摘要
基于TIGGE集合预报资料分析评估了欧洲中期天气预报中心(ECMWF)、日本气象厅(JMA)、美国国家环境预报中心(NCEP)、中国气象局(CMA)4个模式系统在湖南2008年低温雨雪冰冻天气过程中的气温预报技巧,并对湖南地面气温和欧亚地区500 hPa天气形势进行了超级集合预报试验。结果表明,在湖南地区,ECMWF的预报效果最好,CMA的预报效果最差,并且ECMWF的168 h预报误差小于CMA的24 h预报误差。滑动训练期超级集合预报误差比较稳定,预报效果优于最好的单中心模式和固定训练期超级集合预报。对于24～72 h预报时效滑动窗口可选取50 d左右,而对于96～168 h预报时效的滑动窗口有必要选取2个月以上。此外,滑动训练期超级集合预报各时效对500 hPa天气形势的预报技巧都比单中心的预报技巧高,并且和实况资料相比,其预报效果也比较好。
Based on the TIGGE ensemble forecast data, the forecast skills of the surface temperature forecasts in Hunan province from the ensemble forecast systems of the European Centre for Medium-Range Forecasts （ ECMWF）, Japan Meteoro- logical Agency （ JMA ） , US National Centers of Environment Prediction （ NCEP ） and China Meteorological Administration （CMA） have been analyzed and evaluated, and a multimodel superensemble forecast experiment of the surface temperature in Hunan province and the 500 hPa geopotential height over Eurasia have been conducted. The results show that the root-mean- square error of the ECMWF forecast for the surface temperature in Hunan province is the smallest of the four forecast systems, while the CMA forecast system has the largest forecast error. In addition, the forecast error of the ECMWF 168 h tbrecast is smaller than that of the CMA 24 h forecast. The superensemble forecast error with the sliding training period varies little with time, and its forecast skill is superior to that of each single model and of the superensemble forecast with the fixed training peri- od. The optimal length of the sliding training period for 24 - 72 h forecast is about 50 days, while it becomes more than 2 months for the 96 - 168 h forecast. In addition, the multimodel superensemble forecast skill of the 500 hPa geopotential height for different leading time is higher than that of each single model, and it is reasonably good compared with the observed data.
出处
《广东气象》
2013年第2期21-26,共6页
Guangdong Meteorology
基金
公益性行业(气象)科研专项-冬季降水相态研究-南方区域(GYHY201006010-5)
公益性行业(气象)科研专项"面向TIGGE的集合预报关键应用技术研究"(GYHY(QX)2007-6-1)
江苏高校优势学科建设工程资助项目(PAPD)
关键词
天气学
超级集合预报
滑动训练期
地面气温预报
湖南
multimodel superensemble forecast
sliding training period
surface temperature forecast
Hunan province
