海洋变分同化背景场误差协方差矩阵构造技术

Construction of background error covariance matrix in oceanic variational data assimilation

为改进背景场误差协方差矩阵构造技术,提高海洋资料变分同化效果,首先将背景场误差协方差矩阵分解为背景误差标准偏差矩阵、非平衡变量空间相关算子以及多变量间平衡算子3部分,然后利用广义扩散方程模拟空间相关算子,联合引入温盐约束关系、线性状态方程、海表高度诊断方程、静力平衡、地转平衡关系模拟多变量平衡,最后给出了一种适用于海洋资料变分同化的背景场误差协方差矩阵构造方案。单点海表温度观测模拟同化试验表明,该方案可实现观测信息在三维空间、多变量间的传播,充分体现了背景场误差协方差矩阵在同化中的作用。

To improve the construction technology of error covariance matrix and promote the effectiveness of variational oceanic data assimilation,ageneralized scheme for the construction of the background error covariance matrix in the ocean data assimilation was firstly presented by decomposing the matrix into three parts such as the standard error bias,the space correlation matrix for independent variables,and the multivariate balance operator.And then a realistic scheme for the constructing of the matrix was proposed by modeling the space correlation matrix using the general diffusion equation,and the multivariate balance operator using the temperature-salinity relations,the linear state equation,the sea surface height diagnose equation,the hydrostatic equation,and the geostrophic velocity balance equation.The experiment results of single SST observation show that the scheme can spread the information of single variable in single place to all variables in three dimensional space,which fully embodies the effectiveness of the matrix in variational data assimilation.