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

net/publication/255950595

Seismic passive resistance by pseudo-dynamic method

Article  in  Géotechnique · November 2005

DOI: 10.1680/geot.2005.55.9.699

CITATIONS READS

168 2,799

2 authors:

Deepankar Choudhury Sanjay Shrawan Nimbalkar

Indian Institute of Technology Bombay University of Technology Sydney
273 PUBLICATIONS   3,710 CITATIONS    149 PUBLICATIONS   2,465 CITATIONS   

SEE PROFILE SEE PROFILE

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

Liquefaction and Post-liquefaction Resistance of Lightly Cemented Sands View project

Liquefaction and Cyclic mobility analysis of clayey soils and GIS mapping of hazards View project

All content following this page was uploaded by Deepankar Choudhury on 28 May 2014.

The user has requested enhancement of the downloaded file.

Choudhury, D. & Nimbalkar, S. (2005). Géotechnique 55, No. 9, 699–702

TECHNICAL NOTE

Seismic passive resistance by pseudo-dynamic method

D. C H O U D H U RY * a n d S . N I M BA L K A R *

KEYWORDS: dynamics; earth pressure; earthquakes; retaining A C

walls; soil–structure interaction; time dependence Qv

Qh z
z
INTRODUCTION Ppe W
The pioneering work on earthquake-induced lateral earth H F
1ä ö
pressure under active and passive conditions acting on a dz
retaining wall was reported by Okabe (1926) and Mononobe
& Matsuo (1929), based on a pseudo-static approach by h Vs, Vp
following Coulomb’s earth pressure analysis under static á
B
conditions, and is known as the Mononobe–Okabe method
(Kramer, 1996). But in the pseudo-static method, the dy-
namic nature of earthquake loading is considered in a very Fig. 1. Model for computation of pseudo-dynamic passive earth
approximate way. The phase difference due to finite shear pressure
wave propagation behind a retaining wall can be considered
using a simple and more realistic pseudo-dynamic method,
proposed by Steedman & Zeng (1990). Again Zeng & Fig. 1. For most geological materials Vp /Vs ¼ 1.87 (Das,
Steedman (1993) compared the theoretical results with cen- 1993). A period of lateral shaking T ¼ 2/ø ¼ 4H/Vs
trifuge model test results to validate the pseudo-dynamic (Kramer, 1996) is considered in the analysis. A planar
method. rupture surface BC, inclined at an angle Æ to the horizontal,
Steedman & Zeng (1990) and Choudhury & Nimbalkar is assumed for the analysis to avoid further complication of
(2005) considered the case of active earth pressure behind a the problem. The assumption of a planar rupture surface for
retaining wall by a pseudo-dynamic method of analysis. But seismic passive earth pressure was also made by previous
the case of passive earth pressure or passive resistance researchers such as Okabe (1926) and Mononobe & Matsuo
behind a retaining wall by this pseudo-dynamic method has (1929) (see Kramer, 1996), and Davies et al. (1986).
not received any attention so far. Hence, in this paper, the If the base is subjected to harmonic horizontal and
pseudo-dynamic method is applied to determine the seismic vertical accelerations of amplitudes ah and av , the accelera-
passive resistance behind a rigid retaining wall. tions at depth z below the top of the wall can be expressed
Steedman & Zeng (1990) did not consider the effect of as
vertical seismic acceleration on the active earth pressure,

which was corrected by Choudhury & Nimbalkar (2005). In H
z
ah ð z, tÞ ¼ ah sin t
(1)
this paper, the effects of both horizontal and vertical seismic Vs
accelerations on passive earth pressure are considered. Also,

H
z
as an improvement over the Steedman & Zeng (1990) analy- av ð z, tÞ ¼ av sin t
(2)
Vp
sis, the effect of variation of different parameters such as
wall friction angle
, period of lateral shaking T, soil friction
For a thin element of thickness dz at depth z, as shown in
angle , horizontal and vertical seismic coefficients kh , kv ,
Fig. 1, the mass is given by
shear wave velocity Vs and primary wave velocity Vp are
considered in the present analysis. ªH
z
mð zÞ ¼ dz (3)
g tan Æ
The weight of the whole wedge is
METHOD OF ANALYSIS
The pseudo-dynamic analysis considers finite shear velo- 1 ªH2
city. It is assumed that the shear modulus is constant with W¼ (4)
2 tan Æ
depth through the backfill, and that only the phase and not
the magnitude of acceleration is varying. The total horizontal inertia force acting on the wall can be
Consider a fixed-base vertical cantilever wall AB of height expressed as
H, supporting horizontal cohesionless backfill as shown in ðH
Fig. 1. Under earthquake conditions, the shear wave velocity Qh ð tÞ ¼ mð zÞah ð z, tÞdz
Vs ¼ (G/r)1=2 and primary wave velocity Vp ¼ [G(2
2)/ 0
r(1
2)]1=2 are assumed to act in the direction shown in
ºªah
¼ ½2 H cos ø þ ºð sin ø
sin øtÞ
Manuscript received 7 June 2004; revised manuscript accepted 19 4 g tan Æ
2
August 2005. (5)
Discussion on this paper closes on 2 May 2006, for further details
see p. ii.
* Department of Civil Engineering, Indian Institute of Technology The total vertical inertia force acting on the wall can be
Bombay, India. expressed as

699
700 CHOUDHURY AND NIMBALKAR
ðH
Table 1. Typical result of seismic passive earth pressure coef-
Qv ð tÞ ¼ mð zÞav ð z, tÞdz ficient (Kpe ) using optimisation technique (Data used:
308,
0

/2, kh 0.2, kv 0.1, H/TVs 0.3 and H/TVp 0.16)
ªav
¼ ½2 H cos øł þ ºð sin øł
sin øtÞ
42 g tan Æ Parameter Value*
(6) Angle of failure surface with horizontal, Æ 22.58 (25.58)
Dimensionless parameter, t/T 0.231
where º ¼ TVs is the wavelength of the vertically propagat- Seismic passive earth pressure coefficient, Kpe 3.220 (4.120)
ing shear wave;  ¼ TVp is the wavelength of the vertically
propagating primary wave;  ¼ t
H/Vs; and ł ¼ t
H/ *Values in parenthesis are by the Mononobe–Okabe method of
Vp. pseudo-static analysis for the given data
The special case of a rigid wedge is given, in the limit, as
ª H 2 ah ah
lim ð Qh Þmax ¼ ¼ W ¼ kh W (7) for a typical set of values such as  ¼ 308,
¼ /2, kh ¼
Vs !1 2 g tan Æ g 0.2, kv ¼ 0.5kh , H/º ¼ 0.3 and H/ ¼ 0.16. It can be seen
ª H 2 av av that the dynamic wedge computed by the present study (Æ ¼
lim ð Qv Þmax ¼ ¼ W ¼ kv W (8) 22.58) is larger than that obtained by the Mononobe–Okabe
Vp !1 2 g tan Æ g
method (Æ ¼ 25.58), and both the pseudo-static and pseudo-
which is equivalent to the pseudo-static forces assumed in dynamic methods result in smaller wedges than Coulomb’s
the Mononobe–Okabe method. The total (static and dy- static wedge (Æ ¼ 20.78) (Kramer, 1996).
namic) passive resistance can be obtained by resolving The seismic passive earth pressure distribution is obtained
forces on the wedge: that is, by differentiating the total seismic passive resistance acting
W sinðÆ þ Þ
Qh cosðÆ þ Þ
Qv sinðÆ þ Þ on the retaining wall with respect to the depth of the wall.
Ppe ¼ This is similar to the methodology described by Steedman &
cosðÆ þ
þ Þ Zeng (1990) and Choudhury & Nimbalkar (2005) for the
(9) seismic active earth pressure distribution. The distribution of
the seismic passive earth pressure, ppe , can be expressed as
The seismic passive earth pressure coefficient, Kpe , is defined dPpe ð tÞ ªz sinðÆ þ Þ
as ppe ð tÞ ¼ ¼
dz tan Æ cosðÆ þ
þ 
Þ 
2Ppe k h ªz cosðÆ þ Þ z
K pe ¼ (10)
sin ø t

ªH2 tan Æ cosðÆ þ
þ Þ Vs
(12)


k v ªz sinðÆ þ Þ z

sin ø t

Substituting for Qh and Qv in equation (9), an expression for tan Æ cosðÆ þ
þ Þ Vp
Kpe in terms of Qh , Qv and W can be derived. From equation
(11) it is seen that Kpe is a function of the dimensionless From equation (12) it can be seen that the seismic passive
parameters H/TVs , H/TVp , t/T and the wedge angle Æ. pressure changes as a non-linear function of depth with a

 shape that depends on the ratio H/º and H/. Thus the
1 sinðÆ þ Þ kh TVs
K pe ¼
2 seismic passive earth pressure distribution is clearly non-
tan Æ cosð
þ  þ ÆÞ 2 tan Æ H linear.


cosðÆ þ Þ kv TVp
3 3 m1
2
cosð
þ  þ ÆÞ 2 tan Æ H RESULTS AND DISCUSSIONS
sinðÆ þ Þ In Fig. 2, the typical variations of seismic passive earth
3 3 m2 (11) pressure distribution for different horizontal seismic accel-
cosð
þ  þ ÆÞ
0·0
where



t H TVs
m1 ¼ 2 cos 2
þ 0·2
T TVs H kh 5 0
"

# kh 5 0·2
t H t
3 sin 2

sin 2 kh 5 0·4
T TVs T 0·4



z/H

t H TVp
m2 ¼ 2 cos 2
þ
T TVp H 0·6
"

#
t H t
3 sin 2

sin 2
T TVp T 0·8
kv 5 0
······ kv 5 0·5 kh
The minimum value of Kpe is obtained by optimising it
1·0
with respect to t/T and Æ. Table 1 shows a typical output of
0 1 2 3 4 5
the Kpe values. It is found that Kpe is a function of H/TVs ppe/ãH
and H/TVp , which is the ratio of time for a shear wave and
primary wave to travel the full height of the wall to the Fig. 2. Typical variation of seismic passive earth pressure
period of lateral shaking. Table 1 also gives the inclination distribution for different kh and kv (
308, 
/2, H/º
of slip wedges to the horizontal (Æ) by the present method 0.3, H/h 0.16)
 SEISMIC PASSIVE RESISTANCE BY PSEUDO-DYNAMIC METHOD 701
erations (kh ) and vertical seismic accelerations (kv ) are 0
presented for  ¼ 308,
¼ /2, H/º ¼ 0.3 and H/ ¼ 0.16.
From these results it is clear that the presence of seismic
forces, with either kh or kv , induces a reduction in the seismic 0·2
passive earth pressure behind the retaining wall. From the
plot, it may be seen that the seismic passive earth pressures at
the base of the retaining wall for kh ¼ 0.4 and kh ¼ 0.2 are 0·4
respectively about 60% and 23% smaller than those for kh ¼
0 with kv ¼ 0.5kh . Again, from Fig. 2, it can be observed that

z/H
the seismic passive earth pressures at the base of the retaining 0·6
wall for kv ¼ 0.5kh are about 8% and 20.8% smaller than
those for kv ¼ 0 for kh ¼ 0.2 and 0.4 respectively. Though
T 5 0·2 s
the effect of vertical seismic acceleration on seismic passive
0·8 T 5 0·3 s
resistance is hardly considered in the analysis by many
T 5 0·4 s
researchers, the present study reveals the significant influence
T 5 0·5 s
of vertical seismic acceleration on the seismic passive resis-
tance. 1·0
Figure 3 shows a plot of the normalised distribution of 0 1 2 3 4 5 6
ppe/ãH
seismic passive earth pressure with the height of the retain-
ing wall for different values of soil friction angle  and wall Fig. 4. Effect of period of lateral shaking T on seismic passive
friction angle
for kh ¼ 0.2, kv ¼ 0.5kh , H/º ¼ 0.3 and H/ earth pressure distribution for 
/2,
308, kh 0.2 and
 ¼ 0.16. From the plot, it may be seen that the seismic kv 0.5kh
passive earth pressures at the base of the retaining wall for
 ¼ 408 are about 133% and 360% greater than that for 
¼ 208 for
/ ¼ 0.0 and 0.5 respectively. Thus seismic T ¼ 0.3 s, T ¼ 0.4 s and T ¼ 0.5 s. But as the period of
passive resistance decreases significantly with decrease of lateral shaking increases, the total seismic passive resistance
soil friction angle. From the plot, it can also be observed decreases. From Fig. 4 it may also be seen that the differ-
that the seismic passive earth pressures at the base of the ence in seismic passive earth pressure distribution along the
retaining wall for
¼ /2 are about 24% and 150% greater height of the retaining wall is marginal when T varies from
than those at
¼ 0 for  ¼ 208 and 408 respectively. 0.3 to 0.5 s. Now for most geotechnical structures T ¼ 0.3 s
Hence, similar to the static case, in the seismic case passive closely describes the practical state (Prakash, 1981). Hence
resistance also increases with increase in wall friction angle. in the present study, for all other results reported in this
Figure 4 shows a plot of the normalised distribution of paper, a value of T ¼ 0.3 s is considered.
seismic passive earth pressure with the height of the retain-
ing wall for different values of period of lateral shaking T
for
¼ /2,  ¼ 308, kh ¼ 0.2 and kv ¼ 0.5kh . From the
plot it can be seen that seismic passive earth pressure near COMPARISON OF RESULTS
the top of the retaining wall corresponding to T ¼ 0.2 s is Results obtained by the pseudo-dynamic method for seis-
8%, 15% and 26% smaller than those corresponding to T ¼ mic passive resistance are still scarce. Hence, in Fig. 5, the
0.3 s, T ¼ 0.4 s and T ¼ 0.5 s respectively. It becomes larger present results for seismic passive earth pressure coefficient
near the mid-height of the retaining wall and, further, it Kpe are compared with those obtained by the pseudo-static
shows a substantial increase in the passive earth pressure, method of analysis by Davies et al. (1986) by considering
which is nearly 27%, 37% and 36% larger than those for the Mononobe–Okabe method for different values of kh and
 with
¼ /2, kv ¼ 0.5kh , H/º ¼ 0.3 and H/ ¼ 0.16.
The present pseudo-dynamic method gives the minimum
0 seismic passive resistance that is safe for the design of a
rigid retaining wall under passive conditions. Also, the
highly non-linear nature of the seismic passive earth pressure
0·2 ä/ö 5 0
ä/ö 5 0·5 ö 5 20°
12
ö 5 30°
0·4 Davies et al. (1986) ö 5 40°
ö 5 20° 10
z/H

5 40°
······ ö
8
0·6
Kpe

Present study

0·8 4

2
1·0
0 1 2 3 4 5 6 7 8 9 10 11 0 0·1 0·2 0·3 0·4 0·5
ppe/ãH kh

Fig. 3. Effect of soil friction angle
and wall friction angle  Fig. 5. Comparison of seismic passive earth pressure coefficient
on seismic passive earth pressure distribution for kh 0.2, kv Kpe for different kh and
(
/2, kv 0.5kh , H/º 0.3, H/h
0.5kh , H/º 0.3 and H/h 0.16 0.16)
 702 CHOUDHURY AND NIMBALKAR
distribution is seen from the present study compared with T period of lateral shaking
the pseudo-static method, which gives a linear passive t time
pressure distribution under seismic conditions. Hence the Vp primary wave velocity
point of application of total seismic passive resistance will Vs shear wave velocity
also differ in the two methods. Æ angle of inclination of failure surface to the horizontal
ª unit weight of soil

wall friction angle
 t
H/Vs
CONCLUSIONS  Poisson’s ratio
Using a pseudo-dynamic method, the seismic passive earth r density of soil
pressure distribution behind a vertical retaining wall is  soil friction angle
obtained in terms of the variation of parameters such as soil ł t
H/Vp
friction angle, wall friction angle, horizontal and vertical ø angular frequency of base shaking
seismic accelerations, period of lateral shaking, and veloci-
ties of shear and primary waves. The seismic passive earth
pressure reduces with increase in both kh and kv . Typical
results show the highly non-linear nature of the seismic REFERENCES
passive earth pressure distribution by this pseudo-dynamic Choudhury, D. & Nimbalkar, S. (2005). Pseudo-dynamic approach
method compared with the existing linear seismic passive of seismic active earth pressure behind retaining wall. Geotech.
earth pressure distribution using a pseudo-static approach. Geol. Engng (in press).
Comparisons of the present method with the available Das, B. M. (1993). Principles of soil dynamics. Boston, MA: PWS-
Kent.
pseudo-static methods are reported, leading to the minimum Davies, T. G., Richards, R. & Chen, K. H. (1986). Passive pressure
seismic passive resistance by the present study. during seismic loading. J. Geotech. Engng ASCE 112, No. 4,
479–484.
Kramer, S. L. (1996). Geotechnical earthquake engineering. Upper
NOTATION Saddle River, NJ: Prentice Hall.
a(z, t) acceleration at depth z, time t Mononobe, N. & Matsuo, H. (1929). On the determination of earth
ah , av horizontal and vertical seismic acceleration pressure during earthquakes. Proc. World Engng Congress, 9,
G shear modulus of soil 274–280.
g acceleration due to gravity Okabe, S. (1926). General theory of earth pressure. J. Japanese
H height of retaining wall Soc. Civ. Engrs, 12, No. 1.
Kpe seismic passive earth pressure coefficient Prakash, S. (1981). Soil dynamics. New York: McGraw-Hill.
kh , kv seismic acceleration coefficient in horizontal and vertical Steedman, R. S. & Zeng, X. (1990). The influence of phase on the
directions calculation of pseudo-static earth pressure on a retaining wall.
Ppe pseudo-dynamic passive resistance Géotechnique 40, No. 1, 103–112.
Qh , Qv horizontal and vertical inertia forces due to seismic Zeng, X. & Steedman, R. S. (1993). On the behaviour of quay
acceleration walls in earthquakes. Géotechnique 43, No. 3, 417–431.

View publication stats