综合指标方法在降水分级预报中的应用

An Application of Composite Index Method to Precipitation Categorical Forecast

利用 1995年 1月～ 1997年 12月国家气象中心的 T10 6分析场资料 ,采用因子组合、相关分析等手段 ,确定与降水关系密切 ,且有、无雨分类较明确的物理量因子 ,将样本内每个因子由小到大排列后分段 ,计算各段上的降水频率 ,根据各物理量对降水贡献的大小 ,确定各因子的权重系数 ,将每个样本中的所有因子所在段上对应的降水频率加权后累加 ,得到一个综合降水频率值 ,通过历史拟合 ,确定降水 TS评分最大的降水频率值为综合预报指标 ,分别建立小雨、中雨、大雨、暴雨的短期降水综合预报指标。利用该方法进行了降水分级预报 ,并对预报效果进行了分析对比。

By using of the T106 analysis data of National Meteorological Center during January 1995 December 1997,factors grouped and correlation analysis the physical factors close correlation with precipitation and definitely differentiating precipitation were defined.Every factor in sample is arranged from low to high order, then divided in section, and calculated precipitation frequency in every section.According to every facter contribution to precipitation, the weighing coefficient of every factor was calculated.The precipitation frequency of every section of every facter of every sample are weighed, and then added.A composite precipitation frequency value is given by history sample fitting.The TS score maximum value of the precipitation is defined as composite forecast index.The composite index of short range forecast about light rain,moderate rain, heavy rain and storm rain were derived respectively.Precipitation categorical forecasts were conducted by using the method and the forecast effect was analysed.