Joint Assimilation of FY-4B AGRI and GIIRS Infrared Radiance and Its Effects on Typhoon Prediction

Current research on Fengyun-4B (FY-4B) satellite data assimilation has primarily focused on individual instrument. The Advanced Geostationary Radiation Imager (AGRI) water vapor channels data effectively extract atmospheric water vapor information, while the Geostationary Interferometric Infrared Sounder (GIIRS) temperature channels data provide high vertical resolution atmospheric temperature information. To achieve complementary advantages in temperature–humidity detection and space, this study develops a joint assimilation framework for FY-4B AGRI all-sky data and GIIRS clear-sky data, and investigates their joint assimilation characteristics and impacts on the forecasts of Typhoons Gaemi and Kong-Rey. The results show that, in the joint assimilation, AGRI supplements observations in cloudy regions and plays a dominant role in adjusting the moisture analysis field. Compared with the GIIRS experiment, the AGRI experiment produces more stable track errors, achieves better performance in the late-stage typhoon track forecast, and obtains higher precipitation scores. In contrast, GIIRS in the joint assimilation mainly contributes to the temperature analysis field and exhibits finer vertical structures of temperature increments. Compared with the AGRI experiment, the GIIRS experiment shows more pronounced improvements in the first 24-h typhoon track forecast and in the intensity forecast. The joint assimilation integrates the strengths of both datasets, maintaining relatively small and stable track errors throughout the forecast period, while also showing superior performance in the late-stage central pressure forecast, warm-core structure simulation, and precipitation forecast, thereby achieving comprehensive optimization of typhoon prediction.