Assimilation of INSAT-3D imager retrieved aerosol optical depth for enhanced dust storm prediction using WRF-Chem model

Abstract Frequent dust storms during the pre-monsoon season in northwest India significantly impact weather, air quality, and health, necessitating accurate predictions. This study demonstrates the first-time assimilation of aerosol optical depth (AOD) data from INSAT-3D into the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem), alongside previously assimilated AOD data from MODIS AQUA. Focusing on an extreme dust storm over the Indo-Gangetic Plain (IGP) from June 12–17, 2018, identical experiments were conducted with and without AOD assimilation. The Gridpoint Statistical Interpolation (GSI) three-dimensional variational (3DVAR) method was utilized for data assimilation, with simulations performed for 144 h, adjusting for uncertainties in the model’s background error covariance using different inflation factors. The model’s performance was validated using the ECMWF Atmospheric Composition Reanalysis 4 (EAC4) product. The assimilated AOD analysis significantly improved bias, root mean square difference (RMSD), and correlation from [0.47, 0.71, 0.40] to [0.07, 0.38, 0.80]. The 24 h AOD forecast improved by approximately 56% and 30% with INSAT-3D and MODIS AOD assimilation, respectively. AOD assimilation directly impacted cloud properties and radiative forcing, enhancing the 24 h forecast of net shortwave downward flux (SWD) by approximately 51 W m −2 and 24 W m −2 for INSAT-3D and MODIS, respectively. These changes also affected vertical levels up to 5 km, modifying the rainwater mixing ratio (QRAIN) and surface rainfall forecasts. This study shows that integrating INSAT-3D AOD data provides substantial improvements in aerosol forecasts, crucial for making reliable and accurate forecasts of severe dust storms and air pollution episodes over the Indian region.