Global Forest Management Type Map at 10m Resolution for 2020 Part 1

Zenodo dataset for Global Forest Management Type Map at 10m Resolution for 2020 Part 1

Source
Zenodo
Published
2026
Region
Global
Format
tif
Access
Open access
License
cc-by-4.0
Themes
Geospatial & GIS Data, Sustainable Cities, Plants & Vegetation, Biodiversity & Ecosystems, Forests

Get the data from Zenodo

About this dataset

This is an experimantal dataset that represents the update d version of the global forest management layer f or 2015 developed by Lesiv et al. ( 202 2 ). It is based on a new training dataset for the year 2020 that was collected and expanded with two additional classes: rubber plantations and tree crops. The VITO team trained a hybrid classification framework combining a deep learning and pixel-based classifi cation approaches using Sentinel-1 and Sentinel-2 imagery f or the year 2020. The resulting product provides a global wall-to-wall map of forest management types at 10 m spatial resolution. The map is spit in two parts and uploaded as 2 seperate records due to the zenodo size limitations. Part 1 includes Americas. Tree cover mask comes from the LCFM annual land cover map 2020. The map distinguishes the following c lasses: 0 – No tree cover 11 - Naturally regenerating forests without any signs of management, including primary forests; 20 – Naturally regenerated forests where there are clearly visible indications of human activities, such as selective logging, shifting cultivation, etc.; 31 – Planted forest 32 – Plantation f orest 33 – Rubber plantation 40 – Oil palm plantations 50 – Tree crops ( mono cultu re plantations) 53 - Agroforestry 100- Other trees(e.g. trees in urban areas) D etailed class d efinitions are provided in a separate file . The modelling framework follows the methodology developed for global land cover mapping and described in the Copernicus Global Land Cover documentation . The work was funded by WRI.

How the data was collected

This is an experimantal dataset that represents the update d version of the global forest management layer f or 2015 developed by Lesiv et al. ( 202 2 ). It is based on a new training dataset for the year 2020 that was collected and expanded with two additional classes: rubber plantations and tree crops. The VITO team trained a hybrid classification framework combining a deep learning and pixel-based classifi cation approaches using Sentinel-1 and Sentinel-2 imagery f or the year 2020. The resulting product provides a global wall-to-wall map of forest management types at 10 m spatial resolution. The map is spit in two parts and uploaded as 2 seperate records due to the zenodo size limitations. Part 1 includes Americas. Tree cover mask comes from the LCFM annual land cover map 2020. The map distinguishes the following c lasses: 0 – No tree cover 11 - Naturally regenerating forests without any signs of management, including primary forests; 20 – Naturally regenerated forests where there are clearly visible indications of human activities, such as selective logging, shifting cultivation, etc.; 31 – Planted forest 32 – Plantation f orest 33 – Rubber plantation 40 – Oil palm plantations 50 – Tree crops ( mono cultu re plantations) 53 - Agroforestry 100- Other trees(e.g. trees in urban areas) D etailed class d efinitions are provided in a separate file . The modelling framework follows the methodology developed for global land cover…

Browse sustainability datasets on SusanHub