In support of the Designing Sustainable Landscapes project, avian habitat models were developed for 40 species representing 12 distinct habitat types within the South Atlantic Migratory Bird Initiative region (see Table 1 ). These models were used to assess the current and future capability of habitats to support sustainable bird populations in the face of complex landscape changes including urban growth and the effects of climate change on sea level rise and plant community succession. The basis of the habitat models was derived from the Southeast Gap Analysis Project which utilized a deductive approach to develop boolean (presence/absence) predictive habitat maps within a species known range (see Appendix A and Appendix B for SE-GAP [...]
Summary
In support of the Designing Sustainable Landscapes project, avian habitat models were developed for 40 species representing 12 distinct habitat types within the South Atlantic Migratory Bird Initiative region (see Table 1 ). These models were used to assess the current and future capability of habitats to support sustainable bird populations in the face of complex landscape changes including urban growth and the effects of climate change on sea level rise and plant community succession.
The basis of the habitat models was derived from the Southeast Gap Analysis Project which utilized a deductive approach to develop boolean (presence/absence) predictive habitat maps within a species known range (see Appendix A and Appendix B for SE-GAP methodology, see Table 2 for species model reports). Models were developed using both ESRI GRID and ERDAS Imagine software. Batch scripting was employed in both software programs to achieve standardization and efficiency. The original SE-GAP habitat models were further refined by the inclusion of vegetative structure (i.e. open, closed) and stand age parameters (i.e. early, mid, late) into both the land cover classification and the habitat association database (see Appendix C and Appendix D ).
Three scenarios of climate change (A2, A1B, and B1) were incorporated into landscape succession modeling and subsequently used as the land cover data input for avian habitat models. All other model input data layers were consistent for each species habitat model. Each scenario has 11 decadal time steps from 2000-2100 resulting in 33 model outputs for each species (1,386 total data layers).
This raster map represents the presence/absence of predicted habitat of a single avian species. Since it is one of 40 species whose predicted habitats were modeled, the processes used to create this raster map may differ from other species in this dataset. Processes were unique because while a majority of models incorporated land use/land cover data as the primary input, several other environmental features were incorporated for specific species depending on life history and habitat requirements noted both in published literature and expert review with biologists.