Impacts of Assimilating All‐Sky FY‐4A AGRI Satellite Infrared Radiances on the Prediction of Super Typhoon In‐Fa During the Period With Abnormal Changes

Abstract The capabilities of assimilating the all‐sky Fengyun‐4A Advanced Geostationary Radiation Imager (AGRI) infrared radiances (IR) are completed by including hydrometers in the observation operator, its adjoint, and tangent linear model. This allows the three‐dimensional variational data assimilation model to include cloud‐precipitation information from infrared IR observations. Advanced as the all‐sky data assimilation methodologies are, the assimilation of cloudy scene IR radiances for tropical cyclone (TC) systems has not led to consistently better results, especially for the intensity of TCs. This work explores the effects of all‐sky AGRI radiance assimilation on a Super Typhoon In‐Fa (2106) during its stage experiencing abnormal changes in the intensity and the track. It is shown that the all‐sky assimilation of AGRI two channels 9–10 brings no obviously better TC forecasts than the all‐sky AGRI single‐channel assimilation does. Besides, the O − B (observation minus background) bias was corrected to be even larger with the variational bias correction method for the pixels with relatively lower or higher cloud impact. This indicates that traditional bias correction schemes with linear fitting functions are suboptimal if the relations between the predictor and O − B biases are non‐linear. When the conventional observation and IR radiances are assimilated in two steps, the wind in the inner‐core region is better described to properly capture the changes in the typhoon intensity. Generally, the analyses and forecasts of Typhoon In‐Fa from experiments with the all‐sky IR observations are enhanced compared to those with only the clear‐sky IR observations.