Cl'NDKKSON POPULATION BIOLOO I >!■ SEBASTES M UTUS 



I O-i 



£ 08- 



• 



04 



1 



<C 0.2- 



• OCS 



A WVI 



10 II 12 13 



Age (years) 



— i — 

 15 



— i 

 16 



FIGURE 15. — Estimated proportion of each age group recruited 

 to the fishing grounds, for the QCS and WVI stocks of Pacific 

 ocean perch. 



MORTALITY 



All estimates of Pacific ocean perch mortality 

 rates depend on a knowledge of the age structure 

 of the population. Virtually all fish caught die 

 from the effects of decompression, so that no suc- 

 cessful tagging studies have ever been carried out. 



In this section, data on number caught per hour 

 by age-group were used to estimate the survival of 

 14- to 18-yr-old Pacific ocean perch in year n to 

 ages 15-19 in year n + 1, or to ages 16-20 in year 

 n + 2. These age-groups were chosen so as to strike 

 a balance between problems with incomplete 

 recruitment on the one hand and age determina- 

 tion problems on the other. Previous analysis has 

 suggested that recruitment is not complete until 

 age 16, about the same time that age determina- 

 tion becomes difficult (Table 7) and the ages 

 of some individuals are presumably underesti- 

 mated. No age-groups older than 20 should be 

 included in survival estimates, and restricting 

 mortality estimates to fully recruited age-groups 

 (16-19 yr olds) could result in underestimation of 

 survival rates. Inclusion of the incompletely re- 



cruited 14 and 15 yr olds offset this to some degree 

 and had the additional benefit of basing the sur- 

 vival estimates on five age-groups rather than 

 three. 



All survival estimates were expressed on an an- 

 nual basis (S ), and then used to estimate Z. On the 

 assumption that M is density independent and 

 thatF is a linear function of total hours trawled, 

 the model Z — qf + M, where q = proportion of 

 population caught by trawling 1 h and f = mean 

 annual number of hours trawled, was employed. 

 Linear regression ofZ on /yields estimates of q 

 and M where the model is appropriate. 



Total international fishing effort (f) was esti- 

 mated by dividing the total international catch in 

 a given year by the corresponding CPUE for the 

 Washington trawl fleet (after Gulland 1969). The 

 value of/" was obtained by averaging f over the 

 years that each estimate of Z pertained to. 



Queen Charlotte Sound 



Calculation of total international fishing effort 

 is outlined in Table 8. The 1967-72 Soviet catch 

 data for the INPFC Charlotte area was taken from 

 unpublished analyses by T. A. Dark and N. B. 

 Parks. These data were derived from analysis of 

 fleet location and catch by quarter and give the 

 most detailed breakdown of the Soviet catch that 

 is currently available. Soviet catch estimates for 

 1965, 1966, 1972, and 1973, as well as all Japanese 

 and North American data for the years 1963-73, 

 were derived from Westrheim et al. (1972) and a 

 recent update of that report. 



Estimates of Z are plotted against mean inter- 

 national fishing effort (Table 8) in Figure 16 and 

 the results indicate that the information collected 

 so far can provide only tentative estimates of M. 

 Pacific ocean perch vary widely in their availabil- 

 ity to on-bottom trawls and the CPUE indices used 

 in mortality estimation are consequently suscep- 



TABLE 7. — Deviations of Canada's final otolith readings from those of United States, by age-group, for a collection of 



Pacific ocean perch from Queen Charlotte Sound, June 1972. ' 



'S. J. Westrheim and W. R. Harlmg. 1973. Report on the 1972 comparison of Pacific ocean perch otolith and scale interpretations. 

 Unpubl. manuscr., 24 p. 



383 



