QUEENSLAND ROADS Edition No 4 September 2007

39
THE USE OF SLOPE STABILITY RADAR IN MONITORING SLOPES AND
MANAGING SLOPE INSTABILITY HAZARDS

Dr Neal Harries
1
BSc (Geol), PhD (Rock mechanics)
Principal Geotechnical Engineer
Mr Michael Holmstrom
1
BEng (Elec), MSc (Elec)
Senior Systems Engineer
Abstract
Assessing and managing instability hazards is an
essential activity when working with both unstable
natural slopes and engineered slopes. The slope
stability radar (SSR) is a state-of-the art tool to
manage risks related to slope instability. The SSR is
an all-weather system that remotely scans slopes to
continuously measure surface movement with a sub-
millimetre precision. The combination of near real-time
measurement, sub-millimetre precision and broad area
coverage to quickly identify the size of developing
failures provides ideal parameters for the management
of slope instability hazards. It allows users to detect
and alert personnel of rock wall movements that may
result in instability. The success of using radar in
monitoring slopes is veried by the rapid adoption of
this Queensland invented and produced technology
throughout the world.
Introduction
The instability of engineered and natural slopes is often
of high importance, due to potential for loss of life
and property. The use of SSR in applications such as
in open-cut mines allows greater productivity without
compromising safety. The SSR was developed to aid in
the management of risks associated with unstable slopes.
This paper describes the SSR system and gives an
example of rock slide monitoring on a major highway.
Rock slope monitoring
Rock slope monitoring is required to identify precursor
phenomena that occur prior to a slope failure. A number
of signs may occur prior to collapse, including dilation
of cracks on the wall, new fracturing seen on the face,
audible noise (cracking and popping), dislocation
(shearing) along fractures, and increased dust or rilling
of loose material from the rock face. However, not all
signs will be observed prior to collapse and it is difcult
to predict the progression of failure towards collapse
by reviewing such qualitative phenomena. This usually
results in overcautious decisions that impact on road
access or productivity, while the reverse places lives at
higher risk.
1 Current employees of GroundProbe, South Brisbane, Australia. www.groundprobe.com
QUEENSLAND ROADS Edition No 4 September 2007
40
A more reliable indicator of instability involves the
quantitative measurement of outward movement and
acceleration of material as an instability occurs for an
extended period prior to rock slope collapse (1). The
standard methods for dealing with slope instability (2)
are:
development of a monitoring system
adopting acceptable slope deformation criteria
warning systems
design of stabilisation works
risk mitigation strategies.
Qualied personnel can interpret the pattern and
history of movement to improve prediction of the
failure processes, and to advise appropriate and timely
stabilisation or safety management actions.
Figure 1. Interferometric process
Radar Monitoring Process
The SSR concept is based on the considerable success
achieved by differential interferometry obtained with
synthetic aperture radar (SAR) which can measure small
movements of land masses from satellites (for example:
along-fault slippage associated with earthquakes, ground
subsidence associated with oil extraction etc). The slope
stability radar utilises the differential interferometry
processing technique on a stationary platform positioned
50m to 1700m back from the foot of the wall. The
system scans a region of the wall in a grid pattern and
compares the phase measurement of each footprint (pixel)
with the rst scan to determine the nature of movement
on the rock face (Figure 1), and hence the stability of the
slope. The SSR system has a pixel size of 1m x 1m at a
range of 50 metres and of 15m x 15m at 850 metres. The
SSR-X system (High Denition Slope Stability Radar)
has an improved spatial resolution by a factor of four
over the SSR system. At 50 metres, the pixel size of the
SSR-X system is 0.5m x 0.5m, and at 850 metres is 7.5m
x 7.5m.
QUEENSLAND ROADS Edition No 4 September 2007
41
The system offers sub-millimetre precision of wall
movements without being adversely affected by rain, fog,
dust, smoke and haze, although reduced precision occurs
in pixels where there is low phase correlation between
scans. Low phase correlation and loss of precision is
caused by vegetation growing on the scanned slope.
Slope Stability Radar (SSR) Equipment
The slope stability radar system is housed on a
self-contained trailer that can be easily and quickly
moved around the site (Figure 2). The radar dish,
electronics and computer module are located on a tripod
which detaches from the trailer during deployment so that
vibrations from the trailer and generator do not degrade
reliability and performance. The two-axis mount will
allow the radar dish to scan through 320O in azimuth,
and 60O vertically from the horizontal. The area
to be scanned is set manually using a digital camera
image. The display and interface module contains a
keyboard touchpad and display which are mounted in a
weatherproof enclosure (Figure 3). Power is provided by
a remote area power supply which consists of a battery
pack that is automatically charged during a 3.5 hour
period each day by a diesel generator.
The system provides immediate monitoring of slope
movement without calibration and prior history. Scan
times are typically every 1 to 10 minutes, depending on
the size of the selected scanned area. Data can be viewed
directly on the SSR unit or uploaded to a remote location
via a dedicated radio link. Custom software (SSRViewer)
enables the user to set movement thresholds to warn
of unstable conditions. Data from the SSR is usually
presented in two formats. Firstly, a colour rainbow plot
of the slope representing total movement quickly enables
the user to determine the extent of the failure and the area
where the greatest movement is occurring (Figure 4).
Secondly, time vs displacement graphs can be selected at
any location to evaluate displacement rates (Figure 5).
Figure 2. GroundProbes Slope Stability Radar Unit
Figure 3. SSR functional components
QUEENSLAND ROADS Edition No 4 September 2007
42
Figure 5. Typical gure plots
Figure 4. SSRViewer visual and deformation image
QUEENSLAND ROADS Edition No 4 September 2007
43
Figure 7. Rock slide over main access road to
Yosemite National Park
Figure 6. Typical alarm response (5)
Figure 8. Deformation data and movement gures
QUEENSLAND ROADS Edition No 4 September 2007
44
decelerating, or completely at rest. GroundProbes
SSR was rst deployed at Yosemite in J une 2006 (3).
The system was set up adjacent to the rock slide on the
opposite side of the Merced River. Within an hour
of deployment the movement of the land slide was
characterised, with faster and slower moving parts of
the slide easily identied by the systems real-time
capabilities and sub-millimetre precision (Figure 8).
The SSR scan gave the California Department of
Transportation great condence as it dened the stability
of the area. The California Department of Transportation
continually monitored the highway slide area for over 12
months so that they could be informed of any movements
occurring in the affected area and respond accordingly.
At the same time work proceeded to establish an
alternative access route into the park.
Conclusions
The SSR is the state-of-the-art development for
monitoring slope movement in rock slopes, structures
and non-vegetated slopes. It offers the unprecedented
advantage of sub-millimetre precision and broad area
coverage of wall movements through rain, dust and
smoke. The real-time display of the movement of rock
faces has allowed continuous management of the risk
related to slope instability at a large number of mining
operations and at several infrastructure projects. The
SSR technology has enabled a radical change in the
management of risks in rock slopes, which has resulted in
a rapid adoption of the technology throughout the world.
References
1. Hoek E, Bray J W. Rock Slope Engineering, The
Institute of Mining and Metallurgy. 1981
2. Glastonbury J , Fell R. Report on the analysis of the
deformation behaviour of excavated rock slopes.
Uniciv report number R-403, The University of
New South Wales, Australia. 2002.
3. GroundProbe. Safety Monitoring of rock slide at
Yosemite National Park. 2006.
A critical component of the SSR software is the alarms.
These provide warning for work crews of impending
instability so that appropriate actions can be taken.
Alarms can be based on amount of deformation on a
slope or the rate of deformation. Numerous warning
alarm options are available such as:
SMS text messages sent to engineering staff
mobiles
e-mail to appropriate staff
software alarms on remote computers
alarms activating audible sirens and ashing lights,
and
alarms activating trafc lights to stop trafc
moving into an area of land-sliding instability.
A typical system is to use a remote computer system
(e.g. trafc control ofce) which can raise an alarm in
real time. Alarms will have detailed instructions for each
alarm response type (Figure 6).
In April 2006, a major rock slide buried the main highway
access into Californias world famous Yosemite National
Park (Figure 7). The slide blocked Californias State
Highway 140, the main all-weather route to the park.
The local community of Mariposa and neighbouring
communities were concerned about the consequences of
future land slides. These concerns included public safety,
signicant nancial losses to businesses due to limited
park entry, trafc congestion on alternative park access
routes as well as fears of destruction of property and
camping grounds.
What initially appeared to be a problem of a few rocks
rolling onto the highway from the adjacent slope quickly
turned into a complex rock slide with a volume of
approximately 1.5 to 2.0 million cubic metres. With
signicant media coverage and public involvement, this
event required the effort of a number of state agencies
and professionals to address the growing public concerns.
On 12 J une 2006, the U.S. Geological Survey visited
the area and concluded it would be difcult to dene
whether the slide mass was slowly creeping, accelerating,