?!W^ 55.302:F 53/9 ^H^f Jgfrl.. ^J^ mU ' JA W ^m WWk^ T ' wftlHkHl GSYSll M- ' ifi"!^^ --■^V • :.r. ^''''^■1 WlS^«^^[V^K ip '^ ^wifei)^ ■V A ReDort ■■\ s Principles y Panel ♦ >' V V 'V \ i; ceanic and Jsmospheric Administration "^^.^^^.-A* National Marine FisheriesService ECOSYSTEM-BASED FISHERY MANAGEMENT nil 0 a ^^^^ A Report to Congress ^^^ by the ?S..eP--^«^^ Ecosystem Principles Advisory Panel As mandated by the Sustainable Fisheries Act amendments to the Magnuson-Stevens Fishery Conservation and Management Act of 1996 U.S. DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration National Marine Fisheries Service April 1999 NATIONAL MARINE FISHERIES SERVICE ECOSYSTEM PRINCIPLES ADVISORY PANEL Chair, David Fluharty University of Washington/North Pacific Fishery IVIanagement Council Pete Aparicio Texas Shrimpers Association/Gulf of IVIexico Fishery IVIanagement Council Christine Blackburn Alaska Groundfish Data Bank George Boehlert NMFS, Pacific Fisheries Environmental Laboratory Felicia Coleman Florida State University/Gulf of Mexico Fishery Management Council Philip Conkling The Island Institute Robert Costanza University of Maryland Paul Dayton University of California, San Diego Robert Francis University of Washington Doyle Hanan California Department of Fish and Game Ken Hinman National Coalition for Marine Conservation Edward Houde University of Maryland Center for Environmental Science James Kitchell University of Wisconsin Rich Langton Maine Department of Marine Resources Jane Lubchenco Oregon State University Marc Mangel University of California, Santa Cruz Russell Nelson Florida Marine Fisheries Commission/Gulf of Mexico and South Atlantic Fishery Management Councils Victoria O'Connell Alaska Department of Fish and Game Michael Orbach Duke University Michael Sissenwine NMFS, Northeast Fisheries Science Center NMFS Staff: Coordinator, Ned Cyr NMFS, Office of Science & Technology David Detlor NMFS, Office of Science & Technology Aligon Morgan Atlantic States Marine Fisheries Commission i TABLE OF CONTENTS Acknowledgments v Preface w/ Executive Summary 1 Section One: Introduction 9 Section Two: Ecosystem Principles, Goals, and Policies 73 Section Three: Current Application of the Ecosystem Principles, Goals, and Policies... 23 Section Four: Recommendations for Implementing the Ecosystem Principles, Goals, and Policies in U.S. Fisheries Conservation, Management, and Research 27 Section Five: Summary and Conclusions 37 Glossary 39 Literature Cited 41 Appendix A: Charter — NMFS Ecosystem Principles Advisory Panel 47 Appendix B: MSFCMA Section 406 Fisheries Systems Research 51 Appendix C: Meeting Participants 53 ACKNOWLEDGMENTS While the Ecosystem Principles Advisory Panel takes full responsibility for the content of this re- port, we would like to give thanks and credit to oth- ers for the assistance they so generously provided to us. The first thanks goes to members of Congress who responded to public and agency interests in ex- panding the use of ecosystem-based management in the fishery management processes in the United States. Next, we appreciate the help given to the National Marine Fisheries Service (NMFS) by the National Research Council in nominations for Panel membership. The Panel is extremely grateful to the NMFS staff, its regional Science Centers, regional administrative staffs, and Council staffs for their technical support and advice during this process. Similarly, a significant boost to our deliberations came from State and other agencies, individuals, and organizations who met with us (Appendix C) and provided considerable insight. A special thanks is due to Alec MacCall and four other (anonymous) reviewers of the report. Ned Cyr, David Detlor, and Alison Morgan, NMFS Office of Science and Tech- nology, composed the core team who coordinated meetings, produced drafts, and attended to all the details of text manipulation. Willis Hobart and David Stanton, NMFS Scientific Publication Office, de- serve special recognition for their editing assistance and development of a format for this presentation. Panel members owe a collective debt of gratitude to our respective institutions, colleagues, friends, and families who have supported and encouraged our par- ticipation in this endeavor. PREFACE Seeking solutions to reverse the decline of New England's fisheries in 1871, Congress created the U.S. Commission of Fish and Fisheries (Hobart 1995). The first appointed Commissioner, Spencer Baird, initiated marine ecological studies as one of his first priorities. According to Baird, our under- standing of fish "... would not be complete without a thorough knowledge of their associates in the sea, especially of such as prey upon them or constitute their food...." He understood that the presence or absence of fish was related not only to removal by fishing, but also to the dynamics of physical and chemical oceanography. Despite this historical, fundamental understand- ing of fisheries as part of ecosystems, we have con- tinued to struggle to manage fish harvests while si- multaneously sustaining the ecosystem. Recogniz- ing the need for a more holistic management ap- proach. Congress charged the National Marine Fish- eries Service (a direct descendant of the U.S. Com- mission of Fish and Fisheries) with establishing an Ecosystem Principles Advisory Panel to assess the extent that ecosystem principles are used in fisher- ies management and research, and to recommend how such principles can be further implemented to improve our Nation's management of living marine resources. The resulting Panel was composed of members of industry, academia, conservation orga- nizations, and fishery management agencies. The Panel's diversity played a substantial role in the de- velopment of a pragmatic approach to expand eco- system-based fishery management within the con- text of the existing fishery management system. The Panel attempted to build on the progress of past efforts, namely the 1996 Sustainable Fisheries Act's (SFA) amendments to the Magnuson-Stevens Fishery Conservation and Management Act (MSFCMA) (NMFS 1996). The provisions of the SFA require the Regional Fishery Management Councils to set harvest rates at or below maximum sustained yield levels; develop rebuilding plans for those species that are currently below the long-term sustainable yield; better account for and minimize bycatch and discard of fish; identify essential fish habitat and take measures to protect it; and deter- mine the effects of fishing on the environment. These actions are being implemented and are vital to achieving ecosystem-based management. Still, it will take years to decades before the results are fully realized. The Panel forged a consensus on how to expand the use of ecosystem principles in fishery manage- ment. We do not have a magic formula, but we offer a practical combination of principles and actions that we believe will propel management onto ecologi- cally sustainable pathways. By asking more encom- passing questions about fisheries management such as, "What are the effects of fishing on other ecosys- tem components?" and "What are acceptable stan- dards for fisheries removals from ecosystems?" we are broadening the scope of management and ulti- mately making fisheries sustainable. Ecosystem-based fishery management is likely to contribute to increased abundance of those spe- cies that have been overfished. It may, however, require reduced harvest of species of critical impor- tance to the ecosystem. We expect that ecosystem- based fishery management will contribute to the sta- bility of employment and economic activity in the fishing industry and to the protection of marine biodiversity on which fisheries depend. As a soci- ety, we are recognizing the limits of the sea to pro- vide resources and of our abilities to stay within those limits. What are acceptable levels of change in ma- rine environments due to fishing? This Report does not answer that question for society, but it does set a framework for beginning to take actions based on the insight of Baird 127 years ago. David Fluharty Chair, Ecosystem Principles Advisory Panel Seattle. Washington November 15, 1998 VII Digitized by the Internet Archive in 2012 with funding from LYRASIS IVIembers and Sloan Foundation http://archive.org/details/ecosystembasedfiOOunit EXECUTIVE SUMMARY Ecosystem-based management can be an important complement to existing fisheries management approaches. When fishery managers understand the complex ecological and socioeco- nomic environments in which fish and fisheries ex- ist, they may be able to anticipate the effects that fishery management will have on the ecosystem and the effects that ecosystem change will have on fish- eries. However, ecosystem-based management can- not resolve all of the underlying problems of the existing fisheries management regimes. Absent the political will to stop overfishing, protect habitat, and support expanded research and monitoring programs, an ecosystem-based approach cannot be effective. A comprehensive ecosystem-based fisheries management approach would require managers to consider all interactions that a target fish stock has with predators, competitors, and prey species; the effects of weather and climate on fisheries biology and ecology; the complex interactions between fishes and their habitat; and the effects of fishing on fish stocks and their habitat. However, the approach need not be endlessly complicated. An initial step may require only that managers consider how the har- vesting of one species might impact other species in the ecosystem. Fishery management decisions made at this level of understanding can prevent signifi- cant and potentially irreversible changes in marine ecosystems caused by fishing. Recognizing the potential of an ecosystem-based management approach to improve fisheries manage- ment. Congress requested that the National Marine Fisheries Service (NMFS) convene a panel of ex- perts to: 1) assess the extent to which ecosystem principles are currently applied in fisheries research and management; and 2) recommend how best to integrate ecosystem principles into future fisheries management and research. In response. NMFS cre- ated the National Marine Fisheries Service Ecosys- tem Principles Advisory Panel (Panel). WHAT BASIC ECOSYSTEM PRINCIPLES, GOALS, AND POLICIES CAN BE APPLIED TO FISHERIES MANAGEMENT AND RESEARCH? To guide our deliberations, we developed a set of eight ecosystem operating principles (Principles) with societal goals for ecosystems (Goals), and a set of six management policies (Policies). These Prin- ciples, Goals, and Policies were used to evaluate the current application of ecosystem-based fisheries management and to develop recommendations for further implementation of such approaches. BASIC ECOSYSTEM PRINCIPLES, GOALS, AND POLICIES Based on the Panel's experience and review of the fisheries ecosystem literature, we suggest that the following Principles, Goals, and Policies embody key elements for ecosystem-based management of fisheries. Principles • The ability to predict ecosystem behavior is lim- ited. • Ecosystems have real thresholds and limits which, when exceeded, can effect major system restruc- turing. • Once thresholds and limits have been exceeded, changes can be irreversible. • Diversity is important to ecosystem functioning. • Multiple scales interact within and among eco- systems. • Components of ecosystems are hnked. • Ecosystem boundaries are open. • Ecosystems change with time. Goals • Maintain ecosystem health and sustainability. Policies • Change the burden of proof. • Apply the precautionary approach. • Purchase "insurance" against unforeseen, adverse ecosystem impacts. • Learn from management experiences. ECOSYSTEM-BASED FISHERY MANAGEMENT Make local incentives compatible with global goals. Promote participation, fairness, and equity in policy and management. TO WHAT EXTENT ARE ECOSYSTEM PRIN- CIPLES, GOALS, AND POLICIES CURRENTLY APPLIED IN RESEARCH AND MANAGEMENT? The Panel considered a management system based on the ecosystem Principles, Goals, and Poli- cies, as a framework with which to evaluate the cur- rent application in U.S. marine fisheries management and research. This model was then compared to the current state of research and management. We conclude that NMFS and the Regional Fish- ery Management Councils (Councils) already con- sider and apply some of the Principles, Goals, and Policies outlined above, but they are not applied comprehensively or evenly across Council jurisdic- tions, NMFS Regions, or ecosystems. The fact that the Principles are not applied consistently in U.S. fisheries management and research should not be interpreted as reluctance or intransigence on the part of these entities to adopt ecosystem approaches. Rather, these agencies lack both a clear mandate and resources from Congress to carry out this more com- prehensive, but ultimately more sustainable ap- proach. Furthermore, the ecosystem-based manage- ment of fisheries is a relatively new concept and there are considerable gaps in knowledge and practice. HOW CAN WE EXPAND THE APPLICATION OF ECOSYSTEM PRINCIPLES, GOALS, AND POLICIES TO FISHERIES RESEARCH AND MANAGEMENT? Several practical measures can be implemented immediately to make U.S. fisheries management and research more consistent with the ecosystem Prin- ciples (see Summary of Recommendations). These measures comprise an incremental strategy for mov- ing toward ecosystem-based fisheries research and management. Councils should continue to use existing Fish- ery Management Plans (FMP) for single species or species complexes, but these should be amended to incorporate ecosystem approaches consistent with an overall Fisheries Ecosystem Plan (FEP). The FEP, to be developed for each major ecosystem under Council jurisdiction, is a mechanism for incorporat- ing the Principles, Goals, and Policies into the present regulatory structure. The objectives of FEPs are to: • Provide Council members with a clear descrip- tion and understanding of the fundamental physi- cal, biological, and human/institutional context of ecosystems within which fisheries are managed; • Direct how that information should be used in the context of FMPs; and • Set policies by which management options would be developed and implemented. Fisheries management based on the ecosystem Principles, Goals, and Policies must be supported by comprehensive research. Significant ecosystem research is now conducted by the National Oceanic and Atmospheric Administration (NOAA) and other agencies, as well as the academic community. This research is critical and must continue, but must ex- pand into several key areas. First, we must better understand the long-term dynamics of marine eco- systems and how they respond to human-induced change, particularly changes brought about by fish- ing. Second, we must develop governance systems which have ecosystem health and sustainability, rather than short-term economic gain, as their pri- mary goals. THE FUTURE OF ECOSYSTEM APPROACHES IN U.S. FISHERIES MANAGEMENT Fisheries scientists and managers are beginning to grasp the potential of ecosystem-based fishery management to improve the sustainability of fisher- ies resources. Given the depressed state of many U.S. fisheries, this awareness must be expanded and actions taken to implement this approach. Our man- agement recommendations and research actions pro- vide a pragmatic framework within which to apply the ecosystem Principles, Goals, and Policies. The success of this approach depends on full implemen- tation of measures already underway as a result of the passage of the Magnuson-Stevens Fishery Con- servation and Management Act (MSFCMA) (NMFS 1996), particularly the essential fish habitat (EFH) requirements and strengthened national standards. The recommendations contained in this report pro- vide the required next steps. EXECUTIVE SUMMARY While some of the recommended actions can start immediately, we believe that legislation is required to implement measures like the FEP. Given that leg- islative processes may require three to five years to enact the proposed regulations, we recommend in- terim actions by the Secretary of Commerce to de- velop demonstration FEPs and to encourage volun- tary adoption by management Councils of the Prin- ciples, Goals, and Policies proposed herein. We also are aware that these new tasks will require additional human and financial resources for full implementa- tion. The benefits of adopting ecosystem-based fish- ery management and research are more sustainable fisheries and marine ecosystems, as well as more economically-healthy coastal communities. We have identified the actions required to realize these ben- efits. We urge the Secretary and Congress to make those resources available. SUMMARY OF RECOMMENDATIONS Fisheries management and policy recommenda- tions are directed toward Congress for implementa- tion by NMFS and the Councils. Interim measures and research recommendations are directed toward the Secretary of Commerce for implementation by NMFS and other appropriate agencies. management must be to identify and bound the eco- system. Hydrography, bathymetry, productivity, and trophic structure must be considered; as well as how climate influences the physical, chemical, and bio- logical oceanography of the ecosystem; and how, in turn, the food web structure and dynamics are af- fected. Transfers across ecosystem boundaries should be noted. Within each identified ecosystem. Councils should use a zone-based management approach to designate geographic areas for prescribed uses. Such zones could include marine protected areas, areas particularly sensitive to gear impacts, and areas where fishing is known to negatively affect the trophic food web. 2. Develop a conceptual model of the food web. For each targeted species, there should be a cor- responding description of both predator and prey species at each life history stage over time. FEPs can then address the anticipated effects of the al- lowed harvest on predator-prey dynamics. 3. Describe the habitat needs of different life his- tory stages for all plants and animals that rep- resent the "significant food web" and how they are considered in conservation and man- agement measures. Develop a Fisheries Ecosystem Plan (FEP) Require each Council to develop an FEP for the ecosystem(s) under its jurisdiction. The FEP is an umbrella document containing information on the structure and function of the ecosystem in which fish- ing activities occur, so that managers can be aware of the effects their decisions have on the ecosystem, and the effects other components of the ecosystem may have on fisheries. Each FEP should require the Councils to take, at least, the following eight actions: 1. Delineate the geographic extent of the ecosystem(s) that occur(s) within Council au- thority, including characterization of the bio- logical, chemical, and physical dynamics of those ecosystems, and "zone" the area for alternative uses. The first step in using an ecosystem approach to Essential fish habitat (EFH) for target and non- target species at different life stages should be iden- tified and described. Using habitat and other eco- system information. Councils should develop zone- based management regimes, whereby geographic areas within an ecosystem would be reserved for prescribed uses. FEPs should identify existing and potential gear alternatives that would alleviate gear- induced damage to EFH, as well as restrict gears which have adverse affects. Further, FEPs should evaluate the use of harvest refugia as a management tool to satisfy habitat needs. 4. Calculate total removals — including inciden- tal mortality — and show how they relate to standing biomass, production, optimum yields, natural mortality, and trophic struc- ture. Total removals (i.e., reported landings, unre- ported landings, discards, and mortality to fish that come into contact with fishing gear but are not cap- ECOSYSTEM-BASED FISHERY MANAGEMENT tured) should be incorporated into qualitative food web and quantitative stock assessment models. These models will allow managers to reduce uncer- tainty, monitor ecosystem health and better predict relative abundance of species affected by the har- vest of target species. 5. Assess how uncertainty is characterized and what kind of buffers against uncertainty are included in conservation and management actions. Given the variability associated with ecosystems, managers should be cognizant of the high likelihood for unanticipated outcomes. Management should acknowledge and account for this uncertainty by de- veloping risk-averse management strategies that are flexible and adaptive. Councils and DOC have authority over a limited range of the human, institutional, and natural com- ponents of a marine ecosystem. It is important to recognize those components of the ecosystem over which fisheries managers have no direct control, and to develop strategies to address them in concert with appropriate international. Federal, State, Tribes, and local entities. Measures to Implement FEPs The following are general recommendations to ensure effective development and implementation of FEPs: 1 . Encourage the Councils to apply ecosystem Principles, Goals, and Policies to ongoing ac- tivities. 6. Develop indices of ecosystem health as tar- gets for management. Ecosystem health refers to a balanced, integrated, adaptive community of organisms having a species composition, diversity, and functional organization that has evolved naturally. Provided that a healthy state can be determined or inferred, management should strive to generate and maintain such a state in a given ecosystem. Inherent in this management strategy would be specific goals for the ecosystem, including a description of "unhealthy" states to be avoided. 7. Describe available long-term monitoring data and how they are used. In preparation for FEP implementation. Coun- cils should begin to apply the ecosystem Principles, Goals, and Policies to the conservation and manage- ment measures of existing and future FMPs. Three actions are particularly important; specifically, each FMP's conservation and management measures should: • Consider predator-prey interactions affected by fishing allowed under the FMP. • Consider bycatch taken during allowed fishing op- erations and the impacts such removals have on the affected species and the ecosystem as a whole, in terms of food web interactions and community structure. Changes to the ecosystem cannot be determined without long-term monitoring of biological indices and climate. Long-term monitoring of chemical, physical and biological characteristics will provide a better understanding of oceanic variability and how climate changes affect the abundance of commer- cially important species and their corresponding food webs. 8. Assess the ecological, human, and institu- tional elements of the ecosystem which most significantly affect fisheries, and are outside Council/Department of Commerce (DOC) au- thority. Included should be a strategy to ad- dress those influences in order to achieve both FMP and FEP objectives. • Minimize impacts of fisheries operations on EFH identified within the FEP. 2. Provide training to Council members and staff. To facilitate an ecosystem approach and to aid the development and implementation of FEPs, NMFS should provide all Council members with basic instruction in ecological principles. Further, training materials should be made available to the fishing industry, environmental organizations, and other interested parties. EXECUTIVE SUMMARY 3. Prepare guidelines for FEPs. Researcii Required to Support Management The Secretary of Commerce should charge NMFS and the Councils with estabhshing guidelines for FEP development, including an amendment pro- cess. NMFS and the Councils should conduct a de- liberative process — similar to the process of devel- oping National Standards Guidelines — to ensure that FEPs are realistic and adaptive. Require and provide support for NMFS and other appropriate agencies to initiate or continue research on three critical research themes which will provide the information necessary to support ecosystem- based fisheries management. These themes are: 1. Determine the ecosystem effects of fishing. 4. Develop demonstration FEPs. While encouraging all Councils to develop framework FEPs, the Secretary of Commerce should designate a Council or Councils to develop a dem- onstration FEP, as a model to facilitate rapid imple- mentation of the full FEP when required in MSFCMA reauthorization. 5. Provide oversight to ensure development of and compliance with FEPs. To ensure compliance with the development of FEPs, the Secretary of Commerce should establish a review panel for FEP implementation oversight. Implicit in this action is the establishment of a time- table for development of a draft FEP, its review by the panel, and any necessary revisions before the draft FEP becomes a basis for policy. 6. Enact legislation requiring FEPs. To provide NMFS and the Councils with the mandated responsibility of designing and implement- ing FEPs, Congress should require full FEP imple- mentation in the next reauthorization of the MSFCMA. Fishing affects target species, non-target species, habitat, and potentially marine ecosystems as a whole. A directed program must be initiated to de- termine all effects of fishing on marine ecosystems. 2. Monitor trends and dynamics in marine eco- systems (ECOWATCH). In order to detect, understand, and react appro- priately to ecosystem changes, a broad-scale eco- system research and monitoring program must be undertaken based on the best available technology. We refer to this program as "ECOWATCH" because it will enable scientists and managers to observe eco- system changes in a comprehensive manner. 3. Explore ecosystem-based approaches to gov- ernance. Many of today's fisheries problems stem from governance systems which create incentives that are incompatible with, or inimical to, ecosystem-level Goals (e.g., health and sustainability ). Alternate gov- ernance systems must be identified which provide fishermen and others with incentives to consider the health and sustainability of the ecosystem as primary goals. ECOSYSTEM-BASED FISHERY MANAGEMENT SECTION ONE: INTRODUCTION The National Marine Fisheries Service (NMFS) was charged by Congress to establish an Ecosystem Principles Advisory Panel (Panel) to identify eco- system principles, evaluate how those principles are currently used in fishery management and research, and then to recommend measures that would expand their use in fishery management and research. Our Charter (Appendix A) describes the rationale for our effort and provides the charge to this Panel. Here we outline our views of the historical developments and current issues leading to this charge. We lay out a conceptual framework that includes management actions and research on marine resources and fish- eries in an ecosystem context. THE PROBLEM The world's oceans are at or near maximum sus- tainable fishery yields. The number of overexploited stocks increased by 2.5 times between 1980 and 1 990 (Alverson and Larkin 1994). Much of the global sustained yield is being accomplished by increased fishing for species at progressively lower trophic levels (Pauly et al. 1998). The prospect of increas- ing total sustained yield is unlikely (Pauly and Christensen 1995). Although fisheries provide di- rect or indirect employment to about 200 million people (Garcia and Newton 1997), overfishing is the most commonly observed result of fishery develop- ment. The consequences of overharvesting are ex- pressed in social, economic, cultural, and ecologi- cal changes. The ecological consequences of over- fishing often are undocumented and may be poorly known or overlooked. identified as overfished" under the new definition of overfishing in the Magnuson-Stevens Fisheries Conservation and Management Act (MSFCMA). While there are some encouraging recoveries (e.g., striped bass in the Atlantic and Pacific sardine), record-setting yields (e.g., Alaska salmon), and man- agement successes (e.g.. Pacific halibut), those cases are the exceptions rather than the rule. As in the global case, we should be concerned that overfish- ing will be a common consequence for most fisher- ies (Ludwig et al. 1993, Mooney 1998), although this need not be the case (Rosenberg et al. 1993). This issue is urgent because the current harvest levels are high and because new fisheries will rise, be fully capitalized, and reach unsustainable levels of catch levels before the management process can establish effective constraints. That, unfortunately, is the too-common lesson of history (Ludwig et al. 1993). In many cases, the ecological correlates of changing fish populations could have served as evi- dence of intensified exploitation effects. Frequently, the advent of a fishery and implementation of catch restrictions have unknown ecological consequences. Too often, we learn about ecological consequences after the fact, because we do not consider them in our decision-making, nor do we monitor ecosystem changes due to increased exploitation. Those les- sons are not unique to fisheries. Many Federal, re- gional and State resource management agencies are now moving toward or considering an ecosystem ap- proach in their attempt to provide a holistic frame- work for resource management. Fisheries must do so as well (Langton and Haedrich 1997). Since 1990, annual harvests by U.S. fleets have been slightly in excess of 4.5 million metric tons, with nearly half of that coming from two fisheries — menhaden and Alaska pollock. In its annual report to Congress on the status of the fisheries of the U. S., NMFS states that of the 727 managed stocks in the United States, 86 are overfished, 10 are ap- proaching overfished status, and 183 are not over- fished (NMFS 1997). This leaves 448 stocks, for which the status is virtually unknown. NMFS ( 1 997 ) also indicates that "additional stocks will likely be FISHERIES IN AN ECOSYSTEM CONTEXT Much of the foundation of fisheries science pro- vides a basis for determining maximum yields so that fishing can safely remove surplus production (Hilborn and Walters 1992). However, when fish- ing is examined in an ecosystem context, the ratio- nale for harvesting surplus production is unclear. Marine ecosystems are effective at capturing energy. ECOSYSTEM-BASED FISHERY MANAGEMENT cycling nutrients, and producing biomass. Very little, if any of this biomass, is truly "surplus" to an eco- system; before the advent of fisheries, it was recycled within the ecosystem. Consequently, our societal decision to harvest fish induces ecological changes among competitors, prey, and predators as the sys- tem responds to fishing and the trophically-induced changes fishing causes in ecosystems. These changes affect future levels of surplus production of the har- vested population, including the possibility that there may be none. We understand that fisheries must continue, be- cause they provide food, desirable social and eco- nomic benefits, and be- cause the cultural tradi- tions of fishing are highly valued. How- ever, we also understand that overutilized fisher- ies are a serious threat to those traditions and benefits (National Re- search Council 1999). Conflict thus develops when management agencies (e.g., NMFS, Regional Fishery Man- agement Councils, etc.) seek to implement sus- tainable yield policies for open-access re- sources, when fishery effects extend to animals pro- tected by our Endangered Species Act or Marine Mammal Protection Act, and, most recently, when conservation and management interests assert that the burden of proof should be placed on the fishing industry (i.e., to demonstrate that exploitation does not produce large-scale and long-term ecological changes) (Dayton 1998). Finding the balance be- tween competing interests is a difficult challenge, and each fishery will have its unique solutions. On the Federal level, NMFS will be expected to pro- vide the ecological insights that are essential for long- term protection offish stocks and their ecosystems. Decisions regarding fishing practices derive from our social, economic, political, and cultural context, and only secondarily from the ecological context that supports fisheries (Mooney 1998). A holistic view requires that we recognize fishery management and exploitation as a real and integral part of the marine ecosystem (Langton and Haedrich 1997). Because Nature has limits If nature is a shifting mosaic or in essentially con- tinuous flux, then it may be wrong to conclude tliat what- ever societies clioose to do in or to the natural world is fine. The question can be stated as, "If the state of na- ture is flux, then is any luiman-generated cimnge okay? " ... The answer to this question is a resounding "No!" ... Human-generated changes must be constrained because nature lias functioned, historical, and evolutionary lim- its. Nature has a range of ways to be, but there is a limit to those ways, and therefore, human changes must be within those limits (Pickett et al. 1992). fishing actively removes a percentage of one or sev- eral species, it can affect the predators and prey of those species, their physical habitat, and it can change the growth and mortality rates of target and non-tar- get species alike. In short, fishing can and is likely to alter the structure and function of marine ecosys- tems (Dayton 1998, Pauly et al. 1998). Humans are at the top of the global marine food chain. We thus have the obligation and opportunity to make choices to affect the marine environment positively. While fishing has a long history, it is a relatively new force in the scales of evolutionary time. Fish- ing is typically a species-selective and size-selec- tive agent of mortality and, therefore, is unlike the natural causes of mortality. Most of the fish removed by fish- ing activities are in the middle or near the top of their respective food webs. Fishing can be viewed as a keystone predator; the ecologi- cal effects of fishing are therefore substan- tially greater and more complex than simply the biomass removed. Thus, we should expect that substantial changes have or could occur in those ecosystems due to fish- ing. We have witnessed changes in the landscape around us with the advent of technology evolved from the axe and the plow. We should expect equally profound ecological changes from modern, large- scale uses of the hook and net. MANAGING FISHERIES IN AN ECOSYSTEM CONTEXT Ecosystem-based fisheries management does not require that we understand all things about all com- ponents of the ecosystem. We know that the tradi- tional single-species approach of fisheries manage- ment is tractable, but we also know that it may not be sufficient. We know that an ecosystem perspec- tive is desirable, but it is complex and unpredict- able. There simply is not enough money, time, or talent to develop a synthetic and completely informed view of how fisheries operate in an ecosystem con- 10 SECTION ONE: INTRODUCTION text. There will always be unmeasured entities, ran- dom effects, and substantial uncertainties, but these are not acceptable excuses to delay implementing an ecosystem-based management strategy. Each fishery and each ecosystem is unique and yet, in all cases, we are confronted with four funda- mental problems: • We do not have a complete understanding of the ecological system that produces and supports fishes. We cannot forecast weather or climate and their effects on ecosystems. Systems evolve over time and knowing how the system works does not nec- essarily mean that an ecosystem would re- spond predictably to future changes in weather, climate, or fisheries. Our institutions are not configured to manage at the eco- system scale. Fish and the fisheries that pursue them are not easily aligned with our political and ju- risdictional bound- aries. These constraints are not unique to fisheries, they confront all attempts to manage natural resources in an ecosystem context. We know that the removal of one species can and does affect others, but rarely have we developed management plans that ad- equately account for those direct and indirect effects. We know that ecosystems have a limited carrying capacity that results in bounds on fish yields. We know that habitat loss contributes to declines in spe- cies abundance, but too often we only regulate catch, gear, or effort for one target species as a way to com- pensate for habitat loss and its effects on other spe- cies. We know that major, unexpected events (e.g.. Legal Authorities for Ecosystem Management of Fisheries The Magnuson-Stevens Fishery Conservation and Management Act allows fishery managers to consider ecosystems in setting tncmagement objectives. National Standard 1 requires consen^ation and management mea- sures to "prevent overfishing white achieving, on a con- tinuing basis, the optimum yield from each fishery " {Sec. 30l(a)(I )). The "optimum " yield is defined as provid- ing "the greatest overall benefit to the Nation, particu- larly with respect to food production and recreational opportunities, and taking into account the protection of marine ecosystems" (Sec. 3(28)(A)). Moreover the op- timum yield is prescribed as "the maximum sustainable yield from each fisheiy, as reduced by any relevant eco- nomic, social or ecological fiictor" (Sec. 3(28 )(B)). In addition, tlie Act states as one of its purposes "to pro- mote the protection of essential fish habitat" (Sec. 2(b)(7)). To the extent that ecosystems are not being adecjuately considered in FMPs. it is not because of a lack of statutory authority so much as it is a lack of di- rection about wliat information is required and how it should be put into operation. El Nifio) can alter ecosystem processes, thus affect- ing species targeted by fisheries, but we have no method for integrating these events into our assess- ments of target species population trends (Mantua et al. 1997. Francis et al. 1998). What are the potential gains of implementing an ecosystem approach to management, and how do we develop a holistic view that is both sufficient and tractable? In this report, we develop a strategy for implementing ecosystem-based management. First, we develop a conceptual model that sets fisheries in the con- text of what we know about ecosystem theory (which is provided in the section on Ecosys- tem Principles, Goals, and Policies). Second, we provide a brief as- sessment of the extent to which ecosystem principles, goals, and policies are applied in U.S. fisheries research and management (Cur- rent Applications of the Principles, Goals, and Policies). Third, we offer a series of spe- cific recommendations for applying these prin- ciples to the operational context of NMFS, the Regional Fishery Man- agement Councils (Councils), their ad- ministrative structure and their management activi- ties (Recommendations for Implementing the Ecosystem Principles, Goals, and Policies in U.S. Fisheries Conservation, Management, and Re- search). Finally, we recommend a comprehensive research program to provide the ecological and gov- ernance underpinnings for ecosystem-based fishery management. Taken as a whole, the report presents our best advice about innovative approaches that can help set fisheries in an ecosystem context. Ecosystem-based management is an important new challenge. We expect that NMFS, Council managers, and scientists 11 ECOSYSTEM-BASED FISHERY MANAGEMENT will develop creative ways to help meet that chal- agement. It will also demand a strong poHtical will lenge. But these new approaches cannot substitute expressed through Congress. NMFS and the Coun- for compliance with existing mandates. Ecosystem- cils — one based on a broader appreciation of the based management will require re-evaluation of the ecosystem context within which we prosecute our institutional structure necessary for effective man- fisheries (Hutchings et al. 1997). 12 SECTION TWO: ECOSYSTEM PRINCIPLES, GOALS, AND POLICIES There are two requirements for managing human interactions with marine ecosystems. One is to de- velop an understanding of the basic characteristics and principles of these ecosystems — what patterns they exhibit and how they function in space and time. The second is to develop an ability to manage ac- tivities that impact marine ecosystems, consistent with both their basic principles and with societal goals concerning the kinds of behavior we would like ecosystems to exhibit (i.e., health and sustainability). This section lists eight basic ecosystem principles (Principles) and their parallels in human systems that are part of marine ecosystems. A discussion of soci- etal goals (Goals) for ecosystem-based management follows. Finally, a list of general management poH- cies (Policies) to achieve the Goals is provided. BASIC ECOSYSTEM PRINCIPLES Marine ecosystems are complex, adaptive sys- tems composed of interconnected groups of living organisms and their habitats. Living organisms are constantly adapting and evolving to their environ- ment (both to the physical environment, which var- ies on multiple scales, and to other living organisms with which they co-exist); this evolution leads to complex, sometimes chaotic dynamics. Marine ecosystems are generally extensive and open. Their fluid environments are subject to vari- ability in both local and remote inputs of energy (a consequence of physics operating on many spatial and temporal scales) which may dominate such sys- tems. Highly variable and chaotic dynamics of liv- ing resources are often observed as well. Today, humans are a major component in most ecosystems. The human component of the ecosys- tem includes the humans themselves, their artifacts and manufactured goods (economies), and their in- stitutions and cultures. The human imposition of fishing mortality, at rates often higher than natural mortality, can have major impacts not only on tar- geted species but on the ecosystem itself. The following eight Principles have analogs in both the human and nonhuman aspect of ecosystems: 1. The ability to predict ecosystem behavior is limited. Uncertainty and indeterminacy are fundamen- tal characteristics of the dynamics of com- plex adaptive systems. Predicting the behav- iors of these systems cannot be done with absolute certainty, regardless of the amount of scientific effort invested. We can, however, learn the boundaries of expected behavior and improve our understanding of the under- lying dynamics. Thus, while ecosystems are neither totally predictable nor totally unpre- dictable, they can be managed within the lim- its of their predictability. Properties characterizing marine ecosystems may vary within wide bounds on decadal and longer time scales (Fig. 1). For example. El Nino events and decadal climate changes may displace species, re- structure communities, and alter overall productiv- ity in broad oceanic areas. Other phenomena, some- times operating on smaller time scales, may precipi- tate regime shifts characterized by major fluctuations in constituent species (Steele 1996), but our ability to predict such events is only now evolving (Langton et al. 1996) and will always be shrouded in a degree of uncertainty. Nevertheless, management policies can be guided by the broad understanding we pos- sess of marine ecosystem boundaries and produc- tion potential limits. The ability to predict human behavior in fishery systems is also limited, but evolving. Many fisher- men pass through rounds of fishing in regular an- nual patterns, markets respond in predictable ways to price changes, and fishermen often have predict- able responses to policy proposals or regulatory changes. Fisheries systems respond to global mar- ts ECOSYSTEM-BASED FISHERY MANAGEMENT SCALES OF PHYSICAL VARIABILITY AFFECTING MARINE FISHERIES POPULATIONS GLOBAL BASIN-WIDE u 09 REGIONAL LOCAL Atmospheric conditions (wind/pressure) affect direction and speed of regional currents Currents and winds affect location, duration and intensity of upwelling in coastal areas T 37N ■ Shifts occur in upper atmosphere circulation creating ocean temperature anomalies (e.g. El Nino) Ocean-Atmosphere Energy Exchanges DAILY TO MONTHLY SEASONAL INTERANNUAL DECADAL MULTIDECADAL EFFECTS ON FISHERIES: SHORT-TERM FISHING SUCCESS FISH REPRODUCTION FISH YEAR-CLASS SUCCESS REGIONAL-SCALE FISH POPULATION FLUCTUATIONS TIME SCALE Figure 1 . Scales of physical variability affecting marine resources. Variability in marine ecosystems is linked to variability in the physical environment on a continuum of time and space scales. We are often constrained to work on scales at which data are available, and long-term monitoring must be carefully designed to address appropriate scales. Figure courtesy of NMFS Pacific Fisheries Environmental Laboratory. ket trends and economic changes, social preferences and philosophies. The ability to describe, explain, and predict these human behaviors, although the be- haviors vary according to circumstance, is increas- ing with the growing body of social scientific data and information on fishery systems. 2. Ecosystems have real thresholds and limits which, when exceeded, can effect major sys- tem restructuring (Holling and Meffe 1996). Ecosystems are finite and exhaustible, but they usually have a high buffering capacity and are fairly resilient to stress. Often, as stress is applied to an ecosystem, its struc- ture and behavior may at first not change no- ticeably. Only after a critical threshold is passed does the system begin to deteriorate rapidly. Because there is little initial change in behavior with increasing stress, these thresholds are very difficult to predict. The nonlinear dynamics which cause this kind of behavior are a basic characteristic of ecosys- tems. 14 SECTION TWO: ECOSYSTEM PRINCIPLES, GOALS. AND POLICIES The concepts of limits and thresholds have been misused in single-species fishery management in the sense that they have been viewed as targets for fish catches rather than levels to be avoided. Because single-species management has prevailed, limits and thresholds rarely have been applied in a broader eco- system context. Limits in fisheries management of- ten have been biological reference points such as pre- scribed fishing mortality rates or yields, that are set without concern for other components in the eco- system. Many limits are in fact thresholds that, when exceeded, challenge the resilience of the managed stock and associated species. Experience has shown that some past target levels used by managers (e.g., maximum sustainable yield) ultimately lead to stock declines or damage to ecological communities be- cause they are too close to critical thresholds (Caddy and Mahon 1995). Thresholds are to be avoided to maintain resilience at the species and community levels. Fishery targets should be set conservatively, well below the limits and critical thresholds that com- promise the productive potential and stability of the ecosystem. Limits and thresholds of non-targeted organisms have only recently been considered through mandates of the Marine Mammal Protec- tion Act, the Endangered Species Act, and in the new MSFCMA definitions of overfishing levels and pro- visions for bycatch and essential fish habitat. Human systems (fishermen, their communities, and fishery management systems) are both resilient and generally resistant to change. Thresholds of prof- itability, tolerance of regulatory conditions, and risk or uncertainty-induced stress on fishery-dependent human communities are real. Thresholds must be determined through both constituent advice and in- dependent research on individual and group re- sponses to stress. Identification of reference points for the limits of human resilience may be possible. non-target organisms may radically alter communi- ties and ecosystems. It is too soon to know whether heavily fished systems, such as Georges Bank, will return to their previous states when fishing effort is relaxed (Fogarty and Murawski 1998). Fisheries scientists and managers have demonstrated an abid- ing faith in the ability of fish stocks to compensate for fishing effects by increasing their level of pro- ductivity. Implicitly, that faith is extended to eco- systems which support exploited stocks. Up to a point, recoveries are possible. In some coastal eco- systems, however, resilience and limits have been exceeded, often by the combined effects of habitat destruction and fishing, and it is doubtful if they will return to their original condition. Changes in ecosystems may permanently alter human behaviors. When a fisherman goes out of business, when an annual season of fishing is dis- turbed, or when market flow is interrupted, it is of- ten not possible to reestablish the former business, pattern, or market. Some aspects of human systems and behavior can be reestablished given enough time and attention, whereas changes in natural compo- nents of ecosystems are typically more enduring. In contrast, policy and management systems are con- tinually subject to change and reversal. 4. Diversity is important to ecosystem function- ing. The diversity of components at the individual, species, and landscapes scales strongly af- fects ecosystem behavior. Although the over- all productivity of ecosystems may not change significantly when particular species are added or removed, their stability and re- silience may be affected. 3. Once thresholds and limits have been ex- ceeded, changes can be irreversible. When an ecosystem is radically altered, it may never return to its original condition, even after the stress is removed. This phenomenon is common in many complex, adaptive sys- tems. It is probable that some estuaries, coral reefs (Hughes 1994), and mangrove ecosystems have been irreversibly altered by fishing, aquaculture, and other habitat-destructive activities. Farther offshore, ef- fects of fishing itself on abundances of target and Long-term consequences of diversity losses due to overfishing or poor fishing practices in marine systems are largely unknown. It is clear, however, that the economic value of specific components of catch change dramatically as some stocks are over- fished, to be replaced in the ecosystem by lower- valued species (Deimling and Liss 1994, Fogarty and Murawski 1998). At the ecosystem level, drastic al- terations of diversity certainly have occurred, and biological productivity has been redirected to alter- native species, but it is not clear that these ecosys- tems are less productive or less efficient. However, such ecosystems are often valued less; witness the loss of tourist revenue in areas that have suffered 15 ECOSYSTEM-BASED FISHERY MANAGEMENT damage to coral reef systems. It is prudent to pre- sume that changes in biodiversity will decrease re- siliency of species, communities and ecosystems, especially with perturbations that occur over long time scales (Boehlert 1996). This principle also applies to the human element. An economy with more than one sector, a commu- nity with more than one industry, a fishing family with more than one income from different sources, or an industry large enough to foster technological innovation, are all aspects of the strength in diver- sity found in human society. Communities which lose such diversity are more susceptible to stress and unexpected sources of change. 5. Multiple scales interact within and among ecosystems. Ecosystems cannot be understood from the perspective of a single time, space, or com- plexity scale. At minimum, both the next larger scale and the next lower scale of inter- est must be considered when effects of per- turbations are analyzed. Consequences of perturbations at one scale in marine systems may be magnified at larger and smaller scales (Langton et al. 1995). For example, destruction of a species' spawning habitat — typically a small fraction of its range — may translate into major impacts on species associations and trophic interactions in the broader feeding areas of recruited fish. Likewise, effects of fishing on a broad ecosys- tem scale may have profound impacts on compo- nents of ecosystems far removed in space and time — scientists are investigating the relationship between pollock fishing and the general decline of Steller sea lion populations in the eastern Bering Sea and Gulf of Alaska. Seemingly small human perturbations, applied at a point in time or in one part of a marine ecosystem, may have unforeseen impacts because of the open nature and fluid environment that char- acterize marine ecosystems. These features elevate the probability that a stress applied at one scale will be transmitted and may have unforeseen effects at other scales in the ecosystem. Human impacts on ecosystems cannot be under- stood from the perspective of a single time, space, or complexity scale. A fishing community is sub- ject to perturbations both from its own members and from outside forces. Fishery systems in one loca- tion are subject to environmental, social, economic, and regulatory forces far removed in time and space, especially with respect to markets. 6. Components of ecosystems are linked. The components within ecosystems are linked by flows of material, energy, and infor- mation in complex patterns. Critical linkages in marine ecosystems are sus- tained by key predator-prey relationships. Large, long-lived predators and small, short-lived prey (e.g., forage fishes) both contribute in major ways to ma- rine fish catches. Heavy fishing may precipitate spe- cies replacements, both at lower trophic levels (e.g., sand lance replacing herring and vice-versa) and at upper trophic levels (e.g., sharks and rays replacing Atlantic cod) (Fogarty and Murawski 1998). Loss from ecosystems of large and long-lived predators is of particular concern because they potentially ex- ercise top-down control of processes at lower trophic levels. Global data sets have indicated that the mean trophic level of fish caught declined significantly from 1950-1994 (Pauly et al. 1998). Fishing down food webs (i.e., fishing at lower trophic levels) dis- rupts natural predator-prey relationships and may lead first to increasing catches, but then to stagnat- ing or declining yields. Disruption of ecosystem linkages clearly may have resounding impacts on human economies and, in the worst cases, ecosystem stability and produc- tivity are compromised. Components of human sys- tems are linked by flows of material, energy, and information. The collapse of a market may drasti- cally change fishing behavior. A technological in- novation or entry of a new segment of a fishing fleet may cause far-reaching changes in dependent hu- man communities. 7. Ecosystem boundaries are open. Ecosystems are far from equilibrium and can- not be adequately understood without knowl- edge of their boundary conditions, energy flows, and internal cycling of nutrients and other materials. Environmental variability can alter spatial boundaries and energy inputs to ecosystems. Productive potential of marine ecosystems is es- 16 SECTION TWO: ECOSYSTEM PRINCIPLES, GOALS, AND POLICIES pecially sensitive to environmental variability over a spectrum of temporal and spatial scales. The un- bounded structure of marine communities provides the backdrop for the high (relative to terrestrial) vari- ability that is observed (Steele 1991). Boundaries of ecosystems, or productive regions, shift with weather and longer-term climate change. Species abundances and distributions vary in accord with annual to decadal shifts in ocean features (e.g., Pearcy and Schoener 1987, Polovina et al. 1995, Roemmich and McGowan 1 995, Francis et al. 1 998, McGowan et al. 1998). In open systems, local heavy fishing in combination with major changes in ocean conditions (e.g.. El Niiio) can lead to fishery col- lapses and associated shifts in the partitioning of energy or biomass among trophic levels (e.g., Walsh 1981, Barber and Chavez 1983). Human behavioral systems are also subject to variability over a spectrum of temporal and spatial scales, and cannot be understood without knowledge of their boundary conditions. Certain components of human systems (people) are closely related and interact regularly over time; others are only sporadi- cally in contact and interact in cyclical or irregular patterns. The more intermittent or sporadic the con- tact or interaction, the less stable the human system (Axelrod 1984). 8. Ecosystems change with time. Ecosystems change with time in response to natural and anthropogenic influences. Dif- ferent components of ecosystems change at different rates and can influence the overall structure of the ecosystem itself and affect the services provided to society in the form of fish catch, income and employment. Marine ecosystems experience directional changes. Shifts in climate are responsible for many such changes, but the role of biological interactions in the absence of human influence are largely un- known. Dramatic changes in coastal and estuarine ecosystems, attributable to long-term geological and erosional processes are easily observed (e.g., Chesa- peake Bay, see Mountford 1996). Anthropogenic changes are all too common, especially in neritic and estuarine ecosystems or enclosed seas (e.g., San Francisco Bay (Nichols et al. 1986), Great Lakes, Black Sea, Aral Sea, Chesapeake Bay). Species in- troductions, excess nutrient loading, damming of tributaries, poor stewardship of bordering forests, bad agricultural practices, and poorly-managed fisher- ies are examples of factors that cause change. Rapid advances in fishing technologies (e.g., vessel power, navigation, sensing-locating, and harvest efficiency), the propensity for fisheries to selectively remove species, failure to control bycatch, and unintended damage to the physical structure of ecosystems, have changed the character of heavily fished ecosystems (e.g., Georges Bank) (Fogarty and Murawski 1998). Selective fishing, that often targets long-lived preda- tors, can have cascading effects on community struc- ture (Marten 1979, Laws 1977), while heavy indus- trial fishing on forage species may have unintended impacts on top predators, especially those (e.g., ma- rine mammals) unable to adapt quickly to changes in the forage base. Removal of large whales through past whaling practices, likewise, may have linger- ing effects on the nature of ecosystem structures to- day (National Research Council 1996). Deteriora- tion of coastal ecosystems may also generate active attempts at remediation or enhancement through aquaculture and other means (Morikawa 1994), which can also generate pollution and wastes (Wu 1995). Human activities dependent on ecosystems may change in response to environmental change and changes induced by fishing and other activities. In the short run, these impacts may be considered the normal consequences of a highly variable activity. However, humans adapt to long-term changes in composition of fisheries by stopping fishing or shift- ing effort to other species: changes which may pro- duce adverse impacts. In addition, changes in per- ception, values, preferences, patterns of use, and ac- cumulation of knowledge or expertise may cause changes over time in the ways humans interact within ecosystems. Human components of ecosystems (es- pecially technology and institutions) can change rap- idly in ways that outstrip the capacity for change of other ecosystem components. Communities may continue to grow and consumption rates increase, for example, yet the capacity of the seas to increase yields of living marine resources is limited. Thus, fishery management policies must be prepared to take into account these factors. BROADENING SOCIETAL GOALS FOR ECOSYSTEMS Traditionally, societal goals have emphasized benefits to humans resulting from extractive uses of 17 ECOSYSTEM-BASED FISHERY MANAGEMENT ecosystem components. For example, fishery man- agement has typically had revenues, employment, recreational fishing opportunities, and/or mainte- nance of traditional lifestyles as explicit or implicit goals. From an ecosystem perspective, these goals need to be broadened to include concepts of health and sustainability (Lubchenco et al. 1991, National Research Council 1999). Ecosystem health is the capability of an ecosystem to support and maintain a balanced, integrated, adaptive community of or- ganisms having a species composition, diversity and functional organization comparable to that of the natural habitat of the region (Sparks 1995). This concept is also referred to as biotic integrity, which is defined as a system's wholeness, including the presence of all appropriate elements and occurrence of all processes at appropriate rates (Angermeier and Karr 1994, Angermeier 1997). While the concept of health applied to marine ecosystems is relatively new and untested, it has become a guiding framework in several areas, including forest ecosystems (Kolb et al. 1994), agroecosystems (Gallopin 1995), desert ecosystems (Whitford 1995), and others (Rapport etal. 1995). A healthy ecosystem provides certain ecosystem goods and services, such as food, fiber, the capacity for assimilating and recycling wastes, potable wa- ter, clean air, etc. (International Society for Ecosys- tem Health, 1998). How do we extract from, and otherwise utilize ecosystems, while maintaining their health and the array of non-use services that they also provide (Costanza et al. 1997) into the indefi- nite future? The challenge to scientists and managers is to develop useful, quantitative measures of ecosystem health which can guide management. What level of fishing, for example, can a "healthy" ecosystem sus- tain? How can vigor and resilience be expressed quantitatively so that managers can maintain them within healthy limits? These are difficult questions which will not be answered in their entirety in the foreseeable future, but incremental implementation of ecosystem-based fisheries management will be- gin to identify ecosystem variables (or indicators) that are unacceptable. These could be used to guide management away from unhealthy ecosystem states. GENERAL ECOSYSTEM- BASED MANAGEMENT POLICIES Ecosystem Principles to achieve societal Goals must be implemented through ecosystem-based man- agement Policies. There are three overriding aspects of the Principles that are taken into account in the six Policies discussed below. These are the exhaust- ibility of ecosystems (reflected in Principles 2 and 3), uncertainty about ecosystems (reflected in Prin- ciples 1, 2, 4, and 8), and the role of humans within ecosystems (reflected in all of the Principles). The exhaustibility of the ecosystem requires a policy to change the burden of proof (Policy 1). Both the ex- haustibility of ecosystems and uncertainty about eco- systems require policies to manage by a precaution- ary approach (Policy 2) and to "purchase insurance" (Policy 3) against adverse ecosystem impacts. Un- certainty about ecosystems also dictates that there is learning from management experiences (Policy 4). The role of humans within ecosystems requires poli- cies to make incentives for human behavior consis- tent with societal goals for ecosystems (Policy 5). Acceptance and effective implementation of the poli- cies and management is served by promoting par- ticipation, fairness and equity (Policy 6). Each of the Policies is discussed below. 1 . Change the burden of proof. We live in a world where humans are an im- portant component of almost all ecosystems. Thus, it is reasonable to assume that human activities will impact ecosystems. The modus operandi for fisheries management should change from the traditional mode of restrict- ing fishing activity only after it has demon- strated an unacceptable impact, to a future mode of only allowing fishing activity that can be reasonably expected to operate without unacceptable impacts. To date, almost any type of fishing activity has been allowed until problems arise and regulations are established to solve them. Decision makers have to be convinced that management restrictions are needed. As W. F. Thompson (1919) wrote ". . . proof that seeks to change the way of commerce and sport must be overwhelming." Several authors have ar- gued that a change is needed in this "burden of proof ' (Sissenwine 1987, Mangel etal. 1996, Dayton 1998). The key elements of the change are: 1 ) that future fishing activity should be allowed, if and only if it is 18 SECTION TWO: ECOSYSTEM PRINCIPLES, GOALS, AND POLICIES explicitly provided for by fishing regulations which take into account risk and uncertainty and are pro- mulgated to protect all elements of the ecosystem, and 2) that to a substantial degree the responsibility for providing the information and other support (e.g., the cost of management) necessary to manage fish- eries in a sustainable manner, lies with participants in the fishery. The first part of the change is analogous to chang- ing the "null" hypothesis from "marine fisheries are inexhaustible" (Huxley 1883), to today's reality that marine fisheries will usually evolve to a state of over- fishing unless they are carefully managed (Garcia and Newton 1997). The second element of the change makes clear that the direct beneficiaries from fishing should accept a greater share of the burden (i.e., costs) of fishery management. The standard of proof associated with the change (i.e., how much certainty is needed before a fishing activity is al- lowed) should be commensurate with the severity of the risk of a mistake. Applying the proper stan- dard of proof is implicitly an element of the precau- tionary approach (see Policy 2). In practice, changing the burden of proof will mean that, when the effects of fishing on either the target fish population, associated species, or the eco- system are poorly known (relative to the severity of the potential outcome), fishery managers should not expand existing fisheries by increasing allowable catch levels or permitting the introduction of new effort and should not promote or develop new fish- eries for so-called "underutilized species." 2. Apply the precautionary approach. The precautionary approach is a l»* •**^'>* Ml ** -V. ■"A* .') ■ •# ^ . J **1 t t1 It -f1 'of l^^.^k^-y ■:^'^ff'^ * 1 > ■« ■ p- 2 'J'A Iwy.tjl, |- 4> 1^ ^4s "^ i ■«• *# KiA" -*,: u PENN STATE UNIVERSITY LIBRARIES ADDDDMb313D7E