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

net/publication/268522721

Seismic Effects on Structures Located in the Region of Palestine: A Critical

Review on International Building Regulations

Article in Journal of Civil and Structural Engineering · April 2015

CITATIONS READS

0 2,458

2 authors:

Fahed H. Salahat Pedro Martinez-Vazquez

Kansas State University University of Birmingham
2 PUBLICATIONS 0 CITATIONS 50 PUBLICATIONS 173 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Nonlinear modelling parameters of shear critical reinforced concrete columns View project

Energy methods on wind turbine tower buckling study View project

All content following this page was uploaded by Fahed H. Salahat on 03 January 2022.

The user has requested enhancement of the downloaded file.

 Proc. of the Second Intl. Conf. on Advances In Civil, Structural and Mechanical Engineering- CSM 2014.
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-006-4 doi: 10.15224/ 978-1-63248-006-4-108

Seismic Effects on Structures Located in the Region

of Palestine
A Critical Review on International Building Regulations
F.H. Salahat, P. Martinez-Vazquez

Abstract - A qualitative assessment of seismic effects on a

regular building located in Palestine is presented. The assessment
is based on two distinct approaches, one by following
II. Development of local
recommendations from the International and Uniform Building
Codes (IBC and UBC) and the other using response spectra
response spectra
inferred from international historical records whose level of
seismicity compares to those in Palestine. The study shows that
A. Selecting representative earthquakes
both the IBC and UBC tend to underestimate the maximum Linear and nonlinear dynamic analysis for seismic design
response of structures whose fundamental period of vibration and evaluation of structures require seismic loads to be
falls between 0.1s and 3s. The difference reduces gradually for represented either by a response spectrum or time history
periods above 3s. The rate of underestimating the seismic effects records. In this paper, a response spectrum is developed based
varies with the fundamental period of vibration of the structure on the maximum response of a SDOF oscillator subjected to
and its distance from the Dead Sea Fault Line, but it can be up to time history acceleration records. The acceleration records
~ 50%. It thus appears necessary to address such differences by were selected based on three criteria that define their
taking into account local seismic conditions in the region of compatibility with the local seismicity in Palestine.
Palestine through a new local design code.
The factors for selecting representative records are the
Keywords—seismic response of buildings, response spectra, magnitude (M) and the distance from the geological fault (R)
time history analysis, international standards. acting as the source of seismic waves [1], in this study, the soil
type (S) is used to account for the local seismicity in the
I. Introduction region. Therefore, an M-R-S relationship is used to select
acceleration records for earthquake events that have occurred
In developing countries the impact of strong earthquake in different places in the world and which are likely to
events on infrastructures is as important as the phenomenon reproduce in Palestine.
itself. As a developing country, Palestine has limited
resources to deal with emergencies caused by such a naturally The magnitude (M) was set to the range of 6.0-7.5 in
occurring phenomenon. Therefore, mitigating earthquake Richter scale in order to consider the maximum seismic
risks should be done by correctly evaluating the seismic loads demand that was experienced in the region according to the
and improving current design practices. In addition to other available earthquake catalogue, which was used to develop the
measures that involve the retrofit of old structures, urban currently used seismic hazard map [2]. Palestine is located in
development and contingency plans. Structural resilience to the area that is adjacent to the Dead Sea Fault Line (See Figure
earthquake events depends on the accuracy in estimating the 1), the distance from the fault is considered normal to tangent
seismic demand and so the structural response. The use of of the fault line and it covers 74 km on average all over the
response spectra is the most common method to estimate the country. Therefore, R falls within the range 0-74 km. while
seismic demand and accordingly design earthquake-resisting previous studies [3] show that the shear velocity through the
infrastructures in Palestine and elsewhere. soil media has a range of 500-1500 m/s which corresponds to
soil type SC-SB according to the UBC soil classification
This paper considers the use of some of the adopted design system.
codes, namely: the Uniform Building Code (UBC) and the
International Building Code (IBC), and their suitability to Based on the selection criteria described above, ten
represent the seismic demand in Palestine using the response earthquake acceleration records were downloaded from the
spectrum analysis method. Pacific Earthquake Engineering Research Centre (PEER)
ground motion data base [4] and selected to represent the local
seismic demand in Palestine, the characteristics of the
F.H. Salahat P. Martinez-Vazquez
earthquakes are presented in Table 1 – where g represents the
University of Birmingham University of Birmingham gravitational constant.
United Kingdom United Kingdom
fahedsalahat@yahoo.com P.Vazquez@bham.ac.uk

209
 Proc. of the Second Intl. Conf. on Advances In Civil, Structural and Mechanical Engineering- CSM 2014.
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-006-4 doi: 10.15224/ 978-1-63248-006-4-108

The maximum elastic displacement (MED) response of the

oscillators for each of the ten earthquakes was obtained by
using SAP2000 [5]. Then the displacement response was
converted to acceleration using Equation 1.
Sa = (2π/T)2 . Sd (1)
Where:
Sa: The spectral acceleration
Sd: The spectral displacement
T: The period of vibration.
Figure 3 shows the spectral accelerations estimated for the
ten selected earthquakes.

Figure 1: Determination of (R) range

Table 1: The representative earthquake events

PGA R
ID Event name Date M
(g) (km)
1 Parkfield–California 28-06-1966 6.1 0.375 14.7
2 Morgan Hill- San Francisco 24-04-1984 6.2 0.292 11.8
3 Victoria - Mexico 09-06-1980 6.4 0.621 34.8 Figure 3: The spectral accelerations for the ten earthquakes
4 Imperial Valley – California 15-10-1979 6.5 0.204 14.2
It is observed in Figure 3 that response spectra cover a
5 Superstition Hills – California 24-11-1987 6.7 0.455 0.7
wide range of spectral ordinates. Therefore two groups of
6 Kobe - Japan 16-01-1995 6.9 0.821 0.6
spectra are distinguished, the first includes earthquakes within
7 Duzce - Turkey 12-11-1999 7.1 0.134 15.6
and beyond 5 km from the fault (all the records) to represent
8 Gulf of Aqaba – Palestine 22-11-1995 7.2 0.100 44.1
near fault seismicity. The seconds includes earthquakes
9 Landers – California 28-06-1992 7.3 0.785 1.1
beyond the 5 km border (earthquakes 1, 2, 3, 4, 7 and 8 – see
10 Kocaeli – Turkey 17-08-1999 7.4 0.220 4.8
Table 1) to represent the lower seismicity region. Figure 4
shows the average spectra estimated for each one of these
Typically, two horizontal components and one vertical groups together with their idealised version. M1 represents the
component are recorded for an event. However, for each of the mean spectrum of the first group, which is idealised as IM1
records that are shown in Table 1, the largest horizontal whilst M2 represents the mean spectrum of the second group
component in terms of the peak ground acceleration (PGA) which is idealised as IM2.
was considered whilst vertical component of the seismic
motion was neglected. All the records are considered to cause
vibration in the same direction.

B. Spectral representation of the

selected records
SDOF oscillators characterised by
vibration periods in the range of 1-10s
and 5% damping were subjected to the
seismic records described above. The
period of the oscillator was modified by
varying the structural mass (see Figure
2). The increment of the period dT was
0.1s until T=1s, thereafter an increment Figure 4: The mean response spectra with their idealisation
dT of 1s applied until T=10s. Eq. (2) and (3) define the idealisation of IM1 and IM2.
Figure 2: SDOF oscillator

210
 Proc. of the Second Intl. Conf. on Advances In Civil, Structural and Mechanical Engineering- CSM 2014.
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-006-4 doi: 10.15224/ 978-1-63248-006-4-108

B. Comparing IM1, IM2 response

spectra with the UBC and IBC
{ spectra
Figure 6 compares the UBC97 and IBC09 response spectra
with the developed IM1 and IM2 response spectra for 5%
damping.

Note that at T = 0, Sa/g is taken as the average PGA for

the considered earthquakes (0.4 for M1 and 0.29 for M2).

III. The currently adopted code

response spectra in
Palestine
Figure 6: Comparing the adopted design response spectra with IM1 and IM2
A. UBC and IBC response spectra The comparison amongst the spectra presented in Figure 6
The UBC97 [6] and IBC09 [7] response spectra are the will be done by considering different period ranges. The
most popular codes of practice in Palestine. The UBC ordinates of the short period range are presented in Table 2.
response spectrum is shown in Figure 5.
Table 2: Comparison of the spectral ordinates for various values of T
UBC IBC IM1 IM2

Period (s) Sa/g x 102

T=0 0.2 0.16 0.4 0.29 50 69 40 55.2
T = 0.1 0.5 0.345 0.6 0.465 83.3 107.5 57.5 74.2
T=0.2-0.4 0.5 0.4 0.8 0.63 62.5 79.4 50 63.5
T = 0.5 0.4 0.4 0.8 0.63 50 63.5 50 63.5

From Table 2 it can be observed that both UBC and IBC

response spectra underestimate the local earthquake demand in
the period range of 0-0.4s except at T= 0.1s when the UBC
ordinate slightly exceeds that of IM2. The differences are
Figure 5: UBC 97 design response spectrum
clearer on sites within 5km from the fault line. The UBC and
Where: IBC recommend using peak spectral accelerations (PSA) of
TS = CV/2.5 Ca 0.5g and 0.4g respectively. This represents 62.5% (UBC) and
To = 0.2 TS 50% (IBC) of the PSA in IM1 and correspondingly 79.4%,
T = Fundamental period of Vibration 63.5% of the PSA in IM2. The UBC is closer to IM1and IM2
than the IBC in this period range. At T = 0.5s, the UBC equals
The parameters Ca and CV depend on the seismic zone
the IBC ordinates at PSA of 0.4g which represents 50% and
factor (Z) and the soil type (S) [8]. In Palestine the value of Z 63.5% of IM1 and IM2 respectively. After T = 0.5s, the IBC
is taken as 0.2 and the soil type is considered as SB. For Z of ordinates become closer than the UBC and almost match IM2.
0.2 and soil type SB, the values for Ca and CV are 0.2 and 0.2 However, for the near fault sites represented by IM1
respectively. Similarly, the IBC response spectrum also differences remain significant until about T=3s.
depends on the site seismicity and the soil type. In the next
section, the UBC97 and the IBC09 spectra are presented and IV. Case studies
compared with the IM1 and IM2 response spectra i.e. those In the previous sections it was pointed out that applying
inferred from the real time history records. the UBC and IBC regulations in Palestine results in
underestimating the local seismic demand. This section
considers the effect of such underestimation by observing the

211
 Proc. of the Second Intl. Conf. on Advances In Civil, Structural and Mechanical Engineering- CSM 2014.
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-006-4 doi: 10.15224/ 978-1-63248-006-4-108

maximum storey drift of 5% damped reinforced concrete Figure 9 shows that both the IBC and UBC codes
buildings that were analysed by using SAP2000. Four case underestimate the local seismic demand. At the 6 th storey, the
studies are considered for that purpose. These are IBC and UBC induce a maximum elastic displacement (MED)
characterised by having periods of vibration of 0.4s, 0.5s, 1.5s, of about 2.2, 2.7 cm respectively whereas using the IM2 and
and 3s. The selected range would cover all possible scenarios IM1 the MED at the 6th storey is of 3.4, 4.3 cm respectively.
of incompatibility amongst the international building codes Thus for locations beyond 5 km from the fault line, the
and the spectra derived from historical data. The general estimated MED associated to the IBC and UBC response
characteristics of the building model with T=1.5s are shown in spectra are of about 64% and 80% of the IM2 respectively i.e.
Table 3 whilst Figures 7 and 8 show a plan and 3D view of the the UBC provides a closer approximation for Case Study 1.
model respectively. However, if this building model is located within 5 km from
Table 3: Characteristics of the building model with T = 1.5 s
the fault line, the proportion between IBC and UBC with
Model properties Description respect to IM1 becomes 51% and 63% respectively, i.e. the
underestimation becomes more critical near fault sites. These
Structural system Dual (frames + shear walls) results are consistent with the differences amongst spectral
Number of storeys 6 ordinates presented in Table 2.
Storey height (m) 4 Case 2: T = 0.5 seconds (Figure 10).
Number of bays in each direction 3
Bay width (m) 5.35
Columns section (cm) 45x45
Beams section (cm) 30x40
External shear walls section (cm) 200x30
Internal shear walls section (cm) 150x20
Non-structural mass (Kg) 613938

Figure 10: Maximum storey drift of the model with T=0.5 s

Figure 10 suggests that there is compatibility between the
IBC and UBC ordinates when T=0.5s and shows that their
approach tends to underestimate the MED when compared to
IM1 and IM2. At the 6th storey, IM1 and IM2 result on MED
of 6.8cm and 5.3cm respectively, whereas the IBC and UBC
give 3.5 cm. this represents ~51% compared to IM1 and 66%
compared to IM2 results which again are consistent with the
spectral ordinates presented in Table 2.
Figure 7: Plan view of the model Figure 8: The 3D model Case 3: T = 1.5 seconds (Figure 11).
The following section discusses the results of the analyses.
Case 1: T = 0.4 seconds (Figure 9).

Figure 11: Maximum storey drift of the model with T=1.5 s

Figure 11 shows that the IBC and UBC still underestimate
Figure 9: Maximum storey drift of the model with T=0.4 s the MED; however, the underestimation is less critical when

212
 Proc. of the Second Intl. Conf. on Advances In Civil, Structural and Mechanical Engineering- CSM 2014.
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-006-4 doi: 10.15224/ 978-1-63248-006-4-108

T=1.5s. The MED at the 6th storey by IBC and UBC is 13.1cm References
and 10.1cm respectively; which translates into 87.3% and
67.3% of the MED (15cm) given by the IM2. If the same
comparison is done against IM1 results (MED=24.3 cm), the [1] Katsanos, E., Sextos, A. and Manolis, G. (2010) Selection of earthquake
ground motion records: A state-of-the-art review from a structural
percentage of UBC/IM1 and IBC/IM1 are of 54% and 41.5% engineering perspective. Soil Dynamics and Earthquake Engineering, 30
respectively. This shows that for T=1.5s the IBC provides the (4): 157—169
better approximations. However, the differences are still [2] Shapira, A., Hofstetter, R., Geophysical Institute of Israel (2007)
considerable. Earthquake Hazard Assessment for Building Codes
[3] J. Al-Dabbeek and ESSEC team, (2011) Assessment for seimic site
Case 4: T = 3 seconds (Figure 12). effect for Rawabi first Palestinian planned city
(http://www.preventionweb.net/files/globalplatform/entry_presentation~
12h00[1].pdf). (Visited October 13, 2014)
[4] Pacific Earthquake Engineering Research Centre
(http://peer.berkeley.edu/) (visited July 10, 2014)
[5] SAP2000, V.14.2
[6] Uniform Building Code 1997 (UBC97)
[7] International Building Code 2009 (IBC09)
[8] Table 16-Q and Table 16-R in UBC1997

About Authors:

Fahed H. Salahat, MSc. Civil Engineering.

Figure 12: Maximum storey drift of the model with T=3 s
Research interests include: Earthquake Engineering
In this case the differences between IBC and IM2 are small and computer applications, performance-based
whilst UBC results still underestimates the MED given by seismic design and assessment.
IM2. At the 6th storey, the ratios UBC/IM2 and UBC/IM1 are Dr P. Martinez-Vazquez.
Research interests include random vibrations,
of 0.77 and 0.63 respectively whereas the ratios IBC/IM2 and artificial intelligence, image recognition techniques,
IBC/IM1 are of 0.98, 0.82. Thus, although UBC approaches to structural analysis, and structural design
IM1 and IM2 as T increases the rate of variation is slower than
that of the IBC which gives the best estimation after T=0.5s.

V. Conclusion
The various analyses show that the current design practice
in Palestine, which follows recommendations by the UBC and
IBC, underestimate the structural response of buildings whose
period of vibrations falls within the range 0.1s-3.0s. The rate
of variation changes with the natural period of vibration and
distance to the fault parameter. The analyses show that
compared to response spectra derived from real earthquakes
records, the UBC gives a better estimation than the IBC in the
period range of 0.0-0.4s. Whereas the displacement demand
given by the IBC and UBC match at T=0.5s. For T>0.5s the
IBC consistently provides the best approximation to the real
spectral case. It is important to note that the majority of the
current infrastructure in Palestine includes low to medium rise
buildings, i.e. structures whose fundamental period fall within
the critical period region (0.1s-3.0s) as derived from this
analysis. Therefore given the possibility that earthquake events
of a magnitude of 6.0-7.5 in the Richter scale or above may hit
the region makes it necessary to ponder the amount of risk that
underdeveloped countries can accept, given the limited
resources and strategic plans that are currently in place to deal
with natural disasters caused by earthquakes.

213

View publication stats