See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/223247408

Seismic stability analysis of gravity retaining

walls

Article in Soil Dynamics and Earthquake Engineering October 2010

DOI: 10.1016/j.soildyn.2010.04.005

CITATIONS READS

18 231

3 authors:

Xinpo Li Yong Wu
Chinese Academy of Sciences 170 PUBLICATIONS 1,915 CITATIONS
56 PUBLICATIONS 261 CITATIONS
SEE PROFILE

SEE PROFILE

Siming He
Chinese Academy of Sciences
140 PUBLICATIONS 585 CITATIONS

SEE PROFILE

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

Unsaturated soil mechanics View project

All content following this page was uploaded by Siming He on 30 August 2015.

The user has requested enhancement of the downloaded file.

 Soil Dynamics and Earthquake Engineering 30 (2010) 875878

Contents lists available at ScienceDirect

Soil Dynamics and Earthquake Engineering

journal homepage: www.elsevier.com/locate/soildyn

Technical Note

Seismic stability analysis of gravity retaining walls

Xinpo Li a,b, Yong Wu b, Siming Hea,b,
a
Key Laboratory of Mountain Hazards and Surface Process, CAS, Chengdu 610041, China
b
Institute of Mountain Hazards and Environment, CAS, Chengdu 610041, China

a r t i c l e in fo abstract

Article history: A new approach based on the category of upper bound theorem of limit analysis is presented in this
Received 22 September 2009 study to consider the seismic stability of gravity retaining walls. The retaining wall and the backll soil
Received in revised form were taken as a whole system. For a translational failure mechanism assumed, formulas are provided to
1 February 2010
calculate directly the yield acceleration and the inclination of the failure surface. An example is shown
Accepted 3 April 2010
to illustrate the method. Comparisons are made with limit equilibrium method, and the results are
found consistent. Based on a limited parametric study, it is shown that the wall roughness has
Keywords: remarkable inuence on the yield acceleration.
Seismic stability & 2010 Elsevier Ltd. All rights reserved.
Limit analysis
Gravity retaining walls
Yield acceleration

1. Introduction
Gravity walls are widely used as earth retaining systems
supporting ll slopes adjacent to roads and residential areas, also
in regions prone to earthquake. Many researchers have developed
design methods for retaining walls during earthquakes by using
different approaches. Though the quest for rational design
methods of retaining structures has been pursued for several
decades, deformations ranging from slight displacement to
catastrophic failure have been observed in many earth retaining
structures during the recent major earthquakes, including the
1999 Ji-Ji earthquake [1], the 2004 Chuetsu earthquake [2], and
the 2008 Wenchuan earthquake [3].
To date, the theoretical approach is the most widely used method
to analyze seismic stability of earth retaining structures. The majority
of the methods used by practitioners usually require estimating the
earth pressure behind the wall and expressing the stability of soil
structure by a factor of safety. The effect of earthquake is represented
pseudo-statically by an approximate static force acting in the
horizontal direction. To compute the active earth thrust acting
against retaining walls in seismic conditions, the MononobeOkabe
method or its extensions are most widely used [46]. The Mononobe
Okabe solution treats earthquake loads as pseudo-dynamic, generated
by uniform acceleration in the backll. The retained soil is considered
as perfect plastic material, which fails along a planar surface, thereby
exerting a limit thrust on the wall. The method has prevailed mainly
 Corresponding author.
E-mail address: hsm112003@yahoo.com.cn (S. He).
due to its simplicity and the familiarity of engineers with the
Coulomb method. However, the MononobeOkabe method presents
a basic shortcoming: the solution is based on the limit equilibrium of
the soil wedge without taking into account the presence of the wall.
Caltabiano et al. [7] suggested a new solution based on the pseudo-
static equilibrium of the soilwall system and applied it to soilwall
systems with surcharged backlls. More recently, Mylonakis et al. [8]
presented a stress plasticity solution for determining gravitational and
earthquake-induced earth pressures on gravity walls retaining
cohesionless soil. The solution is essentially an approximate slip line
approach, based on the theory of discontinuous stress elds, and takes
into account the following parameters: (1) weight and friction angle
of the soil material, (2) wall inclination, (3) backll inclination,
(4) wall roughness, (5) surcharge at soil surface, and (6) horizontal
and vertical seismic acceleration.
The nite element method is certainly the most comprehen-
sive approach to analyze the performance of soil structures
subjected to seismic loading. Psarropoulos et al. [9] utilized the
nite-element method to study the dynamic earth pressures
developed on rigid or exible non-sliding retaining walls. And
more recently, a two-dimensional, effective-stress nite element
procedure in conjunction with a generalized elasto-plastic
constitutive model, with slight modications, was conducted by
Alyami et al. [10]. Certainly, the nite element method has some
advantages in considering the natural failure mechanisms and the
interaction of structuresoil system, however, its use usually
requires high numerical costs and accurate measurements of the
properties of the component materials, which are often difcult to
achieve. This makes the use of nite element method not very
attractive for current applications.

0267-7261/$ - see front matter & 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.soildyn.2010.04.005
876 X. Li et al. / Soil Dynamics and Earthquake Engineering 30 (2010) 875878
As well known, design based on pseudo-static approach is
generally considered conservative, since even when the safety
factor drops below one the soil structure could experience only a
nite displacement rather than a complete failure [11,12]. Over
the past several decades, analytical methods have been developed
to estimate the displacement of retaining walls under earthquake
loading for specic applications. Richards and Elms [4] formerly
employed the Newmark procedure in evaluations of earthquake-
induced displacements of gravity retaining walls. A design
procedure based on a tolerable displacement against sliding was
proposed by Ling et al. [13]. They employed the Newmarks
sliding block method to evaluate permanent displacement of
reinforced slopes during earthquakes. And the same approach has
been extended by Ling and Leshchinsky [14] to include the
vertical component of ground acceleration. More recently,
Trandar et al. [2] conducted a displacement approach for the
seismic stability of gravity retaining walls based on the sliding
block concept.
In order to solve a limit state problem, limit analysis is also an
effective methodology. The upper bound approach of limit
analysis was used by Chen [15], Skrabl and Macuh [16], and Yang
[17] to consider the problem of earth pressures. And more
recently, a method based on limit analysis for calculations of yield
acceleration and seismic displacements of multi-block structures
(including retaining wall) was suggested by Michalowski [18].
In the present study, the upper bound approach of limit
analysis is applied to calculate the yield acceleration for gravity
retaining walls undergoing direct sliding failure. The retaining
wall and the backll soil were taken as a whole system. In the
framework of limit analysis, a simple block approach was used to
solve the seismic stability of the system. The computed results of
yield acceleration are compared with that of limit equilibrium
method. The closed-form solution may be found useful by
engineers in the displacement-based seismic design of retaining
walls.
2. Theoretical model of analysis
The analysis of the seismic stability of walls retaining backll
soil is based on the following assumptions: (1) the wallsoil
system is long enough for the end effects to be ignored (plane
strain condition); (2) the soil is homogeneous, dry, and cohesion-
less; (3) the retaining wall can be subjected only to horizontal
displacements; (4) the seismic action is uniform horizontally
distributed in the whole mass of the system; and (5) the failure
wedge is a plain. Furthermore, the upper bound limit analysis is
Fig. 1. (a) Translational failure mechanism of retaining wall with backll soil and (b) velocity compatibility between adjacent blocks and velocity hodograph.
based on the assumption that soil will be deformed according to
the associated ow rule and the convexity of the soil yield
condition. In the following analysis, we assumed that these
conditions are met.
The possible failure modes considered in this work is direct
sliding mechanism which is illustrated in Fig. 1(a). In this failure
mechanism, the gravity wall slides over the bottom layer once the
ground acceleration threshold for slide is exceeded. It should be
noted that the retaining wall herein is idealized as constructing
directly on the ground surface, but in reality, massive retaining
walls always have some embedment depth.
The upper bound theorem of limit analysis states that the soil
wall system will start to slide under its own weight plus inertia
force induced by earthquake and any other loads, if the rate of
work done by the external forces exceed the rate of internal
energy dissipation for any assumed kinematically admissible
failure mechanism. So, the yield acceleration factor kc can be
given by equating the rate of external work to the rate of energy
dissipation.
The failure mode shown in Fig. 1(a) consists of two wedges, the
soil wedge and the retaining wall wedge. This failure mechanism
is geometrically specied by wall height H, backll soil slope a,
and the angle that planar failure surface makes with the
horizontal b. The rate of work done by the gravity forces is the
vertical component of the velocities multiplied by the weight of
the wedges
_ g Ws V0 sinbjWw V1 sin db
W 1
where Ws and Ww indicate the weight of soil wedge and retaining
wall, respectively, and db is friction angle between retaining wall
and the base.
Once the system is subjected to horizontal seismic loading, the
rate of the inertial force needs to be accounted for in the energy
balance equation. It can be calculated analogously to the rate of
work of the soil weight, and be written as
_ s kh Ws V0 cosbj kh Ww V1 cos db
W 2
where kh is seismic coefcient representing horizontal accelera-
tion as a fraction of the gravity acceleration.
Since there is no cohesion along the slip surface and the
retaining wall base, energy dissipation is zero, the energy balance
equation yields
Ws V0 sinbj ky Ws V0 cosbjWw V1 sin db ky Ww V1 cos db 0
3
where ky is the yield acceleration coefcient of the failure
mechanism with respect to angle b.
 X. Li et al. / Soil Dynamics and Earthquake Engineering 30 (2010) 875878 877

Fig. 2. Forces and acceleration on gravity wall. (a) MononobeOkabe method and (b) soilwall system after Caltabiano et al. [7].

For a kinematically admissible failure mechanism, some

relationship should be satised between the velocities V0 and
the V1. Let us observe the two adjoining wedges as shown in
Fig. 1(b). The left and right wedges move with the absolute
velocities V1 and V0 which incline at angles db and j to their
bases, respectively. The relative velocity of the left wedge with
respect to the right one along the interface is represented as V01,
which inclines at an angle d. To allow the velocities assigned to
the wedge failure mechanism to be kinematically compatible, the
two adjoining wedges must not move to cause overlap or
indentation. This implies that the velocity hodograph must be
closed, i.e.,
V0 V01 V1 4
From Eq. (4) and Fig. 1(b), we obtain
cosdb d
V0 V1 5
cosj db Fig. 3. Retaining wall with dry backll considered in the analysis.

where d is friction angle between retaining wall and backll soil.

Substituting Eq. (5) into Eq. (3) and rearranging the terms
leads to the following expression for ky:
Ww sin db Ws cosdb d=cosj dbsinbj
ky
Ww cos db Ws cosdb d=cosj dbcosbj
The critical seismic coefcient is obtained by minimising ky,
with respect to b. This means that taking the rst derivatives of ky
and equating them to zero, i.e. (qky/qb)0. Solving this equation
and substituting the value of b, the least upper bound value of
yield acceleration factor is calculated. This critical value of ky is
indicated in the following text as kc.
3. Comparable models in literatures
It is instructive to compare the values of kc calculated by
means of the expressions derived in the previous sections to those
obtained from other solutions publishing in the literature in order
to show the possible differences. Several plots have been recently
published to illustrate the effect of seismic forces on the stability
of gravity retaining walls.
A pseudostatic type of the rotating block method was
developed by Zeng and Steedman [6] to calculate the permanent
rotational displacement of gravity retaining walls under earth-
quake loading. The model is similar to the sliding block method of
Newmark, which denes a threshold acceleration in rotation, and
each time this threshold acceleration is exceeded, permanent
rotation is accumulated. And the model was validated by
comparison with data from centrifuge model tests. At the same
time, they also gave a method to calculate the critical acceleration
coefcient for lateral sliding. For the wall shown in Fig. 2(a), at the
instant when sliding starts, equilibrium of forces in the horizontal
direction gives
6
ky Ww PAE cos d Ww PAE sin dtan db 7
where PAE is the dynamic earth pressure acting on the retaining
wall. PAE and its acting point can be determined by the
MononobeOkabe solution.
Caltabiano et al. [7] presented a new solution based on the
pseudostatic equilibrium analysis of the soilwall system with
surcharge backlls. The soilwall system considered in their
analysis is as shown in Fig. 2(b). At failure, the equilibrium
between resisting and driving forces is given by the expression:
Ww tan db ky Ww Ws ky tanbj 8
The critical seismic coefcient kc is obtained similarly by
minimising ky, with respect to b.
4. Illustrative examples and discussions
4.1. Illustrative example
A gravity wall made of concrete and supporting dry backll is
shown in Fig. 3. The wall is constrained to horizontal slide only. The
yield acceleration factors for slide can be determined Using Eqs. (7)
and (8), and the approach of this study are 0.236, 0.112, and 0.236,
respectively. It is noted that limit analysis gives the same value of
878 X. Li et al. / Soil Dynamics and Earthquake Engineering 30 (2010) 875878
Fig. 4. Effects of wall roughness (coefcient l): (a) on kc and (b) on the corresponding b.
yield acceleration factor with Eq. (7) in which the dynamic earth
pressure was calculated by MononobeOkabes solution.
Actually, the results of Eq. (7) equal to that of limit analysis using
a translational failure mechanism. Though, in this case, the upper-
bound theorem of limit analysis yields the same results as the limit
equilibrium method, the two methods are very different. The
equivalence of the results for limit load problems obtained using
the upper-bound theorem of limit analysis and the limit equilibrium
method has been mentioned in the literature on a few occasions
[15,19,20], and was discussed in detail by Michalowski in one of his
studies [20].
4.2. Effect of friction angle between retaining wall and backll soil
Fig. 4 shows the inuence of the friction angle between
retaining wall and backll soil on the critical seismic coefcient kc
and failure surface in the cases of a 01, db j 251, 301, 351, and
401. In the gure, kc is calculated using Eq. (6) and plotted against
the dimensionless abscissa l which is dened as
d the Geotechnical Engineering Division 1979;105(GT4):44964.
l 9
j Journal of Geotechnical Engineering 1983;109(7):91531.
where 0plp1. That is to say, the friction of wall back is
limited to a range between 0 and j. As can be expected, kc
increases with increasing l, and b decreases with increasing l. It
can be seen that the roughness of wall back has a remarkable
effect on the yield acceleration factors of the gravity retaining
wall. As the case of j 301, kc increases dramatically from 0.112
for smooth wall with l 0 to 0.236 for rough wall with l 1.0. So
in the design of retaining wall under earthquake loading, wall
roughness should be taken into account properly.
5. Conclusions
This work attempts to develop a method to analyze the seismic
stability of gravity retaining walls with backll under the category
of upper bound theorem of limit analysis. For a translational
failure mechanism assumed, closed-form solutions are derived
that are based on the soilwall system analysis. The expressions
derived in this paper can be conveniently used to calculate the
critical acceleration factor, kc, which is a key parameter to
evaluate the seismic stability of gravity retaining walls. The
conducted illustrative example calculation showed coincidence
between the proposed method and that from classical Mono-
nobeOkabe solutions. Ultimately, the inuence of wall roughness
on the critical acceleration factor was analyzed. It is expected that
this limit analysis method can be useful in current applications
because of its simplicity and reasonability.
View publication stats
Acknowledgements
This research has received nancial supports from the 973
Program of China (Grant no. 2008CB425802), and the West Light
Foundation of The CAS (Grant no. 09R2200200). These supports
are greatly appreciated.
References
[1] Ling HI, Leshchinsky D, Chou NNS. Post-earthquake investigation on several
geosynthetic-reinforced soil retaining walls and slopes during the Ji-Ji
earthquake of Taiwan. Soil Dynamics and Earthquake Engineering 2001;21:
297313.
[2] Trandar AC, Kamai T, Sidle RC. Earthquake-induced displacements of gravity
retaining walls and anchor-reinforced slopes. Soil Dynamics and Earthquake
Engineering 2009;29:42837.
[3] Han Q, Du X, Liu J, Li Z, Li L, Zhao J. Seismic damage of highway bridges during
the 2008 Wenchuan earthquake. Earthquake Engineering and Engineering
Vibration 2009;8(2):26373.
[4] Richards R, Elms DG. Seismic behavior of gravity retaining walls. Journal of
[5] Nadim F, Whitman RV. Seismically induced displacement of retaining walls.
[6] Zeng X, Steedman RS. Rotating block method for seismic displacement of
gravity walls. Journal of Geotechnical and Geoenvironmental Engineering
2000;126(8):70917.
[7] Caltabiano S, Cascone E, Maugeri M. Seismic stability of retaining walls with
surcharge. Soil Dynamics and Earthquake Engineering 2000;20:46976.
[8] Mylonakis G, Kloukinas P, Papantonopoulos C. An alternative to the
MononobeOkabe equations for seismic earth pressures. Soil Dynamics and
Earthquake Engineering 2007;27:95769.
[9] Psarropoulos PN, Klonaris G, Gazetas G. Seismic earth pressures on rigid and
exible retaining walls. Soil Dynamics and Earthquake Engineering 2005;25:
795805.
[10] Alyami M, Rouainia M, Wilkinson SM. Numerical analysis of deformation
behaviour of quay walls under earthquake loading. Soil Dynamics and
Earthquake Engineering 2009;29:52536.
[11] Newmark NM. Effects of earthquake on dams and embankments. Geotechni-
que 1965;15:13960.
[12] Ausilio E, Conte E, Dente G. Seismic stability analysis of reinforced slopes. Soil
Dynamics and Earthquake Engineering 2000;19:15972.
[13] Ling HI, Leshchinsky D, Perry EB. Seismic design and performance of
geosynthetic-reinforced soil structures. Geotechnique 1997;47(5):93352.
[14] Ling HI, Leshchinsky D. Effects of vertical acceleration on seismic design of
geosynthetic-reinforced soil structures. Geotechnique 1998;48(3):34773.
[15] Chen WF. Limit analysis and soil plasticity. Amsterdam (The Netherlands):
Elsevier Science; 1975.
[16] Skrabl S, Macuh B. Upper-bound solutions of three-dimensional passive earth
pressures. Canadian Geotechnical Journal 2005;42:144960.
[17] Yang XL. Upper bound limit analysis of active earth pressure with different
fracture surface and nonlinear yield criterion. Theoretical and Applied
Fracture Mechanics 2007;47:4656.
[18] Michalowski RL. Displacements of multiblock geotechnical structures sub-
jected to seismic excitation. Journal of Geotechnical and Geoenvironmental
Engineering ASCE 2007;133(11):14329.
[19] Yu HS. Plasticity and geotechnics. New York: Springer; 2006.
[20] Michalowski RL. Three-dimensional analysis of locally loaded slopes.
Geotechnique 1989;39(1):2738.