Correlation imaging method based on local wavenumber for interpreting magnetic data

Depth estimation is a general task in the interpretation of magnetic data, and local wavenumber is an effective tool to accomplish this task, but this method requires the structural index of causative source when applies it to compute the depth of the source, which is hard to obtain for an unknown area. In this paper, we suggested a correlation imaging method to interpret magnetic data, which uses the correlation coefficient of local wavenumber of real magnetic data and transformative local wavenumber of synthetic magnetic data generated by assumed source to estimate the location of the source, and this method does not require any priori information of the source and does not require solving any matrix. The computation steps as follows, first, we assume that the causative sources are distributed regularly as a rectangular grid, and then separately compute the correlation coefficient of the local wavenumber of real data and local wavenumber of the anomaly generated by each assumed source, and the correlation coefficient gets maximum when the location parameters of the assumed source are in accord with the true locations of real sources. The synthetic tested results show that this method can obtain the location of magnetic source effectively and correctly, and is insensitive to magnetization direction and noise. This method is also applied to measured magnetic data, and get the location parameters of the source.