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

net/publication/232423495

Methods for the seismic analysis of transverse section of circular tunnels in

soft ground

Conference Paper · January 2007

CITATIONS READS

4 102

5 authors, including:

Emilio Bilotta Giovanni Lanzano

University of Naples Federico II National Institute of Geophysics and Volcanology
90 PUBLICATIONS   448 CITATIONS    91 PUBLICATIONS   532 CITATIONS   

SEE PROFILE SEE PROFILE

G. Russo Filippo Santucci de Magistris

University of Naples Federico II Università degli Studi del Molise
88 PUBLICATIONS   716 CITATIONS    105 PUBLICATIONS   731 CITATIONS   

SEE PROFILE SEE PROFILE

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

Field Instrumentation and Monitoring of Embedded Retaining Wall View project

case histories View project

All content following this page was uploaded by Giovanni Lanzano on 23 January 2014.

The user has requested enhancement of the downloaded file.

Methods for the seismic analysis of transverse section of circular tunnels
in soft ground
Méthodes pour l'analyse séismique de la section transversale des tunnels circulaires en terre

Emilio Bilotta, Giovanni Lanzano, Gianpiero Russo

DIG, University of Napoli Federico II bilotta@unina.it, g.lanzano@unina.it, pierusso@unina.it .
Filippo Santucci de Magistris
SAVA, Engineering & Environmental Div., University of Molise, filippo.santucci@unimol.it
Francesco Silvestri
DDS, University of Calabria, f.silvestri@unical.it

ABSTRACT
The load increments induced by an earthquake in a tunnel lining can be ascribed to two different motion com-
ponents: the ovalisation of the transversal section, induced by soil shear straining in the vertical plane, and the
asynchronous movement of the ground-lining system along the longitudinal axis of the tunnel.
Forces induced by earthquakes in a tunnel lining can be assessed with several procedures at different levels of
complexity. In this paper, the different levels of analysis are developed on idealised geometry and soil condi-
tions, considered representative of soil classes specified by EC8 and the new draft Italian seismic code.
Equivalent linear analyses along the transversal direction were carried out following two different design ap-
proaches, corresponding to increasing levels of complexity of analytical models, soil characterisation and de-
scription of seismic input:
- pseudo-static analysis, where the seismic input is reduced to an equivalent inertia force or a peak strain am-
plitude, computed through a free-field pseudo-static analysis of the ground and then considered acting on the
tunnel lining in static conditions as well;
- full dynamic analysis, where the soil and tunnel responses are mechanically coupled and analysed via nu-
merical modelling, such as finite element methods.
The sample problem is a circular tunnel of 6 m diameter, with the axis at a depth of 15 m in a 30 m thick layer
of medium dense gravel, sand or soft clay, overlying a relatively stiff bedrock.
The force increments as calculated with the different procedures in the transverse section have been com-
pared, showing a fair approximation to the full dynamic analysis achieved with the pseudo-static approach.

RÉSUMÉ
Les incréments de charge induits par un tremblement de terre dans un revêtement de tunnel peuvent être attri-
bués à deux composants différents de mouvement: l'ovalisation de la section transversale, induite par le cisail-
lement du sol dans le plan vertical, et le mouvement asynchrone du système de sol-revêtement le long de l'axe
longitudinal du tunnel. Ces forces peuvent être évaluées avec plusieurs procédures à différents niveaux de
complexité. En cet article, les différents niveaux d'analyse sont développés sur une géométrie idéalisée et
conditions de sous-sol considérées représentant des classes de sol indiquées par EC8 et le nouveau code séis-
mique italienne. Des analyses linéaires équivalentes le long de la direction transversale ont été effectuées
après deux approches de conception différentes, correspondant à différents niveaux de complexité des mo-
dèles analytiques, de la caractérisation du sol et de la description du signal séismique:
- analyse pseudo-statique, où le signal séismique est réduite à une force équivalente d'inertie ou à une ampli-
tude maximale de contrainte, calculée par une analyse pseudo-statique du mouvement séismique à champ
libre et puis considérée agir sur le revêtement du tunnel en conditions statiques.
- analyse dynamique complète, où les réponses du sol et du tunnel sont mécaniquement couplées et analysées
par modélisation numérique, comme les méthodes d'éléments finis.
Le problème analysé est d’un tunnel circulaire de diamètre de 6 m, avec l'axe à une profondeur de 15 m dans
une couche épaisse 30 m de gravier moyen dense, de sable ou d'argile molle, recouvrant une roche en place
relativement raide. Les incréments de force calculés avec les différentes procédures dans la section transver-
sale ont été comparés, montrant que l'approche pseudo-statique réalise une juste approximation avec l’analyse
dynamique.

Keywords: underground structures, tunnels, seismic loads, pseudo-static analysis, dynamic analysis
1 INTRODUCTION ships, linking the shear modulus (G0) and the damp-
ing ratio (D 0) to the lithostatic stress, the void ratio
It was recently understood that a sustainable devel- and intrinsic soil properties, such as particle size and
opment of urban areas requires an increasingly large plasticity index IP (Santucci de Magistris, 2005;
use of underground facilities, especially to ensure a d’Onofrio & Silvestri, 2001). Starting from G0(z), it
greater transportation demand, preserving the envi- is possible to obtain the shear wave velocity profile
ronmental quality. On the other hand, it was long- VS(z) as:
time understood that, in seismic areas, civil infra-
structures and lifelines should be designed to support G 0 (z )
VS ( z)  (1)
the extra loading produced by earthquakes. In spite 
of the above, no provisions are given in EC8 (EN where is the soil density. The variations of VS with
1998-1, 2003) on how to evaluate seismic loading on depth for each soil model are shown in Fig. 1, where
underground structures. On the other hand, some in- the dashed lines represent the value of the so called
dication can be found in Owen & Scholl (1981), ‘equivalent velocity’ VS,30 (EN 1998-1, 2003).Table
JSCE (1992), AFPS/AFTES Guidelines (2001), ISO 1 summarizes the geotechnical parameters and the
TC 98 (2003). To bridge this gap some research ac- ground type according to EC8
tivities are in progress in Italy, aimed at developing Table 1: Ground parameters and classification according to
reliable methods for the analysis of the behaviour of EC8
tunnels under seismic actions (e.g. Bilotta et al., Ground type ’ IP  D0 VS,30
2007). This paper describes the most recent results
achieved in this research project. (°) (%) (kN/m 3) (%) (m/s)
During an earthquake, ovaling or racking defor- Clay D 25 30 18 2.5 124
mations of a transverse section of a tunnel are
mostly due to shear waves propagating perpendicu- Sand C 35 - 20 1.0 239
larly to the tunnel axis, resulting in a distortion of Gravel B 44 - 21 1.0 401
the cross-section of the structure. In this paper,
forces induced by ground shaking in the tunnel lin-
ing are evaluated using several calculation proce- To account for soil non-linearity and cyclic energy
dures at different levels of complexity, which fall dissipation, in the computational models it is also
into the class of the pseudo-static and dynamic necessary to introduce the variation of shear
methods. modulus G and damping ratio D with the shear strain
The analytical simulations have been performed level . Therefore, the curves G( )/G0 and D(
) for
on three idealized ground conditions (Fig. 1): a 30 m the three materials (Figure 2) have been assumed ac-
thick layer of soft clay, medium dense sand or cording to literature indications:
gravel, overlying a soft rock half-space as a bedrock - for clay and sand, the curves suggested by
(Vr= 800 m/s,  =22 kN/m 3, D0=0.5%). The tunnel Vucetic & Dobry (1991) for IP =30% and 0%;
has the following characteristics: - for gravel, the relationships reported by Sto-
- circular shape with reinforced concrete lining koe (2004) for D50 = 10mm.
(variable thickness, diameter D=6 m);
- axis depth z0=15 m, length 1000 m. 1 25

VS (m/s) 0.8 20
0 200 400 600
norm al ised shear st iff ness, G/Go

damping ratio, D (%)

0 clay stiffness
0.6 15
sand stiffness
gravel stiffness
5 clay damping
12 m 0.4 10
sand damping
gravel damping
10
t=0.1
t = 0.31.3
mm
z(m)

0.2 5
argilla
clay (D)(D)
6m 15 sabbia
sand (C)(C)
ghiaia (B)
gravel (B)
0 0
0.0001 0.001 0.01 0.1 1
(riv. Cls Rbk 450 20
concrete lining Rckkg/cmq)
=45 MPa shear strain, g (%)

12 m Figure 2. Variation of shear modulus and damping with shear

25
strain level
30
Bedrock==soft
bedrock roccia tenera
rock V S = 800 m/s 
= 22 kN/m 3 D0 = 0.5%

Figure 1. Ground conditions 2 METHODOLOGY

The values of small strain soil parameters have been In a full dynamic analysis, the force increments in
chosen by means of literature empirical relation- the lining due to an earthquake are directly obtained
from the simulation of the shaking of the coupled
ground-tunnel system, using a numerical method. results of the series of SSR analyses, at least for the
On the other hand, in simplified methods the kine- subsoil profiles considered.
matic soil-structure interaction is neglected and free- All the pseudo-static methods considered require
field displacements are applied to the tunnel bound- a preliminary evaluation of the peak acceleration at
ary (e.g., Hashash et al., 2001). Moreover, the ef- surface; its value has been computed as:
fects of compression waves are also neglected, as a max,s S 
ag (4)
only shear waves propagating in vertical planes
might induce shear strain  . Thereafter, distortions where a g is the peak acceleration on outcropping
are used to calculate seismic force increments in the rock site and S the site response factor. Its value has
tunnel lining by means of closed-form elastic solu- been either assumed constant, as specified by EC8
tions, such as those by Wang (1993) or Penzien and its subsequent proposals of updating (i.e. Italian
(2000). OPCM 3274, 2003; ETC12, 2006) or varying with
the ground motion amplitude, as proposed by Au-
silio et al. (2007). In the latter case, a non-linear re-
2.1 Pseudo-static analysis
sponse factor (FRN) was assumed, for each soil
Bilotta et al. (2007) discussed the results of four dif- class, as a negative power law of a g as reported in
ferent methods developed in the framework of a Table 2.
pseudo-static analysis, to evaluate the maximum
shear stress  max. They were all based on the equilib-
Table 2: Response factors in pseudo-static methods
rium of a deformable soil column from the surface to Soil EC OPC ETC1 FRN
a given depth z. 8 M 2 (Ausilio et al., 2007)
A first class of methods (namely, method 1 and 3274
method 2 in the above referenced paper) needed to
specify a vertical profile of peak acceleration amax Clay 1.35 1.35 1.1 0.539( a g /g)-0.4171
and the maximum shear stress  max was computed by -0.2362
Sand 1.15 1.25 1.15 1.0624( a g /g)
integration as:
Gravel 1.2 1.25 1.3 1.0177( a g /g)-0.2017
z
max ( z) 
amax ( z) dz (2)
0
The maximum shear strain at a depth z is therefore
Another class of methods (i.e. method 3 and method
calculated from the maximum shear stress,  max(z),
4) followed an approach similar to the simplified
according to the Ramberg & Osgood (1943) model:
procedures used in the evaluation of liquefaction
susceptibility by reducing with depth the maximum τ (z) 
R
τ (z)
value of surface acceleration amax,s. Hence the shear γmax (z)  max C max  (5)
stress distribution with depth was calculated accord- G0  G0 
ing to the following equation of dynamic equilib- where the parameters C and R have been calibrated
rium of a soil column: on the curves of Fig. 2 (Valentino, 2006).
a max, s
max (z ) rd (z) v (z ) (3) 2.2 Dynamic analysis
g
A set of input acceleration time histories has been
In Eq. (3), v is the total vertical stress, and rd is a selected from a database of records of Italian seismic
reduction parameter which takes into account the de- events (Scasserra et al., 2006). All the signals have
formability of the soil column. This latter can be been scaled to a value of ag equal to 0.35g and ap-
calculated as a function of depth z, for instance plied at the base of the subsoil models.
through the relationships given by Iwasaki et al. The finite elements (FE) software Plaxis v8
(1978) and Liao & Whitman (1986), or including the (Brinkgreve, 2002) has been used to perform two-
effects of magnitude, as suggested by Idriss & Bou- dimensional free-field and soil-structure interaction
langer (2004). Further values of rd(z) can be derived dynamic analyses. The code allows defining the
by the reduction factors of a max,s with depth given by damping tensor [C] through the Rayleigh formula-
Power et al. (1996). tion, i.e. as a linear combination of the mass tensor
Seismic site response (SSR) analyses were also [M] and the stiffness tensor [K]:
performed by Bilotta et al. (2007) with different in-
put signals to evaluate an average function of the re- [C] R [M] R [K] (6)
duction factor with depth. The results showed that
method 3, which used a reduction factor Coefficients R and R have been calculated accord-
rd 10.015 z (m) according to Iwasaki (1978), ing to the double frequency method, assuming that
provided a fairly good agreement with the average the damping ratio is about constant between the first
natural frequency of the deposit and a frequency n
times larger; n is the first odd integer which ap-
 proximates by excess the ratio between the funda- most overlapped. This is relatively true for sand al-
mental frequency of the seismic signal and the first so: for soil class C, EC8 and ETC12 specify the
natural frequency of the deposit. This method avoids same site response factor, smaller than OPCM 3274;
the overestimate of damping throughout the consid- for a g =0.35g, the non-linear response factor is larg-
ered frequency range (Lanzo et al., 2004). The bed- er (1.36). In the case of clay, OPCM 3274 and EC8
rock has been assumed as a rigid boundary, whereas specify the same value, whereas ECT12 proposed a
lateral mesh boundaries, about 40 D aside from the lower value and the FRN is even lower than unity
tunnel, are modelled as dampers according to the (0.83) for stronger ground motions.
Lysmer & Kuhlemeyer (1969) formulation (Fig. 3). In all, the computed profiles of shear strain appear
The FE analyses were performed with a linear more disperse and by using the non-linear response
elastic model for the soil. The dependency of the soil factor all the computed shear strains are at the lower
stiffness and damping ratio on the strain level has bound of the cluster of the profiles computed with
been first considered by equivalent linear analysis. the same method. This last result should not to be in-
Therefore, preliminary one-dimensional SSR analy- tended as general: the non-linear response factor is
ses have been performed by means of the code in fact higher than S for weaker motions (see Ausilio
EERA (Bardet et al., 2000), which operates in the et al., 2007).
frequency domain. Ground conditions and soil be- The shear strains calculated at the tunnel depth
haviour have been modelled according to Figs. 1 and with the four pseudo-static methods,  PS, have been
2. The material properties calculated as output from averaged for each soil and plotted in Fig. 5 against
the SSR analysis were hence used as input to the FE the corresponding values,  dyn, calculated by Plaxis
analyses. for the same soil profiles in free-field conditions.
The lines in the figure have been obtained by adding
and subtracting the standard deviation to the mean
values, plotted as markers. The solid black line
represent the condition  PS = dyn. It can be noted that
only method 4 underestimates the shear strain com-
puted by Plaxis. Furthermore, the scatter of data and
Figure 3. Sketch of the mesh used for FE Plaxis analyses the deviation of pseudo-static predictions from the
results of dynamic analyses increase with soil de-
formability. On the whole, both for gravel and sand,
3 RESULTS method 3 and 4 give fairly consistent results, whe-
reas for clay, the best prediction is by method 4.
Figure 4 shows the profiles of shear strain for the
different soil conditions, as calculated by means of 10.0% PS = dyn
the four pseudo-static methods (1 to 4) and the site method 1
response factors listed in Table 2 . The darker lines method 2
refer to calculations performed with non-linear re- method 3
sponse factors (FRN) whereas the thin grey lines re- method 4
PS

1.0%
method 1
fer to constant site factors. gravel method 2
sand
   clay method 3
0% 1% 2% 0% 2% 4% 0% 2% 4% 6%
method 4
0 0 0
Gravel (B) Sand (C) Clay (D) 0.1%
5
5 5 0.1% 1.0% 10.0%
10 10 10

dyn
depth (m)
depth (m)

depth (m)

15 15 15
Figure 5. Free-field shear strains
20 20 20

25 25 25
In pseudo-static analyses, it is commonly as-
30
4 3 21
30
4 3 21
30
43 1 2
sumed that the internal forces in the lining can be
Figure 4. Shear strain with depth according to all the pseudo- calculated from the free-field average shear strain,
static calculations performed i.e. neglecting the kinematic interaction between the
lining and the ground. The analytical expressions by
The methods based on the integration of the a,max(z)
Wang (1993) have been used, which refer to a tunnel
profile (i.e. 1 and 2) give very close results and tend
with diameter D, lining thickness t, and elastic pa-
to overestimate the shear strain with respect to those
based on the reduction of a,max,s (i.e. 3 and 4). In the rameters El and l . The tunnel is surrounded by a
case of gravel (soil type B) the different site factors homogeneous and isotropic half-space, with linear
have similar values (being FRN=1.26), therefore the elastic parameters Gm and m.
profiles corresponding to the same method are al-
 Under the hypothesis of rough interface between 300 PS = dyn
t=0.1 m
method 1
the lining and the soil, the variation of hoop (N) and method 2
t=0.3 m
t=0.7 m
bending moment (M) with the angle θis given by: method 3 t=1.3 m

Nmax, PS/PS (MN/m)

method 4
1   200 method 1
N 
  K 2 Gm DPS cos 2
  (7a) method 2
2  4 method 3
method 4
1  
M 
 K 1GmD 2 PS cos 2
  (7b) 100
12  4
where:
12
1 m  (8a)
0
K1  0 100 200 300
2 F 5 6m
(a) Nmax,dyn /dyn,ff (MN/m)
1
F12m 
12m C 12m  2
2

2 (8b)
K2 1
3 2m 12m CC
F
5 

PS ( MNm/m)
 8m 6m2 6 8m t=0.1 m
2  100 t=0.3 m
t=0.7 m
t=1.3 m
The dimensionless parameters: PS = dyn

F m

G 1  D l
2
 3

(9)
method 1
method 2

M max, PS/
2 El t 3
10 method 3
method 4

C m
 
G 1 l2 D
(10)
method 1
method 2
El t
1 2m  method 3
method 4
1
represent the relative soil/tunnel stiffness, i.e. are in- 1 10 100
versely proportional to the influence of soil–
structure interaction. (b) Mmax,dyn / dyn,ff (MNm/m)
The maximum lining internal forces (i.e. at =/4) Figure 6. Comparison between dynamic and pseudo-static
calculated by uncoupled soil–tunnel analyses can be analyses in terms of hoop (a) and bending moment (b)
normalised dividing eqns. 7a,b by  PS in the follow- The normalised internal forces increase with the lin-
ing way: ing thickness, indicating the increasing effect of soil-
1 structure interaction on the seismic load increment in
N max, PS / PS  K 2 Gm D (11a)
2 the tunnel. Accordingly, the maximum bending
moment and hoop computed by full dynamic analy-
1 sis result increasingly higher than those obtained by
M max, PS / PS  K1Gm D2 (11b)
12 the uncoupled pseudo-static approach, although the
difference appears acceptable for flexible lining (up
In a similar way, the lining internal forces calcu-
lated in FE analyses, accounting for soil-tunnel in- to about 30 cm). Among the pseudo-static proce-
dures, method 4 again provides a better agreement
teraction, can be divided by the shear strain calcu-
with the FE analysis.
lated in dynamic free-field analyses, obtaining
N max,dyn / dyn, ff and M max,dyn / dyn , ff . In Figs. 7a,b, the ratios between the maximum dy-
namic and pseudo-static bending moments and
In Figs. 6a-b the average values (markers) and hoops computed for sand have been plotted against
bands of deviation (lines) of normalised internal the lining thickness, t.
forces obtained for the case of sandy subsoil from
the pseudo-static methods are plotted against those
calculated from the FE analyses. The different data
colours refer to a lining thickness increasing from
0.1 to 1.3 m.
 1 method 1 The variable accounts for the variation of strains
NmaxPS /NmaxDYN
method 2 (a)
method 3 calculated according to different pseudo-static meth-
0.8
method 4 ods. In Table 3 the values of are shown, as com-
0.6 puted by eq. (12) on the average shear strain of each
method.
0.4
Table 3: Average values of ratio 
0.2
Gravel/Sand Clay
0 method 1 2.5 5
method 2 2.3 4.7
0 0.5 1 1.5
method 3 1.4 3
t (m) method 4 0.4 0.8
3 method 1
method 2 (b) The values of for sand and gravel are the same.
M maxPS /MmaxDYN

2.5 method 3
method 4 Different is the case of clay, for which the average
2
values are about twice as larger.
1.5 The ratios k M and kN are different for different
1 pseudo-static methods, as they depend on the values
of Gm (i.e. on the shear strain level). In Figs. 8 and 9
0.5 the values k *N =kN/Gm,PS and k *M =kM/Gm,PS are plot-
0 ted against the thickness t: note that k N* seems sub-
0 0.5 1 1.5 stantially independent on the choice of pseudo-static
t (m) method.
Figure 7. Ratio between dynamic and pseudo-static maximum
hoop (a) and bending moment (b) vs lining thickness 0.03 method 1
method 2
Note that, whatever the pseudo-static method 0.025 method 3
method 4
adopted, beyond a threshold lining thickness such 0.02
ratios does not depend on the value of t. Moreover,
kN*

the maximum hoop computed by pseudo-static 0.015

analyses, NmaxPS, always underestimates the corre-
0.01
sponding maximum dynamic hoop, NmaxDYN.
0.005

4 FINAL REMARKS 0
0 0.5 1 1.5
To summarize, the following dimensionless parame-
ters can be defined: t (m)
Figure 8. Ratio k*N =kN /Gm,PS vs lining thickness

 PS (12)
DYN ,FF 0.1 method 1
method 2
N PS /  0.08 method 3
k N  DYNmax PS (13a) method 4
N max / DYN ,FF
0.06
/ PS
k M*

PS
M max
k M  DYN (13b)
M max / DYN ,FF 0.04

For the ratios plotted in Fig. 7 it follows: 0.02

PS
N max 0
DYN

kN (14a)
Nmax 0 0.5 1 1.5
PS
M t (m)
max
DYN

kM (14b)
M max Figure 9. Ratio k*M=kM/Gm,PS vs lining thickness
By means of any of the above mentioned pseudo- Brinkgreve R.B.J., Plaxis 2D version8. A.A. Balkema
static methods, the following expressions may be Publisher, Lisse, 2002
used to evaluate the maximum bending moments d’Onofrio A., and Silvestri F., “Influence of micro-
and hoops, taking into account the possible kine- structure on small-strain stiffness and damping of
matic interaction: fine grained soils and effects on local site response”,
Proc. IV Int. Conf. on ’Recent Advances in Geothec-
PS
N max nical Earthquake Engineering and Soil Dynamics’.
DYN
N max  (15a)

k *N  GmPS San Diego, Paper 1.19, 2001
PS
EN 1998-1, “Eurocode 8: Design of structure for earth-
M max quake resistance,. Part 1: General rules, seismic ac-
DYN
M max  (15b)

k M*  GmPS tions and rules for buildings”. CEN European Com-
mittee for Standardisation, Bruxelles, Belgium, 2003
PS PS PS PS
where N max / Gm and M max / Gm are computed by Idriss I.M., and Boulanger R., “Semi-empirical proce-
means of one of the pseudo-static methods (1 to 4) dures for evaluating liquefaction potential during
according to a straightforward rearrangement of earthquakes”, Proc. V ICSDEE & III ICEGE, Berke-
ley, USA (1) 32 -56, 2004
(11a-b); follows for each method from Table 3;
* * ISO TC 98/SC 3 N229, “Bases for design of structures –
k N and k M can be obtained, for sand, from the Seismic actions for designing geotechnical works”,
charts in Figs. 10 and 11, for a given thickness t. 2003
Hashash, Y.M.A., Hook, J.J., Schmidt, B., and Yao, J.I-
C., “Seismic design and analysis of underground
structures”, Tunnelling and Underground Space
5 ACKNOWLEDGMENTS Technology, 16, 247-293, 2001Iwasaki T., Tatsuoka
F., Tokida K., and Yasuda S. “A practical method for
This work is a part of a Research Project funded by assessing soil liquefaction potential based on case
ReLUIS (Italian University Network of Seismic En- studies at various sites in Japan”, Proc. II Int. Conf.
gineering Laboratories) Consortium. The Authors on Microzonation, San Francisco, 1978
wish to thank the coordinator, prof. Stefano Aversa, Lanzo G., Pagliaroli A., D’Elia B., “Influenza della mo-
for his continuous support and the fruitful discus- dellazione di Rayleigh dello smorzamento viscoso
sions. The strong motion database used in this study nelle analisi di risposta sismica locale”, Proc. XI Ital-
was developed as part of an ongoing joint project in- ian National Conference on “L’ingegneria Sismica in
volving researchers from the University of Rome La Italia”, 2004 (in Italian)
Sapienza and the University of California, Los An- Liao S.S.C., and Whitman R.V., “Overburden correction
geles, with support from the Pacific Earthquake En- factors for SPT in sand”, Journal of Geotechnical
gineering Research Center. Preliminary results from Engineering, ASCE, 112(3):373-377, 1986
Lysmer J. and Kuhlemeyer R.L, “Finite dynamic model
this group were reported by Scasserra et al. (2006),
for infinite media”, J. of Eng. Mech. Div., ASCE:
but the data utilized here have not been published. 859-877, 1969
OPCM 3274, “Primi elementi in materia di criteri genera-
li per la classificazione sismica del territorio naziona-
le e di normative tecniche per le costruzioni in zona
6 REFERENCES sismica”, GU Repubblica Italiana, 105-8/5/03, 2003
Owen G.N., and Scholl R.E., “Earthquake engineering of
AFPS/AFTES, Guidelines on earthquake design and pro- large underground structures”, Report no. FHWA/RD-
tection of underground structures, 2001 80/195. Federal Highway Administration and National
Ausilio E., Silvestri F., Troncone A., Tropeano G., Science Foundation, 1981
“Seismic displacement analysis of homogeneous Penzien J., “Seismically induced racking of tunnel lin-
ings”, Int. J. Earthquake Eng. Struct. Dynamics 29,
slopes: a review of existing simplified methods with
reference to Italian seismicity”, IV ICEGE, Thessa- pp. 683–691, 2000
loniki, 2007 Power M.S., Rosidi D., and Kaneshiro J., Vol.III
“Strawman: screening, evaluation and retrofit design
Bardet J. P., Ichii K., and Lin C. H., “EERA a Computer
Program for Equivalent-linear Earthquake site Re- of tunnels”, Report Draft, Nat. Center for Earthquake
sponse Analyses of Layered Soil Deposits”, Univ. of Engineering Research, Buffalo, New York. 1996
Ramberg W., and Osgood W.R., “Description of
Southern California, Dep. of Civil Eng., 2000
Bilotta E., Lanzano G., Russo G., Santucci de Magistris stress-strain curves by three parameters”, Technical
F., Aiello V., Conte E., Silvestri F., Valentino M., Note 902, National Advisory Committee for Aero-
nautics, Washington, D.C. 1943
“Pseudo-static and dynamic analyses of tunnels in
transversal and longitudinal directions”, IV ICEGE, Santucci de Magistris F., “Fattori di influenza sul com-
Thessaloniki, 2007 portamento meccanico dei terreni”, App. B in 'As-
 petti geotecnici della progettazione in zona sismica -
Linee Guida AGI', Associazione Geotecnica Italiana,
Patron, Bologna, 2005 (in Italian)
Scasserra G., Lanzo G., Mollaioli F., Stewart J.P.,
Bazzurro P., and Decanini L.D., “Preliminary com-
parison of ground motions from earthquakes in Italy
with ground motion prediction equations for active
tectonic regions”, Proc. 8th U.S. National Conference
on Earthquake Engineering, San Francisco, 2006
Stokoe K.H., “Comparison of Linear and Nonlinear Dy-
namic Properties of Gravel, Sand, Silts and Clays”,
Proc. V ICSDEE & III ICEGE, Berkeley, USA, 2004
JSCE - The Japanese Society of Civil Engineers, “Earth-
quake resistant design for civil engineering structures
in Japan”, 1992
Valentino M., “Metodi dinamici e pseudostatici per
l’analisi sismica di infrastrutture interrate”, M.Sc.
Thesis, University of Calabria, 2006 (in Italian)
Vucetic M., and Dobry R., “Effects of the soil plasticity
on cyclic response”, Journal of Geotechnical Engi-
neering, ASCE, 117(1):89-107, 1991
Wang J., “Seismic Design of Tunnels: A Simple State-of-
the-art Design Approach”, Monograph 7, Parsons,
Brinckerhoff, Quade and Douglas Inc, New York,
1993

View publication stats