Google Research Expands Heat Resilience Data to Over 50 Cities Worldwide

Google Research has released an expanded dataset on roof reflectivity (albedo) at the building level for more than 50 cities worldwide to help urban planners implement "cool roof" solutions and mitigate the effects of extreme heat. About 500,000 deaths annually are linked to extreme heat, worsened by the urban heat island effect, which causes temperatures in megacities to rise twice as fast as the global average. Dark roofing and pavement materials, as well as a lack of vegetation, exacerbate local warming. "Cool roofs" significantly reduce solar energy absorption and local temperatures by increasing albedo. In 2024, Google Research piloted this approach with 14 cities, providing them with roof albedo data, leading to the adoption of regulations and adaptation plans. Now the data covers more than 50 cities in 9 countries, including London, Athens, Barcelona, Rio de Janeiro, São Paulo, Los Angeles, Austin, and New York. The methodology is described in a paper in Nature Communications: researchers combined data from the Sentinel-2 satellite (10 meters) with commercial Airbus Pléiades Neo imagery (30 centimeters) using machine learning. This produced albedo maps with 30-centimeter resolution, enabling analysis of individual large buildings. The data was verified using hyperspectral measurements from an aircraft over Boulder, Colorado, with a root mean square error of just 0.04. Modeling showed that targeted implementation of "cool roofs" could reduce extreme urban heat by up to 0.5°C globally. The new Heat Resilience Earth Engine App provides open access to this data.