Proceedings of 8th IOE Graduate Conference

Peer Reviewed
ISSN: 2350-8914 (Online), 2350-8906 (Print)
Year: 2020 Month: June Volume: 8

Analysis of Dam Foundation in Alluvial Deposits

Mabin Dahal a , Mohan Prasad Acharya b , Indra Prasad Acharya c
a, c
Department of Civil Engineering, Pulchowk Campus, IOE, TU, Nepal
bNEA Engineering Company Limited, Thapathalli, Kathmandu, Nepal
Corresponding Email: a 074msgte011.mabin@pcampus.edu.np, b mp.acharya@gmail.com, c indrapd@ioe.edu.np

Abstract
This paper presents a study made to analyse a gravity dam foundation on alluvial deposits. Alluvial deposits
consist of fine fraction and coarse fraction which can be characterised as boulder mixed soil. The percentage
fines fraction of alluvial deposits has direct impact on the engineering characteristics of soil including density
compactness and consequently affect the stability of Gravity dam in alluvial deposits. In this paper, a relation
is established between the percentage of fines fraction with the bulk modulus of elasticity, shear modulus of
elasticity and density. This relation is used to predict the mechanical parameters after the improvement of soil
with consolidation and compaction grouting which decreases the percentage of void with fines.
FDM software is used for the analysis and stability of gravity dam at full reservoir conditions. The effect
of the mechanical parameters for the stability of gravity dam is analysed. The soil material is modeled as
homogeneous and heterogeneous material consisting different fraction of boulders and fines. The results from
FDM analysis demonstrated the decrease deformation in dam foundation with the decrease in percentage fine
fraction in alluvial soil.
Keywords
Alluvial deposits, Fine fraction, Coarse fraction, Gravity Dam, Displacement

1. Introduction failure[1]. The cost of treatment might be extremely

difficult and costly.
The various types of dams according to method of
construction are earthen, rockfill, hardfill, and gravity Rivers in Nepal pass through deep gorges with full of
dam. Usually gravity dam is constructed on stronger sediments and alluvial deposits. The bed rock might
soil having hard strata whereas earthen dam is be buried beneath more than 100m below the river bed
constructed on weak foundation. But in recent times, In such cases the construction of dam foundation on
at Gerdebin dam, Iran there is weak and soft alluvial the bedrock is extremely costly especially for small
deposits beneath it and rockfill dam was constructed hydropower projects. This subject is not studied in
above it[1]. In Clear lake dam, gravity dam was detail in Nepal few dam builds are either directly
constructed above alluvium deposits of cobble, build on top of bed rock. The risk of dam stability due
boulder, gravel, sand, silt, clay with proper design of to deformation in an alluvium and colluviums
filter materials, seepage control mechanism, grouting deposits is to be studied in detail to assure the
materials [2]. long-term stability of dam and to protect from the
sudden catastrophic failure. Therefore, a study to
The significant variations in thickness of the alluvial determine the methodology of accessing the stability
deposits and the presence of a low strength layer are of dam in alluvial foundation is necessary.
the major problem to consider for design of dam on
alluvial deposits [3] [4]. There’s always remain a In this paper, FDM software is used for the analysis of
possibility bed rock can’t be found that is why dam foundation stability of gravity dam at full reservoir
has to build within the alluvial soil layers. Treatment conditions. The effect of the mechanical parameters
of dam foundation in an alluvial deposit is of for the stability of gravity dam in alluvial foundation
importance due to being susceptible to the large is analysed. The model incorporates the methodology
settlements, which may cause the dam body subjected to increase the strength of alluvial soil foundation
to undesirable deformations and even, catastrophic from the combination of consolidation and
compaction grouting. 2D finite difference software is

Pages: 951 – 961

Analysis of Dam Foundation in Alluvial Deposits

used to determine the, deformation, pore pressure, and heterogeneous deposits of boulders, cobbles, rockfall,
stress distribution within the foundation. alluvium and lacustrine deposits proposed a solution
to minimize cracking, control seepage, and bearing
capacity on existing condition (Monley et al. 2018).
2. Dam in Alluvial Foundation The primary difficulties in the design was supporting
dam on the weak alluvial deposits, controlling
Construction and design of dam on alluvial deposits is
seepage gradient, and proving proper filter and drain
a great challenge due to the unpredictability nature of
system at different portion of dam. A roller
material below. Erosion of fine-graded soils through
compacted dam with proper control of seepage
seepage can cause piping. It will result in differential
gradient, filter drainage system, cutoff wall is
settlement and seepage of dam. Moreover, alluvial
constructed to withstand in weak foundation. Another
deposits consist of mixtures of boulders, cobbles,
case study was done at Chapar-Abad dam which have
gravels, silts, coarse sand, fine sand etc. The finer
alluvial deposits of over 60m thickness as base
material between the boulder and cobbles can erode in
material and having various coefficient of
the presence of high-gradient flow known as suffusion
permeability (Uromeihy and Barzegari 2007).
thus resulting the sinkholes in foundation of dam and
Probable seepage on foundation was estimated by
abutment [5]. Alluvial deposit is a mixture of soil
conducting in-situ tests, and numerical modelling.
such as clay, silt, sand, gravel and boulder which has
Based on those results installation of grout curtain
been eroded and carried by fast flowing streams and
was suggested.
later settled when the velocity of flow decreases [6].
The schematic diagram presented below (figure:1)
shows the structure of alluvial soil mass. The Fine 3. Data Collection and Analysis
fraction is defined as the volume of fine particles
except boulder and gravel in an alluvial deposit. It The 60m depth borehole data of Kimathanka Arun
consists of sand and silt. The Coarse fraction is the Hydroelectric Project is collected. It is found that up
volume of boulders and gravels in an alluvial deposit to 60m depth there is no presence of bed rock. The
which can be extracted as a core from drilling. borehole data shows the lithological characteristics of
subsurface which is predominantly consists of sands,
gravel, boulder of varying size. And also, the
sub-surface geological condition of the project area
done by multi-channel analysis of the surface waves
(MASW) at major structural sites of the project is
collected. MASW calculates the shear wave velocity
profile of sub-surface ground condition. Based on
shear-wave velocity profile, the dynamic parameters
of soil: density, moduli of elasticity of each layer can
be estimated. From the 60m depth of borehole log, the
distribution of percentage finer fraction and
percentage coarse fraction per meter of layer is found
Figure 1: Alluvial mixture containing Boulder,Gravel out. Then, the percentage finer fraction of borehole
and Fine fraction. log is compared to shear wave velocity profile of
MASW survey. For this, exact coordinate of MASW
profile and borehole log is scrutinized. And also,
Skhalta Dam is located in south-west Georgia average of seven different shear wave velocity profile
(Pawson and Russell 2014). Due to landslide and of MASW survey and borehole log is examined. In
erosion there occurs deposition of alluvial sand, these two cases, there was not much variation in the
gravels and sediments.And the predominant material percentage finer content relation with shear wave
contains alluvium and lacustrine deposits. At center velocity. From this, the relation of percentage finer
of valley 50m thick alluvial soil is deposited. To fraction with bulk modulus of elasticity, shear
curtail the quantity of grouting faced symmetrical modulus of elasticity and density is found out.
hardfill dam is selected due to its minimal base area. Deformation and seepage on foundation material is
A research was conducted on “Clear lake dam” having calculated. The below figure 1, 2, 3 illustrates the

952
 Proceedings of 8th IOE Graduate Conference

relationship of Percentage fine fraction with moduli of and 20 percentage fine fraction, the random
elasticity and density in an alluvial deposit. heterogeneous material modelling is done. The
density and dynamic properties of soil is shown from
figure 5 to figure 10.

Figure 2: Variation of Shear Modulus of Elasticity

with Percentage Fine Fraction in an Alluvial Deposits. Figure 5: Shear Modulus of Foundation Material at
80 percentage Fine Fraction.

Figure 3: Variation of Density with Percentage Fine

Fraction in an Alluvial Deposits.

Figure 6: Bulk Modulus of Foundation Material at 80

percentage Fine Fraction.

Figure 4: Variation of Bulk Modulus of Elasticity

with Percentage Fine Fraction in an Alluvial Deposits.

The material modelling of heterogeneous material is

performed. For this the depth and width of foundation
material considered is 30m and 60m. At 80 Figure 7: Density of Foundation Material at 80
percentage fine fraction, 35 percentage fine fraction percentage Fine Fraction.

953
Analysis of Dam Foundation in Alluvial Deposits

Figure 8: Shear Modulus of Foundation Material at Figure 11: Shear Modulus of Foundation Material at
35 percentage Fine Fraction. 20 percentage Fine Fraction.

Figure 12: Bulk Modulus of Foundation Material at

Figure 9: Bulk Modulus of Foundation Material at 35 20 percentage Fine Fraction.
percentage Fine Fraction.

Figure 13: Density of Foundation Material at 20

percentage Fine Fraction.
Figure 10: Density of Foundation Material at 35
percentage Fine Fraction.

954
 Proceedings of 8th IOE Graduate Conference

The total height and base width of gravity dam are fraction of boulder is relatively low comparing to
27m and 20m as shown in below figure:14. The sum others. To increasing the bearing capacity of deposits,
of vertical and horizontal forces are found to be it is necessary to increase the stiffness of fine fraction
4218.05KN and 1883.52 KN. The resisting and in alluvial deposits. This is achieved by performing
overturning moment are calculated 75562.94 KNm consolidation and compaction grouting with enough
and 35333.78KNm. Total stress on toe and heel are pressure. Grouting consolidate the fine fraction in
240640 N/m2 and 181166 N/m2 . The eccentricity of deposits and decreases the volume of fine content.
dam is found to be 0.47. The factor of safety against When we make the finer particles stiffness close to the
overturning and sliding are estimated more than 2 and boulder stiffness at that condition, we can be assured
1.5. Thus, the stability of dam is assured from that there won’t be deformation. Since stiffness of
perspective of sliding and overturning. Now, it needs boulder is more than rock.
to be assured on bearing capacity for its stability.
To study the effect of percentage fine fraction on the
stability of Rolwaling Dam in alluvial deposits, a
two-dimensional finite difference-based software is
selected for modelling. The dam is modelled in
plane-strain condition. The material modelling is
done through Mohr-Coulomb material model.
Soil-Gravel: mixture of gravel and sand is chosen.
The dam body is modelled at natural condition of
Rolwaling Khola considering linear variation of
material upto 30m depth. The uniformly varying load
(trapezoidal) is put in model considering full reservoir
condition. The foundation material is considered as
completely saturated. The dam is analysed
Figure 14: Gravity Dam. considering linear distribution of foundation material.
The material modeling of heterogeneous material is
done considering random distribution of 80,35 and 20
4. Modeling Process percentage fine fraction. This decrease in percentage
Secant pile techniques with combination of fine fraction is attained in field by combination of
compaction and consolidation grouting increases the consolidation and compaction grouting, and deep
bearing capacity of dam foundation. Grouting helps to mixing. This is checked in field by drilling core.
enhance the bearing capacity of weak deposits and The different modeling case of gravity dam done are
consequently decrease the degree of settlement. At as follows:
first, secant pile boundary is constructed on outer
periphery to enclose the targeted area. It is necessary 1. Linear Distribution of homogeneous Materials
to ensure boundary doesn’t deform during different (a) Linear distribution of homogeneous material at 80,
combination of grouting. 60 and 40 percentage fine fraction
The assumption is that, with the help of consolidation (b) Linear distribution of homogeneous material at 50,
or compaction grouting the voids will be filled by 40 and 35 percentage fine fraction
cemented material such that the fine fraction will be
decreased and consequently the stiffness/ density of (c) Linear distribution of homogeneous material at 30,
the material will be increased. This is represented by 25 and 20 percentage fine fraction
the increase in K and G values. (d) Linear distribution of homogeneous material at 20
In a mixture of alluvial deposits containing boulders, percentage fine fraction
gravels, silt and sand, boulder has excessively higher
2. heterogeneous random distribution of Material
stiffness comparing to others four. Then, gravel has
(a) Random Distribution of heterogeneous Material at
higher stiffness and the other two has relatively low
35 percentage Fine Fraction
stiffness. The bearing capacity of alluvial deposits
depends upon the overall stiffness of the deposits. (b) Random Distribution of Heterogenus Material at
There is a high chance of excessive deformation if the 20 percentage Fine fraction

955
Analysis of Dam Foundation in Alluvial Deposits

5. Results and Discussion bottom i.e. 30,25, and 20 percentage fine fraction is at
top bottom and middle. The decrease in fine fraction
The following are the results which shows consequently, increase the stiffness of the deposits.
deformation, vertical stresses, deformation histories at This is represented in model through increase value in
different points on two cases: Linear distribution of K, G and ρ. This is illustrated in figure 22 to figure 24.
Material, heterogeneous distribution of material. The The maximum vertical displacement at top of the
first case is linear distribution of homogeneous foundation is found to be 300mm through
material of Rolwaling khola at Natural Conditions displacement histories plot. This shows the decrease
(figure:15 to figure:27) and the second case is random in percentage fine fraction through grouting. The
distribution of heterogeneous material at different displacement contour at top is found to be 7mm. But
percentage fine fractions (figure:28 to figure:33). there is no significant change in vertical stress
Figure 12 to Figure 15 shows the modeling of dam at distribution than second case.
natural conditions. At natural conditions there is 10m In fourth case, the dam is modelled for percentage
thickness of 80 percentage fine fraction at top, then 60 fine fraction of 20. This is achieved when voids are
percentage fine fraction at middle and 40 percentage filled up with Cementous material through
fine fraction at bottom. It shows the vertical consolidation and compaction grouting. This can be
displacement histories at 8 different points, vertical check in field through core drilling. The fine fractions
stress contours, and displacement of dams. The changes into a boulder fraction through binding of
maximum displacement is found to be 1m at top Cementous material. The output of numerical
layers below the toe of dam. The maximum vertical modeling is shown in figure 24 to figure 27. The
displacement contours at top of the foundation is maximum vertical displacement occurs just below the
25mm. Since the stress on toe is more than stress on toe of dam and is found to be 50mm by plotting
heel, maximum displacement on toe. The displacement histories at eight different points. The
displacement histories plot at 8 different points below displacement contours at top of the foundation is
toe of dam shows the maximum deformation occurs at found to be 6mm. The maximum vertical stress is on
top portion of foundation which is more than 1m. Due the region where there occurs maximum dam pressure.
to this dam completely collapses in bearing capacity. The maximum stress occurs below toe of dam which
The vertical stress diagram shows that maximum is found to be 2.5 X 105 N/m2 .
vertical stress at natural conditions is about 3 X 105
N/m2 . The vertical stress distribution decreases with Material modeling at heterogeneous random
respect to depth. This is due to the load of dam. If distribution is performed at 35 percentage fine
there was no structure or load at the top, then the fraction and 20 percentage fine fraction from figure 28
vertical stress distribution would increase with respect to figure 34. The maximum vertical deformation by
to depth. plotting deformation histories at eight different points
is found to be 9mm. The bearing capacity of material
In second case, the dam is modelled as homogeneous is found out to be higher in heterogeneous random
material consisting of 50,40, and 35 percentage fines distribution than homogeneous distribution of
fraction at top, middle and bottom of each layer 10m material. The maximum vertical deformation on dam
thick. The decrease in fine fraction is achieved at 35 percentage fine fraction is found to be 7mm
through combination of consolidation and compaction whereas at 20 percentage fine fraction it is found to be
grouting. The output of numerical modeling is shown 1mm. The maximum vertical stress is found just
in figure 19 to figure 21. The vertical stress diagrams below toe of dam at 20 percentage heterogeneous
show that maximum vertical stress at is about 3 X 105 distribution of fine fraction which is of value 6 X 105
N/m2 . The maximum vertical displacement at top of N/m2 and 3.5 X 105 N/m2 at 35 percentage fine
the foundation is found to be 450mm. The fraction. In both cases maximum deformation occurs
displacement contours are found of 10mm. There is of around 9mm.
slightly decrease in vertical stress distribution.
In sum up, from above results it is found out that
In third case, voids are filled up with cementing mechanical parameters have direct influence on the
material through combination of compaction and stability of gravity dam on weak foundation. The
consolidation grouting and when the percentage mechanical parameters of boulder mix soil is
boulder fraction is 70 at top, 75 at middle and 80 at improved through combination of compaction and

956
 Proceedings of 8th IOE Graduate Conference

consolidation grouting. The model incorporates the

combination of compaction and consolidation
grouting in model through increase value of bulk
modulus of elasticity, shear modulus of elasticity and
density. The increase in mechanical parameters
decreases the settlement problem in alluvial
foundation through decrease in percentage fine
fraction. These calculations provide one of the
methodologies to analyse dam foundation in alluvial
deposits through use of FDM software.

Figure 17: Vertical Stress Contours at layer of 80, 60

and 40 Percentage Fine Fraction.
1. Linear Distribution of homogeneous Materials
(a) modeling of Dam at layer of 80, 60 and 40
Percentage Fine Fraction

Figure 15: Model of Dam at layer of 80,60 and 40 Figure 18: Displacement Contours of Dam at a layer
Percentage Fine Fraction. of 80,60 and 40 Percentage Fine Fraction.

(b) modeling of Dam at layer of 50, 40 and 35

Percentage Fine Fraction

Figure 16: Deformation Histories of Dam at 8 Figure 19: Vertical displacement contours at 50, 40
different points. and 35 percentage fine fraction.

957
Analysis of Dam Foundation in Alluvial Deposits

Figure 20: Vertical Stress contours at 50, 40 and 35 Figure 23: Vertical displacement Contours at 8
percentage fine fraction. different points.

Figure 21: Vertical displacement histories at 8 Figure 24: Vertical displacement contours at 30, 25
different points. and 20 percentage fine fraction.

(c) modeling of Dam at layer of 30, 25 and 20 (d) modeling of Dam at layer at layer of 20 Percentage
Percentage Fine Fraction Fine Fraction

Figure 22: Vertical Stress Contours at 30, 25, and 20 Figure 25: Displacement histories at 8 different
Percentage Fine Fraction. points below toe of dam.

958
 Proceedings of 8th IOE Graduate Conference

Figure 26: Vertical displacement of dam at 20 Figure 29: Vertical Stress Contours at 35 Percentage
percentage fine fraction . fine Fraction considering random heterogeneous
distribution.

Figure 27: Vertical Stress Contours at 20 percentage Figure 30: Vertical deformation of dam at 35
fine fraction. Percentage fine fraction considering random
heterogeneous distribution.

2. heterogeneous random distribution of Material

(a) heterogeneous random distribution of Material at
35 percentage Fine Fraction

Figure 31: Vertical deformation histories of dam at

35 Percentage Fine fraction considering random
Figure 28: Modeling of Gravity dam at 35 heterogeneous distribution.
Percentage fine fraction considering random
heterogeneous distribution (b) heterogeneous random distribution of Material at

959
Analysis of Dam Foundation in Alluvial Deposits

20 percentage Fine Fraction 6. Conclusion

This study presented a methodology to analyse the
stability of dam foundation in alluvial deposit in
natural and improved soil strata by means of
analytical and numerical method. First a relationship
is established between percentage fine fraction with
moduli of elasticity and density from the analysis of
MASW survey data and borehole data on alluvial
deposits. The relationship shows that mechanical
properties of soil increase with decrease in percentage
fine fraction. This relationship is used to find the
mechanical properties of soil having different fine
fractions of alluvial deposits. .The results are used to
Figure 32: Deformation histories of dam at 8 model the dam foundation in alluvial soil deposit. The
different points considering random heterogeneous model considered the reduction in the percentage fine
distribution. fraction by the means of compaction and
consolidation grouting, i.e., the cementation of fine
material. The model simulated the process and the
result obtained showed the decrease in deformation
with the decrease in fine fraction. In total 6 numbers
of models are analysed to represent the reduction in
fine fraction from (0-75)%. The result from the
numerical model representing (10-80)% reduction in
fine fraction clearly depicts that the stability of dam in
alluvial foundation can be achieved with the decrease
in the fine fraction. For the 27m height gravity dam in
assumed alluvial soil, the dam was stable after
reduction in 75% of fine volume. The numerical
model considered the homogeneous and random
distribution of the fines and coarse material. The
homogeneous material assumed uniform value of
Figure 33: Vertical Displacement contours at 20
strength parameters K, G and ρ, i.e., average value of
Percentage fine fractions considering random
boulders and fines. The heterogeneous material model
heterogeneous distribution.
represented the random distribution of different
fraction of boulders and fines and consequently the
strength parameters. When an equal stress is applied
at linear homogeneous material model and random
heterogeneous material model. The bearing capacity
of heterogeneous material model is found higher than
homogeneous material model at equivalent percentage
fine fraction. The vertical stress on alluvial deposits is
found higher in case of heterogeneous material
distribution than homogeneous material distribution at
equivalent percentage fine fraction.
Moreover, the method presented to analyse the stability
of dam foundation in alluvial foundation represented
the load deformation mechanism in the foundation at
Figure 34: Vertical Stress contours at 20 Percentage the natural and improved soil. The numerical model
fine fractions considering random heterogeneous result showed the validity of the method.
distribution.

960
 Proceedings of 8th IOE Graduate Conference

References Conference of the British Dam Society at Queen’s

University, Belfast, from 3–6 September 2014, pages
[1] Fardin Jafarzadeh and Amir Akbari Garakani. 148–160. ICE Publishing, 2014.
Foundation treatment of embankment dams with [4] Hossein Hedayati Talouki, Gholam Reza Lashkaripour,
combination of consolidation and compaction Mohammad Ghafoori, and Abbas Ali Saba.
groutings. 2013. Assessment and presentation of a treatment method
[2] Greg J Monley, Steve Jamieson, Henry Buczek, and to seepage problems of the alluvial foundation of
Don Lopez. Clear lake dam replacement: Rcc dam ghordanloo dam, ne iran. Journal of the Geological
on a challenging soil foundation. In Rocky Mountain Society of India, 85(3):377–384, 2015.
Geo-Conference 2018, pages 76–87. American Society
of Civil Engineers Reston, VA, 2018. [5] JB Burland and MC Burbridge. Settlement of
foundations on sand and gravel. In Institution of Civil
[3] JR PAWSON and E RUSSELL. Skhalta dam–design of Engineers, Proceedings, Pt 1, volume 76, 1985.
a hardfill dam founded on deep alluvium and lacustrine
deposits. In Maintaining the Safety of our Dams [6] Robin Fell. Geotechnical engineering of dams. CRC
and Reservoirs: Proceedings of the 18th Biennial press, 2005.

961