Step-By-Step Downscaling of Land Surface Temperature Considering Urban Spatial Morphological Parameters

Land surface temperature (LST) is one of the most important parameters in urban thermal environmental studies. Compared to natural surfaces, areas more complex, and spatial variability LST higher. Therefore, it obtain a high-spatial-resolution for research. At present, downscaling studies are mostly performed from low resolution directly another high resolution, which often results lower accuracy with larger scale span. First, step-by-step random forest model (SSRFD) proposed this study. In our work, 900-m Sentinel-3 was sequentially downscaled 450 m, 150 m 30 by SSRFD. Then, morphological were introduced into SSRFD, abbreviated as SSRFD-M, compensate deficiency remote-sensing indices driving factors LST. The showed that RMSE value SSRFD reduced 2.6 °C 1.66 compared direct (DRFD); SSRFD-M built-up areas, such Gulou Qinhuai District, approximately 0.5 °C. We also found underestimation caused considering only places flowerbeds streets improved results.