Imaging of Subsurface Conductors using VLF Electromagnetic Data- A Development and Comparison

Abstract:

We demonstrate a new fast imaging technique for very-low-frequency (VLF) electromagnetic data measured on profiles into a current density distribution in the subsurface. An analytical expression has been derived to compute vertical component of magnetic field for a given current density distribution in the subsurface. Since real anomaly of VLF measurement is proportional to the vertical component of magnetic field, an optimization approach is developed based on smoothness constrained least square inversion with a spatially variable regularization parameter to obtain a stable solution. This yields better depth information compared to filtering techniques used to interpret VLF data. Synthetic data from a dyke type structures containing conductor was modeled to show that the current densrity distributions coincide well with the horizontal position of the conductors and images of the depth distribution of the conductors. The imaging results from the new formulation are compared with other imaging techniques in terms of apparent current density and exact resistivity distribution using numerical technique. A comparison with models obtained from regularized inversion of the same data shows good correspondence with present approach. The VLF field data from an area to investigate possible interconnection of fracture zones with very high fluoride contamination in ground water near hot spring are modeled into current density distributions show significant improvement using present approach.