基于频率匹配法的江西省ECMWF降水预报订正研究

Correction tests of ECMWF modeling quantitative precipitation using frequency matching method in Jiangxi provice

利用2016—2019年ECMWF(European Centre for Medium-Range Weather Forecasts)降水预报资料和江西省93个国家气象站降水资料,基于频率匹配法进行降水偏差订正,采用4种方法统计降水频率对降水预报进行订正试验(不分区试验),根据江西省汛期暴雨气候特征对汛期降水进行分区订正试验,并对典型强降水个例进行分析。结果表明:频率匹配法降低了模式预报小雨的空报率和大雨、暴雨的漏报率,预报技巧改进明显。在4种降水频率统计方法中,准对称滑动平均法效果最好。分区试验对强降水的订正效果优于不分区试验,该试验对模式预报正技巧暴雨过程的订正能力大于无技巧过程。对于模式预报效果差(TS=0)、一般(0<TS<0.2)、好(TS≥0.2)的暴雨过程,分区试验改善的概率分别为40.8%、89.1%和65.3%。频率匹配分区订正后强降水面积更加接近实况,但强降水落区不能得到明显的改善。订正方法对模式预报强降水形态、位置与实况较接近的过程,效果较好。

Based on the precipitation forecast data from European Centre for Medium-Range Weather Forecasts(ECMWF)and precipitation observations at 93 meteorological stations in Jiangxi province from 2016 to 2019,we analyzed the rainfall frequency using four statistical methods and conducted the correcting test(without zone division)on precipitation forecast based on a frequency matching method.In addition,we also conducted correcting test on precipitation during flood seasons over different zones according to the climatic characteristics of a rainstorm in Jiangxi province and analyzed some typical heavy precipitation events.The results show that the frequency matching method reduces the false alarm rate of light rain and the missing alarm rate of heavy rain and rainstorm,leading to obvious improvement in the forecast technique.Among the four statistical methods for a precipitation frequency,the quasi-symmetric sliding average method performs the best.The zone-division tests make a better correction than the tests without zone division,and they have better correction capability in the modelling precipitation prediction using positive techniques than that without using techniques.For the rainstorm prediction with poor(Test Sore,TS=0),normal(0<TS<0.2),and good(TS≥0.2)forecast levels,their improvement probabilities using zone-division tests are 40.8%,89.1%,and 66.7%,respectively.The predicted rainfall areas after using zone-division correction based on the frequency matching method are closer to the observation,however,the locations of the heavy rainfall can not be improved.The correction method has a better(poor)performance for the cases with the shape and location of predicted heavy precipitation close to(having a large deviation from)the real situation.