GRAPES区域集合预报条件性台风涡旋重定位方法研究

The study on the method of conditional typhoon vortex relocation for GRAPES regional ensemble prediction

为了在集合预报中更合理描述台风涡旋中心定位的不确定性,采用2009-2018年中国气象局和日本气象厅台风最佳路径数据,分析台风最佳路径涡旋中心定位的不确定性特征,在此基础上设计条件性台风涡旋重定位方法(Conditional Typhoon Vortex Relocation,CTVR),构建集合成员台风涡旋中心重定位阈值条件、台风涡旋分离数学处理及涡旋重定位等数学处理过程,利用中国气象局数值预报中心区域集合预报系统(Global/Regional Assimilation and Prediciton System-Regional Ensemble System,GRAPES-REPS)对2018年西北太平洋上的3个台风(1808号"玛莉亚"、1824号"谭美"和1825号"康妮")进行轴对称结构和轴对称+非对称结构条件性台风涡旋重定位两种方案的集合预报试验和检验评估。结果表明:(1)中国气象局和日本气象厅台风最佳路径误差平均值为13.72 km,可视为台风涡旋中心定位不确定性的合理估计值;(2)统计检验结果和典型个例分析表明,采用轴对称结构和轴对称+非对称结构条件性台风涡旋重定位方法的台风集合预报路径误差及集合预报一致性结果比较接近;(3)条件性台风涡旋重定位方法可以有效改进GRAPES-REPS区域集合预报台风路径概率预报效果,如台风路径集合预报平均误差有所减小,集合预报一致性(路径离散度与路径均方根误差比值)增大,特别是预报初期概率预报效果改进更为显著,而预报中后期改进有限;(4)通过对"玛莉亚"台风集合预报诊断分析发现,经过条件性台风涡旋重定位后,各集合成员的台风路径误差在预报初期明显减小且路径收敛,但随着预报时效的延长台风路径逐渐发散。应用条件性台风涡旋重定位方法后,台风涡旋环流与大尺度环境场仍然比较连续协调,且台风涡旋环流外的大尺度环境场具有一致性特点,最低气压误差、最大风速误差和降水预报技巧基本不变。可见,条件性台风涡旋重定位方法的应用可以提供更准确的台风路径预报不确定性信息,帮助预报员做出更准确的预报决策。

In order to describe the uncertainty of the location of typhoon vortex center more reasonably in ensemble forecast,the uncertainty characteristics of the location of typhoon vortex center in the best track of typhoon were analyzed by using the data of the China Meteorological Administration and Japan Meteorological Agency from 2009 to 2018.On this basis,the method of Conditional Typhoon Vortex Relocation(CTVR)is designed,and the threshold conditions of Typhoon Vortex Center Relocation,the mathematical processing of typhoon vortex separation and the mathematical processing of typhoon vortex relocation are constructed.Based on the Global/Regional Assimilation and Prediction System-Regional Ensemble System(GRAPES-REPS)of Numerical Weather Prediction Center of China Meteorological Administration,the ensemble forecast experiments and test of two schemes of axisymmetric structure and axisymmetric plus asymmetric structure Conditional Typhoon Vortex Relocation for three typhoons(Maria(1808),Tarmi(1824)and Kong(1825))over the Northwest Pacific Ocean in 2018 were carried out.The results show that:(1)The mean error of the observational track of typhoon between China Meteorological Administration and Japan Meteorological Agency is 13.72 km,which can be regarded as a reasonable estimate of the uncertainty of typhoon vortex center location.(2)Statistical test results and typical case analysis show that typhoon ensemble forecast track error and ensemble forecast consistency using axisymmetric structure and axisymmetric plus asymmetric structure CTVR are comparatively close.(3)The method of CTVR can effectively improve the probabilistic prediction effect of typhoon track in GRAPES-REPS.For example,the mean error of ensemble forecast of typhoon track is reduced,and the consistency of ensemble forecast(the ratio of track spread to Root Mean Square Error of track)is increased,especially in the initial stage,the improvement of probabilistic prediction effect is more remarkable,while the improvement in the middle and later stage is limited.(4)Through the diagnosis and analysis of Maria typhoon ensemble forecast,it is found that after the CTVR,the typhoon track error of each member obviously decreases and the track converges in the early stage of forecast,but with the increase of forecast time,the typhoon track gradually diverges.After using the method of CTVR,the typhoon vortex circulation and the large-scale environmental field are still in continuous coordination,and the large-scale environmental field outside the typhoon vortex circulation has the characteristics of consistency.In addition,the error of minimum atmospheric pressure,the error of maximum wind speed and the skill of precipitation forecast are basically unchanged.It can be seen that the application of CTVR method can provide more accurate information of the uncertainty of typhoon track prediction and help forecasters to make more accurate prediction decisions.