Random forest classification data developed from multitemporal Landsat 8 spectral data and phenology metrics for a subregion in Sonoran and Mojave Deserts, April 2013 – December 2020

These data were compiled for the creation of a continuous, transboundary land cover map of Bird Conservation Region 33, Sonoran and Mojave Deserts (BCR 33). Objective(s) of our study were to, 1) develop a machine learning (ML) algorithm trained to classify vegetation land cover using remote sensing spectral data and phenology metrics from 2013-2020, over a large subregion of the Sonoran and Mojave Deserts BCR, 2) Calibrate, validate, and refine the final ML-derived vegetation map using a collection of openly sourced remote sensing and ground-based ancillary data, images, and limited fieldwork, and 3) Harmonize a new transboundary classification system by expanding existing land cover mapping resources from the United States portion of BCR 33 into Mexico. These data represent the final land cover maps produced by the developed random forest model, with additional ancillary labels for urban and agriculture areas. These data were created within a subregion of the Sonoran and Mojave Deserts BCR which spans from Phoenix, Arizona, US to Hermosillo, Sonora, Mexico for the time of April 2013 to December 2020. These data were created by the University of Arizona Vegetation Index and Phenology Lab who collected, processed, and analyzed all of the data and developed the random forest model used to produce the final mapping results. These data can be used to guide land management and conservation decisions within the Sonoran and Mojave Deserts BCR.