利用DINEOF方法重构缺测的卫星遥感海温数据  

Missing Satellite-Based Sea Surface Temperature Data Reconstructed by DINEOF Method

作　　者：盛峥[1] 石汉青[1] 丁又专[2] 

机构地区：[1]解放军理工大学气象学院,江苏南京211101 [2]国家海洋局第二海洋研究所,浙江杭州310012

出　　处：《海洋科学进展》2009年第2期243-249,共7页Advances in Marine Science

基　　金：国家自然科学基金项目--卫星遥感资料变分反演云水廓线的研究(40675020)

摘　　要：针对海温数据补缺问题提出一种基于经验正交函数(EOF)分解的缺测数据重构方法——经验正交函数插值法(DINEOF)。此方法是一种自适应的EOF分解方法,不需要任何先验信息,并且能估算出重构数据的误差大小。奇异值(SVD)分解时采用了Lanczos算子以提高计算速度,并利用了交叉验证的算法以确定出重构时所需最优的EOF阶数。实测数据处理结果表明,DINEOF方法能很好地重构大面积的缺测数据,具有较高精度及实际可操作性。To make up deficiency of sea surface temperature （SST） data, a method used for reconstruction of missing SST data, that is the Data Interpolating Empirical Orthogonal Function （DINEOF） method, has been proposed. This method is established on the basis of and is self-consistent with the decomposition of Empirical Orthogonal Function （EOF）. It does not need any prior information and can estimate the er- ror of the reconstructed data. To improve calculation speed, Lanczos operator is applied for the singular value decomposition （SVD）. A cross-validation method is used for the determination of optimal number of EOFs needed for the data reconstruction. The results from the process of measured data have shown that the DINEOF method can be successfully applied to the reconstruction of missing data in a large area and has a relatively high precision and a practical operability.

关 键 词：海温数据补缺 EOF插值 SVD分解 Lanczos算子 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]