Reconstruction of hourly sea surface temperature based on FY-4B AGRI

The diurnal variation of sea surface temperature (SST) is of great significance for the research on air-sea flux exchange, marine ecology, etc. The diurnal variation of SST in China’s offshore waters is obvious. However, high-resolution, high-coverage, and accurate hourly-level SST data are scarce. Therefore, in this study, based on the quality control and bias correction of the SST products retrieved by the Advanced Geostationary Radiation Imager (AGRI) carried on Fengyun-4B (FY-4B) satellite, data interpolating empirical orthogonal functions (DINEOF) method is used to reconstruct the SST under clouds, and hourly-level SST products for China’s offshore waters are generated. The spatial coverage rate of the reconstructed SST has been significantly improved, and the detailed characteristics of the original observed data have been retained. Compared with the in-situ measured SST, the average bias of the reconstructed SST is 0.04°C, and the root mean square error is 0.84°C. These data can well reflect the diurnal variation process of SST. Compared with the hourly SST produced by Japan, whose spatial coverage mainly lies in the Sea of Japan, the reconstructed SST of FY4B can cover the entire coastal area of China, and it has better accuracy and timeliness. Therefore, it can better meet the application needs of users in China.