基于小波变换模极大值法和阈值法的CSAMT静态校正

采用小波分析方法进行CSAMT静态校正时,传统方法对视电阻率数据进行多尺度小波分解后,将所有尺度的细节系数设置为零,然后进行重构获得校正后的视电阻率.这使得在压制静态效应的同时,会损失一部分大构造异常的信息.针对传统小波分析方法在CSAMT静态校正中存在的问题,本文提出了基于小波变换模极大值法和阈值法相结合的静态效应校正方法.先对视电阻率数据进行多尺度小波变换,得到每一尺度上的模极大值,然后计算李氏指数进行静态效应的判断,通过设置合理的阈值,进行静态效应的压制.借助于正演模型和实测数据,验证了本方法的有效性.结果证明,本文提出的方法在压制静态效应的基础上,能够最大限度地保留大构造异常的信息.

CSAMT static correction method based on wavelet transform modulus maxima and thresholds

When the wavelet analysis method is used in CSAMT static effect correction,traditionally the detail coefficients of each scale are set to zero after multi-scale wavelet decomposition for apparent resistivity data. Then corrected apparent resistivity is obtained through reconstruction. But this method can lose some information of the large structural anomaly while suppressing the static effect. To solve this problem,this paper proposes a method that applies multi-scale wavelet transform modulus maxima and thresholds for CSAMT static correction. Firstly,through multi-scale wavelet transformation of apparent resistivity data,the modulus maximum of each scale is obtained and Lipschitz exponents are calculated to judge the static effect. Then reasonable threshold is set for every scale and the Mallat alternating projection method is used to obtain the reconstructed resistivity data after static correction. The effectiveness of this method is verified by using forward modeling and field data. The method is able to suppress the static effect and to keep the information of large structural anomalies to the utmost at the same time.