Abstract
GOES-16, the first in a series of new generation Geostationary Operational Environmental Satellite (GOES), carries an Advanced Baseline Imager (ABI) with six Visible and Near Infrared (VNIR) channels that are calibrated periodically with an onboard solar diffuser. After its launch in 2016, we found seasonal changes in the time series of some VNIR channel gains that cannot be explained by instrument degradation nor measurement errors. This was thought to be related to the implementation of the different solar calibration algorithm version and the impacts of satellite instrument angles condition. To characterize and quantify this seasonal variation and measurement bias, we firstly reprocessed all GOES-16 solar calibration events since January 15, 2017 with the same calibration algorithm and look-up-table (LUT), which excluded the impact of algorithm difference in implementation. Secondly, to correct for the impact of “imperfect” solar calibration timing (elevation angle not equal to zero) before Dec 12, 2017, we used the extra solar calibration events in 2018 with the same elevation angle as the ones in 2017 to estimate the degradation rate in the first year, and then corrected this for the previous events. After removal of the bias due to non-zero elevation angle, we then derived the relationship between seasonal variation of corrected gains with the instrument beta angle. Our result shows that there is an asymmetrical pattern between beta angle and the gains, especially band 1. This dependence of the gains on beta angle was likely related to Bidirectional Reflectance Distribution Function (BRDF) factor in solar calibration. We finally derive the bias correction coefficient and then applied to correct such a dependence. Our results show clearly that it improves the assessment the GOES-16 VNIR instrument degradation and calibration repeatability. We will present these and other details at the conference.
Correction for Dependence of Instrument Elevation/Beta Angles on GOES-16 ABI Solar Calibration
GOES-16, the first in a series of new generation Geostationary Operational Environmental Satellite (GOES), carries an Advanced Baseline Imager (ABI) with six Visible and Near Infrared (VNIR) channels that are calibrated periodically with an onboard solar diffuser. After its launch in 2016, we found seasonal changes in the time series of some VNIR channel gains that cannot be explained by instrument degradation nor measurement errors. This was thought to be related to the implementation of the different solar calibration algorithm version and the impacts of satellite instrument angles condition. To characterize and quantify this seasonal variation and measurement bias, we firstly reprocessed all GOES-16 solar calibration events since January 15, 2017 with the same calibration algorithm and look-up-table (LUT), which excluded the impact of algorithm difference in implementation. Secondly, to correct for the impact of “imperfect” solar calibration timing (elevation angle not equal to zero) before Dec 12, 2017, we used the extra solar calibration events in 2018 with the same elevation angle as the ones in 2017 to estimate the degradation rate in the first year, and then corrected this for the previous events. After removal of the bias due to non-zero elevation angle, we then derived the relationship between seasonal variation of corrected gains with the instrument beta angle. Our result shows that there is an asymmetrical pattern between beta angle and the gains, especially band 1. This dependence of the gains on beta angle was likely related to Bidirectional Reflectance Distribution Function (BRDF) factor in solar calibration. We finally derive the bias correction coefficient and then applied to correct such a dependence. Our results show clearly that it improves the assessment the GOES-16 VNIR instrument degradation and calibration repeatability. We will present these and other details at the conference.