Calibration of low-level light sensor using deep convective clouds

The National Polar-orbiting Partnership (NPP) satellite was successfully launched in 28 October 2011. The day night band (DNB) is mounted on the Visible Infrared Imaging Radiometer Suite (VIIRS) on board the NPP to greatly improve the night time imaging capabilities. In order to cover this wide dynamic radiation range, the three gain stages are employed to view the earth at daytime, twilight and even nighttime. Especially, the radiance measured at night is relatively low, so the radiance calibration is too difficult to realize. In order to realize radiance calibration at night, the vicarious calibration method of Low-level light detection at night using the deep convection clouds is presented. The lunar irradiance model was introduced to enhance the SCIATRAN to realize the radiance transfer simulation at all time. Then, the reference values of the surface albedo, atmospheric profiles and optical properties of deep convection clouds were inputted to the enhanced model to study the sensitivity of the TOA radiance. The simulation results show that the main factors are the Cloud Optical Thickness (COT) and Effective Radius (Re) and the maximum reflectance errors are no more than 5%. In order to examine the feasibility of the calibration method, the 6 months NPP data range from August 2012 to January 2013 was used. The simulated albedos were calculated using the enhanced radiative transfer model and compared with the observed albedos. The result shows that the radiance uncertainties on a daily basis are in the range of ±10%, which means that the calibration method using Deep convection clouds is very well to calibrate the DNB high gain stage directly. ©, 2015, Chinese Optical Society. All right reserved.