A factor limiting the development of general theories within the field of landscape genetics has been the shortage of models capable of predicting spatial genetic patterns based on complex life history attributes such as dispersal behavior, or resulting from dynamic landscape features. Recently developed modeling software such as Circuitscape (McRae et. al. 2008, Ecology 10:2712-2724), CDPOP (Landguth and Cushman, 2010, Mol. Ecol. Resour. 10:156-161), and UNICOR (Landguth et. al. 2012, Ecography 35:9-14), have begun to be applied to examine underlying assumptions and limitations within the field of landscape genetics. However, the discipline’s long-term growth will require tools that capture the additional behavioral and ecological realism necessary to perform management-relevant forecasting and work quantitatively with adaptive genetic traits. Here, we look at the short and long term goals of the discipline, and ask to what extent current modeling tools will be able to meet these demands.
Results/Conclusions
The principal focus of work in landscape genetics has thus far involved hindcasting possible barriers to gene-flow using the spatial distribution of allelic variation from neutral markers. Simulation work to date has provided valuable insight into sampling design and limitations to hindcasting, and is just beginning to address evolutionary landscape genetics using adaptive markers. However, the need to develop management strategies that ensure species’ viability in the midst of climate change, invasive species, habitat loss, and other anthropogenic disturbances will force landscape geneticists to do more detailed forecasting incorporating species and stressor interactions, and individual behavioral traits into their models. We illustrate an approach for doing just this using the recently developed HexSim model, which includes a landscape genetics toolkit. Our HexSim example involves a simulated predator-prey system in which the benefits of prey capture efficiency impart a selective pressure upon the predator genome. We use this example to illustrate how advances in simulation model development might assist landscape genetics in meeting the practical challenges the future is sure to hold.