Developing geospatial estimates of potential cheatgrass (Bromus tectorum L.) and perennial herbaceous cover following fire and fire surrogate treatments has been challenging due to high variability in climate and site environmental characteristics. The advancement of geospatial climate models and analysis tools, such as BIOCLIM and Random Forest, may provide opportunities for applicability to this problem. We conducted a Random Forest analysis to spatially estimate cheatgrass and perennial herbaceous cover using 43 site environmental characteristics as explanatory variables. Site environmental characteristics were derived from 5 m digital elevation models (DEM), BIOCLIM, and ClimateWNA geospatial data sets. Cheatgrass and perennial herbaceous cover were previously collected at 45 wooded (Juniperus spp.) shrublands (Artemisia spp.) in Utah. We used Random Forest and ModelMap for R statistical software to conduct the analysis. Preliminary analysis indicates a hierarchy of influential site environmental characteristics that included elevation, mean precipitation and mean temperature of the wettest quarter of the year. These site environmental characteristics and a geospatial map of potential cheatgrass and perennial herbaceous cover may aid managers in determining which sites are more or less resilient to woodland reduction treatments or which sites may require the addition of plant materials for successful restoration.
Oral presentation and poster titles, abstracts, and authors from the Society for Range Management (SRM) Annual Meetings and Tradeshows, from 2013 forward.