UC Santa Cruz

The degree of genomic diversity of a species, how that diversity is partitionedover space, and how it has changed over time are critical aspects that inform the continued viability of a species in a changing environment. Once restricted to humans and model species, decreased costs of next generation sequencing and improved analytical methods have enabled genomic studies of threatened and endangered non-model species, contributing to more effective conservation and management. In this dissertation I generate new genomic data and provide insights into four aquatic mammals, each of which have unique natural histories and conservation needs. In chapter one, I used spatially dense spatial genomic sampling to understand the distribution of diversity and inbreeding in southern sea otters. Consistent with other studies, I found evidence of a genomic bottleneck that pre-dates the fur trade, likely due to indigenous hunting. I showed that southern sea otters are less diverse than their northern sister subspecies across all measures, likely a legacy of their long term isolation at the southern end of the sea otter range, sequential bottlenecks, their reduction to a single small population by the maritime fur trade, and their current geographic restriction. My results indicate that although southern sea otters have little spatial variation in neutral genomic diversity, rates of inbreeding and genetic load are significantly higher in the northern part of their small range. These results highlight the vulnerability of southern sea otters - as they are currently a single population and cannot expand their range naturally - and underscore the importance of a metapopulation structure in maintaining and improving the genetic diversity of the species. Translocations of southern sea otters to northern California and Oregon are likely necessary to restore a metapopulation structure. Furthermore, given the ecological importance of sea otters, improving the outlook for southern sea otters is critical to maintaining the viability of coastal kelp forest ecosystems at their more southerly range as the climate continues to change. In chapter two, I assembled a highly contiguous reference genome for the dugong. While a single genome is insufficient to represent the full diversity of this wide-ranging species, it provides initial insights into the demographic history and diversity of a centrally-located population and will serve as an important resource for future studies. I showed that dugongs have relatively high genome-wide heterozygosity compared to other Vulnerable mammals and that they have a dynamic demographic history that likely reflects Pleistocene glacial cycles and resulting sea level change. Future whole genome resequencing studies will provide useful insights into more recent dugong demographic history, as well as how neutral and adaptive variation are partitioned across their large, but discontinuous geographic range, allowing for more targeted management strategies. In chapter three, I use whole genome sequencing from museum samples of historic Alaskan and Russian polar bears to investigate two main questions: 1. How do polar bears from understudied Russian subpopulations fit in the range-wide diversity of the species? And 2. How has Alaskan polar bear diversity changed over the past 150 years in response to human hunting and climate change? For question 1. I found that despite broad geographic sampling across four management units, polar bears from across Russia are closely related to each other and to historic Alaskan bears. This result highlights earlier findings, which indicate that the scale of polar bear population structure is highly variable and does not correspond to management unit boundaries. For question 2. I found that Alaskan polar bear genomic diversity has declined significantly over the past 150 years, with the majority of diversity loss occurring in the second half of the 20th century, likely due to heavy sport hunting. There is also evidence of a potential population replacement in Alaska occurring sometime after 1970, potentially also due to abundance declines from sport hunting. In chapter four, I expand beyond a single species focus to a more holistic paleoecosystem approach by using sedaDNA techniques to investigate the arrival and persistence of beavers in Grand Teton National Park over the last 10 ka and their interactions with the local climate and vegetation. My findings show that beavers arrived surprisingly late to this region following Pleistocene deglaciation, but thereafter persisted at the watershed scale for the last 5 ka, despite periods of environmental change and regional drought. Their arrival coincided with a regional mid-Holocene neoglacial advance, likely due to increased water availability. Beaver arrival was also associated with a shift from a more coniferous vegetation regime to increased riparian vegetation and higher vegetative diversity. Determining the relative contribution of beavers versus climate in structuring the local plant community will require further study. These results suggest that under certain conditions, the positive effects of beaver engineering on local ecosystems may persist over millennia despite drought and other environmental changes, an encouraging finding that suggests that beaver restoration may be an effective long term solution for conserving ecosystems and mitigating the effects of climate change. These chapters provide novel insights into the genomic diversity of these four species, and improved understanding of their spatial and temporal vari- ation, particularly the effects of human exploitation and past and present climate change. Additionally, I have generated high-quality genomic resources which will be made publicly available and will contribute to future studies.