Direct Assimilation of FY-3E Hyperspectral Infrared Atmospheric Sounder-II Radiance Data in CMA-GFS System

Abstract The assimilation of satellite-borne hyperspectral infrared data shows an increasingly important impact on the improvement of numerical weather prediction. Fengyun-3E ( FY-3E ) is China’s first early-morning-orbit satellite, which was launched on 5 July 2021. The second-generation Hyperspectral Infrared Atmospheric Sounder-II (HIRAS-II) on board FY-3E can provide continuous high-spectral-resolution radiation data. In this study, the FY-3E HIRAS-II data assimilation system was established within the framework of the China Meteorological Administration Global Forecast System (CMA-GFS). Data preprocessing involves apodization, preliminary channel selection, data splicing and thinning, etc. Key modules for assimilation include the clear-channel cloud detection based on the differences between observed and simulated brightness temperature ( O − B ), the quality control module according to the characteristics of the data error, and the bias correction module considering the satellite scanning bias and real-time atmospheric state. The impact of HIRAS-II data on the analysis fields and model forecasts was evaluated through four sets of 40-day cycling experiments. The results show that single-instrument assimilation significantly reduced the analysis errors, particularly the root-mean-square error (RMSE) of the geopotential height and temperature fields above 500 hPa in the Southern Hemisphere. HIRAS-II data assimilation in operational settings reduces the analysis RMSE of geopotential height by 0%–2% in the Southern Hemisphere and approximately 1% at most in the Northern Hemisphere. For forecasting, the assimilation of HIRAS-II data has a neutral to slightly positive impact. The improvement is mainly reflected in the Southern Hemisphere, where it reduces the RMSE of 72–120-h forecasts with statistical significance at the 95% level.