Nowcasting of a Warm-Sector Rainfall Event in Southern China With the TRAMS Model: Sensitivity to Different Radar Reflectivity Retrieval Methods and Incremental Updating Strategies

Abstract

To improve the radar data assimilation scheme for the high-resolution Tropical Regional Atmospheric Model System (TRAMS) model, this study investigates the sensitivity of simulating a warm-sector rainfall event in southern China to different radar reflectivity retrieval methods and incremental updating strategies. The findings indicate that the ice cloud retrieval (ICR) method yields more reasonable cloud hydrometeors. However, the impact of different retrieval methods is minimal without corresponding adjustments to the dynamic field. Further assimilation of the wind field effectively reduced the overestimated south winds and successfully simulated the observed low-level convergence in northern Guangdong, significantly improving precipitation forecasts. Both incremental analysis update (IAU) and Nudging methods were able to adjust the forecast to better match the observations, with IAU performing slightly better. These findings are beneficial for further improving the forecast accuracy of precipitation intensity. Extending the IAU relaxation time from 4 to 10 min has almost no impact on the actual forecasting. However, prioritizing the adjustment of the wind field through time-dependent IAU weighting factors, the impact of cloud particle adjustments on the dynamical field can be avoided (e.g., the drag caused by the sinking of cloud particles may offset the upward motion induced by dynamical convergence adjustments). This allows for more realistic low-level wind convergence and precipitation forecasts to be obtained. Overall, the ICR method for retrieving cloud hydrometeors, combined with the IAU method using time-dependent distribution weighting factors appears to be a more suitable option for the radar data assimilation scheme in TRAMS model.