Phylogeography and molecular phylogenetics of the hoary marmot (Marmota caligata)

Abstract

In this dissertation I documented the phylogeographic history of the Hoary Marmot (Marmota caligata) and its phylogenetic relationships with the Vancouver Island (M. vancouverensis) and Olympic (M. olympus) marmots. The Hoary Marmot is an iconic alpine mammal that is broadly distributed throughout the Pacific Northwest (PNW) from Washington and central Idaho in the south to Alaska in the north. Vancouver Island and Olympic marmots have much more restricted geographic distributions, occurring only on Vancouver Island (British Columbia, Canada) and the Olympic Peninsula (Washington, USA), respectively. In my first chapter I used mitochondrial DNA (mtDNA) sequence data to document the existence of 2 mtDNA clades in Hoary Marmots. I also used mtDNA and nuclear sequence data to infer historic gene flow from Hoary into Vancouver Island marmots, which resulted in the latter "capturing" the mitochondrial genome of the former. Analyses of nuclear sequence data also suggested the potential for historic gene flow between Hoary Marmots and Olympic Marmots in Washington. In my second chapter I investigated the origins of Hoary Marmots on Sud Island, Alaska, part of the Maritime National Wildlife Refuge. This island population of marmots was purported to have been introduced by humans and detrimental to nesting seabirds. As a result, the United States Fish and Wildlife Service undertook efforts to eradicate the Hoary Marmot population on Sud Island between 2009-2011. I conducted a literature review of marmot introductions in Alaska and used molecular data to determine the geographic origin of marmots on Sud Island. Through my literature review I found no direct evidence that marmots were introduced to Sud Island or any documentation that they were detrimental to nesting seabirds on this island. Molecular analysis identified the Hoary Marmot population on Sud Island as a distinct genetic cluster, with divergence time estimates similar to those of a naturally occurring island population, suggesting a natural colonization of Sud Island by Hoary Marmots. In my third chapter I investigated potential refugia used by Hoary Marmots during the Last Glacial Maximum (LGM) and the potential effects of climate change on the future distribution of suitable habitat. To address these questions I used species distribution models (SDMs) based on all available museum specimens and population genetic summary statistics calculated from mtDNA sequence data. I found the most likely areas of LGM refugia were located south of the glacial margins of the Pleistocene and along the PNW coast. Habitat in the southernmost portion of the Hoary Marmot current geographic distribution was predicted to be the most negatively impacted by future climate change. Additionally, populations from this region were the most genetically diverse, indicating that these populations may be important for conservation of the species as a whole. In my final chapter I used microsatellite and sequence (mtDNA and nuclear) data to revisit the findings of my first chapter and to test for gene flow between Hoary, Vancouver Island, and Olympic marmots, as well as between the 2 Hoary Marmot mtDNA clades. I also improved the known distribution of the Hoary Marmot mtDNA clades by determining clade membership of 98 museum specimens for which no fresh tissues exist. Analysis of the combined sequence and microsatellite data confirmed previous findings that introgression led to Vancouver Island Marmots capturing the mitochondrial genome of Hoary Marmots. The addition of microsatellite data did not resolve the origin of nuclear alleles shared between Hoary Marmots from Washington and Olympic Marmots. Regarding the 2 Hoary Marmot mtDNA clades, molecular results suggested unidirectional gene flow between the clades and that male-biased dispersal is likely occurring in the species. The additional mtDNA clade membership data from the 98 museum specimens revealed that British Columbia is predominantly occupied by a single mtDNA clade. Overall, my research has shown that populations in the southern portion of the Hoary Marmot's geographic distribution are likely to be the most important for conservation and that additional research in this region is needed. I also documented the existence of introgression between Hoary and Vancouver Island marmots, highlighting the importance of using multiple unlinked loci for phylogenetic and phylogeographic analysis. Lastly, my findings call attention tothe importance of rigorously verifying primary sources of information before undertaking species eradications.