Comparative landscape genetics of two pond-breeding amphibian species in a highly modified agricultural landscape.

Evaluating fine-scale population structure of multiple species in the same landscape increases our ability to identify common patterns as well as discern ecological differences among species' landscape genetic relationships. In the Palouse bioregion of northern Idaho, USA, 99% of the native prairie has been converted to nonirrigated agriculture and exotic grasslands. Columbia spotted frogs (Rana luteiventris) and long-toed salamanders (Ambystoma macrodactylum) in this area breed almost entirely in artificial ponds on private land. We used genetic distances (F(ST) and D(c)) derived from eight microsatellite loci in 783 samples to evaluate the relationships among sympatric breeding populations (N = 20 and 26) of these species in a 213-km(2) landscape. Both species showed a pattern of isolation by distance that was not improved when distance was measured along drainages instead of topographically corrected straight lines (P < 0.01). After testing for autocorrelation among genetic distances, we used an information theoretic approach to model landscape resistance based on slope, soil type, solar insolation, and land cover, and multi-model inference to rank the resistance of landscape surfaces to dispersal (represented by genetic distance). For both species, urban and rural developed land cover provided the highest landscape resistances. Resistance values for long-toed salamanders followed a moisture gradient where forest provided the least resistance, while agriculture and shrub/clearcut provided the least resistance for Columbia spotted frogs. Comparative landscape genetics can be a powerful tool for detecting similarities and differences between codistributed species, and resulting models can be used to predict species-specific responses to landscape change.