29-4 Size-Specific Maturation Probability in California Chinook - Variation Across Years and Among Runs

William Satterthwaite , Department of Applied Mathematics and Statistics, University of California Santa Cruz, Santa Cruz, CA
Michael Mohr , Southwest Fisheries Science Center, National Marine Fisheries Service, Santa Cruz, CA
Michael O'Farrell , Southwest Fisheries Science Center, National Marine Fisheries Service, Santa Cruz, CA
Brian Wells , NOAA Fisheries/Southwest Fisheries Science Center, Santa Cruz, CA
Salmonid life history theory posits that the timing of maturation and spawning reflects a tradeoff between the potential for further growth in the ocean (associated with larger size at spawning, predicted to yield fecundity and competitive advantages) and the associated additional mortality risk and generation time effects.  As a result, we predict that maturation probability of individuals may vary as a function of both their current size and recent growth rate, with the maturation probability at a given size also affected by the advantages conferred large size at spawning as well a the risk of ocean mortality.  We looked for evidence of size-dependent maturation in multiple runs of California Chinook salmon based on estimates of their age-specific size distributions in the ocean and on the spawning grounds.  We estimated ocean size distributions from harvest of coded wire tagged fish, correcting for truncation introduced by size limits in the fishery.  We estimated spawner size distributions from surveys of spawning grounds and hatchery returns.  We then compared ocean and spawner size distributions at appropriate lags in an attempt to infer size-dependent maturation.  We examined variation among runs and across time within runs to explore the relationship between growth history (mean size at age) and size-dependent maturation, testing for the expected relationship between growth rates and maturation and testing for correlations between growth rates, environmental covariates, and inferred maturation size thresholds.