See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/271833055

FRAGILITY ASSESSMENT OF PRE-SEISMIC

CODE BUILDINGS AND EMERGENCY
FACILITIES IN THE UAE

CONFERENCE PAPER · AUGUST 2014

CITATIONS READS

2 14

2 AUTHORS:

Anas Issa Aman Mwafy

United Arab Emirates University United Arab Emirates University
3 PUBLICATIONS 2 CITATIONS 34 PUBLICATIONS 382 CITATIONS

SEE PROFILE SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Anas Issa
letting you access and read them immediately. Retrieved on: 28 December 2015
FRAGILITY ASSESSMENT OF PRE-SEISMIC CODE BUILDINGS AND
EMERGENCY FACILITIES IN THE UAE
Anas ISSA1 and Aman MWAFY2

ABSTRACT
High levels of investments and population densities are represented in the building inventory in the
UAE. The number of buildings in the existing inventory that may be at risk because of the inefficient
seismic design provisions at the time they were constructed cannot be underestimated. Moreover, a
crucial role in the recovery period following an earthquake is played by emergency facilities. Hence,
this study focuses on the seismic fragility assessment of buildings representing substandard and
emergency structures when subjected to potential earthquake scenarios in the UAE. Nine reference
buildings representing the building inventory in a highly populated seismically active area in the UAE
are selected based on an on-ground survey. Detailed structural design and fibre-based modelling are
carried out for the reference structures, and forty earthquake records are selected to represent potential
earthquake scenarios in the study area. A wide range of local and global response parameters are
monitored from static pushover and incremental dynamic analyses. Previous studies are reviewed and
compared with the performance criteria obtained from the present study. Fragility relationships and
damage state probabilities are generated for the nine reference structures under the effect of two seismic
scenarios for the direct implementation in a loss estimation system for the UAE. While the results reflect
the acceptable performance of emergency facilities, particularly under near-field events, they reveal the
need for implementing mitigation measures for pre-code structures to reduce seismic losses.

INTRODUCTION
The United Arab Emirates (UAE) was known to be a region of low seismicity even though being in
close proximity to high seismic zones. Recent earthquakes recorded within the region have emphasized
the significance of investigating the impacts of anticipated seismic scenarios on different types of
structures. Due to the rapid development of the construction industry in the UAE, significant attention
has been drawn to the risk posed by earthquakes to the buildings and infrastructure. More importantly,
since pre-seismic code buildings were constructed when the knowledge regarding the seismicity and
seismic code regulations was limited, they may be vulnerable. This knowledge has advanced since then
and reflected on the improved seismic performance of modern code-design structures. The poor
performance of pre-seismic code buildings will prevent such structures from reaching the desired
seismic performance objectives (ATC, 1996; Buratti et al., 2010). In addition, serious attention should
be paid to the emergency facilities to assess their performance as they play a crucial role in the
earthquake recovery period and should be fully functioning. A great attention has been paid to the
emergency facilities around the world. For instance, the State of California has taken steps in that

1
Graduate Teaching Assistant, United Arab Emirates University, Al Ain, UAE, [email protected]
2
Associate Professor, United Arab Emirates University, Al Ain, UAE, [email protected]
also Professor of Structural Engineering, Zagazig University, Egypt (on leave)

1
direction by enacting a decree to retrofit emergency facilities by 2030 to allow them to be fully
operational subsequent an earthquake (Bruneau and Reinhorn, 2007).
Catastrophic consequences are expected if the seismic risk of the building stock is neglected,
particularly for substandard and emergency facilities. The repeated reports of seismic activities in the
UAE and the devastating social and economic impacts of strong seismic events that have been recorded
in neighbouring territories emphasize the significance of adopting reliable loss estimation and hazard
mitigation strategies for the built environment in this region. This study thus focuses on the probabilistic
seismic vulnerability assessment of buildings representing pre-seismic code and emergency structures,
which enables the direct implementation of a wide range of fragility relationships representing different
structures and earthquake scenarios in a loss estimation and mitigation system for the UAE.

REFERENCE STRUCTURES AND GROUND MOTIONS

Selection of Earthquake Records based on Possible Seismic Scenarios:
A diverse set of earthquake records representing the study area are selected for inelastic dynamic
simulations. Both the PEER Ground Motion Database (Chiou and Youngs, 2008) and the European
Strong-Motion Database (Ambraseys et al., 2004) are thoroughly searched to select 40 natural records
that represent two earthquake scenarios recommended for the study area, namely far-field records and
near-source events (Mwafy et al., 2006). The main criteria of selecting input ground motions are: (i)
epicentral distance, (ii) magnitude, (iii) soil condition, (iv) a/v ratio, and (v) PGA. Table 1 shows a
summary of the selected near source records. The response spectra of the selected records are developed
and scaled to the recommended design intensity of the study area (i.e. a PGA of 0.16g; Mwafy et al.,
2006). Figure 1 compares the response spectra of the 40 input ground motions that represent the two
seismic scenarios with the current design spectra for Dubai for soil classes C and D.

Table 1. Summary of near-field earthquake records

Ref Earthquake Station Comp. Date Mag. Site Ep. Dist. Duration PGA a/v a/v
(Mw) class (km) (sec) (m-s2) g-ms-1 class.
NF01 Coyote Lake San Juan Bautista, 24 polk St 213 06-08-1979 5.74 v. dense 19.7 15 0.991 1.424
NF02 Livermore-02 Livermore-Morgan Terr Park 355 27-01-1980 5.42 v. dense 14.1 14 2.235 2.581
NF03 Hollister-04 City Hall 271 28-11-1974 5.14 v. dense 9.8 21 1.651 1.480
NF04 Whittier Narrows-01 Brea Dam (L Abut) 130 10-01-1987 5.99 v. dense 24.0 26 1.299 1.460
NF05 Whittier Narrows-01 LA-Centry City CC North 90 10-01-1987 5.99 stiff 29.9 9.9 0.851 1.788
NF06 Whittier Narrows-01 LB-Orange Ave 2280 10-01-1987 5.99 stiff 24.5 20 2.111 1.468
NF07 Northridge-06 Panorama City-Roscoe 90 20-03-1994 5.28 stiff 11.8 30 1.141 1.916
NF08 Montenegro Petrovac-Hotel Oliva Y 15-04-1979 5.80 v. dense 24.0 15 0.873 1.426
NF09 Umbria Ma. Castelnuovo-Assisi NE 26-09-1997 6.04 v. dense 16.0 12 1.600 1.254
NF10 Lazio Abr. Y Cassino-Sant Elia EW 07-05-1984 5.93 v. dense 16.0 11 1.123 1.590
NF11 Mammoth Lakes-02 Mammoth Lakes H. S. 344 25-05-1980 5.69 v. dense 3.49 28 4.064 1.957
NF12 Mammoth Lakes-06 Fish & Game (FIS) 0 27-05-1980 5.94 stiff 12.02 7.2 3.979 2.753 high
NF13 Coalinga-04 Anticline Ridge Free-Field 270 07-09-1983 5.18 v. dense 6.34 45 3.220 2.048
NF14 Coalinga-04 Anticline Ridge Pad 270 07-09-1983 5.18 v. dense 6.34 20 3.246 2.350
NF15 Coalinga-05 Burnett Construction 360 22-07-1983 5.77 stiff 12.38 30 2.915 1.988
NF16 Whittier Narrows-01 Alhambra - Fremont School 180 10-01-1987 5.99 v. dense 6.77 21 3.806 1.514
NF17 Whittier Narrows-01 Garvey Res. - Control Bldg 60 10-01-1987 5.99 v. dense 2.86 26 3.775 2.432
NF18 Whittier Narrows-01 LA - 116th St School 360 10-01-1987 5.99 stiff 21.26 38 3.343 1.888
NF19 Whittier Narrows-01 LA - Obregon Park 360 10-01-1987 5.99 stiff 9.05 25 4.161 1.748
NF20 Friuli Breginj-Fabrika IGLI Y 15-09-1976 6 v. dense 21.0 30 4.956 2.333
a/v: PGA/PGV, a/v classification (<0.8 Low & >1.2 high), Vs30 of very dense soil = 360-760 m/s, and for stiff soil = 180-360 m/s

0.9 0.8
0.8 Natural Records 0.7 Natural Records
Design Code Spectrum (D) Design Code Spectrum (D)
0.7 0.6
Design Code Spectrum (C)
Spec. Acc. (g)
Spec. Acc. (g)

0.6 Mean 0.5 Design Code Spectrum (C)

0.5 Mean
0.4
0.4
0.3
0.3
0.2 0.2
0.1 0.1
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.5 1 1.5 2 2.5 3 3.5 4
Period (s) Period (s)
Figure 1. Response spectra of near-field (left) and far-field (right) earthquake records along with their mean and
the current design spectra for Dubai for soil classes C and D

2
 A.Issa and A.Mwafy 3

Selection of Pre-seismic Code and Emergency Structures:

A fundamental step in this study is selecting representative structures for the study area. On-ground
surveying was carried out with several site visits for Dubai, Sharjah and Ajman, which is a highly
populated and seismically active area in the UAE (Figure 2), to select a set of nine reference structures
representing the pre-seismic code buildings and emergency facilities (Mwafy, 2012). Table 2 shows a
summary of the selected representative buildings. Three-dimensional (3D) finite element (FE) models
are developed for the investigated buildings using the structural analysis and design program ETABS
(CSI, 2011). Figure 3 depicts the layouts and 3D models for the set of the reference buildings. For the
pre-code wall structures, the height of all stories is 3.2 m except for the ground storey, which is 4.5 m.
For the pre-code frame structures, the height of all stories is 3.5 m except for the ground storey, which
is 5.0 m. The total heights of the nine reference buildings are shown in Table 2.

Point North East

Ajman H
o o
A 25 04’ 17’’55 07’ 30’’
Sharjah
B 25o 03’ 40’’55o 08’ 15’’
G
C 25o 05’ 48’’55o 10’ 57’’
F
D 25o 04’ 35’’55o 12’ 43’’ E
o o
E 25 12’ 25’’55 18’ 57’’ Dubai
F 25o 13’ 47’’55o 17’ 26’’
G 25o 17’ 23’’55o 21’ 33’’ A C
H 25o 25’ 41’’55o 31’ 14’’ B D
Figure 2. Study area

Table 2. Summary of the selected buildings

Number Buildings description Building Reference Building Height (m)
Selection according to construction date
1 BO-02 8
2 BO-08 28.5
3 Pre-seismic code structures BO-18 58.9
4 BO-26 84.5
Category
5 BO-40 129.3
Selection according to function
6 Fire station FS 9.5
7 Police station PS 9
8 School SC 13
9 Hospital HO 24.5

P
Pre-code Structures

CORE
BO-02, 2 story BO-08, 8 story BO-18 & BO-26,18 & 26 story BO-40, 40 story
Emergemcy

FS, Fire Station PS, Police Station SC, School HO, Hospital

Figure 3. Three-dimensional design models

Design of Reference Buildings:
The five pre-seismic code reference buildings are designed and detailed specifically for the purpose of
the current study according to the building codes that were implemented at the construction time
(BS8110, 1986). Pre-seismic code buildings are defined in this study as those built in the study area
before 1990 when seismic design provisions might be disregarded. It is worth noting that the Uniform
Building Code (UBC, 1997) recommended the use of Seismic Zone ‘0’ for the cities of Abu Dhabi and
Dubai, UAE. Revised seismic design criteria have been adopted in the UAE after the recommendations
of recent seismic hazard studies (e.g. Abdalla and Al-Homoud, 2004; Mwafy et al., 2006). Therefore,
wind loads are the only lateral loads considered in design to represent the real situation before 1990.
The permanent loads used in design include the self-weight and a superimposed dead load of 4.0 kN/m2.
The live load is 2.0 kN/m2 except for stairs and exit ways, which is 4.8 kN/m2. Wind loads are calculated
based on a basic wind speed of 45 m/s (BS-6399-2, 1998). The design is carried out for each building
carefully to obtain the optimum cross sections for different structural elements. Table 3 shows sample
design results for the 26-story building. To accurately represent the pre-code structures, the material
properties that were utilized at the time of construction are considered. The concrete strength ranges
between 20 and 40 MPa, while mild steel is used with a yield strength of 240 MPa. Pre-code buildings
with such material properties are likely to be more vulnerable to earthquake loads due to the large cross
sections, heavy mass and inadequate detailing as compared to modern code-designed structures.
The four emergency buildings are designed and detailed in the current study according to modern
building codes (ACI-318, 2011; ASCE-7, 2010). Wind loads are estimated using ASCE-7 (2010) based
on a basic wind speed of 45 m/s and an exposure category C. The seismic loads are also estimated using
ASCE-7 (2010) with a soil class C, as per the recommendations of recent hazard assessment studies and
design provisions for the UAE. The adopted live load is 3.80 kN/m2, while the permanent loads used in
design include the self-weight of structural members in addition to a uniformly distributed load on slab
of 4.0 kN/m2, which is intended to account for superimposed dead loads. Taking into consideration the
moderate seismicity of the UAE, intermediate moment resisting frames are employed in the design of
such buildings. The design information is used in the numerical modelling phase, as explained below.

Table 3: Design summary of the vertical structural members of the 26-story building
Pier P
Location of section Foundation Floor no. 3 Floor no. 8 Floor no. 12 Floor no. 17
Vertical steel ratio, % 3.9 3.11 1.72 1.34 1.00
VL. Reinforcement 66#40 52#40 42#32 36#26 36#20
HL. Reinforcement #12@200mm #12@200mm #8@200mm #8@200mm #8@200mm
Links #12@510mm #12@510mm #8@380mm #8@340mm #8@260mm
Design/Capacity (D/C) Ratio 0.998 0.996 0.995 0.922 0.969
Pier section, mm x mm 600x3500 600x3500 550x3500 550x3500 400x3500
Concrete strength (fc'), MPa 28 24 24 24 20
CORE
Location of section Foundation Floor no. 8 Floor no. 17
Vertical steel ratio 1.1% 1.0% 1.0%
Vertical Reinforcement 160#20 121#20 97#20
Horizontal Reinforcement #10@200mm #8@200mm #8@200mm
links #10@430mm #8@300mm #8@190mm
Design/Capacity (D/C) Ratio 0.999 0.849 0.572
Core thickness, mm 300 250 200
Concrete strength (fc'), MPa 28 24 20

Modelling and Verifications:

The fibre modelling technique is used to idealize the reference structures for the multi-degree-of-
freedom Inelastic Pushover Analysis (IPA) and Incremental Dynamic Analysis (IDA) using the inelastic
analysis platform Zeus-NL (Elnashai et al., 2012). Reinforcing steel, confined and unconfined concrete
are effectively idealized to enable monitoring of stresses and strains during the inelastic simulations.
Figure 4 illustrates the numerical models of the selected buildings. Each structural member is divided
into three segments to allow for more accurate representation of longitudinal reinforcement and
prediction of steel yielding, concrete crushing, and shear capacity of structural members (Elnashai and
Mwafy, 2002). The verifications of numerical models is conducted by comparing the periods of

4
 A.Issa and A.Mwafy 5

vibration of different models used in both design and inelastic analyses. Table 4 summarizes the obtained
periods of the nine reference buildings as well as those associated with the first two mode shapes for a
sample building (HO). It is clear that the periods of the design models are slightly longer (within 10%
difference) than the periods of the fibre-based models. This difference is justified by the consideration
of actual material strength values and steel reinforcement in the fibre-based models, which increase
stiffness and shorten periods. These sample results verify the numerical models and lend weight to the
results obtained from the present study.

Slab

Core wall
Wall Rigid arm

a. Pre-seismic code structures b. Fiber-based model (Elnashai and Mwafy, 2002) c. Critical Facilities
Figure 4. Fiber-based numerical models developed for two set of reference buildings

Table 4. Summary of buildings fundamental periods (T1) from fibre-based and design models
Building T1, Fibre-based Models T1, Design models
BO-02 0.780 0.880
BO-08 1.344 1.396
BO-18 1.432 1.572
BO-26 2.370 2.457
BO-40 3.901 3.755
FS 0.746 0.764
PS 0.656 0.702
SC 0.817 0.891 Fiber-based model Design model
HO 1.294 1.365 First two mode shapes for the Hospital (HO)

SELECTION OF LIMIT STATES

Seismic performance criteria of structures, which are related to the level of non-structural and structural
damage, have received focused attention. The performance levels considered in this study are as follows
(ASCE-41, 2007; SEAOC, 1995):
(i) Collapse Prevention (CP): which allows for a small margin of safety against collapse during a severe
earthquake.
(ii) Life Safety (LS): which indicates a significant damage to the building's lateral force resisting system,
but keeps a large margin against collapse.
(iii) Immediate Occupancy (IO): where a relatively minor damage may occur to the building, and the
lateral force-resisting elements retain their initial strength and much of their original ductility.
Studying the structural performance at both the global and local response levels would provide a
clear understanding about the behaviour of the structure during an earthquake. The local and global
seismic behaviour of the reference structures are therefore assessed using IPA and IDA to provide
insights into the performance limit states. Figure 5 depicts sample IPA results including mapping the
local response with the global response for one of the reference buildings (BO-08). The 10% reduction
in ultimate strength is also shown on the IPA curve which is observed at 2.96% Interstorey Drift (ID).
It is shown from the results that the steel yielding starts in beams and followed by columns. The global
yielding takes place at 1.16% ID. First concrete crushing occurs in beams at 1.86% ID. Mapping the
plastic hinge distributions of a sample building shows a wide spreading of yielding unlike concrete
crushing which is limited to four locations. Despite the acceptable overstrength, plastic hinge
distributions and sequence, which are attributable to considering wind loads in design, the results
confirm the rapid strength degradation of pre-seismic code buildings.
Inelastic dynamic simulations are performed using the set of earthquake records selected
previously. Post-processors and spreadsheets are utilized for monitoring the local and global
performance criteria during IDAs. Figure 6(a) presents the distributions of concrete crushing and plastic
hinges for the 2-story frame building. Figure 6(b) also presents sample results for the shear demand
versus capacity of an internal column. The experimentally verified shear strength model used to check
the shear failure possibility in structural members was proposed by Priestley et al. (1994). The shear
strength obtained using the design code is also shown as a reference (BS8110, 1986). The results of the
former model show a significant drop in shear strength due to increasing ductility to the level that
matches the code shear strength. It is clear that from these sample results that columns are dominated
by flexure, rather than shear as the demand does not exceed either of the Priestley or the code shear
strength models. The results obtained from monitoring the local and global response of the nine
reference structures are used for selecting the required limit states for developing fragility curves.

6000
First beam yielding ● First column yielding

5000 Global yielding  Global failure ▲

First beam crushing × First column crushing ○
4000 ID=1.86
Base Shear (kN)

Vy ID=2.96
ID=1.16
3000

2000 10% reduction

Vd
in strength
Design
1000

0
0 200 400 600 800 ∆max 1000
∆y Top Disp. (mm)

Figure 5. Sample IPA results for the 8-story pre-code building (BO-08); mapping the local response with the
global capacity curve (left), plastic hinge distributions (middle) and distributions of concrete crushing (right)

(a) (b)
250
V(Demand)
Shear Deamand & Supply (KN)

V(Design Strength)
200
V(Priestley)

150

100

50

0
0 5 10 15 20 25 30
Time Step (s)

Figure 6. Sample IDA results for the 2-story frame building (BO-02): (a) mapping the distributions of plastic
hinges and concrete crushing, and (b) shear response of an internal column

A literature review is also carried out to confirm the limit states used for deriving fragility curves.
The adopted performance criteria thus take into consideration those recommended by code provisions
and previous experimental and analytical studies as well as the IPA and IDA results obtained from the
current study. For the pre-code frame and wall structures, the IO limit states are determined from the

6
 A.Issa and A.Mwafy 7

IDA curves based on the 16 percentile of the ID values at the first indication of nonlinear response
(Vamvatsikos and Cornell, 2002). The CP limit state of pre-code frames is determined from IPA based
on the 10% strength reduction of ultimate capacity, which is the most conservative CP performance
indicator observed in the present study (Mwafy and Elnashai, 2001; Park, 1988). For pre-code wall
structures, the CP performance limit is determined from Time History Analysis (THA) based on the 16
percentile of the ID values at the first indication of crushing in the confined concrete of shear walls. The
strain corresponding to the crushing of confined concrete is obtained as per Mander et al. (1988). Finally,
for emergency facilities, the IDA results indicated a significantly high limit states compared with
previous studies. Therefore, the IO limit state is selected based on another study that covered a wide
range of well-designed structure with different characteristics (Ashri and Mwafy, 2014). The CP limit
state is determined according to the statistical analysis of several previous test results (Dymiotis et al.,
1999). The LS performance level is generally considered 50% of the CP value (ASCE-41, 2007). Table
5 summarizes the literature review and results of the present study, which are used to select limit states.
All of the selected performance criteria are consistent with the results obtained from the current study,
previous experimental studies (e.g. Dymiotis et al., 1999; Ghobarah et al., 1998; Wood, 1991), previous
analytical studies (Ghobarah et al., 1999; Liel et al., 2010; Ramamoorthy et al., 2008), and the code
provisions (ASCE-41, 2007). Additional information regarding the adopted limit states is available
elsewhere (Issa, 2014).

Table 5. Summary of interstory drift (ID) ratios for different limit states
Reference Structure
Pre-code Frames Pre-code Walls Emergency Facilities
Selection Approach
Limit State - Interstory Drift (%)
IO LS* CP IO LS* CP IO LS* CP
ASCE-41, 2007 0.50 1.00 2.00 1.00 2.00 4.00
Experimental studies

Ghobarah, 1998 1.00 2.00 3.28

Wood, 1991 - 16% 1.36
Wood, 1991 - 50% 1.88
Wood, 1991 - 84% 2.60
Dymiotis et.al., 1999 - 16% 1.90
Dymiotis et.al., 1999 - 50% 4.00
Dymiotis et.al., 1999 - 84% 6.70
Ghobarah et.al., 1999 0.70 1.10 2.50 0.40 1.80 3.00
Ramamoorthy et.al., 2008 - 16 % 0.33 0.56
Analytical

Ramamoorthy et.al., 2008 - 50 % 0.50 0.98

studies

Ramamoorthy et.al., 2008 - 84 % 0.75 1.71

Liel et.al., 2010 - 16 % 3.26
Liel et.al., 2010 - 50 % 4.17
Liel et.al., 2010 - 84 % 5.34
IPA, first yield and crushing 0.67 3.74 0.33 1.59 0.85 3.67
IPA, 10% strength reduction 2.96
Current study

THA - 16% 0.40 2.71 0.32 1.78 0.77 2.91

THA - 50% 0.48 3.30 0.39 2.27 0.98 3.16
THA - 84% 0.57 4.02 0.48 2.89 1.24 3.44
IDA - 16% 0.39 4.13 0.34 2.83 0.65 6.49
IDA - 50% 0.57 5.43 0.62 3.83 1.00 8.79
IDA - 84% 0.84 7.14 1.13 5.18 1.54 11.91
Selected Limit State 0.39 1.48 2.96 0.34 0.89 1.78 1.00 2.00 4.00
IO: Immediate Occupancy, LS: Life Safety, CP: Collapse Prevention
IPA: Pushover Analysis, THA: Time History Analysis, IDA: Incremental Dynamic Analysis
*
LS limit state is considered 50% of the CP counterpart

FRAGILITY ASSESSMENT
A large number of IDAs are performed to derive the vulnerability relationships of the nine reference
structures. In addition to monitoring global response parameters such as ID ratio, top displacement, and
base shear, the developments of plastic hinges and concrete crushing in different structural members are
monitored to provide more understanding into the level of structural damage. As shown in Figure 7, 280
response points (PGA versus ID ratio) are plotted for each of the nine buildings from each of the
considered seismic scenarios. Regression analyses are performed for IDA results to develop fragility
relationships. The presented set of regression analysis results shown in Figure 7 are for three reference
buildings.

Fire Station (Short) Fire Station (Long)

20.00 12.00
10.00
15.00
8.00 y = 4.9707x1.3399
y = 1.1432x1.0726
ID (%)

ID (%)
R² = 0.7828 R² = 0.9421
10.00 6.00
4.00
5.00
2.00
0.00 0.00
0.00 1.00 2.00 3.00 4.00 5.00 0.00 0.20 0.40 0.60 0.80 1.00 1.20
PGA (g) PGA (g)

18-Story (Short) 18-Story (Long)

6.00 6.00

5.00 y = 0.6434x0.9268 5.00 y = 3.3225x1.0141

R² = 0.8208 R² = 0.8886
4.00 4.00
ID (%)
ID (%)

3.00 3.00
2.00 2.00
1.00 1.00
0.00 0.00
0.00 1.00 2.00 3.00 4.00 5.00 0.00 0.20 0.40 0.60 0.80 1.00 1.20
PGA (g) PGA (g)

40-Story (Short) 40-Story (Long)

4.00 6.00
y = 0.4084x + 0.187
R² = 0.5713 5.00 y = 3.1294x0.9625
3.00 R² = 0.8871
4.00
ID (%)
ID (%)

2.00 3.00
2.00
1.00
1.00

0.00 0.00
0.00 1.00 2.00 3.00 4.00 5.00 0.00 0.20 0.40 0.60 0.80 1.00 1.20
PGA (g) PGA (g)

Figure 7. Regression analysis for sample buildings using two seismic scenarios
(short: near-field records; long: far-filed records)

The statistical distributions shown in Figure 7 are employed to estimate the probability of
exceeding each of the selected limit states at different ground motion intensity levels. The vulnerability
curves are derived by plotting the calculated probability values versus PGAs. The derived fragility
relationships of the reference structure are shown in Figure 8. The results are presented to enable the
comparison of the vulnerability curves derived from the two seismic scenarios used in the current study.
The results show that the steepness of the fragilities decreases as the limit state changes from IO to CP.
For pre-seismic code buildings, the probability of exceeding different limit states are higher for low-rise
frame buildings. This indicates that earthquakes have less impact on high-rise wall structures. This
statement is confirmed under the effect of both severe distant and moderate close events. This is
attributable to the efficiency of shear walls in controlling drift and to the lower contribution of the
fundamental mode of vibration to seismic response with increasing the building height.
The vulnerability curves of the five pre-code structures reflect the differences between the
fragilities obtained from the two seismic scenarios employed in the present study. The slopes are sharper

8
 A.Issa and A.Mwafy 9

and the probability of exceeding various limit states is much higher under the effect of the far-field
ground motions compared with near-source events. The seismic response of the pre-code high-rise
buildings (i.e. 18, 26 and 40 storeys) is acceptable at the design PGA when subjected to the near-source
records. The results confirm that the earthquake magnitude and distance have a significant influence on
the seismic risk of multi-story buildings. The findings support the observations discussed above about
the vulnerability of the pre-code buildings to severe distant earthquakes compared with moderate close
events. The low impacts of short-period records on seismic performance is noticeable and it follows the
findings of previous analytical studies (Mwafy, 2012).
The four emergency facilities show satisfactory seismic performance under the two seismic
scenarios considered herein. As discussed above in pre-code structures, far-field records have the highest
impact on emergency facilities. In spite of the good performance of modern code designed emergency
buildings, and taking into consideration their important role during and after an earthquake, it is
preferable to perform a precautionary retrofit for such facilities to minimize seismic losses.
1 1 1

0.8 0.8 0.8

P(Limit State|GMI)

P(Limit State|GMI)

P(Limit State|GMI)
0.6 IO-Long 0.6 IO-Long 0.6 IO-Long
LS-Long LS-Long LS-Long
CP-Long CP-Long CP-Long
0.4 0.4 0.4
IO-Short IO-Short IO-Short
LS-Short LS-Short LS-Short
0.2 CP-Short 0.2 CP-Short 0.2 CP-Short

PGA (g) PGA (g) PGA (g)

0 0 0
0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5

BO-02, 2-storey BO-08, 8-storey BO-18, 18-storey

1 1 1

0.8 0.8 0.8

P(Limit State|GMI)

P(Limit State|GMI)

P(Limit State|GMI)

0.6 IO-Long 0.6 IO-Long 0.6 IO-Long

LS-Long LS-Long LS-Long
CP-Long CP-Long CP-Long
0.4 0.4 0.4
IO-Short IO-Short IO-Short
LS-Short LS-Short LS-Short
0.2 CP-Short 0.2 CP-Short 0.2 CP-Short

PGA (g) PGA (g) PGA (g)

0 0 0
0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5

BO-26, 26-storey BO-40, 40-storey FS, Fire Station

1 1 1

0.8 0.8 0.8

P(Limit State|GMI)

P(Limit State|GMI)

P(Limit State|GMI)

0.6 IO-Long 0.6 IO-Long 0.6 IO-Long

LS-Long LS-Long LS-Long
CP-Long CP-Long CP-Long
0.4 0.4 0.4
IO_Short IO-Short IO-Short
LS-Short LS-Short LS-Short
0.2 CP-Short 0.2 CP-Short 0.2 CP-Short

PGA (g) PGA (g) PGA (g)

0 0 0
0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5

PS, Police Station SC, School HO, Hospital

Figure 8. Fragility curves of nine reference structures using two seismic scenarios (short: near-field records;
long: far-filed records)

To provide more representative results from the derived fragility curves, limit states probabilities
are estimated at the design and twice the design PGAs for far-field records, twice and four times the
design for near-source records (Figure 9). Two main observations are evident; the first observation is
that pre-code structures are significantly more vulnerable compared with emergency facilities. The
second observation is that far-field records have much higher impact on the reference structures over
the near-source records. The large increase in the probabilities of various limit states is also clear when
the buildings are subjected to twice and four times the design intensities for far-field and short-period
records, respectively.
 IO LS CP IO LS CP
1 1

0.8 0.8

Probability (%)
Probability (%)

0.6 0.6

0.4 0.4

0.2 0.2

0 0
BO-02 BO-02 BO-08 BO-08 BO-18 BO-18 BO-26 BO-26 BO-40 BO-40 BO-02 BO-02 BO-08 BO-08 BO-18 BO-18 BO-26 BO-26 BO-40 BO-40
(1D) (2D) (1D) (2D) (1D) (2D) (1D) (2D) (1D) (2D) (2D) (4D) (2D) (4D) (2D) (4D) (2D) (4D) (2D) (4D)
Building Building
Pre-code buildings, Pre-code buildings,
Far-field records at the design (1D) and at twice the design Near-source records at twice the design (2D) and at four times
(2D) the design (4D)
IO LS CP IO LS CP
1 1

0.8 0.8

Probability (%)
Probability (%)

0.6 0.6

0.4 0.4

0.2 0.2

0 0
FS (1D) FS (2D) PS (1D) PS (2D) S (1D) S (2D) H (1D) H (2D) FS (2D) FS (4D) PS (2D) PS (4D) S (2D) S (4D) H (2D) H (4D)
Building Building
Emergency facilities, Emergency facilities,
Far-field records at the design (1D) and at twice the design Near-source records at twice the design (2D) and at four times
(2D) the design (4D)
Figure 9. Limit state probabilities of the nine reference buildings

CONCLUSIONS
Pre-code structures and emergency facilities are of high significance in the potential consequences from
natural hazard events since they represent high levels of vulnerability and readiness of service,
respectively. Due to the significance of such structures along with increasing the frequency of seismic
activities, a detailed vulnerability assessment of seismic risk in the earthquake prone areas of the UAE
is necessary in order to effectively estimate and mitigate earthquake losses. This comprehensive study
enabled the derivation of a wide range of dynamic response simulation-based fragility relationships for
pre-seismic code buildings and emergency facilities. The study included the selection, structural design
and developing verified fibre-based simulation models for nine reference structures representing a wide
range of the pre-code structures and emergency facilities in the UAE. The selection and scaling of 40
input ground motions representing the near-field and far-field earthquake scenarios anticipated in a
highly populated seismically prone area of the UAE were briefly discussed. The limit state criteria
employed for deriving fragility curves were selected based on the mapping of local and global response
parameters from inelastic pushover analyses (IPAs) and incremental dynamic analyses (IDAs) as well
as from previous studies. The measured seismic response from a large number of IDAs was related to
ground motion intensity to derive the fragility relationships of the nine reference structures.
The probability of exceeding limit states was much higher and the slopes were sharper under the
effect of the far-field earthquake scenario compared with the near-source counterpart. The latter seismic
scenario had low impacts on the seismic response of the reference buildings, especially the emergency
facilities at the design and twice the design intensity. The high impacts of the far-field seismic scenario
are attributed to the higher spectral amplifications and effective durations of the long period input ground
motions. At the design intensity, the seismic response of the pre-seismic code wall structures and
emergency facilities was acceptable under the near-field seismic scenario. On the other hand, the
presented study confirmed the vulnerability of a wide range of pre-code buildings to the far-field seismic

10
 A.Issa and A.Mwafy 11

scenario anticipated in the earthquake-prone areas of the UAE. It also showed the lower vulnerability of
emergency buildings due to the adoption of modern code provisions in the design of such facilities. The
results support implementing a mitigation strategy for reducing the seismic risk of pre-code structures
by means of various rehabilitation techniques. Such measures include but not limited to, reinforced
concrete jacketing, fibre-reinforced polymers (FRP) wrapping, and buckling-restrained braces (BRB).
In spite of the acceptable performance of modern code designed emergency buildings under near-field
events, and taking into consideration their crucial role following an earthquake, it is recommended to
perform a precautionary retrofit for such facilities to minimize seismic losses, particularly under the
effect of the far-field seismic scenario. The developed fragility functions in this study are an essential
component for the estimation of possible earthquake losses in the UAE. The outcomes of the present
study motivates interdisciplinary research activities in future to develop a comprehensive earthquake
loss assessment and mitigation system to preserve the large investments in the construction industry in
this region.

AKCNOWLEDGEMENT
This work was supported by the United Arab Emirates University under research grants no. 31N132 and
31N007.

REFERENCES
Abdalla J A, and Al-Homoud A S (2004) "Seismic hazard assessment of United Arab Emirates and its
surroundings," Journal of Earthquake Engineering, 8(6): 817-837
ACI-318 (2011) "Building Code Requirements for Structural Concrete and Commentary (318-11)." American
Concrete Institute, Detroit, Michigan,
Ambraseys N, Smit P, Douglas J, Margaris B, Sigbjörnsson R, Olafsson S, Suhadolc P, and Costa G (2004)
"Internet site for European strong-motion data," Bollettino di Geofisica Teorica ed Applicata, 45(3): 113-
129
ASCE-7 (2010) "Minimum design loads for buildings and other structures, ASCE Standard ASCE/SEI 7-10."
American Society of Civil Engineers, Reston, VA,
ASCE-41 (2007) "Seismic Rehabilitation of Existing Buildings (ASCE/SEI 41-06)." American Society of Civil
Engineers, Reston, VA
Ashri A, and Mwafy A M "3D vulnerability functions for contemporary buildings with varying structural systems
and heights", Proceedings, Second European Conference on Earthquake Engineering and Seismology
(2ECEES)
ATC (1996) Seismic evaluation and retrofit of concrete buildings, Applied Technology Council, report ATC-40.
Redwood City
Bruneau M, and Reinhorn A (2007) "Exploring the concept of seismic resilience for acute care facilities,"
Earthquake Spectra, 23(1): 41-62
BS8110 (1986) "Structural use of concrete." BS8110, British Standard Institution, London,
BS-6399-2 (1998) "6399-2. Loading for buildings, Part 2: Code of practice for wind loads." British Standard
Institution, London,
Buratti N, Ferracuti B, and Savoia M (2010) "Response surface with random factors for seismic fragility of
reinforced concrete frames," Structural Safety, 32(1): 42-51
Chiou B-J, and Youngs R R (2008) "An NGA model for the average horizontal component of peak ground motion
and response spectra," Earthquake Spectra, 24(1): 173-215
CSI (2011) ETABS - Integrated building design software, Computers and Structures, Inc., Berkeley, California,
Dymiotis C, Kappos A J, and Chryssanthopoulos M K (1999) "Seismic reliability of RC frames with uncertain
drift and member capacity," Journal of Structural Engineering, 125(9): 1038-1047
Elnashai A, and Mwafy A (2002) "Overstrength and force reduction factors of multistorey reinforced‐concrete
buildings," The structural design of tall buildings, 11(5): 329-351
Elnashai A S, Papanikolaou V, and Lee D (2012) Zeus-NL - A System for Inelastic Analysis of Structures - User
Manual Mid-America Earthquake Center, University of Illinois at Urbana-Champaign, Urbana, IL,
Ghobarah A, Abou‐Elfath H, and Biddah A (1999) "Response‐based damage assessment of structures,"
Earthquake engineering & structural dynamics, 28(1): 79-104
Ghobarah A, Aly N, and El-Attar M (1998) "Seismic reliability assessment of existing reinforced concrete
buildings," Journal of earthquake engineering, 2(4): 569-592
Issa A (2014) Development of Simulation-based Fragility Relationships for the Seismic Risk Assessment of
Buildings Master thesis, United Arab Emirates University, UAE
Liel A B, Haselton C B, and Deierlein G G (2010) "Seismic collapse safety of reinforced concrete buildings. II:
Comparative assessment of nonductile and ductile moment frames," Journal of Structural Engineering,
137(4): 492-502
Mander J B, Priestley M J, and Park R (1988) "Theoretical stress-strain model for confined concrete," Journal of
structural engineering, 114(8): 1804-1826
Mwafy A (2012) "Analytically derived fragility relationships for the modern high‐rise buildings in the UAE," The
Structural Design of Tall and Special Buildings, 21(11): 824-843
Mwafy A, and Elnashai A (2001) "Static pushover versus dynamic collapse analysis of RC buildings," Engineering
structures, 23(5): 407-424
Mwafy A M "Classification and idealization of the building stock in the UAE for earthquake loss estimation",
Proceedings, 15th World Conference on Earthquake Engineering, International Association for
Earthquake Engineering,
Mwafy A M, Elnashai A S, Sigbjornsson R, and Salama A (2006) "Significance of severe distant and moderate
close earthquakes on design and behavior of tall buildings," The Structural Design of Tall and Special
Buildings, 15(4): 391-416
Park R "Ductility evaluation from laboratory and analytical testing", Proceedings, 9th World Conference on
Earthquake Engineering, Tokyo-Kyoto, Japan: 605-616
Priestley M N, Verma R, and Xiao Y (1994) "Seismic shear strength of reinforced concrete columns," Journal of
structural engineering, 120(8): 2310-2329
Ramamoorthy S K, Gardoni P, and Bracci J M (2008) "Seismic fragility and confidence bounds for gravity load
designed reinforced concrete frames of varying height," Journal of structural engineering, 134(4): 639-
650
SEAOC (1995) "Performance based seismic engineering of buildings." Structural Engineers Association of
California, Sacramento, Calif
UBC (1997) "Vol. 2." International Council of Building Officials, Whittier, CA
Vamvatsikos D, and Cornell C A (2002) "Incremental dynamic analysis," Earthquake Engineering & Structural
Dynamics, 31(3): 491-514
Wood S L (1991) "Performance of reinforced concrete buildings during the 1985 Chile earthquake: implications
for the design of structural walls," Earthquake Spectra, 7(4): 607-638

12