Tunnelling and

Underground Space
Technology
incorporating Trenchless
Technology Research
Tunnelling and Underground Space Technology 20 (2005) 411–417
www.elsevier.com/locate/tust

2-D analysis of circular tunnel against earthquake loading

a,* b
Mohammad C. Pakbaz , Akbar Yareevand
a
Department of Civil Engineering, Shahid Chamran University, Ahwaz, Iran
b
Department of Civil Engineering, Khuramabad University, Khuramabad, Iran

Received 12 February 2004; received in revised form 24 January 2005; accepted 30 January 2005
Available online 14 March 2005

Abstract

The use of underground structures such as tunnel for subways, highways, material storage, and sewage and water transport is
increasing in developed countries. The safety of these facilities during operation in areas with seismic activities such as in Japan,
Taiwan and Turkey in recent earthquakes has been questioned. Dynamic effects on these structures are in the form of deformations
that they experience during earthquakes. In this paper, first latest methods on the subject are reviewed and then the interaction
between the ground and tunnel lining during earthquake excitation is investigated by a finite difference computer program
(CA2). Analysis show that a good agreement between analytical closed form and numerical solutions exist. According to the results
obtained in this study some practical suggestion for using closed form solution are also given.

2005 Elsevier Ltd. All rights reserved.

Keywords: Earthquake; Flexibility; Lining; Seismic; Tunnel

1. Introduction tunnel during the earthquake) proved otherwise.

In recent years because of construction and increas-
ing trend in using large tunnels and underground
spaces, the safety of these structures during operation
and their response against earthquake loading have
been addressed by many researchers. For some times
general belief has been that effect of earthquake on
tunnels and underground spaces is not very important
and have long been assumed to have the ability to sus-
tain earthquakes with little damage. Nevertheless, some
underground structures have experienced significant
damage in recent earthquake including the 1995 Kobe,
Japan earthquake, the 1999 Chi-Chi Taiwan earth-
quake and the 1999 Kocaeli Turkey earthquake (Has-
hash et al., 2001) and recently 2004 Baladeh Iran
earthquake (principal authorÕs personal experience
regarding many rock falls in the vicinity of Kandouvan
*
Corresponding author. Tel./fax: +98 611 333 7010.
E-mail address: mpakbaz@yahoo.com (M.C. Pakbaz).
According to recent studies and observed failures of
underground structures, researchers relate response of
these structures during earthquake to various parame-
ters. According to statistical studies following general
conclusion in this regard are drawn (Dowding and Ro-
zen, 1978; Owen and Scholl, 1981; Wang, 1993; Has-
hash et al., 2001):
1. The effect of earthquake on underground struc-
tures depend on factors such as; depth of tunnel
below the ground surface, type of soil or rock sur-
rounding tunnel, maximum ground acceleration,
intensity of earthquake, distance to earthquake
epicenter and type of tunnel lining.
2. Failures in rock tunnels are much less than those
in earth tunnel.
3. Underground structures as compared to structures
on the ground have more safety.
4. General failures occur when an active fault is
crossed the axis of the tunnel.

0886-7798/$ - see front matter
2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tust.2005.01.006
412 M.C. Pakbaz, A. Yareevand / Tunnelling and Underground Space Technology 20 (2005) 411–417
5. Failures of near surface tunnels such as a cut
and cover tunnel is more than those of deep
tunnels.
6. More damage occurs to portal structures than to
the underground tunnel itself due to slope
instability.
7. Lined and grouted tunnels are safer than unlined
tunnel in rock, shaking damage can be reduced
by stabilizing the ground around the tunnel and
by improving the contact between the lining and
the surrounding ground through grouting.
8. Tunnels are more stable under symmetric load,
which improves ground–lining interaction.
Improving tunnel lining by placing thicker and
stiffer sections without stabilizing surrounding
poor ground may result in excess seismic forces
in the lining. Backfilling with non-cyclically
mobile material and rock-stabilizing measure
may improve the safety and stability of shallow
tunnels.
9. Damage may be related to peak ground accelera-
tion and velocity based on the magnitude and epi-
central distance of the affected earthquake.
10. Duration of strong-motion shaking during earth-
quakes is utmost importance because it may cause
fatigue failure and therefore, large deformation.
11. High frequency motions may explain the local
spalling of rock or concrete along planes of weak-
ness. These frequencies which rapidly attenuate
with distance, may be expected mainly at small dis-
tances from the causative fault.
12. Ground motion may be amplified upon incidence
with a tunnel if wavelength are between one and
four times the tunnel diameter.
Because Iran is in an active seismic zone and because of
increasing number of tunnels and underground con-
struction in recent years in this country, the study of
earthquake response of these underground structures
needs special attention and treatment.
2. Effect of earthquake on tunnel
Earthquake effects on underground structures falls
into two groups: (1) ground shaking; and (2) ground
failures such as liquefaction, fault displacement, and
slope instabilities (Hashash et al., 2001). In this paper
item (1) will be considered. When an earthquake oc-
curs various bodies and surface waves are produced.
Effect of earthquake waves on an underground struc-
ture is related to three factors namely (i) source re-
gion, (ii) transmitting region and (iii) site region
(Wang, 1993). After earthquake occurrence in source
region, waves transmit the produced energy to the
underground structure. Because of damping effects in
the transmitting region,only fraction of total energy
will reach the structure. Distance of the site from
the source region and type of material affect final ef-
fect of earthquake on the structure.
The behavior of a tunnel is sometimes approximated
to that of an elastic beam subject to deformations im-
posed by the surrounding ground. The response of tun-
nels to earthquake shaking appears in three forms of
deformation (Owen and Scholl, 1981): axial compres-
sion and extension (Fig. 1(a),(b)), longitudinal bending
(Fig. 1(c),(d)), and ovaling/racking (Fig. 1(e),(f)). Axial
and curvature deformations in horizontal or near hori-
zontal tunnel occur because of wave propagation paral-
lel or at an angle to the axis of tunnel. Ovaling
deformations on the other hand are due to wave propa-
gating perpendicular or near perpendicular to the axis of
tunnel.
According to Hashash et al. (2001) the assessment of
underground structure seismic response consists of three
major steps:
1. Definition of the seismic environment and develop-
ment of the seismic parameters for analysis.
2. Evaluation of ground response to shaking, which
includes ground failure and ground deformation.
3. Assessment of structure behavior due to seismic shak-
ing including (a) development of seismic design load-
ing criteria, (b) underground structure response to
ground deformations and (c) special design issues in
the following sections items (3a) and (3b) are
discussed.
Fig. 1. Deformation modes of tunnels due to seismic waves (after
Owen and Scholl, 1981).
 M.C. Pakbaz, A. Yareevand / Tunnelling and Underground Space Technology 20 (2005) 411–417 413

2.1. Effect of axial and curvature deformations on circular

tunnel

2.1.1. Behavior of tunnel under free-field deformation

The term Ôfree-field deformationsÕ describes ground
strains caused by seismic waves in the absence of struc-
tures or excavations. These deformations ignore the
interaction between the underground structure and the
surrounding ground, but can provide a first-order esti-
mate of the anticipated deformation of the structure
(Hashash et al., 2001). Newmark (1968) and Kuesel
(1969) proposed a simplified method for calculating
free-field ground strains caused by a harmonic wave
propogating at a given angle of incidence in a homoge-
neous, isotropic, elastic medium (Fig. 2). The most crit-
ical incidence angle yielding maximum strain, is typically
used as a safety measure against the uncertainties of Fig. 2. Simple harmonic wave and tunnel (after Wang, 1993).
earthquake prediction. NewmarkÕs approach provides
an order of magnitude estimate of wave-induced strains where M, flexural moment; V, shear force; Q, trust
while requiring a minimal input, making it useful as force; h, angle of wave impact; Il, moment of inertia of
both an initial design tool and a method of design veri- tunnel lining; El, modulus of elasticity of lining material;
fication (Wang, 1993). D, amplitude of sine wave; L, shear wave length; and Al,
St. John and Zahrah (1987) used NewmarkÕs ap- section area of lining.
proach to develop solutions for free-field axial and cur-
vature strains due to compression, shear and Rayleigh 2.1.2. Behavior of tunnel in light of ground and structure
waves. Solutions for all three wave types are shown in interaction
Table 1. According to this method moment and forces The method presented in Section 2.1.1 based on
generated in tunnel lining are in the following form: upper limit of strains and forces. On the other hand,
 2   in the case of tunnels with rigid lining in loose soils, lin-
2p 3 2p
M¼
ðcos hÞ
El
I l
D
sin ; ð1Þ ing strains and those for soil are not equal and tunnel
L L= cos h
lining resists imposed deformations by the ground shak-
3   ing. Therefore, in this situation taking into account the
2p 2p
V ¼
ðcos4 hÞ
El
I l
D
cos ; ð2Þ soil–lining interaction could show more realistic behav-
L L= cos h
ior of tunnel. In this class of solutions the beam-on elas-
    tic foundation approach is used to model (quasi-static)
2p 2p
Q¼
cos h
sin h
El
Al
D
cos ; soil–structure interaction effects. The solutions ignore
L L= cos h
dynamic (inertial) interaction effects. Under seismic
ð3Þ loading, the cross-section of a tunnel will experience

Table 1
Strains due to body and surface waves (after St. John and Zahrah, 1987)
414 M.C. Pakbaz, A. Yareevand / Tunnelling and Underground Space Technology 20 (2005) 411–417

axial bending and shear strains due to free field axial, where tm is the PoisonÕs ratio of the ground material.
curvature, and shear deformations. The moments and Both of these equations assume the absence of the lin-
forces are evaluated from following relationships (after ing, therefore ignoring tunnel–ground interaction. In
St. John and Zahrah, 1987): the free-field, Eq. (9) would yield a much greater distor-
sion than Eq. (8) sometimes by a factor of 2 or 3. This
K t ðL=2pÞ2 provides a reasonable distortion criterion for a lining
M max ¼ 4

D; ð4Þ
1 þ 2ðK t =El I l Þ
ðL=2pÞ with little stiffness relative to the surrounding ground,
Eq. (8) is appropriate when the lining stiffness is equal
ðK t L=2pÞ to that of the medium (Hashash et al., 2001).
V max ¼ 4

D; ð5Þ
1 þ ðK t =El I l Þ
ðL=2pÞ
2.2.2. Behavior of tunnel in light of ground and lining
ðK a L=2pÞ interaction
Qmax ¼
D: ð6Þ
1 þ ðK a =Ec Ac Þ
ðL=2pÞ
2 In case of tunnel in soft ground the above relation-
ships do not consider the real behavior of the tunnel.
In the above relations, effect of ground–lining interac- In a case of rigid lining no deformation can be produced
tion is directly taken into account. Rigidity of the by the ground, and for a flexible lining, on the other
ground is taken into account by coefficient of axial hand interaction between lining and ground exists. In
and transverse flexibility Ka, Kt of medium and that of this regard, flexibility and compressibility ratio sug-
lining by modulus of lining ElAl and ElIl. gested by Peck et al. (1972) based on earlier work by
Hoeg (1968) can be used to take into account this factor:
2.2. Effect of ovaling deformation on circular tunnel
Em ð1  t2l ÞR3
F ¼ ; ð10Þ
2.2.1. Behavior of tunnel under free-field deformation El I l ð1 þ tm Þ
Ovaling deformations on circular tunnel are pro-
duced by wave propagation perpendicular to tunnel Em ð1  t2l ÞR
C¼ ; ð11Þ
axis and are therefore designed for in transverse direc- El tð1 þ tm Þð1  2tm Þ
tion (typically under 2-D, plane-strain conditions). where F, flexibility ratio; C, compressibility ratio, R,
According to studies made, propagation of vertical tunnel radius, and t, thickness of lining. According to
shear wave (SV); which cause vibration in horizontal Peck et al. (1972) if F < 20, interaction between lining
direction are most effective on producing ovaling and ground must be taken into account, otherwise
deformation around tunnel (Wang, 1993). For a tun- free-field approach explained above can be used.
nel at medium depth and for a deep tunnel in homo- According to Wang (1993):
geneous soil or rock, NewmarkÕs (1968) simple  
method may give rational estimate of such deforma- 1 Em
T max ¼  Kl Rc ; ð12Þ
tions around tunnel free-field maximum shear strain 6 ð1 þ tm Þ max
is obtained as following accordingly:  
Vs 1 Em
M max ¼  Kl R2 cmax ; ð13Þ
cmax ¼ ðh ¼ 00 Þ; ð7Þ 6 ð1 þ tm Þ
Cs
 
where Cs, shear wave velocity; and Vs, particle shear DD 1
wave velocity. By using free-field maximum shear ¼ K l F cmax ; ð14Þ
D 3
strain,produced deformation around tunnel can be ob-
 
tained in two forms: DDlining 2
(1) Assuming no presence of tunnel cavity in the ¼ K lF ; ð15Þ
Dfree-field 3
ground the ovaling effects from ground, in which case
the diametrical strain is a function of maximum free- where Kl is the coefficient of lining response:
field strain only 12ð1  tm Þ
Kl ¼ : ð16Þ
DD cmax 2F þ 5  6tm
¼ : ð8Þ
D 2 This solution is based on full-slip assumption between
(2) Ground deformation occurs in the presence of a lining and ground. This assumption is valid only for
cavity due to tunnel excavation, in which case the dia- the case of very soft soil or earthquake of high intensity.
metrical strain will be In many situation condition of partial or no-slip exist.
Although assumption of full slip create conservative re-
DD sults for Mmax and DD that compensate for errors re-
¼ 2cmax ð1  tm Þ; ð9Þ
D lated for using pseudo-static model for dynamic
 M.C. Pakbaz, A. Yareevand / Tunnelling and Underground Space Technology 20 (2005) 411–417 415

earthquake loading (Wang, 1993). For trust force in lin- 50 · 40 m in size with a tunnel radius of R = 4 m has
ing (T) assumption of no-slip between lining and soil is been used. The record of accelerograph of Naghan Fars
acceptable and following relationship is suggested earthquake (April 7, 1978) with maximum acceleration
(Wang, 1993): of 0.7g and intensity of 7 in scale of Richter (Fig. 3)
Em has been used. Physical properties of tunnel lining:
T max ¼ K 2 smax R ¼ K 2 Rc ; ð17Þ lining (assumed to be reinforced concrete) are:
2ð1 þ tm Þ max
Il = 4.3 · 105–7.6 · 103 m4 and El =ð1  t2l Þ ¼ 27
in which 106 ðkpaÞ.

F ½ð1 þ tm Þ  ð1  2tm ÞC   0:5ð1  2tm Þ2 þ 2

K2 ¼ 1 þ   ; ð18Þ
F ½ð3  2tm Þ þ ð1  2tm ÞC  þ C 2:5  8tm þ 6t2m þ 6  8tm

in order to compensate for possible errors due to using
pseudo-static model for dynamic earthquake loading it
has been suggested that Tmax in Eq. (17) be multiply
by a factor of 1.15 (Wang, 1993). Penzin and Wu
(1998) developed similar closed-form elastic solution
for thrust, shear, and moment in the tunnel lining due
to racking deformations. This solution result in values
of thrust and moment that are very close to those of
Wang (1993) for full-slip condition (Hashash et al.,
2001).
3. Numerical analysis
In order to study effect of earthquake on circular tun-
nel, numerical analysis was performed in two dimen-
sional elasticity-plasticity domain, using the computer
program called CA2 (Fakhimi, 1997). Main purposes
of this analysis is to compare between results obtained
Physical properties and parameters of ground mate-
rial surrounding the tunnel: cm = 2000 kg/m3, Em = 105
kpa, tm = 0.3, C = 1000 kpa, / = 40, T = 300–3000
kpa, f = 10%, in which T = tensile strength and
f = damping coefficient.
3.1. Results
3.1.1. Effect of earthquake on lining
In order to evaluate the effect of earthquake on tun-
nel lining, lining with various flexibility ratio F = 0.5–
89 were examined and results are shown in Fig. 4.
According to these results, effect of earthquake on max-
imum moment at F < 10 becomes significant where not
so for trust force T.
800
600
acceleration

400
by this method and by the closed form solution dis- 200
cussed earlier and to evaluate the significant factors 0
influencing tunnel–ground interaction. In this analysis -200 0 1 2 3 5
following assumptions have been made (Yareevand, -400
2000): -600
-800

(i) Material surrounding tunnel is homogeneous. time (sec)

(ii) Coefficient of lateral earth pressure K = 1. Fig. 3. Record of acceleration of Naghan Fars earthquake.
(iii) Elasto-plastic model for material surrounding tun-
nel and elastic beam element for lining.
(iv) Accelerogram is fed to finite difference mesh at the
bottom. 400
(v) A free-field boundary is used in dynamic analysis. 300
Mmax

(vi) Lining is placed immediately after excavation. 200

100
In order to avoid tensile crack during execution of 0
program, a high tensile strength for material is assumed. 0 10 20 30 40 50 60 70 80 90
Because the effect of earthquake on tunnel at shallow to
medium depth (h < 50 m) is more (Hwang and Lysmer, Flexibility Ratio
1981), depth of tunnel in these analysis is taken at 20 m Fig. 4. Effect of earthquake loading on flexural moment, CA2 (no-
below the ground surface. In the analysis a mesh of slip).
416 M.C. Pakbaz, A. Yareevand / Tunnelling and Underground Space Technology 20 (2005) 411–417

3.1.2. Effect of earthquake maximum acceleration on 500

lining behavior
The effect of earthquake maximum acceleration on 400
tunnel lining behavior is examined at F = 20 for
300
amax = (0.3–0.7)g using the time record of Naghan Fars

Tmax
earthquake. The results are shown in Figs. 5 and 6.
200
CLOSED FORM SOLUTION CA2
3.1.3. Comparison of numerical and closed form solution 100
Results of numerical and closed from solutions are
shown in Figs. 7 and 8. The closed form solution re- 0
quired information about cmax at the tunnel section. In 0 10 20 30 40 50 60 70 80 90
order to estimate cmax computer program SHAKE 91 Flexibility Ratio
(Idriss and Sun, 1991) was used and resulted
Fig. 8. Comparison of Tmax of CA2 (no-slip) and closed form (full-
slip) solution.

Influence of peak acceleration

45
Mmax and Vmax

40 Mmax (KN.m) Vmax (KN)

35 cmax = 0.00357. This cmax was used to calculate Mmax
30
and Tmax in closed form solution. This approach has
25
20
also been recommended by Hashash et al. (2001).
15
10
0.3 0.4 0.5 0.6 0.7 4. Conclusions
Peak acceleration
According to this study following general results are
Fig. 5. Effect of peak acceleration on Mmax, and Vmax, CA2 (no-slip). concluded:

Effect of earthquake on tunnel–ground interaction

380 depend on various parameters including peak acceler-
340 ation, intensity and duration of earthquake and the
300 relative rigidity between tunnel and ground.
Tmax

260 Increasing structural dimensions of lining in static

220 design can not always be a reliable method against
180 earthquake loading, because this would increase the
140 rigidity of lining (low F) and therefore, would
0.3 0.4 0.5 0.6 0.7 increase the effect of earthquake loading.
Peak acceleration Closed form solutions can be used for analyzing cir-
cular tunnel against earthquake loading. For this pur-
Fig. 6. Effect of peak acceleration on Tmax, CA2 (no-slip).
pose following relationships are suggested for
maximum flexural moment calculation:
400  
1 Em
F < 20 M max ¼ Kl R2 cmax ; ð19Þ
300 CLOSED FORM SOLUTION CA2
6 ð1 þ tm Þ
 
1 Em
Mmax

200 F > 20 M max ¼ Kl R2 cmax : ð20Þ

10 ð1 þ tm Þ

100

References
0
0 10 20 30 40 50 60 70 80 90 Dowding, C.H., Rozen, A., 1978. Damage to rock tunnels from
Flexibility Ratio earthquake shaking. Journal of the Geotechnical Engineering
Division – ASCE 104 (GT2), 175–191.
Fig. 7. Comparison of Mmax of CA2 (no-slip) and closed form (full- Fakhimi, A.A., 1997. Theory and Manual of CA2 Software. Center for
slip) solution. Housing and Building Research of Iran.
 M.C. Pakbaz, A. Yareevand / Tunnelling and Underground Space Technology 20 (2005) 411–417
Hashash, Y.M.A., Hook, J.J., Schmidt, B., Yao, J.I., 2001. Seismic
design and analysis of underground structures. Tunnelling and
Underground Space Technology 16, 247–293.
Hoeg, K., 1968. Stresses against underground structural cylinders.
Journal of Soil Mechanics and Foundation Division – ASCE 94
(SM4), 833–858.
Hwang, R.N., Lysmer, J., 1981. Response of burried structure to
traveling waves. Journal of the Geotechnical Engineering Division
– ASCE 107 (Gt2), 727–746.
Idriss, I.M., Sun, J.I., 1991. SHAKE 91, A Computer Program for
Conducting Equivalent Linear Seismic Response Analysis of
Horizontally Layered Soil Deposit. UserÕs Guide. University of
California, Davis, pp. 13.
Kuesel, T.R., 1969. Earthquake design criteria for subway. Journal of
Structural Division – ASCE 95 (ST6), 1213–1231.
Newmark, N.M., 1968. Problems in wave propagation in soil and
rock. Proceedings of the International Symposium on Wave
Propagation and Dynamic Properties of Earth Materials.
417
Owen, G.N., Scholl, R.E., 1981. Earthquake Engineering of Large
Underground Structures. Federal Highway Administration, Wash-
ington, DC.
Peck, R.B., Hendron, A.J., Mohraz, B., 1972. Sate of the art of soft
ground tunnelling. In: Proceedings of Rapid Excavation and
Tunnelling Conference, Chicago, IL, vol. 1.
Penzin, J., Wu, C., 1998. Stresses in linings of bored tunnels.
International Journal of Earthquake Engineering and Structural
Dynamic 27, 283–300.
St. John, C.M., Zahrah, T.f., 1987. Aseismic design of underground
structures. Tunnelling and Underground Space Technology 2
(2).
Wang, J.N., 1993. Seismic Design of Tunnels: a State of the Art
Approach, Monograph, Monograph 7. Parsons, Brinckerhoff,
Quade and Douglas Inc., New York.
Yareevand, A., 2000. Behaviour of Earth and tunnel lining under
dynamic loads, M.S. Thesis, University of Shahid Chamran,
Ahvaz, Lran.