The US Department of Interior’s (USDI) Bureau of Land Management (BLM) has a long history of soil and vegetation monitoring of public rangelands it manages. However, historical monitoring data have been stored and managed at the field, district, or state level, making them difficult to compile and analyze. BLM’s Soil Vegetation Inventory Method (hereafter SVIM) program occurred between 1977 and 1983 in Arizona, California, Colorado, Idaho, Montana, New Mexico, Nevada, Oregon, Utah, and Wyoming. Our objective was to extract and decode vegetation cover data from the SVIM dataset in a georeferenced, digital format. Vegetation cover data are available for 22,578 SVIM site write-up areas in nine states. The wide geographic coverage of the SVIM dataset will make it an outstanding resource for researchers interested in quantifying vegetation change through time in rangelands of the western US. For several studies, the USGS created a crosswalk table to convert SVIM species codes into their modern counterparts for species within the Great Basin, USA. This crosswalk table also converted questionable or incorrect species codes (codes with mistyped codes or species that did not occur in the area) into their most likely species code based on the area in which they were sampled. This conversion was completed using a local expertise along with the USDA Plants Database (https://plants.sc.egov.usda.gov/java/).