Applying geophysics to mining geology problems

A R King1

1. Geophysicist, CSIRO, Kensington WA 6151. Andrew.king@csiro.au

ABSTRACT
Although geophysics is routinely applied to mineral exploration and has been responsible for most
large mineral deposit discoveries in recent years, its use in mining is still limited. However,
geophysics has two major advantages: First, in contrast to results from borehole drilling which are
point or line measurements that only sample a small fraction of the rock, it is able to produce three-
dimensional images of the sub-surface; Second, most geophysical measurements are in fact
effectively volume averages of rock properties, and so are less affected by spatial variation than are
the results from boreholes.
While the physical properties that geophysics is able to image might not always be directly relevant
to a mining engineer, they are often correlated with properties of interest. A relatively small number
of samples can be used to derive a relationship between the physical properties measured by
geophysics and quantities relevant to mining. This allows full three-dimensional inferences to be
made about the desired parameters.
These principles are illustrated using the examples of electrical and seismic properties. The
conductivity of the ground, obtained using electromagnetics, can used to infer information about
groundwater amount and quality, as well as geology. Seismic velocity images obtained both using
active sources and from passive seismics, where images are generated using seismic “noise”, are
used to infer rock stress, and other geotechnical information.
Use of geophysics to improve the interpolation of ore-grade values or impurity content between
grade holes, or to yield information about rock hardness, crushability and blastability to allow
optimisation of blast-hole drilling for example, could potentially have significant financial implications.

INTRODUCTION
Efficient mining requires the separation of ore from waste rock, and rock fragmentation into sizes
appropriate for excavation, transport, and processing. Well-designed blasting, informed by good rock
mass characterization, is therefore crucial. Current methods for gathering rock mass information to
delineate ore bodies and inform blast design rely heavily on sparse exploration data, interpolation
between ore-grade drill holes and crude blast hole cuttings assessment.
Better characterisation of the rock mass would allow improved blasting via optimisation of blast-hole
drilling and explosive charge distribution, based on better ore grade and rock blastability maps.
Fragmentation control affects processes such as excavation, loading, hauling and processing, as
well as allowing better ore-waste separation. Improved blasting potentially results in ore upgrading
through better size-based separation of ore and waste, improved material transfer and handling, and
more productive digging and loading, which would result in less machine wear. There are potential
gains from improved blasting of tens of millions of dollars per year per mine site.
Another issue in both open-pit and underground mines is the presence of unforeseen geological
structures such as faults, whose presence can have significant geotechnical consequences. Good
advance mapping of such structures allows a mine to be planned in such a way as to mitigate
negative effects.
Geophysical methods, while being used extensively for mineral exploration, are still not routinely
used for these kinds of mining problems. While some of the rock parameters of interest for blast
design, such as rock hardness or blastability, may not be directly measurable using geophysics, they
are typically corelated with other properties, such as seismic velocity, which can be measured.
Geophysical imaging therefore potentially provides a means of obtaining improved maps of
properties relevant to mine blast design.

Mining Geology 2019 / Perth, WA, 25-26 November 2019 375

GEOPHYSICAL METHODS
The most appropriate geophysical techniques to use for mining operations planning purposes will
be site dependant, because of the different physical properties of different orebodies and host rocks.

Seismics
Seismic waves in the earth respond to bulk modulus, shear modulus and density of the rocks. Any
variation in these properties will therefore be detectable using seismics. Metalliferous ores, for
example, tend to be denser and have higher velocity than most other rocks. Different seismic
techniques exist, exploiting different aspects of the seismic wavefield, and having different costs and
different resulting resolution: reflection seismics images reflections off of velocity and density
contrasts; refraction seismics uses first-arriving waves, resulting in a lower-resolution velocity image;
surface-wave dispersion techniques use the frequency-dependant velocity of surface waves to
produce depth sections of seismic velocity.
Most seismic methods use active sources, such as explosives, weight drops, or vibroseis, but
techniques also exist for doing passive seismics without these sources. Multichannel Analysis of
Surface Waves (MASW) and Microtremor techniques can use surface waves generated by road
traffic, for example (Louie, 2001; Park and Miller, 2012). Seismic while drilling uses energy from a
drillbit as a seismic source (Poletto, 2000; Rector and Marion, 1991), and mining machinery has also
been used as a source for seismic tomography (King and Luo, 2009). Cross correlation of seismic
noise can produce “virtual sources” – the signal that would have been produced had a source been
present at one of the receiver locations (Roux et al., 2005; Ryberg, 2011). These passive methods
could be advantageous for mining applications, because they are operationally easier to implement.

Electrical Methods
There are various electrical geophysical methods, which respond to the rock’s electrical conductivity
and polarizability. These include electromagnetics, resistivity, and induced polarisation. With the
exception of a few rocks such as massive sulphides and graphite, most minerals are poor electrical
conductors. The conductivity of most rocks is due to the conductivity of water in pore spaces, which
means that conductivity is determined by porosity, and by the conductivity of the pore fluid. The
presence of clay minerals can substantially increase the conductivity. For these reasons, basement
rocks, which tend to have low porosity, have much lower conductivities than the overlying weathered
material, which has higher porosity and, usually, the presence of clay minerals.
Electromagnetics has been used extensively in the exploration for massive sulphide orebodies (see
e.g. (Legault, 2015) for a review) but has also seen application in detailed shallow work for
groundwater and regolith studies (Lawrie et al., 2015; Munday et al., 2001).

INFERENCE OF PHYSICAL PROPERTIES VIA CORRELATION

Often, the properties of primary interest will not be directly measurable using geophysics. But in
these cases, it is often possible to find a proxy measurement, or a combination of other physical
properties, which can be used to predict the quantity of interest. For example, rock blastability is
influenced by a combination of rock strength and the density and orientation of joints and fractures
in the rock. Methods have been developed to estimate rock strength from seismic properties
measured down-hole and from reflection surveys (Hatherly et al., 2010)
Two examples are shown below, where geophysical imaging of one property can be used to make
inferences about another property once a physical relationship has been established.
The first example shows how mining-induced microseismic events can be used as sources for
constructing a velocity tomogram image. Figure 1 shows the reconstructed velocity in a sandstone
layer above a coal seam which is being mined. The velocity was determined by simultaneously
inverting picked travel-times at an array of seismic sensors for both the locations of the seismic
fracture events, and the velocity structure of the rock (King, 2013). Since this is a reasonably
homogeneous sandstone, variations in velocity are thought to be caused by different stress states,
and this interpretation is consistent with what is seen in the figure: mined-out panels have caved and
destressed, and are associated with lower velocities. Pillars between the panels are intact, and are
carrying the weight of the overburden, and so are under high stress. These regions are associated

Mining Geology 2019 / Perth, WA, 25-26 November 2019 376

with high velocities. So this is an example where a velocity image can be used as a proxy for a stress
image. A sufficient number of measurements of stress (or even values from numerical modelling)
could, in principle, be correlated with the seismic velocity in order to make quantitative predictions.

Figure 1. Seismic velocity tomography using mining-induced microseismic events as sources.

Since seismic velocity is correlated with rock stress, the stress is being indirectly imaged, which is
useful for geotechnical purposes. Velocity is coloured from blue (low) to red (high). The mine
gateroads and pillars are shown in grey. Circles = seismic events; triangles = seismic sensors.

A second example, illustrated in figure 2, shows the use of airborne electromagnetic (AEM) data for
inferring information about aquifer properties. While not directly mining-related, the same kinds of
ideas could be transferred to a mining application.
AEM data can be used to determine the conductivity of the ground as a function of depth. Figure
2(a) shows an example, where the conductivity of the ground at the depth of an important aquifer is
shown. The conductivity is effectively mapping the groundwater salinity: blue, low-conductivity
regions correspond to fresh water due to groundwater recharge from the river on surface; more-
distant regions are coloured red, indicating high conductivity associated with more-saline
groundwater. The relationship between conductivity and salinity is illustrated in figure 2(b), which
shows salinity values measured in groundwater wells across the survey, plotted against the
conductivity determined from the AEM. A best-fit correlation line from this figure is then used to
convert the conductivity map of figure 2(a) into an inferred salinity map, shown in figure 2(c). As a
final example, aquifer thickness, also inferred from the AEM data, is combined with the salinity map
to produce an estimate of the total salt load, shown in figure 2(d).

CONCLUSIONS
The use of geophysics for improved mapping of rock mass properties ahead of mining could result
in significant cost reductions. In particular, optimisation of blasting could be improved by better maps
of ore grade, rock blastability, and presence of impurities. Better fragmentation and ore/waste sorting
results in savings in excavation, transport, crushing and processing.
Although the properties of interest, such as blastability, may not be directly mappable using
geophysics, it is likely that they are correlated with other properties which can be measured, as has
been illustrated with examples of velocity - stress and conductivity - salinity relationships. Given such
a correlation, only a few calibration measurements are required in order to use such a proxy to infer
values of the desired parameter, and produce maps across the mining block.

Mining Geology 2019 / Perth, WA, 25-26 November 2019 377

 a) b)

c) d)

Figure 2. Example showing aquifer characterisation using AEM data. (a) conductivity of the aquifer
layer; (b) cross plot of conductivity with salinity measured in boreholes; (c) inferred salinity using
the cross-plot best-fit line as a calibration; (d) total salt content using (c) along with aquifer
thickness, also from AEM data.

REFERENCES
Hatherly, P., Medhurst, T., Zhou, B., 2010. The geophysical strata rating and its derivation from geophysicial logs and
seismic reflection data, in: Sydney Basin Symposium, pp. 1–10.
King, A., 2013. Imaging using mining machinery as a source. ASEG Extended Abstracts 2013, 1–4.
King, A.R., Luo, X., 2009. Methodology for tomographic imaging ahead of mining using the shearer as a seismic source.
Geophysics 74, M1–M8.
Lawrie, K., Christensen, N.B., Brodie, R.S., Abraham, J., Halas, L., Tan, K., Brodie, R.C., Magee, J., 2019. Optimizing
Airborne Electromagnetic (AEM) Inversions for Hydrogeological Investigations using a Transdisciplinary Approach.
ASEG Extended Abstracts 2015, 1–4.
Legault, J.M., 2015. Airborne Electromagnetic Systems–State of the Art and Future Directions. CSEG Recorder, June
2015, 38-49.
Louie, J.N., 2001. Faster, Better: Shear-Wave Velocity to 100 Meters Depth from Refraction Microtremor Arrays. Bulletin
of the Seismological Society. 91 (2), 347–364.
Munday, T.J., Macnae, J.C., Bishop, J., Sattel, D., 2001. A geological interpretation of observed electrical structures in the
regolith: Lawlers, Western Australia. Explor. Geophys. 32, 36–47.
Park, C.B., Miller, R.D., 2012. Roadside Passive Multichannel Analysis of Surface Waves (MASW). J. Environ. Eng.
Geophysics 13, 1–11.
Poletto, F., 2000. A blind interpretation of drill-bit signals. Geophysics 65, 970–978.
Rector, J.W., III, Marion, B.P., 1991. The Use of Drill-Bit Energy as a Downhole Seismic Source. Geophysics 56, 628–634.

Mining Geology 2019 / Perth, WA, 25-26 November 2019 378