Linking genetic structure, landscape genetics, and species distribution modeling for regional conservation of a threatened freshwater turtle

• Published in: Landscape ecology 2022-04, Vol.37 (4), p.1017-1034
• Main Authors: McCluskey, Eric M., Lulla, Vijay, Peterman, William E., Stryszowska-Hill, Kinga M., Denton, Robert D., Fries, Anthony C., Langen, Tom A., Johnson, Glenn, Mockford, Stephen W., Gonser, Rusty A.
• Format: Article
• Language: English
• Subjects: Aquaporins; Biomedical and Life Sciences; Connectivity; Conservation; Cultivated lands; Ecology; Environmental Management; Gene flow; Genetic analysis; Genetic distance; Genetic structure; Genetics; Geographical distribution; Habitats; Land cover; Landsat; Landscape; Landscape architecture; Landscape Ecology; Landscape/Regional and Urban Planning; Life Sciences; Modelling; Nature Conservation; Population genetics; Remote sensing; Research Article; Species; Sustainable Development; Turtles
• ISSN: 0921-2973; 1572-9761
• DOI: 10.1007/s10980-022-01420-0

Context Regional conservation efforts should incorporate fine scale landscape genetic and habitat suitability data for management decisions. This information permits conservation measures to be tailored to a specific landscape. Objectives We investigated the landscape determinants of gene flow and habitat suitability for the state-threatened Blanding’s turtle ( Emydoidea blandingii ) in northeastern New York (NNY). We applied the results from each to examine their complementary contributions to local connectivity and genetic structuring. Methods We conducted population and individual-based genetic analyses with microsatellite data to evaluate genetic structuring and landscape genetics in NNY. We coupled these genetic analyses with species distribution modeling (SDM) to estimate the extent of suitable habitat across this important region for species persistence in the state. Results Gene flow was strongly associated with open water and cultivated land, indicating the role open water channels play in connecting neighboring activity centers, and the propensity of females to select cultivated land to nest. Species distribution models based on Landsat-derived vegetation indices and percentage of scrub-shrub wetlands accurately identified Blanding’s turtle habitat. Connectivity estimates from our NNY focal area using landscape genetic and SDM resistance surfaces showed potential movement constraints between the two genetic clusters. Conclusions Land cover better explained genetic distance data than geographic distance for Blanding’s turtles in our focal area. Accurate SDMs were developed for our focal area with a small number of occurrences (