Exploration Geophysics - Volume 19, Issue 1-2, 1988

A Seismic Reflection Amplitude study carried out over the Goodwyn Gas Field found that reflection amplitudes at the major erosional unconformity bounding the reservoir sequence could be directly related to the lithology and fluid content of the subcropping reservoirs. A colour coded amplitude map was generated and has been used for the prediction of lateral sand distribution/continuity within the reservoir units, both in the interpretations leading to the successful 1986 Goodwyn North appraisal drilling campaign and subsequently in planning for development of the field.

The tau-p transform is still a relatively new and novel seismic data processing technique. As with other transforms used in digital seismic data processing, problems are encountered in choosing the appropriate discretization to avoid aliasing. A diagrammatic method of understanding the aliasing concept in tau-p domain is presented. Examples of the correctly applied tau-p transform velocity filters are presented and show the excellent capabilities of the technique. Convenient methods for enhancing events on the basis of their coherency are also made available by the transform. An example of one of these methods is presented along with a more novel application of enhancing events with characteristic travel times as well as moveouts.

An example of inversion modelling and interpretation which can be done on a personal computer, and three examples of significant seismic data improvement achieved by simple application of oft-overlooked basic procedures, are presented.

Acoustic impedance traces containing only those frequencies recorded in the field can be easily generated from zero-phase seismic data. Forward modelling of these traces using a personal computer can determine lithology and porosity. A short case history of modelling a Miocene reef is presented.

Shallow anamolies of limited areal extent can sometimes distort raypaths to the degree that desired reflections are not received. Designing field setups to avoid such a problem is common on land. An example of successful application of this technique offshore is presented.

Another problem more common onshore than offshore is statics. An example of a marine statics problem is presented.

Finally, the most neglected process of all, display, has been addressed. I intend to show, that contrary to the beliefs of many, this is not a totally subjective topic.

Although our understanding of vibrator source characteristics continues to increase, the fundamental issues of signal polarity and phase remain controversial. Improvements in the accuracy of vibrator control have reduced the potential for phase inconsistencies in processed data, but field setup procedures and variations in data processing still create confusion.

This paper describes an experiment comparing a vibrator and two impulsive sources. Downgoing and reflected waveforms from these three sources are analyzed at selected stages in the processing sequence. Emphasis is placed on polarity procedures during data collection and phase considerations during digital processing. Polarity issues are examined in detail.

Downhole tests compare first-arrival energy from dynamite, an ARIS* impulse source, and a 45,000 pounds peak force vibrator. Stacked sections and VSP data recorded using the ARIS source and vibrator over a well are compared to a synthetic seismogram derived from the lithology of the well. The results show that controlling the vibrator ground force permits the polarity of the vibrator signal to be defined consistently with respect to impulsive signals for both downgoing and reflected energy.

The Wilga Park 1 gas discovery well, drilled in 1985, was the first valid structural test for hydrocarbons in the Gunnedah Basin in northern NSW. A further untested structure (Nyora) is located some 30 km to the west. Both structures were located and delineated solely by the use of seismic reflection techniques.

Both structures are relatively small (7–10 sq km) and to assist in evaluation of optimal survey procedures the Department of Mineral Resources conducted trial surveys with alternative geophysical methods.

The shallow Wilga Park structure (600 m to gas trap) was delineated by resistivity profiling due to the resistive Pilliga Sandstone (70 to 100 ohmm) contrasting with the deeper Purlawaugh Formation (7 to 10 ohmm). The resistivity data were duplicated using the dipole-dipole array (galvanic) and coincident loop SIROTEM (inductive) method. An induced polarization anomaly was observed at depth over strong faulting, which may have been a source of hydrocarbon gas seepage. The gravity method at Wilga Park was only partially successful as a regional gradient masked the anomaly.

The untested Nyora structure was well delineated by gravity and magnetic methods. The magnetic anomaly was interpreted to be due to a fault bounded teschenite sill at 400 m depth, which was also defined by strong seismic reflections. The gravity anomaly, however, was modelled to be due to a basement high. Induced polarization data indicated anomalous responses over the fault margins of the structure and a depleted zone over the sill, which conforms with the model for an alteration/sulphide plume over gas and oil pools.

It is concluded that potential field techniques could be cost effectively applied to exploration for these small structures, to supplement seismic data. Electric methods provide a useful adjunct data set that has positive exploration incentives if drilling proves the ‘plume’ to be a hydrocarbon indicator.

The published gravity and magnetic maps of Australia provide a means to extend and refine the generally accepted regional geological framework of the continent. Major crustal units can be recognised by areas of predominantly high or low gravity values. The boundaries of most units are outlined by steep gravity gradients.

Magnetics are used in defining more closely the boundaries of crustal units and features within the units such as igneous rocks, particularly granitic bodies, basic and ultrabasic masses, major faults and other lineaments.

The mechanism of low-frequency Borehole Coupling from the literature is here applied to a borehole of non-uniform radius. It is noted that in addition to the pressure signal in a fluid-filled borehole coincident with the passage of an external plane compressional wave, there are waves created at any change in radius which arrive later as coherent noise. This is illustrated first with a single step-change in radius. An example using the caliper log from an oil well shows that these delayed waves create a noise train of substantial amplitude relative to the direct compressional wave. It is pointed out that a down-hole pressure source will radiate a noise train due to caliper changes which will follow the direct compressional wave with exactly the same waveform as that encountered in Borehole Coupling.

Definition of gas reservoirs in several fields in the offshore Gippsland Basin, Victoria has been enhanced by use of amplitude variation with offset, direct hydrocarbon indicators and seismic modelling. Direct indicators include bright spots, flat spots and phase reversals associated with reduced acoustic impedance in the gas zones.

The techniques have generally worked successfully at the Golden Beach Field and at the Baleen and Patricia gas discoveries at subsea depths from 600–700 m.

In 1985, the Australian Bureau of Mineral Resources recorded 500 km of deep seismic reflection data across the Arunta Block and the Ngalia and Amadeus Basins in central Australia. Large offset shots and expanding reflection spreads were also recorded coincident with the near-vertical incidence reflection data in the Arunta Block and the northern part of the Amadeus Basin to better define seismic velocity variations throughout the crust. In the northern part of the Amadeus Basin lateral and vertical variations in seismic velocity or variable dip in the reflectors can be detected throughout the 10 km thick sedimentary sequence, thus providing useful constraints on lithological or facies variations. A new method of mapping lateral velocity variations below weathering has been developed for interpreting first break data from conventional reflection shots, and has enabled the presence of a northerly-dipping fault in crystalline rocks to be detected in a part of the Arunta Block with few outcrops. The detailed velocity variations measured in selected localities is assisting with the proper migration and interpretation of both shallow and deep reflections.

Oil has recently been discovered in the Hugoton Embayment, Kansas, through the use of stratigraphic interpretation of seismic data. This oil is located in traps of channel sands in the Morrow and Chester Formations.

Current seismic data, from which producing well locations were selected, were ‘deprocessed’ to simulate older data. Field acquisition and seismic processing parameters were duplicated for six of the prior twenty-four years. This simulated data shows the logical progression of field acquisition and processing technology. This method also demonstrates the stratigraphic appearance of the anomaly on the data during those years.

Both field acquisition and processing parameters and techniques are reviewed on each of two lines. The resulting impact of the changes on each of the sections and its anomaly is noted as we progress through time.

In addition to showing full sections for each of the six years, blowups of the actual anomaly location on each of the simulations are shown with the original data for easy comparison.

A new digital fibre optic streamer and recording system continues the trend towards finer spatial sampling, and more comprehensive and sophisticated electronic and quality control systems. However, at the same time it provides greater flexibility by giving a choice of hydrophone array lengths, and the option to use multiple streamers. Streamer noise analysis led to a revised design, with new materials, giving a very low tow noise level. The system can be configured as 1, 2 or 4 streamers, the multiple streamers giving increased efficiency in 3D data collection. For increased coherent noise attenuation, 3 or 5 streamer groups can be mixed before digitisation, in a weighted overlapping manner. The on-board streamer control system performs many in-water electronic tests on the streamer, while the field computer/recording system has a comprehensive range of seismic quality control and display functions. The system gives the ability to improve the signal-to-noise ratio of the recorded data over a wide variety of conditions.

Ground water contamination is a major problem facing industrial nations. Electrical methods seem particularly promising in mapping and monitoring ground water regimes since the electrical conductivity of rocks depends almost entirely on the fluid saturation, salinity, and distribution. The most important recent developments in resistivity include the use of numerical modelling and resistivity mapping using subsurface electrodes. The latter yields far greater accuracy and resolution than can be obtained with surface arrays. To illustrate the power of subsurface-surface arrays we have studied an idealised two dimensional model of a contaminated zone. Since we are interested in emphasising the anomaly caused by the repository, or subsequent changes over time in its vicinity, we have discovered that it is very useful to express the apparent resistivity results as percentage differences from either the background (for surface arrays) or from the apparent resistivities observed at a particular depth of the current source (for subsurface arrays). Percent differencing with respect to data at the repository depth dramatically reduce near-surface and topographic effects that usually confound quantitative interpretation of surface surveys. Thus, dc resistivity appears to have great potential for mapping and monitoring zones of impaired ground water.

Processing seismic data to extract information in addition to the structural configuration requires careful attention to the effect the processing sequence has on the trace attributes of phase, frequency and amplitude.

This paper is primarily concerced with processing noisy onshore data to produce relative amplitude stack sections.

Particular emphasis is given to compensation for near-surface effects on trace amplitudes. The derivation of compensation scalers from the data is demonstrated. The application of these scalers alone will produce a relative amplitude stack section, but if significant noise is present, the stack will not have the optimum signal-to-noise ratio. The combination of compensation scalers and power diversity stack is shown to produce a good relative amplitude stack.

The Tasmanian Mines Department recently initiated a large-scale project to assist exploration in Western Tasmania, one of the world’s most mineralised regions. The geophysical component of the project was designed to determine the geophysical signatures of the type deposits and the effectiveness of the techniques being used in Tasmania, particularly in areas of thick cover.

The study has shown that the replacement tin and polymetallic volcanogenic deposits are good targets for TEM surveys, although small occurrences of non-conducting zinc-rich deposits have been found. A comparison of FEM, TEM and CSAMT surveys over a series of moderate conductors emphasised the limitations of fixed transmitter surveys and demonstrated the potential of CSAMT for discriminating deep targets.

After a brief discussion on why source patterns are now an integral part of marine acquisition parameters the argument is developed that these very same arrays inhibit the response of high frequencies with offset, even at depth, which cannot be restored even if signature deconvolution takes source directivity into account.

A controlled experiment is described where the only variable is the critical ‘in-line’ dimension of the source pattern. The source in this case was waterguns but it could equally have been sleeve airguns or marine vibroseis, each with enough bandwidth to demonstrate the effects revealed in this paper. Raw field records are compared by exclusive filter suites, normalised onto each other, up to the limit of the 4 ms anti-alias filter.

Thereafter signature deconvolution is compared, again on the same field records, but over a short single cover section of constant offset, in the first case from the middle of the streamer, around 1500 m offset and in the second case near the end of the streamer approximately 3000 m offset.

Geophysical methods provide the capability to map the lateral and vertical distribution of pollution in areas between boreholes drilled to determine the spread of contamination from a waste disposal site. In particular, when contamination causes an increase in dissolved salts in the groundwater, the consequent increase in its conductivity may be detected and monitored by electromagnetic (EM) methods.

Transient electromagnetic (TEM) and Schlumberger DC sounding measurements have been made at a liquid waste disposal site and a landfill site. At both these sites, significant changes in conductivity near the sources of contamination have been detected by the TEM and DC methods, and the results generally correlate with the conductivity of groundwater sampled in the boreholes. The TEM and DC data are inverted assuming a layered earth model. Measurements at the landfill site are being made at approximately six-monthly intervals to test the application of the methods to the monitoring of the spread of the pollution plume. Data are also being collected over the same time intervals at another site prior to its being used as a landfill site. This will enable the discrimination of any seasonal change in the sub-surface conductivity from a change in conductivity caused by groundwater contamination once disposal of wastes begins.

Laboratory electrical conductivity (σ) anisotropy measurements were made on water saturated Palaeozoic black shales, mineralized schists, and banded and massive sulphides from the Mt Lyell region Tasmania. Macro- and micro-fabric and magnetic anisotropy studies were also carried out. Directional 1 kHz conductivity values were obtained in 9 directions for each of 10 samples. Reduction of these measurements produced the electrical conductivity tensor principal values and directions which were correlated with the fabric and magnetic data. The study suggests that: (i) a water saturated layered rock may be characterised by large σ1:σ2:σ3 anisotropy; (ii) the conductivity tensor representation may reflect petrofabric; and (iii) anisotropy may cause the current density to deflect significantly from the electric field direction.

Although magnetics is the most widely used geophysical method in hard rock mineral exploration, interpretation of the data suffers from inherent non-uniqueness. Ignorance of the magnetic properties of magnetic rock units may lead to faulty interpretation, with expensive consequences. Several examples of interpretations which have been assisted by magnetic property measurements will be presented. Applications include siting drill holes in order to maximise the probability of intersecting the target and testing whether intersected material explains the anomaly. Factors which may have to be considered include the remanence direction and Koenigsberger ratio (particularly for volcanics, basic plutonic rocks and pyrrhotite-bearing rocks), the age of remanence relative to folding, and susceptibility anisotropy (particularly for banded-iron formations and certain orebodies). Palaeomagnetic cleaning and data analysis techniques are essential for extracting maximum information from samples.

Over the past several years little attention has been paid to the electrode arrays used in induced polarisation (IP) surveys. However, arrays are very important in relation to anomaly detection and interpretation and to survey cost.

Currently popular arrays are gradient and dipole-dipole. Other arrays are used for resistivity surveying, but generally suffer more inductive coupling. Some new arrays have been investigated in order to look at:

1. the advantages of current or potential dipoles parallel to strike,
2. the virtues of broadside arrays,
3. minimisation of inductive coupling, and
4. a possible reduction in survey costs.

A perpendicular dipole-dipole array can reduce inductive coupling with horizontal layers and give an IP anomaly similar to that from a collinear array, if run across strike with both dipoles oblique to strike. Rather different, crossover type, responses are obtained with an L-array, in which one dipole is parallel to the strike of a target zone. This arrangement may locate the target more clearly.

A broadside parallel dipole-dipole array can take advantage of the common strike elongation of a target deposit, and relative to a collinear array run across strike, can provide improved resolution, dip information, and efficiency in reconnaissance.

Some of the features and benefits of various arrays are demonstrated by modelling results. With further work, including development of interpretation methods, attempts can be made to optimise array geometry for a specific exploration scenario.