Genetic clustering methods reveal bull trout (Salvelinus confluentus) fine-scale population structure as a spatially nested hierarchy

Abstract

Many conservation genetics studies in fishes define populations based on capture location. In salmonid fishes, this traditional a priori designation is made by spawning stream, with subsequent post hoc approaches used to define units of conservation. In this study of bull trout from southwestern Alberta, we provide evidence that a model-based Bayesian genetic clustering method may provide a more parsimonious alternative to designating population structure and units of conservation in comparison to traditional methods. The clustering method captured a hierarchical model of population structure, in which seven local populations were nested within three genetic archipelagos. This was in contrast to using simple F ST based approaches between thirteen a priori designated populations, which found significant differences for nearly every pairwise comparison. In addition, assignment tests results from Bayesian clustering revealed that movement may be common between sampling locations. These clustering methods are easy to use, intuitive and provide substantial information on populations of fish; this study provides an example of their utility for local fisheries management and conservation.