On-board radiometric calibration for thermal emission band of FY-3C/MERSI

On-board radiometric calibration is the most efficient method to improve the measurement accuracy of satellite-borne sensors. Chinese Medium Resolution Spectral Imager (MERSI), loaded on the Fengyun-3C (FY-3C) satellite, uses one satellite-borne fixed-temperature blackbody (BB) for its thermal emission band's (TEB) radiometric calibration, but its optimal calibration algorithm is not determined. By using MERSI's on-board calibration data taken on November 2013, this article investigates two algorithms of linear and semi-nonlinear calibration for MERSI TEB's on-board radiometric calibration and finds that the linear calibration is more reasonable than semi-nonlinear calibration because linear coefficients' variation tendency can reflect MERSI's inherent systematic properties better. The relative difference between linear properties and inherent properties for pixel variability being 9.5%, mirror-side variability being 21.5% and scan variability being 17.8% are all smaller than those between semi-nonlinear case and inherent case. All of those suggest that the linear calibration is coincident with the inherent systematic properties. By using MERSI's calibration data at June 2014, the performance of those two algorithms is validated by comparing the difference between inferred BB radiance L-I and standard BB radiance L-S (0.01%), and between inferred BB brightness temperature T-I and standard BB temperature T-S (0.25 K).