Skip to main content
U.S. flag

An official website of the United States government

This site is currently in beta, and your feedback is helping shape its ongoing development.

Return to search results

Tree Cover Estimates at 30 m Resolution for Mexico, 2016-2018

Published by ORNL_DAAC | National Aeronautics and Space Administration | Metadata Last Checked: February 21, 2026 | Last Modified: 2026-02-17
The data set provides multi-year (2016-2018) percent tree cover (TC) estimates for entire Mexico at 30 m spatial resolution. The TC data (hereafter, NEX-TC) was derived from the 30 m Landsat Collection 1 product and a hierarchical deep learning approach (U-Net) developed in a previous CMS effort for the conterminous United States (CONUS) (Park et al., 2022). The hierarchical U-Net framework first developed a U-Net model for very high-resolution aerial images (NAIP) using training labels derived from previous work based on an interactive image segmentation tool and iterative updates with expert knowledge (Basu et al., 2015). The developed NAIP U-Net model and NAIP data produced 1-m NAIP TC across all lower 48 CONUS states. A Landsat U-Net model was developed for multi-year and large-scale TC mapping based on the very high-resolution NAIP TC made in the earlier stage. The Landsat U-Net model developed was adopted over the CONUS for testing its transferability, validation, and improvement across Mexico. This dataset provides national-scale percent tree cover estimates over Mexico and can be helpful for studies of carbon cycling, land cover and land use change, etc. The team has been working on improving temporal stability of the product and will update the product once the next version is ready to be shared.

data.gov

An official website of the GSA's Technology Transformation Services

Looking for U.S. government information and services?
Visit USA.gov