Assimilation of FY-3D MWHS-2 Radiances with the Observation Operator ARMS for Typhoon Prediction

Effective assimilation of satellite microwave radiances remains challenging in regional numerical weather prediction (NWP), particularly for improving tropical cyclone forecasting. Existing radiative transfer models may exhibit inherent uncertainties in simulating Fengyun microwave humidity sounder data, which constrains the fidelity of typhoon initialization and subsequent forecasts. To address these limitations, the Advanced Radiative Transfer Modeling System (ARMS) is integrated into the Weather Research and Forecasting model Data Assimilation (WRFDA) assimilation framework, to enhance the utility of Fengyun satellite observations in regional NWP. This study compares the performance of the ARMS, Community Radiative Transfer Model (CRTM), and Radiative Transfer for Tovs (RTTOV) models as observation operators for the clear-sky assimilation of FY-3D Micro-Wave Humidity Sounder-II (MWHS-2) radiances, quantifying their respective impacts on the initialization and subsequent forecasting of Typhoon Chaba (2022). The results indicate that the CRTM and ARMS models exhibit high consistency in simulating the radiance from the water vapor channels, with their sensitivity to water vapor lower than that of the RTTOV model. This characteristic could lead the CRTM and ARMS experiments to simulate higher brightness temperatures compared to the RTTOV. Specifically, in channels 11 and 13, the brightness temperatures simulated using the ARMS and CRTM produces warmer simulated radiance than the RTTOV does. These differences contribute to more effective adjustments in humidity and geopotential height in the CRTM and ARMS experiments than RTTOV, yielding a reduction in the forecast error for both typhoon track and intensity. Further analysis reveals that ARMS model demonstrates better performance in the track prediction, whereas the use of the CRTM and ARMS model produces comparably smallest errors in the intensity forecast. Additional experiments with Typhoon Muifa (2022) further validate the more favorable performance of CRTM and ARMS in typhoon track forecasting.