Published May 26, 2026 | Version v1

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

Description

This is an experimantal dataset that represents the updated version of the global forest management layer for 2015 developed by Lesiv et al. (2022). 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 classification approaches using Sentinel-1 and Sentinel-2 imagery for 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 classes: 

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 forest 

33 – Rubber plantation 

40 – Oil palm plantations 

50 – Tree crops (monoculture plantations) 

53 - Agroforestry 

100- Other trees(e.g. trees in urban areas)

 

Detailed class definitions 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.

Files

global_forest_management_2020_10m_v1_part1.tif

Files (17.1 GB)

Name Size
md5:b7192e03a098d060372ca734a66aec85
17.1 kB Download
md5:739d9d590b62fc21c55da0e516b0d94a
17.1 GB Preview Download