利用正则化方法对三维风场的变分调整

VARIATIONAL ADJUSTMENT OF 3-D WIND FIELD COMBINING WITH THE REGULARIZATION METHOD

以三维风场为例,首先分析了Sasaki的变分最佳分析处理三维风场的方法,当观测资料没有高频成分时,变分最佳分析方法能收到良好效果,但当观测资料有高频成分及随机扰动时,变分最佳分析方法往往无法抑制高频成分,然后利用正则化思想,结合变分最佳分析方法及滤波技术对三维风场采用了广义变分最佳调整,并进行了数值试验,数值试验表明该方法(广义变分最佳分析)比通常的变分最佳分析方法具有较大的优越性.

The variational analysis method （VAM） suggested by Sasaki is reinvestigated first. It turns out that when observational data contain no high frequency components, VAM is a powerful tool; when containing high frequency components, VAM often does not restrain high frequency components and appears to be less efficient. In the present paper, based on Sasaki＇s VAM, a generalized variational analysis method （GVAM） is proposed to analyze 3-D observational wind combining with the regularization method and filtering. A numerical test shows that observational wind can be both variationally adjusted and filtered, and therefore GVAM is an efficient method.