Modelling the spatial distribution of snake species in northwestern Tunisia using maximum entropy (Maxent) and Geographic Information System (GIS)

We used GIS and maximum entropy to predict the potential distribution of six snake species belong to three families in Kroumiria (Northwestern Tunisia): Natricidae (Natrix maura and Natrix astreptophora), Colubridae (Hemorrhois hippocrepis, Coronella girondica and Macroprotodon mauritanicus), and Lamprophiidae (Malpolon insignitus). The suitable habitat for each species was modelled using the maximum entropy algorithm, combining presence field data (collected during 16 years: 2000-2015) with a set of seven environmental variables (mean annual precipitation, elevation, slope gradient, aspect, distance to watercourses, land surface temperature and normalized Differential Vegetation Index. The relative importance of these environmental variables was evaluated by jackknife tests and the predictive power of our models was assessed using the area under the receiver operating characteristic. The main explicative variables of the species distribution were distance from streams and elevation, with contributions ranging from 60 to 77 and from 10 to 25%, respectively. Our study provided the first habitat suitability models for snakes in Kroumiria and this information can be used by conservation biologists and land managers concerned with preserving snakes in Kroumiria.