基于WRF模式构建集合更新预报系统，利用Haar小波分解方法分析了北京“7·21”特大暴雨过程中三种初始扰动方案所构造集合扰动的多尺度特征，基于此探讨了混合初始扰动方法的可行性，并对比了三种扰动对误差的模拟能力。其中扰动方案一是由集合转换卡尔曼滤波方法（ETKF）对NCEP全球集合预报的分析扰动更新后得到，扰动方案二（DOWN）是直接由NCEP全球集合预报扰动插值到试验所设置的模式网格中得到，而扰动方案三（BLEND）则是将上述二者通过Barnes滤波进行尺度混合后得到。结果表明：各组扰动的能量均随时间增长，其中包含分析不确定性的ETKF扰动在预报中前期有较高的中小尺度能量，而DOWN扰动有较高的大尺度能量且能量增长速度明显快于ETKF，二者能量的大值区最终都向中尺度（64~128 km）部分发展，混合后的扰动（BLEND）能量在预报中前期增长速度最快，综合表现最优。从扰动成分来看，ETKF和DOWN中在预报前期可以快速增长的部分均集中在8~32 km的小尺度上，64~128 km部分的中小尺度的扰动信息增长缓慢，而256 km的中尺度信息则很快被耗散，这为如何选取合理的滤波波段构造多尺度混合扰动提供了依据。从降水预报结果来看，控制预报误差主要集中在降水的大值区，虚假初始扰动会导致预报初期产生虚假降水区；在暖区降水阶段，扰动对误差的模拟能力较弱，而在锋面降水阶段，扰动对误差的模拟能力明显提高，总体来看大尺度的误差较难模拟，三种方案中BLEND对误差的模拟能力最强；根据扰动-误差的相关分析同样验证了BLEND在误差模拟能力方面的优势；在降水预报TS评分方面，各组集合试验均优于控制试验，其中BLEND的效果略优于ETKF和DOWN。
A storm-scale ensemble was conducted by WRF model during Beijing "7·21" extreme precipitation event. Three initial perturbation methods is tested. The first one is produced by ETKF update and forecast cycle which contained analysis uncertainty. The second method (DOWN) is downscaled from NCEP global forecast perturbation, and the third one is produced by blending ETKF and DOWN using barnes filter with wavelength of 180 km (80~280 km). Results show that each perturbation energy can grow with time, in which ETKF has more medium and small scale energy due to flow-dependent analysis uncertainty and DOWN has more large scale energy in early time. BLEND has the most perturbation energy during most forecast time. Energy from each perturbation all grow to medium scale (64~128 km) and the fastest growing composition are focused on small scale at early forecast hours, while the medium scale component grow slowly. These results motivate further studies on how to choose properly wavelength to construct a blending initial perturbation. When coming to the error of precipitation, spurious perturbation may lead to small spurious precipitation in early hours. Forecast perturbations for different methods all have better performance in sampling error during front precipitation than warm area precipitation. All in all, ETKF has advantage in small scale and lead time error sample and DOWN is better at large scale in the later forecast time, BLEND has both advantages of ETKF and DOWN during the whole forecast time. The threat score also show that BLEND has the best overall performance.
[1]Bowler N E, Mylne K R. 2009. Ensemble transform Kalman filter perturbations for a regional ensemble prediction system[J]. Quart J Roy Meteor Soc, 135(640):757-766.
[2]Caron J F. 2013. Mismatching perturbations at the lateral boundaries in limited-area ensemble forecasting:a case study[J]. Mon Wea Rev, 141(1):356-374.
[3]Casati B, Ross G, Stephenson D B. 2004. A new intensity scale approach for the verification of spatial precipitation forecasts[J]. Meteor Appl, 11(2):141-154.
[4]Eckel F, Mass C F. 2005. Aspects of effective mesoscale, short-range ensemble forecasting[J]. Wea Forecasting, 20:328-350.
[5]Gao F, Childs P P, Huang X Y, et al. 2014. A relocation-based initialization scheme to improve track-forecasting of tropical cyclones[J]. Adv Atmos Sci, 31(1):27-36.
[6]Hohenegger C, LuthiD, Schar C. 2006. Predictability mysteries in cloud-resolving models[J]. Mon Wea Rev, 134:2095-2107.
[7]Hohenegger C, Schar C. 2007a. Atmospheric predictability at synoptic versus cloud-resolving scales[J]. Bull Amer Meteor Soc, 88(11):1783-1793.
[8]Hohenegger C, Schar C. 2007b. Predictability and error growth dynamics in cloud-resolving models[J]. J Atmos Sci, 64(12):4467-4478.
[9]Hohenegger C, Walser A, Langhans W, et al. 2008. Cloud-resolving ensemble simulations of the August 2005 Alpine flood[J]. Quart J Roy Meteor Soc, 134(633):889-904.
[10]Hollan M A, Ancell B C. 2015. Ensemble Mean Storm-Scale Performance in the Presence of Nonlinearity[J]. Mon Wea Rev, 143(12):5115-5133.
[11]Johnson A, Wang X G, Xue M, et al. 2014. Multiscale characteristics and evolution of perturbations for warm season convection-allowing precipitation forecast:Dependence on background flow and method of perturbation[J]. Mon Wea Rev, 142(3):1053-1073.
[12]Lorenz E N. 1969. The predictability of a flow which possesses many scales of motion[J]. Tellus, 21(3):289-307.
[13]Ma J, Zhu Y, Hou D, et al. 2014. Ensemble transform with 3D rescaling initialization method[J]. Mon Wea Rea, 142:4053-4072.
[14]Montani A, Cesari D, Marsigli C, et al. 2011. Seven years of activity in the field of mesoscale ensemble forecasting by the Cosmo-Leps system:Main achievements and open challenges[J]. Tellus, 63(3):605-624.
[15]Rodwell M J, Magnusson L, Bauer P, et al. 2013. Characteristics of occasional poor medium-range weather forecasts for Europe[J]. Bull Amer Meteor Soc, 94(9):1393-1405.
[16]Szintai B, Ihász I. 2006. The dynamical downscaling of ECMWF EPS products with the ALADIN mesoscale limited area model:preliminary evaluation[J]. Quarterly Journal of the Hungarian Meteorological Service, 110(3/4):253-277.
[17]Toth Z, Kalnay E. 1997. Ensemble forecasting at NCEP and the breeding method[J]. Mon Wea Rea, 125:3297-3319.
[18]Wang Y, Bellus M, Geleyn J F, et al. 2014. A new method for generating initial condition perturbations in a regional ensemble prediction system:Blending[J]. Mon Wea Rev, 142(5):2043-2059.
[19]Wei M, Toth Z. 2003. A new measure of ensemble performance:Perturbation versus error correlation analysis (PECA)[J]. Mon Wea Rev, 131(8):1549-1565.
[20]Zhang F, Snyder C, Rotunno R. 2003. Effects of moist convection on mesoscale predictability[J]. J Atmos Sci, 60(9):1173-1185.
[21]Zhang F, Odins A M, Nielsen-Gammon J W. 2006. Mesoscale predictability of an extreme warm-season precipitation event[J]. Wea Forecasting, 21:149-166.
[22]Gao Feng, Min Jinzhong, Kong Fanyou. 2010. Experiment of the storm-scale ensemble forecast based on breeding of growing mode[J]. Plateau Meteor, 29(2):429-436.<br/>高峰, 闵锦忠, 孔凡铀. 2010.基于增长模繁殖法的风暴尺度集合预报试验[J].高原气象, 29(2):429-436.
[23]Min Jinzhong, Gao Feng, Kong Fanyou. 2009. The Dynamics of Error Growth and Propagation in Storm-Scale System[C]//. The 7th Conference ofNnational Dynamic Meteorology, Jiangdezhen, Jiangxi Province.<br/>闵锦忠, 高峰, 孔凡铀. 2009. 风暴尺度系统中初始误差增长和传播的动力机制分析[C]//第七次全国动力气象会议, 江西景德镇.
[24]Sun Jianhua, Zhao Sixiong, Fu Shenming, et al. 2013. Multi-scale characteristics of record heavy rainfall over Beijing area on July 21, 2012[J]. Chinese J Atmos Sci, 37(3):703-718.<br/>孙建华, 赵思雄, 傅慎明, 等. 2013. 2012年7月21日北京特大暴雨的多尺度特征[J].大气科学, 37(3):705-718.
[25]Tang Pengyu, He Hongrang, Yang Xiangrong, et al. 2015. Research and analysis of dry intrusion during Beijing '7·21' extreme torrential rain[J]. Plateau Meteor, 34(1):210-219. DOI:10. 7522/j. issn. 1000-0534. 2013. 00128.<br/>汤鹏宇, 何宏让, 阳向荣, 等. 2015.北京"7·21"特大暴雨中的干侵入分析研究[J].高原气象, 34(1):210-219.
[26]Yu Xiaoding. 2012. Investigation of Beijing extreme flooding event on 21 July 2012[J]. Meteor Mon, 38(11):1313-1329.<br/>俞小鼎. 2012. 2012年7月21日北京特大暴雨成因分析[J].气象, 38(11):1313-1329.
[27]Zhang Linna, Guo Rui, He Na, et al. 2015. Study on whether a tornado occurred of '7·21' rainstorm in Beijing[J]. Plateau Meteor, 34(4):1074-1083. DOI:10. 7522/j. issn. 1000-0534. 2014. 00025.<br/>张琳娜, 郭锐, 何娜, 等. 2015. "7·21"北京特大暴雨过程龙卷形成可能性探究[J].高原气象, 34(4):1074-1083.
[28]Zhuang Xiaoran, Min Jinzhong, Cai Yuanchen, et al. 2017. Optimal design of lateral boundary condition perturbation method in storm-scale ensemble forecast:A case study[J]. J Meteor Sci, 37(1):21-29.<br/>庄潇然, 闵锦忠, 蔡沅辰, 等. 2017.风暴尺度集合预报最优侧边界条件扰动方法设计:个例分析[J].气象科学, 37(1):21-29.
[29]Zhuang Xiaoran, Min Jinzhong, Cai Yuanchen, et al. 2016. Accounting for initial and lateral boundary condition uncertainties under different synoptic-scale forcing in convection-allowing ensemble prediction[J]. Acta Meteor Sinica, 74(2):244-258.<br/>庄潇然, 闵锦忠, 蔡沅辰, 等. 2016.不同大尺度强迫条件下考虑初始场与侧边界条件不确定性的对流尺度集合预报试验[J].气象学报, 74(2):244-258.
