Fine-scale genetic structure of brown trout (Salmo trutta)

Abstract

Many freshwater fishes are fragmented into local populations. Brown trout (Salmo trutta) is a fish that express high genetic heterogeneity and are known to be structured into genetically differentiated subpopulations over small limited areas. In this study, five sections in a river (Julussa) and three sections in a stream (Bellbekken) were sampled in order to investigate the genetic population structure. It was predicted that the two rivers would be genetically differentiated as they are separated by a small waterfall that could be a potential migration barrier hindering gene flow. It was also predicted that sample sites in Julussa, as this river had longer distances between sites, would be more differentiated than in Bellbekken, according to the isolation by distance model. Both predictions were investigated using FST values (fixation index) and two different assignment tests; one partially Bayesian and one fully Bayesian, in order to get information about migration between sample sites and population structure. The results confirmed a clear pattern of subdivision between the two rivers; the differentiation values between trout in Julussa and trout in Bellbekken were high (FST = 0.0216-0.0522), the migration was low and the structure test clearly divided the two rivers into two subpopulations. There was weak evidence for an isolation by distance genetic structure (r2 = 0.069, p= 0.0790), indicating that other reasons than distance are more important for the pronounced subdivision. There was clear evidence for significant differentiation between sites in Bellbekken (FST = 0.0051 - 0.0213) and relatively high site specific assignments. This indicates genetic divergence in this stream, which is probably due to a low effective population size and low effective migration, resulting in genetic drift. There was no clear evidence for substructure within Julussa (FST = 0.0013 - 0.0100), which could indicate a larger effective population size and a higher rate of gene flow. Samples from the same site taken during sampling sessions were not significantly different, indicating a stable population structure, at least on a short time scale.