Pitcher et al,: Abundance and distribution of Eumetopias /ubatus 



111 



(Wade, 1998; Harkonen et al.. 2002). This annual rate 

 of increase indicates that either some factor or factors 

 are still limiting the growth rate of this population or 

 that the growth potential of this otariid is less than the 

 theoretical maximum, which was derived from phocid 

 population growth rates. We have observed Steller sea 

 lions that have been shot or entangled in marine debris, 

 and this undocumented mortality could be preventing 

 the population from increasing at a higher rate. In addi- 

 tion, the Steller sea lion tends to have a longer period of 

 maternal investment and a lower reproductive rate than 

 most phocids (Pitcher et al., 1998), both of which may 

 limit the growth potential of populations. 



Although the three geographic regions supporting the 

 largest rookeries all increased at about the same rate, 

 individual rookeries often exhibited different population 

 growth rates or temporal changes in growth rates. At the 

 northern end of the range, Forrester Island accounted for 

 essentially all of the population growth until the 1970s; 

 however the observed rate of change has slowed since 

 the 1980s. At the same time, some of the rookeries to the 

 south of Forrester Island in British Columbia and to the 

 north of it in central-northern southeastern Alaska have 

 exhibited higher-than-average growth rates since the 

 1980s. The mechanism causing these geographic patterns 

 is unknown, but could involve 1) dispersal of breeding 

 animals between rookeries, 2) differences in local condi- 

 tions that affect reproduction and survival, or 3) a shift 

 in distribution of prey resources. Some dispersal of breed- 

 ing females from their natal rookeries has been shown 

 to occur. Six of 31 females that were marked as pups on 

 the Forrester Island rookery were subsequently observed 

 to have given birth on other rookeries (Raum-Suryan et 

 al., 2002). The authors of that study concluded that the 

 Steller sea lion generally conformed to the metapopula- 

 tion concept as depicted by Hanski and Simberloff (1997), 

 in that local breeding populations (rookeries) and move- 

 ments among these local populations have the potential 

 of affecting local dynamics. 



For our assessment of long-term historic population 

 trends, we relied mainly on counts of non-pups (or oc- 

 casionally pups and nonpups combined) on rookeries, as 

 few reliable pup counts were available prior to the 1970s. 

 The 2002 population-wide survey (Fig. 2) and the last 30 

 years of counts in British Columbia indicated there is a 

 relationship between the numbers of nonpups and pups 

 on rookeries. However, departures from this relation- 

 ship can occur, especially where existing rookeries are 

 being abandoned or new rookeries are being formed. For 

 example, the Farallon Islands, which no longer meet our 

 definition of a rookery, now serves largely as a haulout 

 site (Le Boeuf et al., 1991). The historical rookery on the 

 Sea Otter Group in British Columbia, the only rookery 

 known to have been extirpated by control efforts, is also 

 still used during the breeding season as a haulout by 

 nonbreeding animals. Conversely, in southeastern Alas- 

 ka, the new rookeries were established at sites previously 

 used as major haulouts by nonbreeding animals. The 

 lack of accurate pup counts may, thus, have influenced 

 our historical interpretation of historical data and our 



depiction of the exact breeding range, but there is a gen- 

 eral consensus that the breeding range has shifted. Pup 

 production in southern California has disappeared and 

 in central California has dropped to less than one-fifth 

 of what it was in the 1920s. Few, if any, pups were born 

 in southeastern Alaska in the early 1900s, whereas this 

 area now accounts for nearly half of total pup production 

 in the eastern North Pacific population. 



Control programs and harvesting clearly depleted the 

 eastern Steller sea lion population and may have con- 

 tributed to its redistribution, but the kills cannot fully 

 explain the shift in the distribution. For example, while 

 control efforts were underway in British Columbia dur- 

 ing the 1950s and 1960s, animals may have taken ref- 

 uge just north of the British Columbia-Alaska border 

 at Forrester Island, or animals breeding on Forrester 

 Island may have benefited from reduced competition as 

 a result of the reductions on British Columbia rookeries. 

 However, the northward expansion of the breeding range 

 in southeastern Alaska continued through the 1980s 

 and 1990s, even though killing of sea lions in British 

 Columbia ceased in the 1960s. At the southern end of 

 their range, sea lions were apparently very abundant in 

 California before the 1860s, but were depleted during the 

 1870s because of intense hunts of sea lions for oil and 

 hides (Bonnot, 1929). The last organized kills were made 

 in 1909, although hunting, especially of bulls for trim- 

 mings (genitals, lips with whiskers, and gall bladders) 

 continued into the 1930s. Nevertheless, the population 

 declines in southern California began in the late 1930s, 

 and in central California began in the late 1960s and 

 early 1970s, well after major kills by humans had ended 

 (Hastings and Sydeman, 2002). 



The reason for the northward shift in the overall 

 breeding distribution is unknown, and different factors 

 may have been in play at the southern and northern 

 ends of the range. In the south, competition with in- 

 creasing populations of other pinnipeds may have been 

 a factor in range constriction (Stewart et al., 1993). In 

 particular, the number of California sea lions breeding 

 in California increased from at most a few thousand 

 in the 1920s (Bonnot, 1928) to about 240,000 in 2000 

 (Lowry and Maravilla-Chavez, 2005). It is likely that 

 California sea lions and Steller sea lions compete with 

 each other because 1) their ranges overlap, 2) they share 

 the same haulout sites, and 3) they probably consume 

 many of the same prey species. On San Miguel Island 

 and the Farallon Islands, where Steller sea lions used to 

 predominate (Bartholomew and Boolootian, 1960; Ripley 

 et al., 1962; Stewart et al., 1993), the decHnes in Steller 

 sea lions coincided with large increases in numbers of 

 California sea lions (Stewart et al., 1993; Hasting and 

 Sydeman, 2002). 



For unknown reasons, southeastern Alaska represents 

 the only area throughout the range of the eastern North 

 Pacific population where new Steller sea lion rooker- 

 ies have been established. Steller sea lion rookeries are 

 normally located on remote, offshore islands or reefs and 

 require adequate areas above high water levels where 

 young pups can survive most weather conditions. There 



