An Evaluation of the Uncertainty of the GSICS SEVIRI-IASI Intercalibration Products

Global Space-based Inter-Calibration System (GSICS) products to correct the calibration of the infrared channels of the Meteosat/SEVIRI (Spinning Enhanced Visible and Infrared Imager) geostationary imagers are based on comparisons of collocated observations with Metop/IASI (Infrared Atmospheric Sounding Interferometer) as a reference instrument. Each step of the cross-calibration algorithm is analyzed to produce a comprehensive error budget, following the Guide to the Expression of Uncertainty in Measurement. This paper aims to validate the quality indicators provided as uncertainty estimates with the GSICS correction. The methodology presented provides a framework to allow quantitative tradeoffs between the collocation criteria and the number of collocations generated to recommend further algorithm improvements. It is shown that random errors dominate systematic ones and that combined standard uncertainties (with coverage factor k = 1) in the corrected brightness temperatures are ~ 0.01 K for typical clear sky conditions but increase rapidly for low radiances - by more than one order of magnitude for 210 K scenes, corresponding to cold cloud tops.