Distribution and habitat models of the Eurasian otter, Lutra lutra, in South Korea

Abstract

Eastern populations of the Eurasian otter (Lutra lutra) have declined throughout Asia. Although the South Korean government designated the Eurasian otter as an Endangered Species and Natural Monument, limited information exists on the distribution and habitat of this top predator in riparian ecosystems at this terminus of its distribution. We documented the distribution and modeled the habitat of this species using sign survey data collected in 2009 and 2010. We analyzed otter presence with 21 raster data sets using environmental niche modeling. We evaluated 31 variables as explanatory variables for logistic regression models that considered detection and non-detection data. We detected otter sign at 83% (3,288 of 3,957) of survey sites within 63% (680 of 1,074) of survey cells. In our MaxEnt presence-only model, land cover type was the main contributing factor to predicting habitat suitability, with secondary influences from isothermality, precipitation in the wettest month, elevation, and temperature seasonality. Water quality and human disturbance were the main variables significantly associated with otter occurrence based on the regression model that included detection and non-detection data. Direct measures of human activity, such as traffic volume and land price, were better predictors than stationary factors such as density of buildings. Our results suggest that otters and humans can coexist in highly populated areas. Our information on the distribution and habitat characteristics of Eurasian otters in South Korea will assist habitat management and recovery projects in natural and anthropogenic environments.