Retrieval of FY-4A Land Surface Temperature for Operational Application

The FY-4A satellite is the first quantitative remote sensing satellite of a three-axis stabilization structure on geostationary orbit launched by China in 2016. The capability and efficiency are improved with the new variety of instruments in earth observation. The thermal infrared (TIR) channels of Advanced Geosynchronous Radiation Imager (AGRI) which one of the new instruments onboard FY-4A provides high-frequency, high-precision, and quantitative observation data to obtain diurnal variation of land surface temperature (LST). In this paper, nine candidate split window LST algorithms were applied to evaluate the applicability of the different algorithm for FY-4A, and the Ulivieri & Cannizzaro (1985) algorithm was selected. Then different algorithms in day and night for dry and wet atmosphere conditions are developed and optimized to improve the accuracy of retrieved FY-4A LST. Then, the operational products of FY-4A LST are produced in clear sky. Results show that the diurnal variation characteristics of LST can be efficiently obtained. And the validation shows the following: 1) the root mean square errors (RMSE) is 3.5 K in day and 4.24 K in night when compared with the Himawari8 LST product; 2) compared with ground-measured data, the average accuracy of our algorithm is 2.5 K in day and 3.56 K in night. Sensitivity analysis shows that emissivity is lower sensitive to the algorithm accuracy, and the atmospheric water vapor content is higher sensitive to the algorithm accuracy with the view angle increased. In general, our algorithm exhibits good accuracy and is an easy retrieval approach to produce LST operational products on a regional scale. And the FY-4A LST operational products will help to the weather services, climate change, land–atmosphere interaction research, and ecological civilization construction, and so on.