UBC Faculty Research and Publications

Food web models and data for studying fisheries and environmental impacts on Eastern Pacific ecosystems Guénette, Sylvie; Christensen, Villy 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-FCRR_2005_13-1.pdf [ 3.77MB ]
Metadata
JSON: 52383-1.0074785.json
JSON-LD: 52383-1.0074785-ld.json
RDF/XML (Pretty): 52383-1.0074785-rdf.xml
RDF/JSON: 52383-1.0074785-rdf.json
Turtle: 52383-1.0074785-turtle.txt
N-Triples: 52383-1.0074785-rdf-ntriples.txt
Original Record: 52383-1.0074785-source.json
Full Text
52383-1.0074785-fulltext.txt
Citation
52383-1.0074785.ris

Full Text

$*+,$ *$																 !	""#$%&'	(#)*+,	$'--$ !"#"$%%&'(#))*++,-./+,01,,121+%23,456%1,7%+/210/1$2+12%1$$$+42/	587		$		,9 :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::!	587		$ :::::::::::::::::::::::::::::::::::::::::::::::::::::::"	-$	1		!&;"	<:<:6 :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::	1		5	1	=!&	<:<:6 ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::&/	1	=	58 :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::)!,	+	,= :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::"/		$	$<	% ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::#)"6			58>	 ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::#"	95?$8	<:<:6 ::::::::::::::::::::::::::::::::::::::::::::::##*.&(%' !$%&'	($""#6%$2%%+%21%1.2%1$2,261+1@312$$$1,6%1.2%1$2'11( !"#"6 UBC Fisheries Centre Reports, Vol. 13, No. 1Director's forewordThere are various ways ecosystem "control", and two of these are 'top-down control' and 'bottom-up control', usually setas alternatives. This dichotomy has various incarnations; in the Pacific Northwest it is referred to as the'Thompson-Burkenroad debate', with the former associated with top-down control (here: of halibut biomass, by fishing),and the latter bottom-up control (with environmental variability responsible for changes in the recruitment, andeventually, the biomass of halibut). When applied to ecosystems, more often than not, the 'bottom-up' part of thisdichotomy has more evidence in its favour, particularly in the Pacific Northwest, where 'regime shifts' tend to be invokedalmost exclusively to explain ecosystem changes. The main reason for this asymmetry, however, is that it is easier tomeasure temperature and its variability, or chlorophyll and its variability, than to construct and fit ecosystem models andtest how much they explain of the variability at hand. However, it has now become possible to straightforwardly constructmodels of ecosystems, and to fit them with time-series data, and thus to test top-down control hypotheses, i.e., to separateout top-down from bottom-up effects. These tests, which required ecosystem models such as documented in this report,have not shown regime shifts to be unimportant. Rather, they have shown, at least for the North Pacific, that bottom-upand top-down processes occur simultaneously, and that both must be taken in account when modelling these ecosystems.Thus, this document is part of what will take us beyond the dichotomy, toward the complex hypotheses that thesecomplex ecosystems deserve. Daniel PaulyDirector, Fisheries Centre, UBCUBC Fisheries Centre Research Reports, Vol 13, No. 1 7Fisheries, the environment, or what? An introductionSylvie Gu?nette and Villy ChristensenThe North Pacific is a hot-bed for understanding how marine populations are impacted by humans as well as byenvironmental conditions. The ?Thompson-Burkenroad debate? has been ongoing since the late-1940s: what drives themarked fluctuations in Pacific halibut that has been observed over the past century? Dr William Thompson, who startedup the work of the International Pacific Halibut Commission, IPHC, argued that the changes in halibut abundance couldbe fully explained by changes in fishing pressure, i.e. that they were the result of successful management on the part ofIPHC, while his adversary, Dr Martin Burkenroad questioned if the populations trends could be accounted for by fishingpressure on its own, or if wasn?t rather a question of environmental factors impacting halibut recruitment. WhileThompson and Burkenroad actually never debated the relative role of fisheries and the environment ? indeed it may wellbe that they would actually agree that one factor in itself would not suffice to give us the full explanation their debatehas lived on, and both sides still have proponents arguing for one over the other. Examining the Pacific halibut trendsnow, nearly 60 years after the debate started, still yields inconclusive answers only. We cannot name the culprit.The debate has widened since Thompson and Burkenroad?s days, and we regularly hear about regime shifts in connectionwith the North Pacific. A notable debate in this context has emerged, seeking explanations for why the Steller sea lionshave declined to become threatened in major parts of their North Pacific distribution area, while increasing in others. Amultitude of explanations have been brought forward, and considerable research has been aimed at understanding theimportance of nutritional conditions, of predators and of prey, of competition with commercial fisheries, of parasites anddiseases, of the Pacific Decadal Oscillation Index, and of the potential impact of incidental culls, to mention some. Asfor the halibut, no conclusive explanation has emerged.Asking then, if the non-emergence of a single clear explanation may be due to the Steller sea lion being impacted by acombination of factors the North Pacific Universities Marine Mammal Research Consortium and the North PacificMarine Science Foundation initiated a project ?Ecosystem analysis of Steller sea lion dynamics and their prey? throughNOAA funding. The project, which was the brain child of Andrew Trites (Director of the Marine Mammal Research Unit,Fisheries Centre), employs ecosystem modelling of North Pacific ecosystems (Southeast Alaska, the Central Gulf ofAlaska and the Western Aleutian Islands) in an attempt to evaluate (quantify!) the relative role the various factors mayhave played in determining population trends. The methodologies applied for the modelling along with some of thepreliminary findings from the study are described in this report. Notably, the models indicate that no single factor by itselfcan explain the population trends of Steller sea lion, several have to be invoked. In parallel to the work centred on Steller sea lion, the UBC ?Sea Around Us? project (www.seaaroundus.org) throughfunding from the Pew Charitable Trusts initiated a series of workshops aimed at evaluating the relative role of fisheriesand environmental factors for North Pacific ecosystems. Bringing together researchers from the Department of Fisheriesand Oceans,  Pacific Biological Station, Nanaimo; the NOAA Alaska Fisheries Science Center, Seattle; the Universityof Washington,  School of Fisheries, Seattle; and the University of British Columbia, Fisheries Centre, Vancouver, toanalyse a series of ecosystems ranging from the Bering Sea to the Northern California Current, and coordinate themethodologies. We present descriptions of some studies in this report, while most of the findings are publishedseparately. The present report also includes a reconstruction of North Pacific whale catches for the 20th century, whichserved to estimate the whale population at different periods in Southeast Alaska and the Western Aleutians. Finally, inthe interest of preparing future work related to evaluating nutritional aspects of North Pacific ecosystems we haveincluded a compilation of the energy content of invertebrates, fish and mammals in the Gulf of Alaska.The present report is freely available at the website of the Fisheries Centre of the University of British Columbia(www.fisheries.ubc.ca/publications/reports/fcrr.php).8 Aleutian Islands models; Heymans 1Cite as: Heymans, Sheila J.J., 2005. Ecosystem models of the Western and Central Aleutian Islands in 1963, 1979 and 1991. In: Gu?nette, S.,and V. Christensen (editors). 2005. Food web models and data for studying fisheries and environmental impacts on Eastern Pacific ecosystems FisheriesCentre Research Reports 13(1): 8-82.Ecosystem models of the Western and Central Aleutian Islands in 1963, 1979 and 19911Sheila J.J. HeymansFisheries Centre, UBC, Vancouver, BC; s.heymans@fisheries.ubc.caABSTRACTThis paper describes the data and methodology used to construct three models for the Aleutian Islands for 1963, 1979and 1991 as well as how the 1963 model was fitted to time series data. The models were built to examine the decline inthe western stock of Steller sea lions, Eumetopias jubatus, and reflect that purpose in the breakdown of groups, e.g. itincludes 4 different groups for Steller sea lions. The models also include 5 other mammal groups and there were 21groups of fish, 6 invertebrate, 2 primary producers, sea birds and detritus for a total of 40 groups. INTRODUCTIONThe reasons for the Steller sea lion decline in the Aleutian Islands were investigated by using three models of the systemfrom the early 1960s (1963), the late 1970s (1979), and the early 1990s (1991) and fitting the 1963 model to time seriesdata to obtain the best fit of the model parameters to the time series data. The models consist of 40 compartments, ofwhich 1 is detritus, 2 primary producers (phytoplankton and macrophytes), 6 invertebrates, 9 marine mammals, 1 seabirdand 21 fish. A brief overview of the fisheries is given below, followed by the description of each compartment and theirspecific time series of biomass and catches, model balancing and finally model fitting. Various hypotheses for the decline in Steller sea lion have been given: disease, orca predation, junk food hypothesis,entanglement in marine debris, climate change, pollution and fisheries interactions (Alverson 1992). The interactionsbetween fisheries and Steller sea lions include competition for food sources, bycatch mortalities, interruption of normalfeeding patterns, shooting in defence of gear, as well as indiscriminate shooting (Alverson 1992). Fishermen have beenobserved killing adult sea lions at rookeries, haul-outs and in the water near boats and trawl fishermen commonly shootsea lions during haul back operations (Merrick et al. 1987). Sea lions were also used as bait for crab traps and Merricket al. (1987) suggested that it is not a coincidence that the sea lions declined during the peak landing for Bering Sea king,Lithodes spp., and snow crabs, Chionoecetes opilio. Sea lions tend to sink when shot, and Fiscus and Baines (1966) foundthat 68% of the sea lions killed sank when they were collecting them at sea. Thus, the shooting of sea lions in defenceof gear or indiscriminately would not be noticeable as stranding. STUDY SITEThe Aleutian Island chain is 1,100 miles (1,770 km) long and stretches from the Alaskan Peninsula to close to Siberia(Figure 1A) (Murie 1959). It consists of 70 named islands, and the southernmost island (Amatignak) lies not far northof 51N, which is the same latitude as the northern tip of Vancouver Islands (Murie 1959). Very few of the islands areflat and have lakes (Amchitka, Agattu ans Semichi), and many of the islands are volcanic (Murie 1959). The waters ofthe Aleutian Islands are generally sea ice free, and the weather is usually cloudy or foggy with an abundance of rain inthe summer (Murie 1959). The specific area of this model covers administrative areas 541, 542 and 543 in the western and central Aleutian Islands,from 170W to 170E around the Aleutian Islands, to the 500 metre depth contour, and it encompasses an area of 56,936km? (Figure 1B). The westernmost island in the model area is Attu Island in the ?Near Islands? group, and the easterncutoff to the model is Carlisle Island, halfway through the ?Islands of Four Mountains? group (Figure 1B). Most of thefish species are managed for the Bering Sea and Aleutian Islands combined (BSAI), while very few species, such as Atkamackerel, Pleurogrammus monopterygius, are managed specifically for the Aleutian Islands (170W to 170E).UBC Fisheries Centre Research Reports, Vol 13, No. 1 9ABFigure 1. A. Map of the North Pacific showing the Aleutian Islands, Bering Sea and Gulf of Alaska.B. Study area for the Western and Central Aleutian Islands models showing the approximate area ofthe model down to the 500 m depth contour.FISHERIESHuman activity in the north Pacific can be divided into four distinct periods (Figure 2): the subsistence period from28,000 years ago to present, the northern fur seal period (1786?1984), the whaling period (1845?1914) and thecommercial fishing period which started in 1952 in the Bering Sea (Loughlin et al. 1999) and in 1960 in the Aleutians.During the subsistence period aboriginal fishermen fished from large dugout canoes to capture abundant flatfish, dogfish,Squalus acanthias, rockfish, Sebastes spp., lingcod, Ophiodon elongatus, Pacific cod, Gadus macrocephalus, herring,Clupea pallasii, and blackcod or sablefish, Anoplopoma fimbria (Forrester et al. 1978). Faunal analyses of archaeologicalsites throughout the Aleutian chain confirmed the heavy use of sea lion and other marine mammal species by prehistoricAleuts (McCartney (1984) in Wolfe et al. 2002). From four well-preserved sites on southwest Umnak Island in the easternAleutians, about 70% of the archaeological biomass (meat weight) was represented by sea lions, compared to only 12%by fur seals, Callorhinus ursinus, and 3% by sea otters, Enhydra lutris (Yesner (1981) in Wolfe et al. 2002). The primaryuses for sea lions were for food and clothing, and sea lion whiskers were sold to the Chinese in San Francisco (Wolfeet al. 2002). In addition to marine mammals, native fishing also included seabirds and fish (Loughlin et al. 1999). In therecent time, sea lion hunting was done primarily from skiffs along the coast of Atka and Amlia Islands during the early1980s (Wolfe et al. 2002).Sealing started when northern fur seals were discovered on the Probilof Islands by the crew of a Russian ship, and from1786 to 1984 they were commercially harvested for their fur (Figure 2) while at present there is only a subsistence fisheryby Native fishers for meat (Loughlin et al. 1999). The harvest of northern fur seals was the sole commercial activity until1845, when whaling ships moved from whaling grounds near Kodiak and south of the Aleutian Islands into the BeringSea Hunt (1975 in Loughlin et al. 1999). Between 1889 and 1907, predominantly blue whales, Balaenoptera musculus,10 Aleutian Islands models; Heymans 28,000 BP 1780 1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000SubsistenceSealingWhalingCommercialFigure 2. The four periods of human exploitation in the Aleutian Islands adapted from Loughlin etal. (1999).050,000100,000150,000200,000250,0001960 1965 1970 1975 1980 1985 1990 1995 2000Atka Pollock Pcod POP/Rockfish OtherUSA halibutCanadaJapanUSSRKoreaJoint VentureUSA DomesticFigure 3. Catches (tonnes) made in the Aleutian Islands from 1960-2002 and time-frame offishing nations that took those catches. The category ?other? includes marine mammals,sablefish, all flounders, halibut, cephalopods and bycatch/discards of sharks, skates, sculpins,etc.were taken (Mackintosh 1965) and by 1914 whaling became uneconomical in the Bering Sea. However, whalingcontinued in the northeast Pacific until 1974 (Figure 2) when the last fin whale, Balaenoptera physalus, was caught andgray whales, Eschrichthius robustus, are still taken as subsistence (Gu?nette and Salter, this volume). By the late 1960s,only Japan and the USSR were whaling and they were only catching sperm, Physeter macrocephalus, fin and sei whales,Belaenoptera borealis (Merrell 1977). In the Aleutian Islands, the commercial fishery started in 1960-1962 when Canada and the USA began fishing for halibut,Hippoglossus stenolepis (Figure 2) (Forrester et al. 1978). The Japanese fishery for Pacific Ocean perch (POP), Sebastesalutus, started in 1962 (Forrester et al. 1978) and the USSR conducted experimental fisheries for both sablefish andhalibut from 1962-1964 (USFWS 1965). The Japanese longline fisheries for Pacific halibut, cod and sablefish started inthe 1960s and increased their effort in the 1970s (Figure 3). The target species off the Aleutian Islands was mainly POPand other rockfish through the 1960s and early 1970s, but POP and rockfish ceased to be important after the mid-1970s(Alverson 1992). In 1974, South Korea expanded their fleet into the central Aleutian Islands where they fished forpollock, Theragra chalcogramma, arrowtooth flounder, Reinhardtius stomias, POP, cod and halibut (USFWS 1967;1974). The foreign trawl fishery depleted stocks of POP in the 1960s and 1970s, with peak landings in 1965 (115,000tonnes (t), Figure 3) (National Research Council 2003). UBC Fisheries Centre Research Reports, Vol 13, No. 1 11From 1960, the freezer fleets began to catch halibut, sablefish, POP, herring etc. and together with the longline and gillnetfleets, they extended operations to the continental slope in the Aleutians (Forrester et al. 1978). Pollock dominated theJapanese catches after 1964, when the surimi production was introduced to factory trawlers and by 1970 it constituted88% of the Japanese total groundfish catch (Forrester et al. 1978). A foreign fishery for Atka mackerel developed in the1970s with mean annual landings of 13,000 tonnes (t) during 1972-1979 (National Research Council 2003). By 1975,the USSR catch for Atka mackerel and other rockfish (Figure 3) had increased significantly and the effort in the Aleutianswas increasing although the overall effort in the Bering Sea/Aleutian Island (BSAI) area had decreased (USFWS 1975).The BSAI groundfish fishery in 1971-1976 consisted of mostly Japanese and USSR fishers, with Japan taking 80% ofthe catch and the USSR taking the remaining 20% (Forrester et al. 1983). The Japanese groundfish fishery includedmothership-type groundfish fisheries, trawl fisheries, longline gillnet fisheries and land based dragnet fisheries (Forresteret al. 1983).USA commercial fishery operations were instituted in 1978 and increased rapidly as 'joint venture' fisheries (Alverson1992). Joint venture fisheries dominated in the 1980s with average landings of 24,000 tonnes (Figure 3), and by 1990the USA domestic fishery took over. At present, the main fisheries are for Atka mackerel, pollock, and flatfish, whichare mainly caught by trawl gear (National Research Council 2003). Atka mackerel landings increased from 47,000 tonnesin 1992 to 103,000 tonnes in 1996 (Fritz (1993) in National Research Council 2003) and decreased to approximately45,000 tonnes in 2002 (Figure 3), while total catches (including discards) were approximately 200,000 tonnes in 1996and had been reduced to about 100,000 tonnes in 2002.Since the inception of the fishery, large amounts of undesirable groundfish were discarded and estimates of these discardswere included in ?other? in Figure 3. The species most often discarded include skates (Rajidae), sharks, sculpins(Cottidae) and squids (Gaichas 2003). Most of the shark bycatch occurred in the midwater trawl pollock fishery and inthe hook and line fisheries for sablefish, Greenland turbot, Reinhardtius hippoglossoides, and Pacific cod along the outercontinental shelf and slope of the Bering Sea (Gaichas 2003). While skates were caught in almost all fisheries and areas,most of the skate bycatch were taken in the hook and line fishery for Pacific cod, with trawl fisheries for pollock, rocksole, Lepidopsetta bilineata, and yellowfin sole, Limanda aspera, also caught in significant amounts. Sculpins werecaught by a wide variety of fisheries, but trawl fisheries for yellowfin sole, Pacific cod, pollock, Atka mackerel and rocksole had the biggest impact (Gaichas 2003). Squids were mainly caught as bycatch in the midwater trawl pollock fisheryprimarily over the shelf break and slope or in deep waters of the Aleutian basin, while octopuses were mostly caught bybottom trawlers for pollock and all three of the Pacific cod fisheries (pots, longlines and trawls) (Gaichas 2003). MODEL DESCRIPTIONFor most fish and invertebrate groups, time series data on biomass and catches were obtained from stock assessmentreports or from the literature, while diets, estimates of annual P/B and Q/B ratios were obtained from a preliminary modelobtained from NMFS (Yvonne Ortiz, University of Washington and NOAA, Seattle, Wa, pers. comm.) for the AleutianIslands (NMFS model). The NMFS diet database included diet data specifically for the Aleutians, which would be betterthan the literature in most cases. As the NMFS model consisted of ~ 150 groups, with most species being split into adultand juvenile groups, I combined most of the prey groups for adults and juveniles, as well as combining the different preygroups as defined in my model. For the predator compartments, I usually used the adult diet only, as most of the biomassestimates that I had were only for the spawning stock or adult biomass; thus juvenile diets were excluded if they werenot considered in the biomass. Similarly, the annual P/B and Q/B ratios of adult fish only were mostly used. For mosttop predators the diets, annual P/B and Q/B ratios, etc., were obtained from the literature, although in some cases, I hadto revert to those given in the NMFS model.1. Transient killer whalesKiller whales, Orcinus orca, in the Aleutian Islands are divided into transient and resident killer whales. Resident killerwhales were grouped with toothed whales, while transients were placed in their own group as they feed on Steller sealions. There are two proposed communities of transients; the West Coast community (from California to SE Alaska) andthe Gulf of Alaska community, which includes the transients in our model (Ford and Ellis 1999). According to Rice(1968), killer whales were seen frequently in the Aleutian Islands, where there are many large rookeries of Steller sealions. Murie (1959) found that killer whales were common along the Aleutians in the late 1930s (1936-1938) and they generallyfound them in small groups (average size 3 per group), although they did see a pod of 25 animals. He quoted variousunpublished notes and recordings made by captains and others of large groups of 500-1,500 killer whales in the early12 Aleutian Islands models; Heymans 1900s (1913-1922) that were apparently migrating northward (p. 336 in Murie 1959), and Turner (1886 in Murie 1959)saw as many as 150 killer whales at one time in the Aleutian Islands. Murie (1959) also reported that there were ?a greatdeal of fighting accompanied by leaping at a remarkable assemblage of various whales, seals and other {sea} life?.Fiscus et al. (1981) counted 63 killer whales in the central Aleutian Islands (from the Rat Islands to the Fox Islands),which includes the Fox Islands (east of the study area) where 6 animals were counted, and exclude the Near Islands. Theystated that their records may underestimate the number of cetaceans because their emphasis was on surveying close toshore for pinnipeds (Fiscus et al. 1981). They reported groups of 30 and 11 killer whales, two groups of 7 and one groupof 2 killer whales, which could indicate that the group of 30 and 11 were residents, while the groups of 2 and 7 mightbe transients. They found that a group of 27 killer whales near Seguam pass was feeding with minke whales on a commonfood source (probably fish), thus this group was definitely not transients (Fiscus et al. 1981). They also found that therewas no close association between killer whales and sea lion haul out sites (Fiscus et al. 1981). I assumed that the estimateof killer whales in 1979 would probably be similar to the 63 counted by Fiscus et al. (1981), or 0.0003 tkm-2 transientsand 0.002 tkm-2 resident killer whales, using a 1:9 ratio. These could be underestimates, but I also used them for the 1963model.For the 1990s, Waite et al. (2002) gave an estimate of 391 killer whales in the Eastern Bering Sea, and by assuming that25% of this population is in the Aleutian Islands, that 10% of that population is actually transient killer whales(approximately 10 animals), and that the average body weight is 2,435 kg, a biomass of 0.0004 tkm-2 was obtained. Thisis similar to using the 60 animals suggested by Ford and Ellis (1999) for the Gulf of Alaska population (from just northof SE Alaska to the Aleutians), and the area of the Gulf of Alaska (429,000 km? in Aydin et al. 2003) minus that of SEAlaska (91,351 km?, S. Gu?nette, Fisheries Centre, UBC,  pers. comm.), which also gave a biomass of 0.0004 tkm-2. The annual P/B estimate for transient killer whales (0.025) was obtained from NMFS, and is marginally higher than the0.02 used by Gu?nette (this volume) i.e. 50% of rmax. The annual Q/B ratio (7.5) was estimated by using the average dailyrequirement of 73 kgday-1 for transient orcas feeding on mammals and an average weight of 3,550 kg (Barrett-Lennardet al. 1994). This estimate was used for the 1991 and 1979 models. For 1963, I increased the estimate to 10.83 as theannual food requirements were only for captive animals (see Gu?nette, this volume). Killer whales are known to feed on fur seals, walrusses, sea lions, elephant seals, harbour porpoises, Dall's porpoise,minke whales, cod, flatfishes and salmon (Parsons 1987), while the diet of the British Columbia community of transientkiller whales also includes sea otters, harbour seal, seabirds, Steller sea lions, California sea lion and Pacific white-sideddolphin (Ford and Ellis 1999), and some baleen whales (gray and minke whales) have also been reported (Ford et al.1998). Additionally, in a study by Heise et al. (2003), stomach content analysis showed that harbour seals were thepredominant prey item in all killer whale stomachs that contained marine mammals, and that they were likely a moreimportant prey item for killer whales than Steller sea lions.Barrett-Lennard et al. (1994) suggested that the proportion of Steller sea lions in the diet of transients was between 10-15% (mean 12.5%). In a survey of fishers, tour operators and others, Heise et al. (2003) found that of the 492 killerwhale/sea lion interactions, only 32 attacks were fatal, and that the ratio of pups, sub-adults, adults and unknown in theattacks were 6% pups, 16% sub-adults, 50% adults (mostly young adults) and 9% not stated. However, the highpercentage of adults in the diet could be due to the fact that attacks on adults would be more visible and last longer (Heiseet al. 2003), and is probably an overestimate. Harbour seals, which are about the same size as a small sea lion, are usuallyattacked and killed under water, with blood, oil and fragments of blubber being the only evidence of a fatal attack (Heiseet al. 2003). However, the observed kills would not necessarily represent the diet, as many of the smaller mammals(juvenile Steller sea lions and harbour seals) would not necessarily be observed. I therefore adapted the diet used byGu?nette (this volume) to include 78% small mammals, 1% birds, 4% sea otters, 16% Steller sea lions and 1% baleenwhales. I reduced the baleen whales from 3% (in Gu?nette, this volume) to 1% and used the 2% she had as import in hermodel to get a value of 4% for sea otters. The 16% of sea lions in the diet was broken down into 1% pups, 9% juvenilesand 6% adults. The fishing mortality of transient killer whales was reported at 0.4 and 0.2 killer whales respectively by the groundfishand longline fisheries (Angliss and Lodge 2002), out of a population of 346 animals, which gave a catch of 0.0000005tkm-2year-1 and 0.0000002 tkm-2year-1 respectively for those fisheries. I used this value for the 1991 model. Mackintosh(1965) suggested that killer whales were captured in small numbers, and that killer whales were a nuisance to whalersUBC Fisheries Centre Research Reports, Vol 13, No. 1 13Table 1A. Biomass estimates for beaked whales in the AleutianIslands obtained from Trites et al. (1997).BeakedwhalesAvg. bodyweight (t)PacificpopulationProportionin area 67Biomass(tkm-2)Baird's 3.1365 30,000 0.3 0.00376Cuvier's 0.8285 16,500 0.1 0.00018Stejneger's 0.455 3,000 0.5 0.00009Table 1B. Biomass of toothed whales in the Aleutian Islands in1963, 1979 and 1991 (tkm-2).Species Avg. bodyweight (t)1963 1979 1991Resident killers 2.435 0.002 0.002 0.004Sperm whale 18.519 0.007 0.004 0.004Beaked whales 0.004 0.004 0.004Total 0.013 0.010 0.012as they attacked the carcasses of larger whales before they were hauled out of the water. However, as no quantitativeinformation on catches were available or the 1979 and 1963 models, I did not include any catches for those two models.2. Toothed whalesThe toothed whales that occur in the Aleutian Islands include resident killer whales, sperm whales, Physetermacrocephalus, Baird?s beaked whales, Berardius bairdii, Cuvier?s beaked whales, Ziphius cavirostris and Stejneger?sbeaked whales, Mesoplodon stejnegeri. Belugas, Delphinapterus leucas, are rare visitors to the Aleutian Islands(Abegglen 1977) and were not included in our estimates. The only species for which relatively good estimates wereavailable are sperm whales, with the estimates of resident killer whales (see section above) and the beaked whales beingmarginal. The estimates of Baird?s, Cuvier?s and Stejneger?s beaked whales were obtained from Trites et al. (1997) andtheir average weight from Trites and Pauly (1998) (Table 1A). The calculation of biomass of these whales in theNortheast Pacific (area 7,503,000 km?) is given in Table 1B, and was used for all three time periods (1991, 1979 and1963).Waite et al. (2002) estimated the killer whale population in the Eastern Bering Sea at approximately 391 animals, andassuming that 25% of the population occurs in the Aleutians, and that 90% of the population were resident killer whales,the biomass was estimated at 0.004 tkm-2 for the Aleutian Islands. For the 1979 model, estimates of 63 killer whales weremade by Fiscus et al. (1981), and 90% of that population was assumed to be resident killer whales. This estimate (0.002tkm-2) was also used for the 1963 model. The North Pacific sperm whales are divided into the NE and NW stocks and both stocks migrate to the Aleutian Islands(Gosho and Rice 1984). The NE Pacific (Eastern temperate) stock currently consists of approximately 24,000 whales andthe NW Pacific stock 29,674 whales (Whitehead 2002). Perry et al. (1999) suggested that only the mature male spermwhales move north into the Aleutian Islands waters in the summer, although Nishiwaki (1966) did find that a few femaleswere caught in years when the water temperature was above normal, and they caught females around Attu and KiskaIslands in 1961. However, Nishiwaki (1966) also stated that the presence of females is very rare, so I assumed that onlylarge males go that far north. Using a ratio of 72% adults (from Gu?nette, this volume), the male to female ratios for NEand NW Pacific from Gosho and Rice (1984), and assuming that the population is in the North Pacific for 120 days a year(Calkins 1987), yielded an estimate of about 9 large male sperm whales in the Aleutian Islands. Using the average weightof large males (26,939 kg) the biomass of sperm whales in the Aleutian Islands in 1999 was 0.004 tkm-2. The globalpopulation had decreased from 1963 to 1979 and increased thereafter (Whitehead 2002), and using the same ratio ofdecrease and increase as in the global population, gave a biomass of 0.004 t km-2 in 1991 and 0.0036 tkm-2 in 1979 and0.007 tkm-2 for 1963 (Table 1B). The total biomass for toothed whales was estimated at 0.012, 0.010 and 0.013 tkm-2for 1991, 1979 and 1963 respectively.Estimates of the annual P/B ratio for sperm whales in the western sub-Arctic region were obtained from Aydin et al.(2003) and that of resident killer whales from NMFS, and prorated by biomass to give average estimates of 0.029 year-1for the 1991 model, 0.028 year-1 for 1979 and 0.036 year-1 for 1963. The annual Q/B estimates for sperm whales (9.4)and Stejneger?s beaked whales (14.4) were calculated from the energy requirements, energetic values of their food, andaverage weight acquired from Perez and McAllister(1993). The annual Q/B ratio for resident killerwhales was estimated by using the average dailyrequirement of 84.3 kgday-1 for resident killerwhales feeding on fish and an average weight of3,550 kg (Barrett-Lennard et al. 1994), to give anannual Q/B of 10.8. These Q/B estimates were thenprorated by biomass to give annual Q/B ratios forthe group of 10.1, 11.7 and 11.1 respectively for the1991, 1979 and 1963 models.The diet of resident killer whales in the West CoastCommunity (SE Alaska and south) includedchinook, pink, coho, chum and sockeye salmon, aswell as steelhead and other fish such as herring,rockfish and halibut (Ford and Ellis 1999). Perez(1990) and Ford et al. (1998) suggested that14 Aleutian Islands models; Heymans resident killer whales consume herring, salmon, capelin, smelts, Pacific cod, Arctic cod, saffron cod, Atka mackerel,Pacific halibut, other flatfish, sharks, skates, cephalopods, euphausiids, copepods, amphipods, other invertebrates, androckfish. Deep-water cephalopods are the main food for sperm whales (Okutani and Nemoto 1964; Gosho and Rice1984), but their diet also included salmon, lanternfish, lancetfish, Pacific cod, pollock, saffron cod, rockfish, sablefish,Atka mackerel, sculpins, lumpsuckers, lamprey, skates, rattails, cephalopods, amphipods and other invertebrates (Perez1990). The diet of Stejneger?s beaked whales is not well known, except that they fed predominantly on squids (Loughlinand Perez 1985). The diet of all toothed whales combined was estimated by using the diets of the various species inproportion of their biomass and the proportion of biomass of the prey species where available (Table 2). For the differentmodels, different biomass estimates of both whales and fish were used to calculate different diets, specifically for 1991and 1979, while the 1979 diet was also used for 1963 as very little data was available on prey density (Table 2). Inaddition, the proportion of small pelagics (5%), small demersals (3%) and myctophids (2%) in the diet of toothed whaleswere assumed as no biomass values were available for these groups. The different biomass estimates for both toothedwhales and fish in 1979 and 1991 were used to estimate different diets for toothed whales, while the diet in 1963 wasestimated using the 1979 fish biomass and the 1963 toothed whale estimates as no data were available of fish biomassin that time period. Table 2. Diet composition of toothed whales of the Aleutian Islands in 1991 and 1979.Resident killerwhales (%)Sperm whale(%)Toothed whale (proportion)Species/group 1991 1979 1991 1979 1991 1979 1963Skate 0.04 0.08 0.04 0.08 0.0004 0.0008 0.0008Salmon 61.75 61.75  0.2882 0.2468 0.1591    Capelin, sand lance, smelts ? ?      Arctic cod ? ?  Pelagic small invertebrate feeders   0.0500 0.0500 0.0500Atka mackerel 1.767 1.587 1.704 1.520 0.0173 0.0155 0.0154Herring ? ?  0.0080 0.0076 0.0076pollock 0.734 1.152 0.71 1.104 0.0072 0.0112 0.0112Pacific ocean perch  0.08 0.04 0.0004 0.0002 0.0003Rockfish 0.05 0.11 0.05 0.09 0.0005 0.0009 0.0009Sablefish  0.12 0.17 0.0006 0.0010 0.0012Pacific cod 0.57 0.26 0.55 0.25 0.0056 0.0025 0.0025Pacific halibut 0.08 0.08  0.0004 0.0003 0.0002     Saffron cod ? ? ? ?     Sculpin  ? ?S & M demersals   0.0300 0.0300 0.0300     Deep-sea smelt /lanternfish ? ? ? ?  Myctophids   0.0200 0.0200 0.0200Cephalopods 20 20 82 82 0.5307 0.5722 0.6603Euphausiids 1 1  0.0100 0.0100 0.0100Copepods 1 1  0.0100 0.0100 0.0100Other invertebrates 2 2 2 2 0.0200 0.0200 0.0200Catches, obtained from the International Whaling Commission for the Northeast Pacific, the coastal Northwest Pacificand the pelagic whaling fleet (Gu?nette and Salter, this volume), were used to prorate catches in the Aleutian Islands.In 1963, 15,548 sperm whales were caught in the North Pacific, but I only used 8% of this catch as the biomass of spermwhales in the Aleutian Islands was only 8% of the biomass in the North Pacific (72% adults, and 30% males in thewestern stock and 40% males in the eastern stock, prorated by area). Thus, the catches taken from the sperm whales inthis area were 0.0006 tkm-2year-1 in 1963, 0.00012 tkm-2year-1 in 1979, and there were no commercial catches in 1991.However, the fishery catches of resident killer whales was reported to be 1.4 killer whales out of a population of 723animals (0.6 by the groundfish fishery and 0.8 by the longliners) (Angliss and Lodge 2002), which amounted to catchesof 0.000003 tkm-2year-1 and 0.000004 tkm-2year-1 respectively for the groundfish and longline fisheries in 1991. Thetime series of catches for baleen whales (blue, fin, humpback and sei) and toothed whales (sperm) for the whole NorthPacific are given in Figure 4 and were used as proxy for catch time series in the Aleutian Islands. UBC Fisheries Centre Research Reports, Vol 13, No. 1 150100,000200,000300,000400,000500,000600,0001900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000North Pacific catch (tonnes)Baleen ToothedFigure 4. Catch (tonnes) of baleen and toothed whales in the North Pacific from 1900 to 2000.3. Baleen whalesThe baleen whales of the area include minke, Balaenoptera acutorostrata, fin, B. physalus, sei, B. borealis, blue, B.musculus, and humpback whales, Megaptera novaeangliae. Gray whales, Eschrichtius robustus, were not included asthey do not occur in the central and western Aleutian Islands (Murie 1959), but pass through the eastern part of the islands(Abegglen 1977). Pacific right whales, Eubalaena glacialis glacialis, were recorded in the waters of the Aleutian Islandsand two were killed at the Akutan whaling station in 1914 (Murie 1959), but Perry et al. (1999) found that they did occursouth of the Aleutian Islands in the summer. Similarly, bowhead whales, Baelaena mysticetus, used to visit the Aleutians(Murie 1959), but neither of these species are known to frequent the Aleutians anymore (Nishiwaki 1967). I thereforedid not include any information on gray whales, Pacific right whales or bowhead whales in these models.Humpback, fin and right whales feed in both the Gulf of Alaska and the Bering Sea during the summer and early fall,while blue, sei and sperm whales are more restricted to the North Pacific or deeper western Bering Sea (Brueggeman etal. 1987). Nishiwaki (1966) suggested that there were many humpbacks in the Aleutians, but that the number of themtaken was rather small. Humpback whales migrate between the Aleutians and the warm waters of the western NorthPacific (Mackintosh 1965). Most humpback whales (69%) were observed on the continental shelf, while 30% are seenin waters > 2000 m deep and only 1% on the slope (Brueggeman et al. 1987). Most of the humpbacks observed byBrueggeman et al. (1987) were seen in the Shumagin Planning Area (from 156W to 165W south of the Alaskanpeninsula). Angliss and Lodge (2002) estimated that there were 1,175 humpback whales in the central Bering Sea duringthe summer, and also found that the majority of the sightings for humpbacks were close to the Aleutian Islands, whilethe minimum population estimated for the area was 367 humpbacks, assuming that they only stay in the area for thesummer (25% of the time). This estimate is closer to that obtained from Trites et al. (1997) (220 animals), althoughNMFS, quoting estimates made by Zerbini et al. (2003), calculated 268 humpbacks staying for the whole year, thusgiving an estimate of 0.14 tkm-2, which is the estimate I used for the 1991 model. Calkins (1987) estimated the totalNorth Pacific population of humpbacks at approximately 1,200 animals (0.001 tkm-2), which was similar to those ofJohnson and Wolman (1984 in Perry et al. 1999) and I used this estimate for the 1979 model. For 1963, I used an estimateof 1,000 animals (or 0.0008 tkm-2) obtained from Rice (1978 in Perry et al. 1999).The estimates for fin whales (0.01 tkm-2) obtained from Perez (1990) were lower than the estimates obtained by NMFS(0.044 tkm-2) based on Zerbini et al. (2003), but when a more appropriate estimate of body weight (37 vs. 56 tonnes,Nancy Friday pers. comm.) was used, the latter biomass was drastically reduced to 0.03 tkm-2. The 1970s biomass wasestimated as 0.048 tkm-2 byPerry et al. (1999) and the 1963estimate for the North Pacificwas 27,788 fin whales (Gu?netteand Slater, this volume) or 0.03tkm-2.I used the sei whale abundance(21 animals, 0.006 tkm-2)obtained from NMFS quotingZerbini et al. (2003) for 1991,which is higher than that givenby Perez (1990). Calkins (1987)estimated that the total NorthPacific population of sei whalesin the 1970s was approximately8,600, which is similar to the9,110 sei whales estimated byPerry et al. (1999), giving abiomass of 0.004 tkm-2. Calkins (1987) also estimated that the number of sei whales in the North Pacific in the 1960swas approximately 42,000 (0.018 tkm-2), which is similar to the total unfished population given by Perry et al. (1999).Minke whales are divided into two stocks in the North Pacific and the boundary between these stocks runs through theAmchitka Pass (Gosho and Rice 1984; Parsons 1987) and through the middle of this study site. The estimates for minkewhales obtained from Perez (1990) where lower than the estimate of 846 minkes obtained from NMFS (0.09 tkm-2)16 Aleutian Islands models; Heymans calculated based Zerbini et al. (2003). According to Perry et al. (1999), the numbers of minke whales have stayed fairlyconstant over time, thus I assumed that the biomass in all three time periods were 0.09 tkm-2. Blue whales have been recorded in both the central and western Aleutian Islands and have been protected since 1966(Abegglen 1977). The biomass of blue whales (0.003 tkm-2) were estimated from Trites et al. (1997) and their averageweight from Trites and Pauly (1998). This estimate was used for both the 1991 and 1979 model. According to Mackintosh(1965), only about 2,000 blue whales and 40,000 fin whales existed globally by 1964. If we assumed that the ratio of bluewhales in the North Pacific vs. the global population was similar to that of fin whales, the total North Pacific blue whalepopulation in 1963 would be approximately 1,032 blue whales, or 0.0026 tkm-2. This is probably an upper limit, as fewerblue and fin whales frequent the northern hemisphere than the south, and blue whales in particular do not go north of theAleutians according to Mackintosh (1965). Thus, the total biomass for baleen whales were 0.28, 0.153 and 0.145 tkm-2 for 1991, 1979 and 1963 respectively. Currentbest estimates of baleen whales were given by Perry et al. (1999) as 6,000-8,000 humpbacks, 3,300 blue whales, 14,620-18,630 fin whales and 9,110 sei whales in the North Pacific. According to Perry et al. (1999) the numbers of minkewhales have stayed fairly constant over time, thus I assumed that the current minke biomass is similar to that of 1991,which gives a biomass estimate for baleen whales in 1999 of around 0.13 tkm-2. However, the estimates given by Perryet al. (1999) of fin and humpback whales are less than those calculated by NMFS from Zerbini et al. (2003), which wouldincrease the best 1999 estimate to 0.282 tkm-2.The annual P/B and Q/B estimates of fin, sei and minke whales were given by Aydin et al. (2003) and prorated bybiomass to give average annual P/B and Q/B ratios for baleen whales over time. The annual P/B ratios stayed constantat 0.02 as all three whale species had an annual P/B of 0.02, while the annual Q/B ratios differed over time, from 6.99in 1991, to 6.7 in 1979 and 6.99 in 1963. Perry et al. (1999) quoted various authors that give annual natural mortality ratesof 0.04 for blue and fin whales, 0.07 for sei whales, and survival of 0.95 for humpbacks, and prorating these ratios bytheir biomass estimates for the three time periods gave annual natural mortalities of 0.06, 0.04 and 0.05 for the 1963, 1979and 1991 models. Adding to that the annual fishing mortality for each time period (0.04, 0.0003 and 0 for 1963, 1979and 1991) gave annual P/B estimates of 0.1, 0.04 and 0.05 respectively. The diet of this group was obtained by using the average minke, fin, sei and humpback whale diets. The differentabundances of fish and whales between 1991 and 1979 were incorporated into the diet of baleen whales in those twomodels. For 1963, the proportion of fish in 1979 was used as proxy for the proportion of fish in 1963 (Table 3). Minkewhales feed on fish (60%), cephalopods (1%), euphausiids (30%) and copepods (9%) (Perez 1990). Tamura et al. (1998)suggested that minke whales in the central Pacific consumed salmon (1%), pomfrets and other large pelagics (4.5%),saury (80.6%), northern anchovy (7.1%) and some zooplankton (~7%). I used this information to estimate the diet ofminke whales using 30% of the fish from those species for which I had biomass estimates (Atka mackerel and pollock)and assumed that the other 30% of the fish comes from capelin and Arctic cod (10% each), sand lance and saffron cod(5% each). Fin whales consumed fish (16%), cephalopods (2%), euphausiids (55%), copepods (27%) and otherinvertebrates (1%) (Perez 1990). For fin whales, I assumed that the 16% fish consisted of 9% Atka mackerel, rockfishand Pacific cod in the ratio of their biomass, and the other 7% was divided equally between salmon, capelin, Arctic cod,sand lance, herring, juvenile pollock, and saffron cod. The diet of sei whales consisted of 3% fish, 1% cephalopods, 13%euphausiids, 83% copepods, with the 3% fish split equally between smelts, capelin, sand lance, Arctic cod, sardine,pollock, rockfish and greenling (Calkins 1987). Lowry et al. (1989) suggested that the favourite food of humpback whalesin the Aleutian Islands is Atka mackerel, but they also consumed other fish (29%), cephalopods (1%), euphausiids (69%)and copepods (1%) (Perez 1990). The 29% fish was divided into 19% Atka mackerel, pollock and rockfish (in the ratioof their biomass), and 2% each of salmon, capelin, Arctic cod, sand lance and saffron cod. No diet estimates wereavailable for blue whales, but Nemoto (1957) suggested that they feed nearly exclusively on euphausiids and Perry etal. (1999) found that they feed extensively on krill, euphausiids and red crabs.Mackintosh (1965) gave estimates for the average annual global catch of blue, fin, sei and humpback whales and alsostated that 90% of the worlds catch of fin, blue and humpback whales were made in the southern hemisphere, thus givinga catch of 3,133 fin whales, 651 sei whales, 172 blue whales and 360 humpback whales in the northern hemisphere. Theseestimates were only made by the pelagic whaling fleet. From the IWC whaling data (Gu?nette and Salter, this volume),the total catches of whales in the North Pacific in 1963 were 2,140 fin whales, 4,291 sei whales, 2,339 humpbacks and404 blue whales, giving a total catch of 263,946 tonnes in the North Pacific, which included the Northeast Pacific catch,the pelagic catch in the North Pacific and the coastal catch in Northwest Pacific, and gave a catch of 0.006 tkm-2year-1UBC Fisheries Centre Research Reports, Vol 13, No. 1 17for the whole North Pacific (Areas 61, 67 and ? of 77 = 40,203,750 km?). The time series of catches for baleen whalesare shown in Figure 4 above. In 1979, 43 fin whales were caught by the northwest Pacific fishery, with no catch of sei,humpback or blue whales in that year, giving a catch of 0.00004 tkm-2year-1. By 1991 no commercial catches were made,but the mean annual catch of humpback whales in the western north Pacific stock by the groundfish trawl fishery was0.4 animals from a population of 367 (Angliss and Lodge 2002), which gave a catch of 0.0002 tkm-2year-1. The annualcatch of fin whales by the groundfish fishery was 0.6 animals (Angliss and Lodge 2002) or 0.0006 tkm-2year-1, for a totalfishery mortality of baleen whales by the groundfish fishery of 0.0007 tkm-2year-1. Table 3. Diet compositions of baleen whales in the Aleutian Islands in 1991, 1979 and 1963. Note that the diet for sei whalesdid not change in the model between 1979 and 1991.Minke  whale(%)Fin  whale(%)Sei(%)Humpback(%)Baleen whale (proportion)Species/group 1991 1979 1991 1979 1991 1979 1991 1979 1963Salmon  1 1  2 2 0.01 0.01 0.002Capelin, sand lance,   0.4 2 2Capelin 10 10 1 1 0.4  Arctic cod 10 10 1 1 0.4 2 2Sardine/saury   0.4  Pelagic invertebratefeeders     0.09 0.094 0.15Atka mackerel 23.9 20.3 6.6 7.4  14.8 12.4 0.15 0.124 0.137Sand lance 5 5 1 1 0.4 2 2 0.03 0.029 0.038Herring  1 1   0 0 0.001Juv pollock  1 1   0 0 0.001pollock 6.1 9.7  0.4 3.8 5.9 0.06 0.09 0.092Rockfish  0.2 0.4 0.4 0.5 0.7 0 0.004 0.001Pacific cod  2.1 1.2   0 0.001 0.002Saffron cod 5 5 1 1  2 2Greenling   0.4  S & M demersals     0.03 0.029 0.038Cephalopods 1 1 2 2 1 1 1 0.011 0.011 0.01Euphausiids 30 30 55 55 13 69 69 0.524 0.524 0.315Copepods 9 9 27 27 83 1 1 0.09 0.09 0.2114. Steller sea lions (embryo, pups, juveniles and adults; Groups 4-7)Steller sea lions, Eumetropias jubatus, are found throughout southwestern Alaska from Attu Island east to Carlisle Island(Murie 1959). They have been caught for consumption by native fishermen as well as being taken as bycatch by thedomestic and foreign trawl and longline fisheries. According to Alverson (1992), the fisheries that had the greatest impacton Steller sea lions were the Japanese and US salmon fisheries, Japanese, Soviet and US herring fisheries, foreign andUS groundfish, shrimp, longline and crab operations. The average catch by the groundfish trawl and longline fisheriesfrom 1979-2000 were estimated from Perez and Loughlin (1991), Perez (2003) and Berger et al. (1986). The bycatch for1989-1991 was estimated at approximately 6.6 animals by trawlers and 1.8 animals by longliners respectively, assumingthat the unidentified pinnipeds consisted of both harbour seals and Steller sea lions. The estimates of sea lion shootingsin defense of fish gear from 1960-1990 and harvest are not really well known, but the government was seeking ways toreduce the number of Steller sea lions and other marine mammals and had put a bounty on harbour seals (Alverson 1992).Commercial harvests of sea lions were authorized and cannery operators provided ammunition to fishermen. Even marineand wildlife biologists were known to have joined in the shooting, as there was no dishonour in shooting sea lions orusing them for crab bait (Alverson 1992).Alverson (1992) suggested that approximately 150 animals were taken annually statewide as subsistence harvest. Lossesfrom entanglement in marine debris were not assumed to be a major factor, with fewer than 100 animals (he used 97animals) killed each year (Alverson 1992). The average subsistence take was given by Wolfe et al. (2002) for Atka Island(the only community that they studied in the Western Aleutian Islands) from 1992-2002 by sex and for adults, juvenilesand pups (Table 4). The average weight for adult males (430 kg) and females (229 kg), unknown adults (286 kg), juvenilemales (152 kg), juvenile females (123 kg), unknown juveniles (132 kg), male pups (22 kg), female pups (20 kg) andunknown age and sex (162 kg), obtained from the age-structured model (Gu?nette, this volume), were used to calculatethe average weight of the adult, juvenile and pup catch by year.  The number of animals discarded (struck and lost) were18 Aleutian Islands models; Heymans Table 5. Steller sea lions caught andlost (discards) by First Nations in the1979 model of the Aleutian Islands(in kgkm-2year-1). Harvest DiscardsPup 0 0.0001Juvenile 0.023 0.014Adult 0.06 0.036assumed to be in the same proportion as the adults and juveniles (assuming that no pups were lost), and using the averageweight for adults and juveniles. For 1999, no estimate of subsistence catch was available and I assumed that it was theaverage of 1998 and 2000. Table 4. Subsistence catch (in numbers) of Steller sea lions by Western Aleutian Island communities obtainedfrom Wolfe et al. (2002).Group 1992 1993 1994 1995 1996 1997 1998 2000 2001 2002Adult male 8.8 4.4 2.2 10 6.5 7.6 6.1 6.1 10.0 46.5Adult female 9.9 10.9 20.6 12.5 6.5 0 3.1 3.1 1.7 6.0Unknown adults 0 0 5.4 0 0 0 3.1 3.1 0 3.0Juvenile male 2.2 2.2 14.1 2.5 3.2 1.5 3.1 3.1 0 13.5Juvenile females 4.4 2.2 3.3 5 0 3.1 0 0 0 1.5Unknown juv 1.1 0 0 0 1.1 0 1.5 1.5 0 0Male pups 1.1 0 0 0 0 0 0 0 0 0Female pups 1.1 0 0 0 0 0 0 0 0 0Unknown 0 5.5 0 10 0 0 0 0 21.7 3.0Total Harvest 28.6 25.1 45.5 40 17.3 12.2 16.8 16.8 33.3 73.5Struck & Lost 9.9 0 8.7 0 0 0 0 0 11.7 12.0Note: No data available for 1999For the 1979 model, the subsistence estimate of between 15 to 25 animals fromVeltre and Veltre (1983) was used. They stated that sea lions were huntedthroughout the year and that adult and juvenile males were preferred. Veltre andVeltre (1983) also estimated that only 60% of sea lions killed were retrieved, and40% discarded. Thus, using the breakdown of adults, juveniles and pups from the1994 subsistence fishery on Atka Island, Wolfe and Mishler (1995) gave asubsistence catch for 1979 of about 20 harvested and 12 struck and lost (Table 5).I assumed that all subsistence catches prior to 1992 were similar to the estimatesgiven by Veltre and Veltre (1983).Thus, the subsistence catch of Steller sea lions prior to 1992 can be estimated at 32 killed per year (Table 6) and thenumbers for 1992-2002 are given in Table 4. Incidental and intentional kills by the trawl fleet, salmon fisheries and otherfisheries as well as entanglements in marine debris were obtained from Trites and Larkin (1992) and were prorated bythe area of the Aleutian Islands (56,936 km?) to that of the Gulf of Alaska (348,776 km?). Salmon catches were alsoweighted by the ratio of salmon caught in the Aleutians vs. the whole Gulf of Alaska. For intentional shootings I usedthe estimates given by Alverson (1992) prorated by area. Alverson (1992) estimated on average 290 shootings per yearand gave estimates of between 100 and 1,455 animals shot annually by the salmon fisheries between 1960 and 1990. For1991-2002 the kills by the trawl fleet was obtained from Perez (2003) and I assumed that kills by the other fleets, marinedebris and indiscriminate shooting did not change from 1990-2002, except for the salmon fleet which I assumed did notkill any Stellers as they hardly caught any salmon in that time. The bycatch of Steller sea lions by the fishery wasassumed to be juveniles and young adults (years 1-12) in the ratio of the abundances in the general population (Table 6).The catches were prorated for adults and juveniles for each year and gear type, with all incidental harvest by other fleets,indiscriminate shooting and marine debris being attributed to other gear in the model. The population estimates given by Trites and Larkin (1996) were multiplied with the average weight of the animals toobtain the biomass and stanza information given in Table 7. The annual P/B estimates were calculated for each of theyears by calculating the slope of the natural log of numbers at age for pups, juveniles and adults respectively and wereused in conjunction with the annual Q/B estimates to calculate the biomass of the other stanzas in the three models (Table7). The Q/B ratio was calculated from the energy requirements, energetic values of their food, and average weight givenby Perez and McAllister (1993) as 27.4 year-1, which is marginally higher than the 24.1 year-1 estimated by NMFS. Itherefore used a value somewhere in between (25.6 year-1), which is similar to that given by Gu?nette (this volume) forthe SEAK model. The juvenile rate was estimated at between 39.3 year-1and 39.6 year-1  (Table 7) per year which is inthe range of 1.4-1.8 times the Q/B of adults given by Innes et al. (1987). UBC Fisheries Centre Research Reports, Vol 13, No. 1 19Table 6. Known and estimated Steller sea lion kills in the Aleutian Islands by subsistence and commercial fisheries.Year Sub-sistenceIncidental harvest / shooting IndiscriminateshootingMarinedebris TotalDeathsTotal (t)Trawlers Salmon Other Juveniles Adults1960 32 0 52 15 65 8 172 9 291961 32 0 15 15 65 9 137 7 231962 32 0 198 19 82 10 340 18 571963 32 3 16 21 82 11 164 9 281964 32 17 24 18 114 11 217 12 361965 32 24 0 18 114 12 200 11 341966 32 15 20 21 131 13 232 13 391967 32 20 5 22 131 14 223 12 371968 32 21 138 22 131 15 358 20 601969 32 22 41 22 147 16 279 15 471970 32 29 77 24 147 16 325 18 541971 32 43 6 27 163 16 287 16 481972 32 23 1 34 163 16 270 15 451973 32 46 1 35 196 16 327 18 551974 32 46 0 33 196 16 323 18 541975 32 42 38 47 229 16 404 22 681976 32 42 0 55 229 16 374 20 631977 32 20 0 54 229 16 351 19 591978 32 16 5 71 261 16 402 22 671979 32 20 44 73 261 16 447 24 751980 32 23 122 72 261 16 526 29 881981 32 21 21 90 261 16 443 24 741982 32 56 67 92 294 16 557 30 931983 32 28 3 104 294 16 478 26 801984 32 49 96 96 294 16 583 32 981985 32 42 0 81 294 16 466 26 781986 32 17 3 135 278 16 480 26 801987 32 6 0 80 278 16 413 23 691988 32 3 3 53 114 16 221 12 371989 32 1 0 25 33 16 107 6 181990 32 1 2 17 16 16 84 5 141991 32 10 0 17 16 16 91 5 151992 39 0 0 17 16 16 88 5 151993 25 0 0 17 16 16 75 4 121994 54 2 0 17 16 16 106 6 181995 40 2 0 17 16 16 91 5 151996 17 2 0 17 16 16 69 4 121997 12 3 0 17 16 16 65 4 111998 17 4 0 17 16 16 70 4 121999 17 5 0 17 16 16 71 4 122000 17 1 0 17 16 16 67 4 112001 45 0 0 17 16 16 94 5 162002 86 0 0 17 16 16 135 7 2320 Aleutian Islands models; Heymans Attu01000200030004000500060001950 1960 1970 1980 1990 2000Agattu0200040006000800010000120001950 1960 1970 1980 1990 2000Buldir0100020003000400050006000700080001950 1960 1970 1980 1990 2000Kiska0100020003000400050006000700080001950 1960 1970 1980 1990 2000Ayugadak020040060080010001200140016001950 1960 1970 1980 1990 2000Amchitka0500100015002000250030003500400045001950 1960 1970 1980 1990 2000Abundance of sea lionsAbundance of sea lionsFigure 5. Counts of non-pups (open squares) and pups (closed triangles) on the rookeries of the Western AleutianIslands.UBC Fisheries Centre Research Reports, Vol 13, No. 1 21Tag05001000150020001950 1960 1970 1980 1990 2000Gramp Rock050010001500200025001950 1960 1970 1980 1990 2000Adak050010001500200025003000350040001950 1960 1970 1980 1990 2000Kasatochi050010001500200025001950 1960 1970 1980 1990 2000Agligadak05001000150020002500300035001950 1960 1970 1980 1990 2000Seguam01000200030004000500060001950 1960 1970 1980 1990 2000Semisopochnoi050010001500200025003000350040001950 1960 1970 1980 1990 2000Ulak05001000150020002500300035001950 1960 1970 1980 1990 2000Yunaska020040060080010001200140016001950 1960 1970 1980 1990 2000Abundance of sea lionsAbundance of sea lionsFigure 6. Counts of non-pups (open squares) and pups (closed triangles) on the rookeries of the Central AleutianIslands.22 Aleutian Islands models; Heymans 015,00030,00045,0001955 1960 1965 1970 1975 1980 1985 1990 1995 2000Counts020,00040,00060,00080,000100,000Population estimateNon-pups Pups Trites and Larkin (1996) Age structured modelFigure 7. The total counts and population estimates of non-pups (squares) and pups(triangles) in the western and central Aleutian Islands. Open squares and opentriangles are respectively non-pups and pups not included in the model.There was no evidence to suggest that  either growth or body condition was worse for western stock animals than it isfor juveniles from SEAK (Pitcher 2002). On the contrary, although mass at birth were similar between pups in SEAKand the west, growth rates appeared higher in the west, and body composition estimates suggested better conditions inwestern stocks (Pitcher 2002). Andrews et al. (2002) suggested that adult female Stellers found suitable prey morequickly and were able to ingest prey at a much higher rate in the central Aleutians than in SEAK. They suggested thatthis might be the reason why the growth rate for pups measured at the central Aleutian Islands was double that in SEAK(Andrews et al. 2002). Unfortunately, no data was available for the western Aleutians, thus to be conservative, I used aK of 0.282 year-1 and a Wmax/W of 0.00001, (see Gu?nette, this volume). Table 7. Stanza information used in the 1991 and 1979 models. Note values in Italics were estimated by Ecopath.Stanza time(months)Biomass tkm-2 P/B (year-1) Q/B (year-1)1991 1979 1963 1991 1979 1963 1991 1979 1963Embryo 0-6 0 0 0.00005 0 0 0.02 219.614 219.237 220.786Pup 7-19 0 0.003 0.002 0.52 0.51 0.52 82.955 82.822 83.413Juvenile 20-56 0.018 0.042 0.030 0.23 0.24 0.24 39.313 39.279 39.562Adult 57- 0.083 0.184 0.148 0.19 0.18 0.17 25.550 25.550 25.55The diet of Steller sea lions were obtained as % frequency of occurrence from scat samples (Sinclair and Zeppelin 2002)for region 4 (which is mostly the central and western Aleutian Islands), and used as the best representation of their dietin weight. The diet estimates were given for summer and winter, and the average of these two seasons were taken foradult diet (Table 8), while the winter diet was assumed to be more representative for juveniles, as juveniles are not reallypresent on the rookeries much during the summer (Sinclair and Zeppelin 2002). The diet used in this study was probablymostly those of female and young-of-the-year Steller sea lions (Sinclair and Zeppelin 2002). There is no data availableon the diet of Steller sea lions in the 1970s or 1960s in the Central and Western Aleutians and I therefore used this dietfor all three models, although there might have been changes over time. Merrick (1996) found that in the EasternAleutians(around Kodiak Island) the diet consisted of 1.7-42.7 cm fork length (FL) fish, with the average mean FL for juvenilesbeing 20.8 cm and that of adults 27.9 cm, thus I could have both juveniles and adult Steller sea lions feeding on bothjuvenile and adult pollock. Using the average length for 2 year old pollock of 28 cm, the number of prey in each lengthclass from Merrick (1996) and the von Bertalanffy equation, Gu?nette (this volume) found that juvenile sea lions wouldconsume 65% juvenile pollock and 35% adult pollock, while adult sea lions would consume 21% juvenile pollock and79% adult pollock. UBC Fisheries Centre Research Reports, Vol 13, No. 1 23Table 8. Diet for Steller sea lion adults and juveniles(using the winter diet) used for the 1979 and 1991 models.Prey Summer Winter(Juveniles)Average(Adults)Mammal * *Skates & sharks 0.006 0.007 0.007Salmon 0.0934 0.119 0.106Atka mackerel 0.561 0.327 0.427Sand lance 0.006 0.005Pacific herring * *Pollock 0.058 0.014 0.033Rockfishes 0.014 0.017 0.016Pacific cod 0.039 0.085 0.063Halibut * * *Arrowtooth 0.006 0.014 0.010Flatfish 0.012 0.039 0.027Demersals S & M 0.050 0.246 0.162Large demersals 0.027 0.064 0.047Myctophids 0.015 0.010 0.016Cephalopods 0.110 0.058 0.081* <1 Frequency of occurrenceThe diet (Table 8) therefore consisted of sharks and rays(0.7% for both juveniles and adults), salmon (12% juveniles,11% adults), Atka mackerel (33 and 43% respectively), adultand juvenile pollock (1% each for adult and juveniles eaten byadult and juvenile Stellers respectively), rockfish (2%),Pacific cod (9 and 6% respectively), arrowtooth (1% each),flatfish (4% and 3% respectively), small demersals (25% and16%), large demersals (6% and 5%), myctophids (1% and2%), cephalopods (6% and 8% respectively for juvenile andadults Stellers).8. Small mammalsThis group contains both pinnipeds and cetaceans. Thepinnipeds include northern fur seals, Callorhinus ursinus,ribbon seals, Phoca fasciata, spotted seals, P. largha, andharbour seals, P. vitulina stejnegeri. The harbour seals in theAleutian Islands is a different sub species from those in therest of the Gulf of Alaska (P.v. richardii), as the boundarybetween the two subspecies are in the eastern Aleutian Islands(Burns 2002). Murie (1959) suggested that harbour seals werenot particularly abundant in the Aleutians, and they sightedsingle animals or small groups only, but in 1925 they wereabundant and hauled out on kelp covered boulders near the beaches of Adak Island. Fur seals are not normally found in the Aleutian Islands in great numbers (Veltre and Veltre 1983). They generallymigrate between their breeding grounds in the Pribilof Islands and their wintering territories south of the Aleutiansthrough the passes of the eastern Aleutians but as fur seal females can travel up to 800 km between successive nursingperiods (Scheffer et al. 1984), they could feed in the Aleutian Islands. Archaeological information showed that their boneswere found throughout the archipelago, while historic and contemporary reports indicated that they were regularly spottedin low numbers near Atka Island (Veltre and Veltre 1983) and the Aleuts told that fur seals hauled out on Buldir Islandand even bred there (Murie 1959).The cetaceans include Dall?s porpoise, Phocoenoides dalli, and harbour porpoise, Phocoena phocoena. Buckland et al.(1993) did not observe Pacific white-sided dolphins, Lagenorhynchus obliquidens, anywhere near the Aleutian Islands,so I did not include them in the estimates. There are no estimates of harbour porpoise in the Aleutian Islands and so theywere not included in this model. Dall?s porpoise is the most abundant cetacean (>10,000 inidviduals) in the BSAI area(Loughlin et al. 1999). By 1990, there were approximately 400,000 northern fur seals in the summer BSAI population (with an average massof 30 kg) and a tenth of that in winter (70 kg) (Perez 1990), giving a total biomass of 14,785 tonnes or 0.001 tkm-2 if weassumed that they only stay in the Aleutian Islands for 2 months. This estimate is similar to the 33,100 fur seals in thearea between 45N:170E and 55N:170W obtained from the North Pacific Marine Mammal sighting database for 1990(Buckland et al. 1993), which estimated a biomass of 0.0006 tkm-2 using an average adult body weight of 28 kg (Tritesand Pauly 1998). The northern fur seal population in the Eastern Pacific had declined from 1.25 million in 1974 at a rateof 6.5-7.8% per year into the 1980s (York (1987) in Angliss and Lodge 2002). Using this decrease, the total populationin 1979 was estimated at approximately 900,000 (Angliss and Lodge 2002), or a biomass of 0.0005 tkm-2. This is similarto the estimate I used of 0.0007 tkm-2 obtained from Anonymous (2001) and using the ratio between 1979 and 1990 withthe 1990 biomass from Buckland et al. (1993). For 1963, I used the ratio of 1963 to 1991 biomass in the time seriesobtained from Anonymous (2001), to give a biomass of 0.0008 tkm-2.Harbour seal abundance was estimated at 3,437 for the Aleutian Islands in 1994 (Withrow and Loughlin (1995) in Anglissand Lodge 2002) and that included 1,600 animals in a smaller area that had approximately 1,000-2,500 animals in a countin 1975-77 (Small (1996) in Angliss and Lodge 2002). I scaled up this estimate for the smaller area in 1975-77 to thewhole area for an estimate of 3,759 seals. I used the number estimated by the trawl survey for the Aleutian Islands (Small(1996) in Angliss and Lodge 2002), the total area of 56,938 km? and an average body weight of 115 kg for the westernsubspecies, to get a biomass of 0.007 tkm-2 and 0.008 tkm-2 for 1994 and 1979 respectively. According to Kenyon,24 Aleutian Islands models; Heymans 0501001502002501956 1966 1976 1986 1996Otters (tonnes) .02004006008001,0001,2001,400Small mammal (tonnes)  .Otter Small mammalsFigure 8. Biomass (in tonnes) of sea otters and small mammalsavailable from 1959 to 2000.quoted by Sekora (1973 in Veltre and Veltre 1983) the total harbour seal population in 1959 was 11,600 animals, whichis close to the 15,000 reported by Fiscus et al. (1981) in 1979. However, for the early 1960s, Abegglen (1977) suggestedthat the 1965 estimate of harbour seals from Cold Bay (~163E) to Attu was 4,868 animals and the 1962 census was morethan 6,000 animals. I used an estimate of 5,623 harbour seals for 1963, which was the difference between the 6,000animals in 1962 and the 4,868 in 1965 given by Abegglen (1977), or a biomass of 0.011 tkm-2.The biomass of Dall?s porpoise was estimated byBuckland et al. (1993) from the North Pacific MarineMammal sighting database for 1990, and using theabundances for the area between 45N:170E and55N:170W which gave an abundance of 227,098porpoises in that area. Using an average adult weight of61 kg (Trites and Pauly 1998) a biomass of 0.01 tkm-2was estimated. This is substantially lower than the 0.07tkm-2 estimated by NMFS, but I used the 0.01 tkm-2estimate to be conservative. No estimates were availablefor Dall?s porpoise in 1979 or 1963, thus I used the1991 biomass for those two models. The best estimatesof total biomass for small mammals in the three timeperiods were thus 0.017 tkm-2 in 1991, 0.018 tkm-2 in1979, and 0.022 tkm-2 in 1963. The biomass time seriesavailable for small mammals and sea otters are given inFigure 8.The annual P/B (0.24) and Q/B (39.0) estimates of northern fur seals were obtained from Aydin et al. (2003) for thewestern sub-Arctic region. For harbour seal, the annual P/B (0.08) and Q/B (17.4) ratios given in the NMFS model (AI)were used at first, but had to be modified to fit the 1963 model. The annual P/B (0.1) and Q/B (27.5) ratios of Dall?sporpoise were obtained from Aydin et al. (2003), while Gu?nette (this volume) had an estimate of 0.22 year-1 for the P/Bof Dall?s porpoise, which I used and prorated by biomass to get a P/B for small mammals of 0.166 year-1 for 1991, 0.164year-1 for 1979 and 0.150 year-1 for 1963. The annual Q/B estimates were also prorated by biomass for a Q/B of 23.9 year-1 in 1991, 23.7 year-1 in 1979 and 22.7 year-1 in 1963.Harbour seals feed in nearby coastal locations on cephalopods, pollock, sculpins, Atka mackerel, Pacific cod, greenling,capelin, herring, eulachon, sand lance, rockfish, shrimp, crabs, other invertebrates, salmon, Arctic cod and eelpouts (Perez1990) (Table 9). Northern fur seals mainly feed on the shelf-break and offshore waters on pollock, cephalopods, capelin,herring, deep-sea smelts, salmon and Atka mackerel (Perez 1990). The diet of Dall?s porpoise was reported to consistof 50% cephalopods and 50% fish (salmon, capelin, Arctic cod, atka mackerel, sand lance, herring, pollock, rockfish,sablefish, flatfish, eelpouts, deep-sea smelts and lanternfish (Perez 1990). The 50% fish was divided into 25% for speciesfor which the biomass estimates were available (Atka mackerel, pollock, rockfish, sablefish, flatfish), in the ratio of theirbiomass estimates and 25% species for which no biomass was available in equal proportions (salmon, capelin, Arctic cod,sand lance, herring, eelpouts, deep-sea smelts and lanternfish). The diet of Dall?s porpoise was prorated by the biomassof the fish they consumed for the 1979 and 1991 models, and the 1979 ratio of fish was used in the 1963 model. Thesediets were then combined and prorated by the biomass estimates of the mammals and fish (for Dall?s porpoise) for eachmodel, to calculate the overall diet (Table 9). Of all the small mammals, the only species that are taken regularly by First Nations for subsistence are harbour seals,and Wolfe et al. (2002) estimated that 29 harbour seals were taken in 1992, while 10 were struck and lost. The averagesubsistence take given by Wolfe et al. (2002) for Atka Island (the only community that they studied in the WesternAleutian Islands) from 1992-2002 by sex and for adults, juveniles and pups, was used in our model (Table 10). Theaverage body weight for adult male P. v. stejnegeri (128.5 kg), female (101 kg) and pups (19 kg) were obtained fromRidgeway and Harrison (1981) while the juvenile males and females were assumed to be about 50% of adult weights(similar to Steller sea lion females as the males get larget). The average weights used for unknown adults (115 kg),juvenile males (69 kg), juvenile females (54 kg), unknown juveniles (61 kg) and unknown age and sex (65 kg) were usedto calculate the average adult, juvenile and pup catches by year. The number of animals discarded (struck and lost) wereassumed to be in the same proportion as the adults and juveniles (assuming that no pups were lost), and using the averageweight for adults and juveniles (Table 10).UBC Fisheries Centre Research Reports, Vol 13, No. 1 25Table 9. Diet composition of small marine mammals in the Aleutian Islands. Note that only Dall?s porpoise diet isdifferent between 1991 and 1979.Prey Harbourseal (%)Northern furseal (%)Dall's porpoise (%) Total small mammals1991 1979 1991 1979 1963Salmon 1 2 3.1 3.1 0.022 0.022 0.019Capelin, sand lance, smelts   3.1 3.1Capelin 5 16  Arctic cod 1  3.1 3.1Eulachon 4   Pelagic invertebrate feeders    0.081 0.082 0.085Atka mackerel 9 2 18.2 13.5 0.140 0.112 0.101Sand lance 4  3.1 3.1 0.033 0.034 0.035Herring 5 6 3.1 3.1 0.040 0.040 0.042Pollock 12 34 4.6 6.4 0.085 0.098 0.113Rockfish 2  0.6 0.8 0.011 0.013 0.014Sablefish   1.2 1.5 0.007 0.008 0.006Pacific cod 8   0.032 0.033 0.041Flatfishes 3  0.4 2.8 0.014 0.028 0.026Saffron cod 3   Sculpin 9   Eelpouts 1  3.1 3.1Greenling 8   S & M demersals    0.101 0.104 0.122Deep-sea smelt/lanternfish  4 3.1 3.1Lanternfish/myctophids   3.1 3.1Myctophids    0.037 0.036 0.030Cephalopods 19 33 50 50 0.372 0.365 0.335Shrimps 2   0.008 0.008 0.010Euphausiids   0.3 0.3 0.002 0.002 0.001Other invertebrates 2   0.008 0.008 0.010Epibenthic carnivores 2 0.008 0.008 0.010Table 10. Subsistence catch of small mammals (mainly harbour seals) by Western Aleutian Island communities.Group 1992 1993 1994 1995 1996 1997 1998 2000 2001 2002Adult male 4.4 1.1 8.7 27.5 3.2 .0 1.5 1.5 13.3 4.5Adult female 4.4 4.4 20.6 2.5 2.2 .0 3.1 3.1 0 1.5Unknown adults 0 0 8.7 0 1.1 .0 7.6 7.6 0 1.5Juvenile male 6.6 8.7 7.6 2.5 6.5 4.6 0 0 25.0 10.5Juvenile females 8.8 2.2 5.4 0 2.2 4.6 0 0 5.0 12.0Unknown juv 0 0 16.3 0 3.2 .0 0 0 0 1.5Male pups 2.2 0 2.2 0 0 .0 0 0 0 0Female pups 2.2 5.5 1.1 0 0 3.1 0 0 0 0Unknown pups 0 0 3.3 0 0 .0 0 0 0 0Unknown 0 10.9 0 10.0 4.3 13.8 4.6 4.6 20.0 4.5Total Harvest 29 32.7 73.7 42.5 22.7 26.0 16.8 16.8 63.3 36.0Struck & Lost 9.9 0 8.7 0 2.2 1.5 0 0 6.7 0Note: No data available for 1999The bycatch of small mammals in 1992 consisted of 1 northern fur seal and 1 Dall?s porpoise caught by trawlers, 1harbour seal caught by the pot fishery and 3 unidentified pinnipeds caught by the longline fishery (Perez and Loughlin1991). In 1990, 28 Dahl?s porpoises were caught by the Aleutian Islands-Alaska Peninsula salmon driftnet fishery andassuming that 20% of that catch was made in the Aleutian Islands, gave a bycatch of 0.000006 tkm-2year-1 for the salmondriftnet fishery (Angliss and Lodge 2002). The Japanese high seas squid driftnet fishery (20?N-46?N and 170?E-145?W)caught 2,405 fur seals or 1.1% of the population in 1990 (Baba et al. 1993), which gave a bycatch from the squid driftnetfishery of 0.00001 tkm-2year-1. Thus, I assumed that for the 1991 model the total bycatch of small mammals was0.000016 tkm-2year-1 by the driftnet fishery, 0.000002 tkm-2year-1 by the domestic trawl fishery, 0.000004 tkm-2year-1by the longline fishery and 0.000001 tkm-2year-1 by the pot fishery.26 Aleutian Islands models; Heymans 0.0000.0050.0100.0150.0201963 1968 1973 1978 1983 1988 1993 1998Fishing mortality    .0510152025Catch (tonnes)   .Small mammal F Otter F Small mammal F calculatedOtter F calculated Small mammal catch Otter catchFigure 9. Fishing mortality (F, year-1) and catch (tyear-1) of small mammals and sea ottersin the Aleutian Islands.The bycatch of small mammals by the domestic trawl, longline and pot fisheries for the 1979 model was obtained fromPerez (1991), who indicated that no small mammals were caught in that year. However, for the salmon drift net fishery,I assumed that their catch of Dahl?s porpoise would be in the same ratio as for the 1990 catch, giving a catch of 0.002tkm-2year-1 made by the salmon driftnet fishery. For the subsistence catch in the 1970s, Veltre and Veltre (1983)estimated that approximately 30 harbour seals were caught annually and that only 10% was lost, giving a total catch of33 animals. By using the average weight from Ridgeway and Harrison (1981), a total catch of 0.000067 t km2 year-1 wasestimated and a discard estimate of 0.000006 tkm-2year-1. For 1963, I assumed that the fleets fishing in 1963 would have had similar catch rates of small mammals as they did in1991, and using the total catch made by each fleet, I calculated the catch of small mammals to be approximately 39kilograms by the pot fishery, 1 kilogram by the trawl fishery, 283 kilograms by the longlines and 18.3 tonnes by thesalmon driftnet fishery. This large catch by the driftnet fishery is due to the fact that salmon was one of the highestcatches in that year, and it estimated a large fishing mortality for small mammals in 1963 (Figure 9). No estimates ofsubsistence catch was available for the 1963 model, but I assumed that it was similar to the average catch from 1992-1999and 1979 for a catch of 0.000052 tkm2year-1 and discards of 0.000006 tkm-2year-1. These estimates of catch and biomasswere used to calculate the fishing mortality for each of the three models and the estimated fishing mortality from 1963-1990 (Figure 9).9. Sea ottersThe range of the northern sea otter, Enhydra lutris, included the entire southern Alaskan coast and the Aleutian chainbefore they were hunted (Murie 1959). The sea otter population was heavily exploited from 1741 to 1911, but after nearextinction it recovered and recolonised unexploited habitat in its native range in the Aleutians Islands (Palmisano 1975).The Rat Islands were the first to be recolonised in the 1950s, while the Near Island population was only re-establishedin the mid-1960s but the population grew rapidly through the 1970s and 1980s (Konar 1998). The 1911 population estimate was as low as 1,000 - 2,000 animals, but by 1965 the population included approximately25,000 animals (Palmisano 1975). Murie (1959) suggested that the total Aleutian Island population was estimatedconservatively at 2,000 animals and Doroff et al. (2003) indicated that the population increased from 1911 to the 1980s.Doroff et al. (2003) gave estimates of the populations in the Near, Rat, Delarof, Andreanof and Four Mountain Islandgroups for 1959, 1965, 1992 and 2000 (Table 11). For the Islands of the Four Mountains, only the populations onAmukta, Yunaska, Herbert and Carlisle Islands were included in my estimate (Table 11), with the populations onChuginadak, Kagamil and Uliaga Islands being in the Eastern Aleutian area (<170W). These estimates gave a biomassin 1992 of approximately 0.0025 tkm-2 and for 1965 the estimate was 0.0036 tkm-2. Assuming a linear increase over thetime period 1959 to 1965, the 1963 population was approximately 9,620 otters or 0.00363 tkm-2. For the 1979 model,I assumed that the biomass was similar to that in 1965 (0.0036 tkm-2). The decline of the sea otter population in theAleutians started around 1988 at Adak Island, 1991 for Amchitka Island and 1986 for Kagalska in the Andreanof Group(Doroff et al. 2003) but the population was still increasing in Attu by 1986 (Estes 1990). The population declined byapproximately 17.5% per year inthe 1990s (Doroff et al. 2003) andthe cause of the decline in thecentral Aleutian Islands wasfound to be elevated adultmortality (Estes et al. 1998).The annual sea otter reproductionrate was about 16%, with theannual rate of population changebetween 4-5% in densepopulations and 10-12% inunexploited habitat (Palmisano1975), which confirmed that theP/B of approximately 0.12 year-1given by NOAA was realistic fora declining population. Perez andMcAllister (1993) suggested thatthe daily energy requirements forUBC Fisheries Centre Research Reports, Vol 13, No. 1 27Table 11. Estimates of sea otter populations in the Aleutian islands obtainedfrom Doroff et al. (2003).Island group 1959 1960 1965 1992 2000Near Islands 0 27 995 368Rat Islands 3480 3147 1461 192Delarof Islands 4178 2798 995 343Andreanof Islands 1889 3685 3107 8474 Mountains 0 31 30Total 9547 9657 6589 1780Density (tkm-2) 0.0036 0.0036 0.0025 0.0007Table 12. Catch of sea otters byFirst Nations in the Aleutians.Year Number kgkm-2year-11989 50 0.0031990 50 0.0031991 25 0.0021992 50 0.0031993 180 0.0111994 52 0.0031995 50 0.0031996 150 0.0091997 150 0.0091998 50 0.0031999 52 0.0032000 50 0.003otters were 4.9 103 kcal, and that the energy value of their food was 0.9 kcalg-1, which gave a daily food consumptionof 5.4 kg and an annual Q/B of 86.4.Sea otters are fed on by transient orcas in the central Aleutian Islands, with the decline in otters from the mid-1980s tothe mid-1990s possibly being caused by only a few transients (Ford and Ellis 1999). Murie (1959) reported that, in the1930s, killer whales were seen cruising the outer edge of a kelp bed close to sea otter population, but he could not verifythat they were catching otters. They were also preyed on by sharks and according to Estes (1980) by Steller sea lions,but I have not added them to the diet of Stellers. The diet (by volume) of sea otters was estimated at 6% greenlings, 86% sea urchins and 8% other invertebrates (Wilke1957), while Estes et al. (1981) gave the frequency of occurrence in the diet as: 60% sea urchins, 29% other invertebrates,2.5% epibenthic carnivores (crabs), 1% macrophytes, 0.1% octopi, and 6% greenlings, and Yang (1999) suggested thatAtka mackerel has been found in the stomach of sea otters. Watt et al. (2000) calculated the diet of sea otters duringwinter and summer at Amchitka Island in both % frequency and % volume and we used the average % volume for ourdiet in 1991. Thus the diet of sea otters was assumed to be approximately 13.2% greenling/lumpsucker etc. (smalldemersals), 78.9% invertebrates (mostly urchins), 3.9% epibenthic carnivores, 0.3% cephalopods, 1.2% Atka mackereland 1.2% sand lance.Estes (1990) suggested that sea otters start offby eating large sea urchins, eventuallyreducing the number and size of sea urchinsto such an extent that the habitat goes fromurchin barren to kelp bed. In response tothese changes (lack of large urchins and theincrease in kelp associated species) the dietwould change from one dominated byinvertebrates to one dominated by fish (Estes1990). Monson et al. (2000) also indicatedthat by 1993 the body condition of the remaining sea otters at Amchitka Island had improved, because the decline inotters increased the sea urchin population, and they were also feeding on smooth lumpsuckers (small demersal speciesfrom the epipelagic zone) which increased their body condition. Thus, the above mentioned diet would probably be goodfor the 1991 model, but the 1979 model would include more fish (greenlings, lumpsuckers and Atka mackerel). Watt etal. (2000) suggested that by the early 1970s kelp-forest fishes were the single most important prey of otters at Amchitkaand most fish eaten by otters were inshore, kelp-associated species such as greenlings, rockfishes, gunnels, pricklebacksand sculpins. Thus, for the 1979 model I included more fish (30% small demersals, 4% Atka mackerel and 4% sandlance). I also used this diet for the 1963 model.Sea otters were caught as bycatch by the Aleutian Islands black cod pot fishery in 1992 (Angliss and Lodge 2002), andBerger et al. (1986) estimated that 18 sea otters were taken that year, giving a bycatch of 0.000007 tkm-2year-1 for thatfishery, which I used in the 1991 model. No sea otters were caught by the pot fishery in 1979, and as I have no estimateof catch for the 1963 fishery, I assumed that the bycatch by the pot fishery was proportional to the catch, giving a catchof 0.000009 tkm-2year-1.The estimated catch by First Nations from 1989-2000 were obtained fromAngliss and Lodge (2002). The average catch by First Nations for the westernstock (from the Western Gulf of Alaska to the Aleutian Islands and including thePribilof Islands) for 1996-2000 was 97 animals (Angliss and Lodge 2002). Thetotal number of otters in the western stock consisted of 33,203 otters, of which7,309 animals were in the Aleutian Islands (Angliss and Lodge 2002). I used thisratio (33,203:7,309) to calculate the First Nations catch in the Aleutian Islands.Thus, in 1991 approximately 25 otters (or 0.002 kgkm-2year-1) were caught inthe western stock (Table 12). There were no estimates of subsistence catches for1979 or 1963, but the population was higher, therefore I used the average catchfrom 1989 to 2000 in the total Western Gulf of 76 animals or 0.005 kgkm-2year-1 for both time periods (Angliss and Lodge 2002). Thus, the total catch of seaotters amounted to 0.000008 in 1991, 0.000005 in 1979 and 0.000014 tkm-28 Aleutian Islands models; Heymans 2year-1 in 1963. The estimates of catch and biomass were used to calculate a fishing mortality rate for each of the threemodels and the extrapolated fishing mortality from 1963-1990 (Figure 9).10. BirdsThe various bird species of of the Aleutian Islands are given in Table 13 and consist of invertebrate feeders andpiscivorous birds. In addition to the species for which biomass estimates were available (Table 13), invertebrate feedingbirds also include ancient murrelet, Synthliboramphus antiquus, short-tailed albatross, D. albatrus, Cassin's auklet,Ptychoramphus aleuticus, whiskered auklet, A. pygmaea and parakeet auklet, Cyclorrhynchus psittacula (Anonymous2001). Similarly, additional piscivorous bird species include the Aleutian tern, Sterna aleutica, Arctic tern, S. paradisaea,black guillemot, Cepphus grille, red-legged kittiwake, R. brevirostris, Bonaparte's gull, Larus Philadelphia, glacous gull,L. hyperboreus, glaucous-winged gull, L. glaucescens, herring gull, L. argentatus, Mew gull, L. canus, ivory gull,Pagophila eburnean, common murre, Uria aalge, Kittlitz's murrelet, Brachyearamphus brevirostris, marbled murrelet,B. marmoratus, pigeon guillemot, Cepphus columba, rhinoceros auklet, Cerorhinca monocerata, Sabine's gull, Xemasabini (Anonymous 2001). The species for which biomass estimates were available in the Western Sub-Arctic (USA), their residency (92 days), bodymass and occupancy (Table 13) were obtained from Hunt et al. (2000). The total area of the WSA (2,168,000 km?) wasused to calculate the total biomass per unit area (0.09 tkm-2) for the 1991 model. I assumed that the Aleutian Island birdbiomass was similar to that of the western sub-arctic region and this estimate was then a lower limit to the biomass asnot all the species were represented. For the 1979 and 1963 models no biomass was available and they were estimatedby Ecopath. The annual P/B estimate (0.113) was obtained from the NMFS model, while the annual Q/B ratio (65.4) wasestimated by using the daily ration (R), the average weight (W) for each species, and the empirical equation:log R = -0.293 + 0.85   log W(g) obtained from Nilsson and Nilsson (1976) in Wada (1996). Table 13. Estimates of invertebrate feeding and piscivorous bird numbers, mean weight and biomassin the Western Sub-Arctic region; all birds have a residency time of 92 days (Hunt et al. 2000). Common name Species Abundance Body mass(kg)Weight(t)Invertebrate feedersBlack-footed albatross Diomedea nigripes 5,000 3.148 1,448Crested auklet Aethia cristatella 380,000 0.264 9,229Fork-tailed storm petrel Oceanogroma furcata 3,600,000 0.055 18,315Leach's storm petrel Oceanogroma  leucorrhoa 3,500,000 0.040 12,816Least auklet Aethia  pusilla 47,000 0.084 363Northern fulmar Fulmarus glacialis 600,000 0.544 30,029Red phalarope Phalaropus fulicaria 87,000 0.056 446Short-tailed shearwater Puffinus tenuirostris 430,000 0.543 21,481Thick-billed murre Uria lomvia 47,000 0.964 4,168Piscivorous birdsBlack-legged kittiwake Rissa tridactyla 610,000 0.407 22,841Buller's shearwater Puffin bulleri 5,000 0.380 175Cormorants Phalacrocorax spp. 1,000 2.822 260Horned puffin Fratercula corniculata 85,000 0.619 4,841Laysan albatross Diomedea immutabilis 1,100,000 3.042 307,850Long-tailed jaeger Sterocorarius longicaudus 38,000 0.297 1,037Parasitic jaeger Sterocorarius  parasiticus 76,000 0.465 3,248Pomarine jaeger Sterocorarius  pomarinus 190,000 0.694 12,131Sooty shearwater Puffinus griseus 3,100,000 0.787 224,452South polar skua Stercorarius maccormicki 150,000 1.156 15,953Tufted puffin Fratercula cirrhata 892,000 0.779 63928The diet of birds were obtained from NMFS data for shearwaters, murre, kittiwakes, auklets, puffins, fulmars, stormpetrels, cormorants and albatrosses. The diets of all other species were prorated by the biomass of each group. A diet forgulls was given but not used in my calculation, as I had no biomass to prorate their diet. For many species a preferencediet was given, and was either prorated by the biomass of their prey (if those were available) or by taking a straightpercentage of the preference given, so for instance, for albatrosses 50% of the diet consisted of salmon, small pelagics,sand lance, herring, myctophids, juvenile pollock and Pacific cod, and that 50% was divided equally (7.1% each) betweenUBC Fisheries Centre Research Reports, Vol 13, No. 1 29these species. For cormorants 45% of their diet was divided between myctophids and small pelagics (22.5%). Thecontribution of shrimp, benthic invertebrates and epibenthic carnivores were 2.5% combined or 0.8% each. The 3%preference for Atka mackerel, juvenile pollock, Pacific Ocean perch, rockfish and Pacific cod was divided based on thebiomass of these species, and the 2.5% small and large zooplankton was prorated on their biomass. Finally the remaining27% of the diet was divided equally between small pelagics and myctophids.For storm petrels, the 32% allocated to zooplankton was prorated by their biomass, the 1.7% allocated to Atka mackerel,juvenile pollock, Pacific Ocean perch, rockfish and Pacific cod was prorated by their biomass and the 4.2% allocated tosmall pelagics and sand lance was divided equally between them. For fulmars the 3.4% allocated to zooplankton and the30.6% allocated to Atka mackerel, juvenile pollock, Pacific Ocean perch, rockfish and Pacific cod was prorated by theirbiomasses, and the 7.15% allocated to small pelagics and sand lance was divided equally between them. The NMFS dataalso had fulmars eating 0.15% transient killer whales, toothed whales, juvenile and adult Steller sea lions, smallmammals, sea otters and birds. I redirected this portion of the diet to Steller sea lion pups, although it could also beredirected to detritus, as this could be dead mammals. For puffins, the 7% allocated to zooplankton and the 11.6% allocated to Atka mackerel, juvenile pollock, Pacific Oceanperch, rockfish and Pacific cod was prorated by their biomass, and the 71% allocated to small pelagics and sand lancewas divided equally between them. For auklets, the 93% allocated to zooplankton and the 0.3% allocated to Atkamackerel, juvenile pollock, Pacific Ocean perch and Pacific cod was prorated by their biomasses, and the 5.4% allocatedto small pelagics and sand lance was divided equally between them. For kittiwakes, the 8.2% allocated to zooplanktonand the 23.3% allocated to Atka mackerel, juvenile pollock, Pacific Ocean perch, rockfish and Pacific cod was proratedby their biomasses. The 50.9% allocated to small pelagics and sand lance, and the 12.2% allocated to herring andmyctophids were divided equally between them. For murres, the 11.3% allocated to zooplankton and the 20% allocatedto Atka mackerel, juvenile pollock, Pacific Ocean perch, rockfish and Pacific cod was prorated by their biomass. The 37.4% allocated to small pelagics and sand lance and the 0.2% allocated to herring and myctophids were divided equally. For shearwaters, the 1.7% allocated to zooplankton was prorated by their biomass and the 29% allocated to herring andmyctophids were divided equally. The diet of birds in 1991 and 1979 were prorated for the biomass of their prey wherepossible and are given in Table 14 and 15 respectively. For the 1963 model, I used the same diet as for the 1979 modelas no biomass estimates were available for fish during that time. Table 14. Diet composition of birds (% weight) in the Aleutian Islands in 1991.Group Shearwater MurreKitti-wake Auklet Puffin FulmarStormPetrelCormo-rantAlbatrossJaeger TotalBiomass 0.114 0 0.011 0.004 0.032 0.014 0.01 0 0.15SSL Pups 0 0 0 0 0 0.2 0 0 0 0Salmon 0 0 0 0 0 0 0 0 7.1 3.1Small Pelagics 41.6 18.7 25.4 2.7 35.7 3.6 2.1 13.5 7.1 21.5Atka mackerel 0 13.4 15.6 0.2 7.8 20.4 1.1 2 0 2.2Sand lance 14.5 18.7 25.4 2.7 35.7 3.6 2.1 50 7.1 12.5Herring 0 0.1 6.1 0 0 0 0 15 7.1 3.3Juv. pollock 0 27.1 1.5 0 0.8 2 0.1 0.2 7.1 3.5A. pollock 0 0 0 0 0 0 0 0 0 0POP 0 0.6 0.7 0 0.4 0.9 0.1 0.1 0 0.1Rockfish 0 0.4 0.5 0 0.2 0.6 0 0.1 0 0.1Pcod 0 4.3 5 0.1 2.5 6.6 0.4 0.6 7.1 3.8Myctophids 14.5 0.1 6.1 0 0 0 0 13.5 7.1 8.2Shrimp 0 0 0 0 0 0 0 0.8 0 0Benthic inverts 1 0.6 4.7 0.9 6.3 0.1 1.4 0.8 0 1.1Epi.carnivores 0 0 0 0 0 0 0 0.8 0 0Cephalopods 26.6 3.5 0.7 0.1 3.8 58.6 60.7 0 50 36.3L zooplankton 1.2 8.1 5.9 66.4 5 2.4 22.8 1.8 0 3S Zooplankton 0.5 3.2 2.3 26.6 2 1 9.2 0.7 0 1.2Both piscivorous and invertebrate feeding birds were taken as bycatch in the longline and trawler fisheries(Anonymous2001). Bycatch by the BSAI longline fleet and the BSAI and GOA trawl fleets were given for 1993-1999 by Anonymous(2001). The area of the Bering Sea and Aleutian Islands is approximately 552,000 km?, while the BSAI and Gulf ofAlaska combined is approximately 844,000 km? giving a bycatch in 1993 of 0.00013 tkm-2 year-1 by the longline fleetand 0.000002 tkm-2 year-1 by the trawlers in 1993, which was used for bycatch in the 1991 model. Estimates of bycatch30 Aleutian Islands models; Heymans for the 1979 and 1963 models were not available, and I used the ratio of the total catch by trawlers and longlines in 1979and 1963 compared to 1993 to get the proportion of bycatch in 1979 and 1963. The total catch made by trawlers for 1979was 83,000 tonnes, or 63% of the 1993 catch (~ 132,000 tonnes), resulting in a bycatch of 0.000001 tkm-2year-1.Similarly, the total catch made by trawlers and longlines in 1963 were 12,325 tonnes and 664 tonnes respectively,indicating catches of 0.0000005 and 0.0000004 tkm-2year-1 made by the longliners and trawlers respectively.Table 15. Diet composition of birds (% weight) in the Aleutian Islands in 1979.Group Shear-water MurreKitti-wake Auklet Puffin FulmarStormpetrelCormo-rantAlbatrossJaeger TotalBiomass* 0.114 0 0.011 0.004 0.032 0.014 0.01 0 0.15SSL Pups 0 0 0 0 0 0.2 0 0 0 0Salmon 0 0 0 0 0 0 0 0 7.1 3.1Small Pelagics 41.6 18.7 25.4 2.7 35.7 3.6 2.1 13.5 7.1 21.5Atka mackerel 0 13.5 15.7 0.2 7.8 20.6 1.1 2 0 2.2Sand lance 14.5 18.7 25.4 2.7 35.7 3.6 2.1 50 7.1 12.5Herring 0 0.1 6.1 0 0 0 0 15 7.1 3.3Juv. pollock 0 27.1 3.7 0 1.8 4.8 0.3 0.5 7.1 3.8Ad.pollock 0 0 0 0 0 0 0 0 0 0POP 0 0.4 0.4 0 0.2 0.6 0 0.1 0 0.1Rockfish 0 0.8 1 0 0.5 1.2 0.1 0.1 0 0.1Pcod 0 2.2 2.6 0 1.3 3.4 0.2 0.3 7.1 3.5Myctophids 14.5 0.1 6.1 0 0 0 0 13.5 7.1 8.2Shrimp 0 0 0 0 0 0 0 0.8 0 0Benthic inverts 1 0.6 4.7 0.9 6.3 0.1 1.4 0.8 0 1.1Epi.carnivores 0 0 0 0 0 0 0 0.8 0 0Cephalopods 26.6 3.5 0.7 0.1 3.8 58.6 60.7 0 50 36.3L zooplankton 1 6.9 5 56.9 4.3 2.1 19.6 1.5 0 2.6S Zooplankton 0.7 4.4 3.2 36.1 2.7 1.3 12.4 1 0 1.6* Assuming that biomass has the same ratio as in 1991.11. Mammal eating sharksMammal eating sharks include the Pacific sleeper shark, Somniosus pacificus, the great white shark, Carcharodoncarcharias, and the bluenose sixgill shark, Hexanchus griseus. There are some indications that salmon sharks, Lamnaditropis, also feed on sea lions (Loughlin and York 2000), but this species was not added to the mammal eating sharkgroup. Little biological information was available for Pacific sleeper sharks, although they were considered common inboreal and temperate regions of shelf and slope waters of the North Pacific (Hare et al. 2003). Sleeper sharks are foundin relatively shallow waters at higher latitudes, and in deeper habitats in temperate waters and large concentrations ofsleeper sharks were found during the 2000 pilot Bering Sea slope survey, but hardly any were found in the Eastern BeringSea shelf survey (Hare et al. 2003). Orlov and Moiseev (1998) found that they occurred in depths from 85 m to 717 m(average about 450 m) in the Western Bering Sea and the Northwestern Pacific (close to the Kuril Islands andKamchatka). Great whites have been reported off the Aleutians Islands (http://www.sharkresearch committee.com/dist.htm) but very little other information is available.No data was available on the biomass of mammal eating sharks in the Aleutians for any of the models (1963, 1979 or1991). The annual P/B (0.1) and Q/B (3.0) ratios of Pacific sleeper sharks were obtained from the NMFS model and usedfor the 1979 and 1991 models (and the Q/B in 1963). The annual P/B ratio in 1963 of 0.13 was obtained from the naturalmortality given by Gu?nette (this volume). The diet of Pacific sleeper sharks included harbour seals and cetaceans(Hulbert et al. 2002), although the diet estimates obtained from NMFS for sleeper sharks did not include any mammals.I therefore used the initial diet used by Gu?nette (this volume) for mammal eating sharks (including white and sixgillsharks), which included 3.5% Steller sea lion juveniles, and added the percentage of slope and shelf rockfish into myrockfish group, as well as adding the small demersals to myctophids (see Table A1). This diet was used for all threemodels.According to the Predator Conservation Network website, sleeper sharks were caught year-round on commercial sablefishlong-line gear in Alaska, with tagged sharks were usually recaptured near where they were originally caught. Fishermenreported few catches of sleeper sharks in the late 1980's but catches have increased since the early 1990's(http://www.conservationinstitute.org/pcnpacificsleepershark.htm). According to Orlov and Moiseev (1998), PacificUBC Fisheries Centre Research Reports, Vol 13, No. 1 31sleeper sharks were more abundant in the western Bering Sea than in the Kuril Islands, with bottom trawl catches in theBering Sea usually being represented by 1-10 (maximum 25) specimens and caught at a frequency of 42.7%. Off thenorthern Kuril Islands and southestern Kamchatka they were caught mostly as single specimens with frequency ofoccurrence in bottom trawl catches of about 3.5% (Orlov and Moiseev 1998). However, there was no clear estimate ofthe catch of sharks and therefore I estimated the catch from the ?other groundfish? group given by Gaichas (2003). Thisestimate for other groundfish was divided using the proportion of sharks, skates, sculpins and octopuses in the 1999 catchobtained from Anonymous (2001). The proportion of the catch allotted to sharks was then divided equally betweenmammal eating and other sharks, and the other shark catch was added to the skate and shark group (Table 16). Thedomestic catches for sharks and skates were divided into trawl, pot and longline gear (Anonymous 2001) and I assumedthat all of these catches were discarded. For 1963-1976, no estimates of catches or discards were available, and I assumedthat the bycatch of sharks (both mammal eaters and sharks and skates) were in the same ratio of the bycatch to catch ratiofor the 1979 model.Table 16. Catch (t) of mammal eating sharks and sharks and skates in the Aleutians.Mammal eating sharks Sharks and skates TotalYear Pot Trawl Longline Total Pot Trawl Longline1963 0 56 1 57 0 4262 82 4,3451964 0 223 2 225 0 17061 151 17,2111965 0 270 1 271 0 20631 97 20,7281966 0 220 8 228 0 16823 633 174561967 0 154 14 168 0 11800 1,070 12,8701968 0 125 9 134 0 9553 682 10,2361969 0 109 14 123 0 8327 1,066 9,3931970 0 198 14 212 0 15167 1,061 16,2281971 0 72 7 79 0 5489 544 6,0331972 0 115 17 132 0 8829 1,303 10,1321973 0 75 16 91 0 5727 1,216 6,9431974 0 129 16 145 0 9863 1,196 11,0601975 0 115 7 122 0 8797 567 9,3641976 0 88 8 96 0 6726 615 7,3411977 0 134 0 134 0 10,234 38 4,3191978 0 101 2 103 0 7,719 181 7,9001979 0 104 4 107 0 7,931 285 8,2161980 0 108 0 108 0 8,243 32 8,2761981 0 60 0 60 0 4,587 34 4,6211982 0 42 1 43 0 3,230 53 3,2821983 0.02 30 1 31 0 2,265 70 2,3341984 0.00 14 0 14 0 1,037 24 1,0611985 0.01 17 0 17 0 1,269 33 1,3021986 0.03 12 0 13 0 941 18 9591987 0.00 10 0 10 0 731 3 7341988 0.06 3 0 4 0 236 42 2781989 0.04 1 0 1 0 40 29 691990 1.92 34 3 39 0.13 1,720 1,261 2,9811991 0.38 7 1 8 0.03 344 252 5961992 1.26 22 2 26 0.08 1,129 828 1,9571993 1.34 23 2 27 0.09 1,201 881 2,0821994 0.45 8 1 9 0.03 403 295 6981995 0.53 9 1 11 0.03 473 347 8191996 0.70 12 1 14 0.05 625 458 1,0841997 0.62 11 1 13 0.04 557 408 9661998 1.01 18 2 20 0.07 900 660 1,5591999 0.69 12 1 14 0.05 615 451 1,0662000 1.23 22 2 25 0.08 1,103 809 1,9122001 1.65 29 3 33 0.11 1,477 1,083 2,5592002 0.81 14 1 16 0.05 726 532 1,25812. Sharks and skatesAll sharks and skates, excluding the three mammal eating shark species above are given in this group. The sharks includesalmon sharks, Lamna ditropis, and spiny dogfish, Squalus acanthias, while skates include the white skate, Bathyearaja32 Aleutian Islands models; Heymans 0102030401963 1968 1973 1978 1983 1988 1993 1998 2003Catch and Biomass (thousand tonnes)     .00.20.40.60.8Biomass Catch F calculated FAnnual fishing mortality (F)Catch and Biomass (thousand tonnes)     .Annual fishing mortality (F)Figure 10. Catch (tonnes), biomass (tonnes) and fishing mortality (year-1 observed andinterpolated) for sharks and skates in the Aleutian Islands.spinosissima, deepsea skate, B. abyssicola, big skate, Raja binoculata, Bering skate, B. interrupta, longnose skate, R.rhina, starry skate, R. stellulata, mud skate, B. taranetzi / Rhinoraja longii, black skate, B. trachura, Alaska skate, B.parmifera, Aleutian skate, B. aleutica, commander skate, B. lindbergi, whiteblotched skate, B. maculate, whitebrow skate,B. minispinosa, golden skate, B. smirnovi and Okhotsk skate, B. violacea (Hare et al. 2003). The skate community in theAleutian Islands appeared to be different from that of the Eastern Bering Sea (Hare et al. 2003). The most abundantspecies in the 1997 survey of the Aleutian Islands was the whiteblotched skate, while Alaska and Aleutian skates werealso common (Hare et al. 2003). The mud skate was relatively common but represented a lower proportion of totalbiomass because it is a smaller skate and all seven other skate species identified in the 1997 survey made up about 7%of aggregate skate complex biomass (Hare et al. 2003). The biomass for sharks and skates (Figure 10) for 1991 (0.31 tkm-2) was obtained from the survey (Gaichas 2002), butno estimates were available for either the 1979 or 1963 model so they were estimated by Ecopath. The annual P/B (0.18)and Q/B (2.5) ratios for this group were obtained by using the average of the ratios given in the NMFS model for salmonsharks, dogfish, Alaska skate, Bering skate, Aleutian skate, whiteblotched skate, mud skate, longnosed skate, big skateand black skate, and were used for all three models. For 1963, an average annual P/B of 0.16 was calculated based onthe natural mortality of sharks and skates given by Gaichas (2003).The diet for sharks and skates was obtained from NMFS for salmon sharks, dogfish, Alaska skate, Aleutian skate, Beringskate, whiteblotched skate, mud skate, big skate, longnose skate and black skate. In the NMFS database, when noinformation was available about the proportion in the diet of the different prey, it was assumed that prey were consumedin proportion to their abundance. As with birds, only preference diet was given for some prey species, and these prey wereeither prorated by their biomass (if those were available, and I used the different biomass for 1979 and 1991 to getdifferent diets for those time periods) or by taking a straight percentage of the preference given. The diet of salmon sharks (Table 17) included a preference of 1% for all rockfish, and 4% for greenlings, sculpins, etc.which were all grouped into the small demersal group. Dogfish diet had a preference of 9% for small zooplankton andsome large zooplankton species (viz. mysids, chaetognaths, pelagic amphipods), and I prorated this preference betweenthe large and small zooplankton based on the biomass small and large zooplankton (adding the value to the 16.3% alreadygiven for euphausiids and jellies). The invertebrate portion of the diet included non-pandalid shrimps (shrimps),anemones, hydroids, clams, polychaetes (benthic invertebrates), snails and sea stars (carnivorous epibenthos), and as nobiomass estimates existed for shrimp and carnivorous epibenthos, I divided the 2.9% equally into these three groups,adding it to the 7.6% shrimps, 2.2% benthic invertebrates and 6.7% epibenthic carnivores already consumed. Apreference of 15.4% was divided between Pacific cod, juvenile and adult pollock in the ratio of their biomass estimatesand 11.1% was divided between halibut, arrowtooth and flatfish in the ratio of their biomass, while the 2.8% preferenceUBC Fisheries Centre Research Reports, Vol 13, No. 1 33for all rockfish was allocated to the one rockfish group, and the 1.9% preference for dogfish and skates was allocated tothe shark and skate group.Table 17. Diet composition for sharks and skates (in % weight) in the Aleutian Islands for 1991. Data from NMFS except forAlaska skate, which was adapted from Gu?nette (this volume). Diets for longnose, big and black skates were prorated to getproportions.Group SalmonsharksDog-fishAlaskaskateBeringskateAleutianskateWhite-blotch Mud skateLongnose Big skateBlackskate TotalSharks/skates 1 2 0 0 0 0 0 0 0 0 0.3Salmon 39.6 0 0 0 0 0 0 4.8 1.7 0 4.4Small Pelagics 0 3.7 1 0 0 0 0 4.8 1.7 1 0.9Atka mackerel 0 0 5 0 0 27.8 0 4.8 1.7 34.9 4.1Sand lance 0 1.1 1 0 0 0 0 4.8 12.1 1 0.9Herring 0.4 14.4 1 0 0 0 0 4.8 1.7 1 2Juv. pollock 0 6.3 1 0 25.6 0 0 4.8 1.7 1 3.7A. pollock 0 4 0 0 0 10.2 0 4.8 1.7 94.8 3POP 0 0 0 0 0 0 0 0 0 1 0.01Rockfish 1 2.8 0 0 0 0 0 0 0 22.7 0.7Sablefish 36 0 5 0 0 0 0 0 0 1 4.1Pcod 0 5.1 5 0 0 0 0 4.8 1.7 1 1.5Halibut 11 1.7 1.7 0 0 0 0 8.3 5.8 1 2.3Arrowtooth 0 4 1.7 0 0 0 0 8.3 5.8 1 1.4L demersals 0 0 0 0 0 0 0 0 0 1 0.01Flatfish 0 5.5 1.7 0 0 0 0 33.7 5.8 1 3.7S demersals 4 0 10 0 0 8.8 0 4.8 103.3 59.1 5.8Large deep 0 0 0 0 0 0 0 0 0 15.5 0.2Myctophids 0 0 10 0 0 0.1 1.2 4.8 0 101.4 2.8Shrimp 0 8.5 4 8.7 74.4 32.2 0 21 44.1 171.2 17.6Benthic inverts 0 3.1 12 90.3 0 0.6 1.9 0 76.7 106.8 13.8Epi.carnivores 0 7.6 8 0 0 12.4 0 0 10 90.3 4.1Cephalopods 7 4.7 24 0 0 2.4 83.4 0 51.7 55.3 14Lzooplankton 0 22.8 8 1 0 0.1 1.7 0 95 43.6 6.1SZooplankton 0 2.6 0 0 0 0 0 0 0 0 0.3Detritus 0 0 0 0 0 5.4 11.8 0 13.2 26.6 2.3Import 0 0 0 0 0 0 0 0 0 0 0Total 100 100 100 100 100 100 100 119 435 833 100The diet of Alaska skates included 79% small pelagics, sand lance, herring, sablefish, Pacific cod, halibut, arrowtoothand flatfish, and the total diet summed to >1, so I adapted the diet used by Gu?nette (this volume) for Alaska skates, toinclude 5% Atka mackerel instead of rockfish, 5% sablefish and Pacific cod, and the 5% allocated to flatfish I dividedbetween halibut, arrowtooth and flatfish. The diet also consisted of 1% each for sand lance, herring, small pelagics andjuvenile pollock, 10% each for small and medium demersals and myctophids, 4% shrimps, 12% benthic invertebrates,8% epibenthic carnivores, 24% cephalopods and 8% large zooplankton.For big skates, the diet preference for small pelagics, salmon, Atka mackerel, herring, juvenile pollock, adult pollock andPacific cod was divided equally between these species, while the preference for halibut, arrowtooth flounder and flatfishwas also divided equally. Similarly, for longnose skates the preference for halibut, arrowtooth and flatfish was dividedbetween these three groups, and the value added to the consumption of rex sole by longnose skates and the preferencefor salmon, small pelagics, Atka mackerel, sand lance, herring, pollock adult and juveniles, Pacific cod, myctophids andsmall and medium demersals was divided equally between these groups. Likewise, the preference by black skates forsmall pelagics, sand lance, herring, adult and juvenile pollock, Pacific Ocean perch, rockfish, sablefish, Pacific cod,halibut, arrowtooth, demersal large predators, flatfish, demersal small/medium predators and large deep water fish weredivided equally between the groups. This proportion was added to the known percentage for shortraker rockfish in therockfish group, to the known percentages for Irish lord and sculpins in the demersal small/medium group and to the othermacrourids in the large deep group. The total 1991 and 1979 diet breakdowns for sharks and skates are given in Tables17 and 18 respectively. For 1963 very few estimates of fish biomass were available, thus I used the 1979 diet for thismodel. Spiny dogfish were commonly taken by the pelagic pollock trawl fishery and in the longline fisheries for sablefish,halibut, Greenland turbot, and Pacific cod, and their catch rates have increased five-fold in Prince William Sound and34 Aleutian Islands models; Heymans Table 19. Diet composition (in proportion) for salmon inthe Aleutian Islands obtained from the NMFS model.Groups ReturningsalmonOutgoingsalmonAverageCephalopods 0.2 0 0.1Large zooplankton 0.4 0.25 0.325Small zooplankton 0.2 0.25 0.225Algae 0.2 0.5 0.35Import 0 0 0Total 1 1 1Table 18. Diet composition for sharks and skates (% weight) in the Aleutian Islands for 1979. Data from NMFS except for Alaskaskate, which was adapted from Gu?nette (this volume). Diets for longnose, big and black skates were prorated to get proportions.Group SalmonsharksDog-fishAlaskaskateBeringskateAleutianskateWhite-blotchMud skateLongnoseBig skateBlackskate TotalSharks and skates 1 2 0 0 0 0 0 0 0 0 0.3Salmon 39.6 0 0 0 0 0 0 4.8 1.7 0 4.4Small Pelagics 0 3.7 1 0 0 0 0 4.8 1.7 1 0.9Atka mackerel 0 0 5 0 0 27.8 0 4.8 1.7 34.9 4.1Sand lance 0 1.1 1 0 0 0 0 4.8 12.1 1 0.9Herring 0.4 14.4 1 0 0 0 0 4.8 1.7 1 2Juv. pollock 0 8.3 1 0 25.6 0 0 4.8 1.7 1 3.9A. pollock 0 5.3 0 0 0 10.2 0 4.8 1.7 94.8 3.1POP 0 0 0 0 0 0 0 0 0 1 0.01Rockfish 1 2.8 0 0 0 0 0 0 0 22.7 0.7Sablefish 36 0 5 0 0 0 0 0 0 1 4.1Pcod 0 1.8 5 0 0 0 0 4.8 1.7 1 1.1Halibut 11 1.6 1.7 0 0 0 0 8.3 5.8 1 2.3Arrowtooth 0 2.8 1.7 0 0 0 0 8.3 5.8 1 1.3L demersals 0 0 0 0 0 0 0 0 0 1 0.01Flatfish 0 6.8 1.7 0 0 0 0 33.7 5.8 1 3.8S demersals 4 0 10 0 0 8.8 0 4.8 103.3 59.1 5.8Large deep 0 0 0 0 0 0 0 0 0 15.5 0.2Myctophids 0 0 10 0 0 0.1 1.2 4.8 0 101.4 2.8Shrimp 0 8.5 4 8.7 74.4 32.2 0 21 44.1 171.2 17.6Benthic inverts 0 3.1 12 90.3 0 0.6 1.9 0 76.7 106.8 13.8Epiben.carnivores 0 7.6 8 0 0 12.4 0 0 10 90.3 4.1Cephalopods 7 4.7 24 0 0 2.4 83.4 0 51.7 55.3 14Lzooplankton 0 21.8 8 1 0 0.1 1.7 0 95 43.6 6SZooplankton 0 3.5 0 0 0 0 0 0 0 0 0.4Detritus 0 0 0 0 0 5.4 11.8 0 13.2 26.6 2.3Import 0 0 0 0 0 0 0 0 0 0 0Total 100 100 100 100 100 100 100 119 435 833 100twenty-fold in the central Gulf of Alaska since 1994 (http://www.conservationinstitute.org/spinydogfish.htm). Thebycatch of sharks and skates were estimated from the stock assessment of ?other groundfish? (Gaichas 2003) using thebreakdown of other groundfish in the 1999 estimates from Anonymous (2001). From 1963-1976 no estimates of catchesor discards were available, and I assumed that the bycatch of sharks and skates were in the same ratio of the bycatch tothe catch