data api · Global

WRI/Google DeepMind Global Drivers of Forest Loss (1km)

This dataset maps the dominant driver of tree cover loss globally from 2001 to 2025 globally. Produced by the World Resources Institute (WRI) and Google Deepmind. Using satellite imagery and a neural network to classify seven categories of drivers. Source & citation: Zenodo record 14163025; also on the WRI Data Explorer.

Visit source ↗

Claim this listing

Represent WRI/Google DeepMind Global Drivers of Forest Loss (1km)? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.

Explore paid sponsorship →

What's in it

Data offered docs-verified: dominant driver of tree cover loss (classification, 7 classes) · per-class pseudo-probabilities (probability_1 to probability_7)
Coverage docs-verified: Global, longitude -180 to 180, latitude -56 to 84 · 2001 to 2025
Resolution docs-verified: 0.01° (~1 km) grid; one dominant driver per cell for the whole 2001–2025 period
Formats docs-verified: GeoTIFF (COG) · Icechunk virtual store (xarray) · STAC 1.1 catalog
Access docs-verified: Public, anonymous access via HTTPS/S3 (data.source.coop); CC-BY 4.0 licence, attribution required

Useful for

Use cases llm-inferred: Attributing historical tree-cover loss drivers (e.g. logging, agriculture, wildfire) for a site-scale ecological or land-use baseline · Screening upstream catchments for past forest-loss drivers as context for sediment or water-quality risk assessments · Environmental due diligence for overseas project sites where UK-style land-use records are unavailable
Audience llm-inferred: ecologists · GIS analysts · hydrologists · planners · environmental consultants

Limitations

Constraints docs-verified: Does not separate natural forest loss from the clearing of planted trees · Only one dominant driver per 1 km cell for the whole period; co-occurring drivers and their sequence are not recorded · Loss detection inherits the accuracy of Global Forest Change v1.13 · Training and validation data from the original study are not included in the mirror · Icechunk access may require downgrading zarr to 3.2.1 due to virtual_tiff incompatibility

Licence: see product README
Authentication: Public; anonymous HTTPS/S3 access via data.source.coop
GitHub stars: Unknown
Last commit: Unknown
Last successful page check: 14/10/2026
Page health: page reachable
Package: Not recorded
Verification: Not independently verified

Provenance & evidence

Verified via Source Cooperative API /api/v1/products/epoch (public product, updated 2026-09-11).

Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (source.coop)