Predicting multi-species bark beetle (Coleoptera: Curculionidae: Scolytinae) occurrence in Alaska: First use of open access big data mining and open source GIS to provide robust inference and a role model for progress in forest conservation

Native bark beetles (Coleoptera: Curculionidae: Scolytinae) are a multi-species complex that rank among the key disturbances of coniferous forests western North America. Many landscape-level variables known to influence beetle outbreaks, such as suitable climatic conditions, spatial arrangement incipient populations, topography, abundance mature host trees, and disturbance history include former outbreaks fire. We assembled first open access data, which can be used in source GIS platforms, for understanding ecology organism Alaska. boosted classification regression tree machine learning data mining algorithm model-predict relationship between 14 environmental variables, model predictors, 838 occurrence records 68 species compared pseudo-absence locations across state The predictors topography- climate-related well feature proximities anthropogenic factors. were able model, predict, map occurrences Alaska on 1-km resolution addition providing good quality dataset freely accessible public. About 16% mixed forest 59% evergreen expected occupied by based current conditions biophysical attributes landscape. we prepared, modeling approach used, provide foundation future research not only scolytines but other questions concern, defoliators, small big game wildlife worldwide.