15th ISRM Congress 2023 & 72nd Geomechanics Colloquium. Schubert & Kluckner (eds.

) © ÖGG

Rock slope design and residual risk management for Aktogay

Copper Mine, Kazakhstan

Neil Bar
KAZ Minerals, Almaty, Kazakhstan

Nurkhair Teleu
KAZ Minerals, Almaty, Kazakhstan

Herman Zlobin
KAZ Minerals, Aktogay, Kazakhstan

Philipp Mohr
KAZ Minerals, Aktogay, Kazakhstan

Alison McQuillan
Rocscience Inc, Phuket, Thailand

ABSTRACT: Aktogay is a large open pit copper mine in Kazakhstan. Several geotechnical
investigations have been completed from initial scoping and feasibility studies to continuous
improvement studies during mining. Rock slope design has involved several phases of stability
analyses for various stages and the life-of-mine open pit. With the progression of mining in the pit,
ground conditions at the mine are becoming more apparent and uncertainty is progressively reduced.
This paper briefly describes the slope stability modelling approaches used to understand and
communicate geotechnical risks and opportunities for integration with a complex mine plan and
sequence. A combination of 3D limit equilibrium and finite element analysis was incorporated into
slope stability assessments for future slope excavation stages. The models have been developed with
the intention of being digital twins to actual mine progression, with routine updating as engineering
geological features are validated and updated with new exposures and additional site investigations.

Keywords: 3D slope stability analysis, risk management, digital twin, mining.

1 INTRODUCTION

Aktogay copper mine is located in south-eastern Kazakhstan, approximately 470 km to the north-
east of Almaty, and 250 km west of the border with China. Mining operations commenced in 2015
and mine life is expected to be at least another 25 years.
The Aktogay deposit is located on the western part of the Central Asian Orogenic Belt (Li et al.
2018). It is confined to the eastern part of the Central Aktogai raft of volcanics and the enclosing
pluton. The raft is intruded by Early Carboniferous, porphyritic granodiorite which also cross-cuts
diorite and granodiorite of the Koldar pluton with the orebody striking west to north-west in an
elliptical shape (Zvezdov et al. 1993). Major geological structures present in current and future pit
slopes also follow this trend (Figure 1). Argillic-sericite alteration is focused on the core of the
porphyry system and decreases outward (LeRiche et al. 2022).
A comprehensive geotechnical model has been developed from lithology, alteration, major
structure and rock mass models as illustrated in Figure 1. These models were developed from a
comprehensive data set from several site investigation campaigns including 162 geotechnically-

-2880-
logged diamond cored boreholes, 130 boreholes with machine learning core photograph
classification and over 1700 geomechanical laboratory tests (LeRiche et al. 2022).

Figure 1. Aktogay Life of Mine Pit Shell and Geotechnical Model developed from Lithology, Structural and
Alteration models using Rock Mass Data and Machine Learning (after LeRiche et al. 2022).

Average annual precipitation at Aktogay is less than 200 mm with a maximum of 297 mm being
recorded in 1957 (Sagintayev et al. 2015). Snowfall at Aktogay usually lasts for approximately five
months between November and March with an average annual snowfall of 268 mm.
Groundwater management during initial mining stages only involves surface water management
(sumps and pumps to remove water from the pit floor). Horizontal drains are planned for deeper
mining stages. Pore pressures behind the slopes are well understood through a network of vibrating
wire piezometers. Higher pore pressures are observed during the snowmelt period (Spring).
During the initial mining stages at Aktogay, slope performance has been excellent with only a
few multi-bench instabilities. As shown in Figure 2, bench performance has also been very good as
a result of a focus on operational excellence for blasting and excavation.

-2881-
Slope stability analysis in earlier studies was completed using two dimensional (2D) cross-sectional
analysis. Limitations of 2D analysis for open pit mines include (Bar & Weekes, 2017; Bahsan &
Fakhriyyanti, 2018; Chakraborty & Goswami, 2021; Bar et al. 2022; McQuillan & Bar, 2023):

• Spatially or laterally varying geological conditions as seen in Figure 1.

• Spatially varying material strengths, including anisotropic behavior.
• Non-linear slope geometry and pit curvature as seen in Figure 2.
• Persistent geological structures, striking and intersecting slopes at various angles (i.e. not
representable using a 2D cross-section).
• High variability in 2D results within close spatial proximity to each other.
• Inability to understand potential failure sizes for operational risk management purposes.

This paper describes the development and use of three-dimensional (3D) limit equilibrium (LE) and
finite element (FE) models for slope design review and residual risk management at Aktogay.

Figure 2. Aktogay Copper Mine Stage 2 Pushback: Looking South. Left: September 2021 (Autumn). Right:
December 2022 (Winter).

2 THREE-DIMENSIONAL SLOPE STABILITY ANALYSIS

A 3D slope stability model was developed in 3D LE and 3D FE modelling software packages, Slide3
and RS3 of Rocscience Inc, respectively.
The model geometry was based on the geotechnical model in Figure 1 and isotropic material
properties from Table 1 were applied. The 3D model developed is 2.35 km long, 1.85 km wide, and
approximately 0.65 km deep, which allows for fast assessments of different slope design geometries.
A total of 45 known geological faults were included as weak layers or interfaces (Figure 3).
Pore pressures were simulated based on current mine dewatering practices comprising perimeter
pumping wells and surface water pumping. Monitored pore pressure data, from a network of over 40
vibrating wire piezometers, enabled the derivation Hu coefficients assigned to a phreatic surface that
considered both typical and elevated pore pressures during the snowmelt period during spring.

Table 1. Material Properties.

Geotechnical Domain UCS GSI mi D c' ϕ' Hu
Unit of Measure [MPa] [kPa] [°]
Diorite 50 - 98 42 - 57 23 - 24 0 - 0.7 - - 0.80 - 0.83
Granodiorite 66 - 90 46 - 59 19 - 20 0 - 0.7 - - 0.77 - 0.80
Volcanics 32 - 88 42 - 56 16 - 19 0 - 0.7 - - 0.79 - 0.86
SE Corner & F.D. 24 42 16 0 - 0.7 - - 0.80
Quartz Altered Rock 90 46 20 0 - 0.7 - - 0.77
Transition Rock 15 39 12 0 - 0.7 - - 0.80
Oxide Rock - - - - 100 35 0.80
Fault Damage Zone 24 42 16 0 - 0.7 - - 0.80
Discrete Fault - - - - 15-75 15-25 1.00

-2882-
 Figure 3. 3D Slope Stability Model for Aktogay Ultimate Pit. Semi-translucent red surfaces show discrete
geological faults from the structural model incorporated into the models as weak layers. Semi-translucent
blue surface represents the phreatic surface used to apply Hu coefficients and replicate monitored pore
pressures.

Slope stability analysis using 3D LE methods identified potential risk areas or design hazards on the
southern slopes of the ultimate pit design as illustrated in Figure 4 (based on inputs from Table 1).
The low FoS (Factor of Safety) areas identified were all associated with adversely oriented discrete
faults and formed one-, and two-sided wedges in the upper part of the slope. No significant risk areas
were identified elsewhere in the design.

Figure 4. 3D LE Slope Stability Model for Aktogay Ultimate Pit Southern Slopes. Left: FoS Map with
Potential Risk Areas and Design Hazards. Right: Looking South-East at Wedge ‘A’ formed by two adversely
oriented discrete faults intersecting in the upper slope.

3D FE was used to validate the results of 3D LE modelling. Wedge ‘A’ was identified and is shown
in Figure 5. Elastic analysis was run to determine the slope areas most susceptible to deformation
based on modelled geological structure and material properties.
Results were then compared to 3D LE to verify predicted failure mechanisms and identify
potential risk areas and design hazards. Contours of higher maximum shear strain were observed at

-2883-
the intersections of modeled faults. This confirms expected behavior in such ground conditions.
3D FE modeling identified the potential for ductile deformation on the central north slope (Figure
5), not identified with 3D LE methods, where this upward movement is less likely to be identified
using limit equilibrium methods. 3D LE and 3D FE modeling was completed in tandem to provide a
comprehensive review of potential risks and hazards using multiple analysis methods.

Figure 5. Elastic 3D FEM Slope Stability Model for Aktogay Ultimate Pit Southern Slopes. Left: Total
Displacement. Upper Right: Section cut displaying 4-noded tetrahedral finite element mesh applied to 3D FE
model. Lower Right: Cross-Section Looking West with Wedge ‘A’ on southern slope and ductile
deformation upward on faults on the central north slope.

3D LE and FE models were also analyzed using a series of different sensitivity scenarios to
understand input parameter sensitivity, particularly relating to discrete fault shear strengths and pore
pressures. It was identified that most of the potential risks or design hazards on the southern side of
the ultimate pit could be stabilized with slope depressurization (pore pressure reduction). By way of
example, Figure 6 demonstrates improved FoS and fewer risks, using FoS as hazard criteria,
(compared to Figure 4) with reduced pore pressures (Hu=0.60) that can be achieved using horizontal
drains and confirmed with additional vibrating wire piezometers on successive pushbacks.

Figure 6. 3D LE Slope Stability Model for Aktogay Ultimate Pit Southern Slopes with reduced pore pressure
(Hu=0.60) applied. Left: FoS Map with Potential Risk Areas and Design Hazards. Right: Looking South at
Wedge ‘A’ and a second low-FoS one-sided wedge.

Wedge ‘A’ from Figures 4 and 6 is formed by two adversely oriented discrete faults, which even if
depressurized, remains unstable with a potential failure volume of 0.46 million m3 (approximately

-2884-
1.2 million tonnes). Being positioned at the pit crest (top of the slope), this design hazard can be
easily managed through a minor design change to drastically unload, or completely remove the
wedge. Further site investigations are warranted to improve the characterization of the discrete faults
and confirm their orientation, continuity and character. Doing so has the potential to reduce additional
material movement needs.

3 KEY FINDINGS

The 3D slope stability models have been developed based on the geotechnical model incorporating
major geological structures. These 3D models provide insights into potential design hazards
associated with structurally-driven and complex failure mechanisms, both spatially and in magnitude
(size), which remains impractical to impossible using 2D analysis.
This enhanced understanding of residual risk has helped guide control measure development,
specific future site investigation objectives and sets targets for slope depressurization. It also enables
a cost-benefit analysis of various control measure options (e.g. unloading or shallower slope angles
versus horizontal drilling for depressurization).
3D slope stability models also have the ability to serve as digital twins. That is, they can be
updated quarterly or annually to include any updates to the geotechnical model such as the inclusion
or amendment of any geological faults, pore pressures, etc., updating of mine plans and forecasting
risk and recent slope performance against the models both qualitatively in the case of LE, and
quantitatively for FE.

ACKNOWLEDGEMENTS

The authors sincerely thank the support of KAZ Minerals mine operations, mine planning and
technical departments for their contributions supporting slope optimization and risk management
initiatives. Mr Johan Barnard is thanked for his support of the project and internal peer review. AMC
and SRK consultants are also acknowledged for their insightful previous work.

REFERENCES

Bahsan, E. & Fakhriyyanti, R. 2018. Comparison of 2D and 3D stability analyses for natural slope.
International Journal of Engineering & Technology 7(4.35), pp. 662-667.
Bar, N. & Weekes, G. 2017. Directional shear strength models in 2D and 3D limit equilibrium analyses to
assess the stability of anisotropic rock slopes in the Pilbara Region of Western Australia. Australian
Geomechanics 52(4), pp. 91-104.
Bar, N., Mojica, B., Cobián, J.C., Bautista, M., Hammah, R., McQuillan, A., Yacoub, T., Gaich, A., Pötsch,
M., Coli, N. & Preston, C. 2022. Cumba Slope Failure: A Technical Review. In: Proceedings of Slope
Stability 2022, Tucson USA.
Chakraborty, A. & Goswami, D. 2021. Three-dimensional slope stability analysis using stability charts.
International Journal of Geotechnical Engineering 15(6), pp. 624-649.
LeRiche, A., Tims, S., Saunders, E. & Mohr, P. 2022. Geotechnical Domaining for the Aktogay Porphyry
Deposit Supported by Machine Learning Techniques. In: Proceedings of Slope Stability 2022, Tucson USA.
Li, C., Shen, P., Pan, H., Cao, C. & Seitmuratova, E. 2018. Geology and ore-forming fluid evolution of the
Aktogai gian porphyry Cu deposit, Kazakhstan. Journal of Asian Earth Sciences 165, pp. 192-209.
McQuillan, A. & Bar, N. 2023. The necessity of 3D analysis for open-pit rock slope stability studies: Theory
and practice. Journal of the Southern African Institute of Mining & Metallurgy 123(2), pp. 63-70.
Sagintayev, Z., Yerikuly, Z., Zhaparkhanov, S., Panichkin, V., Miroshnichenko, O. & Mashtayeva, S. 2015.
Groundwater inflow modelling for a Kazakhstan copper ore deposit. J. Environ. Hydrology 23, pp. 1-13.
Zvezdeov, V.S., Migachev, I.F. & Girfanov, M.M. 1993. Porphyry copper deposits of the CIS and the models
of their formation. Ore Geology 7, pp. 511-549.

-2885-