【摘要】 基于安徽省80个国家级气象站与3 510个区域自动站的降水观测数据,系统评估了安徽省气象台对2024年6月19日入梅首场大暴雨过程的预报表现。结果表明,主观预报虽成功预测江南地区存在大暴雨,但落区明显偏南、范围偏小、强度偏弱。通过综合对比ECMWF、WRF、风雷模型等数值预报产品与ERA5再分析资料及雷达观测,发现各模式对暴雨过程具备基本预报能力,但对极端降水强度与精确落区的刻画仍存在显著偏差。分析表明,模式对中高层冷空气活动、低空急流强度及地形强迫效应的模拟不足,是导致预报系统性偏弱的主要原因。进一步分析得出,风雷模型在短时临近预报中表现出良好应用潜力,可作为传统数值预报的有效补充。分析结果可为今后类似天气过程的预报改进与技术发展提供参考依据。更多还原

【Abstract】 Based on precipitation observation data from 80 national meteorological stations and 3 510 regional automatic stations in Anhui Province,this study systematically assesses the forecast performance of the Anhui Provincial Meteorological Observatory for the first heavy rainstorm of the 2024 Meiyu season on June 19.Results show that although the subjective forecast successfully predicted the occurrence of heavy rainstorm in the Jiangnan region,the predicted rainfall was significantly farther south,narrower in coverage and weaker in intensity than the actual rainbelt.By comprehensively comparing numerical forecast products such as ECMWF,WRF,and the Fenglei model with ERA5 reanalysis data and radar observations,it was found that all models have basic forecasting capability for heavy rainstorm events,but there are still significant deviations in depicting extreme rainfall intensity and precise rainfall locations.Analysis indicates that insufficient simulation of mid-and upper-level cold air activity,low-level jet intensity,and orographic forcing effects by the models is the main reason for the systematic underestimation by the forecast.Further analysis shows that the Fenglei model demonstrates good application potential in shortterm nowcasting and can serve as an effective supplement to traditional numerical forecasts.These results provide a reference for the improvement of forecast accuracy and technological development for similar weather events in the future.更多还原