A mobile sensor-based Approach for Analyzing and Mitigating Urban Heat Hazards

High-quality temperature data at a finer spatial-temporal scale is critical for analyzing the risk of heat hazards in urban environments. The variability of urban landscapes makes cities a challenging landscape for quantifying heat exposure. Most of the existing heat hazard studies have inherent limitations on two fronts: the spatial-temporal granularities are too coarse and the ability to track the actual ambient air temperature instead of land surface temperature. Overcoming these limitations requires radically different research approaches, both the paradigms for collecting the temperature data and developing models for high-resolution heat mapping. We present a comprehensive approach for studying urban heat hazards by harnessing a high-quality hyperlocal temperature dataset from a network of mobile sensors and using it to refine the satellite-based temperature products. We mounted vehicle-borne mobile sensors on thirty city buses to collect high-frequency (5 sec) temperature data from June 2018 to Nov 2019. The vehicle-borne data clearly show significant temperature differences across the city, with the largest differences of up to 10℃ and morning-afternoon diurnal changes at a magnitude around 20℃. Then we developed a machine learning approach to derive a hyperlocal ambient air temperature (AAT) product by combining the mobile-sensor temperature data, satellite LST data, and other influential biophysical parameters to map the variability of heat hazard over areas not covered by the buses. The machine learning model output highlighted the high spatio-temporal granularity in AAT within an urban heat island. The seasonal AAT maps derived from the model show a well-defined hyperlocal variability of heat hazards which are not evident from other research approaches. The findings from this study will be beneficial for understanding the heat exposure vulnerabilities for individual communities. It may also create a pathway for policymakers to devise targeted hazard mitigation efforts such as increasing green space and developing better heat-safety policies for workers.