Olesiuk Prey consumption of Phoca vitulina 



493 



The Strait of Georgia also supports important com- 

 mercial and sport fisheries. With respect to salmon, 

 the most valuable fishery, the study area accounts for 

 approximately 28% by number and 31% by weight of 

 total escapement in British Columbia 5 . Ketchen et al. 

 (1983) estimated that the Strait of Georgia accounted 

 for 27% of the total non-salmonid commercial harvest 

 by weight, 34% by landed value, and 26% by whole- 

 sale value. The Strait of Georgia also accounts for about 

 96% of all recreational angler-days expended in the 

 tidal waters of British Columbia (Sinclair 6 ), which rep- 

 resents about 600,000 boat-trips annually (Shardlow 

 and Collicutt, 1989). Harbor seal-fishery conflicts in 

 the Strait of Georgia would therefore be expected to be 

 more intense than in other regions of the province. 



Two distinct types of habitat were recognized within 

 the study area: estuaries and non-estuaries. Estuaries 

 were defined as the large, shallow, soft-bottomed areas 

 that occurred at the mouths of some rivers. Twelve 

 estuaries (Fig. 1) within the study area were inhab- 

 ited by appreciable numbers (>10) of seals. In estuar- 

 ies, seals typically hauled out on logbooms or on tidal 

 sandbars. Seals were widely distributed outside of es- 

 tuaries and utilized 285 different haulout sites, most 

 of which consisted of tidal boulders, reefs, islets, and 

 ledges at the base of bluffs (Olesiuk et al., 1990a). 



The model 



Input parameters for the model consisted of informa- 

 tion on harbor seal abundance, distribution, the sex- 

 and age-structure of the population, diet composition, 

 and daily energy and food requirements. The first four 

 parameters were derived from local studies (Bigg, 1969; 

 Olesiuk et al., 1990a; Olesiuk et al., 1990b; Olesiuk 4 ), 

 whereas the last parameter was deduced from the 

 metabolic and feeding rates reported for phocids in the 

 literature. The primary output from the model con- 

 sisted of estimates of the annual consumption of each 

 prey species. 



The annual consumption of a particular prey was 

 estimated as the product of the abundance of seals, 

 feeding rates, and the proportion of that prey in their 

 overall diet. Daily food requirements were estimated 

 separately for each sex-and age-class in the popula- 

 tion, and mean per capita requirements calculated from 

 energetic life tables by weighting the individual esti- 

 mates according to the sex-and age-structure of the 

 population. Feeding rates were assumed to be the same 

 within and outside of estuaries and constant with sea- 



s Estimated from data provided by G. Hudson, Pacific Biological Sta- 

 tion, Nanaimo, B.C., pers. commun. 1989. 



"Sinclair, W. F. 1972. The British Columbia sport fisherman. Can. 

 Dep. Environ. Fish. Serv., Pac. Reg., 69 p. 



son. However, because the abundance and diet compo- 

 sition of seals differed between estuaries and non- 

 estuaries and varied seasonally in both habitat types 

 (Olesiuk et al, 1990b), the abundance and diet data 

 were stratified by habitat type and integrated over 

 time. The biomass of the Mh prey species consumed 

 annually, B k , was thus estimated as 



B t = FR 



(365 



P Fkl -dt + FR- 



365 



N,„ 

 i 



P nt ,-dt 



[1] 



where FR is the mean per capita daily food require- 

 ment, N El and N ot the number of seals within and out- 

 side of estuaries, respectively, on the tth date ( ^Janu- 

 ary 1st; 365=December 31st), and P Ekl and P m the 

 proportion of the diet within and outside estuaries com- 

 prised of the &th prey species on the tth date. The 

 integrals were solved by calculating finite approxima- 

 tions using daily \t increments. 



Abundance and distribution 



The abundance and distribution of harbor seals in the 

 study area were determined by aerial censuses. In 1988, 

 the entire study area was surveyed twice; once just 

 prior to the pupping season (31 May-16 June) and 

 again at the end of the pupping season (9-26 August). 

 A third estimate of the population size in 1988 was 

 obtained by projecting the population trends observed 

 during censuses conducted between 1973 and 1988. 



Aerial censuses were conducted under standardized 

 conditions under which maximum numbers of seals 

 were hauled out: 1) the lower of semi-diurnal low tides; 

 2) between 08:00-11:30 AM; 3) usually toward the end 

 of the pupping season which extends from early July 

 to mid August (Bigg, 1969); and 4) not during inclem- 

 ent weather such as rough seas or heavy precipitation. 

 Under these conditions, the variability of replicate cen- 

 suses indicated that counts represented, on average, 

 88.4% of the actual population (Olesiuk et al., 1990a). 



The confounding effects of seasonal fluctuations in 

 the size of the population due to recruitment during 

 the protracted pupping season were removed by ad- 

 justing all census counts, C„ to post-pupping levels by 

 applying a correction, P„ to account for pups born sub- 

 sequent to the date, t, of the census. Since births are 

 normally distributed over time with a mean Julian birth 

 date, u, of 208 (27 July) and a of 16.1 days (Bigg, 1969; 

 Olesiuk et al., 1990a), P, was calculated as 



f t- 



P, = 1+3 



(f-u) 2 

 2<j 2 

 2;r 05 -e dt • fj 



[21 



