Google Research has expanded its building-level rooftop albedo dataset to cover more than 50 cities and launched a public Heat Resilience Earth Engine App to help municipalities plan cool-roof interventions. The methodology is described in the Nature Communications paper “Estimating high-resolution albedo for urban applications.” The project aims to give planners actionable, building-scale reflectivity data to reduce urban surface temperatures.
Why this matters
Extreme heat contributes to roughly 500,000 deaths annually worldwide, and the urban heat island effect causes metropolitan areas to warm faster than the global average. Dark pavements and roofs and lack of vegetation concentrate heat locally. Increasing rooftop reflectivity (cool roofs) is a cost-effective mitigation: higher albedo reduces absorbed solar energy, lowers surface temperatures, and helps protect vulnerable populations.
Method: fusing satellite sources for building-scale albedo
As part of the Google Earth AI collection, the team developed a method that fuses Sentinel‑2 satellite data with high-resolution (30‑cm) commercial imagery from Airbus Pléiades Neo. Sentinel‑2 provides radiometric accuracy and global coverage but only at about 10‑meter spatial resolution, which is insufficient to resolve individual roofs. By using machine learning and radiometric calibration to blend the datasets captured across different wavelengths, the method reconstructs a spectral reflectance profile for each urban pixel and produces building-level albedo estimates.
The approach was validated against airborne hyperspectral measurements over Boulder, Colorado. The fused 30‑cm albedo maps achieved a root mean square error (RMSE) of 0.04 relative to ground-truth data, demonstrating sufficient precision to prioritize individual large-footprint buildings for targeted cool-roof retrofits.
Google Research reports that targeted cool-roof planning informed by these data could reduce extreme urban heat by up to 0.5 °C (1.8 °F) globally.
Pilot work and policy impact
In 2024, Google piloted this approach with 14 cities, providing rooftop reflectivity data to identify vulnerable neighborhoods and locations where cool roofs would be most effective. Those data helped inform decisions in several cities, leading to initiatives such as cool-roof ordinances and adaptation plans.
Heat Resilience Earth Engine App: features and access
The Heat Resilience Earth Engine App makes the high-resolution rooftop albedo data available to decision-makers. Key features include:
- Building-level visualization: albedo shown as centroids to locate low-reflectivity surfaces.
- Baseline analysis: tools to understand current reflectivity and monitor changes over time.
- Data portability: ability to download high-resolution data for local analysis and policy development.
- Dynamic zoom: interface that transitions from census-tract aggregates to individual building insights.
The app is live and publicly accessible and includes the expanded coverage of 50+ cities across nine countries.
Coverage details
The expanded dataset includes major urban centers in Europe (including London, Athens, Barcelona), Brazil (including Rio de Janeiro and São Paulo), and the United States (including Los Angeles, Austin, and New York City). By providing open access to building-level albedo data, Google Research intends to help cities accelerate adoption of reflective surfaces to lower urban surface temperatures.
Acknowledgements
The research was developed by Google Research in collaboration with the World Resource Institute (WRI). Contributors named by the project are: Elizabeth J. Wesley (WRI), Salil Banerjee, Vishal Batchu, Aniruddh Chennapragada, Kevin Crossan, Bryce Cronkite-Ratcliff, Ellie Delich, Tristan Goulden (National Ecological Observatory Network), Mansi Kansal, Jonas Kemp, Eric Mackres (WRI), Yael Mayer, Rebecca Milman, John C. Platt, Shruthi Prabhakara, Gautam Prasad, Aaron Bell, Shravya Shetty, Charlotte Stanton, Wayne Sun, and Lucy R. Hutyra.
Where to find the data
The Heat Resilience Earth Engine App is live and available for public use. The interactive platform can be used to visualize rooftop albedo across all 50+ cities in the release, and the project’s technical documentation provides links to download the high-resolution datasets for further analysis and use.
