dataset · Global
10m Annual Land Use Land Cover (9-class)
This dataset, produced by Impact Observatory, Microsoft, and Esri, displays a global map of land use and land cover (LULC) derived from ESA Sentinel-2 imagery at 10 meter resolution for the years 2017 - 2024. Each map is a composite of LULC predictions for 9 classes throughout the year in order to generate a representative snapshot of each year. This dataset was generated by Impact Observatory, which used billions of human-labeled pixels (curated by the National Geographic Society) to train a de
Visit source ↗Source details
Licence: [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)
Evidence quality: 41.2
Listed in aws-odr public catalogue API on 2026-10-16; relevance/value screened (haiku).
Metadata checked: 2026-10-14T13:20:00.000Z
What's in it
- Data offered docs-verified
- land use / land cover (LULC) classification, 9 classes · annual LULC composite maps derived from Sentinel-2
- Coverage docs-verified
- Global · 2017-2024
- Resolution docs-verified
- 10 m spatial; one annual composite per year
- Update frequency docs-verified
- Annual (new year released each January; 2024 data released June 2025)
- Formats docs-verified
- COG · STAC 1.0.0
- Access docs-verified
- Free and open; no AWS account or authentication required (anonymous S3 access). Licence CC BY 4.0. Also available via STAC 1.0.0 endpoint and IO Store for AOIs.
Useful for
- Use cases llm-inferred
- Land cover baseline for FRA runoff/impermeable-area context, 2017-2024 · Change screening for 'Bare' vs vegetated classes (Trees, Crops, Rangeland, Flooded Vegetation) on a site or catchment · Broad habitat context for BNG baseline mapping, to be validated locally
- Audience llm-inferred
- GIS analysts · hydrologists · flood risk consultants · ecologists · planners
Limitations
- Constraints llm-inferred
- Global product with assessed average accuracy of over 75%, so local validation is needed for project use · Only 9 generalised classes; not detailed habitat or ground-truth classification · Anomalous change between Bare and vegetative classes is reduced, not eliminated · Annual snapshots with a release lag (2024 data released June 2025)
Evidence
Check upstream coverage, uncertainty and licence before use. Station observations are not forecasts or design levels.