Moving up: Using climate, physiology, and gene flow to characterize current and future geographic range limits in montane salamanders

Abstract

What causes, maintains, and changes species’ geographic ranges are central questions in ecology and evolution. Geographic ranges are a complex product of both ecological and evolutionary processes, reflecting current biotic and abiotic conditions as well as gene flow, drift, adaptation, and history. It is only through understanding the factors that influence species past and present distributions that we can begin to accurately predict how these distributions will change in the future. Anthropogenic climate change poses a major threat to native biodiversity around the world, but especially in montane systems. Understanding the dynamics at these lower elevation range limits is of particular importance. My dissertation has sought to elucidate why adaptation fails at the range edge and how that influences current and future species distributions. For this work, I focused on mountaintop, terrestrial, lungless salamanders of the genus Plethodon. A commonly invoked hypothesis for the inhibition of range expansion centers around the idea that asymmetrical gene flow from a densely populated range center prevents local adaptation at the range periphery. In Chapter 1, I quantified gene flow and effective population size along a bidirectional elevation transect in the Smoky Mountains, for the species Plethodon jordani. I found evidence for downslope biased gene flow and more dense mountaintop populations. In Chapter 3, I further explored the potential for asymmetric gene flow to limit adaptation by assessing both gene flow and phenotypic differentiation in the species Plethodon ouachitae in the Ouachita Mountains. Unlike my findings in the Smoky Mountains, in the Ouachitas, there was no indication of asymmetrically biased downslope gene flow, even though population density appears to diminish at low elevation. On the majority of transects movement appeared to be biased upslope. Within a single mountain, I found sampling sites were connected by gene flow supporting a single panmictic population within a mountain. Between mountains, I found an overall signature of genetic structure with populations segregating by mountain, supporting prior work that indicated unique mitochondrial lineages on each mountain. Correlative niche models built on occurrence records for each individual mountain indicate that the abiotic conditions occupied by populations on each mountain are different. These same metrics have been used in other work to indicate niche divergence between species and as indication of niche adaptation. However, I found neither differentiation in metabolic rate thermal sensitivity nor differentiation in acclimation ability between populations on different mountains and populations at different elevations. These findings support that mountaintop endemic Plethodon, even in the absence of gene flow shows conservation in these ecophysiological traits. In Chapter 2, I used this species-specific physiology to predict shifts in future distributions for four montane Plethodon in the Southern Appalachians. I was able to predict current range limits with high accuracy using both correlative and mechanistic distribution models for the three mountaintop species. Neither model was able to accurately predict the distribution of the one lower elevation generalist species, most likely because these limits are determined by biotic interactions as well as climate. As hypothesized the mechanistic model forecasted more suitable habitat under almost all future climate scenarios for the three mountaintop species. The choice of global circulation model had an order of magnitude influence on how much suitable habitat was predicted for both distribution modeling methods. All models indicate that these animals will be quickly contracting their distributions upslope.