Re-visit of GOES Visible Lunar Calibration: Error Budget and Scan-Angle Dependent Reflectance

The moon is well recognized as a reference source for the space-based remote sensing instrument calibration. NOAA/NESDIS, in collaboration with USGS, used the Robotic Lunar Observatory (ROLO) model to examine the GOES-10 Imager visible channel calibration degradation in early 2006. A total of 33 suitable lunar observations at various phase angles had been collected over the eight years of GOES-10 Imager operational period between July 1998 and June 2006. The relative calibration accuracy with lunar calibration was about 3%, comparable with the results of other vicarious calibration methods conducted at NOAA/NESDIS. It was believed then that the space background noise, stray-light and the possible incident-angle dependent reflectance were the major components attributed to the calibration uncertainty. The calibration method and trending result was reported in SPIE 2006 and the analysis of possible uncertainty components was presented in CALCON 2006. In this study, we revisit the GOES Imager lunar calibration by assessing the GOES Imager visible lunar calibration error budget for each uncertainty component. As recent research results confirm that GOES Imager visible channel has scan-angle dependent reflectance, we will also try to characterize the angular reflectance of the scan mirror with lunar observation data. Results of the error budget analysis and the scan-angle dependent reflectance will be reported in the coming meeting