AbstractObjectiveGenetic stock identification (GSI) can be an effective tool for fisheries management, but development of reference baselines for species with broad geographic distributions can be challenging. Mixed‐stock fisheries for Chinook Salmon Oncorhynchus tshawytscha have utilized GSI analyses for decades with various genetic baselines, but these have largely become outdated with advances in technology that enable more efficient genotyping. Thus, our goals were to (1) create nested baselines of genotypic data for Chinook Salmon throughout their entire natural range using existing data from multiple sources and (2) evaluate the utility of those nested baselines to conduct accurate hierarchical GSI of mixture proportions or the stock identification of individual fish.MethodsIn this study, we compiled a large genetic baseline of single‐nucleotide polymorphism (SNP) markers for 389 populations that encompass the entire geographic range of Chinook Salmon. We used cross validation and realistic mixture simulations to test the accuracy of the baseline in generating GSI estimates.ResultWe demonstrated that a multi‐tiered assignment approach can provide high accuracy at both tier 1 (broadscale, with three coastwide reporting groups; 97.8% mean accuracy) and tier 2 (fine‐scale regional reporting groups; up to 97.7% mean accuracy) levels. Realistic mixture simulations showed that this multi‐tiered approach can provide highly effective GSI results for several common mixed‐stock fisheries applications in the Pacific Ocean.ConclusionThis new SNP baseline and the multi‐tiered assignment approach provide the most comprehensive rangewide GSI baseline for Chinook Salmon over any previous application and enable highly accurate estimates for GSI purposes.
Read full abstract