Using Serially Collected Specimens to Investigate the Potential Population Genetic Consequences of Reported Declines...

Biodiversity is facing global change at an unprecedented rate. Understanding how populations have responded to accelerated change over the last century is key to informing effective conservation policies. Serially collected specimens from natural history repositories can provide a window into how populations change over time and highlight further vulnerabilities in remaining populations. Changes in observed population abundance during field surveys suggest that some Plethodon salamander populations experienced declines since the 1960s, but the potential population genetic consequences of these declines remain unstudied. Thanks to decades of sustained collection-based efforts, Plethodon salamanders serve as a model to test the utility of historical DNA to identify shifts in genetic diversity at recent time scales. Here, we investigate demographic change in six Plethodon species through time using DNA from formalin-fixed museum specimens (1960s-1970s), historic frozen blood (1980s-1990s), and contemporary sampling. We generated several reduced representation SNP datasets using a target-capture approach to investigate two sites in the Appalachian Range: one with documented declines (Indian Grave Gap) and one without (Skull's Gap). We quantify the impact of bioinformatic choices on estimates of genetic diversity, quantify demographic shifts, and trace changes in allele frequencies in immune-related loci to explore the potential impact of pathogens on putative declines. We found consistent patterns of genetic diversity change across datasets and filtering regimes. At Skull's Gap, our results suggest that populations were stable or expanding, while at Indian Grave Gap, our results suggest contraction in one species and mixed signals of contraction and expansion in the others. Analyses of immune loci suggest that balancing selection is maintaining shared polymorphism through time in all but one species. Our study outlines important considerations for leveraging historical DNA in time series collections to quantify the genomic effects of localized population declines.