Taking the planet’s pulse: AI to empower adaptation to global change
summary
Amazon hosts a cloud program that places petabytes of Landsat, Sentinel and other satellite data in AWS and terraPulse uses that stack plus GEDI lidar and machine learning to produce 30 m global tree canopy map now refined to annual 10 m monitoring. Models train on noisy, multi‑source truth data and output a probabilistic uncertainty per pixel so change detection is significance‑tested. Outputs include biomass and fire‑danger surfaces and a value‑at‑risk biomass × fire probability product that governments and companies use to enter carbon markets and assess supply‑chain and insurance risk.
AI-written summary · about
chapters (10)
- 0:00Introduction to ASDI
- 3:59ASDI program overview
- 9:49Joe Sexton introduction
- 10:33The Anthropocene challenges
- 16:46Landsat and landscape ecology
- 25:07Founding terraPulse
- 27:14Olympic Peninsula case study
- 37:35Carbon markets and fire risk
- 43:28Adaptation to climate change
- 47:04Q&A and closing
related talks
27mSmart Forestry: AI in the Service of Nature
41mAbby Seneor | AI Track | Compass AI & Data Summit 2025
32mMeasuring Vegetation Health to Predict Natural Hazards
36mWatching Millions of Trees with Python: For Impact and Profit
28mGeoAI: Evolution of AI-Based Tree/Vegetation Detection Models Using Imagery
09mGen AI Workflows for Climate Risk Modeling in Food Systems
11mTerraKit: Standardising AI-Ready Geospatial Data...
48mAccelerate science, research, and innovation with AWS