一个简单的集合最优插值方法在太平洋高度计资料同化中的应用

摘要

采用不同季节样本设计了为期一年(2006年)的太平洋海面卫星高度计资料同化实验,并与控制实验及Evensen(2003)提出气候态样本的简单EnOI方法实验进行比较,利用日本海面融合产品MGDSST资料和Argo浮标廓线资料比较了3组实验结果的均方根误差和相关分布.结果表明:不同季节样本同化方法的主要特点是避免了人为给定参数带来的影响,其效果相对气候态样本实验较好.%By using different seasonal ensemble, we designed a simple ensemble optimal interpolation (EnOI) experiment in the Pacific to assimilate sea surface height data during 2006. At the same time, we also performed a control experiment and a climatology ensemble experiment documented by Evensen (2003). We compared the RMS errors and the correlation patterns between the three experiment results and independent reanalysis MGDSST data sets and Argo floats data sets. The comparison results show that the experiment of different season ensemble performs better than that of climatology ensemble, and it can avoid effectively using a experiential parameter.