Information-theoretic model selection affects home-range estimation and habitat preference inference: a case study of male Reeves’s Pheasants Syrmaticus reevesii: Home-range estimation of Reeves’s Pheasants

摘要: Reeves’s Pheasant Syrmaticus reevesii is a vulnerable forest bird inhabiting broadleaved habitats dominated by oaks Quercus spp. in central China. Identifying home-ranges and habitat associations important for understanding the biology of this species developing effective management conservation plans. We used information-theoretic criteria to evaluate relative performance four parametric (exponential power, one-mode bivariate normal, two-mode normal circle) two non-parametric models (adaptive fixed kernel) estimating Pheasants. For models, Akaike’s information criterion (AICc) likelihood cross-validation (CVC) were relatively consistent ranking exponential power model least acceptable, whereas performed better. The CVC suggested that kernel particularly adaptive kernel, best among all six evaluated. average core area 95% contour based on with greatest support 6.1 54.9 ha, respectively, larger than those estimated from other models. discrepancy estimates between highest lowest decreased as size increased; however, home-range shapes differed Minimum convex polygons removed 5% extreme data points (MCP95) roughly half Estimates evaluation not affected sample (> 50 observations each bird). Inference about preference composition analysis overlap varied That strongest Pheasants selected mature fir mixed forest, avoided farmland, had mean among-individual overlaps 20%. recommend methods, method, areas complex multi-polar preferences heterogeneous environments large patches. However, we caution against traditional convenience using single estimate exploration multiple describing ecological processes underlying space use associations.