Majka, D. 2005. Modeling and Mapping Avian Distribution in a Montane Tropical Forest Using Topographic Variables Created From STRM Digital Elevation Model. In GIS-Based Modeling of Avian Distributions in a Montane Tropical Forest. MS Thesis. Purdue Univerisity.

In the second chapter of his MS theses, Majka integrated Generalized Linear Models (GLM) (multiple logistic regression) with Geographical Information System (GIS) to map the distribution of 41 species of Neotropical birds in Costa Rica. The variables used in the modeling process were obtained from a Shuttler Radar Topography Mission (STRM) digital elevation model. The variables were categorized in two groups: two primary gradient variables (elevational gradient) and seven secondary gradient (involving topographic features i.e. soil moisture, streams, and topographic position) variables. The bird data was obtained from field-based work. The author evaluated the models using receiver operating characteristic (ROC) curves. Majka found that for 28 species the variable distance to continental divide was the most important. He also noted that secondary gradients improved the performance of the distribution models using only a primary gradient. Because the lack of number of previous distribution studies that use continental divide as a variable, the author suggest that more research is necessary, to better understand its influence to the distribution of organisms.

            

More about Daniel Majka