Development of high resolution 3D J

o
vertical seismic profiles u
r
by C.C. Pretorius*, M.A. Gibson†, and Q. Snyman‡ n
a
l

P
Introduction a
Vertical seismic profiles (VSPs) are purely p
defined as measurements made in a vertical e
wellbore (borehole) using geophones inside
Synopsis the wellbore and a source at the surface near
r
3D seismic surveys have become a well-accepted method of imaging the well (Wikipedia 2010). In the more general
sub-surface geology in the Bushveld Basin. Because both the context, VSPs vary in the well configuration,
seismic sources and receivers are placed at surface, most surveys the number and location of sources and
are limited to imaging faults with >=7 m throw. However, some geophones, and how they are deployed. In
areas such as shaft infrastructure can benefit from the ability to their most basic application, standard VSPs,
detect smaller structures prior to shaft sinking. Borehole radar and also known as rig-source or zero-offset VSPs
logs such as the acoustic televiewer can identify these structures in (Figure 1 a), fulfil a fundamental role of tying
the shaft barrel, but not within ~200 m from the shaft. Therefore, reflectivity to stratigraphy and providing
Anglo Platinum has developed the method of high-resolution 3D
interval velocity data for time-to-depth
vertical seismic profiles (VSPs) for structural sterilization of shaft
conversion of seismic data. VSPs enjoy a
sites prior to shaft sinking. VSPs utilize a surface source and sub-
surface receivers deployed down a borehole. This means that natural advantage over in-hole acoustic logs
seismic energy goes through the near surface only once, and because they sample reflectivity and velocity at
therefore suffers less frequency loss than when both source and a bulk scale, similar to the surface seismic
receivers are on surface. Furthermore, VSPs can record 3 component methods. Hence their estimates of these
(3C) data which comprises P wave, Sh (horizontal) and Sv (vertical) parameters in new areas inspire confidence,
wavemodes. Because S waves travel at about 0.6 of the velocity of particularly in new feasibility studies.
a P wave, shorter wavelengths and therefore higher resolutions are However, the recently revised interest in
possible using S wave data. VSPs relates to more complex offset and
VSPs were first used in their one dimensional, zero offset, mode
walkaway implementations of the technique in
to correlate borehole geology with seismic stratigraphy within
2D and 3D (Figures 1: c and d). These have
seismic surveys. This showed that VSPs record ~30% higher
frequencies than surface seismic surveys. The program then applied
the potential to profile structure and
two dimensional VSPs to the problem of imaging geological stratigraphy laterally, with higher resolution at
structure close to a borehole. 2D VSPs confirmed that high higher frequencies than surface seismics. This
frequency results could be achieved, and that this enabled a more is mainly due to the fact that most of the
accurate image of the Merensky Reef than could be achieved by downhole receivers in a VSP are placed
surface seismic data. beneath the surface weathered zone, thereby
Finally, the program has achieved the successful recording and partially avoiding the attenuation of high
processing of 3D VSP data, with both compressional and shear frequencies within this zone.
wavemodes giving a high resolution structural result. Although P In addition to recording higher frequency
wave data gives ~30% frequency improvement over surface seismic
data, borehole seismic sensors provide a
surveys, S wave data can give as much as 80% improvement,
number of other advantages: borehole seismic
further improving the structural image achieved.
The high fixed costs of VSP surveys mean that they should
ideally be run in multi-client batches, which can reduce the per
borehole cost by up to 50%. Their application is anticipated to be on
high value capital infrastructure with a geological risk from faults
which are not imaged by surface seismics—generally in the sub-7 m
range.
Keywords * Charles Pretorius Geophysical Consultants CC.
VSP, 3D seismic, bushveld, platinum, resolution, imaging, shear † Anglo Technical.
waves, shaft sinking. ‡ Anglo Platinum.
© The Southern African Institute of Mining and
Metallurgy, 2011. SA ISSN 0038–223X/3.00 +
0.00. This paper was first presented at the,
SAIMM Conference, Platinum in Transition ‘Boom
or Bust’, 11–14 October 2010.
▲

The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 111 FEBRUARY 2011 117
Development of high resolution 3D vertical seismic profiles
Classification and nomenclature of seismic
wavemodes
Figure 2 illustrates the different seismic wavemodes which
will be discussed in this paper:
P waves (Figure 2 a) are compressional (pressure) waves
with particle movement in line with the direction of wave
propagation. S waves (Figure 2 b) are shear waves, with
particle movement perpendicular to the direction of wave
propagation. Vibration is in both the vertical (Sv) and
horizontal (Sh) directions.
The main reflected wave-modes are:
➤ PPup (Figure 2 a)—downgoing compressional P waves
are reflected as upgoing P waves.
➤ PSup (Figure 2 c)—downgoing P waves are mode-
converted to upgoing S waves at the reflector.
➤ PSSup (Figure 2 d)—downgoing P waves are mode-
Figure 1—Different VSP acquisition geometries converted to S waves at the base of the near-surface
weathered zone. The resultant downgoing S waves are
reflected as S waves at the reflector.
Surface seismics using conventional single-component
data typically achieve a much higher signal-to-noise ratio geophones can record only P wave or PSSP data (i.e. PSSup
than surface seismic data due to a combination of a quiet data converted back to P on return through the base of
borehole environment and strong sensor coupling to the weathering). Three-component geophones used for VSP
borehole wall. Surface geophones are generally poorly recording can record both P and S waves. S wave data can
coupled in weathered rock and exposed to cultural and have a shorter wavelength than P wave because S wave
environmental noise at the surface. Good sensor coupling in velocity is generally 0.6 that of P wave velocity and velocity =
the borehole enables acquisition of the full seismic wavetrain frequency* wavelength for both modes. The ratio of P to S
as well as three-component seismic data to be recorded with wave velocity can be used to calculate various geotechnical
high vector fidelity (Paulsson et al., 2004). This ultimately parameters such as shear modulus and Poisson’s ratio.
allows imaging of mode-converted waves and multiples.
Certain of these wave modes, notably the shear or S waves, Anglo Platinum’s 1D and 2D VSP campaigns
have the potential to further improve the imaging of
geological structure, providing higher resolution than surface The 2005 VSP programme
reflection seismics, as elaborated below.
During 2005 a VSP crew comprising CGG, Baker Atlas,
Vertical seismic profiles (VSPs) were extensively
Reeves Wireline Services and VSFusion staff was mobilized
employed in Anglo American’s Gold Division in the 1980s.
to South Africa to conduct a test programme, coinciding with
In 1981/2 the decision to mobilise the first Vibroseis crew to
the completion of one of Anglo Platinum’s surface 3D seismic
South Africa was still pending and it was understood that
surveys. Rig-source and Offset 2D VSPs were conducted in 10
this major commitment required a rigorous feasibility study
surface drillholes on several mining leases in the Bushveld
to manage technical risk. A year of VSP surveying was
Complex.
therefore undertaken to measure the in situ reflectivity of the
Witwatersrand Triad, basement and cover sequences before
approval was granted to mobilize the first surface Vibroseis
crew to South Africa. The resulting surface seismic
programme was one of the most sustainable geophysical
exploration campaigns in history, spanning 30 years and
spreading from gold to platinum clients in the hard-rock
environment. The solid foundation laid by the VSPs played a
significant role in the subsequent success of the surface
b)
seismic programmes. a)

The potential for improved structural resolution, partic-

ularly around new shaft sites, has been the main driver for
VSP research in Anglo American over the last five years.
Anglo Group companies, notably Anglo Platinum and Anglo
Technical, have conducted joint research with geophysical
service companies such as CGG Veritas (CGGV), VSFusion,
Baker Atlas, VSProwess, HiPoint and IFP to pursue these
mutual technical and business goals. This paper summarizes c) d)

the main findings and provides recommendations to

capitalize on some exciting opportunities in the future. Figure 2—Useful VSP wavemodes in the hard-rock environment
▲

118 FEBRUARY 2011 VOLUME 111 The Journal of The Southern African Institute of Mining and Metallurgy
Development of high resolution 3D vertical seismic profiles
J
As part of a shaft feasibility study, 10 Walkaway lines the receiver spacing was 6.25 m over the full borehole length. o
were recorded along surface seismic shot lines centred on a This spacing prevented spatial aliasing of shear wave data.
shaft geotechnical borehole. The main objective of the Source specifications were the same as for the Walkaway
u
Walkaway VSP programme was to determine whether VSPs surveys. r
could offer improved structural resolution over that achieved Figure 4 shows the correlation of synthetics, PPup VSP, n
in surface 3D seismic surveys, particularly within a 100 m to and surface seismics. Focusing on the UG2 event, the a
200 m radius around new shaft sites. A further objective was unfiltered corridor stack offered a promising 30% increase in
to determine whether VSPs would illuminate steeply dipping useable frequencies over the surface seismics. A low pass
l
structures, which are not well imaged on surface seismics. filter of 80 Hz applied to the corridor stack displays a good
The data acquisition, processing and interpretation of these match with the surface seismics on the right-hand panel.
data-sets are briefly discussed below, with a focus on Figure 5 shows a PP Walkaway line (red) overlain on a P
achievement of the above objectives. surface seismic crossline (blue). Referring to the UG2 event
An 8-level, Baker Atlas, MSR600, 3-component geophone the higher frequency VSP response is clear. The VSP also
a
array was assembled for the 2005 programme, capable of shows more reflectivity in the hangingwall. The net result p
deployment in holes down to 76 mm diameter. The was that the PPup VSPs demonstrated the stability of e
Walkaway surveys were a special case because they had to reflectors in the vicinity of the shaft hole, and further r
be zippered into a production 3D surface seismic survey and endorsed the structural suitability of this shaft site. The
incur minimal standing time. The production constraints radial subsurface coverage at a UG2 reef depth of 1 200 m
dictated that the receiver system could not be moved and was 350 m to 400 m. Unfortunately, the VSPs did not provide
would have to be anchored at an optimum level for all a superior image of steeply dipping structures, and this
Walkaway lines. Ray-trace modelling was conducted by requires further investigation.
VSFusion to determine the optimum geophone depth position
in the borehole for target illumination and was estimated to
be approximately two-thirds of the target depth (Figure 3).
Receiver spacing was set at 12.5 m, providing 87.5 m of
vertical geophone coverage in the drillhole. CGG’s Nomad 65
vibrators provided the surface source. Sweep length was 16
seconds and a Log 6 db, 30 Hz to 250 Hz sweep was
employed. Vibroseis source spacing was 25 m.
Following completion of the surface 3D seismic survey,
the vibrators returned to resurvey one of the Walkaway lines,
with a total of 48 geophone levels spaced at 6.25 m. This
data-set was later incrementally decimated to provide an
estimate of the minimum number of receiver positions for
effective imaging, which would be an important input
criterion for future equipment design and survey efficiency. Figure 3—Simultaneous surface seismic and walkaway VSP acquisition
In addition to the Walkaway surveys, 38 multi-offset at a prospective shaft site (TD Approx 1 200 m). Raytracing is shown
survey trials were conducted at the shaft site. For these trials along one Walkaway line (Charles, VSFusion, 2005)

Figure 4—Synthetics, PPup VSP’s & surface seismics at shaft test site
▲

The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 111 FEBRUARY 2011 119
Development of high resolution 3D vertical seismic profiles

Figure 5—PPup Walkaway line at shaft test site overlaid on surface seismic crossline (Mead, VSFusion, 2005)

Figure 6—VSP-CDP transform of reflected P-S (Mead, VSFusion, 2005)

The main difficulty encountered in processing was the A decimation study was undertaken along one of the
separation of PS and PSS mode-converted data. A detailed Walkaway lines. Although the Walkaway quality was best
discussion of the processing challenges is beyond the scope with 48 levels, 24 provided an acceptable result. Below 24
of this paper, but the problem can be summarized as follows: levels, data quality deteriorated significantly. This
Between the down going P wave and down-going S wave information was provided to Sercel, with a recommendation
arrivals, S wave reflections can be confidently separated. to manufacture a 24-level tool, if possible. Ultimately Sercel
However, below the down going S arrival the SS image can manufactured a 12-level SlimWave tool, as an interim
contain spurious PS reflections, falsely mapped to shallower measure to test the mineral market in the short-term. This
depths. Figure 6 shows reflected, offset PSup VSP data was used on the 3D VSP survey described below.
transformed to the common depth point (CDP) domain. Data
in the left ‘oval’ is contaminated with upgoing SS, while The ongoing role of 2D VSPs
further to the right the upgoing PSup separates nicely. What
is particularly interesting on this image is that the Merensky The 2D VSPs conducted during the 2005 programme, supple-
Reef has a better reflection response than on the PPup image. mented by 2D transects from the 3D VSP discussed below,
This was a recurring theme in the VSP R&D programme, have provided encouragement for a future role for 2D VSPs in
highlighting the need to extract value from these mode- the Bushveld Complex. In fact, the role for 2D VSPs could be
converted wavefields. significantly greater than for 3D VSPs, which appear to
▲

120 FEBRUARY 2011 VOLUME 111 The Journal of The Southern African Institute of Mining and Metallurgy
Development of high resolution 3D vertical seismic profiles
J
have a niche application around new shafts. The latter The arrival of mini-vibrators on the South African scene o
opportunities are usually presented only every five years has provided an opportunity to produce this stand-alone
case. These vibrators are small enough to be transported
u
or so.
2D VSPs certainly provide an improvement in structural within the region on standard container transport, rather r
resolution over surface 3D seismics and could therefore fulfil than using the expensive and travel-restricted abnormal load n
a high-resolution follow-up function to the former surface 3D systems which were required by the Nomad-65s. However, a
surveys in structurally complicated areas. This would be swop-out of the Nomads for mini-vibrators alone will not
required mainly when drill testing of the seismic structure improve the economics sufficiently to make a good business
l
plans continues ahead of mining. 2D Walkaway VSPs could case. Bulk work is required to reduce unit costs and this is
be particularly useful close to significant faults and may help best achieved by batching up multi-client work and
to quantify the widths of the fault damage zones. This is an conducting this seasonally, in bulk campaigns. Cooperation P
important parameter for mine planners and cannot be between mining companies on brownfield’s technical
accurately quantified on detailed infill 3D seismic surveys. It campaigns is normally achievable. The cost of a typical a
is possible to envisage a situation where deep drilling Walkaway VSP with a 700 m target depth under a multi- p
programmes could routinely include a provision for a few company batch model, with a batch size beween 5 and 10, e
VSPs, to be held in reserve and used in case structural varies between R300 000 and R350 000. Unfortunately, the
r
complexity is encountered or expected. This kind of advance cost of a stand alone single VSP is double this amount at
planning will help to make 2D Walkaway VSPs economically close to R700 000.
viable. The below discussion on survey economics shows
that the unit price per VSP is halved when moving from 3D VSPS
single campaigns to a group of five. Campaigns could
potentially be shared between several mining companies and Background
conducted in batch mode, thereby achieving significant
The success of the 2D VSPs in 2005 promoted the launch of a
mobilization synergies and shared cost savings.
3D VSP survey campaign for shaft risk management in 2006.
The case made for the Bushveld Complex above can be
Figure 7 illustrates the main objective of 3D VSPs in this role.
generalized to cover other commodities where seismics is
A tremendous amount of infrastructural planning takes place
currently paying a significant role. In addition, 2D VSPs
within a 100 m square area at the base of a new shaft
should continue to be important in the following niche roles:
system, and the best possible structural geological resolution
➤ Seismic feasibility studies, particularly in remote areas. is required to ensure the stability of the chosen site. It was
This may include insights on whether mode–converted hoped that a 3D VSP would deliver this.
energy has a role to play in surface seismics.
➤ As high-resolution transects linking deep drillholes to Combined surface 3D Patch and VSP survey at a new
establish stratigraphic or ore continuity. shaft site
➤ Imaging beneath difficult surficial conditions where
During 2008 a combined surface 3D seismic patch and 3D
surface seismics fails but subsurface techniques could
VSP survey was undertaken at a new Anglo Platinum shaft
succeed (for example over the Malmani Dolomites in
site where the UG2 occurs at a depth of 645 m. The VP
South Africa).
source ring radius was 1 000 m (approximately 1.5 x the
Illustrative survey costs depth to target). Based on prior modelling, a vibrator point
VSPs are generally quite costly because they require scarce, (VP) interval of 40 m was selected, giving a total of 1963 VPs
expensive equipment and highly specialized technical staff. per run. Receiver spacing for the surface 3D patch was 40 m
An added complication is the shortage of slimline downhole (Figure 8).
receiver systems. The main demand for these surveys is in
the oil industry, where much larger holes are drilled and
these can accommodate the more plentiful, larger-diameter
VSP receivers. The new generations of multi-level tools are
mostly being developed in larger diameter versions (i.e. HQ
equivalent and larger).
A further complication is the expense of the sources. For
example, the mobilization, running and maintenance costs
for a large Nomad-65 vibrator are the same whether the
machine is being employed on a substantial 3D seismic
survey or a small VSP programme. If charged at the normal
market rate, the mobilization overheads on their own
(normally running to several million rand) would have
stopped most of the VSP research programmes mentioned
above. Fortunately the seismic service companies concerned
were stakeholders in the R&D outcomes and appended the
VSP work to take place after 3D surface seismic surveys,
thereby reducing mobilization charges and other overheads.
This is not a sustainable business model and a stand-alone Figure 7—Detailed 3D VSP target area at the base of main and vent
VSP business case must be made. shafts
▲

The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 111 FEBRUARY 2011 121
Development of high resolution 3D vertical seismic profiles
➤ 50% drive level
➤ Specific designed ramp applied with 0 dB at 20–90 Hz,
6 dB at 150–170 Hz and 0 dB at 200–220 Hz
➤ 1 ms sample rate.
Figure 10 is a pictorial composite of the overall field
operation.

Images produced by VSProwess

A detailed discussion of data processing is beyond the scope
Figure 8—3D perspective of combined 3D surface seismic patch and
3D VSP (CGGV / VSFusion)
of the current report. VSProwess produced three main 3D
images: PPup, PSup and PSSup, as defined in the
introduction. All of these were considered to provide higher
resolution than the surface seismics. The comparisons
The main tool deployed in the VSP survey was Sercel’s
between synthetic seismograms and all VSP images were
new12-level SlimWave (Figure 10) with a 10 m inter-level
good. The PSup mode was probably the most coherent VSP
spacing. A Sercel downhole recorder synchronized with the
image and will be discussed in detail before considering the
main 408 system recorded the 36 channels of the SlimWave.
relative advantages of all three modes.
Figure 9 shows the modelled positions of the Slim-Wave
Figure 11 shows a PSup migrated VSP spliced into the
tool (marked in violet) corresponding to the three different
runs of the 3D acquisition, with all surface shotpoints being corresponding surface seismic in-line at the new shaft site.
repeated for each run. Note that there is a 1-level overlap The PSup data has been squeezed into equivalent PPup two-
between each run. The deepest level was positioned just way travel times before the merge. Note the improved
above the UG2 target at 635 m and the shallowest level at resolution of the Merensky Reef (MR) on the VSP compared
305 m. The additional blue runs in Figure 9 would be added to the surface seismics. Figure 12 shows a power spectral
before (deepest string) and after the 3D acquisition (2 comparison of the two data-sets. It indicates that the VSP
shallow strings), using a single offset vibrator point. This recovers significantly higher frequency data and offers about
gave extensive single offset VSP coverage over most of the 60% greater vertical resolution than the surface seismics.
borehole, for 2D analytical purposes and later data Figure 13 is a zoom into the composite data-set, showing
processing. auto picks on the UG1 horizon on a seismic interpretation
workstation, extending from the surface cube through the
Data acquisition VSP. A 7 m fault resolution should be comfortably achieved
Three runs of 1963 VPs were completed followed by the and image processing should offer even higher resolution.
offset 2D VSPs. The sweep selected had to be adequate for Imaging of fault damage zones should also be improved, due
both downhole measurements and high-resolution surface 3D to the closer proximity of the downhole receivers to fault
recording. After exhaustive trials the following parameters displacements on the target horizons. Fortunately, no
were selected (Gilot , CGG Veritas): significant faults appear close to the shaft site (centre of
➤ One vibrator delivering two standing sweeps per VP image). This upgrades the structural suitability of this site.
(two separate records) The signal to noise ratios deteriorate visibly on the extreme
➤ 16 seconds sweep length, tapers of 300–500 ms, 4 sec flanks of the VSP volume and autopicks in this zone should
record length not be trusted. Further editing of input records would
➤ Frequency range from 20 Hz to 220 Hz, with negative probably improve the coherency, and this will be the subject
time 20–30Hz of future research.

Figure 9—SlimWave tool positions in shaft hole for 3D runs (violet) and 2D supplements (blue) (Gilot, CGGV, 2008)
▲

122 FEBRUARY 2011 VOLUME 111 The Journal of The Southern African Institute of Mining and Metallurgy
Development of high resolution 3D vertical seismic profiles
J
o
u
r
n
a
l

P
a
p
e
r

Figure 10—a) CGGV Seismic recording truck (foreground), Weatherford borehole geophysical logging truck (middle) and crane for VSP tool deployment
(background); b) CGGV Nomad 65 Vibrator; c) 6 levels of CGGV/Sercel SlimWave tool ready for deployment; d) sidewall calliper brace on SlimWave element
in open position; e) Engineer inside CGGV seismic recording truck; f) Survey simulation and design using 3D raytracing (VSFusion)

Figure 11—Migrated PSup VSP section spliced into surface 3D seismic line through new shaft site (Humphries, VSProwess, 2010)
▲

The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 111 FEBRUARY 2011 123
Development of high resolution 3D vertical seismic profiles

Figure 12—Waveforms (top) and power spectra (bottom) of the PSup VSP (red) and surface seismic data (blue) shown in Figure 11 (Humphries, VSProwess,
2010)

Figure 13—Top-UG1 picks in section through surface seismic patch and PSup VSP

Figure 14 shows a composite of merged PPup, PSup and a typical VSP plus 3D surface seismic patch, imaging a
PSSup data with surface seismic backdrops, indicating the primary target between 700 m and 1 000 m, would cost about
incremental increases in structural resolution moving from P R10 million to acquire, process, and interpret in 2010.
wave data to the two mode-converted shear wave images.
Clearly, the first prize in the structural resolution stakes Conclusions
should be the PSSup data-set. Unfortunately, the signal to During 2009/2010 a number of breakthroughs were achieved
noise ratio for this mode was significantly lower than PSup , in the reprocessing of 3D VSP data. Most of these were
the deterioration on the image flanks was greater and the achieved by Mary Humphries of VSProwess. Her results
map image showed more gaps. Improvement in acquisition complement the 2D VSP achievements of the CGG/Baker
and processing of PSSup data will be the subject of future Hughes/VSFusion consortium in 2005 and both have
R&D. provided a basis for drafting good practice guidelines for the
future. P wave and mode-converted S wave VSP images
Illustrative survey costs returned higher structural resolution than surface seismics in
Unlike the 2D VSP case, it is difficult to arrive at a generic localized apertures around the drillholes. Future surface
cost for a 3D VSP survey. As a rough estimate, it is likely that seismic programmes should take this into account and design
▲

124 FEBRUARY 2011 VOLUME 111 The Journal of The Southern African Institute of Mining and Metallurgy
Development of high resolution 3D vertical seismic profiles
J
o
u
r
n
a
l

P
a
p
e
r

Figure 14—Structural resolution improvements over surface 3D seismics offered by PPup, PSup and PSSup VSPs (Humphries, VSProwess, 2010)

for S as well as P wave recording. Significant progress was seismic patch and VSP would be about R10 million, and this
also made with innovative processing approaches by HiPoint approach is recommended only at new shaft sites.
Reservoir Imaging (Fuller et al., 2008) and IFP. These may
be discussed at some stage in the future. Acknowledgements
The authors are grateful to Anglo Platinum and Anglo
Recommendations
American Plc for their diligent sponsorship of five years of
2D Walkaway VSPs should fulfil an important role in high VSP research and development. Particular thanks are due to
resolution follow-up of surface seismic surveys, particularly the following colleagues: Kazek Trofimczyk, Gordon
when structurally complex areas are drill-tested ahead of Chunnett, Ron Hieber, Fatheela Kaldine, Andy Rompel and
mining. They may be particularly useful for quantifying the Agnes Jikelo. We also acknowledge the excellent research
potential width of fault damage zones. In such complex and project work undertaken by consultants in CGG Veritas,
areas, provision should be made in drilling budgets for VSProwess, HiPoint Reservoir Imaging, IFP, Baker Hughes
Walkaway VSP surveys to be conducted in strategically and VSFusion. Particular thanks in this group are due to
important drillholes. 2D VSPs are also recommended in Michel Denis, Eric Gilot, Renee Daures, Frederic Moinet, and
seismic feasibility studies, particularly in remote areas. 2D Frederic Naud of CGG Veritas for their championing and
Walkaway VSPs are also potentially useful for imaging support throughout the R&D programmes and to Mary
beneath difficult surficial conditions, where surface seismics Humphries of VSProwess for her breakthroughs on the 3 D
fails, but subsurface techniques could succeed. 3D VSPs are VSP processing. All of the reports produced are internal to
most appropriate for risk management at new shaft sites, Anglo American, but acknowledgements of consultants’
where they should be conducted with a coincident, high contributions are included wherever possible on figures in
resolution, surface 3D seismic patch. An important the text.
requirement for all VSP surveys is that the drillhole diameter
must be at least 76 mm. References
In order to reduce costs, it is recommended that 2D VSPs
FULLER, B., STERLING, M., and VAN DOK, R. Time domain 2D VSP and 3D VSP
should be conducted in multi-client, collaborative batch processing. Proceedings of the Society of Exploration Geophysicists (SEG)
campaigns. An optimal 2D VSP batch size is about five Conference, 2008.
surveys, which would halve the unit cost of a typical
PAULSSON, B., KARRENBACH, M., MILLIGAN, P., GOERTZ, A., and HARDIN, A. High
Bushveld Walkaway VSP from about R700 000 to approxi- resolution 3D seismic imaging using 3C data from large downhole seismic
mately R350 000. The typical cost of a combined 3D surface arrays, First Break, vol. 23, 2004. ◆
▲

The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 111 FEBRUARY 2011 125