Winter habitat and travel route selection by wolves in the northern Apennines, Italy

摘要: To assess the wolf-habitat relationship on home range scale (i.e., third order selection), we developed a resource selection probability function (RSPF) through multiple logistic regression model based winter travel routes of wolf pack in northern Apennines, Italy (1991-95). Both (240 km) and habitat variables were mapped at 1:10000 scale, digitised as Geographical Information System (GIS) layers, overlaid with 100 x m pixel grid to census all used unused units. Out 15 covariates, full included 10 4 interaction terms. According model, by wolves not randomly located within but clearly associated selected bio-physical factors, including human-related modifications roads), which appeared affect wolves' and, ultimately, quality. Using jackknife procedure, correctly classified 73.1% units 63.2% A Monte Carlo test showed non-significant effect coefficients increasing sub-sampling levels units, suggesting that autocorrelation snow-tracking data exerted little influence point estimates coefficients. However, given increased standard error higher levels, might have caused an underestimation theoretical variance. Although are generally considered generalists, this study shows patterns disclosed liner scales analysis. In perspective, functions offer different complementary insight respect regional landscape applications, provide useful management tool for assessment quality local scale.