An Urban Single-Channel Algorithm for Retrieving Land Surface Temperature From Chinese ZY1-02E Data

Abstract

Urban land surface temperature (ULST) is a key parameter for characterizing urban thermal environments and surface energy processes. However, the satellite-based ULST retrieval remains constrained by the coarse spatial resolution of conventional thermal infrared (TIR) sensors. The Chinese ZY1-02E satellite provides a unique opportunity for fine-scale thermal observation with its 15-m resolution TIR sensor (IRS), but its broadband spectral response (7– $12~\mu $ m) challenges the atmospheric assumptions underlying conventional mono-window algorithms. In this study, an urban single-channel (USC) algorithm is developed for retrieving land surface temperature (LST) from ZY1-02E IRS data. Unlike traditional mono-window approaches designed for narrow atmospheric-window sensors, the proposed method explicitly considers the broadband spectral characteristics of IRS and separately characterizes the upward and downward atmospheric radiative contributions in the surface–atmosphere radiation transfer process. The algorithm was evaluated against in situ measurements and cross-compared with satellite-derived LST products and existing single-channel (SC) methods. The results show that the proposed USC algorithm achieves mean absolute error (MAE) and root mean square error (RMSE) of 1.65 and 2.31 K, respectively, outperforming the Qin_SC and JM_SC algorithms. Cross-validation with Landsat 8 LST products demonstrates consistent spatial patterns over both urban and natural surfaces, with MAEs of 1.75 and 2.09 K, respectively. Furthermore, the comparisons with other USC algorithms show that the USC algorithm achieves the lowest retrieval errors. The proposed framework provides a physically based approach for LST retrieval from broadband TIR observations and facilitates high-resolution characterization of urban thermal environments.