Nothing endures but change.
CHALLENGES IN UNDERSTANDING THE CAUSE OF POPULATION DECLINES
The Gulf of Alaska, Aleutian Islands, and Bering Sea encompass a vast and spatially heterogeneous territory. The biological richness of this region has been exploited by humans for at least 5,000 years. Groundfish, such as Pacific halibut (Hippoglossus stenolepis)1 and Pacific cod (Gadus macrocephalus), were first harvested in nearshore waters by Alaskan natives for subsistence. Beginning in the mid- to late 19th century, domestic fixed-gear (hook-and-line) fisheries began for Pacific cod, Pacific halibut, and sablefish (Anoplopoma fimbria). In the 1930s and 1940s, Japanese trawl fisheries developed for walleye pollock (Theragra chalcogramma) and flatfish (primarily yellowfin sole, Pleuronectes aspera) in the eastern Bering Sea. During World War II foreign fishing ceased. After the war, large multinational fisheries developed off Alaska. These included drift gillnet fleets for Pacific salmon; tangle net fisheries for crabs; longline fisheries for Pacific halibut, Pacific cod, sablefish, and Greenland turbot (Reinhardtius hippoglossoides) in the Bering Sea; and trawl fisheries for groundfish,
Scientific names for species mentioned in the text are given in Appendix H.
herring (Clupea pallasi), and shrimp (Pandulus spp.). Foreign fleets largely comprised of small- to medium-sized (90 to 180 foot) trawlers and Danish seiners, some large factory stern trawlers (up to 270 feet), motherships for at-sea processing, and many support vessels were deployed to exploit groundfish resources off Alaska. Fishing effort shifted among species partly as a result of overfishing and changes in product demand by countries participating in the international groundfish fleet. Conflicts, including those among foreign trawlers and American fixed-gear vessels, resulted in increasing restrictions on foreign fleets. The Magnuson Fishery Conservation and Management Act of 1976 promoted domestic fishing by limiting the total allowable level of foreign fishing to be that portion of the optimum yield that was not expected to be harvested by domestic vessels. During the 1980s, foreign operations were converted to joint ventures between domestic catcher vessels and foreign processing vessels. In the 1990s the Alaska groundfish fishery became fully “Americanized.”
The current domestic groundfish fisheries target walleye pollock (67% of total groundfish catch in 2000), Pacific cod, Atka mackerel (Pleurogrammus monopterygius), sablefish, and a variety of rockfish and flatfish species. In 2000 the groundfish fleet was comprised of 1,261 catcher vessels, 90 catcher processors, and 69 inshore processors and motherships. The majority of catcher vessels are less than 60 feet in length, but vessels in this size class accounted for only 20% of the exvessel value of groundfish harvested in 2000. The 15 catcher processor vessels that landed pollock in the eastern Bering Sea ranged in length from 201 to 376 feet.
Species composition and abundance in the North Pacific have undergone substantial variation over time. Small-mesh trawl survey data show a major shift in the relative abundance of benthic species from shrimps in the 1970s to groundfish in the 1980s (Anderson and Piatt, 1999). Increases in many flatfish and gadid species were particularly conspicuous (North Pacific Fishery Management Council, 2001a, 2001b) and overlapped with major declines of some crab populations, such as red king crabs (Paralithodes camtschaticus) throughout the Gulf of Alaska and Bristol Bay in the eastern Bering Sea (Zheng and Kruse, 2000). Variation also exists in the distribution and abundance of mammals and marine birds (Springer et al., 1999). Coincident with these changes, the western stock of Steller sea lions (Eumetopias jubatus) has undergone a persistent decline. The largest loss of animals occurred in the late 1970s and 1980s, but a more gradual decrease (about 5% per year) continued through the 1990s to 2000, with some signs of recovery in the Gulf of Alaska region seen in the recently released 2002 counts (see Figures 1.1 and 1.2). This decline, from a population estimated to be in the hundreds of thousands in the 1960s to roughly 30,000 in 2001, caused the National Marine Fisheries Service (NMFS) to list Steller sea lions as threatened in 1990 under the Endan-
gered Species Act (ESA). In the face of continued decline in the west, this stock was proposed as endangered in October 1995 and listed as such in May 1997. It is not possible to determine whether the 1960s sea lion counts assessed the population above, at, or below the long-term average because reliable population data are not available to establish this type of baseline. A 19th-century description of Steller sea lions suggests that the population previously experienced large fluctuations in abundance, although again the cause is unknown but could be partially the result of human disturbance and hunting (see Appendix D).
Concerns about conflicts between Steller sea lions and commercial fisheries have a long history of scientific studies that started well before the listing of the western stock as endangered. Alverson (1992) provided a list of possible factors that contributed to the decline of Steller sea lions based on data from Loughlin (1987), and this list was revisited in the
Bowen et al. (2001) critique of NMFS’s November 2000 Biological Opinion (BiOp #3). The hypothesized factors include changes in the species composition and abundance of Steller sea lion prey, disease, toxins, killer whale predation, intentional and incidental (or illegal) killing of Steller sea lions, and regime shifts in the physical environment. From the first documentation of a population decline in the 1980s to fiscal year 2001, there has been an escalation of funding for research aimed at identifying plausible causes of the decline (see Appendix E). In fiscal year 1992, research funding was a modest $1.43 million, compared with $43.15 million in fiscal year 2001 and $40.145 million in 2002 (DeMaster and Fritz, 2001; Ferrero and Fritz, 2002).
Every ecosystem presents a multitude of complex processes and interactions that defy straightforward description and explication. With respect to this particular ecosystem, spatial structures are important features,
whether at the scale of sea lion rookeries and haulouts, oceanic fronts, or entire oceanic basins. Spatial ecology (Tilman and Kareiva, 1997) provides a conceptual perspective that helps to unite existing data on rookery-specific analyses and pollock stock dynamics and movement (Shima et al., 2002). The challenge here is to bring together these spatial perspectives—regional and local—in a structured approach so that it is possible to generate and test hypotheses about the causes of the continuing sea lion population decline.
At present there is insufficient data to conclusively identify the cause or causes of the decline. However, the available information can be organized both to identify the most fruitful avenues for future research and to evaluate the likely efficacy of current and future management actions. This weight of evidence approach, described in Box 1.1, provides an analytical framework for timely decision making when scientific certainty is constrained by limited historical data, high system variability, and extremely challenging field conditions.
THE POLICY CONTEXT
Is it possible to reach definitive answers about the cause of the Steller sea lion decline? Complex systems, by definition, are not amenable to simple description and analysis. Not only is the Steller sea lion situation complex, there is a dearth of information about both the ecosystem and the sea lions before and during the decline that could be used to reduce uncertainty about the cause or causes of the population decrease. When that complexity includes an important economic activity—in this case the groundfish fisheries—policy made in the absence of certainty often results in various legal challenges.
Science proceeds by seeking to rule out hypotheses that do not seem credible and by identifying those hypotheses that appear plausible and supportable. In this process there is often a reluctance to promote specific hypotheses as causal, but there is often less reluctance to rule out alternative hypotheses as unimportant. Hence, one focus of this report is to evaluate evidence both for and against the various hypothesized causes of the Steller sea lion decline.
Several pieces of legislation form the context for the Steller sea lion controversy. The Magnuson-Stevens Fishery Conservation and Management Act of 1996 calls for the protection of marine ecosystems, assuring that “irreversible or long-term adverse effects on fishery resources and the marine environment are avoided”(16 U.S.C. 1802 (5)). The Marine Mammal Protection Act states: “The primary objective of (marine mammal) management should be to maintain the health and stability of the marine ecosystem” (16 U.S.C. 1361). The National Environmental Policy
BOX 1.1 Nature of Evidence and Standards of Inference
The strength of scientific inference varies greatly depending on both the nature of the problem and the availability of relevant reliable data. Strong inferences can be made when there is a hypothesis testable by experimentation. The standard procedure is to establish a decision criterion (usually termed the critical value) against which the hypothesis is either accepted or rejected based on the experimental result. Usually the decision criterion is chosen to reduce the probability of incorrectly accepting the hypothesis. This procedure is the preferred approach of scientific inquiry.
In natural resource management and conservation, there may be obstacles to experimentation. Achieving this high standard of inferential rigor might require more time or resources than would be practical for management of the affected resources. In some cases, the necessary experiments may even be deemed unethical or illegal or the complexity of the system may reduce the likelihood of obtaining unequivocal results. In these situations, information from various sources can be assembled to make a weight of evidence argument among the competing hypotheses. In general, the strength of a weight of evidence inference depends on the number of independent lines of evidence and the relevance of each. The strength of the inference is typically weaker when there are a number of plausible alternative hypotheses and the problem may be caused by multiple factors.
No fewer than eight hypotheses have been proposed to explain the Steller sea lion decline (see Chapter 6). A wide array of data is available on sea lions and their associated ecosystems, including spatiotemporal patterns of population change, information on body condition and foraging behavior, fish stock assessments and fishing effort, population trends in other co-occurring species, and a variety of historical records. Given various inferences about how these data relate to each hypothesis, a weight of evidence assessment can be established. Furthermore, hypotheses that are inconsistent with multiple lines of evidence can be ranked as less likely than hypotheses consistent with existing evidence. As with all types of analyses, the robustness of a conclusion is based on the quality of the data and potential biases introduced during the analysis.
Act seeks “efforts which will prevent or eliminate damage to the environment and biosphere” by requiring federal agencies to “identify and develop methods and procedures that will insure that presently undocumented environmental amenities and values may be given appropriate consideration in decisionmaking”(42 U.S.C. 4321). The ESA provides “a means whereby the ecosystems upon which endangered species and threatened species depend may be conserved [and] a program for the conservation of such endangered species and threatened species”(16 U.S.C. 1531 (b)). While these legislative mandates approach protection from different perspectives, careful management of human use of living marine resources remains their overarching goal.
Under the ESA, federal agencies are required to ensure that their actions, or actions authorized or funded by them, are not likely to jeopardize the survival or recovery of protected species or damage their critical habitat. Section 7 of the ESA requires that when an action may affect a listed species or its critical habitat, the federal agency conducting or authorizing that action (the action agency) must consult with the federal agency charged with overseeing recovery efforts for the listed species (the expert agency). In cases where a federally managed fishery may interfere with the survival or recovery of certain marine mammals (seals, sea lions, porpoises, and whales), NMFS is both the action and the expert agency. These responsibilities are segregated within NMFS between the Office of Sustainable Fisheries, which has responsibility for reviewing the fishery management plans, and the Office of Protected Resources, which has responsibility for implementing ESA regulations for listed species under its jurisdiction.
The consultation process allows for the possibility that the agency will be unable to develop reasonable and prudent alternatives (RPAs) that will remove jeopardy (50 CFR 402.15[h], 2001). The biological opinion requires consideration of the extent to which listed species are vulnerable to “incidental takes” as an unintended consequence of a particular action under consideration. These statements on incidental takes must include reasonable and prudent measures to minimize the probable impacts arising from such takes and must set forth terms and conditions that will minimize probable impacts. On the basis of this information, the federal action agency (NMFS Office of Sustainable Fisheries) will determine whether or not to proceed with the proposed action (in this case the groundfish fishery in the Bering Sea/Aleutian Islands and the Gulf of Alaska (Figure 1.3).
While neither “jeopardy” nor “adverse modification” is defined in the ESA, the Code of Federal Regulations defines jeopardy as follows: “Jeopardize the continued existence of means to engage in an action that reasonably would be expected, directly or indirectly, to reduce appreciably the likelihood of both the survival and recovery of a listed species in the wild by reducing the reproduction, numbers, or distribution of that species” 50 CFR 402.02, p. 38 (2001).
Scientists and managers cannot ensure specific outcomes of regulatory actions even when legislation requires a particular outcome, such as the preservation of a particular species. In some cases, federal agencies have been forced to relinquish discretion in the matter of endangered species policy to the courts. If the public is not satisfied as to whether or not an agency’s actions seem reasonable in light of its obligations to an endangered species, the determination is made by the courts. This reflects the role of courts in the U.S. constitutional system of judicial oversight of
legislative intent and language and executive branch action in response to that language.
If a jeopardy decision is rendered, the agency must issue a set of RPAs that would remove the finding of jeopardy. These RPAs must meet four conditions: (1) they must be identified during formal consultation and must be implemented in a manner consistent with the intended purpose of the action; (2) they must be implemented in a manner consistent with the agency’s legal authority and jurisdiction; (3) they must be economically and technically feasible; and (4) they must be actions that the agency believes will ameliorate the original finding of jeopardy to the listed species or its habitat. A successful RPA must meet all four of these conditions. When uncertainty exists, the benefit of the doubt must be given to the listed species (Greenpeace v. National Marine Fisheries Service, 55 F. Supp. 2d, at 1262).
The third condition for a successful RPA is that actions taken should be economically and technically feasible. This does not mean that an agency is required to balance benefits to the endangered species against
costs falling on an industry. To do so would be “fundamentally inconsistent with the purposes of the ESA and with the case law interpreting the ESA” (55 F. Supp. 2d, at 1267). Instead, case law suggests the application of lexicographic choice. In lexicographic choice one first satisfies the paramount goal and then chooses among alternatives consistent with that goal. In the current context, lexicographic choice requires that the NMFS develop sets of reasonable and prudent alternatives that comply with the ESA mandate of no jeopardy. An evaluation of benefits and costs can then be conducted to select a preferred option from among these RPAs.
HISTORY OF ESA LISTINGS AND COURT CHALLENGES
As part of the authorization of the fishery management plans for the commercial groundfish fisheries in the Bering Sea/Aleutian Islands and the Gulf of Alaska regions, NMFS summarized the consultation in a biological opinion as required under Section 7 of the ESA. The purpose of the biological opinion is to ascertain if the groundfish fisheries, as prosecuted under the fishery management plans, are likely to imperil the continued existence of Steller sea lions (and other listed species) or are likely to destroy or adversely modify critical habitat. In April 1998, Greenpeace filed a complaint in U.S. District Court that NMFS had failed to revise the environmental impact statement relating to federal groundfish fisheries in Alaska and had violated the ESA because the biological opinions regarding the impacts of these fisheries on sea lions were inadequate. In the biological opinion issued in December 1998 (known as BiOp #1), NMFS concluded that the groundfish fisheries, excepting pollock, were unlikely to cause harm to listed species. In the case of the pollock fishery, there was a finding of jeopardy based on competition between the fishery and sea lions for pollock. In response to this finding, a set of RPAs was developed in consultation with the North Pacific Fishery Management Council (NPFMC) that spread fishing effort out spatially and temporally and closed the Aleutian Islands management area to pollock fishing. These restrictions were implemented in the 1999-2000 fishery management plans. After the RPAs went into effect in January 1999, NMFS issued another biological opinion (BiOp #2), which analyzed the effects of the entire groundfish fishery management plan on sea lions and found no jeopardy from the pollock fishery based on a review of the total allowable catch levels proposed for the Gulf of Alaska and Bering Sea/Aleutian Islands management areas.
Greenpeace filed suit in response to the new biological opinion and on July 9, 1999, U.S. District Court Judge Thomas Zilly found the RPAs to be arbitrary and capricious because there was no explanation of how the proposed restrictions mitigated jeopardy for the pollock fishery. He also
found that the environmental impact statements were inadequate and directed NMFS to prepare a more comprehensive analysis of the Bering Sea, Aleutian Islands, and Gulf of Alaska groundfish fisheries (Greenpeace v. National Marine Fisheries Service, 80 F. Supp. 2d 1137 WD. Wash., 2000). In January 2000, Judge Zilly ruled that the “no jeopardy” finding in BiOp #2 was inadequate under ESA because it only considered the total allowable catch levels for individual groundfish fisheries and failed to consider the combined and cumulative impacts of all groundfish fisheries on sea lion populations. Based on the January ruling, Greenpeace filed for an injunction prohibiting groundfish trawling in sea lion critical habitat until a new comprehensive biological opinion was prepared by NMFS. The injunction was granted in July and implemented in August 2000.
NMFS released the revised biological opinion on November 30, 2000 (BiOp #3). It concluded that Steller sea lion populations are jeopardized by the Alaska groundfish fisheries, including Atka mackerel, Pacific cod, and pollock, due to competition for prey and modification of prey distribution in critical habitat. This revised biological opinion found jeopardy with regard to pollock even under the restrictions imposed by the 1999 RPAs. The opinion included a comprehensive set of new RPAs that incorporated adaptive management to assess the efficacy of the groundfish restrictions. The western population was divided into 13 management areas designated as either open—with fishing allowed under the 1999 restrictions—or closed—with no fishing allowed in critical habitat. However, these new regulations were resisted as being too costly to the groundfish fisheries, and Alaska Senator Ted Stevens attached an amendment to the December 2000 omnibus appropriations bill that delayed full implementation of the RPAs and provided the NPFMC with an opportunity to develop an alternative set of RPAs. In addition, the amendment provided $30 million for economic relief to offset losses incurred by sea lion protection measures, $28 million for research on the causes of the decline of sea lions, and $2 million for scientific review of BiOp #3, including the review by the National Academy of Sciences that is the subject of this report.
In February 2001 the NPFMC appointed an RPA committee to develop alternatives to the RPA in BiOp #3 that removed the potential for jeopardy from the pollock, Atka mackerel, and Pacific cod fisheries but had less impact on the fishing industry and associated communities. In June the RPA committee proposed an alternate set of measures that discarded the earlier adaptive management approach and used new telemetry data to justify restricting fishing primarily in the first 10 nm of the 20-nm radius delineating critical habitat areas. The telemetry data suggest that sea lions spend most of their time at sea within 10 nm of the rookeries. The revised RPA assumes that the telemetry data reflect the foraging behavior
of sea lions, and hence 10 nm delimits the zone with the maximum likelihood of competition with the fisheries. By moving most fishing activities beyond 10 nm, with some further restrictions between 10 and 20 nm, the RPA committee was able to reach the same theoretical reduction of jeopardy as in BiOp #3. In August 2001, NMFS released BiOp #4, which evaluates the new RPA measures and includes a supplemental environmental impact statement that compares the various RPA measures. NMFS concludes in BiOp #4 that the June 2001 RPAs provide adequate protection for Steller sea lions with regard to the groundfish fisheries.
SCOPE AND ORGANIZATION OF THE REPORT
In response to the congressional request for a review of the Steller sea lion decline and the Alaska groundfish fishery, the Ocean Studies Board and the Polar Research Board of the National Academies agreed to undertake a study of the issue, which was funded through the NPFMC. A committee of experts in marine mammal biology, marine ecology, and fisheries science was convened in the summer of 2001 to address the issues of concern listed in the study’s statement of task (see Box 1.2). See Appendix A for committee biographies.
The committee held three public meetings, two in Seattle and one in Anchorage, to receive input from the NPFMC, NMFS, academic scientists, fishermen, Alaska natives, environmentalists, and other concerned members of the public.
The report is organized to provide a rationale for analyzing the causes of the sea lion decline based on what is known about the demographics of
BOX 1.2 Statement of Task
This study will examine interactions between Alaska groundfish fisheries and Steller sea lions (Eumetopias jubatus) and the role of these fisheries in the evolving status of the sea lion population. The focus of the study will be (1) the status of current knowledge about the decline of the Steller sea lion population in the Bering Sea and Gulf of Alaska ecosystems; (2) the relative importance of food competition and other possible causes of population decline and impediments to recovery; (3) the critical information gaps in understanding the interactions between Steller sea lions and Alaska fisheries; (4) the type of research programs needed to identify and assess potential human and natural causes of sea lion decline; and (5) the components of an effective monitoring program, with yardsticks for evaluating the efficacy of various management approaches.
the population and the known linkages of Steller sea lions, human activities, and various features of food webs in the North Pacific region. Chapter 2 describes the general features of the environmental setting based on the climatic, oceanographic, and biological features of the region defined by the range of the western Steller sea lion population. An accounting of sea lion mortality based on population and ecosystem models is presented in Chapter 3 to evaluate how much of the sea lion decline cannot be explained by known factors. This analysis also helps identify the types of mortality that could most readily explain the observed pattern of sea lion population decline. Readers unfamiliar with Steller sea lion biology may prefer to consult the review of sea lion biology presented in Chapter 4 prior to reading the modeling discussion in Chapter 3. Chapter 5 describes the North Pacific commercial fisheries in reference to potential interactions with Steller sea lions. These summaries provide the basis for the weight of evidence approach used in Chapter 6 to evaluate each of the eight major hypotheses proposed to explain the Steller sea lion decline. Identification of critical information gaps, research approaches, and recommendations for future monitoring programs are presented in Chapter 7.