Emerging technologies and the future of geotechnical instrumentation

J.M. Girard and E.L. McHugh

NIOSH Spokane Research Laboratory, Spokane, Washington, USA

ABSTRACT: This paper discusses the results of two recent case studies in which high-tech prototype
instruments were used. The first case study describes the results of a ground-based hyperspectral imaging
tool used to map the geology of an open-pit mine highwall. The second case study outlines the use of
interferometric synthetic aperture radar and how this type of radar can be used in a variety of geotechnical
situations. These case studies are part of an on-going research project at the National Institute for
Occupational Safety and Health’s (NIOSH) Spokane Research Laboratory. The project objective is to
implement engineering controls and design methods in order to reduce the number of injuries and fatalities
associated with slope failures at mining operations. Project personnel are hopeful that these new
developments in technology will lead to better geotechnical monitoring and design in slope stability and other
important areas of rock mechanics.

1 INTRODUCTION 2 HYPERSPECTRAL IMAGING

The National Institute for Occupational Safety and A contributing factor in many slope failures is the
Health (NIOSH), Spokane Research Laboratory is presence of mechanically incompetent, hydro-
involved in research to improve methods for thermally-altered rock (Watters & Delahaut,
detecting conditions in open pit mines that could 1995). Major structures are generally well
lead to catastrophic slope failures. Since 1995, 33 mapped, but weak rock units may be much more
miners have died in slope failure accidents at U.S. difficult to identify and the degree of clay
mines. Better methods for monitoring and design alteration is oftentimes difficult to determine by
are needed to ensure the safety of mine workers. visual examination alone. Geologic maps can
This paper highlights two emerging technologies vary greatly in quality and detail due to the
that could potentially be adapted to improve mine subjectivity of various geologists and the extreme
safety. Applications to rock mechanics issues complexity of many deposits. In addition, there
related to surface and underground mining, are financial and practical limits to the number of
petroleum, geological, civil engineering and samples that can be taken for geochemical or
geohazards monitoring are also included. engineering analyses. Inevitably, a large
percentage of data shown on geologic maps is
subject to scrutiny. High turnover rates of
engineers and geologists at many companies,
financial and time limitations, and unfamiliarity
with complex formations can contribute to crucial
errors or omissions on geologic maps.

2.1 Prototype Spectral Imager

NIOSH recently completed field tests of a

prototype imaging spectrometer that was built by
Carnegie Mellon Research Institute (CMRI). The
purpose of the tests was to evaluate a new method
for objectively mapping the geology of an open-pit
mine from a distance.
Every material on the earth’s surface reflects
light in a characteristic pattern. The manner in
which light of different wavelengths is reflected or
absorbed by each material is known as its
reflectance spectrum. By filtering light of specific
wavelengths, images can be created that
differentiate specific materials. Figure 1 illustrates
examples of the spectral signatures of Neodymium
Oxide and Kaolinite (Clark et al., 1998). A
spectrometer is a device that collects these
diagnostic spectral absorption features;
identification of the minerals is then achieved by
comparing the data to spectral libraries.
Field portable spectrometers are commercially
available that can analyze light reflected from a
single point, but the CMRI instrument is
distinctive because it collects spectral reflectance Figure 1. Examples of reflectance spectrum for
data in a two-dimensional array. The instrument Neodymium Oxide (top) and Kaolinite (bottom)
operates in the visible and near infrared ranges and (adapted from Clark et al., 1993)
employs an acousto-optical tunable filter to control
wavelength, a phase retarder to measure
polarization signatures, and a digital camera linked
to a laptop computer to capture data (Figure 2).
The instrument is very robust since it has no
moving parts, and is also field portable (Denes et
al., 1998; Gupta et al., 1999; Sabine et al., 1999;
McHugh et al., 2000).

2.2 Field Site Selection

Since the current configuration of CMRI’s

instrument is limited to the visible and near ~10 cm
infrared ranges (0.45 – 1.0 µm), mineralization
with diagnostic features within this particular
Figure 2. Spectro-polarimetric imager
spectral range was required for the field tests. For developed by Carnegie Mellon Research
Institute.
this reason, the Mountain Pass Mine, which is ENVI software Spectral profiles were compared
approximately 60 miles southwest of Las Vegas, to data in the U.S. Geological Survey’s Digital
Nevada, was selected as the test site for the CMRI Spectral Library (Clark et al. 1993) to correlate the
prototype instrument. The Mountain Pass Mine images with specific minerals. Detailed
geology is comprised of rare-earth elements of the information regarding the spectral processing can
lanthanide series (cerium, lanthanum, neodymium, be found in McHugh, et al., 2000. The results of
europium, and others), which have at least eight the image analyses are illustrated in Figure 4. The
distinctive absorption features between 0.45 – 1.0 white areas correspond to bastnaesite ore and the
µm. Spectral data for the deposit were published darker areas are waste rock or unclassified spectra.
by Rowan & others (1986; 1996) and Kingston
(1993).

2.3 Field Tests

Twelve multispectral images -- including pit

highwalls (Figure 3), outcrops, drill core from the
ore zone, and hand samples -- were collected at
the site. Supplemental images of hand specimens
under artificial light sources were collected at
CMRI to aid in calibration and evaluation of the
instrument. Thin sections were made from both
the ore and the host rock to assist with
interpretation of the images.
Figure 4. Results from the CMRI spectro-
polarimetric imager. Light areas correspond to ore;
darker areas to waste or unclassified spectra.

The results from this field study successfully

illustrate the capability of a field-portable imaging
spectrometer to discriminate minerals within
images of highwalls, outcrops, rock samples, and
drill core. However, to effectively apply the
AOTF-based technology to the characterization of
alteration in rock masses caused by clays and
other minerals, the instrument would need to be
redesigned to extend into the shortwave and mid-
Figure 3. West highwall of the Mountain infrared region of the spectrum. At this time,
Pass Main Pit. Study area is outlined. funding for this advancement in technology is not
available at NIOSH or at CMRI.
There are many advantages to using spectral
2.4 Analyses and Results analyses for geologic mapping. First, spectral
identification of minerals would remove the
After the spectro-polarimetric images of the area human error and subjectivity of trying to visually
were collected, substantial processing, filtering, determine the degree of alteration in a rock mass.
and computer analyses were performed using Secondly, workers would be able to map mine
highwalls, or other inaccessible or precarious INSAR’s have many advantages over current
areas, from a safe distance. Finally, spectral types of monitoring systems. Able to work in
analyses could be used to define faults, shear nearly all weather, an INSAR can acquire imagery
zones, and fracture systems in rock masses. through fog, mist, rain, haze, or cloud cover, and
can operate day or night. Also, an INSAR can
sample large areas for ground displacement, which
3 SYNTHETIC APERTURE RADAR gives them a tremendous advantage over survey
networks, extensometers, and other instruments
3.1 Overview of Technology which sample movement on a discrete set of
points. Recent developments in instruments such
The second type of remote sensing technology that as prismless laser range finders partially address
is being studied by NIOSH researchers is synthetic the problem of under sampling large areas for
aperture radar (SAR). SAR is a type of ground- movement. However, the range and accuracy of
mapping radar that was originally designed to be these units can vary greatly depending on the
used from aircraft and satellites. SAR can be used reflectivity of the rock, the angle of the rock face,
to generate high quality digital elevation maps and weather, and other factors.
to detect disturbances of the earth’s surface. A
variation of SAR – Interferometric Synthetic 3.2 Field Tests
Aperture Radar (also known as IFSAR or INSAR)
– uses differences in successive SAR images to The Microwave Earth Remote Sensing Laboratory
generate maps of displacements. One of the first at Brigham Young University (BYU) has recently
applications that tracked ground motion without designed and built a small synthetic aperture radar
the aid of ground control points used Earth system capable of operating from light aircraft or
Resource Satellite (ERS-1) INSAR data to
estimate ice-stream velocity in Anarctica
(Goldstein et al., 1993). Although radar
technology has been in widespread use since the
1940’s, interferometric techniques have mostly
been developed during the past decade. INSAR
has also been successfully applied to produce the
first displacement maps of ground movement
caused by natural hazards such as earthquakes
(Massonnet et al., 1993 & 1994; Zebker et al.,
1994), volcanic activity (Massonnet, 1995), and
landslides (Fruneau & Achache, 1996). This
technique can also be used to monitor
displacement of unstable slopes or landslides
(Reeves et al., 1997; Sabine et al., 1999), and
subsidence caused by the extraction of ground
water, oil & gas, or minerals (Fielding, et al.,
1998; Carnec, 1996; Stow 1996; Dixon, 1994).
These pioneering studies have generated
Figure 5. Photograph of the prototype
enormous interest in the Earth science community
interferometric radar system. The transmit
because they point to an entirely new way to study antenna is seen in the center and the receive
the surface of the Earth. antennas are on the top-right and bottom-
right.
from stationary ground-based locations highwalls at mines. Research and development
(Thompson, 1998.) The prototype system, as seen needs to be completed and the prototype system is
in Figure 5, consists of a transmitter, two currently undergoing engineering tests. As soon
receivers, three antennas, and an embedded as BYU has completed modifications to the
computer. instrument, NIOSH is planning to perform
The first field tests related to geohazard ground-based field tests of an open-pit highwall at
monitoring using this instrument were performed an active mine site. An established survey
by BYU for the Canadian National Railways network of prisms and GPS data will be used to
(CN). The railways were interested in finding for calibrate the radar results.
a method to accurately detect rockfalls and
washouts on railroad tracks before trains
approached those dangerous areas. CN reports
that approximately 20% of all their train 4 CONCLUSIONS
derailments are caused by slope failures -- either
rockslides or washouts (Figure 6. Arnold & Remote sensing, at the present time, cannot
Clegg, 1999). Corrective action requires a lot of replace conventional geotechnical methods of
time and can expose repair personnel to the risk of investigation. However, emerging technologies
injury from additional falling debris. CN needed such as hyperspectral imaging and interferometric
a system that could detect problems along the synthetic aperture radar should not be ignored.
railroad track without having to send out The satellite versions of this technology are
personnel. already widely used to monitor ground
In the field tests for CN, the BYU subsidence, landslides, volcanoes, and active
interferometric radar system generated a terrain faults. As computer technology continues to
topography map accurate to within a few advance and processing speeds increase,
centimeters. Successive radar images were integration of remote sensing tools into mining,
compared to watch for sudden changes in the petroleum, and civil engineering applications will
topography caused by rockfalls or washouts along also increase.
the track. The potential benefits of using multiple remote
sensing methods to better understand slope
failures is merely one aspect of the current
research to reduce the number of deaths caused by
ground instability at surface mines. The NIOSH
Slope Stability Hazard Recognition Team is
pursuing many other options to minimize the
dangers associated with surface ground control
problems. Results of this research will hopefully
benefit others involved in the study of landslides,
rock falls, avalanches, volcanic activity, and other
geohazards.

Figure 6. Rail track demolished by a slope

failure.

The initial results from the CN railways tests

were positive, and BYU researchers are confident
that their system can be adapted to monitor
REFERENCES

Arnold, D. & E. Clegg (1999). Natural Hazard Goldstein, R.M., H. Engelhardt, B. Kamb, and
Detection along the Right-of-Way Using R.M. Frolich (1993). Satellite Interferometry for
Interferometric Radar. Proceedings of the Monitoring Ice Sheet Motion: Application to an
American Railway Engineering and Maintenance Antarctic Ice Stream, Science, 262, pp. 1525-
of Way Associations Annual Meeting. Baltimore,
1530.
MD.

Carnec, C., C. King, D. Massonnet, & T. Rabaute Gupta, N., R. Dahmani, M. Gottlieb, L. Denes, B.
(1996): SAR Interferometry for monitoring land Kaminsky, & P. Metes (1999). Hyper-spectral
subsidence: application to areas of underground Imaging Using Acousto-optical Tunable Filters.
earth resources mapping. In FRINGE 96: Proceedings of SPIE, v. 3718, 10 pp.
Proceedings ESA Workshop on Applications of
ERS SAR Interferometry. Zurich, Switzerland. Kingston, M.J. (1993). Evaluation of AVIRIS
http://www.geo.unizh.ch/rsl/fringe96/papers/ Data for Mineral Mapping at Mountain Pass, the
East Mohave Desert, California. In Proceedings
Clark, R.N., G.A. Swayze, A.J. Gallagher, T.V.V. of the ninth Thematic Conference on Geologic
King, and W.M. Calvin (1993). The U.S. Remote Sensing, Pasadena, CA. pp. 181-182.
Geological Survey, Digital Spectral Library:
Version 1: 0.2 to 3.0 microns. U.S. Geological Massonnet, D., M. Rossi, C. Carmona, F.
Survey Open File Report 93-592, 1340 pp. Adragna, G. Peltzer, K. Feigl, and T. Rabaute
http://speclab.cr.usgs.gov (1993). The Displacement Field of the Landers
Earthquake Mapped by Radar Interferometry.
Denes, L.J., M. Gottlieb, & B. Kaminsky (1998). Nature, 364, pp. 138-142.
Acousto-Optic Tunable Filters in Imaging
Applications. Optical Engineering, 37: 1262- Massonnet, D., K. Feigl, M. Rossi, and F. Adragna
1267. (1994). Radar Interferometric Mapping of
Deformation in the Year after the Landers
Dixon, T., editor (1994) SAR Interferometry and Earthquake. Nature, 369, pp. 227-230.
Surface Change Detection: Report of a Workshop.
Boulder, Colorado. Massonnet, D., P. Briole, and A. Arnaud (1995).
Deflation of Mount Etna Monitored by
Fielding, E.J., R.G. Blom, and R.M. Goldstein Spaceborne Radar Interferometry. Nature, 375,
(1998). Rapid Subsidence over Oil Fields pp. 567-570.
Measured by SAR Interferometry. Geophysical
Research Letters, vol. 25, no. 17, pp. 3215-3218. McHugh, E.L., J.M. Girard, L.J. Denes, P. Metes,
& C. Sabine (2000). Current Research on Slope
Fruneau, B. and J. Achache (1996). Satellite Movement in Mines: Use of Hyperspectral
Monitoring of Landslides Using SAR Imagery. Proceedings of the 14th International
Interferometry. News Journal, International Conference on Applied Geologic Remote Sensing.
Society for Rock Mechanics, vol. 3, no. 3. Las Vegas, NV. pp. 584-591.
Reeves, B., D. Noon, G. Stickley, and D. Watters, R.J., & W.D. Delahaut (1995). Effect of
Longstaff (1997). Monitoring Rock Slope Argillic Alteration on Rock Mass Stability. In
Deformation by Radar Interferometry. In Haneberg, W.C., & Anderson, S.A., editors, Clay
Proceedings of the Workshop on Applications of and Shale Slope Instability: Boulder Colorado,
Radio Science WARS'97, A. Kulessa (ed.) Geological Society of America Reviews in
Australian Academy of Science. Engineering Geology, volume X.

Rowan, L.C., M.J. Kingston, & J.K. Crowley Zebker, H., P. Rosen, R. Goldstein, A. Gabriel,
(1986). Spectral Reflectance of Carbonatites and and C. Werner (1994). On the Derivation of
Related Alkalic Igneous Rocks: Selected Samples Coseismic Displacement Fields using Differential
from Four North American Localities. Economic Radar Interferometry: The Landers Earthquake.
Geology, v. 81, pp. 857-871. J. Geophys. Res., 99, pp.19617-19634.

Rowan, L.C., R.N. Clark, & R.O. Green (1996).

Mapping Minerals in the Mountain Pass,
California Area using the Airborne
Visible/Infrared Imaging Spectrometer (AVIRIS).
In Proceedings of the Eleventh Thematic
Conference and Workshops on Applied Geologic
Remote Sensing. Las Vegas, NV. pp.175-176.

Sabine, C., L.J. Denes, M. Gottlieb, B. Kaminsky,

P. Metes, R.T. Mayerle, & J.M. Girard (1999). A
Portable Spectro-Polarimetric Imager: Potential
Mine Safety and Geologic Applications. In
Proceedings of the 13th International Conference
on Applied Geologic Remote Sensing. Vancouver,
Canada, vol. I: 190-194.

Stow, R. (1997) Application of SAR

Interferometry to the Imaging and Measurement of
Neotectonic Movement Applied to Mining and
other Subsidence/Downwarp Modeling. FRINGE
96: Proceedings ESA Workshop on Applications
of ERS SAR Interferometry. Zurich, Switzerland.
http://www.geo.unizh.ch/rsl/fringe96/papers/

Thompson, D., D. Arnold, D. Long, G.F. Miner,

M.A. Jensen, T.W. Karlinsey, A.E. Robertson, J.S.
Bates (1998). YINSAR: A Compact, Low-Cost
Interferometric Synthetic Aperture Radar. In
Proceedings of the 1998 International Geoscience
and Remote Sensing Symposium, Seattle, WA. pp.
1920-1922.