Landscape and Yellowstone amphibians: Insights from genetic, developmental and community variation through space and time

As reports of amphibian declines become increasingly common worldwide, the amphibian decline crisis has become evident even in remote and highly protected areas. In addition to the threats posed by emerging infectious disease and anthropogenic destruction of habitat and other challenges, amphibian populations are specifically threatened by climate change-induced wetland destruction. In the first chapter of this thesis, I documented a substantial decline in amphibian community presence in northern Yellowstone National Park. The declines in the three most common regional amphibian species were linked to disappearances of regional wetlands and pond breeding environments, as well as disease. Using remote sensing of wetland presence, in situ surveys of pond condition and analysis of historical climate records, I attributed the disappearances of pond habitats to extended drought and trends of long-term climate change. The Lamar Valley in northern Yellowstone supports dozens of highly variable glacial kettle ponds where amphibians live and breed. In the remaining chapters, I used the naturally occurring environmental variation of this region to determine how the genetics and developmental biology of the blotched tiger salamander Ambystoma tigrinum melanostictum varied with spatial environment. In the second chapter of my thesis, I examined developmental strategies of A. t. melanostictum under different developmental conditions. Individuals in the most rapidly drying ponds showed the most rapid developmental profiles, and truncated juvenile development to metamorphose at extremely small sizes. As the landscape continues to dry, I predict an increase in the frequency of these lower-quality adults, as well as decreases in adult recruitment and reproductive output of the entire population. To examine the structure of the A. t. melanostictum metapopulation in the context of this environmentally-mediated developmental heterogeneity, in the third chapter, I used 8 neutral microsatellite markers to characterize the genetics of the population. A nested geographic analysis revealed that an isolation-by-distance pattern was prevalent throughout the population at multiple spatial scales. However, I found that at local geographic scales, gene flow patterns corresponded with the conditions of individual pond environments rather than with distance, landscape features or geographic barriers. Even with rapid developmental periods and low reproductive output, networks of ephemeral ponds showed robust metapopulation structure. Nonetheless, changing environmental conditions continue to decrease the reproductive output of metapopulation patches, and destruction of permanent ponds that serve as source populations can destabilize the entire population network. To examine the historical trends of this population, I analyzed the genetics of the species through time. In the final chapter of my thesis, I analyzed a 700 base-pair region of the mitochondrial genome across space and through 3300 of years of recent geologic time. I found that, although there have been many fluctuations in climate and population dynamics, mitochondrial genetics have remained in stasis through time. Using serial coalescent modeling, I found that the Ambystoma population of the Lamar Valley has remained at an extremely low effective population size for millennia, possibly due to regular landscape and climate-mediated population bottlenecks. These four studies constitute a detailed analysis of development, migration and genetics of an amphibian population through both short and long-term environmental changes. Correlating amphibian metapopulation responses on multiple spatial and temporal scales, the insights of these manuscripts can contribute to conservation strategies that will preserve amphibian populations in a rapidly changing world.