Urban land surface temperature retrieval with high-spatial resolution SDGSAT-1 thermal infrared data

In the context of global urbanization, unplanned urban expansion renders cities particularly susceptible to the impacts of climate change, natural disasters, and extreme heat and humidity. Monitoring the land surface temperature (LST) in urban areas is crucial for assessing the urban thermal environment. Fine-scale (<100 m) LST products are essential for comprehensively understanding urban environments because of their detailed thermal distribution patterns. The Sustainable Development Science Satellite 1 (SDGSAT-1), launched on November 5, 2021, possesses the capability to capture fine-scale urban LST imagery at a 30-m resolution, both day and night. Based on the characteristics of the SDGSAT-1 thermal infrared data, we implemented two methods (the multiband-based (MBB) method and the two-band-based (TBB) method) to generate 30-m urban LSTs. The derived LSTs are evaluated against the MODIS LST product and in situ measurements. Furthermore, various simulation datasets are constructed based on the spectral characteristics of SDGSAT-1/TIS and utilized to derive LSTs using the MBB and TBB methods to further clarify the feasibility of the two methods. The results indicate that the MBB method exhibits superior performance in urban areas, with an RMSE that is 0.74 K lower than that of the TBB method. In contrast, the TBB method is suitable in areas with lower emissivity fluctuations, such as dense vegetated areas, with an RMSE that is 0.96 K lower than that of the MBB method. These two methods are planned for incorporation into the SDGSAT-1 LST production framework, thereby contributing to the advancement of accurate LST retrieval and achieving sustainable development goals in the future.