Influence of Ground‐Based Microwave Radiometer Profile Assimilation on Fog Genesis Forecasts in the Winter Boundary Layer of Northern India

Abstract This study investigates the synergistic impacts of conventional and non‐conventional atmospheric data assimilation (DA) and fine‐gridded soil state assimilation on wintertime fog formation over the Indo‐Gangetic Plain (IGP), with a specific focus on Delhi. Two DA experiments were conducted using the Weather Research and Forecasting (WRF) model: the first (DA1) assimilated temperature and humidity profiles from a microwave radiometer (MWR) using 3DVar/GSI‐based system, while second (DA2) extended DA1 by incorporating fine‐gridded initial soil fields from the High‐Resolution Land Data Assimilation System (HRLDAS). The effectiveness of these data sets in improving the forecast accuracy of wintertime meteorological parameters within the boundary layer was evaluated. MWR profiles were validated against simultaneous radiosonde (RS) measurements during the winter seasons of 2017–2019, and bias correction using RS data was implemented to enhance MWR profile accuracy. The results indicated that the assimilation of MWR profiles (DA1) improves the accuracy of near surface temperature and humidity forecasts, conducive for fog conditions. The inclusion of soil state assimilation (DA2) further improves the representation of soil states, thereby better capturing the physical processes associated with fog formation. With DA2, biases in near‐surface meteorological and soil variables were significantly reduced (50% in T 2 , 16% in RH 2 , 66% in SM, 46% in ST) compared to DA1. DA2 also improved the representation of surface fog heterogeneity and lifecycle across the IGP, with a spatial skill score of 0.36, versus 0.29 for DA1 and 0.24 without assimilation. Additionally, DA2 achieved a higher critical success index (CSI) of 0.75, compared to 0.50 for DA1.