Simulation Impact of Radar Data Assimilation on a Squall Line Process

Radar data assimilation can improve the forecasts of severe convective weather, but different model configurations generate different results. This paper studies a squall line process that occurred in southern China on 4 March 2018. ARPS (Advanced Regional Prediction System) 3DVAR (three-dimensional variational) data assimilation system is applied to assimilate Doppler radar radial velocity, and cloud analysis is used to process radar reflectivity data. Considering the assimilation interval, frequency, and different parameter adjustments in cloud analysis, different assimilation schemes are designed by adopting a 1-h assimilation window. Using WRF (Weather Research and Forecasting) model forced by the GFS (Global Forecast System) analyses and forecasts, the influence of radar data assimilation on the triggering and development mechanism of the squall line system is investigated. The results show that when the assimilation interval is too short, false echoes are generated because of an imbalance in the model thermodynamic variables. When the assimilation interval is too long, the system triggering and development characteristics are generally weak. The best initial field is obtained using a 12-min interval assimilation, and the higher the assimilation frequency, the better the precipitation forecast results. In addition, ARPS cloud analysis can greatly improve the initial field and reduce the model spin-up time. Among the parameter adjustments, those of humidity, temperature, rainwater, and water vapor have a greater impact on the dynamic process of the system and the initial field distribution of the hydrometeors, while parameter adjustments related to vertical velocity have less impact.