334 



Fishery Bulletin 103(2) 



42 



41° - 



40 



39 c 



38 c 



37 c - 



36° 



35' 



34 c 



Point Conception 



J I I I I I I I I I I I I I L 



N. 



nA 



WLg(O) 



(1) 



127° 126° 125° 124° 123° 122° 121° 120° 119° 

 Longitude (°W) 



Figure 3 



A posteriori stratification of study area 

 into "offshore" stratum and into seven 

 zones (A through G) within the "inshore" 

 stratum for estimating abundance of Cali- 

 fornia sea lions (Zalophus californianus) 

 from strip-transect data and haulout 

 count data. 



(56,526 km 2 total surface area), using transect intersect 

 points as the dividing line (Fig. 3). Differences between 

 the definition of haulout sites for the surveys in this 

 study and during previous surveys in 1980-82 and 1995 

 (Bonnell et al. 1 , and Beeson and Hanan 8 ) made it neces- 

 sary to create additional zones within the inshore stra- 

 tum to allow comparisons of the three data sets. The 

 inshore stratum was thus divided into seven zones ("A" 

 through "G"), separated at the following latitudes: 1) 

 35°25'N; 2) 36°15'N; 3) 37°20'N; 4) 38°10'N; 5) 39°30'N; 

 and 6) 40°50'N (Fig. 3). The zones were separated where 

 gaps occurred in the distribution of haulout areas along 

 the coastline. Total area sizes for the seven zones were 

 the following: A: 7647 km 2 ; B: 7206 km 2 ; C: 8025 km 2 ; 

 D: 6153 km 2 ; E: 7790 km 2 , F: 6030 km 2 , and G: 7695 

 km 2 . At-sea abundance was obtained separately for 

 offshore and inshore strata, and for each zone within 

 the inshore stratum, by using a modified strip-transect 

 formula that included a correction, g(0), for diving ani- 

 mals that were not available to be seen: 



where N c = corrected total abundance (corrected for 



animals below the surface); 

 n = number of individuals sighted within the 



strip-transect; 

 A = total size of study area (in km 2 ); 

 W = the strip width (in km); 

 L = distance surveyed (in km) calculated as the 



sum of the great circle distances between 



position fixes', and 

 g(0) = probability that a sea lion will be visible 



at the surface within the strip viewed by 



the observer as the aircraft passes over the 



water. 



Coefficients of variation (CV) and lognormal 95% con- 

 fidence limits of these abundance estimates were cal- 

 culated by using standard formulae (Buckland et al., 

 1993). 



Probability of missing submerged sea lions 



We estimated the probability of seeing sea lions at the 

 surface, g(0), from dive data in Feldkamp et al. (1989) 

 derived from 14 foraging trips made by seven lactating 

 adult female California sea lions during late breeding- 

 season: 



g<0) = 



t+s + r 



t+s+r+d 



(2) 



where t = average time (hours) spent at the surface 



between dives within diving bouts by an adult 



female sea lion; 

 s = average time (h) spent swimming near the 



surface between diving bouts by an adult 



female sea lion; 

 r = average time (h) spent resting at the surface 



between diving bouts by an adult female sea 



lion; and 

 d= average time (h) spent diving during diving 



bouts by an adult female sea lion. 



From seven female sea lions, Feldkamp et al. (1989) 

 calculated averages of 12.0 hours (no SD given) spent at 

 the surface between dives within diving bouts (t), 21.9 

 hours (SD = 9.5 hours) spent swimming near the surface 

 between diving bouts (s), 1.6 hours (SD = 1.6) spent rest- 

 ing at the surface between diving bouts (r), and 17.3 

 hours (SD = 6.7) spent diving during diving bouts (d). We 

 calculated the CV forg(0) from the standard deviations 

 of diving data. In using these data we assumed that 

 between dives, sea lions swam near the surface and at 

 a depth where they would be seen by an observer in 

 the aircraft and that sea lions were not visible to an 

 observer in the aircraft during dives. Dive data were 

 not available for other age and sex classes; therefore, 



