全覆盖高分辨率SST融合方法概述

A review of merging methods of all covered high resolution SST

现有各种SST观测资料均有各自的优缺点,为了得到接近实际分布的全覆盖高分辨率的SST场,就必须合理地融合所有SST观测资料。本文综述了海冰资料换算、卫星SST资料偏差校正以及各种融合方法,其中融合方法包括逐步订正法、混合分析、最优插值(Optimum Interpolation,OI)分析、三维变分同化(3-Dimensional Varia-tional Assimilation,3DVAR)分析、统计最优估计以及使用截断滤波窗的变分分析。通过总结对比各融合方法优缺点,探讨了在SST融合中需要注意以及需要进一步改进的各方面问题。

To obtain the most exact all covered high resolution SST （sea surface temperature）, all kinds of SST observations must be merged since they have advantages and disadvantages of themselves. The paper reviews the conversion of sea ice data, the bias correction of satellite SST data, and all kinds of assimilation methods for merging SST. The merging SST methods includes step-by-step revision, blended analysis, analysis using optimum interpolation （OI）, 3D variational assimilation （3DVAR）, statistically optimal estimate, and variational assimilation using filter windows. Some issues are discussed for further improvement of SST merging by comparing the advantages and disadvantages of all mentioned methods.