两种定量集合降水预报方法对比分析

Comparative Analysis of Two Kinds of Quantitative Ensemble Forecasting Methods

文章基于EC集合平均预报产品,选取EC集合预报的51个成员包括温度(500h Pa和700hpa)及降水量,以内蒙古2014年8月份5次暴雨降水过程为例,用择优法(以500h Pa和700h Pa的温度差即稳定度相关系数作为对集合成员的筛选因子)进行试验,并与集合平均预报方法进行对比,结果如下:(1)在短期24h预报中集合成员个数增加到3时预报效果得到一定改善,并优于集合平均法的预报,成员个数增加到较多(如4或5时)或减少到2时预报效果反而较差。因此分析集合预报中集合成员的个数对预报结果的影响有非常重要的意义;(2)无论是集合平均法还是择优法对小量级降水预报效果好于大雨以上降水预报,因此有必要去分析更好的集合预报方法使得集合预报在灾害性天气预报中应用得到进一步发展。

This paper selected EC ensemble prediction of 51 members including temperature（500h Pa and 700hpa） and precipitation based on EC ensemble average forecast products, the optimization method（selects as filtered factors the correlation coefficients of temperature differences of 500 h Pa and 700 h Pa（T（500-700）） between previous 24 h forecasts from the ensemble members and the observations） is discussed and tested with the 5 rainstorm precipitation process in Inner Mongolia in August, 2014 and were compared with the ensemble averaging method,the results are as follows：（1）In the short 24 hours forecast, Forecast effect is improved when the number of members of the ensemble will be improved to 3, and better than the ensemble average method,when the number of members increases to more（such as 4 or 5） or reduced to 2, the prediction effect is poor. Therefore, it is very important to study the influence of the number of members in the ensemble prediction on the prediction results.（2） The effect of a small amount of precipitation forecast is better than precipitation forecast of heavy rainfall whether the ensemble average method or the optimization method, Therefore, it is necessary to study the better ensemble forecasting method, which makes the application of ensemble forecast in the disastrous weather forecast has been further developed.