Exploration Geophysics - Volume 22, Issue 2, 1991

A velocity analysis process, SIVA, is described which combines the information in CMP gathered trace data with interpreted seismic data (e.g. time maps) to create a model of structure and P wave interval velocity. The technique works one depth interval at a time, in a top down fashion. Raytracing is used to estimate the shape of the reflecting surface and determine a family of non-hyperbolic moveout trajectories associated with a range of trial interval velocities. The trace data are stacked along the computed moveout curves, and semblance techniques are used to determine the best-fit velocity for the interval being analyzed. The analysis is performed for a number of gathers to yield a spatially varying interval velocity field. This field is then used to re-determine the shape and depth of the reflector, the model is updated with the velocity field and structural information, and analysis continues to the next lower interpreted horizon. The resulting depth/velocity model is fully consistent with all the information in the seismic trace data, as well as the interpreted time picks, and is fully compensated for overburden structure as well as structure and reflection-point smear on the reflector. SIVA analysis may be performed to produce a 2D model from a single line, or to create a 3D model for input data from multiple crossing lines or a 3D survey.

Bouguer gravity maps and aeromagnetic maps of the Koonenberry 1:500 000 map area in northwestern New South Wales have been compiled from Bureau of Mineral Resources data bases.

Low regional magnetic field strength within the area of the Precambrian Adelaide Fold Belt extends about 90 kilometres north of the northernmost Precambrian outcrop, suggesting some continuity of crustal structure through this region.

On the western edge of the area, in the vicinity of 30°30’ S, magnetic response from shallow subcrop indicates that a detailed aeromagnetic survey in this vicinity could clearly delineate the shallow subsurface structure.

The Bancannia Trough is underlain by strongly magnetised rock units, possibly dacite/andesite. These magnetized rock units are terminated at their southern end by an ENE trending lineament which extends through the Wonominta Block into the basement of the Darling Depression. ERTS-1 images indicate that this lineament also extends WSW through Broken Hill.

Bouguer gravity values over the Wonominta Block are markedly higher than over the Adelaide Fold Belt, reflecting a probable convergent plate margin associated with the Wonominta Block.

The full extent of the Koonenberry Fault (which in this area is a major component of the boundary between the Kanmantoo and Lachlan Fold Belts) is clearly delineated by the magnetic data.

The few near-surface magnetized rock units east of the Koonenberry Fault generally give rise only to isolated anomalies, except in the restricted area of Ordovician metasediments surrounding the Tibooburra Granite.

The Eromanga Basin is distinguished from the Darling Depression by higher Bouguer values (indicating shallower sedimentary basement depths), and basement structures with generally northwesterly trends (compared with no discernable basement trend directions in the Darling Depression).

DC resistivity sounding data usually are interpreted in terms of a layered earth by an iterative inversion scheme. This paper presents an alternative mechanism for the inversion process by formulating the resistivity inverse problem as an optimisation problem. By invoking the commercial mathematical programming software package called GAMS, an inversion routine has been developed without having to linearise the problem explicitly and without having to know details of the convergence algorithm. The GAMS high-level programming language provides a structure for coding the forward problem, for specifying an initial model and for declaring reasonable bounds on model parameters. GAMS inversion is extremely versatile in allowing ad hoc constraints of any complexity, but converges more slowly than special-purpose resistivity inversion algorithms. After familiarity is gained with the GAMS language, new code for more complex resistivity models or different geophysical methods can be developed in much less time than is required for the development of dedicated geophysical inversion software.

During 1989, a closely spaced offshore seismic grid was acquired over an area of thrust faulting along the Taranaki Boundary Fault Zone, New Zealand, and complemented by simultaneous gravity recording along a number of profiles.

Production seismic data processing included DMO and steep dip migration. Subsequent reprocessing of selected lines with full prestack time migration improved the imaging in the zone of complex structuring.

Integrated interpretation of the seismic and gravity data developed a model for the geometry of the Taranaki Boundary Fault in this region. From north to south, the structural style is shown to change from low to high angle overthrusting of Mesozoic basement with truncation of the (younger) sediments against the basement.

Stacked profiles of aeromagnetic data often fail to resolve and display detail as this can be obliterated by larger amplitude features on adjacent flight lines. Contours and images of interpolated data can also obscure detail because they are sensitive to survey levelling errors and can show erroneous distortions caused by the interpolation method. Mclntyre (1981) showed that shaded stacked profiles of the second horizontal differences of flight-line data are effective in detecting and displaying detail in aeromagnetic data. Mclntyre used a square-root function to control the dynamic range of the computed difference so that both small and large amplitude features could be displayed without being obliterated by larger amplitude features on adjacent flight-lines. The shaded profile preferentially enhances positive values. Improvements on Mclntyre’s shaded stacked profiles of horizontal differences form the basis of a new method of presenting horizontal differences of aeromagnetic data: the bipole map. This is a series of stacked flight-line profiles of bar-graphs of the horizontal difference values. The heights of the bar-graphs are controlled by the amplitude of the difference values whilst polarity is depicted with colour: red for positive values and blue for negative values. The bar-graphs are centred and located precisely on the curved flight-line; the flight-line forming the centre-line of the profile. Wavelength dependent automatic gain control (AGC) filters are used to modify the dynamic range of the horizontal differences so that both small and large amplitude anomalies can be displayed. The bipole map visually enhances both negative and positive values, more precisely positions the data and does not depend upon dubious magnetic values interpolated between the flight-lines. Bipole maps of the first, third and fourth horizontal differences are effective in resolving detail in other parts of the frequency spectrum. They supplement contours and images of longer wavelength features. Magnetic models show the responses of the horizontal differences for various inclinations of the Earth’s magnetic field and different widths of the magnetic source. In areas of steep inclination, the horizontal differences delineate the strike axis of thin bodies and the edges of thick bodies. Bipole maps of horizontal differences may find useful application in displaying other types of geophysical data such as VLF-EM and aeromagnetic gradiometer data. Bipole maps of higher order differences are an effective means of resolving high frequency (instrument) noise in geophysical data. For aeromagnetic data from the Mt Magnet goldfield, located in the Archean greenstone belt of Western Australia, bipole maps of the second horizontal differences revealed that an anomaly from a banded iron formation might be caused by several close-spaced banded iron bodies rather than a single body, as might first be interpreted from the contours and images of the gridded data. Bipole maps of the fourth horizontal differences revealed a series of NNE striking magnetic lineaments which are not apparent in the contours and images of the gridded data.

The Murray Basin has an area of 320 000 km² and is one of the most important sedimentary basins in Australia in terms of agricultural and cultural development. It is essentially a closed groundwater basin which consists of a thin cover (200-600 m) of sediments containing 4600 million megalitres of water varying in salinity from 500 mg/L to over 35 000 mg/L total dissolved salts.

This study of the central Murray Basin covers an area of 200 x 300 km, and is based on 1700 km of seismic refraction, over 300 vertical electrical soundings and the analysis of several thousand bore logs.

The pre-Tertiary basement surface was found to have highly variable relief of up to 300 m. A rectilinear system of crustal fractures has been responsible for the formation of a basement ridge complex, identified as the Ivanhoe Block. A three dimensional (3D) representation of the basement has been developed to highlight dominant trends. Occlusion of aquifers across the Ivanhoe Block is a controlling factor affecting the regional groundwater flow, and is a contributing cause of the outflow of saline groundwater to the land surface and the river system.

Hydrocarbon reservoirs sometimes occur within low velocity channels, thus forming potential seismic waveguides. The waveguide might be excited at its discontinuities (faults) by seismic waves travelling downward from a surface source. Given that a waveguide would confine scattered energy to the horizontal plane, it should be possible to map fault scatterers at large range by means of a downhole triaxial geophone placed within the reservoir.

A simple laboratory scale seismic model of a Southern North Sea gas reservoir was constructed to study the problem. Multicomponent seismic measurements were taken in both walkaway VSP and offset VSP shooting geometries. The results validate the presence of strong waveguiding in a low velocity layer disrupted by faulting. The diffracted arrivals can be imaged to locate the fault ‘source’. The model results, when scaled to the field situation, suggest that faults within a reservok can be detected at horizontal distances greater than several kilometres from the drill hole.

Xenolith-derived paleogeotherms for the lower crust and upper mantle beneath the eastern margin of the Australian craton (EMAC) and beneath the Phanerozoic Tasman Fold Belt (south-eastern Australia — SEA) indicate distinct thermal regimes. The SEA geotherm passes through 800°C at 7 kbar and 1200°C at 25 kbar. The strong curvature of the constructed geotherm from 10 to 30 kbar indicates significant advective heat transfer in the lower crust and upper mantle. The EMAC geotherm lies some 150 to 200°C below the SEA curve and stretches from 600°C at 7 kbar to 1000°C at 21 kbar. This is well above the steady state conductive geotherm calculated from (low) surface heat flow for cratonic areas.

The lower crustal suite xenoliths from EMAC and SEA are dominated by mafic lithologies. The EMAC mafic xenoliths represent igneous intrusions into the lower crust and uppermost mantle metamorphosed to granulites and eclogites. The large lateral variations in temperature at the base of the crust (depending on tectonic environment) are significant to the relative stability of eclogite and granulite mineral assemblages. The restriction of lower crustal eclogite suites to cratons or craton margins, and their apparent absence from younger terranes, is a consequence of this temperature difference.

The definition of distinct thermal regimes in different tectonic settings provides a basis for examining the response of magnetic properties and Vp-depth profiles with changes in temperature. The seismic Moho and the crust mantle boundary do not coincide in areas that have high geothermal gradients. Differences in the seismic characteristics of cratonic areas and areas with high geothermal gradients can be modelled as the effect of temperature on similar lithological columns and do not necessarily imply large petrographic differences. A lower crust beneath cratonic areas dominated by eclogite facies assemblages may be misinterpreted seismically as mantle.

During the 1980s Outokumpu Exploration carried out several exploration programs for nickel sulphides in eastern Finland. Geophysical methods have been significant both in locating nickel bearing intrusives hosted by mica gneiss and in detailed downhole investigations of the dimensions of the ore zones.

In-hole 3-component magnetic and mise-a-la-masse measurements contributed to the discovery of a small but high grade orebody beneath a worked out shallow deposit at Telkkala, southeastern Finland.

The high contrast in magnetization and conductivity between the ore mineralization and the host gneisses provides a favourable situation for quantitative interpretation of the geophysical data.

The Kerna gas field is located in the south-central part of the Cooper Basin close to the large Dullingari and Toolachee Fields. Kerna is a domal anticline and contains gas structurally trapped within sandstones of the Early Permian Patchawarra Formation. The overlying Epsilon Formation sands also contain gas which may be stratigraphically trapped around the south and west flanks of the field.

Seismic reflection amplitudes are used to map the extent of this Epsilon gas sand. Seismic modelling studies show that the gas sand displays an amplitude-versus-offset (AVO) response which distinguishes it from a wet sand or from a coal bed at the same stratigraphic level. Constructive and destructive interference from the top and base of the gas sand, and from overlying coal seams, have a marked effect on the overall seismic stack response of the target horizon. The modelled AVO response shows a good correlation with AVO changes on cd.p. gathers from seismic lines across the area.

The spatial distribution of the AVO anomalies (AVO increases), and of the overall seismic stack response, has been mapped across the field and can be used to locate future appraisal and exploration wells. This study has implications for other areas of the Cooper Basin where adequate separation between coal beds and gas sands allows the AVO effect of the latter to be observed. These AVO effects can then be used as a direct indicator of gas in stratigraphic or even structural traps.

The North West Shelf of Australia presents problems in data acquisition and processing. In particular the hard water bottom and near surface prograding carbonate wedge set up a standing wave multiple train analogous to that of a closed organ pipe. This strong multiple energy and the observed rapid attenuation of the high frequencies in the source spectrum degrade the quality of the recorded data. The data quality can be improved by careful selection of the acquisition parameters. The multiple energy appears to be reduced by reducing the output power of the source and restricting the high frequency end of the source spectrum to useable frequencies. The frequencies in the low range of the seismic spectrum can be used to improve the seismic data through broadening of the bandwidth and improved penetration.

The discovery of the Scuddles volcanogenic massive zinc-copper sulphide deposit (400 km NNE of Perth) in 1979 was the result of an integrated exploration programme. The initial diamond drill hole was proposed on the basis of anomalous copper and lead geochemistry in chloritised tuff and chert in RAB drilling across a 120 nanoTesla aeromagnetic anomaly. The first 3 holes intersected interesting but sub-economic grades of copper, lead, zinc and silver mineralisation. A Pulse EM survey was then completed and identified a moderate conductor associated with the mineralised horizon. Diamond drill hose SC-4 was drilled beneath SC-3 and intersected massive sulphide which included a section of 25.5% zinc over a true width of 6 metres. Although the next 2 holes were unmineralised, subsequent drilling quickly established that a discovery of economic significance had been made.

The discovery was made by the application of a well-planned and carefully executed exploration programme with an emphasis on direct field geological input. This result emphasises the need for flexibility in an exploration programme so that encouragement in one area can be followed up while continuing to generate and evaluate new targets.

The Tertiary uplift of the Eromanga Basin over PELs 5&6 (SA) and ATP 259P (Queensland), has significantly complicated depth-porosity relationships in the important oil-bearing Hutton Sandstone, as well as altering migration pathways.

Because seismic reflection data in the Eromanga Basin are focused on the lower part of the succession, they are not suitable for imaging and mapping the Tertiary unconformities. In this study, an estimate of the magnitude of Tertiary uplift across the acreage has been made by cross-plotting interval velocity and depth for the lithologically homogenous Cretaceous marine shales near the top of the succession. Deviations from a depth-velocity baseline fitted to data from the basin depocentres have enabled quantification of uplift.

The uplift estimates have been used to produce pre-uplift depth structure maps which can be used to map pre-uplift hydrocarbon migration pathways. Maximum burial depth can also be calculated, providing a better predictor of porosity in the Hutton Sandstone than present burial depth.

This paper describes a high-resolution seismic survey in the Southern Coalfields of New South Wales over rugged terrain in areas yet to be mined. The aim of the survey was to identify faults in the working seam, which are very disruptive to longwall mining operations. Rapidly changing topography with deep gorges, swamps and dense bush create access and environmental difficulties for seismic operations. For the first time, attempts have been made to acquire seismic data across difficult gorge and swamp country. The techniques used were successfully adapted from previous surveys carried out in less difficult terrain. Current indications are that there will be minimal impact on the environment. The data collection and processing parameters developed enabled good coherent data to be obtained across deeply incised terrain. Faults with various throws have been interpreted.

Recently, novel geophysical technology has been developed which is capable of determining groundwater content at various depths without drilling wells. The instrument is the Hydroscope and relies on the principles of Nuclear Magnetic Resonance (NMR) tomography. Using the Hydroscope, it is now possible to undertake a regional hydrogeological survey to map the distribution of groundwater reserves both in plan and in depth with high efficiency, thus ensuring a reliable choice of the most productive areas for water supply bores. Some results of Hydroscope tests on sedimentary, water-containing rocks in South-Eastern Australia, undertaken during February and March 1990, together with several theoretical and experimental results obtained previously by the developers of the NMR water prospecting technology, are presented.

Southwest Queensland is covered by sediments of the Eromanga Basin. Although deposition occurred during a structurally quiet time the basin shows extensive folding, faulting, uplift, and sequence truncation. This structuring is attributed to post-depositional (post-Cenomanian) tectonism involving essentially mild, east-west directed, basement compression. The detailed chronology of this tectonism is vague but uplift began during the Late Cretaceous and at least two discrete episodes of Tertiary folding are recognised.

In addition to regional uplift, deformation involved the reactivation of pre-existing basement trends. The style of deformation varies systematically but is always dominantly compressional, although along the east-southeast oriented Gidgealpa-Merrimelia-lnnamincka (GMI) trend wrenching is conspicuous. Across areas of shallow basement, deformation has been preferentially expressed as vertical displacements involving the formation of reverse fault bounded, tilted, blocks, pop-up blocks, and trap door structures. As hydrocarbon traps these structures have limited prospectivity. Formation of these structures post-dates the peak phase of hydrocarbon generation and critical closure is often fault dependent.

Where sediments of the Adavale Basin are preserved Tertiary reactivation of Kanimblan structural trends has resulted in the development of folds within the overlying Eromanga Basin sequence. Folding may enhance pre-existing drape related closure. Where folding is associated with the rejuvenation of basement-related faults the trapping potential of the reactivated structures is often reduced.

The common occurrence of multi-storey hydrocarbon pools not filled to spill-point, and located adjacent to reactivated fault escarpments provides strong evidence that Tertiary, reactivated, reverse faults do not seal. The most prospective structural targets appear to be those involving reactivated, large palaeo-highs. By absorbing much of the Tertiary deformation along their flanks the crestal integrity of the structure remains relatively intact. At the same time reactivated faults along the flanks provide conduits for vertical migration of hydrocarbons up into reservoirs lying in crestal locations.

Uplift of the eastern margin established it as the major intake area for the recharge of the basin aquifers and the southwest flow of groundwater. The widespread formation of prominent anticlinal fold trends and reverse faults during the Tertiary established new barriers to ground water movement and horizontal migration of hydrocarbons. Tertiary uplift also retarded maturation along several key trends. Contemporary hydrocarbon drainage patterns were probably established during the Middle Tertiary and after peak hydrocarbon generation. Thus predictions of hydrocarbon drainage areas based on structural maps, such as the ‘C or ‘P’ horizons, can lead to erroneous conclusions. ‘C-P’ isopachs, corrected for post-depositional thinning, provide the best indication of drainage patterns during the critical period of peak generation.

In 1985 the Bureau of Mineral Resources undertook deep seismic reflection and refraction profiling experiments across the Early Proterozoic to Late Palaeozoic Amadeus Basin and Arunta Block in central Australia. The experiments were designed to test various models of intracratonic basin formation which involved tectonic interaction with the surrounding basement.

The seismic reflection results show the thrust belt at the northern margin of the Amadeus Basin to be dominated by a major, south-directed, basement thrust feature, the Redbank Thrust Zone, which was imaged to mantle depths. Other, more southerly, basement-cored thrusts appear to splay southwards from the Redbank Thrust Zone towards the Amadeus Basin. The seismic reflection results support a ‘thick-skinned’ Laramide or Wind River style of overthrusting for the final compressive Late Devonian-Carboniferous event, the Alice Springs Orogeny. It was the accumulation of thick, synorogenic conglomeratic sediments in a foreland-like setting at the time of thrusting that appears to have been the dominant control on maturation of the underlying Early Ordovician source rocks.

In contrast to the northern margin, in the central and southern parts of the Amadeus Basin, the seismic reflection results indicate north-directed, ‘thin-skinned’ overthrusting on shallow detachments that were accompanied by Jura-style box folding. The detachments sole out in a salt horizon near the base of the succession. Major repetition of part of the succession has occurred, particularly at the leading northern edge of the overthrust region. Basement involvement at the southern basin margin is enigmatic, but appears to involve both Late Proterozoic and Devonian-Carboniferous tectonism.

Since 1981 CRA Exploration (CRAE) has flown in excess of 43,000 line km of airborne magnetics in the Cobar district. The data, which were acquired from a number of surveys, are being used to assist in the exploration for base metals and gold. Examples of the targets sought are the Elura lead-zinc-silver deposit, the CSA copper deposit and the Cobar gold deposits which all have associated magnetic features.

A 1974 aeromagnetic survey flown at 90 m terrain clearance gave a discrete 40 riT response over the Elura orebody. Such an anomaly could be detected in areas of low magnetic relief within the detailed surveys. Unfortunately significant parts of these surveys are affected by the variable responses of maghemite concentrations in shallow channels prevalent in the Cobar district which could mask small discrete anomalies.

To test this, a process of upward and downward continuation was applied to the original 90 m data set. These data sets were added to a grid containing a maghemite channel response. The resulting grids were then filtered in an attempt to extract the original data.

The results suggest that even without filtering an Elura anomaly could be detected under a maghemite channel to a depth of about 200 m. Using a Butterworth low pass filter it was considered that at depths to 400 m below a channel, an Elura anomaly could be recognised.

The paper concludes that this practical approach is applicable to other data sets that may require filtering.

Techniques for automatic determination of the depths of sources from magnetic profiles represent fast and inexpensive interpretation tools. The automatic technique proposed by Naudy is widely used. This technique requires accurate locations of the centers of anomalies in order to obtain reliable final estimation of the depths.

In the proposed improvement in the Naudy technique, the total magnetic fields on the profiles are split into horizontal and vertical components. It is demonstrated that this procedure leads to accurate locations of the centers of anomalies. Final estimates of the depths are less ambiguous and agree well with data obtained from drilling.

Application of the Naudy technique to vertical gradient data is discussed.