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

net/publication/257523459

Analysis and interpretation of the seismic response of RC buildings in

Concepcin during the February 27, 2010, Chile earthquake

Article in Bulletin of Earthquake Engineering · February 2012

DOI: 10.1007/s10518-012-9404-5

CITATIONS READS

61 1,639

9 authors, including:

Juan Carlos de la Llera Rosita Jünemann

Pontificia Universidad Católica de Chile Pontificia Universidad Católica de Chile
102 PUBLICATIONS 2,378 CITATIONS 30 PUBLICATIONS 413 CITATIONS

SEE PROFILE SEE PROFILE

Matias Hube Jose Antonio Inaudi

Pontificia Universidad Católica de Chile National University of Cordoba, Argentina
92 PUBLICATIONS 1,198 CITATIONS 41 PUBLICATIONS 1,259 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Matias Hube on 22 March 2015.

The user has requested enhancement of the downloaded file.

Bull Earthquake Eng (2013) 11:69–91
DOI 10.1007/s10518-012-9404-5

ORIGINAL RESEARCH PAPER

Analysis and interpretation of the seismic response of RC

buildings in Concepción during the February 27, 2010,
Chile earthquake

Benjamín Westenenk · Juan Carlos de la Llera · Rosita Jünemann ·

Matías A. Hube · Juan José Besa · Carl Lüders · José Antonio Inaudi ·
Rafael Riddell · Rodrigo Jordán

Received: 16 January 2012 / Accepted: 9 November 2012 / Published online: 2 December 2012
© Springer Science+Business Media Dordrecht 2012

Abstract Observed trends in the seismic performance of eight severely damaged reinforced
concrete (RC) structures after the February 27, 2010, Chile earthquake are presented in this
article. After a reconnaissance and surveying process conducted immediately after the earth-
quake, several aspects not conventionally considered in building design were observed in the
field. Most of the considered structures showed extensive localized damage in walls of lower
stories and first basements. Several factors indicate that damage was brittle, and occurred
mainly in recent RC structures supported on soft soils with some degree of vertical and/or
horizontal irregularity. Non-ductile behavior has been inferred due to the lack of evidence of
spread damage in the structure, and the fact that very similar structural configurations existed
nearby without apparent damage. Some key aspects in understanding the observed damage
are: geographical orientation of the building, presence of vertical and horizontal irregularities,
wall thickness and reinforcement detailing, and lack of sources for energy dissipation. Addi-
tionally, results of a building-code type analysis are presented for the 4 most critical buildings,
and Demand/Capacity ratios are calculated and compared with the observed behavior. It is
concluded that the design codes must be revised relative to wall design provisions.

Keywords Reinforced concrete · Shear wall · Seismic behavior · Design code ·

Chile earthquake

B. Westenenk · J. C. de la Llera · R. Jünemann · M. A. Hube (B) · J. J. Besa · C. Lüders ·

R. Riddell · R. Jordán
Department of Structural and Geotechnical Engineering, Pontificia Universidad Católica de Chile,
Av. Vicuña Mackenna 4860, Santiago, Chile
e-mail: mhube@ing.puc.cl

J. A. Inaudi
Department of Structures, Universidad Nacional de Córdoba,
1611 Av. Vélez Sarsfield, Córdoba, Argentina

123
70 Bull Earthquake Eng (2013) 11:69–91

1 Introduction

The great February 27, 2010, Chile earthquake of moment magnitude Mw = 8.8, affected
approximately 13 million people and the most industrialized part of the country, causing
serious damage to residential buildings, industry, lifelines, and other infrastructure. Although
less than 1 % of the medium-rise RC buildings exposed to the earthquake showed severe
damage, such damage was brittle and showed repetitive patterns in different structures and
locations.
The seismic performance of some shear wall buildings was certainly unexpected. These
type of buildings, so positively evaluated after the 1985 Chile earthquake (Riddell 1992),
showed this time to be vulnerable and most likely incapable of dissipating the energy con-
sistent with the strength reduction factor used in their design. This behavior is relevant
because there is little information available in the literature of severely damaged medium-
rise shear-wall buildings. Thus, information presented in this article is useful for increas-
ing our seismic database of cases and serves also as good start for future more in-depth
research.
This article presents results obtained from a larger initiative that involved different field
visits, detailed damage surveys, and the initial structural evaluation of eight RC buildings
located in the city of Concepción, 440 km south-west from Santiago. The first inspection of
the site was performed in March 2010, soon after the earthquake by a team composed of
EERI members, local researchers, engineers, and students. Later on, the Chilean Ministry
of Public Works (MOP) requested experts from Pontificia Universidad Católica de Chile to
perform an in-depth study by collecting evidence and data for further analysis of the true
structural condition of these buildings. It is apparent that there are substantial aspects to
be investigated regarding the seismic performance of these structures. More detailed infor-
mation such as structural drawings with damage identification, structural models, material
logs, and other documentation, is available and may be used for future research purposes
(Westenenk 2011).
A first global analysis of the field information mentioned is presented elsewhere (West-
enenk et al. 2012). The scope of this study is to anticipate observed trends in the seismic
performance, and to propose some general interpretation of the building response as a result
of the reconnaissance and surveying process. A more detailed analysis, but still preliminary,
for the considered building named TO-9 is also presented here. Finally, results of linear
response and code-type analysis are presented for buildings that showed more critical struc-
tural conditions.
There is little doubt that structural damage during an earthquake is often a multi-factorial
problem. Building performance was probably conditioned by factors such as: (i) the long
duration of ground motion (2 min and 45 s), (ii) ground motion directivity effects, (iii) simul-
taneous action of horizontal and vertical ground motion excitations, (iv) considerable vertical
component of ground motions, (v) local soil conditions, (vi) structural orientation, (vii) wrong
characterization of design spectra for long periods, (viii) excessive axial loads, (ix) lack of
detailing, and (x) construction aspects, among other factors. Many of these interesting aspects
may trigger questions, such as: why did shear failures occur in some walls of the buildings in
Concepción, but not in Santiago (Moehle 2010)?, why were there exceptions in the general
correlation of structural damage with soil-type and topographical accidents?, and what made
these buildings different from others with similar structures and no damage?. These questions
will not all be answered in this research since they require a larger bulk of research on the
subject; their purpose is to serve simply as guidelines of our work.

123
Bull Earthquake Eng (2013) 11:69–91 71

2 Trends in seismic building behavior

This section describes some relevant and recurrent structural aspects that have been identified
by analyzing damage in eight buildings located in the city of Concepción, Table 1. All
buildings are RC wall structures commonly denoted as “fish-bone” (except building TO-9),
characterized by a typical plan with a central longitudinal corridor enclosed by shear walls,
and transverse walls separating building apartments and interior spaces (Westenenk et al.
2012).
Shown in Table 1 are the number of stories above grade and the number of basements;
the year of construction; the longitudinal axis orientation, defined in clockwise direction
from the north cardinal direction; the soil type classification based on Chilean seismic code
(INN 1996) and obtained from detailed geotechnical reports performed after the earthquake
by this team; the percentage of total damage defined as the ratio between the total volume
of damaged elements over the total volume of elements, including walls, columns, beams
and slabs with light, moderate, or severe damage (Westenenk et al. 2012); the fundamental
building period; the ordinate of the elastic design spectrum (INN 1996) at the fundamental
period of the building; and the ordinate of the response spectrum at the fundamental period
of the building corresponding to the closest site with a recorded ground motion (San Pedro
record for AH-2 building and Concepción record for all other buildings).
In the case of CM-3 and TO-9, soil classification results differ from the original classifi-
cation during building design. Soil type II represents a dense gravel or clay with shear wave
velocity larger than 400 m/s in the upper 10 m, and type III is unsaturated gravel or clay with
shear wave velocity less than 400 m/s (INN 1996).
Although the eight buildings considered in this article are localized in the same region,
damage patterns observed in them also occurred in other buildings at very distant places with
different base motions and soil conditions. This tends to suggest that the observed repetitive
wall damage is somewhat invariant to local anomalies present in the ground motions and soil
conditions, and that there is probably a more general design explanation to this damage.
This section analyzes five critical aspects that may be correlated with the observed damage:
(i) geographical orientation of damage; (ii) existence of vertical and horizontal irregularities;

Table 1 RC buildings considered in this investigation

ID # of Stories Year Longitudinal Soil Total Building Design Response

(above built axis type damage period spectrum spectrum
grade + orientation (%) (sec) (m/sec2 ) (m/sec2 )
below grade) (◦ )

AA-1 20 + 1 2002 64 II 12.31 0.71 4.79 9.51

AH-2 15 + 2 2009 175 III 2.44 0.78 10.49 12.56
CM-3 18 + 1 2005 152 IIa 6.60 0.80 4.05 7.74
TL-4 17 + 1 1973 152 III 6.38 0.88 9.42 6.51
PR-6 12 2006 152 III 9.79 0.50 12.11 7.06
PP-7 10 2004 4 III 6.07 0.36 11.16 8.68
RT-8 10 2006 4 III 7.73 0.36 11.16 8.68
TO-9 21 + 2 2008 152 IIIb 10.23 0.93 8.89 5.04
a Original classification was type III.
b Original classification was type II.

123
72 Bull Earthquake Eng (2013) 11:69–91

(iii) detailing of walls and construction aspects; (iv) damage distribution within the structure;
and (v) energy dissipation capacity of the structure.

2.1 Orientation of damage

It is apparent that observed damage in walls tends to concentrate mainly in one principal
direction of the building. The percentage of damaged walls oriented in both N–S and E–W
direction for each building considered is summarized in Table 2. Damage is concentrated in
walls oriented in the E–W direction for all cases, with exception of AA-1 and PP-7. This
behavior can be attributed to the fact that the E–W direction coincides with the transverse
axis of most buildings, except for building AA-1 that has an orthogonal orientation. Building
AA-1 suffered localized damage in the N–S direction; however, this direction also coincides
with the transverse direction of this building. Finally, the peculiarities of the behavior of
buildings PP-7 and TO-9 are discussed in detail in later sections.
Please note that the peak ground accelerations recorded in Concepción were similar in
both directions, 0.61 and 0.65 g, for the E–W and the N–S direction, respectively in the San
Pedro record (Barrientos 2010).
Shown in Fig. 1 are the cases of buildings AH-2 and TL-4, which experienced severe
damage in the walls located in the first two stories and in the E–W direction, which coincides
with the transverse axis of these buildings. For these two buildings the damage observed in the
longitudinal walls was small. It is quite remarkable that both buildings were able to survive
the earthquake without collapsing in spite of the severe damage in the E–W walls. This
behavior may be attributed to the role played by the less damaged longitudinal walls which
helped stabilize the structure in the transverse direction as well as the longitudinal one. This
hypothesis is corroborated by the observed tension and compression driven damage observed
at each side of the longitudinal walls indicating that the building ended up balancing about the
longitudinal corridor. Damage in the transverse walls was also observed in buildings located
in the cities of Santiago and Viña del Mar.
Building PR-6 is also a clear example of this balancing effect. It is composed of two
structures perpendicularly oriented in plan, forming an “L-shaped” footprint (Fig. 2). The
damage of this building was also concentrated in the stories 1 and 2 (Westenenk et al.
2012), and despite the similarity of the two structures, only the building with its short axis

Table 2 Distribution of damage in walls

ID % of damaged % of damaged Approximate

walls N–S walls E–W orientation of
the longitudinal
axis

AA-1 66 34 E–W
AH-2 27 73 N–S
CM-3 17 83 N–S
TL-4 18 82 N–S
PR-6 36 64 N–S
PP-7 66 34 N–S
RT-8 5 95 N–S
TO-9 49 51 N–S

123
Bull Earthquake Eng (2013) 11:69–91 73

Story 1 Story -2 Story -1

Story 1 Story 1 Story 2

Transverse
Transverse direction
direction Story -2

Story 1

N
N

Story -2

Story 2 Story 2 Story 1 Story 1 Story -1

Story 2
(a) (b)
Fig. 1 Examples of severely damaged walls in buildings: a TL-4 and b AH-2

oriented in the E–W direction (shaded region in Fig. 2b) suffered severe damage. These
walls evidenced flexure, axial, and shear damage, indicative of the lateral cycles undergone
by the structure. The walls oriented in the E–W direction of the non shaded structure suffered
minor damage because this direction coincides with the longitudinal axis of the building. The
shaded structure of building PR-6 also evidenced diagonal cracks in the slab located close to
axis N (Fig. 2c). The presence of these cracks and the fact that the most damaged walls were
located along axis V, indicates that the building rotated about a vertical axis passing near the
much stiffer shear wall core (axis N). The shaded structure of building PR-6 also showed
tilting towards the east with a residual roof displacement of about 35 cm (Fig. 2a). The rigid
body displacement shape observed in this building may be explained by foundation rotation,
which is consistent with the foundation uplift and ground settlement observed around the
building (Fig. 2d, e).

2.2 Vertical and horizontal irregularities

Vertical and horizontal irregularities were a common factor present in most of the damaged
buildings, though they are currently not explicitly incorporated in the Chilean design code
(INN 1996). Buildings TO-9 and AA-1, which suffered the greatest percentage of damaged
elements (Table 1), were characterized by having large horizontal and vertical irregularities.
Figure 3a, b show schematic plan views of these buildings, where the different dash-line traces

123
74 Bull Earthquake Eng (2013) 11:69–91

~35.5cm Minimum damage

24 27 31 34
2 21
C

L
N
Evidence of cracks

Severely damaged
due to torsion

N
(c)
V W
24 34

Pendulum type movement

(b)

(a) Ground settlement (d) (e)

Fig. 2 Building PR-6: a schematic elevation of axes V and W; b story-plan and damaged area; c cracking in
axis N; d ground settlement near axis V; and e ground settlement observed around the building

identify the geometry changes in plan along the building height. As a result of both vertical
and horizontal irregularities, severe damage was observed at the 11th story of building TO-9
(Fig. 3c). In this story, a beam that was supporting an inverted wall—i.e., a wall that starts
on the story above and lacks continuity in the present story and below—dropped about 1.2 m
leading to the observed damage.
A common vertical irregularity pattern observed in most damaged buildings was the flag-
shaped walls. This type of walls are characterized by having reduced length at lower levels,
mainly in the first stories and parking levels due to architectural and functional constraints
(Fig. 4a). The flag-shaped walls produce high stress concentration at the discontinuity region
of the wall, which leads to localization of damage (Fig. 4b). The observed damage in flag-
shaped walls of buildings AA-1 and AH-2 are shown in Fig. 4c, d, respectively. In these walls
the bending moment and axial load are expected to be similar above and below the slab, but
the axial stresses in the lower wall are much higher, facilitating its damage. In flag-shaped
walls, the usual design assumption that the floor slab at the discontinuity level would be
capable of distributing shear from the weaker and more flexible flag-shaped wall to adjacent
walls in a sudden form is not warranted.
Another aspect of vertical irregularity is found where two different walls become coupled
through lintels in upper and/or lower stories. This irregularity is common as openings are
closed in some stories (Fig. 5a). The observed damage in coupled walls for several buildings
is shown in Fig. 5b–f. As both walls try to move independently, the coupling shear stresses
induced at the continuous wall in top and bottom stories are very high causing both lintels to
fracture (Fig. 5a). Therefore, natural openings tend to form in such continuous walls. This

123
Bull Earthquake Eng (2013) 11:69–91 75

Most damaged story

Story -1
Collapsed story
Stories -1, -2 Stories 1-5
Story 1 Stories 6-14
Story 2 Stories 15-16
Stories 3-4 Story 17
Story 5 Stories 18-20
10 Story 6
Stories 7, 8
(b)
Stories 9-11
Stories 12-15
Story 16
Stories 17-19
Story 20 ~ 1.20 m
Story 21

(a) (c)
Fig. 3 Diagram of severe irregularities in buildings: a TO-9; and b AA-1. Example of (c) inverted wall and
damage above beam at the 11th story of TO-9 along axis 10

Flag-shaped wall

High stress concentration

(c)

(d)
(a) (b)
Fig. 4 Flag-shaped vertical irregularity: a typical flag-shaped wall; and b damage in flag-shaped walls.
Examples in buildings: c AA-1; and d AH-2

type of damage may affect the vertical load carrying capacity of slabs located on the damaged
floors causing serious local stability issues.
Moreover, abrupt changes in wall thickness were also a source of irregularity that induced
damage, especially in building AH-2. The plan views of the 1st and 2nd stories of building
AH-2 are shown in Fig. 6a, b, respectively. The corner walls have variable thickness of 20
and 30 cm. The regions of these walls with 30 cm thickness are circled with solid lines in
Fig. 6a, b. As shown in Fig. 6c, shear dominated damage tends to concentrate at the coupling
interface of the two walls of different thickness, mainly due to their relative displacement
and stress concentrations generated in this interface. This building also evidenced punching
shear slab damage at the four corners of the building generating permanent slab deformations
(Fig. 6e). This damage was due to the large cantilever slabs located at the building corners that

123
76 Bull Earthquake Eng (2013) 11:69–91

Opening is formed at
continuous wall in
top stories

Door
(b) (c) (d)

Opening is formed at
continuous wall in
bottom stories
(e) (f)
(a)
Fig. 5 Other relevant vertical irregularities: a schematic cracking of top and bottom lintels; damage in upper
stories b AA-1; c AH-2; and d CM-3; and damage in lower stories e CM-3; and f building in Chillan

2 4 13 15 2 4 13 15

C C

E E

W W

Y Y

(a) (b) (c)

(d) (e)
Fig. 6 Behavior of building AH-2: a story 1; b story 2; c shear damage in first story wall; d slab movement;
and e punching slabs by walls

123
Bull Earthquake Eng (2013) 11:69–91 77

Beam supported by orthogonal wall

Orthogonal wall

(a) (c) (d)

Direction of movement

Spandrel beam

Transverse
shear wall

Transverse
shear wall
(b) (e) (f)
Fig. 7 Example of beam damage: a schematic plan view of beams supported by orthogonal walls; b beams
coupled with transverse shear walls; beam punching for: c RT-8; d TO-9; e PP-7; f and damage in coupling
beams of PP-7

flapped up and down freely leaving permanent deformation in all stories above 2 (Fig. 6d).
This damage was increased by the vertical discontinuity of the walls circled with dashed
lines in Fig. 6b that run from stories 2 through 15 and are discontinued in story 1. Although
this damage in the slabs is localized, repair costs are very high.
Other structural configuration issues also induced damage to the observed buildings. Some
beams supported by orthogonal walls (Fig. 7a) induced damage to the walls because the
cyclic motion tends to pull out the beam and punch the orthogonal walls. Such damage was
observed in walls of buildings RT-8, TO-9 and PP-7 (Fig. 7c–e). Furthermore, damage was
also observed in spandrel beams coupled with transverse shear walls (Fig. 7b), as observed
in building RT-8 (Fig. 7f). Clearly, the common modeling assumption of the spandrel beams
as frame elements in a structural model does not necessarily capture well the real behavior
of this connection.

2.3 Structural detailing of walls and construction issues

Typical Chilean construction changed after the generally good performance of shear walls
during the 1985 Chile earthquake (Riddell et al. 1987; Wallace and Moehle 1989, 1992).
Taller buildings have been constructed with thinner walls as a result of not limiting the wall
axial stresses. Prior to 1985, wall thicknesses of 30 cm and above were common in mid-rise
RC buildings (Wood 1991; Wood et al. 1987), while thicknesses of 20 cm or less, 15 cm,
are common in newer construction (Westenenk et al. 2012; Massone et al. 2012; Jünemann
et al. 2012). This proved to be a critical aspect in this earthquake, as thin walls subject to

123
78 Bull Earthquake Eng (2013) 11:69–91

(a) (b) (c)

Fig. 8 “Unzipping” wall damage in buildings; a AH-2; b CM-3; and c PR-6

high axial-flexural demands became overcrowded with longitudinal steel, poorly confined,
and brittle.
Due to the ground shaking, high axial stresses are imposed at the wall boundaries due to
axial-flexural interaction. Since only a small area of these walls was partially confined—and
most of the time poorly confined because the Chilean seismic code (INN 1996) excluded
the ACI (1995) confinement requirements— cover concrete spalls-off at wall boundaries and
longitudinal boundary reinforcement was exposed and buckled due to the large spacing of
the transverse steel, which was poorly hooked into the longitudinal bars. Consequently, the
wall lost considerable flexural capacity, while maintaining the axial load. As building cycles
continued, the longitudinal reinforcement of the web needed to resist the bending moments,
but the demand exceeded the flexural capacity of the wall. Therefore, the lateral cover suffered
spalling, the longitudinal web reinforcement yielded and fractured, the concrete core crushed,
and the wall decreased its capacity to sustain axial loads. From the field observation, the
estimated high axial load demands, and the obvious lack of confinement within the boundary
element, it is hypothesized that this “unzipping” crack quickly propagated throughout the
length of the wall in a brittle manner. This “unzipping” wall damage observed in buildings
AH-2, CM-3 and PR-6 is shown in Fig. 8a–c. Due to the localized nature of the failure,
the building was not capable to develop large ductility or energy dissipation making the
assumption of the strength reduction factor used in design unwarranted. For T-walls with
large flange in compressions, no damage occurred on the flange, which also demonstrates
that compression stresses played, as would be expected, an important role in the observed
brittle or ductile failure pattern.
As mentioned earlier, confinement of the boundary elements of shear walls was very poor
or nonexistent. The typical detailing in wall boundaries considered transverse steel with legs
bent in 90◦ , but without hooking the longitudinal bars (Fig. 9d). Although, this confinement
proved to be highly insufficient and brittle behavior was observed in such walls, the question
that remains is whether better confinement could have limited the observed damage.
Other construction problems and detailing issues that may have triggered the observed
damage were identified in the analyzed buildings, especially for building TO-9 (Fig. 9).
Poor anchorages between beams and walls were frequently observed, where longitudinal
reinforcement of beams was sometimes anchored in the concrete cover of the wall, i.e.,
outside of the confined concrete core (Fig. 9b). Since the beams and walls were long and had
the same thickness as the wall, there was no or little space inside the concrete core of the
wall to anchor such large longitudinal reinforcement of the beams. Additional construction
problems were observed also in building TO-9 such as beams with cut stirrups or even without
stirrups (Fig. 9c).

123
Bull Earthquake Eng (2013) 11:69–91 79

(a) (b) (c) (d)

Fig. 9 Poor anchorage of beams: a orthogonal beam-wall connection; b lateral beam wall connection with
reinforcement outside of the confined area; c unconfined longitudinal reinforcement in beam and lack of
stirrups; and d poor confinement of transverse reinforcement in shear wall

2.4 Damage propagation within the structure

Building CM-3 evidenced multistory critical sections with damage, where three of the four
corners of the building experienced damage involving more than one story (Fig. 10). For this
building, the fracture along the critical section followed the simplest and least demanding
failure mode. This type of damage is somewhat difficult to capture in design because struc-
tural models and design procedures treat building elements as single-story units. Therefore,
buildings should always be analyzed and designed as 3D systems, considering all possible
failure mechanisms crossing stories and connecting weak elements in neighboring stories.
Multistory damage propagation is also apparent in building PR-6 as shown in Fig. 11e.
Propagation of damage into orthogonal wall elements and slabs was observed in PR-6
(Fig. 11). More specifically, Fig. 11b shows how shear damage started at the bottom of
wall V1 along axis V, propagated through the slab (zoom at Fig. 11c) and reached wall W1
along axis W. The effect of shear stress in the center of wall V2 combined with axial-flexure

B 13

1 3.2
3.1 13

A B
D

3.1 D A (b)

1 3.2
(a) (c) (d)
Fig. 10 Multistory damage of CM-3: a structural plan of story 2; b N–E corner; c N–W corner; and d S–W
corner

123
80 Bull Earthquake Eng (2013) 11:69–91

24 28 34

V2
(c)

V
V1 W (d)
W1
(a) (b)
24 27 28 31 34

(e)
Fig. 11 Building PR-6: a structural plan of 1st story; b shear damage propagation through the slab; c zoom
of damaged slab; d superimposed axial-flexural and shear stress cracking; and e global multistory damage of
resisting plane V–W

interaction at the boundaries of this wall, leads to the peculiar trace of the crack shown in
Fig. 11d. Based on this observation, it is apparent that it is not enough to analyze and design
each resisting element individually, since damage in one element could lead to a different
behavior and severe damage in other elements. This observation and behavior in building
PR-6 repeats in several other buildings.
For building CM-3, damage occurred not only between stories and elements, but it also
showed a variable pattern behavior within the element. Shown in Fig. 12b, c, the damage
propagated from the edges of the wall due to axial-flexural interaction, toward the center of
the wall on axis F (Fig. 12a) causing shear damage.

2.5 Energy dissipation sources

The fifth critical aspect discussed here is the true energy dissipation capacity of current
Chilean residential shear-wall buildings. After the 1985 earthquake, these buildings became
taller without increasing the wall density, which led to higher vertical stresses in the walls
(Jünemann et al. 2012). Moreover, the architectural topology has changed by drastically
reducing or even eliminating lintels or coupling beams, moving toward a flat-slab configu-
ration. Lintels were eliminated because they are essentially impossible to design due to the
coupling of walls which leads to large amounts of reinforcement. Unfortunately, with this

123
Bull Earthquake Eng (2013) 11:69–91 81

3 8 14

(b) (c)

(a)
3 8 9 14 3 8 9 14 3 8 9 14

Axial-flexural damage starts Damage starts at Damage spreads Damage spreads through middle
at one edge opposite edge wall as shear failure
(d)
Fig. 12 Combined damage in CM-3: a location of walls in plan CM-3; b observed damage in wall at axis F;
c shear failure of center of wall; and d schematic view of induced damage

elimination a very good source for energy dissipation in the building was also eliminated.
Thus, the strength reduction factor (R-factor) in the Chilean code should have been reduced
accordingly, since it was calibrated for a different structural configuration with coupling
beams, as typically used in the past (prior to 1985). Since the building must dissipate energy
to reduce forces, in many cases the slabs turned into flat coupling beams and underwent
severe structural damage. Examples of such behavior are shown in Fig. 13 for buildings
AA-1, AH-2 and PP-7. The elimination of lintels also increased the flexibility of buildings
inducing heavy damage in some non-structural elements.
The absence of coupling beams makes non-structural components in a building to increase
their role as energy dissipators. Indeed, the performance of non-structural components
may well explain the different intensities in the observed structural damage for the iden-
tical buildings PP-7 and RT-8. As mentioned elsewhere (Westenenk et al. 2012), building
RT-8 showed larger structural damage than PP-7, although they have identical nominal plans,
identical orientation, and are located next to each other. However, building PP-7 contained

123
82 Bull Earthquake Eng (2013) 11:69–91

(a) (b) (c)

Fig. 13 Evidence of slabs working as coupling beams in buildings: a AA-1; b AH-2; and c PP-7

very stiff partition elements made of plaster while RT-8 contained very flexible partitions.
Non-structural elements in building PP-7 were practically shattered, and in RT-8 they were
practically undamaged. It is a plausible hypothesis that the energy dissipation coming from
damaged non-structural elements in PP-7 contributed to reduce structural damage as opposed
to RT-8. The influence of the non-structural components in these cases was critical since
building PP-7 will be repaired and RT-8 will be demolished.

3 Interpretation of TO-9 building

This section describes the particular behavior of building TO-9, due to some additional aspects
of interest to researchers and practitioners. Although its stability condition is precarious,
building TO-9 was thoroughly inspected, element by element, and detailed videos from the
state of the structure were recorded by the authors. Internal inspection was possible by hanging
safety lines from the central core, which did not showed significant damage throughout the
building height.
Building TO-9 suffered its main collapse at story 12, where the east section of this story
collapsed and dragged the upper stories down (Fig. 14b, c). The north and west facades
(Fig. 14a) evidenced very little damage as compared with the east and south ones (Fig. 14b,
c). In addition to the main collapse in story 12, two other partial collapses occurred in stories
16 and 20, which can be observed from the East elevation (Fig. 14b). A close up of the main
collapse at story 12 (axis J) is shown in Fig. 14e (axes in story 12 are defined in Fig. 15a).
The collapsed region shown is between axes 4 and 7, and G and J. The slabs crushed on
to the ceiling of the 11th story; Fig. 14e shows that the slab between axis 4 and 7 dropped
one complete story, while the slab between axis 9 and 10 just half a story. Columns were
completely collapsed as observed in Fig. 15e and f and all other structural elements deformed
as shown from the interior in Fig. 15d. One relevant observation is that the wall thickness in
axis J changes from 25 to 20 cm in the 12th story (Westenenk et al. 2012). At axis J, upper
stories displaced westward also by 13 cm, approximately (Fig. 14c).
Stories above level 12 only collapsed partially because stiffer walls, enclosed in blue
lines in Fig. 15a, remained essentially with light or no damage. Actually, the west elevation
(axis C) had negligible damage as shown in Fig. 14a. Therefore, the collapsed part of the
structure corresponds to that farther from the core with several short walls and columns, some
with the thickness irregularity present at the 12th story. Tensile cracks due to torsion were
also observed in the wall located along axis C (refer to the close-up figure to the top left of
Fig. 15), which may be attributed to significant aftershocks.

123
Bull Earthquake Eng (2013) 11:69–91 83

Main collapse
Partial collapse

Story 20

Story 16

Story 12

(a) (b)
N

~13cm

(d)
(e)
(c)

Lineof severely
damaged columns

(f) (g)
Fig. 14 Overview of building TO-9: a west side view; b east side view; c south side view and close-up of
deformed axis J; d plan view; e details of the main collapse at story 12; f damage in short columns in axis 1A;
and g overall condition of the 21st story

The partial collapse at story 16 is shown in Fig. 15b from the inside and the partial collapse
at story 20 is shown in Fig. 15c from the outside. For these collapses to occur columns were
completely displaced out of their axes, losing their vertical capacity. Additionally, both stories
had important setbacks, which coincided with the regions mostly affected.
Building TO-9 also experienced severe shear damage in squat walls. In the upper stories,
these walls are connected by spandrel beams as shown in Fig. 14f for axis 1A. Once the short
walls failed in shear, the span of the beams increased as well as their M/Vd ratio, causing
plastic hinges at both ends of these beams (Fig. 14f).

123
84 Bull Earthquake Eng (2013) 11:69–91

C G J
Stiff section of building

9
(b)
7

5
Evidence of torsion
post-collapse 4

9 (a) (c)

(d) (e) (f)

Fig. 15 Internal damage of building TO-9: a structural plan of the 12th story; b partial collapse of the 16th
story; c partial collapse of the 20th story; d damaged beams of the 12th story; e column in axes G-5 of the
12th story; and f column in axes G-4

The slabs played an important role by holding axially the collapsed regions to the core
walls, thus preventing a larger collapse. Additionally, Fig. 14g shows the critical condition of
stories 20 and 21 as all columns farther from the core of shear walls were severely damaged;
the slab partially collapsed by losing the vertical support along axes J and G.
The shear wall in axis 4 between axes B1 and D1 and stories 11 and 12, was the only
wall from the core with severe damage, as shown in Fig. 16. This resisting plane has a flag-
shaped geometry. The wall showed axial-flexural interaction localizing damage first in the
top corner of story 11 (Fig. 16a), which leads to a much smaller section of the wall given the
“unzipping” type of failure. As a result a shear failure was induced at story 12. This brittle
damage induced a sudden larger lateral and torsional flexibility at the 12th story; based on
judgment all these types of irregularities and setbacks in the building played a crucial role in
its collapse.
In general, building damage was observed to increase with height. This fact contradicts
a capacity design philosophy of wall buildings in which plastic hinges should occur at the
first stories. Little to no damage was observed between stories −2 through 4; severe damage
occurred in a single wall in the 5th story; and stories 6 through 12 showed increasing damage
in height; with a main collapse in the 12th story. Local damage to structural elements such
as slabs and columns increases in upper stories; damage of stories 16, 18 and 21 is shown in
Fig. 17a and c, respectively.

123
Bull Earthquake Eng (2013) 11:69–91 85

Movement

B1 C D1 B1 C D1 B1 C D1

Story 14

Story 13

Story 12

Story 11

Story 10

Axial-flexural damage Shear damage at story 12

at story 11

B1 C D1
(a)
4 D1

(b) (c) (d)

Fig. 16 Important damage in the shear-wall core of building TO-9: a schematic explanation of the observed
behavior; b wall location in story 12; c wall biaxial bending and damage in story 11; d shear damage of wall
in story 12

4 Building code type analysis

A complete code review of the structural design was performed for buildings AA-1,
CM-3, PR-6 and TO-9 in order to provide an idea of the nominal structural condition
of each building. These buildings were selected because they have the largest percentage
of severely damage elements, Table 1. Theoretical Demand/Capacity (D/C) ratios were
computed for all elements, and they were classified according to the four categories pre-
sented in Table 3. The results obtained are analyzed and contrasted with observed damage
to validate any correlation with strength deficit. Displacement and interstory drifts ratios
were also calculated for all the 8 buildings listed in Table 1 and compared with code
provisions.
The three-dimensional finite element linear models of each building were developed using
commercial software (ETABS 2011). The models considered the gross section of elements,
material properties obtained from experimental results performed on the buildings, and rigid

123
86 Bull Earthquake Eng (2013) 11:69–91

(a) (b) (c)

Fig. 17 Internal damage of building TO-9: a slab deformation in story 16; b overview of damage in story 18;
and c severely damaged column in story 21

in-plane floor diaphragms. Models were fixed at the foundation level except for buildings
AH-2 and TO-9, which included a full soil-foundation-structure interaction model. Seismic
weight with 100 % of dead load and 25 % of live load was considered.
The design spectrum given by the Chilean design code (INN 1996) was used for modal
analysis. Ninety percent of the total seismic weight was considered in each direction of
analysis; and modal superposition used the CQC method and a constant damping ratio of
0.05. Shown in Table 1 is a comparison of the probable demand produced by this earthquake
relative to the one considered by the design of the building (INN 1996).
Please note that the seismic code used in this review (INN 1996), is the same as the one
used for the design of these buildings. This code refers to ACI318-95 (1995) for the design of
RC elements. Furthermore, all code analyses were based on the soil classification performed
after the earthquake by the authors, Table 1. The original soil classification differs from ours
in buildings CM-3 and TO-9 (Table 1). For building TO-9 the soil was originally classified
as type II, and according to this study it should be type III (softer), implying a consider-
able larger seismic demand on the structure. Moreover, the original soil classification for
CM-3 is soil type III and our study indicates soil type II. Therefore D/C ratios would increase
considerably from the ones that will be shown later. A proper soil classification is particularly
critical in these projects.

4.1 Results of wall analysis

Walls were evaluated in shear and axial-flexural interaction. A similar analysis was performed
for all beams (Westenenk 2011), but for sake of brevity the results are omitted here. For
shear, D/C ratios are defined as Vu /Vn . The shear D/C ratios, its mean value and standard
deviation for a representative sample of walls for each of the four considered buildings are
summarized in Table 4. The analysis was focused on the most damaged building sectors,
which for buildings CM-3 and PR-6 coincides with the lower stories.
It is apparent from Table 4 that all buildings present walls with D/C ratios greater than
1.0, which implies a deficit relative to the code design. However, the number of elements

Table 3 Categories of D/C ratios

Properly designed Slightly deficient Moderately deficient Significant

elements elements elements strength deficit

D/C ratio < 1 1 < D/C ratio ≤ 1.25 1.25 < D/C ratio ≤ 1.5 D/C ratio > 1.5

123
Bull Earthquake Eng (2013) 11:69–91 87

Table 4 Total results for shear behavior in walls

ID No of walls D/C < 1 (%) 1 < D/C ≤ 1.25 (%) 1.25 < D/C ≤ 1.5 (%) D/C > 1.5 (%) μD/C σD/C
analyzed

AA-1 1,458 85.87 5.76 3.43 4.94 0.58 0.48

CM-3 1,383 67.25 14.68 9.69 8.39 0.82 0.47
PR-6 593 90.89 5.23 3.54 0.34 0.50 0.33
TO-9 1,152 89.67 4.86 2.08 3.39 0.52 0.38

AA-1 CM-3 PR-6 TO-9

1
Cumulative histogram

0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.00 0.50 1.00 1.50 2.00 2.50 3.00
Shear D/C ratio in walls
Fig. 18 Cumulative histogram of shear D/C ratios in walls

with D/C greater than 1.0 is not large in general (about 10 % for buildings AA-1, PR-6 and
TO-9), which implies, as indicated earlier that the design of these structures is not balanced
in terms of D/C requirements and probably just a few elements caused the high level of
damage observed. Note that mean values for buildings AA-1, PR-6 and TO-9 range from
0.50 to 0.58, while for CM-3 is 0.82. The standard deviation is also significant for the CM-3
building. This implied that the design of every element in building CM-3 is very close to 1,
which leaves little space for overstrength in the structure, especially considering the brittle
failures observed.
CM-3 is the building with the largest proportion of shear D/C ratios greater than 1.0 and
also greater than 1.5. Results show that CM-3 elements with significant strength deficit are
located in stories 2 and 3, and generally correspond to short perimeter walls coupled with
spandrel beams, thus leading to high shear stresses. In the case of building PR-6 walls with
D/C> 1.5 are concentrated in stories 1 and 2, matching also the observed damage (Westenenk
et al. 2012). Building AA-1 has the walls with severe strength deficit between stories 5 and
13, and TO-9 shows maximum code deficiencies on the 3rd story and axis J, which is the one
that concentrates the highest D/C ratios. This axis is the same as the one that collapsed in
story 12, and for buildings AA-1, CM-3 and TO-9, the highest D/C ratios are found in zones
with vertical irregularities.
The cumulative histogram of shear D/C ratios in walls for each building analyzed is shown
in Fig. 18. Although buildings AA-1, PR-6 and TO-9 show similar D/C distributions, the
CM-3 curve is well below the others. This emphasizes the fact that CM-3 has a weaker design
condition.
In the case of axial-flexural behavior in walls, interaction curves (Mn , Pn ) were com-
puted for a representative sample of walls in each building. The analysis focused on the most
damaged sectors; in particular, for buildings CM-3 and PR-6 this occurs in the lower stories.

123
88 Bull Earthquake Eng (2013) 11:69–91

Table 5 Total results for axial-flexural behavior in walls

ID No of walls D/C < 1 (%) 1 < D/C ≤ 1.25 (%) 1.25 < D/C ≤ 1.5 (%) D/C > 1.5 (%)
analyzed

AA-1 281 91.46 4.63 1.42 2.49

CM-3 69 68.12 8.70 10.14 13.04
PR-6 50 80.00 4.00 6.00 10.00
TO-9 205 93.17 2.44 2.93 1.46

The design verification considered the complete macro-element of the wall, be that C-shaped,
T-shaped, L-shaped, or other more complex shapes. To find the D/C ratio in this composite
walls, demand (D) is defined as the radial distance between the origin of the interaction curve
and the farthest point of demand (Mu , Pu ); capacity (C) is defined as the distance between
the origin and the point of the interaction curve (Mn , Pn ) along the same radius. Global
axial-flexural D/C ratios for the considered buildings are shown in Table 5.
Once again, building CM-3 shows the largest percentage of walls over demanded in design
with D/C ≥1.5—13.04 % of the verified analyzed elements versus 10 % for PR-6, and less
than 3 % for AA-1 and TO-9. This may be the case because the sample of walls analyzed for
CM-3 and PR-6 was concentrated in the lower stories, which are the critical ones, while in
the case of AA-1 and TO-9 the sample of walls was distributed all over the structure.
Special damage drawings were prepared in order to physically compare the observed dam-
age with that resulting from code analysis results. Drawings showing the observed damage
were created for the four buildings using the following color convention: lightly damaged
elements, green; moderately damaged elements, blue; and severely damaged elements, red.
Moreover, drawings showing maximum D/C ratios were created by assigning green to D/C
ratios between 1 and 1.25, blue to D/C ratios between 1.25 and 1.5, and red for 1.5 or higher.
Fig. 19 shows both types of drawings for buildings CM-3 (Fig. 19a, b) and PR-6 (Fig. 19c, d).
The comparison revealed that there are several cases in which the design deficit, represented
by high D/C ratios, match the damaged elements. However, there are also cases in which
damage cannot be explained by high D/C ratios. This occurs due to the load transfer that
occurs between elements: as one element collapses, load is redistributed within the structure.
Indeed, there is evidence of significant redistributions of loads in these buildings.
As a final and global design verification, nominal interstory drifts ratios and roof dis-
placements were checked for the eight considered buildings (Table 1). The values are listed
in Table 6 and were obtained from the three-dimensional structural models using the reduced
spectrum. Note that the Chilean code (INN 1996) states that the interstory drift ratio at the
center of mass (CM) must be less than 0.002, and that the maximum interstory drift ratio at
any location on the floor plan must not exceed by more than 0.001 that of the CM. Please note
that buildings PP-7 and RT-8 are each composed of two different blocks which are separated
by a construction joint, and hence, the need to differentiate both blocks with indices “a” and
“b” (Westenenk 2011).
In Table 6, δCM_X and δCM_Y are the maximum interstory drift ratios at the CM for the
X and Y directions, respectively; δO_X and δO_Y are the maximum interstory drift ratios
anywhere in the floor plan for the X and Y direction; and CM_X and CM_Y are the roof
displacements for the X and Y directions, respectively.
As it is shown in Table 6, maximum interstory drifts ratios in the Y-direction (δO_Y ) of
building AA-1 are 68 % larger than interstory drifts at the CM in the Y-direction (δCM_Y ) due

123
Bull Earthquake Eng (2013) 11:69–91 89

(a) (b) (c) (d)

Fig. 19 Identification of damaged walls in plan: a damage in 2nd story of CM-3; b D/C ratios in the 2nd
story of CM-3; c damage in 1st story of PR-6; and d D/C ratios in the 1st story of PR-6

Table 6 Interstory drifts and roof displacements in the buildings studied

       
ID δCM_X 0/00 δCM_Y 0/00 δO_X 0/00 δO_Y 0/00 CM_X (cm) CM_Y (cm)

AA-1 0.83 0.81 1.03 1.36 3.15 3.72

AH-2 0.90 0.73 1.14 0.83 3.26 2.59
CM-3 0.60 0.97 0.69 1.13 2.18 3.76
TL-4 0.88 1.17 0.94 1.25 3.01 3.80
PR-6 0.77 0.34 1.06 0.35 1.94 0.91
PP-7a 0.21 0.66 0.25 0.89 0.44 1.34
PP-7b 0.21 0.51 0.23 0.58 0.43 1.02
RT-8a 0.21 0.66 0.25 0.89 0.44 1.34
RT-8b 0.21 0.51 0.23 0.58 0.43 1.02
TO-9 1.10 0.40 1.51 1.09 4.69 0.84

to plan asymmetry. Please note that building CM-3 has very different deformations (in mag-
nitude) in both directions, which is probably due to its narrow plan. Maximum interstory drift
ratios in the Y-direction for this building are similar to those at the CM, implying that there
are no significant torsional coupling effects. For PR-6, the maximum interstory drift ratio
in the X-direction (δO_X ) is 38 % larger than that at the CM (δCM_X ), implying significant
torsional coupling in this direction. Building TO-9 also shows a marked difference in seis-
mic deformations in both directions with a roof displacement of 4.7 cm in the X-direction
(CM_X ) and 0.84 cm in the Y-direction (CM_Y ). In upper stories of this building, there is
a greater difference between the maximum interstory drift ratios and those at the CM for the
X-direction, implying larger coupling in this direction at higher levels. Buildings PP-7a, b,
RT-8a, b showed also very different seismic deformations in both directions. The difference
between maximum interstory drift ratios and those at the CM for buildings PP-7a and RT-8a
is indicative also of torsional coupling. Finally, it is apparent that almost every building with
a predominant shear-wall configuration meets the displacement and drift requirements of the
code; which says in the light of the results of this earthquake, that meeting these requirements

123
90 Bull Earthquake Eng (2013) 11:69–91

is not enough for a no-damage status in the building, especially for irregular buildings in plan
and height.

5 Summary and conclusions

This article summarizes the results of the visual observation of 8 reinforced concrete shear-
wall buildings that were severely damaged during the February 27, 2010 Chile earthquake
in the city of Concepción. Most of the buildings were less than 10-years old.
Damage observed in these buildings was probably the result of multiple physical factors.
There is some positive correlation between damage and building orientation, soil quality,
plan and height building irregularities, slenderness and plan aspect ratio, high levels of com-
pression stresses, lack of boundary element confinement, and small wall thicknesses.
Observed damage in the buildings analyzed was mainly concentrated in the E–W direction,
which coincides usually with the narrower transverse axis of these buildings. It may also
be inferred that changes in building configurations and construction since the 1985 Chile
earthquake, may have affected their seismic performance. For instance, there is a direct
relationship between the observed damage and the presence of horizontal and/or vertical
irregularities or setbacks produced by “flag-shaped” walls, which are frequently found in
the lower stories of buildings that result from vehicular circulations. These cause stress
concentrations that correlate well with the observed damage. Moreover, the unrestrained
value for the axial stresses led to an important decrease in wall thickness, reaching 15 cm
walls, which are very difficult to reinforce and confine properly.
Critical sections at collapsed elements show that damage propagates within the structure
involving multiple stories and elements, thus requiring that the designer verifies the elements
not only as individual units but also as a system of interacting structural components. A good
example for this behavior occurred at PR-6.
The reduction and/or elimination of lintels in Chilean buildings removed a good source of
energy dissipation within the structure, making it necessary to at least reconsider the seismic
reduction coefficient, R = 6, used in ductile wall seismic design. This reduction is also encour-
aged by the observed brittle shear-wall behavior, which is inconsistent with a high R value.
Based on the building-code type analysis developed it is apparent that there are D/C ratios
that greatly exceed one, hence, there is a design deficit. In some cases such deficit matches
the damaged elements, but in others damage cannot be explained only by high D/C ratios.
Although this observation implies that there might be under design and construction issues
in these structures; it is apparent that non-linear analysis would be needed in order to better
understand how damage really propagated within these structures.
It is concluded that design codes must be revised relative to wall design provisions. Those
revisions could include many aspects, from the ground motion characterization in longer
periods to stricter confinement requirements. Two of these aspects were already implemented
in the Chilean code after the 2010 earthquake: (i) to limit the maximum axial loads to a point
well below the point of balance in the axial-moment interaction curve, and (ii) to limit
minimum wall thicknesses in wall boundaries that requires boundary reinforcement.
Although confinement of boundary elements appears as a plausible cause of the brittle
behavior observed in the damaged shear walls, it really needs to be investigated further since
there is reasonable doubt whether well confined boundary elements in walls 15–17 cm thick,
could have performed better and in a ductile manner.
Other aspects such as the consideration in design of the simultaneous action of the different
ground motions components in wall design, horizontal and vertical building irregularities—

123
Bull Earthquake Eng (2013) 11:69–91 91

like it is done by the US code and Eurocode—, contribution and design of the non-structural
components, reincorporation of energy dissipation components such as lintels, and new inter-
story drift limits, may all need to be explicitly considered into the new codes. Furthermore,
seismic codes should try to promote the concept of 3D integral building design, suggesting
procedures to include in design verifications not only individual elements, but more complex
substructures that can develop mechanisms with critical sections crossing several stories.

Acknowledgments The authors are grateful of many students, people and institutions who helped throughout
this study. This research has been funded by the Chilean Fondo Nacional de Ciencia y Tecnología, Fondecyt,
through Grant #1110377. The authors are also grateful to professors: Hernán Santa María, Christian Ledezma,
Raúl Alvarez, José Luis Almazán; engineers: Carlos Medel, Alfredo Payer, Guillermo Gerbaudo, Narciso
Novillo, Mario Alvarez, Roberto Mollinedo, Jorge Quintanilla, Gabriel Sanhueza, Rocío Rivera, Jaime Arria-
gada, Emiliano Pinto, Walter Quintana, Fabiola Santibañez, Jeff Dragovich, Carlos Sempere, Lorena Larrea,
Felipe Cuevas, Cesar Ahumada, Sebastián De la Fuente, Francisco Vega, Germán Pavez, Paula Valderrama;
students: Santiago Brunet, Alan Sternberg; and companies: SIRVE S.A., and DICTUC S.A., for their relevant
support and work.

References

American Concrete Institute ACI (1995) Building code requirements for structural concrete (ACI 318–95)
and commentary. American Concrete Institute, Detroit
Barrientos S (2010) Terremoto Cauquenes 27 Febrero 2010. Servicio Sismológico Universidad de Chile, Chile,
Informe Técnico (in spanish)
ETABS (2011) ETABS, v. 9.6. Computers & Structures, Inc., Berkeley
Instituto Nacional de Normalización, INN (1996) NCh433, Of96. Earthquake Resistant Design of Buildings,
Santiago (in spanish)
Jünemann R, Hube M, De la Llera JC, Kausel (2012) Characteristics of reinforced concrete shear wall buildings
damaged during 2010 Chile earthquake. In: Proceedings 15th world conference of earthquake engineering,
September 24–28, Lisbon, Portugal, Paper No 2265
Massone LM, Bonelli P, Lagos R, Lüders C, Moehle J, Wallace JW (2012) Seismic design and construction
practices for RC structural wall buildings. Earthq Spectra 28(S1):S245–S256
Moehle JP (2010) Briefing on Chile earthquake of February 27, 2010. Earthquake Engineering Research
Institute
Riddell R (1992) Performance of RC buildings in the 1985 Chile earthquake. In: Proceedings of the 10th world
conference on earthquake engineering, vol 8. Madrid, Spain, pp 4251–4256
Riddell R, Wood SL, de la Llera JC (1987) The 1985 Chile earthquake: structural characteristics and damage
statistics for the building inventory in Viña del Mar. Civil Engineering Studies, Structural Research No.
534, University of Illinois, Urbana
Wallace JW, Moehle JP (1989) The 1985 Chile earthquake: an evaluation of structural requirements for bearing
wall buildings. Report No. UCB/EERC-89-05, Earthquake Engineering Research Center, University of
California at Berkeley
Wallace JW, Moehle JP (1992) Ductility and detailing requirements of bearing wall buildings. J Struct Eng
118(6):1625–1644
Westenenk B (2011) Seismic response in RC buildings in Concepción during the February 27. Master of
Science Thesis, Pontificia Universidad Católica de Chile, Chile, Maule Earthquake
Westenenk B, de la Llera JC, Besa JJ, Jünemann R, Moehle J, Lüders C, Inaudi JA, Elwood KJ, Hwang
S (2012) Response of reinforced concrete buildings in Concepción during the Maule earthquake. Earthq
Spectra 28(S1):S257–S280
Wood SL (1991) Performance of reinforced concrete buildings during the 1985 Chile earthquake: implications
for the design of structural walls. Earthq Spectra 7(4):607–638
Wood SL, Wight JK, and Moehle JP (1987) The 1985 Chile earthquake: observations on earthquake-resistant
construction in Viña del Mar. Civil Engineering Studies, Structural Research Series No. 532, University of
Illinois, Urbana

123
View publication stats