Seismic Vulnerability Assessment of

Low-Rise Reinforced Concrete Buildings

Affected by the 2015 Gorkha, Nepal,
Earthquake
Svetlana Brzev,a) M.EERI, Bishnu Pandey,b) M.EERI,
Dev Kumar Maharjan,c) and Carlos Ventura,a) M.EERI

Low-rise reinforced concrete (RC) frames with brick masonry infill walls up
to five stories high have been used for housing construction in Nepal since the late
1980s. Many buildings of this type were damaged and/or collapsed in the 25
April 2015 Gorkha earthquake (M 7.8), even in areas characterized with moderate
shaking intensity such as Kathmandu Valley. Due to inadequate design and/or
construction of RC frame components, these buildings essentially behave like
masonry shear wall structures with a shear-dominant failure mechanism. The
paper presents the findings of a field survey of 98 RC buildings affected by
the 2015 earthquake. The main objective of the study was to correlate the
observed damage in the buildings using the modified European macroseismic
scale (EMS)-98 and the wall index (defined as the wall area in the direction
of shaking divided by the total building plan area above the level of interest).
The results can be used to help establish recommendations regarding the required
wall index for low-rise RC buildings in Nepal. [DOI: 10.1193/120116EQS218M]

INTRODUCTION
On 25 April 2015, Nepal was affected by a devastating earthquake of magnitude (Mw )
7.8 with the epicenter at Barpak, Gorkha District, a remote hilly area of the country. The
earthquake and subsequent aftershocks caused more than 8,700 fatalities and damaged or
collapsed more than 700,000 buildings, including several UNESCO World Heritage sites
(GoN 2015a, GoN 2015b). The earthquake had a significant impact on housing, institutional
facilities, heritage buildings, schools, hospitals, and lifelines. Most deaths were caused by the
collapse of vulnerable unreinforced masonry dwellings (adobe and stone masonry). Damage
and collapse of several reinforced concrete (RC) buildings revealed the negative effect of
building irregularities and inadequate design and construction practice on the seismic per-
formance of these buildings. RC frame construction is the most prominent building typology
in urban and suburban areas of Nepal. This practice originally started in the late 1970s,

a)
Indian Institute of Technology Gandhinagar, Palaj, Gandhinagar 382355, India
b)
Department of Civil Engineering, British Columbia Institute of Technology, 3700 Willingdon Avenue, Burnaby,
BC, V5G3H2, Canada
c)
National Society for Earthquake Technology-Nepal, Lalitpur, P.O. Box Number 13775, Kathmandu, Nepal
d)
Department of Civil Engineering, the University of British Columbia, 6250 Applied Science Ln, Vancouver, BC
V6T 1Z4, Canada

S275
Earthquake Spectra, Volume 33, No. S1, pages S275–S298, December 2017; © 2017, Earthquake Engineering Research Institute
S276 BRZEV ET AL.

but the rate of RC construction increased after the 1988 Udaipur earthquake (M 6.6), mostly
due to poor performance of unreinforced masonry buildings. RC construction did not experi-
ence significant damage in the 1988 earthquake, but relatively few RC buildings were
exposed to the earthquake (since the construction practice was not common at that time).
An increasing use of RC construction is also associated with economic development in
urban areas and social factors (aspirations of residents to live in RC buildings). Surveys
of building construction in the Kathmandu Valley showed that about 49% of buildings con-
structed in the 1990s were of RC construction, while only 11% of buildings of the 1970s
vintage were of that construction type (JICA 2002). Most RC buildings in urban and sub-
urban areas of Nepal are of low-rise construction. They are used as residential buildings for
extended families, which is a common housing pattern in Nepal. The space at the ground
floor level in these buildings is often used for commercial purposes (small retail stores). Also,
many buildings of this type in the Kathmandu region are used as hostels for workers from
rural areas who have migrated to the capital region.
Due to excessively small RC column size, as well as an inadequate amount and detailing
of reinforcement, RC components in many buildings were not able to effectively resist
earthquake-induced forces due to the 2015 earthquake. These RC frame buildings behaved
essentially like effective shear wall structures, hence their lateral load resistance depended
mostly on the shear capacity of masonry walls. This paper describes the results of a study on 98
RC buildings at three different locations in Nepal that were exposed to the April 2015 earth-
quake and its aftershocks. The main objective of the study was to correlate the observed
damage in individual buildings with the wall index as an indicator of earthquake damage
for low-rise RC buildings with masonry infill walls. A summary of the study and its results
was reported by Brzev et al. (2017a), but this paper presents a more detailed description of the
surveyed buildings and discusses in more detail the concept and application of the wall index
for assessing seismic vulnerability of RC buildings in Nepal.

DESIGN AND CONSTRUCTION PRACTICE

Most low-rise RC buildings are three- to five-story high residential buildings, but some of
these buildings are also used as hotels/hostels or commercial buildings. Many buildings are
of mixed function, with the ground floor used for commercial purposes and upper floors used
for residential purposes. These buildings are known as open storefront buildings and have
one or two open sides in plan, as shown in Figure 1a. Open storefront buildings have a
rectangular plan shape with variable plan dimensions. Typically, stores at the ground
floor level are 3-m-wide rooms separated by brick masonry walls. Fully residential buildings
of this type usually have smaller plan dimensions, with an approximate length of 9 to 12 m
and an approximate width of 6 to 8 m.
Most buildings of this type have one or more structural irregularities. For example, RC
buildings with an open storefront are characterized by a torsional irregularity in plan (due to
the absence of walls on one or two sides). These buildings are also characterized by a weak
story irregularity because the shear capacity of the bottom story is less than the upper stories.
Very often, the top floor in these buildings has a setback with significantly smaller plan area
than the lower floors and is considered a half-floor (see Figure 1b).
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S277

Figure 1. Reinforced concrete (RC) buildings with irregularities: (a) open storefront buildings
and (b) a setback at the top floor level (photos: S. Brzev).

RC buildings have infilled moment frame lateral force-resisting system, but the effect of
masonry infills is often ignored in seismic design. The following two alternative construction
practices are common for low-rise RC buildings in Nepal: (a) RC construction (frames and
floor slabs) is completed first, followed by the construction of masonry infills, and (c) RC
columns are constructed up to the beam soffit level, followed by the masonry infill construc-
tion; subsequently, RC beams (spanning between the RC columns) are constructed on top of
the walls. In the former case, there may be a horizontal gap between the wall and the beam,
while in the latter case there is no gap.
A post-earthquake reconnaissance study after the 2015 earthquake has revealed nonduc-
tile design and construction practice of low-rise RC buildings in the earthquake-affected
areas (Brzev et al. 2017b). This is not a surprise, since most low-rise RC residential buildings
were owner-built and not designed by engineers—these are nonengineered buildings. Even
when engineers were involved, prescriptive design provisions, known as Mandatory Rules of
Thumb NBC 201-1994 (MRT, DUDBC 1994b), have been followed for the purpose of
obtaining building permits from the municipalities. MRT are intended for pre-engineered
design, where the sizes of key structural components, reinforcement details, and standard
design drawings are included. In many cases, reinforcement detailing provisions prescribed
by MRT are not followed at construction sites, as construction quality assurance is nonexis-
tent. Rigorous seismic analysis and design are not required for construction of low-rise RC
buildings up to three-story high with a built-up plinth area less than 92.9 m2 (1000 sq. ft.).
These rules are applicable only to regular buildings, but in practice they have been used for
the design of buildings with various irregularities. In some cases, buildings taller than three
stories were constructed following the same rules. Even when RC frame buildings were
designed by engineers, it is likely that the effect of masonry infills was neglected in the
design.
It is important to acknowledge scale and relevance of seismic vulnerability associated
with nonengineered RC buildings, which are the scope of this paper. A significant
S278 BRZEV ET AL.

fraction of the global building stock (particularly urban and suburban areas in developing
countries) consists of nonengineered RC buildings, which usually do not conform to local
building standards and would be considered significantly deficient per international
standards. Earthquake engineering studies focused on such buildings are highly relevant
due to the extent of their exposure to earthquakes in many countries and the associated
seismic risk.

OBSERVED DAMAGE AND FAILURE MECHANISMS

Many low-rise RC buildings were exposed to the 2015 earthquake and its aftershocks.
Fortunately, most buildings, especially those located in the Kathmandu area, remained
undamaged. This could be expected, based on the available acceleration records, which
showed that the peak ground acceleration (PGA) in Kathmandu was on the order of
0.15 g, that is, significantly less than the design PGA of 0.32 g for the Kathmandu Valley
sites, corresponding to 300-year return period earthquake as prescribed by the Nepali code
NBC 000-1994 (DUDBC 1994d). Many RC buildings were affected by the earthquake,
with the damage extent ranging from minor damage (cracks in the masonry walls and
RC columns) to complete building collapse, particularly in Kathmandu and smaller
communities located closer to the epicenter of either the main shock (25 April 2015)
or the 12 May 2015 aftershock (e.g., Dolakha and Sindupalchok districts). It should be
noted that severely damaged RC buildings in Kathmandu were found in a few localized
areas.
RC frame buildings subjected to severe earthquake ground shaking can experience either
a flexural or a shear failure. A flexural failure mechanism is characterized by the development
of flexural hinges in RC columns and/or beams. Alternatively, RC frames with masonry
infills can experience a shear failure, which is characterized by diagonal shear cracking
of masonry infill walls and adjacent RC columns. Shear failure mechanism may take
place in nonductile RC frames with strong masonry infills (Martín Tempestti and Stavridis
2017). An infill panel can be characterized as “strong” based on the relative stiffness of infill
and adjacent RC columns. Shear failure of RC columns, which is a prerequisite for the shear
failure mechanism, will take place provided that its shear capacity is less than the shear force
corresponding to the column’s flexural capacity (which corresponds to the flexural failure
mechanism). These criteria were established based on the experimental and analytical studies
on RC frames with infills subjected to reversed cyclic loading (Martín Tempestti and
Stavridis 2017, Mehrabi et al. 1994). A recent experimental research study on nonductile
RC frames with infills with shear failure mechanism (Basha and Kaushik 2016) showed
a damage pattern similar to that observed in RC buildings affected by the 2015 Gorkha earth-
quake (Karmacharya et al. 2017). The capacity of an RC frame with a shear failure mechan-
ism is largely governed by the shear capacity of masonry walls. Most extensive damage
usually occurs at the ground floor level of a building where the seismic demand is largest,
and it may lead to the collapse at that level once the base shear capacity is exhausted
(see Figure 2).
A conceptual force-deformation curve (backbone curve) shown in Figure 3 illustrates
a shear-dominant behaviour of RC frame with a masonry infill subjected to a lateral
seismic load; note that the diagram was originally developed for confined masonry
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S279

Figure 2. Shear failure of a RC frame with masonry infill walls: (a) an illustration of the failure
mechanism (Meli et al. 2011) and (b) a shear failure of a RC frame building with masonry infills
due to the 2015 Nepal earthquake (photo: D.K. Maharjan).

Figure 3. Shear failure mechanism in RC frame with masonry infills (Meli et al. 2011).

by Meli et al. (2011). The following two stages are critical for this behavior: (a) an
onset of cracking in the masonry infill (point 1 on the diagram), and (b) the maximum
load-resisting capacity (point 2), which is characterized by extensive diagonal cracking in
the masonry infill and the adjacent RC columns. It is expected that a drop in the lateral
load-resisting capacity will occur after point 2. This is accompanied by increasing lateral
drift and damage before the collapse takes place.
S280 BRZEV ET AL.

DAMAGE CLASSIFICATION
Various approaches for post-earthquake building assessment have been proposed to
determine the severity of damage in structural and nonstructural components and verify struc-
tural integrity after a damaging earthquake (Anagnostopoulous et al. 2004, Baggio et al.
2007, Grünthal 1998, NSET 2009). The findings of post-earthquake damage assessments
influence important decisions, such as whether a building can remain occupied or if it should
be vacated after a damaging earthquake. Also, it is important to determine whether a damaged
building should be repaired and retrofitted or demolished. Damage classification, which char-
acterizes the type and severity of damage, is a critical aspect of post-earthquake damage
assessment. Some publications outline general damage patterns for each damage grade
(Grünthal 1998, NSET 2009), while others offer comprehensive recommendations regarding
the extent of damage, (e.g., size of crack widths in structural components, Anagnostopoulous
2004). The evidence from research studies has also been used to characterize the severity of
damage in structural components of masonry and RC structures (FEMA 1999).
Most damage classifications have identified 3 to 5 Damage Grades (DGs), ranging from
minor damage to total destruction (collapse). These classifications are applicable to various
lateral load-resisting systems (e.g., loadbearing masonry walls or RC frames with masonry
infills). Damage grades for masonry buildings are associated with an increasing extent of
cracking in masonry walls; however, the damage grades in RC buildings with masonry infills
are characterized by the damage both in the RC components and masonry infill walls. Very
few publications recognize the difference between flexural and shear failure mechanisms for
RC frame buildings (Anagnostopoulous et al. 2004). Damage classifications for composite
masonry and RC buildings are presented in Table 1. It can be seen from the table that the
existing damage scales for RC frame structures (e.g., European macroseismic scale [EMS]-92
and EMS-98 scale, Grünthal 1998), describe damage in RC columns mostly due to flexural
behavior. The authors of this paper have proposed a revised damage classification for RC
buildings with a shear-dominant failure mechanism, which is based on the EMS-98 scale. It is
assumed that RC columns experience predominantly shear damage and/or failure due to the
nature of this failure mechanism. RC beams may also experience shear damage and/or failure,
but it is not described in the proposed damage classification. Examples of damage grades are
illustrated in Figures 4 and 5. The authors have found it challenging to make a distinction
between the damage grades 1 and 2 (DG1 and DG2) as defined by the EMS-98 scale. It is
suggested to combine these two damage grades, in a manner similar to a Chilean damage
classification (Moroni et al. 2004).

BUILDING SURVEY
Post-earthquake building surveys are important for assessing the extent and type of
damage in specific building typologies and assessing earthquake intensities. Several
researchers surveyed RC buildings after the 2015 Gorkha earthquake. A survey of more
than 300 buildings (including 76 low-rise RC buildings) was performed to derive macroseis-
mic intensities at several locations (McGowan et al. 2016). A survey of 146 low-rise
damaged RC buildings was performed to document building damage and collect building
plans or primary dimensions (Shah et al. 2017). In the present study, a survey of 98
low-rise RC buildings was performed in July 2015 (less than 3 months after the earthquake).
Table 1. Damage classifications for reinforced concrete frame buildings with masonry infills: shear- and flexure-dominant mechanisms
(RC = reinforced concrete)
Damage Grade
Damage Grade 1 (DG1):
negligible to slight damage
Damage Grade 2 (DG2):
moderate damage
Damage Grade 3 (DG3):
substantial to heavy damage
Damage Grade 4 (DG4):
very heavy damage
Damage grade 5 (DG5):
destruction
a
No damage in RC beams.
b
Many infill walls will have failed at this stage.
Shear-dominant behavior
(proposed damage classification)
Masonry infills
Hairline cracks in very few
walls (plaster cracks only).
Hairline cracks in many walls
(mostly plaster cracks); cracks
along the wall-to-frame
interface; crushing in the
corners of masonry walls.
Diagonal cracks in most walls,
but they are not severe and
there is no sign of fallen
bricks/blocks and wall
segments.
Wide diagonal shear cracks in
walls; severe damage at the
wall intersections; partial
structural failure of roof and
floor structures.
Severe damage and possible
collapse of several walls, usually
at the ground floor level.
Flexure-dominant behavior
(EMS-98; Grünthal 1998)
RC columnsa Masonry infills RC columns and beams
Fine cracks in partitions Fine cracks in plaster over
and infills. frame members at the base.
Plaster cracks in Cracks in partitions and Cracks in columns and beams.
a few RC infill walls; fall of plaster;
columns. fall of brittle cladding and
plaster; falling mortar from
the joints of wall panels.
Visible shear Large cracks in partitions Cracks in columns and beam-
cracks in RC and infill walls; failure of column joints of frames at the
columns. individual infill panels. base; spalling of concrete cover,
buckling of reinforced rods
Wide shear Not describedb Large cracks in structural
cracks and/or elements with compression
tilting of RC failure of concrete and fracture
columns. of rebars; bond failure of beam
reinforced bars; tilting of
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS
columns; collapse of a few
columns or a single upper floor.
Collapse of ground floor or
parts (e.g., wings) of buildings.
S281
S282 BRZEV ET AL.

Figure 4. Examples of Damage Grade 3 (DG3) from Nepal: (a) cracks in a masonry wall and b)
shear cracking in a RC column (photos: S. Brzev).

Figure 5. Examples of Damage Grade 4 (DG4) from Nepal: (a) shear failure of a wall at the
ground floor level of a severely damaged building (photo: S. Brzev) and (b) a vertical separation
crack between the wall and the RC column and a major diagonal shear crack extending from the
wall into the column (photo: B. Pandey).

The following three sites were selected: Sitapaila and Balaju in the Kathmandu Valley and
Batar in Nuwakot District. A map showing the epicenter of the 25 April 2015 and the 12 May
2015 earthquakes and the locations where the building survey was conducted are shown in
Figure 6a. Note that the Kathmandu sites (Sitapaila and Balaju) are about 80 km away (aerial
distance) from the epicenter of the April 2015 earthquake, while the Nuwakot site (Batar) is
about 55 km away from the epicenter. Buildings at all three sites were primarily affected by
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S283

Figure 6. A map showing the damage survey sites and the epicenter locations for the April and
May 2015 earthquakes: (a) a map of Nepal and (b) a map of Kathmandu showing Sitapaila and
Balaju sites (Adhikari et al. 2016).
S284 BRZEV ET AL.

the April 2015 earthquake. Sitapaila and Balaju sites at Kathmandu experienced Modified
Mercalli Intensity (MMI) earthquake intensity of VII (Adhikari et al. 2016), as shown in
Figure 6b. These sites were selected for the survey because of the similar shaking intensity
and reported damage of RC buildings (which was more significant than most other localities
within the Kathmandu Valley). Also, buildings with different damage grades were available
within a small area. In contrast, Batar is a semi-urban area in Nuwakot District away from the
Kathmandu Valley, which was subjected to different ground motion than the Kathmandu
sites. Although there is no reported evidence on shaking intensity at Batar, it is believed
that it could be characterized as MMI VII. This site was selected because the damage of
three- to five-story RC buildings was observed by the authors.
The data collection was performed on a Samsung Galaxy Tab 3 Lite tablet using an
electronic survey form, which was developed in the framework of the Global Earthquake
Model (GEM) for use with the OpenQuake platform (Rosser et al. 2014). Each building
is characterized by its location (latitude and longitude) and 13 attributes describing the details
of the lateral load-resisting system, materials, height, shape of the building plan, type of floor/
roof etc., according to the GEM Building Taxonomy V 2.0 (Brzev et al. 2013). The research
team also took physical measurements of building plan dimensions and wall and column
dimensions. Multiple earthquake damage photographs were taken for each building.
In total, 98 buildings were surveyed. Out of these, 46 buildings were surveyed at Sita-
paila, 21 at Balaju (both in Kathmandu), and 31 at Batar (Nuwakot District); these numbers
correspond to 47%, 21%, and 32% of all surveyed buildings. Therefore, most surveyed build-
ings (68%) were located in Kathmandu Valley. The number of stories varied from two to five;
however, 59 buildings (60%) were three stories high, while an additional 37 buildings (38%)
were two stories high. Only two buildings were taller than three stories (a four- and a five-
story building). Most buildings were of recent vintage, with an average age of 11 years (as of
July 2015). The oldest building was built in 1989, while the most recent one was built
in 2014.
All surveyed buildings were RC buildings with brick masonry infill walls. Most buildings
were characterized by a regular (usually rectangular) plan shape. The south façade of a typical
building surveyed in Sitapaila, Kathmandu is shown in Figure 7a, and its ground floor plan is
shown in Figure 7b; this building was labeled as KAT-SIT-02 in the building survey. Its plan
dimensions (12.50 m
8.08 m length
width) are typical for the surveyed buildings. The
average area of the ground floor plan for the surveyed buildings was 70.3 m. sq., with the
standard deviation of 22.04 m. sq. The plan area of the ground floor in the surveyed buildings
ranged from 22.23 to 175.6 m. sq. Figure 7 shows an example of an open storefront building.
The building had several window openings at the perimeter. The windows were 1,220 mm
high with variable width (either 600 or 1,220 mm). The doors were 2,130 mm high by 780 mm
wide. All exterior walls were 230 mm (one-brick) thick brick masonry, while all interior
walls were 115 mm (half-brick) thick. Typical RC columns were 227 mm square, and
RC beams were 227 mm wide and 305 mm deep. This building was irregular in elevation.
The top floor had a terrace that covered approximately 25% of the plan area, and the south
wall at the same level was offset with regard to the lower floors (see Figure 7c). The damage
was in the form of diagonal cracks at the two bottom stories (the top floor experienced less
damage), and it was classified as DG2 by the survey team. The cracks in mortar joints along
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S285

Figure 7. An example of a surveyed building in Kathmandu: (a) a photo showing the

south façade (entrance), (b) ground floor plan, and (c) top floor plan (size of openings
W1∶ 1.525 width
1.22 m height; D∶ 0.78 m width
2.13 m height; V∶ 0.914 m width

0.78 m height).

the wall-to-frame interface were also observed. The damage mostly occurred in the walls
aligned in the north-south (NS) direction.
It is believed that configuration issues (irregularities in building plan and elevation) and
the quality of masonry construction significantly influenced the extent of damage in these
buildings. This will be illustrated on an example of two three-story buildings in the Sitapaila
area of Kathmandu that were located within 1.62 km aerial distance of one another. Building
KAT-SIT-04, constructed in 1991, was characterized by an irregular plan shape and had a
staircase located at the southeast corner (Figure 8a). The plan dimensions are 6.91 m by
10.155 m. It was observed that the quality of masonry construction was poor, both in
terms of materials and construction practice; for example, very thick mortar joints had
been used, as shown in Figure 8b). Interior walls in the NS direction experienced damage
at all floor levels (note that the damage was more significant at the second than the first-floor
level). A wall adjacent to the staircase (aligned in the east-west [EW] direction) was also
damaged. Based on the observed damage, this building was classified as DG3 in the survey.
Another building (KAT-SIT-10) was constructed in 2006, and it has two identical apartment
units separated by a long wall in the NS direction (Figure 8c). The building plan is almost
S286 BRZEV ET AL.

Figure 8. The effect of configuration irregularities and construction quality: (a) ground floor
plan for KAT-SIT-04, (b) ground floor plan for KAT-SIT-10, (c) a damaged masonry wall in
KAT-SIT-04, and (d) north façade of the building KAT-SIT-10.
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S287

square in shape (with 10.44 by 11.98 m plan dimensions), and it has a centrally located
staircase. There is an open storefront on the north façade at the ground floor level and
large windows along the same façade at the upper floor levels. All walls in the building
are 115 mm (half-brick) thick. The building experienced minor damage in the form of diag-
onal cracks extending from the corners of door openings and was classified as DG1 in the
survey. A north façade of the building is shown in Figure 8d.

WALL INDEX
Wall index (also known as wall density index in Latin American countries) is a measure
of shear capacity of the building for the seismic force direction under consideration. It has
been used for buildings with effective shear wall lateral load-resisting system (such as
masonry and RC walls), and it is usually determined as a ratio of the sum of cross-sectional
areas for all walls aligned in the specific direction of the building plan and the floor plan area.
Wall index has been used for seismic design of low-rise masonry buildings (unreinforced,
reinforced, and confined) in Europe since 1990s (CEN 2004). Several research studies in
countries like Mexico and Chile have confirmed a correlation between the magnitude of
wall index and the extent of earthquake damage in loadbearing masonry and RC shear
wall buildings (Astroza et al. 2012, Moroni et al. 2004). Chilean researchers have correlated
the actual wall index and the observed damage in more than 280 masonry buildings affected
by the 1985 Llolleo, Chile earthquake (M 7.8). The surveyed buildings were of reinforced,
confined, and hybrid masonry construction (consisting of a mix of confined masonry and RC
shear walls). It was concluded that a minimum wall index per floor of 1.15% or higher was
required to minimize chances of significant earthquake damage in these buildings. Masonry
buildings with a wall index per floor in the range from 0.85% to 1.15% suffered moderate
damage, while buildings with a wall index per floor in the range from 0.5% to 0.85% suffered
severe damage. Buildings with a wall index per floor of less than 0.50% suffered heavy
damage, according to the damage classification proposed by the authors (Moroni et al.
2004). A study of masonry buildings affected by the 2010 Maule, Chile earthquake (M
8.8) showed that, in general, confined masonry buildings with a wall index per floor of
0.9% or higher in both directions were considered safe at all earthquake intensities (VI
to IX) observed in the 2010 earthquake, while buildings with a wall index per floor of
less than 0.75% experienced more extensive damage at MSK-64 shaking intensities greater
than VII (Astroza et al. 2012). It was also observed that high wall index in one direction only
may not be sufficient to ensure satisfactory seismic performance. One of the collapsed build-
ings had wall indices of 1.4% and 0.7% in longitudinal and transverse directions respectively.
The building collapsed after experiencing extensive damage in the transverse direction and
losing its bottom floor.
The concept of wall index was also used to develop an approach for assessing seismic
vulnerability of RC frame buildings with masonry infills in Turkey (Gulkan and Sozen 1999,
Hassan and Sozen 1997). Gulkan and Sozen (1999) established a relationship between the
wall and column indices and the drift demand in RC buildings. The column index (CI) was
determined as a ratio of effective column area at the base of the building and the total floor
area. The results of these previous studies have motivated our research team to apply the
concept of wall index in the study described in this paper.
S288 BRZEV ET AL.

Note that the following two different (but interrelated) wall indices can be used:
a. Wall index per floor, WI f , which is based on the ratio of wall area Aw and a plan area
for a typical floor, Ap , that is, WI f ¼ Aw ∕Ap ;
b. Overall wall index, WI, which is the ratio of Aw and the sum of floor plan areas
above the base level of a building, Aptotal , that is, WI ¼ Aw ∕Aptotal . For a regular
n-story building it can be assumed that Aptotal ¼ n · Ap , thus WI ¼ WI f ∕n.
The overall WI accounts for the number of stories in the building, and it needs to be used
when buildings of different heights are compared, which is the case of surveyed buildings
considered in this study.
It should be noted that only shear walls that are a part of the seismic force resisting system
in one principal plan direction are accounted for in calculating the wall area Aw . In this study,
only walls and/or piers with height-to-length ratio of 1.5 or less that are expected to demon-
strate shear-dominant behavior have been considered in the calculations. For example, walls
with openings on the north façade of the building shown in Figure 7b were disregarded in the
wall index calculations because the piers in these walls do not meet the above-mentioned
criteria.
In the study presented in this paper, a WI was calculated for each building in NS and EW
directions. For example, for the building shown in Figure 7, WI for EW direction was
calculated by considering four walls labeled as X1 in Figure 7b, thus the wall area is
AEW ¼ 4ð3.5 · 0.1Þ ¼ 1.40 m2 . On the other hand, the total area of walls in NS direction
can be determined considering walls Y1 to Y4 as follows: ANS ¼ 2ð3.5 · 0.1Þ þ
4.22 · 0.1 þ 1.76 · 0.1 þ 5.23 · 0.23 ¼ 2.90 m2 . The total floor plan area for the building
is Aptotal ¼ 2 · 101.0 þ 55.95 ¼ 257.95 m2 . Note that the first and second floor have the
same plan area of 101.0 m2 (as shown on Figure 7b, while the top floor has a smaller
area [55.95 m2 ], as shown in Figure 7c). Therefore, the WI values are as follows:
WI EW ¼ 1.40∕257.95 ¼ 0.54% WI f EW ¼ 1.40∕101.0 ¼ 1.39%
WI NS ¼ 2.90∕257.95 ¼ 1.12% WI f NS ¼ 2.90∕101.0 ¼ 2.87%
Note that the WI f was determined using the plan area for the first floor (101.0 m2 ), while
the sum of plan areas for all floors (257.95 m2 ) was used to determine the overall WI. It is
interesting to note that the WI values for buildings KAT-SIT-04 and KAT-SIT-10 shown in
Figure 8 are similar for the EW direction (0.41% and 0.50%, respectively). Note that the WI
value for NS direction is significantly lower for KAT-SIT-04 (0.87%) than for KAT-SIT-10
(2.85%). As expected, the former building experienced significant damage in the walls along
the NS direction.

RESULTS AND DISCUSSION

Plan measurements and wall dimensions were recorded for all surveyed buildings, and
it was possible to determine the wall index values for the two principal horizontal directions
in each building. Note that an overall WI has been used for all charts and discussions in
this section. An average wall index, WI, value for all surveyed buildings (98 in total) was
1.38%, with a standard deviation of 1.01% and a coefficient of variation of 0.703. The WI
in the surveyed buildings ranged from 0.19% to 5.65%. Figure 9 shows a histogram
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S289

Figure 9. A histogram showing wall index values for all surveyed buildings (98 total); note that
an average of the two wall index values was taken for each building.

illustrating a distribution of the WI values among the surveyed buildings. It can be seen that
a majority of buildings (53 out of 98, that is, 54%) are characterized by an average WI value
less than 1.36%. Figure 10 shows a histogram that illustrates a distribution of DG for
the surveyed buildings. It can be seen that a majority of the surveyed buildings (51 out
of 98, that is, 52%) sustained only a minor damage (DG1) in the 2015 Gorkha earthquake.
A majority of buildings (around 80%) sustained a minor to moderate damage (DG1 to DG3).

Figure 10. A histogram showing DG distribution for the surveyed buildings (98 total).
S290 BRZEV ET AL.

Figure 11. DG versus wall index in two horizontal directions for the surveyed buildings (sample:
98 buildings - note two WI values per building).

Only a small fraction of the surveyed buildings (three out of 98, that is, 3%) sustained a
significant damage (DG4).
Figure 11 illustrates how surveyed buildings are clustered in DG1 to 4, depending on the
WI; note that the WI values for two horizontal directions in each building are presented on
the chart. The chart shows that the cluster with DG1 is spread over a wide range of WI values,
whereas buildings with other damage grades are clustered within smaller WI ranges.
Figure 12 shows average WI values for the clusters of damaged buildings classified based
on the DG (1 to 4). An average WI value for 51 buildings that experienced DG1 is 1.38%;
this is significantly higher than WI value of 0.97%, characteristic for three buildings that
experienced DG4. This chart again shows a trend of increasing extent of damage (DG)
in buildings with decreasing WI values. Unfortunately, a sample of severely damaged build-
ings (DG4) was relatively small. It is interesting to observe that an average WI value was
similar for buildings classified under DG2 (1.30%) and DG3 (1.28%).
Figure 13 shows average WI values for all surveyed buildings. A unique average WI
value was assigned to each building, based on the WI values corresponding to the two prin-
cipal horizontal plan directions. The values were clustered (grouped) based on the DG, as
shown on the chart. A trend line shows that buildings characterized by a lower average WI
value were more severely damaged (a higher DG was assigned to those buildings).
A significant fraction of the surveyed buildings (46 out of 98) were located in Sitapaila, a
neighborhood in the capital Kathmandu. The trend line for WI variation for buildings in
Sitapaila with different DGs shown by a dashed line in Figure 13—a lower average WI
than the overall building sample (shown by the solid line). The buildings in Sitapaila experi-
enced more significant damage than buildings at most other localities in Kathmandu. This can
be explained by lower WI values and also poor quality of RC construction in the Sitapaila area.
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S291

Figure 12. Average wall index for buildings characterized by different DGs (sample: 98
buildings).

Figure 13. Average wall index values clustered based on the DG (sample: 98 buildings); note
two trend lines—one for all surveyed buildings and other for the Sitapaila site.
S292 BRZEV ET AL.

The authors have reviewed floor plans of the surveyed buildings to assess the effect of
irregularities, such as open storefront and others, which may lead to a soft story collapse. It
was concluded that the buildings that suffered the most extensive damage (DG3 and DG4)
were not characterized by the above-mentioned irregularities. It is interesting to note that the
three most severely damaged buildings, classified as DG4, had a fairly regular plan and con-
tinuous masonry infills up the building height. However, masonry infill walls in these build-
ings were thin (100-mm thickness), and this resulted in low WI values, which likely
contributed to severe damage in these buildings. The mean WI values for the NS and
EW directions were 1.12% and 0.7%, respectively.
The data was also analyzed to understand the effect of RC columns on the extent of
damage in the surveyed buildings. The main indicator is CI, which was determined as
the sum of cross-sectional areas for all columns at the base of the building and the total
floor plan area Aptotal (which was also used to find the WI). An average CI value for all
surveyed buildings was 0.37%, with the standard deviation of 0.169% and the coefficient
of variation of 0.461. The minimum reported CI value is 0.13% and the maximum value is
1.20%. It was observed that there is no significant correlation between the CI and the cor-
responding damage grade for the surveyed buildings. This can be explained by the fact that,
for most of these buildings, masonry walls are the main lateral load-resisting elements, thus
the overall seismic response is governed by the wall shear capacity. It should be noted that
RC column areas were also taken into account in the wall index calculations.

REQUIRED WALL INDEX FOR LOW-RISE RC BUILDINGS WITH

SHEAR-DOMINANT BEHAVIOR IN NEPAL
The results of this study have shown an empirical correlation between the wall index values
and DG for a sample of 98 buildings. It is possible to determine the required wall index for a
building with the given seismic hazard level, type of soil, masonry shear strength, expected
seismic performance (ductility), average story weight, and the number of stories (Brzev et al.
2014, Meli et al. 2011). This procedure can be applied to buildings with regular plan shapes and
wall layout, that is, without significant torsional effects. It is assumed that all walls at a specific
story level have shear-dominant behavior and that they are going to reach their shear capacity
simultaneously. The required wall index for a specific building can be determined from the
fundamental seismic design requirement that the lateral load-resisting capacity at its base, Vc,
should exceed the corresponding seismic demand, that is,
V c ≥ LF · V d
EQ-TARGET;temp:intralink-;e1;41;227 (1)
The seismic demand can be considered equal to base shear force, V d , multiplied by
the applicable load factor, LF, (1.25) when the limit states design approach is used per
NBC 105-1994 (DUDBC 1994c), and
V d ¼ Cd · W
EQ-TARGET;temp:intralink-;e2;41;162 (2)
where C d denotes seismic coefficient and W denotes seismic weight. It is assumed that the
seismic weight of a building with n stories and typical floor plan area, Ap , can be expressed
as follows:
W ¼ n · wavg · Ap
EQ-TARGET;temp:intralink-;sec8;41;98
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S293

where wavg is average floor weight (including the weight of floor slabs and tributary
wall weight). In Nepal, coefficient, Cd , can be determined from NBC 105-1994 as
follows:

Cd ¼ C · Z · I · K
EQ-TARGET;temp:intralink-;sec8;62;603

where C is the basic seismic coefficient (maximum value 0.08 for buildings with low
fundamental period), Z is seismic zoning factor (equal to 1 for Kathmandu Valley
towns), I is building importance factor (equal to 1 for residential buildings), and K is
structural performance factor, which depends on the structural system and expected duc-
tility potential (ranging from 1.0 for ductile moment-resisting frames to 4.0 for structures
of minimal ductility). For low-rise RC buildings located at Kathmandu Valley,
C d ¼ 0.32 based on C ¼ 0.08, Z ¼ 1, and I ¼ 1. The value K ¼ 4 reflects expected
nonductile seismic response of RC frames and significant influence of unreinforced
masonry infills on seismic response of the buildings under consideration.
The seismic capacity at the base of the building, V c , can be assumed to be equal to the
sum of the shear capacities of all walls aligned in the direction of the seismic force. It is
assumed that the shear capacity of wall, i, is equal to the product of its cross-sectional
area, Awi , and the factored shear strength, vr , (when the limit states design approach is
used), that is,

X
n
Vc ¼
EQ-TARGET;temp:intralink-;e3;62;390 V rwi (3)
i¼1

where:
V rwi ¼ vr · Awi
EQ-TARGET;temp:intralink-;sec8;62;322

In some countries, factored shear strength for masonry is prescribed by empirical equa-
tions, but that is not the case in Nepal, where masonry design needs to be performed accord-
ing to NBC 109-1994 code (DUDBC 1994a). In general, masonry shear strength depends on
the masonry compressive strength, f m0 . In the absence of experimental testing, several inter-
national masonry design codes have adopted an equation for estimating masonry shear
strength as a function of f m0 and other factors (CSA 2014, Paulay and Priestley 1992).
The proposed empirical equation is considered conservative since it disregards some poten-
tially influential factors, such as the wall aspect ratio and axial precompression (which causes
an increase in the masonry shear strength):
pffiffiffiffiffi
vr ¼ ð1∕γÞ0.18 f m0
EQ-TARGET;temp:intralink-;e4;62;194 (4)

where γ is partial material resistance factor (taken as 2.0 for masonry). It was not possible to
determine masonry mechanical properties for the buildings considered in the survey. Based
on the local knowledge of brick compressive strengths (ranging from 7.0 to 10.0 MPa) and
typical 1:4 cement:sand mortar mix, it was assumed that the f m0 value is on the order of
3.5 MPa, which results in vr value of 0.17 MPa. It should be acknowledged that the masonry
S294 BRZEV ET AL.

shear strength significantly influences the results, and it is very important to have a good
knowledge of masonry properties.
Finally, the required wall index per floor can be derived from Equations 1 to 3 as
follows:
LF · Cd · n · wavg
W If ≥ (5)
vr
EQ-TARGET;temp:intralink-;e5;41;585

For a three-story building in Kathmandu Valley, the required wall index per floor is equal
to 3.9% (since n ¼ 3); this corresponds to an overall wall index (WI) of 1.3%. It should be
noted that the required values are comparable to the actual wall index values obtained from
the building survey, where an average overall wall index was found to be 1.38% (based on
the sample of 98 buildings). It should be noted that the survey showed a significant variation
in wall index values (from 0.19% to 5.65%).
A probabilistic seismic risk study, motivated by the findings of the current study, exam-
ined the relationship of wall index and the corresponding damage state in low-rise RC build-
ings in Nepal (Karmacharya et al. 2017). In total, 13 cities in Nepal were considered, each
characterized by a different seismic hazard. The study has shown that a wall index value
below 1.0% led consistently to an Average Annual Collapse Probability (AACP) below
an acceptable level (set to 1.0
104 for Nepal).
It is important to discuss acceptable damage in buildings characterized by higher than
required wall index value—assuming that those buildings would comply with the seismic
design requirements of NBC 105-1994. Since the main performance objective of the code is
Life Safety (LS), it can be argued that DG3 would correspond to the LS objective. It is true
that shaking intensities at the surveyed sites were less than expected for the code-prescribed
design earthquake at those sites. For that reason, it would be expected that most buildings
would either remain undamaged or experience minimal damage (corresponding to DG1 or
DG2). As mentioned earlier in the paper, more than 80% of all surveyed buildings experi-
enced damage less than DG3. It should be also noted that most buildings at those sites
remained undamaged, but the survey was limited to damaged buildings.

APPLICABILITY OF THE WALL INDEX CONCEPT

Based on the empirical findings of the building survey and the rational parameters influ-
encing wall index that were discussed in the previous section, applicability of the wall index
concept to the RC building typology considered in this study is discussed below:
1. Effect of building irregularity. It is true that many low-rise RC buildings in Nepal are
irregular in plan and elevation (e.g., open storefront buildings). The resulting torsional
effects cannot be easily accounted for in the wall index approach, but some attempts to
account for this have been made (Porst et al. 2017, Tena-Colunga and Cano-Licona
2010). A consequence of the application of the present wall index approach to irre-
gular buildings with an open storefront is that the extent of damage (damage grade) is
expected to be higher compared with regular buildings with similar wall index. The
authors believe that wall index is still a useful tool for approximate seismic vulner-
ability verification of regular and moderately irregular buildings.
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S295

2. Effect of construction quality. Based on the authors’ assessment, construction qual-

ity in the surveyed buildings (and many other similar buildings) is generally poor.
The masons are usually not properly trained and the quality of masonry materials is
substandard. There is no quality control/supervision by qualified technicians during
the construction. Poor construction quality is not considered to be a major constraint,
since the surveyed buildings are of similar construction quality. But the difference in
construction quality and the corresponding masonry mechanical properties may
impact the ability to compare the findings of this study with other regions charac-
terized by different masonry construction practices and material properties (e.g.,
Latin America).
3. Applicability of the findings for earthquake-affected areas with higher shaking
intensities. The computed wall index values for the surveyed buildings and the cor-
responding damage are reflective of the shaking intensity to which the buildings
were exposed. The authors believe that the surveyed buildings would experience
more significant damage if subjected to higher shaking intensities. It is expected
that the extent of damage would depend on the wall index magnitude, that is, a
building with a higher wall index is expected to experience less damage than similar
building with a lower wall index value. The authors have also presented a rational
approach for estimating the wall index for different seismic hazard levels (and
shaking intensities). Equation 5 in the paper can be used to estimate the required
(recommended) wall index value, and it shows that a higher wall index is required
for buildings at sites with higher seismic hazard (and shaking intensity).
4. Applicability of the findings in other regions of the world. The context of the present
study is different from previous studies involving wall index from other countries
(e.g., Latin America), both in terms of the shaking intensity and the masonry
mechanical properties (due to different masonry materials and construction practices).
It is important to note that international standards have significantly different require-
ments with regard to the minimum acceptable masonry material properties, such as
brick compressive strength. These minimum requirements vary between design codes,
and may be very low in developing countries. Since brick compressive strength
significantly influences masonry compression and shear strength, this also limits
an ability to compare wall index values and the corresponding damage in masonry
buildings from different countries and regions of the world, even when these buildings
are exposed to similar shaking intensities.
CONCLUSIONS
This paper presents the results of one of the first reported post-earthquake surveys of wall
index characteristics in low-rise RC buildings and the corresponding damage grade after an
earthquake in Nepal and neighboring countries (India, Pakistan). The survey of 98 low-rise
RC buildings that experienced damage due to the 2015 Gorkha, Nepal earthquake has shown
a strong relationship between the extent of damage and the corresponding wall index
characteristics, that is, buildings with lower wall index suffered more extensive damage.
The buildings demonstrated shear-dominant behavior that was governed by the masonry
wall shear capacity, as opposed to the flexural capacity of RC frames. The relationship between
the CI and the extent of damage in these buildings was found to be weak. Based on the limited
S296 BRZEV ET AL.

data considered in this study, it can be concluded that wall index can be used as an indicator of
seismic vulnerability for existing RC buildings with masonry walls that are characterized by the
shear-dominant behavior. The wall index can also be used in preliminary design of new RC
buildings with masonry infill walls with the shear-dominant behavior. The required wall index
value for a specific building can be calculated for the given seismic hazard and soil type
parameters, number of stories, and masonry mechanical properties.

ACKNOWLEDGMENTS
The authors are grateful for the funding provided by the 12th CMS Education and
Research Fund and the support provided by Mr. Bill McEwen, Executive Director of
Masonry Institute of BC, Canada. The study was initiated during a post-earthquake recon-
naissance visit to Nepal in June 2015 that was sponsored by the Canadian Association for
Earthquake Engineering, the British Columbia Institute of Technology, and the UBC Earth-
quake Engineering Research Facility. The survey was performed by Mr. Rudra Bahadur Oli
and Mr. Kabin Sapkota, and their contribution to the study is gratefully acknowledged. The
support of NSET, Nepal is greatly appreciated. The authors are grateful to Ms. Natalia
Leposavic for preparing the drawings for the paper.

REFERENCES
Adhikari, S. R., Guragain, R., and Murakami, H., 2016. Study on shaking intensity distribution of the
2015 Gorkha earthquake in Nepal. Proceedings from the International Workshop on the First
Memorial Day of 25 April, 2015, Gorkha Earthquake, 24–25 April 2016, Kathmandu, Nepal.
Anagnostopoulos, S. A., Moretti, M., Panoutsopoulou, M., Panagiotopoulou, D., and Thoma, T.,
2004. Post Earthquake Damage and Usability Assessment of Buildings: Further Development
and Applications, Final Report, European Commission-D.G. Environment, and Civil Protec-
tion EPPO, Patras, Greece.
Astroza, M., Moroni, O., Brzev, S., and Tanner, J., 2012. Seismic performance of engineered
masonry buildings in the 2010 Maule earthquake, special issue on the 2010 Chile earthquake,
Earthquake Spectra 28(S1), S385–S406.
Baggio, C., Bernardini, A., Colozza, R., Corazza, L., Della Bella, M., Di Pasquale, G., Dolce, M.,
Goretti, A., Martinelli, A., Orsini, G., Papa, F., and Zuccaro, G., 2007. Field Manual for
Post-Earthquake Damage and Safety Assessment and Short Term Countermeasures
(AeDES), European Commission, Joint Research Centre, Institute for the Protection and Secur-
ity of the Citizen, Italy.
Basha, S. H., and Kaushik, H. B., 2016. Behavior and failure mechanisms of masonry-infilled RC
frames subject to lateral loading, Engineering Structures 111, 233–245.
Brzev, S., Pandey, B., Maharjan, D. K., and Ventura, C., 2017a. Seismic vulnerability index for
low rise composite reinforced concrete and masonry buildings in Nepal, paper no. 3522,
in Proceedings of 16th World Conference on Earthquake Engineering, 9–13 January 2017,
Santiago, Chile.
Brzev, S., Pandey, B., and Pao, J., 2017b. Urban Housing: Performance of Reinforced Concrete
Buildings, Chapter 5, The April 25, 2015 Gorkha, Nepal Earthquake Reconnaissance Report,
Canadian Association for Earthquake Engineering, Vancouver, BC, Canada.
Brzev, S., Perez Gavilan, J. J., and Rangwani, K., 2014. Simplified method for seismic design of
confined masonry buildings, in Proceedings Short Course on Seismic Design of Reinforced
and Confined Masonry Buildings, Indian Institute of Technology Gandhinagar, India.
SEISMIC VULNERABILITY ASSESSMENT OF LOW-RISE REINFORCED CONCRETE BUILDINGS S297

Brzev, S., Scawthorn, C., Charleson, A. W., Allen, L., Greene, M., Jaiswal, K., and Silva, V.,
2013. GEM Building Taxonomy Version 2.0, GEM Technical Report 2013-02 V1.0.0, GEM
Foundation, Pavia, Italy.
CEN, 2004. Eurocode 8: Design of Structures for Earthquake Resistance, Part 1: General Rules,
Seismic Actions and Rules for Buildings, EN 1998-1:2004, European Committee for
Standardization (CEN), Brussels, Belgium.
CSA, 2014. CSA S304 Design of Masonry Structures, Canadian Standards Association,
Mississauga, ON, Canada.
Department of Urban Development and Building Construction (DUDBC), 1994. NBC 109-
1994 Masonry: Unreinforced, Nepal National Building Code, Kathmandu, Nepal.
Department of Urban Development and Building Construction (DUDBC), 1994a. NBC
201-1994 Mandatory Rules of Thumb - Reinforced Concrete Buildings with Masonry Infill,
Nepal National Building Code, Kathmandu, Nepal.
Department of Urban Development and Building Construction (DUDBC), 1994b. NBC 105-1994
Seismic Design of Buildings in Nepal, Nepal National Building Code, Kathmandu, Nepal.
Department of Urban Development and Building Construction (DUDBC), 1994c. NBC 000-1994
Requirements for State-of-the-Art Design: an Introduction, Nepal National Building Code,
Kathmandu, Nepal.
Federal Emergency Management Agency (FEMA), 1999. Evaluation of Earthquake Damaged
Concrete and Masonry Wall Buildings - Basic Procedures Manual, FEMA 306,
Washington, DC.
Government of Nepal (GoN), 2015a. Nepal Earthquake 2015 – Post Disaster Needs Assessment,
Volume A: Key Findings, Government of Nepal, National Planning Commission, Kathmandu,
Nepal.
Government of Nepal (GoN), 2015b. Nepal Earthquake 2015 – Post Disaster Needs Assessment,
Volume B: Sector Reports, Government of Nepal, National Planning Commission,
Kathmandu, Nepal.
Grünthal, G. (ed.), 1998. European Macroseismic Scale 1998 (EMS-98), Centre Europèen de
Géodynamique et de Séismologie, Luxembourg.
Gulkan, P., and Sozen, M. A., 1999. Procedure for determining seismic vulnerability of building
Structures, ACI Structural Journal 96(3), 336–342.
Hassan, A. F., and Sozen, M. A., 1997. Seismic vulnerability assessment of low-rise buildings in
regions with infrequent earthquakes, ACI Structural Journal 94(1), 31–38.
Japan International Cooperation Agency (JICA), 2002. The Study on Earthquake Disaster
Mitigation in the Kathmandu Valley, The Final Report, Tokyo, Japan.
Karmacharya, U., Silva, V., Brzev, S., and Martins, L., 2017. Improving the Nepalese building
code based on lessons learnt from the 2015 M7.8 Gorkha earthquake, Impacts and Insights of
Gorkha Earthquake in Nepal, Elsevier (under review).
Martín Tempestti, J., and Stavridis, A., 2017. Simplified analytical method to predict the failure
pattern of infilled RC frames, paper no. 1263, Proceedings of the 16th World Conference on
Earthquake Engineering, Santiago, Chile.
McGowan, S. M., Jaiswal, K. S., and Wald, D. J., 2016. Using structural damage statistics to
derive macroseismic intensity within the Kathmandu Valley for the 2015 M7.8 Gorkha, Nepal
earthquake, Tectonophysics. doi:10.1016/j.tecto.2016.08.002.
S298 BRZEV ET AL.

Mehrabi, A. B., Shing, P. B., Schuller, M. P., and Noland, J. L., 1994. Performance of Masonry-
Infilled R/C Frames Under In-plane Lateral Loads. Report No. CU/SR-94/6, University of
Colorado at Boulder, CO.
Meli, R., Brzev, S., Astroza, M., Boen, T., Crisafulli, F., Dai, J., Farsi, M., Hart, T., Mebarki, A.,
Moghadam, A. S., Quiun, D., Tomazevic, M., and Yamin, L., 2011. Seismic Design Guide
for Low-Rise Confined Masonry Buildings, Earthquake Engineering Research Institute,
Oakland, CA.
Moroni, M. O., Astroza, M., and Acevedo, C., 2004. Performance and seismic vulnerability
of masonry housing types used in Chile, Journal of Performance of Constructed Facilities
18, 173–179.
National Society for Earthquake Technology-Nepal (NSET), 2009. Seismic Vulnerability Eva-
luation Guideline for Private and Public Buildings, Part II: Post Disaster Damage Assessment,
Kathmandu, Nepal.
Paulay, T., and Priestley, M. J. N., 1992. Seismic Design of Concrete and Masonry Buildings,
John Wiley and Sons, Inc., New York, NY, 744 pp.
Porst, C., Brzev, S., and Oschendorf, J., 2017. Confined masonry for resilient low-cost housing in
India: a design and analysis method, Proceedings of the 16th World Conference on Earth-
quake Engineering, paper no. 3876, Santiago, Chile.
Rosser, J., Morley, J. G., and Vicini, A., 2014. User Guide: Android Mobile Tool for Field Data
Collection, GEM Technical Report 2014—03 V1.0.0, GEM Foundation, Pavia, Italy.
Shah, P., Pujol, S., Kreger, M., and Irfanoglu, A., 2017. 2015 Nepal Earthquake, Concrete Inter-
national 39(3), 42–49.
Tena-Colunga, A., and Cano-Licona, J., 2010. Simplified method for the seismic analysis of
masonry shear-wall buildings, Journal of Structural Engineering 136(5), 511–520.
(Received 1 December 2016; accepted 30 October 2017)