This study assesses the impact of the assimilation of Doppler Weather Radar (DWR) reflectivity and radial velocity on various meteorological parameters and their influence on the simulation of thunderstorm formation and lightning occurrence during short-duration weather events over Odisha, India. The regional version of the National Centre for Medium Range Weather Forecasting (NCMRWF) Unified Modelling (NCUM-R) system is used in this study. Observing System Experiments (OSEs) are designed to estimate the impact of DWR data through the 4D-VAR framework. The analysis is performed on a baseline OSE (CTL), which assimilates conventional and satellite observations (winds and radiances) and a second OSE (RAD), which assimilates the DWR data in addition to the observations assimilated in CTL. The experiments reveal significant improvements in simulating atmospheric dynamics and convective processes following DWR data assimilation. Results indicate that RAD analyses exhibit more intense wind flow and higher relative humidity compared to CTL, suggesting a higher likelihood of thunderstorm formation. The RAD experiment demonstrates better simulation of wind, relative humidity, vorticity, convergence, and vertical velocity, closely resembling ERA5 reanalysis. Enhanced convective instability, evidenced by higher equivalent potential temperature in RAD, correlates with increased lightning activity. Model performance in reproducing lightning flashes is additionally improved in RAD, with closer agreement with observed lightning flash counts and spatial distribution. Furthermore, RAD simulations show better representation of convective rainfall patterns, with considerable improvement in Equitable Threat Score (ETS) across rainfall thresholds compared to CTL. Assimilation of DWR data into the model enhances forecast skill, with RAD simulations consistently outperforming CTL in terms of ETS and Mean Square Error (MSE). Also, RAD simulations exhibit lower error in volume, displacement and pattern, indicating improved simulation of rainfall intensity and distribution. Overall, assimilating DWR data significantly enhances model capabilities in simulating a convective system, leading to improved forecasts of thunderstorm activity, lightning occurrence, and heavy rainfall.
Dutta et al. (Mon,) studied this question.