See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268278633

The effect of a weak foundations rock joint on the stability of a concrete gravity
dam

Conference Paper · October 2014

DOI: 10.13140/2.1.4285.4086

CITATIONS READS
0 1,717

3 authors:

Younis Alshkane Alec M Marshall

University of Sulaimani University of Nottingham
50 PUBLICATIONS   107 CITATIONS    149 PUBLICATIONS   1,622 CITATIONS   

SEE PROFILE SEE PROFILE

Rod Stace

68 PUBLICATIONS   280 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Long-term stability of abandoned mines View project

Tunnelling beneath piled structures in urban areas (TUBEURB) - MARIE SKŁODOWSKA-CURIE ACTIONS Individual Fellowship (IF) View project

All content following this page was uploaded by Younis Alshkane on 15 November 2014.

The user has requested enhancement of the downloaded file.

DFNE 2014 - 207

The effect of a weak foundations rock joint on the stability of a concrete

gravity dam
* 1
Alshkane , Y. M.1 , Marshall, A.M. , & Stace, L. R.1
1
Department of Civil Engineering, University of Nottingham, University Park, Nottingham NG7 2RD, UK
*
Department of Civil Engineering, University of Sulaimani, Sulaimani, Kurdistan Region, Iraq

ABSTRACT: The stability of a dam on a jointed rock foundation might be endangered by weak joints that may be present in the fracture
network of the bed rock. A review of the literature shows that there are few studies of the effect of a weak joint in the foundation rock on
the stability of dams. This research uses the finite difference numerical modelling software FLAC to model a gravity dam, the foundation
rock, and the influence of a weak joint (using interfaces), which may exist in the upstream or downstream direction, on the stability of the
dam. To do this, a conceptual model has been developed and one representative joint of a discrete fracture network was examined. Dif-
ferent locations of a weak joint with varying dip angles were studied in order to investigate the critical configuration that has the most
significant effect on the dam’s stability. The paper shows that both the location and orientation of the weak joint have an important effect
on the predicted deformation of the dam.

1. INTRODUCTION
There are many issues that affect the stability of concrete
dams. One of these issues is the effect of weak layers in the
foundation rock. Rock masses under dams often contain
weak elements such as bedding planes, fractures, joints,
faults and seams of weaker materials. They may be weath-
ered by the flow of water which can lead to a reduction of
strength, and in turn, may affect the stability of the dam.
Generally, most of the incidents of instability have oc-
curred in the foundations of dams [1, 2]. An example of
such an incident is the failure of the Malpasset dam in
France (Figure 1). Although the Malpasset dam is an arch
dam, the failure mechanism of the foundation is directly
relevant to this study.
Fig.1 Weak layer ( Fault ) under the Malpasset Dam [3]
A review of the literature shows that there are few studies
of the effect of a weak joint in the foundation rock on the
stability of a dam. Using the finite element technique,
Varshney [4] studied the effect of seams of weaker rock
and faults on the stress distribution in a concrete gravity
dam. He utilized a vertical fault at the heel, toe, and the
center of the dam base. He showed that the location and
orientation of the weak layer affect the stress arrangement
in a dam to a great extent. However, the effect of variations
in the dip angle and fault location (both upstream and
downstream from the dam) was not considered in his pa-
per. Singh and Varshney [5] utilized the finite element
method to study the influence of faults with dip angles of
0, 45, and -45 degrees on the behaviour of a concrete
gravity dam. In addition, they used a fault at the heel, mid-
base, and toe so as to determine which joint location and
dip angle have the greatest effect on the dam behaviour.
They concluded that faults located at the toe should be
avoided especially when dipping upstream at a 45 degree
angle. However, only one fault inclination with a dip angle
of 45 degrees was studied in both upstream and down-
stream directions. Snell and Knight [6] used a generalized
theoretical model to study the influence of faults dipping
upstream within ranges from 0-25 degrees of dip under two
types of dams: rockfill and a typical concrete gravity dam.
They used the limit equilibrium method to calculate the
factor of safety against sliding along the stratum or weak
layer. They showed that the possibility of failure due to
sliding along these strata should not be ignored.
Alshkane et al. [7] applied both a continuum model
(FLAC) and a discontinuum model (UDEC) to evaluate
which code gives the most realistic prediction on the stabil-
ity of a concrete dam on a rock mass foundation. It was
concluded that both codes can give similar results and this
similarity was found to depend on joint strength and con-
figuration. As the joint strength is reduced, UDEC was
found to provide a better representation of the behavior of
a jointed system and arguably a more realistic prediction of
the stress distribution under the dam. Also the authors
showed that the failure mechanism for a discontinuum
model is sliding along a joint dipped 15 degree in the up-
stream direction. As a result of this previous work this
study investigated the influence of only one joint within a
fracture network of the dam’s foundation, on its stability.
A weak joint within a fracture discrete network on either
side of the concrete gravity dam may affect the dam’s sta-
bility; therefore this study has focused on finding out the
effect of a weak layer upstream and downstream on dam’s
stability. The term joint will be used to represent all types
of weak layers.
Numerous techniques have been developed to study the
stability of dams, such as the limit equilibrium method,
physical modelling, and numerical modelling. The limit
equilibrium method has proven to be very useful for design
but the level of complexity that can be considered in the
analysis is limited. Physical modelling is an effective re-
search tool, however, limitations including model size,
scaling, cost, and time make it difficult to conduct a thor-
ough parametric study of dam stability. Numerical model-
ling is arguably the most popular and powerful method
used by researchers and designers because of its capability
to accurately and economically evaluates the stability of
dams, including complex geometry and material behav-
iour. There are many numerical methods which have been
developed to analyze different types of structures, such as
the Finite Element and Finite Difference Methods (FEM,
FDM), the Boundary Elements method (BEM), and the
Discrete Element Method (DEM). This paper is based on
results obtained using the two-dimensional FDM code
FLAC to study the effect of a joint in a sandstone rock
foundation on the stability of a concrete gravity dam.
The dam and foundation rock were analyzed in a plane
strain condition for simplicity. The seepage, which was
treated as steady state according to Darcy’s law, was cou-
pled with stresses in order to obtain a realistic analysis.
The behaviour of the intact material (concrete dam and
sandstone) and joints (weak layer and dam-foundation in-
terfaces) were considered as elastic perfectly plastic fol-
lowing the Mohr-Coulomb yield criteria.
This paper is organized into five main sections following
this introduction. Section 2 covers a conventional design of
the gravity concrete dam with a reasonable factor of safety
against sliding and overturning. In sections 3 and 4, the
method of analysis and results are presented. The paper
finishes with a review of the main conclusions of the work.
2. DESIGN OF THE BASE MODEL
Current dam design methods typically rely heavily on the
conventional limit equilibrium analysis method. In this
study a theoretical model was used which was developed
and designed with a reasonable factor of safety against
sliding and overturning by using the conventional method.
Generally a concrete gravity dam is constructed in three
stages: preparation of foundation, construction of the
dam’s body, and filling of the reservoir with water. After
the reservoir has been filled the water induces a lateral
pressure on the upstream face of the dam that should be
resisted by the shear strength of the rock in the foundation
along the joint or weak layer to prohibit sliding failure [8].
The basic stability requirements for a gravity dam for all
conditions of loading are [9]:
 that it be safe against overturning at any horizontal
plane within the structure, at the base, or at a plane
below the base,
 that it be safe against sliding on any horizontal or
near-horizontal plane within the structure, at the
base, or in any rock in the foundation, and
 that the allowable unit stresses in the concrete or
in the foundation material shall not be exceeded.
In the analysis, the dam was assumed to have a height of
50m and a base length of 45m (Figure 2). The height of the
water in the reservoir was equal to the dam height. The
dimensions of the dam were established using the conven-
tional method such that the resultant vertical force beneath
the dam was within the middle third of the base in order to
ensure no tensile stresses existed beneath the dam [10].
There are two methods used for assessment of sliding sta-
bility. The first method is given by Equation 1, which
gives the shear friction factor (SFF) at the contact between
the dam and the foundation. The SFF should be greater
than 3 according to the American Engineering Army Corps
design criteria [11]. The second method for determining a
sliding factor ( ) can be calculated by dividing the sum of
the horizontal forces due to reservoir load (V) by the sum
of the vertical forces (N), as illustrated in Equation 2 [12].
The value of should be smaller than 0.75 for usual load-
ing.
( - )
(1)
(2)
-
where = cohesion, = contact area between the dam and In order to simulate the weak joint, FLAC provides inter-
the foundation, = sum of normal forces, = sum of faces that are characterized by Coulomb sliding and/or ten-
uplift forces, = angle of internal friction, = sum of sile separation. Interfaces have the properties of friction,
horizontal forces. cohesion, dilation, normal and shear stiffness, and tensile
strength. Although there is no restriction on the number of
In reality, there is the potential for many loads to be acting
interfaces or the complexity of their intersections, it is gen-
on a dam. For the conceptual model used here, only the
erally not reasonable to model more than a few simple in-
dead load and reservoir load are taken into account. For
terfaces with FLAC because it is awkward to specify com-
real design of dams, all types of loads should be consid-
plicated interface geometry.
ered, such as silt, ice, thermal effects, and dynamic load-
ing. Full details about the potential loads that can affect 3.1 Simulation of the base model
dams can be found in a variety of texts, for example The simulation of the construction of the gravity concrete
Thomas [10]. dam was achieved in three stages. Firstly, the foundation
rock was constructed; in this stage the in-situ stresses were
The stability of the dam was computed using Equations 1
and 2, the results of which are shown in Table 1. A linear initialized and the displacements (after reaching equilibri-
um) were reset to zero. Secondly, the concrete dam was
variation of hydrostatic water pressure was assumed under
installed; in this stage the stresses and displacements under
the dam between the heel and toe.
the dam were recorded. Also the displacement of the dam
crest was calculated. Thirdly, the water load due to the res-
Table 1 Stability assessment results ervoir was applied and the same recordings from stage two
were made again.
Stability Type Ratio
The shear friction factor, SFF (Eq. 1) 3.410 3.2 Geometry and Properties
Effective sliding factor, Fs (Eq. 2) 0.625 Figure 2 shows the base model geometry which was used
in this research. The foundation rock dimensions were
500m in width and 150m in depth. The boundary condi-
The resultant force was computed and it was found to be
tions were as follows; the sides were restrained in the hori-
within the middle third of the dam’s base (eccentricity of
zontal direction and the base was fixed in both directions.
resultant force from the centreline of the dam base,
Tables 2 and 3 show the properties for the model compo-
e = 3.177m). The induced distribution of stresses due to
nents. These properties are as follows: bulk stiffness (K),
dam weight and reservoir load was calculated according to
shear stiffness (G), joint normal stiffness (kn), joint shear
USACE [11]. The minimum effective vertical stress at the
stiffness (ks), cohesion (C), tensile strength ( ), and coef-
heel was found to be 0.256 MPa; the maximum value at
ficient of friction ( .
the toe of the dam was 0.633 MPa after applying the reser-
voir load. Also, the stability of the dam against overturning
(computed as the ratio of resistant moment to the disturb- Table 2 Material properties
ing moment about the toe of the dam) was found to be 1.7.
Density
K G c
Material (
MPa MPa MPa Degree MPa
3. NUMERICAL METHODOLOGY kg/m3
Concrete 2400 12200 10300 - - -
Rock 2416 26800 7000 0.6 35 0.3
FLAC was specially developed for geotechnical and min-
ing engineering mechanics computation [13]. The formula-
tion is based on the treatment of the problem domain as a Table 3 Material properties of joints
continuum that responds in accordance to one or more con- kn ks c
stitutive relationships that can be selected. FLAC can be MPa/m MPa/m MPa degrees MPa
used to solve complex problems in rock mechanics, in par- Rock-
ticular dam construction. Three fundamental stages should dam con- 3000 1000 0.6 40 0.3
be specified during the set-up of a numerical model: (i) tact
creation of a finite difference grid and boundary condi- Joint
tions, (ii) initiation of initial stress conditions, and (iii) se- Case1 2700 900 0.2 25 0.2
Case2 2700 900 0 15 0
lection of constitutive model and material properties. Once
these components are described in the model, the initial
equilibrium state can be computed.
 Fig.2 Model geometry

3.3 Simulation of the foundation with weak layer base model after filling the reservoir. It can be seen that the
To simulate the effect of a weak joint within the founda- stress distributions are not as linear as the conventional
tion on the stability of the gravity dam, several joint loca- method results. However, the average values for normal
tions selected in the upstream and downstream directions and shear stresses are similar.
with different dip angles were prepared, as shown in Fig-
ure 2. The models generated were named according to the
location and dip angle of the joint, for example a joint lo-
cated in downstream direction at =5m from the toe of
the dam with a dip angle of 30 degrees was named as
DJ30at5 and a joint located at upstream direction at
=4.5m from the heel with a dip angle of 60 was named
as UJ60at4.5. Each case was modelled in two conditions;
Case 1 considered the ‘intact’ joint strengths whilst Case 2
used the ‘reduced’ strengths (setting joint cohesion and
tensile strength to zero while reducing the joint friction
angle to 15 degrees).
For a weak layer located in the upstream direction, three
joint locations were chosen as follows: at the heel of the
dam, at 4.5m from the heel, and at 35m from the heel. For
each location, six dip angles were checked: 15, 30, 45, 60,
75, and 90 degrees. Eighteen models in all were generated Fig. 3 Stress distributions under the dam (base model)
for the upstream side. For a weak layer located in the
downstream direction, four locations were chosen: at the 4.2 Displacements
toe of the dam and at 5, 20, and 50 m from the toe, respec- To illustrate the effect of a weak layer on the stability of
tively. For each location the same six dip angles used in dams, the displacements were measured along the contact
the upstream cases were checked, with the exception that between the rock and foundation. Also the horizontal dis-
for joints located at the toe, only dip angles of 15, 30 and placements were measured at the top (crest) of the dam on
90 degrees were checked. Twenty one models were gener- the upstream face.
ated for the downstream side. 4.2.1 Downstream scenario
The horizontal displacement at the crest of the dam versus
4 RESULTS the dip angle of the joint modelled is presented in Figure 4.
4.1 Stresses under the dam for base model The horizontal displacement for the base model is shown
Normal and shear stresses were calculated under the dam for comparison. It can be seen that the joint location and its
for the base model. Figure 3 shows the stress distribution at dip angle have an important effect on the dam’s stability.
the contact between the dam and the foundation for the The results indicate that the dam is not stable when the dip
angle of the weak layer (Case 2) is 15 degrees and is locat-
ed at a distance of of 0.1 and 0.4 (the circled ‘Failed
model’ results in Figure 4 represent models that did not
reach numerical equilibrium, indicating dam instability).
To confirm this observation, the vertical displacement at
the dam-foundation contact was computed for the models
located in the downstream direction, as shown in Figures
5-8. These figures again show that the critical cases (where
the model failed to reach numerical equilibrium) are where
dip angle is 15 degrees and the location of the joint is be-
tween 0.4 and 0.6 times the height of the dam measured
from the toe ( model was also run for a
joint dip of 15 degrees which did reach a stable state simi-
lar to the model). It can be seen that when the
distance between the toe and the joint location is more than Fig.6 Vertical displacement for joint located at 5m ( )
half the dam’s height, the effect of reducing the shear from the toe of the dam – downstream scenario
strength (Case 2) on the dam deformation is not serious.
Generally, when the strength of the joint is reduced (Case
2), the horizontal displacement at the crest of the dam de-
creases with an increase in dip angle up to 45 degrees and
then the displacement increases when the dip of the joint
becomes larger than 45 degrees.

Fig.7 Vertical displacement for joint located at 20m (

) from the toe of the dam – downstream scenario

Fig.4 Horizontal displacements at the crest of the dam with vari-

ation of joint angle - downstream scenario

Fig.8 Vertical displacement for joint located at 50m ( )

from the toe of the dam – downstream scenario

Fig.5 Joints located at toe ( ) of the dam – downstream

scenario
4.2.2 Upstream scenario
Figure 9 shows the results for crest horizontal displace-
ments versus the joint dip angle for the joint positions con-
sidered. Generally the horizontal displacement tends to
reduce with increasing joint dip angle up to an angle of
about 60 degrees. It can be seen that the case where joint
dip angle is 15 degrees gives the maximum displacement,
especially when the shear strength of the joint is reduced
(Case 2) whereas a joint with a dip angle of 60 has the
minimum displacement when the joint’s strength is re-
duced.
Figures 10-12 show vertical displacements beneath the
dam for the joints located upstream of the dam. Overall,
the models have similar trends to the downstream scenario
and the maximum displacement occurs when the joint dip
angle is between 30o and 60o for the reduced joint strength
(Case 2). It can be seen that the joint dip has a similar ef- Fig.10 Vertical displacements for joint at the heel ( )–
fect on vertical displacement irrespective of joint location upstream scenario
in the upstream direction. This is due to the fact that the
resultant force induced by the reservoir and dam loading
acts on the downstream side of the joint and causes slip-
ping along the joint, without causing a global instability
Furthermore, the location of the maximum vertical dis-
placement for Case 2 is approximately at one-quarter of the
dam base measured from the heel, whereas the maximum
vertical displacement is at the mid-base of the dam for
Case 1 and the shape of deflection line is similar to that of
the base model.

Fig.11 Vertical displacement for joint at 4.5m ( )

from the heel of dam – upstream scenario

Fig.9 Horizontal displacements at crest of dam with joint angle –

upstream scenario

Fig.12 Vertical displacements for joint at 35m ( )

from the heel of dam – upstream scenario
4.3 Stability assessment and Discussion location (δ) and the toe of dam was less than half of dam’s
The assessment of a dam’s stability is not an easy task in height.
engineering. Bieniawski and Orr [14] proposed Table 4 to
Figure 13 shows the velocity vectors for model DJ15at5 at
assess the stability of dams according to the dip angle of
an arbitrary stage of the numerical simulation (the model
the main joint set. In addition, they used this to classify
did not reach equilibrium, indicating instability) and after
rock masses under dams according to the Rock Mass Rat-
significant movements had occurred. The vectors illustrate
ing (RMR) system, which is a geomechanical classification
the failure mechanism of the dam (sliding along the weak
system proposed originally by Bieniawski [15]. Table 4
layer). According to Table 5, the assessment of model
was based on experience and on the stress distribution in
DJ15at5 is fair whereas according to Table 4 and the nu-
foundation rock masses [16] as well as on an assumption
merical models presented in this paper, it is unfavourable.
that both the arch and the gravity effects were present in a
Therefore, this paper suggests that Table 5 should also be
dam structure. This table should be used with caution, as
used with caution and should be checked by using numeri-
indicated within the original text[14]: “this table is tenta-
cal modelling or other appropriate means. These results
tive only and represents a simplification of a problem
illustrate the need for a more rigorous rock mass classifica-
which is justified for rock mass classification purposes on-
tion system for dam construction.
ly.” Since publication of this paper in 1976, there has been
very little work published to prove the reliability of the Figure 14 shows the shear friction factor (SFF) predicted
factors presented in Table 4. Also this table is not repro- using Equation 1 for model DJ15at5 versus maximum dis-
duced in the RMR system version published in 1989 [17]. placement vector. The SFF was increased by increasing the
density of concrete in the dam. As can be seen, increasing
the factor of safety against sliding will stabilize the weak
Table 4 Adjusting factors for dam foundation and stability layer within the foundation of the dam. According to these
assessment [14] results, the SFF factor should be at least 4 or greater for
Joint In- Dip 10 - 30 concrete gravity dams on jointed rock so as to stabilize a
Dip Dip
formation Dip Dip Direction weak layer that may exist under the dam.
30 - 60-
0-10 Up- Down-
60 90 It should be noted that the maximum displacement vector
stream stream
Assess- for SFF=3.41 was calculated at an arbitrary step after sig-
VF U FA F VU
ment nificant movement of the dam along the joint (case 2) indi-
Rating 0 -15 -7 -2 -25 cating failure of the model.
Note: VF is Very favorable, U is Unfavorable, FA is fair, and
VU is Very Unfavorable.

Snell and Knight [6] studied systematically the problem of

dam stability taking account of all the forces and stresses
acting on the dam by using the limit equilibrium method.
Based on their study, Romana [18] proposed a new set of
adjusting factors, as shown in Table 5.
Table 5 Adjusting factors for the dam stability (RSTA) (after
Romana[18] )
Type of VF F FA U VU
Dam
Fill Others 10-30 DS 0-10 A - -
Gravity 10-60 30-60 US 10-30 US 0-10 -
DS 60-90 A A
Arch 30-60 10-30 DS 30-60 US 10-30 0-10
DS 60-90 A US A
RSTA 0 -2 -7 -15 -25
Note: DS is Dip Downstream, US is Dip Upstream, and A is
Fig.13 Velocity vectors (m/sec) for model DJ15at5.
any direction. Other notation as in Table 4.

In this paper, displacements have been measured at select-

ed points under the dam in the vertical direction and at the
crest of the dam in the horizontal direction, as shown in
Figures 4-12. Also, a state of numerical equilibrium was
checked in order to identify whether the model was stable
or not. It was shown that the models DJ15 did not reach a
state of equilibrium where the distance between the joint

Fig.14 Shear friction factor along the dam-foundation contact for
model DJ15at5 from heel.
5 CONCLUSIONS AND DISCUSSION
In this paper the effect of a weak joint within a discrete
fracture network of rock foundation in both the upstream
and downstream direction on the stability of a concrete
gravity dam has been studied using numerical modelling.
Only one joint was selected from the discrete fracture net-
work in order to demonstrate the mechanisms involved
within a simple model format. The authors have attempted
to demonstrate that both the location and orientation of the
weak joint have an important effect on the stability of a
dam. The modelling described suggests that a joint located
on the downstream side of the dam dipping at 15 degrees is
a critical case for the dam’s stability, especially where the
distance between the toe and the joint location is less than
half of the dam’s height. This study suggests that the shear
friction factor for the dam foundation contact should have
a value of at least 4. These results have implications for
engineers involved in future dam design and may be used
as a guide to develop a thorough DFN model to study the
stability of dams. Additionally this study concludes that
further development of a rock mass classification system
specifically for dam construction would be beneficial.
ACKNOWLEDGEMENT
The authors wish to acknowledge the Ministry of Higher
Education and Scientific Research in Federal Government
of Iraq for providing full scholarship to the lead author.
REFERENCES
1. Douglas, K.J., The shear strength of rock masses.
2002, PhD Thesis.The University of New South
Wales, Sydney, Australia.
2. Douglas D. Boyer , P.E., Geological factors
influencing dam foundation failure modes, in The
Role of Dams in the 21st Century, 26th Annual
USSD Conference, . 2006, U.S. Society on Dams:
San Antonio, Texas, USA. p. 105-112.
View publication stats
3. Poisel, R., W. Steger, and W. Unterberger. The
Malpasset dam failure-comparison between
continuum and discontinuum mechanics. in 7th ISRM
Congress. 1991. International Society for Rock
Mechanics.
4. Varshney, R.S., Rock foundation characteristics and
dam design. International Journal of Rock Mechanics
and Mining Sciences & Geomechanics Abstracts,
1972. 9(5): p. 645-659.
5. Singh, B. and R.S. Varshney, Effect of faults in the
foundations on dam behaviour, in Irrigation and
power. 1974, New Delhi: Central Board of Irrigation
and power.
6. Snell, E. and K. Knight, Susceptibility of dams to
failure by sliding on sub-foundation strata that dip
upstream., in Seventeenth International Congress on
Large dams (ICOLD), Vienna, 1991
7. Alshkane, Y.M., A.M. Marshall, and L.R. Stace,
Continuum and discontinuum modelling of gravity
dams on jointed rock foundations, in Rock
Engineering and Rock Mechanics: Structures in and
on Rock Masses. 2014, CRC Press. p. 745-750.
8. Wyllie, D.C., Foundations on Rock. 1999, London
and New York: The Bath Press, Bath.
9. Varshney, R.S., Concrete Dams. 1982, New Delhi:
Mohan Primlani.
10. Thomas, H.H., The Engineering of Large Dams.
1976, UK: John Wiley & Sons, Ltd.
11. USACE, (US Army Corps of Engineers), Engineering
and design: Gravity dam design, in Report EM1110-
2-2000. 1995: Washington, D.C.
12. Jansen, R.B., Advanced Dam Engineering for Design,
Construction, and Rehabilition. 1988, New York:
Van Nostrand Reinhold.
13. Itasca, FLAC Code (Fast Lagrangian Analysis of
Continuum. Minneapolis, USA. 2008.
14. Bieniawski, Z.T. and C.M. Orr, Rapid site appraisal
for dam foundation by geomechanics classification.
1976, 12th ICOLD: Mexico.
15. Bieniawski, Z., Engineering classification of jointed
rock masses. Civil Engineering in South Africa 1973.
15(12).
16. Gaziev, E.G. and S.A. Erlikham, Stresses and strains
in anisotropic rock foundation (model studies). in
proceeding Symposium of Rock Fracture. 1971,
International Society for Rock Mechanics (ISRM):
Report II-I, Nancy, France
17. Bieniawski, Z.T., Engineering rock mass
classifications. 1989: Wiley.
18. Romana, DMR (Dam Mass Rating). An Adaptation of
RMR Geomechanics Classification for use in Dams
Foundation, Int. Cong. on Rock Mechanics, (
Technology Roadmap for rock Mechanics), in South
African Institute of Mining and Metallurgy. 2003. p.
977-980.