GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING

John Simmons
Sherwood Geotechnical and Research Services, Peregian Beach, Queensland

ABSTRACT
Geotechnical considerations for support of coal mining are described in this review, which is an updated version of an
earlier review published in 1995. A thorough working knowledge of the geological environment and of modern coal
mining operations is required to provide specialist geotechnical advice. The mining industry operates within a range of
constraints and drivers that are quite different in some respects to those encountered in the wider geotechnical
community that supports civil projects. Fundamental geotechnical requirements remain the same: a sound working
knowledge of applicable geotechnical parameters and groundwater conditions and reliable analytical tools.
Opportunities for data gathering are limited and much reliance is placed on experience, judgement and consideration of
mine slope forming processes and the operating requirements of equipment. Principles for stability assessment of both
rock and spoil slopes are outlined. The implications of modern risk management procedures are discussed along with
future developments that are anticipated.

1 INTRODUCTION
Large-scale open pit coal mining has contributed significantly to the Queensland economy for over forty years. Over
that period market conditions have been cyclic and there have been episodes of major, capital-intensive expansion.
Geotechnical support has been required for mine infrastructure development and also for operational management of
slope stability hazards, with the industry having sponsored research and development projects in response to needs that
have arisen through operational experience.
Following several years of application of major research and development activities, Simmons (1995) outlined
considerations for slope stability in open pit coal mining as understood and practised at that time. Many of those
understandings and practices remain valid, but in certain respects there have been technological changes and also
cultural changes in the terminology and practices of risk management. The following is an updated version of the 1995
paper, written primarily to recognise these changes, bearing in mind that generational changes also mean that the
industry is rapidly losing the continuity of geotechnical engineering experience that has been a key feature of
operational support.

2 BACKGROUND
To participate effectively in the coal mining industry the geotechnical specialist must have a good working knowledge
of coal geology and mining operations and be familiar with site conditions. This may not easily be achieved because
continuous specialist input is not normally required. Geotechnical specialists are most likely to be involved in
feasibility projects, slope profile design, assessment of risks and management of risk controls particularly when slopes
fail.
This review covers issues which the geotechnical specialist must be prepared to encounter. It is written from a
background of practices in large-scale predominantly strip mining in the Bowen Basin of Central Queensland, the
Hunter Coalfield of New South Wales, and applies equally to open cut coal mining in Kalimantan Timur, Indonesia.
No geotechnical work can be undertaken on the scale of an open pit coal mine without an appreciation of the geological
environment. The geotechnical perspective is necessarily different from that of a coal quality or exploration geologist,
since what is sought is the response of materials to the processes of slope formation. The coal is usually of marginal
interest compared to the overburden, which is almost the opposite perspective from that of the geologist or miner.
The processes of slope formation must be well understood since the operating characteristics and limitations of
machinery have to be appreciated. Mining operations are tightly scheduled to maximise productivity. Slope stability
issues directly affect productivity and must be approached accordingly. Obviously, there must be an appreciation of the
geotechnical properties of the materials, but this is by no means as simple as it seems owing to the nature of the
materials and the conditions under which these must be dealt with.
Slope stability assessments must be undertaken for intact and broken rock masses as well as dumped spoil. While the
mechanics of analysis may be relatively straightforward, there are many special issues to be recognised and addressed
particularly regarding the roles of water and time.

Australian Geomechanics Vol 48 No 1 March 2013 85

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

3 GEOLOGICAL ENVIRONMENT
A specialist geological literature on coal basin geology is available and serves to provide an understanding of where to
find coal, what sort of coal to expect and (through the processes of basin development and subsequent history) what
stands between the miner and the coal. Deposition processes, bedding and structural features can be utilised by
geotechnical specialists to help understand what sort of materials to expect in slopes, what are the defects which control
stability of these materials, and what shear strength behaviour should be expected. Ward et al. (1995) provides review
material for all significant Australian coal basins. For the Bowen Basin specifically, Henderson and Stephenson (1980),
and the Bowen Basin Geologists Group (1990, 1995, 2000, 2005, and 2010) provide a body of reference information
from which geotechnical contexts can be derived.

3.1 DEPOSITION AND BEDDING

Each sedimentary basin has specific features which determine the nature and distribution of coal measures materials.
Exploration geologists establish models of sedimentation processes and cycles, and rock units are described
accordingly. Likely overburden and spoil material characteristics can be interpreted from sedimentary environment
descriptions for geotechnical purposes.
Coal seams are more likely than not to be encapsulated within fine-grained sediments which generally produce lower
strength rock materials. Such sedimentary packages may also be more likely to form locations for compaction or
flexural slip shearing and preferred glide surfaces for thrust faulting. Coal measures rocks immediate to seams are
formed by flood-plain processes and will include various combinations of fine- and coarse-grained sediments. The
provenance of the sediments may vary, with no guarantee that coarser particle-sizes are quartz-rich. The cycles of coal
formation may be interrupted by marine transgressions, uplift or erosion.
In the Bowen Basin, a distinguishing feature of the coal measures sequences is lateral migration of channel-sand bodies
associated with differential compaction of off-stream deposits during burial. Sedimentary structure dip may therefore
be discordant to the average seam structure dip. This implies a potential for mining to encounter slip along features
inclined at a low- to high-angle to average seam structure dip.
Coal seams form in areas of generally low-energy deposition and, in many cases, contain thin zones of fine-grained
sediment. Some of these zones may lithify while others remain as clay bands. The clays may be the product of ash
falls, but in any event would have been subjected to circulation of organic fluids at possibly elevated temperatures,
leading to sometimes unusual mineralogies with very low strengths. Because of intense compaction during formation
of the coal seam and low strength zones being loci for deformation during tectonic processes, clay bands in and around
coal seams are often found in a sheared state.

3.2 STRUCTURAL FEATURES

Rocks exposed by open cut coal mining will have experienced significant unloading by geological-scale uplift and
erosion processes. Due to the anisotropic nature of sedimentary deposits, directional unloading will almost inevitably
result in the more pronounced bedding contacts becoming mechanical defects. Inevitably there will also be across-
bedding tensile fracturing which should be exposed as two sets of mutually orthogonal joints that are also typically
orthogonal to bedding defects.
Coal measures will also typically exhibit structural disturbance in the forms of faults and other joint orientations related
to the strain changes imposed by tectonic events. Igneous processes subsequent to lithification may also result in many
other types of disturbance to the coal seams and surrounding strata. Typical igneous-related disturbance includes
faulting, often with thrusting that glides along bedding surfaces and can be very difficult to identify. In older coal
basins there may be multiple episodes of such disturbances. It should be noted that many apparently normal and reverse
faults may actually be lateral ramp features of extensive thrust sheets. These factors may all be significant when
considering the stability of rock slopes.
Cycles of weathering and erosion have two effects on coal measures rocks. There may be deep erosion and oxidation
which tends to concentrate along structurally disturbed areas, leading to unconformities with unpleasant geotechnical
characteristics. Younger, typically Palaeogene to Neogene channels with sand or gravel beds, subsequently infilled or
overlain by volcanic deposits, are widespread in the Bowen Basin and other Queensland coal deposits. These
circumstances are associated with some of the most challenging geotechnical problems for mining. Both erosion and
tectonic thrusting lead to concentrations of high horizontal stresses particularly where there are contrasts in rock
strength and stiffness and around structurally disturbed zones.

86 Australian Geomechanics Vol 48 No 1 March 2013

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

3.3 EXPLORATION DATA GATHERING

The coal industry has not been very good at collecting geological information of sufficient quantity or quality for
geotechnical interpretation purposes. At the same time, civil-based geotechnical specialists have not been widely
represented in data collection for the industry. It is therefore important, for any coal mining project, to carefully review
data requirements and the means by which data can be collected to meet those requirements. Rosengren et al., 2010
provides a suitable overview of geotechnical requirements for coal mining exploration purposes.
Across the coal industry, routine collection of geotechnical data is generally limited by the predominant use of open-
hole or chip-hole drilling methods for exploration purposes. Specific geotechnical drilling is undertaken for project
purposes, generally but not always with fully cored boreholes. Deliberately non-vertical drilling is still very unusual
despite its advantages for mapping absolute defect orientations. All drillholes tend to drift directionally within
anisotropic rock masses because the rotating cutting tip responds to and interacts with the predominant joints.
Exploration geologists therefore place significant reliance on borehole geophysics to provide detail to maximise the
value of open-hole drilling. A wide range of geophysical tools have been in use for decades but their interpretation for
geotechnical purposes has generally been limited to measurement of density, broad-scale identification of lithology and
sedimentary sequences, and estimation of rock material strength through correlation with UCS tests on core (Davies and
McManus, 1990). Particular care must be taken that the core used for UCS testing is preserved in the least-disturbed
state, which mainly applies to preserving the as-cored moisture content. Recently, considerable progress has been made
in deriving a rock mass characterisation parameter from combinations of different geophysical data channels (Hatherley
et al., 2008).
Routine exploration drilling provides negligible evidence of defects other than detecting faulted or intensely fractured
intervals. Because of borehole deflection, and the non-oriented coring procedures that are in general use, only a relative
sense of structure orientations can be inferred from most cored geotechnical drilling. More recently, oriented structure
features can now be routinely interpreted using borehole scanners which are deployed during geophysical logging.
Borehole data does not, however, provide any information on the lateral extent and connectivity of defects.
While the civil-based community generally logs geotechnical data in accordance with recognised standards, most
logging for the coal industry is undertaken by exploration geologists with generally minimal understanding of
geotechnical issues. Accordingly, the industry has until recently applied practices and coding systems that have been
developed to suit the particular geological modelling systems of individual organisations, which have confusing and
conflicting codes, and which often do not conform to recognised standards. The industry-based initiative, CoalLog
(Larkin and Green, 2012), has addressed these issues, but widespread adoption will take time and the existence of a
large body of legacy data collected under former schemes will complicate interpretations for some time to come.

4 SLOPE FORMATION AND MINING PERFORMANCE ISSUES

Slopes are formed by blasting and digging where the characteristics of available equipment and the mining schedule
may be controlling factors for slope design. The geotechnical specialist must develop a good working knowledge of
slope formation processes. Not only is this necessary in order to understand the hazards that arise from both excavated
and dumped spoil slopes, but the ability to communicate geotechnical conditions and requirements relies on being
conversant with mining practices and terminology. A general review of many mining issues related to coal basins may
be found in AusIMM (2009). The following matters relating to blasting, excavation and dumping are of particular
importance to slope stability, with additional consideration of the effects of stress conditions on material behaviour.

4.1 BLASTING
In spite of decades of experience, mine operators routinely achieve minimal connection between geotechnical
knowledge of ground conditions and designs for productive blasts that minimise adverse impacts on excavated faces.
Modern understanding of blast design requirements in bedded and jointed rock masses is still dominated by
requirements for rock material breakage (typically in terms of UCS and burden), effectiveness of relief or cast (typically
in terms of pattern layout and timing), and compliance with operating constraints (typically in terms of vibration and
noise levels). A useful guide to blasting and blast design is Scott et al. (1996).
Inadequate blast design can have a major impact on mining costs and may also contribute to hazardous or unstable
excavated slope conditions. Slope angles and bench layouts can have a large influence on the effectiveness of blast
design, so geotechnical design recommendations for slope profiles should include consideration of the capabilities and
limitations of blasthole drilling rigs and site work practices.
Blastholes must be drilled, stand open and be successfully charged with explosives. Usually it is not possible to
optimise blast design for all of the variations of actual ground conditions and many compromises must be accepted. In
addition to rock material strength, some measure of rock mass characteristics such as RQD or GSI should be

Australian Geomechanics Vol 48 No 1 March 2013 87

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

considered. However, the varying lithology and anisotropy of coal measures rock masses means that specific
lithological units, jointing patterns and their responses to blasting must also be understood in order to achieve a
successful blast outcome. Major localised features such as faults and shear zones, if unrecognised, may also adversely
affect fragmentation and heave.
Presplitting is a well-understood methodology for creating a controlled blast outcome that minimises backbreak and
damage. However, environmental constraints may require presplits to be stemmed for noise control. Presplits based on
production blasthole diameters can be problematic because wider blasthole spacings lead to a greater likelihood of
intersecting joints that dampen the intended inter-hole fracture effect. Stemming of presplits results in unwanted
confinement with uncontrolled backbreak of intended crestlines and gas-pressure opening of connected, adversely
oriented structures. Techniques for overcoming excessive wall damage associated with stemming of presplits are very
limited and are reliant on site factors which may change for every blast pattern.
The practical constraints of tight schedules, impacts of weather delays on loaded holes and other operational
considerations are often in conflict with requirements for good blast outcomes. Unsatisfactory outcomes often require
responses from geotechnical specialists for hazard management purposes. Geotechnical specialists in such roles should
have a working knowledge of blasting performance, which can best be developed from examination of slope exposures
and discussions with mining personnel. Explosives suppliers can also provide specialist advice regarding blast design
issues and unplanned blast outcomes.

4.2 EXCAVATION
Excavated slope profiles for rock overburden materials and coal must be designed in accordance with safe, practical and
efficient working methods and a clear understanding of the hazards that will remain from the excavation process.
Overburden stripping and coal mining excavation is carried out by excavators, shovels, and draglines, assisted by
dozers. Bucket-wheel excavators or continuous mining machines are not currently used in the Bowen Basin or Hunter
Coalfields. Diggability characteristics, and capabilities and limitations for dig-face trimming, vary markedly for these
different classes of machines. Field trials with experienced and skilled operators are most reliable for assessing digging
and trimming outcomes. Guidance may also be sought from equipment specialists and suppliers, but such advice must
be tempered by observations of achieved outcomes.
Both excavators and draglines dig downwards, which provides better options for trimming excavated faces to a tight
condition with minimised potential for subsequent falls of loose ground. Shovels and bucket-wheel excavators dig
upwards with an increased likelihood of leaving a loose face condition. In some dragline digging methods a dozer is
used to cut the final excavated face, and for certain materials and blast outcomes dozer-cutting can result in a loose face
condition with an elevated risk of rockfall impact on the dozer.
Efficient excavated batter angles are generally relatively steep, in the order of 60° to 80°, in all cases. If geotechnical
strength design considerations require a flatter slope then this may be difficult to achieve in practice without creating
rockfall or facefall hazards. Significantly, this may be more of a problem for weathered ground and free-dig conditions
where blasting is not undertaken and natural sub-vertical or low-angle fractures and fissures can result in batter-scale or
multi-bench instability. These ground conditions can be particularly risky for excavators, which operate most
efficiently by “top-loading” where the machine sits on the top of its digging pass which is typically 4 m to 5 m high,
and called a “flitch” in mining parlance. Double-benching is a related method where two flitches are excavated closely
side-by-side. For both top-loading and double benching the excavator may be exposed to sudden and damaging face
collapse, including collapse of the bench and the machine.
Shovels are less exposed to such hazards, and can dig single batters that are very steep and notionally stable to a height
approximating that of the boom sheaves (about 15 m to 18 m for typical large modern shovels). Such shovel-excavated
batters create hazards for subsequent work below, particularly for drilling and blasthole loading activities. Loose and/or
oversteepened face hazards can be managed by creating an exclusion zone demarcated by a windrow on the operating
bench. Effectively this turns a steep loose face into a notionally flatter inter-bench angle defined by the windrow and
the crest, with the width behind the windrow dimensioned to achieve inter-bench and overall slope profile stability as
well as batter-scale rockfall catch-capacity.
Draglines are primarily deployed to expose coal by digging and dumping to limits defined by the machine dig depth,
dump height and swing radius. Thus draglines are most cost-effective for relatively flat coal seam dips and relatively
long strips of limited width. Highwall batter heights in the range of 40 m to 65 m can be excavated with the two
dragline size classes currently in operation in the Bowen Basin and Hunter Coalfield. A range of prestrip excavation
techniques is required for greater pit depths, with the prestrip spoil surcharging the dragline spoil dump profile.
Dragline stripping requires excavation of a “dig” or “undercut” batter below the level of the operating bench. This
batter may be for a temporary key cut as part of an in-pit bench operation, or for the final spoil lowwall profile.

88 Australian Geomechanics Vol 48 No 1 March 2013

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

Excavated lowwall batter angles typically range from 35° to 45° depending on the nature of the spoil material.
Sometimes a dragline is used to create additional dumping space by means of a pull-back operation which involves
sitting the machine much higher than a normal operating bench level where it is exposed to a potential multi-bench
lowwall collapse hazard. Pull-back slope profiles may have multiple excavated spoil batters and require appropriate
stability assessment
Spoil batters may also be excavated as part of dozer-push stripping methods, or when older spoil dumps have to be
rehandled in order to access coal that has become economically attractive due to changes in market conditions.
Irrespective of the reason, excavated spoil batters should be dug as steep as possible. Some types of spoil can sustain
cut batters steeper than the natural rill angle of the dumped spoil, but this is dependent on the height of the batter,
nonlinearity of the shear strength envelope, and the significance of coarse particle interlock and/or matrix suction within
the spoil fines.
Overall excavated pit wall heights currently range from 80 m to 300 m and planning has been undertaken for final wall
heights approaching 500 m.

4.3 DRAGLINE IN-PIT BENCHES

In deeper pits, dragline in-pit benches are frequently required to maximise stripping efficiency but this also introduces a
range of stability hazards for the machine. Typically the dragline will make a key cut to form the final highwall batter
and a temporary batter on the bench side. The bench batter is dug in shot rock, and if low strength bedding-parallel
surfaces have not been disrupted by the blasting process there is a potential for bench instability to occur under dragline
loading. Low strength surfaces include shears and clay bands within the overburden, coal seam and immediate floor.
Draglines are also subjected to sliding hazards when there is insufficient frictional strength at the bench-tub contact and
bench collapse hazards when weak mud is left in the previous strip void under the bench profile. Figure 1 shows the
range of potential instability hazards for an idealised in-pit bench configuration.
Dragline tub loading is complex, consisting of torsional shear during the swinging and plugging phases of the dumping
cycle combined with normal stresses that vary with the eccentricity of the bucket load during the dumping cycle. While
strictly a 3D loading problem, dragline in-pit bench stability can be adequately approximated by 2D modelling that
includes the average tub contact pressure.

Figure 1: Potential geotechnical hazards for dragline in-pit bench operations

The risks associated with the hazards of draglines operating on in-pit benches can be managed in two ways. Stability
analysis can be used to determine acceptably low likelihoods of instability. Alternatively, the instability consequences
can be limited by positioning the machine tub so that potential instability mechanisms do not encroach on the tub
position. Based on the observation that backscarps for bench sliding mechanisms are inclined typically at about 60°, a
tub exclusion zone can be defined by extending a line upslope at this inclination from the coal edge toeline. This so-
called 60° line rule is particularly useful where there are significant uncertainties associated with bench instability
hazards.

Australian Geomechanics Vol 48 No 1 March 2013 89

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

4.4 SPOIL DUMPING

Dumped spoil materials form slope angles that are controlled by particle rilling. The dumping process translates energy
into initially linear kinetic energy which is then dissipated by sliding resistance and inter-particle collisions but also
translated into rotational kinetic energy. The net effect is for the finer particle to form a thin mantle over the upper rill
slope and the coarser fraction to segregate towards the lower slope. Rill angles are thus a function of particle
kinematics as much as interparticle friction, and are typically within a few degrees of 37°. The interparticle fabric of
the dump is complex, depending on the sequence of dumping and the range of source materials that become mixed at a
particular location within the dump.
Typical dump batter heights for draglines are in the range of 40 m to 60 m above operating bench level. The overall
profile of a typical dragline spoil lowwall combines the dig and dump batters with a remnant operating bench and is
typically 80 m to about 110 m high. Typical truck or continuous spreader dump lift batter heights range from 15 m to
30 m, with the overall dump slope designed to achieve adequate stability as well as adequate control of rolling rock
hazards by the provision of benches between lifts.
Stability of truck dump tip-heads is also strictly a 3D loading problem that is compounded by the time taken for freshly
placed spoil material to settle and achieve long-term strength. Due to the dynamic effects of reversing and braking, the
load on the rear axle of a truck can exceed 90% of the gross vehicle weight. Specialised 3D stability analysis (USBM,
1991) is available for design of truck dump tip-head loading, but reliable outcomes are reliant on adequate allowance
for the reduced short-term strength of the dumped material.
Overall dumped spoil slope heights currently range from 80 m to 360 m and planning has been undertaken for final
dump heights in excess of 500 m.

4.5 STRESS PATHS AND STRESS CHANGES

Due to the non-linear characteristics of material shear strength development, temporary matrix suction effects and the
stress paths induced during the cutting of faces, it is possible to achieve cut batters much steeper than the rill angle.
This apparent stability may be temporary and its limits with height and time are unpredictable. Oversteepening may
thus lead to shallow-seated instability which can be a significant hazard where it directly affects the safety of mining
personnel and equipment.
The formation of slopes, either excavated or dumped, causes significant stress changes. In the case of excavations,
relief of stress causes relaxation deformations of walls and heaving of floors. Dump development causes settlement and
bulging of spoil under self-weight loading as well as floor settlement. Unloading and loading deformations have a
time-dependent component, but only the settlement and bulging of spoil is likely to be large enough to be noticeable by
visual observation. Many mining personnel are unaware of the distributions of stresses and shear strength mobilisation
in slopes, and the degree to which water plays a role in slope stability. Concepts of slope stabilisation by buttressing,
crest unloading and drainage may therefore not be readily understood by site personnel and require careful explanation
by geotechnical specialists in order to achieve desired outcomes.
As pits become deeper the stresses near the base of excavated faces and dumps become higher, deformation effects are
more likely to be observed and concerns about stability are more likely to be raised. Deformation effects and stability
concerns are also more likely where adverse geological or groundwater conditions are encountered. Modern monitoring
tools can provide almost real-time information on deformations, but interpretation in terms of stability requires
reference models. A wide variety of deformation and water pressure modelling techniques are now available and in
many circumstances the models are principally limited by availability of reliable parameters.

5 GEOTECHNICAL PROPERTIES
5.1 DENSITY
The bulk density of in-place and spoil materials varies between certain limits and accurate knowledge of density values
is not critical for stability analysis. Bulk density is, however, of great concern for productivity measurement and
reconciliation. In situ rock density can be reliably determined from geophysical logs and there is circumstantial
evidence that spoil density can be measured reliably by similar methods for some materials. The changes in volume of
a given mass of material, from its original in situ state through a chain of handling processes to its final placement state,
are described by bulking factor or swell parameters. Too often these parameters are guesstimated, even when they are
critically important for planning and budgeting purposes, as measurement under field conditions is challenging.

90 Australian Geomechanics Vol 48 No 1 March 2013

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

5.2 SIGNIFICANCE OF POTENTIAL INSTABILITY MECHANISMS

On a single-batter scale, structure-controlled (kinematic) instability is likely to be critical but as the slope scale
increases, rock mass strength-controlled instability becomes more significant.
In order to assess structure controlled stability it is necessary to have reliable data. While directional attributes may be
observed or interpreted from boreholes, it is the position, spacing, continuity and persistence of defects which determine
the likelihood for an instability mechanism to occur at a particular location on a pit wall. The lateral extent of joints in
coal measures rock masses varies, with continuity and spacing being related to bedding parting separation. Tectonic
overprint defects are more likely to be continuous and persistent than “normal” across-bedding joints. Realistic
assessment of these lateral extent attributes requires field observation and measurement. Currently, the most productive
methods for data collection are traditional face mapping and stereophotography, with the latter preferred in
circumstances where safe access restrictions apply.
For rock mass strength-controlled instability a regressive step-path mechanism is most likely to develop in practice
(Baczynski, 2000; Stacey (2006). For design purposes it is currently impractical to simulate such a process in detail,
and simplistic arcuate mechanisms are usually invoked together with equally simplified shear strength models.
Observations of collapse mechanisms for dumped spoil slopes are critical for reliable stability assessment. On the scale
of a single rilled batter (a truck-dump lift or a dragline dump slope), collapse is observed to follow an arcuate path when
external loads such as mining equipment are the trigger. The other triggers for batter-scale collapse are superficial
wetting-up of the batter, or relative weakness of the basal layer, leading to slope-parallel sliding that Simmons and
McManus (2004) described as a rill-wedge mechanism. For dumps with complex internal structure, such as a multi-lift
truck dump or a full dragline spoil lowwall profile, a deep-seated wedge sliding mechanism with a steep backscarp and
a reverse midscarp is most likely to develop. Such mechanisms have been observed for some time (Gonano, 1980) and
have recently been re-examined in some detail with the benefit of modern analytical tools (Duran, 2012). What is less
clearly understood still is why wedge-shaped mechanisms develop in preference to arcuate mechanisms that appear
(erroneously) to be more critical when applying 2D limit equilibrium analyses.

5.3 SHEAR STRENGTH

Determination of the shear strength of coal measures rock masses is problematic and requires informed judgement. For
individual rock defects there are many proposed models for shear strength and laboratory tests can be undertaken on
field lump or core samples, but in most coal mine situations there is very little reliable data and reasonable assumptions
are required more often than not.
For coal measures rock masses it is clearly unrealistic to model each variation in lithology and, in practice, a “rock mass
unit” simplification is generally adopted. Each rock mass unit has a representative lithology, UCS and general defect
fabric attributes. A simple linear Mohr-Coulomb shear strength model can be adopted for each unit, with the
parameters determined from an envelope of backanalyses for stable rock slopes. Alternatively, the nonlinear
Generalised Hoek-Brown (GHB) shear strength model (Hoek and Marinos, 2007) may be utilised. This requires
adoption of several parameters including GSI. For anisotropic rock masses the meaning of GSI is not obvious and both
Hoek and Brown have repeatedly stated in public communications that application of the GHB criterion to relatively
“soft” bedded rocks was never intended and is not recommended without extreme caution. In the absence of any better
alternative, however, the ROCLAB freeware (RocScience, 2005) is an attractive method for constructing a nonlinear
strength envelope but should be applied with great care.
Bedded rock mass anisotropy implies different strength parameters applied to different directions. Several
mathematical models for implementing such a scheme have been developed, initially for soil mechanics purposes, and
are now implemented in stability analysis codes. The application of such models is a matter of judgement and
hopefully future case history studies will provide guidance. The shear strength of blasted rock, prior to excavation and
dumping, is a matter of judgement since it is likely to contain anisotropic features but also exhibit some soil-like
properties.
Once handled and dumped, however, both blasted and free-dug materials become spoil in a general sense. The linear
Mohr-Coulomb shear strength criterion is the simplest and most convenient starting-point for expressing spoil strength,
but selection of appropriate parameters for such a wide range of source materials has always been a matter requiring a
combination of testing and judgment. BHP Australia Coal, now BMA Coal, developed an empirical method for relating
observations of source material characteristics to laboratory test data and backanalyses of full-scale dragline lowwall
slope profile collapses. This is now known as the BMA Coal spoil categorisation scheme (Simmons and McManus,
2004).
Several considerations are involved in applying the BMA Coal spoil categorisation scheme. Identification of the
category of spoil is based on visual-tactile field observations. The shear strength mobilised in the material depends on

Australian Geomechanics Vol 48 No 1 March 2013 91

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

its moisture state. The normal, unsaturated condition applies within the spoil mass but a saturated condition is believed
to apply to a zone typically within 5 m of the floor of the dump. In addition, saturated material which has been
sufficiently deformed in shear will mobilise a sheared state which corresponds to the residual strength of the fines.
Table 1 shows the BMA Coal shear strength parameters as currently (since 2004) understood. It should be noted that
the parameters listed in Simmons (1995) were updated prior to 2004.
Since 1995 the heights of spoil dumps have increased substantially. Simmons and McManus (2004) specifically noted
that spoil shear strength envelopes are most likely nonlinear and that the parameters in Table 1 strictly apply to a limited
normal stress range only. This means that for current practice based on analysis codes it would be appropriate to
modify the parameters in some manner to account for expected strength envelope curvature. At present any
modifications can only be empirical, but the author has used reasonable judgement for application to several dumps
with design heights up to 450 m and one where the dump height is currently 360 m. It is expected that a more
systematic approach to strength of spoil will emerge from research including large-scale testing that is currently in
progress.
Table 1 BMA Coal Shear Strength Parameters for Categories and Mobilisation Modes
(Parameter standard deviations in italicised parentheses)

Category Unsaturated Saturated Remoulded

γ (kN/m3) c' (kPa) φ’ (deg) γ (kN/m3) c' (kPa) φ’ (deg) c’=0 kPa, φ’ (deg)
1 18 20 25 20 0 18 18
(1) (10) (2.5) (1) (0) (3) (1.5)
2 18 30 28 20 15 23 18
(1) (15) (3) (1) (7.5) (2.5) (1.5)
3 18 50 30 20 20 25 18
(1) (15) (2) (1) (10) (2.5) (1.5)
4 18 50 35 20 0 30 28
(1) (15) (2.5) (1) (0) (1.5) (2)

5.4 STIFFNESS AND DEFORMATION CHARACTERISTICS

Sophisticated numerical tools for modelling the stress-deformation behaviour of pit slopes are now available and
research into improved material models and solution methods is likely to continue for some time. In addition to shear
strength, such models require elastic stiffness parameters as well as parameters to describe non-elastic yield
deformations. Empirical stiffness parameters for spoil materials can be guesstimated from visual-tactile observations,
laboratory tests, and published information. It is also well understood that stiffness for soil (and by association, spoil)
materials is nonlinear, with different test methods measuring small- to large-strain parameters. In the absence of
reliable data, it is a matter of judgement whether modelling of such nonlinearity can be justified. Stiffness for rock
masses can be determined from consideration of modular ratios for typical rock materials, plus the influences of defect
fabric (Hoek and Diederichs, 2006).

6 ROLES OF WATER AND TIME IN SLOPES

Water affects slope stability in a number of ways: physical and chemical effects on the materials, pore water pressures,
and cleft water pressures. There is always some time scale implied in these processes, but the scale can be uncertain
and highly variable. The geotechnical specialist must be aware of the potential time effects and evaluate them on a
needs basis. The time scale that is important for design purposes may range from weeks, for a localised pit
requirement, to “forever” in the case of mine closure planning.
There are no easy methods for measurement of water pressures in many mining situations, so the normal state-of-
practice is to make reasonable assumptions. It is necessary to understand how critical the outcomes are to such
assumptions, and to take into account the sensitivity of the outcomes to the assumptions made. Mining personnel tend
to be over-optimistic about water conditions, and are more conditioned to working in dry rather than wet climatic
conditions. Geotechnical specialists must recognise this, but still be realistic about assumptions regarding hazards
involving water pressures.

6.1 PHYSICAL AND CHEMICAL EFFECTS

Many coal measure rocks are sensitive to water. The natural pore water may be of high salinity, so that access to rain
water or fresh water runoff can affect clay minerals in highly deleterious ways. Many softer rocks are poorly lithified,
with cementitious agents that are readily attacked by percolating water. The net effects are tendencies to slake and/or

92 Australian Geomechanics Vol 48 No 1 March 2013

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

disperse, releasing weak clays and forming pit mud deposits that must be removed to promote spoil dump stability.
Rock walls also deteriorate under prolonged exposure to air and water, becoming looser and more susceptible to runoff
erosion. For this reason it is wise to stand batters as steep as possible to minimise runoff exposure, and to control
overall slope angles by well-drained benches.
Physical deterioration of rock materials can occur faster in spoils that are readily exposed to air and water. There is
little detailed knowledge of long term spoil mass evolution processes, other than in selected cases relevant to soil
generation. Recent experiences and current research indicate that spoil is generally of high and variable permeability so
that internal drainage tends to be rapid, provided that the lower spoil does not become clogged with migrated or locally
generated fines.
Pyritic materials will oxidise on exposure to air, forming acid drainage when water is allowed to percolate. The general
subject of acid rock drainage is not addressed here, other than to recognise that it may be a serious design issue which is
likely to involve the geotechnical specialist.
The engineering implications of these water effects must be understood, with the most difficult matter being the
identification of appropriate shear strengths. If reasonable assumptions cannot be made then shear strength testing must
be undertaken. Selection and preparation of representative samples may be difficult so the justification for such work
must be well founded in terms of risks, particularly contrasted to the risks associated with not sampling and testing.

6.2 PORE PRESSURE AND CLEFT PRESSURE

Pore pressures in coal measures rock masses are not simple matters to assess. Water is present within the rock material
matrix but is effectively bound compared to “free” water that circulates within fractures. Coal seams are often the
major aquifers within coal measures strata, but many intervals may act as leaky aquitards. Measurement of water
pressure within the rock mass may thus lead to complications, rather than clarifications, in determining groundwater
flow regimes. It is essential to undertake measurements carefully and to include sufficient redundancy and detail so that
interpretations can be made with as much certainty as possible. Crude standpipes and open slotted wells are unsuitable
for water pressure measurements. The preferred method is to install multiple vibrating-wire sensors by fully grouting
strings into boreholes and providing surface facilities for automated logging and data transmission. Pore pressures
within spoil materials are also not easily assessed. Apart from the significant difficulties of drilling and installation,
spoil may contain large localised voids and may be subject to rapid drainage.
Cleft water pressures may be considered as local manifestations of water which enters specific defects by means of
pathways to the surface. Of principal concern are defects which are opened by the effects of blast vibration or by
deformation that causes the rock mass to dilate. It is essential for hazard management purposes to insist on effective
runoff management and methods for assisting drainage at the base of spoil dumps. Good practices include asking
“what-if” questions such as levees or ditches overtopping after accumulation of sediment. The potential for instability
to be triggered by water entering cracks, or sediment blocking drainages, must be recognised and communicated to
relevant mine personnel.

6.3 MODELLING OF GROUNDWATER PRESSURE DISTRIBUTIONS

The simplest representation of groundwater pressure is a phreatic surface which is considered to be the line where the
pore water pressure is zero and below which materials are saturated and positive pore water pressures can be
determined with reference to depth below the line. Phreatic surface elevation changes correspond to pore water flow. It
is generally not realistic to model flow in anisotropic and heterogeneous media using a simple phreatic surface, but the
phreatic surface approach may be adequate for relatively simple material configurations and slope profiles. More
complex flow conditions, as well as perched aquifers, can be modelled by piezometric lines which are specific to a
defined zone.
Geotechnical specialists now have ready access to finite element codes for modelling groundwater pressure
distributions. As with other aspects of geotechnical assessments, such models are only reliable if they are based on
credible data and realistic boundary conditions. While it is generally understood that permeability is anisotropic, the
degree of such anisotropy, the absolute value and direction of the principal permeability and the relative values of
principal permeabilities for other material zones, together form a large body of mostly unknown parameters.
Confidence in the reliability of finite element groundwater modelling can only be justified from backanalysis and from
calibration to monitoring data.

Australian Geomechanics Vol 48 No 1 March 2013 93

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

7 SLOPE STABILITY ASSESSMENT

7.1 ANALYTICAL METHODS AND STABILITY CRITERIA
Methods of slope stability were first developed prior to the advent of computers. Assumptions and simplifications of
many methods have been incorporated into modern codes and it is arguable that only the more recent and rigorous
methods based on realistic assumptions should be used in the modern era. Rigorous methods require checks and
balances to determine if the assumptions are valid and if the resulting stress distributions are reasonable. Some codes
do not provide ready access for such checks, and it is now incumbent on geotechnical specialists to demonstrate that
sufficient checks have been carried out to establish the reliability of a computational outcome.
Almost all slope stability analysis codes are based on methods of vertical slices where assumptions about stresses are
used to calculate stresses on vertical boundaries and the base and top of each slice. This is perfectly reasonable except
for spoil dump assessments, where there is overwhelming evidence of stresses being critical and sliding being localised
on non-vertical boundaries. Only the Sarma method is both rigorous and formulated in terms of arbitrary slice
boundary orientations.
Another aspect of slope stability assessment using the usual 2D limit equilibrium techniques is that the outcome is
consistent with plasticity theory. By imposing a mechanism and calculating the equilibrium of forces, the solution is
merely a bound to the true solution. Limit equilibrium models do not consider compatibility of displacements, but
alternative stability formulations are now available which work in terms of compatibility and which therefore impose a
different bound on the true solution. Limit equilibrium methods also compute a single parameter, the Factor of Safety
(FOS), which is the factor by which shear strength parameters must be scaled simultaneously in all parts of the model in
order to bring the proposed mechanism to the point at which it will collapse. Needless to say, the advent of powerful
computational facilities means that uncertainty of some input parameters can be modelled, thus leading to a frequency
distribution for FOS outcomes which may, not correctly, be called a Probability of Failure (POF). In modern mining
risk management terminology, the POF is inherently more attractive than the FOS because it can be directly related to
likelihood and connected with consequence to measure risk. It is incumbent on geotechnical specialists to be realistic
when presenting the results of analyses and when accounting for uncertainties and likelihoods of collapse.
Alternatives to the FOS are now available based on stress-deformation analysis. These are more demanding to model
and also much more computationally intensive than limit equilibrium analysis but allow for variable mobilisation of
strength and identify a parameter, the Strength Reduction Factor, which applies globally to a critical deformation
mechanism. The SRF is analogous, but not necessarily equivalent, to the FOS. The value of SRF methods, relative to
widely established FOS methods, is yet to be convincingly demonstrated for coal mine slope configurations because of
the general reliance on assumed or empirical input parameters and a lack, to date, of comprehensive case histories.
Much has been written in the geotechnical literature about the use of computed FOS values as stability acceptance
criteria. As in many engineering fields where judgements are made in the face of uncertainty, there has been a reliance
on indirect association of a parameter with an indirectly linked outcome. Limit equilibrium analysis makes no reference
to deformations, yet the computed FOS is typically used as a proxy for limiting slope deformations. Miners typically
will accept what appear to be relatively high levels of uncertainty and risk because consequences can usually be
managed more effectively in the controlled environment of a mine site, in comparison with geotechnical engineering in
the wider community. Table 2, after Simmons (1995) has been indirectly proven to be a reliable framework for
acceptance of stability acceptance outcomes. It is based on FOS calculations for apparently stable, marginally moving,
and spectacularly failed slopes and is consistent with the wider geotechnical literature.
Table 2: Suggested Factor of Safety Relationships for Open Pit Coal Mining
FOS POF Indications of Movement Action/Response
1.5 to 1.2 < 0.001 minor, stationary cracking acceptable
surface cracks opening, some check sensitivity to
1.2 to 1.1 0.001 to 0.02 ravelling, observable bulging, assumptions, review risk
minor scarp formation assessment
significant observable movement, seek stabilisation options,
1.1 to 1.0 0.2 to 0.8 loosening and isolated rolling of introduce access restrictions,
rocks, scarps, mechanism forming upgrade monitoring
continuing movement, mechanism manage consequences,
< 1.0 > 0.8
well defined implement stabilisation options
In Table 2, the FOS value ranges are regarded as more useful and reliable than the corresponding POF values. This is
because, despite ready availability of computational power, the relationships between FOS and movement are much

94 Australian Geomechanics Vol 48 No 1 March 2013

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

better understood than any relationship, when it is proven to exist, between POF and movement. In summary, a FOS
not less than 1.2 is considered to be a good general target for all pit slope design, but a FOS more like 1.5 may be
justified in brittle rock slopes where the consequences of collapse are highest. FOS values as low as 1.05 may be
justified in cases where the likelihoods and consequences of collapse are well considered and where mining personnel
are satisfied that appropriate risk controls can be implemented. What is essential, in any reporting of stability
assessments, is a clear statement of the data and methodologies used for calculations and the acceptance criteria applied
to the outcomes.

7.2 LOADING EFFECTS

Gravitational body forces are straightforward to define and as accurate as the knowledge of density for both in situ
materials and dumped spoil. Complete initial states of stress are more complex. Virgin lithostatic stresses are broadly
understood for Australian conditions (Hillis et al., 1999), but in many mining areas local factors including prior mining
history will cause significant changes to the in situ stress for a particular project location and context. Careful and
thoughtful numerical modelling may be required to define the local project initial stresses in anisotropic, dipping rock
masses. Virgin states of stress in dumped slopes are not well understood but initial spoil placement is equivalent to an
active stress condition and subsequent loading is equivalent to confined compression. Lateral pressure estimates based
on the “at-rest” condition are probably reasonable for most purposes.
Dynamic loadings are generated by machine operations, blasting and earthquake. The principles of earthquake loading
are set out in AS1170.4-2007 but are formulated primarily for structural engineering purposes. For open pit slope
purposes the procedures used for earthquake design of embankment dams can be adopted. A design vibration level is
determined, typically as plots of peak acceleration versus likelihood of occurrence for different event intensities, from
which the designer can select an appropriate level of horizontal acceleration. Appropriate levels of vibration associated
with blasting and machine trafficking may be measured or guesstimated. In practice, for the majority of vibrating load
cases, a pseudo-static assessment can be undertaken using the USACE screening methodology (Hynes-Griffin and
Franklin, 1984).
Discrete machine loads are primarily static, but are strictly 3D and localised to the machine contacts. For trucks and
dozers an equivalent larger-scale 2D pressure can be derived from the gross operating weight and the projected gross
contact area. The exception is for draglines where the tub area is large enough, and the consequences of collapse severe
enough, that the true 3D contact stress may be adopted as an equivalent 2D strip loading.

7.3 DUMPED SPOIL SLOPES

Dump slopes are designed to acceptably low likelihoods of instability, but many factors cannot be fully understood or
controlled and dumped slope instability risks have to be actively controlled. While materials generally become stronger
in deeper pits, this benefit is counteracted by higher spoil loadings and larger exposure to the effects of runoff and
seepage water. A continuing risk management issue arises because the common understanding of stability is based on
simple superficial observations, and things that look satisfactory may be in fact be only marginally so.
Stability calculations for dumped spoil slopes are straightforward provided that appropriate mechanisms and proven
methods of analysis are used. Because of the predominance of experience and judgement based on reasonable
assumptions and empirical correlations, it is critical to view the materials, groundwater model, shear strengths and
potential instability mechanism as a linked set. Adoption of a different mechanism while maintaining the same phreatic
surface and strengths may lead to a different outcome that is inconsistent with observations and experience (Simmons
and McManus, 2004).
Almost all stability assessments are now made with sophisticated software with the capability for automated searching
for critical mechanisms. The rules used for such searches are mathematical, and outcomes that are inconsistent with
geological conditions (including the internal stratigraphy of a complex dump) and observed slope behaviour are not
acceptable.
A particular problem with in-pit spoil dumps arises when there are relatively weak floor conditions. While spoil masses
are generally ductile and deformable, the floor rocks may be stiff and brittle. Spoil dumps can fail by sliding within the
floor rock, causing heave and breakout where the slope can fail very rapidly and with little warning. Dunbavan (1990)
showed that 90 m dragline lowwall dump slopes have failed on layers as deep as 5m below the pit floor. The depth of
potential heave will depend critically on the breakout strength, which is usually understood except by backanalysis of
failure events. Considerable caution is warranted in stability analyses for very high dumps if there are known or
suspected weak floor conditions. In such situations stress-deformation modelling and the SRF approach may be
warranted.

Australian Geomechanics Vol 48 No 1 March 2013 95

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

Part of the weak floor hazard arises because of data limitations. Even the most careful core drilling may result in loss of
weak and sheared material by flushing. Also, coal samples are normally used for product analysis, and may not be
made available for geotechnical testing of weak layers within the seam and immediate roof and floor strata.
Geophysical logs may provide important clues to supplement information missing from core drilling and sampling. The
geotechnical specialist must search carefully for all forms of information that may indicate sub-floor weakness.

Figure 2 : Example stability backanalysis for dragline lowwall collapse by floor shearing and heaving
Figure 2 shows the geometry of a backanalysis for a spoil dump failure that occurred along a weak band between the
Dysart Lower 1 and Lower 2 seams at Saraji Mine. This weak band continues to be a notorious destabilising influence
when not treated appropriately. The BMA Coal spoil categorisation process was used to assign spoil strengths and the
shear strength of the weak seam was backanalysed taking into account groundwater observations. Several such
backanalyses for different spoil and floor geometries were undertaken, yielding consistent shear strengths for the weak
band that were (much) later confirmed by laboratory testing.

7.4 ROCK SLOPES

For typical Bowen Basin or Hunter Coalfield lithologies, rock material strengths are such that material strength is
unlikely to be a significant factor until pit depths approach 300 m. Several pits currently are close to 300 m deep, and
much greater depths are planned. To date, rock slopes of this magnitude have demonstrated adequate stability with no
signs of unexpected material damage or mass deformation. In East Kalimantan, where rock material strengths are
typically much lower, rock slopes of over 300 m have been designed and successfully excavated. In the author’s
experience most major coal mine rock slope failures are related to faults, shear zones, and tectonically imprinted
defects.

Figure 3: Example stability analysis outcome for bedded rock mass slope with modelled groundwater pressures.

96 Australian Geomechanics Vol 48 No 1 March 2013

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

Figure 3 shows an example result from a 2D limit equilibrium analysis for a 350 m high rock slope design. The
phreatic surface and associated pressure head contours were computed using finite element analysis, with permeabilities
calibrated to piezometer measurements and other groundwater observations from adjacent pits.
Rock slope stability assessment therefore consists of several considerations. Batter- and multi-bench scale instability is
primarily controlled by structure and requires kinematic assessment. Rock mass-controlled instability must also be
assessed, typically using 2D limit equilibrium methods and becomes more critical for higher slopes and more complex
slope profiles and/or geological conditions. As slopes become higher and more geologically complex, the likelihood of
deformation-driven mechanisms increases and assessment should be extended to include stress-deformation modelling.
Because rock slope excavation is an unloading process, the unloading stress paths close to the slope face are also
associated with significant extensional straining. Under these circumstances brittle extensional fracture is inevitable in
most geological materials because of the presence of bonding. Observations of coal measure rock slope instability
confirm that rock mass strength failure is usually brittle and rapid after a relatively stiff precursor phase. Industry-
sponsored research (Sherwood Geotechnical and Research Services, 2011) has provided strong evidence that collapse is
most likely during the final phase of coal removal at the toe of a rock wall, and that stable deformation response has
particular characteristics that can be measured during mining activities. It is more likely than not that the trigger for
development of a collapse mechanism is some combination of geological conditions hidden behind the slope face that
were not detected during exploration, and rock mass damage related to mining processes. This research has highlighted
the value of stress-deformation modelling as a tool for deformation predictions which should be compared to
monitoring data to search for abnormally large movements that may signal impending collapse.

7.5 BOXCUT LOWWALLS

A special case of rock slope stability arises when boxcuts are excavated. In coal mining, the boxcut lowwall is located
at or adjacent to the LOX (limit of oxidation) line. Boxcut economics is driven by minimising the volume of
excavation required to start generating cash flow from coal mining. The ground surface updip of the LOX line is also a
convenient place to dump boxcut spoil where it will not interfere with access to coal. In geotechnical terms this means
boxcut lowwalls are often dug as steeply as possible and surcharged with spoil dumps, even if this means paying only
lip-service to geotechnical design advice.
Geotechnical models for boxcuts are critically dependent on knowing the extent of weathered coal and any associated
weak bands which underlie the boxcut lowwall. Coal typically weathers to a sooty clay that may also have local
flexural shears caused by the deformations accompanying any volume changes involved in the weathering processes.
Boxcuts also require adequate understanding of initial groundwater drawdown effects, since the boxcut is effectively a
large drain.

8 HAZARD MANAGEMENT
8.1 LEGISLATIVE FRAMEWORK
All Australian mining jurisdictions are state-based, and relevant safety and health management requirements follow
international standards that are based on risk assessment. Because the consequences of slope failures can be
catastrophic, there are complex processes involved in seeking and granting approval for mining activities. A key
consideration under these circumstances is the definition of acceptable standards for slope design. Currently all
Australian states are participating in the adoption of uniform workplace safety and health legislation. One of the
matters that have arisen is agreement regarding an acceptable standard for slope design.
Geotechnical guidelines for open pit slope design have been developed recently through a large project involving
international collaboration (Read and Stacey, 2009). These guidelines were developed primarily for large metalliferous
pits, and specific considerations for coal mining and for spoil dumping generally are not represented with the level of
detail that most geotechnical specialists would desire. A proposal for an Australian code of practice for pit slope design
is currently being considered, but is unlikely to be adopted unless it can address the specific requirements and practices
of coal and metalliferous pit design issues.

8.2 RISK ASSESSMENT METHODOLOGY

Under risk-based legislative requirements, mine operators administer safety and health management systems that are
driven by assessment of risks which in turn are generated by likelihoods and consequences of hazardous events. Risk
controls are the systems and processes developed to manage the assessed risks at acceptable levels. Unfortunately, and
unlike the quantitative fault-tree methodologies applied to risk management of major dams, it has become acceptable in
the coal mining industry to assess risks on a qualitative basis.

Australian Geomechanics Vol 48 No 1 March 2013 97

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

Understanding of slope hazards for coal mining is therefore reliant on observation, experience and judgement.
Geotechnical specialists are unlikely to be exposed to slope hazards to the same extent and with the same frequency as
coal mine workers. Slope design is an engineering-level risk control, intended to identify consequences and provide
reliable information on likelihoods. Risk management methodology requires constant attention to monitoring, in the
sense that controls are monitored to check that they are working as intended and to identify changed conditions where
controls need to be revised.

8.3 OBSERVATION AND MONITORING

Most risk monitoring is based on observations, which can range from visual checks to complex movement measurement
technologies involving radar or optical survey. The key principles are to identify unexpected conditions which signal a
level of likelihood that is higher than expected, and to act on observations which exceed tolerable conditions. Most
open pit coal mines now operate with trigger action response plans, or TARPS, which define the actions to be taken and
the persons responsible, when triggers are identified.
But how are triggers identified? And are the current trigger levels appropriate for the risks that are being controlled in
this manner? Most site-based geotechnical personnel are finding their roles increasingly defined by observations and
documentation of pit slope conditions, participation in the issuing of hazard alerts and reports, and updating of risk
assessments. Sometimes this does not leave enough time for rational slope stability assessment, so that likelihood
judgements have to be made on the basis of experience and judgement.
Radar and optical survey movement measurement technologies continue to be implemented. Maintenance and
management of such equipment is often vested in geotechnical personnel and can absorb a significant amount of
resources and time. Setting of movement trigger levels is ultimately a role for geotechnical specialists, and requires
both technical conviction and communication skills. All of these matters constitute an interesting and challenging
environment for career development, but it also takes time and experience to develop the necessary judgement as well
as the technical skills that the industry demands.

9 FUTURE DEVELOPMENTS
Coal mining also generates waste from the coal handling and preparation plants (CHPP) that turn raw mined coal into
marketable products. Traditional CHPP waste streams consist of coarse breaker and plant rejects (CPR) and “tailings”
fines. CPR is essentially a wet slightly silty and sandy gravel that can be handled and dumped in much the same
manner as spoil. Tailings is traditionally a wet slurry that is pumped to tailings storages that may be conventional dams
or purpose-managed pit voids. For the past two decades CPR and tailings streams have also been co-disposed through a
common pipeline system. Some sites are not suited to conventional tailings storages and have implemented mechanical
dewatering of tailings slurry, predominantly using belt press filters, to form a paste-like belt filter reject (BFR) product
that can be handled and dumped by trucks. BFR, and super-thickened tailings slurries, are typically dumped in
purpose-built containment cells within overburden dumps.
Capping and rehabilitation of tailings dams and BFR or slurry cells involves consideration of the soil mechanics
principles developed for soft ground engineering, except that cost minimisation generally precludes ground treatment
by wick drains or vacuum consolidation. Some mines are now actively implementing truck placement of run-of-CHPP
mixes of CPR and BFR directly into spoil dumps. These so-called MPR (mixed plant rejects) schemes are more likely
than not to become the norm for CHPP waste management in the future.
Slope stability considerations for spoil dumps with randomised MPR placement are now emerging. Current designs are
based on fundamental geotechnical principles together with a suitable high degree of conservatism about short-term
(“undrained”) strength behaviour. Consideration of the moisture transfer between MPR and spoil, and the associated
gains and losses of shear strength in each of these materials, is a subject of current research.
As dumps become higher, irrespective of whether or not more complex materials such as MPR are present, there is a
likelihood that particle breakdown together with internal stresses and moisture conditions in the dump may generate a
saturated condition towards the dump base. Deformations associated with placement may therefore create a condition
where dump stability assessments will have to include evaluation and analysis of undrained strength responses.
As excavated slopes become higher, the stress states and scales of natural and stress-induced defects are likely to
generate a class of rock mass response which is not yet well appreciated in the coal mining industry. The metalliferous
mining industry has documented and analysed complex slope deformations where even very high strength rock
materials generate highly fractured, semi-ductile masses. It is expected that the future increased deployment of large-
scale movement measurement systems will provide sufficient case histories for extending geotechnical understanding of
coal measures rock masses to include stress-induced response mechanisms in high slopes.

98 Australian Geomechanics Vol 48 No 1 March 2013

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

10 ACKNOWLEDGEMENTS
The original 1995 version of this paper summarised working knowledge gained from a number of Bowen Basin mine
sites. Alan Davies, Dennis McManus, Michael Dunbavan, John Eckersley and Peter Gruythuysen were acknowledged
at that time for their advice and assistance. Since that time other individuals whose advice, comments, and criticisms
have been particularly instrumental in shaping the author’s current views are Kevin Rosengren, Paul Maconochie,
Graeme Boyd, and Dick Stacey, and this counsel is gratefully acknowledged.
The Bowen Basin Geologists Group remains a fertile source for background knowledge of the geological environments
where open pit coal mining is carried out. Although originating from Bowen Basin experience in Queensland, the
principles outlined in this review have been proven by experience to be equally applicable to open pit coal mining
generally, and this has been confirmed in publications originating from other coal mining regions.
Where possible this review is based upon the prevailing common views of the geotechnical community within the open
pit coal mining industry in Queensland and New South Wales. Notwithstanding, all matters described and opinions that
are expressed here remain the sole responsibility of the author.

11 REFERENCES
Australian Standard AS1170.4-2007. Structural design actions – Earthquake actions in Australia. Standards Australia.
Australasian Institute of Mining and Metallurgy (2009). Australasian Coal Mining Practice. Third Edition,
Monograph 12, ISBN: 978 1 921522 07 9, Melbourne.
Baczynski, N R P, 2000. STEPSIM4 ‘step-path’ method for slope risks. Proceedings, GeoEng2000, an International
Conference on Geotechnical and Geological Engineering, Australian Geomechanics Society, Melbourne, 2000.
CD paper reference SNES0213, 6 p.
Bowen Basin Geologists Group, 1990. Proceedings, Bowen Basin Symposium 1990, Mackay. Geological Society of
Australia.
Bowen Basin Geologists Group, 1995. Proceedings, Bowen Basin Symposium 1995, Mackay. Geological Society of
Australia.
Bowen Basin Geologists Group, 2000. Proceedings, Bowen Basin Symposium 2000, Rockhampton. Geological Society
of Australia.
Bowen Basin Geologists Group, 2005. Proceedings, Bowen Basin Symposium 2005, Yeppoon. Geological Society of
Australia.
Bowen Basin Geologists Group, 2010. Proceedings, Bowen Basin Symposium 2010, Mackay. Geological Society of
Australia.
Davies, A L and McManus, D A, 1990. Geotechnical applications of downhole sonic and neutron logging for surface
coal mining. Exploration Geophysics, Volume 21, pp 73 – 81.
Dunbavan,   M.   1990.     Spoil   instability   caused   by   sub-­‐floor   weakness.     In   Bowen   Basin   Geologists   Group   (1990)  
above.  
Duran,   A,   2012.     Spoil   piles   –   comparison   of   limit   equilibrium   stability   analyses.     Proceedings,   Australia   –   New  
Zealand   Geomechanics   Conference,   Australian   Geomechanics   Society   and   New   Zealand   Geotechnical  
Society,  Melbourne,  pp  529  –  534.  
Gonano,   L,   1980:   An   integrated   report   on   slope   failure   mechanisms   at   Goonyella   -­‐   November   1976.   Technical  
Report  No.  114,  CSIRO  Division  of  Geomechanics,  Melbourne.  
Hatherly, P, Medhurst, T, and MacGregor, S, 2008. Geophysical Strata Rating. End of Grant Report, ACARP Project
C15019, Australian Coal Association Research Program.
Hillis, R R, Enever, J R, and Reynolds, S D, 1999. In situ stress field of eastern Australia. Australian Journal of Earth
Sciences, Volume 46, pp 813 – 825.
Hoek, E, and Diederichs, M S, 2006. Empirical estimation of rock mass modulus. International Journal of Rock
Mechanics and Mining Sciences, Volume 43, pp 203–215.
Hoek, E, and Marinos, P, 2007. A brief history of the development of the Hoek-Brown failure criterion. First published
in the new Brazilian Journal of Soils and Rocks, No. 2, November, 12 p. Available from
<[Link] [Accessed: 28 January 2013].
Hynes-Griffin, M E, and Franklin, A G, 1984. Rationalizing the seismic coefficient method. US Army Corps of
Engineers Miscellaneous Paper GL-84-13, Waterways Experiment Station, Vicksburg Mississippi, July.
Larkin, B and Green, D, 2012. CoalLog borehole data standard for the Australian coal industry. ACARP Project
C21003. Available from <http//:[Link]/content/[Link]?ID=451>
[Accessed: 2 November 2012].
Read, J, and Stacey, P, 2009. Guidelines for Open Pit Slope Design. CSIRO Publishing, Melbourne, 496 p.

Australian Geomechanics Vol 48 No 1 March 2013 99

GEOTECHNICAL SUPPORT FOR OPEN PIT COAL MINING JOHN SIMMONS

RocScience, 2005. ROCLAB. Software to support application of the Generalised Hoek-Brown criterion to laboratory
test data. Available from <[Link]
[Accessed: 28 January 2013].
Rosengren, K, Simmons, J, Maconochie, A P, and Sullivan, T, 2010. Geotechnical investigations for open pit mines –
250m and beyond. Proceedings, Bowen Basin Symposium 2010, Mackay, Bowen Basin Geologists Group and
Geological Society of Australia, pp 169 – 179.
Scott, A (ed.), 1996. Open Pit Blast Design: Analysis and Optimisation. Julius Kruttschnitt Mineral Research Centre,
The University of Queensland.
Sherwood Geotechnical and Research Services, 2011. ESFC-Based Trigger Levels for Monitoring Pit Wall Stability.
Australian Coal Association Research Program, Project C15033 final report, November.
Simmons, J V, 1995. Slope stability for open pit coal mining. Technical paper 9506, Australian Geomechanics,
December, pp 45 – 57.
Simmons, J V, and McManus, D A, 2004. Shear strength framework for design of dumped spoil slopes for open pit coal
mines. Proceedings, Advances in Geotechnical Engineering The Skempton Conference, London, March 2004,
R J Jardine, D M Potts, and K G Higgins eds, Thomas Telford Limited, Volume 2, pp 981-991.
Stacey, T R, 2006. Considerations of failure mechanisms associated with rock slope instability and consequences for
stability analysis. Journal of the South African Institute for Mining and Metallurgy, Volume 106, Number 7,
pp 485-493.
Ward, C R, Harrington, H J, Mallett, C W, Beeston, J W, 1995. Geology of Australian Coal Basins. Geological
Society of Australia, Coal Geology Group Special Publication.

100 Australian Geomechanics Vol 48 No 1 March 2013