On-orbit radiometric calibration of FY-3E MERSI-LL low light band (LLB) using the Moon

The Medium Resolution Spectral Imager-Low Light (MERSI-LL) on board Fengyun-3E (FY-3E) is equipped with a low-light band (LLB), marking the first achievement of low-light remote sensing capabilities for Fengyun meteorological satellites. In response to the complexity of on-orbit radiometric calibration for LLB, a calibration method using the Moon as a stable natural target was adopted. This study compiled ten active lunar observations from MERSI-LL under the operational mode from 2022 to 2024, covering phase angles from -17.3° to 5.4°. Qualitative comparisons and quantitative analyses were conducted between the satellite observations and the Robotic Lunar Observatory (ROLO) and Miller-Turner 2009 (MT2009) lunar irradiance models, validating the accuracy of the calibration data and the reliability of the calibration methods. The results indicated a phase-angle dependence of lunar irradiance, with relative differences between the ROLO and MT2009 models consistently below 2% when the phase angle exceeded 5°. Furthermore, this study corrected the influences of lunar phase and variations in the solar spectrum when applying the MT2009 model to satellite observations. The modified MT2009 model exhibited improved consistency with MERSI-LL observations, particularly at phase angles exceeding 10°. Finally, the calibration coefficients were derived by calibrating the digital number (DN) of the MERSI-LL lunar observations against the radiance simulated by the models. The ROLO model yielded the best fitting result, with R² of 0.99796, followed by the MT2009 model and the modified model, with R² values of 0.98765 and 0.9876, respectively. We found that the ROLO model is more suitable for the on-orbit radiometric calibration of satellites compared to the MT2009 model. Meanwhile, the total uncertainty of the calibration is approximately 5.78%. This study provides a validated methodology for lunar-based calibration applicable to other low-light remote sensing systems and a reference for the optimization of lunar irradiance models.