Exploration Geophysics - Volume 26, Issue 2-3, 1995

Airborne transient electromagnetic (TEM) methods initially developed for the detection of deep conductive targets are continuously being upgraded to improve near-surface resolution of both vertical and lateral conductivity distributions suitable for salinity mapping. However, since it is a combination of location, proximity to the surface, as well as the mobility of saline groundwater that constitutes the problem, it is important to the hydrogeologist to map the complete subsurface structure responsible for the aquifer framework.

A QUESTEM survey was flown over the Jemalong-Wyldes Plains area, New South Wales, by World Geoscience Corporation with the help of funding by the National Soil Conservation Program. The data have been re-evaluated with respect to ground TEM, DC resistivity, GEONICS EM 34-3, shallow drill hole sampling and a comprehensive hydro-geological investigation. Apparent conductance maps were used to guide the location of specific ground surveys, and conductivity-depth sections calculated from the airborne data were used for semi-quantitative comparison with the ground methods.

Comparison with TEM and DC resistivity results showed correlation with true conductivity trends down to ~140 m. Comparison with GEONICS EM 34-3 data and electrical conductivity (EC) measurements on drill cutting samples to a depth of 10 m also showed good correlation with the near-surface features of the airborne TEM conductivity-depth section.

The significant correlation with both shallow and deep conductivity trends observed in ground surveys suggests that the presentation of complete conductivity-depth sections calculated from airborne TEM, should become routine procedure in hydrogeological investigation.

A seismic experiment was undertaken in order to investigate the effect of an observed (outcropping) fracture set on shear-wave propagation, and more specifically whether the observed fracture set induced shear-wave birefringence. The source for the experiment was a hammer swung to impact horizontally on a source block coupled to the ground by two large metal pins. Eight three-component geophones recorded arrivals up to 40m offset from the hammer source. Impacts of opposite polarity in the in-line and cross-line direction were differenced in order to maximise S-wave, and minimise P-wave energy. After vector gain, filtering and rotation of the recorded data into in-line and cross-line directions, Alford rotation was undertaken on selected windowed events in order to detect shear-wave birefringence. In-line and cross-line energy is maximised by a co-ordinate rotation to the north - south direction, parallel to the observed fracture direction, suggesting that the observed fracture direction results in an anisotropy that induces shear-wave birefringence.

ADI Services, in conjunction with the Geophysical Research Institute of the University of New England, have been utilising the TM-4 Imaging Alkali-Vapour Magnetometer to locate and validate the presence of buried ferrous items in contaminated areas. Unlike many conventional systems, the TM-4 enables the rapid collection of high resolution magnetic data. The data can be displayed as colour images and interpreted semi-automatically and/or manually. Flexibility within the TM-4 system enables irregular blocks to be surveyed without requiring the user to know the exact beginning and end coordinates of the survey lines.

Two case studies are presented which demonstrate the effectiveness of high resolution magnetic surveys for contam-nated site assessment. The first case study shows how the TM-4 system has been used for the detection of unexploded ordnance, associated ammunition components and fragmentation. The objective for the second case study was to locate one edge of a former waste burial pit which was known to contain significant quantities of ferrous waste. The contrast in the total magnetic field between the burial pit and the surrounding bedrock was easily recognised.

The qualitative interpretation of aeromagnetic data relies on amplitude and textural variations within the data to map different geological units. An interpretation is commonly based on a suite of images where the data have been subjected to various filtering operations designed to highlight different characteristics. Most of these filtering operations respond primarily to amplitude variations within the data. An alternative approach is to use filters that respond to variations in the texture, that is spatial variation of amplitude. There are a number of different methods of filtering in this manner including simple statistical transforms, methods based on grey-level co-occurrence matrices, methods using fractal measures and a method based on the textural spectrum. These methods have mainly been developed for use in the medical and cartographic fields. Preliminary trials suggest such filters can extract additional structural and stratigraphic information from aeromagnetic data which complements that derived from the standard amplitude-based filtering operations.

The Black Hill Norite is a mafic intrusion which formed around 487 Ma ago. It intruded sediments of the Kanmantoo Group and Adelaide Supergroup which were deformed and metamorphosed during the Delamerian Orogeny. The unusual aeromagnetic anomaly caused by the norite gave the first indication of the presence of natural remanent magnetisation (NRM) totally different from the present field direction.

The Black Hill Norite exhibits a strong remanent magnetisation (declination = 221.2 degrees, inclination = 7.6 degrees, and intensity =4.9 A/m) indicating that it was intruded at equatorial latitudes. The remanence is probably thermal in origin. The known amount of coarse-grained multi-domain magnetite present is sufficient to explain the high magnetic susceptibility. The stable and intense remanence is carried by fine-grained single-domain pure magnetite.

Demagnetisation and hysteresis studies suggests the presence of two populations of single-domain magnetite with nearly parallel but consistently different remanent directions. Preliminary electron probe work has confirmed the presence of randomly oriented fine-grained magnetite in the rock matrix as well as magnetite exsolved out of pyroxenes and feldspars. The bimodal NRM direction can be explained as due to secular variation associated with slightly different times of NRM acquisition or due to a shape anisotropy (caused by the orientations of the exsolved single-domain grains being confined to particular crystallographic directions within the host silicates, which in turn, show a preferential orientation, possibly indicating magma flow directions).

The 3D analytic signal map and the reduced to the pole map, both made from calculations on the total magnetic intensity data, are similar, thus demonstrating the dual application of the analytic signal in the interpretation of low latitude surveys and the interpretation of magnetic sources suspected of carrying a strong, but unknown, remanent component.

The Black Hill Norite is an excellent palaeomagnetic recorder. It provides evidence for the Early Ordovician palaeomagnetic pole position for Australia (interpreted as being near the African Bight in a standard Gondwana reconstruction) which was previously in doubt because of the absence of reliable data. The success of this study illustrates how aeromagnetic maps, by indicating the presence of remanently magnetised rock units, can be constructively used to select sampling sites for NRM studies.

2D or 3D gravity and magnetic modelling procedures usually involve comparison of the response from a few, or simple, sources with an observed profile. Common software may also allow complex source distributions typical of a complete geological section.

When several sources are involved the observed profile may be the result of complex interference and individual contributions may not be obvious. Gradients may be transformed by such interference. Much trial and error may be required to resolve or understand the source and effect interactions. This is not a simple process if only the resultant calculated from all sources is compared with the reference profile.

A source component profile represents the response due to a specific source within a model.

Assessment of critical geometry or contrast changes, crucial source segments and curve fitting parameters - and hence model revision, can be greatly accelerated if the actual responses from each part of the model are viewed in addition to the resultant of their effects. Review of the source contributions in a model leads to more satisfactory interpretations since it reveals how the observed profile may, or may not, be composed. Component profiles also demonstrate how observed and resultant gradients may be modified by interference and thus provide some salutory cautions about the use and reliability of simple procedures. Software should be modified to allow this more complex display and a declaration of how the resultant and observed curves were fitted. Most currently available modelling software either generates, stores or requests the required information.

The paper includes examples from the Mount Isa Inlier of NW Queensland to illustrate the use of source components.

Aeromagnetic data from the highly structured and metamorphosed geology of the Highland Rocks 1:250 000 map sheet of the Arunta Block of the Northern Territory is virtually impossible to interpret in its original form by virtue of complexity due to the inclination of the Earth’s magnetic field, interference between anomalies, regional effects and an extreme range of anomaly amplitudes. Only by employing a comprehensive suite of enhancement techniques is it possible to produce an interpreted geological map of the area. Although virtually no geological control exists for the area it has been possible delineate lithologic units and a complex system of faults in which thrusting appears to predominate.

A case study is presented from the Southern Cross Greenstone Belt, Western Australia, where regional scale gravity data have been collected in order to study the belt and associated granitoid bodies. Differential Global Positioning System (GPS) satellite surveying techniques have been used to coordinate the gravity observations, and automated processing techniques, such as terrain corrections based on digital elevation models (DEMs), have been used as part of the study. These approaches greatly reduce the manual effort and time needed to collect and process the gravity data.

While the use of GPS methods introduces further considerations in data reduction, such as the geoid-spheroid separation, it can provide up to a 250% increase in efficiency in the field while collecting survey data. The computer based terrain correction calculation using a DEM provides a considerable saving in the time expended on data reduction. With the introduction of such field practices and reduction methods, gravity as a semi-regional scale exploration tool is becoming more attractive.

The present study has shown that station spacings of 1 km should be considered a maximum when working on the semi-regional scale (50 - 60 km square areas), with closer station spacings (500 m maximum) more appropriate close to expected geological contacts.

Measurements of horizontal or vertical gradients of the magnetic field intensity can add a new dimension to high resolution aeromagnetic surveys of shallow basement areas. Although a number of vertical, horizontal and triaxial aeromagnetic gradiometers are in operation throughout the world, very little has been documented on the relative merits of horizontal and vertical gradiometer measurements or of measured versus calculated gradients. Our results demonstrate that there are differences in spatial resolution and bandwidth between measured and calculated gradients and that horizontal and vertical gradients provide complementary information. Finally our results indicate that better geo-referencing of horizontal gradiometer data is needed before we can derive reliable vertical gradient data from the measured longitudinal and transverse gradients

Development of short-baseline aeromagnetic gradiometers began in 1975 with construction of the Geological Survey of Canada vertical gradiometer. The success of current short-baseline gradiometers depends on the use of high sensitivity optical pumping magnetometers, usually caesium vapour sensors and effective compensation for aircraft manoeuvre noise.

Practical advantages of measured gradients include elimination of the diurnal problem and improved spatial resolution of small shallow sources. Transient signals have the same effect on all the magnetometer sensors so the effects are cancelled out in the gradients and diurnal-free total magnetic intensity can then be reconstructed by integrating the gradient data. The gradient data effectively remove the main field of the earth and enhance smaller scale, shallow anomalies while attenuating longer wavelength, deeper seated anomalies. In addition, measurement of the transverse horizontal gradient provides extra information between flight lines, leading to a reduction in flight line dependency of magnetic anomalies and a reduction in aliasing effects. The vertical gradient can also be derived from the longitudinal and transverse horizontal gradients with less line dependence and probably lower noise levels than the measured vertical gradient.

Data from a conventional Canadian vertical gradiometer system, an Australian horizontal gradiometer system and a Southern African triaxial gradiometer system are evaluated. The measured vertical gradient data appear to provide better resolution of shallow sources than the vertical gradient calculated from the total field. The horizontal gradient data allow us to derive the vertical gradient from the horizontal gradients and provide much improved enhanced total magnetic intensity grids.

The choice of a vertical or horizontal gradiometer system depends on survey objectives and both systems have their advantages. The vertical gradient appears to be marginally easier to measure and simpler to interpret. A triaxial gradiometer system provides the advantages of both at the cost of an additional sensor and retractable boom installation, compared to a horizontal gradiometer. Unfortunately, in Australia, CAA certification of a retractable tail boom may prove difficult. The viability of deriving the vertical gradient profiles from measured horizontal gradients needs more research, but if this can be done routinely, we think this would tilt the balance towards the horizontal gradiometer system. The improvement in total magnetic intensity grids using the transverse gradient information can be quite dramatic in many cases and there is always some improvement in the data. The combination of transverse gradient enhanced total magnetic intensity data and high spatial resolution ‘texture’ filters provides excellent definition of lower amplitude anomalies. In contrast, the calculated vertical magnetic gradient is often very similar to the measured data and the benefits of the measured vertical gradient may be small in many cases.

Gradiometer performance deteriorates rapidly as the source becomes deeper and existing gradiometers are probably close to the noise level at 500m depth. Rugged topography also causes fixed wing gradiometer performance to deteriorate.

Classical fast Fourier transformation methods for reduction to the pole cease to produce realistic results at low magnetic inclinations because the factor by which the frequency domain field representation must be multiplied becomes infinitely large due to the denominator of this term approaching zero. The problems of the Fourier transformation method however can be avoided by performing the transformation in the space domain by convolving the field with a set of filter coefficients which perform the desired transformation. Such filter coefficients may be calculated using the Wiener design principle which produces filter coefficients for transformations such that a known input is transformed to a desired output in a manner that the mean square error between an actual output and a desired output is minimised.

A large portion of the southern Joseph Bonaparte Gulf area, northwestern Australia, was originally underlain by extensive contiguous Proterozoic sills and volcanic rocks. Subsequent extension and basin formation in this area during the Palaeozoic fractured this horizontal basic magnetic sheet. Evidence of this fracture patten and the extensional movements are obvious in images of the magnetic field over the basin. A series of transfer fault accommodation zones trending at 035 degrees and an associated series of perpendicular normal faults have been mapped. Onshore geology, gravity and seismic data support the transfer fault model. An underexplored graben, 75 kilometres long and 50 kilometres wide, which formed as a result of the extension process, has been mapped on the western edge of the Joseph Bonaparte Gulf adjacent to the Proterozoic Kimberley Basin.

Aeromagnetic surveying carried out in the Abminga area by Mines and Energy South Australia (MESA) and published in 1993 identified several groups of isolated magnetic anomalies which have a strong prima facie resemblance to kimberlite or lamproite diatreme fields. Aeromagnetic anomalies produced by diatremes are characterised as being generally of low amplitude, isolated, compact, di-polar, of short strike length and indicative of a shallow source.

This paper compares data flown for diamond explorer Caldera Resources NL at 100 metre line spacing with the published 400 metre line spaced data. Modelling of several of the anomalies using an inversion algorithm gives very good fits to shallow pipe-like bodies with appreciable depth extent. Conventional field exploration techniques such as loaming have not been successful but floaters of ultramafic rocks collected at 3 anomaly sites have been petrographically identified as extremely weathered kimberlites. Chromite grains recovered from these carry up to 59% CrzCX Diamonds have not yet been recovered but active exploration is in progress.

It is concluded that while most of the magnetic anomalies were successfully identified from the 400 metre line spaced data, useful modelling can only be done using more detailed data. Although positive identification of the source of these magnetic anomalies is still awaited they constitute some very intriguing exploration targets.

In the last few years, the use of airborne gamma-ray spectrometry for geological mapping and mineral exploration has shown considerable growth. With this growth there has developed an increasing need to standardise the airborne measurements so that they will be independent of survey parameters. This paper describes the various calibration and processing procedures that have to be followed to achieve this goal through the conversion of the airborne measurements to ground concentrations of potassium, uranium and thorium. One of the most critical steps in this process is in the use of airborne calibration ranges whose potassium, uranium and thorium content must be measured at the time of the calibration flights. Two calibration ranges have been set up in South Australia and Queensland. However, these ranges are radioactively inhomogeneous which has made this calibration task more complicated than it otherwise would have been.

Airborne radiometric systems must be calibrated to present data devoid of artefacts peculiar to their acquisition. The IAEA(1991) model of the components of airborne radiation is outlined. This features: aircraft background radiation; cosmic-effect (related to survey altitude); Stripping Ratios which account for the spectral overlap of radiation from different radioelement sources; and height attenuation coefficients, which quantify the effect of Compton Scattering of gamma rays passing through the air. Resources of calibration pads and radiometric test range sites are listed. The facets of calibration procedure are discussed according to their application to the IAEA model. The resulting calibration parameters are used in a process to generate airborne radiometric system sensitivities by comparison of airborne and ground spectrometer data. This method can avoid errors in system sensitivity caused by environmental factors.

A new method for determining the optimum channel combinations for reducing the number of channels of data for multichannel airborne gamma-ray spectrometry is developed. The concept of ‘generalised’ channels, which are the summation of one or more 12 keV data channels that are not necessarily contiguous in the spectrum is introduced. The method uses a simulated annealing minimisation technique to find the best generalised channels such that the sum of the fractional errors for each of the 3 radioelements is minimised. All methods of summing channels lead to an increase in the fractional errors of the radioelements. But the new method is an improvement on existing methods, with the greatest improvement occurring as the number of generalised channels is decreased. A simple method of determining optimum window boundaries for conventional 3-channel airborne gamma-ray spectrometry is also demonstrated. Equivalent thorium fractional concentration errors can be reduced by broadening the conventional ²⁰⁸TI window.

High resolution digital elevation models (DEMs) can now be produced for large areas of Australia. This paper will describe the method of producing DEMs and some of the applications.

The change in technology which has made possible routine generation of height data for large areas, is soft photo-grammetry. In the past, the method of determining height was by matching points on two airphoto stereopairs. This was accomplished manually which made for a tedious and time consuming process. The advent of computer software that can match features on stereopairs has meant that DEMs can be created much more quickly.

The procedure is to scan the diapositives of the air photographs with a geometrically accurate scanner at high resolution (15-30 microns). This will give a pixel size of 1.5 metres for a 1:50,000 scale photo. Internal and external distortion is then removed from the scanned images by internal, relative and absolute orientation procedures. The images are then resampled as epipolar pairs. The final procedure is to automatically match common features on both of the stereopairs. Changes in the X parallax give the height or Z value.

The DEMs are accurate to within +/-1 pixel i.e. 1.5 metres for standard 1:50,000 scale photography or 25 cm for 1:10,000 photography. This high resolution allows for a number of novel applications, including:- identify topographic anomalies associated with mineralisation (e.g. kimberlite lows, silicified highs)

- georectify photographs to produce orthophotos

- 3D visualisation, 3D Photomaps, and terrain draping

- structural analysis using shaded relief images

- landform analysis including palaeo surface reconstruction

- slope vector maps for soil sampling and geochemical dispersion

- drainage maps and drainage divides for stream sampling programs

- height corrections for high resolution gravity surveys

Images for these applications will be shown for the Laverton 1:50,000 sheet area in Western Australia.

A large portion of the western offshore Victorian portion of the Otway Basin appears to have originally been underlain by an extensive horizontal basic magnetic sheet emplaced during the initial stages of basin formation. Subsequent basin extension along transfer faults trending at 210 degrees has fractured this sheet, and evidence of the transfer fault fracture system is given by the present day outlines of the magnetic anomalies arising from the fragments of the sheet. The transfer fault system is further indicated by linear magnetic anomalies within the sedimentary section due to magnetic material accumulated in the fault planes. Gravity and seismic data support the transfer fault model which can be used to explain features of the depositional history and structural development of the area.

The Victorian Government through the Department of Agriculture, Energy and Minerals is undertaking a 3-year, $16.5 million Victorian Initiative for Minerals and Petroleum (VIMP) to support the state’s mineral and petroleum exploration industry. Projects being undertaken as part of the initiative include the acquisition of airborne magnetic and radiometric, gravity and regional seismic data, stratigraphic drilling and geophysical interpretation and geological mapping products. Major airborne surveys are being carried out in the north west of the State, the Eastern Highlands and the Otway Basin.

Geological mapping of the highly prospective Lachlan Fold Belt in Eastern Victoria has been hampered by difficult mountainous terrain. High level regional magnetic surveys contribute little to the detailed geological understanding. As part of VIMP, the Victorian Government has undertaking systematic helicopter geophysical helicopter surveys with a line spacing of 200m over an area of 30 000 km in the Eastern Highlands of Victoria to assist the detailed geological mapping required for mineral exploration.

A case study from the Orbost area of the Eastern Highlands illustrates the advances available in geological mapping with the timely collection of detailed airborne geophysics and the implications this has for future exploration in Eastern Victoria. The Orbost magnetic and radiometric data are of a high quality and detailed interpretation enables the production of accurate lithological and structural maps in an area which has significant potential for gold and base metals.

The magnetic data differentiate between various types of volcanics in the Buchan Rift, indicate multiple intrusions, show bedding and faults in many areas within the Palaeozoic sediments, and reveal a previously unknown pervasive west-northwest to northwest fault direction. The radiometric data show compositional differences between the outcropping granites, correlate well with the magnetic data in identifying different types of volcanics in the Buchan Rift, and provide a number of lithological boundaries in areas with little magnetic contrast between the rock types.

A crosswell tomography (CWT) experiment in a hard-rock environment was carried out to image the subsurface between two 240 m wells separated by 28 m. Detonators were used as downhole sources and a 24-channel hydrophone streamer provided the downhole receivers. The receiver spacing in the streamer is 2 m. An effective 1 m receiver spacing was achieved by two deployments of the streamer with a shift of 1 m. High quality seismic data were recorded at a sampling interval of 0.1 ms.

Seismic first-break times from the experiment were used to construct the velocity tomogram between the two wells by a least squares algorithm. Large source statics were identified in many receiver gathers. These statics can have strong adverse effects on the tomographic imaging. The statics can, however, be mostly recovered and compensated for in the analysis. In a tomographic inversion, they are treated as part of the model parameters and are reconstructed simultaneously with the velocity distribution.

Tomographic images agree well with the known geological structure obtained from the two imaging wells and a third one between them. They also agree with other independent geophysical survey results (e.g. Applied Potential and TEM). The difference between the CWT and the other geophysical methods is that the CWT can provide a high-resolution 2-D image between the wells rather than simple detection and localisation of the anomaly using the other geophysical techniques. CWT can offer vital information for further exploration and may have great economic potential.