1

EARTHQUAKE DESIGN OF RECTANGULAR

UNDERGROUND STRUCTURES

J.H. Wood 1

ABSTRACT

Severe damage to six out of a total of 21 subway stations in the Kobe area during the 1995 Hyogoken-nanbu
earthquake indicated a need for more attention to be given to the earthquake design of rectangular underground
structures. This paper presents work undertaken to extend the present knowledge of the dynamic interaction of box-
section structures with the surrounding soil, and a design method for predicting the earthquake loads on underground
structures such as basement walls, tanks, subways, utility boxes, highway underpasses, and culverts.

1 INTRODUCTION behaviour of a long structure subjected to a component of

parallel wave deformation is similar to that of an elastic
Seismic design methods for underground structures are not beam embedded in the soil. In simplified analyses, the
adequately covered in New Zealand design codes and structure is assumed to be flexible relative to the
guidelines. For example, the current Transit New Zealand surrounding soil or rock, and to respond with the same
Bridge Manual identifies the problem of the seismic deformation pattern as in the free-field elastic seismic
design of underground structures for culverts and subways waves. These simplified analyses are often employed for
with moderate depths of cover, but suggests using an out- pipelines that have relatively small cross-sectional areas
of-date analysis procedure known as the stresses at and for the preliminary analyses of tunnels. When the
infinity method. structure is stiff in the longitudinal direction relative to the
Many small underground structures exist in urban areas in surrounding soil, it will not be compliant with the soil or
New Zealand. For example there are about 13 rock deformations. For this case, interaction effects need
underground subways, box culverts and highway to be considered by employing either numerical methods
underpasses in the Lower Hutt City area located in or approximate solutions developed from wave
sediments close to the Wellington Fault. None of these propagation theory for beams on an elastic foundation
have been specifically designed for earthquake effects. (Wang, 1993).
Large underground car-parking buildings and buildings Ovaling or racking deformations develop in an
with significant underground basement structures have underground structure when the seismic waves propagate
been constructed in major New Zealand cities. Earthquake in a direction perpendicular to, or with a significant
loads have usually been considered in their design, but the component perpendicular to, the longitudinal axis,
analyses have been based on simplified theory developed resulting in distortion of the cross-section (see Figure 1b).
for retaining walls with the wall top at the ground surface For this case, and if the structure is relatively long, a
and the foundation assumed to be rigid. When the plane-strain two-dimensional analysis may be employed.
structure is constructed below the surface in deep soil
layers, these simplified methods have obvious limitations.
In the past, earthquake analysis of underground structures
Underground structures are constrained by the surrounding focused on circular cavities in elastic continua and then
soil or rock and cannot move independently so are not advanced to consider lined cavities located within elastic
generally subjected to significant dynamic amplification half-spaces. These shapes were a reasonable
effects. They are affected by the deformation of the approximation to typical tunnel cross-sections and could
surrounding ground and not by the inertia forces acting on be analysed by analytical theory of elasticity and
the structure. In contrast, above-ground structures often approximate elasticity methods. With the rapid advance
have natural frequencies that are within the range of the of numerical methods in the last decade, satisfactory
predominant frequencies of earthquake ground motions, analysis procedures have been developed for rectangular
resulting in resonant effects with the accelerations acting structures. Prior to this, rectangular cross-sections were
on the structure amplified with respect to the ground designed by assuming that they were subjected to the free-
surface. field soil or rock shear strains and by the application of
The earthquake response of underground structures is standard structural frame analysis methods. Rectangular
usually considered with reference to the three principal underground structures are suitable for cut-and-cover
types of deformations: axial, curvature and racking construction and are therefore often constructed close to
(rectangular cross-sections), or ovaling (circular cross- the surface in relatively soft soils. In this situation, the
sections). Axial and curvature deformations develop when assumption of compatibility with the free-field shear
seismic waves propagate either parallel or obliquely to the strains may in some cases lead to very conservative
longitudinal axis of the structure (Figure 1a). The general designs.

1
John Wood Consulting, Lower Hutt (Fellow).

BULLETIN OF THE NEW ZEALAND SOCIETY FOR EARTHQUAKE ENGINEERING, Vol. 40, No. 1, March 2007
2

Figure 1a. Axial and curvature deformations. Figure 1b. Ovaling and racking deformations.
(Source Owen and Scholl, 1981) (Source Yang, 1993)

2 ANALYSIS METHOD FOR RACKING shear waves in a layer of uniform thickness. These
solutions provide sufficiently good approximations for the
The evaluation of the racking performance of the cross- design of smaller underground structures, particularly
section of a rectangular underground structure subjected to where the site soil properties are not know in any great
earthquake ground motions can be undertaken using the detail. They are also helpful in providing quick solutions
following steps: for preliminary analysis work, and provide a useful
(a) Evaluation of the free-field peak shear strain in verification method for more sophisticated numerical
the soil at the average depth of the structure. analysis.
(b) Evaluation of the elastic and post-elastic In the present study, analytical solutions were evaluated
stiffness of the structure. for a uniform elastic shear layer of infinite horizontal
(c) Evaluation of the racking deformation of the extent overlying rock assumed to form a rigid lower
structure from the free-field strain, structure boundary. Solutions were evaluated for the following
stiffness and soil-structure interaction curves. depth profiles of the elastic shear modulus, G.
(d) Evaluation of the member forces in the structure (a) Uniform.
from the racking deformation. (b) Parabolic with the surface value Gt = 0.
2.1 Free-Field Strain (c) Linear variations with G increasing with depth.
G defined by G(y) = Gb(1 q y/H),
The shear distortion of the ground from vertically where Gb is the shear modulus at the
propagating shear waves is usually considered to be the base of the layer, y is the height
most critical and predominant effect producing racking above the base, H is the thickness of the layer
type deformations in underground rectangular structures. and q is a parameter varying between 0 and 1.
Numerical methods have often been applied to estimate
the free-field shear distortions, particularly in sites with
variable stratigraphy. Computer codes such as SHAKE Plots of G versus depth for the layers investigated are
based on one-dimensional wave propagation theory for shown in Figure 2. Layers included uniform, parabolic,
equivalent linear systems have been developed to carry out and linear variations with q = 0.75, 0.95, 0.9 and 0.95.
these types of analyses. Analytical solutions by Wood (1973) were used to
evaluate first horizontal mode displacement and shear
An alternative to undertaking numerical analysis to predict strain responses for H = 50 m, Gave = 100 MPa, and
free-field shear deformations is to evaluate theory of uniform soil density = 2.0 t/m2. First mode displacement
elasticity analytical solutions for vertically propagating and shear strain responses calculated for a one-g static

2
 3

base acceleration are shown in Figures 3 and 4. First mode may occur leading to plastic hinge formation in the walls,
periods of vibration for the analysed layers are given in roof or floor members. Progressive development of the
Table 1. The response to a dynamic base acceleration hinges can be investigated by a pushover analysis using
input is obtained by scaling the one-g static responses by nonlinear frame analysis software.
Sa(T1)/g where Sa(T1) is the spectral acceleration of the
input acceleration response spectrum at T1 , the first
horizontal mode period of the layer. 2.3 Soil-Structure Interaction

Because of both soil-structure interaction and dynamic

inertial effects, the soil shear strains in the vicinity of the
Table 1. First mode periods structure are generally significantly different to the free-
Layer Period, field shear strain at the corresponding depth in the soil
Description T1 sec layer. Often the mass change at the cavity created by the
structure is small in relation to the total mass in the layer
Uniform 0.894 acting in unison with the structure during dynamic
loading, and previous research has indicated that dynamic
Parabolic 0.811 inertial effects are sufficiently small to be neglected.
Linear, q = 1.0 0.826 In contrast to the influence of soil inertial effects, soil-
Linear, q = 0.95 0.819 structure interaction effects may produce significant
changes in the shear strains near the structure. If a cavity
Linear, q = 0.90 0.818 in the soil is unlined, then the shear strains in the soil near
the cavity would clearly be greater than the free-field
Linear, q = 0.75 0.826 shear strain. If a stiff structure is inserted in the soil cavity,
then the shear strains may be less than the free-field.
Conversely, with a very flexible structure, the shear strains
2.2 Structure Stiffness and Flexibility may be greater than in the free-field.

Most underground structures of rectangular shape are In assessing soil-structure interaction effects on
designed to act as rigid-jointed box structures. The underground structures it is usual to define shear strain
simplest example of this type of structure is the single- deformation and flexibility ratios. The shear strain
barrel box shown in Figure 5. To assess the racking deformation ratio, R, is defined by:
stiffness or flexibility, the structure is loaded with a
horizontal load P at the roof level to produce a racking or Shear deformation of structure
shear deformation of . The structure is assumed to be of embedded in soil (including interaction) sti
R= = (5)
sufficient extent normal to the plane of the section for Free-field shear deformation ff
plane-strain conditions to exist. Simple closed-form over height of structure
solutions for the racking stiffness and flexibility of this
single-barrel structure have been published previously
The flexibility ratio, Fr is defined by:
(Shepherd & Wood, 1966). The elastic flexibility can be
written as: Shear flexibility of free standing
f st
F =
R
structure without soil interaction
= (6)
H 2
1 Shear flexibility of soil block of same fs
f st = = (1) overall dimensions as structure
P 24K w 2 + 3r + 3 jr
1 2 + 2r(2 + 2 j + 3 jr ) Soil shear stiffness is defined by:

Where:
Ks = G = (7)
Kr
EI w EI r r=
Kw = Kr = Where: G = soil shear modulus in the soil at the level
H L Kw under consideration, = soil shear stress and = soil shear
strain.
E is Youngs modulus for the structure material and Iw and From the soil shear stiffness definition it follows that the
Ir are the moments of inertia per unit length of the wall and shear flexibility of a block of soil of height H and length L
roof respectively. H, L and j are defined in Figure 5. is given by:
There are no simple closed-form solutions available for
multi-barrel structures, but reducing the multi-barrel H
structure to an equivalent single-barrel structure can fs = (8)
provide a good approximation (within 6%) for the stiffness LG
or flexibility. Each wall element of the equivalent single- The flexibility ratio Fr can be readily computed from the
barrel structure has one-half of the sum of the flexural soil shear modulus G and the structure flexibility
stiffness of all the wall elements in the multi-barrel coefficient fst. Methods for calculating the shear
structure, and the roof and floor elements of the equivalent deformation ratio R from the flexibility ratio have been
structure have the sum of the corresponding roof or floor investigated in several previous research projects (Wang,
elements in the multi-barrel structure. 1993; Penzien, 2000; Nishioka and Unjoh, 2003) and also
in the present study. The structure shear deformation
Under high racking deformations, yield in the members (including soil-structure interaction) sti can be readily
4

calculated from the shear deformation ratio R and

6Kw j(3 + q)
estimates of the peak free-field shear deformation ff Mf =
H j (2 j + q)(2 + q)
(9)
during the design earthquake event. The structure shear
deformation can be used to calculate the earthquake-
Where
induced forces in the members of the structure using
analytical or numerical methods. For example, the Kw
moment at the base joints of a single-barrel box are given q= (10)
by Wood, 2004 as: Kr
1.0 1.0
Uniform Variation
0.9 0.9
Parabolic Variation
0.8 Linear q = 1 0.8

Height Above Base, y/H

Height Above Base, y/H

0.7 Linear q = 0.95 0.7

Linear q = 0.9
0.6 Linear q = 0.75 0.6

0.5 0.5

0.4 0.4 G uniform

0.3 G parabolic variation
0.3
G linear q = 1
0.2 0.2 G linear: q = 0.95
0.1 0.1 G linear: q = 0.9
G linear: q = 0.75
0.0 0.0
0 50 100 150 200 0.00 0.05 0.10 0.15 0.20 0.25 0.30
Shear Modulus, MPa Displacement, m

Figure 2. Variation of layer shear modulus, G. Figure 3. First shear mode displacement response.

1.0

0.9

0.8 Kr
Height Above Base, y/H

P
0.7

0.6 G uniform
G parabolic
0.5
G linear: q = 1.0 Kw Kw H
0.4 G linear q = 0.95
G linear: q = 0.9
0.3
G linear: q = 0.75
0.2 j Kr
0.1
L
0.0
0.000 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008
Strain

Figure 4. First shear mode shear strain response. Figure 5. Single-barrel box frame.

5m 5m 5m

Thickness varied
5m

5m

Thickness varied from 200 to 800 mm

from 200 to 800 mm

Single-Barrel Structure Double-Barrel Structure

Figure 6a. Single-barrel box model Figure 6b. Double-barrel box model.
 5

In the present study, R versus Fr relationships were of 0.4, depth of cover of 5 m and a 50 m deep soil layer
evaluated for box structures by simplifying a dynamic with a shear modulus increasing parabolicaly from zero at
finite element analysis procedure originally used by Wang the surface are compared in Figure 7 with the results of
(1993). The dynamic analysis was simplified by Wang (1993), Penzien (2000), and Nishioka and Unjoh
undertaking a modal analysis and investigating the strain (2003), for similar assumptions. There is good general
field in the first mode of the soil layer. This simplification agreement between the R versus Fr curves produced by the
enabled a wide range of the important parameters to be various methods, particularly at Fr ratios less than 2.0.
investigated. By using a refined mesh, it was also possible
to reliably calculate the forces in the members of the The results in Figure 7 show that when the flexibility ratio
structure. The rigid-frame single-barrel and twin-barrel approaches zero, representing a perfectly-rigid structure,
box structures shown in Figure 6a and 6b were analysed the R value also reduces to zero. At Fr = 1, the structure
using a range of layer depth and shear modulus profiles. has the same stiffness as the soil, and the structure has a
Plane-strain conditions, and linearly-elastic material similar racking distortion to the ground in the free-field,
properties for soil and structure were assumed. resulting in an R value of approximately 1.0. When the
flexibility ratio is greater than 1.0, the structure is flexible
with respect to the ground, and the racking distortions
3. RESULTS OF ANALYSES become magnified with respect to the free-field shear
R versus Fr curves for the single and twin-barrel structures distortion.
analysed in the present study using a soil Poissons Ratio

2.5 2.5
Poissons Ratio Soil = 0.4
Strain Ratio, R (Interaction / Free-Field)
Poissons Ratio for Soil = 0.4
Strain Ratio, R (Interaction / Free-Field)

Uniform Layer

2.0 2

1.5 1.5

1.0 Nishioka & Unjoh (2003) 1 Top of structure at surface

Penzien (2000) 5 m Cover to Structure
Wang (1993) - Numerical Results 10 m Cover to Structure
0.5 Wang (1993) Theoretical for Circular Section 0.5 15 m Cover to Structure
Present Study Twin Barrel 20 m Cover to Structure
Present Study Single Barrel Structure on rock base
0.0 0
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Flexibility Ratio Fr (Structure/Soil) Flexibility Ratio, Fr (Structure / Soil)

Figure 7. Interaction curve comparison. Figure 8. Interaction for single barrel at various depths.

R versus Fr curves for the single-barrel structure in a structures without surrounding soil. Similar variations
uniform 50 m thick layer are shown in Figure 8 for a range with Fr were obtained for the moments and shears at the
of different cover depths. There is little variation between tops of the walls (Wood, 2004).
the curves for cover depths between 5 and 20 m. Surface-
layer effects cause moderate changes in the interaction for
From Figures 9 and 10 it is apparent that the interaction
shallow covers at high Fr ratios, but once the structure is
effects significantly modify the bending moments and
covered by more than its height of soil, the interaction
shear forces in the box members for large Fr ratios.
does not change significantly with increasing depth of the
structure within the layer.
The effects of the depth of the layer, strain profile, Poisson 4. BOX CULVERT ANALYSIS EXAMPLE
Ratio and the box shape on the interaction curves are Table 2 summarises calculations based on the above
presented by Wood (2004) in a more detailed report on the analysis procedure for a large single-barrel box culvert
present study. Provided the depth of cover is greater than with outside dimensions 4 x 4 m and uniform wall
the height of the structure, it was found that the variation thickness of 400 mm. The structure was assumed to be
in the shear modulus or strain field has little influence on covered by 5 m of soil in a 50 m deep soil layer with a
the interaction curves. Poissons ratio of 0.4 and shear modulus varying
The bending moments and shear forces at the bottom of parabolicaly from zero at the surface to 200 MPa at a rigid
the side-wall members for the single and twin- barrel box base. It was assumed that the cracked stiffness of the walls
structures with a 5 m depth of cover are shown in Figures was 50% of the elastic stiffness. A ground-surface design
9 and 10. Both the plotted interaction bending moments acceleration of 0.4 g was adopted.
and shear forces have been normalised by dividing the For the assumed soil profile, the structure has a flexibility
computed values by the corresponding values for the ratio Fr of 2.27, indicating that it is more flexible than the
6

2.2 6
Soil Poisson's Ratio = 0.4 Soil Poisson's Ratio = 0.4
Wall Base Moment Ratio (Soil / No Soil)

Wall Base Shear Ratio (Soil / No Soil)

Cover Depth = 5 m Cover Depth = 5 m
2.0
5
1.8
4
1.6

1.4 3

1.2
2
Twin-barrel box: uniform layer Twin-barrel box: uniform layer
1.0
Twin-barrel box: parabolic varation G with depth Twin-barrel box: parabolic varation G with depth
1
0.8 Single-barrel box: uniform layer Single-barrel box: uniform layer
Single-barrel box: parabolic variation G with depth Single-barrel box: parabolic variation G with depth
0.6 0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Flexibility Ratio Fr (Structure / Soil) Flexibility Ratio Fr (Structure / Soil)

Figure 9. Bending moments at base of side wall. Figure 10. Shear force at base of side wall.

replaced soil, with the induced strain greater than the free- 7. REFERENCES
field strain by a moderate margin (strain ratio R = 1.4).
The maximum earthquake-induced moment at the base of Nishioka, T. & Unjoh, S. (2003). A simplified evaluation
the walls was 133 kN m/m which is greater than the method for the seismic performance of underground
cracking moment of about 100 kN m/m, and of similar common utility boxes. Proceedings 2003 Pacific
order to gravity and water pressure induced moments Conference on Earthquake Engineering, NZSEE,
(Wood, 2004). Auckland.

A refined version of the above analysis procedure is Owen, G.N. & Scholl, R.E. (1981). Earthquake
necessary when the nonlinear response of the structure engineering of large underground structures. Report
becomes significant. In this case, it is necessary to carry prepared for the Federal Highway Administration,
out an inelastic pushover analysis of the structure (without FHWAIRD-801195.
soil interaction) to define the structure force/displacement Penzien, J. (2000). Seismically induced racking of tunnel
relationship. The analysis is then advanced by estimating a linings. Earthquake Engineering and Structural
ductility factor, calculating the equivalent flexibility Dynamics, Vol 29, 683-691.
factor, and undertaking a standard interaction analysis to
estimate the total shear deformation of the structure. From Shepherd, R. & Wood, J.H. (1966). Normal mode
this deformation and the yield displacement, an improved properties of multi-storey frameworks. Journal of
ductility factor can be calculated and the process repeated Sound and Vibration, Vol. 3, No. 3, 300-314.
until convergence (Wood, 2004).
Wang, J. (1993). Seismic design of tunnels. Monograph
7, Parsons Brinckerhoff Quade & Douglas Inc, New
5. CONCLUSION York, June 1993.

Large earthquake-induced forces in the culvert structure of Wood, J.H. (2004). Earthquake design procedures for
the above example indicate that earthquake racking effects rectangular underground structures. Project Report to
need to be considered in the design of rectangular Earthquake Commission, EQC Project No. 01/470, Rev
underground structures. The analysis method first B July 2004.
described by Yang (1993) and advanced in the present Wood, J.H. (1973). Earthquake-induced soil pressures on
study provides a suitable design approach for small structures. EERL 73-05, Earthquake Engineering
structures and preliminary design for large structures. Research Laboratory, California Institute of
Technology.

6. ACKNOWLEDGEMENT

Financial support from the Earthquake Commission

Research Foundation to undertake this project is gratefully
acknowledged.