Three-dimensional variational assimilation method for satellite observation data considering channel correlation

Abstract

The invention provides a three-dimensional variational assimilation method for satellite observation data considering channel correlation. The method comprises the following steps: 1) diagnosing channel correlation and an error standard deviation of satellite observation errors by using posterior information; 2) performing diagonalization decomposition on satellite observation error covariance matrix block; 3) inversing the satellite observation error covariance matrix block; 4) constructing a three-dimensional variational assimilation cost function considering channel correlation of the satellite observation data; 5) solving the constructed cost function and a gradient thereof by using a conjugate gradient minimization algorithm, and obtaining an optimal analysis field; and 6) testing theconvergence of the constructed three-dimensional variational assimilation method. According to the method, the channel correlation and error standard deviation of the satellite observation data are diagnosed by using the posterior information, an observation error covariance matrix in three-dimensional variational assimilation is constructed, and a matrix inversion method is designed, thereby solving the problem that the correlation between satellite observation data channels is not considered in current three-dimensional variational assimilation.

Description

technical field [0001] The invention belongs to the technical field of numerical weather prediction, and in particular relates to a three-dimensional variation assimilation method of satellite observation data considering channel correlation. Background technique [0002] Numerical weather prediction is a typical initial boundary value problem of differential equations. Improving the accuracy of numerical weather prediction requires improving the forecast model on the one hand, and improving the initial conditions of the model on the other hand. With the improvement of the numerical prediction model, the accuracy of the initial field of the model has become a key factor affecting the quality of the numerical prediction. In numerical weather prediction, the dimension of the atmospheric state variables in the governing equations exceeds 10 7 , and the number of observation samples of the atmospheric state in each 6-hour interval is 10 6 . Data assimilation improves the ini...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a three-dimensional variational assimilation method of satellite observation data considering channel correlation, so as to solve the problem that it is difficult to consider the channel correlation of satellite observation data in the satellite observation error covariance matrix in the current data assimilation model

Images