集合资料同化方法的理论框架及其在海洋资料同化的研究展望

发布时间：2018-05-03 14:55

本文选题：资料同化 + 集合卡尔曼滤波器　；参考：《海洋学报》2016年03期

【摘要】：在海洋动力系统的数值模拟中,海洋资料同化是一种能够有效融合多源海洋观测资料和数值模式的方法。它不仅可以显著地提高数值模拟的效果,构造海洋再分析资料场,还能有效减少海洋和气候预报时模式初始条件的不确定性。因此,海洋资料同化对于海洋研究和业务化应用具有非常重要的意义。资料同化方法的研究一直是大气、海洋科学的热门课题之一。其中,集合卡尔曼滤波器(EnKF)是一种有效的资料同化方法,自提出以来经过了20多年的发展和改进,已经在海洋资料同化中得到了广泛的研究和应用。近年来,随着动力模式的不断发展和计算能力的提高,粒子滤波器由于不受模型线性和误差高斯分布假设的约束,也逐渐成为了当前资料同化方法研究的热点。本文分析和总结了目前关于集合卡尔曼滤波器和粒子滤波器的一些最新理论研究结果,在贝叶斯滤波理论的框架下讨论了这两类算法的关联和区别,以及各自在资料同化实践中的优势和不足。在此基础上,我们探讨了粒子滤波器应用于海洋模式资料同化的主要困难和目前可行的一些解决方法,展望了集合资料同化方法研究的新趋势,为集合资料同化方法的进一步发展和应用提供理论基础。
[Abstract]:In the numerical simulation of marine dynamic system, ocean data assimilation is a method which can effectively fuse multi-source ocean observation data and numerical model. It can not only significantly improve the effect of numerical simulation and construct ocean reanalysis data field, but also can effectively reduce the uncertainty of the initial conditions of the model in ocean and climate prediction. Therefore, ocean data assimilation plays an important role in marine research and application. The study of data assimilation method has been one of the hot topics in atmospheric and marine science. The ensemble Kalman filter (EKF) is an effective method for data assimilation. It has been developed and improved for more than 20 years and has been widely studied and applied in ocean data assimilation. In recent years, with the continuous development of dynamic models and the improvement of computational power, particle filters have become a hot topic in the research of data assimilation methods because they are not constrained by the assumption of linear model and the distribution of error Gao Si. In this paper, we analyze and summarize some new theoretical research results about set Kalman filter and particle filter, and discuss the correlation and difference between these two algorithms under the framework of Bayesian filtering theory. And their advantages and disadvantages in the practice of data assimilation. On this basis, we discuss the main difficulties in the application of particle filter to ocean model data assimilation and some feasible solutions, and look forward to the new trends in the research of aggregate data assimilation methods. It provides a theoretical basis for the further development and application of ensemble data assimilation methods.
【作者单位】：国家海洋局第二海洋研究所卫星海洋环境动力学国家重点实验室;北大不列颠哥伦比亚大学环境科学及工程系;
【基金】：国家自然科学基金项目(41276029;41321004) 科技部国家基础科研项目(2013CB430302) 卫星海洋环境动力学国家重点实验室自主课题(SOEDZZ1404;SOEDZZ1518)
【分类号】：P717

【相似文献】