Tropical Tree Cover 10 m

WRI — version v20220922 (January 2024)

Description

This dataset classifies tree extent for the year 2020, with coverage over the tropics at 10 m spatial resolution.

Dataset ID
72cfdb7e
Category
Land use & land cover
Type
Raster
CRS
EPSG:4326
Spatial coverage
Tropics
Spatial resolution
10 m
Temporal coverage
2020
Temporal resolution
Single time point

Usage notes

Spatial coverage is non-desert land surfaces between 23.44ºN and 23.44ºS. Inconsistent imagery usage and normalisation between adjacent tiles may cause inaccurate predictions or visual artefacts. Data availability is limited in regions with high cloud cover. Model versions may vary between regions due to selective back-processing of affected areas.

Variables

tree_extent_probability

Typeuint16Units%

Description

The probability of one or more tree canopies intersecting a pixel centroid. Data is generated using a convolutional neural network trained to predict tree extent in 18,000 plots from Sentinel-1 and Sentinel-2 satellite data.

Usage notes

Values are binned probabilities rounded to the nearest 10% (0 , 10 , 20 , 30 , 40 , 50 , 60 , 70 , 80 , 90 , 100). A tree is defined as any woody vegetation with > 5 m canopy height or 3-5 m canopy height plus > 5 m crown diameter. Includes tree-based plantations (e.g. eucalyptus, avocado) but excludes herbaceous vegetation (e.g. sugarcane, bananas, cacti) and short woody crops (e.g. tea, coffee). Recommended cut-off for converting probabilities to binary tree/no-tree classifications is 35%. Overall accuracy is 94% ± 0.1% for binary classifications. Accuracy for non-forested areas is between 70% and 95%, with lowest accuracy in urban and arid areas.

Pricing

This dataset has no data acquisition cost.

Resources

Provider

WRI

www.wri.org

The World Resources Institute (WRI) is a global research organisation that provides authoritative data, research, and tools to address critical environmental and development challenges. Through initiatives like Global Forest Watch and Aqueduct, WRI combines research with innovative technology to make environmental data accessible and actionable for governments, businesses, and civil society worldwide.

GFW

www.globalforestwatch.org

Global Forest Watch (GFW) is an online platform that provides data and tools for monitoring forests. Established by the World Resources Institute (WRI), GFW harnesses cutting-edge geospatial technology to allow anybody to access near realtime information about where and how forests are changing around the world.

UMD

glad.umd.edu

The Global Land Analysis and Discovery (GLAD) laboratory at the University of Maryland (UMD) studies and develops tools for assessing global land surface change using satellite imagery. GLAD also collaborates on major initiatives including Global Forest Watch and the WRI Land and Carbon Lab.