Remote reference magnetotelluric processing algorithm based on magnetic field correlation

Abstract

Valid interpretations require precise and accurate determination of magnetotelluric impedance. Although remote reference magnetotellurics has been extensively investigated, majority of these studies have focused on the non-correlation between noise and signal or between the noise in the base station and that in the reference station. Few works have explored the correlation between magnetic signals in the base station and in the reference station. This study analyzes the effects of remote reference magnetotellurics on the sounding curve under different noise intensities in the base station. Results showed that regular remote reference magnetotellurics induce a limited quality-improving effect on the sounding curve and fail to satisfy the further data processing requirements at a low signal-to-noise ratio (SNR), suggesting that regular remote reference magnetotelluric methods cannot obtain an accurate transfer function under a low SNR for a time series. Comparison of various magnetic field data revealed that a strong correlation exists among magnetic signals 60 km apart at the Longmenshan area. Thus, the remote reference magnetotelluric method based on the magnetic field correlation between the base and reference stations is proposed to screen the power spectrum and undo the noise. The effectiveness and correctness of the proposed method are validated by the results of the theoretical and field data processing and of the intermediate data analysis, further proving that the remote reference magnetotelluric method based on magnetic field correlation is superior to the regular remote reference magnetotelluric method.