资源环境科技发展态势分析平台

The Suomi National Polar-orbiting Partnership Program (NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) Thermal Emissive Bands (TEBs) have been performing well since the data became available on 20 January 2012, and the Sensor Data Record data reached validated maturity on 18 March 2014. While overall the validation has shown that these channels have an estimated absolute uncertainty on the order of 0.1 K based on extensive comparisons, there is a remaining issue that persisted over the years. A calibration bias on the order of 0.1 K is introduced in channels such as M15 during the quarterly blackbody temperature warm-up/cooldown, and the bias is further amplified by the sea surface temperature (SST) retrieval algorithm up to 0.3 K in the global daily-averaged products which causes an apparent spike in the SST time series. Our investigation reveals that this bias is caused by a fundamental but flawed theoretical assumption in the VIIRS calibration equation, which states that the shape of the calibration curve is assumed unchanged from prelaunch to postlaunch without any constrains. While the assumption may work to account for long-term degradation, it has a shortcoming during the blackbody unsteady state. In this study, we present a diagnostic and correction method with a compensatory term (L-trace) to reconcile the assumption such that it removes the calibration bias during the blackbody temperature changes. The methodology has been tested using historical data, and the results are very positive. The implementation has minimal impacts on the operational data processing system and is readily available for use in operations.

Plain Language Summary This paper studies an anomaly with the VIIRS instrument in the infrared channels which affects sea surface temperature retrievals. It diagnoses the traceability of the calibration and provides a correction to resolve the issue to improve the quality of the data for all users.