NOTE Lowry et al Counts of Mirounga angustirostris from aerial photographs 



181 



photographs differed from that of the small-vessel 

 and ground survey by —36%, and the number of adult 

 females counted during peak breeding season differed 

 from the number of pups counted 18 days later by small- 

 vessel and ground survey by +15T7c (Table 5). 



Analysis of the trend in pup production over time 

 for San Miguel Island, based on counts obtained from 

 aerial photographic surveys or from ground surveys, 

 showed that 1) a log-linear relationship between 

 count and year (Fig. 2) fits counts from photographs 

 better (R 2 =0.979) than counts from the ground 

 (i? 2 =0.788 ); 2 ) the standard errors of the constant and 

 the slope from each regression indicated that the 

 counts from photographs were less variable; 3) each 

 had a significant slope and intercept (P<0.03); and 4) 

 there was no significant difference between the slopes 

 (P=0.66) or the elevations of the regressions (P=0.32). 



Discussion 



Large-format color transparency photographs were 

 useful for obtaining accurate and precise counts of 

 northern elephant seal pups and adults. The preci- 



Photographtc 

 count 



o Ground 

 count 



1967 1988 19 



Year 



Figure 2 



Comparison of northern elephant seal, Mirounga angus- 

 tirostris, birth estimates from 1985 to 1991 for San Miguel 

 Island, California, based on large-format aerial color pho- 

 tographs with estimates based on ground surveys iStewart, 

 1992). The solid line represents the regression line for pho- 

 tographic counts (ft-=0.979) and the dashed line represents 

 the regression for ground counts (fl 2 =0.788). 



