Fragility-based seismic performance

assessment of modular underground

arch bridges Van-Toan Nguyen
Visit to download the full and correct content document:
https://ebookmass.com/product/fragility-based-seismic-performance-assessment-of-
modular-underground-arch-bridges-van-toan-nguyen/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Finite element analysis-aided seismic behavior

examination of modular underground arch bridge Toan Van
Nguyen

https://ebookmass.com/product/finite-element-analysis-aided-
seismic-behavior-examination-of-modular-underground-arch-bridge-
toan-van-nguyen/

Armstrong’s Handbook of Performance Management: An

Evidence Based Guide to Delivering High Performance
(Ebook PDF)

https://ebookmass.com/product/armstrongs-handbook-of-performance-
management-an-evidence-based-guide-to-delivering-high-
performance-ebook-pdf/

Prince of Chaos: A Chicago Underground Novel (The

Chicago Underground Book 1) Angie Cottingham

https://ebookmass.com/product/prince-of-chaos-a-chicago-
underground-novel-the-chicago-underground-book-1-angie-
cottingham/

SPT-based probabilistic and deterministic assessment of

seismic soil liquefaction triggering hazard K. Onder
Cetin & Raymond B. Seed & Robert E. Kayen & Robb E.S.
Moss & H. Tolga Bilge & Makbule Ilgac & Khaled
Chowdhury
https://ebookmass.com/product/spt-based-probabilistic-and-
deterministic-assessment-of-seismic-soil-liquefaction-triggering-
hazard-k-onder-cetin-raymond-b-seed-robert-e-kayen-robb-e-s-moss-
The Underground Library Jennifer Ryan

https://ebookmass.com/product/the-underground-library-jennifer-
ryan/

To Kiss a Dragon: Lords of Forbidden Fantasy Miranda

Bridges

https://ebookmass.com/product/to-kiss-a-dragon-lords-of-
forbidden-fantasy-miranda-bridges/

The AFS Textbook of Foregut Disease Ninh T. Nguyen

https://ebookmass.com/product/the-afs-textbook-of-foregut-
disease-ninh-t-nguyen/

Children’s Speech: An Evidence-Based Approach to

Assessment and Intervention

https://ebookmass.com/product/childrens-speech-an-evidence-based-
approach-to-assessment-and-intervention/

Time-frequency Analysis of Seismic Signals Yanghua Wang

https://ebookmass.com/product/time-frequency-analysis-of-seismic-
signals-yanghua-wang/
 Structures 39 (2022) 1218–1230

Contents lists available at ScienceDirect

Structures
journal homepage: www.elsevier.com/locate/structures

Fragility-based seismic performance assessment of modular underground

arch bridges
Van-Toan Nguyen a, Jin-Hee Ahn b, Achintya Haldar c, Jungwon Huh a, *
a
Department of Architecture and Civil Engineering, Chonnam National University, Gwangju 61186, South Korea
b
Department of Civil and Infrastructure Engineering, Gyeongsang National University, Jinju 52725, South Korea
c
Department of Civil Engineering and Engineering Mechanics, University of Arizona, Tucson, AZ 85721, USA

A R T I C L E I N F O A B S T R A C T

Keywords: Seismic performance assessment of modular underground arch bridges (MUABs) can help mitigate structural
Modular underground arch bridge damage to MUABs during earthquakes. This study proposed an efficient procedure to evaluate the seismic
Seismic performance performance of a three-hinge MUAB by using fragility-based analysis. Firstly, a novel proposal for damage states
Stratum-uncertainty characteristics
was presented using the relationship between the crack depth and thickness of the arch component. The rela­
Damage states
Fragility-based analysis
tionship between ground motion intensity and structural failure probability was then described using stratum-
uncertainty characteristics. The multiple stripe analysis approach with the maximum likelihood function was
introduced to detail the uncertainty characteristics of strata by allowing various ground motions at distinct
seismic intensity levels. Finally, the collapse probability considering stratum-uncertainty characteristics was
specified for seismic intensity and probability levels. The results revealed that the proposed procedure to
randomly select ground motions from the strong-motion database of the Pacific Earthquake Engineering
Research Center is feasible for evaluating the seismic vulnerability of MUABs. Therefore, the seismic perfor­
mance of MUABs can be reasonably predicted under earthquakes.

1. Introduction have been investigated for such structural constructions [29,37–40].

Globally, sophisticated construction projects both underground and
aboveground are increasingly being undertaken. Therefore, the seismic
performance assessment of multistory buildings, bridges, dams, and
underground structures is a critical research topic [1–19].
Underground structures have been widely used to precast segmental
components to incorporate salient technical and construction features
[20–22]. The precast arch bridge with modular components, namely
modular underground arch bridges (MUABs), is a typical example of
advanced construction [23]. Many cross-section shapes of the arch and
numerous hinges have been used. A novel design of the multi-hinged
type has been investigated and applied in Korea [24–26]. Two-hinge
and three-hinge types are widely used in Japan, with more than 200
sets constructed along arterial roads [22].
Although underground structures are less vulnerable to earthquake-
induced damage compared with aboveground structures [27–29],
earthquake damage to underground facilities is still necessary
[28,30–32] because of reported damage to many MUABs because of
earthquakes [20,22,33–36]. Seismic analysis and design procedures
Thus, accurate evaluation of the seismic performance of MUABs is
critical for the risk mitigation of MUABs.
MUABs subjected to ground motions have been investigated using
experimental [41–44] and numerical approaches [23,45,46]. The results
have revealed that the seismic behavior of MUABs differs under distinct
excitation directions because of the difference in the seismic capacity
between the longitudinal and transverse directions of MUABs
[22,23,46]. Furthermore, the seismic behavior of MUABs is dependent
on the peak ground accelerations (PGAs) and ground motion directions
and is sensitive to the variance in arch length and stiffness of prestressed
rebar [23]. Although these studies have evaluated seismic behaviors of
MUABs, limited studies have been conducted on the vulnerability
assessment of MUABs.
The surrounding soil plays a role in the stability of MUABs. Homaei
and Yazdani [47] pointed out that the soil–structure interaction affects
the confidence level of bridge safety; therefore, the soil–structure
interaction should be adopted to estimate arch bridge capacity. Miya­
zaki et al. investigated the effect of embankment shape patterns on the
seismic behavior of MUABs [41], and Nguyen et al. investigated the

* Corresponding author.
E-mail address: jwonhuh@chonnam.ac.kr (J. Huh).

https://doi.org/10.1016/j.istruc.2022.04.005
Received 7 January 2022; Received in revised form 30 March 2022; Accepted 1 April 2022
2352-0124/© 2022 Institution of Structural Engineers. Published by Elsevier Ltd. All rights reserved.
V.-T. Nguyen et al.
influence of soil–structure interaction on the seismic responses of
MUABs [23]. The static loads on the arch bridge increased approxi­
mately linearly with the depth of cover, whereas the loads were only
mildly affected by the cover depth under earthquake loadings [45]
because phasing occurred between the motion of the side fill and the
cover. Studies conducted on MUABs have considered limited ground
motions. In some studies, MUABs have been subjected to one ground
motion with no change in the intensity level [46]. Nguyen et al. [23]
provided considerable insights on the seismic behavior examination for
MUABs by using the three-dimensional (3D) finite element analysis
(FEA) model. However, limited number of ground motions have been
investigated. Nguyen et al. [23] indicated that further study is necessary
for investigating the seismic vulnerability of MUABs.
This study proposed a feasible and efficient fragility-based procedure
to assess the seismic performance of MUABs. First, previous literature is
discussed, and a sophisticated 3D finite element modeling of three-hinge
MUABs was simulated considering contact interactions and prestressed
rebar. The concrete damaged plasticity model was used to investigate
the micro-damage-cracking behavior of concrete by using the dissipated
fracture energy. A methodology of fragility-based analysis for MUABs
was subsequently derived. Next, we described a random selection pro­
cedure of ground motions for the 3D FEA model of MUABs. An extensive
database of ground motions was inputted in the FEA model, and the
uncertainty in the ground motion time history was assessed. Next, we
discussed critical findings from seismic fragility analysis, followed by
crucial conclusions drawn from this research.
Fig. 1. Schematic of the studied MUAB [23].
1219
Structures 39 (2022) 1218–1230
2. 3D finite element model of MUAB
A 3D FEA model of three-hinge MUAB was constructed based on the
previous study, as illustrated in Fig. 1. Numerical simulation was per­
formed according to Nguyen et al. [23].
In the FEA model, incompatible-mode eight-node brick elements
from ABAQUS [48] were used to represent the arch-foundation-
wingwall-soil system of the MUAB. The most optimal meshing of
MUAB was obtained through numerous trial analyses. The arch was
finely discretized for seismic damage analysis of concrete. Boundaries of
the FEA model were truncated with an approximate ratio of 1.5 from the
center of the arch to reduce computational cost. The truncated bound­
aries were based on the viscous-spring artificial boundary model [49]
applied in ABAQUS [12,50]. A viscous-spring artificial boundary was
imposed by installing springs and dampers on the truncated boundaries
of the surrounding soil corresponding to excitation directions [23].
Springs and viscous dampers are installed in the normal and tangential
directions on each node of the truncated boundaries. Consequently, the
FEA model closely reproduced the actual response of far-field soil
because of the truncated boundaries; the seismic waves were absorbed
on reaching the boundary.
The concrete damaged plasticity (CDP) model [51,52] was used for
concrete material considering its nonlinearity [53]. Two damage vari­
ables, namely failure mode and tensile cracking and compression
crushing, were adopted for the concrete model. Detailed can be further
referred to Nguyen et al. [23] and Yazdani and Habibi [54]. The CDP
V.-T. Nguyen et al.
model represents the damage-cracking behavior with the formation of
micro-cracks. M40 grade concrete was used for the segmented arch and
the wingwall, whereas M30 grade concrete used in ABAQUS was used
for invert foundation [23]. The concrete of segmented arch has the
following properties: initial elastic modulus of 30,000 MPa, Poisson’s
ratio of 0.2, the mass density of 2450 kg/m3, compressive strength of 40
MPa, the tensile strength of 4 MPa, the maximum compressive strain of
0.00363, and maximum tensile strain of 0.00133. The dynamic prop­
erties of concrete were used in nonlinear seismic analysis. The structural
damping of MUABs was incorporated using Rayleigh damping and 5%
damping ratio in the fundamental vibration modes to investigate mass-
and stiffness-proportional damping factors.
The damaged plasticity parameters were considered in the numerical
simulation of the CDP model in ABAQUS [23,48]. The dilation angle
considerably affected the entire model because it was the ratio between
volume strain and shear strain. The dilation angle depended on two
parameters, namely plastic strain and confining pressure. An increase in
plastic strain and confined pressure resulted in a decrease in the dilation
angle. Thus, concrete was assumed to exhibit a constant dilation angle of
31◦ for numerous pressure stresses used for the confinement of concrete.
The default flow potential eccentricity was 0.1. The default value of fb0/
fc0 was 1.16, representing the ratio of initial equibiaxial compressive
yield stress to starting uniaxial compressive yield stress. The ratio of the
second stress invariant on the tensile meridian to the compressive me­
ridian was 0.67. A null default viscosity parameter was used so that
viscoplastic regularization did not occur [48]. This parameter value
enhanced the convergence rate of the model when the softening process
occurred, and it provided excellent results in the seismic analysis
[23,55].
3. Methodology of fragility-based analysis for MUABs
The fragility curve is a potent tool for identifying the potential
seismic risk and consequences of MUABs during and after an earthquake
[56,57]. This study involved defining damage states of MUABs, fragility
curve function, and random ground motions compatible with the
seismic hazard used as input excitations. An overview of the method­
ology of fragility-based analysis for MUABs is presented in this section.
3.1. Damage states of MUABs
In the fragility-based approach, seismic fragility curves correspond­
ing to the variance in the damage index with an increase in the intensity
measurement for various levels of damage states (DSi) are plotted. Thus,
the damage states represent the damage condition of MUABs associated
with a limit state of the structural response.
DSi and corresponding indices have not been identified for under­
ground structures like MUABs. Numerous methods have been proposed
to indicate the damage levels of underground structures [58,59]. For
underground box bridges (UBBs), the number of plastic hinges could
represent damage levels [59]. Furthermore, the damage index (DI) could
be assumed to be the ratio of the elastic moment demand to the yield
moment (also known as the capacity bending moment) [58–60]. An
assumption for this definition is that the seismic behavior of UBBs is
approximated to that of an elastic beam subjected to deformations
imposed by surrounding soil because of seismic waves propagating
perpendicular to the longitudinal direction of UBBs.
For underground arch bridges (UABs), Argyroudis et al. [61] used
the same method with UBBs. However, the studies mentioned above are
based on two-dimensional (2D) FEA models of underground bridges
[58–61]. Hence, ground motions in all horizontal directions have not
been considered. The seismic resistance of MUABs between the longi­
tude and transversal direction differs considerably [23,46]. The ratio of
the maximum bending moment on the arch to its capacity is widely used
to determine damage states. The arch structure cannot withstand a
loading combination of static and dynamic loads at a single bending
1220
Structures 39 (2022) 1218–1230
moment. A load scenario can be a combination of axis force (N), shear
force (V), and moment (M) (Fig. 2). Thus, limited studies have been
conducted on the DI of MUABs because the criterion for the probabilistic
seismic analysis of MUABs does not exist.
Furthermore, recent efforts delved into investigating the seismic
resistance of masonry arch bridges (MABs) [5,6,18,47,54]. The MAB is
the sibling structure of MUABs due to forming from modular concrete
arch structures and using the soil-arching effect in the overall structural
stability. The maximum displacement of the arch (aka the crown) was
considered as the structural engineering demand parameter or repre­
sented the damage state of MABs [6,47] instead of other structural re­
sponses because of similarity and accessibility with the well-known drift
[62]. The probability of damage state presented by the maximum
displacement was adopted using the incremental dynamic analysis [6].
Yazdani and Habibi [54] proposed and employed the extended finite
element method, which is efficient for quantitative crack growth pat­
terns and failure of MABs. Therefore, the crack is a manifestation of the
failure state that can be quantified using estimation to evaluate the loss
of the structure. The quantities of the structural response (i.e., moment,
displacement, stress, crack) could use to represent the damage condition
of different soil–arch structure systems.
The crack depth can reflect the damage level of the concrete arch. At
every cross-section of each strip of the arch, tensile and compressive
stresses resulting from thrust and/or shear force and/or moments are
combined; the internal-force combination depends on the geometric
shapes of the arch and the external-force combination (Fig. 2). The
equivalent plastic strain (PEEQ) results from stress on the cross-section.
Crack propagation with crack depth corresponds to PEEQ because the
macrocracks propagate through the arch when the PEEQ value exceeds a
threshold value [23].
In this study, the DI is assumed to be based on the classification but
not on the definition of Pitilakis [63] and Argyroudis and Pitilakis [60].
The DI is the ratio between the crack depth (DCrk) and the thickness of
the arch component (DThks), as expressed in Eq. (1), followed by a suit of
damage, as presented in Table 1. The DI, ranging from 0 to 1, is not
dependent on loading directions. Furthermore, the DI is not dependent
on the loading combination of N-V-M scenarios. The damage state
classifications of MUABs, from DS1 to DS3, corresponded to the well-
known three performance levels [immediately occupancy (IO), life
safety (LS), and collapse prevention (CP)] [64,65]. Therefore, in
approximation theory, the minor, moderate, and extensive damage can
be considered serviceability performance, repairable level, and collapse
prevention. Definitions in studies have been subjective, resulting in high
uncertainty of qualification in a particular damaged state. The defini­
tions presented in this study can be used for analytical analysis.
crack depth DCrk
DI = = (1)
thickness of arch component DThks
The probabilistic seismic demand model could be based on regres­
sion analysis [56,66–70] in accordance with adaptable and sufficient
data regarding the demands and/or capacities of MUABs. Thus, the
relationship between the structural engineering demand parameter
(EDP) and the ground motion parameter (IM) in the study could be
expressed as follows:
EDP = a × IM b (2)
where IM is the intensity measure of ground motions; a and b are un­
known regression coefficients obtained from linear regression analysis.
For a database of EDP, the median value (EDPmedian) of the EDP-data
set is well-known as a special parameter. EDPmedian is a value separating
the higher half from the lower half of the EDP-data set. The relationship
between EDPmedian and IM can be expressed as follows:
EDPmedian = a × IM b (3)
Using a two-sided logarithm, Eq. (3) becomes a new form similar to a
V.-T. Nguyen et al. Structures 39 (2022) 1218–1230

Fig. 2. Various cross-section shapes of MUABs correspond with their internal-force combination from thrust (N) and/or shear force (V) and/or moment (M).

Table 1
Definition of damages states for MUABs.
Damage Meaning Range of Median value of Performance
state level damage damage index level [64,65]
index (DI or (DImedian or
EDP) EDPmedian)

DS0 None DI ≤ 0.05 – –

DS1 Minor/ 0.05 < DI ≤ 0.125 IO
slight 0.2
DS2 Moderate 0.2 < DI ≤ 0.325 LS
0.45
DS3 Extensive 0.45 < DI ≤ 0.625 CP
0.8
DS4 Collapse DI > 0.8 – –

power model [66].

ln(EDPmedian ) = ln(a) + b × ln(IM) (4)

3.2. Fragility curve development

The seismic fragility curve of MUABs can represent the conditional

probability of a seismic demand (EDP) exceeding the structural seismic
capacity of MUAB (C), defined by damage states, given a specific level of
ground motion IM [62,71–75], as expressed in Eq. (5). When MUAB is
subjected to ground motion having a specific seismic intensity level, the
possibility of MUAB P(DSi|IM) being damaged corresponds with the
probability of being equal to or exceeding a specific damage state (DSi)
presented in Table 1.
Pfragility = P(EDP⩾C|IM)
The fragility curves can be obtained using various methodologies,
such as expert-based, empirical, analytical, and hybrid methods. Many
researchers developed fragility curves to evaluate the seismic perfor­
mance of different structures [76–81]. These studies prove that the
fragility-based curve could be a powerful approach to estimate the
seismic probability risk of structures, including underground structures.
The fragility curve in this study was based on the analytical fragility
approach developed via dynamic structural analysis (Fig. 3). The
analytical method used for developing vulnerability curves exhibited
considerable advantages and disadvantages. The analytical method is
the most accurate and considers all uncertainties. However, this
Fig. 3. Methodology of fragility curve based on the analytical approach.
analytical method incurs considerable computational costs. Further­
more, developing adaptive numerical modeling with high sensitivity is
challenging. Nevertheless, the analytical method is more suitable than
(5)
the empirical method for seismic vulnerability assessment of MUABs
[82] because observation data of seismic damage cannot be used to
develop MUABs.
The fragility function is defined using the lognormal cumulative
distribution function [4,56,60,61] as expressed in Eq. (6). The
lognormal distribution of the seismic intensity of earthquakes resulting
in the collapse of MUABs can be expressed as follows:
[ ]
ln(x/θ)
Pfragility = P(C|IM = x) = Φ (6)
β
where P(C|IM = x) is the probability that the ground motion with

1221
V.-T. Nguyen et al. Structures 39 (2022) 1218–1230

seismic intensity IM = x will cause MUAB to collapse, Φ[ ] is the stan­ underground structures, Huh et al. [4] considered 23 seismic PGA levels
dard normal cumulative distribution function, θ is the median of ranging from 0.02 to 1.3 g and 50 artificial acceleration time histories
fragility function (the IM level with 50% probability of collapse), and β is for each PGA level according to the Korean design spectrum codes
the standard deviation of ln(IM). Therefore, θ and β are fragility pa­ [89,90]. Huh et al. [4] recommended that the number of ground motions
rameters of the distribution function. per PGA was equal to or greater than 20, which reasonably satisfies the
A multiple stripe analysis was conducted using seismic IM levels. fragility curves of UBBs. Consequently, 1,150 ground motion time his­
Structural analyses of collapses were conducted for numerous ground tories were applied to derive vulnerability curves of a two-story UBB
motions at intensity levels IM = xj. Assuming that the observation of based on the ground response acceleration method [4]. The 1D linear
collapse or no-collapse from each ground motion is independent of the site response analysis was adopted to perform the free-field soil response
observations from other ground motions, the probability mass function for each soil profile [4,86].
of observing zj collapses or no collapses out of nj ground motions with Hu et al. [91] selected fifteen-actual ground motions with ten in­
IM = xj can be calculated by the binomial distribution. The probability pj tensities up-and-down scaled to acquire the internal forces of UABs. Bao
that a ground motion with IM = xj causes the collapse of MUABs is et al. [2] used a dataset of 10 earthquakes with a large magnitude of 6.5
expressed as follows: to 6.9 from the Pacific Earthquake Engineering Research Center (PEER)
( ) database to perform a seismic vulnerability analysis of buildings. The
( ) nj zj ( )n − z
P zj collapse in nj ground motions = pj 1 − pj j j (7) selected earthquakes were then scaled according to their intensity in­
zj
dexes to generate 200 ground motion samples. Notably, Chen et al. [92]
The maximum likelihood function was obtained by considering the used 42 near-fault pulse-like ground motions records. The 42-real
product of the binomial probabilities at each IM level to obtain the ground motions having PGA from 0.13 to 1.34 g were applied to cap­
likelihood for the entire data set as Eq. (8). Therefore, the fragility ture seismic hazard levels for pier bridges.
function provided the highest probability of collapse data based on the In particular, the prior techniques to build up ground motion sam­
maximum likelihood approach. The observation data of collapse was plings are dissimilar in terms of selecting criteria, earthquake charac­
obtained from the dynamic structural analysis, which was conducted teristics, the number of ground motions, and generating/calculating
using various IM levels (m is the number of various IM levels). methods for accelerograms [2,4,88,91,92]. These studies made suitable
m ( ) choices to solve the research objectives following degrees of an
∏ nj zj ( )n − z
Likelihood = pj 1 − pj j j (8) acceptable level. The artificial accelerograms can be useful when the
j=1
z j
approach using real accelerograms is challenging or inappropriate. The
use of real records could be a potential selection when there is an in­
The fragility parameters (θ and β) are explicit in the substitution
crease in the availability of strong-motion accelerograms. Therefore,
likelihood function, as expressed in Eq. (9). Estimates of the fragility
using real records and spectrum matching techniques was recommended
function parameters are obtained by maximizing the logarithm of the
for deriving suits of records for use in nonlinear dynamic analysis of
likelihood function as follows:
structures [93]. No specific guidance or reference exists for the number
m ( ) ( )z [ ( ) ]nj − zj
∏ nj ln(x/θ) j ln(x/θ) of random ground motions selected from the current earthquake data­
Likelihood = Φ 1− Φ (9)
zj β β base as the PEER for MUABs. Thus, this study proposed the following
j=1
procedure to select ground motions from the strong-motion database of
{⌢ ⌢} ∑m { (
nj
) [ (
ln(x/θ)
)]
( ) PEER randomly:
θ , β = argmax ln + zj × ln Φ + n j − zj
z β
(1) Each earthquake candidate has distinct characteristics. First, the
θ,β j=1 j
[ ( )]}
× ln 1 − Φ
ln(x/θ) Richter magnitude (Mw) is equal to or greater than 4.0 [94].
β Second, the epicentral distance is greater than 10 km to reduce
(10) the influence of near-field effects. Third, the travel-time averaged
near-surface shear wave velocity corresponding to 30 m depth of
3.3. Input ground motions and intensity measurements Vs30 is smaller than 1500 m/s to cover most classification of
subsoil classes [95–97].
A database of ground motion records is required for seismic fragility (2) In the NGA databases [98,99], only the RotD50 component [100]
analysis. Therefore, random selections of ground motions are critical for is available. Initially, the record component of the RotD50 mea­
estimating seismic demands because the selections must ensure an sure was adopted to perform searches for records based on the
adequate number of ground motions to satisfy site characteristics [23]. median value of the distribution of each IM.
Many scholars have evaluated the influence of different types of (3) For each seismic event, only horizontal seismic waves should be
seismic IM in earthquake performance assessment for underground and selected for each station to avoid the effect of the same focal
semi-underground structures [83–87]. Nguyen et al. [85] revealed that mechanism. Next, the seismic waves are selected for greater PGA
the PGA might not be the best option for seismic response analysis in the priority [91].
high-frequency seismic regions for nuclear power plants. However, for (4) Kohrangi et al. [101] noted that when analyzing 3D structural
shallow UBBs, the PGA is an efficient and appropriate IM of the ground models subjected to ground motions in the two horizontal di­
motions in constructing seismic fragility curves [83]. Zhang et al. [84] rections, alternative components definitions, such as GeoMean or
concluded that the optimal IM depends on stratum characteristics in the SRSS, should be considered. Thus, the spectral ordinate of SRSS
fragility analysis of UBBs. The PGA is more adaptable for the stratum with a damping ratio of 5% was adopted for selecting control.
having high equivalent shear wave velocity, while the velocity spectrum (5) To define the target spectral ordinate, the record selections
intensity is a priority for low equivalent shear wave velocity. However, should match the Korean code-based spectrum [89,90]. This
the peak ground velocity is better correlated with seismic damage to study was first based on a return period of 500 years, alongside
deep underground structures [86,87]. Therefore, the PGA was adopted seismic zone 1.
as seismic IM to assess the seismic performance of MUABs without loss of (6) At least 40 best-fit records were selected to be ground motions.
generality because the MUABs were designed as shallow underground
structures. A set of selected ground motion records is listed in Table 2, which
A range from 10 to 20 records is sufficient to provide high accuracy matched the Korean design spectrum. The selections are scattered in an
in the seismic demand analysis for mid-rise buildings [88]. Regarding approximately 50-year period. The selected records were scaled into 12

1222
V.-T. Nguyen et al. Structures 39 (2022) 1218–1230

Table 2
Selected ground motions from the PEER ground motion database.
No. RSN Event Year Mw PGA (g) PGV (cm/s) PGD (cm) Vs30 (m/s)

1 1091 Northridge-01 17/01/1994 6.69 0.15 13.69 2.87 996.43

2 1161 Kocaeli, Turkey 18/01/1994 7.51 0.19 40.77 36.22 792
3 3954 Tottori, Japan 19/01/1994 6.61 0.2 15.67 9.28 967.27
4 1011 Northridge-01 20/01/1994 6.69 0.14 12.2 2.07 1222.52
5 5618 Iwate 21/01/1994 6.9 0.27 24.87 12.76 825.83
6 80 San Fernando 22/01/1994 6.61 0.14 9.86 1.05 969.07
7 680 Whittier Narrows-01 23/01/1994 5.99 0.1 7.92 0.99 969.07
8 3925 Tottori, Japan 24/01/1994 6.61 0.15 11.09 12.83 940.2
9 1256 Chi-Chi, Taiwan 25/01/1994 7.62 0.07 8.25 5.49 789.18
10 788 Loma Prieta 26/01/1994 6.93 0.08 9.23 3.72 895.36
11 243 Mammoth Lakes-04 25/05/1980 5.7 0.25 10.14 1.22 537.16
12 1762 Hector Mine 16/10/1999 7.13 0.19 23.78 14.82 382.93
13 1135 Kozani, Greece-04 19/05/1995 5.1 0.23 14.62 1.34 511.16
14 3979 San Simeon, CA 22/12/2003 6.5 0.16 13.43 7.69 362.42
15 548 Chalfant Valley-02 21/07/1986 6.19 0.19 15.16 2.93 370.94
16 15 Kern County 21/07/1952 7.36 0.16 16.96 8.04 385.43
17 6876 Joshua Tree, CA 23/04/1992 6.1 0.21 10.12 1.7 425.02
18 164 Imperial Valley-06 15/10/1979 6.53 0.17 15.88 6.6 471.53
19 590 Whittier Narrows-01 01/10/1987 5.99 0.23 8.9 1.12 375.16
20 572 Taiwan SMART1(45) 14/11/1986 7.3 0.14 13.68 6.75 671.52
21 161 Imperial Valley-06 15/10/1979 6.53 0.18 37.41 21.82 208.71
22 730 Spitak, Armenia 07/12/1988 6.77 0.19 20.58 7.35 343.53
23 1100 Kobe, Japan 16/01/1995 6.9 0.21 21.15 10.59 256
24 1754 Northwest China-04 15/04/1997 5.8 0.21 11.92 2.11 240.09
25 850 Landers 28/06/1992 7.28 0.15 19.92 8.06 359
26 6886 Darfield, New Zealand 03/09/2010 7 0.19 41.66 41.91 280.26
27 733 Loma Prieta 18/10/1989 6.93 0.16 14.88 6.99 271.06
28 2943 Chi-Chi, Taiwan-05 22/09/1999 6.2 0.17 8.6 2.86 277.5
29 6874 Joshua Tree, CA 23/04/1992 6.1 0.2 12.5 3.24 333.89
30 826 Cape Mendocino 25/04/1992 7.01 0.17 24.99 8.61 337.46
31 1209 Chi-Chi, Taiwan 20/09/1999 7.62 0.18 25.37 12.17 169.52
32 3934 Tottori, Japan 06/10/2000 6.61 0.17 21.84 11.04 138.76
33 5989 El Mayor-Cucapah 04/04/2010 7.2 0.16 25.42 18.6 162.94
34 729 Superstition Hills-02 24/11/1987 6.54 0.2 31.61 26.14 179
35 3302 Chi-Chi, Taiwan-06 25/09/1999 6.3 0.15 10.17 5.44 169.84
36 3282 Chi-Chi, Taiwan-06 25/09/1999 6.3 0.2 14.49 5.83 169.52
37 718 Superstition Hills-01 24/11/1987 6.22 0.13 13.8 5.02 179
38 2715 Chi-Chi, Taiwan-04 20/09/1999 6.2 0.13 15.43 8.97 169.52
39 5471 Iwate 13/06/2008 6.9 0.11 9.61 4.24 158.16
40 608 Whittier Narrows-01 01/10/1987 5.99 0.11 10.09 1.62 160.58

intensity levels (0.03, 0.09, 0.15, 0.3, 0.45, 0.6, 0.75, 0.9, 1.05, 1.2,
1.35, and 1.5 g) according to their intensity indexes. Therefore, a dataset
of 480-time history samples can represent random input ground motion
for the fragility analysis of MUAB.
The response spectrum of the 40 selected ground motion records is
displayed in Fig. 4. A deconvolution procedure was adopted for
matching the earthquakes [23]. The average response spectrum was
similar to the design response spectrum, which could play a critical role
in seismic behavior and response of MUABs. Each acceleration time
history from the data set of time history samples was separately applied
to the 3D FEA model at the bottom face.
Differences in seismic resistance and seismic response of MUABs
resulting from the horizontal ground motion effects and/or the vertical
ground motion effects are available in the literature [23,45]. The
maximum thrust in the arch was significantly affected by the depth of
cover during the vertical seismic loading. In contrast, the effect of the
horizontal seismic loadings on the arch was more detrimental due to the
combination of thrust and moment in the arch structure. The maximum
moment was caused by the horizontal seismic motions rather than the
ones in the vertical direction [45]. The detrimental force acting on the
arch with shallow cover depth was influenced mainly by the effect of the
horizontal ground motions compared to the ones in the vertical direction
[23]. Therefore, this seismic fragility analysis focused on the combined
effects of the vertical and horizontal ground motion components. In
every application, the PGA of ground motion was assumed to be iden­
tical in both the longitudinal and transverse directions and two-thirds in
the vertical direction. The assumption is similar to the hypothesis for the
seismic performance assessment of arch dams [12].
The types of distribution in the data of earthquake characteristics are
displayed in Fig. 5. Linear regression results from the least-squares an­
alyses revealed that Vs30 data fits with lognormal distribution, whereas
PGA is distributed in the normal distribution. Similarly, moment
magnitude Mw and the epicentral distance of the earthquakes are rep­
resented in normal and lognormal distributions, respectively. Although
the selection was random from the PEER to eliminate stratum-
uncertainty characteristics, the analyses indicated that earthquake
characteristics have correlations between earthquake quantities and
uncertainty of stratum in their distribution (i.e., Vs30, PGA, Mw, and
epicentral distance). There is a similarity in the distribution shape of
PGA and Mw, which represents the earthquake magnitude. Furthermore,
there is a prominent correspondence of the convergence degree in dis­
tribution between Vs30 and the epicentral distance. The lognormal dis­
tribution is more consistent with the geological properties, and hence
Vs30 and epicentral distance can be used to estimate geological structure.
Therefore, the data on earthquake characteristics can range between the
upper and lower bound of the probabilistic seismic hazard curve. In
comparison with previous studies [102,103], this proposal of ground
motion selection provided accurate results because uncertainties of the
parameter of the soil deposit structures were considered without any
random modeling by random parameters. These random selections
covered uncertainties of the soil parameters. Therefore, seismic hazards
defined in terms of acceleration spectra of a set of 40-earthquake records
represented the seismic environment.

1223
V.-T. Nguyen et al.
Fig. 4. The response spectrum of the selected ground motion records.
4. Results and discussion
4.1. Probabilistic seismic demand model
Many seismic standards are used; therefore, site classification criteria
for the seismic response of soil profile differ for countries and continents,
as presented in Table 3. This study clarified the selection of criteria.
Specifically, the EUROCODE-8 standard [104] was used to classify the
ground with the average shear wave velocity of the upper 30 m soil layer
in the same approach as NBCC [105] and ASCE-7-16 [106]. However, in
Japan and Mexico, the site period is applied as a ground classification
method because the amplification of seismic motion by the ground oc­
curs in the site period and increases the seismic load of structures similar
to the site period. The average shear wave velocity of the upper 30 m soil
layer and the site period are mixed according to Australia, New Zealand,
and China standards. Most classification standards are based on shear
wave velocity. Therefore, the shear wave velocity was adopted in this
study to classify the seismic site responses of soil deposit profiles.
Uncertainties in soil deposits were considered according to set
standards [89,104–106,112]. The soil deposits were classified into four
groups based on their shear wave velocities Vs30 and corresponded to a
decrease in soil stiffness, as listed in Table 4. This classification revealed
that it was in accordance with various standards. Therefore, the classi­
fication can represent the variation of shear wave velocities of the soil
1224
Structures 39 (2022) 1218–1230
deposits and uncertainties in soil properties. Multiple stripe analysis
[56] by loading a data set of ground motion covering all types of strata
can comprehensively assess the seismic performance of MUABs.
The results of linear regression analysis between DI and PGA with
various soil deposits are presented in Fig. 6. The goodness-of-fit (R2) of
all analyses is more significant than 0.9. Thus, the discreteness of
regression results is small, and these fitting results are satisfactory. This
analysis revealed that the correlation between the seismic intensity of
ground motions and the DI of MUAB is the lognormal distribution.
However, the increase in PGA did not rapidly increase the crack depth
when the crack depth was extensive. As mentioned before, the damage
index is the ratio of the crack depth and the thickness of the arch
component. Crack propagated in the arch depth depends on the ground
motion intensity level (i.e., PGA) [23]. Generally, the crack occurs at
specific intensity then progresses with increments in the PGA. Interest­
ingly, the relationship between the crack depth and the PGA was not
perfect linearity from the initial crack to the completed crack. An in­
crease in crack depth propagation slowed down when the concrete
reached nearly complete plasticity [23]. As a result, the increase in ln
(DI) significantly decreased when the PGA increased over a threshold
value (PGA = 1 g). Therefore, the relationship between ln(PGA) and ln
(DI) is linear for each PGA interval, although the confidence level of the
fitting is high.
The higher the Vs30 value is, the lower vulnerability in the MUABs is;
V.-T. Nguyen et al. Structures 39 (2022) 1218–1230

Fig. 5. Distribution function in data of earthquake record characteristics: (a) Travel-time averaged near-surface shear wave velocity Vs30; (b) Peak ground accel­
eration PGA; (c) Moment magnitude Mw; and (d) Epicentral distance.

Table 3 Table 4
Site classification criteria for the seismic site response of soil profile corre­ Variation of shear wave velocities of the soil deposits corresponding to earth­
sponding to various standards. quake standards.
Continent Representative Standard Classification criteria Group Vs30 (m/ Mean Equivalent site class Equivalent site
territory soil s) EUROCODE-8 [104], class in KDS 17
ASCE-7-16 [106], NEHRP 10 00 [89]
Asia Japan BCJ 1997 [107] Site period
[112], and NBCC [105]
Korea KDS 17 10 00 [89] Shear wave velocity
China GB-50011-2010 Shear wave velocity and 1 760–1500 Rock Class B S1
[108] site period 2 360–760 Soft Class C S1, S2
Europe EU EUROCODE-8 Shear wave velocity rock
[104] 3 180–360 Stiff Class D S2, S3, S4
Oceania Australia AS-1170.4-2007 Shear wave velocity and soil
[109] site period 4 <180 Soft Class E S5
New Zealand NZS 1170.5-2004 Shear wave velocity and soil
[110] site period
America Canada NBCC [105] Shear wave velocity
USA ASCE-7-16 [106] Shear wave velocity 4.2. Fragility curve
Mexico MOC-2008 [111] Site period

The maximum likelihood method is the most appropriate fitting

which is consistent with critical findings for buried circular structures
[86] because the Vs30 value decreases with the decrease in the hardness
in soil properties from group soil 1 to 4, resulting in the reduction of the
resistance of the ground to the effects of earthquakes. This result is
consistent with that displayed in Fig. 7, in which a correlative com­
parison exists in the median values of PGA, causing each damage state
level to MUAB for various soil deposits. The MUAB exhibited superior
results in the stiff soil deposit group because of a higher value of PGA to
reach damage state levels.
technique [73,113,114]. Seismic fragility curves represent the condi­
tional probability of exceeding a predefined damage state for various
soil types. The validation of the MUAB seismic performance distribution
model was performed through the hypothesis testing of good-of-fit for
determining the correlation between IM and EDP, as displayed in Fig. 6.
Although a limited number of IM levels exist, the fragility fitting using
multiple stripe analysis provides considerable advantages in evaluating
MUAB because of an effective strategy [56,115].
Figs. 8 and 9 depict the fragility curves of the seismic performance of
MUAB under the three damage levels for various soil deposits. The

1225
V.-T. Nguyen et al.
Fig. 6. DI (EDP) versus various IMs for four soil profile classes in accordance with the Korean standard to consider the effects of local soil conditions, geological
conditions, and surface and subterranean topography on the seismic demand of the MUAB corresponding to the classification of soil deposits as presented in Table 4.
Fig. 7. Relationship between the damage state levels of MUAB for various soil
deposits and intensity levels of ground motions.
vulnerability curve of the MUAB indicates the probability that the
seismic demand of MUAB exceeds a specific failure state under the
loading of array IM of earthquakes. The abscissa (horizontal axis) in­
dicates the PGAs representing the magnitude of randomly seismic
ground motions. As the intensity of ground motion increases, the
collapse probability of MUAB increases. However, the curves differed
considerably in various types of soil deposits. The higher performance
level progresses from the minor to extensive levels; the curve slope is
1226
Structures 39 (2022) 1218–1230
gentler. Notably, the better the soil deposit, the smoother the curve. The
fragility curve of the minor level is the steepest, which revealed the
probability of MUAB surpassing the regular use performance level
subjected to the earthquake effect.
Fig. 9 displays each seismic performance level’s probability thresh­
olds (upper and lower thresholds). These probability-bounded regions
determine the variation degree of collapse probability for the damage
state levels. Furthermore, the variation degree of collapse probability
covering uncertainties in soil deposit was dependent on both PGAs and
probability levels.
The estimated fragility parameters are listed in Table 5. The esti­
mation is appropriate and reasonable because of the accurate fragility
function obtained from structural analyses. The fragility function pa­
rameters were calculated by using the maximizing the likelihood func­
tion. Results of the estimation revealed that the standard deviation
parameter (β) and the median parameter (θ) decreased with a decrease
in the soil property of Vs30 (soil 1 to soil 4). Variation of the median
parameter is displayed in Fig. 10. The median of the fragility function
corresponded with the IM level with a 50% probability of collapse, and β
reflected the standard deviation of ln(IM). The effective IM values are
approximately centered around the median of the fragility function.
Analysis of the seismic behavior of MUAB [23] revealed that the
segment components of MUAB are repairable (corresponding to the
crack depth ratio of 20%) when PGA was below 0.47 g for Vs30 = 620 m/
s with individually longitudinal/transverse ground motion. This per­
formance level of MUAB can be equal to the minor damage state (IO).
However, the analyzed results revealed that the MUAB could reach the
IO level (50% probability) with combined ground motions having PGA
V.-T. Nguyen et al. Structures 39 (2022) 1218–1230

Fig. 8. Multiple stripe analysis results for observed fractions of collapse, fragility curves based on fragility-estimated function according to maximum likelihood
method: (a) Soil 1; (b) Soil 2; (c) Soil 3; (d) Soil 4.

Fig. 9. Probability thresholds for each seismic performance level considering

stratum-uncertainty characteristics. Fig. 10. Variation in the IM level with a 50% probability of collapse for soil
classification.

Table 5 ranging from 0.21 to 0.25 g, regardless of soil deposit profiles.

Fragility function-related parameters for the soil deposit classification corre­ The level design of earthquake ground motion corresponds with the
sponding with the damage state levels. seismic performance level of MUABs. The probabilistic ground motion in
Damage state Soil 1 Soil 2 Soil 3 Soil 4 modern earthquake-resistant design standards [89] was used for the
level expected performance of MUABs displayed in Fig. 11. Three basic-
θ β θ β θ β θ β design earthquake levels are serviceability earthquake (SE)—10%
Minor 0.25 0.32 0.23 0.26 0.22 0.24 0.21 0.26 probability of exceedance in 10 years (100-year return period), design
Moderate 0.60 0.31 0.53 0.29 0.48 0.27 0.39 0.25 basis earthquake (DBE)—10% probability of exceedance in 50 years
Extensive 1.05 0.28 0.99 0.27 0.87 0.24 0.80 0.20
(500-year return period), and maximum considered earthquake

1227
V.-T. Nguyen et al. Structures 39 (2022) 1218–1230

MUAB can satisfy the seismic design philosophy; the MUAB can resist
random SEs with a PGA of 0.3 g and random DBEs with a PGA of 0.5 g.
Future research should consider the potential effects of the
geometrical arch properties, such as arch cross-section shapes, because
arch geometry considerably affects MUAB stability.

Declaration of Competing Interest

The authors declare that they have no known competing financial

interests or personal relationships that could have appeared to influence
the work reported in this paper.

Acknowledgments

The authors acknowledge the support by the Korea Agency for

Infrastructure Technology Advancement grant funded by the Ministry of
Fig. 11. PGAs of the probabilistic ground motion in modern earthquake-
Land, Infrastructure and Transport (20CTAP-C151892-02). The authors
resistant design standard [89].
also thank PEER for the databases of ground motion records used in this
study.
(MCE)—10% probability of exceedance in 250 years (2400-year return
period). The seismic performance of MUABs was classified as MUABs
References
that can resist SE without damage, resist DBE with damage that may or
may not be economically repaired but without causing extensive loss of
life, and resist the most vigorous earthquake expected MCE without
collapse, but potentially with extreme damage. Therefore, the MUAB
can satisfy the seismic design philosophy [89]. The MUAB can resist
random SEs with a PGA of 0.3 g and random DBEs with a PGA of 0.5 g.
5. Conclusions
This study proposed an efficient procedure to evaluate the seismic
performance of a three-hinge MUAB using multiple stripes analysis to
develop fragility curves with 3D finite element models. The MUAB
(having geometrical properties of segment arch: inner radius of 5 m, the
identical thickness of 0.25 m, widths of each segmental component 1.25
and 0.625 m, length of arch 7.5 m) was used to assess the seismic per­
formance of the arch. As a result, we obtained the following conclusions:
This study indicates that the proposed procedure to select ground
motions from the strong-motion database of PEER randomly is feasible
to evaluate the seismic vulnerability of MUABs. Damage states can be
investigated using the relationship between the crack depth and thick­
ness of the arch component. The multiple stripe analysis approach with
the maximum likelihood function can consider the uncertainty charac­
teristics of soil deposits by allowing various ground motions at varying
seismic intensity levels. A dataset of ground motions was developed to
represent random input ground motion for the fragility analysis of
MUABs. Thus, the vulnerability assessment of MUABs to random ground
motions was performed.
The likelihood of MUAB collapsing increased with the increase in
ground motion intensity. However, the fragility curves in various types
of soil deposits differed considerably. The curve slope became gentle as
the performance level moved from minor to extensive. The better the
soil deposit is, the smoother the curve. The fragility curve of the minor
level was the steepest, which revealed the probability of MUAB sur­
passing the regular use performance level subjected to the earthquake
effect. The varying degrees of collapse probability that covered un­
certainties in a soil deposit were dependent on both PGAs and proba­
bility levels.
The standard deviation and median parameters of the fragility curve
decreased with a decrease in the soil property of Vs30. The MUAB could
reach the IO level (50% probability) with combined ground motions
having PGA ranging from 0.21 to 0.25 g, regardless of soil deposit
profiles. The level design of earthquake ground motion corresponds with
the seismic performance level of MUABs. The probabilistic ground mo­
tion should be used for predicting the performance of MUABs. The
[1] Argyroudis SA, Mitoulis SA. Vulnerability of bridges to individual and multiple
hazards-floods and earthquakes. Reliability Eng Systm Safety 2021;210:107564.
https://doi.org/10.1016/j.ress.2021.107564.
[2] Bao C, Xu F, Chen G, Ma X, Mao M, Zhang S. Seismic vulnerability analysis of
structure subjected to uneven foundation settlement. Appl Sci 2019;9:3507.
https://doi.org/10.3390/app9173507.
[3] Huh J, Haldar A. Stochastic finite-element-based seismic risk of nonlinear
structures. J Struct Eng 2001;127:323–9. https://doi.org/10.1061/(ASCE)0733-
9445(2001)127:3(323).
[4] Huh J, Tran QH, Haldar A, Park I, Ahn J-H. Seismic vulnerability assessment of a
shallow two-story underground RC box structure. Appl Sci 2017;7:735. https://
doi.org/10.3390/app7070735.
[5] Jahangiri V, Yazdani M, Marefat MS. Intensity measures for the seismic response
assessment of plain concrete arch bridges. Bull Earthq Eng 2018;16:4225–48.
https://doi.org/10.1007/s10518-018-0334-8.
[6] Jahangiri V, Yazdani M. Seismic reliability and limit state risk evaluation of plain
concrete arch bridges. Struct Infrastruct Eng 2021;17:170–90. https://doi.org/
10.1080/15732479.2020.1733030.
[7] Kabtamu HG, Peng G, Chen D. Dynamic analysis of soil structure interaction
effect on multi story RC frame. Open J Civil Eng 2018;08(04):426–46.
[8] Katona MG. Seismic design and analysis of buried culverts and structures.
J Pipeline Syst Eng Practice 2010;1:111–9. https://doi.org/10.1061/(ASCE)
PS.1949-1204.0000057.
[9] Kang J, Im H, Park JS. The effect of load reduction on underground concrete arch
structures in embedded trench installations. Tunnell Underground Space Technol
2020;98:103240. https://doi.org/10.1016/j.tust.2019.103240.
[10] Lanzano G, Salzano E, de Magistris FS, Fabbrocino G. Seismic vulnerability of
natural gas pipelines. Reliab Eng Syst Saf 2013;117:73–80. https://doi.org/
10.1016/j.ress.2013.03.019.
[11] Mangalathu S, Jeon JS, DesRoches R. Critical uncertainty parameters influencing
seismic performance of bridges using Lasso regression. Earthquake Eng Struct
Dynam 2018;47:784–801. https://doi.org/10.1002/eqe.2991.
[12] Pan J, Xu Y, Jin F. Seismic performance assessment of arch dams using
incremental nonlinear dynamic analysis. Eur J Environ Civil Eng 2015;19:
305–26. https://doi.org/10.1080/19648189.2014.960950.
[13] Perdomo C, Monteiro R, Sucuoğlu H. Development of fragility curves for single-
column RC Italian bridges using nonlinear static analysis. J Earthquake Eng 2020:
1–25. https://doi.org/10.1080/13632469.2020.1760153.
[14] Sevieri G, De Falco A, Andreini M, Matthies HG. Hierarchical Bayesian framework
for uncertainty reduction in the seismic fragility analysis of concrete gravity
dams. Eng Struct 2021;246:113001. https://doi.org/10.1016/j.
engstruct.2021.113001.
[15] Seo J, Linzell DG. Horizontally curved steel bridge seismic vulnerability
assessment. Eng Struct 2012;34:21–32. https://doi.org/10.1016/j.
engstruct.2011.09.008.
[16] Seo J, Rogers LP. Comparison of curved prestressed concrete bridge population
response between area and spine modeling approaches toward efficient seismic
vulnerability analysis. Eng Struct 2017;150:176–89. https://doi.org/10.1016/j.
engstruct.2017.07.033.
[17] Tsinidis G. Response characteristics of rectangular tunnels in soft soil subjected to
transversal ground shaking. Tunnelling Underground Space Technol 2017;62:
1–22. https://doi.org/10.1016/j.tust.2016.11.003.
[18] Yazdani M, Jahangiri V. Intensity measure-based probabilistic seismic evaluation
and vulnerability assessment of ageing bridges. Earthq Struct 2020;19:379–93.
10.12989/eas.2020.19.5.379.
[19] Ren L, He S, Yuan H, Zhu Z. Seismic fragility analysis of bridge system based on
fuzzy failure criteria. Adv Civil Eng 2019;2019:1–13.

1228
V.-T. Nguyen et al.
[20] Abe T, Nakamura M. The use of and the caution in the application of the culvert
constructed by large pre-cast element in the expressway construction. Found Eng
Equip 2014;42:8–11.
[21] Fairless G, Kirkaldie D. Earthquake performance of long-span arch culverts. New
Zealand Transport Agency Research Report 2008.
[22] Miyazaki Y. Fundamental study on seismic behavior of hinge types of precast arch
culverts in culvert longitudinal direction [Doctoral dissertation]: Kyoto
University; 2019. 10.14989/doctor.k21736.
[23] Nguyen V-T, Seo J, Ahn J-H, Haldar A, Huh J. Finite element analysis-aided
seismic behavior examination of modular underground arch bridge. Tunnell
Underground Space Technol 2021;118:104166. https://doi.org/10.1016/j.
tust.2021.104166.
[24] Jeon SH, Cho K-I, Huh J, Ahn J-H. The performance assessment of a precast,
panel-segmented arch bridge with outriggers. Appl Sci 2019;9:4646. https://doi.
org/10.3390/app9214646.
[25] Jeon SH, Cho K-I, Lee W-H, Huh J, Ahn J-H. Lifting test and analysis of a
segmented arch system with outrigger ribs and flexural loading tests of precast
panels. KSCE J Civ Eng 2021;25:4285–303. https://doi.org/10.1007/s12205-
021-1553-0.
[26] Jeon SH, Moon HD, Sim C, Ahn J-H. Construction stage analysis of a precast
concrete buried arch bridge with steel outriggers from full-scale field test.
Structures 2021;29:1671–89. https://doi.org/10.1016/j.istruc.2020.12.050.
[27] Anderson DG, Martin GR, Lam IP, Wang JNJ. NCHRP REPORT 611: Seismic
analysis and design of retaining walls, buried structures, slopes, and
embankments: Transportation Research Board U; 2008.
[28] Dowding CH, Rozan A. Damage to rock tunnels from earthquake shaking.
J Geotech Eng Division 1978;104:175–91. https://doi.org/10.1061/
AJGEB6.0000580.
[29] Hashash YM, Hook JJ, Schmidt B, John I, Yao C. Seismic design and analysis of
underground structures. Tunnelling Underground Space Technol 2001;16:
247–93. https://doi.org/10.1016/S0886-7798(01)00051-7.
[30] ASCE. Earthquake damage evaluation and design considera-tions for
underground structures, February. American Society of Civil Engineers, Los
Angeles Section 1974.
[31] JSCE. Earthquake resistant design for civil engineering structures in Japan.
Tokyo: Japanese Society of Civil Engineers; 1988.
[32] Kaneshiro J, Power M, Rosidi D. Empirical correlations of tunnel performance
during earthquakes and aseismic aspects of tunnel design. Proceedings of
Proceedings of the conference on lessons learned from recent earthquakes On
earthquakes in Turkey. 1999.
[33] Jing-Ming W, Litehiser Jr JJ. The distribution of earthquake damage to
underground facilities during the 1976 Tang-Shan earthquake. Earthquake
Spectra 1985;1:741–57. https://doi.org/10.1193/1.1585291.
[34] Yu H, Chen J, Bobet A, Yuan Y. Damage observation and assessment of the Longxi
tunnel during the Wenchuan earthquake. Tunnelling Underground Space Technol
2016;54:102–16. https://doi.org/10.1016/j.tust.2016.02.008.
[35] Wang W, Wang T, Su J, Lin C, Seng C, Huang T. Assessment of damage in
mountain tunnels due to the Taiwan Chi-Chi earthquake. Tunnelling
Underground Space Technol 2001;16:133–50. https://doi.org/10.1016/S0886-
7798(01)00047-5.
[36] Wang Z, Zhang Z. Seismic damage classification and risk assessment of mountain
tunnels with a validation for the 2008 Wenchuan earthquake. Soil Dyn
Earthquake Eng 2013;45:45–55. https://doi.org/10.1016/j.soildyn.2012.11.002.
[37] Huh J, Haldar A. Uncertainty in seismic analysis and design. J Struct Eng 2002;
29:1–7.
[38] Tyrtaiou M, Papaleonidas A, Elenas A, Iliadis L. Accomplished reliability level for
seismic structural damage prediction using artificial neural networks.
Proceedings of international conference on engineering applications of neural
networks. Springer; 2020. 10.1007/978-3-030-48791-1_6.
[39] Seo J, Hatlestad AJ, Kimn J-H, Hu JW. Application of mathematical functions for
seismic increment fragility determination. Eur J Environ Civil Eng 2022;26(2):
473–80.
[40] Wang J. Seismic design of tunnels: a state-of-the-art approach, monograph,
monograph 7. New York: Parsons, Brinckerhoff, Quade Douglas Inc; 1993.
[41] Miyazaki Y, Sawamura Y, Kishida K, Kimura M. Dynamic centrifuge model tests
on seismic performance in culvert longitudinal direction of hinge-type arch
culverts due to patterns of embankment shape. J Japan Soc Civil Eng, Ser C
(Geosphere Eng) 2017;73(4):429–41.
[42] Miyazaki Y, Sawamura Y, Kishida K, Kimura M. Dynamic behaviour of three-
hinge-type precast arch culverts with various patterns of overburden in culvert
longitudinal direction. Physical modelling in geotechnics, Volume 2: CRC Press;
2018. p. 915–20. 10.1201/9780429438646.
[43] Sawamura Y, Ishihara H, Kishida K, Kimura M. Experimental study on damage
morphology and critical state of three-hinge precast arch culvert through shaking
table tests. Proc Eng, Adv Transport Geotechnics 2016;III(143):522–9. https://
doi.org/10.1016/j.proeng.2016.06.066.
[44] Sawamura Y, Kishida K, Kimura M. Centrifuge model test and FEM analysis of
dynamic interactive behavior between embankments and installed culverts in
multiarch culvert embankments. Int J Geomech 2015;15:04014050. https://doi.
org/10.1061/(ASCE)GM.1943-5622.0000361.
[45] Byrne PM, Anderson DL, Jitno H. Seismic analysis of large buried culvert
structures. Transp Res Rec 1996;1541:133–9. https://doi.org/10.1177/
0361198196154100117.
[46] Miyazaki Y, Sawamura Y, Kishida K, Kimura M. Elasto-plastic 3D FE analysis of
the seismic behavior in culvert longitudinal direction of three-hinge type of
1229
Structures 39 (2022) 1218–1230
precast arch culverts. In: Advances in Computer Methods and Geomechanics.
Springer; 2020. p. 223–35. 10.1007/978-981-15-0886-8_18.
[47] Homaei F, Yazdani M. The probabilistic seismic assessment of aged concrete arch
bridges: the role of soil-structure interaction. Structures 2020;28:894–904.
https://doi.org/10.1016/j.istruc.2020.09.038.
[48] Abaqus V. 6.14 Documentation. Dassault Systemes Simulia Corporation; 2014.
[49] Liu J, Du Y, Du X, Wang Z, Wu J. 3D viscous-spring artificial boundary in time
domain. Earthquake Eng Eng Vibr 2006;5:93–102. 10.1007/s11803-006-0585-2.
[50] Zhang W, Seylabi EE, Taciroglu E. An ABAQUS toolbox for soil-structure
interaction analysis. Computers Geotechnics 2019;114:103143. https://doi.org/
10.1016/j.compgeo.2019.103143.
[51] Lee J, Fenves GL. A plastic-damage concrete model for earthquake analysis of
dams. Earthquake Eng Struct Dyn 1998;27:937–56. https://doi.org/10.1002/
(SICI)1096-9845(199809)27:9<937::AID-EQE764>3.0.CO;2-5.
[52] Lee J, Fenves GL. Plastic-damage model for cyclic loading of concrete structures.
J Eng Mech 1998;124:892–900. https://doi.org/10.1061/(ASCE)0733-9399
(1998)124:8(892).
[53] Hognestad E. Study of combined bending and axial load in reinforced concrete
members1951. http://hdl.handle.net/2142/4360.
[54] Yazdani M, Habibi H. Residual capacity evaluation of masonry arch bridges by
extended finite element method. Struct Eng Int 2021:1–12. https://doi.org/
10.1080/10168664.2021.1944454.
[55] Zhang S, Wang G, Sa W. Damage evaluation of concrete gravity dams under
mainshock–aftershock seismic sequences. Soil Dyn Earthquake Eng 2013;50:
16–27. https://doi.org/10.1016/j.soildyn.2013.02.021.
[56] Baker JW. Efficient analytical fragility function fitting using dynamic structural
analysis. Earthquake Spectra 2015;31:579–99. https://doi.org/10.1193/
021113EQS025M.
[57] Muntasir Billah A, Shahria AM. Seismic fragility assessment of multi-span
concrete highway bridges in British Columbia considering soil–structure
interaction. Can J Civ Eng 2020;48:39–51. https://doi.org/10.1139/cjce-2018-
0667.
[58] Le TS, Huh J, Park J-H. Earthquake fragility assessment of the underground
tunnel using an efficient SSI analysis approach. J Appl Mathem Phys 2014;2:
1073. https://doi.org/10.4236/jamp.2014.212123.
[59] Lee T-H, Park D, Nguyen DD, Park J-S. Damage analysis of cut-and-cover tunnel
structures under seismic loading. Bull Earthq Eng 2016;14:413–31. https://doi.
org/10.1007/s10518-015-9835-x.
[60] Argyroudis S, Pitilakis K. Seismic fragility curves of shallow tunnels in alluvial
deposits. Soil Dyn Earthquake Eng 2012;35:1–12. https://doi.org/10.1016/j.
soildyn.2011.11.004.
[61] Argyroudis S, Tsinidis G, Gatti F, Pitilakis K. Effects of SSI and lining corrosion on
the seismic vulnerability of shallow circular tunnels. Soil Dyn Earthquake Eng
2017;98:244–56. https://doi.org/10.1016/j.soildyn.2017.04.016.
[62] HAZUS-MH. Technical Manuals. Federal Emergency Management Agency and
National Institute of Building Science: Washington, DC, USA 2004.
[63] Pitilakis K. Fragility Function for Roadway System Elements. SYNER-G; 1995.
[64] ATC-40. Seismic Evaluation and Retrofit of Reinforced Concrete Buildings.
Applied Technology Council 1996.
[65] FEMA-356. Prestandard and Commentary for the Seismic Rehabilitation of
Buildings. Washington (DC)2000.
[66] Cornell CA, Jalayer F, Hamburger RO, Foutch DA. Probabilistic basis for 2000
SAC federal emergency management agency steel moment frame guidelines.
J Struct Eng 2002;128:526–33. https://doi.org/10.1061/(ASCE)0733-9445
(2002)128:4(526).
[67] Dutta A, Mander J. Seismic fragility analysis of highway bridges. Proceedings of
Proceedings of the INCEDE-MCEER center-to-center project workshop on
earthquake engineering Frontiers in transportation systems1998.
[68] Jernigan J, Hwang H. Development of bridge fragility curves. Proceedings of 7th
US National Conference on Earthquake Engineering. MA: EERI Boston; 2002.
[69] Kunnath SK, Larson L, Miranda E. Modelling considerations in probabilistic
performance-based seismic evaluation: case study of the I-880 viaduct.
Earthquake Eng Struct Dynmics 2006;35:57–75. https://doi.org/10.1002/
eqe.531.
[70] Muntasir Billah A, Shahria AM. Seismic fragility assessment of concrete bridge
pier reinforced with superelastic shape memory alloy. Earthquake Spectra 2015;
31:1515–41. https://doi.org/10.1193/112512EQS337M.
[71] Kappos AJ, Panagopoulos G, Panagiotopoulos C, Penelis G. A hybrid method for
the vulnerability assessment of R/C and URM buildings. Bull Earthq Eng 2006;4:
391–413. https://doi.org/10.1007/s10518-006-9023-0.
[72] Padgett JE, DesRoches R. Methodology for the development of analytical fragility
curves for retrofitted bridges. Earthquake Eng Struct Dyn 2008;37:1157–74.
https://doi.org/10.1002/eqe.801.
[73] Shinozuka M, Feng MQ, Lee J, Naganuma T. Statistical analysis of fragility curves.
J Eng Mech 2000;126:1224–31. https://doi.org/10.1061/(ASCE)0733-9399
(2000)126:12(1224).
[74] Salmon M, Wang J, Jones D, Wu C. Fragility formulations for the BART system.
Advancing mitigation technologies and disaster response for lifeline systems;
2003. p. 183–92. 10.1061/40687(2003)19.
[75] Tekie PB, Ellingwood BR. Seismic fragility assessment of concrete gravity dams.
Earthquake Eng Struct Dyn 2003;32:2221–40. https://doi.org/10.1002/eqe.325.
[76] Lee Y-J, Moon D-S. A new methodology of the development of seismic fragility
curves. Smart Struct Syst 2014;14(5):847–67.
[77] Zhao C, Yu N, Peng T. Probabilistic seismic fragility assessment of isolated
nuclear power plant structure using IDA and MSA methods. Structures 2021;34:
1300–11. https://doi.org/10.1016/j.istruc.2021.08.034.
V.-T. Nguyen et al.
[78] Pang R, Xu B, Zhou Y, Zhang X, Wang X. Fragility analysis of high CFRDs
subjected to mainshock-aftershock sequences based on plastic failure. Eng Struct
2020;206:110152. https://doi.org/10.1016/j.engstruct.2019.110152.
[79] Zhao C, Yu N, Peng T, Gautam A, Mo YL. Vulnerability assessment of AP1000 NPP
under mainshock-aftershock sequences. Eng Struct 2020;208:110348. https://
doi.org/10.1016/j.engstruct.2020.110348.
[80] Xu B, Pang R, Zhou Y. Verification of stochastic seismic analysis method and
seismic performance evaluation based on multi-indices for high CFRDs. Eng Geol
2020;264:105412. https://doi.org/10.1016/j.enggeo.2019.105412.
[81] Pang R, Xu B, Zhou Y, Song L. Seismic time-history response and system
reliability analysis of slopes considering uncertainty of multi-parameters and
earthquake excitations. Comput Geotech 2021;136:104245. https://doi.org/
10.1016/j.compgeo.2021.104245.
[82] Kassem MM, Nazri FM, Farsangi EN. The seismic vulnerability assessment
methodologies: A state-of-the-art review. Ain Shams Eng J 2020;11:849–64.
https://doi.org/10.1016/j.asej.2020.04.001.
[83] Zhong Z, Shen Y, Zhao M, Li L, Du X, Hao H. Seismic fragility assessment of the
Daikai subway station in layered soil. Soil Dyn Earthquake Eng 2020;132:106044.
https://doi.org/10.1016/j.soildyn.2020.106044.
[84] Zhang C, Zhao M, Zhong Z, Du X. Seismic intensity measures and fragility analysis
for subway stations subjected to near-fault ground motions with velocity pulses.
J Earthquake Eng 2021:1–27. https://doi.org/10.1080/
13632469.2021.1994056.
[85] Nguyen D-D, Thusa B, Han T-S, Lee T-H. Identifying significant earthquake
intensity measures for evaluating seismic damage and fragility of nuclear power
plant structures. Nucl Eng Technol 2020;52:192–205. https://doi.org/10.1016/j.
net.2019.06.013.
[86] Zhang WA. Quantitative seismic behavior assessment of buried structures
[Doctoral dissertation]. Los Angeles: University of California; 2019.
[87] Corigliano M, Lai CG, Barla G. Seismic vulnerability of rock tunnels using fragility
curves. Proceedings of Congress of the International Society for Rock Mechanics:
OnePetro. 2007.
[88] Shome N. Probabilistic seismic demand analysis of nonlinear structures. Stanford
University; 1999.
[89] KDS. KDS 17 10 00:2018 Earthquake-resistant design. Korea Construction
Standards Center; 2018.
[90] KDS. KDS 41 17 00:2019 Building Seismic Design Code. Korea Construction
Standards Center; 2019.
[91] Hu X, Zhou Z, Chen H, Ren Y. Seismic fragility analysis of tunnels with different
buried depths in a soft soil. Sustainability 2020;12:892. https://doi.org/10.3390/
su12030892.
[92] Chen Xu, Xiang N, Li J, Guan Z. Influence of near-fault pulse-like motion
characteristics on seismic performance of tall pier bridges with fragility analysis.
J Earthquake Eng 2022;26(4):2001–22.
[93] Bommer JJ, Acevedo AB. The use of real accelerograms as input to dynamic
analysis. J Earthquake Eng 2004;8:43–91. https://doi.org/10.1080/
13632460409350521.
[94] Richter CF. An instrumental earthquake magnitude scale. Bull Seismol Soc Am
1935;25:1–32. https://doi.org/10.1785/BSSA0250010001.
[95] Borcherdt RD. Estimates of site-dependent response spectra for design
(methodology and justification). Earthquake spectra 1994;10:617–53. https://
doi.org/10.1193/1.1585791.
1230
Structures 39 (2022) 1218–1230
[96] Hollender F, Cornou C, Dechamp A, Oghalaei K, Renalier F, Maufroy E, et al.
Characterization of site conditions (soil class, V S30, velocity profiles) for 33
stations from the French permanent accelerometric network (RAP) using surface-
wave methods. Bull Earthq Eng 2018;16(6):2337–65.
[97] Wang HY, Wang SY. A new method for estimating VS(30) from a shallow shear-
wave velocity profile (depth < 30 m). Bull Seismol Soc Am 2015;105:1359–70.
https://doi.org/10.1785/0120140103.
[98] Ancheta TD, Darragh RB, Stewart JP, Seyhan E, Silva WJ, Chiou B-J, et al. NGA-
West2 database. Earthquake Spectra Bull Seismol Soc Am 2014;30(3):989–1005.
[99] Goulet CA, Kishida T, Ancheta TD, Cramer CH, Darragh RB, Silva WJ, et al. PEER
NGA-East database. Earthquake Spectra 2021;37(1_suppl):1331–53.
[100] Boore DM. Orientation-independent, nongeometric-mean measures of seismic
intensity from two horizontal components of motion. Bull Seismol Soc Am 2010;
100:1830–5. https://doi.org/10.1785/0120090400.
[101] Kohrangi M, Bazzurro P, Vamvatsikos D. Conditional spectrum bidirectional
record selection for risk assessment of 3D structures using scalar and vector IMs.
Earthquake Eng Struct Dyn 2019;48:1066–82. https://doi.org/10.1002/
eqe.3177.
[102] Pehlivan M, Rathje EM, Gilbert RB. Factors influencing soil surface seismic hazard
curves. Soil Dyn Earthquake Eng 2016;83:180–90. https://doi.org/10.1016/j.
soildyn.2016.01.009.
[103] Trevlopoulos K, Zentner I. Seismic fragility curve assessment based on synthetic
ground motions with conditional spectra. Pure Appl Geophys 2020;177:2375–90.
https://doi.org/10.1007/s00024-019-02245-w.
[104] EUROCODE-8. Design of structures for earthquake resistance. 1998-1:2004.
European Committee for Standardiztion; c2004 299 p 2004.
[105] NBCC. National Building Code of Canada. National Research Council of Canada,
Ottawa, Ontario 2005.
[106] ASCE-7-16. Minimum Design Loads and Associated Critera for Buildings and
Other Structures. 7-16. USA. American Society of Civil Engineers; c2016 2016.
[107] BCJ-1997. Structural provisions for building structures. 1997 edition Tokyo:
Building Center of Japan. Japan 1997.
[108] GB-50011-2010. Code for Seismic Design of Buildings, Beijing. Ministry of
Housing and Urban-Rural Development of the People’s Republic of China; c2010
2010.
[109] AS-1170.4-2007. Structural design actions, part 4: Earthquake actions in
Australia. American Society of Civil Engineers, Los Angeles Section 2007.
[110] NZS-1170.5:2004. Structural design actions - Part 5: Earthquake actions - New
Zealand. New Zealand 2004.
[111] MOC-2008. Manual of Civil Structures. Mexico 2008.
[112] NEHRP. National Earthquake Hazard Reduction Program, (1994). Recommended
provisions for seismic regulations of new buildings: Part I, Provisions. FEMA
222A, Federal Emergency Management Agency, Washington, DC 1994.
[113] Baker JW, Cornell AC. A vector-valued ground motion intensity measure
consisting of spectral acceleration and epsilon. Earthquake Eng Struct Dyn 2005;
34:1193–217. https://doi.org/10.1002/eqe.474.
[114] Straub D, Kiureghian AD. Improved seismic fragility modeling from empirical
data. Struct Saf 2008;30:320–36. https://doi.org/10.1016/j.
strusafe.2007.05.004.
[115] Eads L, Miranda E, Krawinkler H, Lignos DG. An efficient method for estimating
the collapse risk of structures in seismic regions. Earthquake Eng Struct Dyn 2013;
42:25–41. https://doi.org/10.1002/eqe.2191.
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Lady
Chatterley's lover
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.

Title: Lady Chatterley's lover

Author: D. H. Lawrence

Release date: March 11, 2024 [eBook #73144]

Language: English

Original publication: Garden City, NY: Nelson Doubleday, Inc, 1928

Credits: Greg Weeks, Mary Meehan and the Online Distributed

Proofreading Team at http://www.pgdp.net (This file was
produced from images generously made available by The
Internet Archive)

*** START OF THE PROJECT GUTENBERG EBOOK LADY

CHATTERLEY'S LOVER ***
Lady Chatterley's Lover

By D. H. LAWRENCE

NELSON DOUBLEDAY, Inc.

Garden City, New York

The text is the third manuscript version,

published in Florence in 1928 by Orioli

Distributed in Canada by permission

of Laurence Pollinger Limited

Printed in the United States of America

 CONTENTS
CHAPTER I
CHAPTER II
CHAPTER III
CHAPTER IV
CHAPTER V
CHAPTER VI
CHAPTER VII
CHAPTER VIII
CHAPTER IX
CHAPTER X
CHAPTER XI
CHAPTER XII
CHAPTER XIII
CHAPTER XIV
CHAPTER XV
CHAPTER XVI
CHAPTER XVII
CHAPTER XVIII
CHAPTER XIX

CHAPTER I
Ours is essentially a tragic age, so we refuse to take it tragically. The
cataclysm has happened, we are among the ruins, we start to build
up new little habitats, to have new little hopes. It is rather hard work:
there is now no smooth road into the future: but we go round, or
scramble over the obstacles. We've got to live, no matter how many
skies have fallen.
This was more or less Constance Chatterley's position. The war had
brought the roof down over her head. And she had realised that one
must live and learn.
She married Clifford Chatterley in 1917, when he was home for a
month on leave. They had a month's honeymoon. Then he went
back to Flanders: to be shipped over to England again six months
later, more or less in bits. Constance, his wife, was then twenty-three
years old, and he was twenty-nine.
His hold on life was marvellous. He didn't die, and the bits seemed to
grow together again. For two years he remained in the doctor's
hands. Then he was pronounced a cure, and could return to life
again, with the lower half of his body, from the hips down, paralysed
for ever.
This was in 1920. They returned, Clifford and Constance, to his
home, Wragby Hall, the family "seat." His father had died, Clifford
was now a baronet, Sir Clifford, and Constance was Lady Chatterley.
They came to start housekeeping and married life in the rather
forlorn home of the Chatterleys on a rather inadequate income.
Clifford had a sister, but she had departed. Otherwise there were no
near relatives. The elder brother was dead in the war. Crippled for
ever, knowing he could never have any children, Clifford came home
to the smoky Midlands to keep the Chatterley name alive while he
could.
He was not really downcast. He could wheel himself about in a
wheeled chair, and he had a bath-chair with a small motor
attachment, so he could drive himself slowly round the garden and
into the fine melancholy park, of which he was really so proud,
though he pretended to be flippant about it.
Having suffered so much, the capacity for suffering had to some
extent left him. He remained strange and bright and cheerful, almost,
one might say, chirpy, with his ruddy, healthy-looking face, and his
pale-blue, challenging bright eyes. His shoulders were broad and
strong, his hands were very strong. He was expensively dressed,
and wore handsome neckties from Bond Street. Yet still in his face
one saw the watchful look, the slight vacancy of a cripple.
He had so very nearly lost his life, that what remained was
wonderfully precious to him. It was obvious in the anxious brightness
of his eyes, how proud he was, after the great shock, of being alive.
But he had been so much hurt that something inside him had
perished, some of his feelings had gone. There was a blank of
insentience.
Constance, his wife, was a ruddy, country-looking girl with soft brown
hair and sturdy body, and slow movements, full of unusual energy.
She had big, wondering eyes, and a soft mild voice, and seemed just
to have come from her native village. It was not so at all. Her father
was the once well-known R. A., old Sir Malcolm Reid. Her mother
had been one of the cultivated Fabians in the palmy, rather pre-
Raphaelite days. Between artists and cultured socialists, Constance
and her sister Hilda had what might be called an aesthetically
unconventional upbringing. They had been taken to Paris and
Florence and Rome to breathe in art, and they had been taken also
in the other direction, to the Hague and Berlin, to great Socialist
conventions, where the speakers spoke in every civilised tongue,
and no one was abashed.
The two girls, therefore, were from an early age not the least
daunted by either art or ideal politics. It was their natural
atmosphere. They were at once cosmopolitan and provincial, with
the cosmopolitan provincialism of art that goes with pure social
ideals.
They had been sent to Dresden at the age of fifteen, for music
among other things. And they had had a good time there. They lived
freely among the students, they argued with the men over
philosophical, sociological and artistic matters, they were just as
good as the men themselves: only better, since they were women.
And they tramped off to the forests with sturdy youths bearing
guitars, twang-twang! They sang the Wandervogel songs, and they
were free. Free! That was the great word. Out in the open world, out
in the forests of the morning, with lusty and splendid-throated young
fellows, free to do as they liked, and—above all—to say what they
liked. It was the talk that mattered supremely: the impassioned
interchange of talk. Love was only a minor accompaniment.
Both Hilda and Constance had had their tentative love affairs by the
time they were eighteen. The young men with whom they talked so
passionately and sang so lustily and camped under the trees in such
freedom wanted, of course, the love connection. The girls were
doubtful, but then the thing was so much talked about, it was
supposed to be so important. And the men were so humble and
craving. Why couldn't a girl be queenly, and give the gift of herself?
So they had given the gift of themselves, each to the youth with
whom she had the most subtle and intimate arguments. The
arguments, the discussions were the great thing: the love-making
and connection were only a sort of primitive reversion and a bit of an
anticlimax. One was less in love with the boy afterwards, and a little
inclined to hate him, as if he had trespassed on one's privacy and
inner freedom. For, of course, being a girl, one's whole dignity and
meaning in life consisted in the achievement of an absolute, a
perfect, a pure and noble freedom. What else did a girl's life mean?
To shake off the old and sordid connections and subjections.
And however one might sentimentalise it, this sex business was one
of the most ancient, sordid connections and subjections. Poets who
glorified it were mostly men. Women had always known there was
something better, something higher. And now they knew it more
definitely than ever. The beautiful pure freedom of a woman was
infinitely more wonderful than any sexual love. The only unfortunate
thing was that men lagged so far behind women in the matter. They
insisted on the sex thing like dogs.
And a woman had to yield. A man was like a child with his appetites.
A woman had to yield him what he wanted, or like a child he would
probably turn nasty and flounce away and spoil what was a very
pleasant connection. But a woman could yield to a man without
yielding her inner, free self. That the poets and talkers about sex did
not seem to have taken sufficiently into account. A woman could
take a man without really giving herself away. Certainly she could
take him without giving herself into his power. Rather she could use
this sex thing to have power over him. For she only had to hold
herself back in sexual intercourse, and let him finish and expend
himself without herself coming to the crisis: and then she could
prolong the connection and achieve her orgasm and her crisis while
he was merely her tool.
Both sisters had had their love experience by the time the war came,
and they were hurried home. Neither was ever in love with a young
man unless he and she were verbally very near: that is unless they
were profoundly interested, TALKING to one another. The amazing,
the profound, the unbelievable thrill there was in passionately talking
to some really clever young man by the hour, resuming day after day
for months ... this they had never realised till it happened! The
paradisal promise: Thou shalt have men to talk to!—had never been
uttered. It was fulfilled before they knew what a promise it was.
And if after the roused intimacy of these vivid and soul-enlightened
discussions the sex thing became more or less inevitable, then let it.
It marked the end of a chapter. It had a thrill of its own too: a queer
vibrating thrill inside the body, a final spasm of self-assertion, like the
last word, exciting, and very like the row of asterisks that can be put
to show the end of a paragraph, and a break in the theme.
When the girls came home for the summer holidays of 1913, when
Hilda was twenty and Connie eighteen, their father could see plainly
that they had had the love experience.
L'amour avait passé par là, as somebody puts it. But he was a man
of experience himself, and let life take its course. As for the mother,
a nervous invalid in the last few months of her life, she only wanted
her girls to be "free," and to "fulfil themselves." She herself had
never been able to be altogether herself: it had been denied her.
Heaven knows why, for she was a woman who had her own income
and her own way. She blamed her husband. But as a matter of fact,
it was some old impression of authority on her own mind or soul that
she could not get rid of. It had nothing to do with Sir Malcolm, who
left his nervously hostile, high-spirited wife to rule her own roost,
while he went his own way.
So the girls were "free," and went back to Dresden, and their music,
and the university and the young men. They loved their respective
young men, and their respective young men loved them with all the
passion of mental attraction. All the wonderful things the young men
thought and expressed and wrote, they thought and expressed and
wrote for the young women. Connie's young man was musical,
Hilda's was technical. But they simply lived for their young women. In
their minds and their mental excitements, that is. Somewhere else
they were a little rebuffed, though they did not know it.
It was obvious in them too that love had gone through them: that is,
the physical experience. It is curious what a subtle but unmistakable
transmutation it makes, both in the body of men and women: the
woman more blooming, more subtly rounded, her young angularities
softened, and her expression either anxious or triumphant: the man
much quieter, more inward, the very shapes of his shoulders and his
buttocks less assertive, more hesitant.
In the actual sex-thrill within the body, the sisters nearly succumbed
to the strange male power. But quickly they recovered themselves,
took the sex-thrill as a sensation, and remained free. Whereas the
men, in gratitude to the women for the sex experience, let their souls
go out to her. And afterwards looked rather as if they had lost a
shilling and found sixpence. Connie's man could be a bit sulky, and
Hilda's a bit jeering. But that is how men are! Ungrateful and never
satisfied. When you don't have them they hate you because you
won't; and when you do have them they hate you again, for some
other reason. Or for no reason at all, except that they are
discontented children, and can't be satisfied whatever they get, let a
woman do what she may.
However, came the war, Hilda and Connie were rushed home again
after having been home already in May, to their mother's funeral.
Before Christmas of 1914 both their German young men were dead:
whereupon the sisters wept, and loved the young men passionately,
but underneath forgot them. They didn't exist any more.
Both sisters lived in their father's, really their mother's Kensington
house, and mixed with the young Cambridge group, the group that
stood for "freedom" and flannel trousers, and flannel shirts open at
the neck, and a well-bred sort of emotional anarchy, and a
whispering, murmuring sort of voice, and an ultra-sensitive sort of
manner. Hilda, however, suddenly married a man ten years older
than herself, an elder member of the same Cambridge group, a man
with a fair amount of money, and a comfortable family job in the
government: he also wrote philosophical essays. She lived with him
in a smallish house in Westminster, and moved in that good sort of
society of people in the government who are not tip-toppers, but who
are, or would be, the real intelligent power in the nation: people who
know what they're talking about, or talk as if they did.
Connie did a mild form of war-work, and consorted with the flannel-
trousers Cambridge intransigeants, who gently mocked at
everything, so far. Her "friend" was a Clifford Chatterley, a young
man of twenty-two, who had hurried home from Bonn, where he was
studying the technicalities of coal-mining. He had previously spent
two years at Cambridge. Now he had become a first lieutenant in a
smart regiment, so he could mock at everything more becomingly in
uniform.
Clifford Chatterley was more upper-class than Connie. Connie was
well-to-do intelligentsia, but he was aristocracy. Not the big sort, but
still it. His father was a baronet, and his mother had been a
viscount's daughter.
But Clifford, while he was better bred than Connie, and more
"society," was in his own way more provincial and more timid. He
was at his ease in the narrow "great world," that is, landed
aristocracy society, but he was shy and nervous of all that other big
world which consists of the vast hordes of the middle and lower
classes, and foreigners. If the truth must be told, he was just a little
bit frightened of middle and lower class humanity, and of foreigners
not of his own class. He was, in some paralysing way, conscious of
his own defencelessness, though he had all the defence of privilege.
Which is curious, but a phenomenon of our day.
Therefore the peculiar soft assurance of a girl like Constance Reid
fascinated him. She was so much more mistress of herself in that
outer world of chaos than he was master of himself.
Nevertheless he too was a rebel: rebelling even against his class. Or
perhaps rebel is too strong a word; far too strong. He was only
caught in the general, popular recoil of the young against convention
and against any sort of real authority. Fathers were ridiculous: his
own obstinate one supremely so. And governments were ridiculous:
our own wait-and-see sort especially so. And armies were ridiculous,
and old duffers of generals altogether, the red-faced Kitchener
supremely. Even the war was ridiculous, though it did kill rather a lot
of people.
In fact everything was a little ridiculous, or very ridiculous: certainly
everything connected with authority, whether it were in the army or
the government or the universities, was ridiculous to a degree. And
as far as the governing class made any pretensions to govern, they
were ridiculous too. Sir Geoffrey, Clifford's father, was intensely
ridiculous, chopping down his trees, and weeding men out of his
colliery to shove them into the war; and himself being so safe and
patriotic; but also, spending more money on his country than he'd
got.
When Miss Chatterley—Emma—came down to London from the
Midlands to do some nursing work, she was very witty in a quiet way
about Sir Geoffrey and his determined patriotism. Herbert, the elder
brother and heir, laughed outright, though it was his trees that were
falling for trench props. But Clifford only smiled a little uneasily.
Everything was ridiculous, quite true. But when it came too close and
oneself became ridiculous too...? At least people of a different class,
like Connie, were earnest about something. They believed in
something.
They were rather earnest about the Tommies, and the threat of
conscription, and the shortage of sugar and toffee for the children. In
all these things, of course, the authorities were ridiculously at fault.
But Clifford could not take it to heart. To him the authorities were
ridiculous ab ovo, not because of toffee or Tommies.
And the authorities felt ridiculous, and behaved in a rather ridiculous
fashion, and it was all a mad hatter's tea party for a while. Till things
developed over there, and Lloyd George came to save the situation
over here. And this surpassed even ridicule, the flippant young
laughed no more.
In 1916 Herbert Chatterley was killed, so Clifford became heir. He
was terrified even of this. His importance as son of Sir Geoffrey and
child of Wragby was so ingrained in him, he could never escape it.
And yet he knew that this too, in the eyes of the vast seething world,
was ridiculous. Now he was heir and responsible for Wragby. Was
that not terrible? And also splendid at the same time, perhaps, purely
absurd?
Sir Geoffrey would have none of the absurdity. He was pale and
tense, withdrawn into himself, and obstinately determined to save his
country and his own position, let it be Lloyd George or who it might.
So cut off he was, so divorced from the England that was really
England, so utterly incapable, that he even thought well of Horatio
Bottomley. Sir Geoffrey stood for England and Lloyd George as his
forebears had stood for England and St. George: and he never knew
there was a difference. So Sir Geoffrey felled timber and stood for
Lloyd George and England, England and Lloyd George.
And he wanted Clifford to marry and produce an heir. Clifford felt his
father was a hopeless anachronism. But wherein was he himself any
further ahead, except in a wincing sense of the ridiculousness of
everything, and the paramount ridiculousness of his own position.
For willy-nilly he took his baronetcy and Wragby with the last
seriousness.
The gay excitement had gone out of the war ... dead. Too much
death and horror. A man needed support and comfort. A man
needed to have an anchor in the safe world. A man needed a wife.
The Chatterleys, two brothers and a sister, had lived curiously
isolated, shut in with one another at Wragby, in spite of all their
connections. A sense of isolation intensified the family tie, a sense of
the weakness of their position, a sense of defencelessness, in spite
of, or because of the title and the land. They were cut off from those
industrial Midlands in which they passed their lives. And they were
cut off from their own class by the brooding, obstinate, shut-up
nature of Sir Geoffrey, their father, whom they ridiculed, but whom
they were so sensitive about.
The three had said they would all live together always. But now
Herbert was dead, and Sir Geoffrey wanted Clifford to marry. Sir
Geoffrey barely mentioned it: he spoke very little. But his silent,
brooding insistence that it should be so was hard for Clifford to bear
up against.
But Emma said No! She was ten years older than Clifford, and she
felt his marrying would be a desertion and a betrayal of what the
young ones of the family had stood for.
Clifford married Connie, nevertheless, and had his month's
honeymoon with her. It was the terrible year 1917, and they were
intimate as two people who stand together on a sinking ship. He had
been virgin when he married: and the sex part did not mean much to
him. They were so close, he and she, apart from that. And Connie
exulted a little in this intimacy which was beyond sex, and beyond a
man's "satisfaction." Clifford anyhow was not just keen on his
"satisfaction," as so many men seemed to be. No, the intimacy was
deeper, more personal than that. And sex was merely an accident, or
an adjunct: one of the curious obsolete, organic processes which
persisted in its own clumsiness, but was not really necessary.
Though Connie did want children: if only to fortify her against her
sister-in-law Emma.
But early in 1918 Clifford was shipped home smashed, and there
was no child. And Sir Geoffrey died of chagrin.

CHAPTER II
Connie and Clifford came home to Wragby in the autumn of 1920.
Miss Chatterley, still disgusted at her brother's defection, had
departed and was living in a little flat in London.
Wragby was a long low old house in brown stone, begun about the
middle of the eighteenth century, and added on to, till it was a warren
of a place without much distinction. It stood on an eminence in a
rather fine old park of oak trees, but alas, one could see in the near
distance the chimney of Tevershall pit, with its clouds of steam and
smoke, and on the damp, hazy distance of the hill the raw straggle of
Tevershall village, a village which began almost at the park gates,
and trailed in utter hopeless ugliness for a long and gruesome mile:
houses, rows of wretched, small, begrimed, brick houses, with black
slate roofs for lids, sharp angles and wilful, blank dreariness.
Connie was accustomed to Kensington or the Scotch hills or the
Sussex downs: that was her England. With the stoicism of the young
she took in the utter, soulless ugliness of the coal-and-iron Midlands
at a glance, and left it at what it was: unbelievable and not to be
thought about. From the rather dismal rooms at Wragby she heard
the rattle-rattle of the screens at the pit, the puff of the winding-
engine, the clink-clink of shunting trucks, and the hoarse little whistle
of the colliery locomotives. Tevershall pit-bank was burning, had
been burning for years, and it would cost thousands to put it out. So
it had to burn. And when the wind was that way, which was often, the
house was full of the stench of this sulphureous combustion of the
earth's excrement. But even on windless days the air always smelt of
something under-earth: sulphur, iron, coal, or acid. And even on the
Christmas roses the smuts settled persistently, incredible, like black
manna from skies of doom.
Well, there it was: fated like the rest of things! It was rather awful, but
why kick? You couldn't kick it away. It just went on. Life, like all the
rest! On the low dark ceiling of cloud at night red blotches burned
and quavered, dappling and swelling and contracting, like burns that
give pain. It was the furnaces. At first they fascinated Connie with a
sort of horror; she felt she was living underground. Then she got
used to them. And in the morning it rained.
Clifford professed to like Wragby better than London. This country
had a grim will of its own, and the people had guts. Connie
wondered what else they had: certainly neither eyes nor minds. The
people were as haggard, shapeless, and dreary as the countryside,
and as unfriendly. Only there was something in their deep-mouthed
slurring of the dialect, and the thresh-thresh of their hobnailed pit-
boots as they trailed home in gangs on the asphalt from work, that
was terrible and a bit mysterious.
There had been no welcome home for the young squire, no
festivities, no deputation, not even a single flower. Only a dank ride
in a motorcar up a dark, damp drive, burrowing through gloomy
trees, out to the slope of the park where grey damp sheep were
feeding, to the knoll where the house spread its dark brown façade,
and the housekeeper and her husband were hovering, like unsure
tenants on the face of the earth, ready to stammer a welcome.
There was no communication between Wragby Hall and Tevershall
village, none. No caps were touched, no curtseys bobbed. The
colliers merely stared; the tradesmen lifted their caps to Connie as to
an acquaintance, and nodded awkwardly to Clifford; that was all.
Gulf impassable, and a quiet sort of resentment on either side. At
first Connie suffered from the steady drizzle of resentment that came
from the village. Then she hardened herself to it, and it became a
sort of tonic, something to live up to. It was not that she and Clifford
were unpopular, they merely belonged to another species altogether
from the colliers. Gulf impassable, breach indescribable, such as is
perhaps non-existent south of the Trent. But in the Midlands and the
industrial North gulf impassable, across which no communication
could take place. You stick to your side, I'll stick to mine! A strange
denial of the common pulse of humanity.
Yet the village sympathised with Clifford and Connie in the abstract.
In the flesh it was—You leave me alone!—on either side.
The rector was a nice man of about sixty, full of his duty, and
reduced, personally, almost to a nonentity by the silent—You leave
me alone!—of the village. The miners' wives were nearly all
Methodists. The miners were nothing. But even so much official
uniform as the clergyman wore was enough to obscure entirely the
fact that he was a man like any other man. No, he was Mester
Ashby, a sort of automatic preaching and praying concern.
This stubborn, instinctive—We think ourselves as good as you, if you
are Lady Chatterley!—puzzled and baffled Connie at first extremely.
The curious, suspicious, false amiability with which the miners' wives
met her overtures; the curiously offensive tinge of—Oh dear me! I
am somebody now, with Lady Chatterley talking to me! But she
needn't think I'm not as good as her for all that!—which she always
heard twanging in the women's half-fawning voices, was impossible.
There was no getting past it. It was hopelessly and offensively
nonconformist.
Clifford left them alone, and she learnt to do the same: she just went
by without looking at them, and they stared as if she were a walking
wax figure. When he had to deal with them, Clifford was rather
haughty and contemptuous; one could no longer afford to be friendly.
In fact he was altogether rather supercilious and contemptuous of
anyone not in his own class. He stood his ground, without any
attempt at conciliation. And he was neither liked nor disliked by the
people: he was just part of things, like the pit-bank and Wragby itself.
But Clifford was really extremely shy and self-conscious now he was
lamed. He hated seeing anyone except just the personal servants.
For he had to sit in a wheeled chair or a sort of bath-chair.
Nevertheless he was just as carefully dressed as ever, by his
expensive tailors, and he wore the careful Bond Street neckties just
as before, and from the top he looked just as smart and impressive
as ever. He had never been one of the modern lady-like young men:
rather bucolic even, with his ruddy face and broad shoulders. But his
very quiet, hesitating voice, and his eyes, at the same time bold and
frightened, assured and uncertain, revealed his nature. His manner
was often offensively supercilious, and then again modest and self-
effacing, almost tremulous.
Connie and he were attached to one another, in the aloof modern
way. He was much too hurt in himself, the great shock of his
maiming, to be easy and flippant. He was a hurt thing. And as such
Connie stuck to him passionately.
But she could not help feeling how little connection he really had with
people. The miners were, in a sense, his own men; but he saw them
as objects rather than men, parts of the pit rather than parts of life,
crude raw phenomena rather than human beings along with him. He
was in some way afraid of them, he could not bear to have them look
at him now he was lame. And their queer, crude life seemed as
unnatural as that of hedgehogs.
He was remotely interested; but like a man looking down a
microscope, or up a telescope. He was not in touch. He was not in
actual touch with anybody, save, traditionally, with Wragby, and,
through the close bond of family defence, with Emma. Beyond this
nothing really touched him. Connie felt that she herself didn't really,
not really touch him; perhaps there was nothing to get at ultimately;
just a negation of human contact.
Yet he was absolutely dependent on her, he needed her every
moment. Big and strong as he was, he was helpless. He could wheel
himself about in a wheeled chair, and he had a sort of bath-chair with
a motor attachment, in which he could puff slowly round the park.
But alone he was like a lost thing. He needed Connie to be there, to
assure him he existed at all.
Still he was ambitious. He had taken to writing stories; curious, very
personal stories about people he had known. Clever, rather spiteful,
and yet, in some mysterious way, meaningless. The observation was
extraordinary and peculiar. But there was no touch, no actual
contact. It was as if the whole thing took place in a vacuum. And
since the field of life is largely an artificially-lighted stage today, the
stories were curiously true to modern life, to the modern psychology,
that is.
Clifford was almost morbidly sensitive about these stories. He
wanted everyone to think them good, of the best, ne plus ultra. They
appeared in the most modern magazines, and were praised and
blamed as usual. But to Clifford the blame was torture, like knives
goading him. It was as if the whole of his being were in his stories.
Connie helped him as much as she could. At first she was thrilled.
He talked every thing over with her monotonously, insistently,
persistently, and she had to respond with all her might. It was as if