The borehole controlled‐source audiomagnetotelluric response of a three‐dimensional fracture zone

Abstract

Approximate theoretical borehole CSAMT profiles of the normalized vertical electric and magnetic fields [Formula: see text] and [Formula: see text] were computed for several models of practical interest using a 3-D MT computer program based on the method of integral equations. [Formula: see text] and [Formula: see text] are the horizontal fields measured at the surface. A conductive tabular prism in a layered and homogeneous half‐space was chosen to simulate a 3-D fracture zone composed of individual, interconnected fractures. Model parameters varied during this study were depth and dip of the tabular body, the conductivity and layering of the half‐space, the frequency of the plane‐wave source, and the separation between the borehole and target. In addition, a model composed of two horizontal prisms was investigated. Decreasing the host conductivity, the depth of the prism, or the separation between the borehole and prism increases the magnitude of the subsurface normalized vertical electric and magnetic fields. Depth to the top of a dipping prism in a half‐space can be determined from the crossover of the profile of the [Formula: see text] real component. Peak amplitudes of the [Formula: see text] profiles provide information about the location of the maximum current density within the prism, which is quite variable for the imaginary component. There is no simple relationship between a small borehole‐to‐prism separation and the separation between the antisymmetric peaks of the [Formula: see text] profiles. Without knowledge of the borehole‐to‐prism separation, the dip of a nonhorizontal prism cannot be determined accurately. However, the down‐dip side of a dipping prism is indicated by the larger peak anomaly in the real component of the profile. The normalized [Formula: see text] anomalies of the conductive prism seem more sensitive to body position and also to variation in the host resistivity than are the normalized [Formula: see text] anomalies, thus making the former parameter more susceptible to geologic noise. The resistivity of the half‐space and overburden and the frequency of the source significantly influence the amplitude and phase of the secondary electric fields. Current channeling is a significant contributor to the response of the prism within a half‐space of even moderate resistivity at frequencies of 10 to 100 Hz. The [Formula: see text] field is little influenced by horizontal layering, so the borehole profiles reflect mainly subsurface inhomogeneities. The two main advantages of this technique are that the signal is much larger than the level of natural‐field noise and data acquisition is rapid because of the high frequency of the source (10–1000 Hz). However, the borehole profiles can be significantly affected by nearby inhomogeneities and by the incident field when the borehole and sensor are not vertically aligned.