998. Feeding the Ten Billion: Plants andPopulation Growth. Cambridge University Press.Cambridge.

Ewel, J.J. 1999. Natural systems as models for thedesign of sustainable systems of land use. Agro-forestry Systems 45:1–21.

Fabricius, C., C. Folke, G. Cundill, and L. Schultz.2007. Powerless spectators, coping actors, andadaptive co-managers: A synthesis of therole of communities in ecosystem manage-ment. Ecology and Society 12(1):29. http://www.ecologyandsociety.org/vol12/iss1/art29/.

Falcon, W.P. 1991. Whither food aid? Pages 237–246in C.P. Timmer, editor. Agriculture and the State:Growth, Employment, and Poverty in DevelopingCountries. Cornell University Press. Ithaca.

Falcon, W.P., and R.L. Naylor. 2005. Rethinkingfood security for the 21st century. American Jour-nal of Agricultural Economics 87:1113–1127.

Falkenmark, M., and J. Rockstrom. 2004. Balanc-ing Water for Humans and Nature: The NewApproach in Ecohydrology. Earthscan, London.

FAO (Food and Agriculture Organization). 1995.Code of Conduct for Responsible Fisheries. FAO,Rome.

FAO (Food and Agriculture Organization). 2004.The State of Food Insecurity in the World. Mon-itoring Progress towards the World Food Sum-mit and the Millennium Development Goals. FAO.Rome.

FAO (Food and Agriculture Organization). 2006.The State of the World Fisheries and Aquaculture.United Nations FAO. Rome.

FAOSTAT. 2007. FAO Statistical Database. FAO,Rome.

Farber, S., C. Costanza, D.L. Childers, J. Erickson,K. Gross, et al. 2006. Linking ecology and eco-nomics for ecosystem management: A services-based approach with illustrations from LTERsites. BioScience 56:117–129.

Fargher, J.D., B.M. Howard, D.G. Burnside, andM.H. Andrew. 2003. The economy of Australianrangelands: Myth or mystery? Rangeland Journal25:140–156.

Felson, A.J., and S.T.A. Pickett. 2005. Designedexperiments: New approaches to studying urbanecosystems. Frontiers in Ecology and the Environ-ment 10:549–586.

FEMAT (Forest Ecosystem Management Assess-ment Team). 1993. Forest ecosystem manage-ment: an ecological, economic, and social assess-ment. USDA Forest Service: U.S. Department ofCommerce, National Oceanic and AtmosphericAdministration, National Marine Fisheries Ser-vice; USDI Bureau of Land Management, Fishand Wildlife Service, National Park Service, Envi-ronmental Protection Agency, Portland, OR.

Fernandez, M., and J.C. Castilla. 2005. Marineconservation in Chile: Historical perspective,lessons, and challenges. Conservation Biology 19:1752–1762.

Fernandez, R., E.R.M. Archer, A.J. Ash, H.Dowlatabadi, P.H.Y. Hiernaux, et al. 2002. Degra-dation and recovery in socio-ecological systems:A view from the household/farm level. Pages 297–323 in J.F. Reynolds and D.M. Stafford Smith,editors. Global Desertification: Do HumansCause Deserts? Dahlem University Press,Berlin.

Fernholz, K. 2007. TIMOs & REITs: What,why, & how they might impact sustain-able forestry. Dovetail Partners. http://www.dovetailinc.org/reportView.php?action=displayReport&reportID=78.

Field, C.B., D.B. Lobell, H.A. Peters, and N.R.Chiariello. 2007. Feedbacks of terrestrial ecosys-tems to climate change. Annual Review of Envi-ronment and Resources 32:1–29.

Field, D.R., P.R. Voss, T.K. Kuczenski, R.B.Hammer, and V.C. Radeloff. 2003. Reaffirm-

References 365

ing social landscape analysis in landscape ecol-ogy: A conceptual framework. Society & NaturalResources 16:349–361.

Finlayson, A.C., and B.J. McCay. 1998. Crossing thethreshold of ecosystem resilience: The commer-cial extinction of Northern cod. Pages 311–337 inF. Berkes and C. Folke, editors. Linking Socialand Ecological Systems: Management Practicesand Social Mechanisms for Building Resilience.Cambridge University Press, Cambridge.

Finlayson, C.M., R. D’Cruz, N. Aladin, D.R. Barker,G. Beltram, et al. 2005. Inland water systems.Pages 551–583 in R. Hassan, R. Scholes, and N.Ash, editors. Ecosystems and Human Well-Being:Current State and Trends. Millennium EcosystemAssessment. Island Press, Washington.

Fischer, F. 1993. Citizen participation and thedemocratization of policy expertise: From theo-retical inquiry to practical cases. Policy Sciences26:165–187.

Fischer, F. 2000a. Citizens, Experts, and the Environ-ment: The Politics of Local Knowledge. Duke Uni-versity Press, Durham, NC.

Fischer, J. 2000b. Participatory Research in Ecolog-ical Fieldwork: A Nicaraguan Study. Pages 41–54in B. Neis, and L. Felt, editors. Finding Our SeaLegs. Institute of Social and Economic Research,St. Johns, Canada.

Fischer, J., A.D. Manning, W. Steffen, D.B. Rose, K.Daniell, et al. 2007. Mind the sustainability gap.Trends in Ecology and Evolution 22:621–624.

Foley, J.A., M.T. Coe, M. Scheffer, and G.Wang. 2003a. Regime shifts in the Sahara andSahel: Interactions between ecological and cli-matic systems in Northern Africa. Ecosystems 6:524–539.

Foley, J.A., M.H. Costa, C. Delire, N. Ramankutty,and P. Snyder. 2003b. Green surprise? How terres-trial ecosystems could affect earth’s climate. Fron-tiers of Ecology and the Environment 1:38–44.

Foley, J.A., R. DeFries, G.P. Asner, C. Barford, G.Bonan, et al. 2005. Global consequences of landuse. Science 309:570–574.

Foley, J.A., J.E. Kutzbach, M.T. Coe, and S.Levis. 1994. Feedbacks between climate andboreal forests during the Holocene epoch. Nature371:52–54.

Folke, C. 2003. Freshwater for resilience: A shift inthinking. Philosophical Transactions of the RoyalSociety of London, Series B 358:2027–2036.

Folke, C. 2006. Resilience: The emergence of aperspective for social-ecological systems analysis.Global Environmental Change 16:253–267.

Folke, C., S. Carpenter, T. Elmqvist, L. Gunderson,C.S. Holling and B. Walker. 2002. Resilience

and sustainable development: Building adaptivecapacity in a world of transformations. Ambio31:437–440.

Folke, C., S.R. Carpenter, B.H. Walker, M. Scheffer,T. Elmqvist. 2004. Regime shifts, resilience andbiodiversity in ecosystem management. AnnualReview in Ecology Evolution and Systematics35:557–581.

Folke, C., J. Colding, and F. Berkes. 2003. Syn-thesis: Building resilience and adaptive capacityin social-ecological systems. Pages 352–387 in F.Berkes, J. Colding, and C. Folke, editors. Navigat-ing Social-Ecological Systems: Building Resiliencefor Complexity and Change. Cambridge Univer-sity Press, Cambridge.

Folke, C., T. Hahn, P. Olsson, and J. Norberg. 2005.Adaptive governance of social-ecological systems.Annual Review of Environment and Resources30:441–473.

Folke, C., M. Hammer, R. Costanza, and A. Jansson.1994. Investing in natural capital: Why, what, andhow? Pages 1–20 in A. Jansson, M. Hammer, C.Folke, and R. Costanza, editors. Investing in Nat-ural Capital. Island Press, Washington.

Folke, C., A. Jansson, J. Larsson, and R. Costanza.1997. Ecosystem appropriation by cities. Ambio26:167–172.

Folke, C., J.L. Prichard, F. Berkes, J. Colding, and U.Svedin. 1998. The Problem of Fit Between Ecosys-tems and Institutions. Institutional Dimensions ofGlobal Environmental Change Program, Bonn,Germany.

Folke, C., J.L. Prichard, F. Berkes, J. Colding,and U. Svedin. 2007. The Problem of Fitbetween Ecosystems and Institutions: TenYears Later. Ecology and Society 12(1):30.http://www.ecologyandsociety.org/vol12/iss1/art30/.

Ford, J.D., and B. Smit. 2004. A framework forassessing the vulnerability of communities in theCanadian Arctic to risks associated with climatechange. Arctic 57:389–400.

Fore, L.S., K. Paulsen, and K. O’Laughlin. 2001.Assessing the performance of volunteers inmonitoring streams. Freshwater Biology 46:109–123.

Foster, D.R., and J.D. Aber. 2004. Forests in Time:The Environmental Consequences of 1,000 Yearsof Change in New England. Yale University Press,New Haven.

Foster, S.S.D., and P J. Chilton. 2003. Groundwater:The processes and global significance of aquiferdegradation. Philosophical Transactions of theRoyal Society of London, Series B 358:1957–1972.

Fowler, C., and T. Hodgkin. 2004. Plant geneticresources for food and agriculture: Assessing

366 References

global availability. Annual Review of Environ-ment and Resources 29:143–179.

Fox, J., J. Krummel, S. Yarnasarn, M. Ekasingh, andN. Podger. 1995. Land-use and landscape dynam-ics in Northern Thailand: Assessing change inthree upland watersheds. Ambio 24:328–34.

Fox, J.J. 1991. Managing the ecology of rice produc-tion in Indonesia. Pages 61–84 in J. Hardjono, edi-tor. Indonesia: Resources, Ecology, and Environ-ment. Oxford University Press. Oxford.

Francis, R., and R. Shotton. 1997. “Risk” in fisheriesmanagement. Canadian Journal of Fisheries andAquatic Sciences 54:1699–1715.

Frank, K.T., B. Petrie, J.S. Choi, and W.C. Leggett.2005. Trophic cascades in a formerly cod-dominated ecosystem. Science 308:1621–1623.

Freeman, M.M.R., and U.P. Kreuter, editors. 1994.Elephants and Whales Resources for Whom? Gor-don and Breach Publishers, Basel, Switzerland.

Friedlingstein, P., K.C. Prentice, I.Y. Fung, J.G.John, and G.P. Brasseur. 1995. Carbon-biosphere-climate interactions in the last glacial maxi-mum climate. Journal of Geophysical Research100:7203–7221.

Frost, P., E. Medina, J.-C. Menaut, O. Solbrig, M.Swift, et al. 1986. Response of savannas to stressand disturbance: A proposal for a collaborativeprogramme of research. International Union ofBiological Sciences, Special Issue 10. Paris.

Funke, N., K. Nortje, K. Findlater, M. Burns,A. Turton, et al. 2007. Redressing inequality:South Africa’s new water policy. Environment 49:10–23.

Galaz, V., P. Olsson, T. Hahn, C. Folke, and U.Svedin. 2008. The problem of fit among biophysi-cal systems, environmental and resource regimes,and broader governance systems: Insights andemerging challenges. Pages 147–186 in O.R.Young, L A. King, and H. Schroder, editors.Institutions and Environmental Change: Princi-pal Findings, Applications, and Research Fron-tiers. MIT Press, Cambridge, MA.

Gallant, A.L., E.F. Binnian, J.M. Omernik, and M.B.AQ16Shasby. 1995. Ecoregions of Alaska. USGS Pro-fessional Paper 1567. USGS, Washington. http://agdcftp1.wr.usgs.gov/pub/projects/fhm/ecoreg.gif.

Gallopin, G.C. 2006. Linkages between vulnerabil-ity, resilience, and adaptive capacity. Global Envi-ronmental Change 16:293–303.

Galloway, J., M. Burke, E. Bradford, W. Falcon, J.Gaskell, et al. 2007. International trade in meat:The tip of the pork chop. Ambio 36:622–629.

Galvin, K.A., R.S. Reid, R.H.J. Behnke, and N.T.AQ17Hobbs, editors. 2008. Fragmentation in Semi-Arid

and Arid Landscapes: Consequences for Humanand Natural Systems. Springer, Dordrecht.

Garreau, J. 1991. Edge City: Life on the New Fron-tier. Doubleday, New York.

Geertz, C. 1963. Agricultural Involution: The Pro-cess of Ecological Change in Indonesia. Universityof California Press. Berkeley.

Gelcich, S., G. Edwards-Jones, M.J. Kaiser, and J.C.Castilla. 2006. Co-management policy can reduceresilience in traditionally managed marine ecosys-tems. Ecosystems 9:951–966.

Getz, W.M., L. Fortmann, D. Cumming, J. du Toit, J.Hilty, et al. 1999. Sustaining natural and humancapital: Villagers and scientists. Science 283:1855–1856.

Geist, H.J., and E.F. Lambin. 2004. Dynamic causalpatterns of desertification. BioScience 53:817–829.

Ghassemi, F., A.J. Jakeman, and H.A. Nix. 1995.Salinisation of Land and Water Resources: HumanCauses, Extent, Management, and Case Studies.CAB International. Wallingford, UK.

Giller, K., and C. Palm. 2004. Cropping systems:Slash-and-burn cropping systems of the tropics.Pages 262–366 in R.M. Goodman and M. Dekker,editors. Encyclopedia of Plant and Crop Science.CPLScientificPublishingServicesLtd.,Berks,UK.

Gillson, L., and K.I. Duffin. 2007. Thresholds ofpotential concern as benchmarks in the manage-ment of African savannahs. Philosophical Trans-actions of the Royal Society of London, Series B362:309–319.

Ginn, W.J. 2005. Investing in Nature: Case Studiesof Land Conservation in Collaboration with Busi-ness. Island Press, Washington.

Gladwell, M. 2000. The Tipping Point: How Lit-tle Things Can Make a Big Difference. Abacus,London.

Gleick, P.H. 1996. Basic water requirements forhuman activities: Meeting basic needs. WaterInternational 21:83–92.

Gleick, P.H. 2000. The World’s Water 2000–2001.Island Press, Washington.

Gleick, P.H. 2002. The World’s Water: The Bien-nial Report on Freshwater Resources 2002–2003.Island Press, Washington.

Gleick, P.H., H. Cooley, D. Katz, E. Lee, J. Morrison,et al. 2006. The Worlds Water: The Biennial Reporton Freshwater Resources 2006–2007. Island Press,Washington.

Goodwin, N.B., A. Grant, A.L. Perry, N.K. Dulvy,and J.D. Reynolds. 2006. Life history correlates ofdensity-dependent recruitment in marine fishes.Canadian Journal of Fisheries and Aquatic Sci-ences 63:494–509.

References 367

Gordon, L.J., G.D. Peterson, and E.M. Bennett.2008. Agricultural modifications of hydrologicalflows create ecological surprises. Trends in Ecol-ogy and Evolution 23:211–219.

Gordon, M.R., J.C. Mueller, and C.C. Walden. 1980.Effect of biotreatment on fish tainting propensityof bleach kraft whole mill effluent. Transactionsof the Technical Association of Pulp and PaperIndustry 6:TR2-TR8.

Gorgens, A.H.M., and B.W. van Wilgen. 2004. Inva-sive alien plants and water resources in SouthAfrica: Current understanding, predictive abilityand research challenges. South African Journal ofScience 100:27–33.

Gottdiener, M., and R. Hutchinson. 2001. TheNew Urban Sociology. 2nd Edition. McGraw-HillHigher Education, New York.

Grafton, R.Q., R. Arnason, T. Bjørndal, D.AQ18Campbell, H.F. Campbell, et al. 2006. Incentive-based approaches to sustainable fisheries. Cana-dian Journal of Fisheries and Aquatic Sciences 63:699–710.

Granovetter, M.D. 2004. The impact of social struc-tures on economic development. Journal of Eco-nomic Perspectives 19:33–50.

Gray, B. 1989. Collaborating: Finding CommonGround for Multiple Problems. Jossey-Bass, SanFrancisco.

Gray, B., and D.J. Wood. 1991. Collaborativealliances: Moving from practice to theory. Journalof Applied Behavioral Sciences 27:3–22.

Griffith, B., D.C. Douglas, N.E. Walsh, D.D. Young,T.R. McCabe, et al. 2002. The Porcupine caribouherd. Pages 8–37 in D.C. Douglas, P.E. Reynolds,and E.B. Rhode, editors. Arctic Refuge CoastalPlain Terrestrial Wildlife Research Summaries.U.S. Geological Survey, Biological ResourcesDivision, Biological Sciences Report USGS/BRDBSR-2002–0001, Alaska.

Grimm, N., J.M. Grove, S.T.A. Pickett, and C.L.Redman. 2000. Integrated approaches to long-term studies of urban ecological systems. Bio-Science 50:571–584.

Grimm, N.B., S.H. Faeth, N.E. Golubiewski, C.L.Redman, J. Wu, et al. 2008. Global change and theecology of cities. Science 319:756–760.

Groffman, P., N.L. Law, K.T. Belt, L.E. Band, andG.T. Fisher. 2004. Nitrogen fluxes and retentionin urban watershed ecosystems. Ecosystems 7:393–403.

Groffman, P.M., D.J. Bain, L.E. Band, K.T. Belt,G.S. Brush, et al. 2003. Down by the riverside:Urban riparian ecology. Frontiers in Ecology andthe Environment 1:315–321.

Groffman, P.M., J.S. Baron, T. Blett, A.J. Gold, M.A.Palmer, et al. 2006. Ecological thresholds: The keyto successful environmental management or animportant concept with no practical application?Ecosystems 9:1–13.

Grove, J.M. 1995. Excuse me, Could I speak tothe property owner, please? Common PropertyResources Digest 35:7–8.

Grove, J.M., and W.R. Burch. 1997. A social ecol-ogy approach to urban ecosystems and landscapeanalysis. Urban Ecosystems 1:185–199.

Grumbine, R.E. 1994. What is ecosystem manage-ment? Conservation Biology 8:27–38.

Gunderson, L.H. 1999. Resilience, flexibility andadaptive management: antidotes for spuriouscertitude? Conservation Ecology 3(1):7. http://www.consecol.org/vol3/iss1/art7/.

Gunderson, L.H. 2003. Adaptive dancing: Inter-actions between social resilience and ecologicalcrises. Pages 33–52 in F. Berkes, J. Colding, andC. Folke, editors. Navigating Social–EcologicalSystems: Building Resilience for Complexity andChange. Cambridge University Press, Cambridge.

Gunderson, L.H., S.R. Carpenter, C. Folke,P. Olsson, and G.D. Peterson. 2006. WaterRATs (resilience, adaptability, and transforma-bility) in lake and wetland social-ecologicalsystems. Ecology and Society 11(1):16. http://www.ecologyandsociety.org/vol11/iss1/art16/.


Gunderson, L.H., C.S. Holling, and S.S. Light. 1995.Barriers and Bridges to the Renewal of Ecosystemsand Institutions. Columbia University Press, NewYork.

Gunderson, L.H., and S.S. Light. 2006. Adaptive AQ19management and adaptive governance in theEverglades ecosystem. Policy Science 39:323–334.

Gunderson, L.H., and L. Pritchard, editors. 2002.Resilience and the Behavior of Large-ScaleEcosystems. Island Press, Washington.

Gunter, J. 2004. The Community Forestry Guide-book: Tools and Techniques for Communities inBritish Columbia. FORREX, Kamloops, Canada.

Hahn, T., P. Olsson, C. Folke, and K. Johansson.2006. Trust-building, knowledge generation andorganizational innovations: The role of a bridg-ing organization for adaptive co-management of awetland landscape around Kristianstad, Sweden.Human Ecology 34:573–592.

Hahn, T., L. Schultz, C. Folke, and P. Olsson. 2008. AQ20Social networks as sources of resilience in social-

368 References

ecological systems. Pages 119–148 in J. Norbergand G.S. Cumming, editors. Complexity Theoryfor a Sustainable Future. Columbia UniversityPress, New York.

Hanna, S.S., C. Folke, and K.-G. Maler, editors. 1996.Rights to Nature: Ecological, Economic, Cultural,and Political Principles of Institutions for the Envi-ronment. Island Press, Washington, DC.

Hannah, L., G. Midgley, T. Lovejoy, W.J. Bond, M.Bush, et al. 2002. Conservation of bioidversityin a changing climate. Conservation Biology 16:264–268.

Hara, M., and J. Nielsen. 2003. Experiences withfisheries co-management in Africa. Pages 81–97in D.C. Wilson, C. Douglas, J.R. Nielsen, and P.Degnbol, editors. The Fisheries Co-ManagementExperience: Accomplishments, Challenges andProspects. Kluwer, Dordrecht.

Hardin, G. 1968. Tragedy of the Commons. Science162:1243–1248.

Harley, S.J., R.A. Myers, and A. Dunn. 2001.Is catch-per-effort proportional to abundance?Canadian Journal of Fisheries and Aquatic Sci-ences 58:1760–1772.

Harmon, M.E., W.K. Ferrell, and J.F. Franklin.1990. Effects on carbon storage of conversion ofold-growth forests to young forests. Science 247:699–702.

Hawley, A.H. 1950. Human Ecology: A Theory ofCommunity Structure. Ronald Press, New York.

Haynes, R.W., B.T. Bormann, D.C. Lee, and J.R.Martin, editors. 2006. Northwest Forest Plan –The First 10 Years (1994–2003): Synthesis ofMonitoring and Research Results. General Tech-nical Report PNW-GTR-651. U.S. Departmentof Agriculture, Forest Service, Pacific NorthwestResearch Station, Portland.

Heal, G. 2000. Nature and the Marketplace: Captur-ing the Value of Ecosystem Services. Island Press,Washington.

Heaslip, R. 2008. Monitoring salmon aquaculturewaste: The contribution of First Nations’ rights,knowledge, and practices in British Columbia.Marine Policy.AQ21

Held, D.A. McGrew, D. Goldblatt, and J. Perraton.1999. Global Transformations: Politics, Eco-nomics, and Culture. Stanford University Press,Stanford.

Hennessey, T., and M. Healey. 2000. Ludwig’sratchet and the collapse of New England ground-fish stocks. Coastal Management 28:187–213.

Hewitt, K. 1997. Regions of Risk: A GeographicalIntroduction to Disasters. Longman, Harlow, UK.

Hibbard, K.A., P.J. Crutzen, E.F. Lambin, D.Liverman, N.J. Mantua, et al. 2006. Decadal inter-

actions of humans and the environment. Pages341–375 in R. Costanza, L. Graumlich, and W.Steffen, editors. Integrated History and Future ofPeople on Earth. Dahlem Workshop Report 96.MIT Press, Cambridge, MA.

Hiernaux, P., and M.D. Turner. 2002. The influenceof farmer and postoralist management practiceson desertification processes in the Sahel. Pages135–148 in J.F. Reynolds and D.M. Stafford Smith,editors. Global Desertification: Do HumansCause Deserts? Dahlem University Press,Berlin.

Hilborn, R. 1992. Hatcheries and the future ofsalmon in the northwest. Fisheries 17:5–8.

Hilborn, R., and D. Eggers. 2000. A review of thehatchery programs for pink salmon in PrinceWilliam Sound. Transactions of the AmericanFisheries Society 129:333–350.

Hilborn, R., J.M. Orensanz, and A.M. Parma. 2005. AQ22Institutions, incentives, and the future of fisheries.Philosophical Transactions of the Royal Society,Series B: Biological Sciences 360:47–57.

Hilborn, R., R.H. Parrish, and C.J. Walters. 2006.Peer Review, California Marine Life ProtectionAct (MLPA) Science Advice and MPA NetworkProposals. California Fisheries Coalition, Sacra-mento, CA.

Hilborn, R., and C.J. Walters. 1992. Quantita- AQ23tive Fisheries Stock Assessment and Management.Chapman-Hall, New York.

Himmel, M.E., S. Ding, D.K. Johnson, W.S. Adney,M.R. Nimlos, J.W. Brady, and T.D. Foust. 2007.Biomass recalcitrance: Engineering plants andenzymes for biofuels production. Science 315:804–807.

Hoekstra, A.Y., and A.K. Chapagain. 2007. Waterfootprints of nations: Water use by people asa function of their consumption pattern. WaterResources Management 21:35–48.

Holling, C.S. 1973. Resilience and stability of eco-logical systems. Annual Review of Ecology andSystematics 4:1–23.

Holling, C.S., editor. 1978. Adaptive EnvironmentalAssessment and Management. John Wiley & Sons,New York.

Holling, C.S. 1986. Resilience of ecosystems: Localsurprise and global change. Pages 292–317 in W.C.Clark and R.E. Munn, editors. Sustainable Devel-opment and the Biosphere. Cambridge UniversityPress, Cambridge.

Holling, C.S. 1995. What barriers? What bridges?Pages 3–34 in L.H. Gunderson, C.S. Holling, andS.S. Light, editors. 1995. Barriers and Bridgesto the Renewal of Ecosystems and Institutions.Columbia University Press, New York.

References 369

Holling, C.S. 1998. Two cultures of ecology. Conser-vation Ecology 2(2):4. http://www.consecol.org/vol2/iss2/art4/.

Holling, C.S., F. Berkes, and C. Folke. 1998. Sci-ence, sustainability, and resource management.Pages 342–362 in F. Berkes, and C. Folke, edi-tors. Linking Social and Ecological Systems: Man-agement Practices and Social Mechanisms forBuilding Resilience. Cambridge University Press,Cambridge.

Holling, C.S., and L.H. Gunderson. 2002. Resilienceand adaptive cycles. Pages 25–62 in L.H. Gunder-son and C S. Holling, editors. Panarchy: Under-standing Transformations in Human and NaturalSystems. Island Press, Washington.

Holling, C.S., L.H. Gunderson, and D. Ludwig.2002a. In quest of a theory of adaptive change.Pages 3–22 in L.H. Gunderson and C.S. Holling,editors. Panarchy: Understanding Transforma-tions in Human and Natural Systems. Island Press,Washington.

Holling, C.S., L.H. Gunderson, and G.D. Peter-son. 2002b. Sustainability and panarchies. Pages63–102 in L.H. Gunderson and C.S. Holling, edi-tors. Panarchy: Understanding Transformationsin Human and Natural Systems. Island Press,Washington.

Holling, C.S., and G.K. Meffe. 1996. Commandand control and the pathology of naturalresource management. Conservation Biology 10:328–337.

Holling, C.S., and S. Sanderson. 1996. Dynamicsof (dis)harmony in ecological and social systems.Pages 57–85 in S. Hanna, C. Folke, K.-G. Maler,and A. Jansson, editors. Rights to Nature: Ecolog-ical, Economic, Cultural, and Political Principlesof Institutions for the Environment. Island Press,Washington.

Homer-Dixon, T.F. 1999. Environment, Scarcity, andAQ24Violence. Princeton University Press, Princeton.

Hostetler, M. 1999. Scale, birds, and human deci-sions: A potential for integrative research inurban ecosystems. Landscape and Urban Plan-ning 45:15–19.

Hough, M. 1984. City Form and Natural Process:Towards a New urban Vernacular. Van NostrandReinhold Company, New York.

Hughes, T.P., D.R. Bellwood, C. Folke, L.J. McCook,and J.M. Pandolfi. 2007. No-take areas: Herbivoryand coral reef resilience. Trends in Ecology andEvolution 22:1–3.

Huitric, M. 2005. Lobster and conch fisheriesof Belize: A history of sequential exploita-tion. Ecology and Society 10(1):21. http://www.ecologyandsociety.org/vol10/iss1/art21/.

Hulme, D., and M. Murphree, editors. 2001. AfricanWildlife and Livelihoods: The Promise and Per-formance of African Conservation. James CurreyPublishers, Oxford.

Hutton, G., and L. Haller. 2004. Evaluation ofthe Costs and Benefits of Water and SanitationImprovements at the global level. World HealthOrganization, Geneva.

Imhoff, D. 2007. Food Fight: A Citizen’s Guide toa Food and Farm Bill. University of CaliforniaPress. Berkeley.

Imhoff, M.L., L. Bounoua, T. Ricketts, C. Loucks, R.Harriss, and W.T. Lawrence. 2004. Global patternsin human consumption of net primary production.Nature 429:870–873.

IPCC (Intergovernmental Panel on ClimateChange). 2007a. Climate Change 2007: ThePhysical Science Basis, Contribution of WorkingGroup I to the Fourth Assessment Report of theIntergovernmental Panel on Climate Change.Cambridge University Press, Cambridge.

IPCC (Intergovernmental Panel on ClimateChange). 2007b. Climate Change 2007: Impacts,Adaptation and Vulnerability. in M.L. Parry, O.F.Canziani, J.P. Palutikof, P.J. van der Linden, andC.E. Hanson, editors. Contribution of WorkingGroup II to the Fourth Assessment Report ofthe Intergovernmental Panel on Climate Change.Cambridge University Press, Cambridge.

IPCC (Intergovernmental Panel on Climate AQ25Change). 2007c. Climate Change 2007: SynthesisReport.

IFPRI (International Food Policy Research Insti-tute). 2002. Reaching Sustainable Food Securityfor All by 2020: Getting the Priorities and Respon-sibilities Right. IFPRI, Washington.

IUCN (International Union for the Conservation ofNature). 1991. Caring for the Earth: A Strategy forSustainable Living. World Conservation Union,United Nations Environment Programme, WorldWildlife Fund for Nature, Gland, Switzerland.

IUCN (International Union for the Conserva- AQ26tion of Nature). 2008. Protected Area Cate-gories Task Force. http://www.iucn.org/themes/wcpa/theme/categories/categories.htm.

Ives, A.R., K. Gross, and J.L. Klug. 1999. Stabilityand variability in competitive communities. Sci-ence 286:542–544.

Jackson, J.B.C. 2001. What was natural in the coastaloceans? Proceedings of the National Academy ofSciences 98:5411–5418.

Jackson, J.B.C., M.X. Kirby, W.H. Berger, K.A.Bjorndal, L.V. Botsford, et al. 2001. Historicaloverfishing and the recent collapse of coastalecosystems. Science 293:629–638.

370 References

Jacobs, J. 1961. The Death and Life of Great Ameri-can Cities. Vintage Books, New York.

James, C.D., J. Landsberg, and S.R. Morton. 1999.Provision of watering points in the Australian aridzone: A review of effects on biota. Journal of AridEnvironments 41:87–121.

Janssen, M.A., T.A. Kohler, and M. Scheffer. 2003.Sunk-cost effects and vulnerability to collapsein ancient societies. Current Anthropology 44:722–728.

Janssen, M.A., and M. Scheffer. 2004. Over-exploitation of renewable resources byancient societies and the role of sunk-costeffects. Ecology and Society 9(1):6. http://www.ecologyandsociety.org/vol9/iss1/art6/.

Janssen, M.A., M.L. Schoon, W. Ke, and K. Borner.2006. Scholarly networks on resilience, vulnera-bility and adaptation within the human dimen-sions of global environmental change. GlobalEnvironmental Change 16:240–252.

Jansson, A., C. Folke, J. Rockstrom, L. Gordon, andM. Falkemark. 1999. Linking freshwater flows andecosystem services appropriated by people: Thecase of the Baltic Sea Drainage Basin. Ecosystems2:351–366.

Jarvis, A., A. Lane, and R. Hijmans. 2008. Theeffect of climate change on crop wild rela-tives. Agriculture, Ecosystems and Environment(in press).AQ27

Jenkins, J.C., and R. Riemann. 2003. What does non-forest land contribute to the global C balance?Pages 173–179 in R. McRoberts, G.A. Reams,P.A. Van Deusen, and J.W. Moser, editors. Pro-ceedings, Third Annual FIA Science Symposium.USDA Forest Service General Technical ReportNC-230. St. Paul, MN.

Jentoft, S. 1993. Dangling Lines: The Fisheries Crisisand the Future of Coastal Communities: The Nor-wegian Experience. Institute for Social and Eco-nomic Studies, St. Johns, Canada.

Jiang, H. 2002. Culture, ecology, and Nature’schanging balance: Sandification on Mu Us SandLand, Inner Mongolia, China. Pages 181–196in J.F. Reynolds and D.M. Stafford Smith, edi-tors. Global Desertification: Do Humans CauseDeserts? Dahlem University Press, Berlin.

Johannes, R. 1978. Traditional marineconservationmethods in Oceania and their demise. AnnualReview of Ecological Systems 9:349–364.

Johannes, R. 1981. Words of the Lagoon: Fishing andMarine Lore in the Palau District of Micronesia.University of California Press, Berkeley.

Johnson, E.A. 1992. Fire and Vegetation Dynamics:Studies from the North American Boreal Forest.Cambridge University Press, Cambridge.

Johnson, K.N., F J. Swanson, M. Herring, and S. AQ28Greene. 1999. Bioregional Assessments: Science atthe Crossroads of Management and Policy. IslandPress, Washington.

Johnson, S. 2001. Emergence: The Connected Livesof Ants, Brains, Cities, and Software. Scribner,New York.

Johnson, S. 2006. The Ghost Map: The Story ofLondon’s Most Terrifying Epidemic – and How ItChanged Science, Cities, and the Modern World.Riverhead Books, New York.

Jones, C.G., J.H. Lawton, and M. Shachak. 1994.Organisms as ecosystem engineers. Oikos 69:373–386.

Justice, M. 2007. Volunteering in Fish-HabitatRehabilitation Projects in British Columbia.MRM Research Report No. 435. (thesis) Schoolof Resource and Environmental Management,Simon Fraser University, Vancouver.

Kahneman, D., and A. Tversky. 1979. Prospect the-ory: An analysis of decisions under risk. Econo-metrica 47:263–291.

Kaldor, M. 2003. Global Civil Society: An Answer toWar. Polity Press, Cambridge.

Kanter, R. 1983. The Change Masters: CorporateEntrepreneurs at Work. Simon and Schuster, NewYork.

Kaplan, S., and R. Kaplan. 1989. The Experience ofNature: A Physchological Perspective. CambridgeUniversity Press, New York.

Kasperson, R.E., K. Dow, E.R.M. Archer, D.Caceres, T.E. Downing, et al. 2005. Vulnerablepeoples and places. Pages 143–164 in R. Hassan,R. Scholes, and N. Ash, editors. Ecosystems andHuman Well-Being: Current State and Trends.Millennium Ecosystem Assessment. Island Press,Washington.

Katz, B., and J. Bradley. 1999. Divided we sprawl.Atlantic Monthly 284:26–42.

Kaufman, L. 1992. Catastrophic change in species-rich freshwater ecosystems: The lessons of LakeVictoria. BioScience 42:846–858.

Keane, J. 2003. Global Civil Society. CambridgeUniversity Press, Cambridge.

Keen, I. 2004. Aboriginal Economy and Society:Australia at the Threshold of Colonisation. OxfordUniversity Press, South Melbourne.

Keene, E. 2002. Beyond the Anarchical Society:Grotius, Colonialism and World Order. Cam-bridge University Press, Cambridge.

Keesing, J.K., R.H. Bradbury, L.M. DeVantier, M.J.Riddle, and G. De’ath. 1992. Geological evidencefor recurring outbreaks of the crown-of-thornsstarfish: A reassessment from an ecological per-spective. Coral Reefs 11:79–85.

References 371

Kennedy, V.S., and L.L. Breisch. 1983. Sixteendecades of political mismanagement of the oysterfishery in Maryland’s Chesapeake Bay. Journal ofEnvironmental Management 16:153–171.

Kent, M., R.A. Stevens, and L. Zhang. 1999.Urban plant ecology patterns and processes: Acase study of the flora of the City of Ply-mouth, Devon, UK. Journal of Biogeography 26:1281–1298.

Khagram, S. 2004. Dams and Development:Transnational Struggles for Water and Power.Cornell University Press. Ithaca.

Khagram, S., W.C. Clark, and D.F. Raad. 2003. Fromthe environment and human security to sustain-able security and development. Journal of HumanDevelopment 4:289–313.

Kidder, T. 1981. The Soul of a New Machine. AronBooks, New York.

King, J.M., R.E. Tharme, and M.S. de Villers. 2000.Environmental flow assessments for rivers: Man-ual for the building block methodology. WaterResearch Commission, Pretoria, South Africa.

Kinzig, A., D. Starrett, K. Arrow, B. Bolin, P.Dasgupta, et al. 2003. Coping with uncertainty:A call for a new science-policy forum. Ambio 32:330–335.

Kinzig, A.P., P. Ryan, M. Etienne, T. Elmqvist, H.E.Allison, et al. 2006. Resilience and regime shifts:Assessing cascading effects. Ecology and Soci-ety 11(1):20. http://www.ecologyandsociety.org/vol11/iss1/art20/.

Kirwan, J. 2008. Using Indigenous Fire Practices toManage Coastal Wetlands. Paper read to the Con-ference of the Society for Applied Anthropology,Memphis, Tennessee.

Kitchell, J.F. editor. 1992. Food Web Management:A Case Study of Lake Mendota. Springer-Verlag,New York.

Klare, M.T. 2001. Resource Wars: The New Land-AQ29scape of Global Conflict. Henry Holt and Com-pany, New York.

Kocher, T.D. 2004. Adaptive evolution and explo-sive speciation: The cichlid fish model. NatureReviews Genetics 5:288–298.

Koehane, R.O., and E. Ostrom, editors. 1995. LocalCommons and Global Interdependence. SagePress, London.

Kofinas, G.P. 1998. The Costs of Power Shar-ing: Community Involvement in Canadian Porcu-pine Caribou Co-Management. PhD Dissertation,University of British Columbia, Vancouver.

Kofinas, G.P. 2005. Hunters and researchers at theco-management interface: Emergent dilemmasand the problem of legitimacy in power sharing.Anthropologica 47:179–196.

Kofinas, G.P., and Communities of Aklavik, Arc-tic Village, Old Crow, and Fort McPherson. 2002.Community contributions to ecological monitor-ing: Knowledge co-production in the U.S.-CanadaArctic Borderlands. Pages 54–91 in I. Krupnikand D. Jolly, editors. The Earth Is Faster Now:Indigenous Observations of Arctic Environmen-tal Change. Arctic Research Consortium of theUnited States, Fairbanks, Alaska.

Kofinas, G.P., and J.R. Griggs. 1996. Collaborationand the B.C. Round Table: Analysis of a “bet-ter way” of deciding. Environments: A Journal ofInterdisciplinary Studies 23:17–40.

Kofinas, G.P., S.J. Herman, and C.L. Meek. 2007.Novel problems require novel solutions: Innova-tion as an outcome of adaptive co-management.Pages 249–267 in D. Armitage, F. Berkes, andN. Doubleday, editors. Adaptive Co-Management:Collaboration, Learning, and Multi-Level Gov-ernance. University of British Columbia Press,Vancouver.

Kofinas, G., P. Lyver, D. Russell, R. White, A.Nelson, et al. 2003. Towards a protocol for com-munity monitoring of caribou body condition.Rangifer 14:43–52.

Kokou, K., and N. Sokpon. 2006. Les foretssacrees du culoir du Dahomey. Bois et Forets desTropiques 288:15–23.

Kolar, C., and D.M. Lodge. 2000. Freshwater non-indigenous species: Interactions with other globalchanges. Pages 3–30 in H.A. Mooney and R.J.Hobbs, editors. Invasive Species in a ChangingWorld. Island Press, Washington.

Kolata, G. 2000. Flu: The Story of the GreatInfluenza Pandemic of 1918 and the Search for theVirus that Caused It. Macmillan, London.

Kolstad, C. 1999. Environmental Economics. OxfordUniversity Press. UK.

Krauss, M.E. 1982. Native Languages of AQ30Alaska (Map). Alaska Native LanguageCenter, University of Alaska, Fairbanks.http://www.uaf.edu/anlc/.

Krupnik, I., and D. Jolly, editors. 2002. TheEarth is Faster Now: Indigenous Observations ofArctic Environmental Change. Arctic ResearchConsortium of the United States, Fairbanks,Alaska.

Kundzewicz, Z.W., and H J. Schellnhuber. 2004.Floods in the IPCC perspective. Natural Hazards31:111–128.

Kursar, T.A., C.C. Caballero-George, T.L. Capson,L. Cubilla-Rios, W.H. Gerwick, et al. 2006. Secur-ing economic benefits and promoting conserva-tion through bioprospecting. BioScience 56:1005–1012.

372 References

Lach, D., P. List, B. Steel, and B. Shindler. 2003.Advocacy and credibility of ecological scientistsin resource decision making: A regional study.BioScience 53:171–179.

Lambin, E.F. 2007. The Middle Path: Avoiding Envi-ronmental Catastrophe. University of ChicagoPress, Chicago.

Lambin, E.F., B.L. Turner, H.J. Geist, S.B. Agbola,A. Angelsen, et al. 2001. The causes of land-use and land-cover change: Moving beyond themyths. Global Environmental Change 11:261–269.

Landsberg, J., C.D. James, S.R. Morton, W.J. Muller,and J. Stol. 2003. Abundance and composition ofplant species along grazing gradients in Australianrangelands. Journal of Applied Ecology 40:1008–1024.

Lane, M.B., and G. McDonald. 2002. Towards a gen-eral model of forest management through time:Evidence from Australia, USA, and Canada.Land Use Policy 19:193–206.

Langdon, S. 1999. Communities and Quotas. Alter-natives in the North Pacific Fisheries. Presentationto the Pacific Marine States Fisheries Commis-sion. Semiahmoo, Washington.

Langdon, S. 2008. The Community Quota Programin the Gulf of Alaska: a vehicle for Alaska Nativevillage sustainability. Pages 155–194 in M.E. Loweand C. Carothers, editors. Enclosing the Fisheries:People, Places, and Power. American FisheriesSociety. Bethesda, MD.

Lansing, S. 2006. Perfect Order: Recognizing Com-plexity in Bali. Princeton University Press,Princeton.

Lansing, S.J. 1991. Priests and Programmers: Tech-AQ31nologies of Power in the Engineered Landscape ofBali. Princeton University Press, Princeton.

Larson, A.M., and J.C. Ribot. 2007. The poverty offorestry policy: Double standards on an unevenplaying field. Sustainability Science 2:189–204.

Larsson, S., and K. Danell. 2001. Science and themanagement of boreal forest biodiversity. Scan-dinavian Journal of Forest Research Supplement3:5–9.

Lathrop, R.C. 1992a. Nutrient loadings, lake nutri-ents and water clarity. Pages 69–96 in J.F. Kitchell,editor. Food Web Management: A case study ofLake Mendota. Springer-Verlag, New York.

Lathrop, R.C. 1992b. Decline in zoobenthos den-sities in the profundal sediments of Lake Men-dota (Wisconsin, USA). Hydrobiologia 235/236:353–361.

Lathrop, R.C., B.M. Johnson, T.B. Johnson, M.T.Vogelsang, S.R. Carpenter, et al. 2002. Stockingpiscivores to improve fishing and water clarity: A

synthesis of the Lake Mendota biomanipulationproject. Freshwater Biology 47:2410–2424.

Lebel, L., N.H. Tri, A. Saengnoree, S. Pasong, U.Buatama, et al. 2002. Industrial transformationand shrimp aquaculture in Thailand and Vietnam:Pathways to ecological, social, and economic sus-tainability? Ambio 31:311–323.

Lee, K.N. 1993. Compass and Gyroscope: IntegratingScience and Politics for the Environment. IslandPress, Washington.

Lee, K.N. 1999. Appraising adaptive manage-ment. Conservation Ecology 3(2):3. http://www.consecol.org/vol3/iss2/art3/.

Lee, R.B. 1979. The !Kung San: Men, Women andWork in a Foraging Society. Cambridge UniversityPress, Cambridge.

Lee, R., and I. DeVore, editors. 1968. Man theHunter. Aldine, Chicago.

Lee, S., and C. Webster. 2006. Enclosure of theurban commons. GeoJournal 66:27–42.

Leemans, R., and A. Kleidon. 2002. Regional and AQ32global assessment of the dimensions of desertifi-cation. Pages 215–231 in J.F. Reynolds and D.M.Stafford Smith, editors. Global Desertification:Do Humans Cause Deserts? Dahlem UniversityPress, Berlin.

Leopold, A. 1949. A Sand County Almanac. OxfordUniversity Press, Oxford.

Levin, S.A. 1998. Ecosystems and the biosphere ascomplex adaptive systems. Ecosystems 1:431–436.

Levin, S.A. 1999. Fragile Dominion: Complexity andthe Commons. Perseus Books, Reading, MA.

Levy, M., S. Babu, K. Hamilton, V. Rhoe, A.Catenazzi, et al. 2005. Ecosystem conditions andhuman well-being. Pages 123–142 in R. Hassan,R. Scholes, and N. Ash, editors. Ecosystems andHuman Well-Being: Current State and Trends.Millennium Ecosystem Assessment. Island Press,Washington.

Lewis, H.T. 1989. Ecological and technologicalknowledge of fire: Aborigines versus park man-agers in Northern Australia. American Anthro-pologist 91:940–61.

Lewis, J. 2002. Agrarian change and privitization ofejido land in Northern Mexico. Journal of Agrar-ian Affairs 3:402–420.

Lewis, H.T., and T.A. Ferguson. 1988. Yards, corri-dors and mosaics: How to burn a boreal forest.Human Ecology 16:57–77.

Liebman, M., and C.P. Staver. 2001. Crop diversifica-tion for weed management. Pages 322–374 in M.Liebman, C.L. Mohler, and C.P. Staver, editors.Ecological Management of Agricultural Weeds.Cambridge University Press, Cambridge.

References 373

Likens, G.E., F.H. Bormann, R.S. Pierce, J.S. Eaton,and N.M. Johnson. 1977. Biogeochemistry of aForested Ecosystem. Springer-Verlag, New York.

Lindblom, C.E. 1959. The science of “muddlingthrough”. Public Administration Review 19:70–88.

Lindblom, C.E. 1972. Still muddling, not yet through.Public Administration Review 39:517–526.

Lindgren, E., and R. Gustafson. 2001. Tick-borneencephalitis in Sweden and climate change.Lancet 358:16–18.

Lindgren, E., L. Talleklint, and T. Polfeldt. 2000.The impact of climate change on the northern lat-itude limit and population density of the disease-transmitting European tick Ixodes ricinus. Envi-ronmental Health Perspectives 108:119–123.

Lobell, D., M. Burke, C. Tebaldi, M. Mastrandrea,W. Falcon, et al. 2008. Prioritizing climate adap-tation needs for food security in 2030. Science319:607–610.

Lobell, D.B., and J.I. Ortiz-Monasterio. 2008. Satel-lite monitoring of yield responses to irrigationpractices across thousands of fields. AgronomyJournal in press.AQ33

Longhurst, A. 2006. The sustainability myth. Fish-eries Research 81:107–112.

Loucks, L.A. 2005. The Evolution of the Area 19Snow Crab Co-Management Agreement: Under-standing the Inter-relationship Between Transac-tion Costs, Credible Commitment, and CollectiveAction. PhD. dissertation. Simon Fraser Univer-sity, Vancouver.

Lovelock, J.E. 1979. Gaia: A New Look at Life onEarth. Oxford University Press, Oxford.

Low, B., E. Ostrom, C. Simon, and J. Wilson.2003. Redundancy and diversity: Do they influ-ence optimal management? Pages 83–114 in F.Berkes, J. Colding, and C. Folke, editors. Navigat-ing Social-Ecological Systems: Building Resiliencefor Complexity and Change. Cambridge Univer-sity Press, Cambridge.

Ludwig, D., W.A. Brock, and S.R. Carpenter. 2005.Uncertainty in discount models and environ-mental accounting. Ecology and Society 10(2):13.http://www.ecologyandsociety.org/vol10/iss2/art13/.

Ludwig, D., R. Hilborn, and C. Walters. 1993. Uncer-tainty, resource exploitation, and conservation:Lessons from history. Science 260:17, 36.

Ludwig, D., M. Mangel, and B. Haddad. 2001.Ecology, conservation, and public policy.Annual Review of Ecology and Systematics 32:481–517.

Ludwig, J.A., D.J. Tongway, G.N. Bastin, and C.D.James. 2004. Monitoring ecological indicators ofrangeland functional integrity and their relation

to biodiversity at local to regional scales. AustralEcology 29:108–120.

Ludwig, J.A., D.J. Tongway, and S.G. Marsden. 1999.Stripes, strands or stipples: Modelling the influ-ence of three landscape banding patterns onresource capture and productivity in semi-aridwoodlands, Australia. CATENA 37:257–273.

Ludwig, J.A., B.P. Wilcox, D.D. Breshears, D.J.Tongway, and A.C. Imeson. 2005. Vegetationpatches and runoff-erosion as interacting ecohy-drological processes in semiarid landscapes. Ecol-ogy 86:288–297.

Lukes, S. 2002. Power: A Radial View. PalgraveMacmillan, New York.

Luthar, S.S., and D. Cicchetti. 2000. The construct AQ34of resilience: Implications for interventions andsocial policies. Development and Psychopathology12:857–885.

MacCall, A. 1990. Dynamic geography of fish Popu-lations. University of Washington Press, Seattle.

Machlis, G.E., J.E. Force, and W.R. Burch, Jr.1997. The human ecosystem. Part I: The humanecosystem as an organizing concept in ecosystemmanagement. Society and Natural Resources 10:347–367.

Magadlela, D., and N. Mdzeke. 2004. Social benefitsin the Working for Water programme as a publicworks initiative. South African Journal of Science100:94–96.

Magnuson, J.J., T.K. Kratz, and B.J. Benson, editors.2006. Long-Term Dynamics of Lakes in the Land-scape. Oxford University Press, London.

Magnuson, J.J., and R.C. Lathrop. 1992. Histori-cal changes in the fish community. Pages 193–232in J.F. Kitchell, editor. Food Web Management.Springer-Verlag, New York.

Mahler, J. 1997. Influences of organizational cul-ture on learning in public agencies. Journal ofPublic Administration Research and Theory 7:519–540.

Makse, H.A., S. Havlin, and H.E. Stanley. 1995.Modeling urban growth patterns. Nature 377:608–612.

Malthus, T.R. 1798. An Essay on the Principle ofPopulation. Johnson, London.

Manning, R. 2002. Agriculture versus biodiversity:Will market solutions suffice? Conservation inPractice 3:18–27.

Manning, R. 2004. Against the Grain: How Agricul-ture Has Hijacked Civilization. North Point Press,New York.

Mansuri, G., and V. Rao. 2004. Community-based(and –driven) development: A critical review. TheWorld Bank Research Observer 19(1):1–39.

374 References

Martin, P.S. and R.G. Klein, editors. 1984. Quater-nary Extinctions: A Prehistoric Revolution. Uni-versity of Arizona Press, Tucson.

Maskrey, A. 1989. Disaster Mitigation: ACommunity-based Approach. Oxfam, Oxford.

Maslow, A.H. 1943. A theory of human motivation.Psychological Review 50:370–396.

Matson, P., A. Luers, K. Seto, and R. Naylor.2005. People, land use, and environment in theYaqui Valley, Sonora, Mexico. Pages 238–264in B. Entwisle, P.C. Stern, editors. Population,Land Use, and Environment: Research Direc-tions. National Resource Council of the NationalAcademies. The National Academies Press, Wash-ington, D.C.

Matson, P.A., R.L. Naylor, and I. Ortiz-Monasterio.1998. Integration of environmental, agronomic,and economic aspects of fertilizer management.Science 280:112–115.

Matson, P.A., W.J. Parton, A.G. Power, andM.J. Swift. 1997. Agricultural intensification andecosystem properties. Science 227:504–509.

McAllister, R.R.J., I.J. Gordon, M.A. Janssen, andN. Abel. 2006. Pastoralists’ responses to variationof rangeland resources in time and space. Ecolog-ical Applications 16:572–583.

McCarthy, J.J., M.L. Martello, R. Corell, N.E. Selin,S. Fox, et al. 2005. Climate change in the contextof multiple stressors and resilience. Pages 945–988in ACIA, editor. Arctic Climate Impact Assess-ment. Cambridge University Press.

McCay, B.J. 2002. Emergence of institutions for thecommons: Contexts, situations, and events. Pages361–402 in NRC. The Drama of the Commons.National Academy Press, Washington.

McCay, B.J., and J. Acheson, editors. 1987. TheQuestion of the Commons: The Culture and Ecol-ogy of Communal Resources. University of Ari-zona Press, Tucson.

McGinnis, M.D. editor. 1999. Polycentric Gover-nance and Development: Readings from the Work-shop in Political Theory and Policy Analysis. Uni-versity of Michigan Press, Ann Arbor.

McGoodwin, J. 1994. “Nowadays, nobody has anyrespect”: The demise of folk management in arural Mexican fishery. Pages 43–54 in C.L. Dyerand J.R. McGoodwin, editors. Folk Managementin the World’s Fisheries. Lessons for Modern Fish-eries Management. University Press of Colorado,Niwot, CO.

McGrath, B., V. Marshall, M.L. Cadenasso, J.M.Grove, S.T.A. Pickett, et al., editors. 2008.Designing Patch Dynamics. Columbia University,New York.

McKeon, G., W. Hall, B. Henry, G. Stone, andI. Watson, editors. 2004. Pasture Degradationand Recovery in Australia’s Rangelands: Learn-ing from History. Queensland Department ofNatural Resources, Mines and Energy, Brisbane,Australia.

McLachlan, J.S., J. Hellmann, and M. Schwartz.2007. A framework for debate of assisted migra-tion in an era of climate change. ConservationBiology 21:297–302.

McNaughton, S.J. 1977. Diversity and stability ofecological communities: A comment on the roleof empiricism in ecology. American Naturalist111:515–525.

McNeill, J. 2000. Something New Under the Sun: AnEnvironmental History of the Twentieth Century.W.W. Norton, New York.

MEA (Millennium Ecosystem Assessment). 2005a.Ecosystems and Human Well-being: Current Sta-tus and Trends. Island Press, Washington.

MEA (Millennium Ecosystem Assessment). 2005b.Ecosystems and Human Well-being: Deserti-fication Synthesis. World Resources Institute,Washington.

MEA (Millennium Ecosystem Assessment). 2005c.Ecosystems and Human Well Being: Scenarios.Island Press, Washington.

MEA (Millennium Ecosystem Assessment). 2005d.Ecosystems and Human Well-being: Synthesis.Island Press, Washington, D.C.

Meadows, D.H, J. Randers, D.L. Meadows, and W.W.Behrens. 1972. The Limits to Ggrowth: A Reportfor the Club of Rome’s Project on the Predicamentof Mankind. Universe Books, New York.

Mears, L.A. 1981. The New Rice Economy ofIndonesia. Gadja Mada University Press,Yogyakarta, Indonesia.

Melosi, M.V. 2000. The Sanitary City: Urban Infras-tructure in America from Colonial Times tothe Present. Johns Hopkins University Press,Baltimore.

Memmott, J. 1997. The structure of a plant-pollinator food web. Ecology Letters 2:276–280.

Middleton, N.J., and D.S.G. Thomas, editors. 1997.World Atlas of Desertification. 2nd Edition.U.N. Environment Programme, Edward Arnold,New York.

Millar, C.I., N.L. Stephenson, and S.L. Stephens.2007. Climate change and forests of the future:Managing in the face of uncertainty. EcologicalApplications 17:2145–2151.

The Millennium Development Goals Report.2007. United Nations. New York, NY. http://www.un.org/millenniumgoals/pdf/mdg2007.pdf.

References 375

Mintzberg, H., and F. Westley. 1992. Cycles of orga-nizational change. Strategic Management Journal13:39–59.

Mitchell, C.E., and A.G. Power. 2003. Release ofinvasive plants from fungal and viral pathogens.Nature 421:625–627.

Mollenhoff, D. 2004. Madison - A History of theFormative Years. University of Wisconsin Press,Madison.

Morrow, P., and C. Hensel. 1992. Hidden dissen-sion: Minority-majority relationships and the useof contested terminology. Arctic Anthropology29(1):38–53.

Mulder, M.B., and P. Coppolillo. 2005. Conservation:Linking Ecology, Economics, and Culture. Prince-ton University Press, Princeton.

Mullon, C., P. Freon, and P. Cury. 2005. The dynam-ics of collapse in world fisheries. Fish and Fisheries6:111–120.

Munch, R., and N.J. Smesler. 1987. Relating themicro and macro. Pages 356–387 in J.C Alexander,B. Giesen, R. Munch, and N.J. Smesler, editors.The Micro-Macro Link. University of CaliforniaPress, Berkeley.

Mwangi, E. 2006. The footprints of history: Pathdependence in the transformation of propertyrights in Kenya’s Maasailand. Journal of Institu-tional Economics 2:157–180.

Myers, N.A. 1993. Tropical forests: The main defor-estation fronts. Environmental Conservation 20:9–16.

Myers, N.A., R.A. Mittermeier, C.G. Mittermeier,G.A.B. da Fonseca, and J. Kent. 2000. Biodiver-sity hotspots for conservation priorities. Nature403:853–858

Myers, R.A. 1998. When do environment-recruitcorrelations work? Reviews in Fish Biology andFisheries 8:285–305.

Myers, R.A., K.G. Bowen, and N.J. Barrowman.1999. Maximum reproductive rate of fish at lowpopulation sizes. Canadian Journal of Fisheriesand Aquatic Sciences 56:2404–2419.

Myers, R.A., and B. Worm. 2003. Rapid worldwidedepletion of predatory fish communiies. Nature423:283.

Naeem, S. 1998. Species redundancy and ecosystemreliability. Conservation Biology 12:39–45.

Nagendra, H. 2007. Drivers of reforestation inhuman-dominated forests. Proceedings of theNational Academy of Sciences 104:15218–15223.

Naiman, R.J., H. Decamps, and M.E. McClain. 2005.AQ35Riparia. Elsevier Academic Press, Amsterdam.

Narayan, D., R. Chambers, M.K. Shah, and P.Petesch. 2000. Voices of the Poor: Crying Out forChange. Oxford University Press, New York.

Naylor, R.L. 1991. The rural labor market inIndonesia. Pages 58–98 in S. Pearson, W. Falcon,P. Heytens, E. Monke, and R. Naylor, editors. RicePolicies in Indonesia. Cornell University Press.Ithaca.

Naylor, R.L. 1994. Culture and agriculture: employ-ment practices affecting women in Java’s riceeconomy. Economic Development and CulturalChange 42:509–535.

Naylor, R.L. 2000. Agriculture and global change.Pages 462–475 in G. Ernst, editor. Earth Systems:Processes and Issues. Cambridge University Press,Cambridge.

Naylor, R.L., D.S. Battisti, D.J. Vimont, W.P. Falcon,and M.B. Burke, 2007a. Assessing risks of climatevariability and climate change for Indonesian riceagriculture. Proceedings of the National Academyof Sciences 104:7752–7757.

Naylor, R.L., and M. Burke. 2005. Aquacultureand ocean resources: Raising tigers of the sea.Annual Review of Environment and Resources 30:185–218.

Naylor, R.L., and P.R. Ehrlich. 1997. Natural pestcontrol services and agriculture. Pages 151–174in G.C. Daily, editor. Nature’s Services: SocietalDependence on Natural Ecosystems. Island Press,Washington.

Naylor, R.L., and W.P. Falcon. 2008. The role ofpolicy in agricultural transition. In preparation.Pages ??? in P.A. Matson, R.L. Naylor, I. Ortiz- AQ36Monasterio, and W.P. Falcon, editors. Agriculture,Development and the Environment in the YaquiValley (Mexico). Island Press, Washington.

Naylor, R.L., W. Falcon, and C. Fowler. 2007b. Theconservation of global crop genetic resources inthe face of climate change. Summary report fromPlant Genetic Resources in the Face of ClimateChange Conference, Bellagio, Italy, September3–7, 2007. http://fse.stanford.edu/.

Naylor, R.L., W.P Falcon, R.M. Goodman, M.M.Jahn, T. Sengooba, et al. 2004. Biotechnologyin the developing world: A case for increasedinvestments in orphan crops. Food Policy 29(1):15–44.

Naylor, R.L., R.J. Goldburg, J.H. Primavera, N.Kautsky, M.C.M. Beveridge, et al. 2000. Effectof aquaculture on world fish supplies. Nature405:1017–1024.

Naylor, R.L., A. Liska, M. Burke, W. Falcon, J.Gaskell, et al. 2007c. The ripple effect: Biofuels,food security and the environment. Environment49(9):30–43.

Naylor, R.L., H. Steinfeld, W. Falcon, J. Galloway,V. Smil, et al. 2005. Losing the links between live-stock and land. Science 310:1621–1622.

376 References

Neis, B., and L. Felt. 2000. Finding Our Sea Legs:Linking Fishery People and Their Knowledge withScience and Management. Institute for Social andEconomic Research, St. Johns, Canada.

Nelson, D.R., W.N. Adger, and K. Brown. 2007.Adaptation to environmental change: Contribu-tions of a resilience framework. Annual Review ofEnvironment and Resources 32:395–419.

Neuwirth, R. 2006. Shadow Cities: A BillionSquatters, An Urban New World. Routledge,New York.

Nicholls, N., and K.K. Wong. 1990. Dependenceof rainfall variability on mean rainfall, latitude,and the southern oscillation. Journal of Climate 3:163–170.

Niemela, J. 1999. Ecology and urban planning. Bio-diversity and Planning 8:118–131.

NOAA (National Oceanic and AtmosphericAdministration). 2004. Population Trends Alongthe Coastal United States: 1980–2008. U.S.Department of Commerce, Washington.

Norberg, J., J. Wilson, B.H. Walker, and E. Ostrom.2008. Diversity and resilience of social-ecologicalsystems. Pages 46–79 in J. Norberg and G.S.Cumming, editors. Complexity Theory for a Sus-tainable Future. Columbia University Press, NewYork.

Norgaard, R.B. 1989. The case for methodologicalpluralism. Ecological Economics 1:35–57.

North, D.C. 1990. Institutions, Institutional Changeand Economic Performance. Cambridge Univer-sity Press, New York.

North, D.C. 1991. Institutions. Journal of EconomicPerspectives 5:97–112.

Northridge, M.E., E.D. Sclar, and P. Biswas. 2003.Sorting out the connection between the built envi-ronment and health: A conceptual framework fornavigating pathways and planning healthy cities.Journal of Urban Health: Bulletin of the New YorkAcademy of Medicine 80:556–568.

Nowak, D.J. 1994. Atmospheric carbon dioxidereduction by Chicago’s urban forest. Pages 83–94 in E.G. McPherson, editor. Chicago’s UrbanForest Ecosystem: Results of the Chicago UrbanClimate Project. Northeastern Research Station,USDA Forest Service, Radnor, PA.

NRC (National Research Council). 1996. TheBering Sea Ecosystem. National Academy Press,Washington.

NRC (National Research Council). 1999. Sharingthe Fish: Toward a National Policy on Individ-ual Fishing Quotas. NRC (Committee on Ecosys-tem Management and Sustaining Marine Fish-eries, Ocean Studies Board, Commission on Geo-

sciences, Environment, and Resources). NationalAcademy Press, Washington.

NRC (National Research Council). 2000. WatershedManagement for Potable Water Supply: Assessingthe New York City Strategy. National AcademyPress, Washington.

NRC (National Research Council). 2002. TheDrama of the Commons. in E. Ostrom, T Dietz,N. Dolsak, P.C. Stern, S. Stovich, et al., editors.Committee on the Human Dimensions of GlobalChange. National Academy Press, Washington.

NRC (National Research Council). 2003. Decline ofthe Stellar Sea Lion in Alaskan Waters: UntanglingFood Webs and Fishing Nets. National AcademyPress, Washington.

Nystrom, M. 2006. Redundancy and response diver-sity of functional groups: Implications for theresilience of coral reefs. Ambio 35:30–35.

Nystrom, M., and C. Folke. 2001. Spatial resilienceof coral reefs. Ecosystems 4:406–417.

Odum, E.P. 1989. Ecology and Our EndangeredLife-Support Systems. Sinauer Associates, Sun-derland, MA.

OECD (Organization for Economic Cooperationand Development). 1997. Towards SustainableFisheries: Economic Aspects of the Manage-ment of Living Marine Resources. Organizationfor Economic Cooperation and Development,Paris.

O’Flaherty, R., I.J. Davidson-Hunt, and M.Manseau. 2008. Indigenous knowledge andvalues in planning for sustainable forestry:Pikangikum First Nation and the WhitefeatherForest Initiative. Ecology and Society 13(1):6.http://www.ecologyandsociety.org/vol13/iss1/art6/

Oki, T., and S. Kanae. 2006. Global hydrologi-cal cycles and world water resources. Science313:1068–1072.

Oldfield, F., and W. Steffen. 2004. The earth sys-tem. Page 7 in W. Steffen, A. Sanderson, P.Tyson, J. Jager, P. Matson, et al., editors. GlobalChange and the Earth System: A Planet UnderPressure. IGBP Global Change Series. Springer-Verlag, New York.

Olsen, J.P. 2001. Garbage cans, new institutionalismand the study of politics. American Political Sci-ence Review 95:191–198.

Olsson, P., C. Folke, and F. Berkes. 2004a. Adaptiveco-management for building resilience in social-ecological systems. Environmental Management34:75–90.

Olsson, P., C. Folke, and T. Hahn. 2004b. Social-ecological transformation for ecosystemmanagement: The development of adaptive

References 377

co-management of a wetland landscape insouthern Sweden. Ecology and Society 9(4):2.www.ecologyandsociety.org/vol9/iss4/art2/.

Olsson, P., C. Folke, and T.P. Hughes. 2008. Navi-gating the transition to ecosystem-based manage-ment of the Great Barrier Reef, Australia. Pro-ceeding of the National Academy of Sciences.AQ37

Olsson, P., L.H. Gunderson, S.R. Carpenter, P.Ryan, L. Lebel, et al. 2006. Shooting the rapids:Navigating transitions to adaptive governanceof social-ecological systems. Ecology and Soci-ety 11(1): 18. http://www.ecologyandsociety.org/vol11/iss1/art18/.

Oppenheimer, S. 2004. Out of Eden: The Peopling ofthe World. Constable, London.

Ostfeld, R.S., and F. Keesing. 2000. Biodiversity anddisease risk: The case of Lyme disease. Conserva-tion Biology 14:722–728.

Ostrom, E. 1990. Governing the Commons: TheEvolution of Institutions for Collective Action.Cambridge University Press, Cambridge.

Ostrom, E. 1999a. Coping with the tragedies ofthe commons. Annual Review of Political Science2:493–535.

Ostrom, E. 1999b. Institutional rational choice: Anassessment of the institutional analysis and devel-opment framework. Pages 35–71 in P.A. Sabatier,editor. Theories of the Policy Processes. WestviewPress, Boulder.

Ostrom, E., 2005. Understanding Institutional Diver-sity. Princeton University Press, Princeton.

Ostrom, E., M.A. Janssen, and J.M. Anderies.2007. Going beyond panaceas. Proceeding of theNational Academy of Sciences 104:15176–15178.

Otterman, J., A. Manes, S. Rubin, P. Alpert, andD.O. Starr. 1990. An increase of early rainsin southern Israel following land-use change.Boundary-Layer Meteorology 53:333–351.

Page, S.E. 2007. Uncertainty, difficulty, and com-plexity. Journal of Theoretical Politics 20:115–149.

Palm, C.A., S.A. Vosti, P.A. Sanchez, and P.J.Ericksen. 2005. Slash-and-Burn agriculture: TheSearch for Alternatives. Columbia UniversityPress. New York.

Paloheimo, J.E., and L. Dickie. 1964. Abun-dance and fishing success. Rapports et Proces-verbaux des Reunions, Council International pourl’Exploration de la Mer 155:152–63.

Parris, T.A., and R.W. Kates. 2003. Character-AQ38izing and measuring sustainable development.Annual Review of Environment and Resources 28:559–586.

Parry, M., C. Rosenzweig, A. Iglesias, M. Livermore,and G. Fischer. 2004. Effects of climate change

on global food production under SRES emissionsand socio-economic scenarios. Global Environ-mental Change 14:53–67.

Patz, J.A., U.E.C. Confalonieri, F.P. Amerasinghe,K.B. Chua, P. Daszak, et al. 2005. Human health:Ecosystem regulation of infectious diseases. Pages391–415 in R. Hassan, R. Scholes, and N. Ash, edi-tors. Ecosystems and Human Well-Being: CurrentState and Trends. Millennium Ecosystem Assess-ment. Island Press, Washington.

Patz, J.A., P. Daszak, G.M. Tabor, A.A. Aguirre,and M. Pearl. 2004. Unhealthy landscapes: Policyrecommendations on land use change and infec-tious disease emergence. Environmental HealthPerspectives 101:1092–1098.

Pauly, D. 1995. Anecdotes and the shifting baselinesyndrome. Trends in Ecology & Evolution 10:430.

Pauly, D., J. Alder, A. Bakun, S. Heileman, K.-H.Kock, et al. 2005. Marine fisheries systems. Pages477–511 in R. Hassan, R. Scholes, and N. Ash, edi-tors. Ecosystems and Human Well-Being: CurrentState and Trends. Millennium Ecosystem Assess-ment. Island Press, Washington.

Pellow, D.N. 2000. Environmental inequality forma-tion: Toward a theory of environmental injustice.American Behavioral Scientist 43:581–601.

Perera, A.J., L.J. Buse, and M.G. Weber. 2004. Emu-lating Natural Forest Landscape Disturbances:Concepts and Applications. Columbia UniversityPress, New York.

Peterson, G.D., G.S. Cumming, and S.R. Carpen-ter. 2003. Scenario planning: A tool for conserva-tion in an uncertain world. Conservation Biology17:358–366.

Peterson, J.H. 1994. Sustainable wildlife use forcommunity development in Zimbabwe. Pages 99–111 in M.M.R. Freeman and U.P. Kreuter, edi-tors. Elephants and Whales: Resources for Whom?Gordon and Breach Publishers, Basel.

Petit, J.R., J. Jouzel, D. Raynaud, N.I. Barkov, J.M.Barnola, et al. 1999. Climate and atmospheric his-tory of the past 420,000 years from the Vostok icecore, Antarctica. Nature 399:429–436.

Petschel-Held, G., A. Block, M. Cassel-Gintz,J. Kropp, M.K.B. Ludeke, et al. 1999. Syn-dromes of global change: A qualitative modelingapproach to assist global environmental manage-ment. Environmental Modeling and Assessment 4:295–314.

Pfeffer, J. 1981. Power in Organizations. HarvardBusiness, Stanford.

Pickett, S.T.A., W.R. Burch, and S. Dalton. 1997a.Integrated urban ecosystem research. UrbanEcosystems 1:183–184.

378 References

Pickett, S.T.A., W.R. Burch, S. Dalton, T. Foresman,J.M. Grove, et al. 1997b. A conceptual frameworkfor the study of human ecosystems in urban areas.Urban Ecosystems 1:185–199.

Pickett, S.T.A., and M. Cadenasso. 2007. Link-ing ecological and built components of urbanmosaics: An open cycle of ecological design. Ecol-ogy Future Directions 1:1–5.

Pickett, S.T.A., and M. Cadenasso. 2008. Patchdynamics as a conceptual tool to link ecologyand design. Pages 94–103 in B. McGrath, V.Marshall, M.L. Cadenasso, J.M. Grove, S.T.A.Pickett, et al., editors. Designing Patch Dynamics.Columbia University, New York.

Pickett, S.T.A., M.L. Cadenasso, and J.M. Grove.2004. Resilient cities: Meaning, models andmetaphor for integrating the ecological, socio-economic, and planning realms. Landscape andUrban Planning 69:369–384.

Pickett, S.T.A., M.L. Cadenasso, J.M. Grove, C.H.Nilon, R.V. Pouyat, et al. 2001. Urban ecologicalsystems: Linking terrestrial ecological, physical,and socioeconomic components of metropolitanareas. Annual Review of Ecology and Systematics32:127–157.

Pickett, S.T.A., P. Groffman, M.L. Cadenasso, J.M.Grove, L.E. Band, et al. 2008. Beyond urban leg-ends: An emerging framework of urban ecologyas illustrated by the Baltimore Ecosystem Study.BioScience 58:139–150.

Pickett, S.T.A., and P.S. White. 1985. The Ecologyof Natural Disturbance as Patch Dynamics. Aca-demic Press, New York.

Pimentel, D., L. Lach, R. Nuniga, and D. Morrison.2000. Environmental and economic costs of non-indigenous species in the United States. Bio-Science 50:53–65.

Pinkerton, E.W. 1987. The fishing-dependent com-munity. Pages 293–325 in P. Marchak, N. Guppy,and J. McMullan, editors. Uncommon Property:The Fishing and Fish Processing Industries ofBritish Columbia. University of British ColumbiaPress, Vancouver.

Pinkerton, E.W. 1989a. Attaining better fisheriesmanagement through co-management: Prospects,problems, and propositions. Pages 3–33 inE.W. Pinkerton, editor. Co-operative Manage-ment of Local Fisheries: New Directions forImproved Management and Community Devel-opment. University of British Columbia Press,Vancouver.

Pinkerton, E.W., editor. 1989b. Co-operative Man-agement of Local Fisheries: New Directions forImproved Management and Community Devel-

opment. University of British Columbia Press,Vancouver.

Pinkerton, E.W. 1991. Locally based water qualityplanning: Contributions to fish habitat protection.Canadian Journal of Fisheries and Aquatic Sci-ences 48:1326–1333.

Pinkerton, E.W. 1992. Translating legal rights intomanagement practice: Overcoming barriers to theexercise of co-management. Human Organization51:330–341.

Pinkerton, E.W. 1993. Analyzing co-managementefforts as social movements: The Tin-Wis coali-tion and the drive for forest practice legislation inBritish Columbia. Alternatives 19(3):33–38.

Pinkerton, E.W. 2003a. Toward specificity in com-plexity: Understanding co-management from asocial science perspective. Pages 61–77 in D.C.Wilson, J.R. Nielsen, and P. Degnbol, editors. TheFisheries Co-Management Experience: Accom-plishments, Challenges and Prospects. Kluwer,Dordrecht.

Pinkerton, E. and D. Edwards. 2009. The Elephantin the Room: the Hidden Costs of Leasing Indi-vidual Transferable Fishing Quotas. Marine Pol-icy. in press. AQ39

Pinkerton, E.W., and L. John. 2008. Creatinglocal management legitimacy. Marine Policy 32:680–691.

Pinkerton, E.W., and M. Weinstein. 1995. Fisheriesthat Work: Sustainability through Community-Based Management. The David Suzuki Founda-tion, Vancouver.

Plummer, R., and D. Fennel. 2007. Exploring co-management theory: prospects for sociobiologyand reciprocal altruism. Journal of Environmen-tal Management 85:944–955.

Poff, N.L., J.D. Allan, M.A. Palmer, D.D. Hart,B. Richter, et al. 2003. River flows and waterwars: Emerging science for environmental deci-sion making. Frontiers in Ecology and the Envi-ronment 1:298–306.

Pomeroy, R.S. 2003. The government as a partnerin co-management. Pages 247–261 in D.C.Wilson.,J.R. Nielsen, and P. Degnbol, editors. TheFisheries Co-Management Experience: Accom-plishments, Challenges and Prospects. Kluwer,Dordrecht.

Pomeroy, R.S., and F. Berkes. 1997. Two totango: The role of government in fisheries co-management. Marine Policy 21:465–480.

Pomeroy, R.S., and K. Viswanathan. 2003. Experi-ences with fisheries co-management in SoutheastAsia and Bangladesh. Pages 99–117 in D.C. Wil-son, J.R. Nielsen, and P. Degnbol, editors. The

References 379

Fisheries Co-Management Experience: Accom-plishments, Challenges and Prospects. Kluwer,Dordrecht.

Portney, P., and J. Weyant, editors. 1999. Discountingand Intergenerational Equity. Resources for theFuture, Washington.

Posey, D.A., editor. 1999. Cultural and SpiritualValues of Biodiversity. UNEP and IntermediateTechnology Publications, Nairobi.

Posey, D.A. 2000. Ethnobiology and ethnoecol-AQ40ogy in the context of national laws and interna-tional agreements affecting indigenous and localknowledge, traditional resources and intellectualproperty rights. Pages 35–54 in R. Ellen and H.Harris, editors. Indigenous Environmental Knowl-edge and its Transformations. Harwood AcademicPublishers, Amsterdam.

Post, J.R., M. Sullivan, S.P. Cox, N.P. Lester, C.J.Walters, et al. 2002. Canada’s recreational fish-eries: The invisible collapse? Fisheries 27:6–17.

Postel, S.L. 1999. Pillar of Salt: Can the IrrigationMiracle Last? W.W. Norton, New York.

Postel, S.L. 2001. Growing more food with lesswater. Scientific American 284(2):46–49.

Postel, S.L. 2005. Liquid Assets. Worldwatch Insti-tute, Washington.

Postel, S.L., and S.R. Carpenter. 1997. FreshwaterAQ41ecosystem services. Pages 195–214 in G.C. Daily,editor. Nature’s Services: Societal Dependence onNatural Ecosystems. Island Press, Washington.

Postel, S.L., and B. Richter. 2003. Rivers for Life:Managing Water for People and Nature. IslandPress, Washington.

Postel, S.L., G.C. Daily, and P.R. Ehrlich. 1996.Human appropriation of renewable fresh water.Science 271:785–788.

Power, A.G. 1999. Linking ecological sustainabil-ity and world food needs. Environment, Develop-ment, and Sustainability 1:185–196.

Power,A.G.,and A.S.Flecker.1996.Theroleofbiodi-versity in tropical managed ecosystems. Pages 173–194 in G.H. Orians, R. Dirzo, and J.H. Cushman,editors. Biodiversity and Ecosystem Processes inTropical Forests. Springer-Verlag, New York.

Pretty, J. 2002. Agriculture: Reconnecting People,Land and Nature. Earthscan Publications Ltd.,London.

Pringle, H.J.R., and K.L. Tinley. 2003. Are we over-looking critical geomorphic determinants of land-scape change in Australian rangelands? Ecologi-cal Management and Restoration 4:180–186.

Pringle, R.M. 2005a. The Nile perch in Lake Vic-toria: Local responses and adaptations. Africa75:510–538.

Pringle, R.M. 2005b. The origins of the Nile Perch inLake Victoria. BioScience 55:780–787.

Putnam, R.D. 2000. Bowling Alone: The Collapseand Revival of American Community. Simon &Schuster, New York.

Pyne, S.J. 1997. World Fire: The Culture of Fire onEarth. University of Washington Press, Seattle.

Quinn, J. 1985. Managing innovation: Controlledchaos. Harvard Business Review, May-June 63:73–84.

Rabalais, N.N., R.E. Turner, and W.J. Wiseman, Jr.2002. Gulf of Mexico hypoxia, A.K.A. “The deadzone”. Annual Review of Ecology and Systematics33:235–263.

Radeloff, V.C., R.B. Hammer, S.I. Stewart, J.S. Fried,S.S. Holcomb, et al. 2005. The wildland-urbaninterface in the United States. Ecological Appli-cations 15:799–805.

Rahel, F. 2000. Homogenization of fish faunas acrossthe United States. Science 288:854–856.

Raivio, S., E. Normark, B. Pettersson, and P.Salpakivi-Salomaa. 2001. Science and the man-agement of boreal forest biodiversity: Forestindustries’ views. Scandinavian Journal of ForestResearch Supplement 3:99–104.

Ramakrishnan, P.S. 1992. Shifting Agriculture andSustainable Development: An InterdisciplinaryStudy from North-Eastern India. Parthenon Pub-lishing Group, Park Ridge, NJ.

Ramakrishnan, P.S., K.G. Saxena, and U.M.Chandrashekara, editors. 1998. Conserving theSacred for Biodiversity Management. Oxford andIBH, New Delhi.

Rahmstorf, S., A. Cazenave, J.A. Church, J.E.Hansen, R.F. Keeling, et al. 2007. Recent cli-mate observations compared to projections. Sci-ence 316:709.

Ranjan, R., and V.P. Upadhyay. 1999. Ecologicalproblems due to shifting cultivation. Current Sci-ence 77:1246–50.

Rappaport, R.A. 1967. Pigs for the Ancestors. YaleUniversity Press, New Haven.

Raupach, M.R., G. Marland, G., P. Ciais, P., C. LeQuere, J.G. Canadell, et al. 2007. Global andregional drivers of accelerating CO2 emissions.Proceeding of the National Academy of Sciences104:10288–10293.

Reardon, T., and C.P Timmer. 2007. Transfor-mation of markets for agricultural output indeveloping countries since 1950: How has thethinking changed? Pages 2807–2855 in R. Even-son and P. Pingali, editors. Handbook of Agri-cultural Economics, Volume 3. Elsevier Press,Amsterdam.

380 References

Redman, C.L. 1999. Human Impact on AncientEnvironments. University of Arizona Press,Tucson.

Regier, H.A., and G.L. Baskerville.1986. Sus-tainable redevelopment of regional ecosystemsdegraded by exploitive development. Pages75–101 in W.C. Clark, and R.E. Munn, editors.Sustainable Development of the Biosphere. Cam-bridge University Press, London.

Renault, D., and W.W. Wallender. 2000. Nutritionalwater productivity and diets. Agricultural WaterManagement 45:275–296.

Repetto, R., editor. 2006. Punctuated Equilibriumand the Dynamics of U.S. Environmental Policy.Yale University Press, New Haven.

Rerkasem, B., editor. 1996.Montane MainlandSoutheast Asia in Transition Chiang Mai Univer-sity Consortium. Chiang Mai, Thailand.

Reynolds, J.E., and D.F. Greboval. 1988. Socio-economic effects of the evolution of the Nile perchfisheries in Lake Victoria: A review. FAO, Rome,Italy.

Reynolds, J.F., and D.M. Stafford Smith, editors.2002. Global Desertification: Do Humans CauseDeserts? Dahlem University Press, Berlin.

Reynolds, J.F., D.M. Stafford Smith, E.F. Lambin,B.L. Turner, II, M. Mortimore, et al. 2007. Globaldesertification: Building a science for drylanddevelopment. Science 316:847–851.

Ricketts, T.H., G.C. Daily, P.R. Ehrlich, and C.D.Michener. 2004. Economic value of tropical forestto coffee production. Proceedings of the NationalAcademy of Sciences 101:12579–12582.

Ridgwell, A.J., and A.J. Watson. 2002. Feedbackbetween aeolian dust, climate, and atmosphericCO2 in glacial time. Paleoceanography 17:1059,doi:10.1029/2001PA000729, 2002

Rietkerk, M., S.C. Dekker, P.C. de Ruiter, andJ. van de Koppel. 2004. Self-organized patchi-ness and catastrophic shifts in ecosystems. Science305:2004–2009.

Robertson, G.P., and S.M. Swinton. 2005. Reconcil-ing agricultural productivity and environmentalintegrity: A grand challenge for agriculture. Fron-tiers in Ecology and the Environment 3:38–46.

Robinson, J.B., editor. 1996. Life in 2030: Explor-ing a Sustainable Future for Canada. University ofBritish Columbia Press, Vancouver.

Robinson, J.G., and E.L. Bennett, editors. 2000.Hunting for Sustainability in Tropical Forests.Columbia University Press, New York.

Rockstrom, J., L. Gordon, C. Folke, M. Falken-mark, and M. Engwall. 1999. Linkages amongwater vapor flows, food production, and terres-

trial ecosystem services. Conservation Ecology3:5. http://www.consecol.org/vol3/iss2/art5/.

Rose, G.A., B. deYoung, D.W. Kulka, S.V. Goddard,and G.L. Fletcher. 2000. Distribution shifts andoverfishing the northern cod (Gadus morhua): Aview from the ocean. Canadian Journal of Fish-eries and Aquatic Sciences 57:644–663.

Ross, M.L. 2008. Blood barrels: Why oil wealth fuelsconflict. Foreign Affairs 87:2–8.

Rothschild, B.J., J.S. Ault, P. Goulletquer, and M.Heral. 1994. Decline of the Chesapeake Bay oys-ter population: A century of habitat destructionand overfishing.Marine Ecology Progress Series111:29–39.

Ruthenberg, H., J.D. MacArthur, H.D. Zandstra,and M.P. Collinson, 1980. Farming Systems in theTropics. 3rd Edition. Clarendon, Oxford.

Ruttan, V.W., and Y. Hyami. 1984. Induced inno-vation model of agricultural development. Pages97–114 in C.K. Eicher and J.M. Staatz, editors.Agricultural Development in the Third World.Johns Hopkins University Press, Baltimore.

Safriel, U., Z. Adeel, D. Niemeijer, J. Puigdefabre-gas, R. White, et al. 2005. Dryland Systems. Pages623–662 in R. Hassan, R. Scholes, and N. Ash, edi-tors. Ecosystems and Human Well-Being: CurrentState and Trends. Millennium Ecosystem Assess-ment. Island Press, Washington.

Sainsbury, K.J. 1988. The ecological basis of multi-species fisheries and management of a demersalfishery in Tropical Australia. Pages 349–382 in J.Gulland, editor. Fish Population Dynamics. JohnWiley & Sons, New York.

Sainsbury, K.J., R.A. Campbell, R. Lindholm, andA.W. Whitelaw. 1997. Experimental managementof an Australian multispecies fishery: Examin-ing the possibility of trawl-induced habitat mod-ification. Pages 107–112 in E.K. Pikitch, D.D.Huppert, and M.P. Sissenwine, editors. GlobalTrends in Fisheries Management. American Fish-eries Society, Bethesda, MD.

Sampson, R.N., N. Bystriakova, S. Brown, P. Gonza-lez, L.C. Irland, et al. 2005. Timber, fuel, and fiber.Pages 585–621 in R. Hassan, R. Scholes, and N.Ash, editors. Ecosystems and Human Well-Being:Current State and Trends. Millennium EcosystemAssessment. Island Press, Washington.

Sandford, S. 1983. Management of Pastoral Develop-ment in the Third World. John Wiley, London.

Sarkar, S., R.L. Pressey, D.P. Faith, C.R. Margules,T. Fuller, et al. 2006. Biodiversity conservationplanning tools: Present status and challenges forthe future. Annual Review of Environment andResources 31:123–159.

References 381

Sass, G.G., J.F. Kitchell, S.R. Carpenter, T.R.Hrabik, A.E. Marburg, et al. 2006. Fish com-munity and food web responses to a whole-lakeremoval of coarse woody habitat. Fisheries 31:321–330.

Schaaf, T., X. Zhao, and G. Keil. 1995. Towards aSustainable City: Methods of Urban EcologicalPlanning and Its Application in Tianjin, China.Urban System Consult GmbH, Berlin.

Scheffer, M., S.R. Carpenter, J. Foley, C. Folke, andB. Walker. 2001. Catastrophic shifts in ecosystems.Nature 413:591–596.

Scheffer, M., and E.H. van Nes. 2004. Self-organizedsimilarity: The evolutionary emergence of groupsof similar species. Proceedings of the NationalAcademy of Sciences 103:6230–6235.

Scheffer, M., and F.R. Westley. 2007. The evolu-tionary basis of rigidity: Locks in cells, minds,and society. Ecology and Society 12(2):36. http://www.ecologyandsociety.org/vol12/iss2/art36/.

Schelling, T.C. 1978. Micromotives and Macrobehav-ior. W.W. Norton, New York.

Schellnhuber, H.-J. 2002. Coping with Earth SystemAQ42complexity and irregularity. Pages 151–156 in W.Steffen, J. Jaeger, D.J. Carson, and C. Bradshaw,Clare, editors. Challenges of a Changing Earth.Springer-Verlag, Berlin.

Schlager, E., and E. Ostrom. 1993. Property rightsregimes and coastal fisheries: An empirical anal-ysis. Pages 13–41 in T.L. Anderson and R.T.Simmons, editors. The Political Economy of Cus-toms and Culture: Informal Solutions to the Com-mons Problem. Rowman and Littlefield Publish-ers, Lantham, MD.

Schlesinger, W.H. 1997. Biogeochemistry: An Anal-ysis of Global Change. 2nd Edition. AcademicPress, San Diego.

Schmidhuber, J. 2007. Biofuels: An emergingthreat to Europe’s food security? Notre-Europe.aris, France. http://www.notre-europe.eu/uploads/tx publication/Policypaper-Schmidhuber-EN.pdf.

Schnore, L.F. 1958. Social morphology and humanecology. American Journal of Sociology 63:620–624, 629–634.

Schoennagel, T., T.T. Veblen, and W.H. Romme.2004. The interaction of fire, fuels, and cli-mate across Rocky Mountain forests. BioScience54:661–676.

Schreiber, D. 2003. Salmon farming and salmonpeople: identity and environment in the LeggattInquiry. American Indian Culture and ResearchJournal 27(4):79–103.

Schreiber, D. 2006. First Nations, consultation, andthe rule of law: Salmon farming and colonialism

in British Columbia. American Indian Culture andResearch Journal 30(4):19–40.

Schreiber, E.S.G., A.R. Bearlin, S.J. Nicol, and C.R.Todd. 2004. Adaptive management: A synthe-sis of current understanding and effective appli-cation. Ecological Management and Restoration5:177–182.

Schultz, L., C. Folke, and P. Olsson. 2007. Enhancingecosystem management through social-ecologicalinventories: Lessons from Kristianstads Vat-tenrike, Sweden. Environmental Conservation34:140–152.

Schulze, E.-D., C. Wirth, and M. Heimann. 2000. Cli-mate change: Managing forests after Kyoto. Sci-ence 289:2058–2059.

Scoones, I. 1995. Exploiting heterogeneity: Habitatuse in cattle in dryland Zimbabwe. Journal of AridEnvironments 29:221–237.

Scott, W.R. 2000. Institutions and Organizations.Sage Press, Thousand Oaks, CA.

Selby, M.J. 1993. Hillslope Materials and Processes.2nd Edition. Oxford University Press, Oxford.

Sen, A.K. 1981. Poverty and Famines: An Essay onEntitlement and Deprivation. Clarendon, Oxford.

Senge, P.M. 1990. The Fifth Discipline: The Art andPractice of the Learning Organization. Double-day/Currency, New York.

Shah, T., D. Molden, R. Sakthivadivel, and D.Seckler. 2000. The global groundwater situation:Overview of opportunities and challenges. Inter-national Water Management Institute, Colombo,Sri Lanka.

Shane, G.D. 2005. Recombinant Urbanism: Concep-tual Modeling in Architecture, Urban Design, andCity Theory. John Wiley & Sons Ltd., Chichester,UK.

Shertzer, K.W., and M.H. Prager. 2007. Delay infishery management: Diminished yield, longerrebuilding, and increased probability of stock col-lapse. ICES Journal of Marine Science 64:149–159.

Shindler, B.A., and L.A. Cramer. 1999. Shifting pub-lic values for forest management: Making senseof wicked problems. Western Journal of AppliedForestry 14:28–34.

Shumway, D.L., and G.G. Chandwick. 1971. Influ-ence of kraft mill effluent on the flavor of salmonflesh. Water Research 5:997–1003.

Shvidenko, A., D.V. Barber, R. Persson, P.Gonzalez, R. Hassan, et al. 2005. Forest andwoodland systems. Pages 585–621 in R. Hassan,R. Scholes, and N. Ash, editors. Ecosystems andHuman Well-Being: Current State and Trends.Millennium Ecosystem Assessment. Island Press,Washington.

382 References

Singer, P. 1993. Practical Ethics. Cambridge Univer-sity Press, Cambridge.

Singleton, S. 1998. Constructing Cooperation: TheEvolution of Institutions of Co-Management. Uni-versity of Michigan Press, Ann Arbor.

Smil, V. 2000. Feeding the World: A Challenge for the21st Century. MIT Press, Cambridge, MA.

Smil, V. 2002. Eating meat: Evolution, patterns,and consequences. Population and DevelopmentReview 28:599–639.

Smit, B., and O. Pilifosova. 2003. From adaptationto adaptive capacity and vulnerability reduction.Pages 9–28 in J.B. Smith, R.J.T. Klein, and S.Huq, editors. Climate Change, Adaptive Capac-ity and Development. Imperial College Press,London.

Smit, B., and J. Wandel. 2006. Adaptation, adaptivecapacity and vulnerability. Global EnvironmentalChange 16:282–292.

Smith, B.D. 1998. The Emergence of Agriculture. Sci-entific American Library, New York.

Solow, R.M. 1991. Sustainability: An economist’sperspective. Eighteenth J. Seward Johnson Lec-ture, Marine Policy Center, Woods Hole Oceano-graphic Institution, Woods Holes, MA. Pages179–187 in R. Dorfman and N.S. Dorfman, edi-tors. Economics of the Environment: SelectedReadings. W.W. Norton, New York.

Songorwa, A.N., T. Buhrs, and K.F.D. Hughey.2000. Community-based wildlife management inAfrica: A critical assessment of the literature. Nat-ural Resources Journal 40:603–643.

Spirn, A.W. 1984. The Granite Garden: UrbanNature and Human Design. Basic Books, Inc.,New York.

Stafford Smith, D.M. 2003. Linking environments,decision-making and policy in handling climaticvariability. Pages 131–151 in L. Botterill and M.Fisher, editors. Beyond Drought in Australia: Peo-ple, Policy and Perspectives. CSIRO Publishing,Melbourne.

Stafford Smith, D.M. 2008. The ′desert syndrome′ -causally-linked factors that characterise outbackAustralia. Rangeland Journal 30:3–14.

Stafford Smith, M. 2008. The ′desert syndrome:′

Causally-linked factors that characterise outbackAustralia. The Rangeland Journal 30:3–14.

Stafford Smith, D.M., G.M. McKeon, I.W. Watson,B.K. Henry, G.S. Stone, et al. 2007. Learning fromepisodes of degradation and recovery in vari-able Australian rangelands. Proceedings of theNational Academy of Sciences 104:20690–20695.

Stankey, G.H., B.T. Bormann, C. Ryan, B. Shindler,V. Sturtevant, et al. 2003. Adaptive management

and the Northwest Forest Plan: Rhetoric and real-ity. Journal of Forestry 101:40–46.

Starfield, A.M., K.A. Smith, and A.L. Bleloch. 1990.How to Model it: Problem Solving for the Com-puter Age. McGraw Hill Inc., New York.

Stearns, F., and T. Montag, editors. 1974. TheUrban Ecosystem: A Holistic Approach. Down-den, Hutchinson & Ross, Inc., Stroudsburg, PA.

Steffen, W.L., A. Sanderson, P.D. Tyson, J. Jager,P.A. Matson, et al. 2004. Global Change and theEarth System: A Planet Under Pressure. Springer-Verlag, New York.

Steinfeld, H., P. Gerber, T. Wassenaar, V. Castel,M. Rosales, et al. 2006. Livestock’s Long Shadow:Environmental Issues and Options. U.N. Food andAgricultural Organization, Rome, Italy.

Steinfeld, H., and T. Wassenaar. 2007. The roleof livestock production in carbon and nitro-gen cycles. Annual Review of Environment andResources 32:271–294.

Stephens, S.L. 1998. Evaluation of the effects ofsilviculture and fuels treatments on potentialfire behavior in Sierra Nevada mixed-coniferforests. Forest Ecology and Management 105:21–35.

Stern, N.H. 2007. The Economics of ClimateChange: The Stern Review. Cambridge UniversityPress, Cambridge.

Stevens, S., editor. 1997. Conservation Through Cul-tural Survival: Indigenous Peoples and ProtectedAreas. Island Press, Washington.

Stiassny, M.L., and A. Meyer. 1999. Cichlids of therift lakes. Scientific American 280:64–69.

Stiglitz, J.E. 2002. Globalization and Its Discontents.W.W. Norton, New York.

Stiglitz, J.E. 2006. Making Globalization Work.W.W. Norton, New York.

Stoffle, B.W., D. Halmo, R. Stoffle, and G. Burpee.1994. Folk management and conservation ethicsamong small-scale fishers of Buen Hombre,Dominican Republic. Pages 115–138 in C.L. Dyerand J.R. McGoodwin, editors. Folk Managementin the World’s Fisheries: Lessons for Modern Fish-eries Management. University Press of Colorado,Niwot, CO.

Straussfogel, D. 1997. World-systems theory:Toward a heuristic and pedagogic conceptualtool. Economic Geography 73:118–130.

Suding, K.N., S. Lavorel, F.S. Chapin, III, J.H.C.Cornelissen, S. Diaz, et al. 2008. Scaling envi-ronmental change through the communitylevel: A trait-based response- and-effect frame-work for plants. Global Change Biology 14:1125–1140.

References 383

Sukopp, H., S. Hejny, and I. Kowarik, editors. 1990.Urban Ecology: Plants and Plant Communities inUrban Environments. SPB Academic Publishing,The Hague.

Swain, D.P., and Sinclair, A.F. 1994. Fish distributionand catchability: What is the appropriate measureof distribution? Canadian Journal of Fisheries andAquatic Sciences 51:1046–1054.

Swanson, F.J. 2004. Roles of scientists in forestrypolicy and management: Views from the PacificNorthwest. Pages 112–126 in K. Arabas and J.Bowersox, editors. Forest Futures: Science, Poli-tics, and Policy for the Next Century. Rowman &Littlefield Publishers, Inc., Lanham, MD.

Swanson, F.J., C. Goodrich, and K.D. Moore. 2008.Bridging boundaries: Scientists, creative writers,and the long view of the forest. Frontiers in Ecol-ogy and the Environment 6:doi:10.1890/070076.

Swezey, S., and R. Heizer. 1977. Ritual managementof salmonid fish resources in California. Journalof California Anthropology 4(1):6–29.


Tainter, J. 1988. The Collapse of Complex Societies.Cambridge University Press, Cambridge.

Tansley, A.G. 1935. The use and abuse of vegeta-tional concepts and terms. Ecology 16:284–307.

Walters, C.J., V. Christensen, and D. Pauly. 1999.ECOSPACE: Prediction of mesoscale spatial pat-terns in trophic relationships of exploited ecosys-tems. Ecosystems 2:539–544.

Taylor, B.R., editor. 2005. Encyclopedia of Religionand Nature. Thoemmes Continuum, London.

Taylor, D.E. 2000. The rise of the environmentaljustice paradigm: Injustice framing and the socialconstruction of environmental discourses. Ameri-can Behavioral Scientist 43:508–580.

Taylor, L.H., S.M. Latham, and M.E.J. Woolhouse.2001. Risk factors for human disease emergence.Philosophical Transactions of the Royal Society ofLondon, Series B 356:983–989.

Taylor, M., and S. Singleton. 1993. The communalresource: Transaction costs and the solution ofcollective action problems. Politics and Society21:195–214.

Thalenberg, E. 1998. Fisheries Beyond the Crisis.Canadian Broadcasting Company documentaryfilm. 55 minutes. Box 500 Station A, Toronto,Ontario M5W 1E6.

Thom, R.M. 2000. Adaptive management of coastalecosystem restoration projects. Ecological Engi-neering 15:365–372.

Thomas, S.J. 1994. Seeking equity in common prop-erty wildlife in Zimbabwe. Pages 129–142 inM.M.R. Freeman and U.P. Kreuter, editors. Ele-phants and Whales: Resources for Whom? Gordonand Breach Publishers, Basel.

Thompson, G.E., and F.R. Steiner, editors. 1997.Ecological Design and Planning. John Wiley &Sons, Inc., New York.

Tietenberg,T.2002.Thetradablepermitsapproach toprotecting the commons: What have we learned?”Pages 197–232 in NRC. The Drama of theCommons. National Academy Press, Washington.

Tiffin, M., and M. Mortimore. 1994. Malthus con-troverted: The role of capital and technologyin growth and environment recovery in Kenya.World Development 22:997–1010.

Tilman, D., K. Cassman, P. Matson, R. Naylor,and S. Polasky. 2002. Agricultural sustainabilityand intensive production practices. Nature 418:671–677.

Tilman, D., J. Hill, and C. Lehman. 2006. Carbon-negative biofuels from low-input high-diversitygrassland biomass. Science 314:1598–1600.

Timmer, C.P. 1975. The political economy of rice inAsia: Indonesia. Food Research Institute Studies14:197–231.

Timmer, C.P. 2005. Operationalizing pro-poorgrowth: A case study for the World Bank onIndonesia. The World Bank, Washington.

Timmer, C.P., W. Falcon, and S. Pearson. 1983. FoodPolicy Analysis. The Johns Hopkins UniversityPress, Baltimore.

Timmer, V., and C. Juma. 2005. Biodiversity con-servation and poverty reduction come together inthe tropics: Lessons from the Equator Initiative.Environment 47(4):24–47.

Toepfer, K. 2005. From the Desk of Klaus Toepfer,United Nations Under-Secretary-General andExecutive Director, UNEP. Our Planet.

Tongway, D.J., and J.A. Ludwig. 1997. The conser-vation of water and nutrients within landscapes.Pages 13–22 in J.A. Ludwig, D.J. Tongway, D.O.Freudenberger, J.C. Noble, and K.C. Hodgkinson,editors. Landscape Ecology Function and Man-agement: Principles from Australia’s Rangelands.CSIRO Publishing, Melbourne.

Troy, A., and J.M. Grove. In Press. Property values, AQ43parks, and crime: A hedonic analysis in Balitmore,MD. Landscape and Urban Planning.

Turner, B.L., II, W.C. Clark, R.W.Kates, J.F.Richards,J.T. Mathews, et al., editors. 1990. The Earthas Transformed by Human Action: Global andRegional Changes in the Biosphere over the Past 300Years. Cambridge University Press, Cambridge.

384 References


Turner, B.L., II, and S.R. McCandless. 2004. Howhumankind came to rival nature: A brief his-tory of the human-environment condition and thelessons learned. Pages 227–243 in W.C. Clark, P.Crutzen, and H.-J. Schellnhuber, editors. EarthSystem Analysis for Sustainability. MIT Press,Cambridge, MA.

Turner, M.G., R.H. Gardner, and R.V. O’Neill. 2001.Landscape Ecology in Theory and Practice: Pat-tern and Process. Springer-Verlag, New York.

Turner, N.J., I.J. Davidson-Hunt, and M. O’Flaherty.2003. Living on the edge: Ecological and culturaledges as sources of diversity for social-ecologicalresilience. Human Ecology 31:439–463.

Turner, N.J., M.B. Ignace, and R. Ignace. 2000. Tra-ditional ecological knowledge and wisdom of abo-riginal peoples in British Columbia. EcologicalApplications 10:1275–1287.

Tversky, A., and D. Kahneman. 1974. Judgmentunder uncertainty: Heuristics and biases. Science185:1124–1131

Ulrich, R.S. 1983. Aesthetic and affective responseto natural environment. Pages 85–125 in I. Alt-man and J.F. Wohlwill, editors. Human Behav-ior and Environment: Advances in Theory andResearch. Plenum Press, New York.

UN (United Nations). 2003. Water for people: Waterfor life. UN World Water Development Report.United Nations World Water Assessment Pro-gram. UNESCO, Paris.

UN (United Nations). 2007a. The MillenniumDevelopment Goals Report. United Nations, NewYork.

UN (United Nations). 2007b. Press Release onThe Millennium Development Goals Report. 2July

UN (United Nations). 2008. Population Statistics.http://unstats.un.org/unsd/demographic/products/vitstats/. Accessed January 15, 2008.

UNDP (United Nations Development Programme).2003. Human Development Report 2003: Millen-nium Development Goals: A Compact amongNations to End Human Poverty. Oxford Univer-sity Press, New York.

UNDP (United Nations Development Pro-gramme). 2008. United Nations Develop-ment Programme. The Equator Initiative.http://www.undp.org/equatorinitiative.htm.

USDA and USDI (US Department of Agricultureand US Department of Interior). 1994. Recordof decision for amendments to Forest Service andBureau of Land Management planning documentswithin the range of the northern spotted owl.USDA Forest Service Regional Office, Portland,OR.

Valiela, I. 1995. Marine Ecological Processes. 2ndEdition. Springer-Verlag, New York.

van der Brugge, R., and R. van Raak. 2007. Fac-ing the adaptive management challenge: insightsfrom transition management. Ecology and Soci-ety 12(2):33. http://www.ecologyandsociety.org/vol12/iss2/art33/.

Vannote, R.I., G.W. Minshall, K.W. Cummings, J.R.Sedell, and C.E. Cushing. 1980. The river contin-uum concept. Canadian Journal of Fisheries andAquatic Sciences 37:120–137.

Vetter, S. 2005. Rangelands at equilibrium and non-equilibrium: Recent developments in the debate.Journal of Arid Environments 62:321–341.

Vincent, G., and Y. Bergeron. 1985. Weed synecol-ogy and dynamics in urban environment. UrbanEcology 9:161–175.

Vitousek, P.M. 2004. Nutrient Cycling and Limita-tion: Hawai’i as a Model System. Princeton Uni-versity Press, Princeton.

Vitousek, P.M., J.D. Aber, R.W. Howarth, G.E.Likens, P.A. Matson, et al. 1997. Human alter-ation of the global nitrogen cycle: Sources andconsequences. Ecological Applications 7:737–750.

Vitousek, P.M., P.R. Ehrlich, A.H. Ehrlich, and P.A.Matson. 1986. Human appropriation of the prod-ucts of photosynthesis. BioScience 36:368–373.

Vitousek, P.M., H.A. Mooney, J. Lubchenco, andJ.M. Melillo. 1997. Human domination of Earth’secosystems. Science 277:494–499.

Vorosmarty, C.J., C. Leveque, C. Revenga, R. Bos,C. Caudill, et al. 2005. Fresh water. Pages 165–207 in R. Hassan, R. Scholes, and N. Ash, edi-tors. Ecosystems and Human Well-Being: CurrentState and Trends. Millennium Ecosystem Assess-ment. Island Press, Washington.

Walbran, P.D., R.A. Henderson, J.W. Faithful, H.A.Polach, R.J. Sparks, et al. 1989. Crown-of-thornsstarfish outbreaks on the Great Barrier Reef: Ageological perspective based upon the sedimentrecord. Coral Reefs 8:67–78.

Walker, B.H. 1993. Rangeland ecology: Understand-ing and managing change. Ambio 22:80–87.

Walker, B.H. 1995. Conserving biological diversitythrough ecosystem resilience. Conservation Biol-ogy 9:747–752.

References 385

Walker, B.H., L.H. Gunderson, A.P. Kinzig, C.Folke, S.R. Carpenter, et al. 2006. A hand-ful of heuristics and some propositions forunderstanding resilience in social-ecologicalsystems. Ecology and Society 11(1):13. http://www.ecologyandsociety.org/vol11/iss11/art13/.

Walker, B.H., C.S. Holling, S.R. Carpenter,and A.P. Kinzig. 2004. Resilience, adaptabil-ity and transformability in social–ecologicalsystems. Ecology and Society 9(2):5 http://www.ecologyandsociety.org/vol9/iss2/art5/.

Walker, B.H., and M.A. Janssen. 2002. Rangelands,AQ44pastoralists and governments: Interlinked systemsof people and nature. Philosophical Transactionsof the Royal Society of London, Series B 357:719–725.

Walker, B.H., A. Kinzig, and J. Langridge. 1999.Plant attribute diversity, resilience, and ecosystemfunction: The nature and significance of dominantand minor species. Ecosystems 2:95–113.

Walker, B.H., and J.L. Langridge. 1997. Predict-ing savanna vegetation structure on the basis ofplant available moisture (PAM) and plant avail-able nutrients (PAN): a case study from Australia.Journal of Biogeography 24:813–25

Walker, B.H., and J.A. Meyers. 2004. Thresholdsin ecological and social–ecological systems: adeveloping database. Ecology and Society 9(2):3.http://www.ecologyandsociety.org/vol9/iss2/art3/.


Walker, B.H., D.M. Stafford Smith, N. Abel, andJ. Langridge. 2002. A framework for the deter-minants of degradation in arid ecosystems. Pages75–94 in J.F. Reynolds and D.M. Stafford Smith,editors. Global Desertification: Do Humans CauseDeserts? Dahlem University Press, Berlin.

Walter, E., M. M’Gonigle, and C. McKay. 2000. Fish-AQ45ing around the law: the Pacific salmon manage-ment system as a “structural infringement” ofaboriginal rights. McGill Law Journal 45:263–314.

Walters, C.J. 1986. Adaptive Management of Renew-able Resources. Blackburn Press, Caldwell, NJ.

Walters, C.J. 1997. Challenges in adap-tive management of riparian and coastalecosystems. Conservation Ecology 1(2):1.http://www.consecol.org/vol1/iss2/art1/.

Walters, C.J. 2007. Is adaptive management helpingto solve fisheries problems? Ambio 36:304–307.

Walters, C.J., V. Christensen, and D. Pauly. 1999.ECOSPACE: Prediction of mesoscale spatial pat-terns in trophic relationships of exploited ecosys-tems. Ecosystems 2:539–544.

Walters, C.J., and R. Hilborn. 1976. Adaptive con-trol of fishing systems. Journal of the FisheriesResearch Board of Canada 33:145–159.

Walters, C.J., and C.S. Holling. 1990. Large-scalemanagement experiments and learning by doing.Ecology 71:2060–2068.

Walters, C.J., and S.J.D. Martell. 2004. FisheriesEcology and Management. Princeton UniversityPress, Princeton.

Walters, C.J., R. Goruk, and D. Radford. 1993.Rivers Inlet Sockeye salmon: An experiment inadaptive management. North American Journalof Fisheries Management 13:253–262.

Walter, E., M. M’Gonigle, and C. McKay. 2000.Fishing around the law: the Pacific salmon man-agement system as a “structural infringement”of aboriginal rights. McGill Law Journal 45(1):263–314.

Walters, C.J., and McGuire, J.J. 1996. Lessons forstock assessment from the northern cod col-lapse. Reviews in Fish Biology and Fisheries 6:125–137.

Wapner, P. 1997. Governance in global civil society.Pages 65–84 in O.R. Young, editor. Global Gover-nance: Drawing Insights from the EnvironmentalExperience. MIT Press, Cambridge, MA.

Ward, J.M., and J.G. Sutinen. 1994. Vessel entry- AQ46exit behavior in the Gulf of Mexico shrimp fish-ery. American Journal of Agricultural Economics76:916–923.

Warren-Rhodes, K., and A. Koenig. 2001. Escalat-ing trends in the urban metabolism of Hong Kong:1971–1997. Ambio 30:429–438.

Wasserman, S., and K. Faust. 1994. Social NetworkAnalysis: Methods and Applications. CambridgeUniversity Press, Cambridge.

Watts, D.J. 2003. Six Degrees: The Science of a Con-nected Age. Vintage, London.

WCD (World Commission on Dams). 2000. Damsand development: A new framework for decision-making. Earthscan, London.

WCED (World Commission on Environment andDevelopment). 1987. Our Common Future.Oxford University Press, Oxford.

Weber, J.R., and C.S. Word. 2001. The communi-cation process as evaluative context: What non-scientists hear when scientists speak? BioScience51:487–495.

Weinstein, M. 2006. Not “either-or”: a presenta-tion about traditional knowledge and regulationin relation to salmon farm management in theBroughton Archipelago. Legislative Assembly ofBritish Columbia, Special Committee on Sustain-able Aquaculture. June 26. 9 pp.

386 References

Weinstein, M. 2007. Five years of progress?Understanding changes to clam beaches inthe Broughton. Presentation to the Workshopon Clam Bed Research in the BroughtonArchipelago, Malaspina University-College, Insti-tute for Coastal Research, In partnership with BCMinistry of Agriculture and Lands, BC Ministryof Environment, and Fisheries & Oceans Canada.April 12.

Westley, F. 1995. Governing design: The manage-ment of social systems and ecosystems manage-ment. Pages 391–427 in L.H. Gunderson, C.S.Holling, and S. Light, editors. Barriers and Bridgesto the Renewal of Ecosystems and Institutions.Columbia University Press, New York.

Westley, F. 2002. The devil in the dynamics: Adap-tive management on the front lines. Pages 333–360 in L.H. Gunderson and C.S. Holling, edi-tors. Panarchy: Understanding Transformationsin Human and Natural Systems. Island Press,Washington.

Westley, F., S.R. Carpenter, W.A. Brock, C.S.Holling and L. Gunderson. 2002. Why systems ofpeople and nature are not just social and ecolog-ical systems. Pages 103–119 in L.H. Gundersonand C.S. Holling, editors. Panarchy: Understand-ing Transformations in Human and Natural Sys-tems. Island Press, Washington.

Westoby, M., B. Walker, and I. Noy-Meir. 1989.Opportunistic management for rangelands notat equilibrium. Journal of Range Management42:266–274.

White, W.B., G. McKeon, and J. Syktus. 2003. Aus-tralian drought: The interference of multi-spectralglobal standing modes and travelling waves. Inter-national Journal of Climatology 23:631–662.

Wibe, S., and T. Jones, editors. 1992. Forests: Marketand Intervention Failures. Earthscan PublicationsLtd., London.

Wilson, D.C., J.R. Nielsen, and P. Degnbol, editors.2003. The Fisheries Co-Management Experience.Kluwer Academic Publishers, Dordrecht.

Wilson, E.O. 1998. Consilience: The Unity of Knowl-edge. Alfred A. Knopf, New York.

Wilson, J. 2002. Scientific uncertainty, complex sys-tems, and the design of common-pool institutions.Pages 327–360 in NRC (National Research Coun-cil) in E. Ostrom, T. Dietz, N. Dolsak, P.C. Stern,S. Stovich et al., editors. 2002. The Drama ofthe commons. Committee on the Human Dimen-sions of Global Change. National Academy Press,Washington.

Wilson, J.A., J. Acheson, M. Metcalf, and P.Kleban. 1994. Chaos, complexity, and commu-

nity management of fisheries. Marine Policy19:291–305.

Williamson, O.E. 1993. Transaction cost economicsand organizational theory. Industrial and Corpo-rate Change 2:107–155.

Wolf, A.T. 1998. Conflict and cooperation alonginternational waterways. Water Policy 1:251–265.

Wollenberg, E., D. Edmunds, and L. Buc. 2000.Scenarios as a tool for adaptive forest manage-ment. Center for International Forestry Research,Bogor, Indonesia. http://www.cifor.cgiar.org/ pub-lications/pdf files/Other/SCENARIO.pdf.

Wondolleck, J.M., and S.L. Yaffee. 2000. Mak-ing collaboration work: Lessons from innova-tion in natural resource management. Island Press,Washington.

World Bank. 2007. World Development Indicators. AQ47[online database] URL: http://devdata. world-bank.org/dataonline/. Accessed June 1, 2007.

World Bank. 2004. World Development Report.World Bank. Washington.

Worm, B., E.B. Barbier, N. Beaumont, J.E. Duffy, C.Folke, et al. 2006. Impacts of biodiversity loss onocean ecosystem services. Science 314:787–790.

WRI (World Resources Institute). 2005. WorldResources 2005: The Wealth of the Poor - Manag-ing Ecosystems to Fight Poverty. World ResourcesInstitute, Washington.

Wright, M.M., and R.C. Brown. 2007. Comparativeeconomics of biorefineries based on the biochem-ical and thermochemical platforms. Biofuels, Bio-products and Biorefining 1(1):49–56.

Xue, Y.K., H.M.H. Juang, W.P. Li, S. Prince,R. DeFries, et al. 2004. Role of land sur-face processes in monsoon development:East Asia and West Africa. Journal of Geo-physical Research-Atmospheres 109:D03105,doi:10.1029/2003JD003556.

Yaffee, S.L. 1997. Why environmental policy night-mares recur. Conservation Biology 11:328–337.

Yodzis, P. 2001. Must top predators be culled for thesake of fisheries? Trends in Ecology and Evolu-tion 16:78–84.

Young, M.D. 1985. The influence of farm size on veg-etation condition in an arid area. Journal of Envi-ronmental Management 21:193–203.

Young, O.R. 1982. Resource Regimes NaturalResources and Social Institutions. University ofCalifornia Press, Berkeley.

Young, O.R. 1994. The problem of scale inhuman/environment relationships. Journal ofTheoretical Politics 6:429–447.

Young, O.R. 2002a. Institutional interplay: Theenvironmental consequence of cross-scale

References 387

interactions Pages 263–292 in NRC. The Dramaof the Commons. National Academy Press,Washington.

Young, O.R. 2002b. The Institutional Dimensions ofEnvironmental Change: Fit, Interplay, and Scale.MIT Press, Cambridge, MA.

Young, O.R. 2007. Rights, rules, and common poolsor the perils of studying human/environment rela-tions. Natural Resources Journal 47:1–16.

Young, O.R., F. Berkhout, G.C. Gallopin, M.A.Janssen, E. Ostrom, and S. van der Leeuw. 2006.The globalization of socio-ecological systems: Anagenda for scientific research. Global Environ-mental Change 16:304–316.

Young, O.R., L.A. King, and H. Schroeder, editors.2008. Institutions and Environmental Change:

Principal Findings, Applications, and ResearchFrontiers. MIT Press, Cambridge

Zhu, Y., H. Chen, J. Fan, Y. Wang, Y. Li, J. Chen,J. Fan, et al. 2000. Genetic diversity and diseasecontrol in rice. Nature 406:718–722.

Zimov, S.A., V.I. Chuprynin, A.P. Oreshko, F.S.Chapin, III, J.F. Reynolds, et al. 1995. Steppe-tundra transition: An herbivore-driven biomeshift at the end of the Pleistocene. American Nat-uralist 146:765–794.

Zwanenburg, K.C.T., D. Bowen, A. Bundy, K.Drinkwater, K. Frank, et al. 2002. Decadalchanges in the Scotian Shelf large marine ecosys-tem. Pages 105–150 in K. Sherman and H.R.Skjoldal, editors. Large Marine Ecosystems of theNorth Atlantic. Elsevier, Amsterdam.

BM2

Q. No. Query

AQ1 This reference is not cited in the text part. Please provide.




AQ5 Please update.

AQ6 Please provide page numbers.









AQ15 We have inserted year “2008” as per reference citation. Is this ok?






AQ21 Please provide volume and page numbers.






AQ27 Please update.






AQ33 Please update



AQ36 Please update.

AQ37 Please provide volume and page numbers.


AQ39 Please update this reference.




AQ43 Please update.





Index

AAbel, N.M., 171, 181,

183–185, 319Aber, J.D., 168Accelerated soil erosion, 32Acheson, J.M., 83, 90, 242Adams, L.W., 285Adaptation, 8, 67Adaptive capacity,

20, 22–23, 67depends on, 22–23and resilience

components of, 68fostering, 109

Adaptive comanagement, 77,141, 242

building and maintainingstrong community-basedsocial– ecologicalgovernance, 89–93

design principles forsuccessful management ofcommons, 91

general principles for robustgovernance of resourceregimes, 92

building responsive institutionsproviding collective action,80–82

conditions facilitatinginnovation in, 100

cultural lens, 85–86economic lens, 84–85facilitating adaptive learning,

96–99finding good institutional fit

with social–ecologicalsystem, 87–89

institutionalarrangements, 87

generating innovativeoutcomes, 99–100

linking scales of governance forcommunication,responsiveness, andaccountability, 93–96

no one model, discipline, orsolution, 82–84

lens for assessment ofinstitutional performance,84

political lens, 86–87process, 78social–ecological governance,

78–80recurring nightmares, 79

Adaptive cycles, 21conservation phase, 16and cross-scale linkages, 15growth phase, 16release phase, 15renewal phase, 15

Adaptive governance, 118, 120framework, 120–121

Adaptive learningin governance, 96

Adaptive management, 78, 96active and passive, 98

Adaptive Management Areas(AMAs), 156

Addams, L., 272Adger, W.N., 13, 20, 22, 24, 56, 58,

61, 62, 63, 65, 67, 72Agardy, T., 223Agar, N., 211Agency capture, 81

Agistment, 184Agrawal, A., 90, 158–159,

243, 251Agricultural involution, 269Agricultural supply and demand

market dynamics of, 266Agricultural systems, 335Air and water pollution, 64Albedo, 43Alimaev, I.I., 184Allan, C., 237Allan, J.A., 215Allison, E.H., 55, 73Allison, H.E., 112Alternative stable states, 11, 14Amelung, T., 261–262American national parks model

limitations, 131–132Amplifying feedbacks, 10

See also Stabilizing feedbacksAnderies, J.M., 82Anderson, E.N., 242Andersson, E., 288, 291Andow, D.A., 45Angelsen, A., 262Anthropocene, 55, 299, 302, 306Anthropogenic fixation

of nitrogen, 35Aquaculture, 259Arable land, 263Argyris, C., 80, 97Aridity index, 175Armitage, D., 77, 78, 104, 144,

145, 242Arrow, K., 18, 19Artisanal fisheries, 244Assisted migration, 39Attorre, F., 284

389

390 Index

BBackcasting, 98

See also ForecastingBaker, L.A., 285Bala, G., 43, 152Baland, J.-M., 243Baliga, S., 69Balirwa, J.S., 209Band, L.E., 285Bandura, A., 256Banerjee, A.V., 265Barbier, E.B., 19, 214Bardhan, P., 304Baskerville, G.L., 112Bates, R.H., 304Batty, M., 286, 287, 288, 291, 292Bawa, K.S., 133, 159, 262Bayliss, K., 214Befu, H., 251Behnke, R.H.J., 178, 184Bellwood, D.R., 38, 233Bengtsson, J., 117Benjamin, C., 248, 249Bennett, E.M., 115, 137, 208,

215–216Bennett’s Law, 264Benthic dominance state, 223Berger, J.R., 138Bergeron, Y., 284Berger, W.H., 138Berkes, F., 8, 13, 17, 20, 24, 25, 46,

69–72, 80, 88, 90, 93–94,104, 113, 129, 132–133,135, 141, 144, 146, 242, 250

Berkowitz, A.R., 13Bernstein, S.F., 56Berry, W., 279Bhagwat, S., 133Biermann, F., 93, 306Biggs, H.C., 189, 190, 191–193Binswanger, H., 113Biodiversity, 181“Biodiversity hotspots”, 133Bioeconomic equilibrium, 226Biological diversity, 116

maintenance of, 36–38Biosphere Reserves, 130Birkeland, P.W., 32Blaikie, P., 181Blending Forest Production and

Conservation intheWestern USA, 155–157

Blue water, 33, 199and green-water flows, 201

Bodin, O, 104Bolund, P., 291Bonfil, R., 224Boone, C., 290–291Boreal case study

of indigenous people keepingland healthy, 135–137

Borgstrom, S.T., 292Bormann, B.T., 155Bormann, F.H., 283, 285Bottleneck breaking, 266Boudouris, K., 8Bowles, S., 109, 116Boyden, S., 283, 285Bradley, J., 281Bradshaw, A.D., 39Brady, N.C., 32Brand, S., 281–283, 288, 289–291Bredenkamp, G.J., 284Breisch, L.L., 232Bridging organizations

connecting different levels ofgovernance, 120

Brock, W.A., 104, 198, 213Brodie, J., 233Bromley, D.W., 80, 84Brousse tigre or banded shrub

landscapefunctioning of, 177

Brouzes, R.J.P., 255Brown, K., 133, 159, 276Brunner, R.D., 78Buckles, D., 250Built capital, 18Burch, W.R., 283, 285–287Bureaucracies, 81Bureaucratization, 81Burke, E.J., 173, 179Burke, M.B., 259, 278BurnSilver, 69Burton, I., 63Burton, P.J., 154, 167Busch, D.E., 98Butzer, K.W., 8

CCaddy, J.F., 228Cadenasso, M.L., 284, 286Campbell, B.D., 182Campbell, R.A., 236CAMPFIRE, see Communal

Areas ManagementProgram for IndigenousResources (CAMPFIRE)

Canadell, J.G., 152“Cap-and-trade”, 214Carbon and nutrient cycling

aquatic ecosystems, 35categories of climate

feedbacks, 34major controls, 34terrestrial ecosystems, 34water column of deep ocean, 36

Carpenter, S.R., 12, 13, 15, 17, 25,35, 104, 105–106, 115, 197,198, 206–209, 213, 214, 319

Carr, C., 178Carson, R., 8, 44, 56Cassman, K.G., 202–203, 215, 262,

275–276, 279–280Castilla, J.C., 250Catchability coefficient, 229Centralized vs. decentralized

organizations, 81Cereal consumption and energy

consumption, percapita, 275

Chambers, R., 55Chambers, S., 43Change

critical implications ofmanaging, 51

Chan, S., 281, 283,290, 293

Chaotic behavior, 10–11Chapagain, A.K., 215Chapin, F.S.3, 7, 10, 12, 20, 22, 29,

32, 34, 36, 39, 43, 55, 68, 72,103, 119, 129, 149,164–166, 171, 183, 319

Chase-Dunn, C., 13Chemical pollution, 208Chen, S., 263Chesapeake Bay oyster fishery

development and declineof, 232

Chevalier, J., 250Chilton, P J., 203Choi, J.S., 223Choi, Y.D., 41Christensen, N.L., 29–30Christensen, V., 221Christy, F., 224, 251Cinner, J., 249Cissel, J.H., 167Cities, 281, 335

as functional social–ecologicalsystems

Index 391

ecosystem services anddynamics of cities, 291–292

management of complexadaptive systems, 292–293

importance of urbanizationand, 281–282

principlescomplex adaptive systems

cities, 286–291heterogeneous cities and

ecosystem composites,285–286

open cities and multiscalesystems, 284–285

urban ecosystem, 283urbanization – dynamic, social,

and ecologicalphenomenon, 282–283

use of approaches, 283–284CitiStat-type

programs, 292Civil society, 248Clark, W.C., 8, 18, 226Clay, G., 285Climate change

challenges, 167–168, 308Climate system

feedbacks in, 298Clinton, B.D., 156Coastal marine systems

benefits of involvingcommunities in, 253–256

general conditions under whichcoastal communitiesconserve marine resources

characteristics of communitylend to comanagement,251–252

nature of community’srelationship with outsidegroups and government,252–253

nature of resource lend tocomanagementarrangements, 251

strong access rights to localmarine resources, 245

strong rights to participate inmanagement decisions,245–251

importance of, 241–242international recognition of

local fishing communities,243–244

problems of privatizing,242–243

Coastal oceans, 334Coefficient of variation, 180Cognitive or mental maps, 85Cohen, F., 247Cohen, M.D., 11Colding, J., 39, 41, 286, 288, 291Cole, J.J., 203Coleman, J., 18, 61Cole, M.M., 175Collier, P., 312, 313Collins, S., 283Comanagement, 78, 94, 141,

242, 246Case of Porcupine Caribou

Herd ComanagementAcross an InternationalBorder, 142–144

contributing to communitylivelihoods andsocial–ecologicalresilience, 144

functions, 141Command-and-control approach,

39, 79Commercial forest, 137Common-pool resources

tragedy of the commons, 83Communal Areas Management

Program for IndigenousResources (CAMPFIRE)

Elephant Conservation andCommunity Livelihoods inSouthern Africa, 138–140

Communities of practice/learningnetworks, 96

Community-based wildlifemanagement, 132, 138, 140

CAMPFIRE program of, 138Community forestry, 165Community partnership

programs, 95Community quotas, 243Complex adaptive systems, 9, 14

implications for resourcestewardship, 15

Concurrent management, 246Congestible resources, 88Conservation, 247

locally driven approaches,133–134

principles, 130strategies, 133

Conservation, community andlivelihoods

challenges, 130conservation issues

resource comanagement,141–145

single-species managementof wild resources, 145–146

threatened species andcommunity-basedconservation, 137–141

traditional management ofbiodiversity, 134–137

dimensions of linking, 134social–ecological properties

supporting, 132–134biodiversity of groups of

organisms in forestreserves, sacred groves,and coffee agroforestryplantations, 134

strategies for conservation andbiodiversity, 130–132

categories of protected areas,130–131

Conservation Reserve Program(CRP), 276

Conservation by utilization, 139Contreras-Hermosilla, A., 157Conventional resource

management, 78See also Steady-state resource

managementConvention on International

Trade in EndangeredSpecies (CITES), 139

Conway, G., 55, 261, 264, 268–269Cooke, G.D., 217Coping, 67

range and extreme events, 67Coppolillo, P., 132, 146Corcoran, P.J., 236Cordell, J., 242Cork, S.J., 115Costa, M.H., 43, 205Costanza, R., 13, 19, 46, 79, 104Couzin, J., 282Cramer, L.A., 18, 49Crisis

leading to opportunities, 331Cross-scale linkages, 13

categories, 93strategies for, 90

Cross-scale surprises, 106

392 Index

Crutzen, P.J., 55, 299Cultural diversity, 69Cultural ecology/human

ecology, 241Cultural identity and heritage, 46Cultural services, 31, 46–49Cumulative processes, 299Curran, S., 276Curtis, A., 237Cutter, S.L., 65

DDaily, G.C., 19, 269Daly, H., 19Danell, K., 164D’Antonio, C.M., 36D’Antonio, C.M., 36Dasgupta, P., 18, 57Davidson-Hunt, I.J., 135, 137Davidson, R., 171Davis, M., 289Dayanandan, S., 262DeAngelis, D.L., 10, 14Decision-support tools, 98Decomposition, 34Deforestation, 60, 153

relative extent of reforestationand, 160

Degnbol, P., 253de Groot, R., 46–48DeLuca, D.R., 283, 285–287Depensatory, 228Dercon, S., 278Desalinization, 217Desertification, 32Deutsch, B., 112, 277, 284Deutsch, L., 112, 277, 282, 284Development and empowerment

organizations, 95DeVore, I., 241deYoung, B., 243de Young, B., 253, 255Diamond, J., 3, 25, 29, 111,

273, 310Dickie, L., 229Dickson, N.M., 8, 18Diehl, M., 262Diener, E., 59Dietz, T., 78, 90–93, 120, 141, 158,

210, 218Direct and indirect grain

consumptionas function of income, 274

Dirzo, R., 43, 152

Discount rate, 213Disturbance regime, 12, 38Disturbances, 38Diversity, 67Double-loop learning, 97

See also Single-loop learningDouglas, M., 85Dow, K., 286Downing, J.A., 212Drip irrigation, 216Driscoll, C.T., 166Dryland aridity gradient

statistics, 172Dryland degradation

biophysical or social driverscausing, 179–180

Drylands, 175, 332case studies

crop–livestock system inSouthern Sahel, WesternNiger, 186–187

immigration and off-siteimpacts in inner Mongolia,188–189

learning from century ofdegradation episodes inAustralia, 187–188

managing under uncertaintyin Kruger National Park,South Africa, 189–193

causes of degradation andimpacts

biodiversity loss, 181–182commercialization and

diversification, 185–186disruptive subsidies, 184–185effect of scale, 180expanding aridity, 180–181intensification and

abandonment, 182–183loss of local community

governance structures,183–184

features and functioning,173–178

biological, 175–177physical, 175socioeconomic, 177–178

Drylands development paradigm(DDP)

principles, 174Drylands syndrome

factor interactions andfeedbacks in, 173

Dryland subtypeskey features of, 172

Duffin, K.I., 191Duflo, E., 265Duit, A., 119Duncan, O.D., 283Dustbowl, 33du Toit, J.T., 189, 190, 193Dynamics of denial, 230Dysfunctional states, 113Dyson, M., 211

EEagle, J., 265Earth

as social–ecologicalsystem, 295

Earth’s social-ecological systemssources of vulnerability,

resilience andtransformations of, 333

Earth system, 335applications – climate and

MDGscontrolling climate change,

308–311fulfilling millennium

development goals,311–313

challenge of earth systemgovernance, 306–308

characteristics, 296common threads, 305–306directional changes at

planetary scaleenvironmental–social

interactions, 304–305Global Environmental

Changes (GECs), 299–302Global Social Changes

(GSCs), 302–304Easterling, W.E., 277Easterlin, R.A., 59Ebbin, S., 248Ecological engineering, 116Ecological reserve, 212Ecological restoration, 116Ecological thresholds, 8Economic diversity, 68–69Economic rent, 104Ecosystem, 5, 30, 43, 44,

131, 283Ecosystem management, 51,

154, 165

Index 393

attributes of, 30goal of, 29

Ecosystem services, 5, 8approaches to material needs

of society and flow of, 59biodiversity effects on, 37critical, 49general categories and societal

benefits, 31See also Cultural services;

Provisioning services;Regulating services

Ecosystem stewardship,6, 155, 320

challenges to, 4linkages among ecosystem

services, well-being ofsociety, and, 30

practical approaches toimplementing, 321

resilience-based, 5steady-state resource

management vs, 320Ecosystem stewardship,

transformations incommand-and-control to

complex systems, 104dealing with uncertainty and

surprise, 105–106possible futures of

social–ecologicalsystems, 106

fostering desirablesocial–ecologicaltransformations

creating thresholds fortransformation fromundesirable states, 116–118

identifying dysfunctionalstates, 113–114

recognizing impendingthresholds fordegradation, 114–116

integrated social–ecologicalperspective, 103–105

navigating transformationthrough adaptivegovernance

interplay between micro andmacro, 120

leadership, actor groups,social networks, andbridging organizations,119–120

learning platforms as part ofadaptive governance,120–121

path dependence andwindows of opportunity,118–119

three phases ofsocial–ecologicaltransformation, 121–124

preventing social–ecologicalcollapse of degradingsystems

institutional misfits withecosystems, 111–113

interplay between gradualand abrupt change,107–109

multiple regimes, 109–110thresholds and cascading

changes, 110–111Ecotones, 137Effect diversity, 36Eggers, D., 238Ehrlich, A.H., 265Ehrlich, P.R., 45, 153Ellis, F., 55, 69, 73Ellis, E.C., 301Ellis, J., 180Ellison, A.M., 166Elmqvist, T., 23, 36, 39, 47, 117,

153, 181, 283El Nino-Southern Oscillation

(ENSO), 175Endogenous stresses, 63

See also Exogenous hazardsEndogenous stresses, 64, 65Engel’s Law, 264, 267Engerman, S., 116Engineering policy

approach, 238Entitlements, 65Environmental events, 8Epstein, M.J., 121Estuaries and oceans, 35Eutrophication, 33, 198, 206Evans, L.T., 261, 264, 266Evolution, 22

of resource-managementparadigms, 320

Ewel, J.J., 261Excludable resources, 88Exogenous controls, 12Exogenous factors, 6Exogenous hazards, 63

Exposure, 63Exposure to hazards and stresses

endogenous, 64–65exogenous, 63–64

Extensification, 42, 261

FFabricius, C., 119Falcon, W.P., 267, 268, 271, 272,

280Falkenmark, M., 199Farber, S., 291Fargher, J.D., 174Fast variables, 12Faust, K., 120Feedback policies, 233Felson, A.J., 286Felt, L., 253“Fence-and-fine” approach, 146Fennel, D., 242Ferguson, T.A., 135, 137Fernandez, M., 250Fernandez, R., 186Fernholz, K., 168Field, C.B., 151, 153Field, D.R., 286Finlayson, A.C., 112, 204Finlayson, C.M., 203, 204, 205Fischer, F., 49, 163Fischer, G., 254Fischer, J., 6Fisheries

long-term changes in, 227loss of resilience and

adaptability of, 237Fisheries management

indicators of erosion ofresilience in, 235

Fishery and Aquatic ResourceManagement Councils(FARMCs), 248

aspects of process, 249Fishery management rights,

246–247Fishery subsystems/

components, 225Fixed escapement rules, 233Fixed exploitation rate, 233Flecker, A.S., 45Foley, J.A., 3, 13, 32, 43, 59, 110,

151, 175Folke, C., 3, 6, 11, 13, 19, 22–23,

38, 40, 69, 70, 71, 77, 79,87–89, 103–104, 108–109,

394 Index

111, 112, 116, 118, 120, 166,282, 290, 319

Food security, 264Ford, J.D., 21Forecasting, 98Foreign direct investment

(FDI), 303Fore, L.S., 254Forest certification, 151Forest exploitation, 154Forest management

changes in views andapproaches to, 155

multiple uses, 165Forest protection, 166Forests, 332

as ecological systems, 6sustaining, 150

Forest systemsconversion to new land uses,

168–169managing under conditions of

rapid changedisturbance management,

166–168global change, 166

multiple use forestry, 165plantation forests, 150as social–ecological systems,

150–151supply and use of ecosystem

services, 151–153sustainable timber harvest

evolving views of people inforests, 153–157

forest exploitation and illegallogging, 157–159

production forestry, 163–165timber production from

natural forests, 159–163Formal institutions, 80

See also Informal institutionsFormal knowledge, 46Fossil groundwater, 42Foster, D.R., 168Foster, S.S.D., 203Fowler, C., 278Fox, J.J., 261, 270Framework for understanding

changeintegrated social–ecological

adaptive cycles, 15–17incorporating scale, human

agency, and uncertainty

into dynamic systems,13–15

issues of scale, 12–13linking physical, ecological,

and social processes, 6–9systems perspective, 9–12

sustainability in directionallychanging world

challenges to, 25–26change management to

foster sustainability, 19–20conceptual framework for

sustainability science,17–18

types and substitutability ofcapital, 18–19

Francis, R., 96Frank, K.T., 223Freeman, M.M.R., 137, 138Free-rider problem, 84Freshwater ecosystems, 334Freshwaters, 197

challenges in watermanagement

cross-generational equityand long-term change,213–214

habitats, fisheries, andmanaging forheterogeneity, 212–213

human versus environmentalflows, 211

upstream/downstream flows,210–211

degradation of freshwaterecosystems

draining, fragmentation, andhabitat loss, 204–206

eutrophication, salinization,and chemical pollution,206–209

invasive species andoverharvesting, 209–210

institutional mechanisms forsolving conflicts

conservation, 216decentralization and

integrationof management, 218

markets and benefit–costcomparisons, 214–215

technology, 216–218virtual water and trade,

215–216

key principles for, 198sustainability of, 198–199

Friedlingstein, P., 151Frost, P., 175Fugitive resources, 88Functional redundancy, 36Functional types, 12, 36Funke, N., 212Future generations

depend on components ofnatural capital, 50

GGalaz, V., 87, 89, 118, 119, 308Gallopin, G.C., 67Galloway, J., 273, 274Garreau, J., 281Geertz, C., 269Geist, H.J., 179Gelcich, S., 250Genuine investment, 18GeoSentinel, 293Getz, W.M., 138, 140Ghassemi, F., 263Giller, K., 261Gillson, L., 191Ginn, W.J., 168Gladwell, M., 119Gleick, P.H., 201, 217, 263, 302Global agricultural land use, 263Global CO2 emissions

compared with emissionscenarios and stabilizationtrajectories, 309

Global environmentalchanges, 299

Global Environment Facility(GEF), 313

Global hydrological fluxes andstorages, 200

Globalization, 3“GLOBE”, 96Goal displacement, 250Goal displacement, 81Goodwin, N.B., 226Gordon, L.J., 41Gordon, M.R., 255Gorgens, A.H.M., 216Gottdiener, M., 286Governance, 183

misfits in, 89polycentric, 93

Governance systems, 118transformations, 119

Index 395

Granovetter, M.D., 82Gray, B., 95Grazing, 176Great Acceleration, 299The Great Barrier Reef Marine

Park (GBRMP), 250The Great Barrier Reef study, 117

strategies, 122Greboval, D.F., 209Greener, Greater New York, 290Green Revolution, 269Green water, 33, 199Griffith, B., 142Griggs, J.R., 95Grimm, N.B., 282–284, 286, 292Groffman, P.M., 25, 283, 285Gross domestic product (GDP),

57, 174, 303Grove, J.M., 281, 283, 285,

286, 291Grumbine, R.E., 154, 160, 165Gunderson, L.H., 8, 11, 15, 16, 17,

23, 25, 71, 72, 79, 96, 105,107, 109, 112, 122, 153, 161,167, 190, 233, 290

Gunter, J., 165Gustafson, R., 46

HHabitat/Species Management

Area, 131Hahn, T., 120Haller, L., 202Hannah, L., 164Hanna, S.S., 85Hara, M., 250Hardin, G., 83, 90Harley, S.J., 229Harmon, M.E., 164Hawley, A.H., 283Haynes, R.W., 156, 157Healey, M., 228Heal, G., 50Heaslip, R., 254Heizer, R., 242Held, D.A., 303Hennessey, T., 228Hensel, C., 144Hewitt, K., 60Hibbard, K.A., 299Hiernaux, P.H.Y., 182, 186Hilborn, R., 223, 236, 238Himmel, M.E., 276Hinterlands, 332

Historic range of variability, 167Hobbs, R.J., 112Hodgkin, T., 278Hoekstra, A.Y., 215Holling, C.S., 11, 13, 15–17, 23, 25,

39, 45, 64, 78–79, 107, 109,111, 156–157, 161, 167,190, 218, 290

Homo sapiens, 319Horizontal linkages, 93

See also Vertical linkagesHostetler, M., 286Hough, M., 284Hughes, T.P., 233Huitric, M., 104Hulme, D., 138, 140Human actions

causing loss of resilience ofecosystems, 108–109

and Earth System responses,change in, 300

Human agency, 14, 82, 119Human capital, 18, 23Human–environment

relationships, 132Human–environment system, 6Human impacts

on erodable phosphorus, 208Human organizations, 81Human use of freshwater

resourcesavailable, 199–201for drinking and sanitation,

201–202for food production, 202–203other services, 203–204

Human–wildlife–landinteractions, 129

Hunhammar, S., 291Hutchinson, R., 286Hutton, G., 202Hyami, Y., 267Hydroponics, 216

IIconic species, 48Illegal logging, 157

actions reducing likelihoodof, 158

contributing factors, 158Imhoff, D., 276Imhoff, M.L., 42Incentive systems, 84Inclusive wealth, 18, 50

Income elasticity of demand, 274Indecision

as rational choice, 225Indecision as rational choice, 230Individual transferable quotas

(ITQs), 242Induced innovation, 267Informal institutions, 80Institutions, 80, 85

mismatches betweenecosystems and, 89

Integrated pest management(IPM), 270

Intensification, 42, 261See also Extensification

Intergovernmental Panel onClimate Change (IPCC),276, 301

Invasive species, 45Investment, 18IPAT Law, 265Irreversibility, 109Ives, A.R., 36

JJackson, J.B.C., 3, 111, 221Jacobs, J., 285, 292James, C.D., 182Janssen, M.A., 20, 56, 111Jansson, A., 112Jarvis, A., 278Jenkins, J.C., 285Jentoft, S., 253Jiang, H., 188, 189Johannes, R., 242John, L., 251Johnson, E.A., 38Johnson, S., 281, 282, 283, 289–293Jolly, D., 64Jones, C.G., 36Jones, T., 113Juma, C., 133Justice, M., 256

KKahneman, D., 306Kaimowitz, D., 262Kaldor, M., 303Kalimtzis, K., 8Kanae, S., 199, 200, 203, 215Kanter, R., 121Kaplan, R., 48Kaplan, S., 48Kasperson, R.E., 22, 25, 58, 63, 64

396 Index

Katz, B., 281Kaufman, L., 209Keane, J., 303Keene, E., 303Keen, I., 104, 184Keesing, F., 46, 146Keesing, J.K., 233Kennedy, V.S., 232Kent, M., 284Keystone species, 36Khagram, S., 8, 263Kidder, T., 121Kiester, A.R., 155King, J.M., 211Kinzig, A.P., 49, 105, 109, 110Kirwan, J., 241Kitchell, J.F., 207, 209, 210Kocher, T.D., 209Koehane, R.O., 93Koenig, A., 283Kofinas, G.P., 3, 55, 77, 95, 98, 99,

100, 129, 141, 142, 144, 145,253, 319

Kokou, K., 153Kolar, C., 209Kolata, G., 305Kolstad, C., 267Kreuter, U.P., 137–138Kruger National Park (KNP), 189

communities of practice,192–193

ecological theory, 191monitoring and adaptive

decision-making, 191–192outcomes and challenges, 193science-management

system, 191strategic approach to river

management, 192thresholds of probable

concern, 191thresholds of probable concern

(TPCs), 192vision statement and objectives

hierarchy, 191Krupnik, I., 64Kundzewicz, Z.W., 205Kursar, T.A., 43, 152

LLach, D., 163Lake Chad, 202Lake Mendota, 206

managing, 207

Lake Victoria, 209–210fish landing on, 210

Lambin, E.F., 113, 179, 279Landsberg, J., 182Landscape function, 176Land use intensification pathways

and changes in stocks, 262Lane, A., 153Langdon, S., 11Langridge, J.L., 175Lansing, S.J., 62Larson, A.M., 154Larsson, S., 164Lathrop, R.C., 207Lebel, L., 112Lee, K.N., 71, 78, 82, 96, 99Lee, R.B., 241Lee, S., 288Legacies, 14Leopold, A., 6, 8, 62Levin, S.A., 14, 22,

110, 295Levy, M., 57, 58Lewis, H.T., 6, 241, 271Liebman, M., 45Life-support system, 3Likens, G.E., 41, 283, 285Limits to Growth, 56Lindblom, C.E., 97Lindgren, E., 46Liska, A.J., 276Litter, 32Livelihoods, 55

diversification of, 69Lobell, D.B., 277, 280Local communities, 89Local knowledge, 47, 70Local management

resource characteristicsaffecting ease/difficultyof, 251

Local surprises, 105–106Lodge, D.M., 209Long-term trends in population

and economy, 287Loucks, L.A., 243Lovelock, J.E., 295Low, B., 73“Lucky hunter”, 143Ludwig, D., 7, 15, 97, 104,

176, 228Ludwig, J.A., 176, 177, 182,

213, 214Lukes, S., 86

MMacCall, A., 228Machlis, G.E., 283, 287, 288Magadlela, D., 216Magnuson, J.J., 197, 207Mahler, J., 86Makse, H.A., 281Malthus, T.R., 8Managed Resource-Protected

Area, 131Management and policy

roles of scientists in, 161–163Managing ecosystems sustainably

cultural servicescultural identity and cultural

heritage, 46–47recreation and

ecotourism, 48spiritual, inspirational, and

aesthetic services, 47–48ecosystem restoration,

39–41biodiversity in ecosystem

renewal after disturbance,roles of, 40

facing realities of ecosystemmanagement

adjusting to change, 51multiple use, 50–51regulations and politics,

51–52sustainability, 49–50

pollination services, 46provisioning services

food, fiber, and fuelwood,42–43

fresh water, 41–42other products, 43

regulating servicesclimate regulation, 43–44natural-hazard reduction,

44–45regulation of erosion, water

quantity, water quality, andpollution, 44

regulation of pests, invasions,and diseases, 45–46

supporting servicescarbon and nutrient cycling,

34–36maintenance of biological

diversity, 36–38maintenance of disturbance

regime, 38–39

Index 397

maintenance of soilresources, 32–33

water cycling, 33synergies and tradeoffs among

ecosystem services, 48–49Managing food production

systems for resilienceagricultural enterprise in

perspectivechallenge, 264hunger, 263intensification, 261–263

case studies on resiliencechallenge

managing for globalization inYaqui Valley, Mexico,271–273

managing for propoorgrowth in Indonesia,269–270

emerging issues forresilience-basedmanagement

biofuels boom, 274–276global climate change,

276–278industrial livestock, 273–274

structural dynamicsdeterminants of demand and

supply, 264–265implications for

resilience-basedmanagement, 268

longer-run adjustments,266–267

role of policy, 267–268short-run adjustments,

265–266Manning, R., 264, 272Mansuri, G., 138Marine-protected areas

(MPAs), 249Martell, S.J.D., 223, 225,

230, 236Martin, J.R., 296Maskin, E., 69Maskrey, A., 56Maslow, A.H., 57, 61, 62Matson, P.A., 33, 42, 259, 271Maximum sustained yield (MSY),

50, 154, 319limitation, 51

McAllister, R.R.J., 184McCandless, S.R., 25

McCarthy, J.J., 22, 66McCay, B.J., 85, 90, 112, 124,

204, 242McDonald, G., 153McGinnis, M.D., 218McGoodwin, J., 251McGrath, B., 286McGrew, D., 303McGuire, J.J., 230McKeon, G.M., 187McKinley, T., 214McLachlan, J.S., 39McNaughton, S.J., 36Mdzeke, N., 216Meadows, D.H., 56Mears, L.A., 270Meffe, G.K., 16, 39, 45, 64,

111, 167Melosi, M.V., 283Memmott, J., 46Mexican Fertilizer Company

(FERTIMEX), 272Meyer, A., 209Meyers, J.A., 189Middleton, N.J., 175Millar, C.I., 164, 167Millennium Ecosystem

Assessment (MEA),115, 204

Mintzberg, H., 121The Mississippi River

Delta, 35Mitchell, C.E., 45Mitigation, 22, 63Models, 98Mollenhoff, D., 207Montag, T., 283, 285Montreal Protocol, 64Mookerjee, D., 304Morrow, P., 144Mortimore, M., 183Mulder, M.B., 132, 146Mullon, C., 222, 228, 229Multiple use management, 154Multistakeholder bodies, 94Murphree, M., 138, 140Mwangi, E., 72Myers, N.A., 42, 261, 263Myers, R.A., 226, 235

NNaeem, S., 117Nagendra, H., 158, 159, 160Narayan, D., 57

National Park, 131Natural capital, 18, 23Natural enemies of pests, 45Natural forests, 159

See also Production forestsNatural Monument, 131Nature Conservation

Act, 140Navigation of transformations

strategies for, 329Naylor, R.L., 45, 153, 203, 259,

260, 265–280, 319Neis, B., 253Nelson, D.R., 67, 104Neoliberalism, 242Neuwirth, R., 289Newfoundland “northern” cod

stockrecent history of, 230

Nicholls, N., 175Nielsen, J.R., 250Niemela, J., 286Nitrogen-fixing species, 34–36, 39,

42, 160Nongovernmental organizations

(NGOs), 137Nonpoint pollution, 206Norberg, J., 23, 104Norgaard, R.B., 86North, D.C., 14, 80Northridge, M.E., 283Northwest Forest Plan

(NWFP), 156Nowak, D.J., 285Nusa Tenggara Timor’s

(NTT’s), 269Nystrom, M., 38, 117

OOceans and estuaries

adaptive policies in fisheriesand coastal zonemanagement, 236–237

dynamics and policy optionsfor sustainability, 237–238

fisheries collapse, 228–231preventing fisheries collapse,

226–228resilience in biologically

variable systems, 233–234slow-variable dynamics and

shifting baselines, 231–233social–ecological structure of

fishery, 222–226

398 Index

social sources of resilience infisheries systems, 234–235

Odum, E.P., 3, 8, 103O’Flaherty, R., 135Oki, T., 199, 200, 203, 215Oldfield, F., 296Olsen, J.P., 11Olsson, P., 47, 77, 78, 94, 103, 114,

119, 120, 121, 122,250, 319

“One-size fits-all” policy, 213Ontogenetic habitat shifts, 222Open oceans, 334Operation Windfall, 139Oppenheimer, S., 297Optimal production of single

resources, 111Organization for Economic

Cooperation andDevelopment (OECD),304

Ortiz-Monasterio, J.I., 280Ostfeld, R.S., 46, 146Ostrom, E., 10, 13, 78, 80, 83,

90–93, 117–118, 183,234, 242

Otterman, J., 175Our Common Future, 56Overland flow, 32Oyster wars, 232

PPage, S.E., 117Palm, C.A., 261–262Paloheimo, J.E., 229Panarchy, 17, 21Parry, M., 277Participatory vulnerability

analysisassessments of, 73

Path dependence, 14Patz, J.A., 46, 60, 153Pauly, D., 221, 224, 225, 233Pelagic dominance state,

221, 223Pellow, D.N., 66Perera, A.J., 167Permanent crops, 263Permanent pastures, 263Peterson, G.D., 99Peterson, J.H., 138Petit, J.R., 296, 297Petschel-Held, G., 61Pfeffer, J., 86

Phillips, T.J., 121Pickett, S.T.A., 38, 281,

283–284, 286Pilifosova, O., 67Pimentel, D., 45Pinkerton, E.W., 78, 94, 142, 144,

145, 235, 241–242, 245–246,248, 251–252, 255, 319

Piscivores, 223Planktivores, 223Plant-available nutrients

(PAN), 175Platteau, J.-P., 243Plummer, R., 242Pluralism, 86Poff, N.L., 218Policy communities, 95Polycentric governance

contributing to urbanresilience, 292

structure, 118Polychlorinated-hydrocarbon

pesticides (PCBs), 44Pomeroy, R.S., 248–250Population rank, trajectory of

relative, 287, 288Porporato, A., 175Portney, P., 267Posey, D.A., 132Postel, S.L., 41, 210, 211, 214,

216, 263Post, J.R., 209Post, W.M., 10, 14Poverty trap, 109Power, A.G., 45, 264Power relations, 86Prager, M.H., 229Precautionary policies, 238Pringle, H.J.R., 177Pringle, R.M., 209Pritchard, L., 72Production forests, 159Protected areas, 130

categories of, 130–131Protected Landscape/

Seascape, 131Provisioning services, 31, 41–43Public consultation processes,

94–95Putnam, R.D., 19, 82Pyne, S.J., 296

QQuinn, J., 121

RRabalais, N.N., 35Radeloff, V.C., 167Rahel, F., 206Rahmstorf, S., 301Raivio, S., 164Ramakrishnan, P.S., 42, 46–47,

133, 152–153Ramankutty, N., 301Ranjan, R., 261Rao, V., 138Rappaport, R.A., 8, 56Raupach, M.R., 301, 308Ravallion, M., 263Raven, P.H., 43, 152Real Estate Investment Trusts

(REITs), 168Reardon, T., 265Redman, C.L., 3Redundancy, 68, 117Reflexive action, 22Reflexive behavior, 14Reforestation/afforestation, 150Regier, H.A., 112Regime shifts, 16, 24–25, 219

in ecosystems, 107–108Regulating services, 31, 43–46Renault, D., 215Rent seeking, 84Repetto, R., 56, 109, 119Rerkasem, B., 261Reserve, 212Resilience, 5, 9, 23–24, 56

approaches to enhance, 324depends on, 23, 290understanding, 71

Resilience-based ecosystemstewardship, 5, 20, 29, 52

on ecosystem management, 29enhancing adaptive capacity

and resilienceadapt governance to

changing conditions, 328foster biological, economic,

and cultural diversity,324–326

foster innovation and sociallearning, 327–328

foster mix of stabilizingfeedbacks and creativerenewal, 326–327

navigating transformationsbuilding resilience of new

regime, 331–332

Index 399

navigating transition, 331preparing for

transformation, 329–330opportunities and challenges

for future, 336–337reducing vulnerability

reduce exposure to hazardsand stresses, 321–322

reduce sensitivity to adverseimpacts of hazards andstresses, 322–323

steady-state resourcemanagement, ecosystemmanagement, and, 5

vulnerability, adaptive capacity,and resilience

contrasts among types ofsocial–ecological systems,332–335

general patterns, 335–336Resilience learning

components of, 71Resilience trajectory, 268Resource management, 96

for poverty reduction, 117Resource regimes, 82Resource stewardship

challenges, 27, 150policies, 7sustainable, 17

Response diversity, 36Restoration, 217Reynolds, J.F., 32, 173, 174,

179, 193Reynolds, J.E., 209Ribot, J.C., 154Richter, B., 211Ricketts, T.H., 46, 153Ridgwell, A.J., 296, 298Riemann, R., 285Rietkerk, M., 176, 177Rigidity traps, 109, 116

See also Poverty trapRiver management

strategic approach to, 192Robertson, G.P., 26–269, 264, 279Robinson, J.B., 98Robinson, J.G., 137Rockstrom, J., L., 201Rogers, K.H., 189–193Rose, G.A., 230Ross, M.L., 60Rothschild, B.J., 232Roving bandits, 93, 113

Runoff, 33Ruthenberg, H., 182, 183Ruttan, V.W., 267

SSacred groves, 47Safriel, U., 171, 172, 178–180, 182Sainsbury, K.J., 236Salinization, 208Salt, D., 16, 56, 114, 273, 290, 305Sampson, R.N., 151, 157, 159Sanchez, P.A., 262Sanderson, S., 13Sandford, S., 183–184Sanitary City, 283Sarkar, S., 181Sass, G.G., 206Scenarios, 114

accepted, 98, 114characteristics, 115key lessons, 116shortcomings of, 114

Schaaf, T., 284Scheffer, M., 36, 56, 107, 116, 198Schelling, T.C., 14Schellnhuber, H.-J., 205Schlager, E., 234, 242Schlesinger, W.H., 35Schmidhuber, J., 275Schnore, L.F., 283Schoennagel, T., 167Schreiber, D., 237, 254Schreiber, E.S.G., 237, 254Schultz, L., 70Schulze, E.-D., 164Science-management system, 191Scoones, I., 178Scott, W.R., 81Selby, M.J., 32Self-efficacy, 256Self-organization, 14Self-regulation, 14Seligman, M.E.P., 59Senge, P.M., 97, 256Sen, A.K., 59, 65Shadow networks, 94, 122,

161, 329Shah, M.K., 203Shah, T., 203Shane, G.D., 286, 292Shaver, G.R., 36Shertzer, K.W., 229Shifting baseline syndrome, 233Shifting cultivation, 261–262

Shindler, B.A., 18, 49Shotton, R., 96Shumway, D.L., 255Shvidenko, A., 149, 152Silent Spring, 56Sinclair, A.F., 230Singer, P., 211Single-loop learning, 97Singleton, S., 85, 248Six degrees of separation, 82Slash-and-burn or swidden

agriculture, 152Slow variables

critical, 12See also Fast variables

soils and sediments, 32Slums, 289Small-scale vs. large-scale

fisheries, 224Smil, V., 259, 264, 274Smit, B., 21, 56, 67, 73Smith, B.D., 259, 273Social and biophysical drivers

analysis of, 187Social capital, 18, 23,

61, 85Social and ecological aspects

of Australian pastoral grazingsystem, 188

Social–ecological barriers andbridges

challenges to navigating, 8–9Social–ecological challenges, 336Social–ecological governance,

77, 82challenges of adaptive, 78

Social–ecological interactionspatterns of, 61

Social–ecological monitoring, 98Social–ecological processes, 10Social–ecological resilience, 83

social systems impact, 108–109Social–ecological systems, 6, 9, 66

amplifying and stabilizingfeedbacks in, 10

directional changes, 26open systems, 10spatial interdependence of,

112–113temporal misfits in, 111–112transformations of

actions fostering, 124key factors for preparing, 123

vulnerable, 16

400 Index

Social-ecological thresholds, 8cascading consequences

of, 110Social–ecological vulnerability,

60–61Social learning, 71, 91Social legacies, 69Social memory, 24, 69–70Social movements, 95Social networks, 61, 82, 120

diversity of actor groups and,119–120

Social relations, 61Social slow variables, 72Societal engagement with

forests, 153Soil erosion, 33Sokoloff, K.L., 116Sokpon, N., 153Solow, R.M., 280Songorwa, A.N., 137, 138,

140, 146Special interest groups, 95Spirn, A.W., 284, 285Stabilizing feedbacks, 12Stafford Smith, D.M., 32, 171,

173, 177, 179, 184, 187,188, 319

Stakeholders, 25Stankey, G.H., 157Starfield, A.M., 98State of system to

ecological/socialvariable, 26

Staver, C.P., 45Steady-state

mosaic, 17resource management, 79

Steady state, 32Stearns, F., 283, 285Steffen, W.L., 3, 6, 22, 107,

295–300, 319Steiner, F.R., 284Steinfeld, H., 273, 274, 277Steneck, R.S., 112Stephens, S.L., 136Stern, N., 114, 301, 308Stevens, S., 129, 130Stewardship, 6

strategiesto enhance adaptive capacity

and resilience, 324to reduce vulnerability, 321

Stiassny, M.L., 209

Stiglitz, J.E., 303, 313Stoffle, B.W., 252Strategic adaptive management

communities of practice,192–193

ecological theory, 191monitoring and adaptive

decision-making, 191–192outcomes and challenges, 193thresholds of probable

concern, 191vision statement and objectives

hierarchy, 191Straussfogel, D., 13Subtractable resources, 87

See also Congestible resources;Excludable resources;Fugitive resources

Succession, 38Suding, K.N., 36Supporting services, 31–32

See also Cultural services;Provisioning services;Regulating services

Surette, T., 228Survey and manage protocols, 156Sustainability, 18

definition, 18factors governing properties

of, 19general approaches to foster

assumptions offrameworks, 20

enhanced adaptive capacity,22–23

enhanced transformability,24–25

increased resilience, 23–24principal approaches and

mechanisms, 22reduced vulnerability, 20–22

principal approaches andmechanisms, 22

social-ecological, 6Sustainability indicator

programs, 98Sustainable development, 25, 56Sustainable management, 5Sustaining livelihoods and human

well-being duringsocial–ecological change

assessing vulnerabilities towell-being, 62–63

cycles of dependency, 65–66

exposure to hazards andstresses, 63–65

incorporating novelty insocial–ecological systems,72–73

sensitivity to hazards andstresses, 65

social and environmentaljustice, 66

from human ecology tosustainable developmentto social-ecologicalresilience, 56–57

participatory vulnerabilityanalysis and resiliencebuilding, 73–74

sources and strategies ofresilience and adaptivecapacity, 66

fostering diversity, 67–69incorporating novelty in

social-ecological systems,72–73

innovation and sociallearning, 70–72

legacies and social memory,69–70

spectrum of adaptiveresponses, 66–67

well-being in changing worldmaterial basis, 57–59safety and security, 59–61self-esteem and

actualization, 62social relations, 61–62

Swain, D.P., 230Swanson, F. J., 48, 149, 161Swezey, S., 242Swidden (slash-and-burn)

agriculture, 39, 42Swinton, S.M., 264, 269, 279Synergies, 48, 72

examples of, 49Systemic processes, 299Systems models, 9Szaro, R.C., 156, 161, 165

TTainter, J., 310Tansley, A.G., 283Taylor, B.R., 132Taylor, D.E., 70, 71Taylor, L.H., 46Taylor, M., 85

Index 401

Technology transfer, 161Temporal change

in catch, 229dynamics of, 10–12

Temporal tradeoffs, 49Territorial Use Rights Fisheries

(TURFs), 251Thalenberg, E., 245Theory of fishing, 231Theory of weak ties, 82Thomas, D.S.G., 175, 245Thomas, S.J., 138Thompson, G.E., 284Thompson, J.R., 214Thom, R.M., 236Tietenberg, T., 243Tiffin, M., 183Tilman, D., 262, 264, 276Timberland Investment

ManagementOrganizations(TIMOs), 168

Timmer, C.P., 133, 264–265,269–270

Timmer, V., 133, 269, 270Tinley, K.L., 177Toepfer, K., 282Tongway, D.J., 177Top-down approach, 79Tradeoffs, 49

examples of, 49See also Temporal tradeoffs

Traditional knowledge,46, 70

See also Formal knowledge;Local knowledge

Tragedy of commons, 83Transaction costs, 85Transformability, 20, 24–25

and Regime Shifts, 24–25Transformations, 67, 104, 121

leadership in, 119phases, 121

Transformative learning, 104Transhumance, 186Transpiration, 33Trexler, J.C., 98Triple-loop learning, 104, 105Trophic cascades, 209Tropical deforestation, 168Troy, A., 286, 291True-novelty surprises, 106Turner, B.L., 3, 6, 29, 56, 297Turner, M.D., 182, 186

Turner, M.G., 12, 17Turner, N.J., 20, 22, 56, 61, 62, 63,

137, 198, 253Tversky, A., 306

UUlrich, R.S., 48Uncertainty, 105UN convention on the Law of the

Sea, 243UN Development Programme

(UNDP), 132UNDP Equator Initiative, 133UN Food and Agriculture

Organization’s Code ofConduct on ResponsibleFisheries, 243

United Nations Conference onthe Environment andDevelopment, 243

United Nations FrameworkConvention on ClimateChange (UNFCCC), 308

UN Millennium DevelopmentGoals, 311

Upadhyay, V.P., 261Upwelling, 36Urban ecology, 284Urban Environmental Accords

environmental categories, 284US Multiple-Use Sustained-Yield

Act of, 165, 1960

VValiela, I., 36van der Brugge, R., 121van Nes, E.H., 36Vannote, R.I., 35van Raak, R., 121van Wilgen, B.W., 216Vertical linkages, 93Vetter, S., 171Vincent, G., 284Virtual water, 215

trade, 215Viswanathan, K., 249Vitousek, P.M., 35–36, 42, 259,

301, 305Vorosmarty, C.J., 152, 199,

201–203, 205, 210–211,216–217, 219

Vostok ice core record, 296, 297Vulnerability, 20, 56, 62

components of, 63

linking human adaptivecapacity, resilience,transformability and, 21

reducing exposure to stress,21–22

reducing sensitivity tostress, 22

Vulnerability analysis, 20, 56evaluative steps, 62

Vulnerability index for Africancountries, 58

WWalbran, P.D., 233Walker, B.H., 15, 16, 22–25, 38,

56, 67, 104, 114, 117, 171,175, 189, 233, 273, 290,305, 319

Wallender, W.W., 215Walter, E., 242Walters, C.J., 78, 96, 99, 160, 218,

223, 225, 230, 233, 235,236–237

Wandel, J., 56, 67, 73Wapner, P., 303Warren-Rhodes, K., 283Washington Department of

Fisheries (WDF), 246Wassenaar, T., 277Wasserman, S., 120Water, 33, 41, 199Water cycle

and freshwater ecosystemshuman use affecting, 205

transpiration and run off, 33Water Laws, 272Water Policy in South Africa,

211–212Watershed, 197The Watershed Trust Fund of

Quito, Ecuador, 210–211Water wars, 211Watson, A.J., 296, 298Watts, D.J., 82Weathering, 32Weber, J.R., 163Webster, C., 288Weil, R.R., 32Weinstein, M., 144, 235,

246, 254Well-being, 55

definition, 18frameworks addressing

long-term human, 20

402 Index

relationship betweenecosystem services and, 60

risk to livelihoods and, 60Westley, F.R., 104, 119–121Westoby, M., 171Weyant, J., 267White, P.S., 38White, R., 187Wibe, S., 113Wilderness Area, 130–131Wildland–urban interface, 167Williamson, O.E., 85Wilson, D.C., 142, 144Wilson, E.O., 243Wilson, J.A., 8, 88, 97, 104, 242,

305, 1994, 1998, 2003,2006, 2008

Windows of opportunity, 119Wolf, A.T., 211, 218Wollenberg, E., 155Wondolleck, J.M.,

87, 94Wong, K.K., 175Wood, D.J., 95Word, C.S., 163The World Wildlife Fund, 139Worm, B., 42, 228, 231, 239Wright, M.M., 276

XXue, Y.K., 175

YYaffee, S.L., 79, 94, 284

Yaffee’s recurring nightmares,79, 284

Yodzis, P., 236Young, D.D.2002Young, M.D., 184Young, O.R., 10, 13, 20, 65, 77,

80, 82, 83, 87, 88, 89, 93,295, 306, 308, 319, 1982,1994, 2002, 2006, 2007,2008, 2009

ZZhu, Y., 45Zimov, S.A., 40Zwanenburg, K.C.T., 223

Principles of Ecosystem Stewardship - ANU Open Research

Documents

Transcript of Principles of Ecosystem Stewardship - ANU Open Research