Exploration Geophysics - Volume 38, Issue 4, 2007

Since their introduction by Nigel Anstey and Tury Taner in the 1970s, attributes have become an integral tool in the interpreter’s arsenal. At present, as emphasised by Taner, no direct relationships have been established between all attributes and physical and geological characteristics of the subsurface. Their discriminatory properties, however, allow very useful classifications to be performed. This paper deals with various attribute-related issues.

First, we consider the theoretical and physical aspects concerning instantaneous attributes, particularly instantaneous phase. This attribute is of central importance since it describes the location of events in the seismic trace and leads to the computation of other instantaneous quantities. Second, we deal with the issue of information content. It has often been implied that attributes convey no more information than that present in the original seismic trace from which they are derived. This, however, is akin to claiming that David contains no more information than the raw marble from which Michelangelo freed him. A seismic attribute section provides that much more information. The attribute in time attempts to enhance resolution, whereas the attribute property in the spatial dimension emphasises continuity. These important and interesting issues will be dealt with theoretically and by example. Finally, we present and illustrate by synthetic and real data examples, a novel, hybrid attribute which has been constructed to provide high resolution information. We must point out that, as is always the case, our attribute is dependent on the phase of the source wavelet.

A classification method based on the use of seismic attribute pattern recognition by means of Hopfield Neural Networks is presented. The method is suitable for exploration projects and it can be used to simultaneously perform the analysis of several references or classes and for constructing seismic similarity maps or volumes. First, a brief description of Hopfield Neural Networks and their operational principles is presented, and the most relevant issues of the proposed classification methodology are described. Then, the method is demonstrated by using a synthetic dataset and two real case studies, illustrating the potential of the method as a useful tool for exploration geophysics and reservoir characterisation.

The spectral-element method provides an accurate alternative to the finite-element method for modelling elastic waves in anisotropic media. With the aim of reducing the high computational overheads of 3D modelling, we have implemented 2.5D spectral-element modelling of elastic waves, initially for vertically transversely isotropic media, and then extended to a tilted transversely isotropic medium with a dipping symmetry-axis. We have investigated different categories of absorbing boundaries to minimise artificial boundary reflections, including viscous boundary conditions and two distinct 2.5D formulations of perfectly matched layers (PMLs). Both PML methods use complex coordinate stretching, but the first method is only applicable in the frequency-domain. The other method uses the decomposed gradient operator and can be applied in both the time-domain and frequency-domain. Traditional 2D modelling exhibits cylindrical wave propagation characteristics from a line source. Our 2.5D modelling results, presented as images of seismic wavefields, illustrate the more realistic spherical propagation from a point source.

Seismic deconvolution technology is important in seismic data processing. We propose a new blind deconvolution algorithm for simultaneous wavelet estimation and deconvolution of seismic data. Optimal seismic deconvolution can be achieved by the combined application of a non-white reflectivity model and the blind deconvolution method. We incorporate a scaled Gaussian noise (SGN) model of seismic reflection coefficients into the seismic blind deconvolution method. The SGN model has the property of scale invariance. We present a framework to generalise the conventional deconvolution procedure to handle reflection coefficients that do not follow the white-noise model. Reflection coefficients are assumed to have an autocorrelation function with a power spectrum roughly proportional to some power of frequency.

The seismic blind deconvolution method consists of three steps: (1) selection of initial reflectivities, (2) determination of the hyper parameters of the problem, and (3) an iterative procedure to solve the resulting equations until a tolerance criterion is satisfied. The alternative relaxation solution and pre-conditioned conjugate gradient algorithm are employed for practical numerical implementation. Seismic blind deconvolution not only can better realise the simultaneous evaluation of the seismic wavelet and the reflection coefficients, but also has advantages of stable algorithm and fast convergence. The test shows that the method is an effective tool of improving the resolution of seismic data that can effectively broaden the useful band of records.

Major exploration difficulties occur when the distinctive characteristics of ore targets are obscured or modified by physical property variations within the regolith. Improved exploration success in the highly variable Australian regolith requires greater understanding of this medium which can only be achieved with new or improved exploration technologies.

The Cobar District of New South Wales is one of the world’s most active mineral exploration regions with a variable regolith. This region has yielded substantial mineral wealth and the Elura ore body is one of the largest massive sulphide deposits to have entered the Cobar regolith. Previous pre-mining, shallow seismic refraction data over this ore body have been re-interpreted using visual interactive ray tracing and wavepath eikonal tomography. This improved interpretation approach has been integrated with the most recent geological knowledge, weathering history and the seismic properties of the shallow ore and host rocks to refine the seismic characteristics of the Elura ore body and regolith. The interpretations have confirmed the earlier qualitative interpretation that the Elura gossan and the altered ore zone form a local, low-velocity plug extending to a depth of ~100 m within the shallower, higher velocity weathered and fresh siltstone host rocks. The margins of this plug are well defined in the refraction interpretation as they form strong seismic wave diffraction sites at the base of the surrounding regolith. The base of this plug, representing the altered massive sulphide ore, also tends to have a lower seismic velocity than the fresh host rocks. Velocity information on the deeper gossan, supergene zone could not be obtained directly from the first-arrival seismic data as this region is laterally hidden. It is clear from this interpretation that the base of the regolith over the Elura ore body and margins are highly irregular and not well represented as a single continuous refractor as required by less sophisticated refraction interpretation approaches.

This case study shows that detailed seismic refraction, supported by improved interpretation techniques and petrophysical testing, provide detailed regolith information and have increased exploration potential for massive sulphide targets that enter or are close to the regolith.

Three-dimensional (3D) high-susceptibility tabular models of Banded Iron Formation (BIF) were built for the area of an aeromagnetic survey at the Musselwhite Mine, Ontario. The study used all available drill information to create geologically plausible models built from a central axis of nodes with geometrical and susceptibility attributes assigned to each node. The forward total magnetic intensity was then computed using an algorithm that includes a calculation for the demagnetisation effect. The model-building process greatly enhanced the understanding of the BIF geometry and demonstrated the importance of the interaction of these strongly magnetic units with the Earth’s magnetic field. As BIF is associated with mineralisation at the mine, the interpretation of the observed magnetic anomalies has strong implications for target generation.

Heat flow has been measured in south-east South Australia at a spatial resolution greater than previously available. The study area contains the Southern Murray Basin, Padthaway Ridge, and Western Otway Basin. An extensive network of groundwater observation wells across the study area was used, along with several petroleum wells in the Otway Basin, to measure thermal gradients and calculate 24 new heat flow values.

Geothermal gradients were either measured directly using a cable, winch, and thermistor in cased or open holes with standing water, or by estimating average geothermal gradients from petroleum well completion temperature data. Thermal conductivity values were measured directly on existing core samples using a divided bar apparatus.

A map constructed from the measured values reveals a heat flow dataset that is non-uniform and variable over relatively short distances. Measured heat flow values range between 42 and 123 mW/m^–2. In particular, a 40 km long, 15 km wide zone of elevated heat flow is identified along the northern margin of the Otway or Gambier Sub-basin that may correspond to the occurrence of Delamerian granitoids units of the Padthaway Ridge. High-resolution surface heat flow mapping provides valuable data for further research into the tectonothermal evolution of south-east Australia.