以ECMWF数值产品为基础的单站气温中短期预报

THE SINGLE-STATION SHORT-TERM TEMPERATURE PREDICTION BASED ON ECMWF NUMERICAL FORECAST PRODUCTS

采用ECMWF数值预报产品资料,首先用PP法和MOS方法建立溪洛渡水电站坝区三坪站气温预报逐步回归方程,寻找消空(漏)指标进行预报后处理,然后根据000～168时效的ECMWF数值预报产品资料制作未来24～144 h三坪站的最高、最低、平均气温预报。并研制出能自动运行的预报业务系统。2011年业务运行效果检验评估表明:系统预报效果较稳定,6 d平均、最高、最低气温平均预报准确率分别为70.8%、62.7%和76.0%,在水电气象预报服务中短期气温预报实时业务中有较好的指导作用。平均、最高、最低气温准确率随着预报时效的延长效果降低,平均气温和最低气温比最高气温的预报准确率高。预报方法对制作单站气温预报是可行的。

Using ECMWF numerical forecast products,stepwise regression equation of 24-hour to 168-hour termperature Forecast are established by means of PPM and MOS,and post-processing are Conduct through finding indicators so as to Eliminate empty forecast.Then,the future 24～144h highest,lowest,average temperature forecast at Sanping meteorological station are made according to 000 to 168 limitation of the ECMWF numerical forecast production.Furthermore,We have developed a automatically run prediction platform.The effect test operation forecast in 2011 indicated that Forecast result is stable,The rate of average forecast accuracy is 70.8%,62.7% and 76.0% for the average,highest and lowest temperature forecast,which has a good instruction for the real-time weather service at the Hydroelectric Power Station.The forecast accuracy rate of the average,highest and lowest temperature decreased with the forecast time increase,and the accuracy rate for the average,lowest temperature is higher than that of the highest temperature.So the forecasting methods is feasible for the single station temperature prediction.