PROCEDURES FOR REAL EARTHQUAKE TIME HISTORIES

SCALING AND APPLICATION TO FIT IRANIAN DESIGN SPECTRA

Y. M. Fahjan 1, Z. Ozdemir 2 and H. Keypour 3

ABSTRACT

With advances in seismic analysis and design of structures, linear and nonlinear time-history analyses are
becoming more common in civil engineering area. One of the most important issues for such analyses is the
selection of acceleration time histories as input parameter. Generally, international design codes require
selection and scaling of acceleration time histories to satisfy the specified design response spectrum at the site
of the structure. In literature, there are three sources of acceleration time histories: design response spectrum
compatible artificial records, synthetic records obtained from seismological models and accelerograms recorded
in real earthquakes. Due to the increase of available strong ground motion database, using and scaling real
recorded accelerograms is becoming one of the most contemporary research issues in this field. Generally, two
methods for scaling actual time histories to match a given design spectrums are used: scaling in time domain
and frequency domain. Selection criteria of proper time history records to fit the design code spectrum account
for geological and seismological conditions at specific site.
In this study, basic methodologies and criteria for selecting strong ground motion time histories are summarized
and the selection and scaling criteria of real time history records to satisfy the Iranian design code are discussed.
The time and frequency domain scaling procedures are utilized to scale a number of the available real records to
match the Iranian design spectra. The resulting time histories of the two procedures are investigated and
compared in terms of suitability as input to time history analyses of civil engineering structures.

Introduction

Seismic design is traditionally performed for most common structures by the means of equivalent lateral static
loading or modal spectrum analyses. Nevertheless, time history analyses are required to define real seismic
response of structure especially for irregular, highly ductile, critical or higher modes induced structures. Seismic
codes specify design spectra for the purpose of the design of buildings and recommend scaling of selected
ground motions matching spectral acceleration within the period range of interest to use in the time history
analyses. Selection and scaling of the input ground motions are critical issues to perform linear and nonlinear
response history analyses. In literature, there are three basic options available in order to select acceleration time
histories: design response spectrum compatible artificial records, synthetic records obtained from seismological

1
Department of Earthquake and Structural Science, Gebze Institute of Technology, 41400 Gebze, Kocaeli,
Turkey, [email protected]
2
Kandilli Observatory and Earthquake Engineering Research Institute, Bogazici University 34684, Cengelkoy,
Istanbul,Turkey, [email protected]
3
Ph.D. Graduated, Kandilli Observatory and Earthquake Research Institute, Bogazici University, 34684
Cengelkoy, Istanbul, Turkey, [email protected]
models and accelerograms recorded in real earthquakes.

The attraction of using natural accelerograms to obtain the design spectrum compatible time histories is due to
the increase of available strong ground motion databases. The drawback of using design spectrum-compatible
time histories lies in trying to match a single ground motion to a design response spectrum that is not intended
to represent the motion from an individual earthquake (Naeim and Kelly, 1999). The design response spectrum
is generally a result of a statistical analysis that considers the influence of several seismic sources
simultaneously, whence the response at different periods may be driven by earthquakes in different sources and
the spectrum is effectively the envelope of spectra corresponding to scenarios in each of the sources (Reiter,
1990, and Bommer et al., 2000).

In this study, all the existing methodologies and criteria for selection strong ground motion time histories are
investigated and summarized. The time and frequency domain scaling procedures are utilized to scale a number
of the available real records to match the Iranian design spectrum. The outcomes results of each procedure are
discussed in terms of The time and frequency domain scaling procedures are utilized to scale a number of the
available real records to match the Iranian design spectrum and suitability as input to time history analyses of
civil engineering structure are discussed.

Types of Accelerograms

Real Accelerograms
Real Strong ground motion accelerograms contain a wealth of information about the nature of the ground
shaking and carry all the ground-motion characteristics (amplitude, frequency, and energy content, duration and
phase characteristics), and reflect all the factors that influence accelerograms (characteristics of the source, path,
and site). Despite the continued growth of the global strong -motion databank, there are many combinations of
earthquake parameters such as magnitude, rupture mechanism, source-to-site distance and site classification that
are not well represented, which can make obtaining suitable records difficult in some circumstances (Bommer et
al., 2003). For many engineering application, the purpose of selection and scaling of real earthquake is to fit the
code design spectrum considering the seismological and geological parameters of the specific site.
Artificial Accelerograms
Artificial accelerograms are generated to match a target response spectrum by obtaining a power spectral
density function from the smoothed response spectrum, and then to derive sinusoidal signals having random
phase angles and amplitudes. The sinusoidal motions are then summed and an iterative procedure can be
invoked to improve the match with the target response spectrum, by calculating the ratio between the target and
actual response ordinates at selected frequencies. In order to get other characteristics of artificial spectrum-
compatible record, such as duration, it is necessary to obtain supplementary information about the expected
earthquake motion apart from the response spectrum. Even though, it is possible to obtain acceleration time-
series that are almost completely compatible with the elastic design spectrum, the generated accelerograms often
have an excessive number of cycles of strong motion, and consequently have unrealistically high-energy
content.
Synthetic Accelerograms
Synthetic accelerograms can be generated from seismological source models and accounting for path and site
effects. In general, there are actual difficulties in defining appropriate input parameters such as the source, path,
and site characteristics. To generate synthetic accelerograms there is a need for a definition of a specific
earthquake scenario in terms of magnitude, rupture mechanism in addition to geological conditions and location
of the site. Generally, most of this information is not often available, particularly when using seismic design
codes (Bommer et al., 2003)

Selection and Scaling of Time History Records

Real earthquake records are selected to match specific features of the ground motion, generally based on either
elastic response spectrum, or an earthquake scenario with the minimum parameters being the magnitude,
distance and site classification. Guidance given in seismic design codes on how to select appropriate real
records is usually focused on compatibility with the response spectrum rather than seismological parameters.
Therefore, records are selected based on strong-motion parameters such as peak ground acceleration, peak
ground velocity, and duration to match a design response spectrum. In seismic design codes, general rules have
been recommended for selecting time history records to have similar characteristics (magnitude, distance, site
condition and faulting type) as the site under consideration. It is desirable to use earthquake magnitudes within
0.25 magnitude units of the target magnitude (Stewart et al, 2001). Selection of records having appropriate
fault-site distances is important especially for near-fault sites. Site conditions have a major effect on the
characteristics and frequency content of the strong ground motion records. Even though the ground motions are
amplified in soft soils, the high frequency motions are attenuated.

Methods of Ground Motion Scaling of Real Accelerograms

There are two methods for modifying actual time histories to match a given design spectrum: Ground motion
simulation in time domain and frequency domain.
Ground Motion Scaling in Time Domain
In this approach, recorded motion is simply scaled up or down uniformly to best match the target spectrum
within a period range of interest, without changing the frequency content. It could be stated that the
accelerograms should only be scaled in terms of amplitude. Nevertheless, scaling on the time axis has been used
to modify the frequency content of real ground-motion records (Kramer, 1996). The procedure is based on
minimizing the differences between the scaled motion’s response spectrum and target spectrum in a least-square
sense. The methodology proposed herein considers as “Difference” the squared scaled-to-target difference,
evaluated by the integral,

[α S ] dT
TB 2
actual
Difference = ∫ a (T) − Sat arg et (T) (1)
TA

Where, Satarget and Saactual are target acceleration response spectrum and acceleration spectrum of the given
(actual) time history, respectively. α, T, TA and TB are is scaling factor , period of oscillator, lower period of
scaling, and upper period of scaling, respectively. In order to minimize the difference, the first derivative of the
Difference function with respect to the scaling factor has to be zero which lead to definition of scaling factor
(α) in a discrete form, as
TB
actual
∑ (Sa (T) Satar get (T) )
d Difference T = TA
min Difference ⇒ =0⇒ α= TB
(2)
dα actual
∑ (Sa (T ) ) 2
T = TA

Ground Motion Scaling in Frequency Domain

This method is based on the concept of using actual records to generate time histories that fit a given target
response spectrum. The physical characteristics of the earthquake motion are retained throughout the procedure,
which makes the technique powerful in comparison with the classical artificial record generation. A frequency
domain scaling methodology uses an actual record to produce a similar motion that matches almost perfectly a
target (design) spectrum. In this method, an actual motion is filtered in frequency domain by its spectral ratio
with the design target spectrum. The Fourier phases of the motions remain unchanged during the entire
procedure. The technique is repeated iteratively until the desired matching is achieved for a certain range of
periods. The method is described below:

(a) A target (design) spectrum Satarget (T), and a single time history record, THactual (t), are selected for
scaling
(b) The response spectrum of selected motion, Saactual(T), is calculated using the damping ratio as the target
spectrum,
(c) The ratio between the response spectrum of the actual time history and the target spectrum, SPR(T), is
computed for the period range scaling limits,

SPR (T) = Sat arg et (T) / Sactual

a (T )
, TA ≤ T ≤ TB (3)
(d) The SPR (T) can be expresses in terms of cyclic frequency ω as
FILT(ω) = SPR(ω) , ωmin ≤ ω ≤ ωmax (4)
where ωmin , ωmax are the minimum and maximum scaling frequencies, respectively.
(e) The Fourier Spectrum amplitude, Factual(ω), and phase, θactual(ω), of THactual (t) are computed using a
Fast Fourier Transform (FFT) algorithm. The amplitude of Factual(ω) is multiplied by the function, FILT(ω)

F filtered (ω) = FILT(ω) ⋅ F actual (ω) (5)

(f) A time history, TH(t), can be generated from the filtered amplitude Factual(ω) and actual phase, θactual(ω)
using an inverse Fast Fourier transform. (6)
(g) The procedures from (c) to (f) are repeated until acceptable difference of the response spectrum of TH(t) to
the target spectrum is obtained.

Scaling of Real Accelerograms to Fit the Iranian Design Spectra

Iranian code of practice for seismic resistant design of buildings (1999), define the design code spectra, using
four different seismic regions (1, 2,3, and 4) and four different soil Groups (I, II, III, and IV). The design
response spectra governed by seismic coefficient ( C) are obtained from the equation
ABI
C= (7)
R
where, A, I and R are design base acceleration (ratio of seismic acceleration to gravity acceleration g), building
importance factor and building behavior factor, respectively. B is the building response factor obtained from the
design response spectrum given by
2/3
⎛T ⎞
B = 2.5⎜ 0 ⎟ ≤ 2.5 (8)
⎝T⎠
Where, T is the fundamental period of vibration of the building (in seconds), T0 is as scalar quantity determined
according to soil group. The code definition of design base acceleration and definition of T0 are given in Table
1. Considering both importance factor and behavior factor to be unity, design response spectra for different
seismic regions and soil types are displayed in Figure 1. The Iranian code of practice for seismic resistant
design of buildings (1999) requires at least three accelerograms to be selected to fit the specific design response
spectrum, with the following characteristics:
a) In case accelerograms belong to the real earthquakes taken place in other areas. Care must he taken that
geological, tectonic and seismological characteristic and especially soil layers specifications at the
accelerograph site are similar to those of the site under consideration, as much as possible.
b) The strong motion duration in accelerograms should be at least equal to 10 seconds and three times the
fundamental period of vibration of the structure under consideration.
C) Selected accelerograms should he normalized in such a manner that the resulting spectrum from each of the
accelerograms with 5 percent damping ratio and a vibration period in the range from Trn-0.5 to Trn+0.5
seconds where m includes all the modes that contribute at least 10 percent to the effective mass of the building.

Table 1. Iranian code spectrum definition

Design Base Acceleration (A) Definition of T0

Region Description A Soil Type T0
1 Zone Of Very High Relative Seismic Hazard 0.35 I 0.4
2 Zone Of High Relative Seismic Hazard 0.30 II 0.5
3 Zone Of Intermediate Relative Seismic Hazard 0.25 III 0.7
4 Zone Of Low Relative Seismic Hazard 0.20 IV 1.0
 Figure 1. Iranian code design response spectra for different seismic regions and soil types considering
importance factor ( I ) and behavior factor ( R ) to be unity

Results of Time and Frequency Domains Scaling

The time and frequency domain scaling procedures are utilized to scale a number of the available real records to
match one of the Iranian design spectrums. The accelerograms used in this study are obtained from Pacific
Earthquake Engineering Research (PEER) Center, NGA strong motion data base (PEER, 2005) In PEER data
base 4062 records from 92 shallow crustal earthquakes in active tectonic regions around the world are existed.
The two procedures are performed using MatLab Code based on the response spectra of linear elastic single
degree of freedom system of 5% damping ratio using piecewise exact method, (Aydinoglu and Fahjan, 2003).
In this study, the scaling procedures are used to fit the design spectrum at first seismic region and soil group II
with building importance factor and building behavior factor to be equal to one.

Time domain scaling is used to fit the acceleration response spectrum of the accelerogram to design response
spectrum of seismic region 1 and soil group II. The accelerogram is chosen arbitrary as the acceleration time
history recorded in 1994 Northridge earthquake (Mw=6.7) at the station (78 Stone Canyon record) which has a
closest to fault rupture distance 22 km and recorded time 40 sec. The results of scaling procedure are shown in
Figure 2, and it is clearly noticed that there are significant differences between the two spectra in the range of
0.01-5.0 sec.

In order to minimize these differences, In literature two procedures are used: The first procedure is to utilize
time domain scaling procedure for all the available real earthquake records that satisfy the seismological and
site characteristics of the site under consideration, and the best-fitted accelerogram can be selected. Simple
algorithm for a single time history scaling (Fahjan et al 2006 ) or more advanced genetic algorithms for multiple
time history scaling (Naeim et al 2004) can be utilized. The simple algorithm for a single time history scaling
can be summarized as follows: real accelerograms from strong motion database (PEER, 2005) satisfy
earthquake magnitude of 5.0-7.6 and soil type conditions (II) are selected. Scaling procedure is used to fit each
horizontal component of the records the respected code spectrum for period range (TA=0.01 sec – TB=5 sec) and
the square root summation error for each record is computed. The output results are organized to eliminate the
scale factor less than 0.3 or greater than 4.0. The first fifty record components that have minimum error values
are selected. The best fitted three (3) records the selected with visual inspection of the scaled and target spectra
for each of them. The results of this procedure are shown in Figure 2 and as it is noticed, a better compatibility
between the two spectra is existed.

The second procedure is frequency domain scaling described above. The accelerogram represented in Figure 2
is used and the results of the procedure after one, three and ten iterations are shown in Figure 4 (a,b). Increasing
the number of iterations, a more compatible response spectrum to the design response spectrum can be
achieved. Nevertheless, the frequency contents and the spectral displacement response of the record are
significantly changed. Generally, the scaled accelerograms have an excessive number of cycles of strong
motion, and consequently have unrealistically high-energy content. A less disturbance to the frequency contents
and the spectral displacement response can be achieved by using records that response spectra are more
compatible to the design spectra then the frequency scaling procedure can be utilized, as it is show in Figure 5.

As a general remark, the first portion of the Iranian design spectra at the range of 0.0-0.2 sec is not possible to
fit physically, because the rigid motion response and maximum response of the system cannot be the same).
Thus, this portion is ignored in all scaling procedures.

Conclusions
Selection and scaling strong ground motion time histories are critical and important to the time history analyses
of structures. The time history scaling procedure is simple and has the advantage that frequency contents of the
records are not change. If the real accelerograms to fit the design spectra are scarce, frequency domain scaling
can be used. Even though, the procedure is robust, the frequency contents and the spectral displacement
response of the record are significantly changed. A less disturbance to the frequency contents and the spectral
displacement response can be achieved by starting the procedure with records that response spectra are more
compatible to the design spectra.

References
Aydinoglu M. N. Fahjan , Y. M., 2003, A unified formulation of the piecewise exact method for inelastic
seismic demand analysis including the P-delta effect, Earthquake Engineering & Structural Dynamics, Volume
32, Issue 6 , Pages 871 - 890.
Bommer, J. J., Acevedo, A. B. and Douglas, J. , 2003, The Selection and Scaling of Real Earthquake
Accelerograms for Use in Seismic Design and Assessment, Proceedings of ACI International Conference on
Seismic Bridge Design and Retrofit, American Concrete Institute.
Bommer, J.J., Scott, S.G., and Sarma, S.K., 2000, Hazard-Consistent Earthquake Scenarios, Soil Dynamics and
Earthquake Engineering, 19, pp.219-231.
Building & Housing Research Center,1999, Iranian Code of Practice For Seismic Resistant Design of Buildings
(Standard No. 2800).
Fahjan Y.M., Ozdemir Z.and Al-Qaryouti M, 2005, Selection and scaling of real earthquake accelerograms to
fit the new Jordanian design spectra, Paper No. 27, The International Earthquake Engineering Conference,
November 21-24, 2005, Dead Sea/Jordan
Kramer, S. L. , 1996, Geotechnical Earthquake Engineering, Prentice Hall.
Naeim, F. and Kelly, J.M., 1999, Design of Seismic Isolated Structures: From Theory to Practice, John Wiley
& Sons, 289 pp.
Naeim, F., Alimoradi A. and Pezeshk S. 2004, Selection and scaling of ground motion time histories for
structural design using genetic algorithms. Earthquake Spectra, Vol. 20 No 2, pp. 413-426
Pacific Earthquake Engineering Research Center, 2005, PEER Strong Motion Database,
http://peer.berkeley.edu/smcat/.
Reiter, L., 1990, Earthquake Hazard Analysis: Issues and Insights, Columbia University Press, 254 pp.
Stewart, J. P., Chiou, S.-J., Bray, J. D., Graves, R. W., Somerville, P. G. and Abrahamson, N. A. , 2001 ,
Ground motion evaluation procedures for performance-based design, PEER Report 2001/09, Pacific
Earthquake Engineering Research Center, University of California, Berkeley.
Figure 2. Scaling of time history of 1994Northridge earthquake (78 Stone Canyon record) to fit design response
spectrum of seismic region 1 and soil group II

Figure 3. Selection of three accelerograms to best fit design response spectrum of seismic region 1 and soil
group II

Original Iteration # 3 Iteration # 10

Figure 4 (a). Iterative frequency scaling of 1994 Northridge earthquake (78 Stone Canyon record) to fit design
response spectrum of seismic region 1 and soil group II (Frequency content)
Iteration # 1
Iteration # 3
Iteration # 10

Figure 4 (b). Iterative frequency scaling of 1994 Northridge earthquake (78 Stone Canyon record) to fit design
response spectrum of seismic region 1 and soil group II (Response Spectra)
Iteration # 1
Iteration # 3

Figure 5. Iterative frequency scaling of Tabas 1978, Iran earthquake (70 Boshrooyeh record) to fit design
response spectrum of seismic region 1 and soil group II