Predicting potential distribution of Quercus suber in Italy based on ecological niche models: Conservation insights and reforestation involvements

Different statistical techniques have been used to model species potential distribution related to environmental variables. This paper provides a comprehensive assessments of GARP and MaxEnt methods, and investigates for the first time the probability of occurrence of cork oak (Quercus suber L.) in Italy based on ecological niche modelling approaches. A detailed distribution of the species was achieved during a 3-year National Project (SuberItalia) and 17 environmental layers were employed to obtain the potential distribution of cork oak. The performance of the models were measured using the receiver operating characteristic (ROC) approach and Cohen's Kappa statistic. Results achieved by GARP and MaxEnt showed as the drought and the cold stresses are the main factors affecting cork oak occurrence in Italy. Moreover, the accuracy of the obtained prediction maps were compared to a specifically calibrated geo-statistical method at regional scale, pointing out a preliminary geographical assessment of the suitable surfaces to set apart for cork oak forest expansion in Italy, thereby useful to address reforestation and conservation concerns to face the ongoing area reduction of these forests.