Lightning Forecasting Using EWRF Model Over North‐Eastern India: Preliminary Results

This study evaluates the integration of an explicit electrification module in the Weather Research and Forecasting model to predict lightning over North‐Eastern India, using inductive and non‐inductive charging mechanisms for hydrometeors and charge exchange during collisions. A multigrid solver computes electric field components, while a bulk discharge scheme reduces charge in regions where the field exceeds the breakdown threshold. Simulations employed nested domains (27 km, 9 km, and 3 km), allowing detailed storm analysis via explicit microphysics. Pre‐monsoon simulations included lightning data assimilation (LDA) from Earth Networks Total Lightning Network sensors, adjusting moisture fields through nudging within the 3 km grid. Over 43 pre‐monsoon days, including 3 April 2017, have been chosen for this study, LDA improved model accuracy, particularly in high‐density regions like Assam and Meghalaya, achieving high Probability of Detection (POD) and Equitable Threat Score (ETS) with low False Alarm Ratio (FAR). LDA enhanced 3–6 hr forecasts (91%–100% accuracy) by refining water vapor fields but struggled beyond 6 hr, where POD dropped and FAR rose. Challenges emerged in low‐density areas like Mizoram, where overprediction increased FAR. The model captured core lightning zones but performance decreases surrounding areas, reducing spatial accuracy. A positive bias in flash density over central Assam and Bangladesh suggests adjusting module parameters to improve performance across varied lightning conditions.