Forest Typology 2020 v1.0 Google DeepMind and Google Research, together with the World Resources Institute (WRI) and the International Institute for Applied Systems Analysis (IIASA), have released Forest Typology (ForTy) 2020, a 10m resolution global map that classifies each land area into six categories according to FAO and EUDR definitions: primary forests producing forests, natural forests, natural forests. Forests, tree crops and forestry agriculture and other lands - as of 2020. Independent validation on 8,190 stratified random plots marked by experts showed an overall accuracy of 90.2% for the six-class scheme, 94.8% for natural forest classification, and 95.5% for forest/non-forest classification. Forest typology data 2020 v1.0 in the Google Earth Engine repository, there are methodology reference datasets: Neumann, M., Raychuk, A., Potapov, P. et al. Global forest typology at 10m resolution for forest and land use monitoring. Sci Data 13, 1434 (2026). Beyond the Tree Cover: Mapping the World's Forest Types is a popular article on the DeepMind blog. New data identifies forest types around the world - an article in the World Nature Clock blog. Forest classification showing the dominant class in cells of 1 km2.

Forest Typology 2020 v1.0 Google DeepMind and Google Research, together with the World Resources Institute (WRI) and the International Institute for Applied Systems Analysis (IIASA), have released Forest Typology (ForTy) 2020, a 10m resolution global map that classifies each land area into six categories according to FAO and EUDR definitions: primary forests producing forests, natural forests, natural forests. Forests, tree crops and forestry agriculture and other lands - as of 2020. Independent validation on 8,190 stratified random plots marked by experts showed an overall accuracy of 90.2% for the six-class scheme, 94.8% for natural forest classification, and 95.5% for forest/non-forest classification. Forest typology data 2020 v1.0 in the Google Earth Engine repository, there are methodology reference datasets: Neumann, M., Raychuk, A., Potapov, P. et al. Global forest typology at 10m resolution for forest and land use monitoring. Sci Data 13, 1434 (2026). Beyond the Tree Cover: Mapping the World's Forest Types is a popular article on the DeepMind blog. New data identifies forest types around the world - an article in the World Nature Clock blog. Forest classification showing the dominant class in cells of 1 km2.