832ddf6a-6708-4b7a-aa55-7ca142f9b413 Global Land Cover and Land Use Change University of Maryland Peter Potapov Research Professor potapov@umd.edu https://orcid.org/0000-0003-3977-0021 University of Maryland Matthew C. Hansen Professor mhansen@umd.edu https://orcid.org/0000-0003-0042-2767 University of Maryland Amy Pickens Assistant Research Professor ahudson2@umd.edu University of Maryland Andres Hernandez-Serna Principal Faculty Specialist andreshs@umd.edu University of Maryland Alexandra Tyukavina Associate Research Professor atyukav@umd.edu https://orcid.org/0000-0003-3872-9844 University of Maryland Svetlana Turubanova Assistant Research Scientist sveta@umd.edu https://orcid.org/0000-0001-6386-156X University of Maryland Viviana Zalles Postdoctoral Researcher viviana.zalles@wri.org University of Maryland Xinyuan Li Doctorate Student lxy95@terpmail.umd.edu University of Maryland Ahmad Khan Assistant Research Professor World Resources Institute Fred Stolle Deputy Director fstolle@wri.org https://orcid.org/0000-0002-3961-8591 World Resources Institute Nancy Harris Research Manager nharris@wri.org https://orcid.org/0000-0001-6661-4249 Texas Tech University Xiao-Peng Song https://orcid.org/0000-0002-5514-0321 University of Maryland Antoine Baggett Researcher University of Maryland Indrani Kommareddy Researcher University of Maryland Anil Kommareddy IT Systems Analyst anilk@umd.edu SIB Swiss Institute of Bioinformatics Chiara Bortoluzzi Data Manager chiara.bortoluzzi@sib.swiss https://orcid.org/0000-0001-6589-6635 2022-04-13 English The GLAD Global Land Cover and Land Use Change dataset quantifies changes in forest extent and height, cropland, built-up lands, surface water, and perennial snow and ice extent from the year 2000 to 2020 at 30-m spatial resolution. The global dataset derived from the GLAD Landsat Analysis Ready Data. Each thematic product was independently derived using state-of-the-art, locally and regionally calibrated machine learning tools. Each thematic layer was validated independently using a statistical sampling. The dataset includes information on: <br><div>Annual maps of land cover and land use (2000, 2005, 2010, 2015, 2020) <br></div><div>Global map with continuous measures of bare ground and tree height inside and outside of wetlands, seasonal water percent, and binary labels of built-up, permanent snow/ice, and cropland. Net change of land cover and land use between 2000 and 2020 (2000-2020change)</div><div>Land cover and land use states of 2020 with transitions relative to 2020 labelled.</div> Land cover Land use Global Terrestrial None This work is licensed under a Creative Commons Attribution 4.0 International License Creative Commons CC0 https://creativecommons.org/licenses/by/4.0/ Annual maps of land cover and land use (2000) https://storage.googleapis.com/earthenginepartners-hansen/GLCLU2000-2020/v2/2000.txt Annual maps of land cover and land use (2005) https://storage.googleapis.com/earthenginepartners-hansen/GLCLU2000-2020/v2/2005.txt Annual maps of land cover and land use (2010) https://storage.googleapis.com/earthenginepartners-hansen/GLCLU2000-2020/v2/2010.txt Annual maps of land cover and land use (2015) https://storage.googleapis.com/earthenginepartners-hansen/GLCLU2000-2020/v2/2015.txt Annual maps of land cover and land use (2020) https://storage.googleapis.com/earthenginepartners-hansen/GLCLU2000-2020/v2/2020.txt Net change of land cover and land use between 2000 and 2020 https://storage.googleapis.com/earthenginepartners-hansen/GLCLU2000-2020/v2/2000-2020change.txt Global Land Cover and Land Use 2000-2020 https://storage.googleapis.com/earthenginepartners-hansen/GLCLU2000-2020/v2/download.html Global -180 180 90 -90 2000-01-01 2020-12-31 University of Maryland Peter Potapov Research Professor potapov@umd.edu https://orcid.org/0000-0003-3977-0021 Most global thematic products, except open water, were derived using consistently processed Landsat Analysis Ready Data produced by the Global Land Analysis and Discovery laboratory (GLAD) at the University of Maryland. The annual GLAD data time series were integrated into a set of phenology metrics that enabled global model calibration and application. They used a separate supervised classification model to map each thematic class. Individual decision tree models calibrated with manually collected training data were implemented for cropland and perennial snow and ice mapping. Forest height was estimated using a regression tree model calibrated with Global Ecosystem Dynamics Investigation Lidar (GEDI) forest structure measurements. Built-up lands were mapped using a deep learning convolution neural network (CNN) algorithm trained with Open Street Map (OSM) data. The models were calibrated locally (for forest height and cropland mapping) or regionally (other products). Surface water mapping utilised pre-scene Landsat data classification and time series analysis in Google Earth Engine. We independently validated each global thematic product using statistical sample analysis. The sample reference data were collected through visual interpretation of the best available high-resolution satellite images and Landsat time series. Files are named as follows: [coordinate]_[coordinate].tif Potapov, P., Hansen, M. C., Pickens, A., Hernandez-Serna, A., Tyukavina, A., Turubanova, S., ... & Kommareddy, A. (2022). The global 2000-2020 land cover and land use change dataset derived from the Landsat archive: first results. Frontiers in Remote Sensing, 3, 856903. https://doi.org/10.3389/frsen.2022.856903. Recent advances in Landsat archive data processing and characterization enhanced our capacity to map land cover and land use globally with higher precision, temporal frequency, and thematic detail. Here, we present the first results from a project aimed at annual multidecadal land monitoring providing critical information for tracking global progress towards sustainable development. The global 30-m spatial resolution dataset quantifies changes in forest extent and height, cropland, built-up lands, surface water, and perennial snow and ice extent from the year 2000 to 2020. Landsat Analysis Ready Data served as an input for land cover and use mapping. Each thematic product was independently derived using locally and regionally calibrated machine learning tools. Thematic maps validation using a statistical sample of reference data confirmed their high accuracy (user’s and producer’s accuracies above 85% for all land cover and land use themes, except for built-up lands). Our results revealed dramatic changes in global land cover and land use over the past 20 years. The bitemporal dataset is publicly available and serves as a first input for the global land monitoring system. 2000 GeoTIFF 2005 GeoTIFF 2010 GeoTIFF 2015 GeoTIFF 2020 GeoTIFF 2000-2020change GeoTIFF