Exploring the population genetic consequences of the colonization process with spatio-temporally explicit models: insights from coupled ecological, demographic and genetic models in montane grasshoppers


L. Lacey Knowles, Fax: +1 734 763 4080; E-mail: knowlesl@umich.edu


Understanding the genetic consequences of shifting species distributions is critical for evaluating the impact of climate-induced distributional changes. However, the demographic expansion associated with the colonization process typically takes place across a heterogeneous environment, with population sizes and migration rates varying across the landscape. Here we describe an approach for coupling ecological-niche models (ENMs) with demographic and genetic models to explore the genetic consequences of distributional shifts across a heterogeneous landscape. Analyses of a flightless grasshopper from the sky islands of the Rocky Mountains of North America are used to show how biologically informed predictions can be generated about the genetic consequences of a colonization process across a spatially and temporally heterogeneous landscape (i.e. the suitability of habitats for the montane species differs across the landscape and is itself not static, with the displacement of contemporary populations into glacial refugia). By using (i) ENMs for current climatic conditions and the last glacial maximum to (ii) parameterize a demographic model of the colonization process, which then (iii) informs coalescent simulations, a set of models can be generated that capture different processes associated with distributional shifts. We discuss how the proposed approach for model generation can be integrated into a statistical framework for estimating key demographic parameters and testing hypotheses about the conditions for which distributional shifts may (or may not) enhance species divergence, including the importance of habitat stability, past gene-flow among currently isolated populations, and maintenance of refugial populations in multiple geographic regions.