Spatio-temporal characteristics and predictions of the endangered leopard cat Prionailirus bengalensis euptilura road-kills in the Republic of Korea

Abstract

Road-kills negatively impact wildlife populations, especially those that are threatened or small in size. The increase of linear structures such as roads or railways causes road-kills. Understanding and knowing where and when road-kill probability is high is important to prevent collisions. The leopard cat (Prionailurus bengalensis euptilrura) is listed on CITES Appendix Ⅱ and is considered as endangered in the Republic of Korea. We used 141 P. b. euptilura road-kill events occurring from 2006 to 2012 in the Republic of Korea with the same number of randomly generated points for spatial analyses. Further, 239 events were used for temporal analyses. Spatial analyses and graph modeling were conducted using geographic information system (GIS) and Graphab software. In landscape analyses, P. b. euptilura road-kills were concentrated around agricultural lands and forest, and less frequent near developed areas. The result of the traffic patterns analyses showed that traffic volume, the number of lanes, and distance from ramps were significantly different between the road-kill points and the random points. The road-kill frequency was significantly different by season, and there were two peaks in winter and fall reflecting the seasonal behavior of P. b. euptilura. Among the 14 candidate models, the best model with the lowest Akaike's Information Criterion (AIC) included five factors: traffic volume, distance from ramps, elevation, patch connectivity index, and distance from water. This study shows that P. b. euptilura road-kills are not randomly distributed, but rather are related to adjacent landscape, traffic patterns, and season. This study can contribute to minimizing future collisions and lead to conservation of the endangered populations through application to road-kill management policy. © 2019