ASEG Extended Abstracts - ASEG2003 - 16th Geophysical Conference, 2003

In-situ and laboratory measurements were used to characterise the electrical properties of the Scuddles VHMS deposit, Western Australia. A major emphasis of this investigation was to study the scale variation of electrical properties.

Massive pyrite and pyrite-chalcopyrite ores were highly conductive and the underlying intense stringer mineralisation only slightly less conductive. Some sphalerite-rich ore zones were moderately resistive as expected but others with similar grade and mineralogy were highly conductive. Laboratory current mapping and connectivity measurements, together with petrographic investigation indicate that the high conductivity Zn-rich ores at Scuddles owe their unusual properties to very small amounts of well-connected pyrrhotite.

At all measurement sites, irrespective of mineralogy, there was a clear trend of decreasing resistivity with increasing electrode spacing. At some sites resistivity variations of more than three orders of magnitude were apparent between small and large-scale measurements.

Laboratory samples from most sites also display clear bimodal resistivity distributions. The most conductive laboratory samples have properties similar to in-situ measurements at large electrode spacings while the resistive end members have properties consistent with small-scale in-situ measurements. The connectivity of the conductive phases clearly increases with increasing measurement scale.

In-situ and laboratory measurements at Scuddles demonstrate the difficulty of using traditional statistical approaches to estimate bulk electrical petrophysical properties from laboratory scale measurements.

The Derwent Estuary is a drowned river valley system that lies within a complex NNW trending structural zone called the Derwent Graben. The graben preserves a poorly understood record of sedimentation and volcanism that extends from at least the early Tertiary to the Holocene. High-resolution marine geophysical data were acquired throughout the estuary to investigate the structural and sedimentary history of the graben.

Magnetic data clearly delineate the geometry of Jurassic dolerite basement units, basement controls on the development of the Graben and the distribution of previously unrecognised Tertiary basaltic centres. Modelling suggests that in excess of 800m of Cainozoic sediments may be present in the centre of the graben adjacent to the Cascades Fault in the lower portion of the estuary.

High-resolution (boomer) seismic data has for the first time provided an insight into Tertiary, Pleistocene and Holocene sedimentation in the basin. Highly reflective Tertiary sediments are extensively disrupted by normal faulting and are overlain by a complex package of Quaternary sediments. The Pleistocene portion of the section is marked by numerous internal unconformities with deep channels cut during periods of fluvial erosion during sea level lowstands and filled by marine deposition during highstands.

An important outcome of the seismic survey was recognition of extensive zones of acoustic turbidity that correlate closely with the inferred late Pleistocene course of the Derwent River. These zones result from high concentrations of biogenic methane in the shallow marine sediments and are indicative of organic-rich anaerobic conditions. The presence of natural anaerobic zones significantly complicates interpretation of recent anthropogenic effects on the estuary.

In this paper a line of dipole-dipole data, from the Prominent Hill Deposit, will be used to show that with minimum time and effort a more suitable inversion model can be produced that improves the geophysicists interpretation. Inversions of forward models are used to support the interpretation and to help refine inversion parameters.

The use of default parameters in an induced polarisation (IP) and resistively inversion programs should be considered the first stage and not the final outcome. The initial inversion model should be assessed critically and compared with the observed data. The inversion should be re-run adjusting default parameters to produce a better inversion.

Areas of complex geology tend to produce IP responses that are complex due to the interaction of multiple sources. The recognition and interpretation of these complex features are essential in designing a drill program to successfully test the sources.

This paper is the second of two companion papers that describe recent advances in the modelling of earthquake hazard in Australia. Earthquake hazard is a function of the source, attenuation and site response models which have been described in Part 1. Part 2 describes how these models are integrated to estimate the earthquake hazard. Estimating the hazard leads to a better understanding of the risk posed by earthquakes and assists in the management of risk.

The paper concludes with an application to the Newcastle and Lake Macquarie region, NSW where the new estimates of hazard are larger than those previously predicted. The attenuation and site response models used in Newcastle and Lake Macquarie both include measures of variability. Variability is incorporated by allowing random variation about the median attenuation and site response models. The results also demonstrate that the presence of regolith increases the earthquake hazard in the Newcastle and Lake Macquarie region.

The first 2-D electrical resistivity tomography (ERT) survey on the well explored (for Nickel ore) lateritic mantle of a new Caledonian ultramafic massif shows a good fit between the geoelectrical sections and the core loggings.

Data collected along several 1 km long profiles, with 10 m electrode spacing to reach about 100 m penetration depth, are reliable and the preliminary interpretation of resistivity models indicates :

- well-defined geoelectrical layers with significant resistivity contrast,

- a very good fit of the above layers with the various units of a weathering mantle, i.e. ferricrete, saprolite and bedrock,

- the suspected geometry of horizons, deduced from core logs, is better defined by the continuous ERT imaging,

- a transverse (section to section) continuity of saprolite troughs and bedrock ridges along N140° strike, which is the main structural feature of Tiebaghi ultramafic massif.

2D Electrical Resistivity Tomography appears to be an appropriate geophysical method to investigate the structure of weathering mantles of ultramafic massifs in New Caledonia. It should become a useful complementary tool to locate favourable zones, i.e. the saprolite horizon where Ni accumulation can be found, or, at least, to economize on drillings.

In March 2002, the New South Wales Department of Mineral Resources contracted an airborne hyperspectral remote sensing survey of the Broken Hill region in New South Wales. It covered 4 000 square kilometres and involved the acquisition at a ground sampling distance of 3 m.

HyMap^tm was selected as the hyperspectral system as it is a 126 band sensor that provides a combination of high spatial resolution, spectral coverage, signal-to-noise ratio and image quality. The HyMap^tm sensor utilises four 32-element detector arrays to provide 126 spectral channels covering the 450 nm to 2 500 nm spectral range over a 512 pixel swath. The HyMap's instantaneous field of view of 3 m x 3 m, corresponded to a flight altitude of 1.4 km above ground level and an aircraft ground speed of 110 knots. The ground swath coverage corresponded to 1.6 km for each flightline resulting in a 20% overlap.

The data processing and data products were radiance calibrated, atmospheric corrected and geometric corrected. Spectral and radiometric calibration of the HyMap^tm sensor was accomplished prior to the survey and this information was used to allow the conversion of the raw DN counts to radiance values in uW/cm²/nm/sr.

Using the ENVI image processing package, preliminary lithology maps have now being generated using the Mixture Tuned Matched Filtered method. The maps identify stratigraphic units; regolith; alteration zones; sulphate; hydroxyl-bearing minerals; iron oxides; and green and dry vegetation. These maps will now be taken into the field for verification and then released to the mineral industry as a GIS set of geo-referenced mineralogical map layers, along with other geological and geophysical data

We describe a new interpretation technique in geophysical inversion that uses Box-Jenkins schema for model selection using Bayesian paradigm while model parameters are estimated using maximum a posteriori estimation procedure. We employ a regularized Gauss-Newton iterative optimization technique for selecting model parameters. Model discrimination and parameter appraisal have also been realized in the Bayesian framework. The model adequacy is tested using Bayesian information criterion (BIC) and logarithmic evidence criterion (LEC). The method is demonstrated with an application to field data.

In southwestern Australia, the prediction of land salinity and stream flow trends requires the accurate construction of groundwater surfaces. This information is essential in the determination of groundwater and solute gradients. Airborne electromagnetic (AEM) data have the potential to be a useful tool in providing a 'snapshot' of current regolith salt content and possible gradient directions. The focus of major previous research on groundwater and solute movement has been on linear breaks in electrical conductivity produced by dykes and faults. Interpreting electrical conductivity information within granitic regolith, with respect to groundwater recharge and salt mobility has, to a large extent, been ignored.

Comparisons of the granitic regolith salt content and electrical conductivity measurement at Broomehill show that two types of regolith salt profile dominate. These profiles show marked differences in mineralogy and chloride content, and interpreted hydraulic properties. Unfortunately they are not readily distinguished in inverted AEM data for the area, making predictions of regolith salt content problematic. If AEM data are to be used in quantifying salt exported from catchments, then their interpretation should proceed with an appreciation of regolith heterogeneity and models of groundwater flow.

Many helicopter-borne electromagnetic (HEM) surveys have been carried out in mountainous areas. However, the effects of topography on the HEM response have not been reported in the literature for a long time. We simulate the response to topography using a staggered-grid finite-difference method. Modelling shows that a hill produces a high-resistivity anomaly over its top and a low-resistivity anomaly over its foot (when the magnetic-field response is transformed into the apparent resistivity) and that topographic effects increase with increasing frequency. In order to reduce those effects, a simple correction procedure is presented and tested on synthetic data. Results indicate that the corrected data do not reproduce the effects of the actual resistivity structure accurately enough to permit the quantitative interpretation assuming a flat-earth model. The reason for this is that the geometrical relationship between the coil system and the subsurface structure changes. The most rigorous and accurate approach to interpreting HEM data with topographic effects is to incorporate a forward-solution scheme capable of modelling topography into inversions. A 3-D inversion method is successfully tested on synthetic data.

The minimum size of shallow horizontal structures to be detectable and resolvable with airborne electromagnetic (AEM) systems is discussed. Synthetic data were generated for the helicopter frequency-domain system RESOLVE and the fixed-wing time-domain systems TEMPEST and GEOTEM. The modelled scenarios include conductive prisms in a resistive host and resistive prisms in a conductive host. The EM responses of these models were computed for a range of prism thicknesses, side-lengths and host conductivities. Gaussian noise in amplitude comparable to actual system noise levels was added to the synthetic data followed by the derivation of conductivity-depth sections via three-layer inversion and EMFLOW conductivity-depth imaging. Where these 1-D algorithms failed to indicate the presence of the prism the data were inspected for prism anomalies in order to evaluate if 2-D or 3-D algorithms might be able to map these structures.

The results indicate that beyond a minimum prism conductance (conductivity-thickness product), prism thickness and host conductivity are much less important than side-length. For horizontal prisms to be detectable with the RESOLVE system they have to be at least 12 m wide. In order to resolve their thicknesses and conductivities with 1-D algorithms, the prisms must be at least 130 m wide. For the TEMPEST and GEOTEM system the application of 1-D algorithms enables the detection of horizontal prisms wider than 180 m and the resolution of their thicknesses and conductivities for prisms wider than 450 m. Profile inspection shows that horizontal structures as narrow as 24 m might be detectable with more advanced algorithms.

As part of a project by the Department of Water Affairs, Botswana to evaluate groundwater resources in the Boteti area in Botswana airborne EM data were acquired with the TEMPEST system. The project area is characterised by a palaeolake system bordered by elevated terrain to the south. Lower-lying areas are covered by a thick layer of Kalahari Beds with saline groundwater generally situated less than 20 m below surface. The elevated areas have Karoo sediments, constisting of sand- and mudstones, at shallow depths.

For the palaeolake terrain the target aquifers are fresh water zones associated with recharge pans above the saline water table. The correlation of AEM-derived conductivity-depth profiles with drillhole records show that the conductivity of Kalahari Beds is primarily a function of clay content, water saturation and water salinity. The shallow conductivity structure outlines several resistive zones located within Boteti River alluvium beneath the present and past river channels. These prospective zones are interpreted to have a low clay content and to be saturated with fresh water.

In the elevated terrain the target aquifers are sandstones sandwiched in-between dry alluvium and mudstone. The comparison of AEM results with drillhole data indicates that sand- and mudstones have low and high conductivities, respectively. Favourable locations for freshwater exploitation include palaeochannels incised into the mudstone and shallow fractured sandstone units.

Seismic waveform inversion involves estimation of elastic properties by iterative fitting of seismic reflection data with synthetic seismograms. Like many other geophysical techniques, the inversion is applied to common mid point gathers assuming locally one-dimensional plane layered earth model. The objective of this paper is two fold. First, we address the critical issues of robustness, stability and uncertainty in which both global and local optimization methods are developed. An adaptive regularization scheme is developed that is used both in local optimization using a truncated Gauss-Newton algorithm and a global optimization based on very fast simulated annealing (VFSA). The 'regularization weight' is analogous to 'model temperature' of VFSA. The VFSA and local algorithm are used to take advantage of the two approaches. We demonstrate that the adaptive regularization is essential to improve on stability and robustness. The algorithm is inherently parallel and we carry out computation in a cluster of personal computers where load balancing is done carefully. Second, we apply our techniques to two sets of data, one from the Gulf of Thailand and the other from offshore Oregon. We are able to obtain very clear indication of free gas zones from the impedance and Poisson's ratio maps. These are then transformed into maps of saturation and porosity using a rock-physics model for use in reservoir characterization.

Marine seismic surveys are often run in rough sea conditions. While this issue is not necessarily regarded as significant for single surveys, it is more important to quantify these effects for use in time-lapse differencing for monitoring of hydrocarbon reservoirs. Sea swell causes the vertical displacement of both the seismic source and the receivers (mainly the receivers). There is also the consideration of the ghost signals that reflect off the water surface above the source and the receivers that can be displaced relative to the primary signal. These displacements can affect the seismic signals so that there is a significant mismatch in the data from different surveys.

A synthetic survey was made on a physical modelling system over a horizontal acrylic block immersed in water in a tank. A shot record was collected using piezoelectric transducers as the source and receiver containing 36 channels. From this, 144 shot gathers were generated to form a 2-D line with two planar seismic horizons. Using this as the master dataset many sub samples were generated, each of which had varying degrees of swell statics introduced. Each sub set was differenced with the master 2-D line to generate difference 2-D sections showing the effects of swell. The statistics analysed for each sub set were the maximum positive, negative and root mean square (RMS) trace amplitude values.

Results indicate that there is a significant effect on the data with a swell of over 2-3 metres. The maximum effects of sea swell appear toward the sea wave amplitude consistent with approximately half the seismic wavelength of 100 Hz showing evidence of "notching" at this frequency. Thus a moderate swell can have a significant effect on data to be differenced in time-lapse studies.

This paper describes recent developments in the manufacturing of synthetic sandstones for a range of geophysical research applications. The artificial rocks, known as CIPS (Calcite In-situ Precipitation System) can be fabricated with systematic, controllable and reproducible variations in a single parameter, while keeping all other parameters constant. CIPS rocks have been shown to reproduce the acoustic and geomechanical response observed in natural sandstones, and provide us with the experimental capability necessary to validate theoretical and numerical modelling predictions of geophysical properties. Existing and potential research applications range from theoretical studies in seismic wave propagation, petrophysics, NMR, geomechanics and analog reservoir modelling.

Geological storage of CO2 will require accurate, high-resolution geophysical monitoring to map the subsurface flow paths and the phase state of the fluids. Existing and potential monitoring methods include seismic (both surface and borehole), electromagnetics, gravity, and well logging.

Seismic methods are expected be the main form of monitoring as it is cost effective and covers the whole area of interest. When used in a time-lapse sense, areal mapping of the injected CO2 is possible with existing technology. However, improvements in resolution are required to detect leakage and identify preferential flow paths. Multi-component seismic acquisition may be necessary to discriminate between changes in saturation and pressure within the CO2 reservoir. Passive monitoring of induced microseismic events may prove to be an effective way to evaluate the effects from the injection of CO2 on the integrity of the cap rock and provide early warning of fracturing or fault reactivation.

Supporting geophysical methods will be needed for accurate quantitative interpretation of the seismic data. The geological setting and the availability of wells and pre-existing baseline data will have a profound influence on the choice of any additional geophysical methods. Electromagnetic methods have lower resolution than seismic, but are far more sensitive to changes in fluid saturation and may detect anomalies that are too subtle to interpret from seismic data alone. Gravity surveys are relatively cheap to acquire on land, but improvements in measurement sensitivity and constrained inversion methods are required before gravity can be of value for monitoring subsurface CO2 in the supercritical state.

A detailed gravity survey has been carried out by the Geological Survey of Western Australia over seismic grids covering the Beharra Springs and Mondarra gas fields in the northern Perth Basin. In this part of the northern Perth Basin, where the near surface Tamala Limestone is present, seismic data is poorly imaged whereas major structural elements are easily recognised using gravity data.

The new gravity data reveal a major transfer zone and three distinctive zones with specific signatures, coincident with the Dongara Terrace, Beharra Springs Terrace, and Allanooka High. The gravity lineaments within these tectonic units correlate strongly with major faults interpreted from detailed seismic data. Additional features interpreted from gravity, but not identified on seismic sections, may lead to a revision of previous seismic interpretation. Furthermore, positive residual gravity anomalies correlate well with seismically mapped structural highs that coincide with known hydrocarbon fields. Other positive gravity anomalies may correspond to yet unidentified fields. Thus, gravity data provide a cost-effective tool in the search for hydrocarbons in areas with little or poor quality seismic data.