Landscape influences on northern leopard frog genetic diversity and effective population size in North Dakota

Background/Question/Methods

Northern leopard frogs (Lithobates [Rana] pipiens) have a complex life history that requires a mosaic of habitats including breeding wetlands, terrestrial uplands, and over-wintering wetlands.  Historically, researchers and managers have examined correlates of these habitats as predictors of breeding success and population presence.  However, these efforts have commonly been hindered by the inability to capture long-term trends due to varying environmental, landscape, and population level fluctuations.  Our research objectives were to utilize genetic measures of effective populations sizes (N_E), allelic richness (A_R) and expected heterozygosities (H_E) of L. pipiens populations to understand the importance of required life history habitats.  During 2010, we surveyed populations representing the two primary biogeographical regions of North Dakota; the Prairie Pothole Region (PPR) and the Northwestern Great Plains (NGP).  We optimized and screened 8 previously published microsatellite markers by Hoffman et al. (2003) as well as 3 additional cross-species microsatellites developed by McKee et al. (2011). Geographic information systems (GIS) were used to extract generalized land use and land cover types as well as densities and average distance of wetlands within 15-km radius of sampled populations.  We utilized a model selection approach to evaluate how our derived GIS data may influence N_E, A_R, and H_E.

Results/Conclusions

Expected heterozygosity (H_E ) ranged from 0.71 to 0.90, with the highest levels of H_E in the eastern PPR and a longitudinal decline toward the more arid NGP portion of western North Dakota.  In addition we found a decline in A_R, again with eastern populations having 12.6 alleles per polymorphic loci and declining to 6.4 alleles in the west.  Our resulting measures of N_E ranged from 1,167 to 6,584 individuals with a generalized trend of PPR populations being larger while NGP populations have reduced N_E sizes.  Our model selection results indicate wetland densities to be the variable of greatest importance among our 18 a priori models.  In addition to wetland densities, the average distance between wetlands was found to be a variable of importance, especially for the more arid NGP region where wetland densities are reduced compared to the PPR.  Other model variables, such as cultivation and grasslands, showed little weight and influence among our models.  These results highlight the importance of wetlands as life history attributes involving breeding and over-wintering across both the PPR and the NGP.