STABILITY OF EMBANKMENT DAMS UNDER STATIC
LOADING CONDITIONS
Amjad Hussain Bhutto
PCE-010/5
A thesis submitted in fulfillment of the requirement
for the award of the degree of
Doctor of Philosophy
In
Civil Engineering
Department of Civil Engineering
Faculty of Engineering
Quaid-e-Awam University of Engineering, Science & Technology
Nawabshah
2020
i
AUTHOR’S DECLARATION
I declare that this thesis entitled “Stability of Embankment Dams Under Static
Loading Conditions” result from my own research except as cited in the
references. The thesis has not been accepted for any degree and is not
concurrently submitted in the candidature of any other degree.
Signature: ____________________________
Name of Student: Amjad Hussain Bhutto
Date: _______________________________
ii
DEDICATION
This dissertation is dedicated to my beloved parents. They always pray for my
success in every step of my life. The completion of this study is also part of their
prayers.
iii
ACKNOWLEDGEMENT
I am thankful to Almighty Allah for successful completion of my Ph.D work. I
am also greatful to my parents for their kindness and motivation for higher
studies.
The research work has been carried out under the supervision of Dr. Muhammad
Auchar Zardari, who provided every possible support which I needed during my
entire study period. I appreciate his patience, encouragement, and am fortunate to
have worked with him as a Ph.D student. Also, I am highly thankful to my co-
supervisor Prof. Dr. Bashir Ahmed Memon, for his guidance and valuable
suggestions from time to time. Especially his moral support and lessons of
patience encouraged me a lot.
Financial support from the Quaid-e-Awam University of Engineering Science
and Technology Nawabshah Pakistan, is greatly acknowledged. There are a
number of people who were involved directly or indirectly throughout my
research, I am thankful to all those for their support and I would like to
acknowledge them all. Thanks are also due to The Directorate of Postgraduate
Studies Quaid-e Awam University of Engineering Science and Technology for
their help and support to complete my work.
I would like to acknowledge Dr. Riaz Bhanbhro for his continuous guidance and
valuable suggestions throughout my research. I am also thankful to my younger
brother Rashid Hussain Bhutto and my colleague Shahnawaz Zardari for their
cooperation during my research.
Last but not least, I would like to thank my parents and my family for their moral
support throughout my study period.
iv
QUAID-E-AWAM UNIVERSITY OF ENGINEERING, SCIENCE
AND TECHNOLOGY NAWABSHAH
CERTIFICATE
This thesis written by Engr. Amjad Hussain Bhutto under the direction of his
supervisors and approved by all the members of the thesis committee, has been
presented to and accepted by the Dean, Faculty of Engineering, in fulfillment of
the requirements for award of the degree of Doctor of Philosophy in Civil
Engineering.
___________________________ ___________________________
Dr. Muhammad Auchar Zardari Prof. Dr. Bashir Ahmed Memon Supervisor Co-Supervisor
_______________________ _____________________________ Prof. Dr. Daddan Khan Bangwar Prof. Dr. Muhammad Munir Babar Internal Examiner External Examiner
___________________________ ______________________
Dr. Abdul Qayoom Jakhrani Prof. Dr. Noor Ahmed Memon
Director, Dean,
Postgraduate Studies& Research Faculty of Engineering
Date: ______________________
v
TABLE OF CONTENTS Description Page
DECLARATION i
DEDICATION ii
ACKNOWLEDGEMENT iii
CERTIFICATE iv
TABLE OF CONTENTS v
LIST OF TABLES ix
LIST OF FIGURES x
LIST OF ABBREVIATIONS xvi
ABSTRACT xviii
Chapter No. 1 Introduction
1.1 General 1
1.2 Problem statement 2
1.3 Research gap 5
1.4 Aim of research 5
1.5 Research objectives 5
1.6 Research methodology and work plan 6
1.7 Research outcomes and significance 7
1.8 Thesis layout 7
Chapter No. 2 Literature review
2.1 General 9
2.2 Historical development of embankment dams 9
2.3 Embankment dams 11
2.3.1 Zoned embankment dam 13
2.3.2 Components of zoned embankment dam 14
2.4 Soils used in construction of embankment dams 15
2.5 Soil varieties for embankments 16
2.6 Causes of slope failure 17
2.7 Causes of failure of embankment dams 19
vi
2.7.1 Rapid Drawdown mechanism in Embankment
Dams
21
2.7.2 Factors affecting slope-stability due to rapid
drawdown
24
2.8 Pore pressures in embankment dams 25
2.9 Stability conditions of embankment dams 26
2.9.1 End of construction and long-time stability 26
2.10 Slope stability analysis methods 29
2.10.1 Advantages of FEM in slope stability analysis 30
2.11 Main steps involved in FEM in Plaxis 2D 31
2.12 2D and 3D models 32
2.13 Factor of safety 34
2.13.1 Allowable safety factor 34
2.13.2 Importance of parametric analysis 35
2.14 Shear strength 35
2.14.1 Undrained shear strength 36
2.14.2 Drained shear strength 36
2.15 Staged construction of embankment dams 39
2.16 Settlement of embankment dams 39
2.17 Arching of embankment dams 41
2.18 Foundation treatment methods 46
2.18.1 Blanket grouting / consolidation grouting 47
2.18.2 Curtain grouting 47
2.19 Rainfall effect on slope stability 47
2.20 Types of embankment dams analysis 51
2.21 Soil models 52
2.21.1 Idealizations of plastic behavior 54
2.21.2 Linear elastic model 57
2.21.3 Mohr Coulomb Model (MCM) 58
2.21.4 Mohr-Coulomb Model Parameters 59
2.21.5 Application of Mohr Coulomb Model 63
vii
2.21.6 Hardening Soil Model 64
2.21.7 Applications of HSM 68
2.22 Summary 69
Chapter No. 3 Materials and Numerical Modelling
3.1 General 71
3.2 Project background and location 71
3.3 Materials and methods 72
3.4 Constitutive models 75
3.5 Stability of the dam 79
3.6 Settlement of the dam 80
3.7 Rapid drawdown of the dam 80
3.8 Effect of rainfall on stability of the dam 82
3.9 Summary 84
Chapter No. 4 Results and Discussion
4.1 General 85
4.2 Factors affecting dam stability 85
4.2.1 Effect of dam height on foundation soil 86
4.2.2 Effect of water filling on potential failure of
dam
88
4.3 Effect of friction angle of sandy gravel 89
4.4 Effect of friction angle of random fill 90
4.5 Effect of friction angle of clay core 91
4.6 Dam settlement comparison using MCM and
HSM models
92
4.6.1 Settlement at EoC and AFR 98
4.6.2 Dam settlement comparison w.r.t. depth 99
4.7 Variation effect of MOE of Sandy silt stone on
settlement w.r.t. depth
113
4.8 Comparison of long-term settlement for sandy
silt stone
115
viii
4.9 Rapid drawdown 116
4.9.1 Dam stability during drawdown under
Undrained Shear Strength of Clay = 20 kN/m2
117
4.9.2 Dam stability during drawdown under
Undrained Shear Strength of Clay = 25 kN/m2
119
4.9.3 Dam stability during drawdown under
Undrained Shear Strength of Clay = 30 kN/m2
121
4.10 Pore pressure under rapid drawdown 124
4.11 Effect of rainfall on slope stability of dam 132
Chapter No. 5 Conclusion and Suggestions
5.1 Conclusion 134
5.2 Recommendations for future studies 135
References
137
ix
LIST OF TABLES
TABLE NO. TITLE PAGE
2.1 Summary of dams constructed in Pakistan 11
2.2 Dam classification based on capacity and height 12
2.3 Recommended values of safety factors 35
2.4 Parameters of Mohr Coulomb Model 58
2.5 Typical values soil permeability 63
2.6 HSM parameters 67
3.1 Material properties for Nai Gaj dam and its foundation 76
3.2 HSM parameters for four main zones of dam 77
3.3 Relationship b/w consistency index and undrained
shear-strength
81
4.1 Summary of dam's settlement comparison for EoC 97
4.2 Summary of dam’s settlement comparison for AFR 97
4.3 Summary of dam’s settlement comparison for EoC 105
4.4 Summary of dam’s settlement comparison for AFR 105
4.5 Variation effect of MoE of sandy siltstone on
settlement w.r.t. depth at EoC
112
4.6 Variation effect of MoE of sandy siltstone on
settlement w.r.t. depth for AFR
113
4.7 Reservoir drawdown rate vs. dam safety factor under
undrained shear strength of clay core = 20 kN/m2
123
4.8 Reservoir drawdown rate vs. dam safety factor under
undrained shear strength of clay core = 25 kN/m2
123
4.9 Reservoir drawdown rate vs. dam safety factor under
undrained shear strength of clay core = 30 kN/m2
123
4.10 Development and dissipation of excess pore pressure
under various drawdown rates
125
x
LIST OF FIGURES
FIGURE NO. TITLE PAGE
2.1 Components of zoned type embankment dam
14
2.2 Slope instability when reservoir is lowered rapidly 23
2.3 Decrease of safety factor under rapid drawdown
scenario of Pilarcitos Dam
23
2.4 (a) Increase in pore pressure during construction time and
decrease during a long time (t)
28
2.4 (b) Change in safety factor during EoC and long-time 28
2.5 Potential failure zone in a dam with rigid foundation 28
2.6 Potential failure zone extending in embankment
and foundation
29
2.7 15 and 6- point nodal scheme and stress point scheme 31
2.8 Plain strain condition 33
2.9 Axisymmetric condition 33
2.10 Effect of increased clay content on bearing capacity of
footings on sand
38
2.11 Influence of increase in clay on slope's safety factor w.r.t
depths of failure surface
38
2.12 Effect of side slopes on arching ratio of clay core and shell 43
2.13 Effect of side slopes of core on arching ratio 43
2.14 Effect of increase of filter thickness on the arching ratio 44
2.15 Effect of shell core stiffness ratio with arching 44
2.16 Influence of inclined core on arching effect in a zoned
embankment dam
45
2.17 Effect of MoE of foundation on arching effect in a zoned
embankment dam
45
2.18 Rainwater infiltration through a dam 48
2.19 Rainfall infiltration through dam 49
2.20 Stress-strain curve of metals 54
xi
2.21 (a) Elastic and plastic behavior of material 55
2.21 (b) Elastic-linear-strain hardening plastic response of material 55
2.21 (c) Rigid plastic response of material 56
2.22 (a) Lightly over-consolidated clays and loose sands 56
2.22 (b) Over-consolidated clays and dense sands 57
2.23 Stress-strain behavior of an elastic, perfectly plastic model 58
2.24 Drained triaxial test results for 59
2.25 Dilatancy angle using the stress-strain curve obtained from
triaxial test
60
2.26 Determination of cohesion and friction angle 61
2.27 CD triaxial compression tests on three over consolidated
clays
62
2.28 Hyperbolic response in primary loading for triaxial test 66
2.29 Determination of from oedometer test 67
3.1 3D view of Nai Gaj Dam with earthen dykes 72
3.2 Cross section of Nai Gaj dam 75
3.3 FEM for Nai Gaj dam 79
3.4 FEM mesh of Nai Gaj dam 79
3.5 Highest reservoir level of Nai Gaj dam 79
3.6 10 m lowering of reservoir @ 1 m/day 82
3.7 Lowering of the reservoir up to 55 m depth @ 0.1 m/day 82
3.8 Average annual rainfall at Nai Gaj dam for four decades 83
3.9 Input of rainfall intensity at Nai Gaj Dam 83
4.1 Increase in pore pressure at 15 m dam raising 86
4.2 Increase in pore pressure at 30 m dam raising 87
4.3 Increase in pore pressure at 45 m dam raising 87
4.4 Increase in pore pressure at 59 m dam raising 88
4.5 Potential failure zone of the dam at the EoC 89
4.6 Potential failure zone of the dam AFR 89
4.7 Effect of friction angle for sandy gravel on dam safety factor 90
xii
4.8 Effect of friction angle for random fill on dam safety factor 91
4.9 Effect of friction angle for clay core on dam safety factor 92
4.10 Dam settlement comparison for clay core at EoC 93
4.11 Dam settlement comparison for clay core at AFR 93
4.12 Dam settlement comparison for sandy gravel at EoC 94
4.13 Dam settlement comparison for sandy gravel at AFR 94
4.14 Dam settlement comparison for random fill at EoC 95
4.15 Dam settlement comparison for random fill at AFR 95
4.16 Dam settlement comparison for sandy silt stone at EoC 96
4.17 Dam settlement comparison for sandy silt stone at AFR 96
4.18 Settlement increase in % when MoE for sandy siltstone
varied from 70,000-125,000 kN/m2
98
4.19 Dam's central axis at which settlement was computed
w.r.t. depth at an interval of 20 m
100
4.20 Dam's settlement comparison w.r.t. depth at EoC applied to
clay core
100
4.21 Dam's settlement comparison w.r.t. depth at AFR applied to
clay core
101
4.22 Dam's settlement comparison w.r.t. depth at EoC applied to
sandy gravel
101
4.23 Dam's settlement comparison w.r.t. depth at AFR applied to
sandy gravel
102
4.24 Dam's settlement comparison w.r.t. depth at EoC applied to
random fill
102
4.25 Dam's settlement comparison w.r.t. depth at AFR applied to
random fill
103
4.26 Dam's settlement comparison w.r.t. depth at EoC applied to
sandy silt stone
104
4.27 Dam's settlement comparison w.r.t. depth at AFR applied to
sandy silt stone
104
4.28 Dam's settlement comparison w.r.t. depth for EoC applied to 106
xiii
sandy silt stone (MoE = 70,000 kN/m2)
4.29 Dam's settlement comparison w.r.t. depth at AFR applied to
sandy silt stone (MoE = 70,000 kN/m2)
107
4.30 Dam's settlement comparison w.r.t. depth for EoC applied to
sandy silt stone (MoE = 80,000 kN/m2)
107
4.31 Dam's settlement comparison w.r.t. depth for AFR applied to
sandy silt stone (MoE = 80,000 kN/m2)
108
4.32 Dam's settlement comparison w.r.t. depth at EoC applied to
sandy silt stone (MoE = 90,000 kN/m2)
108
4.33 Dam's settlement comparison w.r.t. depth for AFR applied to
sandy silt stone (MoE = 90,000 kN/m2)
109
4.34 Dam's settlement comparison w.r.t. depth at EoC applied to
sandy silt stone (MoE = 100,000 kN/m2)
109
4.35 Dam's settlement comparison w.r.t. depth for AFR applied to
sandy silt stone (MoE = 100,000 kN/m2)
110
4.36 Dam's settlement comparison w.r.t. depth at EoC applied to
sandy silt stone (MoE = 110,000 kN/m2)
110
4.37 Dam's settlement comparison w.r.t. depth for AFR applied to
sandy silt stone (MoE = 110,000 kN/m2)
111
4.38 Dam's settlement comparison w.r.t. depth at EoC applied to
sandy silt stone (MoE = 125,000 kN/m2)
111
4.39 Dam's settlement comparison w.r.t. depth for AFR applied to
sandy silt stone (MoE = 125,000 kN/m2)
112
4.40 Settlement at EoC by MCM and HSM models 114
4.41 Settlement for AFR by MCM and HSM models 114
4.42 Prediction of long-term settlement for sandy siltstone (MoE
= 125,000 kN/m2)
115
4.43 Prediction of long-term settlement for sandy siltstone (MoE
= 70,000 kN/m2)
116
4.44 Safety factor vs. time while lowering the reservoir under
undrained shear strength of clay = 20 kN/m2
118
xiv
4.45 Dam failure zone when reservoir is lowered @ 1 m/day to a
depth of 15 m (Su=20 kN/m2)
118
4.46 Dam failure zone when reservoir is lowered @ 0.1 m/day to
a depth of 20 m (Su=20 kN/m2)
119
4.47 Dam safety factor vs. time under undrained shear strength of
clay = 25 kN/m2
120
4.48 Dam failure zone when reservoir is lowered @ 1 m/day for a
depth of 18 m (Su=25 kN/m2)
120
4.49 Dam failure zone when reservoir is lowered @ 0.1 m/day for
a depth of 55 m (Su=25 kN/m2)
121
4.50 Dam safety factor vs. time under undrained shear strength of
clay = 30 kN/m2
122
4.51 Dam failure zone when reservoir is lowered @ 1 m/day for a
depth of 23 m (Su=30 kN/m2)
124
4.52 Dam failure zone when reservoir is lowered @ 0.1 m/day for
a depth of 55 m (Su=30 kN/m2)
124
4.53 Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 10 m depth in 10 days (Su=20
kN/m2)
126
4.54 Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 10 m depth in 100 days (Su=20
kN/m2)
126
4.55 Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 20 m depth in 20 days (Su=20
kN/m2)
127
4.56 Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 20 m depth in 200 days (Su=20
kN/m2)
127
4.57 Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 20 m depth in 20 days (Su=25
kN/m2)
128
xv
4.58 Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 20 m depth in 200 days (Su=25
kN/m2)
128
4.59 Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 55 m depth in 55 days (Su=25
kN/m2)
129
4.60 Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 55 m depth in 550 days (Su=25
kN/m2)
129
4.61 Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 20 m depth in 20 days (Su=30
kN/m2)
130
4.62 Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 20 m depth in 200 days (Su=30
kN/m2)
130
4.63 Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 55 m depth in 55 days (Su=30
kN/m2)
131
4.64 Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 55 m depth in 550 days (Su=30
kN/m2)
131
4.65 Effect of rainfall on dam stability 133
4.66 Dam failure zone due to 1-day precipitation 133
4.67 Dam failure zone due to 5-day precipitation 133
xvi
LIST OF ABBREVIATIONS
MOE = Modulus of elasticity
MCM = Mohr Coulomb Model
HSM = Hardening Soil Model
EOC = End of construction
AFR = After filling of reservoir
uo = Initial pore pressure
Δu = Change in pore pressure
tc = Construction time
t = Long time
LEM = Limit equilibrium method
FEM = Finite element method
2D = 2 Dimensional
3D = 3 Dimensional
FE = Finite element
kN/m2 = kilo Newton per square meter
kPa = kilo Pascal
m/s = Meter per second
= Arching ratio
= Total vertical stress in homogeneous dam
= Total vertical stress in zoned embankment dam
E = Young’s modulus of elasticity
v = Poisson’s ratio
G = Shear modulus
K = Bulk modulus
= Oedometer modulus
= Angle of internal friction
= Angle of dilatancy
c = Cohesion
xvii
= Elastic strain
= Plastic strain
E = Modulus of elasticity
Eo = Tangent modulus
E50 = Secant modulus/ primary loading stiffness modulus
Eur = Unloading-reloading modulus
C = Effective cohesion
= Effective friction angle
cm/sec = Centimeter per second
= Reference value of confining pressure
qf = Ultimate deviatoric stress
qa = Quantity of deviatoric stress
= Failure ratio
= Axial strain
= Stiffness modulus
= Tangent stiffness
Su = Undrained shear strength
= Major principal stress
= Minor principal stress
= Effective minor principal stress
m = Stress dependency power
= Coefficient of normal consolidation
= Uniaxial compressive strength
= Liquidity index
LL = Liquid limit
PL = Plastic limit
W = Natural moisture content
KPK = Khyber Pakhtunkhwa
xviii
ABSTRACT
The embankment dams require a massive quantity of fill materials. Embankment
dams have different zones, and the soils are generally obtained from different borrow
areas having different strength and stiffness properties, which may have effect on
stability and settlement. If the upper limit of the strength and stiffness properties of
various soils is used, there will be an overestimation of stability and underestimation
of settlement. The stability and settlement response of a 59 m high embankment dam
called Nai Gaj dam situated in District Dadu, Sindh, Pakistan is presented in this
study. The parametric analysis was conducted for three main zones of the dam:
upstream shell, core and downstream shell, which consisted of sandy gravel, clay and
random fill, respectively. The friction angle values of these materials were gradually
decreased to represent the soil conditions of different sources. It was observed that
the stability requirements for the end of construction and after filling of the reservoir
could be satisfied if friction angle values of the sandy gravel and random fill be
respectively utilized as 340 and 32
0 instead of 37
0 and 34
0. However, the value of the
friction angle of the core can be used as 300 without any reduction so that the dam
could be safe after filling the reservoir.
The settlement behaviour of the dam was analyzed by using the MCM and
Hardening Soil Model (HSM) models. The MCM model was applied to all material
zones and the HSM was used for four major material zones that occupied significant
volume. The settlement response of the dam was almost similar when computed with
the MCM and HSM for the three material zones (clay core, sandy gravel and random
fill), each having a modulus of elasticity (MOE) in the range of 25000 to 50000 kPa.
However, it was found that after the end of construction the MCM showed about
57% and 50% more settlement as compared to the HSM when MOE of sandy
siltstone was respectively varied from 70000 to 125000 kPa. The rate of increase of
settlement at the crest and at a depth of 120 m computed by MCM compared to HSM
was 53% and 82% respectively for after filling of the reservoir condition. The
settlement computed with MCM and HSM were 2.9% and 1.35% of the dam height.
It can be interpreted that the settlement predicted with MCM is unrealistically high
xix
due to the single constant value of modulus of elasticity (MOE). On the contrary, the
predictions of HSM are in agreement with the literature. besides, the long term
settlement computed using MCM is about 59% higher than that of HSM for after
filling of reservoir condition. The settlement of an embankment dam could be
predicted reliably by using HSM even when a limited number of stiffness data is
available.
The reservoir of the dam was lowered to a depth of 10 m up to 55 m in gradual
increments. It was found that the reservoir could be lowered up to a depth of 55 m at
a rate of 0.1 m/day when the undrained strength of clay core was 25 kN/m2. The dam
was also subjected to average daily rainfall intensity of 0.086 m/day continuously for
five days. The safety factor was reduced to 1.44 when the dam was subjected to the
rainfall duration for five days. This indicates that with the increase of duration of
rainfall, there is a reduction in the safety factor, because water enters through the
embankment and saturates the slopes. However, the stability of the dam is
satisfactory for the above-mentioned intensity and duration of the rainfall.
Keywords: embankment dam, slope stability, settlement, pore pressure,
consolidation, end of construction, filling of the reservoir, rapid drawdown, rainfall-
induced stability, parametric analysis, modulus of elasticity, Mohr Coulomb model,
Hardening Soil model.
CHAPTER 1
INTRODUCTION
1.1 GENERAL
Embankment dams have played a significant role in providing water for drinking and
agricultural purposes and generating electricity since ancient times. It is estimated that
about half of the embankment dams failures have occurred due to the build-up of
excess pore pressure, seepage, internal erosion, and excessive deformations [1]. These
instability problems may occur during routine and unusual conditions (e.g., heavy
rainfall, intense earthquake) during the service life of an embankment dam [2]. The
failure of a dam may cause loss of lives and financial resources. Hence, all dams must
function safely under routine everyday operations and under unusual conditions such
as floods and earthquakes. All embankment dams during their service life should be
systematically evaluated for safe performance under all operational conditions.
Most slope failures of embankment dams have occurred either during construction or
at the end of the construction stage. The increase of pore water pressures depends on
the availability of water content in the fill and construction rate [3]. The construction
of embankments at a rapid pace may increase total stress, as successive layers of fill
are placed. Due to soils' low permeability, the excess pore pressures might gradually
develop during the reservoir's raising and filling. The increase in excess pore
pressures may lead to a dam's slope instability or failure [2].
Water from rainfall may enter through the embankment into the dam body. High-
intensity rain may cause cracks in the crest and enter the embankment soils and join
the water table. As a result, the soils' suction and shear strength may reduce,
ultimately leading to instability conditions in embankment dams. The entry of
rainwater into clay core may have severe consequences on the embankment dams'
stability and settlement.
1
2
Besides the development of pore pressures, excessive deformations may affect dam
safety. The magnitude of deformations depends on several factors, such as soil
properties in various zones, the raising rate, and the dam's geometry. The dimensions
of various zones and their soils are designed in such a way as to minimize
deformations of an embankment dam concerning unique foundation conditions for
each site [4].
Pakistan is an agricultural country whose economy mainly depends on water
availability, which is also one of the primary sources of hydropower electricity. For
this purpose, various dams have been constructed, and some of the dams are planned
to be built in the future. It seems necessary to develop guidelines to ensure the
embankment dams' safety concerning local site conditions during their design life.
1.2 PROBLEM STATEMENT
The increase in pore water pressure during the construction of a dam and later during
filling of the reservoir stage may cause instability. This phenomenon leads to a
reduction in effective stress, which consequently reduces the shear strength. Thus, the
problem may lead to the dam's failure if the increase of pore pressures is not
appropriately controlled. Some of the failures of embankment dams have occurred
due to excessive generation of pore pressures during construction and filling of the
reservoir ( e.g., Teton dam USA 1976, Malpasset Dam France 1959, St. Francis Dam
USA 1928 [5]. The increase in pore pressure in an embankment dam may occur if
there is an intense rainfall for enough time, or the reservoir is lowered rapidly due to
some emergency.
Sometimes, an embankment dam's reservoir may be lowered rapidly due to
unavoidable conditions that may lead to the dam's instability. However, an
embankment dam's reservoir level needs to be lowered carefully at a slow rate to
avoid the upstream slope's instability. The rapid drawdown condition is dangerous for
the upstream slope because the water pressures do not counterbalance seepage forces
[6]. After drawdown, changes in pore pressure may occur due to changes in total
stress. Changes in pore pressure are dependent upon the stress-strain response of soils.
3
If the soils are stiffer, there may be a little change in pore pressure induced due to
applied stresses. The dissipation of pore pressure depends on initial conditions after
drawdown, permeability and stiffness of the soils.
The entry of water from rainfall through the crest of the dam to the phreatic level is
also a severe issue, as it may lead to a decrease in suction and shear strength due to an
increase in pore pressure. A gradual accumulation of rainwater in a slope may finally
result in slope failure [7].
The embankment dams are generally constructed as zoned structures involving
various materials with diverse stiffness and strength properties. Due to the difference
in stiffness of various zones, there may be the possibility of differential settlement. If
the dam is constructed on a weak foundation, there may be the risk of excessive
horizontal deformations and settlement.
Embankment dams' stability is affected by factors such as a change in the dam's
material properties and geometrical shape. The strength properties of embankment
dams depend upon cohesion and angle of internal friction [8].
It is evident from the literature that soil compaction with optimum water content on
the dry side may reduce shear strength significantly, particularly when fill material is
submerged. The resulting low shear strength may lead to slope instability. Also, the
optimum compaction of the dry side may result in the excessive settlement [9]. A
comparison of soil types also reveals that the clay core in an embankment dam
possesses low shear strength and high compressibility [10].
Since the embankment dams consist of a massive volume of fill materials, in such
conditions, generally different localities are chosen as the source of soil to be used as
the fill material in embankment dams. It is commonly observed that there is a
difference in stiffness and strength properties of soils collected from different borrow
areas. This difference in properties of collected soils may lead to problems related to
stability and settlement. Hence, it is highly essential to conduct a parametric study of
4
the soils concerning stiffness and strength to obtain a realistic estimate of an
embankment dam's stability and settlement.
The shear strength parameters, cohesion and friction angle, influence the stability of a
dam. The modulus of elasticity (MoE) of soils is a critical parameter that affects
settlement response. It is also common practice that the MoE of soils is not evaluated
based on laboratory or field tests. On the contrary, the value of MoE of soils is taken
from the literature based on correlations with other parameters. It is also evident that
the computations based on correlations are generally not reliable [11].
Presently, commercial finite element programs like PLAXIS 2D can be utilized to
computation stability and settlement behavior of embankment dams. This type of
software contains a library of soil models. The designer must choose a suitable kind of
constitutive model for a specific analysis performed for either stability or settlement.
Such constitutive models require input properties, mainly related to shear strength,
stiffness and permeability of soils. The stability and settlement of embankment dams
are directly dependent on the given input of material properties. Reliable estimates of
stability and settlement can be obtained if the suitable values of stiffness and strength
of various soils are given as input. As mentioned earlier, when the soils of different
zones of an embankment dam are obtained from different borrow areas, there may be
less reliability of the given input parameters.
First, it is necessary to identify appropriate constitutive models for computation of
stability and settlement response of embankment dams. Various types of constitutive
models are available in commercial finite element programs. If an advanced
constitutive model is selected for numerical analysis, there will be a need for more
advanced laboratory testing to properly evaluate soil properties. It is commonly
observed that there is a great lack of such advanced laboratory or field tests from
which reliable soil properties can be estimated. This is because the cost of advanced
laboratory and field tests may be higher than the allocated budget for the geotechnical
investigation. This implies a need for more studies on numerical modeling issues that
deal with obtaining reliable parameters of advanced constitutive models based on
5
limited laboratory or field data. Secondly, it is essential to select a suitable range of
stiffness and strength properties of soils used in various embankment dam zones.
For a zoned embankment dam, it is necessary to identify which types of soils have
more influence on settlement so that advanced constitutive models can be used for
their stress-strain behavior to obtain the reliable magnitude of settlement.
In addition to stability and settlement of an embankment dam, it is also essential to
numerically analyze (i) the safe rate of lowering of the reservoir if conditions warrant
for such a situation, and (ii) the effect of maximum possible rainfall, in the region
where the dam is located, on the stability of the embankment dam.
1.3 RESEARCH GAP
Based on the literature review, there is a significant need for more case studies on
stability and settlement issues in Pakistan's embankment dams.
1.4 AIM OF RESEARCH
Following is the main aim of the present study:
To address stability issues of embankment dams during construction, filling of the
reservoir and during service life.
1.5 RESEARCH OBJECTIVES
(i) To evaluate the stability of the embankment dam and its foundation during
gradual raising
(ii) To evaluate the stability of the embankment dam during the first filling of
the reservoir
(iii) To evaluate settlement of the dam and its foundation during construction,
filling of the reservoir and throughout the service life of the dam
(iv) To evaluate the stability of the dam for rapid drawdown condition and to
estimate the safe rate of lowering of the reservoir in case of emergency
6
(v) To evaluate the rainfall-induced stability of the embankment dam for
maximum possible rainfall when the reservoir is full.
1.6 RESEARCH METHODOLOGY AND WORK PLAN
A detailed literature review was conducted regarding stability and settlement issues
on embankment dams under static loading conditions. The focus of the literature
review was on the following main topics:
(i) Stability problems of embankment dams during construction, and filling of the
reservoir
(ii) Effect of excess pore pressure generated during gradual raising, lowering of
the reservoir, and rainfall on the stability of the embankment dams
(iii) Identification of material zones that have a significant influence on the
settlement of the dams
(iv) Use of suitable constitutive models for computation of stability and settlement
of embankment dams
(v) Identification of a proper range of strength and stiffness properties used in
various zones of embankment dams
A locally available embankment dam called Nai Gaj dam was identified as a case
study to perform numerical analysis to address stability issues, as mentioned in the
problem statement.
Numerical analysis of the embankment dam for the following conditions was
performed:
(i) Stability during gradual raising and filling of the reservoir
(ii) Stability during the rapid drawdown of the reservoir
(iii) Effect of maximum possible rainfall on the stability of the slopes
(iv) Settlement of the dam during the End of Construction (EoC) and After Filling
of the Reservoir (AFR)
7
1.7 RESEARCH OUTCOME AND SIGNIFICANCE
The proposed research outcome is hoped to benefit personnel working on
embankment dams, particularly concerning stability and settlement issues. The
following are regarded as the outcomes of the proposed research.
(i) Estimation of properties of filling material collected from different sources.
The same will be utilized in the constitutive modeling, especially when less
data regarding stiffness and strength of the soil is available
(ii) Identification of suitable models for different soil zones
(iii) Evaluation of stability and settlement of the dams
1.8 THESIS LAYOUT
The background of the research is presented in chapter No. 1, titled ―Introduction.‖ In
this chapter overall summary of the study, problem statement, the aim of the study,
objectives of the research, research methodology, and outcomes and consequence of
the study and thesis layout are described.
A review of the previously available state of the art by innumerable researchers is
presented in chapter No. 2. In this chapter, previous research work associated with the
research work presented in this thesis was reviewed. The literature review concerning
stability issues, settlement, different ratios of drawdown and rainfall effect of the
embankment dam using finite element program Plaxis 2D with constitutive models
explained in detail.
Complete details of dam location, dam geology, the primary purpose of dam socio-
economic condition materials used in the earthwork and the design of this dam, Finite
Element Model of the dam and associated material properties, stability, settlement, the
effect of lowering of the reservoir and maximum possible rainfall on the stability of
the dam are stated in chapter No. 3 titled ―Materials and Numerical Modelling.‖
8
Results and discussions are presented in chapter No. 4, ―Results and Discussion.‖ This
chapter explained all investigational work in tabular and graphical forms and
discussed them in detail. In the end, this chapter contains a summary of the results.
Chapter No. 5, ―Conclusions,‖ where detailed conclusions and suggestions for future
studies are described. In this chapter, all possible findings coined from the obtained
results are briefly presented along with potential suggestions.
9
CHAPTER 2
LITERATURE REVIEW
2.1 GENERAL
In the following sections, a review of available state-of-art is presented. The review
has been carried out concerning the objectives of the proposed research work. The
review's focus is on the instability conditions of embankment dams during
construction and operation due to increased pore pressure, drawdown and excessive
rainfall intensity, and identification of appropriate soil models that can be utilized to
analyze stability settlement response of embankment dams. Recent and relevant
literature has been included on the following main topics: historical development of
embankment dams zoned embankment dams and their components, materials of
construction of embankment dams, causes of failure of embankment dams,
development and dissipation of pore pressures under various loading conditions,
stability conditions of embankment dams, methods of slope stability analysis based on
limit equilibrium method, and finite element method.
2.2 HISTORICAL DEVELOPMENT OF EMBANKMENT DAMS
Egyptians were the pioneer in dam construction; they built the Sad-el-Kafara dam, 12
m high and 106 m long. After initial years of construction, the dam failed due to poor
construction and water overflow through the crest. Smith and Schnitter [12, 13]
observed that the dam's remains could be seen even today. In ancient times, the
Nimrud dam was constructed near Baghdad. The failure of this dam occurred in 1200
BC [14]. Sad-el-Arim dam, which initially was 4 m high, built along the River Marib
in Yemen. This dam was triangular with side slopes of 1:1. Sad-el-Arim dam was
constructed to provide water for irrigation purposes [15]. The failure of this dam
occurred in the 6th century due to poor maintenance for an extended period. Several
small earthen dams were constructed in eastern Arabia in ancient days.
9
10
In the olden days, several embankment dams were also constructed in India and
Ceylon. In 504 BC, the Padavil dam was built in Ceylon by making an embankment
that was 21.4 m high, having a foundation width of 61 m and a crest width of 8 m,
with a total length of 18 km [16]. Thirteen million m3 of earth fill was utilized for this
dam. The construction of dams had been given top priority during the reign of the
Roman Empire. One of the olden dams constructed by the Roman engineers was the
Cornalvo dam, 200 m long and 20 m in height [17]. The dam's inside was divided into
various sections by constructing stone walls in longitudinal and transverse directions.
The sections were latter filled with locally available materials, clay and stone. The
whole structure was covered with earth.
Masonry stone lining was used to construct dams in Germany, Austria, France, Great
Britain, and Italy. The necessity of dams construction was seriously felt during the
industrial revolution in the United Kingdom and other European countries.
Eventually, several dams were constructed in the United Kingdom and France at the
end of the 18th century and at the beginning of the nineteenth century [18].
In Great Britain, one of the largest dams was Todd Brook dam, which was 21 m high
[19]. In the 19th
century, various embankment dams were built in Europe and the USA
for water supply [20]. During the end of the 19th
century and beginning of the 20th
century, the industrial countries have had not enough experience of constructing
dams; therefore, failure of several dams occurred in those countries [21].
It is a common saying that necessity is the mother of invention. Having more need for
dams construction, attention was given to scientific knowledge for water flow through
soils. In 1856, Darcy presented a law for the flow of water through porous sandy soils.
Later, it was found that the distribution of water pressure through soils follows
Laplace’s differential equation. Karl Terzaghi, in the early 1920s, presented the basic
concepts of Soil Mechanics, which helped in understanding the behavior of dams built
of local construction materials. In Sweden, during that time, slopes' stability was
studied based on Soil Mechanics' principles.
11
Further advancement of Soil Mechanics theories regarding seepage, consolidation,
and settlement has led to better understating of embankment dams' design and
construction safely. The advancement of technology has also helped in the building of
large embankment dams. Modern dams are not only constructed for control of floods
and water supply, but also for the generation of hydroelectricity. Various dams have
been built in different parts of the world. For simplicity, some major dams constructed
in Pakistan are presented in Table 2.1.
Table 2.1: Summary of dams constructed in Pakistan.
Dam Height Location Type of dam
Mangla Dam 147 m Jhelum River Embankment Dam
Tarbela Dam 143.26 m Indus River Earth fill Dam
Gomal Zam dam 133 m Gomal River Curved gravity, roller
compacted concrete
Warsak Dam 76.2 m Kabul River,
Peshawar Gravity Dam
Allai Khwar Dam 51 m Allai Khwar River Reinforced cement concrete
Khan pur Dam 51 m Haro River, KPK Gravity dam
Hub dam 48 m Hub River Earth fill Dam
Mirani Dam 39 m Dasht River Concrete faced rockfill
Sabak Zai Dam 34.7 m Zhob River Earth and rockfill Dam
Duber Khwar Dam 32 m Duber Khwar River Gravity and roller
compacted concrete
2.3 EMBANKMENT DAMS
Dams are considered necessary for the socio-economic development and prosperity of
a nation. They are constructed to serve different functions such as water supply,
electricity, etc. Dams are built to act as a barrier to retain water. There are several
types of dams; embankment dams are the most commonly made [22]. Table 2.2
provides useful information regarding dams classification based on their storage
capacity and height [23].
12
Table 2.2: Dam classification based on capacity and height [23].
Size (m) Capacity (m3) Height (m)
Small Less than - 1,000,000 Less than 8
Medium 1,000,000 – 3,000,000 8 to 15
Large 3,000,000 – 20,000,000 15 to 30
Major Above - 20,000,000 More than 30
Soils and rocks are mainly used as materials for the construction of embankment
dams. Slopes of an embankment dam must remain stable during all operational stages
of its design life. There must be compatibility of soils and geometries with the
abutments of embankment dams [24].
The angle of inclination of embankments should be such that stability is assured. But
care should be taken in the design. Reducing the inclination angle up to 2-3° (to
horizontal) will cause a substantial increase of fill volume for a great dam [25].
There are several advantages to the construction of embankment dams. Some of the
main benefits of the embankment dams are described as follows [26]:
o If properly designed, these dams may be suitable for different climate,
topography and geological conditions.
o If suitable treatment is provided, embankment dams may remain stable even
on alluvial and pervious foundations.
o With the advancements of technology and existing knowledge of geotechnical
principles, a wide range of soils could be used to construct embankment dams.
o The construction of embankment dams does not involve the use of cement and
steel. This results in saving construction time and cost.
o Although an embankment dam may consume 4 to 6 times more fill volume
than a concrete dam, its total construction cost is 1/15 to 1/20 times lower than
the concrete dam.
o It has also been reported that the embankment dams offer more resistance to
seismic waves as compared to concrete dams.
The design of embankment dams should fulfill the following requirements:
13
(i) The main requirement for a dam to be safe is slope stability during all phases
of operation, i.e., EoC, filling of the reservoir, and long-term steady-state
seepage conditions.
(ii) Dams should be designed in such a way as to control excessive seepage,
internal erosion and uplift pressure.
(iii) It is necessary to provide enough freeboard to account for overtopping, post-
construction settlement of embankment and foundation.
(iv) If a dam is located in a seismic zone, it should have sufficient freeboard to
account for deformation due to the earthquake.
(v) The slopes should be protected with suitable and durable construction material
to resist the action of waves, rainfall, and wind.
2.3.1 Zoned Embankment Dam
Embankment dams may be homogenous, modified homogenous and zoned type. The
type of embankment dam is typically selected as zoned if there is the availability of
various types of soils. Embankment dams are heterogeneous, with various zones. The
fill material of a zoned embankment dam is usually obtained from local sites.
Embankment dams have a central impervious core. The upstream slope is usually
covered with a thin layer of rock called riprap, which protects it from water waves or
erosion. The downstream slope is normally vegetated to avoid erosion. Internal
drainage is provided to alleviate the harmful effects of pore water pressure. The
horizontal drainage is used to enhance the reduction of excess pore water pressures
[25]. This type of dam contains more than one material, e.g., clay, sand, and silt.
Various components of a zoned type embankment dam are shown in Figure. 2.1.
14
Figure. 2.1: Components of zoned type embankment dam [27].
2.3.2 Components of Zoned Embankment Dam
An embankment dam may consist of the following components [27]:
Core: It is an impervious wall of clay or fine material. The clay core is extended from
ground level to the highest flood level.
Shell: The particle size of the shell is coarser than the core. It is provided on both
sides of the core. It helps in the distribution of loads in the foundation and provides
support to the core.
Transition filter: It is placed in between the core and shell. With its help, the
movement of fine particles of the core is prevented from entering the shell's voids.
Upstream blanket: It is a layer of impervious material placed on the upstream side to
enhance the flow path and decrease seepage. As a result of this, excess pore pressures
and the seepage rate decrease on the downstream side.
Cut-off: It is an impervious barrier placed below the dam's core up to an impervious
stratum. It controls the flow of water. It is provided in those cases when the
foundation is incapable of resisting seepage.
Internal drain: It is provided to control seepage water, which may enter through the
core and cut-off. With the help of an internal drain, the downstream shell is kept
15
unsaturated. To prevent the carrying of fine particles, protective filters are also used in
addition to the internal drains.
Toe drain: It is generally provided at the downstream toe. The essential function of it
is the same as that of the internal drain.
Riprap: It is a protective layer that is generally provided to upstream sloped face and
sometimes to downstream slope to guard against the action of waves and rainfall,
respectively.
2.4 SOILS USED IN CONSTRUCTION OF EMBANKMENT
DAMS
For the construction of embankment dams, both fine-grained and coarse-grained soils
are used [28, 29]. The material mustn't contain more than 5% chloride salts, 2%
sulfate salts and 5% organic matter. The core of earth dams is constructed of
impervious earth materials having a permeability of 10-5
cm/s. The moisture content
of clay should be close to optimum moisture content or a little bit higher. An
embankment dam's construction work may become complicated and expensive if the
moisture content of clay is very wet or too dry. The shell of the dams could be
constructed with natural coarse-grained soils.
The rockfill materials which meet the following criteria be used for the construction
of embankment dams:
o Particle size smaller than 5 mm should not be more than 5%
o Particle size smaller than 20 mm should not be more than 30%
o The minimum permeability of the rockfill should be at least 10-2
m/s. The
shear strength of the rockfill should be high considering submergence in
water, breaking and fracture at high stresses.
Those, as mentioned above, are the suitable properties of rockfill. Several dams have
been constructed with rock fill, which does not satisfy suitable properties. It is usually
a difficult task for an engineer to opt for such a design in which rock fill having less
16
suitable properties is proposed to be used and to assure that the dam will function
safely over design life.
Transition zones, filter and subsoil drains can be constructed with sand, sandy gravel
material [30]. Sometimes, natural materials do not comply with all the criteria needed
for such zones. Therefore, several fractions of other materials are added, which may
result in 2-3 times more cost [31]. Dam slopes are usually protected from the action of
waves and weathering with stone fragments. Raising the dam's work is performed in
several layers, sprinkled with water around optimum moisture content and adequately
compacted with rollers.
2.5 SOIL VARIETIES FOR EMBANKMENTS
Fine-grained soils such as clays and silts with high water content may cause high pore
pressures to develop in an embankment dam. It is advisable to avoid clay as a
material for constructing an embankment dam if it has a liquid limit greater than 80.
Fine-grained soils are useful whose water contents are in the range of optimum
moisture content for compaction.
The reduction of water content of impermeable soils is generally difficult and
expensive. A borrow area may be rejected if the borrow area's impervious soils are 2–
5 % wet of optimum water content. Well-graded soils are generally given preference
to the soils of uniform particle sizes because the well-graded soils are stronger, less
compressible, and are less vulnerable to piping and erosion. The stones of sizes of
cobbles and boulders are removed from the embankment dam portion if their size is
larger than the layer's compacted thickness. Embankment soils may shrink upon
drying. Therefore, it is necessary to protect those shrinkable soils with non-shrinking
soils. The soils used in the core of embankment dams should have low permeability to
control the seepage rate through the embankments. The coefficient of permeability of
the core must be lower than 1×10 -5
cm/s. The shear strength of the shell material
must be sufficiently higher than that of the impervious core. Embankment
compressibility and foundation soils must be within tolerable limits to reduce the
dam's excessive settlement and deformation. The plasticity of embankment soils
17
should be high enough to curtail the possibility of cracks. The moisture content of
various zones of an embankment dam should be controlled close to the level of
optimum moisture content. The outer zone may be compacted at moisture content
below the optimum level, and the core is compacted at moisture higher than the
optimum one. It is a general recommendation that the content of particles smaller than
0.005 mm should be lower than 30% to maintain compaction quality control. It is
advisable to restrict the content of swelling and shrinking soils in embankment dams.
Generally, soils are considered heterogeneous material with the following properties:
o Soils exhibit stress-dependent stiffness and show nonlinear behavior.
o The stress-strain response of soils is dependent on confining pressure.
o Under loading conditions, the soils show elastoplastic behavior.
o Loose sands and normally consolidated clay show contractant behavior, dense
sands and overconsolidated clays exhibit dilatant response.
o It is generally reported that soils show different responses under primary
loading, unloading, and reloading conditions.
o Soils generally show anisotropic behavior.
2.6 CAUSES OF SLOPE FAILURE
Failure of a slope may occur due to changes in the properties of soils and loading
conditions. The fundamental condition for slope stability is that soils' shear strength
should be higher than the shear stresses that may cause failure. Therefore, it is
necessary to understand the factors that cause a decrease in shear strength and
increased shear stresses.
Factors causing a decrease in shear strength of a slope are [32]:
(i) Increase of pore pressures
Within slopes, a decrease of shear strength may be due to an increase in pore pressure
caused by the rise of groundwater level, heavy rainfall and seepage conditions.
Permeability of soils is the main factor by either increase and decrease of pore
18
pressures. In general, clayey soils have low permeability, but it has been observed that
such types of soils may show high permeability due to the presence of fissures,
cracks, and lenses of highly permeable materials. Consequently, pore pressures in clay
may sometimes change rapidly.
(ii) Cracking
There may be a possibility of the occurrence of cracks in the vicinity of the crest of a
slope. Since soil is weak in tension, these cracks may develop due to excessive tensile
stresses at the surface. With the development of cracks, the shear strength of the soil
is drastically decreased.
(iii) Swelling
Clayey soils may experience swelling if the water is added. The presence of water for
a long time and low confining pressure may cause swelling. It has been reported by
[33] that highway embankments of highly plastic clay have failed within a period of
10 to 20 years. The main reasons for the failure of such embankments were strength
loss, swelling and cracking.
(iv) Creep for constant loads
Continuous deformation may occur in highly plastic clays when acted by constant
loads. Failure of clays may occur under sustained loading, causing lower shear
stresses than the short-term strength. The creep behavior of clay may get intensified
under cyclic conditions of freezing-thawing and wetting-drying.
(v) Strain softening
Brittle soils show strain-softening behavior. As the strains are further increased
beyond the peak strength, a corresponding decrease occurs in the brittle soils' shear
resistance. Such type of response of brittle soils may facilitate the progressive failure
of slopes.
19
(vi) Cyclic loading
Cycling loading may occur due to an earthquake. As a result, loose saturated soils
become dense due to cyclic loading and pore pressure. Eventually, slope stability is
endangered with an increase of pore pressure.
Factors causing an increase in shear stress in a slope are [32]:
(i) Loads on a slope: If there are heavy loads on the crest of a slope, there will be
an increase in shear stress.
(ii) Presence of water in cracks of a slope: Sometimes, cracks may develop in a
slope. When water enters through cracks, it will make the cracks wide and
create increased pore pressure, which may increase shear stress and cause the
slope to destabilize.
(iii) The presence of water causes an increase in soil weight: When there is more
water content in a slope, there may be an increase in soil mass, which may
increase shear stress.
(iv) Excavation of a slope: If a slope is made steeper by excavation or erosion,
shear stress may increase.
(v) The decrease in water level: If the water level in any slope is rapidly
decreased, it may cause the slope's instability due to the removal of water
support. As a result, there will be an increase in shear stresses in the soil,
creating instability conditions.
2.7 CAUSES OF FAILURE OF EMBANKMENT DAMS
Embankment dams are constructed with compacted soil or small rock particles. These
dams are generally cheaper because locally available soils are utilized as the material
of construction. The stability of embankment dams depends upon the design and
material of construction [34]. To reduce the possibility of seepage, the central core of
embankment dams is constructed of fine-grained soil, i.e., clay and silt. Possible
failure modes of embankment dams could be due to hydraulic, seepage and structural
[35].
20
(i) Hydraulic failure: Hydraulic mode of failure of a dam may be due to wave
erosion, which may weaken soils and decrease the section of the dam, wind or
water erosion of the dam crest and downstream slope of the dam [36].
(ii) Seepage failure: Due to excessive seepage, there may be a possibility of slope
instability, uplift pressure, internal erosion, and foundation failure. Due to
extreme floods or storms, there may be chances of overtopping a dam [37].
Overtopping of dams may occur if the incoming flow is more than the
reservoir's storage capacity [38]. Overtopping of a dam may occur due to less
freeboard and excessive settlement of foundation/embankment soils. Due to
the low permeability of soils and placement water content during construction,
pore pressures may increase in the dam's embankment and foundation. If the
soils' permeability is less than 10-6
cm/s, the pore pressures do not dissipate as
rapidly as the construction work is carried out. If the pore pressures gradually
accumulate as the dam's successive layers are placed, it might create
conditions of low shear strength, which may lead to instability or failure of a
dam [39].
(iii) Structural failure: Some common structural failure signs are cracking,
excessive settlement, and slides [40]. This type of failure mode may occur due
to strong earthquakes [41] or erosive forces, which may weaken the
embankment and foundation soils. Besides, the roots of trees along the
embankment may weaken the soils and lead to the dam's instability. Saxena
and Sharma [14] have shown that 40 percent of the embankment dams have
failed due to overtopping.
For both the upstream and downstream slopes, the dam's safety factor must be
determined for the conditions considered to be most critical during the dam's life
span. It may be possible that the failure zone may occur within embankment soils or
may pass through both the dam and foundation. Two conditions are critical for an
upstream slope: EoC and AFR. For a downstream slope, EoC and steady-state
seepage conditions are essential stages. Slope stability is severely influenced by the
presence of pore water pressures in the dam body. In the normal routine, piezometers
21
are fixed in the dam body to record actual pore pressure generation to compare with
pore pressure [25]. Suppose the failure zone is likely to occur in the foundation, which
contains fissures and joints. In that case, there may be chances of progressive failure,
which may also occur due to difference in stress-strain characteristics of different
soils and non-uniformity of induced shear stresses.
Cracking of embankment dams is a severe problem, which may occur due to
differential settlement and compressibility characteristics of various soil zones.
Besides, there may be a likelihood of hydraulic fracturing of clay core under high
water pressures. Hydraulic fracturing may occur at those planes of the dam where
everyday stresses are lower than the water pressure. In the long-term scenario, clay
cores may settle concerning other zones in a dam. As a result, vertical stress in the
core may be decreased, leading to a hydraulic fracturing process [25].
In compacted embankment dams, strains develop due to differential settlement. As a
result, internal stress distribution changes and may increase stress at some portions
and reduce others. With the gradual increase of the reservoir's filling, the water
pressure exceeds the embankment stresses in the impervious zone and forms a
concentrated leak. With more deformation of the impermeable embankment material
due to the reservoir's filling, more cracks may occur. Eventually, the minor principal
stress becomes so low to create conditions to develop hydraulic fracturing [42].
In addition to the above-mentioned mode of failures, it has been reported that the
embankment dams may fail due to rapid drawdown.
2.7.1 Rapid Drawdown Mechanism in Embankment Dams
Sometimes due to an emergency, a reservoir may be lowered. In this situation, excess
pore pressure does not reduce as rapidly as the water is lowered, reducing the
stabilizing force. An embankment dam's upstream slope may become vulnerable to
failure due to the quick lowering of the water level [43]. Sometimes with the lowering
of the reservoir, instability of upstream embankment may not occur immediately.
22
Still, concerning time, seepage forces may develop, which may ultimately lead to
failure of the dam (Figure. 2.2).
When the water on the upstream side of the dam is removed, its effect as a stabilizer is
lost, and there will be an increase of shear stresses in the slope. If the drawdown rate
is fast, and the permeability of soils is low, then the undrained condition will increase
with excess pore pressure [44]. Even if water pressure is removed in unloading, there
may be a decrease in pore pressure, depending on whether soil shows contractant or
dilatant behavior [44]. It is difficult to correctly predict pore water pressures during
undrained loading or unloading conditions; therefore, the total stress method is
generally used [44].
Limit equilibrium analysis, along with undrained triaxial compression tests, has been
used to analyze dams' stability under drawdown conditions [45]. Apart from limit
equilibrium methods, the finite element method based on the total stress approach has
been used for computation of stability of Pilarcitos Dam under drawdown condition
[44]. In 1969, when the reservoir level of the Pilarcitos Dam was lowered by 10.7 m
in 43 days, the upstream slope suffered a rapid drawdown slide, which was about 5 m
deep and 28 m in length. Vandenberge [44] and Fredlund et al. [46] have also
conducted a useful stress analysis of the Pilarcitos Dam to simulate the stability of the
dam under drawdown condition (Figure. 2.3). The authors have also described the
failure mechanism of 60 feet high Walter Bouldin Dam under a rapid drawdown of 32
feet in 5.5 hours using effective stress analysis.
23
Figure. 2.2: Slope instability when reservoir is lowered rapidly [43].
Figure. 2.3: Decrease of safety factor under rapid drawdown
scenario of Pilarcitos Dam [46].
24
2.7.2 Factors Affecting Slope-Stability Due to Rapid Drawdown
When a reservoir is lowered, undrained conditions may develop with an increase in
excess pore water pressure, which may reduce concerning time. As a result, instability
of the upstream slope may occur, which may lead to failure. Several factors cause
instability of a slope during the drawdown process: drawdown rates, slope angles, soil
characteristics [47]. Factors such as geometry, soil characteristics, internal (pore
pressure) and external forces (surface water pressure) directly affect the slope
stability. The slope is affected by internal and external forces, and the water level
change due to partial or total submergence [48, 49]. Because of change in water level,
two types of pore pressures may develop inside a slope: (i) seepage pore pressure due
to transient flow, (ii) excess pore pressure due to loading. As the consolidation occurs,
excess pore pressures reduce with time [48]. Decrease of both the seepage pore
pressure and excess pore pressure depends on compressibility and permeability of
soils, and drawdown rate.
Stress-induced pore pressures mainly reduce during drawdown if the permeability of
soils is high. However, in low permeability soils, the dissipation rate of seepage pore
pressures and stress-induced pore pressures is not usually the same; therefore, the
soils behave as totally or partially undrained. If the external water level is dropped
quickly, it is called rapid drawdown without allowing proper time for drainage to
occur. As a result, soil resistance reduction will happen that may cause slope failure
[48, 50]. Such types of failures have occurred in natural and man-made slopes. Some
examples are Pilarcitos Dam, Walter Boudin Dam, and several river banks [45].
Therefore, it is necessary to analyze those slopes where there are chances of rapid
drawdown. Nowadays, stability analysis of slopes is commonly performed using
advanced numerical tools based on the finite element method, capable of considering
the nonlinear material response, complex boundary and loading conditions [49, 51,
52].
For slope stability under drawdown conditions, consideration is given to external
loads and seepage forces due to transient flow. The consolidation process can occur in
a slope during the drawdown, which depends on the rate and time of drawdown and
25
soils' permeability. During the drawdown, pore pressure changes occur, consisting of
static hydraulic pressure, seepage pressure, and excess pore pressure. These pore
pressures are not steady; they dissipate with time as the consolidation process occurs
[48]. The researchers [53-55] have developed charts based on limited state analysis to
compute safety factors under rapid drawdown conditions. The safety factor of a
partially submerged slope under rapid drawdown condition cannot be calculated using
charts developed by [53, 55]. On the contrary, such slopes' safety factor can be
computed by utilizing the finite element method [49].
2.8 PORE PRESSURE IN EMBANKMENT DAMS
The increase and reduction of excess pore pressures during construction and a long
time are governed by the consolidation process [56]. For large dams, high pore
pressures may occur within the core [57]. The safety of dams depends on how the
pore pressures develop during construction and decreases during a long time as the
consolidation process occurs [58].
In embankment dams, the pore pressures develop particularly in clay core as the
dam's raising is carried out. If the construction work of the dam is performed fast,
then the pore pressures will generate rapidly. The situation may get worse if the
reservoir is filled quickly. This type of scenario may create conditions that can lead to
hydraulic fracturing of the dam [59].The stability of embankment dams reduces with
the generation of pore pressure during construction. With the compaction process and
weight of the subsequent layers, the pore pressures get increased. In this way, the
shear strength and stability of the slope gets reduced. Concerning time, there may be
gradual dissipation of pore pressures, which may increase stability. For large
embankment dams, construction pore pressures are recorded during the raising of the
dam. The magnitude of pore water pressure depends on the dam's height, degree of
saturation, drainage system, and permeability of soils [60]. To determine the shear
strength of slope in terms of effective stress, pore pressures must be evaluated with
reliability. There are several sources of inaccuracy to estimate pore pressures in
embankment fill. When the reservoir level is increased, it causes more portions of the
26
soil to be saturated, increasing pore pressure and a decrease in shear strength and
slope stability [61].
Embankment slope instability and deformation can initiate with the increase in pore
pressures [62]. Deep cracks may develop due to long-term drought. During rainfall,
the developed cracks may enlarge and become channels for infiltration that may result
in internal erosion and reduced dam stability. Drought and flood cycles may cause
instability of embankment dams [63].
2.9 STABILITY CONDITIONS FOR EMBANKMENT DAMS
For the design and safe construction of embankment dams under different loading
conditions, slope stability is essential. The safety factor is evaluated with the help of
the limit equilibrium and finite element method. Slope stability depends upon the
shear strength of soils in embankment and foundation [64].
Under dry conditions, clay has high cohesion and friction angle as compared to the
wet state. Therefore, under dry conditions, embankment dams having central
impervious clay are satisfactory concerning slope stability. Several factors that affect
slope stability are characteristics of soils, side slopes of the embankment, seepage
forces, heavy surcharge, seismic load, excessive rainfall intensity. A natural stable
slope may fail due to the reduction of shear strength, an increase of external loads,
and a change of the slope's geometrical shape.
The stability of embankment slopes fluctuates with respect to time-dependent changes
in loads and shear strength of soils. Consequently, a slope's safety factor may decrease
or increase with respect to time-dependent material and loading conditions that occur
in a slope. Therefore, slope stability analyses are performed for different loading
conditions.
2.9.1 End of Construction and Long-time Stability
During the compaction process of fine-grained soils in the dam body, water is usually
sprinkled to facilitate compaction. As a result, the majority of the pores in the soil get
27
filled with water. Due to a load of incremental compacting layers and the rollers'
movement, there is an increase in pore pressure in those soils whose permeability is
low. The increase in pore pressure depends on several factors such as permeability
and compressibility of the soils, placement moisture content of layers, how fast the
dam is constructed, possible reduction of pore water due to drainage of soils.
The end of the dam construction may be considered a critical condition that may last
for a particular time because there is partial dissipation of pore pressures concerning
time. For embankment dams, a value of safety factor of 1.3 is considered satisfactory.
If there is uncertainty regarding various soil material properties, it is better to adopt a
higher safety factor value.
If the soils are of low permeability in an embankment dam, the dissipation of excess
pore pressures will take time. This implies that the soil would be in an undrained
condition at the end of the construction stage. For the calculation of stability, total
stress analysis will be utilized [25].
Figure. 2.4 (a) illustrates the increase in pore pressure during the construction stage of
a dam. Initially, the pore pressure is u0 at the start. The pore pressure then gradually
increases by an amount (Δu) during construction time (tc). After the end of the
construction stage, the pore pressures gradually dissipate as the soils are allowed for
consolidation to occur for a long time (t). Figure. 2.4 (b) shows that the safety factor
initially decreased at the EoC. The safety factor has increased due to the reduction of
excess pore pressures in a long time. This suggests that the critical stability condition
may occur at the end of a dam's construction stage.
28
Figure. 2.4 (a): Increase in pore pressure during construction
time and decrease during a long time (t) [25].
Figure. 2.4 (b): Change in safety factor during EoC
and long-time [25].
If an embankment foundation is rigid and robust, then a possible failure zone may
only initiate in embankment soils (Figure. 2.5). On the contrary, if the foundation is
weak, then the failure zone can develop both in the embankment and foundation
(Figure. 2.6). This implies that stability depends on both the shear strength of
embankment and foundation soils.
Figure. 2.5: Potential failure zone in a dam with rigid foundation [25].
29
Figure. 2.6: Potential failure zone extending in embankment
and foundation [25].
2.10 SLOPE STABILITY ANALYSIS METHODS
Two types of methods have been used for the calculation of slope stability of
embankments: the limit equilibrium method and finite element method:
(i) Limit Equilibrium Method (LEM)
The LEM is widely used to compute slope stability. Traditionally, slope stability was
performed using different methods based on conditions of equilibrium. Such
techniques consist of: the ordinary method of slices [65], Bishop's Modified Method
[66], force equilibrium methods [67], Janbu's generalized procedure of slices [68],
Morgenstern and Price's method [69] and Spencer's method [70]. One of the
disadvantages of these limit equilibrium methods is that the potential failure zone is
assumed in advance.
Several assumptions regarding normal and shear forces acting on various slices are
made in limit equilibrium methods of slope stability. This is due to several unknowns
as compared to the number of equilibrium equations. If the ordinary slices method is
used for slopes with high pore pressures, the calculated safety factor is 50% smaller
than the correct value. This is because only one condition of equilibrium is satisfied
[71]. Spencer method of slope stability is more accurate than the other limit
equilibrium methods [72].
30
The ordinary method of slices gives conservative values of safety factors in
comparison to other methods. The slopes designed on the standard method of slices
are always considered to be on the safer side. Some realistic assumptions of interslice
forces are made in methods like Ordinary Bishop's Method, Spencer's Method and
Morgenstern and Price as compared to the ordinary method of slices [73].
Morgenstern and Price method was extended by satisfying both moment and force
equilibrium conditions [68-70].
(ii) Finite Element Method (FEM)
With the advance of science and technology, the FEM has been extensively used in
geotechnical engineering to solve various problems [49]. In geotechnical engineering,
calculation of the stability of a slope plays a vital role. Very complex slope stability
problems have been solved with the help of FEM analysis. The stability of
embankment dams depends upon the geometry, properties of materials in
embankment and foundation. Besides, the stability of an embankment dam relies on
the level of the reservoir and its fluctuations. With the help of FEM settlement and
deformations, flow quantities can be determined for embankment dams. Generally,
two-dimensional FEM programs are used for calculation of slope stability of
embankments. At present, many geotechnical engineering software based on the FEM
is available commercially, which can be used to analyze dams' stability under various
operating conditions. The analysis results, such as settlement and deformation of an
embankment dam, mainly depend upon realistic input of material properties,
especially strength and stiffness parameters [74].
2.10.1 Advantages of FEM in Slope Stability Analysis
The following are the main advantages of FEM over the traditional limit equilibrium
methods of slope stability [49]:
(i) No assumptions are made for the position and shape of the failure zone. Failure
zone develops in those regions of the soil having lower shear strength than
applied shear stresses.
31
(ii) Equations of equilibrium and compatibility are satisfied.
(iii) No side forces are assumed because the failure zone is not divided into slices.
(iv) Realistic deformations of a soil mass could be computed if reliable parameters
of compressibility are utilized.
(v) FEM can capture progressive failure as well as overall shear failure.
2.11 MAIN STEPS INVOLVED IN FEM IN PLAXIS 2D
(i) Discretization of elements
The geometry of a geotechnical problem is divided into finite elements. These
elements are triangular in shape. There are two possibilities to select either 15 nodes
triangular element or 6-node triangular element, as shown in Figure. 2.7. There are 12
stress points and 3 stress points in 15 node elements and 6 node elements.
Figure. 2.7: 15 and 6- point nodal scheme and stress point scheme [75].
(ii) Approximation of primary (unknown) variables
Primary variables, such as displacement, is selected. The rules for variation of
primary variables over a finite element are mentioned.
32
(iii) Elemental equations
Element equations are derived and then combined into global equations.
(iv) Boundary conditions
Boundary conditions for displacement, loads and pore pressures are described. The
global equations are then modified with respect to boundary conditions.
(v) Solution of global equations
The global equations are obtained in large numbers as simultaneous equations. The
solution of the equation gives displacements at all nodes. Secondary quantities such as
stresses and strains are obtained from nodal displacements. The factors on which the
finite element analysis accuracy depends are the size of elements and the
displacement variation.
2.12 2D AND 3D MODELS
Generally, geotechnical problems are solved using three-dimensional numerical
techniques. 3D slope stability analysis is performed when there is a curved or
complicated geometry of an embankment dam [41]. Since the 3D problems involve a
huge geometry with a lot of a number of nodes. As a result, the solution of the 3D
problem using computer software takes substantial computational time. To reduce
computing time, the 3D problems are solved using plane strain and axisymmetric
conditions.
(i) Plane strain condition
For simplification, 3D geometry of geotechnical problems such as retaining walls,
continuous footings, and the stability of slopes, can be analyzed using plane strain
condition. It is assumed that the strain in the longitudinal direction is negligible
(Figure. 2.8). Therefore, a 3-dimensional geometrical problem can be reduced to a 2-
dimensional space. It is generally observed that in 2D situations, the factor of safety is
33
lower than 3D. So from a stability point of view, it is appropriate to use the 2D FE
method instead of 3D to save time [76].
Figure. 2.8: Plain strain condition [75].
(ii) Axisymmetric condition
Axisymmetric condition is adopted for circular structures having a radial cross-section
and loading on the central axis (Figure. 2.9). It is assumed that deformation and
stresses are similar in the radial direction. For geotechnical problems, which involve
axisymmetric condition, x-coordinate shows radius and y, and coordinate represents
the axial line of symmetry.
Figure. 2.9: Axisymmetric condition [75].
34
2.13 FACTOR OF SAFETY
Slope stability is usually determined with the help of the safety factor described as the
ratio of available shear strength of the soil to the shear strength required for
equilibrium. In the FE program PLAXIS 2D, the safety factor is defined using the
strength reduction technique. In this strength reduction technique, the tangent of the
friction angle and soils' cohesion is incrementally decreased until the soil mass's
failure. The safety factor is defined as follows [77]:
∑ =
Where is a safety factor, are friction angle, cohesion and undrained
shear strength of the soil.
2.13.1 Allowable Safety Factor
When the safety factor is greater than 1, it shows that resistive forces are more than
the driving forces. There are several uncertainties in determining material properties
with reliability; therefore, it is advisable to use safety factors greater than unity.
The slope stability of embankment dams is determined based on the factor of safety.
The following are the main points which provide basis why factory of safety is
necessary:
o Since the soil is heterogeneous material, there is less reliability of available
experimental techniques for reliable calculation of the strength properties of
soils
o There is some degree of uncertainty in the computation of pore pressures
within soils
o The magnitude of the safety factor gives a possible indication of the effect of a
dam's failure on losing lives and property.
o It depends on the height, storage capacity and foundation conditions of soils
o It also depends on how extensive the geotechnical investigation is performed
and how well the construction work is carried out.
35
The embankment dams' safety factor depends on the input values of shear strength
parameters of different soil types. It is indispensable that representative values of
shear strength parameters be obtained from field or laboratory tests. If the values of
strength parameters are overestimated, consequently, the safety factor can be
unrealistically higher. Allowable safety factors for various operating conditions of
embankment dams are mentioned in Table 2.3.
Table 2.3: Recommended values of safety factor [78].
Condition Safety factor Slope
End-of-construction 1.3 Upstream and Downstream
Long-term 1.5 Downstream
Rapid drawdown 1.1-1.3 Upstream
2.13.2 Importance of Parametric Analysis
A parametric analysis is carried out to investigate the effect of variation of specific
essential parameters on a study's results. For example, in geotechnical engineering, a
parametric analysis may be required for assessing the impact of strength parameters
(cohesion and friction angle of soils) on the overall slope stability of a dam. The
parametric analysis may also be required to assess the effect of stiffness properties on
the settlement response.
2.14 SHEAR STRENGTH
The shear strength of soils is internal resistance against the shear stresses along the
failure plane. The shear strength of soils depends on friction angle and applied normal
stresses. In addition to this, it also depends on the cementation of soils. The shear
strength analysis is conducted for two scenarios: drained and undrained conditions,
depending on soil loading and permeability. The drained condition develops under
loads applied very slowly, so excess pore pressures do not generate in the soil mass.
On the contrary, the undrained condition develops if the loading rate is fast, so there is
not enough time to reduce excess pore pressures [79].
36
2.14.1 Undrained Shear Strength
When soil is loaded to failure under undrained conditions, its corresponding strength
is called undrained shear strength. In the field, the undrained condition refers to loads,
which are applied faster than the soils' rate of drainage. On shearing, normally
consolidated soils exhibit increased pore pressure, and overconsolidated soils show
negative pore pressures under the undrained state. Undrained shear strength is
generally applicable to low permeability soils such as clay and silts [78].
Undrained shear strength is affected by soil type, moisture condition, and stress
history [80]. If the time is allowed for consolidation, the undrained shear strength of
clay during multi-stage loading increases [81].
It depends on the anisotropy of soils, stress history and strain rate [82]. The strength
of soils can be determined by laboratory tests and in-situ tests [82, 83]. It is to be
noted that there is some degree of disturbance observed in laboratory tests; therefore,
such type of tests do not represent actual in-situ conditions. In addition, laboratory
tests may be misleading because of the smaller size of the specimen that involves
fissures [84]. For undrained conditions, water remains in pores of soils. As a result,
with an increase in normal stress or shear stress, excess pore water pressure in fine-
grained soils increase. In practice, undrained shear strength is not determined as a
single value representative of soil conditions—the value of undrained shear strength
changes with stress direction [79].
2.14.2 Drained Shear Strength
Drained shear strength of soils shows the condition in which load is applied very
slowly so that there should not be an increase in excess pore pressures. In the field,
the drained condition refers to soil loading for a long term when pore pressures reach
equilibrium. Dense soils show dilatancy, and loose soils show contracting during
shearing under drained conditions. For conducting stability analysis of various
geotechnical structures, the shear strength is a major parameter. For reliable results of
stability analysis, correct shear strength values must be determined and given as input
37
to the finite element programs. In the case of non-cohesive soils (sands and gravels), it
is essential to assess correct values of effective friction angle from laboratory tests
[85].
Generally, the shear strength parameters of ideal soils, like, clay, and sand, are
presented in the literature [86]. It has been commonly observed in the field that a
variable amount of fines (silt and clay) is mixed with sandy soils. The load response
of sand-clay mixture soils may be lower than pure sands due to a change in shear
strength and stiffness properties. The sand-clay mixture soils may possess properties
which are intermediate between sand and clay [87].
The effect of fines content on the behavior of sandy soils was studied by many
researchers, e.g., [87-91]. In the above-mentioned studies, direct shear/triaxial tests on
different sand-clay mixtures were conducted. It was commonly observed that at a
particular confining pressure, there was a corresponding decrease in effective internal
angle of friction and an increase in effective cohesion by increasing clay content.
Furthermore, it was also observed that the plasticity index and compression index of
the sand-clay mixture increased with the addition of more percentage of clay.
Najjar et al. [85] report that the allowable bearing capacity of 1 to 4 m deep footings
on sand decreases due to the presence of clay content (Figure. 2.10). It can be
observed from the Figure. that there is a relatively decrease in the small percentage of
allowable bearing capacity when the clay content exceeds from 20% to 40%. Najjar,
Yaghi et al. [85] also mention that an infinite slope's safety factor at a given depth of
failure surface reduces with the clay content (Figure. 2.11).
38
Figure. 2.10: Effect of increased clay content on bearing
capacity of footings on sand [85].
Figure. 2.11: Influence of increase in clay on slope's
safety factor w.r.t depths of failure surface [85].
39
2.15 STAGED CONSTRUCTION OF EMBANKMENT DAMS
Staged construction is a technique in which step by step construction of various
geotechnical structures is analyzed using finite element programs. The construction of
dams is also modeled using staged construction techniques. In this approach, an
embankment dam can be raised gradually in layers using a specified period as it is
practiced on-site. One layer is placed during a specific period, and then the successive
layers are raised accordingly. If the permeability of soils involved in foundation and
embankment is low, there may be an increase in excess pore pressure, which can
decrease the dam's shear strength and overall stability. If the foundation of a dam is
not strong enough, then partial consolidation is allowed for some time so that the soils
can gain strength. In this way, with the finite element program, a designer can suggest
a proper time of raising and partial consolidation depending upon the strength and
permeability conditions of the foundation soils of a dam.
2.16 SETTLEMENT OF EMBANKMENT DAMS
Among all types of the number of earth and rockfill dams around the globe are more
than 85% of all the dam types. The excessive settlement is the main problem to which
these types of dams may be subjected. Therefore, it is essential to make a reliable
estimate of the settlement of such dams. Generally, maximum crest settlement of a
dam can be up to 2% of the height. The magnitude of a dam's settlement depends on
the self-weight of the embankment material and foundation geology [92].
The design engineer can predict a dam's behavior such as settlement, slope stability,
stress state and seepage based on geometry, soil properties involved in the dam and
foundation. Parameters such as settlement, stresses, and pore pressures are recorded
during construction and operation stages of a dam to determine whether the dam is
performing as per the predictions made during the design stage. The acceptable range
of behavior depends on the height of the dam and soil characteristics. A general rule is
that an embankment dam's post-construction settlement must not be greater than 1-2
% of the dam's height. After filling the reservoir, lateral deformation may occur at the
40
crest of the dam. If the loading condition is constant, then it is necessary that the rate
of settlement of a dam must decrease with respect to time [93].
If a rockfill dam is not compacted properly, then a settlement of 1 to 2 % of the dam's
height may occur. If the rockfill dams are well compacted, the probable settlement
could be lower from 0.5 to 0.7 % of the dam height [94]. When the settlement of an
embankment dam is analyzed, it must include the soft bedrock's depth. This is
because the total settlement of the embankment dam depends on the foundation
settlement [95].
The density and MoE of soils directly affect the deformation of the dam. This
indicates that the deformation of a dam will be more if the density and MoE of
different soil types are lower. A major portion of deformation occurs during the
construction process of a dam. Usually, in embankment dams, the core is considered
soft material as compared to the shell. When there is a significant settlement in a core,
the shells' load transfer occurs [96].
If the dams are constructed on foundations comprising thick layers of soft clay, there
can be issues related to stability and excessive deformations. Such types of problems
can lead to the development of cracks, fractures or collapse. Such types of issues can
be solved using staged construction in which partial consolidation occurs during the
raising process. The strength of soft clay soils depends on the embankment load, the
speed with which construction activities are performed, and waiting time between
different construction stages [3].
Ozcoban et al. [3] describe the staged construction of the Alibey dam (Turkey) on a
clay's thick deposit. Various instruments were installed on the site to record
settlement, deformation, and excess pore pressures in this dam. Based on the field
data, the dam's construction rate was adjusted, and the dam was safely constructed on
a soft clay foundation. The maximum recorded settlement at the main dam section
was 4.34 m that occurred during a time period of 24 years.
41
Javanmard et al. [97] conducted a settlement analysis of earth-fill dam Taham (Iran),
123 m high. The settlement computed using three different software packages
PLAXIS 2D, FLAC, and Geostudio was about 0.93 %, 0.97%, and 1.38% of the
dam's height, respectively.
Keyvanipour et al. [98] have analyzed the settlement response of a 35.5 m high bar
embankment dam (Iran) and compared the results with the field data. The authors
conclude that the settlement response predicted with the HSM was in close agreement
with the field data as compared to the Mohr-Coulomb Model (MCM). Rashidi and
Haeri [99] have conducted a numerical analysis of the Gavoshan rockfill dam
settlement, which is 123 m high. The calculated and measured settlement was about
2.2% of the height of the dam. The authors further describe that 88% of the dam's
total settlement occurred during the construction period.
According to available literature [93, 94, 100], an acceptable settlement of
embankment dams ranges from 1 to 2% of the dam height. Differential settlement of
embankment dams is also associated with the arching phenomenon, the details of
which are presented below:
2.17 ARCHING OF EMBANKMENT DAMS
The stability of embankment dams is related to stress conditions in the clay core as
compared to shell and other zones [101]. In zoned embankment dams, differential
settlement may occur due to differences in stiffness of clay core and shell soils [102].
As a result, vertical and horizontal stress in the core decreases and the load is
transferred to the shell. This situation may cause horizontal cracking of the core,
which is called arching of embankment dams. The transfer of load from the core to
adjacent soils is due to differences in stiffness of different soils [103, 104]. Compared
to the shell, the core is a soft material; therefore, the load is transferred from the core
to the shell. Due to this load transfer, pore water pressure in the core exceeds than the
total stresses [104, 105]. As a result, excessive settlement may occur in the core as
compared to the shell [101].
42
Arching of dams may occur during construction and may continue during operation.
When the reservoir is filled, water may enter the core and decrease effective stresses.
The combined action of arching and reducing effective stresses could pave the way
for hydraulic fracturing in the dam's clay core [102]. Concentrated leaks may occur
through the cracks in the core. Eventually, there may be gradual erosion of the core,
which may lead to the collapse of the dam [106]. For about 30-50% of embankment
dams' failure, progressive erosion is the main cause [1]. The failure of Teton Dam
occurred in the USA on 5 June 1976, within a few hours of filling the reservoir, which
is possibly considered to be due to leakage and hydraulic fracturing [107].
The arching ratio is defined as follows [108]:
Where arching ratio, total vertical stress in the depth of z in a
homogeneous dam and total vertical stress in depth z in a zoned embankment
dam.
Talebi et al. [106] studied various geometrical factors on a zoned embankment dam's
arching ratio. The authors conclude that a zoned embankment dam's side slopes have
a very negligible influence on the arching ratio (Figure. 2.12). The authors also
describe that if the side slopes of the core are flatter, and then there is a decrease in
the arching effect (Figure. 2.13). This implies that if the cores are thinner and steeper,
there may be chances for more differential settlement between core and shell. The
authors further mention that the arching effect decreases with the increase in filter
thickness (Figure. 2.14). The authors noted that the lower is the shell/core stiffness
ratio; the lower is the effect of arching (Figure. 2.15).
The authors describe that the inclined cores show a less arching effect than the
vertical cores of zoned embankment dams (Figure. 2.16). The authors conclude that
foundation compressibility has a negligible influence on the arching effect of a zoned
embankment dam (Figure. 2.17).
43
Figure. 2.12: Effect of side slopes on arching ratio
of clay core and shell [106].
Figure. 2.13: Effect of side slopes of core on arching ratio [106].
44
Figure. 2.14: Effect of increase of filter thickness on
the arching ratio [106].
Figure. 2.15: Effect of shell core stiffness ratio
with arching [106].
45
Figure. 2.16: Influence of inclined core on arching
Effect in a zoned embankment dam [106].
Figure. 2.17: Effect of MoE of foundation on arching
effect in a zoned embankment dam [106].
46
2.18 FOUNDATION TREATMENT METHODS
If the foundation of an embankment dam is on soft or weak soil/rock, there may be a
possibility of a large settlement and excessive deformations, leading to failure.
Grouting is a technique in which cement slurry is injected under pressure into the
underlying rock to fill joints, fractures, open bedding planes or other openings.
Grouting of dam foundations is carried out to decrease leaks, to reduce uplift pressure
and seepage. Generally, grouting is performed when a dam is newly constructed.
However, grouting could also be used for old dams as a remedial technique. Under a
dam foundation, grout is injected with high pressure to seal the joints to reduce
seepage.
Jafarzadeh and Garakani [109] have conducted the numerical analysis for the
effectiveness of consolidation grouting as foundation treatment of a 58 m high
embankment dam, Gerdebin (Iran). The authors conclude that with the application of
consolidation grouting below the clay core, there was a reduction of about 28% of the
dam's vertical deformation.
The main types of foundation grouting for embankment dams are curtain grouting and
blanket grouting. For new dams, curtain grouting is commonly used to reduce
seepage. In this method, boreholes are drilled into the bedrock parallel to the dam
axis. Blanket grouting is used at those sites where the rock is fractured at the
foundation contact. The purpose of blanket grouting is to make the foundation strong,
reduce seepage from the foundation bedrock and to decrease the possibility of internal
erosion of embankment and foundation materials. Commonly grout cap is provided at
about 3 feet wide and 3 – 8 feet deep.
Due to the presence of joints, cavities, interconnected fractures, and fissures in
bedrock, its permeability may be high. Therefore, grouting is provided to reduce
foundation permeability and decrease the risk of uplift pressure. If the permeability of
rocks is less than 1x10-5
cm/s, there may be less need for grouting.
47
2.18.1 Blanket Grouting/Consolidation Grouting
Blanket grouting is also termed as consolidation grouting. In this type of grouting, the
slurry is injected with low pressure to seal shallow bedrock below the embankment
dam's core. As a result, the bedrock gets harder. Blanket grouting is used in those
areas of the dam where the bedrock is highly fractured, having high permeability,
which may cause excessive seepage and internal erosion when the reservoir is filled.
Blanket grouting is provided only below the impervious core section of the dam. The
depth of blanket grouts is about 20 to 30 feet. The 20 feet depth of a blanket grout
may be satisfactory for a dam 100 feet high. If the dam's height is greater than 100
feet, then 30 feet blanket grout holes are drilled. Besides, the depth of blanket grout
depends on the geological conditions of the site. Twenty feet deep blanket grouts are
filled in one stage. On the contrary, 30 feet deep holes are grouted in two steps: 0 to
10 feet, and 10 to 30 feet.
2.18.2 Curtain Grouting
In curtain grouting, a zone of low permeability is constructed to a certain depth on the
dam's upstream side. The hazards such as piping and uplift pressure can be controlled
with the help of grout curtains in addition to the downstream drainage system.
If there is a possibility of much seepage under the dam's foundation, then curtain
grouts are used. These grouts are not thoroughly watertight. Such type of grouts just
reduces the quantity of seepage water. Usually, some arrangement of drainage is made
to collect the seeping water downstream of the curtain. The purpose of the grout
curtains is to reduce the quantity of seepage so that there is no loss of stored water and
no risk of the erosion of the foundation soils of the dam [110]. Reclamation
recommends grouting holes up to a depth of 0.5 to 1 times the reservoir water level.
2.19 RAINFALL EFFECT ON SLOPE STABILITY
When rainfall water enters through embankments, there will be a decrease in shear
strength due to excessive generation of pore pressures. As a result, there may be a
48
possibility of sliding off the slope [111]. With the entry of rainwater into the slopes,
unit weight, and moisture, the fill material's content can increase. There is a high
variation of rainfall rate with respect to time and space.
The following parameters have an effect on the slope stability due to rainfall:
(i) Permeability of soil along the depth of the embankment
(ii) Rate of infiltration of rainfall through the embankment
(iii) Depth of water that penetrates through the crest and slopes
The following are the conditions on which rainwater infiltration through the soil
depends: properties of fill soil such as permeability, moisture content, and porosity
[112, 113]. Initially, the surface layers of the embankment get saturated due to
rainfall. Pore water pressure develops in newly saturated layers. This process
continues until the rainwater reaches the phreatic level (Figure. 2.18) [42, 61].
Figure. 2.18: Rainwater infiltration through a dam [61].
If the intensity of rainfall is very low, the rainwater is absorbed by the soils. The rate
of infiltration depends upon the type of soils [114]. The depth of wetted soil depends
upon the intensity of rainfall and the rate of infiltration. For a partially saturated fill,
initially, the rate of infiltration is high. The rate of infiltration slowly decreases as the
saturation of fill soil occurs. Figure. 2.19 shows how the embankment soils are
saturated due to rainfall.
49
Figure. 2.19: Rainfall infiltration through dam [61].
In the unsaturated zones of embankment dams, there will be a decrease in suction,
increased permeability, and moisture content due to rainwater penetration. The
penetration of rainwater reduces the negative pore water pressure in slopes where the
groundwater level is initially low. This may cause a decrease in the safety factor.
Slope stability is affected by the intensity of rainfall and the position of the initial
groundwater table. If the initial water level in a slope is low, then there will be less
impact of rains on the slope's stability. High-intensity rainfalls may increase the water
level and decrease the negative pore pressure of slopes [115].
Several slope failures are triggered due to rainfall. In a conventional slope stability
analysis, the influence of rain is simulated by increasing the groundwater level or
changing water flow patterns. Several shallow failures of slopes suggest that there
was not any indication of an increase in water level in those slopes [116]. The main
reasons for failures are decreased shear strength and reduction of suction [117, 118].
Therefore, conventional techniques cannot perform a proper analysis of such failures
[119, 120].
Changes in pore water pressure and seepage forces due to rainfall are responsible for
the failure of slopes. Two different types of failure mechanisms are observed for
rainfall-induced failures.
(i) In the low area of the slope, there may be excessive development of positive
pore pressure. Movement of the soil along the sliding surface may decrease
50
effective stress to such a magnitude near zero, causing the soil mass to liquefy
[121].
(ii) The soil is in unsaturated state, the slope failure occurs due to a reduction in
shear strength, which also decreases soil suction [116, 117].
If the rainfall is more intense, a significant decrease of negative pore pressure is likely
to occur in a slope and an associated rise in the groundwater table [122, 123]. The
increase in water pressure and a decrease in suction may reduce shear strength, which
may cause the slope's failure. Depending upon the intensity of rainfall and the soil
type, the failures may be shallow and deep-seated slips. The safety factor of a slope
decreases if the infiltration of rainwater increases with time.
The pattern of the rise of water in a slope and steady-state pore pressure depends on
hydraulic features of the soil in the slope. If the initial volumetric moisture content of
a slope is high before rainfall, pore pressure will quickly increase if the rainwater
infiltrates in its soil.
Rainfall of high intensity and enough duration may cause the failure of slopes having
low permeability. On the contrary, a rainfall of high intensity for a shorter period may
cause a slope's inability to have high permeability. In coarse-grained soils, the
infiltration rate will be high, causing an increase in pore water pressure. As a result,
seepage forces may develop within the slope, which may ultimately lead to failure.
The infiltration rate is low in fine-grained soils; therefore, there is less development of
positive pore pressures. In this type of situation, loss of suction occurs, which may
cause a reduction in shear strength.
A safety factor of slopes decreases as the amount of infiltrated rainwater increases.
Rainfall duration, the permeability of the soil, and the slope's geometry have the main
influence on slopes' instability. If the soil slopes are highly permeable (10-4
m/s), a
small amount of rainfall is not likely to cause slope failure. If the soil slope's
permeability is very low (10-6
to 10-7
m/s), there may not be a significant pore
pressure change during the rainfall. After the wet period, there may be an increase in
pore pressures.
51
Rainwater can infiltrate into embankment dams. The stability of dams can be affected
depending upon the duration and intensity of the rainfall. Due to rainfall, the phreatic
level of embankment dams may change and create high pore pressures. Therefore, it
is essential to evaluate dams' stability when subjected to rainfall during full reservoir
conditions. Due to rainfall infiltration, pore pressures increase, which may add to the
seepage, resulting in the dam's instability. As compared to the intensity, the duration
of the rainfall is vital for the dams' stability. If the rain is of a long duration, there is
an increase in pore pressure and decreased safety.
2.20 TYPES OF EMBANKMENT DAMS ANALYSIS
The following types of analysis are performed for embankment dams: Consolidation
and settlement and stability analysis.
(i) Consolidation and settlement
Cohesive soils such as clay and silt have very poor permeability. These soils are
compacted at optimum moisture content. The pores of such soils are filled with water
due to impermeable nature. As the soil layers are successively placed and compacted
until the dam is raised to a final height, there is a gradual increase in pore pressures.
This is because the rate of loading in such soils is fast as compared to the rate of
dissipation of excess pore pressures. On the contrary, if enough time is allowed, there
may be a decrease in excess pore pressures due to consolidation. In addition to
deformation, in finite element programs, analysis of generation and dissipation of
pore pressures is also carried out [124].
Soils (in particular clay) generally settle slowly at a variable rate. The clays'
settlement rate depends upon how fast the excess pore pressure is generated and
dissipated due to the consolidation process. The development of excess pore pressures
in an embankment dam depends upon the placement water content during the
compaction process, weight of soil layers, rate of raising, the period over which no
construction work is carried out, permeability of core fill materials, and drainage
layers. The shear strength of various soils relies upon consolidation features of how
52
fast excess pore pressures are reduced. If the foundation is composed of low
permeability soils, pore pressure will increase, and the water will carry the loads. If
the foundation of a dam has low shear strength, then it is advisable to decrease the
loading rate, make the slopes flatter, or expedite the consolidation process by applying
drains. Therefore, consolidation analysis of embankment dams must be carried out to
account for the time-dependent variation of excess pore pressures in the dam body. As
the consolidation process occurs, the excess pore pressures reduce with time and soil
gets compressed under the applied loads.
The fine-grained soils' settlement comprises three phases: immediate, primary
consolidation, and secondary compression. Immediate settlement occurs with the
application of load. The consolidation settlement occurs due to the reduction of excess
pore pressures with time. Permeability of the soils controls the time-dependent
consolidation settlement process of the fine-grained soils. The soils' secondary
compression settlement is due to clays' creep behaviour with respect to time under
constant loading conditions. As the coarse-grained soils' drainage process is speedy,
the only immediate settlement is of concern in such soil types.
(ii) Stability analysis
The stability of embankment dams is normally checked for the following critical
conditions:
(a) Slope stability after the end of construction
(b) Slope stability of upstream side when the reservoir is filled
(c) Slope stability of the upstream side during the rapid drawdown of the reservoir
when it is at its full supply level
(d) Slope stability of both the upstream and downstream sides when an earthquake
shakes the embankment.
2.21 SOIL MODELS
Previously it was considered that soils exhibit elastic behavior. Over time, plasticity
theory was applied to soils, and it is now found that soils show elastoplastic behavior.
53
The stress-strain curve of a metal bar tested in tension is presented in Figure. 2.20
[125]. Up to OA, there is a linear relationship between stress and strain. If the bar is
unloaded at point A, then it will follow the same route during unloading as it was
during loading. If the bar's loading is continued after reaching point A, then stress-
strain relation is non-linear, but unloading is reversible. After the loading crosses
point B, the response of the bar changes from elastic to elastoplastic. Therefore, point
B is termed as the yield point of the bar. If the bar's loading is continued at point C,
and then the load is removed, the stress strain curve follows the path CD. It can be
noticed that even there was unloading, but at this stage, the permanent strain has
occurred in the bar, which is denoted by CD. The permanent strain in the metal
resulting from loading beyond the yield point is called plastic strain.
The behavior of the bar is considered as elastic up to point B. When the bar is loaded
up to point C, the total strain OE consists of two parts: (i) the elastic strain DE which
can be recovered if the bar is unloaded, (ii) the plastic strain OD, which is
unrecoverable. The slope of the unloading line CD is identical to elastic loading
behavior represented by OA.
If the bar is again reloaded at point D, it will follow the same unloading path up to
point C, termed the bar's new yield point. If the bar is further loaded beyond point C,
it will follow the original stress strain curve until failure occurs (point F). At point F,
the bar attains its ultimate tensile strength. If the yield point of a material increases,
then it is said that the material shows hardening behavior.
It is well known that soils show more complex behavior. The elastoplasticity
framework shall provide a reasonable basis for developing new mathematical models
that can realistically capture the soil behavior.
54
Figure. 2.20: Stress-strain curve of metals [125].
2.21.1 Idealizations of Plastic Behavior
Various attempts have been made to identify how metals' plasticity could be related to
soils' plastic behavior. It was assumed that essential properties of soils be captured,
and less critical features can be neglected. Elastic-perfectly plastic response is
presented in Figure. 2.21(a). The stress-strain graph is initially linear and elastic, and
with further loading, the material yields and deforms at constant yield stress by
showing no strain hardening. The elastic-linear-strain hardening plastic response of a
material is illustrated in (Figure. 2.21 b). After yielding the material, there is a
reduction in the linear stress-strain curve slope compared to the initial response.
(Figure. 2.21 c) presents the rigid plastic response of a material.
It is essential to determine how the idealizations, as mentioned above of plastic
behavior, could be applied to soils, showing complex stress-strain response.
55
Figure. 2.21 (a): Elastic and plastic behavior of material [125].
Figure. 2.21 (b): Elastic-linear-strain hardening plastic
response of material [125].
56
Figure. 2.21 (c): Rigid plastic response of material [125].
Depending upon the level of confining pressure, soils show compression, dilation, and
deformation at constant volume. Loose sands and lightly over-consolidated clay
exhibit compression (Figure. 2.22 a). On the contrary, dense to medium-dense sands
and heavily over-consolidated clays shown dilation (Figure. 2.22 b).
Figure. 2.22 (a): Lightly over-consolidated clays and loose sands [125].
57
Figure. 2.22 (b): Over-consolidated clays and dense sands [125].
2.21.2 Linear Elastic Model
This model follows Hooke’s law of isotropic elasticity. There are two fundamental
parameters of this model: (i) Young’s modulus of elasticity (E) and (ii) Poisson’s ratio
(v). This model is not reliable for soils because soils show an elastoplastic response.
This model represents the behavior of concrete and intact rocks. The essential
properties of soil, such as non-linearity and irreversibility, are not captured by the
Linear Elastic model. This model is represented by the following relationship [75]:
[
]
[
]
[
]
With the use of Hooke’s Law, following relationships between Young’s modulus (E),
shear modulus (G), the bulk modulus (K), and Oedometer modulus can be
obtained as follows:
58
2.21.3 Mohr-Coulomb Model (MCM)
This model shows the linear elastic, perfectly plastic behavior (Figure. 2.23). The
linear elastic part of the MC model is based on Hooke’s law of isotropic elasticity.
The perfectly plastic section of the model is based on the Mohr-Coulomb failure
criterion. This model is widely used for the calculation of the stability of slopes. There
are five material parameters which are given in Table 2.4. Most of the soil constitutive
models utilize the Mohr-Coulomb failure criterion. The MCM is not capable of
simulating the softening and effects of changes in density [126]. Geotechnical
practitioners and academics have widely used MCM for sand behavior [127].
Table 2.4: Parameters of Mohr-Coulomb Model [75].
Symbol Description Unit
E Modulus of elasticity [kN/m2]
v Poisson’s ratio [-]
c Cohesion [kN/m2]
Angle of internal friction [o]
Angle of dilatancy [o]
Figure. 2.23: Stress-strain behavior of an elastic,
perfectly plastic model [75].
59
2.21.4 Mohr-Coulomb Model Parameters
(i) Modulus of Elasticity (MoE)
Refer to the drained triaxial test (Figure. 2.24). The initial slope of the stress-strain
curve is represented by ) (tangent modulus) and illustrates the secant modulus
at half of the strength. Generally, (E50) is used as a MoE for loading conditions when
MCM is used. If unloading conditions are encountered, such as excavations and
tunneling, then the unload-reload modulus ( ) is used in place of .
If the confining pressure of soils is increased, there will be an increase in both first
loading modulus ( ) and unload-reload modulus ( ). This indicates that there will
be more stiffness for deep layers of soils than layers lying at the surface. As compared
to primary loading, there will be higher stiffness for unloading and re-loading
conditions. Soils show more stiffness during shearing in comparison to drained
compression. When a constant value of stiffness is used in a calculation, it should be
carefully selected to match stress intensity at a particular depth.
Figure. 2.24: Drained triaxial test results for [75].
60
(ii) Dilatancy angle
The heavily over-consolidated clays show dilatancy [128], whereas normally
consolidated soils' dilatancy is almost negligible. Density and friction angle are the
parameters on which the dilatancy of sands depends. The angle of dilatancy of soils is
lesser as compared to the angle of internal friction. The dilatancy angle of sands is
related to the friction angle as follows:
The dilatancy angle is almost zero if the value of the friction angle of sands is less
than 300. In drained condition, if the value of the dilatancy angle is positive, it
indicates that dilation will continue until the occurrence of shear deformation. This is
contrary to the behavior of soils in which shear deformation occurs at constant
volume. If the dilatancy angle's value is positive in undrained conditions, it will
generate tensile pore pressures with constant volume. Consequently, there will be an
overestimation of strength under undrained conditions. The procedure of finding of
dilatancy angle of soils is illustrated in Figure. 2.25.
Figure. 2.25: Dilatancy angle using the stress-strain curve
obtained from triaxial test [128].
61
(iii) Poisson’s ratio
It is the ratio of lateral strain to axial strain. For soils, it may be obtained from the
triaxial test under axial loading conditions. [129, 130] describe that the range of
Poisson’s ratio for granular soils is from 0.18 to 0.33. Newcomb and Birgisson [131]
mention that the Poisson’s ratio for loose and dense sands is in the range of 0.2- 0.4
and 0.3-0.45, respectively. Marchenko et al. [132] mention that Poisson’s ratio of clay
ranges from 0.2-0.45. The Poisson’s ratio of fine-grained soils is found to be 0.3 and
0.5.
(iv) Friction angle and cohesion of soils
The friction angle and cohesion of soils are termed as shear strength parameters.
Cohesion is a force that holds together particles of cohesive soils (clay and silt). The
strength parameters, friction angle and cohesion of various soils like clay, silt and
sand can be obtained from the shear box (Figure. 2.26) or triaxial test (Figure. 2.27).
The values of frictions and cohesion are generally affected by initial moisture
content, drainage, and saturation conditions. Compaction is the technique that
improves the shear strength of soils [133].
Figure. 2.26: Determination of cohesion and friction angle [134].
62
Figure. 2.27: CD triaxial compression tests on three
overconsolidated clays [135].
(v) Permeability of soils
Permeability refers to the ease with water that can flow through soils. Soils are porous
materials. Water can flow between voids of soils. Permeability is an important
physical property that controls the flow of water in soils. The rate of flow of water
through soils depends upon particle sizes and structural arrangement of particles.
Higher will be the flow of water if the particles are larger in size with more void
spaces. The flow of water will be lower if the particles are denser. Generally, a soil
becomes weaker if the water content is higher. The response of soils to loading
conditions depends upon the permeability of soils. The settlement of soils depends on
permeability and decrease of excess pore pressure.
Due to the random arrangement of solid particles in a soil mass, there may be
variation in permeability from point to point. For realistic determination of
permeability in the laboratory, least disturbed samples should be used. It is common
to realize that some disturbance to soil samples is expected to occur when the samples
are extracted, transported, and prepared in the laboratory. In addition, the value of
permeability depends on the sample size and soil structure. It has been observed that
63
most of the soils show anisotropic behavior in permeability in horizontal and vertical
directions. Generally, horizontal permeability is more as compared to the vertical
direction. For laboratory tests, the flow of water is in vertical direction. It can be
interpreted that the measured value of laboratory permeability is lower than the in-situ
permeability of the soil. In the laboratory, the permeability of a small volume of soil
is determined. Sometimes, disturbed soil samples are also used to find out
permeability. The permeability of a soil sample will be highly misleading if the
structure of the soil is altered. Therefore, for disturbed soil samples, there may be
chances that the value of soil permeability could not match with in situ soil
conditions. Typical values of permeability of soils are shown in Table 2.5 [136].
Table 2.5: Typical values soil permeability [136].
Permeability (cm/sec) Type of soils
Greater than 10-1
Coarse gravel
10-1
-10-3
Sand, fine sand
10-3
-10-5
Silty sand
10-5
-10-7
Silt
Less than 10-7
Clay
(vi) Unit weight
The unit weight of soils depends on material type and its density [135]. In zoned
embankment dams, generally, there are various soils having different unit weights.
The permeability of soils depends on unit weight. The higher the unit weight of a soil,
the lesser the voids will be the permeability [137].
2.21.5 Application of Mohr Coulomb Model
The MCM has been extensively used in the computation of slope stability of several
types of embankment dams.
Rabie [138] concluded that MCM can be used to computation slope stability under
heavy rainfall conditions. During rainfall, slope stability is affected by the reduction
64
of shear strength, the increase of bulk density, the increase of pore pressure and the
action of seepage force.
Wulandari and Tjandra [139] concluded that the MCM can be used to computation
safety factors for geotextile reinforced road embankment. Due to an increase in traffic
volume, road embankments on soft soils have been constructed. Such types of
embankments show large settlement and slope instability problems. To achieve
stability of a road embankment on soft soil, geotextile reinforcement is provided.
Singh et al. [140] conclude that the stability of rock slopes in road cut sections can be
evaluated with MCM's help in the landslide region.
Deliveris et al. [141] conducted a stability analysis of a mine waste dam using three
software, i.e., Plaxis 2D, FLAC and Phase2. The analysis was carried out under
drained and plain strain conditions. They determined that MCM can be utilized for the
safety evaluation of mine waste dams. Ouadif et al. [142] conclude that highway fill
materials' stability can be evaluated using the MCM. The authors analyzed the
stability of an embankment under wet and dry conditions, dynamic loads and
reinforcement. Ghanizadeh and Ghanizadeh [143] have used MCM for embankment
dams that involve soils which consisted of sand and gravel. Afiri and Gabi [144] have
used MCM for the stability of excavations, embankment dams and landfills. Vekli et
al. [145] conclude that the safety of slopes stabilized with stone columns can be
calculated by using MCM. Gaber et al. [146] conclude that the stability of reinforced
embankments on soft clay foundations can be modeled using MCM.
2.21.6 Hardening Soil Model (HSM)
The HSM for soils was developed based on the theory of plasticity [147]. This model
captures the response of clays, silts, gravel and sand [75]. In this model, vertical strain
and deviatoric stress are associated with a hyperbolic relationship (Figure. 2.28)
during primary loading. The constitutive equations of this model are based on a
standard drained triaxial test.
for q qf 2.8
65
Where and are the major and minor principal stresses, is Young’s modulus
at half of the shear strength, is ultimate deviatoric stress, is a quantity of
deviatoric stress, is the failure ratio (0.9).
For primary loading, the stress-strain relationship is nonlinear. Primary loading
stiffness is:
(
)
where is primary loading stiffness modulus, which depends on confining
pressure,
is stiffness modulus associated with confining pressure , m is the
stress dependency power. The value of power m is 1 for soft clays.
Janbu [148] describes that the value of m for Norwegian silts and sands is 0.5. Von
Soos [149] mention that the value of m can be in the range of 0.5 to 1.
For most of the soil conditions, the value of
Confining pressure-dependent stiffness modulus for unloading and reloading is:
(
)
is reference Young’s modulus for unloading and reloading.
66
Figure. 2.28: Hyperbolic response in primary loading
for triaxial test [147].
(
)
is tangent stiffness modulus than that can be taken from the Oedometer test
is tangent stiffness at vertical reference stress (100 kPa), the procedure of its
determination is illustrated in Figure. 2.29. The parameters of the HSM are
summarized in Table 2.6.
67
Table 2.6: HSM parameters [75].
Symbol Description Unit
Cohesion [kN/m2]
Friction angle [0]
Dilatancy angle [0]
Secant stiffness [kN/m2]
Tangent stiffness [kN/m2]
Unloading/reloading stiffness [kN/m2]
Power [-]
Poisson’s ratio [-]
Reference stiffness [-]
Normal consolidation [kN/m
2]
Failure ratio [-]
Figure. 2.29: Determination of
from oedometer test [75].
The HSM can capture the real response of soft and dense soils. For example, when
applied to soft soils and hard soils under undrained loading conditions, the HSM
decreases and increases mean effective stress [150]. With HSM's use, realistic
predictions about displacement and failure of various geotechnical structures can be
made [150]. The HSM limitation is that neither time-dependent behavior (creep) of
clay nor anisotropic strength and stiffness can be modeled. The HSM is also not
suitable for dynamic loading conditions [150].
68
2.21.7 Application of HSM
The HSM has been widely used for several geotechnical problems. A summary of the
main applications of the HSM is presented below.
Celik [151] has used HSM for the estimation of the settlement of tunnels. The author
concludes that deformations were estimated with reliability using the HSM compared
to the MCM. Teo and Wong [152] have analyzed the settlement of deep excavation
using HSM. The authors conclude that a realistic settlement of deep excavations can
be estimated using HSM compared to MCM. The soil behavior in deep excavations
simulated with the MCM can be unrealistic. The wall deflection and ground
movement calculated with the HSM is comparable with the measured data.
Pramthawee et al. [153] analyzed a rockfill dam. They conclude that the HSM can
give a realistic estimate of the deformation response of rockfill material. Ozkuzukiran
et al. [154] have analyzed settlement behavior of 133 m high concrete faced rockfill
dam using HSM. The authors conclude that the settlement of the rockfill dam was
consistent with field data.
In the HSM, the Mohr-Coulomb failure criterion is used. Therefore, the values of
safety factors computed with the HSM and MCM are almost the same [155]. Dickin
and Laman [156] have conducted a numerical analysis of strip anchors' uplift
response in cohesionless soil. They conclude that the HSM is capable of capturing
well the pre-peak uplift behavior of strip anchors in such soil types. Locat et al. [157]
have modeled landslide in Quebec, Canada, using HSM. The authors concluded that
the HSM captured the initial state of stress before failure due to land sliding. Singh
and Babu [158] concluded that the HSM could be applied to soil nail walls
constructed in soft soils if essential structures are adjacent. Phien-Wej et al. [159]
conclude that the deformation behavior of diaphragm walls constructed on soft soils
be modeled with HSM. Tschernutter and Kainrath [160] have used HSM for
numerical analysis of central reinforced concrete core embankment dam. The authors
conclude that the stress and deformation analysis of such type of dams can be
performed using HSM for embankment materials and linear elastic model for
concrete. Vahdati [161] have utilized HSM for computation of displacements in
69
earthfill and rockfill dams. They conclude that with HSM's use, calculated
deformations of embankment dams are consistent with field data. Honkanadavar and
Sharma [162] have compared results of triaxial tests with simulations performed with
HSM. They concluded that there was a good match between the triaxial behavior of
riverbed and blasted quarried rockfill materials and the response predicted with the
HSM. Beiranvand and Komasi [163] concluded that the generation of pore water
pressures in an earthen dam could be computed with HSM, which agrees with filed
data. Sukkarak et al. [164] concluded that HSM could be used for predicting the
behavior of rockfill materials in rockfill dams.
2.22 SUMMARY
The following main conclusions are drawn from the literature review of the stability
of embankment dams:
It is essential to assign different material properties to various soils used in zoned
embankment dams as reliable as possible. Because stress-strain response, stability,
and settlement of embankment dams are highly dependent on material properties'
input. It is necessary to include foundation soils in addition to embankment soils for
numerical modeling of stability and settlement.
The literature review regarding stability and settlement issues in embankment dams
suggests a need for more cases to address the effect of variation of material properties
obtained from different borrow areas. It is also necessary to investigate the impact of
lowering the reservoir on an embankment dam's stability. Numerical evaluations are
required to perform a safe rate of reducing of a reservoir so that the embankment
dams can be stable.
Regional rainfall in which an embankment dam is located may affect the stability of
the dam. Therefore, it is necessary to perform analysis to evaluate the effect of
regional rainfall on an embankment dam's stability.
In the literature, there are different models than can represent the stress-strain
behavior of soils. The MCM can be utilized for the computation of slope stability of
embankment dams. The HSM can be used for the estimation of settlement in
70
embankment dams. The HSM gives more reliable estimate of the settlement of
embankment dams as compared to the MCM.
The type of analysis performed for the embankment dam is stability and settlement
analysis and consolidation analysis. Due to the gradual raising of dam pore-pressure
could increase and reduce the shear strength and stability of embankment dams.
Groundwater level and placement moisture content of fill materials influence the
stability of dams.
Excessive settlement of embankment soils and foundation could be dangerous for the
stability of embankment dams. It is generally accepted that settlement of 1-2% of the
dam's height is considered within acceptable limits. If a dam's foundation is weak, it
may require treatment such as consolidation and curtain grouting.
71
CHAPTER 3
MATERIALS AND NUMERICAL MODELLING
3.1 GENERAL
Stability and settlement analysis of an earth dam called Nai Gaj Dam located in
district Dadu, Sindh Pakistan, is presented in this study. Initially, the project
background and location are described. The typical section of the main dam is
selected for numerical modeling. The Geotechnical Investigation report of the project
is reviewed to know about foundation conditions and different dam zones. Soil
properties are mainly obtained from the geotechnical investigation report, and some of
the properties are also obtained from the literature on similar types of soils. The finite
element model is developed using the finite element program Plaxis 2D. Material
properties and boundary conditions are assigned. Finally, the numerical calculations
were run, and the selected results are analyzed for stability and settlement of the dam.
The dam's stability was calculated for the EoC, after filling the reservoir, rapid
drawdown of the reservoir, and the effect of maximum possible rainfall on the dam's
slope stability.
3.2 PROJECT BACKGROUND AND LOCATION
Nai Gaj is a large stream in the Khirthar range of mountains in Sindh. The Nai Gaj
originates from Baluchistan, and after covering a distance of 160 km, it joins the
Khirthar range of mountains. It is located about 65 km north-west of Dadu city. The
bed material of the Gaj river consists mainly of gravel, pebbles, cobbles, clayey silty
and sand [165]. The rocks at the dam site are composed of conglomerate, sandstone,
siltstone and claystone.
71
72
3.3 MATERIALS AND METHODS
The proposed Nai Gaj Dam is an earth-fill type dam. It has been observed that the
reservoir rim is lower at many locations than the maximum flood level; for that
purpose, nine earthen dykes have been designed to protect the floodwater (Figure.
3.1).
Figure. 3.1: 3D view of Nai Gaj Dam with earthen dykes.
The cross-section of the Nai Gaj dam is shown in Figure. 3.2. The main embankment
of the dam is 59 m high. It is estimated that the total volume of the embankment fill is
about 7.8 million cubic meters. The type of dam is a zoned embankment. The dam has
a central impervious core (cf. Figure. 3.2) of clayey silt. The dam's material zones are
sandy gravel, random fill, central impervious clay core, sand filter, coarse filter, and
drainage blanket. The foundation of the dam mainly consists of sandy siltstone. The
dam is 1,137 m long, having a maximum design water level of 56.6 m. The
foundation of the dam consists mainly of sandy siltstone.
73
The clay core is constructed of clayey silt obtained from the borrow area on the
downstream side. The primary function of the clay core is to control seepage through
the dam. The proposed degree of compaction is 98% of standard proctor dry density.
The core is placed at wet of optimum moisture content to allow for post construction
settlement without cracking. The top width of the core is 4 m. To provide an effective
water barrier, the top of the core is extended to 0.8 m above the maximum water level.
The core's upstream slope is 1H to 3V, and the downstream slope of the core is 1H to
4V.
The upstream shoulder is composed of sandy gravel, which provides stability and
controls erosions. The downstream shoulder consists of random fill material, which
provides support to the core. The chimney drain is provided along the downstream of
the dam. A drainage blanket is used below the downstream shoulder. The inclination
of both upstream and downstream slopes is 1V to 2.5H. The freeboard is about 2.4 m.
The cut-off trench is extended up to sufficient depth in sound bedrock. A series of
curtain grouting and consolidation grouting is provided under the dam's clay core
section to prevent seepage. The diameter of each of the curtain and consolidation
grout is about 76 mm. The curtain grouting length is 16 m, while the length of
consolidation grouting is 45 m.
The crest width of the dam is about 10 m. The chimney drain is provided on the
downstream side of the core. It is 3 m thick and is provided to ease the construction at
the site and also hold the seepage from the cracks, which may usually develop in the
core.
Chimney drain is provided for the following purposes:
(i) To allow seepage water through the core so that there is no subtle particle
erosion.
(ii) To allow seepage water to pass through the drainage blanket without creating
a disturbance in the downstream shoulder.
74
A 0.5 m thick gravel drain is provided below the downstream shoulder between the
sand filter layers. The function blanket's collect seepage water from the chimney drain
and bedrock and dispose of the downstream toe. To reduce the wave action and flow
velocities, riprap is provided on the upstream side. The riprap is extended from crest
to the natural surface layers. Riprap consists of rock fragments that are dense, sound
and durable. A bedding material layer of 0.6 m thick is provided below the riprap
layer. The bedding layer aims to provide support to the riprap and protect the
embankment from the wave action. To protect the downstream slope, a 0.6 m thick
layer of cobbles and gravel is provided to protect the dam from erosion by wind or
rain. The size of cobbles and gravels is in the range of 75 mm to 200 mm. Relief wells
are provided at the downstream toe to decrease pore pressure, which may develop due
to seepage and may otherwise cause instability or piping. The spacing between the
two wells is 15 m center to center.
Drainage features, including a chimney drain and horizontal drainage blanket below
the downstream shell, have been incorporated to relieve the pore water pressures
within the dam body and dispose of the seepage water safely so that erosion and
boiling are prevented.
Numerical analyses were carried out on a typical section of the Nai Gaj dam for the
following conditions:
(i) Stability, settlement, and excess pore pressure analyses have been carried out
using a two-dimensional plane strain condition with finite element program
Plaxis 2D for two possible cases: gradual raising at the EoC and AFR.
(ii) Numerical analysis of the lowering of the reservoir is performed so that the
dam can be lowered safely without causing instability conditions.
(iii) Numerical analysis of the effect of rainfall on the slope stability of the dam is
performed.
75
Figure. 3.2: Cross-section of Nai Gaj dam [165].
3.4 CONSTITUTIVE MODELS
The linear elastic model was used for consolidation and curtain grouts, which are used
in the foundation of the dam. The Linear Elastic Model's material properties are the
MoE and Poisson’s ratio, whose values are 2x105 kN/m
2 and 0.33.
The MCM was used for all layers of the dam to computation pore pressures and safety
factors at the EoC. After filling the reservoir, the reservoir's rapid drawdown and
effect of maximum possible rainfall on the dam's stability. The material properties of
the MCM are described in Table 3.1. Mainly the material properties are obtained from
laboratory and field tests [165]. However, some of the material properties were taken
from the literature.
To ensure the safety of embankment dams, in addition to stability [166, 167],
settlement plays a critical role in being estimated with reliability [168, 169]. It has
been observed that the HSM is more reliable for calculation of settlement of the dam
as compared to the MCM. On the contrary, advanced laboratory tests are required to
evaluate the material properties of the HSM. However, the settlement of embankment
dams predicted with MCM may not be realistic for a different type of soils. The MCM
may show unrealistically more settlement. In such situations, there might be a need
76
for advanced constitutive models depending upon the types of soils involved in
embankment dam. The use of advanced soil models for computation of settlement of
an embankment dam, may require advanced testing which may result into more costs.
To obtain a reliable estimate of the settlement of an embankment dam by utilizing
material properties that are already determined during the geotechnical investigation
phase, it might be necessary to find out those material zones that may show variation
in settlement computed with the MCM in comparison to response predicted with an
advanced constitutive model like the HSM.
Therefore, it was decided to identify the dam zones, which occupy a considerable fill
material volume. The dam's cross-section shows that there are four zones, which
occupy most of the dam’s volume. These zones are clay core, sandy gravel, random
fill, and sandy siltstone. It is now important to investigate the effect of these zones on
the dam's overall settlement. For the computation of settlement in the dam during the
EoC and filling of the reservoir, HSM was used for four major zones: clay core, sandy
gravel, random fill, and sandy siltstone. For the remaining material zones, MCM was
used. The material properties of the HSM are presented in Table 3.2.
Table 3.1: Material properties for Nai Gaj dam and its foundation.
Material
type
Saturated
unit weight
(kN/m3)
Cohesion
(kN/m2)
Friction
angle
(deg)
MoE
(kN/m2)
Permeability
(m/day)
Clay 18.85[165]
9.57[165]
30[165]
50000[165]
0.000263[165]
Sandy
gravel 21.5
[170] 0 37
[171-173] 50000
[174] 86.4
[175]
Random
fill 18.85
[165] 0 34
[165] 50000
[165] 0.0263
[165]
Washed
gravel 21.5
[170] 0 37
[171-173] 45000
[176] 864
[135]
D/s slope
protection 19.5
[177] 0 34
[171-173]
40000[39, 135,
178]
8640[135]
Riprap 19.5[165]
0 34[165]
40000
[39, 175,
178]
8640[135]
Sand filter 18.85[165]
0 36[165]
40220[165]
26.33[165]
Drainage
blanket 21.5
[170] 0 37
[171-173] 45000
[135] 864
[135]
Sandy
siltstone 20.4
[165] 12
[165] 29
[165] 125000
[179] 0.00063
[165]
77
Table 3.2: HSM parameters for four main zones of dam [75].
Material Type
Clay core 50000 50000 150000
Random fill 50000 50000 150000
Sandy gravel 50000 50000 150000
Sandy siltstone 125000 125000 375000
Since the dam's settlement depends on the magnitude of the MoE of different soil
types, the effect of MoE variation of the four major zones of the dam is investigated.
For each of the three materials (clay core, sandy gravel and random fill), the MoE is
varied from 25000 to 50000 kN/m2 separately. While for the fourth material zone
(sandy siltstone), the MoE is increased from 70000 to 125000 kN/m2. Whenever the
effect of a particular material zone on the dam's settlement is investigated, the other
materials' properties are utilized as presented in Table 3.1 and Table 3.2. The MoE of
sandy siltstone was evaluated from its uniaxial compressive strength using the
following correlation [179].
Where E is the MoE (kN/m2) and is the uniaxial compressive strength (kN/m
2).
Most of the results show that the value of sandy siltstone's uniaxial compressive
strength at the dam site is less than 2 MPa. The minimum value is 0.35 MPa [165].
The following values of MoE of sandy siltstone are described for numerical analysis
of the present study:
E=200×0.35=70000 kN/m2 (minimum value) 3.2
E=350×0.35=125000 kN/m2 (maximum value) 3.3
In the above equations, the coefficient 200 is the minimum value of the prescribed
range of 200- 500. The coefficient 350 is the average of 200 and 500. The minimum
value of the MoE is 70,000 kN/m2,
and the maximum value is 125,000 kN/m2.
78
Since foundation material plays an essential role in the dam's settlement, the value of
the MoE was not actually determined from the laboratory or field tests. Preferably, the
value of MoE was obtained from correlation, as mentioned in equation 3.1. Therefore,
parametric analysis was performed to determine a suitable range of MoE of sandy
siltstone.
Coupled deformation and consolidation analysis were carried out to estimate the
development and dissipation of excess pore pressures, as a function of time, in the
dam.
The mechanical behavior of soils in the finite element program Plaxis 2D can be
described by fully drained and undrained conditions. Real soil behavior depends on
time. The equations describing the flow of water through soils are combined with the
equations governing the soil's deformation to capture such behavior. This type of
theoretical response is called coupled deformation and consolidation [180].
Besides, the safety analysis was conducted to compute the safety factors of the dam.
The raising of the embankment dam was decided based on construction work carried
on the site. The dam was mainly raised at a rate of three meters in 30 days.
The dam's finite element model was extended laterally by 100 m on each side to
minimize the effect of boundaries, as shown in (Figure. 3.3). The finite element mesh
consisted of 6619 triangular elements with 53463 nodes. The average element size
was 3.2 m. The mesh was sufficiently refined to such an extent that with further
refinement of the mesh, the results were almost the same (Figure. 3.4). The water
level in the finite element model is 15 m below the surface (Figure. 3.5). The highest
flood level of 56.6 m in the finite element model is shown in Figure. 3.6.
79
Figure. 3.3: FEM for Nai Gaj dam.
Figure. 3.4: FEM mesh of Nai Gaj dam.
Figure. 3.5: Highest reservoir level of Nai Gaj dam.
3.5 STABILITY OF THE DAM
As mentioned earlier, the stability of the dam was analyzed based on the factor of
safety. For this purpose, consolidation and safety analyses were performed using the
80
MCM for all the material zones, except consolidation grouting and grout curtain,
where the Linear Elastic model was used.
3.6 SETTLEMENT OF THE DAM
The settlement with respect to the dam's depth and long-term settlement was
calculated with the MCM and HSM. The purpose of the calculation of the dam's
settlement with both models was to check the reliability of the HSM compared to the
MCM.
It is relevant to describe here that the MoE is vital for the computation of settlement.
Real soils exhibit stress-dependent stiffness, whereas MCM uses a single MoE value,
resulting in the overestimation of settlement for embankment dams. On the other
hand, the HSM uses three different types of stiffness, which are stress-dependent.
The three types of stress-dependent stiffness used in the HSM are described in
Chapter 2.
3.7 RAPID DRAWDOWN OF THE DAM
The dam was also analyzed for rapid drawdown conditions due to the uneven
distribution of rainfall and continuous water discharge for drinking and irrigation
purposes. The stress-strain behavior of all the materials was modeled with the MCM.
The gradual raising process of the dam was carried out with consolidation analysis.
For the computation of safety at various stages during the dam's service life, safety
analysis was performed.
The rapid drawdown was carried out using coupled deformation and consolidation
analysis. Both the vertical boundaries and bottom boundary were closed for
groundwater flow in the finite element model. Before performing rapid drawdown
calculations, the effective strength and stiffness parameters (refer to table 1) of the
clay core were changed with undrained shear strength values of the clay core ranging
between 20 to 30 kN/m2. During the drawdown, the support provided by water to the
upstream embankment is lost. As a result, seepage forces develop on the upstream
81
embankment and extend to the clay core. It depends on the undrained strength of the
clay to resist seepage forces. If the undrained strength of clay is low, there will be
chances of the instability of the upstream embankment depending upon the generation
of pore pressures. The values of undrained shear strength were evaluated based on the
consistency index mentioned in reference [181]. The values were computed as
follows:
and
Based on the consistency index value of 0.726, the undrained shear strength (Su) of
soil lies in the range of 20 to 40 kN/m2 (Table 3.3) [181]. The drawdown calculations
were performed for undrained strength of clay as 20, 25 and 30 kN/m2. Besides, since
the clay became saturated, therefore, its reduced MoE was taken as 30000 kN/m2
instead of the initial value of 50000 kN/m2 [181].
The dam's drawdown was performed for different rates ranging from 1 m/day to 0.1
m/day. The lowest value of the drawdown was 10 m, and the highest value was 55 m,
as shown in Figure. 3.6 and Figure. 3.7, respectively.
Table 3.3: Relationship b/w consistency index and undrained shear-strength.
Strength
classification Consistency index (1-LI)
Undrained shear strength
(kN/m2)
Hard > 1.15 > 300
Very stiff 1.05 - 1.15 150 -300
Stiff 0.92 - 1.05 75 - 150
Firm 0.82 - 0.92 40 - 75
Soft 0.60 - 0.82 20 - 40
Very soft < 0.60 < 20
82
Figure. 3.6: 10 m lowering of reservoir @ 1 m/day.
Figure. 3.7: Lowering of the reservoir up to 55 m depth @ 0.1 m/day.
3.8 EFFECT OF RAINFALL ON STABILITY OF THE DAM
The stability of the dam was assessed for possible rainfall. During the last 40 years,
the maximum rainfall was 435 mm/year [184] Figure. 3.8; this rainfall was converted
to m/day and applied to the dam Figure's crest and downstream boundaries 3.9. The
average annual rainfall of 435 mm/year was used in five equal increments in five
days. The effect of rainfall was calculated on the stability of the dam.
83
Figure. 3.8: Average annual rainfall at Nai Gaj dam for four decades.
Figure. 3.9: Input of rainfall intensity at Nai Gaj Dam.
0
50
100
150
200
250
300
350
400
450
500
Ra
infa
ll i
nte
nsi
ty
Years
Average annual rainfall Nai Gaj Dam
84
3.9 SUMMARY
The main points described in this chapter are summarized as under:
o Background of the project and geological conditions of foundation
o Cross-section of the main dam
o Material properties of various soils used in dam and foundation required for
MCM and HSM
o Stability and the settlement of the dam during staged construction and AFR
o Stability of the dam during rapid drawdown
o Stability of the dam when subjected to rainfall.
85
CHAPTER 4
RESULTS AND DISCUSSION
4.1 GENERAL
Numerical analyses were performed for stability and settlement of an embankment
dam using finite element program PLAXIS 2D. The numerical modeling was utilized
to study the effect of retained water on the dam's stability by evaluating its stability
before and after filling up the reservoir. The other factors examined were the dam's
stability for the condition if the reservoir is lowered rapidly and how the stability is
affected if the maximum possible rainfall occurs in the reservoir area. Besides, the
dam's settlement was computed for the end of the construction stage, after filling the
reservoir and long-term. Parametric analyses are performed to investigate the effect of
change of strength properties (mainly friction angle) of three main material zones
(sandy gravel, random fill, and clay core) on the dam's overall stability during raising
and filling of the reservoir. For evaluation of the stability of the dam, the MCM was
used for all the material zones. For the computation of settlement, HSM was used for
four material zones (sandy gravel, random fill, clay core, and sandy siltstone). The
remaining material zones were analyzed with MCM. Parametric analyses were also
conducted by changing the stiffness parameter (i.e., MoE) of the four main zones to
compute the dam's settlement's reliable magnitude during raising and filling of the
reservoir conditions. Relevant results obtained from the numerical modeling process
are described and discussed in this chapter.
4.2 FACTORS AFFECTING DAM STABILITY
The dam's stability was determined based on the safety factor, which depends on the
development of excess pore pressure during raising and filling the reservoir
conditions.
85
86
4.2.1 Effect of Dam Height on Foundation Soil
The rate of loading by the embankment layers was 3 m in 30 days. The bottom of the
dam, where embankment layers are laid, is 300 m wide, and the dam's crest is 10 m in
width. The magnitude of excess pore pressures increased when the embankment was
raised to a level of 45 m (Figure. 4.1-4.3). When the dam was raised from 45 m to 59
m, the excess pore pressures mainly decreased in magnitude. This is because from the
bottom up to 45 m height (Figure. 4.3), the retained soil's weight is more than the
weight of the soil-filled from 45 to 59 m size of the dam (Figure. 4.4). As the dam is
being raised higher and higher, the width of layers gradually decreased. This indicates
that the weight of the top layers of the dam was lower as compared to the bottom
layers (Figure. 4.4). The increase of excess pore pressure at 15 m, 30 m and 45 m
raising of the dam was 46%, 5.2% and 1.2%, respectively. This shows that the
percentage increase in excess pore pressure decreased as the dam height was
increased. Finally, when the dam was raised from 45 m to 59 m, the excess pore
pressure instead of increasing dissipated at a 5% rate.
Figure. 4.1: Increase in pore pressure at 15 m dam raising.
87
Figure. 4.2: Increase in pore pressure at 30 m dam raising.
Figure. 4.3: Increase in pore pressure at 45 m dam raising.
88
Figure. 4.4: Increase in pore pressure at 59 m dam raising.
4.2.2 Effect of Water Filling on Potential Failure of Dam
The dam's safety was computed to be 1.6 and 1.5 for the EoC and AFR conditions.
Due to the soils' saturation, the factor of safety after filling the reservoir is decreased
compared to the end of the construction condition. According to available guidelines
[78], the safety factors for the EoC and the reservoir at the normal operating levels are
1.3 and 1.5, respectively. This implies that the dam's stability at the EoC and AFR is
considered satisfactory.
In addition to the magnitude of safety factors, it is relevant to observe the
embankment dam's potential failure zones at the EoC and AFR. The dam's failure
zones at the EoC and AFR are shown in Figure’s 4.5 and 4.6, respectively. The
potential failure zone for the EoC developed at the upstream side of the dam.
Whereas, the failure zone after filling of the reservoir developed on the downstream
side. The modes of failure are consistent with the findings of the literature [32].
89
Figure. 4.5: Potential failure zone of the dam at the EoC.
Figure. 4.6: Potential failure zone of the dam AFR.
4.3 EFFECT OF FRICTION ANGLE OF SANDY GRAVEL
As the dam's upstream shell consists of sandy gravel material (cf Figure. 3.2), the
effect of the sandy gravel's friction angle on the dam's safety factor was examined
(Figure. 4.7). As expected, the safety factor increased with increasing values of the
sandy gravel's friction angle. For both the EoC and AFR conditions, the dam's
stability is satisfactory if the minimum value of sandy gravel's friction angle is taken
as 340. This implies that the sandy gravel's friction angle value could be 34
0 instead of
370
to be on the safe side. The safety factor's value at the EoC was higher than AFR
condition. The reason is that when the reservoir was filled with water, the soil became
saturated; as a result, the factor of safety was reduced.
90
Figure. 4.7: Effect of friction angle for sandy gravel on
dam safety factor.
4.4 EFFECT OF FRICTION ANGLE OF RANDOM FILL
As the downstream shell of the dam consists of random fill material (cf Figure. 3.2),
the effect of the friction angle of the random fill on the safety factor was analyzed
(Figure. 4.8). As expected, the safety factor increased with increasing the values of
the random fill's friction angle. After filling the reservoir condition, the friction angle
of random fill exceeding 320 showed no influence on the safety factor. This implies
that the random fill's friction angle could be 320 instead of 34
0 to remain on the safe
side. The safety factor values at the EoC were higher than AFR condition. The reason
is that when the reservoir was filled with water, the soil became saturated; as a result,
the factor of safety was reduced.
1.34
1.39
1.44
1.49
1.54
1.59
1.64
30 32 34 36
Fact
or
of
safe
ty
Friction angle of sandy gravel
End of construction After filling of reservoir
91
Figure. 4.8: Effect of friction angle for random fill on dam safety factor.
4.5 EFFECT OF FRICTION ANGLE OF CLAY CORE
As the center of the dam consists of clay core material (cf Figure. 3.2), the effect of
the clay core's friction angle on the dam's stability was investigated (Figure. 4.9). As
expected, the dam's safety factor increased with increasing values of the clay core's
friction angle for both the EoC and AFR conditions. In the partially dry state, the core
was stronger, and the factor of safety of the dam was more than sufficient, even the
value of friction angle utilized is 200 instead of 30
0. However, clay is more sensitive
to saturation. This implies that the friction angle of the core decreases with an
increase in moisture content. Therefore, it is necessary to fully compact the core and
utilize the maximum value of friction angle for the embankment dam's stability,
which is 300. The safety factor values at the EoC were higher than AFR condition.
The reason is that when the reservoir was filled with water, the soil became saturated;
as a result, the factor of safety was reduced.
1.42
1.44
1.46
1.48
1.5
1.52
1.54
1.56
1.58
1.6
1.62
30 31 32 33 34
Fa
cto
r o
f sa
fety
Friction angle of random fill (deg)
End of construction After filling of reservoir
92
Figure. 4.9: Effect of friction angle for clay core on dam safety factor.
4.6 DAM SETTLEMENT COMPARISON USING MCM AND
HSM MODELS
All materials were modeled with the MCM (condition 1). Each of the four materials
(clay core, sandy gravel, random fill and sandy siltstone) were modeled separately
with HSM, and the rest of the materials were modeled with MCM (condition 2). The
dam's settlement calculations were performed for the EoC and AFR conditions. The
settlement response of the dam computed with MCM and HSM were compared for
the EoC and AFR in Figure’s 4.10-4.17.
The results (Figure’s 4.10-4.15) concluded that there is less variation in settlement
computed with the MCM and HSM for three major material zones (clay core, sandy
gravel and random fill) EoC and AFR conditions. The MCM overestimated the shear
strength for clay core, random fill and sandy gravel [75]. In undrained loading
conditions, the MCM shows a constant effective stress path to failure, while the real
soils show a curved effective stress path [75]. Therefore, the settlement predicted by
the MCM was slightly lower than the one computed with the HSM. On the other
hand, for sandy siltstone (foundation), the settlement predicted with the MCM is
higher than the one predicted with HSM (Figure’s. 4.16-4.17).
1.42
1.46
1.5
1.54
1.58
1.62
20 21 22 23 24 25 26 27 28 29 30
Fact
or
of
Safe
ty
Friction angle of clay core (deg)
End of construction After filling of reservoir
93
A comparison of the dam's settlement computed with MCM and HSM for end of
construction, and after filling the reservoir, conditions are illustrated in Table 4.1 and
4.2, respectively.
Figure. 4.10: Dam settlement comparison for clay core at EoC.
Figure. 4.11: Dam settlement comparison for clay core at AFR.
1.5
1.6
1.7
1.8
1.9
2
2.1
25000 30000 35000 40000 45000 50000
Set
tlem
ent
(m)
MoE of clay core (kN/m2)
HSM MCM
1.5
1.6
1.7
1.8
1.9
2
2.1
25000 30000 35000 40000 45000 50000
Set
tlem
ent
(m)
MoE of clay core (kN/m2)
HSM MCM
94
Figure. 4.12: Dam settlement comparison for sandy gravel at EoC.
Figure. 4.13: Dam settlement comparison for sandy gravel at AFR.
1.5
1.6
1.7
1.8
1.9
2
2.1
25000 30000 35000 40000 45000 50000
Set
tlem
ent
(m)
MoE of sandy gravel (kN/m2)
HSM MCM
1.5
1.6
1.7
1.8
1.9
2
2.1
25000 30000 35000 40000 45000 50000
Set
tlem
ent
(m)
MoE of sandy gravel (kN/m2)
HSM MCM
95
Figure. 4.14: Dam settlement comparison for random fill at EoC.
Figure. 4.15: Dam settlement comparison for random fill at AFR.
1.5
1.6
1.7
1.8
1.9
2
2.1
25000 30000 35000 40000 45000 50000
Set
tlem
ent
(m)
MoE of random fill (kN/m2)
HSM MCM
1.5
1.6
1.7
1.8
1.9
2
2.1
25000 30000 35000 40000 45000 50000
Set
tlem
ent
(m)
MoE of random fill (kN/m2)
HSM MCM
96
Figure. 4.16: Dam settlement comparison for sandy silt stone at EoC.
Figure. 4.17: Dam settlement comparison for sandy silt stone at AFR.
0.6
0.9
1.2
1.5
1.8
2.1
2.4
70000 80000 90000 100000 110000 120000 130000
Set
tlem
ent
(m)
MoE of sandy siltstone (kN/m2)
HSM MCM
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
70000 80000 90000 100000 110000 120000
Set
tlem
ent
(m)
MoE of sandy siltstone (kN/m2)
HSM MCM
97
Table 4.1: Summary of dam's settlement comparison for EoC.
Mat
eria
l MCM (EC) HSM (EC)
E=
25
000
(k
N/m
2)
E=
50
00
0 (
kN
/m2)
Set
tlem
ent
incr
ease
by
MC
M
(%)
as c
om
par
ed t
o H
SM
E=
25
00
0 (
kN
/m2)
Set
tlem
ent
incr
ease
by
MC
M
(%)
as c
om
par
ed t
o H
SM
E=
50
00
0
(kN
/m2)
E=
25
00
0 (
kN
/m2)
E=
50
000
(k
N/m
2)
Set
tlem
ent
incr
ease
by
HS
M
(%)
as c
om
par
ed t
o M
CM
E=
25
00
0 (
kN
/m2)
Set
tlem
ent
incr
ease
by
HS
M
(%)
as c
om
par
ed t
o M
CM
E=
50
000
(k
N/m
2)
Clay core 1.819 m 1.581 m
1.826 m 1.595 m 0.38 0.88
Sandy gravel 1.768 m 1.581 m
0.82 1.812 m 1.568 m 2.43
Random fill 1.724 m 1.581 m
0.89 1.929 m 1.567 m 10.6
Sandy
siltstone
E=70000 to
125000
(kN/m2)
2.33 m 1.581 m 56.65 49.40 1.01 m 0.8 m
Table 4.2: Summary of dam’s settlement comparison for AFR.
Mat
eria
l
MCM (AFR) HSM (AFR)
E=
25
000
(k
N/m
2)
E=
50
00
0 (
kN
/m2)
Set
tlem
ent
incr
ease
by
MC
M (
%)
as
com
par
ed t
o H
SM
E=
25
000
(k
N/m
2)
Set
tlem
ent
incr
ease
by
MC
M (
%)
as
com
par
ed t
o H
SM
E=
50
000
(k
N/m
2)
E=
25
00
0 (
kN
/m2)
E=
50
000
(k
N/m
2)
Set
tlem
ent
incr
ease
by
HS
M (
%)
as
com
par
ed t
o M
CM
E=
25
00
0 (
kN
/m2)
Set
tlem
ent
incr
ease
by
HS
M (
%)
as
com
par
ed t
o M
CM
E=
50
00
0 (
kN
/m2)
Clay core 1.891 m 1.702 m
1.978 m 1.751 m 4.40 2.8
Sandy
gravel 1.799 m 1.702 m
2.003 m 1.775 m 10.18 4.11
Random
fill 1.928 m 1.702 m
2.004 m 1.584 m 3.79 6.93
Sandy
siltstone
E= 70000
to 125000
(kN/m2)
2.547 m 1.702 m 58.85 50.30 1.055 m 0.844 m
98
4.6.1 Settlement at EoC and AFR
Figure. 4.18 compares dam settlement computed with the MCM and HSM for EoC
and AFR conditions. The MoE of sandy siltstone was varied from 70000 to 125000
kPa. It is observed that for EoC, the MCM showed 56.65% and 49.40% more
settlement than the HSM when the MoE of sandy siltstone increased from 70000
kN/m2 to 125000 kN/m
2 respectively. It is observed that for AFR, the MCM showed
58.85% and 50.30% more settlement than the HSM when the MoE of sandy siltstone
increased from 70000 kPa to 125000 kPa, respectively.
The MCM is suitable to determine the bearing capacity of foundations and slope
stability of embankment [151]. On the contrary, the MCM is unable to predict the
settlement of soils reliably [182]. The MCM uses constant soils' constant stiffness and
does not distinguish between initial loading, unloading, and reloading conditions in
soils. As a result, there is a significant difference between real soil settlement and the
predicted settlement with the MCM [151]. On the other hand, the HSM utilizes three
types of stiffness for loading, unloading and reloading situations. Therefore, the
settlement predicted with the HSM is more reliable for real problems [151].
Figure. 4.18: Settlement increase in % when MoE for sandy siltstone
varied from 70,000-125,000 kN/m2.
48
50
52
54
56
58
60
70000 80000 90000 100000 110000 120000 130000
Per
cen
tage
incr
ease
of
sett
lem
ent
of
the
dam
calc
ula
ted
wit
h M
CM
as
com
pare
d t
o H
SM
MoE of sandy siltstone (kN/m2)
EOC AFOR
99
The settlement of the dam was calculated only at the crest of the dam. From the
results, it is concluded that the sandy siltstone (foundation) has more influence on
settlement than the other zones of the dam. According to a study on 134 such
embankments, it was found that most of the embankments settled as high as 1 to 2%
of the dam height [94]. As mentioned above, the sandy siltstone (foundation) is
sensitive to settlement when predictions were made with the MCM and HSM. The
MCM showed a settlement of 2.5 m at AFR when the MoE of sandy siltstone is
70000 kN/m2. This indicates that the predicted settlement with the MCM is about
4.2% of the dam height. The MCM showed a settlement of 1.7 m at AFR when the
MoE of sandy siltstone is 125000 kN/m2. This indicates that the predicted settlement
with the MCM is about 2.9% of the dam height.
The HSM showed a settlement of 1 m at AFR when the MoE of sandy siltstone is
70000 kN/m2, with a predicted settlement of about 1.69%, and a settlement of 0.8 m
at AFR when the MoE of sandy siltstone is 125000 kN/m2, showing the predicted
settlement of about 1.35%. This implies that the 0.8 m settlement of the dam is
considered to be a reasonable estimate.
4.6.2 Dam Settlement Comparison w.r.t. Depth
The settlement of the dam was computed by applying MCM to all zones of the dam.
For the sake of convenience, this is referred to as (condition 1). In the second case,
HSM is applied separately to four main zones (clay core, sandy gravel, random fill,
sandy siltstone), and the MCM was used for remaining materials (condition 2). The
settlements were determined along the dam's central axis and were computed up to a
depth of 120 m with an interval of 20 m, as shown in Figure. 4.19. The settlement
with respect to depth response of the dam when computed with MCM (condition 1)
and HSM applied to clay core, sandy gravel, random fill (condition 2) for EoC and
AFR are respectively given in Figure’s 4.20 to 4.25.
100
Figure. 4.19: Dam's central axis at which settlement was computed
w.r.t. depth at an interval of 20 m.
Figure. 4.20: Dam's settlement comparison w.r.t. depth
at EoC applied to clay core.
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
101
Figure. 4.21: Dam's settlement comparison w.r.t. depth
at AFR applied to clay core.
Figure. 4.22: Dam's settlement comparison w.r.t. depth
at EoC applied to sandy gravel.
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
) Settlement (m)
HSM MCM
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
102
Figure. 4.23: Dam's settlement comparison w.r.t. depth
at AFR applied to sandy gravel.
Figure. 4.24: Dam's settlement comparison w.r.t. depth
at EoC applied to random fill.
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
103
Figure. 4.25: Dam's settlement comparison w.r.t. depth
at AFR applied to random fill.
From the results, it can be observed that there is less variation in the settlement
concerning depth when computed with MCM as compared to HSM for both EoC and
AFR for material zones (clay core, sandy grave and random fill).
Figure’s 4.26 and 4.27 respectively illustrate settlement concerning the depth of the
dam for end of construction and after filling of the reservoir when computed with
MCM (condition 1) and HSM (condition 2) applied to sandy siltstone (Foundation)
only.
Tables 3 and 4 respectively present the settlement concerning the dam's depth
computed with MCM and HSM for the end of construction and after filling the
reservoir.
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
) Settlement (m)
HSM MCM
104
Figure. 4.26: Dam's settlement comparison w.r.t. depth
at EoC applied to sandy silt stone.
Figure. 4.27: Dam's settlement comparison w.r.t. depth
at AFR applied to sandy silt stone.
105
Table 4.3: Summary of dam’s settlement comparison for EoC.
Mat
eria
l MCM (EC) HSM (EC)
Set
tlem
ent
at
cres
t (m
)
Set
tlem
ent
at 1
20
m d
epth
Set
tlem
ent
incr
ease
by
MC
M (
%)
com
par
ed t
o
HS
M a
t cr
est
Set
tlem
ent
incr
ease
by
MC
M (
%)
com
par
ed t
o
HS
M a
t 1
20
m d
epth
Set
tlem
ent
(m)
at
Cre
st
Set
tlem
ent
(m)
at a
dep
th o
f
12
0 m
Set
tlem
ent
incr
ease
by
HS
M (
%)
as c
om
par
ed t
o
MC
M a
t th
e C
rest
Set
tlem
ent
incr
ease
by
HS
M (
%)
as c
om
par
ed t
o
MC
M a
t a
dep
th o
f 1
20
m
Clay core 1.574 0.55 4.18 1.612 0.527 2.36
Sandy
gravel 1.574 0.55 0.44 2.91 1.567 0.534
Random fill 1.574 0.55 0.51 1.566 0.562 2.41
Sandy
siltstone
E=70,000 to
125,000
(kN/m2)
1.574 0.55 49.17 79.64 0.8 0.112
Table 4.4: Summary of dam’s settlement comparison for AFR.
Mat
eria
l
MCM (EoC) HSM (AFR)
Set
tlem
ent
(m)
at t
he
Cre
st
Set
tlem
ent
(m)
dep
th o
f 1
20
m
Set
tlem
ent
incr
ease
by
MC
M (
%)
as c
om
par
ed
to H
SM
at
the
cres
t
Set
tlem
ent
incr
ease
by
MC
M (
%)
as
com
par
ed t
o H
SM
at
a d
epth
of
12
0
m
Set
tlem
ent
(m)
at t
he
Cre
st
Set
tlem
ent
(m)
at a
dep
th o
f 12
0 m
Set
tlem
ent
incr
ease
by
HS
M (
%)
as
com
par
ed t
o M
CM
at
the
Cre
st
Set
tlem
ent
incr
ease
by
HS
M (
%)
as
com
par
ed t
o M
CM
dep
th o
f 12
0 m
Clay core 1.7 0.568 1.733 0.571 1.90 0.53
Sandy gravel 1.7 0.568 0.41 1.693 0.586 3.07
Random fill 1.7 0.568 7.29 1.41 1.576 0.56
Sandy
siltstone
E=70000 to
125000
(kN/m2)
1.7 0.568 52.94 82.39 0.8 0.1
There is apparent variation in settlement both at the crest of the dam and at a depth of
120 m. Because in HSM, the modulus of soil elasticity increases with respect to the
depth, which is a real soil behavior. The available literature on embankment dams
106
suggests that most dams exhibit a settlement of about 1 to 2% of the dam's total height
[93, 94]. This dam's settlement computed with MCM and HSM after filling the
reservoir was found to be 2.9% and 1.35% of the dam height, respectively. It can be
interpreted that by utilizing the HSM parameters as mentioned in this study, the
settlement of the dam is considered to agree with the literature. From the above
results, it is observed that sandy siltstone has more influence on the settlement
concerning the depth of the dam as compared to the other material zones (clay core,
sandy gravel, random fill) when computed with MCM and HSM. Therefore, the
settlement response of sandy siltstone is investigated in more detail in Figure’s 4.28-
4.39. For this purpose, MoE of sandy siltstone was varied from 70000 to 125000
kN/m2 with an increment of 10000 kN/m
2 and settlement concerning depth was
calculated using both MCM and HSM at the EoC and AFR.
Tables 5 and 6 respectively present the effect of variation of MoE of sandy siltstone
on the settlement concerning depth using MCM and HSM for end of construction and
after filling of reservoir conditions.
Figure. 4.28: Dam's settlement comparison w.r.t. depth
For EoC applied to sandy silt stone (MoE = 70,000 kN/m2).
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
107
Figure. 4.29: Dam's settlement comparison w.r.t. depth
at AFR applied to sandy silt stone (MoE = 70,000 kN/m2).
Figure. 4.30: Dam's settlement comparison w.r.t. depth
for EoC applied to sandy silt stone (MoE = 80,000 kN/m2).
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5 3
Dep
th o
f d
am
(m
) Settlement (m)
HSM MCM
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
108
Figure. 4.31: Dam's settlement comparison w.r.t. depth
for AFR applied to sandy silt stone (MoE = 80,000 kN/m2).
Figure. 4.32: Dam's settlement comparison w.r.t. depth
at EoC applied to sandy silt stone (MoE = 90,000 kN/m2).
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5
Dep
th o
f d
am
(m
) Settlement (m)
HSM MCM
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
109
Figure. 4.33: Dam's settlement comparison w.r.t. depth
for AFR applied to sandy silt stone (MoE = 90,000 kN/m2).
Figure. 4.34: Dam's settlement comparison w.r.t. depth
at EoC applied to sandy silt stone (MoE = 100,000 kN/m2).
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5
Dep
th o
f d
am
(m
) Settlement (m)
HSM MCM
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
110
Figure. 4.35: Dam's settlement comparison w.r.t. depth
for AFR applied to sandy silt stone (MoE = 100,000 kN/m2).
Figure. 4.36: Dam's settlement comparison w.r.t. depth
at EoC applied to sandy silt stone (MoE = 110,000 kN/m2).
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5
Dep
th o
f d
am
(m
) Settlement (m)
HSM MCM
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
111
Figure. 4.37: Dam's settlement comparison w.r.t. depth
for AFR applied to sandy silt stone (MoE = 110,000 kN/m2).
Figure. 4.38: Dam's settlement comparison w.r.t. depth
at EoC applied to sandy silt stone (MoE = 125,000 kN/m2).
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2D
epth
of
dam
(m
) Settlement (m)
HSM MCM
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2
Dep
th o
f d
am
(m
)
Settlement (m)
HSM MCM
112
Figure. 4.39: Dam's settlement comparison w.r.t. depth
for AFR applied to sandy silt stone (MoE = 125,000 kN/m2).
Table 4.5: Variation effect of MoE of sandy siltstone on settlement
w.r.t. depth at EoC.
Mat
eria
l
MCM (EoC) HSM (EoC)
E =
(kN
/m2)
at t
he
cres
t (m
)
dep
th o
f 120 (
m)
Set
tlem
ent
incr
ease
by M
CM
(%
) as
com
par
ed t
o H
SM
at t
he
cres
t (m
)
Set
tlem
ent
incr
ease
by M
CM
(%
) as
com
par
ed t
o H
SM
dep
th o
f 120 (
m)
at t
he
cres
t (m
)
dep
th o
f 120 (
m)
Set
tlem
ent
incr
ease
by H
SM
(%
) as
com
par
ed t
o M
CM
at
the
Cre
st (
m)
Set
tlem
ent
incr
ease
by H
SM
(%
) as
com
par
ed t
o M
CM
dep
th o
f 120 (
m)
San
dy s
ilts
tone 70000 2.324 0.902 56.58 77.54 1.009 0.199
80000 2.12 0.79 55.24 77.85 0.949 0.175
90000 1.955 0.73 53.81 78.63 0.903 0.156
100000 1.82 0.673 52.42 79.20 0.866 0.14
110000 1.707 0.619 51.03 79.48 0.863 0.127
125000 1.574 0.55 49.17 79.64 0.8 0.112
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2D
epth
of
dam
(m
)
Settlement (m)
HSM MCM
113
Table 4.6: Variation effect of MoE of sandy siltstone on settlement
w.r.t. depth for AFR. M
ater
ial
MCM (AFR) HSM (AFR)
E =
(kN
/m2)
at C
rest
(m
)
dep
th o
f 120 (
m)
Set
tlem
ent
incr
ease
by M
CM
(%
) as
com
par
ed t
o H
SM
at t
he
cres
t (m
)
Set
tlem
ent
incr
ease
by M
CM
(%
) as
com
par
ed t
o H
SM
dep
th o
f 120 (
m)
at c
rest
(m
)
dep
th o
f 120 (
m)
Set
tlem
ent
incr
ease
by H
SM
(%
) as
com
par
ed t
o M
CM
at
the
Cre
st (
m)
Set
tlem
ent
incr
ease
by H
SM
(%
) as
com
par
ed t
o M
CM
dep
th o
f 120 (
m)
San
dy s
ilts
tone 70000 2.548 1.042 60.40 82.73 1.009 0.18
80000 2.298 0.886 58.70 82.28 0.949 0.157
90000 2.138 0.826 57.81 83.05 0.902 0.14
100000 1.976 0.732 56.22 82.92 0.865 0.125
110000 1.841 0.683 56.64 83.31 0.835 0.114
125000 1.7 0.568 52.94 82.39 0.8 0.1
4.7 VARIATION EFFECT OF MoE OF SANDY SILTSTONE ON
SETTLEMENT w.r.t. DEPTH
In this case, the comparison of the settlement of dam at the crest and up to a depth of
120 m was computed with MCM (condition 1), and HSM (condition 2) applied to
sandy siltstone (foundation ) only for the EoC and AFR as presented in Figure. 4.40
and 4.41 respectively. The MoE of sandy siltstone was varied from 70000 to 125000
kN/m2 with an increment of 10000 kN/m
2. The results indicate that the rate of
increase of settlement computed with MCM (compared with HSM) increased
concerning the dam's depth. At the dam's surface, the rate of increase of settlement
calculated with MCM (compared with HSM) slightly decreased with the rise in MoE
of sandy siltstone from 70000 to 125000 kN/m2. However, at a depth of 80 m and
downwards, the rate of increase of settlement computed with MCM (compared with
HSM) is almost the same irrespective of change in MoE of sandy siltstone.
114
Figure. 4.40: Settlement at EoC by MCM and HSM models.
Figure. 4.41: Settlement for AFR by MCM and HSM models.
0
20
40
60
80
100
120
140
35 40 45 50 55 60 65 70 75 80 85
Dep
th
of
da
m (
m)
Increase of settlement calculated by MCM compared to HSM (%) [EoC]
(MOE) Foundation 70000 kN/m2 (MOE) Foundation 80000 kN/m2
(MOE) Foundation 90000 kN/m2 (MOE) Foundation 100000 kN/m2
(MOE) Foundation 110000 kN/m2 (MOE) Foundation 125000 kN/m2
0
20
40
60
80
100
120
140
35 40 45 50 55 60 65 70 75 80 85
Dep
th
of
da
m (
m)
Increase of settlement calculated by MCM compared to HSM (%) [AFOR]
(MOE) Foundation 70000 kN/m2 (MOE) Foundation 80000 kN/m2
(MOE) Foundation 90000 kN/m2 (MOE) Foundation 100000 kN/m2
(MOE) Foundation 110000 kN/m2 (MOE) Foundation 125000 kN/m2
115
4.8 COMPARISON OF LONG-TERM SETTLEMENT FOR
SANDY SILT STONE
The long-term settlement of the dam was calculated since the dam's foundation
consists of sandy siltstone, which is reported to be weak, fragile with open bedding
planes. The settlement of the dam was calculated for 50 years for both the EoC and
AFR conditions using MCM (condition 1) and HSM (condition 2), as presented in
Figure’s 4.42- 4.43 respectively. The results suggest that the dam's long-term
settlement is almost the same as that occurred AFR. Even though the magnitude of the
settlement predicted with MCM is higher than that of HSM. However, both models
predicted that long-term settlement did not increase concerning the time after filling
the reservoir. The long-term settlement predicted with MCM for both EoC and AFR
conditions was about 51% and 59% higher than HSM. The magnitude of the
settlement is concentrated in clay core, sandy gravel and random fill. The clay core is
compacted during the construction process of the dam. However, clay is considered
compressible material; therefore, the clay's compression mostly occurred during the
construction and after filling the reservoir phases. Therefore, further settlement of the
clay did not happen during the long-term period of 50 years.
Figure. 4.42: Prediction of long-term settlement for sandy
siltstone (MoE = 125,000 kN/m2).
116
Figure. 4.43: Prediction of long-term settlement for sandy
siltstone (MoE = 70,000 kN/m2).
4.9 RAPID DRAWDOWN
The undrained shear strength of embankment soils, particularly the core, consists of
clay; have an essential role during the rapid drawdown process. This is due to less
time for excess pore pressures to dissipate. As a result, instability of the upstream
slope may occur due to the development of pore pressures. This implies that the
undrained condition is applied for both the strength and stiffness properties of clay.
Therefore, the effective strength parameters of clay were changed into undrained
shear strength. The undrained shear strength of clay was calculated based on
consistency index values [181], which are explained in detail in chapter 3. Besides,
the value of modulus of clay elasticity is taken as 30000 kN/m2 instead of 50000
kN/m2 to consider the effect of saturation of the clay when in contact with the
upstream water level.
Three cases have been discussed regarding the undrained shear strength of the clay
core. The values of undrained shear strength utilized are 20, 25 and 30 kN/m2. In
117
every case, the drawdown is shown at different increasing rates of 0.1, 0.125, 0.25,
0.5 and finally 1 m/day. The drawdown rates of 0.1 m/day and 1 m/day are
respectively called slow drawdown and the embankment dam's rapid drawdown.
4.9.1 Dam Stability During Drawdown under Undrained Shear Strength
of Clay = 20 kN/m2
Figure’s 4.44 shows the factor of safety of the embankment dam versus time at
different rates of drawdown varying from 0.1 m/day to 1 m/day, as mentioned above.
It can be observed that for a rapid drawdown of 1 m/day, the dam has shown enough
stability up to 15 days (15 m lowering of the reservoir depth) at which the factor of
safety was 1.2 [183]. When the dam's rapid drawdown at the rate of 1 m/day was
allowed continuously beyond 15 days, the dam showed instability in terms of the
safety factor, which continually reduced up to 0.92 at 35 days (35 m lowering of the
reservoir). If the reservoir was lowered beyond 35 m depth, then the computed results
showed complete collapse of the soils in the dam; this is because of the quick
lowering of the reservoir, resulting in an imbalance of the water's support the
upstream face.
When the drawdown rate was 0.5 m/day, the dam showed stability up to 32 days (16
m lowering of the reservoir), with a safety factor of 1.2. The dam became unstable
continuously when the reservoir was lowered continuously beyond 32 days.
Figure. 4.44 shows a limiting value of the factor of safety of 1.2 as a straight line. The
curves for the factor of safety plotted for a different rate of lowering the reservoir on
or above the limiting value of 1.2 are considered satisfactory. Further, if the dam is
lowered up to 20 m at a rate of 0.1 m/day up to 200 days, the dam would be safe with
a safety factor of 1.2.
118
Figure. 4.44: Safety factor vs. time while lowering the reservoir
under undrained shear strength of clay = 20 kN/m2.
The failure zones for rapid drawdown at a rate of 1 m/day (15 m depth of lowering of
the reservoir) and slow drawdown at a rate of 0.1 m/day (20 m depth of lowering of
the reservoir) are respectively shown in Figure’s 4.45-4.46. For the rapid drawdown
situation, the failure was initiated in the clay core and then extended to the
downstream side, which is less permeable than the upstream side. The failure zone
was created in clay core for slow drawdown and then extended to the dam's upstream
side.
Figure. 4.45: Dam failure zone when reservoir is lowered @ 1 m/day
to a depth of 15 m (Su=20 kN/m2).
0.9
1
1.1
1.2
1.3
1.4
10
35
60
85
110
135
160
185
210
235
260
285
310
335
360
385
410
435
460
485
Fac
tor
of
safe
ty
Time (days) 1 m/day 0.5 m/day
0.25 m/day 0.125 m/day
0.1 m/day Limiting values of factor of safety
119
Figure. 4.46: Dam failure zone when reservoir is lowered @ 0.1 m/day
to a depth of 20 m (Su=20 kN/m2).
4.9.2 Dam Stability During Drawdown under Undrained Shear Strength
of Clay = 25 kN/m2
The factor of safety versus time for the different lowering of the dam's reservoir is
shown in Figure. 4.47. It can be observed that when the dam is lowered at a quick
drawdown rate of 1 m/day, the dam showed stability up to 18 days (18 m lowering of
the reservoir), having a factor of safety of 1.2. Beyond 18 days, the reservoir lowered
at the same rate indicated instability of the dam.
When the drawdown rate was 0.5 m/day, the dam showed stability up to 42 days (22
m lowering of the reservoir), with a safety factor of 1.2. The dam became unstable
when the reservoir was lowered continuously beyond the 22 m lowering of the
reservoir.
For the slow rate of lowering the reservoir at a rate of 0.1 m/day, the dam showed
stability up to 550 days (55 m lowering of the reservoir). This indicated that the
reservoir could be lowered safely up to a depth of 55 m at a rate of 0.1 m/day.
120
Figure. 4.47: Dam safety factor vs. time under undrained
shear strength of clay = 25 kN/m2.
The failure zones for rapid drawdown at a rate of 1 m/day (18 m depth of lowering of
the reservoir) and slow drawdown at a rate of 0.1 m/day (55 m depth of lowering of
the reservoir) are respectively shown in Figure’s 4.48-4.49. The failure was initiated
in clay core for both situations and then extended to the dam's upstream side.
Figure. 4.48: Dam failure zone when reservoir is lowered @ 1 m/day
for a depth of 18 m (Su=25 kN/m2).
0.9
1
1.1
1.2
1.3
1.4
1.5
10
35
60
85
110
135
160
185
210
235
260
285
310
335
360
385
410
435
460
485
510
535
560
Fac
tor
of
safe
ty
Time (days) 1 m/day 0.5 m/day0.25 m/day 0.125 m/day0.1 m/day Limiting values of factor of safety
121
Figure. 4.49: Dam failure zone when reservoir is lowered @ 0.1 m/day
for a depth of 55 m (Su=25 kN/m2).
4.9.3 Dam Stability During Drawdown under Undrained Shear Strength
of Clay = 30 kN/m2
Figure. 4.50 illustrates that the dam is safe if its reservoir is lowered up to a depth of
55 m even at a rate of 0.25 m/day in 220 days. If the reservoir's lowering is performed
quickly at a rate of 1 m/day, the dam was stable up to a depth of 23 m only.
It is also interesting to note that for all rates of lowering of the reservoir, the factor of
safety initially decreased and then gradually increased even when the lowering was
continued beyond a depth of 35 m. This is because the clay core and upstream
material, that is, sandy gravel, are more comprehensive at the base as compared to the
top surface. Due to more width of clay core and upstream sandy gravel at more
profound layers of the dam, the soils became more hardened; therefore, safety was
increased.
In general, a decrease in the drawdown rate showed an increase in the dam's safety
factor. This indicates that the dam's stability became improved when the rate of
drawdown was relatively reduced.
122
Figure. 4.50: Dam safety factor vs. time under undrained
shear strength of clay = 30 kN/m2.
The depth of lowering of the reservoir for different drawdown rates for three cases of
undrained shear strength of clay core 20 kN/m2, 25 kN/m
2 and 30 kN/m
2 were
compared in tables 4.7, 4.8 and 4.9 respectively. It is observed that for a rapid
drawdown of 1 m/day, the permissible depth of the reservoir's lowering was 15 m, 18
m and 23 m, respectively, for the above-mentioned undrained shear strength.
For a slow drawdown of 0.1 m/day, the permissible depth of the reservoir's lowering
was 20 m (undrained shear strength 20 kN/m2) and 55 m for undrained shear strength
of 25 kN/m2 and 30 kN/m
2.
0.9
1
1.1
1.2
1.3
1.4
1.5
10
35
60
85
110
135
160
185
210
235
260
285
310
335
360
385
410
435
460
485
510
535
560
Fac
tor
of
safe
ty
Time (days) 1 m/day 0.5 m/day0.25 m/day 0.125 m/day0.1 m/day Limiting values of factor of safety
123
Table 4.7: Reservoir drawdown rate vs. dam safety factor under
undrained shear strength of clay core = 20 kN/m2.
Drawdown
(m/day) Lowering of the reservoir (m) Days Safety factor
1 15 15 1.2
0.5 16 32 1.2
0.25 18 72 1.2
0.125 19 152 1.2
0.1 20 200 1.2
Table 4.8: Reservoir drawdown rate vs. dam safety factor under
undrained shear strength of clay core = 25 kN/m2.
Drawdown
(m/day) Lowering of the reservoir (m) Days Safety factor
1 18 18 1.2
0.5 22 42 1.2
0.25 24 96 1.2
0.125 55 440 1.22
0.1 55 550 1.24
Table 4.9: Reservoir drawdown rate vs. dam safety factor under
undrained shear strength of clay core = 30 kN/m2.
Drawdown
(m/day) Lowering of the reservoir (m) Days Safety factor
1 23 23 1.2
0.5 26 52 1.2
0.25 55 220 1.2
0.125 55 440 1.26
0.1 55 550 1.3
The failure zones for rapid drawdown at a rate of 1 m/day (23 m depth of lowering of
the reservoir) and slow drawdown at a rate of 0.1 m/day (55 m depth of lowering of
the reservoir) are respectively shown in Figure’s 4.51 and 4.52. The failure was
initiated in clay core for both situations and then extended to the dam's upstream side.
124
Figure. 4.51: Dam failure zone when reservoir is lowered @ 1 m/day
for a depth of 23 m (Su=30 kN/m2).
Figure. 4.52: Dam failure zone when reservoir is lowered @ 0.1 m/day
for a depth of 55 m (Su=30 kN/m2).
4.10 PORE PRESSURE UNDER RAPID DRAWDOWN
In addition to stability during rapid drawdown and slow drawdown of the dam, it is
also necessary to observe how the excess pore pressures have developed and
dissipated concerning time.
Development and dissipation of excess pore pressures for slow drawdown (0.1 m/day)
and rapid drawdown (1 m/day) for various depths and undrained shear strength of 20,
25 and 30 kN/m2 are shown in (Figure’s 4.53-4.64).
It can be observed that for undrained shear strength of 20, 25 and 30 kN/m2, the
excess pore pressure reduced when the drawdown was carried out at a slow rate of
125
(0.1 m/day) up to a depth of 20 m as compared to rapid drawdown rate of (1 m/day).
On the other hand, if the depth of the reservoir's lowering was deeper (i.e., more than
20 m), then for undrained shear strength of 25 and 30 kN/m2, the excess pore
pressures were almost the same during both slow drawdown and rapid drawdown
rates. The increase of excess pore pressures during rapid drawdown (1 m/day) and
decrease of excess pore pressures during slow drawdown (0.1 m/day) is related to the
permeability of clay core [44]. For all the cases of undrained shear strength of 20, 25
and 30 kN/m2, the rate of dissipation of suction is lower during the slow drawdown;
this implies that the dam's slope is stabilized during the slow drawdown.
Table 4.10: Development and dissipation of excess pore pressure
under various drawdown rates.
Undrained
shear strength
Su (kN/m2)
Drawdown
rate
(m/day)
Depth of
drawdown
(m)
Days
Excess pore
pressure
(kN/m2)
Suction
(kN/m2)
20 1.0 10 10 360 280
20 0.1 10 100 225 225
20 1.0 20 20 370 160
20 0.1 20 200 280 140
25 1.0 20 20 325 175
25 0.1 20 200 275 150
25 1.0 55 55 540 20
25 0.1 55 550 540 20
30 1.0 20 20 300 175
30 0.1 20 200 275 150
30 1.0 55 55 540 20
30 0.1 55 550 540 20
126
Figure. 4.53: Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 10 m depth in 10 days (Su=20 kN/m2).
Figure. 4.54: Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 10 m depth in 100 days (Su=20 kN/m2).
127
Figure. 4.55: Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 20 m depth in 20 days (Su=20 kN/m2).
Figure. 4.56: Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 20 m depth in 200 days (Su=20 kN/m2).
128
Figure. 4.57: Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 20 m depth in 20 days (Su=25 kN/m2).
Figure. 4.58: Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 20 m depth in 200 days (Su=25 kN/m2).
129
Figure. 4.59: Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 55 m depth in 55 days (Su=25 kN/m2).
Figure. 4.60: Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 55 m depth in 550 days (Su=25 kN/m2).
130
Figure. 4.61: Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 20 m depth in 20 days (Su=30 kN/m2).
Figure. 4.62: Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 20 m depth in 200 days (Su=30 kN/m2).
131
Figure. 4.63: Development of excess pore pressure under lowering of
reservoir @ 1 m/day for 55 m depth in 55 days (Su=30 kN/m2).
Figure. 4.64: Development of excess pore pressure under lowering of
reservoir @ 0.1 m/day for 55 m depth in 550 days (Su=30 kN/m2).
132
4.11 EFFECT OF RAINFALL ON SLOPE STABILITY OF
DAM
The stability of the dam was assessed for possible rainfall at the dam site. The
maximum rainfall during the last 40 years was 435 mm/year. The rainfall intensity of
435 mm/year was applied equally in 5 days to obtain the rainfall's worst effect on the
dam's stability. It is to be noted that the region in which the dam is located is
generally dry, and low rainfall intensity has occurred there in the last four decades.
This rainfall was converted to m/day and applied to the dam's crest and downstream
boundaries. The effect of rainfall was calculated for 1 to 5 days. The safety factor was
calculated for the impact of rainfall on the dam's stability, as shown in Figure. 4.65.
The dam's safety factor was reduced as the rainfall was continuously poured for 1 to 5
days. The dam's safety factor was 1.494 when it was subjected to rainfall intensity for
one day. The safety factor reduced from 1.494 to 1.439 when the rainfall continuously
poured for five days with the same intensity. The reason for the lowering of the safety
factor is that with the continuous pouring of rainfall on the crest and downstream
portion of the dam, the soils get saturated and increase pore pressures. As a result,
there was a reduction in strength and factor of safety. Overall, the dam's safety is
considered satisfactory even for the continuous rainfall of 5 days with the same
intensity.
The potential failure zones of the dam are shown in Figure’s 4.66-4.67, respectively.
It can be observed that the failure zones due to rainfall along with the highest
reservoir level have developed on the downstream side, and the effect of rainfall can
be seen on the clay core of the dam.
133
Figure. 4.65: Effect of rainfall on dam stability.
Figure. 4.66: Dam failure zone due to 1-day precipitation.
Figure. 4.67: Dam failure zone due to 5-day precipitation.
1.42
1.44
1.46
1.48
1.5
1 2 3 4 5
Fact
or
of
safe
ty
Time (Days)
134
CHAPTER 5
CONCLUSION AND SUGGESTIONS
5.1 CONCLUSION
A numerical analysis of an embankment dam's stability and settlement called Nai Gaj
Dam is presented in this study. Numerical simulations of stability and settlement for
staged construction and filling of the reservoir are illustrated. Parametric analysis of
stability was carried out to determine the strength parameters' effect on the dam's
stability. For the computation of stability, strength parameters such as the friction
angle of the main zones (sandy gravel, random fill, and clay core) were gradually
lowered. The values of the MoE of four zones that occupied more volume (sandy
gravel, random fill, clay core, and sandy siltstone) were incrementally lowered for
calculation of settlement. Besides, the lowering of the reservoir and the maximum
possible rainfall on the dam's stability are investigated. The main conclusions drawn
from this study are presented below:
The stability of the dam was evaluated in terms of the safety factor. The safety factor
values for the end of construction and after filling the reservoir are respectively 1.6
and 1.5. The results suggest that the dam's safety factor is satisfactory for both ends of
construction and after filling of the reservoir conditions [78].
Parametric analysis of the dam was conducted to adopt a reasonable magnitude of
MoE of those zones which have a significant influence on the settlement:
o The results suggest that out of the four major zones of the dam (clay core,
sandy gravel, random fill and sandy siltstone), sandy siltstone (Foundation
soil) has a significant influence on the settlement of the dam when computed
with MCM and HSM.
134
135
o As per the study's correlations, the maximum and minimum value of the sandy
siltstone's MoE was from 70,000 kPa to 125,000 kPa.
o Based on numerical analysis, the dam's settlement is predicted to be 1.35% of
the dam's height when the MoE of the sandy siltstone is 125000 kPa.
o The magnitude of the dam's predicted settlement agrees with existing available
literature when the value of MoE of sandy siltstone is taken as 125,000 kPa.
o It was observed that the safe rate of the lowering of the reservoir mainly
depends on the magnitude of the undrained shear strength of the clay core.
o The recommended value of the clay's undrained shear strength can be taken as
25 kN/m2. The computed results indicate that the dam's reservoir could be
lowered safely up to 20 m and 55 m depth if the drawdown rates are
respectively as 1 m/day (rapid drawdown) and 0.1 m/day (slow drawdown).
o For the drawdown mentioned above rates, the dam's obtained safety factor was
1.2, which is considered satisfactory as per guidelines.
o Based on the maximum possible rainfall intensity of 435 mm/year, the dam's
embankment was subjected to maximum rainfall intensity for 1 to 5 days. The
stability of the dam was affected as the duration of rainfall was extended. The
computed safety factors were 1.494 and 1.439, respectively, for 1 to 5 days of
rainfall. Overall, the stability of the dam was satisfactory during the rainfall-
induced response.
o This study shows that parametric analysis based on the finite element method
plays a vital role in estimating suitable values of MoE of the material zones
whenever the said values are not determined experimentally.
5.2 RECOMMENDATIONS FOR FUTURE STUDIES
Based on the results and discussions of the numerical analysis of stability and
settlement response of Nai Gaj dam, future studies may be conducted on the following
aspects of the numerical modeling:
o The effect of seepage on the dam's stability be assessed numerically, and the
results are compared with the field measurements of steady-state seepage
conditions.
136
o Numerical analysis of any possibility of internal erosion through the
embankment and its foundation be conducted.
o A numerical analysis of possible uplift pressure on the downstream side of the
dam is performed.
o The dam's stability is checked against shaking the maximum possible
earthquake in the dam area, and remedial measures are suggested to enhance
dam stability.
137
REFERENCES
[1] Foster, M., Fell. R., and Spannagle. M. (2000), ―The statistics of embankment
dam failures and accidents‖, Canadian Geotechnical Journal, Vol. 37, No. 5
pp. 1000-1024.
[2] Panulinova, E., and Harabinova. S. (2014), ―Methods for analyzing the
stability of an earthen dam slope‖, Advanced Materials Research, Trans Tech
Publ, Vol. 969, pp. 245-248.
[3] Ozcoban, S., Berilgen. M. M., Kilic. H., Edil. T. B., and Kutay Ozaydin. I.
(2007), ―Staged construction and settlement of a dam founded on soft clay‖,
Journal of geotechnical and geoenvironmental engineering, Vol. 133, No. 8,
pp. 1003-1016.
[4] Al-Homoud, A. S. and Tanash. N. (2001), ―Monitoring and analysis of
settlement and stability of an embankment dam constructed in stages on soft
ground‖, Bulletin of Engineering Geology and the Environment, Vol. 59, No.
4, pp. 259-284.
[5] Waltham, T. (2018), ―St Francis The world's worst dam site‖, Geology Today,
Vol. 34, No. 3, pp. 100-108.
[6] Pinyol, N. M., Alonso. E. E., and Olivella, S. (2008), ―Rapid drawdown in
slopes and embankments‖, Water resources research, Vol. 44, No. 5, pp. 1-22.
[7] Zhang, L. L., Zhang. J., Zhang. L. M., and Tang, W. H. (2011), ―Stability
analysis of rainfall-induced slope failure‖, Proceedings of the Institution of
Civil Engineers-Geotechnical Engineering, Vol. 164, No. 5, pp. 299-316.
[8] Sjodahl, P., Dahlin. T. and Zhou. B. (2006), ―2.5 D resistivity modeling of
embankment dams to assess influence from geometry and material properties
2.5D Resistivity modeling of embankment dams‖, Geophysics, Vol. 71, No. 3,
pp. G107-G114.
[9] Al-Homoud, A. S. and Tanash. N. (2004), ―Modeling uncertainty in stability
analysis for design of embankment dams on difficult foundations‖,
Engineering Geology, Vol. 71, No. 3-4, pp. 323-342.
[10] Ural, N. (2018), ―The Importance of Clay in Geotechnical Engineering‖,
Current Topics in the Utilization of Clay in Industrial and Medical
Applications, ISBN 978-1-78923-728-3, pp. 83.
[11] Yi, P., Liu. J. and Xu, C. (2015), ―Reliability analysis of high rockfill dam
stability‖, Mathematical Problems in Engineering pp. 1-8.
138
[12] Smith. N. (1971), ―A history of dams‖, Publisher Peter Davies, London.
[13] Schnitter. N. (1994), ―A history of dams‖, The useful pyramids Rotterdam,
Brookfield A. A Balkema Verlag, ISBN 9054101490.
[14] Saxena, K. R., and Sharma, V. (2004), ―Dams Incidents and accidents‖, CRC
Press, ISBN 90 5809 701 3.
[15] Tanchev. L. (2014), ―Dams and appurtenant hydraulic structures‖, 2nd
Edition, CRC Press, ISBN 9780429211812.
[16] Hughes, A. K. (1981), ―The erosion resistance of compacted clay fill in
relation to embankment overtopping‖, PhD Thesis, Newcastle University.
[17] Cabrera. E. (2010), ―Water Engineering and Management through Time‖,
Learning from History, 1st edition, ISBN 9780429213182.
[18] Nef, J. U., (1977), ―An early energy crisis and its consequences‖, Scientific
American, Vol. 237, No. 5 pp. 140-151.
[19] Haidvogl, G. (2018), ―Historic milestones of human river uses and ecological
impacts‖, Riverine Ecosystem Management, Springer, Cham, Vol. 8, pp. 19-
39 ISBN 978-3-319-73249-7.
[20] Rigby, P. J., Thompson, A. N., and Jones, D. E. (2014), ―The nearly perfect
nineteenth century embankment dam. In Maintaining the Safety of our Dams
and Reservoirs‖, Proceedings of the 18th Biennial Conference of the British
Dam Society at Queen’s University, Belfast, ICE Publishing, pp. 88-100.
[21] Ho, M., Lall, U., Allaire, M., Devineni, N., Kwon, H. H., Pal, I. and Wegner,
D. (2017), ―The future role of dams in the United States of America‖, Water
Resources Research, Vol. 53, No. 2, pp. 982-998.
[22] Zhang, L., Peng, M., Chang, D. and Xu, Y. (2016), ―Dam failure mechanisms
and risk assessment‖, 1st edition, John Wiley and Sons, ISBN
9781118558515.
[23] Atallah, T. A. (2002), ―A review on dams and breach parameters estimation‖,
Masters Thesis, Virginia Polytechnic Institute and State University
Blacksburg, VA.
[24] Rubertis, K. P. E. de (2018), ―Monitoring Dam Performance Instrumentation
and Measurements‖, 1st edition, ISBN 9780784414828.
[25] Knappett. J. and Craig, R. F. (2012), ―Craig's soil mechanics‖, 8th
edition,
ISBN 978-0-415-56125-9.
139
[26] Chen, S. H. and Chen, M. (2015), ―Hydraulic structures‖, 1st edition, ISBN
978-3-662-47331-3.
[27] Duggal. K. (1996), ―Elements of water resources engineering‖, 1st edition
ISBN 81-224-0807-9.
[28] Kutzner. C. (2018), ―Earth and rockfill dams: Principles for design and
construction‖, 1st edition, ISBN 90 5410 682 4.
[29] Yusof, Z. M., Wahab, A. K. A., Ismail, Z., and Amerudin, S. (2018),
―Characteristics of Non-Cohesive Embankment Failure‖, IOP Conference
Series, Earth and Environmental Science, Vol. 140, No. 1, pp. 1-7.
[30] Khor, C. H. and Woo. H. K. (1989), ―Investigation of crushed rock filters for
dam embankment‖, Journal of geotechnical engineering, Vol. 115, No. 3 pp.
399-412.
[31] Singh, B. and Varshney. R. S., (1995), ―Engineering for embankment dams‖,
A. A. Balkema, Vol. 745, pp. 745.
[32] Duncan, J. M., Wright. S. G., and Brandon. T. L. (2014), ―Soil strength and
slope stability‖, 2nd
edition, ISBN 978-1-118-91796-1.
[33] Kayyal, M. and Wright. S. (1991), ―Investigation of long-term strength
properties of paris and beaumont clays in earth embankments‖, Final
Report (No. FHWA/TX-92+1195-2F).
[34] Imbrogno, D. F. (2014), ―Analysis of dam failures and development of a dam
safety evaluation program‖, Masters Thesis, The Ohio State University.
[35] Saluja, I. S., Athar. M. and Ansari, S. A. (2018), ―Causes of Failure of Earthen
Dams and suggested Remedial Measures‖, Vol. 5, No. 3, pp. 177-185.
[36] Talukdar, P. and Dey. A. (2019), ―Hydraulic failures of earthen dams and
embankments‖, Innovative Infrastructure Solutions, Vol. 4, No. 1, pp. 42.
[37] Ashraf, M., Soliman, El-Ghorab, A. H., E. and Zawahry, A. E. (2018),
―Assessment of embankment dams breaching using large scale physical
modeling and statistical methods‖, Water Science, Vol. 32, No. 2 pp. 362-379.
[38] Dhiman, S. and Patra. K. C. (2018), ―Experimental study of embankment
breach based on its soil properties‖, ISH Journal of Hydraulic Engineering,
Vol. 26, No. 1, pp. 1-11.
140
[39] Ormann, L., Zardari, M. A., Mattsson, H., Bjelkevik, A. and Knutsson, S.
(2013), ―Numerical analysis of strengthening by rockfill embankments on an
upstream tailings dam‖, Canadian Geotechnical Journal, Vol. 50, No. 4 pp.
391-399.
[40] Zardari, M. A., Mattsson, H., Knutsson, S., Khalid, M. S., Ask, M. V. and
Lund, B. (2017), ―Numerical analyses of earthquake induced liquefaction and
deformation behaviour of an upstream tailings dam‖, Advances in Materials
Science and Engineering, Vol. 2017, Article ID 5389308, pp. 1-12.
[41] Zardari. M. A., Mattsson. H., Knutsson. S. and Ormann, L. (2017),
―Comparison of three-dimensional and two-dimensional axisymmetric finite
element analyses of a corner section of a tailings dam‖, Scientia Iranica
International Journal of Science and Technology, Vol. 24, No. 5, pp. 2320-
2331.
[42] Linsley Jr, R. K., Kohler, M. A. and Paulhus, J. L. (1975), ―Hydrology for
engineers‖, 2nd
edition, ISBN 0070379673.
[43] Budhu. M. (2010), ―Soil Mechanics and Foundation‖, 3rd edition, ISBN 978-
040-55684-9.
[44] Vandenberge, D.R., (2014), ―Total stress rapid drawdown analysis of the
Pilarcitos Dam failure using the finite element method‖, Vol. 8, No. 2, pp.
115-123.
[45] Duncan, J. M., Wright, S. and Wong, K. S. (1990), ―Slope stability during
rapid drawdown‖, Vol. 2, pp. 253-272.
[46] Fredlund, M., Lu H. and Feng, T. (2011), ―Combined seepage and slope
stability analysis of rapid drawdown scenarios for levee design‖, Geo-
Frontiers, pp. 1595-1604.
[47] Souliyavong, T., Gallage, C. , Egodawatta, P. and Maher, B. (2012), ―Factors
affecting the stability analysis of earth dam slopes subjected to reservoir
drawdown‖, Proceedings of the Second International Conference on
Geotechnique, Construction Materials and Environment, The GEOMATE
International Society, pp. 507-512.
[48] Berilgen, M. M. (2007), ―Investigation of stability of slopes under drawdown
conditions‖, Vol. 34, No. 2 pp. 81-91.
[49] Griffiths, D. and P. Lane, (1999), ―Slope stability analysis by finite elements‖,
Vol. 49, No. 3, pp. 387-403.
[50] Viratjandr, C. and Michalowski R. L. (2006), ―Limit analysis of submerged
slopes subjected to water drawdown‖, Vol. 43, No. 8, pp. 802-814.
141
[51] Dawson, E., Roth, W. and Drescher, A. (1999), ―Slope stability analysis by
strength reduction‖, Vol. 49, No. 6 pp. 835-840.
[52] Lane, P. and Griffiths D. (2000), ―Assessment of stability of slopes under
drawdown conditions‖, Vol. 126, No. 5, pp. 443-450.
[53] Morgenstern, N. (1963), ―Charts for Earth Slopes During Rapid Drawdown‖,
Vol. 13, No. 2, pp. 121-131.
[54] Desai, C. S. (1977), ―Drawdown analysis of slopes by numerical method‖,
Journal of Geotechnical and Geoenvironmental Engineering, Vol. 103, No.
GT7, pp. 667-676.
[55] Cousins, B. F. (1978), ―Stability charts for simple earth slopes‖, Journal of the
Geotechnical Engineering Division, Vol. 104, No. 2, pp. 267-279.
[56] Kemal, S. and Aoyama, S. (2002), ―Pore Pressure Behavior on Embankment
Dams during Construction‖, Transactions-japanese society of irrigation
drainage and reclamation engineering, pp. 25-34.
[57] St-Arnaud, G. (1995), ―The high pore pressures within embankment dams an
unsaturated soil approach‖, Canadian Geotechnical Journal, Vol. 32, No. 5,
pp. 892-898.
[58] Nasrollahi, S. (2008), ―Determination of Excess Pore Pressure in Earth Dam
after Earthquake‖, 14th World Conference on Earthquake Engineering, pp. 1-
11.
[59] Hosseini, S., Mohammad, M. and Fard, R. A. (2003), ―Pore pressure
development in the core of earth dams during simultaneous construction and
impounding‖, Electronic Journal of Geotechnical Engineering, Vol. 8, No.1,
pp. 1-13.
[60] Vafaeian, M., Mansoorzadeh, S. and Sadeqhi, M. M. (2005), ―Influential
factors in evaluating the construction pore pressure in earth dams by using
UDAM program‖, Geotechnical and Geological Engineering, Vol. 23, No. 1,
pp. 87.
[61] Preziosi, M. C. (2013), ―The probabilistic assessment of small homogeneous
UK earthfill dams affected by climate change Precipitation‖, PhD thesis, City
University London.
[62] Morton, K. L., Muresan, M. C. and Ramsden, F. (2008), ―Importance of pore
pressure monitoring in high walls‖, Journal of the Southern African Institute
of Mining and Metallurgy, Vol. 108, No. 11, pp. 661-667.
142
[63] Li, Z., Ye, W., Marence, M. and Bricker, J. (2019), ―Unsteady Seepage
Behavior of an Earthfill Dam During Drought-Flood Cycles‖, Geosciences,
Vol. 9, No. 1, pp. 1-15.
[64] Abbas, J. M., Aljanabi, Q. A. and Mutiny, Z. A. (2017), ―Slope stability
analysis of an earth dam‖, Diyala journal of engineering sciences, Vol. 10, No.
1, pp. 106-117.
[65] Fellenius, W. (1936), ―Calculation of stability of earth dam‖, Transactions,
2nd Congress Large Dams, Washington, DC, Vol. 4, pp. 445-462.
[66] Morgenstern, N. and Bishop, A. (1960), ―Stability coefficients for earth
slopes‖, Geotechnique, Vol. 10, No. 4, pp. 129-153.
[67] Lowe, J. (1960), ―Stability of earth dams upon drawdown‖, 1st Pan American
Conference on Soil Mechanics and Foundation Engineering, Mexico City,
Vol. 3, OCLC No. 220350245.
[68] Janbu, N. (1968), ―Slope stability computations‖, soil mechanics and
foundation engineering report. Technical University of Norway, Trondheim.
[69] Morgenstern, N. U. and Price, V. E. (1965), ―The analysis of the stability of
general slip surfaces‖, Geotechnique, Vol. 15, No. 1, pp. 79-93.
[70] Spencer, E. (1967), ―A method of analysis of the stability of embankments
assuming parallel inter-slice forces‖, Geotechnique, Vol. 17, No. 1, pp. 11-26.
[71] Duncan, J. and Wright, S. (1980), ―The accuracy of equilibrium methods of
slope stability analysis‖, Engineering geology, Vol. 16, No. 1-2, pp. 5-17.
[72] Pereira, T. D. S., Robaina, A. D., Peiter, M. X., Braga, F. D. V. and Rosso, R.
B. (2016), ―Performance of Analysis Methods of Slope Stability for Different
Geotechnical Classes Soil on Earth Dams‖, Engenharia Agricola, Vol. 36, No.
6, pp. 1027-1036.
[73] Burman, A., Acharya, S. P., Sahay, R. R. and Maity, D. (2015), ―A
comparative study of slope stability analysis using traditional limit equilibrium
method and finite element method‖, Asian journal of civil engineering
(building and housing), Vol. 16, No. 4, pp, 467-492.
[74] Pham, H. T., Oo, H. Z., & Jing, C. (2013). Stability of slope and seepage
analysis in earth dam using numerical finite element model. Study of Civil
Engineering and Architecture (SCEA), 2(4), 104-108.
[75] PLAXIS, 2D. (2017), ―Reference Manual‖, Delft, The Netherlands.
143
[76] Griffiths, D. V., & Marquez, R. M. (2007). Three-dimensional slope stability
analysis by elasto-plastic finite elements. Geotechnique, 57(6), 537-546.
[77] Brinkgreve, R. B. J. and Bakker, H. L. (1991), ―Non-linear finite element
analysis of safety factors‖, International conference on computer methods and
advances in geomechanics. Vol. 7, pp 1117-1122.
[78] Army, U. S. (2003), ―Engineering and design slope stability‖, Engineering
manual, Department of the Army, Corps of Engineers, Office of the Chief of
Engineers, ISBN. 1110-2-1902.
[79] Persson, E. (2017), ―Empirical correlation between undrained shear strength
and preconsolidation pressure in Swedish soft clays‖, Master Thesis, KTH,
School of Architecture and the Built Environment (ABE).
[80] Reeves, G. M., Sims, I. and Cripps, J. (2006), ―Clay materials used in
construction‖, Geological Society of London, Publication. 21, ISBN 978-
186239-184.
[81] Suzuki, K. and Yasuhara, K. (2007), ―Increase in undrained shear strength of
clay with respect to rate of consolidation‖, Soils and foundations, Vol. 47, No.
2, pp. 303-318.
[82] Mayne, P. W. (2007), ―Cone Penetration Test‖, A Synthesis of Highway
Practice‖, Washington, D. C. Transportation Research Board, ISBN 978-0309-
09784-0.
[83] Robertson, P. (2009), ―Interpretation of cone penetration tests a unified
approach‖, Canadian geotechnical journal, Vol. 46, No. 11, pp. 1337-1355.
[84] Hossain, M. I. (2018), ―Evaluation of Undrained Shear Strength and Soil
Classification from Cone Penetration Test‖, Masters Thesis, Louisiana State
University and Agricultural and Mechanical College.
[85] Najjar, S. S., Yaghi, K., Adwan, M. and Jaoude, A. A. R. A. (2015), ―Drained
shear strength of compacted sand with clayey fines‖, International Journal of
Geotechnical Engineering, Vol. 9, No. 5, pp. 513-520.
[86] Tembe, S., Lockner, D. A. and Wong, T. F. (2010), ―Effect of clay content and
mineralogy on frictional sliding behavior of simulated gouges Binary and
ternary mixtures of quartz, illite, and montmorillonite‖, Journal of
Geophysical Research Solid Earth, Vol. 115, No. B3, pp. 1-22.
[87] Akayuli, C., Ofosu, B., Nyako, S. O., and Opuni, K. O. (2013), ―The influence
of observed clay content on shear strength and compressibility of residual
sandy soils‖, IJERA, Vol. 3, No. 4, pp. 2538-2542.
144
[88] Chen, Y. and C. L. Meehan, (2011), ―Undrained strength characteristics of
compacted bentonite/sand mixtures‖, Geo-Frontiers Advances in Geotechnical
Engineering, pp. 2699-2708.
[89] Haider, M. M., Huat, B. B. K., Malek, M. A., and Dis, M. N. (2011), ―Effect
of mixing fine sand on the drained shear strength of completely decomposed
granite soil‖, International journal of Geomate, Vol. 1, No. 1, pp. 10-18.
[90] Dimitrova, R. S. and Yanful, E. K. (2012), ―Factors affecting the shear
strength of mine tailings/clay mixtures with varying clay content and clay
mineralogy‖, Engineering geology, Vol. 125, pp. 11-25.
[91] Otoko, G. R. (2014), ―Dependence of shear strength and compressibility of
tropical lateritic soils on clay content‖, International Journal of Engineering
and Technology Research, Vol. 2, No. 2, pp. 1-9.
[92] Behnia, D., MoEinossadat, S. R., Behnia, B., Behnia, M., Safari-Gorgi, A.,
and Zakerian, P. (2014), ―Prediction of settlement in sloping core rockfill
dams using soft-computing‖, Research in Civil and Environmental
Engineering (RCEE), Vol. 2, No. 2, pp. 55-65.
[93] Task Committee, ASCE. (2000), ―Guidelines for instrumentation and
measurements for monitoring dam performance‖, American Society of Civil
Engineers, ISBN 0-7844-0531-X.
[94] Hunter, G., and Fell, R. (2003), ―The deformation behaviour of embankment
dams‖, University of New South Wales, School of Civil and Environmental
Engineering, editionUNICIV Report No. 416, ISBN, 85841-383-3.
[95] Huzjak, R. J., and Prochaska, A. B. (2009), ―Settlement of embankment dams
don’t forget about the bedrock‖, In Proceding Association of State Dam Safety
Ojjicials, Dam Safety Conference (CD-ROM), ASDSO, Lexington, KY, pp 1-
25.
[96] Nadia, S. and R. Bouchrit, (2017), ―Influence of deformation moduli on the
settlement of earth dams‖, International journal of geomate, Vol. 13, No. 38,
pp. 16-22.
[97] Javanmard, M., Amiri, F., and Safavi, S. M. (2019), ―Instrumentation
Readings versus Numerical Analysis of Taham Dam‖,. International Journal
of Engineering, Vol. 32, No. 1, pp. 28-35.
[98] Keyvanipour, M., Moharrampour, M., and Faghih, S. (2012), ―An Evaluation
and Comparison of Bar Embankment Behavior with Instrumentation Data and
Software PLAXIS‖, International Conference on Environment Science and
Engineering, Vol. 32, pp. 197-200.
145
[99] Rashidi, M. and Haeri, S. M. (2017), ―Evaluation of behaviors of earth and
rockfill dams during construction and initial impounding using
instrumentation data and numerical modeling‖, Journal of Rock Mechanics
and Geotechnical Engineering, Vol. 9, No. 4, pp. 709-725.
[100] Sherard J. L. (1963), ―Earth and earth-rock dams‖, John Wily and sons, ISBN
978-0471785477.
[101] Nayebzadeh, R. and Mohammadi, M. (2011), ―The effect of impervious clay
core shape on the stability of embankment dams‖, Geotechnical and
Geological Engineering, Vol. 29, No. 4, pp. 627-635.
[102] Ghafari, A., Nikraz, H. R., and Sanaeirad, A. (2016), ―Finite element analysis
of deformation and arching inside the core of embankment dams during
construction‖, Australian Journal of Civil Engineering, Vol. 14, No. 1, pp. 13-
22.
[103] Ng, A. K. and Small, J. C. (1999), ―A case study of hydraulic fracturing using
finite element methods‖, Canadian geotechnical journal, Vol. 36, No. 5, pp.
861-875.
[104] Sherard, J. (1991), ―Cracking and hydraulic fracturing in Embankment dams‖,
US Army Corps of Engineers, p. 301-327.
[105] Ono, K., and Yamada, M. (1993), ―Analysis of the arching action in granular
mass‖, Geotechnique, Vol. 43, No. 1, pp. 105-120.
[106] Talebi, M., Vahedifard, F., and Meehan, C. L. (2013), ―Effect of
geomechanical and geometrical factors on soil arching in zoned embankment
dams‖, Geo-Congress, pp. 1056-1065.
[107] Sherard, J. L. (1987), ―Lessons from the Teton Dam failure‖, Engineering
Geology, Vol. 24, No. 1-4, pp. 239-256.
[108] Maksimovic, M. (1973), ―Optimal position of the central clay core of a
rockfill dam in respect to arching and hydraulic fracture‖, International
Congress on Large Dams, Vol. 1973, pp. 789-800.
[109] Jafarzadeh, F. and Garakani, A. A. (2013), ―Foundation Treatment of
Embankment Dams With Combination of Consolidation and Compaction
Groutings‖, Seventh International Conference on Case Histories in
Geotechnical Engineering, pp. 1-7.
[110] Bidasaria, M. (2004), ―Treatment or Unconformity Zone and Curtain Grouting
in Foundation of Almatti Dam on Krishna River‖, 5th
International conference
on case histories in geotechnical engineering, pp. 1-5.
146
[111] Terlien, M. T. (1998), ―The determination of statistical and deterministic
hydrological landslide-triggering thresholds‖, Environmental geology, Vol.
35, No. 2-3, pp. 124-130.
[112] Chow, V. T, Maidment, D. R. and Mays, L.W. (1988), ―Applied
hydrology‖,Tata McGraw Hill, Second reprint edition, ISBN 0-07-010810-2.
[113] Leong, E. C., Low, B. K., and Rahardjo, H. (1999), ―Suction profiles and
stability of residual soil slopes‖, International Symposium on slope stability
engineering, ISBN 90-5809-081-7, pp. 387-392.
[114] Wilson, G. W. (1990). "Soil evaporative fluxes for geotechnical engineering
problems."
[115] Ng, C. W. W., and Shi, Q. (1998), ―Influence of rainfall intensity and duration
on slope stability in unsaturated soils‖, Quarterly Journal of Engineering
Geology and Hydrogeology, Vol. 31, No. 2, pp. 105-113.
[116] Fourie, A. B., Rowe, D., and Blight, G. E. (1999), ―The effect of inÆltration
on the stability of the slopes of a dry ash dump‖, Geotechnique, Vol. 49, No.
1, pp. 1-13.
[117] Fredlund. D. and Rahardjo, H. (1993), ―Soil mechanics for unsaturated soils‖,
A Wiley-Inter science Publication, 1st edition ISBN, 978-0-471-85008-3.
[118] Rahardjo, H., Lim, T. T., Chang, M. F. and Fredlund, D. G. (1995), ―Shear-
strength characteristics of a residual soil‖, Canadian Geotechnical Journal,
Vol. 32, No. 1, pp. 60-77.
[119] Fourie, A. (1996) , Predicting rainfall-induced slope instability‖, Proceedings
of the Institution of Civil Engineers-Geotechnical Engineering, Vol. 119, No.
4, pp. 211-218.
[120] Zhu, J. H. and Anderson, S. (1998), ―Determination of shear strength of
Hawaiian residual soil subjected to rainfall-induced landslides‖,
Geotechnique,Vol. 48, No.1, pp. 73-82.
[121] Wang, G. and Sassa, K. (2001), ―Factors affecting rainfall-induced flowslides
in laboratory flume tests‖, Geotechnique,Vol. 51, No. 7, pp. 587-599.
[122] Ng, C. and Q. Shi. (1998), ―A numerical investigation of the stability of
unsaturated soil slopes subjected to transient seepage‖, Computers and
geotechnics,. Vol. 22, No. 1, pp. 1-28.
147
[123] Rahardjo, H., Ong, T. H., Rezaur, R. B., and Leong, E. C. (2007), ―Factors
controlling instability of homogeneous soil slopes under rainfall‖, Journal of
Geotechnical and Geoenvironmental Engineering, Vol. 133, No. 12, pp. 1532-
1543.
[124] Schrefler, B. A., and Lewis, R. W. (1998), ―The finite element method in the
deformation and consolidation of porous media‖, Chichester Wiley, ISBN
9780471912101.
[125] Britto, A. M., & Gunn, M. J. (1987). Critical state soil mechanics via finite
elements.
[126] Tschuchnigg, F., Medicus, G., and Schneider-Muntau, B. (2019), ―Slope
stability analysis: Barodesy vs linear elastic–perfectly plastic models‖, E3S
Web of Conferences Vol. 92, pp. 1-5.
[127] Choo, J. (2018), ―Mohr Coulomb plasticity for sands incorporating density
effects without parameter calibration‖, International Journal for Numerical and
Analytical Methods in Geomechanics, Vol. 42, No. 18, pp. 2193-2206.
[128] Zhang, J. and Salgado, R. (2010), ―Stress-dilatancy relation for Mohr
Coulomb soils following a non-associated flow rule‖, Geotechnique, Vol. 60,
No. 3, pp. 223–226.
[129] Lambe. T. W. and Whitman, R. V. (2008), ―Soil mechanics‖, SI version, John
Wiley and Sons, ISBN 10: 8126517794.
[130] Hicher, P. Y. (1996), ―Elastic properties of soils‖, Journal of Geotechnical
Engineering, Vol. 122, No. 8, pp. 641-648.
[131] Newcomb, D. E. and Birgisson, B. (1999), ―Measuring in situ mechanical
properties of pavement subgrade soils‖, Transportation Research Board, ISBN
0-309-06857-6.
[132] Marchenko, M., Mosicheva, I. and Aniskin, A. (2018), ―Estimation of
Poisson's ratio of soil using stiffness of loose soils‖, Vol. 9, No. 16, pp. 85-94.
[133] Chowdhury, R. H., and Azam, S. (2016), ―Unsaturated shear strength
properties of a compacted expansive soil from Regina‖, Canada Innovative
Infrastructure Solutions, Vol. 1, No. 47, pp. 1-11.
[134] Das, B. M., and Sobhan, K. (2013), ―Principles of geotechnical engineering‖,
8th edition, ISBN, 978-1-133-11089-7.
[135] Todd. M. K. (2017), ―Handbook of geotechnical investigation and design
tables‖, 2nd
edition, ISBN, 978-1-315-81323-3.
148
[136] Ishibashi. I., and Hazarika, H. (2015), ―Soil Mechanics Fundamentals and
Applications‖, 2nd
edition, ISBN, 978-1-4822-5042-8. CRC Press.
[137] Bell. F.G. (2013), ―Engineering properties of soils and rocks‖, Elsevier 3rd
edition, ISBN, 0-7506-0489-1.
[138] Rabie, M. (2014), ―Comparison study between traditional and finite element
methods for slopes under heavy rainfall‖, HBRC Journal, Vol. 10, No, 2, pp.
160-168.
[139] Wulandari, P. S., and Tjandra, D. (2015), ―Analysis of geotextile reinforced
road embankment using PLAXIS 2D‖, 5th International Conference of Euro
Asia Civil Engineering Forum Vol. 125, pp. 358-362.
[140] Singh, R., Umrao, R. K., and Singh, T. N. (2014), ―Stability evaluation of
road-cut slopes in the Lesser Himalaya of Uttarakhand‖, India conventional
and numerical approaches, Bulletin of Engineering Geology and the
Environment, Vol. 73, No. 3, pp. 845-857.
[141] Deliveris, A. V., Zevgolis, I. E. and Koukouzas, N. C. (2016), ―Numerical
modelling of slope stability in open pit lignite mines a comparative study‖,
Bulletin of the Geological Society of Greece, Vol. 50, No. 2, pp. 671-680.
[142] Ouadif, L., Bahi, L., Baba, K. and Akhssas, A. (2012), ―Study of stability of a
highway fill application of the analytical and the finite elements methods‖,
International Journal of Engineering Research and Applications, Vol. 2, No. 1,
pp. 904-910.
[143] Ghanizadeh, E. and Ghanizadeh, M. (2017), ―Evaluation of Stability and
Performance of Dam Foundations at Static State‖, European Journal of
Sustainable Development,Vol. 6, No. 2, pp. 272-292.
[144] Afiri, R. and Gabi, S. (2018), ―Finite element slope stability analysis of Souk
Tleta dam by shear strength reduction technique‖, Innovative Infrastructure
Solutions, Vol. 3, No. 1, pp. 1-6.
[145] Vekli, M., Aytekin, M., Ikizler, S. B. and Calik, U. (2012), ―Experimental and
numerical investigation of slope stabilization by stone columns‖, Natural
hazards, Vol. 64, No. 1, pp. 797-820.
[146] Gaber, M., Kasa, A., Abdul-Rahman, N. and Alsharef, J. (2019), ―Simulation
of Sequential Construction of Embankments on Reinforced Soft Clay
Foundation‖, Indian Geotechnical Journal, Vol. 49, No. 2, pp. 224-231.
[147] Schanz, T., Vermeer, P. A. and Bonnier, P. G. (1999), ―The HSM: formulation
and verification‖, in Beyond 2000 in Computational Geotechnics, Balkema,
Rotterdam, The Netherlands, pp. 281–296.
149
[148] Janbu, N. (1963), ―Soil compressibility as determined by odometer and triaxial
tests‖, Europian Conference SMFE, Vol. 1, pp. 19-25.
[149] Von Soos, P. (1990), ―Properties of Soil and Rock‖, (in German), Grundbau
Taschenbuch Part 4, Edition 4, Ernst and Sohn, Berlin.
[150] Brinkgreve, R. B. (2005), ―Selection of soil models and parameters for
geotechnical engineering application‖, Soil constitutive models Evaluation,
selection, and calibration, ASCE, pp. 69-98.
[151] Celik, S. (2017), ―An experimental investigation of utilizing waste red mud in
soil grouting‖, KSCE Journal of Civil Engineering, Vol. 21, No. 4, pp. 1191-
1200.
[152] Teo, P. L. and Wong, K. S. (2012), ―Application of the HSM in deep
excavation analysis‖, The IES Journal Part A Civil and Structural Engineering,
Vol. 5, No. 3, pp. 152-165.
[153] Pramthawee, P., Jongpradist, P. and Kongkitkul, W. (2011), ―Evaluation of
HSM on numerical simulation of behaviors of high rockfill dams‖,
Songklanakarin Journal of Science and Technology, Vol. 33, No. 3, pp. 325-
334.
[154] Ozkuzukiran, S., Ozkan, M. Y., Ozyazicioğlu, M. and Yildiz, G. S. (2006),
―Settlement behaviour of a concrete faced rock-fill dam‖, Geotechnical and
Geological Engineering, Vol. 24, No. 6, pp. 1665-1678.
[155] Varga, G. and Czap, Z. (2004), ―Soil models safety factors and settlements‖,
Periodica Polytechnica Civil Engineering, Vol. 48, No. 1-2, pp. 53-63.
[156] Dickin, E. A. and Laman, M. (2007), ―Uplift response of strip anchors in
cohesionless soil‖, Advances in Engineering Software, Vol. 38, No. 8-9, pp.
618-625.
[157] Locat, A., Leroueil, S., Fortin, A., Demers, D. and Jostad, H. P. (2014), ―The
1994 landslide at Sainte-Monique, Quebec geotechnical investigation and
application of progressive failure analysis‖, Canadian Geotechnical Journal,
Vol. 52, No. 4, pp. 490-504.
[158] Singh, V. P. and Babu, G. S. (2010), ―2D numerical simulations of soil nail
walls‖, Geotechnical and Geological Engineering, Vol. 28, No. 4, pp. 299-309.
[159] Phien-Wej, N., Humza, M. and Aye, Z. Z. (2012), ―Numerical modeling of
diaphragm wall behaviour in Bangkok soil using HSM‖, Geotechnical Aspects
of Underground Construction in Soft Ground Viggiani (ed), Taylor and
Francis Group, London, ISBN. 978-0-415-68367-8, pp. 715-722.
150
[160] Tschernutter, P. and Kainrath, A. (2016), ―Design considerations and behavior
of reinforced concrete core dams during construction and impounding‖, Water
Science and Engineering, Vol. 9, No. 3, pp. 212-218.
[161] Vahdati, P. (2014), ―Identification of soil parameters in an embankment dam
by mathematical optimization‖, PhD Thesis, Luleå tekniska universitet,
Swedan.
[162] Honkanadavar, N. P. and Sharma, K. G. (2016), ―Modeling the triaxial
behavior of riverbed and blasted quarried rockfill materials using HSM‖,
Journal of Rock Mechanics and Geotechnical Engineering, Vol. 8, No. 3, pp.
350-365.
[163] Beiranvand, B. and Komasi, M. (2019), ―Monitoring and numerical analysis
of pore water pressure changes Eyvashan dam during the first dewatering
period‖, Journal of Applied Research in Water and Wastewater, Vol. 6, No. 1,
pp. 1-7.
[164] Sukkarak, R., Pramthawee, P. and Jongpradist, P. (2017), ―A modified elasto-
plastic model with double yield surfaces and considering particle breakage for
the settlement analysis of high rockfill dams‖, KSCE Journal of Civil
Engineering, Vol. 21, No. 3, pp. 734-745.
[165] WAPDA. (2006), ―Project Planning Report of Nai Gaj Dam‖, Water And
Power Development Authority, Pakistan.
[166] Jansen. R. B. (2012), ―Advanced dam engineering for design, construction,
and rehabilitation‖, Springer Science and Business Media, 1st edition, ISBN.
13:978-1-4613-0857-7.
[167] Penman. A. D. (2018), ―Instrumentation, monitoring and surveillance
embankment dams‖, 1st Edition, ISBN. 9780203750667.
[168] Chen, Q., Zou, Y. H., Tang, M. and He, C. R. (2014), ―Modelling the
construction of a high embankment dam‖, KSCE Journal of Civil Engineering,
Vol. 18, No. 1, pp. 93-102.
[169] Novak. P., Moffat, A. I. B., Nalluri, C. and Narayanan, R. (2017), ―Hydraulic
structures‖, CRC Press 4th Edition, ISBN 9781315274898.
[170] Schleiss, A. J. and Boes, R. M. (2011), ―Dams and reservoirs under changing
challenges‖, CRC press, 1st edition, ISBN 978-0-203-80409-4.
[171] Carter, M. and Bentley, S. P. (2016), ―Soil properties and their correlations‖,
John Wiley and Sons, 2nd
edition, ISBN 9781119130901.
151
[172] Cobb. F. (2008), ―Structural Engineer's Pocket Book‖, CRC Press, 2nd
Edition,
ISBN 978-0-7506-8686-0.
[173] O'Sullivan, C. (2013), ―Craig's soil mechanics‖, Geotechnique, Vol. 63, No. 8,
pp. 705.
[174] Rajapakse, R. A. (2015), ―Geotechnical engineering calculations and rules of
thumb‖, Butterworth-Heinemann, 2nd
edition, ISBN 978-0-12-804698-2.
[175] Barnes. G. (2010), ―Soil Mechanics‖, Principles and practice, Macmillan Press
Limited, 3rd
edition, ISBN 978-0-230-57980-4.
[176] Justo, J. L. and Saura, J. (1983), ―Three‐dimensional analysis of Infiernillo
dam during construction and filling of the reservoir‖, International Journal for
Numerical and Analytical Methods in Geomechanics, Vol. 7, No. 2, pp. 225-
243.
[177] Sara. M. N. (2003), ―Site assessment and remediation handbook‖, CRC Press,
2nd edition, ISBN 0-56670-577-0.
[178] Burgess, P. J., Sarabia, D., Small, J., Poulos, H. G. and Sinha, J. (2007),
―Settlement behaviour of a major dam‖, Ancold bulletin, Vol. 137, No. 91, pp.
1-10.
[179] Deere, D. U. and Miller, R. P. (1966), ―Engineering classification and index
properties for intact rock‖, Technical report, Illinois University At Urbana
Departement of Civil Engineering.
[180] Potts, D. and K. Axelsson, (2002), ―Guidelines for the use of advanced
numerical analysis‖, Thomas Telford.
[181] Trenter. N. A. (2001), ―Earthworks a guide‖, Thomas Telford, Limited
edition, ISBN 0 7277 2966 7.
[182] Obrzud, R. F. (2010), ―On the use of the Hardening Soil Small Strain model in
geotechnical practice‖, Numerics in Geotechnics and Structures, 16.
[183] Reclamation, U.S.D.o.t.I.B.o., Design Standards No. 13 Embankment Dams.
Chapter 4: Static Stability Analysis Phase 4 (Final) October 2011.
[184] World Weather Online (2019), ―Average annual rainfall amount‖,
URL:https://www.worldweatheronline.com/lang/en-pk/dadu-weather-
averages/ sindh/pk.aspx (Last accessed on 18-02-2020).