用典型相关分析预测福建前汛期降水  

Rainfall Forecast in Rainy Season in Fujian Province Using Canonical Correlation Analysis

作　　者：吴滨[1] 蔡学湛[2] 

机构地区：[1]南京信息工程大学 [2]福建省气象台,福州350001

出　　处：《气象科技》2005年第1期32-36,共5页Meteorological Science and Technology

基　　金：福建省自然科学基金项目"福建省汛期降水量场短期气候预测模式的研究" (D9810 0 10 );国家科技部公益性专项资金项目"福建省灾害气候短期气候预测业务服务系统"共同资助

摘　　要：应用EOF-CCA方法对福建省前汛期 (5～ 6月 )降水量场进行试预测 ,预测因子选取关键区海温场及 13项物理量场 ,降水量场为 35个地面气象站降水资料。采用 3种方案进行试预测 ,分别以关键区海温场为因子 ,13项物理量场为因子 ,或两者相结合。通过 5年的独立样本预测检验表明 ,3种方案对福建省前汛期降水量场进行短期气候预测均有一定的预报技巧 ,预报时效最佳的时期是前一年的 11、12月分 ,以关键区海温为因子的预测效果最好 ,12月份关键区海温结合 13项物理量场为因子预测稳定性最好 ,可用于日常预测业务中。By using the EOF-CCA forecast model, a forecast experiment of rainfall in rainy season in Fujian Province was carried out. The rainfall data came from 35 weather stations. Three schemes were used in the forecasting experiment: selecting ① key areas of the sea surface temperature field, ② 13 physical parameter fields and ③ the combination of both as predict factors,respectively. The results of the forecast experiment with 5-year samples indicate that the three schemes all have certain predictive skills for short-term climatic prediction of rainfall in rainy season in Fujian Province. The forecasting periods with the best lead-time is November or December of the previous year. The scheme using key areas of the sea surface temperature field has the most satisfactory forecasting results. The combined scheme in December has the best stability. It can be used in operational prediction.

关 键 词：前汛期 物理量场 海温场 降水量 短期气候预测 降水资料 EOF 典型相关分析 月份 预报 

分 类 号：P458[天文地球—大气科学及气象学] P457.6