Seasonal foraging movements and migratory patterns of female Lamna ditropis tagged in Prince William Sound, Alaska

Conventional and electronic tags were used to investigate social segregation, distribution, movements and migrations of salmon sharks Lamna ditropis in Prince William Sound, Alaska. Sixteen salmon sharks were tagged with satellite transmitters and 246 with conventional tags following capture, and were then released in Prince William Sound during summer 1999 to 2001. Most salmon sharks sexed during the study were female (95%), suggesting a high degree of sexual segregation in the region. Salmon sharks congregated at adult Pacific salmon Oncorhynchus spp. migration routes and in bays near Pacific salmon spawning grounds in Prince William Sound during July and August. Adult Pacific salmon were the principal prey in 51 salmon shark stomachs collected during summer months in Prince William Sound, but the fish appeared to be opportunistic predators and consumed sablefish Anoplopoma fimbria, gadids, Pacific herring Clupea pallasi, rockfish Sebastes spp. and squid (Teuthoidea) even when adult Pacific salmon were locally abundant. As Pacific salmon migrations declined in late summer, the salmon sharks dispersed; some continued to forage in Prince William Sound and the Gulf of Alaska into autumn and winter months, while others rapidly moved south‐east thousands of kilometres toward the west coasts of Canada and the U.S. Three movement modes are proposed to explain the movement patterns observed in the Gulf of Alaska and eastern North Pacific Ocean: ‘focal foraging’ movements, ‘foraging dispersals’ and ‘direct migrations’. Patterns of salmon shark movement are possibly explained by spatio‐temporal changes in prey quality and density, an energetic trade‐off between prey availability and water temperature, intra‐specific competition for food and reproductive success. Transmissions from the electronic tags also provided data on depth and water temperatures experienced by the salmon sharks. The fish ranged from the surface to a depth of 668 m, encountered water temperatures from 4·0 to 16·8° C and generally spent the most time above 40 m depth and between 6 and 14° C (60 and 73%, respectively).     

Genetic structuring among Alaskan Pacific herring populations identified using microsatellite variation

Five highly variable microsatellite loci were used to investigate population structuring in Pacific herring Clupea pallasi collected from Kodiak Island, two sites in the Bering Sea and four sites within Prince William Sound, Alaska. All loci revealed high levels of variability with heterozygosity estimates ranging from 86 to 97% (mean heterozygosity: 89%). The variation was structured significantly among sites suggesting that the samples investigated were genetically distinct from each other. Genetic divergence was greatest between populations from the Bering Sea and those from Prince William Sound. The Kodiak Island and Point Chalmers samples appeared to be distinct from the Prince William Sound and Bering Sea populations. The observed genetic distance relationships among samples could be explained largely in terms of geographical separation.     

Differential Expression of Immune Response Genes in Steller Sea Lions (Eumetopias jubatus): An Indicator of Ecosystem Health?

Characterization of the polygenic and polymorphic features of the Steller sea lion major histocompatibility complex (MHC) provides an ideal window for evaluating immunologic vigor of the population and identifying emergence of new genotypes that reflect ecosystem pressures. MHC genotyping can be used to measure the potential immunologic vigor of a population. However, since ecosystem-induced changes to MHC genotype can be slow to emerge, measurement of differential expression of these genes can potentially provide real-time evidence of immunologic perturbations. MHC DRB genes were cloned and sequenced using peripheral blood mononuclear leukocytes derived from 10 Steller sea lions from Southeast Alaska, Prince William Sound, and the Aleutian Islands. Nine unique DRB gene sequences were represented in each of 10 animals. MHC DRB gene expression was measured in a subset of six sea lions. Although DRB in genomic DNA was identical in all individuals, relative levels of expressed DRB mRNA was highly variable. Selective suppression of MHC DRB genes could be indicative of geographically disparate environmental pressures, thereby serving as an immediate and sensitive indicator of population and ecosystem health.     

Predation on an upper trophic marine predator, the Steller sea lion: evaluating high juvenile mortality in a density dependent conceptual framework

The endangered western stock of the Steller sea lion (Eumetopias jubatus)--the largest of the eared seals--has declined by 80% from population levels encountered four decades ago. Current overall trends from the Gulf of Alaska to the Aleutian Islands appear neutral with strong regional heterogeneities. A published inferential model has been used to hypothesize a continuous decline in natality and depressed juvenile survival during the height of the decline in the mid-late 1980's, followed by the recent recovery of juvenile survival to pre-decline rates. However, these hypotheses have not been tested by direct means, and causes underlying past and present population trajectories remain unresolved and controversial. We determined post-weaning juvenile survival and causes of mortality using data received post-mortem via satellite from telemetry transmitters implanted into 36 juvenile Steller sea lions from 2005 through 2011. Data show high post-weaning mortality by predation in the eastern Gulf of Alaska region. To evaluate the impact of such high levels of predation, we developed a conceptual framework to integrate density dependent with density independent effects on vital rates and population trajectories. Our data and model do not support the hypothesized recent recovery of juvenile survival rates and reduced natality. Instead, our data demonstrate continued low juvenile survival in the Prince William Sound and Kenai Fjords region of the Gulf of Alaska. Our results on contemporary predation rates combined with the density dependent conceptual framework suggest predation on juvenile sea lions as the largest impediment to recovery of the species in the eastern Gulf of Alaska region. The framework also highlights the necessity for demographic models based on age-structured census data to incorporate the differential impact of predation on multiple vital rates.     

MORTALITY OF SEA OTTERS IN PRINCE WILLIAM SOUND FOLLOWING THE EXXON VALDEZ OIL SPILL

Abstract: This paper presents an estimate of the total number of sea otters that died as a direct consequence of the oil spill that occurred when the T/V Exxon Valdez grounded in Prince William Sound, Alaska on 24 March 1989. We compared sea otter counts conducted from small boats throughout the Sound during the summers of 1984 and 1985 to counts made after the spill during the summer of 1989. We used ratio estimators, corrected for sighting probability, to calculate otter densities and population estimates for portions of the Sound affected by the oil spill. We estimated the otter population in the portion of Prince William Sound affected by the oil was 6,546 at the time of the spill and that the post-spill population in the summer of 1989 was 3,898, yielding a loss estimate of approximately 2,650. Bootstrapping techniques were used to approximate confidence limits on the loss estimate of about 500–5,000 otters. The wide confidence limits are a result of the complex scheme required to estimate losses and limitations of the data. Despite the uncertainty of the loss estimate it is clear that a significant fraction of the otters in the spill zone survived. We observed otters persisting in relatively clean embayments throughout the oil spill zone suggesting that the highly convoluted coastline of Prince William Sound produced refuges that allowed some sea otters in the oil spill area to survive.     

The Macrobenthos of Sites within Prince William Sound,Alaska, Prior to the Exxon Valdez Oil Spill

Benthic fauna within three bays (Rocky and Zaikof Bays, and Port Etches) of outer Prince William Sound, Alaska are examined. The data represent the only detailed benthic faunal information available for the period prior to a major oil spill by the tanker Exxon Valdez within the Sound. The spatial distribution of fauna determined by classification and ordination resulted in eight station groups. Stepwise multiple discriminant analysis demonstrated a relationship between station groups, sediment grain size and nitrogen. Major faunal differences were observed within Rocky and Zaikof Bays between 1982 and 1990. The faunal differences between the two time periods demonstrate the extreme temporal variability that might be expected within Prince William Sound. The study serves as a cautionary note to avoid conclusions about the effects of disturbance to the benthos on a single data set. An explanation for temporal differences within the bays is presented.     

Inter-Population Movements of Steller Sea Lions in Alaska with Implications for Population Separation

Genetic studies and differing population trends support the separation of Steller sea lions (Eumetopias jubatus) into a western distinct population segment (WDPS) and an eastern DPS (EDPS) with the dividing line between populations at 144° W. Despite little exchange for thousands of years, the gap between the breeding ranges narrowed during the past 15–30 years with the formation of new rookeries near the DPS boundary. We analyzed >22,000 sightings of 4,172 sea lions branded as pups in each DPS from 2000–2010 to estimate probabilities of a sea lion born in one DPS being seen within the range of the other DPS (either ‘West’ or ‘East’). Males from both populations regularly traveled across the DPS boundary; probabilities were highest at ages 2–5 and for males born in Prince William Sound and southern Southeast Alaska. The probability of WDPS females being in the East at age 5 was 0.067 but 0 for EDPS females which rarely traveled to the West. Prince William Sound-born females had high probabilities of being in the East during breeding and non-breeding seasons. We present strong evidence that WDPS females have permanently emigrated to the East, reproducing at two ‘mixing zone’ rookeries. We documented breeding bulls that traveled >6,500 km round trip from their natal rookery in southern Alaska to the northern Bering Sea and central Aleutian Islands and back within one year. WDPS animals began moving East in the 1990s, following steep population declines in the central Gulf of Alaska. Results of our study, and others documenting high survival and rapid population growth in northern Southeast Alaska suggest that conditions in this mixing zone region have been optimal for sea lions. It is unclear whether eastward movement across the DPS boundary is due to less-optimal conditions in the West or a reflection of favorable conditions in the East.     

An in situ, individual-based approach to quantify connectivity of marine fish: ontogenetic movements and residency of lingcod

As modern fishery assessments change in an effort to be more accurate and encompass the range of potential ecosystem interactions, critical information on the ecology of species including life history, intra and inter-specific competitive interactions and habitat requirements must be added to the standard fishery-dependent and independent data sets. One species whose movements and habitat associations greatly affects exploitation patterns is lingcod, Ophiodon elongatus, which support an economically important fishery along the coastal waters of the Pacific Coast of North America. High site fidelity and limited movements within nearshore areas are hypothesized to have resulted in high catchability, a major factor that has contributed to overfished stocks. Thus, assessing the level of movement and connectivity among lingcod subpopulations inhabiting nearshore habitats is a prerequisite to determining the condition of lingcod stocks. We used the Pacific Ocean Shelf Tracking (POST) Project acoustic receiver array in Alaska's Prince William Sound to monitor movements and residency of 21 acoustic-tagged lingcod for up to 16 months. Eight of sixteen lingcod (50%) initially aged at 2.5- to 3.5- years-old dispersed from their tag site. Dispersal was highly seasonal, occurring in two, five-week periods from mid-December through January and from mid-April through May. Dispersal in winter may be related to sexually immature lingcod or newly-mature male lingcod being displaced by territorial males. Spring dispersal may be indicative of the onset of migratory behavior where lingcod move out into Prince William Sound and possibly the offshore waters of the Gulf of Alaska. Our results reveal a pattern of ontogenetic dispersal as lingcod approach 4-years-old and exceed 50 cm total length. The large proportion of tagged fish migrating out of Port Gravina, their tagging site, reflects a high level of connectivity among Prince William Sound subpopulations. Our results also support the hypotheses that these subpopulations may be highly susceptible to overfishing because most fish show long residence times.     

Parallel signatures of selection in temporally isolated lineages of pink salmon

Studying the effect of similar environments on diverse genetic backgrounds has long been a goal of evolutionary biologists with studies typically relying on experimental approaches. Pink salmon, a highly abundant and widely ranging salmonid, provide a naturally occurring opportunity to study the effects of similar environments on divergent genetic backgrounds due to a strict two‐year semelparous life history. The species is composed of two reproductively isolated lineages with overlapping ranges that share the same spawning and rearing environments in alternate years. We used restriction‐site‐associated DNA (RAD) sequencing to discover and genotype approximately 8000 SNP loci in three population pairs of even‐ and odd‐year pink salmon along a latitudinal gradient in North America. We found greater differentiation within the odd‐year than within the even‐year lineage and greater differentiation in the southern pair from Puget Sound than in the northern Alaskan population pairs. We identified 15 SNPs reflecting signatures of parallel selection using both a differentiation‐based method (BAYESCAN) and an environmental correlation method (BAYENV). These SNPs represent genomic regions that may be particularly informative in understanding adaptive evolution in pink salmon and exploring how differing genetic backgrounds within a species respond to selection from the same natural environment.     

Efficacy of commercial products in enhancing oil biodegradation in closed laboratory reactors

Summary A laboratory screening protocol was designed and conducted to test the efficacy of eight commercial bacterial cultures and two non-bacterial products in enhancing the biodegradation of weathered Alaska North Slope crude oil in closed flasks. Three lines of evidence were used to support the decision to progress to field testing in Prince William Sound: rapid onset and high rate of oxygen uptake, substantial growth of oil degraders, and significant degradation of the aliphatic and aromatic hydrocarbon fractions of the weathered Alaska North Slope crude oil. A product had to enhance biodegradation greater than that achieved with excess mineral nutrients. Experiments were conducted in closed respirometer flasks and shake flasks, using seawater from Prince William Sound and weathered crude oil from a contaminated beach. Analysis of the data resulted in selection of two of the ten products for field testing. Both were bacterial products. Findings suggested that the indigenous Alaskan microorganisms were primarily responsible for the biodegradation in the closed flasks and respirometer vessels.     

Serologic survey for Brucella spp., phocid herpesvirus-1, phocid herpesvirus-2, and phocine distemper virus in harbor seals from Alaska, 1976-1999

Harbor seals (Phoca vitulina richardsi) were captured in the coastal regions of Southeast Alaska, Gulf of Alaska, Prince William Sound (PWS), and Kodiak Island during 1976-1999. Blood was collected from 286 seals. Sera were tested for evidence of exposure to Brucella spp., phocid herpesvirus-1 (PhoHV-1), phocid herpesvirus-2 (PhHV-2), and phocine distemper virus (PDV). Antibody prevalence rates were 46% (46/100) for Brucella spp., 93% (225/243) for PhoHV-1, 0% (0/286) for PhHV-2, and 1% (2/160) for PDV. Antibody prevalence for Brucella spp. was directly related to host age. Antibody prevalence for PhoHV-1 was higher in PWS as compared to the other three regions. No evidence of mortality attributable to these four agents was observed during the course of this study. Based on the results of this survey, none of these agents is considered a significant mortality factor in harbor seals from the four regions of coastal Alaska included in the study.     

Source-Sink Estimates of Genetic Introgression Show Influence of Hatchery Strays on Wild Chum Salmon Populations in Prince William Sound,Alaska

The extent to which stray, hatchery-reared salmon affect wild populations is much debated. Although experiments show that artificial breeding and culture influence the genetics of hatchery salmon, little is known about the interaction between hatchery and wild salmon in a natural setting. Here, we estimated historical and contemporary genetic population structures of chum salmon (Oncorhynchus keta) in Prince William Sound (PWS), Alaska, with 135 single nucleotide polymorphism (SNP) markers. Historical population structure was inferred from the analysis of DNA from fish scales, which had been archived since the late 1960’s for several populations in PWS. Parallel analyses with microsatellites and a test based on Hardy-Weinberg proportions showed that about 50% of the fish-scale DNA was cross-contaminated with DNA from other fish. These samples were removed from the analysis. We used a novel application of the classical source-sink model to compare SNP allele frequencies in these archived fish-scales (1964–1982) with frequencies in contemporary samples (2008–2010) and found a temporal shift toward hatchery allele frequencies in some wild populations. Other populations showed markedly less introgression, despite moderate amounts of hatchery straying. The extent of introgression may reflect similarities in spawning time and life-history traits between hatchery and wild fish, or the degree that hybrids return to a natal spawning area. The source-sink model is a powerful means of detecting low levels of introgression over several generations.     

Hypotheses concerning the decline and poor recovery of Pacific herring in Prince William Sound, Alaska

This paper updates previous reviews of the 1993 stock decline of Pacific herring (Clupea pallasi) in Prince William Sound, Alaska, and focuses on hypotheses about subsequent poor recovery. Recent age structured assessment modeling with covariate analysis indicates that the population dynamics of the sound’s herring are influenced by oceanic factors, nutrition, and, most substantially, hatchery releases of juvenile pink salmon. For the 1993 decline, poor nutrition remains the most probable cause with disease a secondary response. Concerning poor recovery, we examined 16 potential factors and found three to be causal: oceanic factors, poor nutrition, and hatchery releases of juvenile pink salmon. Absences of strong year classes at both Sitka and Prince William Sound after 1993 indicate the action of large-scale ocean processes. Beyond regional-scale environmental factors, two factors specific to the sound influence the population dynamics of herring and are likely impeding recovery. First, pink salmon fry releases have increased to about 600 million annually and may disrupt feeding in young herring, which require adequate nutrition for growth and overwintering survival. Juvenile pink salmon and age-1 herring co-occur in nearshore areas of bays in late spring and summer, and available data on dietary overlap indicates potential competition between the age-1 juvenile herring and juvenile pink salmon. Field studies demonstrate that juvenile herring reduce food intake substantially in the presence of juvenile pink salmon. Second, overwintering humpback whales may consume potentially large amounts of adult herring, but further studies must confirm to what extent whale predation reduces herring biomass.     

Investigating life‐history traits of Steller sea lions with multistate hidden Markov mark–recapture models: Age at weaning and body size effects

The duration of offspring care is critical to female fitness and population resilience by allowing flexibility in life‐history strategies in a variable environment. Yet, for many mammals capable of extended periods of maternal care, estimates of the duration of offspring dependency are not available and the relative importance of flexibility of this trait on fitness and population viability has rarely been examined. We used data from 4,447 Steller sea lions Eumetopias jubatus from the Gulf of Alaska and multistate hidden Markov mark–recapture models to estimate age‐specific weaning probabilities. Maternal care beyond age 1 was common: Weaning was later for animals from Southeast Alaska (SEAK) and Prince William Sound (PWS, weaning probabilities: 0.536–0.648/0.784–0.873 by age 1/2) compared with animals born to the west (0.714–0.855/0.798–0.938). SEAK/PWS animals were also smaller than those born farther west, suggesting a possible link. Females weaned slightly earlier (+0.080 at age 1 and 2) compared with males in SEAK only. Poor survival for weaned versus unweaned yearlings occurred in southern SEAK (female survival probabilities: 0.609 vs. 0.792) and the central Gulf (0.667 vs. 0.901), suggesting poor conditions for juveniles in these areas. First‐year survival increased with neonatal body mass (NBM) linearly in the Gulf and nonlinearly in SEAK. The probability of weaning at age 1 increased linearly with NBM for SEAK animals only. Rookeries where juveniles weaned at earlier ages had lower adult female survival, but age at weaning was unrelated to population trends. Our results suggest the time to weaning may be optimized for different habitats based on long‐term average conditions (e.g., prey dynamics), that may also shape body size, with limited short‐term plasticity. An apparent trade‐off of adult survival in favor of juvenile survival and large offspring size in the endangered Gulf of Alaska population requires further study.     

A Conceptual Model of Natural and Anthropogenic Drivers and Their Influence on the Prince William Sound,Alaska, Ecosystem

Prince William Sound (PWS) is a semi-enclosed fjord estuary on the coast of Alaska adjoining the northern Gulf of Alaska (GOA). PWS is highly productive and diverse, with primary productivity strongly coupled to nutrient dynamics driven by variability in the climate and oceanography of the GOA and North Pacific Ocean. The pelagic and nearshore primary productivity supports a complex and diverse trophic structure, including large populations of forage and large fish that support many species of marine birds and mammals. High intra-annual, inter-annual, and interdecadal variability in climatic and oceanographic processes as drives high variability in the biological populations. A risk-based conceptual ecosystem model (CEM) is presented describing the natural processes, anthropogenic drivers, and resultant stressors that affect PWS, including stressors caused by the Great Alaska Earthquake of 1964 and the Exxon Valdez oil spill of 1989. A trophodynamic model incorporating PWS valued ecosystem components is integrated into the CEM. By representing the relative strengths of driver/stressors/effects, the CEM graphically demonstrates the fundamental dynamics of the PWS ecosystem, the natural forces that control the ecological condition of the Sound, and the relative contribution of natural processes and human activities to the health of the ecosystem. The CEM illustrates the dominance of natural processes in shaping the structure and functioning of the GOA and PWS ecosystems.     

TRENDS IN ABUNDANCE OF ALASKA HARBOR SEALS, 1983–2001

We estimated trends in abundance of harbor seals ( Phoca vitulina richardsii ) using over dispersed, multinomial models and counts obtained during aerial surveys conducted during 1983–2001 in the Ketchikan, Sitka, Kodiak, and Bristol Bay areas of Alaska. Harbor seal numbers increased significantly at 7.4%/yr during 1983–1998 and 5.6%/yr during 1994–1998 in the Ketchikan area, and 6.6%/yr during 1993–2001 in the Kodiak area. Counts were stable (trends not significant) during 1984–2001 (0.7%/yr) and 1995–2001 (-0.4%/yr) in Sitka, and during 1998–2001 (-1.3%/yr) in Bristol Bay. The influence of covariates ( e.g. , survey date, tide height) on trend estimates was significant and varied among areas and across years, demonstrating the need to include covariates in statistical analyses to accurately estimate trend. Our increasing trend estimate for Kodiak represents the first documented increase in harbor seal numbers over a relatively expansive area in the Gulf of Alaska. However, the trend for the Gulf of Alaska stock is equivocal due to the continued decline in Prince William Sound. Similarly, the trend for the Southeast Alaska stock is equivocal based on our increasing (Ketchikan) and stable (Sitka) trend estimates, and a recent decline reported for Glacier Bay. The Bering Sea stock appears stable after a period of possible decline.     

DISTRIBUTION OF KILLER WHALE PODS IN PRINCE WILLIAM SOUND, ALASKA 1984–1996

Thirteen years of encounter data (1984–1996) were used to examine killer whale distribution within Prince William Sound, Alaska. Four patterns of area use were found, which comprised differences between resident pods and transient groups and differences among resident groups. Resident pods frequented large open passages, while transient groups used the narrow passages and bays in the southwest. This dichotomy likely reflects resident use of salmon and transient use of pinniped prey resources, as well as the different foraging strategies required for these prey types. Four resident pods (AB, AI, AJ, and AN) used Knight Island Passage more than other areas of the Sound; two (AE and AK) used all areas of the Sound more evenly. Use of the Sound by the AT1 transient whales declined in the latter part of the study. Nearshore foraging for pinniped prey by the AT1 transient whales was more common in areas where these whales spend a disproportionate amount of time, suggesting that these areas were critical foraging habitat for them. No similar pattern emerged for Open-water Foraging for cetaceans by AT1 whales, nor for foraging by the resident whales.     

Site fidelity and the demographic implications of winter movements by a migratory bird, the harlequin duck Histrionicus histrionicus

Understanding the degree of demographic connectivity among population segments is increasingly recognized as central to the fields of population ecology and conservation biology. However, delineating discrete population units has proven challenging, particularly for migratory birds as they move through their annual cycle. In this study, radio telemetry was used to assess movement rates among habitats by harlequin ducks Histrionicus histrionicus during the non-breeding season in Prince William Sound, Alaska. A total of 434 females were outfitted with radio transmitters over six years of data collection, and their signals tracked by aircraft. Using a spatially nested design, it was determined that 75% of radioed females remained in the bay or coastline area where they were originally trapped, 94% remained on the same island or mainland region of Prince William Sound where they began the winter period, and 98% remained within the 4500 km² study area as a whole. Home range analyses corroborated these findings, indicating that the scale of individual movements was small, with 95% kernel home range estimates averaging only 11.5±2.2 km². A simple demographic model, which incorporated estimates for population size, survival, and movement rates, was used to infer the degree of independence among population segments. Immigrant females were found to contribute little to population numbers in most areas, accounting for only 4% of the adult female population at a scale of approximately 100 km². These results have important implications for the scale of conservation action for the species and demonstrate that winter movements can have a strong influence local population dynamics.     

STOCK STRUCTURE OF SEA OTTERS (ENHYDRA LUTRIS KENYONI) IN ALASDA

Sea otters in Alaska are recognized as a single subspecies ( Enhydra lutris kenyoni ) and currently managed as a single, interbreeding population. However, geographic and behavioral mechanisms undoubrably constrain sea otter movements on much smaller scales. This paper applies the phylogeographic method (Dizon et al . 1992) and considers distribution, population response, phenotype and genotype data to identify stocks of sea otters within Alaska. The evidence for separate stock identity is genotypic (all stocks), phenotypic (Southcentral and Southwest stocks), and geographic distribution (Southeast stock), whereas population response data are equivocal (all stocks). Differences in genotype frequencies and the presence of unique genotypes among areas indicate restricted gene flow. Genetic exchange may be limited by little or no movement across proposed stock boundaries and discontinuities in distribution at proposed stock boundaries. Skull size differences (phenotypic) between Southwest and Southcentral Alaska populations further support stock separation. Population response information was equivocal in either supporting or refuting stock identity. On the basis of this review, we suggest the following: (1) a Southeast stock extending from Dixon Entrance to Cape Yakataga; (2) a Southcentral stock extending from Cape Yakataga to Cape Douglas including Prince William Sound and Kenai peninsula coast; and (3) a Southwest stock including Alaska Peninsula coast, the Aleutians to Attu Island, Barren, Kodiak, Pribilof Islands, and Bristol Bay.