UCLA Geography Leads New Land Use and Land Cover Mapping Effort in Brazil

/ by

Led by Khalil Ganem, with Dr. Yongkang Xue and researchers from Brazil’s National Institute for Space Research and Associação Plantas do Nordeste, the study developed the first land use and land cover (LULC) classification framework tailored to Northeast Brazil (NEB) – a region nearly four times the size of California, spanning tropical forests and over 90% of Brazil’s drylands. Using remote sensing and machine learning, the paper generated annual LULC maps from 2000 to 2020. Its two-tier classification scheme distinguishes NEB shrublands nationally, aligns with global datasets (7 Level-1 classes), and preserves ecoregional characteristics (16 Level-2 classes).

Published in GIScience & Remote Sensing (link here), the work offers a scalable model for mapping dryland–forest mosaics worldwide. The open-access dataset is available here.