Temporal and Spatial Genetic Structure among Some Pacific Herring Populations in Puget Sound and the Southern Strait of Georgia

We investigated temporal and spatial genetic variation in Pacific herring Clupea pallasii collections from six sites in Puget Sound (PS) and the southern Strait of Georgia (SOG), using 12 microsatellite loci. Loci were highly variable with up to 70 alleles per locus (mean 5 30.67 alleles), and observed heterozygosity was high (mean 5 0.823). Analysis of molecular variance (AMOVA) indicated significant structure, with over twice as much variance among sites as among collection years, although among-site variance was mainly due to Cherry Point and Squaxin Pass collections. In an AMOVA excluding Cherry Point and Squaxin Pass, only temporal variance was significant. With the exception of the Cherry Point and Squaxin Pass collections, pairwise genotypic and F_ST tests show some differences among collection years within a site and some genetic overlap among most PS and SOG collections. The Cherry Point and Squaxin Pass collections had no differences in genotypic distributions among collection years, and in cluster analyses the Cherry Point and Squaxin Pass collections each formed groups separate from other PS and SOG collections. Cherry Point herring have a later spawning time than other PS and SOG herring, and Squaxin Pass is physically isolated in southern Puget Sound. We hypothesize that spawn timing differences and spatial isolation generated genetic structure among some Pacific herring in PS and SOG. We suspect that, as in the case of Atlantic herring C. harengus, population genetic structure in Pacific herring in PS and southern SOG is a combination of a larval retention model and a metapopulation model. Because Cherry Point and Squaxin Pass herring are genetically and behaviorally differentiated from other PS and SOG herring populations, this unique variation should be preserved through careful management.