Case Studies in Structural Engineering 2 (2014) 1–8

Contents lists available at ScienceDirect

Case Studies in Structural Engineering

journal homepage: www.elsevier.com/locate/csse

A correlation between damage and intensity on old masonry

churches in Colima, Mexico by the 2003 M7.5 earthquake
A. Preciado a,⇑, A. Orduña b
a
Polytechnical University of Guadalajara, Mexico
b
University of Colima, Mexico

a r t i c l e i n f o a b s t r a c t

Article history: Strong damage was observed in Colima Mexico on most of the cultural patrimony (mainly
Available online 27 June 2014 churches) after the 2003 M7.5 earthquake. In order to find a correlation between the
observed damage on the historical buildings and the earthquake intensity, the vulnerability
Keywords: is assessed by qualitative methods, including the vulnerability class method (VCM) and the
Vulnerability vulnerability index method (VIM). The latter method is modified and adapted in this
Damage assessment research to assess the seismic vulnerability of historical buildings such as churches and
Earthquakes
cathedrals located in areas from high to very high seismicity. The results are intended to
Old masonry
Intensity
serve as preliminary indicators of expected damage levels that allow the local authorities
Churches to take measures oriented to disaster prevention. The assessment using both methodolo-
gies is developed on 15 historical masonry churches, most of them from XIX century. With
the results, a correlation between damage and intensity taking into account a Macroseis-
mic Scale is developed and the qualitative methodologies to assess the seismic vulnerabil-
ity of historical constructions are compared each other.
Ó 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Introduction

Most of the historical constructions located in the State of Colima, Mexico are churches, mainly built (or re-built) in the
XIX century. They have the same colonial typology with variations in size and architectural sophistication (see Fig. 1). The
local society has a special interest to preserve this cultural patrimony with its original characteristics, due to the architec-
tural and historical importance that these structures represent for the people and authorities. In the seismological context,
Colima is distinguished by its important exposure, being considered one of the Mexican states under most significant seismic
hazard. The historical constructions belong to the groups more vulnerable to earthquakes (EQ), as demonstrated by the great
damage suffered by this type of constructions (see Fig. 2) during the EQ occurred at January 21st, 2003 (M7.5). The Govern-
ment of Mexico had to invest in expensive works of restoration and rebuilding, generating a restitution of the structural
capacity of the church, and in some cases, increasing their strength. Nevertheless, the safety level of each historical building
repaired or not after the 2003 EQ, and the possible damage scenario at the occurrence of a larger magnitude EQ is completely
unknown. Due to these circumstances, a research project was carried out with the main objective to assess the vulnerability
of the historical constructions in Colima State. The main objective of this project is to obtain indicators of expected damage
levels that allow the local authorities to take measures oriented to disaster of this prevention. In a first phase research, the

⇑ Corresponding author. Address: Polytechnical University of Guadalajara, Km 3.5 Carretera Santa Fe – Tlajomulco, Guadalajara, Mexico. Tel.: +52
3316696432.
E-mail address: [email protected] (A. Preciado).

http://dx.doi.org/10.1016/j.csse.2014.05.001
2214-3998/Ó 2014 The Authors. Published by Elsevier Ltd.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
2 A. Preciado, A. Orduña / Case Studies in Structural Engineering 2 (2014) 1–8

Fig. 1. Plan view and façade of a typical Colima church.

Fig. 2. Observed damages on churches after the 2003 M7.5 Colima, Mexico EQ.

seismic vulnerability of the constructions was assessed through qualitative methods and in a subsequent phase, a correlation
between damage and intensity was carried out with the results. This paper presents the procedure used to assess the seismic
vulnerability of 15 selected historical churches located in Colima by means of qualitative methodologies such as vulnerabil-
ity class method (VCM) and the vulnerability index method (VIM). With the results, a correlation between damage and
intensity taking into account the European Macroseismic Scale (EMS) developed by [1] is carried out. [2] affirm that this scale
represents a very significant effort to relate damage over a series of construction types with different seismic intensities.
Here, damage and intensity have different levels assigned in qualitative terms.

Methodology

The seismic vulnerability of 15 historical masonry churches is assessed in a qualitative way by qualitative methods. The
basis of these methods is the past experience about seismic behavior of different building typologies, and the characteriza-
tion of potential structural deficiencies. The qualitative methods include the vulnerability class method (VCM) and the vul-
nerability index method (VIM). The classification of structures used in the VCM is the EMS [1]. Table 1 presents a summary of
the scale, considering unreinforced masonry only. This proposal assigns a vulnerability class to every type of structure and it
is considered as an efficient technique to assess the seismic vulnerability in a fast and satisfactory way. The EMS, classifies
the different structural typologies in six vulnerability classes going from A to F (A: high vulnerability, F: low vulnerability) in
function of the construction materials used in the building (masonry, concrete, steel, or wood) and the level of seismic
design. For masonry buildings the first three classes A, B and C represent structural typologies such as walls made of rubble
stone, fieldstone, adobe, simple stone, massive stone, unreinforced manufactured units and unreinforced masonry with rein-
forced concrete floors. The D class represents reinforced or confined masonry.
The VIM used in this work is based on the proposed by [3] and [4], for unreinforced masonry buildings. This method
allows the user to identify and characterize the potential structural deficiencies of a building, attributing numerical values
 A. Preciado, A. Orduña / Case Studies in Structural Engineering 2 (2014) 1–8 3

Table 1
Summary of the classifications used in the European Macroseismic Scale [1]. Differentiation of structures (buildings) into vulnerability classes.

Type of structure Vulnerability class

A B C D E F
MASONRY Rubble stone, fieldstone X
Adobe (earth brick) X 1
Simple stone 0 X
Massive stone 1 X 0
Unreinforced, with manufactured units 0 X 0
Unreinforced, with reinforced concrete floors 1 X 0
Reinforced or confined 0 X 1

X: most likely vulnerability class, 1: probable range, 0: range of less probable, exceptional cases.

Table 2
Vulnerability index numerical scale (Iv) for unreinforced masonry buildings [4].

i Parameter Ki A Ki B Ki C Ki D Wi
1 Organization of the resistant system 0 5 20 45 1.0
2 Quality of the resistant system 0 5 25 45 0.25
3 Conventional resistance 0 5 25 45 1.5
4 Position and foundation 0 5 25 45 0.75
5 Horizontal diaphragms 0 5 15 45 1.0
6 Floor configuration 0 5 25 45 0.5
7 Configuration of elevation 0 5 25 45 1.0
8 Maximum separation between walls 0 5 25 45 0.25
9 Typology of the roof 0 15 25 45 1.0
10 Non structural elements 0 0 25 45 0.25
11 Conservation level of the building 0 5 25 45 1.0

(points) to each significant component, and determining with this a seismic vulnerability index. The GNDT method has been
widely used in Italy during last year and it has been upgraded because of the continuous experimentation, resulting in an
extensive database of damage and vulnerability. The parameters showed in Table 2, were compiled in a questionnaire to
be applied during the field research. Based on experience, the questionnaire have undergone modifications, an example of
this is the questionnaire developed by [5]. In this participation, a base questionnaire developed by [5] was used. However,
further modifications were proposed by [6] in order to assess buildings under particular conditions such as historical
buildings of patrimonial importance located in high seismic areas. Those modifications consisted particularly in:

 Parameter 3: corresponding to conventional resistance, the proposal by [7] was adopted.

 Parameter 4: soil types were adjusted to Mexican typical soil types (I, II, and III).
 Parameter 7: configuration of elevation. The ratio between total height (T) and bell tower height (H) was used to
assign a vulnerability index: (A) T/H 6 0.2, (B) 0.2 < T/H 6 0.3, (C) 0.3 < T/H 6 0.5, (D) T/H > 0.5.
 Parameter 9: typology of the roof. The possibility of vaults, cupolas, and other types of heavy masonry roofs
characteristic of historical constructions were included.

The use of Table 2 is simple, during the field research is selected one of the four classes A, B, C, or D (A: low vulnerability,
D: high vulnerability). To every class corresponds a numerical value Ki varying between 0 and 45. In addition, every param-
eter is affected for a coefficient of importance Wi varying between 0.25 and 1.5. This coefficient reflects the importance of
each parameter inside the resistant system of the building according to the opinion of experts. Next, the seismic vulnerability
index (Iv) can be assessed with Eq. (1).

X
11
Iv ¼ Ki  Wi ð1Þ
i¼1

Analyzing Eq. (1), it can be concluded that the vulnerability index defines a scale of values from 0 to the maximum value
382.5. It is divided by 3.825 to obtain a normalized value of vulnerability index, being the rank 0 < Iv < 100.

Seismic vulnerability assessment

As a first phase of the research project, the seismic vulnerability assessment by qualitative methods was carried out in 15
historical churches in the State of Colima using two different approaches, the VCM and VIM. The assessment of the churches
was made for two scenarios, before and after the EQ occurred at January 21st, 2003 (M7.5).
4 A. Preciado, A. Orduña / Case Studies in Structural Engineering 2 (2014) 1–8

Seismic vulnerability assessment of the buildings using the VCM

The implementation of the VCM consisted on a detailed survey of every one of the 15 historical buildings in order to
obtain the vulnerability class related with the structural typology according to Table 1. This assessment was developed
on the basis of the building’s resistant system such as construction materials, assigning to every building one of the vulner-
ability classes A, B, C, D, E, or F, being A the highest vulnerability class and F the lowest. It is very important to mention that
the assessment was carried out taking into account additional information such as plans, construction materials character-
istics, historical analysis (damages and restoration), structural configuration and connection, previous interventions and
building’s conservation level. These parameters allowed taking into account in the assessment the probabilistic ranges
shown in Table 1. The seismic vulnerability of the 15 churches using the VCM was assessed in a first instance before the
2003 EQ (Table 3) to consider a non-damage scenario. The results in Table 3 represent that eight of the churches belong
to the class A, which represents a high vulnerability, subsequently, five churches obtained a class B and the rest belong to
the class C. This means that the churches before the EQ had a deficient conservation level, damages suffered for the churches
in previous EQs, heavy and tall bell towers made of original materials. All these parameters contributed with the high seismic
vulnerability of every building. The seismic vulnerability of the churches using the VCM was assessed in a second stage after
the 2003 EQ (Table 3) to consider a damage scenario. The results showed that five of the churches belong to the most vul-
nerable class, A, eight of the churches obtained a class B and the last 2 a class C. The high vulnerability of the churches
decreased due to the restoration of the churches after the 2003 EQ. The conservation level of the churches improved and
the connection between structural elements, the tall and heavy bell towers of some churches collapsed or were removed
after the EQ and changed for light ones. That improved in a certain way the seismic vulnerability of most of the churches.

Seismic vulnerability assessment of the buildings using the VIM

The seismic vulnerability assessment of the 15 buildings was developed by the VIM. The procedure consisted on a
detailed survey of everyone of the 15 historical buildings to identify and characterize potential structural deficiencies as sta-
ted at the eleven parameters shown in Table 2. For every one of the parameters is assigned one of the four classes A, B, C, or D
(A: low vulnerability, D: high vulnerability) attributing numerical values (points) to each significant component to deter-
mine with Eq. (1) the seismic vulnerability index (Iv). Four of the eleven parameters are not so easy to evaluate during
the field surveys. These parameters correspond to the conventional resistance, floor configuration, configuration of elevation
and maximum separation between walls. To assess them, it is necessary the use computational tools to simplify the work as
AutoCAD to obtain dimensions, elevations of the building, areas of the floors and vertical structural elements located in the X
and Y direction, separation between walls, etc (see Fig. 1). The seismic vulnerability of the churches using the VIM was
assessed before and after the 2003 EQ to consider two damage scenarios. As in the VCM assessment, it was necessary to take
into account additional information of every building such as plans, construction materials characteristics, historical analysis
(damages and restoration), structural configuration and connection, previous intervention data and building’s conservation
level. The vulnerability index (Iv) results obtained for all of the 15 historical churches before and after the 2003 EQ are illus-
trated in Table 3. The results represent that the vulnerability index of the buildings was elevated before the EQ and it reduced
after the restoration works. In most of the churches, some of the potential structural deficiencies were corrected, such as
conservation level, connection between structural elements and heavy bell towers. This restoration contributed to decrease
satisfactory the seismic vulnerability index of most of the buildings.

Table 3
Vulnerability class and index for every church before and after the 2003 EQ.

Name of the building Before the EQ After the EQ

V.C. Iv % V.C. Iv %
Convent Ruins of San Francisco de Almoloyan A 59.48 A 59.48
Chapel of Nuestra Señora de la Asuncion A 46.73 B 41.50
Museum of Regional History of Colima B 22.55 B 22.55
Church of San Felipe de Jesus B 49.67 B 44.44
Church of Nuestra Señora de la Merced B 39.22 B 39.22
Cathedral Basilica Menor de Guadalupe B 48.69 B 48.69
Church of San Pedro Apostol A 55.56 A 43.79
Church of San Miguel del Espiritu Santo A 59.80 A 53.27
Church of Sagrado Corazon de Jesus A 50 B 31.70
Chapel of Virgen del Refugio A 50 B 43.46
Church of San Jeronimo de los Santos Angeles A 48.04 A 41.50
Church of Sagrado Corazon de Jesus B 44.12 B 44.12
Church of San Jose C 54.25 C 54.25
Church of San Francisco de Asis A 50.98 A 45.75
Church of Nuestra Señora de la Salud C 49.67 C 44.44

V.C.: vulnerability class, Iv: vulnerability index %.

 A. Preciado, A. Orduña / Case Studies in Structural Engineering 2 (2014) 1–8 5

Correlation between damage and intensity

The intensities after the 2003 Colima EQ (M7.5) were mainly assessed taking into account reports made by the author-
ities, field inspections made by experts, news, and interviews. The Scale used was the Mercalli Modified Intensity (MMI). The
region with the maximum intensity was the state of Colima (VIII). It is important to mention that intensities bigger than VII
have only occurred approximately seven times in 100 years [8]. The results from the assessment using both methodologies
VCM and VIM on the 15 historical masonry churches are taken into account to develop a correlation between damage and
intensity by means of the EMS. The Scale allows correlating damage with the intensity of an EQ in the MMI Scale. The cor-
relation takes into account the seismic vulnerability class of the building (VCM) and the intensity, assigning an expected
damage based on the classification of observed damage in masonry buildings after an EQ. The observed damage in every
building showed after the 2003 EQ (Table 4) could be categorized taking into account the classification of observed damage
in masonry buildings described in the scale EMS. For the assessment of the observed damage, it was necessary a survey of
every building after the EQ and extra information such as photographs, and general information. The EMS classifies the
observed damage in masonry buildings after an EQ into five categories. Grade 1: negligible to slight damage (no structural
damage, slight non-structural damage); Grade 2: moderate damage (slight structural damage, moderate non-structural dam-
age); Grade 3: substantial to heavy damage (moderate structural damage, heavy non-structural damage); Grade 4: very
heavy damage (heavy structural damage, very heavy non-structural damage); Grade 5: destruction (very heavy structural
damage). The results showed in Table 4 represent the estimated intensity in every building before and after the EQ (VIII);
the observed damage; the vulnerability class assessed by the VCM; and the expected damage taking into account the corre-
lation between intensity and vulnerability. The results showed in Fig. 3 represent the observed damage Vs expected damage
before the 2003 EQ. For a better understanding of the results, the number of coincidences between the number of churches
(scale of the graphic points) with an observed damage and an equal or close expected damage were organized in groups. Two
churches obtained an observed damage 4 and an expected damage 4, four churches an observed damage 2 and an expected
damage 3, five churches an observed damage 3 and an expected damage 4, the rest of the churches were dismissed (just one
equal or close coincidence). Taking into account the results, the authors conclude that for this assessment before the EQ, the
expected damage is one grade over the observed damage. For example if the observed damage of the building is 3, it was
expected to be a damage of Grade 4. Fig. 4 represents the results of the observed damage Vs expected damage after the
2003 EQ. Taking into account the same methodology as in Fig. 3, two churches obtained an observed damage 3 and an
expected damage 3, four churches an observed damage 3 and an expected damage 4, five churches an observed damage
2 and an expected damage 3, the rest of the churches were dismissed (just one equal or close coincidence). Analyzing the
results, the authors conclude that in this assessment after the EQ, the expected damage is one grade over the observed dam-
age. The notable difference between the results before and after the EQ is that the seismic vulnerability of the churches
reduced 1 grade of expected damage due to the restoration campaigns. The assessment of the churches made by the VIM
before and after the 2003 EQ could be correlated with the damage and the intensity of the EQ, taking into account as in
the VCM the observed damage in every building based on the classification of the EMS. The results showed in Table 4 rep-
resent the observed damage in every building after the EQ (intensity VIII) and the vulnerability index percentage. The results
of Fig. 5 represent the observed damage vs. vulnerability index percentage before the 2003 EQ. The average vulnerability
index for the churches with an observed damage 2, is 42.54 (40% of the churches), observed damage 3, 52.57 (46.67% of
the churches) and finally observed damage 4, 52.78 (13.33% of the churches). Taking into account the results, it is worth not-
ing that an observed damage or expected damage of 2 may be presented in churches with a vulnerability index close of 40,

Table 4
Comparison of indicators before and after the 2003 EQ for every church.

Name of the building EQ Before the EQ After the EQ

I
O.D. V.C. E.D. O.D. V.C. E.D.
Convent Ruins of San Francisco de Almoloyan VIII 3 A 4 3 A 4
Chapel of Nuestra Señora de la Asuncion VIII 3 A 4 3 B 3
Museum of Regional History of Colima VIII 2 B 3 2 B 3
Church of San Felipe de Jesus VIII 2 B 3 2 B 3
Church of Nuestra Señora de la Merced VIII 2 B 3 2 B 3
Cathedral Basilica Menor de Guadalupe VIII 3 B 3 3 B 3
Church of San Pedro Apostol VIII 4 A 4 4 A 4
Church of San Miguel del Espiritu Santo VIII 3 A 4 3 A 4
Church of Sagrado Corazon de Jesus VIII 4 A 4 4 B 3
Chapel of Virgen del Refugio VIII 2 A 4 2 B 3
Church of San Jeronimo de los Santos Angeles VIII 3 A 4 3 A 4
Church of Sagrado Corazon de Jesus VIII 2 B 3 2 B 3
Church of San Jose VIII 3 C 2 3 C 2
Church of San Francisco de Asis VIII 3 A 4 3 A 4
Church of Nuestra Señora de la Salud VIII 2 C 2 2 C 2

I: 2003 EQ Intensity, O.D.: observed damage, V.C.: vulnerability class, E.D.: expected damage.
6 A. Preciado, A. Orduña / Case Studies in Structural Engineering 2 (2014) 1–8

OBSERVED DAMAGE VS EXPECTED DAMAGE

4.5

EXPECTED DAMAGE 3.5

2.5

1.5

0.5

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
OBSERVED DAMAGE

Fig. 3. Observed damage vs. expected damage before the 2003 EQ.

OBSERVED DAMAGE VS EXPECTED DAMAGE

4.5

3.5
EXPECTED DAMAGE

2.5

1.5

0.5

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
OBSERVED DAMAGE

Fig. 4. Observed damage vs. expected damage after the 2003 EQ.

an observed damage of 3 for a vulnerability index close of 50, and finally an observed damage of 4 for churches with a vul-
nerability index bigger than 50. The scale of the points (Fig. 5) in the graphics represents the churches that were assessed
with the same vulnerability index and observed damage. The results showed in Fig. 6 represent the observed damage vs.
vulnerability index in percentage after the 2003 EQ. The average vulnerability index for the churches with an observed dam-
age 2 is 39.71 (40% of the churches), observed damage 3 is 49.21 (46.67% of the churches) and finally observed damage 4,
37.75 (13.33% of the churches). Analyzing the results, it is important to note that an observed damage or expected damage
of 2 may be presented in churches with a vulnerability index close of 40, an observed damage of 3 for a vulnerability index
close of 50, and finally an observed damage of 4 for churches with a vulnerability index close of 40. The discontinuity
between observed damage 3 and 4 means that the churches with an observed damage of 4 were the most damaged of all
the evaluated conjunct, and due to this, their bell towers suffered strong damage or collapse. Their vulnerability index
was reduced by the implementation of light materials at the heavy and vulnerable towers.
 A. Preciado, A. Orduña / Case Studies in Structural Engineering 2 (2014) 1–8 7

OBSERVED DAMAGE VS VULNERABILITY INDEX %

70

60

50
VULNERABILITY INDEX %

40

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
OBSERVED DAMAGE

Fig. 5. Observed damage vs. vulnerability index before the 2003 EQ.

OBSERVED DAMAGE VS VULNERABILITY INDEX %

70

60

50
VULNERABILITY INDEX %

40

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
OBSERVED DAMAGE

Fig. 6. Observed damage vs. vulnerability index after the 2003 EQ.

Conclusions

The seismic vulnerability of 15 different churches was evaluated for two damage scenarios (before and after the 2003
M7.5 Colima EQ) under two different approaches, VCM and VIM. It is important to mention that a real vulnerability index
of every building was also available from the observed damage after the EQ, according to the classification of damage in
masonry churches of the EMS. The results obtained by both assessment methods were analyzed and it was concluded that
the VCM allows the user to evaluate the seismic vulnerability in a faster way than the VIM, but the results indicate a global
vulnerability, nevertheless, the VIM allows an identification of the most vulnerable components of a historical building dur-
ing an EQ. Both methods could be used to obtain the vulnerability of a church in a qualitative way, in order to obtain indi-
cators of expected damage. The churches identified with a high vulnerability may be assessed in a quantitative way by
means of more accurate and quantitative methods such as FEM and limit analysis. These refined methodologies allow to
be obtained results more realistic and reliable and a detailed location of damage after a certain EQ, assessing the seismic
8 A. Preciado, A. Orduña / Case Studies in Structural Engineering 2 (2014) 1–8

performance of the building and failure mechanisms. A correlation between both methodologies with damage and intensity
was also possible taking into account the EMS. In the VCM, the expected damage was obtained taking into account the vul-
nerability class of every building. It was observed that it is one grade over the observed damage for both cases, before and
after the EQ. It indicates that an expected damage of a building may be assessed in an accurate way, reducing one grade the
obtained expected damage. Based on the correlation between damage and intensity with the VIM assessed before the EQ, it
could be concluded that an observed damage (or expected damage) of 2 could occur in churches with a vulnerability index
close to 40, damage 3 for an index close to 50, and an damage 4 for churches with a vulnerability index bigger than 50. It
indicates as in the VCM results that a defined expected damage could occur depending of the vulnerability index of the build-
ing. For the VIM assessed after the EQ, the results for observed damage 2 and 3 are similar than in the assessment before the
EQ. The only difference is that churches with an observed damage of 4, the vulnerability index is close to 40. The disconti-
nuity between observed damage 3 and 4 means that the churches with an observed damage of 4 were the most damaged of
all, and due to this, their bell towers suffered strong damage or collapsed, and were substituted for others made by conven-
tional and light materials, reducing their seismic vulnerability index. The authors of this paper consider the results of the
assessment after the EQ using the VIM as not completely reliable, because it could be for necessary another EQ to obtain
a new observed damage in every church and compare the results evaluated with the VIM of the repaired churches.

References

[1] Grünthal G. European Macroseismic Scale EMS 1998. European Council. Notes of the European center of geodynamics and seismology; 1998.
[2] Orduña A, Preciado A, Galván JF, Araiza JC. Vulnerability assessment of churches at Colima by 3D limit analysis models. In: Proceedings of the 6th
international conference on structural analysis of historical constructions (SAHC), Bath, UK, July 2–4; 2008.
[3] GNDT. Seismic risk of public buildings. National Council of Investigation. National Group for the defense of earthquakes (GNDT). Methodology Aspects;
1990.
[4] Benedetti D, Petrini V. Seismic vulnerability of masonry buildings: proposition of an assessment methodology. The Construction Industry, No. 149,
Rome, Italy; 1984.
[5] Aguiar R, Barbat A, Caicedo C, Canas J. Seismic vulnerability of buildings (in Spanish). Monographs of seismic engineering. Barcelona, Spain: Published by
the International Center of Numerical Methods (CIMNE); 1994.
[6] Preciado A. Seismic vulnerability assessment of historical constructions in the state of Colima, Mexico (in Spanish). Master Thesis, Universidad de
Colima, Mexico; 2007.
[7] Astroza M, Moroni MO, Küpfer M. Seismic assessment of brick masonry confined with elements of reinforced concrete. 26th South American Congress
on Structural Engineering, Uruguay, vol. 1; 2006. p. 327–38.
[8] EERI and SMIS. The earthquake of Colima, Mexico occurred at January 21, 2003. SMIS and EERI reports (learning from the earthquakes).