Publication 110

Electromagnetic imaging of a complex ore body: three-dimensional forward modeling, sensitivity tests and down-mine measurements

P. Queralt, A.G. Jones, and J. Ledo

Abstract

The main purpose of our research is to study the capability of audio-magnetotellurics (AMT) not only to detect but also to delineate complex conductive ore bodies within mineable depths. A detailed three-dimensional (3-D) numerical electrical resistivity model of the Bathurst No. 12 deposit (New Brunswick, Canada) has been constructed using available geological and geophysical information. This model facilitates studies of the capabilities of AMT at locating and defining mineral targets at depth, and of methods of optimizing data acquisition. Different geological and acquisition conditions were synthesized: presence of overburden, differing geometries, dimensions and positions of the ore body, and varying data sampling. The behavior of the surface 3-D electromagnetic fields is compared with that from bodies of infinite strike extent (the two-dimensional, 2-D, case). As is intuitively expected, the 3-D and 2-D AMT responses are similar at high frequencies, so 2-D modeling is both valid and sufficient. However, at low frequencies only those responses for current flow perpendicular to the body (the Transverse Magnetic mode in a 2-D case) are reasonably alike. The 2-D inversions carried out in this study show that the position and the top of the 3-D ore body are well resolved, in contrast to the bottom and the actual resistivity of the body that are both poorly resolved. In order to increase resolution at depth, and to extend mine life, AMT measurements inside the mine are suggested. Simple cases are simulated to show the responses at depth for 3-D cases, and to undertake “body-stripping” of the overlying structures as a first approach. The tests made show that the conductive structures above the measurement level can have strong influence on imaging conducting zones at depth and produce distortion effects in both apparent resistivities and phases. A “body-stripping” approach reduces these effects and can give an indication if there are conductive structures below, but the resulting image is far from the response of the model without overlying structures. Full 3-D inversion, holding known structures constant, is required.

Source

Geophysics, 72, 85-95, 2007.

Request reprint of publication 110


MTNet Home Page Alan's Home Page


Alan G Jones / 09 November 2006 / alan-at-cp.dias.ie