Habitat suitability and connectivity modeling predict genetic population structure and priority control areas for invasive nutria (Myocastor coypus) in a temperate river basin

The nutria (Myocastor coypus), also known as the coypu, is a semi-aquatic, invasive rodent native to South America that causes damage to natural riverine and wetland habitats in many parts of the world, including South Korea. Understanding habitat use, connectivity, and gene flow of nutria populations is critical for the sound management of local and regional ecosystems. Here, we assessed habitat suitability and connectivity in relation to the genetic structure of nutria populations in the Nakdong River Basin of South Korea. A total of 321 nutria occurrence sites and seven environmental variables were used to perform ensemble habitat suitability modeling using five species distribution models (SDMs), including boosted regression trees, maximum entropy model, random forest, generalized linear model, and multivariate adaptive regression splines. Using graph and circuit theory approaches, we assessed the population gene flow and current flow betweenness centrality (CFBC) of suitable habitats derived from the ensemble SDM. All SDMs performed well with a range of test AUC values from 0.962 to 0.970 (mean = 0.966) with true skill statistic values over 0.8. The minimum temperature of the coldest month, mean temperature of the warmest quarter, precipitation of the driest quarter, and distance from water bodies were important predictors in nutria habitat modeling. Nutria population gene flow was significantly correlated with the least-cost path distance on a cost resistance surface based on ensemble habitat suitability modeling and roads (Mantel’s r = 0.60, p < 0.05). Finally, the CFBC positively correlated with the genetic diversity of nutria populations was used to identify priority control areas. Habitat suitability and connectivity modeling not only revealed environmental conditions and areas that support the survival and spread of nutrias, but also improved our understanding of the animals’ genetic population structure, thereby indicating priority areas to target for eradication.