We utilized variable radius point count data collected in 2004 and 2006 at 255 points to generate presence / absent of bird species within a 200m radius of 255 points. The sampling points are part fo the Kenai Nationa Wildlife Refuge's Long-Term Ecological Montiroing Program (LTEMP). LTEMP points are arranged in a 4.8 km resolution, systematic grid spanning the 7722 km2 spatial extent of Alaska’s Kenai National Wildlife Refuge.We built distribution models for 40 bird species that are present within 200m of 2–56% of the sampling points resulting in models that represent species which are both rare and common on the landscape. All models were built using a common set of 152 environmental predictor variables representing topographical features, climatic space, vegetation, anthropogenic variables, and landscape structure. Random Forests produced strong models (ROC >0.8) for 16 bird species. We also assess models based on thier ability to predict "out-of-bag" data which is a protion of the training data that is withheld in model building.Alder Flycatcher Model InformationROC =0.813,Absent Correct = 66%Present Correct = 82%Map represents an index of probability of occurrence (0-1). Consider 0.75 to be very likely.
For a detailed description of the methodology, please refer to: Magness, D.R., F. Huettmann & J.M. Morton. 2008. Using Random Forests to provide predicted species distribution maps as a metric for ecological inventory & monitoring programs. Pages 209-229 in T.G. Smolinski, M.G. Milanova & A-E. Hassanien (eds.). Applications of Computational Intelligence in Biology: Current Trends and Open Problems. Studies in Computational Intelligence, Vol. 122, Springer-Verlag Berlin Heidelberg. 428pp. doi:10.1007/978-3-540-78534-7_9