AI

Google DeepMind Launches 10-Meter Global Forest Typology Map

Thursday, October 22, 2026Read Original

Details

  • Google DeepMind introduced Forest Typology, or ForTy, a global map distinguishing natural forests, planted forests, plantations, tree crops, agroforestry, and other land.
  • The map classifies land in 10-by-10-meter patches, helping separate ecologically important natural forests from intensively managed tree plantations.
  • ForTy uses AlphaEarth Foundations, Google’s geospatial AI model, to analyze years of satellite observations and detect planting patterns, growth cycles, and harvesting rhythms.
  • The system was developed with Google Research, the World Resources Institute, and the International Institute for Applied Systems Analysis.
  • The 2020 baseline map draws on analysis of approximately 1.7 million sites, while a time series tracks forest and land-use changes from 2017 through 2025.
  • The data is freely available through Google Earth Engine under an open license, allowing governments, companies, and researchers to examine forest change.
  • The classification aligns with categories used by the Food and Agriculture Organization and the European Union’s deforestation regulation, supporting supply-chain monitoring and compliance work.
  • Unlike ordinary tree-cover maps, ForTy indicates what type of forest occupies an area and how that classification has changed over time.

Impact

ForTy could improve deforestation-free supply-chain checks by giving companies and regulators a finer-grained way to distinguish natural forests from plantations and tree crops. Its alignment with EU deforestation-regulation categories makes the dataset particularly relevant as compliance requirements expand. The open Earth Engine release may also lower the cost of forest monitoring, although organizations will still need local validation and operational systems before treating the map as definitive evidence for enforcement or sourcing decisions.

Rift Dispatch