Comparing surface albedo and land surface temperature of reflective roofs across five United States Cities

Abstract Rapid urban expansion and increasingly intense summer heat have heightened interest in reflective roofing as an urban heat-mitigation strategy. This study evaluates associations among roof type, surface albedo, and land surface temperature (LST) across five climatically distinct U.S. urban regions: Chicago, New York City (NYC), Las Vegas Valley (LVV), Los Angeles (LA), and Houston. Cloud-free Landsat 8 observations acquired during May–August from 2015 to 2024 were used to derive surface albedo from Operational Land Imager data and LST from thermal observations. A standardized sampling framework included 100 white-roof and 100 lower-reflectivity brown-roof regions of interest (ROIs) per city. Each 150 × 120 m ROI represented a rooftop-dominated urban area rather than an individual building. Spatial mapping, analysis of variance, and roof-specific linear regression were used to evaluate city-level patterns and differences between roof classes. White roofs exhibited significantly higher mean albedo than brown roofs in all five cities ( p < 0.001). However, this consistent reflectivity contrast did not produce a uniform LST response. White roofs had significantly lower mean LST than brown roofs in Chicago ( p = 0.002), NYC ( p < 0.001), and LVV ( p = 0.016), with NYC exhibiting the largest mean temperature difference. In LA, white roofs were slightly warmer than brown roofs, with a small marginally significant difference ( p = 0.049), while Houston showed no significant difference between roof classes ( p = 0.136). The spatial analysis identified the most coherent increase in surface albedo in Chicago, whereas Houston and NYC exhibited more limited or fragmented temporal changes. Within-class regressions did not demonstrate a universal inverse relationship between albedo and LST. Positive slopes occurred for both roof classes in Chicago, NYC, LA, and Houston, while the LVV relationships were nearly flat. Differences between brown- and white-roof slopes were significant in NYC, LA, and Houston but not in Chicago or LVV. These relationships are interpreted as city- and sample-specific associations rather than causal effects because roof materials, emissivity, shading, vegetation, atmospheric conditions, urban morphology, and mixed-pixel contributions were not controlled independently. The findings demonstrate that reflective roofs can be associated with lower surface temperatures, but their apparent performance varies with climatic and built-environment context. Reflective roofing should therefore be considered one component of locally adapted urban heat-mitigation strategies rather than a uniform solution.