Exploration Geophysics - Volume 34, Issue 4, 2003

There are a variety of noise sources that affect the accuracy of Induced Polarization (IP) data, e.g., cultural noise, varying self- potential, lightning strikes, telluric currents, and electromagnetic (EM) coupling. In this paper, we describe a new IP data collection system that collects full time-series data rather than just a single averaged windowed decay. The recording of the full time series allows for greater flexibility in analysing the data so that noise can be identified and removed.

This paper outlines the process used to collect the time-series data, illustrates some of the many types of noise observed in the field, and demonstrates the superiority of the resulting IP data.

The advent of 3D inversion packages for Resistivity and Induced Polarization has meant that geophysicists are no longer constrained by survey arrays designed to produce data to be plotted manually and interpreted by eye. 3D inversion processing means that there is no longer a need to place receiver and transmitter electrodes in a co-linear array. Electrode arrays can now be designed to optimise target definition and data collection efficiency.

The double offset pole–dipole array offers a way to collect large amounts of data efficiently and has superior inversion sensitivity and depth of investigation to standard arrays.

The downhole magnetometric resistivity (DHMMR) method relies on the measurement of low-amplitude (~100 pT), low- frequency (1–4 Hz) magnetic fields arising from galvanic current flow in the Earth. Normal magnetic sensors have been used extensively in surface magnetometric resistivity (MMR) surveys, but equivalent B-field probes have not been generally available for use in standard mineral exploration drillholes. However, recent advances in magnetic sensor design have led to the manufacture of B-field probes, which provide significant advantages over the alternative TEM probes for DHMMR surveys. Synthetic and laboratory examples highlight the differences between data acquired from B-field and TEM probes. Field examples from the Flying Doctor Prospect, near Broken Hill, New South Wales, demonstrate the improved resolution achieved by using a B-field probe for DHMMR surveys.

We have developed a least-squares approach to determine, successively, the depth, index parameter, and amplitude coefficient of a buried thin dyke, using moving-average residual anomalies obtained from magnetic data using filters of successive graticule spacings. By defining the moving-average residual anomaly value at the origin on the profile, the problem of depth determination is transformed into the problem of solving a nonlinear equation, f(z) = 0. Knowing the depth and applying the least-squares method, the index parameter is determined by solving a nonlinear equation of the form λ(θ) = 0. Finally, knowing the depth and the index parameter, the amplitude coefficient is determined in a least- squares sense using a simple linear equation. In this way, the depth, index parameter, and amplitude coefficient are determined individually from all observed magnetic data. We have developed a procedure for automated interpretation of magnetic anomalies attributable to thin dykes. We apply the method to synthetic data with random errors, complicated regionals, and interference from neighbouring magnetic rocks, and we test it on two field examples from Brazil and Canada.

A number of techniques for downward continuation of potential field data, some already established in the literature and some novel, are tested and compared on synthetic and measured potential field profiles. A combination of Wiener filtering and Translation-Invariant denoising gives best results on synthetic data with added white noise. A Multiscale Edge Transform followed by a mild low-pass filter, and the ISVD method, prove to be the two most stable and robust approaches on measured data.

As mineral exploration activity shifts to regions at low magnetic latitudes, interpretation skills acquired at high latitudes become harder to apply, thus leading to under-utilisation of expensive magnetic survey data.

Changes in anomaly shape, reduction in overall amplitude, and changes in map textures make the ready interpretation of geology from magnetic data difficult. These problems are worst for magnetic inclinations within 20° of the equator.

Reduction to pole (RTP) is the best theoretical solution because it normalizes the effect of induced magnetization and strike on the shape of the magnetic anomaly while preserving dip and textural information (the “normal” magnetization and ambient field direction being the vertical). However, in practice, the standard RTP transform is difficult to apply at very low latitudes and produces variable-quality maps, sometimes dominated by declination-parallel artefacts. Additionally, the transform cannot completely reconstruct north-south trending anomalies.

The 3D Analytic Signal is a function of magnetic gradients and is easy to compute at all latitudes. The magnitude or amplitude of the 3D Analytic Signal can be computed easily and accurately for any ambient and source magnetization and can be readily imaged. The magnitude is almost, but not entirely, independent of magnetization direction. However, it lacks the resolution that horizontal and vertical gradient maps provide, and lacks the dip (and therefore structural) and textural information that total magnetic intensity and RTP maps contain.

Although neither is the perfect solution, both transformations are of use, particularly at low latitudes. In this paper, I illustrate their differences, advantages, and disadvantages.