Proceedings of the 3rd International Conference on Seismic Retrofitting, Tabriz, Iran, 20-22 October 2010

Seismic Performance of Bridges: Lessons Learned From the Past Earthquakes

Kamran Amini1,Kianoush Taghikhani2, Amir Yavari1

Concrete Research Centre, Department of Civil Engineering, Islamic Azad University, Qazvin1
Concrete Research Centre, Department of Civil Engineering, Imam Khomeini International University, Qazvin2

Correspondent Author: Kamran Amini

k.amini88@yahoo.com

ABSTRACT
Lifeline systems consist of Bridges, Roads, Tunnels, Communication networks, Water and Gas Transmissions,
Airports, etc, are known as a general parameter for all cities especially Megacities survival. Any trouble in their
operation particularly in times of crisis (earthquake, flood, tempest, war, etc) will increase irreparable losses of life and
property. Large lifeline systems such as bridges are susceptible to earthquake damages, and their performance, integrity
and stability according to their special application in current-day transportation, during and after an earthquake have
significant implications on public safety. Bridges are one of the most importance lines to control rescue operations
during and after earthquake. So, there is an urgent need to recognition the effective factors on bridge damages in an
unforeseen exigency. Hence, their vulnerability significantly should be under continual surveillance and bridges
stability has to be controlled by retrofitting and rehabilitating. In this study the seismic performance of various types of
bridges during the past earthquakes is presented. The objective of this study is to present typical failure modes of
bridges and providing some clues in order to reduce the seismic vulnerability of such structures. Results of this study
demonstrated that there are some typical failure modes in bridges. Hence, by considering some special practical
consideration the seismic vulnerability of bridges may decrease.

Keywords: vulnerability, earthquake, structural properties, retrofitting.

1 INTRODUCTION

A bridge is a structure built to span a valley, road, railroad track, river, body of water, or
many other physical obstacles, for the purpose of providing passage over the obstacles. Design of
bridges will vary depending on the function of the bridge, nature of the terrain and the seismic
behavior of the zone where the bridge is to be constructed. When an extreme earthquake occurs, it
is essential to detect the severe structural damage that affect the structural safety immediately [1].
The seismic effects at a given site can include a variety of hazards, such as ground vibrations,
lateral ground displacements, compaction of fills and underlying sediments, soil fractures,
liquefaction, liquefaction-induced ground movements, slope failures, landslides and tsunamis.
These effects are influenced by the nature of the fault rupture process, the travel path followed by
the resulting seismic waves as they propagate from the ruptured fault to the site, the distance from
the fault and the local soil conditions at the site [1].
Clearly, the collapse of a bridge places people on or below the bridge at risk, and it must be
replaced after the earthquake unless alternative transportation paths are identified. Earthquake
damage to a bridge can have severe consequences [2]. In the immediate aftermath of an earthquake,
closure of a bridge can impair emergency response operations. Later, the economic impact of a
bridge closure increases with the length of time the bridge is closed, the economic importance of the
traffic using the route, the traffic delay caused by following alternate routes, and the replacement
cost for the bridge [3]. Studies concerning bridges response to earthquakes have noticeably
developed in the 90s, especially in Japan and in the United States. Analyses accuracy has certainly
benefited from the large amount of data that have been systematically gathered in recent disasters,
which affected densely populated metropolitan areas [4–9].
By the destructive damage occurred in 1923 Kanto Earthquake the importance of taking
account of the seismic effects in design of engineering structures has been recognized for the first
time. At those days, bridges were constructed based on the technologies imported from the USA,
UK, German and French with no consideration for the effects of seismic disturbances. Besides more
technical aspects referring to physical features of nodes and networks, a greater attention has been
paid to Seismic countermeasures were initiated after the 1923 Kanto Earthquake. The equivalent
static lateral force method using a seismic coefficient of 0.1-0.3 based on the allowable stress
design approach, Seismic Coefficient Method, was first incorporated in design of highway bridges
in 1927 (MI 1927). Since that time, the seismic design practice and also seismic retrofitting has
been improved and extended based on the seismic damage and the progress of researches. However
it was still preliminary stage for assuring the seismic performance [10, 11]. This study aimed at
recognition the failure modes of bridges to provide some seismic retrofitting solution to reduce such
failures in the future.

2 SEISMIC VULNERABILITY OF BRIDGES

Bridges are one the most important elements in transportation networks which fail and collapse
in most earthquakes. Depending on the ground motion, site conditions, overall configuration, and
specific details of the bridge, the damage induced in a particular bridge can take many forms.
Though exceptions abound, most of the severe damage to bridges has taken one of the following
forms.

2.1. Unseating of superstructure at in-span hinges or simple supports attributable to

inadequate seat lengths or restraint

Damage within the superstructure is rarely the primary cause of collapse. The presence of a
skewed or curved configuration further exacerbates the vulnerability. For simply supported bridges,
these failures are most likely when ground failure induces relative motion between the spans and
their supports. In many areas of the world especially in much of the United States, bridges often
comprise a series of simple spans supported on bents.

Fig 1. Collapse of Wu-Shi Bridge, Chi Chi Earthquake., collapse of Claro River Bridge near, Chile
earthquake

These spans are prone to being toppled from their supporting substructures either due to shaking or
differential support movement associated with ground deformation. Unseating of simple spans was
observed in California in earlier earthquakes, leading in recent decades to development of bridge
construction practices based on monolithic box-girder substructure construction. Problems of
unseating still occur with older bridge construction and with new bridges in regions where simple
spans are still common. As happened in1991, during the Costa Rica earthquake, widespread
liquefaction led to abutment and internal bent rotations, resulting in the collapse of no fewer than
four bridges with simple supports [7]. The collapse of the Wu-Shi Bridge in the 1999 Chi Chi
Earthquake and collapse of Claro River Bridge Chile earthquake demonstrates some results of the
unseating of simple spans (Fig 1). Because superstructures are designed for dead load and traffic
load with large safety factor, their direct damage resulted from the seismic effect is limited.
However the members in the vicinity of bearing and the end of the superstructures are vulnerable,
retrofit is often conducted.
Most common retrofit for superstructures is installation of unseating prevention devices.
The unseating prevention devices were first developed after the 1964 Niigata earthquake. Various
unseating prevention devices are now used worldwide. They may be classified into four groups; (1)
extension of seat length, (2) connection of decks to substructure, (3) connection between adjacent
decks at the edge, and (4) connection of decks at intermediate hinge. There are connectors which
resist only tension such as cable restrainers, and connectors which resist both tension and
compression. Figure 2 shows typical unseating prevention devices.

Fig 2. Unseating Prevention Devices

2.2 Column failure

There are two fundamental decisions at the start of a bridge seismic retrofit. The first is whether
the evaluated risk of failure warrants retrofit. Because retrofitting has various limitations in design
and implementation, it is sometimes technically and economically unfeasible to retrofit a bridge.
Replacement of the bridge with a new bridge can be only a solution for reducing the risk of failure
in such a case. Such failures are attributable to inadequate ductility (toughness).

2.2.1 Premature shear failure of reinforced concrete columns

In reinforced-concrete columns, the inadequate ductility usually stems from inadequate

confinement reinforcement. Premature shear failure at terminations of longitudinal bars with
insufficient development lengths has resulted in collapse of many bridges. Eighteen spans of Fukae
Viaduct, Hanshin Expressway collapsed as shown in Figure 3. The viaduct was designed in
accordance with the 1964 Design Specifications using 0.2 horizontal and 0.1 vertical seismic
coefficients based on the allowable stress design approach.

Fig 3. Collapse of Fukae Viaduct, Hyogo-ken Nanbu 1995

As shown in figure 4 on 1999 occurrence the earthquakes have resulted in destructive damages to
the bridge’s piers in Taiwan 921 Earthquake (chi chi earthquake). Reinforced concrete columns
which were designed in accordance with the practice which did not take account of the importance
of plastic deformation and ductility capacity are commonly deficient in flexural ductility, shear
strength, and flexural strength under strong seismic excitation. Lap splices in critical regions and
premature termination of longitudinal reinforcement are the common practice which affects the
deficiency. A number of column retrofit techniques have been developed and clarified based on
loading tests.

Fig 4. Premature shear failure of reinforced concrete columns; Wushi Bridge, Taiwan 1999
Columns retrofit techniques include steel jacketing, active confinement by wire prestressing, use of
composite materials jacketing involving fiberglass, carbon fiber and other fibers, and jacketing with
reinforced concrete. Of these, the most common retrofit technique implemented to date has been
steel jacketing, with lesser amount of retrofit involving reinforced concrete jackets and composite
materials jackets. Retrofit with emphasis on there two approaches are introduced in the following.

2.2.1.1 Steel jacket for circular columns

Two half shells of steel plate rolled to a radius of 12.5 to 25 mm larger than the column
radius are positioned over the area to be retrofitted and are site-welded up the vertical seams to
provide a continuous tube with a small annular gap around the column as shown in Fig. 5(a) [12].
The gap is grouted with a pure cement grout. Generally a space of about 50 mm is provided
between the jacket and footing or cap beam to avoid excessive flexural strength enhancement of the
plastic hinge.

Fig 5. Confinement of Columns by Steel Jacket

2.2.1.2 Steel jacket for rectangular columns

Because lateral confinement is not provided for rectangular columns, various trials were
conducted [13]. Figure 6 shows the elliptical jacket effect on lap splices in critical regions.

Fig 6. Retrofit for Rectangular Columns

2.2.1.3 Precast Concrete Segment Jacket and Cable-Strand Jacket

Because steel jacket is vulnerable to corrosion and impact with floating materials, it is not
generally used for columns in river, lake and seas. Because construction is faster, precast concrete
segment jacket is also used. Fig 7 shows an example of precast concrete segment jacketing.

Fig 7. Precast Concrete Segment Jacket

2.2.1.4 Composite Material jackets

Composite-materials such as carbon-fiber-sheet (CFS) jackets, aramid fiber reinforced

plastics jackets, high-strength fiberglass jackets in various shapes and forms are used for retrofit of
reinforced concrete columns [12,14]. Composite materials are generally expensive, but the cost of
material is only a pat of the retrofit. Because composite materials are generally light and easy to
handle, it is useful at the sites where construction space for steel jacketing is limited.

2.2.2 Collapse of steel columns

In steel columns, the inadequate ductility usually stems from local buckling, which
progresses to collapse. Collapse of steel bridges’ columns has occurs in Fukae viaduct, Hyogo-ken
Nanbu 1995 (Fig 8), for the first time caused by Extensive soil liquefaction and has resulted in
settlements and tilting of foundations and substructures. Lateral spreading of ground associated
with soil liquefaction caused movements of foundations. Unseating prevention device suffered
damage.

Fig 8. Collapse of Takashio Viaduct, 1995 Kobe Earthquake

Under a strong excitation, local buckling of web and flange plates as well as rupture of welded
corner occurred at the bottom of column. This decreased the bearing capacity of the column in both
lateral and vertical directions, which has progressed the settle down of the column due to dead
weight of the decks. This caused buckling to occur at both sides of the center deck, and this further
progressed the buckling of the column. Hence, some Current seismic retrofitting methods for steel
bridge piers are as follows. Design specifications concerning the seismic retrofitting of existing
steel bridge piers have been drafted by HEPC (HEPC,1998) and etc. In the specification, two
seismic retrofitting methods for existing steel columns are prepared on condition that the foundation
and the anchor bolts of the columns under consideration never damage, as follows; (1) Filling steel
columns with additional concrete over the existing encased concrete and (2) Retrofitting the
component stiffened plate panels of steel columns by using additional steel members.

Fig 9. Damage to the lower part of steel column; (a) Local buckling, (b) Fracture or Elongation of
anchor bolts

First of all, in the seismic retrofitting design, the adoption of the former method is examined
from economical and practical aspects through the comparison of the load carrying capacities of the
steel column and the anchor bolts. The steel column filled with concrete can show great
improvements both in the ductility and ultimate strength capacities, but the location of damage will
shift toward the lower parts of it as illustrated in Figure 9. In case that the concrete filling method
cannot be adopted, that is, the ultimate strength capacity of the steel column filled with concrete
becomes larger than that of the anchor bolts, the latter method is adopted to avoid damage to the
lower part of the steel column and expensive consequent retrofitting of the foundation. In
retrofitting of existing steel bridge piers, therefore, it is preferable to avoid increasing in their
ultimate strength capacity.

2.3 Damage to shear keys at abutments

Shear keys are components that restrain relative displacements (usually in the transverse
direction) between the superstructure and the abutments. External shear keys are located outside of
the superstructure cross section, while internal shear keys are located within the superstructure cross
section.

Fig 10. (a) Damage to external shear key in an abutment in the 1994 Northridge earthquake
; (b) Damage to internal shear key in an abutment in the 1994 Northridge earthquake.

Since these elements are stocky, it is nearly impossible to make them ductile, and they will fail if
their strength is exceeded. Shear key failures were widespread during the 1994 Northridge
earthquake [8]. Figure 10(a) shows a typical failure in which the external shear keys failed and
Figure 10(b) shows a failed internal shear key. It appears that these failures can occur with small
transverse displacement and little energy dissipation. Damage to internal shear keys usually is
accompanied by damage to the interlocking backwall. In seat-type abutments, damage has also
occurred in seat abutments due to pounding of backwalls by the superstructure. In monolithic
abutments, the superstructure is cast monolithically with the abutments. This configuration is
attractive, because it reduces the likelihood of span unseating. However, the abutment can be
damaged as the superstructure displaces in the longitudinal direction away from the abutment. Also,
depending on the geometry and details of the abutment, the wingwall may serve as an external shear
key. In such cases, the wingwall can fail in the same manner as an external shear key. Because
abutments are subjected to the soil-pressure from back-fill side to the front side, abutments are
likely to mobilize in the front side to result in a residual displacement after an earthquake.
Essentially there are two types of abutment movement as shown in Fig 11.

Fig 11. Damage of abutments; (a) Tilt in the river side, (b) Tilt in the back-filled side to instability
of ground

One is movement of an abutment at the top, thus resulting in a tilt in the front side. This failure is
typically resulted from the dynamic earth-pressure acting the abutment from back-fill. Because a
deck resists the abutment movement by a strut action, unseating of the deck does not occur by this
failure. However bearings, expansion joints and girders at the contact face as well as parapet wall of
the abutment would suffer extensive damage. In particular, because parapet wall is generally very
lightly reinforced, it suffers various levels of cracks, and sometimes it breaks into several pieces.
Similar damage occurs when the deck oscillation becomes excessively large to result in pounding
with the parapet wall.
The other is the movement of an abutment under footing which results in a tilt of the
abutment to the back-fill side. This damage is likely developed when sliding of the back-fill as well
as underlying ground occurs resulted from instability of underlying ground. Liquefaction of sandy
soil or sliding of soft clayey soils is a typical cause of the instability of underlying ground. Bearings
would suffer extensive damage.

Fig 12. Seismic retrofit of abutments using EPS

If the tilt of the abutment is large, unseating of the deck from the abutment occurs. In the past, the
first type damage was dominant with several examples of the second damage. Because damage
resulted from failure of abutment is generally restricted to the edges of the girder, expansion joints,
bearings, and parapet wall, it has been regarded as a secondary damage compared to damage of
columns and piers, and deck. However once abutments suffer damage, function of the bridge to
traffic would be lost and it takes long period for repair. Similar to retaining wall, the most essential
retrofit for abutments is to enhance the strength of abutments. Retrofit has been conducted by
enlarging the footing, and increasing number of piles similar to foundations. Increase of soil
strength by soil replacement or cement mixing is also a widely used for tilt resulted from instability
of underlying ground. For the tilt of abutments in the front direction, earth-anchors are frequently
used. A unique retrofit for tilt of abutments in the front direction is to replace some part of the back-
fills with expanded polystyrene (EPS) as shown in Fig 12. Because use of EPS reduces the dynamic
earth-pressure, damage of abutments during an earthquake can be mitigated. At an example, 1.75 m
high and 13 m long EPS replaced the back fill behind the abutment to decrease the dynamic earth-
pressure.

3 CONCLUSION
Survey of the damages is expected to give valuable lessons for better bridge structures in the
future that are capable of minimizing earthquake induced damages. In association with this issue
following cases are concluded, generally:

1. Damage to a bridge can have severe consequences for a local economy, because bridges
provide vital links in the transportation system of a region. In general, the likelihood of
damage increases if the ground motion is particularly intense, the soils are soft and the
bridge was constructed before modern codes were implemented, or the bridge configuration
is irregular. Even a well-designed bridge can suffer damage if nonstructural modifications
and structural deterioration have increased the vulnerability of the bridge.
2. According to the general condition of bridges such as site conditions, overall configuration,
specific details of the bridge and etc, the damage induced in a particular bridge can take
many forms. So, most of the severe damage to bridges are classified as follows:
- Unseating of superstructure at in-span hinges or simple supports attributable to
inadequate seat lengths or restraint.
- Column failure attributable to inadequate ductility (toughness).
- Damage to shear keys at abutments.
3. Most common retrofit for superstructures is installation of unseating prevention devices that
may be classified into four groups; 1) Extension of seat length, 2) Connection of decks to
substructure, 3) Connection between adjacent decks at the edge, and 4) Connection of decks
at intermediate hinge.
4. In reinforced-concrete columns, the inadequate ductility usually stems from inadequate
confinement reinforcement. Premature shear failure at terminations of longitudinal bars with
insufficient development lengths has resulted in collapse of many bridges. To contrast the
concrete columns failures there are some retrofit techniques such as steel jacketing (most
common), active confinement by wire prestressing, use of composite materials jacketing
involving fiberglass, carbon fiber and other fibers, and jacketing with reinforced concrete.
5. In steel columns, the inadequate ductility usually stems from local buckling such as the
reinforced- concrete columns which progresses to collapse. In the specification, two seismic
retrofitting methods for existing steel columns are prepared on condition that the foundation
and the anchor bolts of the columns under consideration never damage, as follows; 1)
Filling steel columns with additional concrete over the existing encased concrete, 2)
Retrofitting the component stiffened plate panels of steel columns by using additional steel
members.
6. Shear key failure is due to the abutments mobilization. Such failures can appear with small
transverse displacement and little energy dissipation. Essentially there are two types of
 abutment movement as follows: 1) Tilt in the river side, 2) Tilt in the back-filled side to
instability of ground. Shear keys Retrofitting has been conducted by enlarging the footing,
and increasing number of piles similar to foundations. Increase of soil strength by soil
replacement or cement mixing is also a widely used for tilt resulted from instability of
underlying ground. For the tilt of abutments in the front direction, earth-anchors are
frequently used. An unique retrofit for tilt of abutments in the front direction is to replace
some part of the back-fills with expanded polystyrene (EPS).

4 REFERENCES

[1].Quick earthquake damage detection method for bridge structures J. Sakai Public Works
Research Institute, Tsukuba, Japan H. Kobayashi Hanshin Expressway Company Limited, Osaka,
Japan S. Unjoh Public Works Research Institute, Tsukuba, Japan).
[2]. Jack P. Moehle and Marc O. Eberhard [2000].’’ Earthquake Damageto Bridges hand book,
chapter 34, Introduction’’.
[3]. (2000 by CRC Press LLC 34 Earthquake Damage to Bridges book)
[4]. AA.VV, The whittier narrows earthquake of October 1, 1987. Earthquake Spectra 1988;4(1).
[5]. AA.VV, Northridge earthquake of January 17 1994 reconnaissance report. Earthquake Spectra
1995; 1995(Suppl. C):11.
[6]. Hassani N, Takada S. Lifelines performance and damage during January 17, 1995 South Hyogo
Great Earthquake. Proceedings of the
first Iran–Japan Workshop on recent earthquakes in Iran and Japan, Iran; 1995. p. 1809–21.
[7]. Le Val Lund PE. Lifeline utilities lessons, Northridge earthquake. Lifeline Earthquake
Engineering, Proceeding of the IV US Conference, Technical Council on Lifeline Earthquake
Engineering; 1995. p. 6.
[8]. Lindell M, Perry R. Hazardous materials releases in the Northridge Earthquake: implications
for seismic risk assessment. Risk Anal 1997;
17(2).
[9]. Takada S, Hassani N, Abdel-Aziz M. Quick report of main damage caused by the Hyogo-Ken
Nambu earthquake. J Nat Disasters 1995; 16:59–70.
[10]. Cole S. Lifelines and livelihood: a social accounting matrix approach to calamity
preparedness. J Contingencies Crisis Mgmt 1995;4(3).
[11]. Nojima N. Lifeline system malfunction and interaction. Proceedings of the World Urban-
Earthquake Conference in Fukui, Japan; 1998.
[12]. Priestley, N.M.J., Seible, F., and Calvi, M. (1996) “Seismic design and retrofit of bridges,”
John Wiley & Sons, New York.
[13]. Chai, Y.H., Priestley, N.M.J., and Seible, F. (1990) “Retrofit of bridge columns for enhanced
seismic performance,” Proc. 1st US-Japan Workshop on Seismic Retrofit of Bridges,
321-340, Tsukuba Science City, Japan.
[14 ]. Ogata, N, Maeda, Y and Ando, H. (1994), “Carbon Fiber Strengthening of Existing Bridges,”
2nd US-Japan
Workshop on Seismic Retrofit of Bridges, pp321-333, Report No. UCB/EERC 97-9, EERC, UCB,
CA, USA