Exploration Geophysics - Volume 21, Issue 1-2, 1990

Thirty years ago some engineering organizations expected geophysics, applied virtually in isolation, to provide cheap answers to engineering questions. After some expensive lessons it is recognized today that geophysical methods are fully effective only when used with sufficient geological understanding and calibration. Common applications most useful in engineering are in two categories. First, is the location of features with anomalous physical properties. Such features include dykes, deeply weathered zones, major fault zones, buried channels, cavities and landslipped masses, all of which are potential hazards during the construction of tunnels and dams. Basic dykes have been located by airborne magnetometer, and where forming ridges on the sea bed, by side-scan sonar. During the planning of Sugarloaf Inlet Tunnel a deeply weathered unit was located by airphoto interpretation and refraction seismic traverses. Early recognition of this unit enabled the tunnel to be located so as to pass beneath the weathered zone. Seismic refraction and magnetic methods have also been successful in locating faults and buried channels. Concealed cavities have proven more difficult to locate, despite the use of micro-gravity, seismic, resistivity, shallow EM, Sirotem and ground-radar techniques. Landslipped masses can usually be recognized by their distinctive geomorphological features, but where such features are absent, a landslipped mass might be detected by its anomalously low seismic wave velocity. The second application category is the delineation of boundaries between features with contrasting physical properties, for example between ‘overburden’ and ‘bedrock’. Where the overburden is geologically young transported material overlying unweathered rock, the boundary is readily determined by the seismic refraction method. Where the overburden comprises weathered rock in-situ, its boundary with the fresh rock is often highly irregular and gradational, and weathered profile reversals can occur. Caution is suggested in the interpretation of travel-time plots obtained from such variable masses.

The author questions the value of geophysical methods for assessing the strength and compressibility of rock masses, and the effectiveness of grouting. He proposes seven other civil and mining engineering applications as more worthy of research, viz: location of hazardous features offshore, location of thin weak seams in rock masses, delineation of fresh rock ‘boulders’ within residual soils, location of cavities, development of remote-reading instruments for monitoring rock subsidence during and after mining, assessment of thicknesses of mechanically loosened rock, and detection of leakage areas in reservoir floors.

Engineering geophysical methods are a useful site investigation tool in Civil Engineering provided they are applied by experienced personnel, and in conjunction with geological and other site investigation methods. Seismic refraction methods (P and S wave) and resistivity methods have the greatest application. Use of seismic refraction method with geological factors can allow estimation of the rippability of rock by bulldozers, but the methods available are only approximate. Contracts should recognise this limited accuracy and allow for separate rates for excavation of rippable and non-rippable rock.

The generalized reciprocal method (GRM) and two special cases, the conventional reciprocal method (CRM) and the intercept time method (ITM), constitute an integrated approach which is powerful and convenient for the great majority of exploration refraction problems. The ITM, the CRM and the GRM form a logical sequence of increasing depth of application, increasing performance and increasing complexity. They can be used as reconnaissance methods with widely spaced detectors or for detailed definition using closely spaced detectors.

These methods are well suited to good quality control practices because the data, the processed data and the final depths can be readily presented. A most important benefit of these presentations, especially for geotechnical applications, is that they help resolve whether lateral changes in the seismic velocity associated with a major change in depth to a refractor, are in fact genuine, or simply an artefact of the interpretation algorithm.

Two case histories illustrate the recognition of both fictitious changes in the seismic velocity associated with changes in the depth of the refractor, as well as genuine changes in the seismic velocity associated with a known fault.

S-wave velocities in soils to depths of 60 m typically range from 50 to 1200 m/s. These velocities may be correlated with the results of other engineering tests, such as the standard penetration test (SPT), in order to provide an independent assessment of the dynamic elastic properties of shallow earth materials. Reliable S-wave velocities in soils may be obtained in cased boreholes by vertical seismic shear-wave profiling (VSSP), which employs a surface linear traction SH-wave seismic source, downhole multi-component geophones pneumatically packed against the hole wall, and special field procedures involving reversal of the source polarization, which are designed to enhance SH waves and assist their identification.

Liquefaction assessment of alluvial materials under earthquake loading may be made independently using both SPT and VSSP. The problems of interpretation when both direct S waves travelling through the alluvial material and refracted S waves arising from the steeply dipping bedrock interface arrive at similar times can be resolved using refraction interpretation techniques.

The S-wave velocities obtained from a dam site in the Solomon Islands suggested that the sandy gravels at the site were mainly in a medium dense state, not in a dense state as was expected from the SPT. The VSSP tests indicated that these materials may be subject to liquefaction under typical earthquake loadings likely to occur in the region.

Distinctive resistivity and induced polarization (IP) logs, together with long spaced density, gamma ray and neutron-neutron logs, are presented from a test borehole in a sedimentary environment typical of that encountered in the Sydney region. The high-resolution continuous IP logs are the initial results from a more extensive study aimed at investigating the application of resistivity and IP logging to non-metalliferous environments in Australia.

The electrical logging arrays include the dipole-dipole array and the Schlumberger array for various electrode spacings. The resistivity log for the small spacing dipole-dipole array (0.25 m) shows a close correlation with the neutron-neutron log and identifies sandstone porosity variations along with the presence of conductive argillaceous horizons. Continuous logs of chargeability windows provide a pictorial display of variations in the rate of polarization decay along the hole.

Computer aided interpretation of the resistivity and IP logs for a key interval has resulted in a model of 28 layers. The theoretical logs show an excellent match with field logs. These results confirm that the dipole-dipole array provides better resolution of thin conductive layers than conventional logging configurations.

Sonic logs have been used successfully for estimating strength and static elastic moduli for coal measures strata at 18 sites in N.S.W. and Queensland, and empirical relationships between these geotechnical parameters and sonic log interval transit time are presented. Sonic logs are particularly sensitive to strength variations in weaker rocks, say less than 60 MPa compressive strength. Neutron log response is also proportional to strength, but the relationship is more qualitative. Sonic and neutron logs tend to overestimate porosity in clay-rich coal measures rock, counting the clay matrix with the air or water-filled voids.

At BHP-Utah Coal Ltd minesites, sonic and neutron logs are obtained in about twenty percent of exploration drill holes to identify the geotechnical properties of the overburden rocks for blast design and diggability assessment. Applications in blast design range from the simple identification of massive, high-strength rock units through to the semi-quantitative measurement of rock strength from correlation with sonic log transit time. At Goonyella Mine a bucketwheel excavator is used to remove overburden, and the overburden diggability is determined from sonic logs. The onset of hard digging conditions is identified by a sonic transit time of 150 μs/ft (equivalent to 492 μs/m), and marginal digging conditions occur at 125 μs/ft (equivalent to 410 μs/m). Direct correlation between geophysically determined digging resistance and monitored production data from the bucketwheel excavator shows that the sonic log method of digging resistance measurement is superior to more conventional geomechanical testing of rock cores.

Downhole high resolution seismic methods have significant application to geotechnical investigations in rock. A new down-hole and crosshole seismic method utilises sophisticated digital acquisition and processing to provide multi-fold subsurface images which show a degree of detail and resolution not attainable previously. Both first and reflected seismic waves are employed in the numerical analysis of the data, which includes beam forming and tube-wave removal.

The method shows great potential in geotechnical investigations including the definition of dykes.

Recent advances in magnetometer technology have resulted in high-definition magnetic mapping becoming a rapid and effective aid to engineering site investigation. Conventional magnetometer systems were usually too slow to be economical for mapping magnetic fields in the detail required to resolve near-surface geological and cultural structures of importance to the engineer. A new, self-contained magnetic exploration system, capable of measuring and image processing on location up to 100 000 measurements per day, is now being used in many engineering applications. Measurement densities of up to four per square metre are necessary to define the Earth’s magnetic field comprehensively in a plane half a metre above the ground surface. The instrumentation described is capable of economically acquiring and processing these data at up to two hectares per day. Geological and cultural features that may be detected and mapped using magnetics, and which are relevant to engineering site investigation, are intrusive dykes, sills and pipes, geological faults and contacts, unexploded ordnance, ferrous industrial waste and utilities, and buried archaeological artifacts. The technique has been effectively applied to engineering problems relating to mine planning, excavation site analysis, ground-water flow control, explosive ordnance detection, industrial waste relocation and archaeological site investigations.

Ground probing radar surveys can be used to image subsurface fractures in granite where weathering has contributed to the formation of oxides on the fracture surfaces. Similarly, with the use of conductive salt solutions as the fracturing medium, radar images of artificial cracks created during hydraulic fracturing can be obtained.

Geophysics plays an important role in geotechnical investigations for projects related to roadworks. Geophysical methods commonly employed are seismic refraction and electrical resistivity, and more recently terrain conductivity (EM). The Geotechnical Group of VICROADS routinely uses geophysics to investigate borrow material sites for roadworks projects. Problems may arise however through ineffective and incomplete communication between the investigating geoscientist and the end users of the information, i.e., the design engineer and especially the earthworks contractor.

Following subsidence of the Snowy Mountains Highway, due to collapse of underground cavities in limestone, electromagnetic (EM) methods were employed to try and define the extent of the cavities. Drilling investigations and some ground-radar probing indicated that these cavities were shallow (3-6 m below road surface) and occurring essentially within the soil profile. The subsequent EM conductivity survey clearly indicated the very changeable sub-surface materials profile, being an intricate array of limestone pinnacles alternating with deep troughs of Terra Rosa soil. It is concluded that such surveys are useful for the above stated purpose, but only to a limited degree, and must be supported by other methods of investigation. That is, the use of this geophysical method will not provide definitive answers to the size and shape of cavities, but it will indicate their probable presence.

Engineering requirements for project development should be met by a combination of investigation methods including both geology and geophysics. The integration of different techniques may lead to successful analysis in situations where separate methods would provide misleading results.