Mortensen et al : Growth of Oncorhynchus gorbuscha in relation to marine water temperature, secondary production, and survival 333 



35 



25. 



15 



emigrants. Despite spending more 

 time in low-gi'owth conditions, 

 they will have the advantage of 

 being larger at a given date than 

 subsequent, faster-growing emi- 

 gi-ants (Fig. 10). Thus, there is 

 an adaptive advantage in less- 

 ening size-selective predation by 

 being on the "leading edge" of the 

 synchronous emigration to marine 

 waters. However, this advantage 

 is constrained by poorer growth 

 early in the spring and poorer sur- 

 vival to adulthood. 



The abundance of potential pred- 

 ators (coho salmon, Dolly Var- 

 den, and sculpins) near shore in- 

 creases rapidly in May, empha- 

 sizing the importance of larger 

 size and larger numbers for pink 

 salmon to avoid predation. Late 

 emigrants, although encountering 

 what appear to be good conditions 

 for growth, may not be able to ef- 

 fectively outgrow increasing pre- 

 dation pressure or may not be abun- 

 dant enough to saturate the preda- 

 tor population. Other factors may 

 combine to affect the gi'owth and 

 ultimately the survival of the late 

 emigrants. The abundance of com- 

 petitors such as young-of-the-year 

 and juvenile heiring and capelin 

 may affect the availability of prey 

 at critical times and could explain 

 the lower growth rates observed for 

 lat«-emerging pink salmon. 



Growth rate of pink salmon juve- 

 niles in Auke Bay was consistently related to sur- 

 vival within a given year. This finding is consistent 

 with the concept that high growth rates during early 

 marine residency give a sui'vival advantage by min- 

 imizing the intensity of predation (Parker, 1971; 

 Heard, 1991). Following the Exxon Valdez oil spill in 

 Prince William Sound in 1989, WiUette ( 1996) found 

 a direct relationship between growth and survival of 

 pink salmon juveniles. However, he did not observe 

 this association for juveniles captured in 1990 or 1991 

 and speculated that changes in the feeding habits 

 of predators due to the absence of alternate prey 

 obscured the relationship between growth and sur- 

 vival. Correlations of parameters for scale growth 

 rates with run size of pink salmon in northern south- 

 eastern Alaska indicated that nearshore gi'owth con- 

 ditions are a significant contributor to the interannual 



05J. 



Weight by June 1 



42 



 40 



Days to reach 1 g 



T" 



14 



April 



21 



— r- 



28 



38 



 36 



34 



 32 



30 



12 



May 



Figure 10 



The predicted size of pink salmon fry by 1 June and the number of days to reach 

 1 g for emigration dates between 1 April and 12 May. 



variation in marine sui'vival. Early scale growth ( Jae- 

 nicke et al., 1994) was positively correlated with 

 survival, indicating that high early marine growth 

 lessens mortality due to predation. In our study, for 

 three of the four years, growth rate was an excellent 

 predictor of survival interannually, explaining 85*^ 

 of the observed variability. The anomalous year was 

 1988 ( 1987 brood), which was also the year of highest 

 overall growth rates. Fish from all release groups of 

 the 1987 brood returned to the weir at consistently 

 lower rates than fish from the other brood years. Fish- 

 ery exploitation on returning adults may be a factor. 

 We corrected the rate of survival each year for those 

 fish taken in the commercial fishery with the correc- 

 tion factor used by the Alaska Department of Fish 

 and Game; however, the correction factor was rela- 

 tively crude with no measure of variance. It is conceiv- 



