Exploration Geophysics - Volume 22, Issue 2, 1991

Reinterpretation of 18 local and 7 regional seismic lines in northeast Java, numerous exploration wells and their integration with newly measured stratigraphic sections has enabled a new structural model to be developed for the Cepu oil fields. The generally shallow water limey-clastic sequence developed in a rifting back-arc basin with many northeast-southwest oriented basement faults.

Deformation in the early Middle Miocene caused reactivation of the basement faults in the Nglobo-Semanggi area with wrenching and the initial development of flower structures. This deformation caused areally restricted erosion of the main reservoir rocks in this area. Later Pliocene deformation accelerated the development of the flower structures in the Nglobo-Semanggi area which were reflected at the surface as a series of en echelon, hydrocarbon-bearing anticlines. The Tambakromo-Kawengan area underwent minor north-over-south thrusting along east-west oriented listric, reverse faults with detachment at shallow depths and the development of hydrocarbon-bearing anticlines in the cover sequence. It is possible that further hydrocarbon-bearing folds exist in the subsurface north of the Tambakromo-Kawengan structure. These folds would be related to blind imbricate thrusts parallel to the Tambakromo-Kawengan thrust.

During the past decade several deep electromagnetic techniques have been successfully applied to hydrocarbon exploration problems. Case histories for magnetotellurics (MT) and long offset transient electromagnetics (LOTEM) have shown both the strengths and the limitations of the individual techniques when applied in a production mode under real field conditions.

When exploring for hydrocarbons it is essential that the resistive units are as well resolved as the conductive ones. This means that one must use an electric dipole transmitter and an electric field receiver because this is the most direct way to measure vertical current systems in a layered earth. Furthermore, to extend the LOTEM results to greater depth and to improve the resolution of MT at shallower depth, a combination of both techniques is required. When using MT and LOTEM together for mapping of lateral changes in resistivities, a priori information such as the interfaces from reflection seismics is needed. This type of integration of the different techniques significantly increases the reliability of the interpretation because the strengths of the different methods compensate for each other’s weaknesses. This concept is underlined using field data from different case histories around the globe.

A typical petroleum exploration problem is a resistive unit at great depth which simulates porosity variations within carbonates. This kind of earth model is investigated using a synthetic data set. The joint interpretation of LOTEM-MT does not yield sufficient resolution. When adding seismic information to the EM interpretation the resolution becomes significantly better and the porosity changes can be resolved.

Today’s computers span a wide range of computing capability from desktop personal computers, through minicomputers and mainframes to the fastest, most powerful machines designed to the limit of existing electronics engineering technology. It is the last category, those that perform millions of floating point operations per second (megaflops) at the peak of current technology that are termed supercomputers.

In seismic data processing extremely large volumes of numbers are processed through many stages, most of which involve complex mathematical operations. A typical seismic survey today is acquired with 300 recording channels to 6 s at a 2 ms sample rate. This recording configuration means that one kilometre of data produces 36 million samples (floating point numbers) for processing. Many seismic surveys today involve the recording of up to 10,000 kilometres of data, particularly 3D surveys.

Processing techniques currently performed have evolved over past decades in accordance with the available electronic hardware of the time. The latest supercomputing technology allows more complex processing algorithms to be tested on data in realistic time, as well as the increased throughput capacity needed to accommodate today’s large multichannel 2D and 3D seismic surveys.

An accurate estimate of impedance functions is essential for the correct interpretation of a magnetotelluric (MT) sounding. Unfortunately, noise is inevitably encountered when the MT observation is conducted. In effect this can strongly influence the results given by the MT process and results in a distortion of the estimate.

This paper proposes an alternative method for making unbiased robust estimates of MT impedance functions. The means for accomplishing this is based on the regression M-estimation and the Hilbert Transform operating on minimum phase MT impedance functions. Using MT data from the Columbia River Plateau and the EMSLAB Lincoln line, it is shown that the method can produce usable MT impedance functions even under conditions of severe noise contamination and in the absence of remote reference data.

With the ever increasing number of 3D surveys being acquired, it is essential that 3D interpretation be both effective and efficient to ensure that maximum information is derived from the data. In this context, Woodside has put considerable effort into the development and application of novel techniques in seismic interpretation over recent years to assist in the optimal interpretation of its existing and planned 3D seismic data sets. These studies have included comprehensive analysis of seismic horizon attributes, coupled with the application of image processing techniques, to enhance subtle features not readily discernible by other means.

Image processing, commonly applied in mineral exploration to enhance Landsat images and potential field data, has been applied to interpreted seismic horizons from 3D data sets with outstanding results. Two aspects of image processing which have been utilised with good effect are data illumination and combined displays.

The illumination technique involves artificially illuminating the horizon by a ‘sun’ and varying the ‘sun’ angle (azimuth and elevation). This tends to enhance features normal to the ‘sun’ direction. Combined displays have allowed the integration of different attributes onto one map using both a black/white scale and an overlay colour scale.

Application of these techniques in the Goodwyn South/Tidepole area has provided new insights into the nature of subtle features of the seismic data. For example small faults, barely discernible using conventional mapping techniques, are clearly revealed and can be traced with confidence along their strike. Moreover, the dip direction of these faults is readily apparent. In particular, the illumination and dip displays enhance the fault trends, while the azimuth displays combined with other attributes highlight the inter-relationship of faults and direction of fault throw. Integration of all these attributes/displays during interpretation has allowed a clearer understanding of this structurally complex region.

Potential also exists to highlight even more subtle features such as cuestas associated with the subcrop of resistant beds. This response was utilised recently to assist in targetting development wells in the North Rankin Field.

Further development and application of these image processing techniques, coupled with horizon attribute mapping (dip and azimuth) will result in more confident and better interpretations, and hence will assist the orderly and optimum development of Goodwyn and other fields during the 1990s.

Conventional sediment transport models are ideal, in theory, for modelling how sedimentary basins fill, but in practice there are rarely adequate data to constrain them. An alternative approach recognises that any basin is a spatially varying sedimentation system, evolving in time towards a steady state, and that its state at any time can be characterised by what can be termed the ‘degradation potential’ field. (The instantaneous degradation potential at any point is a reciprocal function of the expected survival time of any sediment then being deposited there.) The process of basin filling can be modelled as one of potential-driven flow: as the degradation potential field changes in time, so does the configuration of the depositional surface, and this can be calculated. This potential-driven flow model has been implemented using a finite-difference approximation over an irregular triangular mesh. It predicts realistic basin fill patterns, in three dimensions, and is computationally very economical.

A knowledge of the structural and tectonic processes which have controlled the evolution of the Archaean/Proterozoic Gawler Craton in South Australia is crucial to an understanding of the geology of the southern central Australian continent.

A particular feature transecting the southern portion of the Craton is the Polda Lineament, which extends east-west for possibly up to 1600 kilometres. Both vertical and lateral movements have been prevalent within this zone of crustal yielding from – Lower Proterozoic to Cenozoic times. The most obvious indication of the lineament is the Polda Trough, an area of deep sedimentary section. However, extensions to the lineament are now recognised eastwards across lead-zinc mineral fields in the eastern Gawler Craton, and, albeit less prominently, cutting the ?Adelaidean to ?Cambrian basement of the Murray Basin.

The Polda Lineament is now seen as a major zone of crustal yielding, with important connotations for mineral and petroleum exploration.

In reflection seismology, migration may be defined as the transformation of apparent reflector positions to their true positions. In practice this usually means that diffractions are collapsed and reflector dip angles are changed. When a medium is anisotropic, and the axes of velocity symmetry are tilted with respect to the horizontal, horizontal reflection events migrate laterally as well.

The lateral migration of reflections from horizontal layers was studied by constructing an anisotropic scale model representing a medium in which the bedding is inclined to the surface. This situation may be classified as one of transverse isotropy with a tilted axis of symmetry.

The elastic parameters of the anisotropic model were recovered by P- and SH-wave transmission measurements carried out to simulate a walk-away VSR Numerical modelling of a P-wave CMP gather above a horizontal reflector in such a medium indicated that there would be an asymmetrical distribution of reflection points. The zero-offset reflection point was displaced laterally by a distance equal to more than 20% of the depth, and the spread of reflection points was 15% of the depth. The NMO velocity was found to be a function of offset.

A zero-offset reflection survey was carried out on the model. The P-wave data were migrated with an anisotropic migration algorithm which may be applied to wavefronts of any shape. On the resulting section, the horizontal reflection features are moved laterally by a distance equal to 20% of the depth back to their correct positions.

The lateral displacement of reflections from a horizontal reflector beneath a medium with tilted anisotropic velocity characteristics has significance for the selection of drill locations when such a situation is encountered in the field.

The New Caledonia Basin is a major bathymetric feature extending northwest from Taranaki, New Zealand, between the Challenger Plateau/Lord Howe Rise and the Norfolk Ridge. Interpretations of gravity data and seismic refraction velocities indicate that crustal thinning did not proceed to formation of ocean crust. Rifting was pre-Cenozoic, and may be pre-Cretaceous. Major transverse or transform faults may have accommodated oblique rifting. Cenozoic volcanism was influenced by these earlier structures. Channels and turbidites within the Plio-Pleistocene section reveal the continuation of a submarine distributary system from the Taranaki continental shelf. Seismic stratigraphy, the volume and thicknesses of sediment observed, and the likely maturation levels suggest potential for hydrocarbon accumulations.

Due to the growing interest of community and government in conserving natural areas, together with the more general growth in environmental and conservation awareness, environmental management has become an important factor in seismic acquisition.

Public perception of the environmental effects of onshore petroleum exploration mainly relates to seismic activities. It is these techniques which have been, historically, the most visibly destructive to the landscape.

Modified acquisition methods are being implemented, particularly in line preparation procedures. These are now vastly different to those which were employed ten years ago. This paper compares the past and present methods of seismic acquisition and discusses the techniques currently being introduced by operating oil companies and geophysical contractors to obtain high quality data but with minimal impact on the environment.

The introduction of environmentally sound seismic has been gradual; however companies have reported that, with proper supervision, there has been minimal impact on the cost of operations resulting from the adoption of more stringent codes of environmental practice.

Shallow seismic refraction is the principal geophysical method for engineering site investigations. The Roads and Traffic Authority of New South Wales applies the method for roadworks, chiefly in the investigation of road cuttings for design and excavation assessment. The need to obtain reliable and detailed interpreted seismic sections, particularly for intermediate layers with seismic velocities in the range 800 to 2500 m/s, has led to an improved field practice with a 3 to 5 m geophone interval and sourcepoint intervals of 10 to 15 m with offsets up to 50 m. As a result of the deficiencies of currently available interpretation software a new interactive and graphically oriented computer program called REFRACT, for IBM AT compatible computers, has been developed. A field example from a variably weathered, basaltic area demonstrates the effectiveness of the improved field and interpretation practice which is state-of-the-art for the shallow engineering refraction method in its application to roadworks.

The data collected in a well-to-well tomography experiment are inherently incomplete even when augmented by VSP data. The nature of the experiment suggests a geometric limitation to the resolution of any central structure. In particular the Backus-Gilbert method shows that only poor horizontal resolution can be expected (Menke, 1984).

Another major constraint can now be identified within the experiment, namely the borehole itself. A parametric model has now been developed in which the borehole size, transducer standoff, damage-zones, and the degree of velocity attenuation, were examined using a range of host velocities and hole separations.

It was found that there are significant velocity variations caused by random perturbations in borehole size. These errors are particularly significant for boreholes with large diameters and small transducer offsets. For an altered zone, errors in both the dimensions and the degree of velocity alteration, gave measurable velocity variance, particularly with large boreholes. In all cases it is observed that the variance in velocities increases as hole separation is decreased.

In late 1989 the Australian Bureau of Mineral Resources (BMR) carried out a regional aeromagnetic survey by contract over the Vulcan Graben region. Elongate short-wavelength anomalies, with wavelengths of 3 to 12 km, amplitudes of 0.2 to 2.0 nT, and with generally northeast strikes, can be correlated for distances of up to 100 km on the basis of their shape and amplitude. Their source appears to be principally at depths of between 0.7 and 3.0 km. Some of the relatively larger amplitude and wavelength anomalies overlie major northwest-dipping normal faults which have been mapped by seismic. All of the elongate anomalies are thought to be due to structures or magnetic deposits associated either directly with NE-trending normal rift faults, or with faults in the post-rift sediments which are related to rift fault reactivation. ENE fault trends, which developed in the Neogene as a result of the collision with Timor, do not appear on the aeromagnetic data.

We have mapped numerous northwest-trending faults. These have a spacing of approximately 10 to 25 km. They are thought to be strike-slip because some offset the major northeast-trending normal faults. The northwest trending faults may be due to the reactivation of Palaeozoic trends, the reactivation of Mesozoic transfer faults, or possibly some may be younger and independent of the earlier faults. Some of these faults have previously been mapped on seismic and correspond to seismic ‘bad data’ areas. These faults may have a major role in the entrapment of hydrocarbon accumulations in the Timor Sea, with most of the significant fields being close to a northwest-trending fault or its extension.

In the Timor Sea, high resolution aeromagnetics have proven to be a relatively cheap and cost-effective tool which both complements and supplements the existing seismic data. It should prove equally useful in other areas with extensive seismic coverage. In frontier areas, it should rapidly define the geometry of the major tectonic elements of the basin, allowing for better positioning of seismic programs.

The standard common-midpoint (CMP) stack procedure becomes questionable as soon as varying reflector dips at about the same two-way time (TWT) are involved. There are several ways to handle that problem ranging from pre stack migration to dip moveout (DMO). This paper presents a new alternative which employs the same physical model of the medium that is in common use for migration: scatterers embedded in a known reference velocity field. Provided that backscattering can be described by the (first) Born approximation, it is possible to derive a zero-offset trace from a CMP gather by a procedure that relies entirely on known processing steps like normal moveout (NMO) and stacking.

Two reflection seismic lines across the Tertiary Rhine Graben in Central Europe were recorded in 1988 as a joint venture of the French ECORS and the German DEKORP deep seismic programs. As a main result the asymmetry of the graben, as indicated by the sedimentary fill, is accompanied by asymmetric features throughout the entire deep crystalline crust. Moho depth, lower crustal thickness and reflectivity pattern vary significantly across the graben. Extension occurred along shear zones that are documented in the upper crust and at the crust/mantle boundary.

Conventional reflection tomography attempts to reconstruct an image of the subsurface velocity field by minimising the differences between measured travel times and the travel times through the proposed image. Such schemes have been widely documented. Conventional algorithms also parameterise depths to major reflectors in the inversion, and hence necessitate that data times be measured consistently from these reflectors. In areas of complex geology, strong coherent reflectors may be hard to find. In such cases, conventional tomographic imaging would be at least difficult to work with, and may be completely inapplicable.

The angles of emission and reception can be obtained from the standard reflection seismic data to enable ray tracing for travel times without reflector parameterisation. This means any moderately strong event can be included in the inversion and allows simpler automatic data picking routines to be employed.

Any inversion that is controlled by a priori and/or smoothness constraints implicitly selects a solution that fits the user’s preconceptions to some degree. The maximum entropy image, on the other hand, ensures that the solution contains no structure other than that implied in the data. Inversions controlled by a maximum entropy criterion have already been used successfully in astronomy and medical imaging and have similar application in seismic traveltime inversion. To facilitate rapid convergence, this non-linear maximum entropy inversion incorporates subspace searching methods and stages of decreasing parameterisation scale lengths.

Recent advances in high resolution velocity estimation using first-arrivals on multicomponent offset VSP data are crucially dependent upon accurate determination of arrival direction (eg., Esmersoy, 1988). Such methods require post-acquisition orientation of 3-component data by coordinate rotation.

Rotation into the direction of maximum P-wave energy in the horizontal plane at each depth of a VSP survey reorients the coordinate system. A second rotation in the vertical plane towards the source establishes the arrival direction and, in addition, separates the P-, SV-, and SH-transmitted waves. An error in the first rotation influences strongly a separation of the wavemodes.

The desired rotation angles can be found from an expression which involves cross-products between measured components (DiSiena et al., 1981). The expression assumes there is no phase difference between these components. This will certainly be the case for an isolated P-wave arrival measured on a recording system with identical impulse responses for all three channels. However, non-identical channel responses or overlapping P-wave arrivals introduce, in turn, erroneous estimates of P-wave arrival direction.

The present study quantifies the error in arrival direction as a function of phase difference between components and looks closer into the difference between an instantaneous and an average estimate when calculating the rotation angle.

A small nature preserve in Ohio (U.S.A.) has an unusually cool environment due to artesian springs supplying cold ground water. This cool environment supports a collection of rare plants and animals. To understand the water supply so that it can be protected from encroaching development, a series of geophysical studies have been conducted over an extended period of time. The techniques include seismic refraction, seismic reflection, gravity, magnetics, and resistivity. These were supported by drilling and hydrochemical analyses.

The results of these studies show that the nature preserve sits on the edge of a major buried valley which has been filled with outwash and till. The ground water appears to flow from a high area in the northeast through an outwash layer which is capped by till. Near the nature preserve later glacial advances caused erosion of channels through the till layer. These channels were back filled with permeable sand and gravel which allowed the cold confined ground water to upwell and cool the nearby areas.

Several wells in the Timor Sea have encountered Eocene sections with anomalously high velocities. This paper presents case history examples of wells drilled for field appraisal or for up-dip exploration tests. Quantitative analysis of the seismic anomalies and the velocity effects on the time images of deeper targets is presented for each example.

The seismic anomalies observed in the Timor Sea region range in width from 200 m to 600 m and occur over strike lengths of up to 3000 m. The degree of seismic two-way time ‘pull-up’ on deeper seismic events is proportional to the width of the anomaly, to the relative depths of the anomaly and the underlying events. At Jabiru-2 the time image of the Jurassic reservoir zone is unaffected by the anomaly which is undershot by most of the CDP ray paths. Other wells, such as Avocet-2, exhibit considerable time pull-up due to broader high velocity zones. At Keeling-1 the target horizon was only partially affected by the overlying velocity anomaly.

The seismic anomalies can be recognised by a narrow zone of noise below the Base Miocene seismic marker, varying degrees of time pull-up below the high velocity limit, and local thinning of the Eocene time interval. A common attribute of the anomalies is the confinement of the high velocity sediments to the Pre-Miocene section. This suggests that the anomalies are not due to diagenetic changes associated with fluids moving along fault planes. If this were the case, then the Lower Miocene section would be similarly affected. It is proposed that the anomalies are caused by diagenetic changes related to sub-aerial exposure of topographic highs during the Oligocene.