Urban surface-emitted longwave radiation estimation from high spatial resolution thermal infrared images using a hybrid method

Accurate estimation of the surface-emitted longwave radiation (SELR) has important scientific value in understanding its spatiotemporal dynamics and surface thermal environment. Thermal infrared (TIR) images with high spatial resolution offer enhanced data support for studying SELR of complex surfaces, such as urban surface. This study proposes a new urban-oriented hybrid (UoHy) method, which considers multiple scattering and adjacent effects of urban pixel using the term sky view factor, to estimate urban SELR from the top-of-atmosphere radiance of TIR images with high spatial resolution, and performs sensitivity analysis and application. The experimental results for the thermal images of GF-5/VIMS as an example showed that the UoHy method has relatively high accuracy, and obtains SELR errors of less than 12.0 W/m² under low atmospheric water vapor conditions and less than 17.0 W/m² under high atmospheric water vapor conditions. The application of the method during daytime and nighttime also demonstrated the method's validity and effectiveness. Compared with the natural surface-oriented hybrid method, the UoHy method obtains an improvement of about 7.0–10.0 W/m² in SELR estimation, which indicates that it has the potential for practical SELR estimation over urban surface with TIR images of high spatial resolution.