Reevaluating how tree canopy cools the air and surfaces in cities at very high resolution

Abstract Tree canopy mitigates urban heat, yet the magnitude and spatial distribution of the cooling effect, particularly for air temperature, remain poorly understood. While land surface temperature (LST) is frequently used in urban heat studies, it is neither a good proxy for metrics of human thermal comfort nor does it adequately predict air temperature. For both heat metrics, we lack clear guidance on where in a city tree planting or canopy preservation is most efficient, limiting the effectiveness of greening strategies. In this study, we modeled the influence of tree canopy and other land cover variables (e.g., impervious surfaces, water) on both LST and mobile air temperature measurements across 11 northeastern U.S. cities. We found that tree cooling efficiency (TCE) for air temperature was highest in areas with substantial existing canopy, whereas TCE for LST was uniform across urban contexts—challenging recent reports that TCE is maximized in low-canopy areas. We also demonstrate that simple (canopy only) models and moderate-resolution satellite canopy data can lead to overstated tree cooling effectiveness in dense urban settings. Instead, we found that impervious surfaces and water bodies exerted temperature influences equal to or greater than vegetation. These findings highlight the importance of high-quality data and multipronged urban heat mitigation strategies, integrating clustered tree planting, impervious surface removal, and other complementary measures.