Google DeepMind and World Resources Institute map global forest types with ForTy

What happened

Google DeepMind and the World Resources Institute developed ForTy, a global map classifying land in 2020 into 6 categories at 10-meter resolution using AlphaEarth Foundations. Expert evaluation found 90.2 percent overall accuracy, but substantially weaker identification of plantations, planted forests, and tree crops complicates its use for EU deforestation compliance. The map, classification layer, and test data are freely available under CC BY 4.0.

Read next

• More changes in today’s briefingConfigure the ChatGPT desktop app with Amazon Bedrock in AWS GovCloud (US)

Reported by 2 sources

The story so far

Google DeepMind, Google Research, the World Resources Institute and the International Institute for Applied Systems Analysis released Forest Typology (ForTy) 2020, a global forest and land-use map, alongside an early time series tracking changes from 2017 to 2025; both datasets are freely available on Google Earth Engine under an open license. The 2020 map classifies land into six categories at 10-meter resolution. Google DeepMind reported that validation across more than 9,000 sites found 90% accuracy for forest types and 95% for distinguishing forest from non-forest land. The Decoder reported 90.2% overall accuracy in expert evaluation, but said identification of plantations, planted forests and tree crops was substantially weaker, complicating the map’s use for EU deforestation compliance. The time series tracks changes including natural forests becoming plantations or farms. Google DeepMind said it and its partners would continue improving that dataset.