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

Fishing by its very nature alters marine ecosystems by selectively removing fish and invertebrates. Humans have become one of the oceans’ dominant predators, and with human populations continuing to grow, the influence on the marine environment is escalating. Today, a significant portion of energy and protein from fish and invertebrates is directed toward human uses (Watson and Pauly 2001, Food and Agriculture Organization [FAO] 2002) (see Appendix B for list of acronyms). However, the upper limit of potential harvest from wild ocean resources may have been reached (Garcia and Grainger 2005). Seventy-six percent of the world’s stocks are fully exploited, overexploited, or depleted; few resources remain that may provide for the development of additional sustainable fisheries (Hilborn et al. 2003, FAO 2005) (Figure 1.1).

Concerns about overfishing have been expressed by fisheries specialists for decades, but in the last 10 to 15 years, overfishing has become a major public issue (Mace 2004). New analyses emphasize the multi-species nature of fisheries and indicate that the resulting changes in predation and competition should be accounted for in fishery management approaches (May 1984, International Council for the Exploration of the Sea [ICES] 2000, Sinclair and Valdimarsson 2003). The issue has not been whether this should be done, but continues to be how it can be done. A number of recent scientific publications conclude that changes to marine populations and food webs caused by fisheries removals are larger than had been previously believed, raising public and scientific concerns about the true extent of changes in marine ecosystems. At the same time, both the U.S. Commission on Ocean Policy (2004) and the Pew Oceans Commission (2003) recommend managing resources in an ecosystem context, including

fisheries. These events all combine to focus intense interest on the possible ecosystem-level effects of fishing and whether these influences can be addressed in fisheries management.

Fisheries by their very nature are extractive enterprises and therefore reduce the biomass of exploited populations. But sometimes the level of harvest has been too great, and scientists have shown that some stocks have been reduced to low fractions of their unfished levels. This is particularly evident in intensively fished coastal regions like the North Sea, the Gulf of Thailand, and the North Atlantic (Gulland 1988, Pauly and Maclean 2003, Garcia and Grainger 2005). But researchers are also now raising new concerns about the multiple impacts of fishing not only on target stocks, but also on other ecosystem components through bycatch (removal of non-target species), indirect effects (the removal of one species leading to the benefit or detriment of another), and habitat impacts (Dayton et al. 1995; Pauly and Christensen 1995; Pauly et al. 2002, 2005; National Research Council [NRC] 2002; Chuenpagdee et al. 2003; Pauly and Maclean 2003).

Furthermore, and possibly more important to this report, the public has been increasingly concerned that activities associated with fisheries have global impacts beyond those related solely to depletions of local stocks. This perception has been fueled by a series of highly publicized journal articles and reports (e.g., Pauly et al. 1998a, Jackson et al. 2001, Myers and Worm 2003). These papers suggest that fishing wild populations of marine fish and invertebrates has fundamentally altered the structure of marine ecosystems, resulting in severe depletion of populations at high trophic levels and propagating through whole communities of interacting species through indirect effects. However, these papers have not been without criticism. 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. These ongoing discussions in the scientific literature are not as well known outside the fisheries science community as the papers mentioned above, yet they are equally important for deciding a course for future management. This report strives to present and discuss the related scientific literature by putting the range of perspectives in context and weighing the collective evidence on fisheries-induced ecosystem change.

Fisheries management has traditionally focused on the status of individual fish stocks; both the United States and the United Nations have policies regarding rebuilding overfished species. Recently, however, concerns have been raised about whether these approaches can account for the possible broader ecological impacts of fishing. As ocean management begins to embrace ecosystem-based principles, what are the specific concerns for fisheries? The possible ecosystem effects of fishing encompass a wide range of biological interactions including

changes in predator-prey relationships and trophic dynamics (Hughes 1994, Pauly et al. 1998a, Steneck 1998), reductions in non-target species through bycatch (D’Agrosa et al. 2000, Lewison et al. 2004a), “cascading” effects mediated through food-web interactions (Frank et al. 2005, Ware and Thompson 2005), and the loss or degradation of habitats due to fishing pressure (NRC 2002). Fishing activities are also size and species selective, potentially resulting in changes to the genetic makeup of a stock as well as the structure of fished populations and communities.

Many of these effects are indirect and can alter complex interactions in marine communities, ultimately affecting the functioning of the ecosystem itself. However, fisheries impacts are occurring at the same time as other large-scale influences, such as El Niño events, other decadal-scale oscillations, and long-term climate change. In addition, a range of other human actions are also taking place in marine and coastal environments, resulting in the loss of wetlands and coral reefs, eutrophication, and pollution. Isolating the underlying cause of ecosystem effects is extremely difficult. All of these actions could be occurring individually or in concert to drive a relatively pristine marine ecosystem to one that is fully utilized¹ and, eventually, to one that is degraded.² Ultimately these effects, individually or collectively, may result in shifts in marine ecosystems that may or may not be reversible.

Another issue at hand is how to manage and protect marine ecosystems when goals and actions are based on incomplete understanding of ecosystem components and what constitutes an undisturbed ecosystem. Some archeological, paleoecological, and historical data analyses indicate that fishing by humans has altered the structure and function of marine ecosystems for centuries or millennia (e.g., Jackson et al. 2001). Such changes would have predated modern descriptions of ecosystem structure and will affect conclusions about why some ecosystems have undergone dramatic changes, perhaps shifting the scale on which ecosystem alteration is measured.

Many scientists and managers agree that a well-informed discussion is needed on what we consider the “ideal” state of the ocean to be, setting in motion thoughtful plans about how to achieve our common goals. Many questions need to be answered to reset the course of fisheries management: How much rebuilding is enough or desirable? Do levels of productivity observed for the last 50 years provide an adequate baseline for setting current goals? What tradeoffs are needed among species or among uses to allow management of the ecosystem as a whole? Some of these questions are scientific in nature; others are societal choices that need to be made in the public-policy arena.

Ecosystems are inherently complex and any disturbance, such as removals of target species by a fishery, is likely to affect other components of the system; however, assessing these effects is often extremely difficult. Science continues to elucidate these interactions, but gaps in scientific knowledge ultimately limit the ability to fully assess the impacts of fisheries on marine ecosystems and to anticipate the likely response of interacting food webs to fishing.

Confounding the issue is the lack of independent and comprehensive fishery data in most regions, leaving fisheries landings and fishing effort data as the only indicators of ecosystem change in some cases. However, the conclusions that can be drawn from fishery-dependent data are often controversial because fisheries change in response to many factors (e.g., markets, management regulations) in addition to alterations in the ecosystem that they exploit. Alternative interpretations of fishery data sets are possible depending on how the analyses treat biases and limitations in the primary data and on the assumptions scientists make to fill the gaps in available information.

Further, consistent data going back more than 20 to 30 years are often difficult to obtain, a problem that makes characterizations of past conditions as tricky as predicting the future of fished ecosystems. Often, less formal sources of information are employed to reconstruct the history of fisheries and ecosystems prior to the implementation of regular monitoring programs (e.g., Rosenberg et al. 2005). Clearly, these studies are important for understanding the likely impacts of fishing operations on marine ecosystems, but this evidence alone is often inconclusive.

Scientists’ ability to qualitatively and quantitatively model ecosystem dynamics and predict complex ecosystem interactions has rapidly advanced in recent decades. Continuous improvements to species-interaction models, energy-balance models, and ecosystem models have raised the possibility of applying such methods to management. In fact, some have evolved to the state where they have been used on small scales to enumerate policy options (Christensen 2005, Kitchell 2005). And while other models are still in the development stage, the possibility remains that models can be used to construct future scenarios of various ecosystem effects based on initial input conditions including biologic, economic, and management parameters.

Broad stewardship options are available for addressing the general effects of fishing on marine populations, food webs, and communities. As a starting point, one can envision some early dynamic state of a wilderness ocean. In a number of cases, humans began exploiting the living resources of the sea well before written records documented this early state. What now exists is an ocean of living

resources that has been modified by fishing, overfishing, and other factors not directly linked to fishing. Whether this state is described as utilized or degraded differs among regions, fisheries, and perspectives. For example, in the United States, the North Pacific region is often referenced as the example of a productive, utilized system in which fisheries management has generally succeeded at controlling harvest rates, while the New England region exemplifies the opposite type of system, one with significant impacts due to overfishing. However, no locations likely remain in the world’s oceans that are entirely free of human influence. And while achieving the wilderness state may be neither possible nor desirable, understanding what that state may have been, and how much change has taken place, is important for determining and setting future management goals.

Recognizing that fishing practices have changed the ocean, the obvious question presents itself: Where do we, as a society, want to go from here? Several broad policy options are available. The traditional option would be to continue managing fish populations with maximum sustainable yield (MSY)³ serving as the target catch for each individual fishery. However, single-species management has not been successfully implemented in many cases, with many populations overfished as a result. Continuing to overfish would further reduce the productivity of the stocks and make it more difficult to achieve a well-managed, utilized ecosystem.

On the other hand, the consequences of implementing single-species management more effectively than in the past are not easy to predict. It is possible that just eliminating overfishing would allow for some stocks or ecosystems to recover to former, more appropriately utilized states that meet societal goals. Another option would be to manage using more conservative fishing targets than those based on MSY. In fisheries managed by the North Pacific Fisheries Management Council, many species are harvested at levels defined by optimum yield,⁴ well below MSY. However, this approach is often not implemented in other regions because of pressure to raise catch levels to accommodate various sectors of the fishing industry. Alternatively, policy options could incorporate multiple-species management strategies based on analyses of species interactions, food webs, and community dynamics. For systems that are already overutilized or potentially degraded, this approach might be termed ecosystem rehabilitation, sensu Francis et al. (1979). The basic premise of ecosystem rehabilitation in a fisheries context is that a reduction in fishing pressure, informed by knowledge of species interactions, would allow recovery to a less disturbed state. How closely the ecosystem could approach some previous ocean condition would depend to some extent on the magnitude of the fisheries management controls implemented. But, conceptually,

these actions may move the ecosystem closer to that condition, perhaps allowing society to take greater advantage of the overall productivity and to harvest the desired mix of living resources.

However, even for well-managed fisheries, a larger question remains. If managers were able to implement specific management actions to achieve a desired point on the scale from pristine to degraded ecosystems, how would this point be chosen? Commercial fishing, tourism, recreational fishing, and many other uses affect, and are affected by, the state of the marine ecosystem. Each stakeholder group will most likely desire a different mix of resource uses and different outcomes from management activities. In addition, the oceans provide large-scale ecosystem services, such as oxygen generation and nutrient cycling. How are these “values” accounted for when deciding ecosystem goals?

Without the ability to satisfy all constituents, the management of natural resources usually results in a series of tradeoffs between various user groups and between different ecosystem services. For example, one easily defined tradeoff would be associated with the recovery of a single stock. Catch rates would be lower in the short term; however, they presumably would be higher once the stock increased. More generally, tradeoffs involve multiple resources and ecosystem services, and they may create conflicts between different user groups with competing goals. The recovery of a top predator, for example, may negatively impact a fishery based on its preys.

When different users are in conflict (i.e., one use precludes or impacts another), the tradeoffs between uses require resolution of public policy decisions and value judgments. But both natural and social science have an important role in informing management decisions by revealing the range of potential outcomes based on a more complete understanding of ecosystem functioning, human behavior, and the connections between interacting species.

Both the U.S. Commission on Ocean Policy (2004) and the Pew Oceans Commission (2003) stressed the need to move away from sectoral management of the oceans (e.g., fisheries, shipping, water quality, oil and gas, invasive species, critical habitat, protected species, etc.) and toward an ecosystem-based management approach to ocean and coastal resources. The statement of task for this study was specifically to examine the ecosystem effects of fisheries and to consider the implications for fisheries management. The study’s recommendations could be implemented by the National Oceanic and Atmospheric Administration (NOAA) National Marine Fisheries Service (NMFS) and the Regional Fisheries Management Councils to facilitate (1) a move from singles-species management to multi-species management, (2) consideration of marine food webs, and ultimately (3) ecosystem-based fisheries management. But it is essential to make the

distinction between these fisheries-focused management decisions and true ecosystem-based management.

Many sectors other than fisheries impact marine ecosystems (Breitberg and Riedel 2005), and many different regional, federal, state, and county agencies have legislative responsibilities relating to marine systems. For example, decisions with respect to oil and gas extraction and shipping can have impacts on fisheries or protected species, but selection of sites for oil and gas development is not the responsibility of fisheries managers. Furthermore, water quality management—the responsibility of yet another agency—begins in the uppermost reaches of the watershed and has impacts downstream in the coastal zone, potentially affecting fish health or growth, and limiting viable habitat (Craig et al. 2001).

This report outlines the potential ecosystem effects of fishing and discusses the potential role of larger ecosystem impacts in fisheries management planning. But these actions will be only a part of a comprehensive ecosystem-based approach to management. As stated by the U.S. Commission on Ocean Policy (2004), the implementation of ecosystem-based management demands the active involvement of multiple agencies, requiring substantial, perhaps unprecedented, cooperation among management agencies at a number of levels of government and across issues.

The National Research Council (NRC) Committee on Ecosystem Effects of Fishing: Phase II—Assessments of the Extent of Ecosystem Change and the Implications for Policy was charged with reviewing and evaluating the current literature on the impacts of modern fisheries on the composition and productivity of marine ecosystems (see Box 1.1). NMFS, the study sponsor, asked the committee to discuss the relevance of these findings for U.S. fisheries management, identify areas for future research and analysis, and characterize the stewardship implications for living marine resources. The committee took the approach of reviewing the current literature to provide a larger context for the findings of several widely publicized studies and to evaluate whether the weight of the collective evidence is sufficient to justify changes in the U.S. approach to fisheries management. The findings and recommendations of the committee were based on presentations heard at three public meetings (see Appendix C for meeting agendas), published literature, and their own expertise. This report examines the current scientific evidence for ecosystem effects of fishing, including changes in abundance, biodiversity, and genetic structure of populations; food-web effects such as trophic cascades and species interactions; and both physical and fishery-induced regime shifts (Chapter 2). Subsequent chapters discuss how these kinds of effects might be addressed by changing how the United States manages fisheries (Chapter 3) and how interactions among species, uses, sectors, and values could be accounted for in fisheries management decisions (Chapter 4).

Chapter 5 discusses the research needed to better understand multi-species interactions and the social and economic science needed to improve management strategies. Chapter 6 contains the recommendations of the committee, setting forth a research and stewardship agenda that will support a more holistic approach to fisheries management. Primary emphasis is placed on U.S. fisheries management—taking into account that much of the existing literature is global in nature and may not apply to conditions within the U.S. exclusive economic zone. Phase I of the NRC series on the ecosystem effects of fishing, Effects of Trawling and Dredging on Seafloor Habitat (NRC 2002), considered the habitat impacts of these fisheries on the seafloor and therefore will not be discussed in this report.

This page intentionally left blank.