Run reconstruction of mixed-stock salmon fisheries using age-composition data

A method for using age-composition data to determine stock-specific migration timing and abundance in a mixed-stock salmon fishery is developed. The Chignik sockeye salmon (Oncorhynchus nerka) fishery has two stocks, but only aggregate catch and escapement data are available. The age composition of the two stocks, however, is known to be consistently different, and age-composition data are collected from one stock at the beginning of the commercial fishing season and from the commercial catch throughout the season. Using the changes in age composition in the commercial catch throughout the season, we estimate the total abundance and migration timing for the two Chignik stocks using maximum likelihood and Bayesian methods. The outcomes of this stock separation model are highly correlated with that of scale pattern analysis for most years from 1978 to 2002 (r = 0.89). The results suggest that age composition may provide salmon managers with a reliable and inexpensive method for determining stock-specific migration timing and abundance in a mixed-stock fishery.