Gudmundson et al Diet of Callorhinus ursinus 



453 



15 

 14 

 13. 



^^■ 

 11 • 



10 

 9 

 8 

 7 

 6 

 5 

 4 

 3 

 2 



A St. George Island 



 Scats (n=225 samples contlning 2856 beaks) 

 g Spews (n=5030 beaks in 1 1 1 samples) 



JUUUi 



14 

 (55) 



24 

 (94) 



34 

 (133) 



4 4 

 (173) 



54 

 (212) 



64 

 (251) 



B St. Paul Islantj 



04 



0,4 

 (15) 



ll 



Scats (n=71 9 beaks in 1 1 samples) 

 Spews (n=47 beaks in 5 samples) 



u 



(55) 



(94) 



34 

 (133) 



4 4 



(173) 



54 

 (212) 



64 

 (251) 



LRL (DML) 



Figure 4 



Percent minimum number of individuals (MNI%) of Gonatopsis borealis and Berry- 

 teuthis magister (Gb-Bm) lower rostral length (LRL) and dorsal mantle length (DMLl 

 in mm recovered from northern fur seal iCallorhinus ursinus) scats and spews con- 

 taining measurable Gb-Bm beaks collected from rookeries on (A) St. George Island 

 and (B) St. Paul Island, Pribilof Islands, Alaska, during the breeding season (late 

 July-September), 1990-2000. 



The dorsal mantle length size of Gb-Bm estimated from 

 lower beaks was significantly larger in spew samples 

 than scat samples (Fig. 4). Digestive biases relating 

 to cephalopod beak size have also been observed in 

 northern fur seal G.I. tracts (Yonezaki et al., 2003), as 

 well as in Australian fur seal (Arctocephalus pusillus 

 doriferus; Gales et al., 1993) and California sea lion 

 (Zalophus californianus; Lowry and Carretta, 1999) 

 spews and scats. Owing to limited taxonomic resolu- 

 tion of squid species in previous northern fur seal diet 

 studies, direct comparisons of Gb-Bm size were not pos- 

 sible. For instance, Sinclair et al. (1994) presented DML 

 size ranges of beaks recovered from stomachs, but, the 

 cephalopod groups Gb-Bm and Gm-Gm were combined 

 for analysis and were reported as gonatid squid. 



We observed a greater percentage of adult walleye 

 pollock in spew than in scat samples (Fig. 3). In ad- 

 dition, few scat and spewing samples were found to 

 contain otoliths from both juvenile and adult age catego- 

 ries, further demonstrating size-related digestive biases 



of prey remains between sample types. The differences 

 in walleye pollock age classes between scat and spew 

 samples seem to indicate that size estimations of pol- 

 lock consumed by northern fur seals have likely been 

 underestimated in previous studies using G.I. tracts 

 and scats (Sinclair et al., 1994, 1996). Northern fur 

 seal diet studies based on analysis of the entire G.I. 

 tract have shown that adult females consume primar- 

 ily juvenile walleye pollock (Sinclair et al., 1994. 1996). 

 However, biases associated with the retention of large 

 prey remains are inherent in studies based on G.I. tract 

 samples (Pierce and Boyle, 1991; Gales et al., 1993), and 

 size-selective relationships between prey otolith size and 

 sample type have been documented for northern fur 

 seals (Kiyota et al., 1999; Yonezaki et al., 2003). 



Data that accurately describe the ages of walleye 

 pollock consumed by northern fur seals are critical for 

 assessing potential competition with the commercial 

 trawl fishery, which focuses on adult pollock. Previous 

 northern fur seal diet analyses using scat and G.I tract 



