The influence of environmental heterogeneity on neutral and adaptive genetic structure in Dall's sheep (Ovis dalli dalli)

Information obtained from neutral genetic markers such as gene flow and population structure, paired with information from genes affected by natural selection can promote a better understanding of how populations adapt to different environments. Joint analysis of these maker types also allows exploration of how the interaction of evolutionary mechanisms contributes to adaptive differentiation. For my dissertation, I investigated how landscape features and environmental heterogeneity influence the structure of neutral and adaptive genetic variation in Dall's sheep (Ovis dalli dalli ) at both landscape and range-wide scales. I analyzed empirical and simulated genetic data and synthesized genomics and habitat modeling approaches to understand the effect of environmental variation, selection strength, effective population sizes, and dispersal distances on genetic structure and local adaptation. At the landscape level I found that although Dall's sheep dispersal is limited by major terrain features (icefields and valleys), large effective population sizes and contiguous expanses of habitat promotes population connectivity and enables dispersal for both male and female sheep. By directly comparing habitat selection and landscape genetic methods with independent modelling procedures, I demonstrated many habitat types were important for summer resource selection and significantly correlated with gene flow. However, certain preferred habitats (rugged terrain, mid-range elevation) were not influential to breeding dispersal, suggesting habitat selection and landscape genetics models assess different ecological functions at different temporal and spatial scales. Therefore, consideration of both selection of habitat for meeting daily needs and for gene flow ensures resource management will meet survival needs and allow for long-term genetic connectivity of Dall's sheep. When analyzing patterns of differentiation at the range-wide scale in putatively adaptive loci (in combination with neutral loci), I detected genetic sub-structuring in the northwestern and central portions of the Dall's sheep range not identified with neutral loci alone. These results, in addition to significant associations between candidate adaptive loci with climate and landscape variables, suggest local environmental adaptation and natural selection. This work advances research on natural selection in spatially heterogeneous environments by providing insight into factors that influence adaptive vs. neutral population genetic structure and the best methods to detect them.