perspectives on ecosystem-based approaches to the … · warner 2003, lubchenko et al. 2003, pauly...

MARINE ECOLOGY PROGRESS SERIESMar Ecol Prog Ser

Vol. 274: 269–303, 2004 Published June 24

Introduction

Howard I. Browman1,**, Konstantinos I. Stergiou2

1Institute of Marine Research - Austevoll, 5392 Storebø, NorwayEmail: [email protected]

2Aristotle University of Thessaloniki, School of Biology, Department of Zoology, Box 134, Thessaloniki 54124, Greece

Email: [email protected]

The urgent need to reduce the intense pressure anddestructive power that modern fishing practices applyto the world’s fisheries, and the oceans that supportthem, is now widely recognized (e.g. FAO 2002a,Hilborn et al. 2003). However, there is far less agree-ment over the exact levels to which fishing mortalitymust be reduced and over how to reduce the indirecteffects of fishing (e.g. bycatch, destruction of theseafloor), in order to ensure sustainability of catchesand the health of marine ecosystems. And this is to saynothing of disagreements over how these goals mightbe achieved. It has proven all too easy for variousfactions—including some fishery scientists—to blameour having arrived at the current crossroads on theineffectiveness of existing management practices, andon the scientific advice that underlies it. Driven bythese forces, and in recognition of the significant directand collateral impacts that fishing imposes on marineecosystems, an Ecosystem Approach to Fisheries (EAF)is rapidly being adopted by institutions charged withstewardship of the marine environment (e.g. NOAA1999, Brodziak & Link 2002, FAO 2003, Garcia et al.2003, Sinclair & Valdimarsson 2003). In conjunctionwith this EAF is the implementation of Marine Pro-tected Areas (MPAs), including marine reserves. BothEAF and MPAs implicitly recognize that the value (tohumanity) of the whole ecosystem is much greater

than the sum of its parts—a commendable step for-ward in-and-of itself. However, there is some disagree-ment over whether the EAF, and MPAs, truly representalternatives that will be any more effective in assistingus with sustainable management of marine resourcesthan historical practices. Regardless of the approachthat is taken to decide upon catch limits, or on the loca-tion, size and number of MPAs, there will always bethe complicated (and socio-economically-politicallycharged) question of how these policies should beimplemented and enforced; that is, governance (see,for example, Mace 2001, Sissenwine & Mace 2003,Caddy 2004, Cochrane 2004, Stefansson 2004). Toaddress these issues, we solicited essay-style contribu-tions from several of the marine and fishery scientistswho are at the forefront of the ongoing debate. Thoseessays are presented here.

We will not use space summarizing the content ofthis Theme Section (TS)—we encourage you to readthrough it. Rather, we take this opportunity to high-light some of the most important conclusions that issuefrom the essays when they are taken as a whole and toadd some commentary of our own. The acronyms usedin this TS are listed in Table 1.

In the critical recommendation of such fishery man-agement tools as limits on maximum fishing mortality,minimum spawning stock biomass, or total allowablecatch levels, fishery scientists often disagree aboutseemingly subtle (to the layman) aspects of data analy-sis and interpretation. Although debates such as theseare at the core of the scientific process, the fact thatfishery scientists themselves do not always agree hasbeen the focus of socio-political criticism, and is surelyone of the reasons that advice on catch quotas is notoften strictly heeded. In the case of the contributions tothis TS, written by proponents sitting on both sides ofthe fence, there is a convincing consensus on most ofthe key issues. While there is disagreement over just

© Inter-Research 2004 · www.int-res.comResale or republication not permitted without written consent of the publisher

THEME SECTION

Perspectives on ecosystem-based approaches to themanagement of marine resources

Idea and coordination: Howard I. Browman, Konstantinos I. Stergiou

Contributors*: Howard I. Browman, Philippe M. Cury, Ray Hilborn, Simon Jennings, Heike K. Lotze, Pamela M. Mace, Steven Murawski, Daniel Pauly, Michael Sissenwine, Konstantinos I. Stergiou, Dirk Zeller

**Contributions are presented in alphabetical order (by firstauthor)

**The views expressed here are those of the author onlyand do not necessarily reflect the official position of TheInstitute of Marine Research

Mar Ecol Prog Ser 274: 269–303, 2004

how severely depleted some fish stocks are, and onwhether and how quickly they will recover, all agreethat many stocks are overexploited. While there issome disagreement over just how much fishing mustbe reduced, all agree that current levels of overcapa-city in the world’s fishing fleets are not sustainable.While there is disagreement over equating MPAs andEAF, all agree that MPAs will complement other man-agement tools, within an EAF or not. Thus, for eachand every major issue, while there might be disagree-ment on the details, there is unanimity over the press-ing need for action to protect marine ecosystems. Andthat must be made the focus of public attention.

Iles (1980) refers to ‘…a ‘Bio-Energetic Multi-SpeciesEcosystem Dynamics (BEMUSED)… ’ basis for settingcatch quotas. This illustrates how the idea of taking anEAF is really nothing new, and it highlights that,unless we are truly more clever (and richer with data)than we were almost 25 yr ago, following EAF couldleave us just as bemused, and/or muddled (see Hedg-peth 1977). Iles (1980) also stated that ‘…social, politi-cal, and economic factors are at least as important infisheries management as the scientific knowledge ofthe resource.’ This conclusion, arrived at 24 yr ago, isreiterated by several contributors to this TS—gover-nance, and not science, remains the weakest link in themanagement chain (also see Hutchings et al. 1997,Harris 1998, Policansky 1998, FAO 2003, Cochrane2004). Thus, even if we were able to provide managerswith perfect scientific prediction, that alone will nothelp. Following from all of this, if there is any hope ofsucceeding with an EAF, or any real chance of control-ling fishing, the organizations and institutions involvedin the governance of marine resources will have to betotally revamped. The new structure will have to in-clude stakeholders, social and political scientists, econ-omists, lawyers, political lobbyists, educators, journal-ists, civil engineers, ecologists, fishery scientists andoceanographers, all operating in a conciliatory andintegrative environment.

We hope that the following analogy will illustratethat it is untenable to ignore the counsel of fisheryscientists, even when they disagree and/or provideadvice that is based upon highly uncertain assess-ments (also see Stefansson 2004). If meteorologists saythat a major storm is coming, people are relocated tosafer places, and houses and buildings are boardedup. Even if the predictions about when and wherethe storm will hit—provided by extensive networksof expensive ground-based monitoring devices andweather satellites—are not very accurate (because thestorm’s behaviour is unpredictable), precautions arestill taken, often over a very wide geographic area…just in case. This illustrates that society does not expectmeteorologists to predict the weather with any degree

of accuracy, yet we have somehow all learned to livewith that, and take appropriate precautions nonethe-less. In the face of this analogy, we must ask: why doessociety have higher expectations of fishery scientistswith respect to their ability to accurately predict thenumbers of fish that will be in the sea several years intothe future? Further, why is it so difficult for fishery sci-entists to convince society, authorities, and stakehold-ers to take a precautionary approach towards themanagement and conservation of fish stocks (or wholeecosystems) (see Lotze’s contribution to this TS)?Finally, if people are routinely relocated to a safe placewhen a potentially destructive storm is coming, why isit so difficult to recognize the inherent rights thatmarine fauna have to a safe haven (in the form, forexample, of MPAs)? The international treaty repre-sented by the Montreal Protocol on Substances thatDeplete the Ozone Layer is another example of howsociety can respond when the stakes are high and theneed is urgent: society can adopt and implement pre-cautionary approaches to the management of theworld’s resources, even when there are complex mix-tures of stake holders. Hopefully, we will be able toachieve the same for the world’s marine ecosystems.

270

Table 1. Acronyms and their full forms used in the TS

Abbreviation/ Full nameacronym

BEMUSED Bio-Energetic Multi Species Ecosystem Dynamics

CML Census of Marine LifeEAF Ecosystem Approach to FisheriesEEZ Exclusive Economic ZonesFAO Food and Agriculture Organization GIS Geographic Information SystemGLOBEC Global Ocean Ecosystem Dynamic ProgramsGOOS Global Ocean Observing SystemICES International Council for the Exploration of

the SeaICNAF International Convention for the Northwest

Atlantic FisheriesITQ Individual Transferable QuotasIUCN International Union for the Conservation of

NatureLME Large Marine EcosystemMPA Marine Protected AreasMSY Maximum Sustainable YieldMVH Member-Vagrant HypothesisNOAA National Oceanic and Atmospheric

AdministrationOECD Organisation for Economic Co-operation and

DevelopmentPISCO Partnership for Interdisciplinary Studies of

Coastal OceansUNDP United Nations Development PlanUNEP United Nations Environmental ProgrammeTAC Total Allowable Catch

Theme Section: Ecosystem-based approaches to management of marine resources

Marine Protected Areas as a centralelement of ecosystem-based management:defining their location, size and number

Howard I. Browman1,*, Konstantinos I. Stergiou2

1Institute of Marine Research - Austevoll, 5392 Storebø, NorwayEmail: [email protected]

2Aristotle University of Thessaloniki, School of Biology, Department of Zoology, Box 134, Thessaloniki 54124, Greece


Marine Protected Areas (MPAs) include many sub-classes (e.g. marine sanctuaries, marine parks, wildliferefuges, fisheries closures, no-take MPAs, multiple-useMPAs, marine reserves, ecological reserves) all ofwhich can be defined based mainly upon the levelof protection and the primary conservation goal (seewww.mpa.gov; Lubchenco et al. 2003). MPAs, andespecially the marine reserves subclass (i.e. ‘areas of theocean completely protected from all extractive anddestructive activities’; Lubchenco et al. 2003) representthe extreme case of the precautionary approach tomanaging marine resources (e.g. Lauck et al. 1998).

The strong and rapidly growing interest in MPAs(and particularly in marine reserves) is reflected in thedramatic increase in the number of publicationsdevoted to them (reviewed in Jones 2002, Gell &Roberts 2003, and the articles in ‘The Science ofMarine Reserves ’, a supplemental issue of EcologicalApplications, Vol 13, Iss 1, freely available for down-load at www.esa-journals.org/esaonline/?request=get-static&name=s1051-0761-013-01-0001). In addition,there are now a number of sites on the World WideWeb that are either totally devoted to MPAs, or includerelevant information on them: UNEP’s World Con-servation Monitoring Centre (www.unep-wcmc.org/protected_areas), the Partnership for Interdiscipli-nary Studies of Coastal Oceans (PISCO, www.piscoweb.org), and several others. This intense interestis at least partly related to MPAs having been identi-fied and advocated as a conservation (of habitat andbiodiversity) and managerial (of fisheries) tool of cen-tral importance in the Ecosystem Approach to Fisheries(EAF) (e.g. Agardy 2000, Stergiou 2002, Halpern &Warner 2003, Lubchenko et al. 2003, Pauly & MacLean2003, Hilborn et al. 2004). It is hoped that MPAs will bebeneficial in (1) rebuilding overexploited fish stocks,(2) preserving habitat and biodiversity, (3) maintainingecosystem structure, (4) buffering against the effects ofenvironmental variability, (5) serving as a control groupagainst which populations in exploited regions can be

compared, among others. Clearly, the choice of loca-tion, spatial extent (horizontal and vertical), and num-ber of MPAs is critical if they are to meet these goals.It is to this issue that we devote our attention here.

Halpern & Warner (2003) state, ‘Most reserve locationsand boundaries were drawn by a political process thatfocused on economics, logistics, or public acceptance,while largely overlooking or ignoring how the complexecology and biology of an area might be affected by re-serve protection.’ In this sense, establishing the locationsand boundaries of MPAs can be seen as analogous to theimperfect process associated with establishing stockmanagement grids—a process that has never reallymanaged to incorporate the key realities of populationdynamics of the exploited species. While there is agrowing consensus on the need for MPAs, at this point intime there is no clear and well-founded basis uponwhich their location, spatial extent and number can bedecided. In fact, rationales/frameworks that are basedupon principles of theoretical and applied ecology haveonly recently been tapped to address these key ques-tions (e.g. Roff & Evans 2002, Botsford et al. 2003,Roberts et al. 2003a,b, Shanks et al. 2003, Fisher & Frank2004). Much of this work focuses on the manner in whichdifferent aspects of the life histories of marine organ-isms—spawning locations, dispersal, larval retentionand export, juvenile nursery areas, etc.—affect MPAdesign. In this context, we contend that an eco-evolutionary framework already exists, grounded inmarine ecology and fisheries oceanography, that iscompletely consistent with EAF and MPA objectives.

The Member-Vagrant Hypothesis as a framework fordefining the location, size and number of MPAs. TheMember-Vagrant Hypothesis (MVH), the development ofwhich can be traced through a series of publications byMike Sinclair and Derek Iles (Iles & Sinclair 1982, Sinclair1988, 1992, Sinclair & Iles 1988, 1989), defines 4 attributesof populations that are involved in the regulation of theirsize. The ’population richness’ refers to the number ofdiscrete self-sustaining populations (henceforth simply’populations’) exhibited by any given species. Speciessuch as herring, cod, mackerel, the salmonids, and manyothers are population rich. The ‘spatial pattern’ relates tothe geographic distribution of these populations. Popula-tion rich species are usually also broadly distributed (thenorth Atlantic region is so far the best studied in thisregard). Population richness and spatial pattern arespecies-level characters. The ’absolute abundance’ refersto the instantaneous size of the various populations of anygiven species, and this size—which can range overseveral orders of magnitude—varies over time (thus, its’temporal variability’). These last 2 components of theMVH are population-level characteristics. Sinclair & Ileshave applied the MVH to describe the richness, pattern,abundance and variability of several economically im-

271

*The views expressed here are those of the author onlyand do not necessarily reflect the official position of TheInstitute of Marine Research

Mar Ecol Prog Ser 274: 269–303, 2004

portant fish including herring, cod, haddock, mackerel,and several others. For all of these, (1) the populationrichness is directly correlated with the number of reten-tion areas for the species’ early life history stages (alsoimplying that the adults are able to return to the samegeographic locations); (2) the spatial pattern is related tothe number of discrete geographic areas allowing closureof the species’ life cycle; (3) the absolute abundance isscaled according to the size of the geographic area inwhich there is closure of the life cycle (corroborated byMacKenzie et al. [2003], who reported that the biomassof cod spawners and recruits is related to habitat size);(4) the geographic locations referred to in (1), (2) and(3) have distinct oceanographic features; and (5) the tem-poral variability is determined by the intergenerationallosses of individuals from any one population (throughmortality and/or passive processes such as advection orspatial constraints = ’vagrancy’). It is worth noting that theMVH is completely consistent with the metapopulationconcepts that have recently been applied to marine fishpopulations (e.g. Smedbol & Wroblewski 2002)

Exploited populations are subject to intense size-dependent mortality and drastic reductions in biomassover a short time and a large spatial scale (e.g. Chris-tensen et al. 2003, Myers & Worm 2003, Pauly & MacLean2003). With modern fishing practices and equipment, thiscan impact a large proportion of the populations in aspecies’ entire spatial pattern. Thus, commercial fishingimposes new conditions on these populations and, there-fore, drastically affects all 4 MVH population attributes.

The MVH ‘…emphasizes that membership in a popu-lation in the oceans requires being in the appropriateplace during the various parts of the life cycle. It impliesthat animals can be lost from their population, and thusbecome vagrants. Life cycles are considered as continu-ity solutions within particular geographical settingswhich impose spatial constraints.’ (Sinclair & Iles 1989,p. 169). Thus, for many marine fishes, population rich-ness, pattern, absolute abundance and temporal vari-ability are all a function of geography.

Following from the MVH, the location of MPAs shouldbe chosen to include a subset of the populations withina species’ (or species complex) spatial pattern. The sizeof each such MPA would then be assigned based uponthe geographic area within which the correspondingpopulation’s life history can achieve closure. In ourview, applying the MVH in this manner would satisfymany of the objectives of MPAs.

It has only recently been possible to assess whetherMPAs do in fact provide the benefits listed above (re-viewed in e.g. S. J. Hall 1998, Jones 2002, Gell & Roberts2003, Halpern & Warner 2003, Luchenco et al. 2003,Hilborn et al. 2004). These assessments have led to argu-ments over the degree to which MPAs can or will succeed.There is also some concern over the possibility of an im-

balanced reliance upon MPAs as a fisheries managementtool (see Hilborn et al. 2004 and several of the contribu-tions to this TS). Nonetheless, if the choice of their loca-tion, size and number is well grounded in marine ecologyand fisheries oceanography, then MPAs stand to becomean effective tool for conservation and management. In or-der for this to be realized, 2 closely related steps are re-quired. First, an operational spatial unit within whichMPAs will be embedded must be defined. Such a unit al-ready exists: the Large Marine Ecosystem (LME) (e.g.Sherman & Duda 1998). LMEs are large ‘regions of oceanspace encompassing coastal areas from river basins andestuaries to the seaward boundaries of continental shelvesand the outer margins of the major current systems’ char-acterized by ‘distinct: (1) bathymetry, (2) hydrography,(3) productivity, and (4) trophically dependent popula-tions’ (www.lme.noaa.gov). When combined with Long-hurst’s (1998) ‘Biogeochemical Provinces’, which extendout into the open ocean areas, LMEs can provide a veryuseful ecosystem framework for fisheries research (seePauly & MacLean 2003, www.seaaroundus.org). Second,future work in fisheries science could adopt a more eco-logical/oceanographic orientation, by (1) identifying andmapping the key faunistic components and the biodiver-sity ‘hot spots’ (sensu Worm et al. 2003) in the mainecosystems of the world’s oceans (as defined above);(2) describing the life cycles of these key componentswithin the context of the MVH framework; (3) spatiallymapping the life cycles of key species (see Zeller & Pauly2001); and (4) identifying the special oceanographic fea-tures associated with the retention and nursery areas ofthese key components (recent work linking populationgenetics with marine ecology and fisheries oceanographyholds promise in this regard, e.g. Reiss et al. 2000).

Acknowledgements. For their influences on our developmentas marine scientists, we dedicate this essay to Maxwell J.Dunbar (deceased), T. Derek Iles, William C. Leggett, BrianM. Marcotte and Michael Sinclair. We thank K. Erzini, K. T.Frank, J. J Govoni, and D. Pauly for their comments on themanuscript. H.I.B.’s ongoing research, and his editorialactivity for MEPS, is supported by the Institute of MarineResearch, Norway, and by The Research Council of Norway.

Tuning the ecoscope for the EcosystemApproach to Fisheries

Philippe M. Cury

Institut de Recherche pour le Développement (IRD), CRHMTBP 171, 34203 Sète Cedex, France


A multidisciplinary scientific approach is neededfor the Ecosystem Approach to Fisheries (EAF). TheReykjavik Declaration of 2001, reinforced at the World

272


Summit of Sustainable Development in Johannesburgin 2002, requires nations to base policy related tomarine resource exploitation on an ecosystem approach.To fulfil this new requirement, a strategy based uponinnovative science that will address the complexity ofmarine ecosystems, coupled with operational frame-works for an effective EAF is needed. EAF must bebuilt on a scientific rationale that will link ecologicalprocesses to ecosystem-level patterns. In doing so, itwill help managers to recognize and understand eco-logical limits to avoid the loss of ecosystem integrityand to maintain fisheries in viable states (Fowler &Hobbs 2002, Mullon et al. 2004).

This is a challenging task, as marine ecosystems aredifficult to define, having no apparent boundaries,and lacking the clear objective or purpose that canbe ascribed to more tractable biological or ecologicalentities (e.g. individuals or populations). An ecosys-tem contains water, nutrients, detritus, and numerouskinds and sizes of organisms ranging from bacteria,phytoplankton, zooplankton, and fish to mammalsand birds, all with their own life history traits. Theseliving and non-living ecosystem components areinterconnected through continuously changing foodwebs, which make ecological systems extraordinarilycomplex.

Today, the explicit study of complexity is both neces-sary and timely in ecology (Loehle 2004). Emergencehas replaced the earlier mostly theoretical approach toimplementing classical population dynamics in ecol-ogy (Woods 2004). The concept of simple cause andeffect is neither adequate nor sufficient when dealingwith complex systems, particularly if one accepts theprinciple that prediction is a pre-requisite for appliedecological research (Peters 1991). Research in ecologyhas been based mostly on studying processes in detail,resulting in an impressive number of potential cause-effect relationships to explain emergent patterns.Emerging patterns suggest likely tendencies and pos-sible response trajectories. A combination of the pro-cess and emergence approaches has long been advo-cated (Elton 1927), but with relatively little success,despite its promise of ameliorating our understandingof marine ecosystems.

Many tools, information systems and models havebeen developed, particularly during the last decade,such as coastal hydrodynamic models, individual-basedmodels that couple physics and ecology, Geographic In-formation System (GIS) and ecosystem models. Thesevarious techniques, in many cases highly sophisticated,offer a unique opportunity in ecology to address thecomplexity of marine ecosystems in a diverse and con-trasted manner. Despite the variety of techniques thatcan help track spatial and dynamical changes in eco-systems, it is often unclear, however, how these can be

applied to solve specific scientific problems or to respondto questions of importance to society.

Using the telescope and microscope as analogies, theterm ‘ecoscope’ was proposed by Ulanowicz (1993) tocharacterize ecosystem modelling that may be used as atool for resolving patterns, indicative of the key ecosys-tem responses (that may otherwise be obscured withinthe complexity of marine ecosystems). Today there existsno general, unified theory of the functioning of marineecosystems, nor a single tool on which a reliable ‘eco-scope’ can be based. Moreover, in the context of globalchanges (i.e. climate change and overexploitation), theexercise is even more difficult as we are facing changesand fluctuations on a global scale that have not been ex-perienced before (Holling 1995). To respond to thesechallenges, the ecoscope must be operationalized into anintegrative framework for studying marine ecosystemsand responding to the needs of the EAF. I discuss belowhow we can start implementing this approach.

Linking patterns to processes. Strong ecologicalpatterns have been described in marine ecosystems(Parson 2003). The mechanisms explaining alternationbetween different pelagic fish populations, synchronybetween remote fish populations, and regime shiftsstill remains largely speculative in the marine environ-ment contrary to studies in lake ecosystems (Carpenter2003). I will use the example of regime shifts that rep-resent a crucial ecological pattern for the EAF, as theyare sudden changes in structure and functioning ofmarine ecosystems that affect several components,exploited or not. For example, shifts from demersal fishdominated to pelagic fish dominated ecosystems (orshort-lived species such as shrimps, crabs or octopus)have been documented in the Atlantic and the Baltic(Worm & Myers 2003); shifts from fish-dominated tojellyfish-dominated ecosystems have been observed inthe Bering Sea, the Black Sea, the Gulf of Mexico,the western Mediterranean Sea, Tokyo Bay and offNamibia (Parsons & Lalli 2002). These regime shiftshave deeply modified marine ecosystems and the fish-eries they sustain. EAF requires understanding thenature of such ecosystem changes, i.e. the processesthat are involved, the speed at which they act, theirpotential reversibility and periodicity...

Linking processes to patterns. Regime shifts havebeen related mainly to climatic changes, but anthro-pogenic influences also play a major role in inducingecosystem changes. A regime shift may be environmen-tally driven (e.g. through bottom-up control of the foodweb, or via direct effects on recruitment), ecologicallydriven (e.g. through competition, predation), mediatedbehaviourally (e.g. behavioural adaptations to habitatchange) or driven by human exploitation of selectedspecies or preferential fish size classes (Cury & Shannon2004).

273

Mar Ecol Prog Ser 274: 269–303, 2004

Environmental processes act at different scales andprobably simultaneously affect most species withinthe ecosystem. Under bottom-up control, a majorenvironmental change can alter the ecosystem’s pri-mary productivity and, thereby, the flow of energy tohigher trophic levels. Climatic variability can itselftrigger a series of concomitant physical and biologicalprocesses in the form of system wide ‘regime shifts’(Hare & Mantua 2000). Mesoscale events can triggerhuge variability in pelagic fish recruitment success(Roy et al. 2001). In upwelling systems, a small num-ber of pelagic fish species occupy the intermediatetrophic level, feeding mostly on phytoplankton and/orzooplankton. These species can attain huge bio-masses, which can vary radically depending upon thestrength of the environmental factors driving recruit-ment. The role of dominant pelagic fish has beenemphasized as they might exert major control onenergy flow, both up and down the food web; this hasbeen termed ‘wasp-waist control’ (Cury et al. 2000).Predation is a fundamental process that is sometimesas important as resource limitation in controllingecosystem dynamics. As most fish species interactthrough predation, the existence of top-down control,through which the lower levels of the food web areregulated by 1 or several upper-level predators,appears to initiate trophic cascades in several marineecosystems (Cury et al. 2003). Fisheries tend toremove top-down forces by preferentially exploitinglarge top predators in marine ecosystems, a mecha-nism known as ‘fishing down the food web’ (Pauly etal. 2000). This mechanism can result in an increase inthe abundance of small forage fish (or short-livingspecies) and to a stronger effect of climate ondepleted marine resources (Beaugrand et al. 2003,Cury & Shannon 2004). All of the processes that areassociated with environmental or anthropogenic forcesshould be related in a more organized manner to theobserved patterns of change in marine ecosystems. Inorder, for example, to arrive at a useful level of gener-alization, the respective roles of top-down, bottom-upor wasp-waist forces need further exploration.

The ‘ecoscope’ as a multidisciplinary dynamicaltool to move towards an EAF. Theories, models, andobservations of the patterns that are important forecosystem dynamics need to be linked (Scheffer &Carpenter 2003). Ecologists have been analyzing eco-logical interactions in 2 different, and often mutuallyexclusive, ways using reductionist (process-oriented)or holistic (pattern-oriented) approaches. However, asstated by Elton (1927), a combination of the 2 methodswould be better. Seventy-five years later, this remainsthe approach that should be applied in future researchon ecosystem dynamics. The ecoscope could be onesuch set of tools.

We need to encourage research in this direction andassemble processes and patterns in the same frame-work to explore the impact of global changes in timeand space. The ‘ecoscope’ can be tuned to disentanglerealities and speculations by assembling our presentbiological, ecological, modelling, and operational tools(GIS; indicators). The ‘ecoscope’ would not rely on asingle model, but would incorporate a suite of modelsthat can use different assumptions for depicting in arobust manner the relevant processes.

With the rapid development of models, methods andhypotheses, there already exists a large variety ofcomplementary approaches and tools. The ‘ecoscope’encompasses all of our expertise and knowledge onmarine ecosystems; however, it needs to be builtaround key scientific questions and information sys-tems. Global changes that affect marine ecosystems,such as overexploitation and climate change, are rele-vant scientific problems and effectively addressingthese is crucial for sustainable development. Spatialand temporal dynamics that link the different organi-sational levels need to be tackled in any EAF. Dynam-ical information systems should represent the converg-ing point around which specific questions can beraised and discussed within the different disciplines.It is a stimulating task for the future, as it requiresmacroecological studies of the oceans to characterizepatterns of ecosystem components, based on largeamounts of data (Parsons 2003). A suite of field, exper-imental and modelling approaches is required to iden-tify, with a high degree of confidence, the underlyingprocesses and emergent patterns. Gathering of fish-eries and ecosystem data has, to date, mostly beenundertaken separately and by different sub-groupsof marine scientists, with little exchange. Long-termdata series are needed to develop data banks for eco-logical and climatologically quality control. We alsonecessitate developing new observation systems byrecognizing that ecological and biological data that arecollected for single-species fisheries management arenecessary but insufficient for understanding ecosystemdynamics. Ecosystem-based indicators can simplify,quantify and inform about the complexity of marineecosystems. The elaboration and evaluation of ecosys-tem-based indicators—such as the Fishing-in-Balanceindex (Pauly et al. 2000) or those related to size spectra(Shin & Cury 2004)—pertain to a multidisciplinaryfield of research on the marine ecosystem and mayconstitute a central focus for fisheries management.This represents a new framework that would challengethe difficulties of understanding the dynamics of com-plex systems at appropriate scales by enabling repeat-able patterns to be tracked by indicators, and by incor-porating existing scientific knowledge on processesinto models and ultimately into fisheries management.

274


The ecoscope for EAF should rely on 3 complementarycomponents: (1) a clear identification of the long-termobjectives (what we want and do not want to happen inmarine ecosystems and for the exploitation of marineresources); (2) a multidisciplinary scientific expertise(data, theory, experiments, models) to address the im-pact of global changes on marine ecosystems, and that isarticulated around dynamical information systems, suchas maps and indicators, to stimulate interactions be-tween disciplines; and (3) an evaluation of the perfor-mance of the ecoscope to solve scientific questions andto address management objectives for the EAF.

Building ecoscopes is a demanding way of integrat-ing knowledge and the necessary ‘ingredients’ andtools to begin the process are already available. How-ever, our marine and fisheries institutions are not cur-rently organized to undertake this integration and willhave to address ecosystem issues by developing amultidisciplinary scientific approach. This integration,which could be achieved in an incremental way, willsubstantially improve the perception of ecologicalresearch and its usefulness to society. However, it is atask that will compete with other scientific priorities atnational levels, as it will require mobilizing efforts. Oursociety seems to be more interested in, and fascinatedwith, developing ‘telescopes’ rather than building‘ecoscopes’. Marine ecosystems sustain our terrestriallife and deserve priority. We need telescopes andmicroscopes, but we also need ecoscopes. Implement-ing and operationalizing ecoscopes will crystallize ourpresent scientific knowledge. It requires agreementupon clear and perceivable objectives and adjustmentof multiform scientific expertise to societal issues. Thepotential task is overwhelming, and we need to takepragmatic steps before fully implementing an EAF.Tuning the ecoscope should help us to move towards‘ecosystem ecology’ as a discipline in its own right, andtowards an effective EAF.

Acknowledgements. Thanks to Dr. Lynne Shannon, who dis-cussed and elaborated with me the ideas that are contained inthis essay, and Vera Agostini, Yunne Shin, Andy Bakun,Audrey Colomb, Jean Lefur and Ian Perry for their comments.

Ecosystem-based fisheries management: the carrot or the stick?

Ray Hilborn

School of Aquatic and Fishery Sciences, Box 355020, University of Washington, Seattle, Washington 98195, USA


In the last few years, a series of papers have beenpublished in high-profile scientific journals describingthe role of fishing in the collapse of marine ecosystems

(Jackson et al. 2001, Myers & Worm 2003), the destruc-tion of marine habitat (Watling & Norse 1998) andchanges in ecosystems that are possible precursors tofuture collapse (Pauly et al. 1998). The central themeof this ‘Litany’ is that conventional single speciesfisheries management has failed and new approachesare needed. A major element of the proposed newapproaches is a move from conventional single-speciesmanagement to ‘ecosystem-based management’ (NRC1998). The specific proposed solutions that emergefrom the Litany include (1) elimination of subsidies forfishing fleets, (2) reduction of target fishing mortalities,(3) protecting a significant portion (20 to 30%) of theworld’s marine areas from fishing in the form of MarineProtected Areas (MPAs) (Pauly et al. 2002), and (4)elimination of destructive fishing practices (bottomtrawling). These approaches require a powerful cen-tralized government and are, therefore, unlikely tobe implemented in most of the developing world.

While papers subscribing to the Litany seem to havenear exclusive access to the pages of the most presti-gious journals, their conclusions are strongly contestedwithin the scientific community. For example, the con-tention that the predatory fishes of the ocean havedeclined by 90% (Myers & Worm 2003) and, by impli-cation, that these fisheries have collapsed, has beenchallenged on both the technical nature of the analysisof fishermen’s catch records (Walters 2003) anddetailed analysis of the fisheries (www.soest.hawaii.edu/PFRP/large_pelagic_predators.html). More sim-ply, the catch data from these fisheries show that theyare providing increasing yields, quite contrary to whatone would expect from fisheries that Myers & Worm(2003) classify as having collapsed 20 to 30 years ago.

The contention that MPAs would significantly bene-fit fisheries yields is equally contested (Norse et al.2003, Hilborn et al. 2004). Nevertheless, the Litany hasdominated public perception of fisheries problems andother authors citing the Litany frequently say that70% of the world’s fish resources are overexploited orcollapsed, rather than fully exploited, overexploited orcollapsed. For example, ‘According to various officialreports, three-quarters of the world’s fish stocks havebeen depleted. Official statistics may well err on theconservative side: overall catches are declining, yetillegal fishing is increasing. The net result is a crisis fornatural fisheries.’ (O’Riordan 2003). In fact, most of theworld’s fisheries are not overexploited and continue tobe quite productive (FAO 2002a). Within the U.S., onlyabout 16% of potential yield is being lost due to over-fishing (Hilborn et al. 2003).

The scientific objections to the Litany are primarily amatter of degree. No one questions that the majority ofthe world’s fisheries are heavily used, many are over-fished, some have collapsed, and good biological and

275

Mar Ecol Prog Ser 274: 269–303, 2004

economic management suggests substantial reduc-tions in fishing pressure are needed for sustainablemanagement (Hilborn et al. 2003). The major disagree-ments over possible solutions are not so much wherewe would like to be, but how to get there. The form ofecosystem management that emerges from the Litanyis one that concentrates on the ecosystem in which thefish are embedded and relies on strong central govern-ment control. I, and others (Garcia et al. 2003, Sissen-wine & Mace 2003), believe that we need a form ofecosystem management that emphasizes the interac-tion between fish, fishermen and government regula-tors and concentrates on incentives and participationwith user groups. This difference can be considered asa choice between a participatory approach with incen-tives as a ‘carrot’, and a centralized government usingregulations as a ‘stick’.

The key elements of the current fisheries managementapproach used in most regulated fisheries in developedcountries and international agencies include (1) singlespecies stock assessment to calculate the MaximumSustainable Yield (MSY) for each stock, (2) a politicalprocess to set regulations that determine allowabletime, area, gear and catch limits that intertwines alloca-tion between users and conservation, (3) regulationon large spatial scales, (4) a centralized managementstructure for science, decision making and enforcementwith costs paid by governments, and (5) involvementof stakeholders primarily through the political or legalprocess. It should be noted that most stocks world-wideare not managed in any meaningful way, and any pro-posals for management, ecosystem or otherwise, needto be achievable. To argue that we need more data-intensive management and more regulation by centralgovernments in the fisheries of the world that have littledata and little regulation is untenable.

There have been a wide range of papers dealingwith ecosystem management and each of these hasa distinct flavor. The ‘ecosystem management’ I de-scribe here shares elements with the views of others,all of whom emphasize various forms of marine tenureand the dynamics of fishing fleets and regulators. Theprimary difference between the incentives approachand the forms of ecosystem management emergingfrom the Litany is governance. The solutions proposedby the Litany rely on strong top-down control todetermine objectives and management actions and toassure compliance by fishing industries. The incen-tives approach recognizes that fisheries are dynamicsystems comprised of people and fish (Harris 1998),that top-down control is highly limited in most fish-eries, and that good outcomes result from creatingincentives that make the interest of the participants inthe fishery consistent with the interest of society as awhole. What has failed in conventional fisheries man-

agement is not single-species management, but thetop-down control as conventionally practiced. In mostof the world’s fisheries, the commercial and recre-ational fishermen have significant political power and,hence, attempts to impose regulations that are con-trary to their economic interests will most likely fail.Ecosystem management that relies on top-down con-trol for implementation, and makes no allowances forthe social/political dynamics of the regulatory struc-ture, is no more likely to succeed than conventionalsingle species management.

What is missing from the conventional single speciesfisheries management approach is (1) a form of marinetenure—where individuals or groups of fishermen areguaranteed a specific share of future catch—for usersthat reconciles their economic interest with long-termconservation, eliminates the race-for-fish, and reducesor eliminates incentives for overcapitalization of fish-ing fleets, (2) recognition that MSY is a poor fisheriesmanagement objective and that economic and biologi-cal outcomes are better when catches are below MSYand stock sizes consequently higher, (3) direct involve-ment of stakeholders in data collection, data analysis,and decision making, (4) setting the spatial scale of thedata collection, science, and management appropriateto the spatial scales of the fish and the fishermen, and(5) management agencies that explicitly strive forharvesting capacity to match the long-term productivecapacity of the resource.

The central theme of this paper is that, by consider-ing humans in ecosystem management, we recognizethat appropriate incentives can stop the race-for-fishand eliminate or reduce most of the current problemsin fisheries management. In the sections below Iexplore the nature of incentives, and how incentivesinteract with other aspects of fisheries managementincluding MSY, institutional structure, and singlespecies management.

Incentives. When there is a race-for-fish, fishermenincrease their incomes by fishing harder, buildingbigger boats and catching fish before someone elsedoes. There is no individual economic incentive forconservation. With various forms of marine tenure,conservation of the resource is in the individual fisher’seconomic interest. The strongest form of tenure isresource ownership, which is the oldest form of fish-eries management in much of the world, found incommunity control of fishing grounds in the westernPacific (Johannes 2002) and now used as the primarymanagement system in Chilean artisanal fisheries(Castilla & Fernández 1998). A different form of owner-ship is allocation of fishing rights by the state throughhigh access fees or auction as is practiced in the Falk-land Islands (Barton 2002) and in Washington State formanagement of geoduck.

276


This contrasts with conventional management inwhich the state gives away the rights to fish and thenuses tax revenue to manage the fishery. When highaccess fees are charged, the state has both the incen-tive and the revenue to implement stringent top-downcontrol. Tenure granted to cooperatives is anothermechanism to stop the race-for-fish since it allows thecooperatives to concentrate on economic maximizationof yield from the fishery. Coops have been imple-mented for hake and pollock on the west coast of theU.S., for salmon in the Chignik area of Alaska, and forseveral fisheries in Mexico. The most broadly usedform of marine tenure is individual quotas in whicha specific portion of the total catch is allocated toindividuals or vessels. Individual Transferable Quotas(ITQs), under which individuals can catch and/or selltheir right to catch a portion of the total allowablecatch, have now been implemented in New Zealand,Australia, Iceland and several specific fisheries withinthe U.S. and Canada. ITQs, like other forms of marinetenure, provide incentives to reduce fishing capacityto a level appropriate for productive capacity of theresource and to concentrate on minimizing costs andmaximizing value of the catch, since the total catch isdetermined by a science-based public process (NRC1999a).

Single species management. A major element in theLitany is a list of fisheries collapses that includes thesea otter, the great whales, the northern cod, andbluefin tuna (NRC 1999). In fact, none of these reallyillustrate that single species management cannot work.Rather, they are examples of failures to do singlespecies management properly, since the stocks weregenerally fished down to less than 1% of their originalbiomass—far below single species guidelines of 25 to50%. Sea otter, great whales and bluefin tuna werelargely unregulated and highly valuable. The naturaloutcome was to move to the bio-economic equilibriumwhich is near extinction. For these stocks, singlespecies management did not fail, it wasn’t practiced.In northern cod, the scientific/political system failed(Harris 1998). While ecosystem changes may haveresulted from the severe depletion of these stocks,these changes would likely not have happened had thestocks been maintained at the abundances called forunder conventional single species management. Thus,this list of fisheries failures suggests that the problemwas poor implementation of single species manage-ment rather than a need to move beyond it.

MSY. MSY emerged in the 1950s as the defaultmanagement objective within fisheries science. How-ever, by the mid-1970s it had been largely discreditedamong scientists who recognized that maximizing thetons of fish landed was unlikely to be the appropriategoal of fisheries management (Larkin 1977). Yet, be-

ginning with the Law of the Sea, and later throughnational legislation in many countries, MSY becamefirmly enshrined as the default objective of fisheriesmanagement. The result is that management agenciesnow try to determine the maximum yield that couldpossibly be obtained from a fish stock, and regulatoryagencies try to set catch limits at the maximum thatcould be harvested. This ignores the fact that theeconomic optimum is almost always at yields lowerthan the MSY, and involves less fishing pressure. Oncethe race-for-fish is eliminated, the fishing industryrecognizes that it is better served by higher stock sizeand, consequently, higher catch-per-hour fished aswell as lower, but more stable catches. MSY is oftenincompatible with economically viable fisheries.

Political decision making and stakeholder involve-ment. The track record of most fisheries managementagencies is not good, and this failure has often beenblamed on the participation of self-interested stake-holders in the decision-making process. This has ledto frequent calls for ‘science based management,’ inparticular for the elimination of commercial and recre-ational fishermen from the decision making process. Iargue that the major problem with political decisionmaking as commonly practiced is that the allocationbetween competing groups (nations, gear types, com-munities) and the questions of conservation and sus-tainability are not distinguished. As most fisheriesinvolve individuals or groups competing for a share ofthe fish, the agencies often spend almost all theirenergy on allocation between competing users. Oncethe race-for-fish is replaced by some form of tenure,representatives of fishing groups will become an inter-est group with a high vested interest in making deci-sions that will allow for the long-term sustained use ofthe resource. With appropriate incentives, commercialfishing groups have often called for lower catches,have engaged in data collection and analysis, and haveoften even funded the majority of the scientific advis-ing process.

Ecosystem management of fish and fleets. Theimportant elements in incentive-based ecosystemmanagement are fishing fleets and fish, rather thanfish and their ecosystem. The dynamics of investment,fish harvesting, markets, and the incentives for fisher-men to conserve fish are, the most important con-siderations for sustainability. The trophic interactionsbetween species, the dynamics of marine ecosystems,or the scientific approach applied in determining quotarecommendations are secondary considerations. Fol-lowing from this, ecosystem management should havethe following characteristics: (1) incentives in the formof marine tenure will be in place so that the long-termeconomic and social benefits of all participants will bemaximized by sustainable fishing practices; (2) data

277

Mar Ecol Prog Ser 274: 269–303, 2004

collection, analysis, setting regulations, and enforce-ment, will be on the spatial scale appropriate to thebiology of the fish and the structure of the fishing com-munities; (3) stakeholders will be intensively involvedin all levels of science, management and enforcement,and under some circumstances fishing groups willhave complete control over the resource; (4) all costs ofresearch, management and enforcement will be paidby user groups; (5) the primary role of central govern-ments will be to audit the system to assure that thebiology and economics of the fishery are sustained andto ensure that national/international agreements andlaws are respected and enforced; and (6) substantialportions of the marine ecosystem will be protectedfrom fishing activity to provide biodiversity reservesand reference sites (in the sense of an unexploitedcontrol group).

The Pew Oceans Commission identified governancestructure as the key failing in U.S. fisheries policy (PewOceans Commission 2003), and recognized the need toseparate allocation from conservation decisions. How-ever, this commission did not see a significant role forincentives. Rather, it recommended strong, centralized,top-down control. The top-down approach contrastswith the incentives approach in that the former oftenviews the exploiters of marine resources as naturaldestroyers of marine environments who need to beexcluded from decision making as much as possible,while the latter views them as necessary partners inachieving good management.

Where economic incentives are not enough. Thestrict economic incentives associated with marinetenure will not protect all ecosystem components fromthe effects of fishing. For example the following topicswould still need to be addressed: (1) unproductivespecies in mixed species fisheries; (2) by-catch ofthreatened or endangered species; (3) trophic impactsof fishing; (4) habitat impacts of fishing; (5) long-livedspecies where the economic optimum is depletion; and(6) where international jurisdictions makes grantingtenure difficult or impossible. The economic return totenure holders is not increased by avoiding these prob-lems and here I see governmental agencies having animportant auditing role. Consider a theoretical exam-ple in which some group had been granted ownershipand management rights to fishing grounds. The tenureholder should be required to develop a managementplan associated with the areas of concern listed above,that would include monitoring, evaluation and en-forcement. The management plan might involvemandatory by-catch quotas, gear modifications toavoid non-target species, prohibition of destructivefishing gears, or overall catch quotas on some non-target species. For many fisheries, this may requireintensive, perhaps complete, observer coverage. While

this is very expensive, it may well be the true real costof achieving economically sustainable fisheries thatmeet society’s goal to protect biodiversity. Alternativesmight include expanding protected areas as reservesfor by-catch species that would then be unprotected inthe exploited areas. Incentives have an important roleto play because the higher the market value of a spe-cific form of tenure is, the more important it is to thetenure holder not to have the tenure revoked due toviolation of regulations.

By offering user groups marine tenure that givesthem much more direct control of their own destiny,and of a highly valuable asset, governments have beenable to obtain agreements with fishing groups toaccept and maintain industry funding of the costsof fisheries research and management (Australia,New Zealand, Iceland, Chile) as well as intrusive andexpensive observer coverage. I am not advocatingITQs, and the usual allocation based on catch histories,as the primary form of tenure. There are many otherforms of tenure that would achieve the desired goals,among them state ownership with high access fees andcooperatives. However, to achieve a politically viabletransition from our current system to a tenure systemsomething has to be offered to the fishermen. Theobvious solution is a significant portion of the futurecatching rights in the form of ITQs, with the remainderowned and leased by the state.

Summary. Ecosystem management means differentthings to different authors. I present here my vision ofthe key elements of such an approach. The emphasison institutions and the evolution of current single spe-cies management approaches is consistent with manyothers, but differs greatly from the ‘revolutionary’change called for in response to the perceived failureof single species management. I see the failures of fish-eries management as being due to a failure to recog-nize the importance of people and people manage-ment, not due to single species management. I supportthe view of ecosystem management that recognizesthe institutional dynamics between harvesters, man-agers and scientists, and stops the race-for-fish andovercapitalization through incentives rather than stop-ping overfishing through centralized top-down con-trol. I share with the papers of the Litany a commonvision of the world’s fisheries that have smaller fishingfleets, higher stock biomasses and significant areasprotected from fishing. However, I see a very differentway to achieve these goals. In my vision incentives arekey, fishermen are involved in all aspects of manage-ment, and they also pay for the annual costs of fisheriesmanagement.

Acknowledgements. I thank Doug Butterworth, Serge Garcia,Loo Botsford, Dan Huppert, J. J. Maguire and Kevin Stokesfor comments on the manuscript.

278


The ecosystem approach to fisherymanagement: a significant step towards

sustainable use of the marineenvironment?

Simon Jennings

Centre for Environment, Fisheries and Aquaculture Science,Lowestoft Laboratory, NR33 0HT, UK


Environmental managers regulate human activitiesto improve ecological, social or economic sustainabil-ity. Such regulation is not always effective, and mostfisheries are seen as excellent examples of failednatural resource management. While regulation andsocietal pressure have often led to reductions in theenvironmental impacts of shipping, aggregate dredg-ing, waste disposal and the oil and gas industries, fish-ing is widely seen as the remaining pariah, currentlyattracting the attention of the global media and numer-ous conservation and lobby groups.

Today, most fisheries are managed on a stock-by-stock basis. Reference points are established for stockbiomass and fishing mortality and then catch controls,effort controls or technical measures, such as changesin mesh size or area closures, are recommended tomanagers to modify mortality rates. In reality, man-agers have always struggled to reduce fishing mortal-ity, and the biomass of many stocks is below intendedreference points (FAO 2002a). The failures of manage-ment are catalogued in numerous publications and theprincipal ecological, social and economic reasons forfailure are well understood (OECD 1997, FAO 2002b).This understanding has thus far done little to improvethe overall effectiveness of management in ecological,social or economic terms.

Although the depletion or collapse of target stocks isoften the most visible and well-publicised failure of thefisheries management process, fisheries take place inecosystems and have wide ranging ecological impacts.These impacts have become an increasing focus ofresearch effort, as evidenced by recent symposia(Gislason & Sinclair 2000, Kaiser & de Groot 2000,Sinclair & Valdimarsson 2003) and reviews (Gislason1994, Dayton et al. 1995, Jennings & Polunin 1996,Jennings & Kaiser 1998, Hall 1999, NRC 2002). Thisinterest in fisheries ecosystem interactions is not new(e.g. Anderson & Ursin 1977, Pope 1979, Pope et al.1988) but the recent shift in research effort from singlespecies to ecosystem-based concerns reflects thegrowing recognition that an ecosystem approach mayhelp to underpin improved management.

From a policy perspective, the move towards anecosystem approach has been rapid and is consistent

with wider commitments to sustainable development.Indeed, while many commentators are still asking forfishing impacts to be considered in environmentalpolicy, the requirements to protect ecosystems fromthe wider impacts of fishing, and to adopt an ecosys-tem approach, have already been written into most ofthe key policy documents relating to marine environ-mental management (Sainsbury & Sumaila 2003, Rice2004). The ecosystem approach, as described in exist-ing policy documents (e.g. WSSD 2002), contributesto sustainable development, which requires that theneeds of future generations are not compromised bythe actions of people today. The ecosystem approach isvariously defined, but principally puts emphasis on amanagement regime that maintains the health of theecosystem alongside appropriate human use of themarine environment, for the benefit of current andfuture generations.

EAF is part of the ecosystem approach. The broadpurpose of the EAF is to plan, develop and managefisheries in a manner that addresses the multiple needsand desires of societies, without jeopardising theoptions for future generations to benefit from the fullrange of goods and services (including, of course, nonfisheries benefits) provided by marine ecosystems(FAO 2003). The success of an ecosystem approach willdepend on whether these high level and somewhatabstract commitments can be turned into specific,tractable and effective management actions (Sains-bury et al. 2000, Sainsbury & Sumaila 2003).

To assess the potential of the ecosystem approach,we need to ask whether it will nullify the failings ofexisting approaches and change attitudes to use of themarine environment. From ecological, economic andsocial perspectives, existing management methodshave generally failed. Thus, 47% of the world’s mainstocks or species groups are fully exploited, while 18%are overexploited and 10% are severely depleted orrecovering from depletion. Only 25% of stocks areunder- or moderately exploited (FAO 2002a). The FAOconducted one of the most comprehensive analyses ofthe factors contributing to unsustainability in fisheries(FAO 2002b). These were inappropriate incentives andmarket distortions, high demand for limited resources,poverty and lack of alternatives to fishing, complexityand inadequate knowledge, lack of governance, andinteractions of the fishery sector with other sectors andthe environment (FAO 2002b). Their analyses showedthat scientific advice on the status of fish stocks and theeffects of fishing made only a small contribution to acomplex management and decision-making process,and often carried little weight in relation to immediatesocial and economic considerations. Advice on fish-eries exploitation in an ecosystem context will alsomake a small contribution to a larger process that is

279

Mar Ecol Prog Ser 274: 269–303, 2004

influenced by many of the same social and economicfactors. Thus, scientific advice may carry little weightwhen there are very high short-term social and eco-nomic costs associated with moving towards sustain-ability. These costs are common to both single speciesand ecosystem-based approaches (Rice 2004). Theecosystem approach will not remove the very highshort-term costs of protecting the environment unlessincentives are introduced to link conservation andshort-term financial reward.

From an ecological perspective, the ecosystem ap-proach recognises, and aims to remedy, the unwantedimpacts of fishing on non-target species, habitats andecological interactions. The approach recognises thatecosystems provide goods and services other than fishand may change the burden of proof if existing man-agement is not precautionary (Sainsbury & Sumaila2003). However, in the broadest directional terms,scientific advice is consistent from both single-speciesand ecosystem perspectives: significant capacity re-ductions are needed. The most pervasive ecosystemimpacts are still the result of massive over capacity,and scientific advisers on single-species issues havebeen arguing for capacity reductions and time or areaclosures for decades. Managing fisheries in an ecosys-tem context also leads to advice to reduce capacity andimplement time or area closures. True, there are caseswhere otherwise sustainable fisheries have additionaladverse effects on non-target species and habitats(Witherell et al. 2000) but, at the present time, suchfisheries are in the minority relative to those wheremortality has to be cut simply to ensure conservation oftarget stocks. Indeed, the ICES Advisory Committeeon Ecosystems concluded that managers would haveto deal with a much smaller and more tractable set ofecosystem issues if capacity were reduced to the extentthat all target stocks were fished sustainably (ICES2001).

The preceding arguments suggest that the transitionfrom single-species to ecosystem-based approacheswill not alter the high short-term social and economiccosts of reducing capacity nor the general advice thatcapacity should be reduced. Thus, scientific advice onthe North Sea cod fishery that is framed in an eco-system context would not be more stringent than therequest for a zero catch in 2004 (ICES 2003a). Perhapsa more relevant issue is whether the adoption of theecosystem approach will encourage society to exertmore pressure on Governments to bear high short-term costs, and to translate high level political commit-ments into capacity reductions and improvements inthe ecological status of the marine environment. Ulti-mately, society’s willingness to bear these high short-term costs, directly or indirectly, will determine thesuccess or failure of the ecosystem approach. Market

instruments that capture at a private level the socialand global values of relatively undisturbed ecosystemsthrough, for example, premium pricing for fish caughtfrom healthy ecosystems (Phillips et al. 2003), may helpto increase the short-term benefits associated withconservation. However, such instruments will not pro-mote conservation in many areas where unsustainablefisheries provide the main source of food, income andemployment. This requires a willingness of Govern-ments to commit substantial international funding, butthe gap between commitment and available fundingis large and growing (UNDP 2003).

Scientific research has shown that the sea providesessential ecosystem goods and services with high long-term value (Balmford et al. 2002), yet human impactson the sea are rarely an important political issue incomparison with health, poverty, education and mili-tary disputes. Management of the marine environmentis not a top spending priority for Governments becauseit does not have an immediate impact on most voterslives. Public attitudes, rather than new types of scien-tific advice, are most likely to change this. In thisrespect, high profile and media friendly conservationprojects, such as those supported by the Pew Charita-ble Trusts, will have a significant role in changing pub-lic perceptions, and may serve to increase the short-term political costs associated with the failure to movetowards sustainability.

The extent to which society can strengthen the casefor management action was well demonstrated by theeffects of consumer and conservation campaigns onattitudes to marine mammal bycatch. Indeed, pressureon the US Government led to the implementation ofthe Marine Mammal Protection Act in 1972. Thisrequired the adoption of fishing practices that reduceddolphin bycatch and the presence of independentobservers on vessels to monitor and control bycatches(M. A. Hall 1998). By 1972, another bycatch species,the common skate Dipturus batis, was effectivelyextinct in the Irish Sea (Brander 1981). There was littlepublic interest in the common skate, and over 30 yrlater no specific measures have been implemented toprotect this species (Dulvy et al. 2003). Clearly, theinfluence of society on commitments to policy imple-mentation has the potential to create ecosystems thatare dominated by ‘favoured’ species. Although theecosystem approach is intended to take account ofhuman impacts on the whole ecosystem, the first stepstowards implementation may be remarkably piece-meal and have a range of unexpected consequences.

Thus far, attempts to implement an ecosystemapproach have often been characterised by a polariseddebate between ‘ecosystem’ and ‘stock assessment’scientists, and the unwillingness of some advocates ofecosystem based management to accept useful parts of

280


the existing management system. The most effectiveprogress towards an ecosystem approach is likely to beachieved by moving forward collectively, integratingthe useful aspects of existing approaches into newones. There are 2 reasons for this. First, both ecosys-tem- and stock-based approaches, at least at a globalscale, lead to the same advice; to reduce fishing capac-ity and restrict access. Second, it has taken a long timeto improve understanding of the issues that affect thesuccess of environmental management, such as deal-ing with risk and uncertainty (Hilborn 1996, Harwood& Stokes 2003), and such insight should not be wasted.The assumption that solutions are simple, but over-looked or untested, has led to many false dawns inenvironmental management.

One such concern exists in relation to closed areas,which are increasingly proposed as an almost singularsolution to the adverse effects of fishing. While closedareas are an important management tool, and arefrequently not used even when they could mitigateunsustainable fishing impacts (Sainsbury & Sumaila2003), a single-minded focus on area closure asopposed to capacity reduction and other measures isunlikely to reduce significantly the aggregate impactsof fishing. Thus, increased use of closed areas withoutassociated capacity reduction will displace fishingimpacts to places where fisheries regulations are not sostringent, and to more vulnerable areas, such as partsof the deep sea (Koslow et al. 2000). Progress towardseffective ecosystem-based management will ulti-mately depend on both access restriction and effectivecapacity reduction. However, the increasing applica-tion of area closure in supporting aspects of ecosystem-based management (e.g. protection of vulnerable habi-tats, genetic diversity or food web structure) will beginto play an important role in changing perceptionsabout open access to the marine environment.

The ecosystem approach is sometimes seen as end-lessly complicated, and it is a common misconceptionthat we need to understand the structure and functionof entire ecosystems to implement effective ecosystem-based management. While understanding ecosystemsis a worthy intellectual exercise, it can be an inappro-priate and unrealistic use of limited resources thatcould be used to address specific and tractable issues.True, the science required to underpin the ecosystemapproach will be more diverse than that contributing tofisheries stock assessment, but funding for this sciencecannot be expected to increase in proportion to therange of ecosystem issues that scientists will be askedto address. The most significant and cost-effectiveprogress towards the ecosystem approach is mostlikely to be made by appropriate reorientation of exist-ing science and management tools. An emphasis on anevolutionary rather than revolutionary move towards

the ecosystem approach is less likely to paralyse thedecision-making process and will help to maintainbroad based support.

From a practical perspective, the essential diversityof scientific involvement in the ecosystem approachcan readily confuse managers. Thus fisheries man-agers who once turned to stock assessments, now haveto consider genetic and species diversity (Law 2000,Murawski 2000), species rarity (Casey & Myers 1998,Schindler et al. 2002), habitats (Collie et al. 2000,Kaiser et al. 2002), food web properties (Pauly et al.1998, Cury et al. 2003) and the ecology of marinemammals and birds (M. A. Hall 1998, Tasker et al.2000) when managing the marine environment. Fish-ing has become an issue on which most ecologists havestrong opinions, but the breadth of knowledge andexperience required to provide balanced and credibleadvice that can actually be used by decision makers isformidable (Sissenwine & Mace 2003). It will also bedifficult for managers to reconcile the range of advicethey receive in the absence of established guidelineson the implementation of an ecosystem approach;though some management agencies have such guide-lines (Constable et al. 2000, Witherell et al. 2000)and most others are working towards them (FAO 2003,Rice 2003).

One component of the ecosystem approach that mayplay an increasing role in shaping the future of marineenvironmental management is the use of environmen-tal impact assessment. Fisheries are effectively exemptfrom the requirements for impact assessment, even inareas where other users of the marine environment,such as the oil and gas industries, would be required toconduct them. There is a precedent for a move towardsenvironmental impact assessment in the FAO Code ofConduct for Responsible Fisheries (FAO 1995b), whichsuggests that conservation and management should becautious until sufficient data for assessment are avail-able. Impact assessment would usefully deal withsocial and economic as well as ecological factors, butwould need to incorporate an agreed long-term per-spective to reduce the significance of high short-termcosts. Moreover, the application of impact assessmentwould require new management structures that facili-tated collaboration between marine ecologists, socialscientists, lawyers and economists, but did not paralysethe decision-making process.

To conclude, the mechanisms to implement an eco-system approach are increasingly well developedand such an approach will improve sustainability inwealthier nations, provided that society is stronglysupportive. With support from society, managementmethods would be expected to evolve quite rapidlyuntil fisheries are treated on a par with other sectoralactivities that impact the marine environment. It is

281

Mar Ecol Prog Ser 274: 269–303, 2004

expected that the capacity of fishing fleets will bemuch reduced, there will be fewer subsidies, new fish-eries will only be licensed following impact assessmentand habitat and species conservation issues willbecome an increasing focus of management plans.Indeed, the work of the North Pacific Fishery Manage-ment Council suggests that the ecosystem approachcan be implemented effectively when there is suffi-cient commitment (Witherell et al. 2000). In manypoorer nations, prospects for improved sustainabilityare not good, unless the international communitycommits to supporting and financing the ecosystem ap-proach and subsidising the very high short-term socialand economic costs associated with reducing capacity.

Acknowledgements. Many thanks to Colin Bannister, NickDulvy, Joe Horwood, Jake Rice and the editors of this TS forhelpful comments on the text. Their generosity in reading thetext does not necessarily mean that they all agree with myopinions!

Repetitive history of resource depletionand mismanagement: the need for a

shift in perspective

Heike K. Lotze

Wattenmeerstation Sylt, Alfred Wegener Institute for Polar andMarine Research, Hafenstrasse 43, 25992 List, Germany

Present address: Leibniz Institute for Marine Science, Experi-mental Ecology, Düsternbrooker Weg 20, 24105 Kiel, Germany


History tends to repeat itself. Plentiful resourcesalways impressed humans as being inexhaustible. Weexploited them without thinking much about eco-logical consequences and replenishment. Only whenresources declined did we start to implement manage-ment actions such as privatization, quotas, closedseasons and other restrictions. High human demand aswell as economic, social or political pressures, how-ever, often undermined sufficient management prac-tices leading to overexploitation and collapse. Unfortu-nately, human societies usually did not question theiractions or demands when resources collapsed, butmoved on to either (1) exploiting the same speciessomewhere else, (2) exploiting a less preferred specieslocally, or (3) intensifying local resource productionthrough aquaculture. Today, these ‘solutions’ are stillwidely used, but hardly work anymore. We havereached global limits of exploitation at the poles, theopen ocean and the deep sea. We have successivelydepleted lakes, rivers, coastal seas, and finally theopen ocean, leaving many species overexploited,endangered or extinct. Although a potential solution tosubstitute for depleted stocks, aquaculture of high

trophic level species faces limits. Single-species man-agement approaches aiming for maximizing resourceoutput to humans have often failed to prevent deple-tion and collapse. Multiple human impacts that destroyhabitat and environmental quality essential to the sus-tenance of aquatic species need to be considered. If‘ecosystem-level’ management is used just as a newlabel hiding the continuation of ongoing practices andattitudes, we will drive aquatic resources to furtherdepletions, collapses and extinctions, possibly passingthe point where recovery would still be possible. Thereis an alternative. Ecosystem-level management shouldaim for managing ecosystems with the goal of optimalfunctioning of all parts, including ourselves. Thisrequires a shift in perspective. We are faced with thechallenging opportunity to break our historical patterns.

Repetitive history of resource use and management.Apparent inexhaustibility of unexploited resources:Whenever people in the past encountered oceanicregions that were formerly not or only little exploited,the vast richness of large fish, birds, turtles, whales,and other marine animals astonished them. Whetherpeople visiting the Baltic or North Sea 1000 yr ago(Hoffmann 2001, 2002), or Europeans reaching theNew World 500 yr ago (e.g. Cabot 1497/98 cited inHoffmann 2001, Rosier 1605 cited in Steneck 1997),their descriptions are similar. The newly discoveredseas and the bounty of life always seemed inex-haustible. Even in the 19th and 20th century, peoplecontinued to believe in this myth of inexhaustibility(Hutchings & Myers 1995, Pauly et al. 2003). Through-out our history, we have repeatedly proven ourselveswrong.

Human population density and demand: Prehistoricpeople hunted, fished and gathered to sustain them-selves or to trade with neighbors. Archaeological evi-dence suggests that in regions with low populationdensity indigenous people had no or little impact oncommon target species such as marine mammals,birds, fish and shellfish (Steneck 1997, Lotze & Milew-ski 2004). In contrast, in regions with high humanpopulation density, most valued species declined inrelative abundance, size or distribution over time,indicating high exploitation pressure (Broughton 1997,Smith 2004). Thus, in some hunter-gatherer societieshuman population density and demand was alreadyhigh enough to cause severe resource depletion.

Since then, human population has grown exponen-tially and demands have multiplied, not only for food,but increasingly for profit, fashion, and prestige. Forexample, rare sturgeon or salmon were reserved forkings and the elite in the late Middle Ages (Hoffmann2001), whales were hunted for their baleens, whichwere used in ladies’ fashion, and seabirds were killedin the millions to supply the millinery trade in the 19th

282


century (Lotze & Milewski 2004). Excessive exploita-tion has resulted in rapid depletion and extinctionssince the Middle Ages, and especially in the 19th cen-tury (Hoffmann 2002, Lotze & Milewski 2004). Fishers,hunters, traders and entire nations increasingly com-peted in the rush for valued but dwindling resources(Hoffmann 2002).

Shifting values and subsequent conservation effortsin the 20th century led to the recovery of some species(Murawski et al. 2000, Cloern 2001, Lotze & Milewski2004). Today, ocean wildlife is exploited to meet thefood demands of an ever-increasing human popula-tion, as well as to supply global luxury markets. Thisgrowing demand, however, is restrained by anincrease in the number of collapsed or overexploitedfish stocks, and declining global catches (Botsford et al.1997, Pauly et al. 2002, Myers & Worm 2003, 2004).While human population growth in the Middle Ageswas mainly fuelled by the supply of cereals (Hoff-mann 2001), today’s population demands a continuoussupply of fish and meat. Clearly, as a society, we needto adapt our demands to the capacity of marine eco-systems, not vice versa.

Resource depletion and management: Throughouthistory, humans have reacted to local resource deple-tion by implementing management actions that be-longed to 4 major categories: (1) privatization andregulation, (2) expansion to unexploited regions, (3)substitution of depleted target species with lessexploited species, and (4) intensification of local pro-duction through aquaculture.

In the Middle Ages, human population densityincreased markedly throughout Europe and the firstsigns of depletion of preferred aquatic food sourcessuch as sturgeon and salmon were already evident inthe 13th century (Hoffmann 2001). Privatization andregulation with quotas, gear, seasonal and other rest-rictions were implemented by landowners or territorialauthorities (Hoffmann 2002). However, in the MiddleAges, as well as today, these management practiceswere often overridden by socio-economic pressures(Botsford et al. 1997). Therefore, a continued decline inresources led to the expansion of frontiers to formerlyunexploited regions. Fisheries moved from freshwaterto marine environments in medieval Europe (Hoff-mann 2001, 2002), from inshore to offshore in theNorth Sea and North Atlantic beginning in the 1400s(Hutchings & Myers 1995, Steneck 1997, Hoffmann2002, Lotze & Milewski 2004), and to the open ocean,polar and deep seas in the 19th and 20th century(Pauly et al. 2002, 2003, Myers & Worm 2003). Thehistory of whaling shows a similar pattern of spatialexpansion from coastal to offshore and polar regions,as well as serial depletion of one species after another.This successive substitution of depleted target species

with formerly unexploited species that were less val-ued, smaller, harder to catch, or lower in the food webis the third common management practice (Fig. 1).Today, low-trophic level exploitation of crustaceans,mollusks and marine plants dominate most coastalfisheries (Pauly et al. 2002, Lotze & Milewski 2004).The fourth form of management practice is intensifica-tion of local production. Like fishing, aquaculturemoved from freshwater to anadromous to marinefishes. Aquaculture of introduced carp was invented inthe Middle Ages (Hoffmann 2002), that of salmon inthe 1970s, and today farming of marine groundfishsuch as cod or haddock is becoming a reality. In con-trast to herbivorous carp, however, aquaculture of pis-civorous fish faces limits and creates many environ-mental problems (Pauly et al. 2002).

Whether it is privatization and regulation, expan-sion, substitution, or intensification, we still repeathistorical patterns, albeit on a global scale (Botsford etal. 1997, Pauly et al. 2003). Today, depleted aquaticresources are the rule rather than the exception. Large,long-lived species such as northern right and hump-back whale, great auk and Labrador duck, sturgeonand salmon, haddock and cod, sharks and rays areextinct or rare, i.e. at around 1 to 10% of their former

283

Fig. 1. Substitution of depleted resources in the Outer Bayof Fundy, NW Atlantic. (A) Declining catches of traditionalgroundfish (cod, haddock, pollock; dotted line) led to increasinginvertebrate and plant landings (solid line). Note that highgroundfish landings in the 1960s arose from expanding tooffshore fishing grounds and the introduction of otter trawls.(B) Increase in the number of target species in commercialfisheries over time. Data adapted from Lotze & Milewski (2004)

Mar Ecol Prog Ser 274: 269–303, 2004

abundance (Myers & Worm 2003, 2004, Roman &Palumbi 2003, Lotze & Milewski 2004). Traditionalmanagement approaches have failed to ensure sus-tainable use of aquatic resources (Botsford et al. 1997,Pauly et al. 2002), and extrapolation of present trendsinto the future presents us with a grim picture (Pauly etal. 2003). If aquatic wildlife and ecosystems as well asfisheries productivity is to be sustained, our societyneeds to shift to more sustainable management andquestion its demands.

Ecosystem management as a shift in perspective.Towards an ecosystem perspective: For a long time,the goal of single-species management was to managepopulations for maximum possible output for humans.If ecosystem-level management is used in the samesense, it will surely only accelerate present patterns ofdepletion and degradation. Ecosystem-level manage-ment should mean that ecosystems are managed withthe goal of optimal functioning of all parts includingourselves. This requires that (1) all the parts (species,habitats) are kept, (2) all parts are kept in a state (ofabundance, diversity, complexity) that allows long-term persistence and resilience of populations, com-munities and ecosystems, and (3) high environmentalquality is provided to ensure health and survival. Italso requires integrating multiple human impactsinto an ecosystem framework because humans inter-fere with all parts through the cumulative effects ofexploitation, habitat destruction, nutrient loading, pol-lution, and other disturbances. Diverse, productive andfunctioning ecosystems will not only conserve aquaticwildlife and wilderness, but will also likely enhanceproductivity, water quality, economic options and othergoods and services for human societies.

Integrating multiple human impacts: Humans havemultiple impacts on aquatic ecosystems that interactwith one another and must, therefore, be managedtogether. Historically, direct exploitation was the firsthuman impact on aquatic resources. In a food web con-text, humans act as top-predators having ‘top-down’impacts which have increased multi-fold from earlysubsistence cultures to today’s societies (Lotze &Milewski 2004). These direct impacts on populationsare complicated by indirect community effects such asdepensation and trophic cascades. Moreover, humanactivities also affect the food web from the ‘bottom-up’through resource enhancement such as nutrient load-ing (Cloern 2001, Lotze & Milewski 2004). These bot-tom-up impacts interact with top-down impacts.Reduction of consumers and enhancement of nutrientloads, for example, can result in excessive algalblooms, loss of diversity and ecosystem functions(Worm et al. 2002). Each trophic level is furtheraffected by pollution effects on health, habitat destruc-tion, and increasing stress due to disturbance, traffic

and noise. These ‘side-in’ impacts reduce overall avail-ability of high quality habitat and environment, andthe amount of undisturbed space and time (Lotze &Milewski 2004).

The cumulative effects of top-down, bottom-up andside-in impacts can alter species interactions, acceler-ate species declines and impair recovery (Lotze &Milewski 2004). In medieval Europe, deforestation,agricultural expansion, river damming, water pollu-tion, and nutrient loading had already affectedfreshwater fishes in addition to direct exploitation(Hoffmann 2001, 2002). Recovery of Atlantic salmonwas for a long time impaired by river pollution anddestruction of spawning habitats (Lotze & Milewski2004). Starting in the rivers, multiple human impactsalso spread into estuaries and coastal seas, possiblyimpairing recovery of collapsed groundfish stocks(Lotze & Milewski 2004). With climate change andworldwide fishing, humans today affect the oceans ona global scale. Former human civilizations collapsednot only because of food shortage but also because ofthe indirect effects of exploitation such as water andfuel shortage (Hughes 2001). Today, our society hasthe advantage of knowing what we are doing, and theoption of acting upon that knowledge.

Ecosystem-level versus human-impact management:Managing an entire ecosystem will be a difficult taskbecause of our limited understanding of all its partsand the linkages between them. In many cases,however, we have a reasonably good understandingof human impacts and should, therefore, focus on‘human-impact’ management in order to reduce ournegative and enhance our positive influences. Inaddition, marine protected areas (no-take zones) areneeded as controls to measure change against, asinsurance against management failures, to preservediversity, and to ensure the persistence and resilienceof aquatic ecosystems (Palumbi 2001, Worm et al. 2003).

Human-impact management should include techni-cal improvements to minimize negative impacts, pro-tection and restoration of species and habitats, and thereduction of our demand as feasible managementoptions. ‘Top-down’ impacts can be reduced by effortcontrol through quotas and cutback on subsidies, whichwill help to re-balance size of fish stocks and fishingfleets (Botsford et al. 1997, Pauly et al. 2002). Protectedareas reduce the spatial extent of exploitation and pro-tect threatened diversity (Worm et al. 2003). Technicalimprovements of more selective and less destructivegear types reduce bycatch and habitat destruction.‘Bottom-up’ impacts can be reduced by wastewatertreatments at point sources, while restoration of wet-lands as natural buffer and filters will reduce non-pointpollution (Cloern 2001). Reduction of ‘side-in’ impactsrequires technical improvements to reduce chemical

284


pollution, noise stress, and destructive harvesting prac-tices, in addition to protected areas that allow habitatregeneration and species recovery (Murawski et al.2000, Palumbi 2001). Technical improvement of aqua-culture can reduce discharges of chemicals, pharma-ceutics, and wastes into the environment.

Reducing excessive and destructive exploitation andenhancing habitat availability and environmentalhealth were successful measures for recovery of somebirds, mammals, fish and invertebrates in the 20thcentury (Murawski et al. 2000, Cloern 2001, Lotze &Milewski 2004). It is the role of scientists to explore,test, communicate, and insist on implementation of thebest management options. It is the role of society

perspectives on ecosystem-based approaches to the … · warner 2003, lubchenko et al. 2003, pauly...

Documents