基于EC集合预报的渤海灾害性大风风速预报方法研究

EC ensemble forecast on severe gale speed in the Bohai Sea

利用2014—2015年渤海埕北A平台逐小时风场特征对7级以上且持续6 h以上的大风个例进行统计,共出现36次大风过程。利用欧洲中心集合预报51个预报成员分别比较最优成员和最优百分位预报产品在预报海上大风风速预报的效果,结果表明:EC集合预报对于7级以上大风有低估的效果,从优选成员来看,集合预报最优成员并不唯一确定;6种(最大值、95、90、85、80、75百分位)集合预报百分位预报的10 m风速,多数预报与实况相比是偏小的,利用历史偏差对EC集合预报的误差进行校准,反算校准后误差缩小。总之,集合灾害性大风风速的预报有低估的效果,反算后预报水平有明显提高,可利用EC集合预报最优百分位订正后的10 m风场产品应用于渤海灾害性大风的预报。

Based on the hourly wind data at Bohai A flat platform from 2014 to 2015, the numbers of gale defined as higher than 7 level and continuous 6 h are counted with a total number of 36. Using forecasting 51 forecast members of the European Center set, the effects of optimal membership and optimal percentile forecasting products to predict the effect of 10 m wind speed are compared. The results show that the set forecast has the effect of underestimation for the wind above 7, and the optimal members of the set forecast are not only determined from the preferred members. Six sets of forecast percentiles（maximum, 95, 90, 85, 80, 75 percentile）predict wind farms, and most forecasts are smaller than actual. The error of the set forecast is calibrated by the historical deviation, and the error is reduced after the inverse calculation. Therefore, the forecast of the annihilated winds has the effect of underestimation, and the forecasting level of the post-antecedent forecast is obviously improved, and the optimal wind percentage of the wind farm products can be used to forecast the severe winds in the Bohai Sea.