奇异交叉谱分析方法在夏季降水预报中的应用

利用奇异交叉谱(SCSA)分析方法，提取1951～2002年夏季(6～8月)河北省降水和北半球2～3月500hPa高度场关键区相互作用的耦合周期信号及其分量序列年际和年代际时变特征。基于SCSA，重建耦合振荡分量序列(RCCS)，并与自回归分析相结合，建立预测模型，对河北省夏季降水趋势进行预测试验。结果表明：2003年以后河北夏季降水将转入一个相对多雨阶段，若将此方案进一步完善，可望在业务中推广应用。

Application of Singular Cross-Spectrum in Analysis of Summer Precipitation in Hebei Province

With Singular Cross Spectrum (SCS) method, the summer precipitation over Hebei Province from 1951 to 2002 and the coupled periodic signals of interaction between key areas in the 500 hPa height field fron February to March in the North Hemisphere were diagnosed and analyzed, and the annual and decadal variation characteristics of the array series were obtained. Based on SCSA and RCCS (Reconstructed Coupled Oscillation Component Series), in conjunction with regression analysis, the prediction model was established, with which the precipitation prediction of summer in Hebei Province was made and the result is satisfactory.