Mapping of buried faults using the 2D modelling of far-field controlled source radiomagnetotelluric data

Abstract

Controlled source radiomagnetotellurics (CSRMT) is a relatively new geophysical method for near-surface applications. A rectangular signal with base frequencies between 0.1 and 150 kHz is injected through a grounded electric dipole which is used as a transmitter. Electric and magnetic field components are observed at these frequencies and at their subharmonics, usually in the far-field zone so that apparent resistivities and impedance phases can be obtained in a broad frequency range between 1 and 1000 kHz. Inline or broadside configuration can be used for measurements. Similar to the controlled source audiomagnetotelluric method, tensor measurements are also possible when locating two transmitters perpendicular to each other. A scalar CSRMT survey was carried out on the buried faults in the Vuoksa region, 110 km north of St. Petersburg to test the applicability of this method to the mapping of near-surface faults. A 700 m electric dipole with base frequencies of 0.5, 11.3, 30 and 105 kHz was used as a transmitter. Smooth apparent resistivity and phase values as a function of frequency from 1 kHz to 1 MHz were observed in the far-field zone for the inline configuration at 57 stations using a station distance of 20 m. Electric fields observed in the direction of the transmitter were perpendicular to the assumed strike direction of the buried faults so that they could be associated with the TM mode. The observed apparent resistivity and phase TM mode data were interpreted using the 2D inversion algorithm, and a good data fitting could be obtained. The resistivity structure beneath the survey area (down to a depth of 80 m) could be derived and the buried faults could be mapped successfully. In addition to the CSRMT observations, a conventional radiomagnetotelluric (RMT) survey was also carried out on the same profile. An excellent correlation of the observed RMT and CSRMT transfer functions and 2D conductivity models was achieved.