Abstract

We used information about hatchery rearing and release practices for 173 releases of age 0+ smolts between 1982 and 2012, as well as time series of early marine prey biomass and predator abundance/biomass, to investigate the biological basis of age-specific return variability of chinook salmon (Oncorhynchus tshawytscha) from the Robertson Creek Hatchery. We used survival rate as the response variable and considered the rate to be an apparent one because it is the product of the survival and maturation rates. Results of multiple regression analyses (adjusted R² ranging between 0.43 and 0.59) showed that Pacific mackerel (Scomber japonicus) and Steller sea lion (Eumetopias jubatus) abundances accounted consistently for all of the explained variation in age-specific survival rate. We suggest that the persistence of the early marine (predation) effect with age shows that there is no effect of hatchery practice on age at maturity. Apparent survival rate variation was not explained when we used conventional physical oceanographic measurements (temperature, salinity, Pacific Decadal Oscillation Index, Northern Oscillation Index, Arctic Oscillation Index, Aleutian Low Pressure Index, Bakun Upwelling Index, timing of spring transition) in our analyses.