Coherent Doppler Wind Lidar Wind Profile Observation and WRF Model Simulated Hourly Precipitation

Assimilating wind profiler observations can improve short-term precipitation forecasts, but the effectiveness of such assimilation depends strongly on the specification of the assimilation impact radius. This study investigates the impact of assimilating coherent Doppler wind lidar (CDWL) wind profiler observations on quantitative precipitation forecasts in the Weather Research and Forecasting (WRF) model, with an emphasis on dynamically estimating the assimilation impact radius, based on a heavy precipitation event that occurred over Hebei Province, China, during 23–26 July 2025. Sensitivity experiments were conducted using a dynamically diagnosed impact radius (DIR Exp.), the default static impact radius in WRF (SIR Exp.), and a control experiment without CDWL assimilation (CTL Exp.). Forecast performance was evaluated using continuous and categorical verification metrics, including temporal and spatial correlation coefficients of hourly accumulated precipitation and receiver operating characteristic analysis for heavy precipitation events. Results show that assimilating CDWL wind profiler observations improves precipitation forecast skill relative to CTL Exp., with the most pronounced benefits occurring during forecast lead times of 1–8 h. Compared with the static impact radius, the dynamically estimated impact radius yields higher spatial correlation and improved detection skill, particularly in regions with dense CDWL deployments. Diagnostic analyses based on moist potential vorticity indicate that the dynamic impact radius produces a more physically consistent thermodynamic–dynamic structure in the lower troposphere, characterized by enhanced instability and stronger convergence favorable for convective initiation and maintenance. These results highlight the importance of scale-aware assimilation strategies for improving short-term heavy precipitation forecasts.