Large discrepancies between air temperature and land surface temperature in assessing urban heat disparities

Accurately assessing socioeconomic disparities in human-perceived urban heat exposure and vulnerability is crucial for effective urban planning. However, the lack of high-resolution 2-meter air temperature (T_a) data due to insufficient monitoring stations has led to widespread reliance on land surface temperature (T_s) as a proxy, despite T_s not having direct implications for public health. It remains unclear whether, and to what extent, T_s misrepresents such disparities, particularly at decision-making scales. In this study, we leverage our newly developed 1-km resolution daily urban T_a data from 2013 to 2023 to evaluate discrepancies between T_a- and T_s-based assessments of urban heat disparities. Our preliminary results show that at daytime, T_s significantly exaggerates disparities among racial and socioeconomic groups and misidentifies vulnerable communities. Across the contiguous United States, 40.9% of census tracts classified as vulnerable based on T_s are not identified as such by T_a, while 22.4% of tracts that are vulnerable by T_a are overlooked by T_s, as defined by an energy burden above 6% and temperatures above the locally-defined median. These biases in T_s-based assessments could lead to misguided decision-making and inefficient resource allocation. Our findings underscore the need for high-resolution T_a data to ensure more equitable and effective urban heat mitigation strategies.