High-Resolution Vegetation Mapping in the Sonoran and Mojave Deserts using Random Forest Classification of Multi-Temporal Landsat 8 Data and Phenology Metrics

Abstract

Existing geospatial data and tools are at times inadequate for supporting the critical habitat-related work in the Sonoran and Mojave Deserts Bird Conservation Region (BCR 33) due to the area’s complex vegetation communities and the discontinuity in data availability across the United States (US) and Mexico (MX) border. Vegetation distributions and dynamics are critical to guiding land management and conservation decisions throughout BCR 33. This research aimed to produce the first high-resolution, consistent transboundary vegetation community map within BCR 33 by prototyping new methods for desert vegetation classification using the Random Forest (RF) machine learning (ML) method. The developed RF classification model utilizes multitemporal Landsat 8 Operational Land Imager spectral and vegetation index data from the period 2013-2020, augmented with phenology metrics tailored to capture the unique growing seasons of desert vegetation. Our RF model achieved an overall classification f-score of 0.80 and overall accuracy of 91.68%. Our results portrayed vegetation cover at a much finer resolution than existing land cover maps from the US and MX portions of the study area, allowing for the separation and identification of smaller habitat pockets, including riparian communities, which are critically important for desert wildlife and are often misclassified or nonexistent in current maps. This early prototyping effort serves as a proof of concept for the ML and data fusion methods that will be used to generate the final high-resolution land cover map for the entire BCR 33 region.