Landscape Genetics Map for Salmon in the Yukon and Kuskokwim Rivers and Norton Sound US Fish and Wildlife Service Conservation Genetics Laboratory, Anchorage Alaska
Map Downloads Project Description Help

Disclaimer:

The findings and conclusions presented here are those of the author(s) and do not necessarily represent the views of the U.S. Fish and Wildlife Service.

Project Report:

For a copy of the complete report: -Not Available Yet- This report is in PDF format for a free copy of the Acrobat Reader from Adobe, click the icon below:

Contact Information: Conservation Genetics Laboratory US Fish and Wildlife Service 1011 East Tudor Road MS 331 Anchorage, Alaska 99503

Project Description

Introduction: Landscape genetics, an integration of molecular population genetics and landscape ecology, provides a framework to evaluate how habitat, as well as historic events (e.g., glaciation, stream capture), influence genetic diversity at different spatial scales. Multi-species studies using landscape genetic techniques are particularly informative for conservation because they can highlight landscape variables that influence genetic diversity in a variety of species. We used this approach to evaluate the interacting role of life history and habitat on population structure in Chinook, chum and coho salmon from three major watersheds (Norton Sound and the Yukon and Kuskokwim rivers) near the northern extent of the species range in North American. Our objectives were threefold. First, we assessed if the level of population divergence from least to most was chum < coho < Chinook as predicted based on their life history differences. Second, we assessed if the spatial pattern of population structure in each species was organized hierarchically around the three watersheds. More generally, we assessed if the patterns of hierarchical population structure in each species were congruent. Finally, we assessed and compared the extent to which nine habitat features from four general categories (spatial isolation, habitat size, climate, and ecology) explained population structure in each species. The results were evaluated in the context of current management and conservation with an emphasis on environmental perturbations from factors such as climate change.

Brief Summary of Results: The level of population structure in coho salmon was significantly larger than the level of population structure in chum and Chinook salmon. For comparison, open the layers titled Fst for each species and compare the legend values. The levels of population structure for Chinook and chum salmon were not significantly different. The spatial patterns of population structure were not organized hierarchically around the three watersheds. The analysis incorporating genetic and geographic data suggests hierarchical population structure for each species occurs primarily along a latitudinal axis (east – west) which is dominated by the Yukon River, rather than the much shorter longitudinal axis (north-south) that defines the relative position of the three major watersheds. This result is evidenced by the large single coastal population group in each species that includes populations from Norton Sound, the lower Yukon River and most of the Kuskokwim River. The other population groups occur primarily within the Yukon River. For comparison and more information, open the layers titled Landscape Genetic Groups for each species and refer to the meta data. Further evidence of a strong latitudinal (or coastal – inland) pattern for each species was revealed when the first principle component (PC1) of genetic variation from each population was extrapolated across the three watersheds using ordinary Kriging. For comparison and more information, open the layers titled PC1_Kriging for each species and refer to the meta data. Nine variables representing habitat were evaluated: waterway distance to the coast, median pairwise waterway distance from each location to all other locations (similar to connectivity), elevation, migration difficulty (waterway distance to the coast x elevation), home subbasin area, home river length, mean annual precipitation, ecoregion, and permafrost region. No single habitat variable was strongly correlated to population structure (as measured by populations specific Fst) in any species. However, the variable mean annual precipitation was one of two or three variables that, in combination, may partially explain the population structure of each species. The other variables were waterway distance to the coast and home subbasin area (Chinook), elevation (chum), and median pairwise waterway distance from each location to all other locations (coho). For comparison and more information open the environmental background layers and refer to the meta data for the Fst layers.