Retrieval of high-resolution sea surface temperature from FY-3E MERSI-LL data in the Northwest Pacific

Abstract

Fengyun-3E (FY-3E) is an operational meteorological satellite in an early-morning orbit. The Medium Resolution Spectral Imager-Low Light (MERSI-LL) onboard the FY-3E satellite provides observations in the 10.8 and 12.0 μm channels at a spatial resolution of 250 m, offering crucial data for high-resolution sea surface temperature (SST) retrieval. First, the stripe noise in the MERSI-LL Level 1 data was reduced by developing an adaptive decomposition optimization model based on the alternating direction method of multipliers. Second, a multi-threshold collaborative method based on the MERSI-LL brightness temperature was adopted to detect clouds. Then, matchups between clear-sky ocean observations and in situ observations were generated within a temporal window of 1 h and a spatial window of 0.0025°. The nonlinear sea surface temperature algorithm and quality level classification were applied to retrieve SST in the Northwest Pacific Ocean. Finally, validation was conducted using in situ observations and the multisensor supercollated gridded SST product from low-Earth-orbit satellites (L3S-LEO SST). For the best-quality retrieved SST, the bias and standard deviation (STD) relative to the in situ observations were 0 and 0.59 °C, respectively. For the best-quality SST data, the bias and STD relative to the L3S-LEO SST products were 0.14 and 0.33 °C, respectively. The results demonstrate that the regional high-resolution SST values retrieved from FY-3E MERSI-LL data exhibit satisfactory accuracy in the Northwest Pacific and can support operational marine environmental monitoring and meteorological analysis.