Expanding our Heat Resilience data to 50+ global cities

June 30, 2026 | Source: Google Research Blog

Tags: Google Research, climate AI, urban heat, cool roofs, satellite imagery, geospatial AI, Earth Engine

Google Research expanded its AI-driven heat resilience dataset from 14 to 50+ global cities, mapping building-level rooftop reflectivity using Sentinel-2 fused with 30cm Airbus imagery — published in Nature Communications and accessible via a new Earth Engine App.

Details

Google Research's Heat Resilience initiative has scaled from a 14-city pilot in 2024 to over 50 global cities, releasing building-level albedo (reflectivity) data to help urban planners identify where cool-roof interventions have the highest impact. The methodology, published in Nature Communications, fuses Sentinel-2 satellite data with Airbus Pléiades Neo 30cm imagery via AI to produce granular, building-level reflectivity maps — moving beyond neighborhood averages to actionable targeting. The stakes are significant: approximately 500,000 deaths annually are attributed to extreme heat, with cities warming at double the global average due to the urban heat island effect. Record-breaking heat waves across Western Europe earlier this month pushed temperatures past 40°C (104°F), underscoring the urgency. The 2024 pilot data already guided concrete policy responses, including cool-roof ordinances and adaptation plans across several cities. The expanded dataset is now freely accessible through a high-resolution Heat Resilience Earth Engine App, making the tool usable by city planners without AI expertise. This is part of Google's Earth AI geospatial model collection, which applies AI to satellite imagery for planetary-scale intelligence.