BEACHAM and STARR: POPULATION BIOLOGY OF CHUM SALMON 



1971), which could possibly bias adult age com- 

 position, sex ratios, and fecundity estimates. 

 However, we believe that any bias present was 

 not of sufficient magnitude to mask trends in 

 abundance of individual brood years. Daily test 

 fishing 5 d/wk was usually conducted from late 

 September until mid-December, although sam- 

 pling within this period was not conducted when 

 the commercial fishery was operating in the 

 river. Ages of chum salmon were determined 

 from scales. Bilton and Ricker (1965) and 

 LaLanne and Safsten (1969) have outlined the 

 methodology of using scales for aging chum salm- 

 on. Lengths were recorded as either fork length 

 or postorbital-hypural length to the nearest mil- 

 limeter. Fecundity was determined by direct 

 counts of the number of eggs in both ovaries. Age 

 composition in the escapement was assumed to 

 be the same as that in the test fishery. Escape- 

 ment was estimated during the 1960s by visual 

 counts on spawning beds, by test fishing, and by 

 a tagging program (Palmer 1972), whereas in 

 the 1970s, escapement was estimated from visual 

 counts and test fishing only. Total returns for 

 a brood year included the catch plus escape- 

 ment. 



Fraser River chum salmon are taken by the 

 fishery which operates near the river mouth, and 

 by fisheries in Johnstone Strait, the Strait of 

 Georgia, and at Point Roberts in the United 

 States. The proportion of Fraser River chum 

 salmon in the Johnstone Strait fishery was esti- 

 mated seasonally based on the tagging studies 

 reported by Palmer (1972). Since the fishery in 

 the Strait of Georgia exploits mixed stocks of 

 chum salmon, it was not possible to estimate the 

 contribution of Fraser River chum salmon in this 

 fishery, although it is small compared with the 

 catch in Johnstone Strait. Catches in the Fraser 

 River, as well as 100% of the catches off Point 

 Roberts, and the Fraser River contribution to 

 Johnstone Strait were summed to estimate the 

 total annual catch of Fraser River chum salmon. 

 Although not all chum salmon caught off Point 

 Roberts, 20 km south of the Fraser River mouth, 

 may be bound for the Fraser River, any overesti- 

 mation of the Fraser River contribution to the 

 Point Roberts catch is compensated for by under- 

 estimation of the Fraser River contribution to 

 the Strait of Georgia catch. Escapements of pink 

 salmon used in the present study were those 

 listed in the annual report of the International 

 Pacific Salmon Fisheries Commission for 1979 

 (Anonymous 1980). 



RESULTS 



Marine Growth 



High-seas scale and tagging studies have indi- 

 cated that chum salmon from British Columbia 

 and Alaska range from lat. 45°N to 60°N and 

 from long. 130°W to 180° in the North Pacific 

 Ocean (Shepard et al. 1968). Specific distribu- 

 tions of Fraser River chum salmon were not 

 available. However, the effect of variability in 

 ocean water temperatures on growth rate was in- 

 vestigated by comparing the mean monthly tem- 

 perature from March through August (major 

 part of growing season) at Station P (lat. 50°N, 

 long. 145°W) and mean length-at-age of return- 

 ing adults. Fork lengths of returning adults dur- 

 ing 1960-69 were taken from Palmer (1972) and 

 were converted to postorbital-hypural lengths by 

 the regressions: 



Males: H = 1.08 FL - 219 (N = 100, r = 0.72) 

 Females: H = 1 .00 FL - 130 ( N = 100, r = 0.89) 



where FL = fork length in millimeters and H — 

 postorbital-hypural length in millimeters. These 

 regressions were derived from chum salmon 

 taken by the Fraser River commercial fishery in 

 1962 and 1963. Postorbital-hypural lengths of re- 

 turning adults during 1970-78 were derived 

 from the test fishery at Cottonwood Drift. For 

 the period of 1960-78, mean lengths-at-age of 4- 

 yr-old chum salmon were correlated with water 

 temperature at Station P during the penultimate 

 growing season for males (r = 0.49, n — 19, 

 P<0.05) and females (r = 0.44, n = 19, P<0.06). 

 However, there was no correlation for mean 

 length-at-age and water temperature at Station 

 P during the penultimate growing season for 

 age-3 fish (males: r = 0.17, n = 19, P>0.10; fe- 

 males: r = —0.08, n = 19,P>0.10)orintheyearof 

 return and mean length-at-age for age-3 fish 

 (males: r = 0.17, P>0.10; females: r = 0.10, 

 F>0.10). The relatively stable mean length-at- 

 age for returning age-3 chum salmon is undoubt- 

 edly due to selectivity of the sampling gear, with 

 possibly only the larger age-3 fish susceptible to 

 capture in a 16.9 cm mesh gill net. 



Age Composition and Sex Ratios of 

 Returning Adults 



The mean monthly age composition of chum 

 salmon migrating upstream during the years 



815 



