麻豆淫院

Land surface temperature (LST) is a critical parameter for understanding Earth's energy balance and monitoring climate change. However, due to the limited spatial resolution of current satellite observations, LST measurements often reflect a mix of sunlit and shaded surface components. This blending introduces significant uncertainty into key environmental assessments involving LST, such as analyses of global warming trends and evaluations of afforestation impacts.

To address this challenge, a research team from the Aerospace Information Research Institute of the Chinese Academy of Sciences has presented the first satellite-scale retrieval of LST components鈥攕pecifically, sunlit soil, shaded soil, and vegetation鈥攂y integrating multi-source observations and leveraging the thermal "hotspot" effect. Their findings were in Remote Sensing of Environment.

Field experiments have demonstrated that the temperature difference between sunlit and shaded surfaces can exceed 20 K at noon during the summer. Satellite-based analyses indicate that this thermal contrast can result in a variation of 0.6 K in retrieved surface temperature for every 10 degree change in the satellite's viewing angle relative to the sun. By utilizing hotspot information, the new method enhances retrieval accuracy by more than 20%, according to the team.

The new approach enhances angular information by combining data from polar-orbiting satellites, such as the Sea and Land Surface Temperature Radiometer (SLSTR), with data from geostationary satellites, like the Advanced Himawari Imager (AHI) and the Spinning Enhanced Visible and Infrared Imager (SEVIRI), in a virtual constellation.

This method introduces the TIRT model, which utilizes a clumping index to more effectively capture hotspot signals associated with vegetation structure. Additionally, it incorporates differences in component temperatures as a physical prior and employs Bayesian inference to improve retrieval stability. Validations using data from locations such as China's Heihe site, Portugal's EVO site, and Namibia's KAL site demonstrated consistent improvements in accuracy.