Bull Earthquake Eng (2009) 7:273–291

DOI 10.1007/s10518-008-9097-y

ORIGINAL RESEARCH PAPER

Evaluation of reduction factors for high-damping design

response spectra

D. Cardone · M. Dolce · M. Rivelli

Received: 18 February 2008 / Accepted: 9 October 2008 / Published online: 29 October 2008
© Springer Science+Business Media B.V. 2008

Abstract High-damping response spectra are essential tools for the assessment and design
methods based on the equivalent elastic structure concept. They are also often used for the
analysis and design of structures with seismic isolation or energy dissipation systems. Many
formulations of the reduction factors have been proposed and included in seismic codes to
estimate high-damping response spectra from their 5% damping representation. They are
reviewed in the present paper. The accuracy of each of them in estimating the maximum
elastic response of structures with viscous damping ratios greater than 5% is assessed by
comparing exact and approximate displacement response spectra for three different damping
levels, namely 10, 20 and 30%, respectively. The comparison is referred to more than 120
ground motion records, relevant to earthquakes with magnitude between 6 and 8, epicentral
distance ranging from 1 to 100 km and Peak Ground Accelerations (PGA’s) greater than
0.1 g. The comparison between exact and approximate response spectra is carried out for both
single earthquakes and groups of earthquakes with similar magnitude and epicentral distance.
The drawbacks of using the same damping reduction factor to estimate both maximum
displacement response and design seismic forces are also addressed.

Keywords Elastic response spectra · Viscous damping ratio · Damping

reduction factor · Natural earthquakes

1 Introduction

High-damping elastic response spectra have several important applications in the design/
assessment of earthquake-resistant structures. One of them, for instance, regards the design

D. Cardone (B) · M. Rivelli

DiSGG, University of Basilicata, Macchia Romana Campus, 85100 Potenza, Italy
e-mail: donatello.cardone@unibas.it
M. Dolce
Italian Department of Civil Protection,
via Vitorchiano 4, 00189 Rome, Italy

123
274 Bull Earthquake Eng (2009) 7:273–291

of structures equipped with seismic isolation (Naeim and Kelly 1999) or energy dissipation
(Constantinou et al. 1998) systems. Other applications refer to all those design/assessment
methods based on the substitute structure concept (Gulkan and Sozen 1974; Shibata and Sozen
1976), in which the inelastic behavior of the structure is described through an equivalent
single-degree-of-freedom (SDOF) elastic system, characterised by a reduced stiffness and
an increased viscous damping. Noticeable examples of design/assessment methods based on
the substitute structure concept are the Direct Displacement-based Design (DDBD) approach
(Priestley 2003) and the Capacity Spectrum Method, implemented in the ATC-40 for the
evaluation and retrofit of concrete buildings (ATC-40 1996).
In the above mentioned cases, the high-damping response spectra are derived from a
5%-damping reference spectrum by using proper reduction factors, which depend on the
equivalent viscous damping of the structure (>5%) and, in some cases, on its equivalent
period of vibration.
Until now, several different expressions of damping reduction factor have been proposed
and then adopted, in somewhat different form, in various seismic codes. Most of the currently
used damping reduction factors have been derived from the observation of the effects of
viscous damping on the maximum displacement response of elastic SDOF systems subjected
to either artificial or natural earthquakes.
In the USA, the results of the studies from Newmark and Hall (1973, 1982) have been
implemented in the guidelines for the evaluation and retrofit of existing structures (ATC-40
1996; FEMA-273 1997; FEMA-356 2000) and in the seismic codes for the design of structures
with seismic isolation and passive energy dissipation systems (UBC 1997; IBC 2000). The
results derived from Ashour (1987) were adopted in a past edition of the UBC (UBC 1994).
Finally, the results from Ramirez et al. (2002) have been implemented in the current provisions
for Seismic Regulations for New Buildings (NEHRP 2000). In Europe, the results of the
studies from Bommer et al. (2000) have been included in current version of the Eurocode
8 (EC8 EN 2004) and recently in the new Italian seismic codes (OPCM 2003, DM 2008).
Among the other international seismic codes, it is worth to mention the Japanese guidelines
for the design of seismically isolated structures (JPN 2001), which provide damping ratios
significantly lower than those adopted in the European and the US seismic codes. Other
remarkable expressions of reduction factors include those proposed by Wu and Hanson
(1989), Lin and Chang (2003), Priestley (2003), Tolis and Faccioli (1999) and Kawashima
and Aizawa (1986).
Recently, Bommer and Mendis (2005) showed that the damping reduction factor may be
affected by earthquake magnitude, source-to-site distance, and ground motion duration.
The main objective of this study is to evaluate the accuracy of different damping reduction
factors in the estimate of the maximum seismic response of elastic SDOF systems with
equivalent viscous damping greater than 5%.
A number of analogous studies have been conducted in the last years (Naeim and Kircher
2001, Lin 2007). However, they were mainly aimed at statistically deriving suitable damping
reduction factors based on ratios of response spectra of horizontal components of specific
earthquake ground motions (e.g., the 1999 Chi–Chi Earthquake, Taiwan) or earthquakes
recorded in certain regions (e.g., California, Alaska and Hawaii islands). Comparisons of
results from 12 Californian earthquakes with five existing expressions of damping reduction
factors are presented in Lin et al. (2005).
In this study, the response spectra obtained by solving the motion equations are compared
to those derived from the 5% exact counterpart by applying different damping
reduction factors. The analysis is performed based on 120 accelerograms, recorded all over
the world in the last 30 years, during 33 different earthquakes with magnitude between 6 and 8,

123
Bull Earthquake Eng (2009) 7:273–291 275

epicentral distance ranging from 1 to 100 km and Peak Ground Acceleration (PGA)
greater than 0.1 g. The comparison between exact and approximate spectra is carried out
for both single earthquakes and groups of earthquakes, with similar magnitude and epicen-
tral distance, and repeated for three different damping ratios, namely 10, 20, and 30%.

2 Overview of currently used damping reduction factors

The damping reduction factors have been introduced to get an approximate estimate of high-
damping elastic response spectra from their 5% counterpart, using the following equations:

Sd (T, ξ ) = η · Sd (T, 5%) (1)

PSa (T, ξ ) = η · PSa (T, 5%) (2)

in which Sd (T, 5%) and PSa (T, 5%) are the elastic displacement and pseudo-acceleration
response spectra for 5% damping ratio, while Sd (T, ξ ) and PSa (T, ξ ) are the corresponding
quantity for damping ratios ξ greater than 5%. T is the natural period of vibration of the SDOF
system. The same reduction factor η is used for both displacement and pseudo-acceleration
response spectra, since they are mutually related through the well-known approximate re-
lationship PSa = ω2 Sd , ω = 2π /T being the natural circular frequency of vibration of the
SDOF system.
Several different formulations of the damping reduction factors are available. The one
initially proposed by Newmark and Hall (1973), for instance, is made of three relationships:
⎧
⎨ 1.514 − 0.321 · ln(ξ )
η = 1.400 − 0.248 · ln(ξ ) (3)
⎩
1.309 − 0.194 · ln(ξ )

which hold in the constant-velocity, constant-acceleration and constant-displacement regions

of the spectrum, respectively. They were derived from median estimates of the maximum
displacement response of SDOF systems with damping ratios lower than 20%. In the ATC-40
and FEMA-273, the approximate method by Newmark and Hall (1973) is further simplified
by neglecting the relationship valid in the constant-displacement region of the spectrum. In
the FEMA-273, moreover, more conservative values are adopted for damping ratios equal to
or greater than 30%. In the UBC-97 and IBC 2000, finally, the relationships of Newmark-
Hall are reduced to a single value damping reduction factor, corresponding to that adopted
by FEMA-273 in the constant-velocity region of the spectrum.
The damping reduction factor proposed by Ashour (1987) can be expressed as:

0.05 · (1 − e−α·ξ )
η= (4)
ξ · (1 − e−0.05·α )

where α is a coefficient ranging from 18 to 65, according to the earthquake characteristics.

The studies carried out by Wu and Hanson (1989) resulted in the following multifaceted
expression:

λ(ξ, T )
η= (5)
λ(ξ = 5%, T )

123
276 Bull Earthquake Eng (2009) 7:273–291

in which:
λ = −0.349 · ln(0.095 · ξ ) for T = 0.1 s
λ = −0.547 · ln(0.417 · ξ ) for T = 0.5 s
λ = −0.471 · ln(0.524 · ξ ) for 0.5 < T < 3 s (6)
λ = −0.478 · ln(0.475 · ξ ) for T = 3 s
λ = −0.291 · ln(0.0473 · ξ ) for T = 10 s

The studies conducted by Ramirez et al. (2002) led to numerical values of the damping
reduction factor between those suggested by Ashour (see Eq. 4) and Wu-Hanson, for T = 0.1 s
(see Eq. 6).
More recently, Lin and Chang derived a period-dependent expression of the damping
reduction factor, based on the examination of the displacement response of elastic SDOF
systems with damping ratio between 2 and 50% and period of vibration ranging from 0.01 to
10 s, subjected to approximately 1000 different ground acceleration-time histories relevant
to 100 US earthquakes with PGA between 25 and 1600 gal, magnitude between 5.5 and 7.5
and epicentral distance between 0.1 and 180 km. The formula proposed by Lin and Chang
(2003) can be written as:

α · T 0.30
η =1− (7)
(T + 1)0.65
in which:

α = 1.303 + 0.436 · ln(ξ ) (8)

All the aforesaid reduction factors depend, in some way, on the period of vibration of the
structure, in addition to its damping ratio.
In Europe, studies conducted by different researchers led to similar equations, all of them
being independent from the period of vibration.
The damping reduction factor derived by Bommer et al. (2000) is expressed by the follo-
wing formula:

10
η= (9)
5+ξ

It has been adopted in the European seismic code (EC8 EN 2004), replacing the equation
adopted in the pre-norm version of the same code (Eurocode 8 1994):

7
η= (10)
2+ξ

A variant of the previous expressions has been proposed by Tolis and Faccioli (1999),
based on the examination of displacement response spectra derived from the seismic records
of the 1995 Kobe earthquake:

15
η= (11)
10 + ξ

In the recent Italian seismic codes (OPCM 2003, DM 2008), the same formulation as in
the EC8 (EN-2004) has been adopted, while a different equation was recommended in the

123
Bull Earthquake Eng (2009) 7:273–291 277

previous Italian guidelines for the design, construction and testing of seismically isolated
structures (SSN 1998):

7
η= 3 (12)
2+ξ

Recently, Priestley suggested a modified version of the EC8 formula, to account for the
effects of near-source earthquakes (Priestley 2003):

10
η= 4 (13)
5+ξ

It is worth also mentioning the expression proposed by Kawashima and Aizawa (1986):
 
1.5
η= + 0.5 (14)
0.4 · ξ + 1
subsequently included in the Standard Specifications of the California Dept. of transportation
for the seismic design of bridges (Caltrans 2006).
Among the other seismic codes, the Chinese and Japanese guidelines for the design of
seismically isolated structures are worth mentioning, as they provide reduction factors consi-
derably lower than in all the other formulations. The expressions suggested by the Chinese
(Zhou et al. 2003) and Japanese (JPN 2001) guidelines are, respectively:
0.05 − ξ
η = 1+ (15)
0.06 + 1.4 · ξ
1.5
η= (16)
1 + 10 · ξ
Figure 1a shows the trend of the most important period-independent reduction factors as
a function of the damping ratio. According to the figure, the lowest values of η are those
provided by the Japanese seismic code, while the largest, i.e., most conservative, ones are
those obtained from the Priestley equation. In any case, the significant differences among the
available equations emphasize the importance of the problem.

(a) (b)
1.0 1.0

0.8 0.8
Reduction factor

Reduction factor

0.6 Priestley 0.6 Bommer

Ashour Wu-Hanson
0.4 Kawashima-Aizawa 0.4 Lin-Chang
Tolis-Faccioli
Newmark-Hall
Japanese seismic code
0.2 0.2
Bommer
Ramirez
0.0 0.0
5 10 15 20 25 30 0 1 2 3 4
Damping ratio (%) Period (sec)
Fig. 1 a Trend of various period-independent damping reduction factors as a function of the equivalent viscous
damping ratio of the structure and b trend of some period-dependent damping reduction factors as a function
of the period of vibration of the structure for a damping ratio equal to 20%

123
278 Bull Earthquake Eng (2009) 7:273–291

Figure 1b shows the sensitivity of the period-dependent reduction factors to changes in

the period of vibration of the structure, when the damping ratio is taken equal to 20%. For
comparison, the period-independent Bommer’s formula is also reported. As can be seen, the
variability of the reduction factors with the period of vibration is rather limited for all the
expressions under consideration. All the reduction factors tend to unity when the period of
vibration tends to zero.

3 Selected ground acceleration-time histories

A number of seismic ground motions of natural earthquakes have been selected, having the
following characteristics: (i) recorded at free field stations or at the ground floor of low-
rise buildings with negligible soil–structure interaction effects, (ii) magnitude between 6
and 8, (iii) epicentral distance between 0.1 and 100 km, and (iv) PGA greater than 0.1 g.
They were selected from the European, Californian and Japanese earthquake strong-motion
databases, that can be accessed on-line at the following Internet sites: www.isesd.cv.ic.ac.uk,
www.quake.ca.gov, ww.peer.berkeley.edu and www.kik.bosai.go.jp. A total of 60 seismic
records (i.e., 120 horizontal acceleration-time histories) having the required characteristics
have been selected. They are listed in Table 1, together with the associated parameters. The
accelerograms have been primarily divided in 4 groups, according to their magnitude and
epicentral distance (see Table 2).

4 Results

4.1 Use of damping reduction factors for single accelerograms

The first step of the analysis was to derive the real response spectra associated to each
accelerogram (see Table 1), for damping ratios equal to 5, 10, 20, and 30%, respectively.
This was done by solving the equation of motion of SDOF systems with period of vibration
ranging from 0.01 to 4 s, with 0.01 s step. For each seismic component, response spectra
in terms of displacements, pseudo-accelerations and accelerations have been drawn. They
have been compared to the corresponding approximate spectra derived from the 5%-damping
spectra, by applying the reduction factor proposed by Bommer et al. (see Eq. 9).
Figure 2 shows the results relevant to the horizontal records of the 1994 Northridge
earthquake (Rinaldi station) and 1999 Duzce earthquake (Meteoroloji Mudurlugu station).
The thick lines correspond to the real displacement response spectra at the damping ratios
(>5%) considered. The thin lines represent the corresponding approximate displacement
spectra.
The discrepancies between exact and approximate spectra are mainly due to the smoothing
of the spectra when increasing the level of viscous damping: peaks of the spectral curves are
reduced more than dips. The approximate spectra, on the contrary, maintains the shape of the
5%-damping spectra also for high-damping. Similar results, not shown in this paper, have
been obtained for all single seismic records.
The outcomes of this first group of analyses clearly prove that the use of damping reduc-
tion factors for the evaluation of high-damping response spectra of single seismic records
can be suitable only for low levels of viscous damping. For relatively high-damping values
(e.g., greater than 10%), it can lead to considerable overestimates or underestimates of the
maximum seismic response, in certain period intervals characterised by peaks or dips.

123
Bull Earthquake Eng (2009) 7:273–291 279

Table 1 Earthquakes considered in this study

EQ Year Earthquake name Station Epicentral Magnitudo Soil PGAmax

distance (km) (m/s2 )

1 1987 Superstition Hills (CA) Imp. Co. Center 36 6.7 Soft soil 3.34
2 1988 Spitak (Armenia) Gukasian 36 6.7 Soft Soil 1.79
3 1989 Loma Prieta Corralitos 7 7.0 Soft soil 6.40
(California)
4 1989 Loma Prieta Saratoga 27 7.0 Alluvium 5.30
(California)
5 1990 Mantjil (Iran) QAZVIN 94 7.4 Stiff Soil 1.80
6 1990 Manjil (Iran) Abbar 75 7.4 Unknown 2.04
7 1990 Chamoli (India) Gopeshwar 18 6.6 Stiff soil 3.52
8 1990 Chamoli (India) Ukhimath 35 6.6 Unknown 0.97
9 1992 Erzincan (Turkey) Meteorologij 13 6.6 Stiff Soil 5.02
Mudurlugu
10 1992 Big Bear (California) Big Bear Lake 11 6.5 Alluvium 5.60
11 1992 Big Bear (California) Desert Hot Spring 40 6.5 Alluvium 2.21
12 1992 Cape Mendocino (CA) Cape Mendocino 15 7.1 Hard Rock 14.6
13 1992 Cape Mendocino (CA) Petrolia GS 16 7.1 Soft soil 6.49
14 1992 Landers (California) Yermo-FS 76 7.4 Alluvium 2.40
15 1992 Landers (California) Joshua Tree FS 14 7.3 Unknown 2.85
16 1992 Petrolia (California) Shelter Cove 41 7.1 Alluvium 2.39
17 1992 Petrolia (California) Petrolia 8 7.1 Alluvium 6.85
18 1994 Bitola (Macedonia) Florina 39 6.1 Rock 0.81
19 1994 Northridge (California) Rinaldi 10 6.7 Alluvium 8.21
20 1994 Northridge (California) Newhall County 20 6.7 Alluvium 6.30
21 1994 Northridge (California) LACC-S 20 6.7 Alluvium 3.50
22 1994 Northridge (California) UCLA 18 6.7 Alluvium 6.60
23 1994 Northridge (California) Castaic 41 6.7 Rock 5.89
24 1994 Kozani (Greece) Kozani Prefecture 17 6.5 Rock 2.03
25 1994 Aigion (Greece) Patra 43 6.5 Stif soil 0.92
26 1995 Dinar (Turkey) Meteoroloji 8 6.4 Stiff Soil 3.13
Mudurlugu
27 1995 Kobe (Japan) Takatori 2 6.9 Unknown 6.10
28 1995 Kobe (Japan) Tadoka 32 6.9 Unknown 2.91
29 1996 Guerrero (Mexico) Papanoa 37 6.8 Rock 3.20
30 1997 Strofades (Greece) Zakynthos 32 6.6 Stiff Soil 1.28
31 1997 Kalamata (Greece) Koroni 48 6.4 Rock 1.18
32 1998 Adana (Turkey) Ceyhan-Tarim Ilce 30 6.3 Stiff Soil 2.64
Mudurlugu
33 1999 Duzce (Turkey) LDEO C0375 23 7.2 Unknown 9.01
34 1999 Duzce (Turkey) Bolu-Bayindirlik 39 7.2 Unknown 7.85
35 1999 Duzce (Turkey) Meteoroloji 9 7.2 Unknown 5.03
Mudurlugu

123
280 Bull Earthquake Eng (2009) 7:273–291

Table 1 continued
EQ Year Earthquake name Station Epicentral Magnitudo Soil PGAmax
distance (km) (m/s2 )

36 1999 Izmit-Kocaeli (Turkey) Izmit-Karayollari 39 7.6 Soft soil 1.26

37 1999 Izmit-Kocaeli (Turkey) Yarimka-Petkim 20 7.6 Soft soil 2.90
38 1999 Izmit-Kocaeli (Turkey) Gebze Arcelik 55 7.6 Stiff Soil 2.07
39 1999 Izmit-Kocaeli (Turkey) Gebze-Tubitak 48 7.6 Stiff Soil 2.33
40 1999 Izmit-Kocaeli (Turkey) Sakarya 34 7.6 Stiff Soil 3.54
41 1999 Chi-Chi (Taiwan) Taichung TCU068 12 7.3 Unknown 4.94
42 1999 Chi-Chi (Taiwan) Chiayi CHY024 10 7.1 Rock 2.72
43 1999 Hector Mine Joshua Tree FS 52 7.1 Alluvium 1.89
(California)
44 2000 Western Tottori Hakuta SMNH01 16 7.3 Unknown 7.20
(Japan)
45 2000 Western Tottori Hino TTRH02 12 7.3 Unknown 9.20
(Japan)
46 2000 Michoacan (Mexico) Caleta De Campos 18 6.5 Unknown 1.51
47 2003 San Simeon Parkfield 68 6.5 Alluvium 2.21
(California)
48 2003 Miyagi-Oki (Japan) Karakuwa MYGH03 13 7.0 Unknown 8.08
49 2003 Miyagi-Oki (Japan) Fujisawa IWTH05 29 7.0 Unknown 5.69
50 2003 Tokachi-Oki (Japan) Hiroo HKD100 69 8.0 Unknown 9.72
51 2003 Tokachi-Oki (Japan) Taiki HKD098 88 8.0 Unknown 3.66
52 2004 Niigata-ken Chuetsu Koide NIG020 19 6.5 Unknown 5.24
(Japan)
53 2004 Niigata-ken Chuetsu Kamo-NIG06 44 6.5 Unknown 4.08
(Japan)
54 2004 Niigata-ken Chuetsu Tohkamachi NIG021 20 6.5 Unknown 8.13
(Japan)
55 2004 Parkfield (California) Vineyard Canyon 13 6.0 Alluvium 5.50
56 2004 Parkfield (California) Coalinga 32 6.0 Alluvium 3.42
57 2004 Hokkaido (Japan) Hamanaka HKD075 48 7.0 Unknown 2.57
58 2004 Hokkaido (Japan) Nosappu HDK074 83 7.0 Unknown 5.50
59 2004 Hokkaido (Japan) Tsurui KSRH06 73 7.1 Unknown 7.56
60 2006 Kiholo Bay (Hawaii) Kealakekua 35 6.7 Unknown 5.30

Table 2 Classification of accelerograms based on magnitude and epicentral distance

Group n. Epicentral distance (km) Magnitude Record count

1 0–30 6–7 30
2 0–30 7–8 30
3 >30 6–7 30
4 >30 7–8 30

123
Bull Earthquake Eng (2009) 7:273–291 281

(a) (b)
0.3 0.4
ξ = 10% ξ = 10%
EW
NS 0.3
0.2 EW
Sd (m)

NS

Sd (m)
0.2
0.1
0.1

0 0
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)
0.4
0.3
ξ = 20% ξ = 20%
0.3
0.2 EW EW

Sd (m)
Sd (m)

NS
0.2 NS
0.1
0.1

0 0
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)
0.3 0.4
ξ = 30% ξ = 30%
0.3
0.2
Sd (m)
Sd (m)

EW
NS EW 0.2
NS
0.1
0.1

0 0
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)

Fig. 2 Comparison between real (thick lines) and approximate (thin lines) displacement spectra relevant to
the horizontal components of the a Northridge and b Duzce earthquakes, for damping ratios equal to 10, 20
and 30%, respectively. The approximate spectra have been derived by applying the damping reduction factor
derived by Bommer et al. (ref. to Eq. 9)

4.2 Use of damping reduction factors for groups of earthquakes

The damping reduction factors are commonly applied in the seismic codes to smoothed
spectral shapes, obtained from probabilistic estimates of the contribution to seismic risk
of a large number of earthquakes. As a consequence, in the second phase of the analysis
reference has been made to smoothed response spectra, derived from the normalised response
spectra of the single accelerograms previously selected (see Tables 1 and 2). To this end, two
different normalisation criteria have been followed: (i) same PGA, for acceleration response
spectra Sa (T) and (ii) same area under each spectral curve, within a given period range, for
displacement response spectra Sd (T):
Sai (T = 0) Sa j (T = 0)
= =1 (17)
PGAi PGA j

123
282 Bull Earthquake Eng (2009) 7:273–291

 T̄  T̄
Sdi (T )dT = Sd j (T )dT (18)
0 0

The upper limit of the interval of integration (T̄ ) has been alternatively taken equal to 2, 3,
and 4 s. In this paper, however, the results related to T̄ = 4 sec are only considered.
The mean value (µ), standard deviation (σ ) and maximum value of the spectral ordinates
(acceleration and displacement) associated to each period of vibration of the normalized
response spectra have been computed and used to derive five different smoothed response
spectra, for each group of earthquakes, namely: average, envelope, µ+σ , µ+2σ and µ+3σ
spectra. Assuming that the available data follow a normal distribution, the µ + σ , µ + 2σ and
µ + 3σ response spectra provide the values of maximum displacement with a probability of
15.8, 2.2, and 0.15%, respectively, of being exceeded.
For each group of earthquakes, the same procedure used for single accelerograms (see
Sect. 4.1) has been followed. First, the real high-damping response spectra associated to
each record have been normalized, according to the two aforesaid criteria. Subsequently,
the corresponding smoothed spectra have been determined. Finally, the 10-, 20-, and 30%-
damping approximate spectra have been derived from the real 5% spectra, by applying the
corresponding damping reduction factors.
In Fig. 3, real and approximate 30%-damped average spectra are compared, for each
group of earthquakes (see Table 2). The approximate spectra have been obtained using three
different damping reduction factor formulations, namely: (i) that proposed by Bommer et al.,
then adopted in the EC8 EN-2004 and in the new Italian seismic code (RF1, see Eq. 9),
(ii) that used in the Japanese guidelines for seismically isolated structures (RF2, see Eq. 16)
and (iii) that derived by Tolis and Faccioli (RF3, see Eq. 11), which represents one of the
most conservative methods (see Fig. 1). As can be seen, the smoothed shape of the average
spectra significantly improves the accordance between real and approximate high-damping
response spectra. However, the accuracy of the results depends strongly on the formulation
of the damping reduction factor. The Japanese guidelines equation leads to considerable
underestimates of the exact maximum elastic displacement, in the whole period range and
for all the groups of earthquakes considered in this study. The Bommer’s formula appears to
be reasonably conservative, except in the long period range of response spectra associated to
lower magnitude earthquakes (groups 1 and 3 of Fig. 3), where alternative reduction factors
(e.g., that suggested by Tolis and Faccioli) may be preferred.
The aforesaid observations are confirmed by Fig. 4, which shows the ratios of the
approximate to exact average spectra of Fig. 3. Spectral ratios smaller than 1 indicate that
the approximate response spectrum underestimates the real maximum elastic displacements.
The curves are cut off at 0.15 s, to take into account that seismic codes generally assume a
linear variation of the damping reduction factor from the current value at T = 0.1–0.2 s to 1
at T = 0 s.
As can be noted from Fig. 4, the approximate maximum elastic displacements predicted by
the Bommer’s method rarely differ more than 25% from the exact values, for all the groups
of earthquakes, levels of damping ratios and values of period of vibrations considered in
this study. More precisely, the absolute errors between approximate and exact 30%-damped
response spectra result, on average, between 15 and 17% (groups 2–3) and 29% (group 4) in
the short period range (T < 0.5 s), between 7 and 8% (groups 1–2) and 20% (group 4) in the
intermediate period range (0.5 s < T < 2 s) and, finally, between 8–9% (group 2–3) and 20%
(group 4), in the long period range (2 s < T < 4 s).
Similar considerations can be done for the other smoothed response spectra (i.e., envelope,
µ + σ , µ + 2σ , and µ + 3σ response spectra). Fig. 5, for instance, shows the 30%-damped

123
Bull Earthquake Eng (2009) 7:273–291 283

0.30 0.30
RF1
0.25 GROUP 1 0.25 GROUP 2 RF3
RF3
0.20 0.20

Sd (m)
Sd (m)

RF1 EX
0.15 0.15
EX
0.10 0.10 RF2
RF2
0.05 0.05
0.00 0.00
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)

0.30 0.30

0.25 GROUP 3 0.25 GROUP 4

0.20 0.20

Sd (m)
RF1
Sd (m)

0.15 0.15 RF3

RF3 EX
0.10 0.10
0.05 0.05 EX
RF1 RF2
RF2
0.00 0.00
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)
Fig. 3 Comparison of exact (EX) and approximate (RFi) 30%-damped average response spectra for groups
of earthquakes differing in magnitude and epicentral distance (see Table2). Approximate response spectra
derived using the Reduction Factor of Bommer et al. (RF1, see Eq. 9), that of the Japanese guidelines for
seismically isolated structures (RF2, see Eq. 16) and that of Tolis and Faccioli (RF3, see Eq. 11)

2.0 2.0
1.8 RF3 GROUP 1 1.8 GROUP 2
RF3
Spectral Ratio
Spectral Ratio

1.6 RF1 1.6 RF1

1.4 1.4
1.2 1.2
1.0 1.0
0.8 0.8
0.6 0.6
0.4 RF2 0.4 RF2
0.2 0.2
0.0 0.0
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)

2.0 2.0
1.8 GROUP 3 1.8 RF3 GROUP 4
RF3 RF1
Spectral Ratio
Spectral Ratio

1.6 1.6
1.4 RF1 1.4
1.2 1.2
1.0 1.0
0.8 0.8
0.6 0.6
0.4 0.4 RF2
RF2
0.2 0.2
0.0 0.0
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)
Fig. 4 Spectral ratios (approximate to exact) associated to the 30%-damped average response spectra of the
four groups of earthquakes considered in the study (see Fig. 3)

µ + 2σ spectra associated to the four groups of earthquakes considered in this study. The
damping reduction factor of Bommer tends to become more and more conservative while
enlarging the confidence interval of the spectrum, i.e., while increasing the multiplier of the
standard deviation σ . At the same time, the underestimates determined by the use of the
damping reduction factor proposed by the Japanese guidelines tend to reduce.

123
284 Bull Earthquake Eng (2009) 7:273–291

0.40 0.40
0.35 GROUP 1 RF3 0.35 GROUP 2 RF1
0.30 RF1 0.30 RF3

Sd (m)
0.25 0.25
Sd (m)

0.20 0.20
0.15 0.15 RF2
0.10 EX 0.10 EX
0.05 RF2 0.05
0.00 0.00
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)

0.40 0.40
0.35 GROUP 3 0.35 GROUP 4
0.30 0.30

Sd (m)
0.25 RF1
Sd (m)

0.25 RF3
0.20 RF3 RF1 0.20
0.15 0.15
0.10 0.10
RF2 EX
0.05 0.05
EX RF2
0.00 0.00
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)

Fig. 5 Comparison of exact (EX) and approximate (RFi) 30%-damped µ + 2σ response spectra for groups
of earthquakes differing in magnitude and epicentral distance (see Table 2). Approximate response spectra
derived using the Reduction Factor of Bommer et al. (RF1, see Eq. 9), that of the Japanese guidelines for
seismically isolated structures (RF2, see Eq. 16) and that of Tolis and Faccioli (RF3, see Eq. 11)

4.3 Comparison between different damping reduction factors

In order to examine in detail the level of accuracy of several different formulations, two
statistical indexes have been computed, referring to a single group of earthquakes, including
the full sample of accelerograms selected (see Table 1), regardless their magnitude and
epicentral distance. The considered statistical indexes are the mean spectral ratio MSR(T, ξ )
and the standard error SE(T, ξ ) between approximate and exact high-damping displacement
response spectra. They are defined by the following expressions:
n

1  η · Sd,i (T, ξ = 5%)

MSR(T, ξ ) = · (19)
n Sd,i (T, ξ )
i=1


 n  
 1  η · Sd,i (T, ξ = 5%) − Sd,i (T, ξ ) 2
SE(T, ξ ) = · (20)
n−1 Sd,i (T, ξ )
i=1

in which Sd,i (T, ξ ) is the real maximum displacement, obtained through linear time-history
analysis of SDOF with period of vibration T and damping ratio ξ , due to the i-th seismic
ground motion, η = η (T, ξ ) the damping reduction factor under consideration and n the
total number of ground acceleration-time histories taken into account. In the evaluation of
both MSR and SE a constant period increment of 0.01 s has been considered.
Values of MSR smaller than 1.0 indicate that the approximate method underestimates,
on average, the exact maximum elastic displacement, for that period and damping ratio.
The standard error SE measures the scattering of the approximate maximum displacements
around their exact values. Values of SE close to zero imply a good accuracy of the approximate
method in the prediction of the real maximum displacements.

123
Bull Earthquake Eng (2009) 7:273–291 285

The damping reduction factors examined are those proposed in: (i) Bommer et al. (2000),
see Eq. 9; (ii) Newmark and Hall (1973), see Eq. 3; (iii) Ramirez et al. (2002); (iv) the
Japanese guidelines for seismically isolated structures (JPN 2001), see Eq. 16; (v) Tolis and
Faccioli (1999), see Eq. 11; (vi) Wu and Hanson (1989), see Eq. 5 and (vii) Lin and Chang
(2003), see Eq. 7.
Figures 6–12 show the (a) MSR and (b) SE indexes, related to the seven aforesaid damping
reduction factors, as a function of the period of vibration, varied from 0.15 to 4 s, and for
three different damping ratios, equal to 10, 20, and 30%, respectively. As expected, the errors
increase with the increase of viscous damping.

(a) (b)
1.5 60%
1.4 ξ = 10%
ξ = 10%
Mean spectral ratio

1.3 ξ = 20% 50%

ξ = 20%
1.2 ξ = 30%
40% ξ = 30%
1.1
1.0 30%
0.9
0.8 20%
0.7
10%
0.6
0.5 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 (sec) Period (sec)
Fig. 6 a Mean spectral ratio and b Standard error from the approximate spectra derived using the damping
reduction factor of Bommer et al.

(a) (b)
1.5 60%
1.4
Mean spectral ratio

Standard error (%)

1.3 50%
1.2 40%
1.1
1.0 30%
0.9
0.8 ξ = 10% 20%
0.7 ξ = 20% 10%
0.6 ξ = 30%
0.5 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 (sec) Period (sec)
Fig. 7 a Mean spectral ratio and b Standard error from the approximate spectra derived using the damping
reduction factor of Newmark and Hall

(a) (b)
1.5 60%
1.4 ξ = 10%
Standard error (%)
Mean spectral ratio

1.3 50% ξ = 20%

1.2 ξ = 30%
40%
1.1
1.0 30%
0.9
0.8 ξ = 10% 20%
0.7 ξ = 20% 10%
0.6 ξ = 30%
0.5 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 (sec) Period (sec)

Fig. 8 a Mean spectral ratio and b Standard error from the approximate spectra derived using the damping
reduction factor of Ramirez

123
286 Bull Earthquake Eng (2009) 7:273–291

(a) (b)
1.5 60%
ξ = 10% ξ = 10%

Standard error (%)

1.4
Mean spectral ratio

1.3 ξ = 20% 50% ξ = 20%

1.2 ξ = 30% 40% ξ = 30%
1.1
1.0 30%
0.9
0.8 20%
0.7 10%
0.6
0.5 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 (sec) Period (sec)

Fig. 9 a Mean spectral ratio and b Standard error from the approximate spectra derived using the damping
reduction factor adopted in the Japanese guidelines

(a) (b)
1.5 60%
1.4
Mean spectral ratio

Standard error (%) 50%

1.3
1.2 40%
1.1
1.0 30%
0.9
0.8 ξ = 10% 20%
0.7 ξ = 20%
10%
0.6 ξ = 30%
0.5 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 (sec) Period (sec)
Fig. 10 a Mean spectral ratio and b Standard error from the approximate spectra derived using the damping
reduction factor of Tolis and Faccioli

(a) (b)
1.5 60%
1.4 ξ = 10%
ξ = 10%
Mean spectral ratio

Standard error (%)

1.3 ξ = 20% 50%

ξ = 20%
1.2 ξ = 30% 40%
1.1 ξ = 30%
1.0 30%
0.9
0.8 20%
0.7
10%
0.6
0.5 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 (sec) Period (sec)
Fig. 11 a Mean spectral ratio and b Standard error from the approximate spectra derived using the damping
reduction factor of Wu and Hanson

Focusing the attention on applications to structures with passive control systems, e.g.,
buildings equipped with Energy Dissipation Systems (EDS) or Seismic Isolation Systems
(SIS), the period ranges of greater interest are 0.2–0.8 s for buildings with EDS and 2–4 s for
buildings with SIS, respectively. The examination of the diagrams of Figs. 6–12 points out
remarkable differences in the accuracy of the damping reduction factors in the two period
ranges considered.
The reduction factor proposed by Bommer et al. (see Fig. 6), for instance, provides esti-
mates of the real maximum elastic displacements fairly accurate in both the examined period
ranges. In the short period range, in particular, the MSR index at 30% viscous damping result,

123
Bull Earthquake Eng (2009) 7:273–291 287

(a) (b)
1.5 60%
1.4 ξ = 10%
Mean spectral ratio

ξ = 10%

Standard error (%)

1.3 ξ = 20% 50%
1.2 ξ = 20%
ξ = 30% 40%
1.1 ξ = 30%
1.0 30%
0.9
0.8 20%
0.7
10%
0.6
0.5 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 (sec) Period (sec)
Fig. 12 a Mean spectral ratio and b Standard error from the approximate spectra derived using the damping
reduction factor of Lin and Chang

on average, of the order of 1.1 and the SE index lower than 30%, with a few exceptions. In
the long period, the MSR index reduces progressively from approximately 1.1 to 0.9, while
increasing the period from 2 to 4 s. The SE index generally do not exceed 30%.
The methods proposed by Ramirez (see Fig. 8) and Tolis-Faccioli (see Fig. 10) lead to
significant overestimates of the real maximum response in both the examined period ranges.
In the short period range, indeed, the MSR index for 30% damping reaches as high values as
1.3 (see Fig. 10a), being greater than 1.2 on average. In the long period range, it decreases
almost linearly, passing from about 1.2 to 1.05 while the period rises from 2 to 4 s. In both the
period ranges, the SE index varies between 30 and 40%, with peaks of 45% (see Fig. 10b).
The reduction factor of the Japanese guidelines (see Fig. 9) leads to significant underesti-
mates in both the examined period ranges. In the short period range, in particular, the MSR
index at 30% viscous damping varies around 0.75 and the SE index around 30%. In the long
period range, the MSR index reduces progressively to about 0.65 and the SE index increases
up to 40%.
The Newmark-Hall method (see Fig. 7) appears to be suitable for damping ratios of the
order of 10–20% (except some underestimates for periods lower than 0.3 s), while it results
excessively conservative at 30% viscous damping. Actually, this could be expected, as the
original studies from Newmark and Hall were limited to viscous damping ratios lower than
20%.
The Wu-Hanson method (see Fig. 11) turns out to be the most accurate one, among the
seven methods examined in this paper, in the long period range (2–4 s). As a matter of fact,
indeed, the MSR index always remains around 1, regardless the level of viscous damping
considered, and the SE index rarely exceeds 30%. In the short period range (0.2–0.8 s),
instead, the Wu-Hanson method can bring about considerable errors (MSR as high as 1.3
and SE greater than 40% at 30% viscous damping).
The Lin-Chang method (see Fig. 12), finally, results reliable and reasonably conservative
in both the examined period ranges, as testified by the MSR index, varying between 1 and
1.15, and by the SE index, which generally do not exceed 30%.

4.4 Effects of viscous damping on acceleration response

Current seismic codes (e.g., UBC, EC8, etc.) follow a force-based approach in the design of
earthquake resistant structures. Design seismic forces are derived from an elastic 5%-damped
pseudo-acceleration response spectrum (PSa ), reduced by a behavior modification factor (q
in EC8) and a damping reduction factor (η in EC8).

123
288 Bull Earthquake Eng (2009) 7:273–291

As known (Chopra 1995), PSa exactly equals Sa only if ξ is zero. As a matter of fact,
m*PSa corresponds to the peak value of the elastic resisting force while m*Sa is equal to the
peak value of the sum of elastic and damping force (Chopra 1995). As a consequence, pseudo-
accelerations (PSa ) are always lower than accelerations (Sa ). The differences between Sa
and PSa increase while increasing the level of damping and the natural period of vibration
of the structure, as clearly confirmed by Fig. 13, which shows the ratio between pseudo-
acceleration and acceleration spectra for the EW component of the Duzce earthquake (EQ
n. 35 of Table 1) and the EW component of the Northridge earthquake (EQ n. 19 of Table 1)
for three levels of damping (10, 20, and 30%, respectively).
The use of design seismic forces derived from high-damping pseudo-acceleration spectra
may lead to significant underestimates of the maximum seismic effects of the structure,
especially when the structural damping corresponds to an equivalent viscous damping ratio
coming from the hysteretic behaviour of the structure.
In order to investigate the aforesaid aspect, the mean acceleration and pseudo-acceleration
response spectra relevant to the selected ground motions (see Table 1) have been computed.
They are compared in Fig. 14a for four different damping ratios, equal to 5, 10, 20, and 30%,
respectively.

(a) (b)
1.2 1.2

1.0 1.0
10%
0.8 10% 0.8
PSa/Sa

20%
PSa/Sa

0.6 20% 0.6

30%
30%
0.4 0.4

0.2 0.2

0.0 0.0
0.0 1.0 2.0 3.0 4.0 0.0 1.0 2.0 3.0 4.0

Period (sec) Period (sec)

Fig. 13 Ratio between pseudo-acceleration (PSa ) and acceleration (Sa ) response spectra of the a EW com-
ponent of the Duzce earthquake and b EW component of the Northridge earthquake for three different levels
of damping

(a) (b)
2.5 1.0
Damping reduction factor

5%
10%
Sa
2.0 0.8
PSa 20%
10%
Sa vs. PSa

1.5 0.6
20% 30%
1.0 0.4
Sa
0.5 0.2 PSa
30%
EC8
0.0 0.0
0 1 2 3 4 0 1 2 3 4
Period (sec) Period (sec)

Fig. 14 a Comparison of mean acceleration (Sa ) and pseudo-acceleration (PSa ) response spectra at different
damping ratios. b Corresponding damping reduction factors compared to the approximate factor adopted in
the EC8 (EN-2004)

123
Bull Earthquake Eng (2009) 7:273–291 289

Applying the definition given in Eqs. 1–2, exact damping reduction factors have been
obtained from both Sa and PSa . They are shown in Fig. 14b and compared to the approximate
relationship adopted in the EC8 EN-2004. As can be seen, the damping reduction factors
derived from Sa are substantially different from those derived from PSa . Those derived from
PSa , indeed, appears to be substantially independent from the structural period (beyond
the dominant period of 0.3 s) and in fairly good accordance with those adopted in the EC8
EN-2004. The damping reduction factors derived from Sa , on the contrary, increase strongly
as the period of vibration of the structure increases. Moreover, for SDOF systems with periods
lower than about 3.5 s, both reduction factors reduce when damping increases. Beyond 3.5 s,
on the contrary, the reduction factor derived from Sa increases while increasing damping.
The aforesaid observations may suggest the use of different damping reduction factors for
displacements and pseudo-accelerations (hence design forces), when the equivalent viscous
damping of the structure is mainly related to its hysteretic behaviour (e.g., due to formation of
plastic hinges in the primary structural members, presence of hysteretic energy dissipation or
isolation systems, etc.). In this case, the damping reduction factors for pseudo-accelerations
should be greater than those for displacements and strongly dependent on the period. Further
studies are needed to fully investigate this subject.

5 Conclusions

The accuracy of different damping reduction factor equations for the evaluation of approxi-
mate high-damping elastic response spectra has been examined. The reduction factors
taken into account are: (i) the Newmark and Hall formulation, adopted in the ATC-40 and
FEMA-356 for the evaluation of existing structures, (ii) the Bommer equation, adopted in
the Eurocode 8 EN-2004, (iii) the Ramirez equation, adopted in the FEMA-368 for the de-
sign of structures with passive control systems, (iv) the relationship used in the Japanese
guidelines for the design of seismically isolated structures, the formulas recently proposed
by (v) Wu and Hanson, (vi) Tolis and Faccioli and (vii) Lin and Chang. All the considered
damping reduction factors have been derived by studying the effects of viscous damping on
the maximum displacement response of elastic SDOF systems.
The examination of the aforesaid damping reduction factors has been carried out based on
approximately 120 natural seismic records relevant to earthquakes with magnitude between
6 and 8, epicentral distance ranging from 1 km to 100 km and PGA greater than 0.1 g. Exact
and approximate displacement elastic response spectra, for three different damping ratios
(10, 20, and 30, precisely), have been compared. The comparison has been performed for
both single spectra and combinations of normalized spectra.
Based on the results of this study, the following conclusions can be drawn.
The use of damping reduction factors for the evaluation of high-damping response spectra
of single seismic records can be accepted only for damping ratio not greater than 10%,
otherwise it can lead to considerable errors in the estimate of the maximum seismic response.
For response spectra representing groups of earthquakes, the accuracy of the estimate
strongly depends on the damping reduction factor used and on the period range considered.
Based on the results of this study, the most accurate damping reduction factors turn out
to be those proposed by Wu-Hanson and Lin-Chang, which depend on both damping ratio
and period of vibration. The reduction factor of Wu-Hanson leads to excellent predictions
in the long period range (i.e., between 2 and 4 s), with mean spectral ratios (approximate
vs. exact) very close to 1 and standard errors always lower than 30%. On the contrary, it
results excessively conservative in the short-intermediate period range (i.e., between 0.2 and

123
290 Bull Earthquake Eng (2009) 7:273–291

0.8 s). The reduction factor of Lin-Chang provides the best estimates in short-intermediate
period range, with mean spectral ratios less than 1.15 and standard errors lower than 30%,
while it results slightly conservative in the long period range. On the other hand, the damping
reduction factor proposed by Bommer et al., which is adopted in the current version of the
European Seismic code (EC8 EN 2004), probably represents the best compromise between
accuracy of results and simplicity of formulation, as it depends on the damping ratio only.
As far as the acceleration response is concerned, results from this study indicate that two
different damping reduction factors should be used for displacement and pseudo-acceleration
response spectra, in order to get accurate design seismic forces, especially for structural
systems with high-damping ratios and long periods of vibration. The damping reduction
factor for pseudo-accelerations should be greater than those for displacements and dependent
on the period of vibration of the structure.

Acknowledgments This work has been carried out within the DPC-RELUIS 2005-2008 program, Research
Project No. 4.

References

Ashour SA (1987) Elastic seismic response of buildings with supplemental damping. Ph.D. Thesis, Department
of Civil Engineering, Michigan University
ATC-40 (1996) Seismic evaluation and retrofit of concrete buildings. Applied Technology Council, Redwood
City, CA
Bommer JJ, Mendis R (2005) Scaling of spectral displacement ordinates with damping ratios. Earthquake Eng
Struct Dynam 34:145–165. doi:10.1002/eqe.414
Bommer JJ, Elnashai AS, Weir AG (2000) Compatible acceleration and displacement spectra for seismic
design codes. In: Proceedings of the 12th world conference on earthquake engineering, Auckland
Caltrans (2006) California Department of Transportation, Seismic Design Criteria, ver. 1.4., Sacramento,
California
Chopra AK (1995) Dynamics of structures. Prentice Hall International Inc, London
Constantinou MC, Soong TT, Dargush GF (1998) Passive energy dissipation systems for structural design
and retrofit. MCEER, State University of New York at Buffalo
DM Infrastrutture 14.01.2008 (2008) Approval of the new technical standards for constructions (in Italian).
GU n. 29 4-2-2008- Suppl. Ordinario n. 30, Roma
Eurocode 8 (1994). Design provisions for earthquake resistance of structures. Part 1: General rules—Seismic
actions and general requirements for structures. European pre-norm ENV 1998-1, CEN, Brussels
Eurocode 8 (2004) Design of structures for earthquake resistance, Part 1: General rules, seismic actions and
rules for buildings. EN 2004-1-1, CEN, Brussels
FEMA-273 (1997) NEHRP guidelines for the seismic rehabilitation of buildings. Fed. Emergency Management
Agency, Washington DC
FEMA-356 (2000) NEHRP prestandard and commentary for the seismic rehabilitation of buildings. Federal
Emergency Management Agency, Washington D.C
Gulkan P, Sozen M (1974) Inelastic response of reinforced concrete structures to earthquake ground motions.
ACI J 71: 604–610
IBC (2000) International Building Code. International Conference of Building Officials, Whittier, CA
JPN (2001) Ministry of vehicle, infrastructure and transport. Guidelines for calculation procedure and technical
standard on seismically isolated structures (in Japanese). Building Center of Japan
Kawashima K, Aizawa K (1986) Modification of earthquake response spectra with respect to damping ratio.
Proc. 3rd US National Conference on Earthquake Engineering, Charleston, South Caroline, vol. II
Lin YY (2007) Statistical study on damping modification factors adopted in Taiwan’s seismic isolation design
code by using the 21 September 1999 Chi-Chi earthquake, Taiwan. Eng Struct 29:682–693. doi:10.1016/
j.engstruct.2006.06.006
Lin Y, Chang KC (2003) A study on damping reduction factor for buildings under earthquake ground motion.
J Struct Eng (ASCE) 129(2):206–214
Lin YY, Miranda E, Chang KC (2005) Evaluation of damping reduction factors for estimating elastic response
of structures with high-damping. Earthquake Eng Struct Dynam 34(9):375–388

123
Bull Earthquake Eng (2009) 7:273–291 291

Naeim F, Kelly JM (1999) Design of seismic isolated structures. Wiley, Chichester, UK

Naeim F, Kircher CA (2001) On the damping adjustment factors for earthquake response spectra. Struct Des
Tall Build 10:361–369. doi:10.1002/tal.180
Newmark NM, Hall WJ (1973) Seismic design criteria for nuclear reactor facilities. Report No. 46, Building
Practices for Disaster Mitigation, National Bureau of Standards, US Department of Commerce
Newmark NM, Hall WJ (1982) Earthquake spectra and design EERI monograph series. Earth Eng Research
Inst, Oakland, CA
NEHRP (2000) Recomended provisions for seismic regulations for new buildings and other structures. Federal
Emergency Management Agency, Washington, DC
OPCM 3274 (2003) Ordinanza del Presidente del Consiglio dei Ministri n◦ 3274, Primi elementi in materia
di criteri generali per la classificazione sismica del territorio nazionale e di normative tecniche per le
costruzioni in zona sismica, Roma
Priestley MJN (2003) Myths and fallacies in earthquake engineering, revisited. IUSS Press, Pavia (Italy)
Ramirez OM, Constantinou MC, Whittaker AS, Kircher CA, Chrysostomou CZ (2002) Elastic and inelastic
seismic response of buildings with damping systems. Earthq Spectra 18(3):531–547 doi:10.1193/1.
1509762
Shibata A, Sozen M (1976) Substitute-structure method for seismic design in RC. J Struct Div 102:1–18
SSN (1998) Presidenza del Consiglio Superiore dei LL.PP.—Servizio Tecnico Centrale, Linee guida italiane
per la progettazione, esecuzione e collaudo di strutture isolate dal sisma, Roma
Tolis SV, Faccioli E (1999) Displacement design spectra. J Earthquake Eng 3:107–125 doi:10.1142/
S1363246999000065
UBC (1994) Uniform Building Code. International Conference of Building Officials, Whittier, CA
UBC (1997) Uniform Building Code. International Conference of Building Officials, Whittier, CA
Wu JP, Hanson RD (1989) Inelastic response spectra with high-damping. J Struct Div 115(6):1412–1431
Zhou F, Wenguang, L, Xu Z (2003) State of the art on applications, R&D and design rules for seismic isolation
in China. In: Proc. 8th world seminar on seismic isolation, energy dissipation and active vibration control
of structures, Yerevan, Armenia, 2003

123