This paper details a method to create regional models of perennial vegetation cover using pre-existing field data and satellite imagery. Total cover of perennial vegetation is an important ecological attribute of desert ecosystems, including the Mojave Desert, USA, an area of 125,000 km2. Moderate-Resolution Imaging Spectroradiometer Enhanced Vegetation Index (MODIS-EVI) data were coupled with measurements of total perennial cover and plot elevation using stepwise linear regression and linear regression techniques to create two models of cover. The final models produced R2 of 0.82 and 0.81, respectively, and yielded maps of perennial cover distribution in the Mojave Desert at 250 m spatial resolution. Copyright ?? 2008 by Bellwether Publishing, Ltd. All rights reserved.