150 



Fishery Bulletin 101(1) 



Table 1 



Number of Steller sea lion pups counted on rookeries in Alaska in 1998 (Sease and Loughlin, 1999) and minimum total breeding 

 season population size estimates (including pups), assuming pups represent 20.5% of the population (Winship et al., 2002). Areas 

 are defined in Figure 1. 



Area 



Number of rookeries 



Southeast Alaska 



Gulf of Alaska 



Eastern Aleutian Islands 



Central Aleutian Islands (subarea 1) 



Central Aleutian Islands (subarea 2) 



Central Aleutian Islands (subarea 3) 



Western Aleutian Islands 



All 



3 

 9 

 6 



8 

 8 

 3 

 3 



40 



have declined dramatically (Loughlin et al., 1992; Trites 

 and Larkin, 1996). Thus, it is unlikely that population 

 structure was the same in the 1990s as it was in the 1970s 

 and that population structure was the same in all regions 

 of Alaska. Unfortunately, there are very few data avail- 

 able with which to determine the relationship between 

 the structure and the rate of change in size of a population 

 of Steller sea lions. To account for this uncertainty Win- 

 ship et al. (2002) used sampling distributions for survival, 

 maturity, and reproductive rates that approximated the 

 uncertainty in population structure (ranges of sampling 

 distributions were about 10-20'7f ). 



The population size during the breeding season in each 

 region of Alaska was estimated by using pup count data 

 from the U.S. National Marine Fisheries Service and Alas- 

 ka Department of Fish and Game surveys done in June 

 and July 1998 (Table 1: Sease and Loughlin, 1999). We 

 assumed that the actual number of pups born could have 

 been as much as 20'7f greater than the number counted 

 because of pups that were hidden during the surveys, pup 

 mortality before the survey dates, and births after the 

 survey dates (Trites and Larkin, 1996). The number of 

 pups in each region was therefore assumed to range from 

 the values in Table 1 to 1.2 x these values (uniform sam- 

 pling distributions). Total population size was estimated 

 by dividing the number of pups by the proportion of the 

 total population size that they represented as described by 

 Winship etal. (2002). 



Diet parameters 



Prey species were grouped into seven prey categories as 

 defined by Merrick et al. ( 1997): 1 ) cepbalopods: squid and 

 octopus; 2) flatfish: Pleuronectidae; 3) forage fish: Pacific 

 herring iClupea pallasi). Pacific sandlance (Ammodytes 

 hcxaptcrus ), eulachon ( Tlialeiclithys pacificuN ), and capelin 

 iMallotus villosus); 4) gadids: walleye pollock iTheragra 

 chalcogramma). Pacific cod iGadus macroccphalus), and 

 other Gadidae, 5) hexagrammids: Atka mackerel (Pleuro- 

 grammiis monoptcrygius) and other Hexagrammidae; 6) 

 salmon: Pacific salmon iOiuorhynchus spp.); and 7) other: 



rockfish (Sebastes spp.), sculpins (Cottidae), pricklebacks 

 (Stichaeidae), skates (Raja spp.), lamprey (Lampetra 

 spp.), sharks, and other demersal fish. 



The diet composition of Steller sea lions in southeast 

 Alaska was estimated from data reported by Trites and 

 Calkins'* for scat collected in the 1990s on rookeries during 

 the summer breeding season and on nonbreeding haul- 

 outs (in inside waters) during the rest of the year (Table 2). 

 Split-sample frequencies of occurrence (Olesiuk et al., 

 1990) of prey categories were used as the median percent 

 contributions of each prey category to diet biomass. Four 

 seasonal diet compositions were used: a "winter" diet 

 commencing on 1 December, a "spring" diet commencing 

 on 1 March, a "summer" diet commencing on 1 June, and 

 an "autumn" diet commencing on 1 September In order 

 to make the modeled transitions between seasonal diets 

 more gradual, the season dates were sampled from uni- 

 form distributions with upper and lower limits equal to ±1 

 week. It was assumed that all ages and both sexes had the 

 same diet composition. 



Diet compositions for Steller sea lions in all other re- 

 gions of Alaska (Gulf of Alaska-western Aleutian Islands) 

 were estimated from data reported by Merrick et al. ( 1997) 

 for scats collected mainly on breeding rookeries during the 

 summers of the early 1990s (Table 2). As with southeast 

 Alaska, split-sample frequencies of occurrence were used 

 as the median percent contributions of each prey category 

 to diet biomass. We assumed the same diet composition for 

 all ages and both sexes year-round in these regions. The 

 only area not covered by Trites and Calkins^ and Merrick 

 et al. (1997) was the eastern Gulf of Alaska. We assumed 

 that the diet composition of Steller sea lions in the eastern 

 Gulf of Alaska was the same as the diet of Steller .sea lions 

 in the Gulf of Alaska region from Merrick et al. ( 1997). 



We randomly sampled the diet from triangular distribu- 

 tions to incorporate uncertainty in the diet composition 



■• Trites. A. W., and D. G. Calkin.s. 2002. Unpubl data. De- 

 partment of Zoology and Marine Mammal Research LInit. Fish- 

 eries Center, Dniv. British Columbia, Room 18. Hut 15-3. 6248 

 Biological Sciences Road. Vancouver, B.C., Canada, V6T 1Z4. 



