Estimation of stock composition and individual identification of Chinook salmon across the Pacific Rim by use of microsatellite variation

Variation at 13 microsatellite loci was surveyed for over 52,000 Chinook salmon Oncorhynchus tshawytscha sampled from 325 localities ranging from Russia to California; the variation was applied to estimate stock composition in mixed-stock fishery samples. A rapid increase in the accuracy of estimated stock composition in simulated mixtures with respect to population sample size was observed for sample sizes of up to about 75 individuals, at which point a 90% accuracy of assignment to population was achieved. The number of alleles observed at a locus was related to the power of the locus in providing accurate estimates of the stock composition of single-population mixtures. In analysis of single-population mixtures where the Pacific Rim baseline was used for estimation of stock identification, 75% accuracy for the average population was achieved by employing approximately 55 alleles in the analysis. Increasing the accuracy of the estimated stock composition to 90% for the average population required approximately 350 microsatellite alleles. The precision of estimated stock composition increased rapidly for approximately the first 100 alleles used; standard deviations declined from 20.0% to 8.0%. Analysis of known-origin samples indicated that accurate regional estimates of stock composition were obtained. The accuracy of assigning individuals to a specific region or river drainage averaged 84% for 54 populations in multipopulation samples. The estimated stock compositions of mixed-fishery samples from northern and southern locations in British Columbia were quite different among samples and reflected whether samples were derived from migrating or resident Chinook salmon. Microsatellites have the ability to provide accurate estimates of stock composition from many fisheries in the Pacific Rim distribution of Chinook salmon.