COS 42-6
Genetic evidence for the glacial refugia and postglacial migration of mountain hemlock in the Pacific Northwest

Tuesday, August 11, 2015: 3:20 PM
319, Baltimore Convention Center
Matias C. Fernandez, Plant Biology, University of Illinois at Urbana-Champaign, Urbana, IL
Feng Sheng Hu, Department of Plant Biology, Department of Geology, and Program in Ecology, Evolution, and Conservation Biology, University of Illinois at Urbana-Champaign, Urbana, IL
Katy D. Heath, Plant Biology, University of Illinois at Urbana-Champaign, Urbana, IL
Background/Question/Methods

           Many plant species may be unable to migrate or adapt quickly enough to keep pace with anticipated rates of climate change. Climate refugia, locally suitable habitats where species persist through periods of climatic fluctuation, allow for pockets of diversity within a regional climate that is otherwise inhospitable. Identifying the presence of such refugia during the last glaciation is necessary for assessing the pattern and rate of post-glacial plant migration in response to historical climate change. Understanding such vegetation dynamics can provide insight for anticipating future climatic impacts. The conifer species mountain hemlock (Tsuga mertensiana) currently has two disjunct distributions in the Pacific Northwest: a coastal range and an interior range. At coarse spatial scales, pollen records suggest a northward postglacial expansion of mesic forests from a southern coastal refugium and recent recolonization of the interior range in the late Holocene. However, the locations of glacial refugia and migration patterns remain unclear because of the inherent limitations of using pollen to detect small populations of certain plants. We used genotyping-by-sequencing (RAD-Seq) to study patterns of genetic variation across the modern range of mountain hemlock and test for the presence of refugia.

Results/Conclusions

            Analysis of 95 individual trees across 19 sampling sites yielded a total of 2,254 genome-wide SNPs. These data suggest two genetically distinct clusters of mountain hemlock populations. Populations in the northern cluster show deep genetic differentiation from those in the southern cluster (mean FST = 0.48). Among-site genetic differentiation is low in the north (mean FST = 0.05) and moderate in the south (mean FST = 0.14). Furthermore, northern coastal sites cluster with northern interior sites, and southern coastal sites cluster with southern interior sites. The mean number of private alleles is 10.5 across our 19 sites, with higher numbers in north-coastal British Columbia (15 alleles), northern Oregon (19 alleles), eastern Oregon (15 alleles), and northern California (22 alleles). These preliminary results suggest post-glacial expansion from at least two coastal refugia, and the presence of two distinct migration corridors between the coastal and interior ranges. Private alleles and genetic differentiation within the southern population also suggest the possibility of a southern interior refugium. These genetic data offers new insights into the history of mountain hemlock and helps resolve longstanding questions regarding the responses of mesic trees to past climate change in the Pacific Northwest.