Unraveling the interplay between demography and landscape features in shaping connectivity and diversity: Insights from the leopard cat on a subtropical island

Elucidating how demography and contemporary landscape features regulate functional connectivity is crucial to implementing effective conservation strategies. We assessed the impacts of landscape features on the genetic variation of a locally endangered carnivore, the leopard cat (Prionailurus bengalensis) in Taiwan. We aim to evaluate the association between genetic structure and landscape features. We further predicted the changes in genetic diversity and suitable habitats in the future. We genotyped 184 leopard cats in western Taiwan using 12 nuclear microsatellites and a mitochondrial marker. We applied a landscape optimization procedure with two genetic distances to identify major genetic barriers and employed ecological niche modeling to predict the future distribution of the leopard cat. Bayesian demographic inferences revealed a dramatic population decline for all leopard cat populations in Taiwan. Genetic clustering and resistance surface modeling supported that the population connectivity was influenced by highways and high elevation. Niche modeling indicated low temperature was one of the primary factors limiting the occurrence of leopard cats that may inhibit their movement in high elevations. We predicted the suitable habitats of leopard cats would shrink northward and towards higher altitudes with rugged topography in response to global warming. Our study provided genetic evidence that leopard cats in Taiwan had undergone a dramatic population decline that may be associated with anthropogenic impacts. We also inferred the anthropogenic linear feature compromised the connectivity and persistence of leopard cats in human-mediated landscapes. Our finding serves as a model for landscape genetic studies of island carnivores in subtropical regions.