NCEP_ECMWF及CMC全球集合预报业务系统发展综述

NCEP_ECMWF及CMC全球集合预报业务系统发展综述20116Transactions of Atmospheric SciencesJun, 2011

麻巨慧，朱跃建，王盘兴，等, 2011, NCEP、ECMWF及 CMC 全球集合预报业务系统发展综述,J,, 大气科学学报，34( 3) :37 0-380,

Ma Ju-hui，Zhu Yue-jian，Wang Pan-xing，et al, 2011, A Review on the edvelopments of NCEP，ECMWF andCMC global ensemble forecats system ,J,,Tra ns AtmoSsc i，34( 3) :37 0-380,

NCEP、ECMWF CMC 及全球集合预报业务系统发展综述

1,2,3 2 1 1,,, 麻巨慧朱跃建王盘兴段明铿

( 1, ，210044;南京信息工程大学气象灾害省部共建教育部重点实验室江苏南京

2, Envronmenta M odeng Center / NCEP / NOAA，CampS prngs，M D 20746 USA; 3, UCAR，Bouder，CO 80307 USA) illiil

( global ensemble forecast system，GEFS——) —:3 摘要总结了目前最具代表性的个全球集合预报系统美

( National Centers foErn vironmental Prediction，NCEP) 、国国家环境预报中心欧洲中期天气预报中心( European Centre Mfoedr ium-Range Weather Forecasts，ECMWF) ( Canadian Meteoro- 和加拿大气象中心

logical Centre，CMC) 。，建成至今的发展概况由于计算资源的不断扩展各中心集合预报系统的模式分

、。，辨率集合成员数也随之增加同时各中心都在不断地致力于发展和完善初始和模式扰动方法来更

，。好地估计与初值和模式有关的不确定性促进预报技巧的提高其中初始扰动方法从最初的奇异向

( ECMWF) 、( NCEP) ( CMC) —量法增殖向量法和观测扰动法更新为现在的集合资料同化奇异向量法 ( ECMWF)、 ( NCEP) ( CMC)。，重新尺度化集合转换法和集合卡尔曼滤波在估计模式不确定性方面ECMWF CMC ，NCEP 和都修订了各自

的随机参数化方案和多参数化方案最近也在模式中加入了随

。，TIGGE ( the THORPEX interactive grand 机全倾向扰动为提高全球高影响天气预报的准确率计划

global ensemble) 、，的提出增进了国际间对多模式多中心集合预报的

合作研究北美集合预报系统 ( North American ensemble forecast system，NAEFS) ，为建立全球多模式集合预报系统提供了业务框架

。这都将有助于未来全球交互式业务预报系统的构建

:; ; ; 关键词全球集合预报系统初值不确定性模式不确定性多模式和多中心

集合预报

:P456. 7 :A:1674-709(72011)03-0370-11中图分类号文献标识码文章编号

A Review on the developments of NCEP，ECMWF and CMC

global ensemble forecast system 1，2，3 2 1 1MA Ju-hu，ZHU ue-an，WANG Pan-xng，DUAN Mng-keng iYjiii ( 1, Key Laboratory Moeft eorological Disaster of Ministry of Education，NUIST，Nanjing 210044，China;

2, Envronmenta Modeng Center / NCEP / NOAA，CampS prngs，MD 20746，USA; 3, UCAR，Bouder，CO 80307，USA) illiil

Abstract: The papers ummarizes the developments of theN ational Centers forE nvironmental Prediction( NCEP) ，the European Centre Mfored ium-Range WeatherF orecasts( ECMWF) ，and the Canadian Mete- orological

Centre( CMC，)w hich are the mosretp resentative of global ensemble forecast system( GEFS,D) ue to theen larging of computational

resources，themo del resolution and ensemble size of their GEFS subsequently increase, At the sameti me，for promoting the improvement

of the ofrecast skill，they all de- vote to edvelop the initial and model perturbation methods used stiom ulate the

effect of initial and model uncertainties, The initial perturbation methods are updated from the singular vector ( SV ) method(E CMWF ) ，

the rbeeding method N( CEP ) and the perturbed observation ( PO ) method ( CMC ) to then seemble of data ssaimilation and singular vector

( EDA-SV ) method ( ECMWF ) ，

:2010-10-16;:2011-01-19收稿日期改回日期

基金项目:NOAA Hurrcane Forecastm provement ProgramH( FP)iII

作者简介:麻巨慧( 1983—) ，女，黑龙江哈尔滨人，博士生，研究方向为集

合预报，Juhui, Ma@n oaa, gov,

the ensemble transform itwh rescaling( ETR) method ( NCEP) and thee nsemble Kalman filter ( EnKF)method( CMC,)S eve ral attempts areal so made to account mforod el-related uncertainty, ECMWF and CMC have

revised their stochastic physics parameterization tendencies( ECMWF) andm ulti-parameteriza- tion( CMC) schemes，and NCEalPs o develops stochastic total tendency eprturbation to estimate the mod- e-reated uncertanty, To acceerate mprovements on the accuracy g ofob a hgh-mpact weather orfe- llilillii