Satellite-based estimation of monthly mean hourly 1-km urban air temperature using a diurnal temperature cycle model

Abstract

Cities worldwide face escalating climate change risks, underscoring the need for spatially and temporally resolved urban air temperature (T_a) data. While satellite-derived land surface temperature (LST) data have been widely used to estimate T_a, high-resolution hourly T_a estimation in urban areas remains underexplored. Traditional methods typically rely on LST data from geostationary satellites and continuous 24-h T_a observations from weather stations. To address these limitations, we introduce a method that combines a diurnal temperature cycle (DTC) model with a random forest model to estimate monthly mean hourly urban T_a at 1-km resolution. This approach leverages a limited number of diurnal T_a observations from weather stations, MODIS LST data, and ancillary information. The core idea of the proposed method is to transform the estimation of monthly mean hourly 1-km T_a into estimating 1-km DTC model parameters, primarily daily maximum and minimum T_a values. This method capitalizes on MODIS LST's ability to estimate daily T_a extremes and requires only four diurnal T_a observations within a daily cycle to estimate monthly mean hourly 1-km T_a. Station-based five-fold cross-validation yields overall RMSE values consistently below 1.0 °C across nine cities with diverse geographic and climatic contexts. The accuracy achieved with only four diurnal T_a observations rivals that obtained using continuous 24-h T_a observations. Even with a limited training set of ten stations, the overall RMSE remains below 1.0 °C for most cities. The proposed method proves effective for both single-city and multi-city modeling and can estimate daily hourly 1-km T_a under clear-sky conditions. In conclusion, this study offers a feasible, efficient, and versatile method for accurately estimating monthly mean hourly 1-km T_a, which can be readily applied to other cities and holds potential for various applications.