Google DeepMind releases ForTy, a 10m global forest-type map with ~95% natural-forest accuracy
anselm · x · 2026-10-31
Google DeepMind and Google Research, with WRI and IIASA, released ForTy (Forest Typology) 2020, a global map classifying every 10x10m patch of land into six forest/land-use types, built on satellite embeddings and deep learning.
- Natural-forest classification accuracy is around 95%
- Serves as a baseline for the EU deforestation regulation
- An accompanying time series tracks yearly changes from 2017 to 2025, showing not just where forest was lost but what replaced it: timber plantations, tree-crop farms, or other land
- The team argues tree cover alone misses ecological value: primary forests store far more carbon and biodiversity than single-species plantations
More from Research
- CMU paper unifies pretraining, SFT and RL into one likelihood-based training algorithm — rohanpaul_ai · 2026-10-31
- 48-page deep dive: why modern LLM RL is secretly off-policy and 14 ways to fix it — cwolferesearch · 2026-10-31
- Google Health: AI Tools Aim to Support Psychiatrists Where Conversation Is Diagnosis — SRSchmidgall · 2026-10-31
- Benchmarking 'scoping': five ways to confine LLMs to a bounded operational domain — burkov · 2026-10-31
- Meta paper: looping transformer weights adds depth without more parameters, up to 32 loops — iScienceLuvr · 2026-10-31
- Simple neg-log-odds loss swap improves every training attribution method tested — davidbau · 2026-10-31