Multi-season satellite imagery (Landsat ETM+) from 1999-2001 were used in conjunction with digital elevation model (DEM) derived datasets (e.g. elevation, landform) to model natural and semi-natural vegetation. General land cover classes (i.e. water, urban, row crop) were included from the National Land Cover Dataset 2001 (). The minimum mapping unit for this dataset is 0.4 ha (1 acre). Vegetation classes were drawn from NatureServe's Ecological System Classification (Comer et al. 2003; ). Two hundred fourteen (214) land cover classes (140 Ecological Systems, 32 modified classes, and 42 general land cover classes recognized as occurring in the 13 USGS mapping zones comprising the Southeast region were mapped. Land cover classes were mapped with a variety of techniques including decision tree classifiers, unsupervised classification, and expert decision rules. The 13 USGS mapping zones were mapped independently of one another by one of three cooperating spatial analysis laboratories:
1. Biodiversity and Spatial Information Center, North Carolina State University, Raleigh, NC 27695-7617.
2. Alabama Gap Project, Alabama Cooperative Fish & Wildlife Research Unit, Auburn University, Alabama 36839.
3. Natural Resources and Spatial Analysis Laboratory (NARSAL), University of Georgia, Athens Georgia 30602.
An internal validation for modeled classes is in progress. Results of the validation will be presented in the project final report.