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

Four techniques are compared for estimating the transfer function between fields measured at one site with fields measured at another site, or with different fields measured at the same site. The techniques are:

1. Linear, point-to-point time domain estimation (LPP),
2. Conventional frequency domain transfer function estimation (FD),
3. Multichannel Wiener estimation in the time domain (WE), and
4. Adaptive filter processing in the time domain (AFP).

These four transfer function estimation methods have been applied to high precision SQUID magnetometer array data collected in Grass Valley, Nevada and Long Valley, California, using natural fields in the frequency band 10^-4 to 1.0 Hz. To monitor the ‘signal’ introduced by motion of the sensors in the Earth’s large static field, high sensitivity cryogenic tiltmeters were used in each magnetometer. The transfer function estimation methods above have accounted for more than 50 db of the signals received at one site using fields measured at another site. Removal of motion-induced magnetic signals provides as much as 10 db further reduction in the residual signal compared to using solely electromagnetic field signals.

An understanding of the subsurface distribution of intrusive rocks is very important for geothermal exploration since they are often related to fracture zones suitable for geothermal fluid reserves while also acting as heat sources to the reserves.

Concealed intrusive rocks may be estimated by analysing magnetic anomalies. However, the effect of topography must be carefully removed from the observed magnetic anomalies in volcanic areas because surface and shallow volcanic rocks produce noise blanketing deeper information. The Topographic Effect Removed magnetic map (TER magnetic map) was developed based on the magnetisation measurement of rock specimens, geologic data, and CSAMT data in the Hakkoda geothermal field located in the northern part of mainland Japan. The intrusive-shaped magnetic bodies were then extracted using the TER map. Drilling holes which hit the Quaternary intrusive rocks were not available. Therefore, we can not determine whether the intrusive-shaped bodies directly correspond to actual intrusions, although many hot springs occur around the bodies. This may confirm the hypothesis that the bodies correspond to intrusions associated with Quaternary volcanic activities in Hakkoda district.

The estimation and removal of the propagating wavelet in seismic data is essential to obtaining a high-resolution estimate of the earth reflectivity. A well-known approach is to assume that the reflectivity is characterized by a relatively sparse spike train and then attempt to select the deconvolution filter so as to maximize some measure of sparseness (spikiness) of the deconvolved trace. One example of this is the minimum entropy deconvolution (MED) method of Wiggins (1978). However, the spikiness deconvolution approach is not very effective in the presence of noise. The reason for this is that the best a linear deconvolution filter can do for bandlimited data in a noisy environment is to produce the reflectivity convolved with a bandlimited zero-phase waveform. Unfortunately, interference patterns generally preclude us from obtaining this output with a wide-band spikiness criterion since the desired output trace is not recognized as being spiky even though the underlying reflectivity is spiky.

In order to solve this problem, we have developed a new spikiness criterion which can be defined over a given frequency band (bandlimited spikiness). We demonstrate the usefulness of this criterion by showing how it can be used to estimate the wavelet phase for bandlimited data under the assumption that the phase is a constant, independent of frequency.

Field studies for testing the usefulness of geophysical methods in the mapping of karst in the northern part of the ‘Rheinischen Schiefergebirge’ yielded the following results:

(1) Geoelectrical resistivity profiling, gravimetry and refraction seismic can be used for the detection of karstified zones. Low apparent resistivities, low p-wave velocities and negative gravity anomalies characterise the loose fill material or highly jointed limestone in these zones.

(2) Rapid detection of karstified zones can be carried out with the resistivity profiling technique. The approximate extent of these zones can be mapped. The double dipole (linear) array gives the best results.

(3) A more exact determination of the size of the karstified zones can be carried out using refraction seismic and gravimetry. Gravimetry enables an approximate volume estimate whereas refraction seismic allows a more precise volume estimate as well as the determination of the internal structure, nature of material present and boundaries to non-karstified areas.

The method of post-stack wavelet extraction utilizing well log data has been described by Ramsden and Hobson (1987) and Danielson and Karlson (1984). This paper presents the results of the application of this method to two vintages of vibroseis data from the Eromanga Basin, Queensland, Australia. A perennial problem in the mapping of seismic data, recorded over the same locality, is the character and timing mismatch often found between different vintages of data. Utilizing well logs a post stack wavelet is extracted from each vintage of seismic data, and used to deconvolve each vintage’s wavelets to zero phase.

In the area of interest it is found that the wavelets on the two datasets are out of phase by approximately 70 degrees. A wavelet deconvolution operator is designed to correct the data to zero phase. This improves the character and timing misties between the different vintages of data and also improves the vertical resolution of the seismic data. The stability of the seismic wavelet over the survey area is demonstrated by extracting the wavelet from the two wells which are separated by approximately 50 kilometres.

The well known problem with seismic tomography is that limitations of source receiver geometry mean that seismic ray coverage is incomplete. This is one cause of artifacts that disturb reconstructed velocity distribution images. In this paper, we attempt to reduce the occurrence of artifacts by means of the damped least squares method.

In general, tomographic velocity models are obtained by dividing the area of exploration into a number of rectangular cells, and assigning slowness values to each cell. The reconstruction procedure generally used is the iterative method, in which ray tracing and modification of the model are carried out alternately.

The damped least squares method is one way to obtain stable least squares solutions. It is an attempt by the authors to take into account irregularity of ray coverage by determining damping parameters for each cell from the number and directionality of rays passing through each cell.

Numerical experiments showed that if appropriate damping parameters are selected, the damped least squares method can reduce the occurrence of artifacts. The method was also applied with good results to field measurements, taken at a dam site in an area of granite.

This paper presents some field examples of using borehole electrical surveys to search for blind metals deposits, and demonstrates the value of borehole electrical surveys.

In a magnetic intrusion type Ni deposit, downhole SP found an off-hole orebody. A 3-D model was set up and used to find a new orebody. Downhole IP surveys have been carried out in several metal deposits, with good results. Downhole IP, resistivity, PEM, EM wave and applied potential methods have been used to solve different geological problems.

These case histories illustrate that under-ground synthetic electrical surveys not only have more power to detect off-hole ores, but also can provide abundant information for determining geological structure, 3-D mapping and ore genesis. During exploration underground synthetic electrical survey can save drilling effort, decrease the time of exploration and reduce the cost.

If the conductivity of the ground increases with frequency (i.e. the ground is polarisable) sign reversals can occur in coincident-loop transient-electromagnetic (CLTEM) measurements. The CLTEM response of a polarisable ground is comprised of two parts: the positive fundamental inductive response, and a weaker, negative, polarisation decay. The time dependence of the polarisation decay can be calculated using a convolution approximation.

CLTEM sign reversals (negatives) occur when the polarisation decay becomes greater than the fundamental inductive response. For most conductivity structures, negatives will not be observed unless the IP chargeability of the ground is exceptionally large. However, in special circumstances the fundamental inductive response is particularly small, and thus negatives can be produced by conducivity structures with geologically feasible chargeabilities. Examples of such circumstances are: in localised zones between inductively interacting conductors, at the edge of overburdens, and over relatively resistive grounds which have a thin polarisable surficial layer.

The polarisation current in a horizontal layer causes the vertical field response to be enhanced outside the transmitter loop and depressed inside the loop. When the response due to the polarisation current is measurable, but not large enough to cause negatives, this produces a situation known as ‘the loop effect’.

One broad objective of geophysics is to produce subsurface images which represent geology as accurately as possible. The goal in this paper is to construct pictures of lithology in depth. An overview of a tomographic method (traveltime inversion) is presented here which uses well-logs, VSP and surface seismic data to produce a velocity structure in depth. This velocity structure is then used in the surface seismic inversion process and depth migration. Thus, the described processing flow attempts to provide accurate stratigraphic and structural depth pictures.

Two examples of oil field development plays in Alberta are processed in the above manner. A brief discussion of the geology, data acquired and interpreted results is presented here. It is found that tomography-based imaging can provide seismic depth sections which are consistent with well-logs in the regions under consideration. Previous attempts to tie the seismic sections into depth, without VSP and tomographic analysis, had proved inaccurate. Depth sections and tomography-constrained, pseudo-sonic sections assist in the structural and stratigraphic interpretation of the reservoirs in the two cases.

This paper presents a method that uses polarisation characteristics of three component seismic data (recorded by surface reflection profiling, VSP, etc.) for discrimination and filtering of waves.

Discrimination of waves in single (e.g., vertical) component data is based on frequency characteristics and apparent velocities, obtained from line-up observations. When three component records are used, discrimination based on polarisation characteristics is also possible.

In this paper, we first describe the method by which polarisation characteristics are determined from three component records, and how these characteristics are used in polarisation filtering. Polarisation characteristics are estimated by eigenvector analysis of covariance matrices, which are obtained from three component data. The parameters thus obtained are then used in polarisation filtering.

Following this is a discussion of the method. In this discussion, three component data, synthesized from waves having various polarisation characteristics, are used.

Finally, we present an actual application of this method to data obtained in a surface observation. This example demonstrates that by the use of polarisation characteristics, it is possible to (1) discriminate linearly polarised waves (P or S waves) from elliptically polarised waves (Rayleigh type surface waves); and (2) distinguish between P waves and S waves, on the basis of differences in particle motion direction.

The time-domain electromagnetic response of a polarizable sphere in a non-polarizable host rock is influenced more heavily by the chargeability m than by the time constant τ, in the range of delay times commonly observed. These results are obtained from a new mathematical model of the sphere and halfspace system, based on the earlier work of Lee (1983).

Petrophysical sampling of the steeply plunging, roughly pipe- like Elura Zn-Pb-Ag orebody demonstrates a dominantly remanent magnetisation with steep negative inclination; the magnetisation is carried by central monoclinic pyrrhotite lenses. The remanence data suggests a mid-Cretaceous low grade thermal overprint possibly related to final cooling associated with minor uplift. Susceptibilities are high, averaging about 4200 × 10^-6 cgs (about 52 800 × 10^-6 SI), and markedly anisotropic with k1:k2:k3 = 1.6:1.2:1.0. The anisotropy reflects a preferred crystallographic orientation of pyrrhotite grains — the basal planes are vertically oriented suggesting plastic flow or crystal growth in a vertical direction. The high susceptibilities coupled with a Koenigsberger Ratio of 9 4 produce a ground magnetic anomaly 140 gammas (nT) in magnitude. This anomaly can be simply and correctly modelled by vertically plunging ellipsoids using equations that incorporate remanence, anisotropic susceptibility, and demagnetisation.

Two radar systems were used to conduct tests in a European salt mine. Both systems obtained very good results. An electromagnetic wave speed of 122 m/μs in salt was measured in this particular mine. Using the first system, Bravo, for long range probing, several areas that appeared to contain many anhydrite stringers and other areas that appeared to have good, clean salt were found. Bravo obtained a maximum range of 592 m. At a few locations the Bravo antennas were moved in azimuth in order to pin-point the location of a particular reflection. At one of these stations, a reflection that corresponded to a known anhydrite stringer was located.

With the second system, Foxtrot, which was used for short range, high resolution tests, a primary and a multiple signal were recorded from a known anhydrite layer below the mine’s salt floor. At another station, Foxtrot obtained a maximum range of 47 m. The salt used for the Foxtrot tests was not as pure as in other locations within the mine. The authors believe that Foxtrot tests at other locations might yield probing distances up to 70 or 80 m.

Salinity is a serious problem in farmlands of Western Australia where the agricultural land lost due to salinity is estimated at over 300 000 hectares while the annual increase in salt land is in excess of 25 000 hectares.

Using a multidisciplinary approach, geophysics is applied to economically establish the geology and hydrology that control the salinity, and then hydrological and engineering solutions are implemented for the land reclamation.

In 1984, a Cranbrook property of 1297 hectares had 240 hectares of salt-affected land although when the property was purchased in 1967 there was no evidence of water-logging or salinity. Magnetic and electrical sounding surveys indicated a series of dyke formations which obstructed the main ground-water system causing an increase in pressure in the system. This is one of the major causes of salinity in Western Australia. Prior to agricultural development, the groundwater system was in equilibrium. However, with clearing of native vegetation and replacement with species that use less water, it has caused a hydrological imbalance, resulting in the accumulation of salt in the top 2 metres of soil.

The engineering and hydrology solution consisted of water pumping and gravity feed. After 2 years the salt-affected land has been reclaimed to top production.

Measurements of magnetic susceptibility and natural remanent magnetisation of representative rocktypes from the eastern Arunta Inlier reveal the majority of lithologies have distinct magnetic properties. Studies of opaque mineralogy and bulk rock chemistry demonstrate that magnetic rock properties are dependent on total iron content and the manner of its division between Fe, Mg-silicates and the oxides of iron and titanium. The partitioning of iron is controlled primarily by oxygen fugacity and titanium content. All these parameters exhibit a well-defined dependence on weight percent SiO2, and as a result, the magnetic properties of most major rock groups are related to SiO2 content. This relationship appears to be inherited from tectonic environment in the case of igneous rocks and depositional environment in the case of sedimentary rocks. Exceptions are usually due to late-stage alteration involving hydration or dehydration resulting in a change of oxygen fugacity.

The aeromagnetic signature of individual lithologies is consistent with their measured magnetic properties and many may be regarded as magnetic litho-stratigraphic marker horizons. Four major rock groups having distinct signatures are recognised. These comprise a predominantly meta-igneous basement complex, a metasedimentary, metavolcanic cover sequence, igneous bodies which intrude basement and cover groups and narrow linear features related to mylonitic and retrograde shear zones.

Upward continuation is normally carried out by convolving grid operators with the data. Continuation height is restricted to units of data intervals and much information is lost around the boundaries.

Continuation can be carried out in the frequency domain in which case no information is lost around the edges and continuation height is not restricted.

In the case of aeromagnetic surveying where high density data are normally collected along flight lines 10 or 20 data units apart, it is desirable to use a one dimensional operator. This avoids interpolation problems, but assumes infinite strike. This paper describes the results of model and field tests designed to investigate the limitations of one dimensional upward continuation in the frequency domain.

The field tests show that in practice, the technique is less restrictive than indicated by model tests. For routine surveys one dimensional continuation is acceptable.

The model tests show that depth to bottom of source is important and a body may have an infinite response at one level but a finite one at another. This casts doubts on the need for drape flying.

Velocity analysis in current conventional seismic processing is beginning to appear inadequate for dealing with the increasingly complex geological structures of interest today. Tomographic imaging using individual arrival traveltimes from a given reflector may offer the prospect of reconstructing in detail an accurate velocity field above that reflector, and, by extension, the velocity field in a layered medium. Bearing in mind the likely scale of this inverse problem we formulated it as a least-squares optimisation and derived a descents-type algorithm; synthetic tests in which the reflector was assumed to be known demonstrated its potential for recovering lateral variations in particular.

However the more realistic problem in which the reflector is unknown and presumed to have structure on a scale similar to the velocity field, necessitating its inclusion in the inversion, is less tractable. As one might intuitively have expected, there are reflector depth trade-offs with near-reflector slownesses resulting in indeterminacy with severe implications for the usefulness of the reconstruction and which cannot be simply resolved. The robustness of excessively overdetermined problems suggests a multi-stage inversion in which the scale of model parameterisation is successively reduced until e.g. no further gain in resolution is achieved. Synthetic tests here suggest that while such schemes may yield significant improvements, they are dependent on the actual early-stage (large-scale) parameterisations used. Fortunately it turns out that we can use multi-stage, decreasing scale-length smoothing to achieve similar effects without such dependence.

Modern petroleum exploration, incorporating routine acquisition of geophysical logs, began in Australia in the mid 1950s.

Prior to 1976, data were recorded in analog form. By 1982, digital recording was routine for most logging units, and since 1985 few operators have accepted non-digital records.

Approximately 60% of the log data acquired to date in Australia is thus analog. Varying standards of recording practice, evolving hardware and changes in drilling practice have all affected the quality of analog records, and their ease of use. The economics of use are often highly dependent on record quality.

In 1981 Wiltshire Geological Services began systematic high fidelity conversion of analog data to digital. Our Initial objectives were to develop tools for better and more efficient analysis of sedimentary sequences and basin evolution.

By 1985 the developing data set was clearly a major asset. Since then we have built the data set as rapidly as possible whilst maintaining the highest possible data conversion standards.

Resulting benefits include:

data preservation is assured.

data are standardized and enhanced by editing of noise.

data reproduction is easier and cheaper.

end-user working material is routinely tailored to specific job needs.

worker efficiency is dramatically improved.

Intangible benefits may arise from better interpretive use of the data in exploration.

Tomographic methods are used to construct velocity images from measurements of seismic traveltime. Most algorithms make the assumption that seismic energy propagates along straight raypaths. This assumption fails significantly when velocity contrasts are 10–20% or greater. If, however, the rays are allowed to bend, the inversion problem becomes nonlinear as the ray paths are dependent on the unknown velocity distribution. This problem can be addressed by iteratively applying ray-tracing and tomographic inversion, attempting either: (1) to estimate the actual curved raypaths; or (2) to estimate the actual straight-line integrals. An advantage of this second approach is that it enables the use of standard tomographic reconstruction algorithms such as Fourier techniques with are based on straight raypaths. In this paper, we show how the second approach can be incorporated in a Fourier reconstruction algorithm, applying the technique to synthetic data.