Invalid cloud top properties product of Fengyun-4A satellite imager induced by infrared calibration bias

Recently, China successfully launched the new-generation geostationary (GEO) meteorological satellites, Fengyun-4A (FY-4A) in the year of 2016. In general, many mature and useful level-2 science product of Advanced Geostationary Radiation Imager (AGRI) onboard FY-4A were well developed in advance for satellite data users. Cloud top properties (CTP, including cloud top height, temperature, and pressure) product as an important science product is always used to monitor the rapid changes of typhoon and weather systems in the operational application. Unfortunately, we still find some invalid retrievals in CTP product of FY-4A/AGRI, which seriously impact on the use of this product (get some complaints from users). After comparing the invalid pixels with valid pixels and the same CTP product of Japanese Himawari-8 satellite and radiative transfer simulations, we find that the radiometric calibration bias of infrared band at 13.5μm is the primary impact factor on this fault. The pixels with brightness temperature lower than 200K show a bias larger than 7-8K, which directly induce the invalid retrieval process in the CTP product of FY-4A/AGRI. However, this fault in CTP also inspire us to use the invalid CTP pixel to monitor the on-orbit radiometric calibration bias of infrared band in the future.