Amplitude-Scaled versus Spectrum-Matched Ground

Motions for Seismic Performance Assessment

YeongAe Heo1; Sashi K. Kunnath, F.ASCE2; and Norman Abrahamson3
Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.

Abstract: The need to consider only a small number of ground motions combined with the complexities of response sensitivity to
both modeling choices and ground motion variability calls for an assessment of current ground motion selection and modification methods
used in seismic performance evaluation of structures. Since the largest source of uncertainty and variability arises from ground
motion selection, this study examines the suitability of two ground motion modification (GMM) schemes: magnitude scaling (wherein
the ground motion is uniformly scaled so that the resulting spectrum matches the amplitude of the design spectrum at the structural funda-
mental period) and spectrum matching. Comprehensive nonlinear time-history (NTH) simulations of two reinforced concrete moment frame
buildings are carried out to evaluate the GMM approaches in the context of seismic demand prediction. Findings from the investigation
indicate that spectrum matching is generally more stable than scaling both in terms of the bias as well as the resulting dispersion in
the predicted demands. It is also concluded that seven ground motions are inadequate to establish median demands for taller frames where
multiple modes influence structural response. Both methods are found to be sensitive to the choice of records for the cases investigated in this
study. DOI: 10.1061/(ASCE)ST.1943-541X.0000340. © 2011 American Society of Civil Engineers.
CE Database subject headings: Drift; Ground motion; Probability distribution; Reinforced concrete; Seismic effects.
Author keywords: Interstory drift; Ground motion modification; Probability distribution; Reinforced concrete; Seismic simulation.

Introduction
The selection of ground motions for use in nonlinear dynamic
simulations is becoming an increasingly critical component of
performance-based seismic evaluation. Given the limited database
of earthquake records that satisfy the required site parameters, it is
often necessary to select empirical recordings from other similar
sites and alter them suitably to meet the needs of the evaluation.
Current practice in nonlinear seismic evaluation is to select ground
motion records that best represent the characteristics of the
expected event (such as magnitude, distance, and fault type) at
the site. Structural engineers tend to favor the use of intensity scal-
ing methods, but since the intensity (such as spectral acceleration)
at critical periods can vary considerably from record to record, the
selected motions need to be modified to achieve certain target
intensities.
According to code requirements (ASCE 2005), the scaling of
selected ground motions for two-dimensional analysis should be
carried out in a manner such that the average value of the 5%
damped response spectra for the suite of motions is not less than
the design response spectrum for the site for periods ranging from
1 for near-fault sites where the inelastic deformation can be signifi-
Senior Researcher, Marine Research Institute of Samsung Heavy
Industries, Korea; formerly, Graduate Student Researcher, Dept. of Civil
and Environmental Engineering, Univ. of California, Davis, CA 95616.
E-mail: yaheo@ucdavis.edu
2 inelastic SDOF system are more appropriate. Kalkan and Chopra
Professor, Dept. of Civil and Environmental Engineering, Univ. of
California, Davis, CA 95616. E-mail: skkunnath@ucdavis.edu
3
Seismologist, Pacific Gas and Electric, San Francisco, CA 94105;
Adjunct Professor, Dept. of Civil and Environmental Engineering, Univ.
of California, Davis, CA 95616.
Note. This manuscript was submitted on April 20, 2010; approved on
October 27, 2010; published online on October 29, 2010. Discussion period
open until August 1, 2011; separate discussions must be submitted for in-
dividual papers. This paper is part of the Journal of Structural Engineer-
ing, Vol. 137, No. 3, March 1, 2011. ©ASCE, ISSN 0733-9445/2011/3-
278–288/$25.00.
0.2 to 1.5 times the fundamental period, T, of the structure for the
direction of response being analyzed. Depending on whether
three or seven records are used, the maximum or mean value of the
response parameter of interest is to be considered in design or
evaluation. In addition to code-based scaling, modifying ground
motions by scaling them to the spectral acceleration at single spec-
tral periods, such as the first-mode elastic period Sa ðT 1 Þ have also
been used, for example, Kunnath et al. (2006). Shome and Cornell
(1998) and Shome et al. (1998) demonstrated that seismic demands
are strongly correlated with the elastic single-degree-of-freedom
(SDOF) oscillator response acceleration at the fundamental period
of the system. More recently, Baker and Cornell (2006) suggested
that if records are properly selected based on spectral shape, the
reduction in bias and variance of resulting structural response
estimates are comparable to the reductions achieved by using a
vector-valued measure of earthquake intensity. In contrast, certain
scaling methods such as scaling to target peak ground acceleration
(PGA) produce biased estimates with large scatter in response
(Nau and Hall 1984; Vidic et al. 1994; Shome and Cornell
1998). Further, because scaling methods do not explicitly consider
the inelastic behavior of the structure, they may not be appropriate
cantly larger than the deformation of the corresponding linear sys-
tem. For such sites, scaling methods that are based on the inelastic
deformation spectrum or consider the response of the first-mode
(2010) used these concepts to develop a modal-pushover-based-
scaling (MPS) procedure for selecting and scaling earthquake
ground motion records. Other approaches to scaling include ground
motion modification over a selected period band (Alavi and
Krawinkler 2004; Kalkan and Kunnath 2006). A good review of
several other scaling methods including scaling to effective peak
acceleration, Arias intensity based parameter, effective peak veloc-
ity, and maximum incremental velocity is reported by Kurama and
Farrow (2003), where it is shown that scaling methods work well

278 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011

J. Struct. Eng. 2011.137:278-288.

 for ground motions representative of stiff soil and far-field
conditions, but lose their effectiveness for soft soil and near-field
conditions.
As an alternative to scaling by a constant factor, the method of
spectral matching (in time or frequency domain) is gaining atten-
tion among researchers and practitioners. With growing demand in
using nonlinear time-history (NTH) analysis for seismic assess-
ment of structures, this method is attractive for engineers because
it minimizes the number of records needed to obtain reasonable
accuracy of the performance estimate; thereby the computational
cost is significantly reduced. In this method, the frequency content
Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.
and phasing of actual recordings are manipulated to match a
smooth target spectrum (e.g., Silva and Lee 1987; Lilhahand
and Tseng 1989; Bolt and Gregor 1993; Carballo and Cornell
2000; Hancock et al. 2006). While scaling methods keep the fre-
quency content of ground motions intact, spectral matching tech-
niques may alter the physical characteristics of the accelerograms
though more recent methods also try to minimize the alteration of
the overall nonstationary characteristics of the motions.
In the case of scaling approaches, the central questions that need
to be answered include the following: How many ground motions
should be used? What number should form the basis for deciding
whether the average or maximum response measure should be con-
sidered? Should the scaling be limited to a single period or a period
range? If two alternative sets of records produce different perfor-
mance estimates, how should the findings be reconciled? For spec-
trum-matching methods, one of the chief concerns is the
modification of the frequency content that can distort the nonsta-
tionary characteristics of the time series. Other concerns include the
uncertain effects of leveling or flattening all the peaks and troughs
of the spectrum on the computed structural response.
Both scaling and matching methods are considered in the present
study to evaluate not only the effectiveness of each methodology but
also to assess the reliability of structural performance estimates for a
given site hazard. A study by Iervolino and Cornell (2005) showed
that there is little evidence to support the need for a careful site-
specific process of record selection based on magnitude and distance
and that concern over scenario-to-scenario record scaling may not be
justified. Hence, this study focuses primarily on ground motion char-
acteristics such as spectral acceleration that affect structural response
characteristics rather than site-specific sources.
Buildings Considered in Evaluation
The two ground motion modification methods are evaluated by
considering the bias and dispersion of seismic demand predictions
Fig. 1. Building plan and typical frame elevation of buildings
JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011 / 279
J. Struct. Eng. 2011.137:278-288.
of a four-story and 12-story RC frame building. The structures
are assumed to be standard office buildings located in San
Francisco (37.46N, 122.25W). The lowest level of the building
is 4.57 m (15 ft) high while the remaining floor heights are
3.66 m (12 ft). The plan dimensions of the building are 36:6 ×
36:6 m (120 × 120 ft) with five equal bays in each direction.
The choice of a symmetric floor plan allows a single typical frame
in either direction to be analyzed as a two-dimensional frame. The
typical plan and elevation of the buildings is shown in Fig. 1.
The buildings are designed to meet the provisions of Seismic
Design Category D as specified in ASCE 7-05 (2005). The follow-
ing parameters were used in the design:
SS ¼ 1:50 g; S1 ¼ 0:67 g; I ¼ 1:0;
C d ¼ 5:5; R ¼ 8:0
Using the preceeding values, the design base shears are
2,325 kips and 1,386 kips for the 12-story and four-story buildings,
respectively. The approximate code-based fundamental periods are
1.43 s and 0.55 s while the true first mode periods based on input
material and element properties were determined to be 2.1 s and
0.88 s. The final design is based on using normal weight concrete
(150 pcf) with a compressive strength f 0c ¼ 4;000 psi and reinforc-
ing steel with a nominal yield strength f y ¼ 60;000 psi. Section
sizes and details of the required flexural and shear reinforcement,
all of which conform to the requirements of ACI-318 (2009), are
presented in Table 1.
Ground Motion Selection and Modification
A study by Watson-Lamprey (2007) introduced the notion of a
“point of comparison” wherein the “true” (statistically most likely)
response of a system is established. It is proposed that the “true”
structural response measure is best obtained by a high-end predic-
tion, wherein NTH simulations are carried out using an extremely
large number of scaled and unscaled ground motions. A response
model is then derived that relates relevant structural demand mea-
sures to critical ground motion parameters. Such a model not only
provides the best estimate of structural demand for any site but also
provides a convenient baseline against which the validity of differ-
ent ground motion modification methodologies can be assessed.
The notion proposed by Watson-Lamprey (2007) is adopted in
the present study with a change in that only unscaled ground
motions are used. Watson-Lamprey (2007) used scaled ground
motions to drive the structure well into the nonlinear range. In this
 Table 1. Component Sizes and Reinforcement Details for Both Buildings
Story levels
1-2 3 4
Column Size 711 × 711 610 × 610 559 × 559
Long reinf 16#29 16#25 16#25
Trans reinf #9.5@152 #9.5@152 #9.5@127
4-story Beam Size 711 × 558 610 × 508 559 × 457
Long reinfa 16#29 14#29 14#29
Trans reinf #12.7@140 #12.7@127 #12.7@114
Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.

Story levels
1-3 4-6 7-8 9-10 11-12
Column Size 864 × 864 813 × 813 762 × 762 711 × 711 559 × 559
Long reinf 20#36 16#32 16#29 16#29 16#25
Trans reinf #12.7@127 #9.5@152 #9.5@152 #9.5@152 #9.5@140
12-story Beam Size 864 × 762 813 × 711 762 × 660 711 × 559 559 × 457
Long reinfa 14#32 14#32 16#29 16#29 14#29
Trans reinf #9.5@190 #9.5@178 #9.5@165 #9.5@140 #9.5@114
Note: n#d ¼ number of bars “n” and bar # with diameter “d”; #d@s ¼ bar#with diameter “d” at spacing “s”; all dimensions in mm.
a
Total reinforcement (equal top and bottom reinforcement).

paper, ground motions that cause nonlinear response is also of pri-
mary interest, therefore, a subset of 200 ground motions from the
Pacific Earthquake Engineering Research (PEER) Next Generation
Attenuation (NGA) database (PEER 2005), whose PGA exceeded
0.2 g, was used in the simulations. The computed inelastic response
models to these 200 records are used to determine the dependence
of the structural response on the spectral acceleration at a suite of
spectral periods. This model of the structural response is then used
to compute the expected response and the variability of the
response given the design spectral values. This distribution of
the predicted response, given the design spectrum, is termed the
“true” response—from the perspective that record-to-record vari-
ability and other inherent uncertainties in the ground motions have
been more comprehensively considered in this data set than in a
smaller subset of records. A methodology that uses a limited data
set (ground motions) can be classified as being reliable if the
resulting performance model comes reasonably close to the “true”
prediction (in terms of bias and variance). Note that the term pre-
diction model refers to a model that provides a demand or perfor-
mance measure (interstory drift demand or damage index, for
example) as a function of ground motion parameters. Fig. 2 shows
Fig. 2. 5% damped response spectra of selected 200 ground motions
and comparison of mean spectrum with ASCE 7-05 design spectrum
(note: T1, T2, and T3 are the first, second, and third mode periods;
subscripts indicate whether the modes are for the 4- or 12-story frame)
the 5% damped response spectra for all records in the database.
Also shown alongside is the mean spectrum of the 200 records
superimposed on the ASCE 7-05 (2005) design spectrum for the
site. The mean spectrum is observed to be slightly lower than
the design spectrum. Complete details of the ground motion char-
acteristics are described in Heo (2009).
Scaling Methods
Numerous approaches to ground-motion scaling exist in current lit-
erature. The simplest and most commonly used amplitude-scaling
approach is selected in the present study—namely, scaling of the
motion such that the ordinate of the spectral acceleration matches
the design spectral acceleration at the fundamental period of the
structure in the direction of loading. As pointed out in the introduc-
tion, this method is also known to typically have scatter less than
other scaling approaches.
A limited subset of 17 records was randomly selected from the
larger bin of 200 records. Three bins of seven records each were
then created from this set of 17 records and are referred to in this
paper as Set 1, Set 2, and Set 3. The accelerograms in each set
were selected to represent three possible scenarios: records whose
mean spectral value at the fundamental period was lower than
the design spectrum, records with a mean spectral value higher than
the design spectrum (note that this outcome was not achieved for
the four-story structure), and a final set that comprised records
whose mean spectral acceleration was relatively closer to the design
spectrum at the fundamental period. Each of the records in the bin
was scaled to match the spectral acceleration of the design spectra
at the fundamental period of the structure. The 5% damped
response spectrum of the records in each set as well as the mean
spectrum for both the original set and the scaled set is displayed in
Fig. 3. Also shown in these plots are the first-mode periods of the
two buildings. Tables 2–4 list the scale factors applied to each
record.
An examination of the spectra in Fig. 3 indicates that the mean
of the original records in Set 1 was significantly lower than the
corresponding design spectral value at the fundamental period
while the records in Set 3 produce a mean spectral demand
(at T1) that is marginally higher than the design spectrum. The
mean of the Set 2 motions were also lower but closer to the design

280 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011

J. Struct. Eng. 2011.137:278-288.

Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. Spectra of amplitude-scaled records: (a) Set 1; (b) Set 2; (c) Set 3 (Notation ¼ orgAvg: average spectrum of unscaled records; Design:
ASCE 7-05 design spectra; sclAvg04: average spectrum of scaled records for 4-story frame; sclAvg12: average spectrum of scaled records for
12-story frame)

Table 2. Intensity and Scale Factors of Scaled Record Set 1 Table 3. Intensity and Scale Factors of Scaled Record Set 2
4 Story 12 Story 4 Story 12 Story
GM # PGA SaðT1Þ SF SaðT1Þ SF GM # PGA SaðT1Þ SF SaðT1Þ SF
1 0.386 1.482 1.51 0.128 2.49 1 0.386 1.482 1.51 0.128 2.49
3 0.597 1.681 0.80 0.129 2.47 3 0.597 1.681 0.80 0.129 2.47
5 0.447 1.416 1.45 0.135 2.36 5 0.447 1.416 1.45 0.135 2.36
6 0.400 1.350 0.78 0.102 3.12 16 0.464 1.378 1.27 0.520 0.61
26 0.402 1.594 0.69 0.268 1.19 81 0.349 1.566 0.95 0.216 1.47
71 0.241 1.338 2.05 0.093 3.42 82 0.536 1.756 0.75 0.421 0.76
145 0.208 1.749 2.96 0.155 2.05 145 0.208 1.749 2.96 0.155 2.05

JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011 / 281

J. Struct. Eng. 2011.137:278-288.

 Table 4. Intensity and Scale Factors of Scaled Record Set 3 where Pi = the polarity of the peak response of the oscillator.
4 Story 12 Story Hancock et al. (2006) shows that the response of an adjustment
time series should be equal to ΔRi
GM # PGA SaðT1Þ SF SaðT1Þ SF
16 0.464 1.378 1.27 0.520 0.61 X
N

77 0.352 1.323 1.53 0.358 0.89 ΔRi ¼ bj f j ðtÞ ð2Þ

j¼1
78 0.283 1.923 1.12 0.202 1.58
82 0.536 1.756 0.75 0.421 0.76
where f j ðtÞ = a set of adjustment functions and bj = the set of
86 0.498 1.309 1.18 0.360 0.88
amplitudes of the adjustment functions. The modified amplitude
91 0.333 1.214 1.96 0.226 1.41 of the responses to the wavelet (which can be considered as a scale
Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.

98 0.296 2.096 1.83 0.507 0.63 factor) is determined by not only the misfit at each spectral point
but also neighboring spectral points

spectrum at the fundamental period that Set 1 motions. Therefore,
the scaling process results in a mean spectrum that is generally
higher than the design spectrum (particularly at smaller periods,
which influence higher modes) for the first two sets and a mean
spectrum that is somewhat lower than the design spectrum for
Set 3. These facts are also reflected in the scale factors shown
in Tables 1–3. For the four-story frame, since the scaling is anch-
ored at a lower period, the resulting mean spectra for all sets are not
as significantly altered as in the case of the 12-story building. How-
ever, the mean spectral demands at periods higher than the first
mode are lower than the design spectral values for Set 1, almost
similar to the design spectra for Set 2, and higher than the design
spectral values for Set 3. These observations provide valuable in-
sights into examining and evaluating the features of the building
responses to the three record sets.
Spectrum Matching Methods
Although the objective of the scaling approach is to match only a
single spectral value at a target period, the concept of spectral
matching is to modify the original acceleration time series to match
the entire range of the target spectrum with minimal alteration of
the velocity and displacement history of the record. The approach
adopted in this study is based on the time domain spectral matching
procedure proposed by Hancock et al. (2006) and assumes that the
time of the peak response does not change because of wavelet ad-
justment. Given N target spectral points to match, at the ith target
period, the spectral misfit to be altered can be computed by the
difference between the target spectral value (Qi ) and the initial time
series spectral value (Ri )
ΔRi ¼ ðQi
Ri ÞPi ð1Þ
b ¼ C
1 δR ð3Þ
Each component of a square matrix C = the amplitude of the
wavelet response for the jth spectral point at the peak oscillator
time (ti ) of the initial time series response for the ith spectral point.
A sample matched spectrum using this procedure is shown in
Fig. 4(a). The mean spectrum for an ensemble of seven matched
records in illustrated in Fig. 4(b). Since the matching process
results in a spectrum that closely matches the target spectrum, the
mean spectrum of multiple records is almost indistinguishable from
the design spectrum; hence, only a sample mean spectrum is shown.
Development of Seismic Demand Prediction Model
The assessment of GMM approaches begins with the development
of a statistically reliable prediction model of the selected seismic
demand parameter. In the present study, the prediction model is
essentially a regression model based on the analysis of large data
sets that consist of empirical data (such as ground motion param-
eters from recorded motions) and simulated data from detailed
numerical procedures (such as nonlinear structural simulations).
The purpose of a prediction model for structural response is
twofold: It provides a simple means to predict a reliable probabi-
listic structural response quantity with considerable reduction in
computational effort, and it offers a means to compare different ap-
proaches in seismic structural assessment. In the present study, it is
used to evaluate ground motion selection and modification
methods.
The typical procedure for developing a regression model of a
sample model response parameter Y in terms of a set of predictive

Fig. 4. Spectrum matching: (a) sample spectral matched record; (b) mean spectrum (Notation ¼ Design: ASCE 7-05 design spectra; orgAvg: average
spectrum of unscaled records; matAvg: average spectrum of 7 spectrum-matched records for Set 1)

282 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011

J. Struct. Eng. 2011.137:278-288.

 parameters Xi consists of the following steps. An expression of the
following form can be generated from available (simulated) data
lnðYÞ ¼ c0 þ c1 lnðX1Þ þ       þ cn lnðXnÞ ð4Þ
In the preceeding expression, n predictive parameters are
selected. The constants are determined by data analysis. The
importance of the selected variables can be established by examin-
ing the magnitude of the constants, the standard deviation of the
residuals, and the correlation among the predictive parameters.
The correlation can typically be assessed by examining the
Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.
following parameter:
δln Y
εδln Y ¼ ð5Þ
σδln Y
where εln Y = the residual normalized by the standard deviation of
the residual. The residual of the model Y is determined from
δ ln Y ¼ ln Y
Eðln YÞ ð6Þ
Eðln YÞ = the expected value of the parameter Y as established
by the regression model. Once it is confirmed that a correlation
exists between some of the predictive parameters, these variables
should be linked by correlation functions. If, for example, εln X1 ,
εln X2 , and εln X3 are correlated, then the following correlation func-
tions can be developed:
εln X2 ¼ c0 þ c1 εln X1 ð7Þ
εln X3 ¼ c00 þ c01 εln X1 þ c02 εln X2 ð8Þ
Note that in the preceeding expressions, each of the predictive
parameters can be formulated as a separate set of regression mod-
els. Hence, they can be applied with generality to the development
of regression models for both the structural response in terms of
ground motion parameters and ground motion parameters in terms
of earthquake parameters. Ideally, the objective of the probabilistic
assessment is to establish the probability of exceeding a certain
damage threshold given an earthquake scenario. To accomplish this
objective, it is also essential to develop a ground motion prediction
model in terms of earthquake parameters (such as moment magni-
tude of the earthquake and closest distance to the rupture zone)
ln SaT1 ¼ c0 þ c1 · ln M þ c2 · ln R ð9Þ
In the preceding example, only a single ground motion param-
eter and two earthquake parameters are considered. Additional
forms of the preceding model are obviously possible, and the
choice of parameters depends on numerous factors. The develop-
ment of a ground motion prediction model based on earthquake site
and source characteristics is beyond the scope of the present study
because a large body of literature currently exists on this topic in the
field of seismology. In the present study, spectral ordinates are gen-
erated by using seismic design provisions in ASCE 7-05 (2005).
Modeling of True Response
The procedure described in the previous section is applied to
develop predictive models of the response of the two RC buildings
selected for this study. As previously indicated, the true response
prediction model is generated through a complete set of 200 NTH
simulations of each building. All nonlinear simulations were
carried out using the open-source software OpenSEES (2009).
Frame elements were modeled using nonlinear beam-with-hinges
JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011 / 283
J. Struct. Eng. 2011.137:278-288.
elements wherein the hinge lengths were set to 10% of the element
length. This value was based on preliminary numerical validation
studies using observed experimental response of cyclically loaded
columns under constant axial load. Concrete 01 and Steel 02 were
used as the material models: the former ignores the tensile resis-
tance of concrete whereas the latter is a complete cyclic description
of the behavior of reinforcing steel based on the uniaxial Giuffre-
Menegotto-Pinto model with isotropic strain hardening. A 5%
mass-proportional damping in the first mode was specified in all
the dynamic analyses.
The following ground motion intensity measures are considered
in the development of the response model: spectral acceleration at
the fundamental period, SaT1 , the spectral acceleration at the sec-
ond-mode period SaT2 , the spectral acceleration at the third-mode
period SaT3 , the spectral acceleration corresponding to 1.5 times
the fundamental period SaT4 , and the spectral acceleration at 2.0
times the fundamental period SaT5 . The parameters SaT4 and
SaT5 were selected after preliminary studies indicated that the soft-
ened fundamental period (following some structural damage) can
be a reliable parameter, particularly for ground motions that push
the system well into the inelastic response region. The primary
response variable considered in this study is the maximum inter-
story drift (IDR). Before determining a probable form of the
response model, the correlation between peak IDRs and the ground
motion intensity measures are examined independently. The corre-
lations of IDR with respect to the spectral accelerations at the first
three elastic modal periods are plotted in Fig. 5 in natural logarith-
mic scale. For the four-story frame it is evident that the response is
well correlated only with respect to the first-mode spectral accel-
eration while all three intensity measures are reasonably well
correlated with IDR for the 12-story frame. This outcome might
suggest that the inclusion of SaT2 and SaT3 in the response model
may not be necessary for the four-story structure. However, these
direct correlations by themselves only provide a measure of the
linear dependence of the demand parameter on each of the selected
variables. Among the features of the response that are of particular
interest in seismic evaluation is the variation of the response
measure as a function of some intensity measure at higher demand
levels (in the inelastic range). Expressing the demand as a linear or
nonlinear combination of one or more intensity measures and
examining the resulting residuals [Eq. (6)] is a simple way to
assess the effectiveness of a predictive demand model. In this study,
the following regression models, where the peak (IDR) expressed
in percentage values) is selected as the primary response parameter,
were considered:
Model A: Eðln IDRÞ ¼ c0 þ c1 ln SaT1 þ c2 ln SaT2 þ c3 ln SaT3
þ c4 ln SaT4 þ c5 ln SaT5 ð10Þ
Model B: Eðln IDRÞ ¼ c0 þ c1 ln SaT1 þ c2 ln SaT2
þ c3 ln SaT3 þ c4 ln SaT4 ð11Þ
Model C: Eðln IDRÞ ¼ c0 þ c1 ln SaT1 þ c2 ln SaT2 þ c3 ln SaT3
ð12Þ
Model D: Eðln IDRÞ ¼ c0 þ c1 ln SaT1 þ c2 ln SaT2 ð13Þ
Model E: Eðln IDRÞ ¼ c0 þ c1 ln SaT1 ð14Þ
Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.

Fig. 5. Correlation of structural demand measures with primary ground motion parameters [Note: (log) refers to natural logarithm]

Results of the data analysis using the response prediction mod-
els expressed by Eqs. (10)–(14) are presented in Tables 5 and 6.
Note that σ ln(IDR) and μ(IDR) refer to the standard deviation
and median of the IDR. The regression coefficients show that
c1 and c4 have a significant effect on the response of the four-story
building; whereas c1 , c2 , and c3 have a significant effect on the
response of the 12-story building. In general, the introduction of
additional variables improves the reliability of the prediction mod-
els, as shown by the reduction in the standard deviation for models
with more terms; however, it is also evident that the spectral accel-
eration at first mode SaT1 is the most significant parameter control-
ling the response of the 4-story frame, but the spectral acceleration
at the second mode (SaT2 ) is most significant for the 12-story
frame. For the 12-story structure, Model C is adequate to represent
the response because the additional reduction in the standard
deviation for Models A and B is negligible. For the 4-story struc-
ture, model B is adequate to represent the response.
Figs. 6 and 7 show the IDR residuals for the 4- and 12-story
frames, respectively, based on Models A and C for the 200
simulations as a function of selected ground motion intensity
parameters. The regression model used has a linear scaling of
the response with spectral acceleration. The residual plots can
be used to determine if nonlinear scaling effects are observed.
Nonlinear effects would be seen as a curvature in the residuals
at the higher intensity ground motions. These residual plots do
not show a curvature except for the T 4 dependence for the 12-story
building for model A—there is a trend to increase the response for
higher SaT4 values, but this trend is not well constrained. Part of the
reason that nonlinear effects are not seen in the residuals is that
most of the unscaled ground motions used in the study did not
drive the systems well into the inelastic range of response. An
alternative approach would have been to scale the motions, as in
the study by Watson-Lamprey (2007), and observe trends at higher
intensity levels.
The probability density functions of Models A, C, and E for
both structures, shown in Fig. 8, confirm that though Model A
has the least dispersion in both cases, Models C and E in the case
of the 4-story frame, or Models A and C in the case of the 12-story

Table 5. Regression Coefficients for Different Response Models Using Table 6. Regression Coefficients for Different Response Models Using
4-Story Building 12-Story Building
Model A Model B Model C Model D Model E Model A Model B Model C Model D Model E
σlnðIDRÞ 0.2593 0.2601 0.2846 0.2856 0.2883 σlnðIDRÞ 0.3548 0.3563 0.3563 0.3736 0.4469
μðIDRÞ (%) 1.4955 1.5078 1.4036 1.4272 1.4168 μðIDRÞ (%) 1.5119 1.4916 1.4892 1.4425 1.4534
c0 0.7151 0.7414 0.5492 0.5676 0.5667 c0 0.8611 0.8235 0.8189 0.8200 1.0904
c1 0.4921 0.4871 0.7685 0.7746 0.7981 c1 0.2759 0.2758 0.2888 0.2813 0.6264
c2 0.1340 0.1336 0.1297 0.0882 c2 0.3800 0.3793 0.3783 0.5527
c3
0:0564
0:0527
0:0690 c3 0.2889 0.2973 0.2971
c4 0.3547 0.2906 c4
0:0897 0.0114
c5
0:0653 c5 0.0983

284 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011

J. Struct. Eng. 2011.137:278-288.

Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.
Fig. 6. Correlation of IDR residuals with ground motion intensity parameters for 4-story frame
frame, both provide similar measures of the median response and
variability of the response.
Spectrum Matching versus Scaling
The two GMM methods are evaluated against the so-called true
solution represented by the predictive regression model generated
through 200 high-end numerical simulations presented in the
previous section. The regression model considered in the evaluation
corresponds to the function representing Model C [Eq. (12)] but the
findings can be generalized to the other models as well. Recall from
the preceding discussion that a subset of 17 records was randomly
selected from the 200 ground motions used to establish the true
response models. Three records of 7 motions each were then
selected from this bin to create three possible scenarios.
To compare the stability and bias of the results obtained with
sets of seven scaled and matched ground motions, the peak
IDRs for each simulation and arithmetic mean for each set are
superimposed on the probability distribution of the corresponding
true solution for each frame. The comparisons are presented in
Figs. 9(a)–9(f) for each of the three ground motion sets and for both
frame structures. Figs. 9(a) and 9(b) compare computed demands
for both frames subjected to the ground motions comprising Set 1.
In the case of the four-story frame, the mean of the maximum
IDRs using scaled motions is 1.52%, whereas that of the matched
JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011 / 285
J. Struct. Eng. 2011.137:278-288.
motions is 1.42%. Although the dispersion in the computed IDRs is
comparable for both methods, the mean estimate using spectrum-
matched records is closer to the true mean estimate of 1.40% (based
on Model C). In the case of the 12-story frame, the results using
spectrum-matched records are clearly more consistent than scaled
records in terms of both the mean and dispersion of the computed
maximum IDRs. The true mean of the expected peak IDR is
approximately 1.49%, whereas the mean estimate is 1.46% using
matched motions and 2.06% using scaled motions.
Results for ground motion Set 2 and Set 3 are displayed in
Figs. 9(c) and 9(d) and Figs. 9(e) and 9(f), respectively. In both
cases, the observations noted for Set 1 with respect to the mean
and variability of the computed maximum IDR demands remains
valid. The mean IDR for the four-story frame using matched Set 2
ground motions is 1.45%, which is slightly higher than the true
estimate but significantly better than the estimated mean value
of 1.58% for scaled motions. In the 12-story frame, the mean
estimate using spectrum-matched records is 1.33% (which is lower
than the true mean of 1.49%), whereas the mean IDR demand using
scaled records is 1.83%. For the ground motions from Set 3, the
mean IDR estimates for the four-story frame using matched and
scaled motions are 1.52% and 1.57%, respectively. For the 12-story
frame, the mean IDR using matched ground motions is 1.35%
(which is lower than the true mean), but the mean estimate of
1.15% using scaled motions is significantly lower than the
true mean.
Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.

Fig. 7. Correlation of IDR residuals with ground motion intensity parameters for 12-story frame

Fig. 8. IDR distribution for three regression models: (a) 4-story frame; (b) 12-story frame

In addition to the previously noted general observations, it is
observed that the predicted mean for matched motions is closer
to the true mean for the four-story frame than for the 12-story struc-
ture. Of the two frames, there is considerably more variability in the
predictions for the for 12-story frame for both scaled and matched
motions. This outcome indicates that additional parameters (such as
higher modes and modal changes attributable to inelastic effects)
are influencing the response of the taller frame, hence more ground
motions need to be considered to reduce the bias in the computed
demands.
Concluding Remarks
Findings from a comprehensive set of NTH simulations of two
reinforced concrete moment frame structures indicate that peak

286 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011

J. Struct. Eng. 2011.137:278-288.

Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.

Fig. 9. Stability and bias in predicted responses using scaled and matched motions: (a) 4-story frame (Set 1); (b) 12-story frame (Set 1); (c) 4-story
frame (Set 2); (d) 12-story frame (Set 2); (e) 4-story frame (Set 3); (f) 12-story frame (Set 3) (Notation ¼ Model C: probability distribution of true
solution based on Model C; Scaled: Peak IDR for each scaled ground motion; Matched: Peak IDR for each matched ground motion; AVGs : Average
of max IDR of all scaled motions; AVGM : Average of max IDR of all matched motions)

IDR demands estimated from two different ground motion scaling
methods are generally a function both structural parameters and
ground motion characteristics. For the two case studies investi-
gated, the estimates of the peak IDR demands resulting from a
relatively small subset (seven ground motions were used in each
subset in the present study) of spectrum-matched records were
consistently closer to the median estimates of the true solution than
estimates from records of the same subset that were scaled to match
the target design spectral accelerations at the fundamental period of
the structures. More importantly, the dispersion in the computed
IDRs was generally smaller for spectrum-matched records than
for scaled records. On the basis of the observed variability, it
may also be concluded that seven ground motions are inadequate
for the 12-story structure, indicating that taller or more complex
structures that are influenced by higher modes, inelastic effects,
and other ground motion parameters will require a larger subset

JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011 / 287

J. Struct. Eng. 2011.137:278-288.

 of records to arrive at statistically reliable results. Since spectrum-
matched records, like scaled records, produced mean demand
estimates that were both lower and higher than the true mean val-
ues, the relative accuracy or reliability of either approach cannot be
assessed with certainty since all critical ground motion parameters
associated with structural demand are yet to be identified. It can
also be argued that the spectral accelerations of ground motions
at elastic modal periods of the system are not necessarily reliable
ground-motion intensity measures. Further studies using scaled
motions that produce larger inelastic drift demands, a significantly
larger set of ground motion bins, additional demand parameters
Downloaded from ascelibrary.org by SEOUL NATIONAL UNIVERSITY LIB on 10/17/13. Copyright ASCE. For personal use only; all rights reserved.
(such as member plastic rotations and heightwise distribution of
demands), enhanced performance measures (such as damage mod-
els as opposed to IDRs), and demand models that utilize additional
ground motion intensity measures are still needed to further exam-
ine the advantages and shortcomings of the two ground motion
modification methods investigated in this study.
References
ASCE. (2005). Minimum Design Loads for Buildings and Other Structures,
ASCE/SEI 7-05, Reston, VA.
Alavi, B., and Krawinkler, H. (2004). “Behavior of moment-resisting frame
structures subjected to near-fault ground motions.” Earthquake Eng.
Struct. Dyn., 33(6), 687–706.
American Concrete Institute (ACI). (2009). “Building code requirements
for structural concrete,” ACI 318-09, Farmington Hills, MI.
Baker, J. W., and Cornell, C. A. (2006). “Spectral shape, epsilon and record
selection.” Earthquake Eng. Struct. Dyn., 35(9), 1077–1095.
Bolt, B. A., and Gregor, N. J. (1993). “Synthesized strong ground
motions for the seismic condition assessment of the eastern
portion of the San Francisco Bay Bridge.” Rep. UCB/EERC-93/12,
Univ. of California, Earthquake Engineering Research Center,
Berkeley, CA.
Carballo, J. E., and Cornell, C. A. (2000). “Probabilistic seismic
demand analysis: Spectrum matching and design.” Rep. No. RMS-
41, Dept. of Civil and Environmental Engineering, Stanford
Univ., Stanford, CA.
Hancock, J., et al. (2006). “An improved method of matching response
spectra of recorded earthquake ground motion using wavelets.”
J. Earthquake Eng., 10(1), 67–89.
288 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / MARCH 2011
J. Struct. Eng. 2011.137:278-288.
Heo, Y. A. (2009). “Framework for damage-based probabilistic seismic
performance evaluation of reinforced concrete frames.”, Ph.D.
dissertation, Univ. of California, Davis, CA.
Iervolino, I., and Cornell, C. A. (2005). “Record selection for nonlinear
seismic analysis of structures.” Earthquake Spectra, 21(3), 685–713.
Kalkan, E., and Chopra, A. K. (2010). “Practical guidelines to select and
scale earthquake records for nonlinear response history analysis of
structures.” USGS Open-File Rep. 2010-1068, USGS, Washington,
DC, 126.
Kalkan, E., and Kunnath, S. K. (2006). “Effects of fling-step and forward
directivity on the seismic response of buildings.” Earthquake Spectra,
22(2), 367–390.
Kunnath, S. K., Larson, L., and Miranda, E. (2006). “Modeling considera-
tions in probabilistic performance-based seismic evaluation: Case study
of the I-880 viaduct.” Earthquake Eng. Struct. Dyn., 35(1), 57–75.
Kurama, Y., and Farrow, K. (2003). “Ground motion scaling methods for
different site conditions and structure characteristics.” Earthquake Eng.
Struct. Dyn., 32(15), 2425–2450.
Lilhhand, K., and Tseng, W. S. (1989). “Development and application of
realistic earthquake time histories compatible with multiple damping
design spectra.” Proc. of the 9th World Conf. on Earthquake Engineer-
ing, Tokyo-Kyoto, Japan, 2, 819–830.
Nau, J. M., and Hall, W. J. (1984). “Scaling methods for earthquake
response spectra.” J. Struct. Eng., 110(7), 1533–1548.
OpenSEES. (2009). “Open system for earthquake engineering simulation.”
Pacific Earthquake Engineering Research Center (PEER), Univ. of
California Berkeley, Berkely, CA. 〈http://opensees.berkeley.edu〉.
Pacific Earthquake Engineering Research Center (PEER). (2005). “NGA—
Next generation attenuation project,” 〈http://peer.berkeley.edu/nga/〉.
Shome, N., and Cornell, C. (1998). “Normalization and scaling accelero-
grams for nonlinear structural analysis.” Proc. of the 6th U.S. National
Conf. on Earthquake Eng., Earthquake Engineering Research Institute,
Seattle.
Shome, N., Cornell, C., Bazurro, P., and Carballo, J. (1998). “Earthquakes,
records, and nonlinear responses.” Earthquake Spectra, 14, 469–500.
Silva, W., and Lee, K. (1987). “State-of-the-art for assessing earthquake
hazards in the United States.” Rep. 24, WES RASCAL Code for
Synthesizing Earthquake Ground Motions, Miscellaneous Paper
S-73-1, U.S. Army Engineer Waterways Experiment Station,
Vicksburg, MS.
Vidic, T., Fajfar, P., and Fischinger, M. (1994). “Consistent inelastic design
spectra: Strength and displacement.” Earthquake Eng. Struct. Dyn., 23,
507–521.
Watson-Lamprey, J. A. (2007). “Selection and scaling of ground motion
time series.”, Ph.D. dissertation, Univ. of California, Berkeley, CA.