Pitwall Stability Analysis-Case Studies*

M. Affam and K. E. Allen

Affam, M. and Allen, K. E., (2009), “Pitwall Stability Analysis-Case Studies”, Ghana Mining Journal, Vol. 11,
pp. 31 - 36.

Abstract
The importance of pitwall stability in design and economy of open pits has been a subject of considerable research in rock
mechanics for some time now. Imperceptible slow mass movement has been a secret agent in inducing major pitwall fail-
ure. The problem is location specific and is associated with stress development, shear strength, discontinuity orientations
and groundwater regimes. This paper therefore sets out to present pitwwall stability analysis in some selected pits within
the Birimian environment in Ghana. Several attitude measurements were taken particularly on the footwall and along the
slope face of the oxide, transition and fresh (sulphide) rock materials from some selected pits. The geologic data was sub-
jected to the Romana’s Slope Mass Rating (SMR) analytical procedure and the pitwalls monitored through installed
prisms.
The oxide/ transition zones gave an SMR quantitative value of 43.1 in the footwall and 44 within the hangingwall, indicat-
ing a partial stability at 50o. The fresh rock (sulphide) gave a value of 64 in the footwall showing stability at 55o. As usual,
the fresh rock was less problematic as indicated by the SMR assessment compared to the oxide and the transition materials.
Creep movement was observed to predominate, being nearly 250 mm/day within the oxidised zone but lesser in the transi-
tion and fresh rocks. Induced pitwall failure pattern within the Birimian was more of composite failure (ie.,planar and top-
pling) capable of destabilising several pits if unchecked.

1 Introduction the Plant North Pit (PNP) of Golden Star Re-

sources, Bogoso/Prestea Mine, Prestea, both
It is observed that in all slopes, there exists an in- within the Birimian environ, required stability as-
herent tendency to degrade to a more stable form, sessment of the slope walls. For instance, MP to-
ultimately towards horizontal. Instability is seen as tally failed before the ultimate pit bottom was
the tendency to move and failure as an actual mass reached. (Figs.1 and 2). PNP which was expected
movement (Whitlow, 1995). The detachment and to reach 4924 m reduced level (RL) began show-
movement of the earth material occur if the stress ing signs of possible failure at an initial stage of
imposed on the material is greater than the shear 5014 m (RL). The observed structural discontinui-
strength capable of holding it in place. ties are early warning signs capable of triggering
Mass movement is usually slow and largely imper- eventual macro failure as creep effects are more
ceptible to the naked eye. Creep is perhaps, the conducive in environment of numerous tension
slowest of all the types of mass movements, trig- cracks.
gering slope failures and precipitated by variation Due to the fact that any unprecedented failure of
in conditions such as changes in water content, the pitwalls could result in economic penalties
stress development, loading and surface instability such as loss of property and life, the need to look
(e.g. removal of vegetation). In the analysis of cut for remedial measures to maintain the pitwall is
and built slopes, it is realized that failure may also important. Workable stability analysis would not
be associated with geological structure of the host only provide measures to save the pit from even-
rocks itself. These changes may be observed dur- tual collapse but also save the company from in-
ing or immediately after excavation, or may be vestment loss and loss of livelihood for the numer-
imposed suddenly at any time (Whitlow, 1995). ous employees. It is against this background that
The slope of the excavation determines the amount this paper sets out to investigate areas of potential
of stress magnitude occurring within the material. failure induced by movement within the pit walls
Tension cracks and hummocky grounds are some by employing techniques that examine unstable
visible signs of relative movement. benches or walls and monitor their displacement in
The occurrence of visible micro cracks, jointing order to analysis the stability of walls for immedi-
and penetrative tension cracks in the Main Pit ate redress. The study is also intended to develop
(MP) of Central African Mine Ltd., Bibiani and remedial measures to ensure creep-free pitwalls.
* Manuscript received March 30, 2009
Revised version accepted December 3, 2009

31 GMJ Vol. 11, December 2009

1.1 Slope Stability in Open Pit Mining frictional bond between the particles and causes
In open pit mining, mineral deposit are mined from beds to slip past one another. The weight of the
the ground surface downward. Consequently, pit overlying water therefore creates excess pressure
slopes are formed as the ore is extracted. It is that drives the particle apart.
sometimes difficult to maintain stable vertical
slopes or pit walls of substantial height even in 2 Geological Setting
very hard and competent rocks. This is because the 2.1 Brief Local Geology
slope of the surface determines the amount of
stress that occurs on the material and therefore the The Birimian supergroup forms part of the Man
economics of the pit operations are to a large ex- Shield in the southern segment of the West Africa
tent linked to the slope geometry (Ross-Brown, Craton (Leube et al, 1990). Within the study area,
1972). The pit walls must be inclined at some an- the Birimian structural trend is referred to as the
gle to prevent failure of the rock mass. This angle central structural corridor or central fault zone,
is governed by the geo-mechanical conditions at a striking NE-SW.
specific mine and represent an upper bound to the The central structural corridor separates the
overall slope angle. Any miscalculation could trig- Birimian structural domain in the west from the
ger instability and eventual failure sometime later Tarkwaian structural domain in the east with both
(Fig.1 and Fig.2). The actual slope angles used in suites hosting an anastomosing network of faults
the mine depend upon: and imbricate faults, volcaniclastic rocks deposited
• Presence of ramps or haul roads in shallow marine basins separated by a sub-
• Possible blast damage parallel series of north-east trending volcanic belts
• Ore grade and economic constraints (Taylor et al; 1988). The transition between vol-
• canic belts and sedimentary basin is marked by
chemical sediments including cherts, manganese
and carbon–rich sediments.

2.2 Structural Setting

Regional scale deformation characterises mineral-
ised zones within the Birimian Gold belts of
Ghana. This is dominated by folding and thrusting
focusing at the boundary of the volcanic and sedi-
mentary basins. Shear zones are nearly parallel to
bedding and schistosity where developed within
the formation, lies in the axial plane direction.
Fig. 1 Unstable Pitwall showing Tension Cracks Zones of shearing and faulting are locally present
(in red) at Initial Excavation Stage in all the rocks, but are especially pronounced in
the soft metasedimentary rocks. Within the shear
zones the carbon bearing phyllites are converted
into lustrous black graphitic schist, which are often
crinkled or drag folded (Luebe et al, 1990). Isocli-
nal folding with nearly vertical axis occurs, but
detailed fold structure of the Lower Birimian is
difficult to unravel as marker horizons are lacking
and major folds are rarely seen.
Post-mineralised faulting is observed locally at
many places. Fault channels are rarely more than 5
cm wide, usually filled with black graphite gouge
composed of brecciated phyllite and schist. Small
faults containing graphitic–guoge which trend be-
tween east and southeast are occasionally seen in
Fig. 2 Burried Equipment Due to Total Pitwall the mines (Kesse, 1985). Minor structures such as
Failure in Bibiani Main Pit (MP) tension gash, quartz veinlet and jointing abound
near the veins.

Water also destabilises the cut slopes by creating 3 Stability of Pits

pressure in the pore spaces of the material me- 3.1 Factors Influencing Rock Slope Stability
dium. The infiltrated water into the slope face satu-
rates the weaker materials at depth, reduces the Slope instabilities are usually precipitated by a

32 GMJ Vol. 11, December 2009

variation in conditions such as changes in ground- sult of shear failure along an internal surface or
water seepage/drainage, geological discontinuities, weak plane owing to general decrease in effective
blast vibration or seismic shock that results in stress. They are classified according to Whitlow
ground particle acceleration. Such changes may (1995) as:
occur immediately after excavation, develop Plane failure: This failure occurs when discontinu-
slowly over a number of years or may be imposed ity striking approximately parallel to the slope face
suddenly at any time (Whitlow,1995). and dipping at lower angle daylight into the slope
The effect of acceleration on ground particles often face, enabling the material above the discontinuity
induced locally through blasting or seismic shock to slide.
can significantly alter the network of forces which Wedge failure: when two discontinuities intersect
act on the sliding block system. The stability of the in such a way that the wedge of material formed
jointed blocks is influenced principally by gravita- above the discontinuity could slide out in a direc-
tional stresses. Usually, the competence of the tion parallel to the line of intersection of the two
block weight acting parallel to sliding plane acts to discontinuities.
promote failure. Toppling failure: this failure involves rotation of
Where horizontal-directed acceleration pulses may columns or blocks of rock about a fixed base. It
be induced, the block will have a measure of hori- may involve linear movement of rock blocks along
zontal weight impart which will shift the net bedding planes or near to sloping surface.
weight vector component from the sorely vertical Rotational slips: this occurs characteristically in
direction. Should vertically-directed acceleration homogeneous soft rocks or cohesive soils; the
pulses be induced, they will only act to add to or movement taking place along a curved shear sur-
counteract the gravity weight component. It is face in such a way that the slipping mass slumps
therefore not uncommon to observe failure along a down near the top of the slope and bulges up near
newly created slip surface or along a pre-existing the toe.
one (Nordlund and Radberg, 1995).
3.3 Geotechnical Pitwall Mapping
3.2 Modes of Failure Mechanism Affecting Pitwall mapping was carried out to determine the
Open Pit dip amount, dip direction and type of discontinui-
Quite a number of movements take place as a re- ties that were present at the Main Pit (MP) and the

Fig. 3 Graph showing Relative Movement against Displacement Time

33 GMJ Vol. 11, December 2009

 GEOTECHNICAL MAPPING AT AGBL MAINPIT (WEST WALL) CONTOUR PLOT readings were picked at regular time intervals for
the purpose of monitoring vertical and lateral
movement (Fig. 3). The imperceptible slow move-
ment down the slope was dependent on water satu-
ration, stress loading and weak rock material. The
plastic strain response depicted secondary creep,
reflecting time dependent elastic deformation
when stress loading was assumed largely constant.
At a point tertiary creep set in, indicating stress/
strain condition in which accelerated sample strain
developed. The overall readings from movement
verse time curve indicated that rate of movement
could near 250 mm/day down the slope face, and
with no change in applied stress, it was capable of
triggering eminent creep failure.
Fig. 4: Weighted Contour Plot of the Main Pit
(MP) Discontinuities at the Central African 3.5 Stereographic Plot of Discontinuities
Gold Ltd., Bibiani.
The footwalls and hangingwalls of both the MP
and PNP were plotted stereographically using the
GEOTECHNICAL MAPPING AT AGBL MAINPIT (WEST WALL) MAJOR PLANS DIPS software. The weighted plot of the disconti-
nuities is as shown in (Fig.4). The major planes of
failure were identified as 70o/154o, 80o/046o,
80o/300o, 81o/215o, 85o/127o and 70o/120o (Fig.5).
Kinematic slope analysis carried out with the DIPS
software indicated that potential for localised pla-
nar and toppling failure existed along the footwall.
3.6 Slope Configuration
The slope configuration measured along one of the
deepest sections of the pit is as shown in Table 1.
As can be seen, the average batter and berm are X
and Y respectively. However, the pits were ex-
pected to be mined using 65o batter, 3 m berms and
12m bench height.
Fig. 5: Stereographic Plot of Discontinuities, Pit
Slope and Friction Angle in MP Table 1 A Typical Slope Configuration
Overall Batter Berm Bench Inter Maximum Material
o o
Angle ( ) Angle( ) Width(m) Height(m) Ramp slope Type
o
Angle( ) Height(m)
Plant North Pit (PNP). The discontinuities in- 38 50 6.5 12 35.8 60 Oxide/Transition
45 55 4.5 12 42.9 60 Oxide/Transition
cluded joints, fractures and faults. All dip direc- 50 60 4 12 47.8 60 Oxide/Transition
tions were taken with respect to the true north. The 50 60 3.5 12 48.8 100 Fresh
55 65 3.2 12 53.8 100 Fresh
mapping was done on both the hanging and foot-
wall. The values were then used to plot the 3.7 Slope Mass Rating (SMR)
stereonet and predict the slope stability behaviour
using the Romana Slope Mass Rating (SMR). The The Slope Mass Rating (SMR) is an empirical
potential mode of failure of the slope was deduced slope stability analytical method that was intro-
from stereographic analysis using measurement of duced by Romana (1985, 1993), as a logical follow
dip and dip direction of the geological structures up to the Bieniawski Rock Mass Rating (RMR) or
encountered on the pit walls loaded into the DIPS geomechanical classification. RMR widely used in
software. The variation of the attitude of the struc- underground mining, is not the best classification
tures was eminent. The mapping revealed a signifi- system for slopes due to the fact that the stability
cant change in orientation of geological structures of most slopes is structurally controlled. In the
necessitating a design modification. RMR system, thorough emphasis of the orientation
of geological structures appears lacking.
3.4 Observation of Pitwall Movement SMR introduces four adjustment factors to the
To ensure reliable monitoring of pitwalls stability, RMR to make up for these deficiencies. The ad-
systematic surveys of the movements of regions justment factors added depends upon joint-slope
within the walls were organised. About 77 prisms relationships and a factor depending on methods of
were installed along the berm and slope faces. The excavation. The SMR is computed using the rela-

34 GMJ Vol. 11, December 2009

 tionship be- Table 2 Results from SMR Calculations
low
Parameter Value Rating
(Romana, 1985):
Intact rock strength 11.8 Mpa 2
RQD 75% 17
where: Joint spacing 105mm 8
F1 depends on the angle between the strikes of the
joints and the slope face. Its value ranges from
Joint condition 1-5mm cont. 10
0.15 to 1.0. This range was established but after- Groundwater Wet 7
ward was found to match the relationship F1 =(1- RMR 44
sinA) where A is the angle between the strike of Joint Dip Joint Dip Slope Face Slope Face F1 F2 F3
the slope face and the joint. Dip Dip
Direction, ?j ? j Direction ?s ? s
F2 refers to the joint dip angle in the planar mode
of failure. Its value ranges from 0.15 (for joints 345o 60o 302 o 50o 0.15 1.00 -6
dipping less than 20o) to 1.0 (for joints dipping
more than 45o). For toppling mode of failure, F2 • High quality and timely drainage, including
remains 1.0. deep drainage (i.e., boreholes) of the rock
making up the slopes, drainage of the sur-
F3 reflects the relationship between slope face and
face in the vicinity of the mine and surface
joint dips. In the planar mode of failure, F3 again
of the mine walls.
refers to the probability that the joints ‘daylight’
into the slope face. For the toppling mode of fail- • Critical study of geologic and hydrologic
ure, due to it nature, unfavourable condition cannot conditions of the pitwall face, nature and
happen. types of discontinuities with the view of
sealing off active ones, through timely cut-
The adjustment factor for the method of excava- backs.
tion F4 has been fixed empirically. No special fac-
tors have been fixed for the wedge failure. The • Avoidance of overloading unstable pitwall
classification is applied for each of the joint sys- surfaces and maintenance of dry mine wall
tems. The SMR has a total range of 0-100. The faces
adjustment rating for joint is the product of the • Precise execution of designed plan, particu-
three factors F1, F2 and F3 Table 3 Typical SMR for Selected Slopes
Prospect Designation RMR SMR Stability
4 Results and Discussions Assessment
Footwall O/T zone 44.0 43.1 Partially stable @50o
4.1 Result of SMR Determination Plant
North Main Pit Footwall F/R zone 60.3 64.0 Stable @55o

The SMR calculation is based on the criteria dis- Hangingwall O/T zone 44.0 44.0 Partially stable @50o

cussed in Section 3.7. The SMR of the hanging- Hangingwall F/R zone 61.0 61.0 Stable @55o
wall and footwall for the oxide/transition zones Footwall O/T zone 44 44 Partially stable @50o
were determined. The calculation as outlined be-
Footwall F/R zone 64 58 Stable @55o
low in Table 2 was carried out using stability rat- North Shaft Pit
Hangingwall O/T zone 44 39.8 Partially stable @45o

Hangingwall F/R zone 61 43.5 Partially stable @50o

**O/T = Oxide/Transition
F/R = Fresh Rock
ing from Romana’s assessment chart (after Roman,
1993). The overall rating for some selected slopes larly in terms of projected slope angles,
within the Pits is as shown in Table 3. bench geometries, etc.
Method of excavation: Mechanical (F4) = 0
Stability assessment at angle 50o is partially stable 4.3 Discussions
(Fig.4). The potential modes of failure of the pitwalls were
deduced from the stereographic analysis using
4.2 Measures for Pitwall Failure Prevention measurements of dip and dip directions of the geo-
Pitwall movement just like other mass movements, logical structures encountered in the pit. The varia-
could be imperceptibly slow. Being a natural phe- tion of the attitude of beds with depth was quite
nomenon, it is increasingly difficult to stop it once common. This was buttressed with sensitivity
it has started. Preventive measures must therefore analysis of the pitwall stability to the method of
take the lead in control as later remedial action has excavation employing the Romana’s Slope Mass
often proved to be ineffective. Among the preven- Rating. This approach was highly relevant because
tive measures are: it highlighted the impacts of geological disconti-
nuities which dominate failure in pitwall.

35 GMJ Vol. 11, December 2009

The analysis indicated that within the pits, the sta- Nordlund, E. and Radberg G., (1995), Berg-
bility of most of the walls were partially stable at mekanik. Kurskompendium, Tekniska Hogsko-
an average overall angle of 50o, triggering possible lani, Lulea (Sweden), 191pp.
planar or toppling failure. The assessment revealed Romana, M., (1985), New Adjustment Ratings for
that stability could be improved by means of pit Application of Bieniawski Classification to
dewatering and meticulous smooth wall blasting. slopes. Int. Symposium on the role of Rock Me-
Localised planar or toppling failure was a threat, as chanics (ISRM), Zacatecas, pp. 49-53.
sustained wet condition in the pit slopes reduced Romana, M., (1993), A Geomechemical Classifi-
effective stress. Induced penetrative tension cracks cation for Slope: Slope Mass Rating. Pegmon.
then empowered destabilising force. It is impera- Vol.3, pp. 575-600.
tive therefore that the pitwall face be maintained at Rose-Brown, D. M., (1972), Design Considera-
dump-to-dry state. tions for Excavated Mine Slope in HardRock.
Research report No. 21, Department of Civ il
The observed value for the creep movement near- Engineering, Geology, Miningand Mineral
ing 250 mm/day along the Main Pit (MP) pitwall Technology. London Imperial col lege and
was quite alarming. Careful study of geologic and Technology, 21pp.
hydrologic conditions, as well as structural charac- Taylor, P.N., Moorbath, S., Leube, A. and Hirdes
teristics suggested a need for an immediate creep W., (1988), Geochronology and Crustal Evolu-
arrest since it is easier to prevent the creep move- tion of Early Proterozoic Granite-greenstone
ment than to control it once it has started. Terrains in Ghana/West Africa (Abstr.) Inter-
national Special Conference on Geology of
5 Conclusions Ghana with special Emphsis on Gold, Comm.,
75th Anniversary of Ghana Geological Survey
Induced pitwall failure mode, especially within the Department, Accra, pp. 43-45.
Birimian is more of composite failure (i.e., planar Whitlow, R., (1995), Basic Soil Mechanics. 3rd
and/or toppling) and it is critical in oxide/ transi- Edition. Longman, Malaysia, pp. 346-347.
tion zones. Romana’s Slope Mass Rating (SMR) is
effective in early theoretical assessment of the
pitwall stability. Relative movements nearing 250
mm/day threatens stability of pitwalls and damp-to Authors
-dry state pitwall face could be achieved through
effective pit dewatering. Deep drainage, seal-off of M. Affam is a lecturer in Geotechnical Engi-
neering at University of Mines and Technol-
active discontinuities and timely cutbacks are ogy, Tarkwa, Ghana. He holds an MSc de-
among the most effective measures in creep pre- gree in Mineral Exploration and BSc in Geo-
vention that threatens pitwall failure. Indeed, pre- logical Engineering from Kwame Nkrumah
ventive measures must take the lead in stabilising University of Science and Technology, Ku-
masi. He is currently a PhD student research-
pitwall as later remedial action has often proved to ing in rockburst potential in Ashanti deep
be ineffective. mine. He has since graduation worked for a number of mining
and exploration companies in Ghana. He is a member of a team
of experts that provide geotechnical consultancy services to
mining and allied companies. He is a member of GhIE, GhIG
Acknowledgments and CIM/ICM.
The author acknowledges the contribution made
by all those who in different ways provided inputs
in the acquisition of data for this paper. Special K. J. Allen is a graduate from University of
Mines and Technology, Tarkwa. He holds
mention is made of Isaac Korankye-Boateng a MSc degree in Geological Engineering and
former student of University of Mines and Tech- Diploma in Geomatic Engineering. He has
nology, Tarkwa. vast experience in mining covering several
years and currently works as Mining Superin-
tendent at Golden Star Resources, Bogoso/

References
Kesse, G.O., (1985), The Mineral and Rock Re-
sources of Ghana. A A Balkema, Rotterdam
Netherlands, pp. 13-17.
Leube, A., Hirdes, W., Mauer, R and Kese G.O.,
(1990), The Early Proterozoic Birimian Super-
group of Ghana and some Aspects of its Asso-
ciated Gold Mineralization Precambrian
Res.,46, pp. 139-165.

36 GMJ Vol. 11, December 2009