See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/282778247
Clarification on the Methodology of Calculation of Seismic Lateral Earth
Pressure in Cohesive Soils
Article in Electronic Journal of Geotechnical Engineering · January 2010
CITATIONS READS
0 764
4 authors:
Mehdi Jalili Iraj Rahmani
Islamic Azad University Semnan Branch university of iran
12 PUBLICATIONS 44 CITATIONS 11 PUBLICATIONS 212 CITATIONS
SEE PROFILE SEE PROFILE
Hamid Alielahi Hossein Ghiassian
Islamic Azad University of Zanjan (IAUZ) Iran University of Science and Technology
69 PUBLICATIONS 467 CITATIONS 19 PUBLICATIONS 398 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
The Effect of Soil Improvement on Seismic Ground Response Analysis (Case Study) View project
The seismic interaction between aboveground structures and underground cavities and tunnels View project
All content following this page was uploaded by Hamid Alielahi on 17 November 2018.
The user has requested enhancement of the downloaded file.
Clarification on the Methodology of
Calculation of Seismic Lateral Earth
Pressure in Cohesive Soils
Mehdi Jalili
PhD student, School of Civil Engineering, Iran University of Science and
Technology (IUST), P.O. Box 16765-163, Narmak, Tehran, Iran.
e-mail:
[email protected] Iraj Rahmani
Assistance professor of Maritime Transportation & Technology Group,
Transportation Research Institute, Tehran, Iran.
e-mail:
[email protected] Hamid Alielahi
Islamic Azad University, Zanjan Branch, Zanjan, Iran.
H. Ghiassian
Associate Professor, School of Civil Engineering, Iran University of Science
and Technology (IUST), P.O. Box 16765-163, Narmak, Tehran, Iran.
ABSTRACT
This paper offers clarifications for two inconsistencies in lateral earth pressure calculations.
First problem is the inconsistency of the methodology of seismic lateral earth pressure
calculation for soils based on Mononobe-Okabe for Undrained Unconsolidated condition of
clayey soils. The solution of this problem is described based on an analytical-graphical
method. In calculations, the soil properties of Shahid Rajaee Port Project situated on shores of
Persian Gulf near Bandar Abbas, in south of Iran are used as a case history. Second, the
clarification on using effective soil unit weight (γ') in highly permeable and low permeable
soils is described.
KEYWORDS: Earth pressure; cohesive soils, seismic condition, pseudo-static.
INTRODUCTION
In this paper, the methodology of OCDI (2002) for calculation of seismic lateral earth
pressure in cohesive soils has been clarified. As a case history, the geotechnical and seismic
features of ground in Shahid Rajaee Port Complex Development (S.R.P.C.D.) Project is
considered. The site is located on the south coast of Iran near the town of Khunsurkh,
approximately 20 km south-west of Bandar Abbas.
- 1171 -
Vol. 15 [2010], Bund. L 1172
METHODOLOGY OF OCDI (2002) FOR
CALCULATION OF SEISMIC LATERAL EARTH
PRESSURE
As we know, OCDI (2002) is an acceptable guidance to design the quay walls with especial
view on seismic conditions. Based on the proposed method by OCDI (2002), the earth pressure
acting on the structure shall be calculated corresponding to various conditions such as the type of
soils, the groundwater level, and the seismic coefficients. In the following sections, the
methodology of lateral earth pressure calculation for two types of soils is described.
Cohessionless Soils
In the pseudo-static approach, the earth pressure could be estimated using Mononobe-Okabe
equation. For a vertical wall having a friction angle of δ, between the backfill and the vertical
wall and a horizontal backfill with an angle of internal friction of φ, the dynamic active earth
pressure coefficient Kae, shall be calculated using this equation:
Cos 2 (ϕ − θ )
k ae = 2
sin (ϕ + δ )sin (ϕ − θ ) (1)
Cosθ Cos (δ + θ )1 +
cos(δ + θ )
In the above equation, θ is the seismic inertia angle which will be derived as below:
θ = tan −1 kh (2)
where kh is the horizontal seismic coefficient.
Below the water level, apparent seismic coefficient kh′, could be substituted in the above
formulae to calculate θ. It is allowed to calculate the earth pressure acting below the residual
water level during an earthquake according to the procedure stated before by using the apparent
seismic coefficient that is determined by:
2(Σγ t H i + Σγ sat H j + q sur ) + γ sat .H sat
k′ = kh
[ ]
2 Σγ t H i + Σγ `H j + q sur + γ `H sat
(3)
where:
Hj Thickness of soil layer below the residual water level
Hi Thickness of the i-th soil layer above the residual water level
Hsat Thickness of soil layer to calculate earth pressure below the residual water level
γt Unit weight of backfill above the residual water level
- 1172 -
Vol. 15 [2010], Bund. L 1173
γsat Unit weight of saturated soil layer below the residual water level
γ' Buoyant unit weight of backfill below the residual water level
qsur Uniform surcharge on the backfill
kh Horizontal seismic coefficient
Finally, the horizontal earth pressure acting on back-face of vertical wall during earthquake
could be calculated by the following equation:
pae = kae [ΣγH + qsur ] (4)
For passive case, the horizontal earth pressure will be determined as follows:
p pe = k pe [ΣγH + qsur ] (5)
where:
Cos 2 (ϕ − θ )
k pe = 2
sin (ϕ − δ )sin (ϕ − θ ) (6)
Cosθ Cos (δ − θ )1 −
cos(δ − θ )
In the above formula, δ is applied by negative sign in passive condition.
In static condition, the aforementioned formula are valid considering θ equal to zero.
Considering the type of the soil, progress of work, standing time of the trench and migration of
bentonite within voids and formation of filter cake, angle of friction between sandy soils and the
φ φ
wall (δ) is assumed as (+ , − ) respectively for active and passive sides of the wall based on
3 3
EAU (1996).
Cohesive Soils
In static case, the earth pressure of cohesive soil acting on the back face wall could be
estimated as below:
pa = ΣγH + qsur − 2C (7)
p p = ΣγH + qsur + 2C
(8)
If a negative value for active earth pressure is obtained by calculation, then the pressure will
be set to zero.
In pseudo-static analysis in cohesive soils, generally two types of calculation are used in
unconsolidated undrained and consolidated undrained analysis.
- 1173 -
Vol. 15 [2010], Bund. L 1174
In unconsolidated undrained (UU) analysis the undrained soil parameters are used and the
active earth pressure acting on the back face wall during an earthquake Pae and the angle of
sliding surface ζa will be calculated by the following equations:
pa e =
(Σγh + qsur )sin (ς a + θ ) − cu
(9)
cosθ sin ξ a cos ξ a sin ξ a
Σγh + 2q sur
ξ a = tan −1 1 − tan θ (10)
2c u
where:
γ: Unit weight of the soil layer
h : Thickness of the soil layer
qsur : Uniform surcharge
ζa : Angle of failure surface of the layer
In consolidated undrained analysis, an expansion of the above earth pressure can be applied
to materials having both the cohesion c and internal friction angle (f φ ) (OCDI, 2002). Obtained
equation in this condition could be expressed as (OCDI, 2002):
pa = (
(Σγh) cos(ψ − β ) ) + q ).
sin(ζ − φ + θ ) . cos(ψ − ζ )
sur
cosψ cosθ . cos(ψ − ζ + φ + δ ) . sin(ζ − β
c . cos(ψ − β ) . cos φ (11)
−
cos(ψ − ζ + φ + δ ) . sin(ζ − β )
Where
ψ Angle of wall to the vertical (º)
β Angle of ground surface to the horizontal (º),
Other parameters are described in previous equations.
- 1174 -
Vol. 15 [2010], Bund. L 1175
USE OF THE METHOD FOR A CASE STUDY
Description of the site
As a case history, the geotechnical and seismic characterization of ground in Shahid Rajaee
Port Complex Development Project (S.R.P.C.D.) is considered. In this project, the development
of an existing container vessel berth site located in the western part of Shahid Rajaee port is
planned. As shown in Figure 1, development Stage consists of two different zones:
a. Zone 1: The two narrow rectangle zones of 910m in 100 m for new berths Eastern and
western berths with the region behind quay walls for unloading and reloading containers. This
zone includes the berthing diaphragm wall, the anchor wall and the rear crane bored deep
foundation. It should be mentioned that all the walls and piles are implemented in situ reinforced
concrete.
b. Zone 2: The area of an approximately rectangular shaped zone with 1300 m in 480 m
dimensions in the north part of the site where the containers are to be stacked with maximum five
high. Different Zones of the site have been reclaimed at the level of approximately +4.5 mCD.
Stacking yard
Main quay walls
Existing Berth
Figure 1: Plan of the S.R.P.C.D. project: two western and eastern container
berths and northern stacking yard zones.
Geotechnical sublayer condition
The geotechnical site investigation was performed in the site location. These activities
consisted of exploratory boreholes with SPT and vane shear tests, cone penetration tests (CPT),
- 1175 -
Vol. 15 [2010], Bund. L 1176
dynamic probing tests. Generally there are five layers with fluctuating levels which is
summarized in Table 1.
Table 1: Ground profile and soil parameters in Shahid Rajaee Port Project
Strata Type Depth (m) Level Cu (kPa) Friction Density Density
(m,CD) Angle Above Below
(ϕ ,0 ) Water Table Water Table
(kN/m2)* (kN/m2)
Rockfill 0 to 3.5 5.5 to 2 - 40 18 20
Silty/Clayey Sand 3.5 to 8.5 2 to -3 - 37 18.5 21
Silty Clay 8.5 to 19.5 -3 to -14 50 0 - 18
Silty/Clayey Sand 19.5 to 31.5 -14 to -26 - 38 - 21
Silty Clay > 32 < -26 220 0 - 20
* Water level was located at +1.7 m, CD
For the site region, two design levels of ground motions are chosen for seismic analysis based
on performance based design method. According to hazard analysis, the horizontal peak ground
acceleration for level (1) and (2) (75 and 475 years return period) earthquakes are 0.25 g and 0.37
g respectively. For the purpose of pseudo static analysis seismic coefficients recommended by
PIANC (2001) for retaining walls was used and the value of kh has been obtained equal to 0.17
and 0.2 for Level (1) and (2) respectively.
Seismic lateral earth pressure calculation for case study
Considering the above mentioned geotechnical profile, in order to calculate the Seismic
lateral earth pressure for upper Silty Clay layer (with undrained cohesion equal to 50 (kPa)), the
value of the inside of root in Equation (10) for Level (2) earthquake will obtain as follow.
Σγh + 2q sur 3.5 × 18 + 0.3 × 18.5 + 4.7 × 11 + 11.5 × 8 + 2 × 30
1 − tan θ = 1 − × 0.2
2c u 2 × 50
1 − 2.72 × 0.2 = 0.67
Based on the maximum number of containers stacked behind the main quay and anchor wall
the uniform surcharge in seismic case was assigned as half of the static condition according to
OCDI (2002), equal to 30 kN/m2.
If the value of undrained cohesion of clayey layer has been considered less than 27 kN/m2, the
value of the inside of root in Equation (10) will obtain negative. So the use of the above formula
should be clarified in order to solve such problems. In the following sections the clarification of
this problem is described.
- 1176 -
Vol. 15 [2010], Bund. L 1177
CLARIFICATION
Clarification on angle of failure surface (ζa)
Equation (10) represents an approximate value of critical angle of slip surface (ζa) and it is
not valid for all conditions, e.g. the inside of root in Equation (10) is negative. Therefore the
graph of Pae versus ζa should be drawn using Equation (9) to find extreme value of Pae. The graph
could present two general cases which are shown by two examples for different parameters in
Figures (2) and (3). In the following calculations, the soil profile stated in Table (1) are
conducted.
In case 1 (Figure 2) the acceptable range assessed from the plot should be considered as
shown in Figure 1. In such condition using Pae,max as illustrated in this figure is being suggested
conservatively.
Pae,max
Pae,max
Acceptable Range
Figure 2: Case 1: acceptable range and the maximum value of Pae,max used in calculations (graph
of seismic active soil pressure (Pae) versus angle of failure surface of the layer (ζa) ( for undrained
cohesion (Cu) of 50 kN/m2, Overburden pressure of 200 kN/m2, θequal to 16.3° for kh=0.19)
In case 2 (Figure 3) the values of Pae versus ζa presents a maximum value. In such
circumstances, Pae,max could be easily derived from the plots as the maximum value and used in
the analysis conservatively.
Figure 3: Case 2: acceptable range and the maximum value of Pae,max used in calculations (graph
of seismic active soil pressure (Pae) versus angle of failure surface of the layer (ζa) (for undrained
cohesion (Cu) of 70 kN/m2, Overburden pressure of 200 kN/m2, θequal to 16.3° for kh=0.19)
- 1177 -
Vol. 15 [2010], Bund. L 1178
The graph for the acceptable range for each of the examples has been demonstrated using the
defined acceptable range. The value of Pae,max does not significantly change in the acceptable
range. In this case Mononobe-Okabe method can be used conservatively by assuming Ccu=0.
Clarification on using effective soil unit weight (γ') in all
kinds of soils
Considering the fundamental concepts on the methodology of calculating the seismic lateral
earth pressure acting on the wall, in case the soil is submerged, the effective unit weight of the
soil (γsat-γw) should be used in the analysis of the applied pressure. This concept is correct for all
kind of soils such as cohesive or non-cohesive soils.
Distinction of highly permeable and low permeable soils
When backfill is located under the ground water table, the buoyancy force should be taken
into account to calculate vertical force. Therefore, σ'v (vertical effective stress) should be used in
order to calculate lateral earth pressure of any soil layer under the ground water table; Two
distinct concepts are being used in current design procedures to calculate lateral earth pressure
depending on the backfill soil type.
Soils of high permeability, say k > 1.0 cm/sec, where pore water can move independently
from mineral skeleton (Matsuzawa et al., 1985). In free water case (Figure 4(a)) considering the
freely move of the pore water within the soil particles, the total active thrust is made up of the
thrust from mineral the skeleton and hydrodynamic pressure:
A thrust from the mineral skeleton, therefore the apparent angle of seismic coefficient (θ') in
Mononobe-Okabe formula is computed as:
FH γ d kh Gs k
tan θ = = . = . h = kh′ (12)
FV γ ′ 1 ± kv Gs − 1 1 ± kv
Where Gs is the specific gravity of the solids. In this case, the effective unit weight of γ' is
used in the equation for calculating lateral earth pressure.
In soils of low permeability, say k < 10-3 cm/sec, porewater moves together with the mineral
skeleton. In such conditions, it is assumed that the solid portion and the pore water portion of the
soil behave as a unit upon the application of the seismic acceleration (Figure 4(b)). Therefore, the
total unit weight of the soil is subjected to the horizontal seismic acceleration and thus the
horizontal inertia body force FH is equal to γSat.kh that means total stress is used. The vertical
component (FV) can be calculated by subtracting the dynamic buoyancy force acting on the whole
soil from the total dynamic gravitational force of the whole soil and thus the value equals γ'.(1±
kv ). Therefore the apparent angle of seismic coefficient (θ') in Mononobe-Okabe formula
becomes:
- 1178 -
Vol. 15 [2010], Bund. L 1179
FH γ sat kh G − e kh
tan θ ′ = = . = s . = kh′ (13)
FV γ ′ 1 ± kv Gs − 1 1 ± kv
Figure 4(a): Apparent Seismic Coefficient Figure 4(b): Apparent Seismic Coefficient for low
for highly permeable backfill soils (free permeable backfill soils (restricted water case)
water case)
This assumption is completely consistent with the recommendation of OCDI (2002), which
suggests apparent seismic coefficient (k'h) to be used in order to calculate the apparent seismic
angle (θ') and consequently active earth pressure for cohesive soils in earthquake condition:
"It shall be standard to calculate the earth pressure acting below the residual water level
during an earthquake according to the procedures given in 14.3.1. Earth Pressure of Sandy Soil
during Earthquake and 14.3.2. Earth Pressure of Cohesive Soil during Earthquake using the
apparent seismic coefficient that is determined by Equation (14.3.7)."
In another part of OCDI (2002) the method of calculation of the seismic lateral earth pressure is
mentioned and it emphasis on the use of Apparent Seismic Coefficient (k'h) in combination with
Effective Unit Weight (γ') of soil. No limitation has mentioned in such method for the analysis of
the seismic lateral earth pressure of cohesive soil regarding the type of the soil and it can be read:
"In the calculation of earth pressure, the apparent seismic coefficient should be used, because the
submerged unit weight that takes buoyancy into account is used in this case."
Regarding the mentioned points, it can be deducted that, σ'v (vertical effective stress) should
be used for the entire soil layer below water table.
- 1179 -
Vol. 15 [2010], Bund. L 1180
Using the modified horizontal seismic coefficient (k'h) with the effective unit weight (γ') for
low permeable backfill soils has been mentioned in so many other well established references
such as PIANC (2001), Matsuzawa et al. (1985), Wu & Prakash (2000), Ebeling & Morrison
(1992) and McCullough (1998).
CONCLUSION
In order to calculate the lateral earth pressure of cohesive soil OCDI (2002) can be used but
several points should be considered in using methodology of OCDI (2002).
In equation proposed for ξa (Equation 10) in some cases the value inside the root may be
obtained negative. In such cases the acceptable range for Pae could be obtained by an analytical
evaluation the effect of ξa on Pae. Two cases may be occurred in this evaluation. In both cases
(Figure 2 & 3) which is related to the situation of negative value of the inside of root in Equation
(10), the (Pae - ξa ) diagram has two asymptotes. In these cases, the acceptable range for calculating
the Pae is shown on Figures 2 and 3. Using these figures the logical value for Pae could be
obtained, conservatively.
In all types of submerged soils (such as cohesive or non-cohesive soils) the effective soil
density (γ') should be used in calculating the value of lateral earth pressure, and the effect of
water pressure should be calculated individually as static or dynamic water pressure
(Westergaard, 1931), and applied on the wall.
ACKNOWLEDGEMENT
The data reported in this article were performed at the Sahel Consultants as a part of studies
for Shahid Rajaee Port Complex Development Project, employed by the Iranian Port & Marine
Organization. The access to these data is gratefully acknowledged.
REFERENCES
1. EAU (1996), Recommendations of the Committee for Waterfront Structures, Harbours and
Waterways, 7th English Edition (English Translation of 9th German Edition), Ernst & Sohn,
Berlin.
2. Ebeling, R.M. and Morrison E.E. (1992), The Seismic Design of Waterfront Retaining
Structures, US Army Corps of Engineers, Technical Report, ITL-92-11/NCEL TR 939.
3. Matsuzawa, H., Ishibashi, I. and Kawamura, M. (1985), Dynamic Soil and Water Pressures of
Submerged Soils, Journal of Geotechnical Engineering, ASCE 111(10):1161-1176. Pages
1162 & 1163.
4. McCullough, N.J. (1998), The Seismic Vulnerability of Sheet Piles Walls, Master's Thesis
Dissertation, Department of Civil, Construction and Environmental Engineering, Oregon
State University. Pages 43 & 44, Section 4.2.2.1.
5. OCDI (2002), The Overseas Coastal Area Development Institute of Japan, Technical
Standards and Commentaries for Port and Harbour Facilities in Japan, Daikousha Printing
Co., Ltd., Tokyo.
- 1180 -
Vol. 15 [2010], Bund. L 1181
6. PIANC (2001), Permanent International Association for Navigation Congress, Seismic
Design Guidelines for Port Structures, BALKEMA Publications.
7. Westergaard, H.M. 1931. Water pressure on dams during earthquakes. Trans., Paper No
1835, ASCE, New York, 418–433.
8. Wu, Y. and Prakash, Sh. (2000), Seismic Displacements of Rigid Retaining Walls on
Submergence, Proceeding of 12th World Conference on Earthquake Engineering, Paper No.
0562. Pages 2 & 3, Condition 3.
© 2010 ejge
- 1181 -
View publication stats