Modeling sage grouse: progressive computational methods for linking a complex set of local, digital biodiversity and habitat data towards global conservation statements and decision-making systems

摘要: Modern conservation management needs to link biological questions with computational approaches. As a global template, here we present such an approach from local study on sage grouse breeding habitat, leks, in North Natrona County, Wyoming, using remote sensing imagery, digital datasets, spatial statistics, predictive modelling and Geographic Information System (GIS). Four quantitative models that describe habitat selection were developed for multiple scales logistic regression multivariate adaptive splines (MARS-Salford Systems). Based candidate AIC, important predictor variables elevation, distance human development, slope, roads, NDVI water, but not Sagebrush. Some predictors changed when different MARS. For the year 2011, cumulative prediction index is presented how population viability of can be assessed over time space Markov chain deriving future landscape scenarios MARS species predictions.