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

P-wave and S-wave velocities, attenuation and the Vp/Vs ratio show different responses in saturated and dry Carnarvon basin sandstones under increasing differential pressures. At low differential pressures, Vp/Vs ratios are high under oil saturation but low in the same rocks under dry (gas-saturated) conditions. Velocities and attenuation are also shown to be sensitive to pore pressure, not just differential pressure. Under identical low differential pressure conditions, velocity is higher in sandstones when pore pressure is high during simulation of inflationary overpressure conditions. Thus the velocity- differential pressure relationship is not unique. However, the use of the measured Biot effective stress coefficient results in a unique relationship. It is postulated that higher velocity and different Vp/Vs ratios are caused by loss of compliance in microfractures due to stiffening by increasing fluid pressure.

The time-domain thin sheet solution derived by J. C. Maxwell more than a century ago remains an important tool for fast interpretation of airborne electromagnetic data. The solution is exact if a conductive layer is thin and surrounded by an isolator. As shown by Maxwell, the magnetic field of the currents induced by a turned-on magnetic dipole coincides with the field of an auxiliary dipole receding at a constant speed, which depends only on the layer conductance. The solution allows for an easy determination of the layer conductance from the measured magnetic field. The layer conductivity and thickness cannot be directly determined.

Maxwell's solution becomes accurate at the late stage of the decay of induced currents provided that the basement that underlies the layer is non-conductive. The moment when the thin sheet solution becomes applicable is specified by the time necessary for the receding dipole to cross the layer. On the other hand, the secondary magnetic field of the currents induced in a basement of finite resistivity eventually prevails over the magnetic field of currents in the conductive layer. Depending on the layer thickness and layer-to-basement conductivity contrast, the time applicability range of the thin sheet solution narrows down or even disappears.

We derive an asymptotic solution that accounts for the layer thickness as well as the basement resistivity. A special correction makes the solution applicable immediately after the source is turned on. The new solution is by orders of magnitude faster than a numeric solution based on successive wave number-to-space and frequency-to-time domain Fast Hankel Transforms.

The time-domain thin sheet solution derived by J. C. Maxwell more than a century ago remains an important tool for fast interpretation of airborne electromagnetic data. The solution is exact if a conductive layer is thin and surrounded by an isolator. As shown by Maxwell, the magnetic field of the currents induced by a turned-on magnetic dipole coincides with the field of an auxiliary dipole receding at a constant speed, which depends only on the layer conductance. The solution allows for an easy determination of the layer conductance from the measured magnetic field. The layer conductivity and thickness cannot be directly determined.

Maxwell's solution becomes accurate at the late stage of the decay of induced currents provided that the basement that underlies the layer is non-conductive. The moment when the thin sheet solution becomes applicable is specified by the time necessary for the receding dipole to cross the layer. On the other hand, the secondary magnetic field of the currents induced in a basement of finite resistivity eventually prevails over the magnetic field of currents in the conductive layer. Depending on the layer thickness and layer-to-basement conductivity contrast, the time applicability range of the thin sheet solution narrows down or even disappears.

We derive an asymptotic solution that accounts for the layer thickness as well as the basement resistivity. A special correction makes the solution applicable immediately after the source is turned on. The new solution is by orders of magnitude faster than a numeric solution based on successive wave number-to-space and frequency-to-time domain Fast Hankel Transforms.

The St Ives terrain is covered by a thick regolith that is saturated with hyper-saline groundwater. The regolith forms a major barrier for electrical and EM surveys designed to detect gold bearing structures in bedrock

The DHMMR method was trialed at St Ives, and did not detect recognisable responses from known structures. DHMMR does not detect anomalies because the structures have small cross sectional areas for channelling current, and the resistive bedrock does not support large current densities for channelling into targets.

DHEM surveys were trialed at St Ives and detected a strong well-defined anomaly at the Junction gold mine. The anomaly was modelled as a plate in layered earth, and the plate corresponded to a major gold bearing shear. Petrophysical tests suggest that the shear is conductive because of brine-saturated porosity.

DHEM surveys did not detect strong anomalies from a gold bearing shear at the Argo gold mine, despite the shear having low measured resistivity relative to host rock. The absence of DHEM response may be because the shear is not conductive enough to support an anomalous current system, or because the conductivity is not connected on a large scale.

Most irrigation channels in Australia are earth lined and leak water to the surrounding terrain. The cost of lost water can be high in economic and environmental costs. The solution is to completely line the canals or replace them with pipes, both expensive options. An alternative is to focus on areas of highest seepage. In this study we trialed a dipole-dipole resistivity array towed behind a dinghy in the canal. The approach followed trials using electromagnetic techniques in previous years that showed conductivity changes below and around the canals in part reflected seepage. We inverted the resistivity results to sections and statistically analysed the data from various depths in the ground by correlating the resistivity against pondage seepage results. Our conclusion was that the resistivity worked best where there was a diffuse seepage from the canal. The greatest effect was around the watertable. Where seepage rates were low there was no correlation with resistivity. The correlations improved with increasing seepage particularly where the inverted sections indicated the seepage rates were uniform along the channel.

The Adelaide Geosyncline is host to sedimentary copper deposits that led to the formation of a copper mining industry in Australia in the mid 1800's. During the 1960's, historic workings at Kapunda were revisited in the search for possible extensions to mineralisation. Due to poor outcrop, geophysics played a significant role in this program. In particular the induced polarization method was used to look for primary sulphide mineralization of the Kapunda ore. The interpretation of these data relied largely on the calculation of a metal factor. A re-interpretation of these data using inversion of the PFE has revealed further information regarding the structure of the mineralised zones, and led to the identification of several targets for copper mineralisation.

After a decade in which Kirchhoff migration has been the workhorse method for structural imaging, the industry has found that this method fails to image the most structurally challenging geology. Wavefield extrapolation based approaches have shown promise in complex structural imaging, but many of their assumed advantages have been found inadequate. In this paper, we review the progress made by these approaches, and we survey various implementations that are available. We also review several key issues that are critical if the wavefield extrapolation approach is to be useful in generating high-fidelity production-scale images. Some of these issues include: true amplitude formulation, imaging condition aliasing, sail line aliasing, and angle-gather computation. We use results from both synthetic and real data to illustrate the advantages and challenges facing wavefield extrapolation based migrations.

A high resolution gravity traverse (27.2 km-long, 50 m or 100 m station-spacing) across the unconformity-associated McArthur River uranium deposit, Athabasca Basin, Canada was conducted to investigate basement geology and locate alteration zones related to mineralization. In this geological environment, amplitudes of gravity anomalies relevant to uranium exploration (related to offsets of the basement unconformity, basement relief and alteration zones) are typically ˂ 1 mGal, but range up to 2 mGal. Alteration near the McArthur River deposit is commonly present as silicification of the basin sandstones, located above basement quartzite units. Many short to intermediate wavelength (250-1800 m) anomalies, amplitudes ˂ 0.5 mGal, are identified along the traverse profile, superposed on longer wavelength and larger amplitude (up to 6000 m, and 1.8 mGal, respectively) anomalies. Preliminary modeling of anomalies, constrained by magnetic data, drill-hole logs and density information, and reflection seismic data, indicates a variety of geological contributions. Short wavelength anomalies are attributed to changes in thickness of the glacial overburden, thickening in some places being linked to preferential erosion along an underlying fault. Long wavelength anomalies are explained by variations in basement density and offsets of the basement unconformity. Importantly, from an exploration perspective, most intermediate wavelength positive anomalies are explained by a combination of silicification within the Athabasca Sandstone, and associated basement relief produced by underlying basement quartzite units. Desilicification is also modeled, but limited. The gravity method, apparently, has been successful in identifying alteration zones that may be of significant interest for uranium exploration, because of their association with mineralizing fluids, and because patterns of silicification/desilicification are important for studies of fluid flow.

The Internet, Intranets and general globalisation of networking technology have produced a dramatic increase in the type and volume of geo-data that are available to geoscientists. The development of useful protocols and underlying technologies for computers to access and share geo-data, both privately within organizations, and globally on the Internet is key to our ability to use this information efficiently.

In this paper, we describe the Data Access Protocol (DAP), which is a suite of server applications that enable geoscientists to find and evaluate data, and automate windowing, reprojection and reformatting the data to suit a specific requirement. DAP technology addresses a variety of network situations including:

1. Simple web-browser based discovery and retrieval of data of interest in a specified format and coordinate system.
2. Support for Open GIS Consortium Web Map Server (WMS) interface to allow any WMS compatible application to retrieve “images” of the data for use as layers in a GIS application.
3. Direct support for DAP-enabled thick clients, such as Oasis montaj, to optimally retrieve data directly for their own use, and transfer data to a hosting DAP environment.

When communicating with DAP-enabled client applications, DAP addresses the movement of data (lossless compression, encryption and streaming) both to and from a data server over a network. The core DAP protocol effectively abstracts data formats to allow client applications to work in whatever environment is required, and DAP servers to connect to data in whatever native format is in use by a hosting organization. This makes DAP suitable for use in many data storage environments. DAP also includes a number of spatially optimised data stores that can be used to deliver extremely high performance for data extraction and retrieval.

The most commonly cited mechanisms of overpressure generation are burial-driven disequilibrium compaction and hydrocarbon generation. However, it is unlikely that these mechanisms generated the overpressure in the Cooper Basin. The last significant burial event in the Cooper Basin was the Cretaceous deposition of the Winton Formation (-90 Ma). Maximum temperature was attained in the Cretaceous, with cooling beginning prior to 75 Ma. Hence, overpressure related to rapid burial or paleo-maximum temperatures (e.g. hydrocarbon generation) must be at least 75 million years old. To hold overpressure in sedimentary basins over this time scale requires average pressure seal permeabilities at or below the lowest published shale permeability measurements. Moreover, the Cooper succession has cooled since the Late Cretaceous, and cooling is an underpressure generating mechanism. Therefore, it is unlikely that burial or temperature driven processes generated the overpressure witnessed in the Cooper Basin. The Cooper Basin has been subjected to an increase in horizontal compressive stress during the Tertiary, and thus an increase in mean stress. The presence of polygonal faulting the Late Cretaceous Eromanga sequence enabled the paleo-stress to be estimated. Layer bound polygonal faulting occurs in a normal fault regime with low differential stress (σv ˃ σH = σh). The contemporary stress regime in the Cooper Basin, measured using the density log, well tests and wellbore deformation modeling, is on the boundary between reverse and strike-slip (σH ˃ σh = σv). The largest measured overpressure in the Cooper Basin is approximately 14.5 MPa in excess of hydrostatic pressure at 3780 m in the Kirby 1 borehole. Mean stress has increased by 46 MPa at 3780 m in the Cooper Basin since the Late Cretaceous. The maximum increase in pressure that can be generated via disequilibrium compaction is equal to the increase in applied stress. Therefore, the increase in mean stress in the Cooper Basin is sufficient to explain the magnitude of the observed overpressure. Additionally, velocity / effective stress analyses on 9 wells indicate that the overpressure was generated by disequilibrium compaction related to increases in mean stress.

A successful application of the high resolution seismic reflection method for talc exploration, is demonstrated using an experimental seismic line, recorded over the Three Springs talc deposit. Since 1973, a variety of ground and airborne geophysical techniques had been applied to talc exploration at Three Springs, with limited success. However, wireline logging and laboratory petrophysical measurements have provided new insights into the physical properties of talc and contrasts between talc and adjacent dolomite and dolerite rocks. In particular, P-wave velocity measurements demonstrated that the elastic contrast between the talc mineralisation and surrounding rocks may be sufficient for the application of seismic reflection methods to talc exploration. Between 1989 and 1993, seismic reflection and refraction surveys were acquired on a single profile across the Three Springs talc deposit. The results of this work were disappointing, apparently contradicting physical property measurements. Greatly improved results from re-processing of the ‘old’ seismic data, in 2000, suggested seismic reflection may indeed assist in mapping of the talc mineralisation, as well as providing important detailed structural information. Additional experimental data acquisition, undertaken in 2000, confirmed the potential application of the seismic reflection method to deposit - scale talc exploration. However, whether the seismic technique can be routinely applied, in a cost - effective manner, providing the required geological detail, remains to be seen.

A particular method of pre-stack multiple attenuation, based on generalised linear inversion (GLI) and the Haskell-matrix formulation, is investigated. The method uses GLI to obtain a 2-D earth-reflectivity function which produces a synthetic seismic record as close as possible to the observed seismic data. The forward model employs the Haskell-matrix method to compute the entire elastic response, including primaries and multiples, corresponding to an input reflectivity. The final reflectivity is used to generate a multiples-only signal, which is subtracted from the original.

Initial trials have been carried out on noisy synthetic pre-stack gathers. Despite erroneous starting earth models, the inversion iterates robustly to provide an output record exhibiting excellent agreement with the observed record. This leads to effective multiple attenuation.

The Haskell matrix method is naturally formulated in terms of wave slowness, and hence the inversion algorithm is most conveniently carried out in either the τ-p domain, or the f-p domain. Our experimentation suggests that, in the presence of noise, the f-p domain is more robust than τ-p.

This method is computationally more expensive than conventional multiple removal strategies, such as those based on differential moveout or predictive deconvolution. Hence it is likely to have most potential where these approaches fail.

Airborne electromagnetic (AEM) data were collected offshore Busselton, Western Australia, using GEOTEM^TM operating at both 12.5 Hz and 25 Hz over the same survey area. The survey covered a region in shallow water up to 35 m depth, and two overlapping tie lines skirting Cape Naturaliste extending into deeper water. The objectives were to compare 12.5 Hz (8 ms pulse) and 25 Hz (4 ms pulse) AEM data in an area of high conductance using conductivity depth sections to map water depth and seafloor resistivity, and to compare B-field and dB/dt responses.

CDI processing of the data with EMFlow software showed that for this dataset, CDIs based on the B-field are better suited for high-conductance targets than CDIs based on the dB/dt data. Water depth was best resolved using 25 Hz B-field data. Sub-sea material and deeper water depths below 45 m was best resolved with 12.5 Hz Bz field data. The 25 Hz dBz/dt data gave the poorest results, due to changes in the bird position not being accounted for in the CDI algorithm.

Sea trials involving the use of marine electromagnetic sensors from DSTO and Flinders University were carried out in Jervis Bay, NSW (Nov 2001 and May 2002) to measure the amplitude of towed controlled source dipole electric fields over various horizontal ranges. The source simulates corrosion-induced ship electromagnetic fields. Three instruments were deployed in 18 m of water, approximately 100 m into the bay to measure two- or three-component electric fields and three-component magnetic fields sampled at 40 Hz (Flinders) and 1800 Hz (DSTO). The towed array of dipole sources produced a variety of signal waveforms of varying frequency and amplitude, and was tracked using either a GPS unit located a few metres behind the towing vessel, or two GPS units, attached to the towing vessel and the rear of the array.

This paper primarily reports on aspects of the towed field propagation through shallow water. Measurements were used to confirm theoretical predictions of electric field characteristics based on numerical modelling of a linear array of current sources and sinks in a layered half-space. Additionally, a large magnetic storm was also recorded in the May experiment, to provide magnetotelluric estimates of seabed conductivity. We show that instruments used in the experiment have both defence applications in electromagnetic field detection, and can also provide shallow geological structure for exploration of the continental shelf.

Structurally-controlled, meso- and epithermal gold deposition in both compressional and extensional settings is a function of local and regional stresses, rheological contrasts, and thermochemical gradients. The influence of these factors can be illustrated through their effects on electrical geophysical structure, since this structure reflects fluid composition, porosity, interconnection and pathways. In the compressional, amagmatic New Zealand South Island, magnetotelluric (MT) data imply a concave-upward (“U”-shaped), middle to lower crustal conductive zone beneath the west-central portion of the island. The deep crustal conductor suggests a volume of fluids arising from prograde metamorphism and radiogenesis within a thickening crust of paleo deep-water clastic rocks. Change of the conductor to near-vertical orientation at middle-upper crustal depths is interpreted to occur as fluids cross the brittle-ductile transition during uplift, and approach the surface through induced hydrofractures. Near the brittle-ductile pressure breakthrough are deposited modern hydrothermal veining, gold mineralization, and graphite of deep crustal provenance, subsequently exposed by erosion. In Nevada, Carlin Trend deposits appear to overlie central intrusives of late Eocene age which occupy the transition from conductive paleo-abyssal pelitic sediments (potential gold source rocks) eastward to more resistive shelf carbonate/quartzite sequences along an ancient continental margin normal fault. The intrusive is flanked by conductive, apparent accommodation fault zones which may possess higher porosity as well as possible graphite flushed from sediments near the high-T system core and redeposited in the periphery. Exposed deposits are modeled to form at fluid pressure breakthroughs across permeability barriers such as organic shales or thrust planes, producing strong and favorable gradients in temperature, pressure and fluid oxidation.

This paper compares the results from 2D inversion of 2D data, 3D inversion of 2D data and 3D inversion of 3D data from a 3D pole-dipole MIMDAS IP survey conducted near Copper Hill NSW.

Six 1.5km lines were laid out over the area of interest at 200m spacings. Data were collected using ‘pole-dipole’ geometry with 100m dipoles. All receiver stations on all lines were read simultaneously for each transmitter station.

Data were processed using standard (proprietary) MIMDAS processing. 2D data were inverted using UBC’s dcip2d software while 2D and 3D data were inverted using UBC’s dcip3d code ported to an NEC SX-5 supercomputer. For the 3D models, resistivities were determined by inversion of primary voltages while secondary voltages, after Time Domain Cole-Cole Inversion, were inverted for the IP model. All errors were based linearly upon observational errors.

Results highlight the differences between inversion models for the two datasets. In particular, the models demonstrate the greater resolution of the 3D survey and the restricted ability of 2D surveys to define boundaries subparallel to the survey lines is emphasized.

Webster (1984) showed “that hardrock tin deposits in south-east Australia have a characteristic magnetic signature in the vicinity of each deposit”. This magnetic signature was concluded to be due to the high temperature of related granitoids imposing a remanent magnetism on rocks within the metamorphic aureole. The paper also discussed the main granitoid classifications and associated mineralisation in relationship to geophysical parameters and the use of the regional data for geological mapping in New South Wales and Tasmania.

The case examples utilised in the original paper have been re-examined using modern airborne geophysical data acquired by Mineral Resources Tasmania (MRT) and the NSW Department of Mineral Resources. The magnetic and radiometric character of the granitoids and local lithology patterns are better contrasted in the new data sets, which allows for mapping of geology and structure to finer resolution.

A significant feature of the new MRT data is the magnetic and radiometric aureole, up to 2km wide, now observed to encircle S-type granites, in particular the Meredith Granite. A metamorphic aureole was known from earlier geological mapping, however, the geophysical response was only poorly expressed in earlier datasets. The new data in the vicinity of these tin deposits are presented and profile modelling is used to confirm that remanence effects are involved in the source of the anomalies. The radiometric data sets show complimentary patterns in the vicinity of tin deposits that enhance the recognition of the mineralised target zones.

In the central NSW tin-belt the S-type granites are intruded into weakly magnetic Ordovician meta-sediments, however, the zero magnetic response of the granites is well contrasted with the north-south linear fabric of the meta-sediments. The tin and gold deposits generally occur in quartz reefs located along sheared contacts of the granite with the meta-sediments and cross cutting east-west structures appear to be related to the source of the tin and gold mineralisation. The application of the radiometric data is restricted by the lack of outcrop.