Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 191 (2017) 744 – 749

Symposium of the International Society for Rock Mechanics

Static and Dynamic Analysis of Rock Slope – a Case Study

Jagadish Kundua,*, Kripamoy Sarkara, T.N. Singhb
a
Department of Applied Geology, Indian Institute of Technology (Indian School of Mines), Dhanbad 826004, India
b
Department of Earth Sciences, Indian Institute of Technology Bombay, Mumbai 400076, India

Abstract

Two dimensional plain strain distinct element method has been used to analyze a curved slope under static and dynamic loading
condition. Most vulnerable cross profile of the slope has been selected through kinematic analysis and slope mass rating specially
devised for curved rock slopes. Static loading has been induced in the model to simulate conditions of the prevailing rock mass.
Again dynamic loading has been performed according to the expected peak ground acceleration in the area in order to simulate
the slope mass condition during earthquake. According to kinematic result and slope mass rating, the slope is more vulnerable
towards 70° face direction. Static analysis results an unstable portion at the base of the slope. The whole slope is unstable under
dynamic loading and is expected to threat the region with the risk of rock fall as well as failure of slope.
© 2017
© 2017TheTheAuthors.
Authors. Published
Published by by Elsevier
Elsevier Ltd.Ltd.
This is an open access article under the CC BY-NC-ND license
Peer-review under responsibility of the organizing committee of EUROCK 2017.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of EUROCK 2017
Keywords: Slope stability; Kinematic; Distinct element modelling; Static; Dynamic

1. Introduction

Failures in rock slopes are controlled by both rock material and discontinuities. In highly jointed rock mass with
weak rock material, failure surface cuts through intact rock but in rock masses, with less number of joints and strong
rock material, failure is controlled mainly by joint planes [1]. With the presence of joints, rock mass responses better
to dynamic and static loading [2]. There are many numerical codes available to simulate dynamic loading in both
continuous and jointed rock mass. Among them continuum method is well-known among geotechnical fraternity
[3–5]. But despite of its astounding capability to handle any complex problem with ease and without any previous
assumptions, this method lacks its compatibility in representing discontinuous medium [6]. A discontinuous medium

* Corresponding author. Tel.: +91-326-223-5111.

E-mail address: jagadishkundu92@gmail.com

1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of EUROCK 2017
doi:10.1016/j.proeng.2017.05.240
 Jagadish Kundua et al. / Procedia Engineering 191 (2017) 744 – 749 745

is defined as an assemblage of discrete blocks separated by interfaces. Though continuum methods use special
techniques (e.g. joint elements in FEM), they are only applicable for less jointed rock mass where little displacement
is expected [7, 8]. To deal with inherent discontinuities in the rock mass and possible rupture in the intact rock,
discontinuum methods prove to be best till date. Discontinuum methods are based on dividing rigid and/or
deformable blocks by deformable interfaces. Among other discontinuum methods, distinct element method has an
advantage of adopting both rigid and deformable blocks along with deformable contacts [9]. This method being
explicit, directly solves the equations of motion by time marching scheme [10].
This paper represents an analysis of a jointed curved slope along the treacherous road of Himalaya hills.
The slope is located in the Jeori area along National Highway (NH) 22, Himachal Pradesh, India. To predict the rock
mass condition and probable modes of failure in the slope, SMR and kinematic analysis have been performed. As
the slope has a curved face, special techniques are adopted to the conventional kinematic analysis for better
prediction of potential failure zones. Based on the slope mass rating (SMR) and kinematic results, a vulnerable two
dimensional slope profile has been selected to simulate the static and dynamic response to loading of the slope under
its self-weight. The Universal Distinct Element Code (UDEC) has been used to simulate this two dimensional rock
slope problem.

2. Geological description

The rocks of the area comes under Jutogh group of rocks in the main central thrust zone (MCTZ) which belong to
the higher Himalayan crystalline zone of Himachal Pradesh. The area has experienced a local detachment in Jutogh
group of rocks as a result of ductile shearing and is named as Jeori dislocation. Common rock types in the area are
quartz mica schist, muscovite biotite schist, biotite gneiss, augen migmatite, garnet bearing quartz mica schist which
are highly truncated by joints and fractures [11]. The location under investigation consists of biotite gneiss and is
intersected by three joint sets. Geological map of the area (left) and the location photograph (right) is presented in
figure 1.

Fig. 1. Regional geological map of the area (left) (after Singh et al. [20]) and photograph of the location (right).
746 Jagadish Kundua et al. / Procedia Engineering 191 (2017) 744 – 749

3. Kinematic analysis and slope mass rating

Stereographic projection is an important and simple tool to analyze structural instability in jointed rock slopes. In
order to obtain kinematic feasibility of a wedge failure in a curved slope, a special technique has been devised by
[12]. The paper adopts a method from [13] to analyze wedge failure of curved slope in primary failure zone
extended the work for both primary and secondary failure. The studied slope has a constant inclination of 75° and
a curved slope face with direction varying from 60° to 95°. Details of the joint orientations in the slope is provided
in table 1 and the stereographic projection for the kinematic analysis is illustrated in figure 1. It shows two potential
wedge failures with trend of intersection line towards 70° and 4°. The wedge with intersection line towards 70°,
being in the primary failure zone, represents the critical wedge sliding on the intersection line (double plane failure).
The rock mass has been classified through SMR. Correction factors for SMR have been determined using
continuous functions suggested by [14]. The F1 parameter in SMR, in case of wedge failure depends on difference
between dip direction of slope and trend of line of intersection. For slope with curved face, F1 changes continuously
from one direction of the curvature to the other. So varying the F1 value, a graph has been obtained to represent
SMR in different direction of the slope. (Fig. 2b). As interpreted from the graph, the slope is very unstable at
the face directing towards 70° and increases its stability on either of its sides.

Table 1. Joint orientations of the studied slope.

Joint Dip(degree) Dip direction
J1 62 293
J2 61 79
J3 70 21

Fig. 2. (a) Kinematic analysis of the studied slope and (b) Slope mass rating with respect to slope face direction.

4. Distinct Element Modelling (DEM)

The distinct element method is most popular for solving discontinuous problems and is extensively being used
for stability analysis of rock slopes [15, 16]. A two dimensional cross sectional profile was selected at 70° of
the slope face direction on the basis of kinematic result and slope mass rating for numerical modelling purpose.
 Jagadish Kundua et al. / Procedia Engineering 191 (2017) 744 – 749 747

A 2D plain strain distinct element rock slope model was created using distinct element code. DEM is most suitable
for discontinuous media as it allows large displacements along the discontinuities and rotation of blocks [7]. For
small block sizes the method allows translational shear followed by rotation but only translational shear occurs in
case of large block sizes [17]. The study area contains three joint sets out of which J1 dips inward and where as J2
and J3 daylights on the face. Considering the wedge failure to slide on the intersection line, only plunge of the line is
considered as the interface instead of J 2 and J3. So an apparent inward dip of 55° for J1 and a plunge angle of 55°
daylighting on the slope surface have been incorporated in the model. The slope height is of 70 m with an overall
slope angle of 60° except at the base where the slope is almost vertical. The rock mass is represented by deformable
discrete blocks and the discontinuities as deformable boundary condition which separate the blocks. Deformable
blocks are created with finite difference mesh along with triangular zones to allow deformation in the matrix. Mohr
Coulomb material properties and Mohr coulomb slip criterion are used for the rock matrix and interfaces
respectively. The properties incorporated in the model are listed in the table 2. Rock mass cohesion, angle of friction
and elastic modulus were obtained from rock mass ratting value while the joint cohesion and friction angle were
obtained from direct shear test.

Table 2. Material and joint properties used in the model.

Rock mass Properties Value Joint properties value
Unit weight 2700 kg/m3 Normal stiffness 58 GPa/m
Bulk modulus 19 GPa Shear stiffness 28 GPa/m
Shear modulus 12 GPa cohesion 0.05 MPa
Cohesion 0.4 Mpa Friction angle 34°
Internal friction angle 42° dilation 0°

In order to simulate static loading, rollers are implemented at the lateral boundary restricting its motion along
x axis and pins at the bottom to fix its base. The joint rock slope model brought to an equilibrium state under its own
weight. Once the model reached equilibrium, dynamic loading was initiated with proper boundary condition.
The lateral boundary and the base of the model were applied viscous condition so that they retain their
non-reflecting properties. Dynamic excitation in the model was achieved by applying a sinusoidal shear stress at
the base and the applied shear stress history was calculated from the following equation:

W Q S u U uQ max (1)

and

Q max D max 2Sf (2)

Where, νs is the shear wave velocity, ρ is the rock mass density and νmax is the ground particle velocity, αmax is
the peak ground acceleration and 2πƒ = ω is the angular velocity.
Most dominant frequency range for the rock slopes due to earthquake vibration is 2–5 Hz [17]. However for
the studied region a frequency of 4 Hz is assumed. The considered area falls in seismic zone IV and therefore
maximum peak ground acceleration is assumed to be 0.25 g [18]. The shear wave velocity calculated from
the laboratory test found to be 2670 m/s. Hence the calculated shear stress is 0.57 m/s. To compensate the effect if
viscous boundary, shear stress has been applied double the calculated value i.e. 1.14 MPa.

5. Results and discussion

Results of the analysis have been documented in terms of displacement velocity and shear stress. Four monitoring
points as indicated in figure 3a were installed from top to bottom along the slope surface to record displacement
748 Jagadish Kundua et al. / Procedia Engineering 191 (2017) 744 – 749

history. Again two joints, J1 and J2 are selected two observe stress history at fixed monitoring points. Results of
the analysis and interpretation of obtained data are as follows.

5.1. Static loading

The model was brought to equilibrium under gravity loading after 1.256 sec (13451 cycles). Maximum velocity
vector and displacement vectors are 0.37 m/s and 0.25 m respectively. Maximum displacement is observed towards
the base of the slope (Fig. 3) where the inclination is almost vertical. Figure 4a shows, there is an increase in
x displacement from top to bottom of the slope face in the static region. In figure 4, the graph of shear stress with
time along J1 represents an increase in negative shear stress (downward movement of hanging block) but at the base,
it is positively increasing (upward movement of hanging wall). This case may be attributed to block rotation at
the slope base which caused some grid points to move in positive y direction. Figure 4c illustrates a graph of shear
stress along J2 which exhibits an increasing shear stress towards inner slope.

Fig. 3. Displacement vectors with Monitoring points (left) and failed blocks at the base of slope (right).

Fig. 4. (a) x-displacement vs. time; (b) shear stress vs. time along joint 1; (c) shear stress vs. time along joint 2.

5.2. Dynamic loading

The model was applied with a dynamic loading for 0.57 s after which the program terminated due to great contact
overlap. The loading resulted into maximum velocity vector and displacement vector to 0.53 m/s and 0.45 m
 Jagadish Kundua et al. / Procedia Engineering 191 (2017) 744 – 749 749

respectively. The rate of displacement increases rapidly in the x direction (fig. 4a.). This large displacement suggest
highly unstable rock mass at the slope base which may lead to collapse. There has also been a considerable
displacement towards the top of the slope. Block detachments are also observed throughout the slope surface which
may trigger huge rock fall in the area [19]. Along J1, shear stress is increasing towards negative direction similar to
static loading condition and block rotation at the base is noticeable as evident from monitoring point 8 (Fig. 4b).
For J2, shear stress behaves in a similar way as of static condition.

6. Conclusion

The study analyzed a rock slope condition through two dimensional static distinct element modelling and
possible effect of earthquake through dynamic loading. Through specially devised kinematic analysis and slope
mass rating, a maximum vulnerable cross profile was found to be towards 70° to north of the convex slope surface.
Prevailing rock mass condition demonstrates the slope is prone to small scale failures at the base and earthquake
vibration for small time may initiate rock fall in the region. Being a curved slope, rock mass condition is not uniform
across the selected cross profile. So, three dimensional model may help in further understanding of the conditions
and mechanism in greater detail.

References

[1] L. Siad, M. Megueddem, Stability analysis of jointed rock slope, Mech Res Commun 25 (1998) 661–70.
[2] S.G.G. Chen, J. Zhao, A study of UDEC modelling for blast wave propagation in jointed rock masses, Int J Rock Mech Min Sci 35 (1998)
93–99. doi:10.1016/S0148-9062(97)00322-7.
[3] J. Kundu, K. Sarkar, A.K. Singh, Integrating structural and numerical solutions for road cut slope stability analysis – A case study, India,
Rock Dyn. From Res. to Eng., CRC Press; 2016, pp. 457–462. doi:10.1201/b21378-69.
[4] P.K. Singh, A. Kainthola, S. Prasad, T.N. Singh, Protection measures on the failed cut-slope along the free expressway, Chembur, Mumbai,
India. J Geol Soc India 86 (2015) 687–695. doi:10.1007/s12594-015-0361-x.
[5] B. Mahanta, H.O. Singh, P.K. Singh, A. Kainthola, T.N. Singh, Stability analysis of potential failure zones along NH-305, India. Nat Hazards
83 (2016) 1341–1357. doi:10.1007/s11069-016-2396-8.
[6] L. Jing, A review of techniques, advances and outstanding issues in numerical modelling for rock mechanics and rock engineering, Int J Rock
Mech Min Sci 40 (2003) 283–353. doi:10.1016/S1365-1609(03)00013-3.
[7] A. Bobet, A. Fakhimi, S. Johnson, J. Morris, F. Tonon, M.R. Yeung, Numerical models in discontinuous media: Review of advances for rock
mechanics applications, J Geotech Geoenvironmental Eng 135 (2009) 1547–61. doi:10.1061/(ASCE)GT.1943-5606.0000133.
[8] R. Hammah, T. Yacoub, B. Corkum, The practical modelling of discontinuous rock masses with finite element analysis, 42nd US Rock Mech
Symp 2nd US – Canada Rock Mech Symp, 2008, pp. 56–63.
[9] P.A. Cundall, R.D. Hart, Numerical Modelling of Discontinua, Eng Comput 9 (1992) 101–13. doi:10.1108/eb023851.
[10] Itasca Manual, Dynamic analysis, User’s guide, UDEC version 5.0, Itasca Consulting Group Inc, 2011.
[11] K. Singh, Deformation history of the rocks around Sarahan Bushair, Himachal Pradesh, in: P.S. Saklani (Ed.), Struct. Geol. Himalaya, 1979,
pp. 163–182.
[12] J. Kundu, B. Mahanta, A. Tripathy, K. Sarkar, T.N. Singh, Stability Evaluation of Jointed Rock Slope with Curved Face, Indorock 2016,
2016, pp. 971–978.
[13] W. Yoon, U. Jeong, J. Kim, Kinematic analysis for sliding failure of multi-faced rock slopes. Eng Geol 67 (2002) 51–61.
[14] R. Tomás, J. Delgado, J.B. Serón, Modification of slope mass rating (SMR) by continuous functions. Int J Rock Mech Min Sci 44 (2007)
1062–1069. doi:10.1016/j.ijrmms.2007.02.004.
[15] Singh PK, Kainthola A, Singh TN. Earthquake induced rockfall along cut slopes - A case study. EUROCK 2015 64th Geomech Colloquium
Schubert Kluckner © ÖGG, 2015, pp. 1127–1132.
[16] V. Kveldsvik, A.M. Kaynia, F. Nadim, R. Bhasin, B. Nilsen, Einstein HH. Dynamic distinct-element analysis of the 800 m high Aknes rock
slope, Int J Rock Mech Min Sci 46 (2009) 686–698. doi:10.1016/j.ijrmms.2008.10.007.
[17] R. Bhasin, A.M. Kaynia, Static and dynamic simulation of a 700-m high rock slope in western Norway, Eng Geol 71 (2004) 213–226.
doi:10.1016/S0013-7952(03)00135-2.
[18] Bureau of Indian Standards, IS 1893 (Part 1): 2002 Indian Standard Criteria for Earthquake Resistant Design of Structures Part 1 General
Provisions and Buildings (Fifth Revision) 2002.
[19] D.K. Keefer, Landslides caused by earthquakes, Geol Soc Am Bull 95 (1984) 406–421.
[20] S. Singh, A.K. Jain, M.E. Barley, SHRIMP U-Pb c. 1860 Ma anorogenic magmatic signatures from the NW Himalaya: implications for
Palaeoproterozoic assembly of the Columbia Supercontinent, Geol. Soc. London, Spec. Publ. 323 (2009) 283–300.