<span style="mso-fareast-language: ZH-CN;">Synergistic 3D-Var Assimilation of FY-4B Satellite AOD and Surface Aerosols Enhances Dust Forecasts and Radiative Brightness Temperature Accuracy in Northern China

This study investigates a severe sandstorm event in northern China during 21-23 March 2023 using a three-dimensional variational (3D-Var) aerosol data assimilation system coupled with the WRF-Chem model. By assimilating FY-4B geostationary satellite aerosol optical depth (AOD) retrievals and surface PM2.5/PM10 mass concentration observations, the research evaluates improvements in aerosol initialization and quantifies aerosol-meteorology feedback through radiative impacts on brightness temperature (BT). A novel approach integrates the Rapid Radiative Transfer Model (RTTOV) to simulate FY-4A satellite infrared BT, enabling independent validation of aerosol-radiation interactions. Results demonstrate that aerosol assimilation significantly enhances initial field accuracy, reducing PM2.5 and PM10 root mean square errors (RMSE) by 56.3% and 63.4%, respectively, with forecast improvements persisting over 40 hours. For meteorological fields, assimilation optimizes aerosol radiative effects, reducing BT biases in dust-affected regions (e.g., Beijing-Tianjin-Hebei). Statistical metrics reveal a 11.5% decrease in BT RMSE and an increase in the index of agreement (IOA) from 0.533 (control) to 0.812 (assimilation), highlighting enhanced representation of aerosol scattering-absorption coupling for coarse-mode dust particles. The study underscores that constraining aerosol fields through multi-source data assimilation not only refines pollutant predictions but also indirectly improves meteorological simulations via radiation-mediated pathways. These findings advance understanding of bidirectional aerosol-meteorology feedback mechanisms and demonstrate the value of geostationary satellite BT products in validating coupled chemistry-climate models.