【摘要】 利用L波段探空和降水数据资料对人工影响天气数值模式产品开展本地化的定量化检验，分析其模拟性能，结果表明：(1)云降水显式预报系统(cloud precipitation explicit forecast system, CPEFS)与睿图数值预报系统(rapid-refresh multi-scale analysis and prediction system, Rmaps)对云底高度/温度的预报均偏高，且CPEFS优于Rmaps;对云顶高度/温度预报，Rmaps略优于CPEFS。(2)CPEFS与Rmaps对特征温度层(0、-5、-10、-20℃)高度的预报值和观测值均比较接近，Rmaps对-20℃高度的预报优于其他温度层高度。(3)随着降水量级增大，两种模式对降水预报的TS评分减小，漏报率和空报率增加，CPEFS对于小雨和中雨的空报率小于漏报率，Rmaps对于小雨和中雨的空报率大于漏报率，两种模式对于大雨和暴雨的漏报率均小于空报率。更多还原

【Abstract】 Localized quantitative verification was conducted on the numerical weather prediction(NWP) model products for weather modification using L-band radiosonde and precipitation data, with an analysis of their simulation performance. The analysis results indicated:(1) Both the Cloud-Resolving Ensemble Forecast System(CPEFS) and Rmaps tended to overestimate the forecasted cloud base height/temperature, with CPEFS outperforming Rmaps in this regard. For cloud top height/temperature forecasts, Rmaps slightly outperformed CPEFS.(2) The forecasted values of CPEFS and Rmaps for the heights of characteristic temperature layers(0 ℃,-5 ℃,-10 ℃,-20 ℃) were relatively close to the observed values. Rmaps provided better forecasts for the height at-20 ℃ compared to other temperature layers.(3) As precipitation intensity increased, the Threat Score(TS) of both models for precipitation forecasts decreased, accompanied by an increase in both the miss rate and false alarm rate. For light and moderate rain, CPEFS had a lower false alarm rate than the miss rate, while Rmaps had a higher false alarm rate than the miss rate. For heavy rain and torrential rain, both models exhibited a miss rate lower than the false alarm rate.更多还原