Genetically-informed population models improve climate change vulnerability assessments

Context Climate change will cause species extinctions that will be exacerbated by human-caused landscape changes, preventing species from tracking shifting climatic niches. Although incorporating functional connectivity into prospective population models has proven challenging, the field of landscape genetics provides underutilized tools for characterizing functional connectivity. Objectives The aim of this study was to explore how genetically-derived representations of dispersal affect assessments of environmental change impacts using a spatially-explicit population modelling approach. We illustrated the utility of this approach to test hypotheses related to the effects of dispersal representation and environmental change for the IUCN-threatened Blanding's Turtle (Emydoidea blandingii). Methods We integrated existing demographic and genetic datasets into a spatially-explicit metapopulation modelling framework. We ran several sets of simulations with varying dispersal representations (distance-based, landscape resistance-based with either static or changing land cover) to explore how landscape genetic estimates of connectivity impact estimates of extinction risk. Results Models incorporating land cover-based dispersal resulted in lower patch occupancy than simulations where dispersal was only a function of interpatch distance. Furthermore, both climate change-induced declines in habitat suitability and land use change-induced declines in connectivity reduced abundance and patch occupancy. Conclusions Incorporating landscape genetics into population models revealed that choices involved in dispersal representation alter both extinction risk and path occupancy, often altering the distribution of extant patches by the end of simulations. As technological advances continue to increase access to landscape genetic datasets, we suggest that researchers carefully consider how genetic resources can be used to improve climate vulnerability assessments.