Dynamic Changes in Marine Ecosystems: Fishing, Food Webs, and Future Options (2006)

Recent scientific literature has raised many concerns about whether fisheries have caused more extensive changes to marine populations and ecosystems than previously realized or predicted. Due to its extractive nature, fishing reduces stocks of harvested species. However, in many cases, stocks have been exploited far beyond management targets, ultimately reducing the potential productivity of the fishery. In addition, new analyses indicate that the abundance and composition of non-targeted organisms in marine ecosystems are radically changing as a result of fishing pressure expressed through food-web interactions.

Several scientific papers suggest that populations of high-trophic-level fishes have been severely depleted and that fishing has fundamentally altered the structure of marine ecosystems in many locations. But the conclusions drawn in these scientific papers often have been controversial. Subsequent articles have disputed the findings of these papers, and others have disputed the implications (or the broad application) of the conclusions presented, while still others continue to provide additional analyses. Arguments on all sides acknowledge the paucity of fishery-independent data as a major roadblock to properly analyzing the current state of fisheries and ecosystems. Instead, the analyses rely on the more readily available landing and catch statistics. These fishery-dependent data are subject to various interpretations because fisheries landings change in response to many factors other than the abundance of the fished stocks (e.g., markets, management regulations, fishing methods, technology, and climate).

While the fisheries science community continues to analyze and debate these issues, several of these publications have been widely publicized. This has increased public awareness and raised concern that fisheries resources are not being effectively

managed, including the impacts of fishing on non-target resources and habitat. In response to this growing concern, the National Oceanic and Atmospheric Administration asked The National Academies’ Ocean Studies Board to form a committee of experts to review recent scientific reports and weigh the collective evidence for fisheries-induced changes to the dynamics of marine ecosystems. The committee was asked to discuss the relevance of these scientific findings for U.S. fisheries management, to identify areas for future research and analysis, and to characterize the stewardship implications for living marine resources. To help accomplish these tasks, the committee met publicly three times to hear presentations on relevant subjects ranging from fisheries biology and fisheries governance mechanisms to current modeling and analysis techniques, among others.

Fishing can alter a wide range of biological interactions, causing changes in predator-prey relationships, cascading effects mediated through food-web interactions, and the loss or degradation of essential habitats. These impacts, along with natural fluctuations in the physical state of the ocean, can interact to intensify fishing impacts beyond targeted species. Fishing is also generally size and species selective, potentially changing the genetic structure and age composition of fished stocks, as well as decreasing the diversity of marine communities. Examples of all these effects have been documented. Although some changes are expected outcomes of management actions, in many instances the measured effects are quantitatively and qualitatively more severe than anticipated by management.

Declines in stock abundance have been measured for many species throughout the world’s oceans, but the extent and severity of these declines differ across stocks and geographical areas. Changes to food-web interactions are expected because fisheries reduce the abundance of one or more components of the food web, simultaneously altering the interactions among species and the strength of these interactions. Direct predator-prey relationships have changed—either releasing lower trophic levels from predation or reducing the availability of prey for higher-level predators—and these effects may spread to successive trophic levels up and down the food web. Such cascading effects are often unforeseen and management actions frequently have unexpected results, especially if the target species plays a critical role in the ecosystem. Some of the greatest long-term impacts of fishing have been observed in non-targeted ecosystem components. Many species, including marine mammals, seabirds, sea turtles, sharks, oysters, kelps, and sea grasses, have been negatively affected by fisheries either directly through bycatch or habitat damage, or indirectly through altered food-web interactions.

One area of active inquiry is the underlying cause for the measured reduction in mean trophic level of landings seen in many of the world’s oceans. Originally, these reductions were attributed to sequentially fishing lower trophic levels as

higher ones were depleted, a process termed “fishing down the food web.” A more recent analysis has offered an alternative hypothesis of “fishing through the food web,” where fisheries add lower trophic species while continuing to catch higher trophic level species. These differing conclusions underscore two of the important issues addressed in this report. The first is the need for new models and new data to identify the underlying cause of change in marine ecosystems. The second is the recognition that the implications for management will differ based on this underlying cause. Fishing down the food web indicates that lower-trophic-level-species are harvested due to the depletion of the higher level predators. Fishing through the food web indicates that multiple trophic levels are being fished simultaneously—and perhaps sustainably. The appropriate management action for each of these cases should be crafted based on the specifics of the ecosystem, species, fishing methods, and values involved.

Whether the unwanted, negative influences of fishing on marine food webs and communities can be reversed is generally unknown. While some stocks have experienced recovery when fishing pressure was reduced, others have not. The overall productivity and composition of marine ecosystems may change for systems exploited beyond a certain threshold with no guarantee of reversibility—new states may persist and even resist return to earlier conditions. In addition, environmental changes, such as climate-driven regime shifts, affect fishery productivity, creating conditions where recovery is even more uncertain.

Management decisions for a particular targeted stock will have impacts on the productivity of other interacting species. Accounting for species linkages in a management context requires that harvest strategies for each species be chosen in ways that recognize the interconnectedness of marine ecosystems. In addition, other consumptive uses, nonconsumptive uses (e.g., recreation and scenic opportunities), and ecosystem services (e.g., nutrient cycling and climate and weather regulation) should be considered when formulating ecosystem goals. Because it is unlikely that value and yield can be simultaneously maximized for all services, tradeoffs are inevitable among various uses and services provided by the marine environment.

Scientific knowledge, from both natural and social sciences, is important for delineating options and illuminating choices, but allocation tradeoffs are public policy decisions. Various stakeholder groups will value a different mix of resource uses and desire different outcomes from management activities. Decisions about what mix of services the ocean will provide and what protections will be afforded to ocean species should be made with input from a broad range of stakeholders. Ultimately, a flexible management structure is needed to adapt to shifting ecosystem dynamics and changing stakeholder values, as well as to integrate decision making across all sectors of human activity. If decisions about tradeoffs in eco-

system services are to be equitable, fisheries management decisions will require consideration of other nonfishery uses of the marine environment.

Ecosystem-level effects of fishing are well supported in the scientific literature, including changes in food-web interactions and fluctuations in ecosystem productivity. Stock biomass and abundance have been reduced by fishing, and the size structure of populations has been altered. Furthermore, changes in trophic structure, species interactions, and biodiversity have been discovered, and fisheries-induced alternative ecosystem states (defined by a different species composition or productivity than that of the prefishing condition) are possible. Assuming that the upper level of harvest productivity from wild ocean resources is at or approaching the theoretical limit, and recognizing the inability to change one ecosystem component without affecting numerous others, food-web interactions will become increasingly important in future fisheries management decisions. Society will need to determine which ecosystem components are the most desirable for harvest, and then managers will need to implement policies designed to maximize this desired production while recognizing that this will affect other species.

If the United States is to manage fisheries within an ecosystem context, food-web interactions, life-history strategies, and trophic effects will need to be explicitly accounted for when developing harvesting strategies. Other uses and values derived from marine resources should also be considered, because fishing activities directly or indirectly impact other ecosystem components and the goods and services they provide. A modeling framework is necessary to examine ecosystem interactions and to compare the possible outcomes of different fishery management actions. Decisions about management strategies should be made in a manner that accounts for the range of uses involved and their relative social, ecological, and economic values.

Currently, fisheries management approaches in the United States do not explicitly account for the ecosystem-level impacts discussed in this report. Furthermore, existing policies do not generally consider the possible effects of fisheries on other services provided by the ocean environment.

Multiple-species harvest strategies should be evaluated to account for species interactions and food-web dynamics.

Setting multi-species harvest strategies requires taking into account food-web interactions, changes in trophic structure and species interactions, life-history strategies, and bycatch. If management is to account for the ecological inter-

dependence among harvest targets and other food-web components, it will be necessary to quantitatively and qualitatively examine these interactions. Increased application of food-web, species-interaction, and ecosystem models, and development of new models could provide a better understanding of food-web effects and the impacts of fishing on ecosystem components and help to develop multi-species harvest strategies.

Food-web, species-interaction, and ecosystem models should be used to evaluate alternative policy and management scenarios. These scenarios should inform the choice of multi-species harvest strategies by elucidating the tradeoffs that will be required from the various user communities to manage in a multi-species context.

Presently, fishery management policies employed in the Unites States are prescriptive, defined in terms of nonspecific biological reference points used to set target and limit harvest rates and to specify biomass thresholds for single species. The basic stock assessment process largely informs tactical decisions, rather than evaluating the consequences of different policy choices for all stakeholders. However, in an ecosystem context, management decisions will reflect value judgments and tradeoffs between uses; hence, scientific advice should provide strategic options about different management scenarios that can then be debated in the public-policy arena.

Ecosystem and food-web models exist that can provide useful tools for evaluating policy alternatives. The challenge for scientists and managers is to identify and assign probabilities to a range of scenarios that capture existing uncertainties about food-web dynamics and responses of food webs to various fishing strategies. The proposed approach includes the creation of appropriate model scenarios for managed systems, the generation of a number of management strategies to be evaluated, and the determination of performance statistics for measuring policy outcomes that will reflect the interests of all stakeholders. The alternative scenarios may represent different structural models for the dynamics and current status of the interactive system of species, different levels of productivity, and maximum population sizes under various climate regimes, as well as different relationships between the performance indicators.

The creation of integrated biologic-socioeconomic models will help to make tradeoff decisions even more explicit and informative. Ideally, new models will be able to capture important biophysical linkages and human impacts via economic market valuation methods. The most useful models will be those that include not only the best depictions of ecosystem links, but also accurate depictions of fishermen’s behaviors and responsiveness to changes in governance systems.

Scenario analyses and the corresponding management actions are best applied in an iterative and adaptive process (Figure S.1). Monitoring programs should be

an integral component of management. Data are necessary to evaluate how both marine organisms and fishermen respond to changing management actions. Models will improve as more is learned and greater levels of complexity are added, requiring an adaptive approach to management.

Interdisciplinary working groups should be considered as a mechanism for developing appropriate models for each management area and for generating the series of scenarios needed to test proposed management actions.

Building models relevant for fisheries management will require the cooperation of many specialists and the integration of information from many sources. Working groups can provide a mechanism for bringing together scientists from government and academia as well as natural and social sciences in order to examine particular areas or fisheries of concern. Including social and economic scientists at the beginning of this process will ensure that these issues are incorporated as the base model is created.

Such working groups would facilitate consolidation of existing information, generate new syntheses with existing models, and develop new models and other new approaches to inform scenario development and forecasting under alternative management strategies. Working groups could meet with a variety of stakeholders—including fishermen and other consumptive and nonconsumptive users—to identify important tradeoffs that should be considered when creating models and evaluating feasible candidate policies. The simulations created should be quantitative when possible, but even rigorous qualitative scenarios would be useful in some situations. Iterative analyses by the working group might be expected as the ecosystem responds to management actions.

Fisheries are primarily managed by direct or indirect controls on either inputs (e.g., effort, gear type and configuration, time, and area closures or openings) or outputs (e.g., catch in weight or numbers; limitations on landing according to size, sex, or species; and maximum bycatch amounts). Most fisheries management in the United States and internationally relies on output controls with catch quotas as a primary regulatory objective, accomplished by some input controls on gear, areas, and seasons. From an ecosystem perspective, addressing the manner in which fishing is conducted via input controls may be more important than limiting the outputs. This is because ecosystem effects often result from the specifics of the fishing methods, rather than the absolute level of target-species removal.

Two approaches exist for regulating fishing effort to achieve either single-species or multi-species management objectives. By far, the most common method used in both the United States and internationally is top-down control. In this report, top-down control refers to a system in which harvest targets or limits are

set by a management body, often with stakeholder participation, and then input controls are chosen and implemented to achieve these goals. Alternatively, bottom-up management approaches that confer secure access privileges are available. These types of management instruments still require that harvest targets be set by a centralized management body, but the details of effort and input decisions are decentralized.

In principle, existing top-down regulatory procedures can be adapted to account for ecosystem-level effects of fishing. However, a potential benefit of secure access privileges is that they can foster a stewardship ethic among fishermen motivated by concern for the long-term health and productivity of the fishery. These approaches may also promote fishing innovations that reduce impacts to other ecosystem components if access to the fishery is predicated on limiting impacts on non-target species.

New governance and management instruments that create stewardship incentives among user groups should be evaluated and considered for adoption in the United States for multi-species management.

Individual quotas, harvester cooperatives, community cooperatives, and territorially defined cooperatives exist in a handful of fisheries in the United States and in other countries. However, new research is needed to understand how these systems affect incentives in a multi-species setting, and how they might be adapted to handle more inclusive ecosystem goals associated with fisheries management in the United States.

Consumptive uses are those that rely on the removal or harvest of ocean resources, such as fishing, and therefore their value is readily measured based on market demand. On the other hand, nonconsumptive values and the value of ecosystem services are much more difficult to measure. The most common nonconsumptive values are those related to tourism, research, and education, in which values are expected to positively correlate with healthy ecosystems. The value provided by ecosystem services such as nutrient cycling and weather regulation are extremely difficult to quantify, but may be proportionally more important—ecosystem services are experienced across society, although the values are often overlooked. In order to make informed decisions about the suite of services provided by ocean ecosystems, increased understanding is needed about the range of values generated by these systems and about how these values affect different stakeholder groups.

Fisheries management structures should ensure that a broad spectrum of social values is included in policy and management decisions.

A diverse cross-section of constituents may be needed to advise decision makers on the desired balance of ecosystem values and uses that management should try to achieve. An important public policy issue is how to ensure that nonconsumptive and public-good values receive proper consideration when making tradeoffs among ocean services. Further, incorporating a broader range of values will require input from fisheries, ecosystem, and social scientists to help understand how various ecosystem configurations generate services that are valued by different stakeholder groups. Melding fisheries science and social science will be important for understanding how behavior might be modified in response to changing priorities and management actions.

Implementing scenario-based analysis; considering alternative management instruments; and integrating ecological, social, and economic values into fisheries management decisions require enhanced research in a number of areas. Scientific advances will need to incorporate new ideas, analyses, models, and data; perhaps, more importantly, new social and institutional climates will need to be established that catalyze a creative, long-term, comparative, and synthetic science of food webs and communities. Data needed to support ecosystem-based management will likely be more than the simple sum of currently available single-species information. Where species interact and to what extent will be as important as determining a stock biomass. Furthermore, a rich array of social science, economic science, and policy considerations will be essential, because many more tradeoffs are likely to be apparent among ecosystem components and stakeholders.

Research is needed to improve our understanding of the extent of fishing effects on marine ecosystems and to promote the development of ecosystem, food-web, and species-interaction models for incorporation into management decisions.

Promising results have come from analyses and models at levels of synthesis above individual populations and individual food-web components. However, if these models are to be applied in a management setting, greater knowledge of trophic effects and species interactions is necessary. Modification of existing models and/or the development of new models are needed to better account for uncertainty in model output, to elucidate indicators of regime shifts and other interacting factors, and to evaluate monitoring schemes necessary to provide adequate information on ecosystem structure and function.

Support of research in a number of areas will help to improve the utility of current and future models, including quantifying important food-web interactions, per capita and population interaction strengths, and baseline abundance data on a number of non-target and lower-trophic-level species. Much more can be learned

about food-web linkages and interactions, including the strength of linkages between species and life-history stages and how these interactions change over time. Because so little is known about the complexities of marine ecosystems, data needs should be prioritized both for near- and long-term efforts, and for species and areas of concern.

Spatial analyses may be one of the greatest obstacles faced by fishery managers, yet new developments in measurement and analysis methods allow for the explicit consideration of spatial variability in marine systems. Collecting spatially explicit biological data will be essential for monitoring and assessing both large-scale population trends and changes at finer scales. Patterns of interaction and the strength of these interactions vary in time and space. Collecting data in both dimensions will increase understanding about the potential variability in these interactions and advance the ability of models to represent future scenarios. Furthermore, biologically relevant boundaries in the marine environment are virtually impossible to identify. Research is needed to determine whether ecosystem boundaries, for both modeling endeavors and management, could be better defined based on known interactions. In addition, analyzing population trends and species interactions on finer spatial scales may lead to new ideas about temporally and spatially flexible, area-based management.

Looking back in time is as important as assessing current ecosystem status. Assessments of historical data can provide new insights about former species abundances and interactions. A historical perspective is important for many reasons, chief among them is avoiding the shifting baselines phenomena. If recovery goals are to be established, it is wise to comprehend the levels of ecosystem productivity that were once possible. Further, synthesizing these types of data using models may allow for the examination of past interactions and their relative importance, indicating when it might be desirable to try to restore these interactions. Landings data, narratives and descriptions, fisheries-independent data, phytoplankton and plankton records, satellite data, archived specimens, empirical knowledge, and many other sources of information should all be considered when conducting these types of analyses.

Research is needed to expand relevant social and economic information and to integrate this knowledge into fisheries management actions.

Understanding social and ethical values linked to the broad suite of services provided by marine ecosystems is essential and will require measuring and scaling of those values in relation to other uses. While some valuation analyses have been completed for terrestrial systems, little comparable work exists for marine systems. Once it is hypothesized how various fishing strategies affect the structure and functioning of marine ecosystems, methods can be designed to evaluate how these changes affect humans directly and indirectly, elucidating those policy options that reflect the most desirable choices.

Evaluating management options will require integrated models that incorporate not only the best depictions of ecosystem links, but also the most accurate depictions of fishermen’s behaviors and responsiveness to changes in regulations and governance systems. As mentioned previously, integrated biologic-socioeconomic models should be explored for their capacity to capture important biophysical linkages that are translated through human impacts via economic market valuation methods. Understanding behavior is a particularly under-researched area, even behavior associated with conventional management systems. Information should be collected to examine how different kinds of governance mechanisms could potentially change fishermen’s behavior. Integrating this information into combined models would allow for the explicit consideration of all aspects of the management process—the how and where of biological resource interactions and the how, where, and why of human actions.

Finally, research should be conducted on how new governance mechanisms might better align fishing incentives to address more encompassing ecosystem management objectives. Most existing incentive-based systems are primarily single-species focused, but some are also beginning to address broader ecosystem objectives. Existing experiences with individual quotas, harvester cooperatives, community cooperatives, and territorially defined cooperatives should be examined. New research is also needed on management strategies that might best address food-web linkages, bycatch questions, and broader portfolios of ecosystem services. Analyzing available experiences worldwide can indicate whether these systems might be appropriate for adoption in U.S. fisheries to reduce ecosystem-level impacts of fishing.

This page intentionally left blank.