联合资料同化理论及其研究进展  

The Theory and Research Progress of Hybrid Data Assimilation

作　　者：李崇银[1,2] 黄红艳[1] 陈超辉[1] 桂发银[1] 

机构地区：[1]解放军理工大学气象海洋学院,江苏南京211101 [2]中国科学院大气物理研究所,北京100029

出　　处：《测绘科学技术学报》2013年第4期343-348,共6页Journal of Geomatics Science and Technology

基　　金：国家自然科学基金项目(41205073;41275099)

摘　　要：联合资料同化是在"集合预报与资料同化一体化"的思想框架下研究发展的一种数值预报新理论和新方法。它在变分同化系统框架的基础上,充分利用集合预报结果,构造"流依赖"的背景误差协方差,提高对不同尺度天气系统背景误差协方差的刻画能力,进而改善分析和预报的质量。目前该方法已引起国际各主要气象业务中心广泛关注,极具业务应用潜力。本文以联合ETKF-3D-Var为例简要介绍了联合资料同化理论及其研究进展,也说明了联合资料同化的应用前景。The hybrid data assimilation scheme comes from the idea that couples ensemble forecast and variational data assimilation systems. It has emerged as a new theory or a new approach to research on the numerical weather forecast. The new system has a flow-dependent background-error covariance which is based on the variational data assimilation framework and making full use of the products of ensemble forecast. The flow-dependent uncertainties considered in the hybrid data assimilation can enhance the ability to identify the various scales of background-er- ror, thus improving the performance of the analysis and forecast. At present, the method has received many mete- orological operation centers around the world more and more attentions. It seems that the hybrid data assimilation scheme has a great potential for operational applications. In this paper, a hybrid ETKF-3D-Var analysis scheme is described as an example to explain the hybrid data assimilation theory. The progress of the hybrid data assimilation scheme is briefly reviewed and the possible developments are pointed out as well.

关 键 词：联合资料同化 集合卡尔曼滤波变换 三维 四维变分同化 集合预报 

分 类 号：P40[天文地球—大气科学及气象学]