PACIFIC EARTHQUAKE ENGINEERING

RESEARCH CENTER

Update of the Chiou and Youngs NGA Ground Motion

Model for Average Horizontal Component of Peak
Ground Motion and Response Spectra

Brian S.J. Chiou

California Department of Transportation
Sacramento, California

Robert R. Youngs
AMEC Environment and Infrastructure
Oakland, California

PEER 2013/07
MAY 2013
 Disclaimer

The opinions, findings, and conclusions or recommendations

expressed in this publication are those of the author(s) and
do not necessarily reflect the views of the study sponsor(s)
or the Pacific Earthquake Engineering Research Center.
Update of the Chiou and Youngs NGA Ground Motion
Model for Average Horizontal Component of Peak
Ground Motion and Response Spectra

Brian S.J. Chiou

California Department of Transportation
Sacramento, California

Robert R. Youngs
AMEC Environment and Infrastructure
Oakland, California

PEER Report 2013/07

Pacific Earthquake Engineering Research Center
Headquarters at the University of California, Berkeley
May 2013
ii
 ABSTRACT

This report presents an update to the Chiou and Youngs [2008a] NGA model for the prediction
of horizontal peak ground acceleration and 5%-damped pseudo spectral acceleration for
earthquakes in active tectonic regions such as California. The update is based on analysis of the
greatly expanded NGA-West2 strong motion database. The updated model contains minor
adjustments to the functional form developed by Chiou and Youngs [2008a] related to style of
faulting effect, hanging wall scaling, scaling with the depth to the top of rupture, the effect of
sediment depth, and the inclusion of two additional terms for the effects of fault dip and rupture
directivity.

iii
iv
 ACKNOWLEDGMENTS

This study was sponsored by the Pacific Earthquake Engineering Research Center (PEER) and
funded by the California Earthquake Authority, the California Department of Transportation, and
the Pacific Gas & Electric Company. Any opinions, findings, and conclusions or
recommendations expressed in this material are those of the authors and do not necessarily
reflect those of the sponsoring agencies.
We would like to thank all of the members of the PEER NGA-West2 project for many
helpful interactions. We would also like to thank Rob Graves and Ronnie Kamai for conducting
the ground-motion simulation that helped us in formulating our model. Some of the graphics in
this report were prepared using R (R Development Core Team 2013).

v
vi
 CONTENTS

ABSTRACT .................................................................................................................................. iii 

ACKNOWLEDGMENTS .............................................................................................................v 

TABLE OF CONTENTS ........................................................................................................... vii 

LIST OF FIGURES ..................................................................................................................... ix 

LIST OF TABLES ..................................................................................................................... xiii 

1  INTRODUCTION..............................................................................................................1 

2  GROUND MOTION DATA USED TO DEVELOP THE UPDATED

MODEL ..............................................................................................................................3 
2.1  Selection of Empirical Data...................................................................................3 
2.2  Use of Ground Motion Simulations ......................................................................6 
2.3  Supplement of Ground Motion Metadata ...........................................................7 
2.3.1  Sediment Thickness (Z1.0) ............................................................................7 
2.3.2  Depth to Top of Rupture (ZTOR) .................................................................10 

3  UPDATED GMPE ...........................................................................................................13 

3.1  Modification of Model Formulation ...................................................................13 
3.1.1  Style of Faulting Effect for Reverse and Normal Faulting ........................14 
3.1.2  Depth to the Top of Rupture ......................................................................15 
3.1.3  Z1.0 Scaling .................................................................................................16 
3.1.4  Hanging Wall Scaling ................................................................................19 
3.1.5  Effect of Rupture Dip Angle ......................................................................23 
3.1.6  Directivity Effect .......................................................................................23 
3.2  Refinement of Model Coefficients ......................................................................25 
3.2.1  Magnitude Scaling .....................................................................................25 
3.2.2  Distance Scaling.........................................................................................25 
3.2.3  VS30 Scaling ................................................................................................34 
3.3  Updated GMPE ....................................................................................................36 
3.3.1  Median Ground Motion .............................................................................36 
3.3.2  Aleatory Variability ...................................................................................38 

vii
4  EVALUATION OF UPDATED MODEL .....................................................................49 
4.1  Between-Event Residuals ....................................................................................49 
4.2  Within-Event Residuals .......................................................................................50 
4.3  Soil Nonlinearity...................................................................................................54 
4.4  Comparison with the 2008 Choiu and Youngs GMPE .....................................56 
4.4.1  Comparison of Median Motion..................................................................57 
4.4.2  Comparison of Aleatory Variability ..........................................................66 

5  EXAMPLE CALCULATIONS ......................................................................................69 

6  MODEL APPLICABILITY............................................................................................71 

REFERENCES.............................................................................................................................73 

APPENDIX A:  MODEL COEFFICIENTS FOR NON-CALIFORNIA REGIONS..........75 

viii
 LIST OF FIGURES

Figure 2.1 Magnitude-distance-region distribution of selected records. ...................................5

Figure 2.2 Number of usable records as a function of spectral period. .....................................6
Figure 2.3 Relationships between VS30 and Z1.0 for California. .................................................8
Figure 2.4 Relationship between VS30 and Z1.0 for Japan. .........................................................9
Figure 2.5 A scatter plot showing Z TOR and M of California earthquakes in the
NGA-West2 database. Solid lines are the regression models developed in
this study for reverse faulting (blue line) and combined strike-slip/normal
(red line) faulting. Dot-dashed lines are the models of Kaklamano et al.
[2011]. ....................................................................................................................11

Figure 3.1 Event terms computed from an interim GMPE demonstrating the
difference in style of faulting effect between small-to-moderate and large
magnitude earthquakes...........................................................................................14
Figure 3.2 Plots showing the variation of estimated coefficients of ZTOR scaling with
earthquake magnitude. The estimated scaling coefficients for earthquakes
in nine 0.5-unit magnitude bins are shown as solid circles, and their 95%
confidence intervals as vertical bars. The solid blue curve is the nonlinear
least-squares fit to the bin coefficients by Equation (3.1). ....................................16
Figure 3.3 Plots showing site terms versus ΔZ1.0. Solid curves are the smooth of site
terms over ΔZ1.0 by local linear regression. The two dashed curves are the
fitted models of Equation (3.4) to the site terms of Japan (green) and
California (red).......................................................................................................18
Figure 3.4 Residuals for simulated ground motion data at sites on top of the hanging
wall (RJB=0) for reverse faulting earthquakes with surface rupture (ZTOR =
0). Solid curves indicate locally weighted least squares (Loess) fits to the
data for each magnitude and the dashed line indicates the overall fitted
model......................................................................................................................21
Figure 3.5 Comparison of revised hanging wall scaling formulation with that in
Chiou and Youngs [2008]. New model based on Equation (3.3) is shown
in red and 2008 formulation in black. ....................................................................22
Figure 3.6 Distance attenuation of hanging wall effect. Red solid curve is a loess fit
to the residuals and the red dashed curves indicate the 90% confidence
interval of the fit.....................................................................................................23
Figure 3.7 Comparison of additive distance coefficient obtained from fits to
individual earthquakes added to the NGA West2 database to relationships
developed in Chiou and Youngs [2008a] (shown by the solid curves). ................26

ix
Figure 3.8 Fit of the distance scaling function form Equation (3.10) to the data for the
2007/10/31 M5.45 earthquake in Northern California. Solid colored
curves indicate the fitted model plotted for the average VS30 and Z1.0 of the
data and the corresponding dashed horizontal colored lines indicate the
truncation level used in the fit. ...............................................................................28
Figure 3.9 Fit of the distance scaling function form Equation (3.10) to the data for the
1994/01/17 M6.69 Northridge earthquake in Southern California. Solid
colored curves indicate the fitted model plotted for the average VS30 and
Z1.0 of the data and the corresponding dashed horizontal colored lines
indicate the truncation level used in the fit. ...........................................................29
Figure 3.10 Values of  obtained from fits of the distance scaling function form
Equation (3.10) to the data for individual California earthquakes. The
vertical lines indicate 90% confidence intervals for the individual
earthquake  values. ..............................................................................................30
Figure 3.11 Average values of  for California earthquakes in three magnitude
intervals obtained from fits of the distance scaling function form Equation
(3.10) to the data for individual California earthquakes. The data points
represent variance weighted averages of the fits for individual earthquakes
and the vertical lines indicate 90% confidence intervals for the individual
earthquake values. ...............................................................................................31
Figure 3.12 Initial model for  (M) as a function of spectral period for California. .................32
Figure 3.13 Values of obtained from fits of the distance scaling function form
Equation (3.2) to the data for individual earthquakes in other regions. The
vertical lines indicate 90% confidence intervals for the individual
earthquake  values. The black dashed line indicates the California model
for  shown on Figure 3.12. ..................................................................................33
Figure 3.14 Values of Ci (c5cosh{c6max(M-cHM,0)} for an individual earthquake)
obtained from fits of the distance scaling function form Equation (3.10) to
the data for individual earthquakes. The vertical lines indicate 90%
confidence intervals for the individual earthquake Ci values. The black
dashed line indicates relationship developed in Chiou and Youngs
[2008a]. ..................................................................................................................34
Figure 3.15 Comparison of linear VS30-scaling coefficient obtained from fits to the
dataset of this study to relationships developed in Chiou and Youngs
[2008a] (shown by the dashed curves)...................................................................35
Figure 3.16 Values of τ (top) and σ computed for PGA residuals in magnitude bins 0.5
magnitude units wide. ............................................................................................40
Figure 3.17 Values of τ (top) and σ computed for 0.17-sec PSA residuals in magnitude
bins 0.5 magnitude units wide. ..............................................................................41
Figure 3.18 Values of τ (top) and σ computed for 1-sec PSA residuals in magnitude
bins 0.5 magnitude units wide. ..............................................................................42

x
Figure 3.19 Values of τ (top) and σ computed for 3-sec PSA residuals in magnitude
bins 0.5 magnitude units wide. ..............................................................................43
Figure 3.20 Values of τ1, τ2, σ1, and σ2 computed from the residuals excluding non-
linear soil effects. ...................................................................................................44

Figure 4.1 Between-event residuals (event terms) for spectral periods of 0.01 (PGA),
0.2, 1, and 3 sec. .....................................................................................................50
Figure 4.2 Within-event residuals for spectral period of 0.01 sec (PGA) plotted
against M, RRUP, VS30, and Z1.0. ............................................................................51
Figure 4.3 Within-event residuals for spectral period of 0.2 sec plotted against M,
RRUP, VS30, and Z1.0. ..............................................................................................52
Figure 4.4 Within-event residuals for spectral period of 1 sec plotted against M, RRUP,
VS30, and Z1.0. .......................................................................................................53
Figure 4.5 Within-event residuals for spectral period of 3 sec plotted against M, RRUP,
VS30, and Z1.0. .......................................................................................................54
Figure 4.6(a) Intra-event residuals plotted as a function of VS30. These residuals are
computed for the updated model for the VS30=1130 m/sec condition. Data
from M > 6 earthquake are shown in black; data from M < 6 are in blue.
The smoothed VS30 trend of these residuals is shown as the thick orange
curve. The thick solid red curve is the predicted nonlinear soil
amplification by our updated model and the thin read solid line is the
predicted linear amplification. The range of event-specific median motion
on reference condition is shown in the lower left corner of each plot. ..................55
Figure 4.6(b) (Continued) ............................................................................................................56
Figure 4.7(a) Predicted median amplitude vs. magnitude for vertical strike slip
earthquake (ZTOR=0, Z1.0= 0, and DPP=0)......................................................58
Figure 4.7(b) (Continued) ............................................................................................................59
Figure 4.8(a) Predicted median amplitude vs. distance for vertical strike slip earthquake
(ZTOR=0, Z1.0= 0, and DPP=0). .......................................................................60
Figure 4.8(b) (Continued) ............................................................................................................61
Figure 4.9(a) Predicted median amplitude vs. distance for reverse slip earthquake of 45
dip. Note that for this comparison is for surface rupture event, ZTOR=0.
Z1.0 and DPP are equal to 0. ..............................................................................62
Figure 4.9(b) (Continued) ............................................................................................................63
Figure 4.10 Median response spectra predicted by the 2008 Chiou and Youngs NGA
model (thin lines) and the updated model (thick lines). Predictions are
made for vertical strike-slip earthquakes and NEHRP B-C boundary
(VS30=760 m/sec). ...................................................................................................64

xi
Figure 4.11 Median response spectra predicted by the 2008 Chiou and Youngs NGA
model thin lines) and the updated model (thick lines). Predictions are
made for vertical strike-slip earthquakes and firm soil condition
(VS30=310 m/sec). ...................................................................................................65
Figure 4.12 Total standard deviations for the updated model with effects for linear soil
response..................................................................................................................66
Figure 4.13 Total standard deviations for the updated model with effects of nonlinear
soil response. ..........................................................................................................67

xii
 LIST OF TABLES

Table 3.1 Period-independent coefficients of model for ln(y) – Equation (3.11). .................45
Table 3.2(a) Period-dependent coefficients of model for ln(yref) – Equation (3.11). .................45
Table 3.2(b) (continued) .............................................................................................................46
Table 3.3 Coefficients of site response model for ln(y) – Equation (3.12). ...........................47
Table 3.4 Coefficients of variance model – Equation (3.13). ................................................48

Table A.1 Model coefficients for non-California regions. .....................................................76

xiii
xiv
1 Introduction

This report presents an update to the Next Generation Attenuation (NGA) ground motion
prediction equation (GMPE) developed by Chiou and Youngs [2008a]. The updated GMPE is
based on analysis of the greatly expanded strong motion database developed by the Pacific
Earthquake Engineering Research Center’s (PEER) NGA-West2 Project, augmented by
extensive numerical ground motion simulations conducted as part of the NGA-West2 Project. As
was the case for Chiou and Youngs [2008a], the update model is for estimating horizontal
ground motions caused by shallow crustal earthquakes occurring in active tectonic environments.
Our model provides predictive equations for the orientation-independent average horizontal
component of ground motions (RotD50) [Boore 2010]. Equations are provided for peak
acceleration and 5%-damped pseudo-spectral acceleration for spectral periods of 0.01 to 10
seconds. The focus of this report is on developing GMPE for California, which is the main
deliverable most useful to the sponsors (CEA, Caltrans, and PG&E), using exclusively California
data from Class 1 earthquakes (mainshocks). In the final step of model development, the
California data were supplemented with recordings from large earthquakes in other active
tectonic regions to verify and refine the magnitude scaling and to provide more robust estimates
of aleatory variability. Regional difference in ground-motion scaling between active tectonic
regions in terms of site effects and distance attenuation were addressed as part of inclusion of the
additional data. The updated Chiou and Youngs model developed for this report contains some
modifications to the functional form developed by Chiou and Youngs [2008a] and the
introduction of two terms to model dependence of ground-motion amplitude on fault dip and
rupture directivity.

1
2
2 Ground Motion Data Used to Develop the
Updated Model

The empirical data set used in this update was selected from the PEER NGA-West2 database
[Ancheta et al. 2013]. We supplemented the NGA-West2 flat file with imputed values of missing
metadata. These supplemental data, as well as record selection criteria, are discussed in this
chapter.

2.1 SELECTION OF EMPIRICAL DATA

The data selection criteria used in this study is the same as that used previously as described in
Chiou and Youngs [2008a, 2008b]. The dataset was restricted to free-field motions from shallow
crustal earthquakes in active tectonic regions, principally California. Recordings made in large
buildings and at depth were removed. We kept records from sites that have been characterized as
having topographic effects (e.g., Tarzana Cedar Hill Nursery, Pacoima Dam left abutment). Our
rationale for including these records is that the effect of topography has not been systematically
studied for all of the records in the database and many other recording stations may have
topographic enhancement or suppression of ground motions. Topographic effects are considered
to be part of the variability introduced into ground motions by travel path and site effects. The
ground motion model developed in this study explicitly accounts for site conditions. Therefore,
recordings from sites for which there is no available information of the local soil conditions were
excluded.
In Chiou and Youngs [2008a, 2008b] we used a combined empirical dataset from a
number of active tectonic regions (principally from California, Taiwan, and Turkey, with limited
data from Iran, Italy, and Japan). We did not find notable differences in the data from these
regions. The expanded PEER NGA-West2 database included additional data from earthquakes in
Italy and Japan, as well as data from earthquakes in New Zealand and China (Wenchuan).
Preliminary evaluations of these data by Chiou and Youngs [2012] suggested that there may be
regional differences in anelastic attenuation and VS30 scaling of ground motion amplitudes among
the various active tectonic regions. Such differences were further examined and quantified in
Section 3.1.3 and 3.2.2 as part of the development of the updated GMPE.
In the early stage of model development, we decided to focus on a California model and
used only California data, therefore avoided the needs for addressing regional difference in
ground motion scaling. After completing the California-specific GMPE, we decided to bring
back relatively well-recorded large non-California earthquakes because the number of California

3
earthquakes with magnitude greater than 7 is smaller than desired. Selection criteria of the non-
California earthquakes were: (1) magnitude is M6 or greater; (2) number of recording is five or
greater. These earthquakes were brought back to help verify and refine the magnitude scaling,
style of faulting effects, and hanging wall effects. They also help provide more robust estimates
of aleatory variability.
Data from earthquakes that occurred in oceanic crust offshore of California and Taiwan
in oceanic crust were excluded because ground motions from these types of events have been
found to be more consistent with ground motions from Wadati-Benioff zone (subduction
intraslab) earthquakes than shallow crustal earthquakes [Geomatrix Consultants 1995]. Data
from the 1992 Cape Mendocino earthquakes were included because the source depth places the
event above the likely interface location [Oppenheimer et al. 1993].
Previously, we developed the GMPE using only data for distance of 70 km or less. This
data cutoff was aimed to limit the impacts of bias in the data sample introduced by data
truncation at large distances. In the updated model, instead of applying the same cutoff distance
to all earthquakes, we developed an assessment of the maximum usable distance for each
earthquake so that we could utilize as much data as possible for well-recorded earthquakes. This
assessment also resulted in a shorter maximum distance (less than 70 km) for some of the older
events, thus actually reduced the number of older data used in regression. More details are
discussed later in Section 3.2.2.
In Chiou and Youngs [2008a] we included data from aftershocks (the Class 2 earthquakes
as defined in Wooddell and Abrahamson [2012]) to help constrain the coefficients of the site
response model. With the NGA-West2 database, it is no longer necessary to do so. Therefore, we
removed data from Class 2 earthquakes located in the vicinity (within 20 km) of a Class 1
earthquake rupture. One notable Class 2 earthquake removed is the 1999 Duzce earthquake.
After applying the selections described above, a total of 12,444 records obtained from
300 earthquakes were selected for the final update of our GMPE. Among them, a total of 2587
records were selected from 18 non-California earthquakes. Figure 2.1 includes a scatter plot of
distance-magnitude-region distribution of the final dataset. Finally, in the regression of response
spectra at a spectral period T, we included only records having a minimum usable spectral
frequency that is lower than 1/T. This selection practice led to a decreased data size as T
increased (Figure 2.2).

4
Figure 2.1 Magnitude-distance-region distribution of selected records.

5
 Figure 2.2 Number of usable records as a function of spectral period.

2.2 USE OF GROUND MOTION SIMULATIONS

Our 2008 GMPE included terms that model an increase in ground motion amplitudes at sites on
the hanging wall of dipping fault ruptures. There were very limited empirical data available from
hanging wall sites and the function form of the hanging wall model was largely based on
conceptual considerations. The expanded NGA-West2 database includes some additional data
from hanging wall sites, but the increase is not large. To augment these data, the NGA-West2
sponsored a set of ground motion simulations focused on investigating the variation of ground
motion amplitudes near and on the hanging wall of dipping fault ruptures. Donahue and

6
Abrahamson [2013] presented a summary of the simulation data and developed a model to
represent the hanging wall effect.

2.3 SUPPLEMENT OF GROUND MOTION METADATA

2.3.1 Sediment Thickness (Z1.0)

The thickness of the near-surface sediments is represented in our NGA GMPE by the depth to
the shear wave velocity horizon of 1.0 km/sec, Z1.0. The completeness of this variable was
greatly improved during the development of NGA-West2 database. The database provided Z1.0
for sites within the Southern California Earthquake Center three-dimensional (3D) basin model,
for sites in the USGS velocity model for the San Francisco Bay area, for sites in Japan, and for
sites where measured shear-wave velocity profiles reached the 1.0 km/sec horizon.
For a site without Z1.0, we estimated its Z1.0 using an empirical correlation with VS30. A
correlation was previously developed from the available (VS30, Z1.0) data in the NGA-West1
database [Chiou and Youngs 2008a, 2008b]. Additional data in the NGA-West2 database and
careful checking (which led to correction of obvious data errors and removal of data with default
Z1.0 value) has resulted in a revised correlation between VS30 and Z1.0.
The majority of data used in the revised relation were from sites in California (Figure 2.1)
and Japan (Figure 2.2). There is a clear difference in VS30 -Z1.0 relation between these two
regions; Z1.0 is smaller in Japan than in California for soft soil sites. Also, for VS30 > 500 m/sec,
the rate of decrease with VS30 is slower in Japan than in California. For these two reasons, we
developed two VS30- Z1.0 relations, one for each region, as follows.
For California:

7.15 V 4  5714
ln( Z1.0 )  ln( S 30 4 )
4 1360  5714 (2.1)

For Japan:

5.23 V 2  412 2
ln( Z1.0 )  ln( S 30 2 ) (2.2)
2 1360  412 2

Based on Figure 2.3, one could argue that there is also a difference between southern
California sites (red circles in the figure) and central California sites (blue circles). However, the
difference is less than the difference between California and Japan. Because we focused on
developing a California wide model, this difference was not included. Future refinements of our
model will explore the need for regionalization within California.

7
Figure 2.3 Relationships between VS30 and Z1.0 for California.

8
Figure 2.4 Relationship between VS30 and Z1.0 for Japan.

9
2.3.2 Depth to Top of Rupture (ZTOR)

Source depth is represented in our GMPE by the depth to the top of rupture, ZTOR. In the NGA-
West2 database, ZTOR was determined either from the available finite fault model or from
hypocentral depth and M using the simulation method described in Appendix B of Chiou and
Youngs [2008b]. As such, ZTOR had been estimated for every earthquake in the database, except
for a few poorly determined (small) earthquakes whose hypocenter location or magnitude is
unknown.
Using California data in the NGA-West2 database, we developed a model relating ZTOR to
M and style of faulting. The developed relationship was used later to center ZTOR (Section 3.1.2).
In the forward application of our updated GMPE, this relationship could also be used as a tool to
estimate the ZTOR of a future California earthquake, given its magnitude and style of faulting.
Following Kaklamanos and others [2011], we applied the square root transformation to
ZTOR to increase the normality of residuals. The scatter plot of Figure 2.5 indicates that, while
being a constant at M < 5, is approximately a linear function of M where M > 6. This
observation, together with the non-negative value of , prompted us to model as a
linear ramp function of M using the relationship:

ZTOR  max[a b max(M  c, 0), 0] (2.3)

In Equation (2.3), parameter c is the threshold magnitude below which is constant a, and
parameter b is the slope for M > c. As noted by Kaklamano et al. [2011], the -M
relationship for reverse and reverse oblique faulting is different from that for the other styles of
faulting. We thus developed two models. One for the reverse and reverse oblique faulting,

ZTOR  max[2.704  1.226 max(M  5.849, 0), 0]2 (2.4)

and the other for the combined strike-slip and normal faulting,

ZTOR  max[2.673 1.136 max(M  4.970, 0), 0]2 (2.5)

ZTOR values predicted by these two models are shown as solid lines in Figure 2.5. The
dot-dashed lines in the figure are predictions by the models of Kaklamano et al. [2011]
developed using worldwide earthquakes in the NGA-West1 database. It should be noted that
these models are similar at M > 5.5, where the datasets used in both models overlap
considerably.

10
Figure 2.5 A scatter plot showing and M of California earthquakes in the
NGA-West2 database. Solid lines are the regression models
developed in this study for reverse faulting (blue line) and
combined strike-slip/normal (red line) faulting. Dot-dashed lines are
the models of Kaklamano et al. [2011].

11
12
3 Updated GMPE

As evidenced by the analysis presented in Choiu et al. [2010], the inclusion of a large data set of
ground motions from small to moderate magnitude earthquakes (SMM) has a significant effect
on ground motion scaling relationships. For this study, we defined SMM events as earthquakes
in the range of 3 < M < 5.5. A large percentage of SMM ground motion records were obtained
from the broad band networks of southern California and central California. Incorporation of
these data as well as data from the additional well recorded large M ≥ 5.5 magnitude (LM)
earthquakes has led to modifications of model formulation and refinements of model
coefficients. These changes and their basis are discussed in the first two sections of this chapter.
The updated models for median motion and aleatory variability are presented in Sections 3.3 and
3.4, respectively.
In several exploratory analyses of the NGA-West2 database, we noted that ground-
motion scaling may be different between SMM and LM earthquakes. After examining several
source scaling in exploratory analysis, we selected the following function to model the observed
differences,

  
f (M i )     (3.1)
 cosh  2  max (M i  4.5, 0)  

Equation (3.1) has a gradual transition from the asymptote (+ ) at M < 4.5 to the other
asymptote  at M6.5, straddling the magnitude boundary between SMM and LM. With this
property, Equation (3.1) decoupled the scaling for LM and SMM and prevented undue influence
on LM scaling by SMM earthquakes. The latter was particularly useful when the metadata in
question (such as magnitude and focal mechanism) was of lesser quality for SMM earthquakes.

3.1 MODIFICATION OF MODEL FORMULATION

In response to the outcomes of several exploratory analyses, we modified the parts of our
previous formulation related to the effects of style of faulting, depth to the top of rupture,
hanging wall, and Z1.0. These modifications and their basis are described in the first four
subsections. We also added two terms to model the effects of rupture dip angle and rupture
directivity; they are presented in the last two subsections.
Given that Class 2 (aftershock) data were no longer needed for constraining site response
model, we decided not to include them in this update in order to simplify our analysis.

13
Consequently, we removed terms related to the effects of Class 2 earthquakes. We plan to
explore and quantify the aftershock effects in a follow-up study.

3.1.1 Style of Faulting Effect for Reverse and Normal Faulting

As stated previously, we noted difference in style-of-faulting effect between SMM and LM

earthquakes. As an example demonstrating such difference, Figure 3.1 shows event terms
computed for an interim GMPE without style-of-faulting effects for 4 spectral periods. As
indicated in the figure, reverse faulting effect (difference in smoothed curve between RV and SS)
is stronger at LM than at SMM. Conclusions about normal faulting effect are less definite
because there are only 8 normal faulting events in our dataset.

Figure 3.1 Event terms computed from an interim GMPE demonstrating the
difference in style of faulting effect between small-to-moderate and
large magnitude earthquakes.

14
 Equation (3.1) tracked the magnitude trend of reverse faulting effect quite well. As a
result, we modified our term for reverse faulting effect

 c1c 
from c1a FRVi to c1a   FRVi
 cosh( 2  max(M i  4.5, 0)) 

and our term for normal faulting effect

 c1d 
from c1b FNMi to c1b   FNMi
 cosh( 2  max(M i  4.5, 0)) 

FRVi and FNMi are reverse faulting and normal faulting flags defined in Chiou and Youngs [2008a,
2008b]. See also Section 3.3.1.

3.1.2 Depth to the Top of Rupture

We implemented two changes to the formulation of source depth effect. First, we centered ZTOR
on the ZTOR-M relationships of Equations (2.4) and (2.5), rather than on a constant value of 4 km
as in our 2008 GMPE. By centering on a M-dependent average, the M scaling of average ZTOR
was transferred to the general magnitude scaling, resulting in a sharp corner in the magnitude
scaling curve, as evidenced by the much increased cn at high frequencies (Table 3.2) compared to
our 2008 model. This change in modeling concept has a small impact on median predictions.
Instead of carrying Equations (2.4) and (2.5) in the equation for median, we created the new
variable ZTOR = ZTOR – E[ZTOR], where E[ZTOR] is the mean ZTOR given by Equations (2.4) and
(2.5).
Secondly, we introduced M dependence into the ZTOR scaling coefficient c7. Similar to
style of faulting effects, we noted a difference in ZTOR scaling between SMM and LM
earthquakes. As an example demonstrating such difference, we computed event terms of
California earthquakes for an interim GMPE without ZTOR effects for 4 spectral periods. We
grouped the event terms into nine non-overlapping magnitude bins. For each bin, we fitted event
terms to a linear function of ZTOR. The slopes of the fitted line (coefficient of ZTOR scaling), along
with their 95% confidence intervals, are shown in Figure 3.2 against magnitude. From the figure,
we concluded that Equation (3.1) adequately model the magnitude dependence of ZTOR scaling.
Based on the above discussions, we modified the ZTOR formulation from

c7 ( Z TORi  4)

to

 c7 b 
c 7   Z TORi
 cosh( 2  max(M i  4.5, 0)) 

15
 Figure 3.2 Plots showing the variation of estimated coefficients of ZTOR scaling
with earthquake magnitude. The estimated scaling coefficients for
earthquakes in nine 0.5-unit magnitude bins are shown as solid
circles, and their 95% confidence intervals as vertical bars. The
solid blue curve is the nonlinear least-squares fit to the bin
coefficients by Equation (3.1).

3.1.3 Z1.0 Scaling

Sediment depth, represented by the depth to shear-wave velocity of 1.0 km/sec (Z1.0), was used
along with VS30 to model the amplification of surface motion by local site condition. Our
previous study revealed ground motion amplification on deep sediment sites, i.e., sites whose Z1.0
is larger than coefficient 7 in 2008 Chiou and Youngs GMPE. However, we did not observe

16
clear evidence of the de-amplification of ground motion on shallow soil sites (Z1.0 smaller than
7) as predicted by site response analysis. Guided by site response analysis, Abrahamson and
Silva [2008] implemented a large de-amplification of long-period motions when a site’s Z1.0 is
far below the average Z1.0 for the site VS30. Encouraged by their work and the benefits of
predictor centering seen earlier, we centered Z1.0 on the average Z1.0 given by Equations (2.1) and
(2.2) and used the centered variable ∆ . as predictor of sediment depth effects.
With ∆ . , de-amplification on shallow soils was clearly illuminated. This modeling
improvement is shown in Figure 3.5 by the negative average of station terms for negative ∆ . .
The station term in the figure was computed as the average residual of records available at a
specific site and the residuals were from interim analyses that included both linear and nonlinear
soil effects. The figure also suggests a stronger amplification for positive Δ . in Japan than in
California, suggesting a need for regionalization of sediment depth effects.
Guided by plots similar to Figure 3.5, we selected the following functional form for Δ .
scaling in our updated model,

Z 
5 1  e 1.0 j 6  (3.4)
 

17
Figure 3.3 Plots showing site terms versus ∆Z1.0. Solid curves are the smooth
of site terms over ∆Z1.0 by local linear regression. The two dashed
curves are the fitted models of Equation (3.4) to the site terms of
Japan (green) and California (red).

18
3.1.4 Hanging Wall Scaling

The hanging wall (HW) scaling model developed in Chiou and Youngs [2008a] involved an
amplification of ground motions for sites on the hanging wall that increased with increasing
absolute value of RX. The amplification function peaked at a value of RX that was independent of
the location of the down dip edge of the rupture. Once the value of RX increased to the point
where the site location moved beyond the surface projection of the bottom of the rupture, then a
distance taper function rapidly decreased the hanging wall amplification. The analyses of ground
motion simulation results presented in Donahue and Abrahamson [2013] show a somewhat more
gradual rate of increase in the hanging wall amplification for sites above the rupture such that the
peak amplification occurs nearer to the down dip edge of the rupture. Because there is very
limited empirical data that can be used to define the RX trends of HW amplification on top of the
rupture, we used the simulation data to develop a revised hanging wall model.
The process used was similar to that employed by Donahue and Abrahamson [2013]. The
footwall and neutral site data for individual simulations were fit using a simple distance
attenuation functional form. Then the residuals with respect to this model were computed for the
data from the simulated earthquake for sites located above the rupture (RJB = 0). The process was
repeated for each simulation, and the resulting sets of residuals combined. Figure 3.4 shows the
resulting sets of residuals. Each plot shows the results for one dip angle. With the exception of
the results from the M6 simulations, the residuals exhibit a similar trend for all of the larger
magnitudes. This trend can be modeled by the function tanhRX / c9b  , where c9b is a constant
independent of magnitude. The residuals for the M6 simulations exhibit a different behavior, in
that they peak at very small values of RX, rather than at values of RX near the down dip extent of
the ruptures, which is the case for the simulations from larger magnitudes. This different
behavior may be the effect of the simulation process for smaller magnitudes. Because of this
different behavior, the residuals for the M6 simulations were not used to develop the hanging
wall model.
The residuals shown on Figure 3.4 show a decrease in amplitude with increasing dip
angle, δ, that can be modeled as a function of cos(δ). The use of cos(δ) rather than δ is motivated
by considering the hanging wall effect to be a geometrical effect representing the location of a
site relative to the projection of the entire rupture plane to the surface, and the this projection is
directly related to cos(δ). The simulation residuals also show a small step at RX = 0. The
simulations were fit with the model form given by Equation (3.5).

 R    2
RJB  ZTOR
2 
f HW  c9 cos( ) c9 a  (1  c9 a ) tanh  X   1   (3.5)
  c9b    RRUP  1 

The fitted values of this function is compared to the residuals for RJB = 0 sites on
Figure 3.4. Figure 3.5 compares the hanging wall scaling factor (without the absolute scaling
coefficient c9) from Equation (3.5) to that for the Chiou and Youngs [2008a] model. The revised
model shows stronger magnitude scaling than the previous model, consistent with the results of
the simulations as modeled by Donahue and Abrahamson [2013]. The magnitude scaling is
introduced by the term tanhRX / c9b  . This term increases with increasing Rx, and this can only
occur with wider ruptures (and corresponding larger magnitudes). Also consistent with Donahue

19
and Abrahamson [2013], we have implemented a step in ground motions at the fault trace for
surface rupturing earthquakes. In our formulation, this step in ground motions disappears as the
depth to top of rupture increases from zero. The constant in the denominator of the term
 2
RJB  ZTOR
2 
1   was increased from 0.001 used in Chiou and Youngs [2008a] to 1 to provide
 RRUP  1 

a smoother transition in amplitude with increasing ZTOR.
The Donahue and Abrahamson [2013] hanging wall model shows less rapid attenuation
of the hanging wall amplification with increasing RX for sites located beyond the down dip edge
of the rupture than the formulation developed by Chiou and Youngs [2008a]. For these types of
site locations there is more empirical data available. These data indicate a faster decay of the
hanging wall amplification with increasing RX than is defined by the Donahue and Abrahamson
[2013] hanging wall model. Figure 3.6 compares the residuals computed using Equation (3.5) for
sites on the hanging wall side (positive Rx) for reverse faulting earthquakes against the distance
 2
RJB  ZTOR
2 
scaling term  1   . The solid and dashed red curves indicate the loess fit to the
 RRUP  1 

residuals and the 90% confidence interval on the fit, respectively. As was found in Chiou and
 2
RJB  ZTOR
2 
Youngs [2008a, 2008b], the data are consistent with a linear trend in the term 1  .
 RRUP 1 

Therefore, this distance decay formulation of Chiou and Youngs [2008a] was retained in the
updated GMPE formulation with the small modification described above.

20
Figure 3.4 Residuals for simulated ground motion data at sites on top of the
hanging wall (RJB=0) for reverse faulting earthquakes with surface
rupture (ZTOR = 0). Solid curves indicate locally weighted least
squares (Loess) fits to the data for each magnitude and the dashed
line indicates the overall fitted model.

21
Figure 3.5 Comparison of revised hanging wall scaling formulation with that in
Chiou and Youngs [2008]. New model based on Equation (3.3) is
shown in red and 2008 formulation in black.

22
 Figure 3.6 Distance attenuation of hanging wall effect. Red solid curve is a
loess fit to the residuals and the red dashed curves indicate the
90% confidence interval of the fit.

3.1.5 Effect of Rupture Dip Angle

In exploratory analysis, we found that the event terms of SMM earthquakes decreased
systematically with decreasing dip angle. We did not observe similar dependency in LM events.
As a result, we added a term for dip angle and used Equation (3.1) to model the absence of this
effect in LM earthquakes. Functional form of the added term is

 c11a 
  cos( i )
2
c
 11
 cosh(2  max(M i  4.5, 0 )) 

Coefficient c11 (dip-angle scaling for LM earthquakes) was fixed to 0 for all periods (see
Table 3.2).

3.1.6 Directivity Effect

The 2008 NGA-West1 GMPEs did not include directivity. Instead, directivity was implemented
as a post facto factor. Implementation of directivity effect in the updated GMPE was one of the
focuses of our model update. Among the five directivity parameterizations discussed in the
Directivity Working Group report [Spudich et al. 2013], we choose the direct point parameter
(DPP) as our predictor for directivity effect. The rationales for choosing DPP over IDP (the

23
ioschrone directivity parameter used in Spudich and Chiou [2008]) were given in Chiou and
Spudich [2013].
As with Z1.0 and ZTOR, we centered DPP on the average DPP of the earthquake in
question. The average DPP is the mean over a suite of sites located at the same distance to that
earthquake. It is a function of distance and is specific to the earthquake rupture being
investigated. Using DPP (centered DPP) forces a GMPE to explicitly scale ground-motion
amplitude relative to the median representing the average directivity at a given distance to fault.
Centering also allows the GMPE users to predict the average directivity effect (by simply setting
DPP to 0) if site- and rupture-specific directivity is not to be explicitly included in the ground
motion prediction.
We adopted the narrow-band formulation proposed by Spudich and Chiou [2013] for its
improved handling of M- and period-dependence. Their form was rearranged so that it could be
incorporated into our regression model one period at a time. The rearranged form is

c8 a ( M  c8 b )2
f D  c8 f R f M e DPP
(3.6)

Relating to the original narrow-band formulation, coefficient c8a is related to the bandwidth
parameter (g), coefficient c8b is the magnitude for which the analyzed period is the period of
peak effect (Tmax), and coefficient c8 is the peak effect (bmax). Note that in the original
formulation c8 is a linear function of M in the M > 5.7 range. In our analysis we found that this
linear dependence was unstable across periods and was statistically insignificant for many
periods; therefore we did not include the linear magnitude dependence in our updated model.
Also in the original formulation, c8 is period independent. This property cannot be implemented
in our one-period-at-a-time regression setting. As a workaround, we estimated c8 for each of the
analyzed periods between 0.75 sec and 10 sec, and took the weighted average of the individual
estimates as our final c8 estimate.
Function fR in Equation (3.6) is the same distance taper used in Spudich and Chiou
[2008].

 max( RRUP  40, 0) 

f R  max 0, (1   (3.7)
 30 

Function fM is a magnitude taper similar to the one used by Spudich and Chiou [2008],

 max(M  5.5, 0) 
f M  min 1,  (3.8)
 0.8

Due to the absence of finite-fault information, directivity effect for M < 5.7 earthquakes cannot
be investigated in this update. We assumed directivity effect at M < 5.5 is negligible, though
some recent studies (for example, Boatwright [2007]) suggested otherwise. This assumption was
implemented via the magnitude taper fM, which reduced c8 from a non-zero value to zero from M
6.3 to M 5.5.

24
3.2 REFINEMENT OF MODEL COEFFICIENTS

The formulation for magnitude and distance scaling of ground motion amplitudes was examined
as part of the update efforts. Our conclusion was that the formulation continues to provide a good
fit to the trends in the NGA-West2 database, although model coefficients needed to be adjusted.

3.2.1 Magnitude Scaling

Development of the magnitude scaling formulation in our 2008 GMPE was guided by the results
of simulations using seismological models for earthquake source spectra and evaluation of
ground motion data of small to moderate California earthquakes. The functional form is given in
Equation (3.9).

1
ln( y )  c2 M  ( c2  c3 )  ln 1  exp cn cM  M 
cn (3.9)

Coefficient c2 is the slope of the magnitude scaling relationship for earthquakes whose
theoretical corner frequency is well above the spectral frequency of interest and c3 is the slope
for earthquakes whose corner frequency is well below the spectral frequency. Coefficient cn
controls the width of the magnitude range over which the transition from c2 scaling to c3 scaling
occurs. Coefficient cM is the magnitude at the midpoint of this transition and its value varies with
the spectral period of the ground motion parameter y. Subsequent analyses by Chiou et al. [2010]
showed that this functional form is able to model magnitude scaling of strong ground motion
over the magnitude range of M3 to M8, requiring only modification of the model coefficients
from those presented in Chiou and Youngs [2008] to extend the GMPE below magnitude M5.5.
The analyses of the extensive data from M3.5 to M5 earthquakes included in the NGA-West2
database confirms that the functional form for magnitude scaling developed in Chiou and
Youngs [2008a] appropriately models ground motion amplitudes over a wide magnitude range.
As discussed above, coefficient c3 and cM were revised to reflect the magnitude scaling of
SMM earthquakes in NGA-West2 database. Due to the centering of ZTOR, coefficient cn was also
revised to accommodate the M-scaling in the average ZTOR, as discussed in Section 3.12.

3.2.2 Distance Scaling

Analyses presented in Chiou and Youngs [2008b] demonstrated that a range of formulations for
distance scaling could be used to satisfactorily model the magnitude-dependent effects of
extended ruptures on the distance scaling of ground motion amplitudes in the distance range of 0
to approximately 100 km. The formulation developed in Chiou and Youngs [2008] is shown by
Equation (3.10) The formulation utilizes a magnitude (and period) independent near source
geometric attenuation coefficient, c4, coupled with a magnitude dependent additive distance
constant (some called it fictitious depth), defined by the expression c5cosh{c6max(M-cHM,0)}, to
capture the effects of extended ruptures. This near source distance scaling then transitions into a
far source distance scaling proportional to R-1/2 (coefficient c4a was set to -0.5) in order to model
the transition from body wave geometric spreading near the source to surface/Lg wave geometric
spreading at larger distances. The far source distance scaling was coupled with a magnitude
dependent attenuation term γ(M)RRUP to model the effects of anelastic attenuation and scattering

25
(i.e., Q). This function form was shown to model the variation in ground motion amplitudes over
the distance range of 0 to several hundred kilometers for data from well recorded earthquakes.
The formulation has the advantage of providing a convenient mechanism to accommodate
differences in Q among different tectonic regions.

ln( y)  c4 ln RRUP  c5 coshc6 max(M  cHM ,0)  (c4a  c4 )ln RRUP

2
 cRB
2
  (M)RRUP (3.10)

The magnitude dependent additive distance constant term in Chiou and Youngs [2008a]
is period dependent as was derived from fitting the general distance scaling functional form to
the ground motion data for individual earthquakes. Figure 3.7 compares the Chiou and Youngs
[2008a] relationships for c5cosh{c6max(M-cHM,0)} with values of the additive constant obtained
from initial fitting the general distance attenuation function to data for individual large
magnitude earthquakes that were added as part of the update of the ground motion database from
NGA-West1 to NGA-West2. As indicated, the values of the additive distance constant for the
new earthquakes are consistent with the relationships developed in Chiou and Youngs [2008a].

Figure 3.7 Comparison of additive distance coefficient obtained from fits to

individual earthquakes added to the NGA West2 database to
relationships developed in Chiou and Youngs [2008a] (shown by
the solid curves).

The Q term, γ(M) RRUP, was assessed in Chiou and Youngs [2008a] by fitting data for
individual California earthquakes using truncated regression (e.g., Toro [1981]; Bragato [2004])
to account for data truncation at low amplitudes and large distances. Fits to limited data for
earthquakes in other regions were shown to produce similar values for γ(M). More recently,
Chiou and Youngs [2012] found from preliminary analysis of the NGA-West2 database that
there are significant regional differences in γ(M). Accordingly, as part of the updated model, the
functional form of Equation (3.10) was fit to the data of individual earthquakes in the NGA
West2 database. In initial fitting these data, all coefficients of the Chiou and Youngs [2008a]
model were fixed at their published values except for the constant term and the value of γ for

26
each magnitude and spectral period. The fixed coefficients included the additive distance
coefficient c5cosh{c6max(M- cHM,0)}, the VS30 scaling coefficients, and sediment depth (Z1.0)
scaling coefficients. The analysis of γ(M) in Chiou and Youngs [2008a] was restricted to PGA
and used the data from three well recorded small California earthquakes to define the variation of
γ(M) with spectral period. The updated NGA West2 database provides sufficient data to extend
this analysis over the full period range for many earthquakes.
Figure 3.8 shows an example of the fit to the data for a single earthquake, the October 31,
2007 M5.45 earthquake 11 km east of Milpitas in northern California. Each panel shows the fit
for a single ground motion parameter. The black circles indicate the data used in the fitting and
the black X symbols indicate the data excluded because it is outside of the usable frequency
range for the specific recording. Truncated regression analyses were performed using the
truncation levels indicated by the horizontal colored dashed lines. The resulting median models
are indicated by the solid colored curves plotted for the average value of VS30 and Z1.0 of the data.
The vertical dashed line indicates a cut off distance set at the point where the truncation level
represents minus 2.5 standard deviations below the fitted median model. These cut off distances
were used to set the distance limits used in the fitting of the entire dataset in the mixed effects
regressions used to develop the updated model coefficients, as will be discussed below. As
indicated on the figure, the function form of Equation (3.10) provides a good representation of
the trends in the data over a wide distance and spectral period range.
Figure 3.9 shows the fit to the data for the January 17, 1994 M6.69 Northridge
earthquake in southern California. For this earthquake, the effects of data truncation are much
more apparent. At high frequencies, the data are sufficient to provide an estimate of γ, although
with greater uncertainty than the estimates for the earthquake data shown in Figure 3.8. There is
also a greater loss of usable data at longer periods due to a narrow range of usable frequencies.

27
Figure 3.8 Fit of the distance scaling function form Equation (3.10) to the data
for the 2007/10/31 M 5.45 earthquake in Northern California. Solid
colored curves indicate the fitted model plotted for the average VS30
and Z1.0 of the data and the corresponding dashed horizontal
colored lines indicate the truncation level used in the fit.

28
Figure 3.9 Fit of the distance scaling function form Equation (3.10) to the data
for the 1994/01/17 M 6.69 Northridge earthquake in Southern
California. Solid colored curves indicate the fitted model plotted for
the average VS30 and Z1.0 of the data and the corresponding dashed
horizontal colored lines indicate the truncation level used in the fit.

29
 The process shown on Figures 3.8 and 3.9 was repeated for all earthquakes in the NGA-
West2 database with sufficient data to define γ. It was found that five or more recordings at
distances less that 100 km and five or more recordings at distances greater 100 km, with the
recordings space over a wide distance range were needed in order to get a stable estimate of γ.
The resulting values of γ obtained from the fits of individual California earthquakes are shown
on Figure 3.10 for PGA and PSA at periods of 0.05, 0.1, 0.2, 0.3, 0.5, 1, and 2 sec. The values
for individual earthquakes are color coded to indicate Southern California and Northern (Central)
California. Tests of the values did not show a statistical difference between the populations of
values for Northern and Southern California earthquakes and the data for all of California were
used to develop a combined model.

Figure 3.10 Values of  obtained from fits of the distance scaling function form
Equation (3.10) to the data for individual California earthquakes.
The vertical lines indicate 90% confidence intervals for the
individual earthquake  values.

The fitted values of γ for individual earthquakes were then used to examine the variation
of γ with magnitude and period. Figure 3.11 shows the variation of γ with period for three
magnitude intervals. For each period and magnitude, a variance weighted average value of the
fitted values of γ for individual earthquakes was computed. The results show that the variation in
γ with period is magnitude dependent. This somewhat complex behavior is likely due to
differences in the frequency content of motions contributing to peak response spectral amplitude
at a given period as the magnitude of the earthquake changes.

30
 Figure 3.11 Average values of  for California earthquakes in three magnitude
intervals obtained from fits of the distance scaling function form
Equation (3.10) to the data for individual California earthquakes.
The data points represent variance weighted averages of the fits for
individual earthquakes and the vertical lines indicate 90%
confidence intervals for the individual earthquake values.

Smooth curves were fit to the values for each magnitude interval, as shown on Figure
3.11. These curves were used to define the starting model for γ(M) for California shown on
Figure 3.12. The γ(M) model parameters were then refined in the final stages of model building
using the combined data set.
The analysis of data from individual earthquakes was extended to data from other
tectonic regions. Figure 3.13 shows the results of fitting the function form of Equation (3.10) to
data for individual earthquakes from Italy, Japan, New Zealand, Taiwan, Turkey, and Wenchuan,
China. The values for New Zealand, Taiwan, and Turkey are similar to those obtain for
California earthquakes. The values for Italy and Japan indicate greater attenuation with distance
(lower Q). The data for the Wenchuan, China, earthquake and its aftershocks shows less distance
attenuation than California (higher Q). These results indicate the need to account for differences
in γ when combining data from multiple regions. These differences were not readily apparent in
our previous analysis due to the limited extent of data from other regions contained in the

31
previous NGA database and the fact that those earthquakes came from regions with similar
attenuation characteristics (e.g., Taiwan and Turkey).
Using the initial γ(M) model, the next step was to examine the parameters that control the
near source distance scaling to verify the conclusions reached by the comparison shown on
Figure 3.7. The fit to data for individual earthquakes was repeated using the initial γ(M) model
shown on Figure 3.12 to compute the additive distance term c5cosh{c6max(M- cHM,0)}, in
Equation (3.10), which for an individual earthquake is a constant Ci. Figure 3.14 shows the
resulting values compared to the relationship developed in Chiou and Youngs [2008a]. The
results indicate that the values of Ci at larger magnitudes are consistent with our previous
relationship, but the results for smaller magnitudes deviate from the previous model.
Accordingly, parameters c5 and cHM were adjusted as part of this update.

Figure 3.12 Initial model for  (M) as a function of spectral period for California.

32
 Figure 3.13 Values of obtained from fits of the distance scaling function form
Equation (3.2) to the data for individual earthquakes in other
regions. The vertical lines indicate 90% confidence intervals for the
individual earthquake  values. The black dashed line indicates the
California model for  shown on Figure 3.12.

As discussed above, as part of the fits to the data for individual earthquakes, a cutoff
distance was defined that represents an estimate of the largest distance for which the data are
unaffected by truncation. This distance is termed Rmax(T)i and varies with period and from
earthquake to earthquake. The value of Rmax(T)i for each earthquake and period was set at the
point where the truncation level used it fitting the data represents minus 2.5 standard deviations
below the fitted median model. For a number of earthquakes, the values of Rmax(T)i are relatively
large. For example, the results shown on Figure 3.8 indicate values in excess of 200 km for many
periods. However, for other earthquakes, the values of Rmax(T)i are less than the 70 km cutoff
distance we used in developing our previous model. An example of this is the data for the
Northridge earthquake, shown on Figure 3.9. Figure 2.1 shows the resulting magnitude and
distances ranges for the selected data.

33
 Figure 3.14 Values of Ci (c5cosh{c6max(M-cHM,0)} for an individual earthquake)
obtained from fits of the distance scaling function form Equation
(3.10) to the data for individual earthquakes. The vertical lines
indicate 90% confidence intervals for the individual earthquake Ci
values. The black dashed line indicates relationship developed in
Chiou and Youngs [2008a].

3.2.3 VS30 Scaling

Our formulation of VS30 scaling was found to adequately model the behavior of the ground
motions in the NGA-West2 database. We felt, however, it was necessary to re-estimate the linear

34
scaling (coefficient 1) because there were much more weak motion data and data at large
distances in the NGA-West2 database than in the previous database. The much larger NGA-
West2 data set would provide a better constrain on 1. Furthermore, the new Z1.0 scaling
(described in Section 3.1.3) would modify the old median amplitude for soft (low-VS30) sites and
thus alter 1. A comparison of the revised 1 against the 2008 values is given in Figure 3.15. The
new estimates indicate a stronger VS30 linear scaling than what we previously estimated, possibly
due to the two factors stated above.

Figure 3.15 Comparison of linear VS30-scaling coefficient obtained from fits to

the dataset of this study to relationships developed in Chiou and
Youngs [2008a] (shown by the dashed curves).

35
3.3 UPDATED GMPE

3.3.1 Median Ground Motion

The revised model formulation for median ground motions is given by Equation (3.11) for
ground motion on the reference site condition (VS30 = 1130 m/sec),

ln( yrefij )  c1
 c1c 
 c1a   FRVi
 cosh(2  max(Mi  4.5, 0)) 
 c1d 
 c1b   FNMi
 cosh(2  max(Mi  4.5, 0)) 
 c7b 
 c7   ZTORi
 cosh(2  max(Mi  4.5, 0)) 
 c11b 
 c11   (cos  i )
2

 cosh(2  max(Mi  4.5, 0)) 

c c cn (cM Mi )
c2 (Mi  6)  2 3 ln 1  e
cn
( )
(
c4 ln RRUPij  c5 cosh  c6  max(Mi  cHM ,0)  )
(
  c4 a  c4  ln 2
RRUPij  cRB
2
)
 c 2 
 c 1   RRUPij
 cosh  max( M i  c3 , 0)  
 max( RRUPij  40, 0)   max(Mi  5.5, 0) 
c8 max 1  , 0  min  , 1 ec 8a (Mi c8b )2
 DPPij (3.11)
 30   0.8 
 
  RXij    ZTORi
2 2
 RJBij 
c9 FHWij cos  i c9a  (1  c9a ) tanh   1 
 c  RRUPij  1

  9b    

and by Equation (3.8) for ground motion on the surface of soil condition,

36
  VS 30 j 
( ) (
ln  yij   ln yrefij  1  min  ln ) ,0
 1130 
 3  min(VS 30 ,1130) 360 
 
 (1130 360)  yrefij  4
2  e 


 e 3  ln

( 4
) (3.12)
 
 Z1.0 j 6 
5 1  e   i   ij
 

The predictor variables are:

M = Moment magnitude.

RRUP = Closest distance (km) to the rupture plane.

RJB = Joyner-Boore distance (km) to the rupture plane.

RX = Site coordinate (km) measured perpendicular to the fault strike from

the fault line, with the down-dip direction being positive (See Figure
3.12 of Chiou and Youngs [2008b]).

FHW = Hanging-wall flag: 1 for RX ≥ 0 and 0 for RX < 0.

δ = Fault dip angle.

ZTOR = Depth (km) to top of rupture.

FRV = Reverse-faulting flag: 1 for 30º ≤ λ ≤ 150º (combined reverse and

reverse-oblique), 0 otherwise; λ is the rake angle.

FNM = Normal faulting flag: 1 for -120º ≤ λ ≤ -60º (excludes normal-

oblique), 0 otherwise.

VS30 = Travel-time averaged shear wave velocity (m/sec) of the top 30 m of

soil.

Z1.0 = Depth (m) to shear wave velocity of 1.0 km/sec.

ΔZ1.0 = Z1.0 (m) centered on the California-specific average Z1.0 model

[Equation (2.1)].

DPP = Direct point parameter for directivity effect [Chiou and Spudich
2013].

ΔDPP = DPP centered on the earthquake-specific average DPP.

37
 Model coefficients (variable names starting with the letter c or ) are listed in Tables 3.1
to 3-3. In the tables, we underlined the coefficients that were unmodified in this update and used
the bold face on those that were added or given a different meaning (such as 5 and6). Because
we excluded data from Class 2 earthquakes (aftershocks), we did not include Class 2 earthquake
terms in Equation (3.11) and Table 3.2.
To simplify, Equations (3.11) and (3.12) were written for application in California,
although our regression analysis included region-dependent terms to account for the known
regional difference in anelastic attenuation and site effects. To apply our GMPE to regions where
regional differences were accounted for, one should use the region-specific coefficients given in
Appendix A. Also, in application to Japanese data, the Japan-specific average Z1.0 model
[Equation (2.2)] should be used to center Z1.0.

3.3.2 Aleatory Variability

The current terminology used to express the components of aleatory variability use the symbol τ
for the inter-event component and the symbol for intra-event variability, with the symbol σ
used for total aleatory variability, such that      . However, to avoid confusion with our
2 2 2

use of the symbol  for the parameters of the site amplification model, we retain the symbols
used in Chiou and Youngs [2008]: τ for the inter-event component, σ for intra-event variability,
with the symbol σT used for total aleatory variability.
The first stage of the variance analysis was to examine the magnitude dependence of τ
and σ. Figures 3.16–3.19 show the values of τ and σ computed from the residuals for overlapping
magnitude bins with width 0.5 magnitude units. Values are shown based on residuals for all
distances and for distances of 100 km or less. At this stage, the values were computed without
consideration of nonlinear soil effects. The figure shows values of τ and σ for PSA residuals at
periods of 0.01, 0.02, 1.0, and 3.0 sec. Slightly lower values of aleatory variability were obtained
using the residuals for distances of 100 km and less than were obtained using residuals from all
distances. Given the small differences, the residuals for all distances were used to develop the
aleatory variability model. The results indicated lower values of τ for larger magnitudes at all
periods. Values of σ were lower for larger magnitude at short periods, but become larger at large
magnitudes at long periods.
The fitted values of τ and σ indicate magnitude dependence at most periods. Therefore,
the tri-linear form used in our previous model was applied. Some experimentation indicated that
the appropriate magnitude break points for τ1, τ2, σ1, and σ2 were at M5 and M7.25. Figure 3.20
shows the resulting values of τ1, τ2, σ1, and σ2. It was found that the magnitude dependence of τ
was influenced by the large event term for the M6.61 Tottori, Japan earthquake. Given that the
event term for this earthquake is well above those for all other earthquakes, the aleatory model
was developed without the residuals for this earthquake.
The final functional form for the total standard deviation, σT, is given by Equation (3.13):

38
  T2  (1  NL0 ) 2  2   NL
2
0

 2  1
  1  min max( M ,5),7.25  5
2.25
  2  1
 NL    1  min max( M ,5),7.25  5   3 Finf erred  0.7 FMeasured  (1  NL0 ) 2
0
 2.25 
 3  min(V S 30 ,1130 )  360    y ref 
N L0  2  e 


 e3 (1130  360 ) 
 y  

   ref 4 
(3.13)

Equation (3.13) implemented the approximate method of Chiou and Youngs [2008b]. The
coefficients 2 and 3 are those in Equation (3.12) and their values are listed in Table 3.3.
Because we excluded data from Class 2 earthquakes (aftershocks), we did not include 4
(increase in σ for Class 2 earthquakes). Unlike the median model, we did not account for regional
differences in τ and σ. The values of σ1 and σ2 listed in Table 3.3 are reduced from those shown
on Figure 3.20 to account for the non-linear soil effects.

39
Figure 3.16 Values of τ (top) and σ computed for PGA residuals in magnitude
bins 0.5 magnitude units wide.

40
Figure 3.17 Values of τ (top) and σ computed for 0.17-sec PSA residuals in
magnitude bins 0.5 magnitude units wide.

41
Figure 3.18 Values of τ (top) and σ computed for 1-sec PSA residuals in
magnitude bins 0.5 magnitude units wide.

42
Figure 3.19 Values of τ (top) and σ computed for 3-sec PSA residuals in
magnitude bins 0.5 magnitude units wide.

43
Figure 3.20 Values of τ1, τ2, σ1, and σ2 computed from the residuals excluding
non-linear soil effects.

44
 Table 3.1 Period-independent coefficients of model for ln(y) – Equation (3.11).

c2  c4  c4a  cRB  c8  c8a 

1.06  ‐2.1  ‐0.5  50  0.2153  0.2695 

Table 3.2(a) Period-dependent coefficients of model for ln(yref) – Equation (3.11).

Period 
(sec)  c1  c1a  c1b  c1c  c1d  c n  c M  c3  c5  cHM  c6 
0.01  ‐1.5070  0.1650  ‐0.3729  ‐0.1650  0.1977  16.0875  4.9993  1.9636  6.4551  3.0956  0.4893 
0.02  ‐1.4788  0.1650  ‐0.3772  ‐0.1650  0.2180  15.7118  4.9993  1.9636  6.4551  3.0963  0.4892 
0.03  ‐1.3116  0.1650  ‐0.4429  ‐0.1650  0.3484  15.8819  4.9993  1.9636  6.4551  3.0974  0.4890 
0.04  ‐1.1159  0.1650  ‐0.5122  ‐0.1650  0.4733  16.4556  4.9993  1.9636  6.4551  3.0988  0.4888 
0.05  ‐0.9464  0.1650  ‐0.5544  ‐0.1650  0.5433  17.6453  4.9993  1.9636  6.4551  3.1011  0.4884 
0.075  ‐0.6749  0.1650  ‐0.5929  ‐0.1650  0.5621  20.1772  5.0031  1.9636  6.4551  3.1094  0.4872 
0.1  ‐0.5735  0.1650  ‐0.5760  ‐0.1650  0.4633  19.9992  5.0172  1.9636  6.8305  3.2381  0.4854 
0.12  ‐0.5439  0.1650  ‐0.5583  ‐0.1650  0.4000  18.7106  5.0315  1.9795  7.1333  3.3407  0.4837 
0.15  ‐0.5543  0.1650  ‐0.5345  ‐0.1650  0.3337  16.6246  5.0547  2.0362  7.3621  3.4300  0.4808 
0.17  ‐0.5933  0.1650  ‐0.5188  ‐0.1650  0.2961  15.3709  5.0704  2.0823  7.4365  3.4688  0.4787 
0.2  ‐0.6865  0.1650  ‐0.4944  ‐0.1650  0.2438  13.7012  5.0939  2.1521  7.4972  3.5146  0.4755 
0.25  ‐0.8720  0.1650  ‐0.4517  ‐0.1650  0.1620  11.2667  5.1315  2.2574  7.5416  3.5746  0.4706 
0.3  ‐1.0558  0.1650  ‐0.4122  ‐0.1650  0.0881  9.1908  5.1670  2.3440  7.5600  3.6232  0.4665 
0.4  ‐1.3800  0.1650  ‐0.3532  ‐0.1650  ‐0.0287  6.5459  5.2317  2.4709  7.5735  3.6945  0.4607 
0.5  ‐1.6472  0.1650  ‐0.3101  ‐0.1650  ‐0.1158  5.2305  5.2893  2.5567  7.5778  3.7401  0.4571 
0.75  ‐2.1421  0.1650  ‐0.2219  ‐0.1650  ‐0.2708  3.7896  5.4109  2.6812  7.5808  3.7941  0.4531 
1  ‐2.5266  0.1650  ‐0.1694  ‐0.1650  ‐0.3527  3.3024  5.5106  2.7474  7.5814  3.8144  0.4517 
1.5  ‐3.0655  0.1650  ‐0.1376  ‐0.1650  ‐0.3454  2.8498  5.6705  2.8161  7.5817  3.8284  0.4507 
2  ‐3.4114  0.1645  ‐0.1218  ‐0.1645  ‐0.2605  2.5417  5.7981  2.8514  7.5818  3.8330  0.4504 
3  ‐3.8330  0.1168  ‐0.1053  ‐0.1168  ‐0.0914  2.1488  5.9983  2.8875  7.5818  3.8361  0.4501 
4  ‐4.1207  0.0732  ‐0.1008  ‐0.0732  ‐0.0306  1.8957  6.1552  2.9058  7.5818  3.8369  0.4501 
5  ‐4.3438  0.0484  ‐0.0996  ‐0.0484  ‐0.0129  1.7228  6.2856  2.9169  7.5818  3.8376  0.4500 
7.5  ‐4.7493  0.0220  ‐0.0994  ‐0.0220  ‐0.0014  1.5737  6.5428  2.9320  7.5818  3.8380  0.4500 
10  ‐5.0370  0.0124  ‐0.1000  ‐0.0124  0.0001  1.5265  6.7415  2.9396  7.5818  3.8380  0.4500 

45
Table 3.2(b) (Continued) Period-dependent coefficients of model for ln(yref) – Equation (3.11).

Period 
(sec)  c7  c7b  c8b  c9  c9a  c9b  c11 c11b  c1  c2  c3 
0.01  0.0352  0.0462  0.4833  0.9228  0.1202  6.8607  0.0  ‐0.4536  ‐0.007141  ‐0.006768  4.1293 
0.02  0.0352  0.0472  1.2144  0.9296  0.1217  6.8697  0.0 ‐0.4536  ‐0.007226  ‐0.006929  4.0715 
0.03  0.0352  0.0533  1.6421  0.9396  0.1194  6.9113  0.0 ‐0.4536  ‐0.007520  ‐0.007160  4.0009 
0.04  0.0352  0.0596  1.9456  0.9661  0.1166  7.0271  0.0 ‐0.4536  ‐0.007881  ‐0.007477  3.9647 
0.05  0.0352  0.0639  2.1810  0.9794  0.1176  7.0959  0.0 ‐0.4536  ‐0.008295  ‐0.007460  3.9832 
0.075  0.0352  0.0630  2.6087  1.0260  0.1171  7.3298  0.0 ‐0.4536  ‐0.009266  ‐0.006565  4.1949 
0.1  0.0352  0.0532  2.9122  1.0177  0.1146  7.2588  0.0 ‐0.4536  ‐0.009710  ‐0.005490  4.4844 
0.12  0.0352  0.0452  3.1045  1.0008  0.1128  7.2372  0.0 ‐0.4536  ‐0.009806  ‐0.004840  4.6730 
0.15  0.0352  0.0345  3.3399  0.9801  0.1106  7.2109  0.0 ‐0.4536  ‐0.009805  ‐0.003897  4.8266 
0.17  0.0352  0.0283  3.4719  0.9652  0.1150  7.2491  0.0 ‐0.4536  ‐0.009733  ‐0.003334  4.8670 
0.2  0.0352  0.0202  3.6434  0.9459  0.1208  7.2988  0.0 ‐0.4440  ‐0.009505  ‐0.002690  4.8796 
0.25  0.0352  0.0090  3.8787  0.9196  0.1208  7.3691  0.0 ‐0.3539  ‐0.008918  ‐0.002127  4.8371 
0.3  0.0352  ‐0.0004  4.0711  0.8829  0.1175  6.8789  0.0 ‐0.2688  ‐0.008251  ‐0.001812  4.7579 
0.4  0.0352  ‐0.0155  4.3745  0.8302  0.1060  6.5334  0.0 ‐0.1793  ‐0.007267  ‐0.001274  4.5523 
0.5  0.0352  ‐0.0278  4.6099  0.7884  0.1061  6.5260  0.0 ‐0.1428  ‐0.006492  ‐0.001074  4.4049 
0.75  0.0352  ‐0.0477  5.0376  0.6754  0.1000  6.5000  0.0 ‐0.1138  ‐0.005147  ‐0.001115  4.2079 
1  0.0352  ‐0.0559  5.3411  0.6196  0.1000  6.5000  0.0 ‐0.1062  ‐0.004277  ‐0.001197  4.0126 
1.5  0.0352  ‐0.0630  5.7688  0.5101  0.1000  6.5000  0.0 ‐0.1020  ‐0.002979  ‐0.001671  3.6733 
2  0.0352  ‐0.0665  6.0723  0.3917  0.1000  6.5000  0.0 ‐0.1009  ‐0.002301  ‐0.002237  3.5073 
3  0.0160  ‐0.0516  6.5000  0.1244  0.1000  6.5000  0.0 ‐0.1003  ‐0.001344  ‐0.003108  3.4748 
4  0.0062  ‐0.0448  6.8035  0.0086  0.1000  6.5000  0.0 ‐0.1001  ‐0.001084  ‐0.003670  3.5179 
5  0.0029  ‐0.0424  7.0389  0.0000  0.1000  6.5000  0.0 ‐0.1001  ‐0.000994  ‐0.003767  3.5479 
7.5  0.0007  ‐0.0348  7.4666  0.0000  0.1000  6.5000  0.0 ‐0.1000  ‐0.000964  ‐0.003784  3.5954 
10  0.0003  ‐0.0253  7.7700  0.0000  0.1000  6.5000  0.0 ‐0.1000  ‐0.000950  ‐0.003747  3.6253 

46
Table 3.3 Coefficients of site response model for ln(y) – Equation (3.12).

Period  
(sec)  1  2  3  4  5  6 
0.01  ‐0.5210  ‐0.1417  ‐0.007010  0.102151  0.0000  300 
0.02  ‐0.5055  ‐0.1364  ‐0.007279  0.108360  0.0000  300 
0.03  ‐0.4368  ‐0.1403  ‐0.007354  0.119888  0.0000  300 
0.04  ‐0.3752  ‐0.1591  ‐0.006977  0.133641  0.0000  300 
0.05  ‐0.3469  ‐0.1862  ‐0.006467  0.148927  0.0000  300 
0.075  ‐0.3747  ‐0.2538  ‐0.005734  0.190596  0.0000  300 
0.1  ‐0.4440  ‐0.2943  ‐0.005604  0.230662  0.0000  300 
0.12  ‐0.4895  ‐0.3077  ‐0.005696  0.253169  0.0000  300 
0.15  ‐0.5477  ‐0.3113  ‐0.005845  0.266468  0.0000  300 
0.17  ‐0.5922  ‐0.3062  ‐0.005959  0.265060  0.0000  300 
0.2  ‐0.6693  ‐0.2927  ‐0.006141  0.255253  0.0000  300 
0.25  ‐0.7766  ‐0.2662  ‐0.006439  0.231541  0.0000  300 
0.3  ‐0.8501  ‐0.2405  ‐0.006704  0.207277  0.0010  300 
0.4  ‐0.9431  ‐0.1975  ‐0.007125  0.165464  0.0040  300 
0.5  ‐1.0044  ‐0.1633  ‐0.007435  0.133828  0.0100  300 
0.75  ‐1.0602  ‐0.1028  ‐0.008120  0.085153  0.0340  300 
1  ‐1.0941  ‐0.0699  ‐0.008444  0.058595  0.0670  300 
1.5  ‐1.1142  ‐0.0425  ‐0.007707  0.031787  0.1430  300 
2  ‐1.1154  ‐0.0302  ‐0.004792  0.019716  0.2030  300 
3  ‐1.1081  ‐0.0129  ‐0.001828  0.009643  0.2770  300 
4  ‐1.0603  ‐0.0016  ‐0.001523  0.005379  0.3090  300 
5  ‐0.9872  0.0000  ‐0.001440  0.003223  0.3210  300 
7.5  ‐0.8274  0.0000  ‐0.001369  0.001134  0.3290  300 
10  ‐0.7053  0.0000  ‐0.001361  0.000515  0.3300  300 

47
 Table 3.4 Coefficients of variance model – Equation (3.13).

Period  
(sec)  1  2  1  2  3 
0.01  0.4010  0.2470  0.5530  0.4106  0.8000 
0.02  0.4080  0.2420  0.5567  0.4107  0.8000 
0.03  0.4320  0.2320  0.5676  0.4136  0.8000 
0.04  0.4460  0.2460  0.5816  0.4197  0.8000 
0.05  0.4670  0.2750  0.5933  0.4265  0.8000 
0.075  0.4860  0.4190  0.5976  0.4544  0.8000 
0.1  0.4680  0.4270  0.5868  0.4761  0.8000 
0.12  0.4530  0.3840  0.5892  0.4795  0.8000 
0.15  0.4170  0.2970  0.5892  0.4798  0.8000 
0.17  0.4030  0.2560  0.5877  0.4839  0.8000 
0.2  0.3860  0.2310  0.5799  0.4848  0.8000 
0.25  0.3690  0.2070  0.5643  0.4796  0.7999 
0.3  0.3680  0.1980  0.5525  0.4813  0.7997 
0.4  0.3760  0.1700  0.5338  0.4824  0.7988 
0.5  0.3750  0.1710  0.5228  0.4837  0.7966 
0.75  0.3880  0.1780  0.4870  0.5089  0.7792 
1  0.4340  0.2330  0.4699  0.5158  0.7504 
1.5  0.4960  0.2180  0.4452  0.5326  0.7136 
2  0.5110  0.2310  0.4221  0.5335  0.7035 
3  0.4770  0.2750  0.4135  0.4866  0.7006 
4  0.5160  0.3300  0.3982  0.4769  0.7001 
5  0.5080  0.3370  0.3857  0.4552  0.7000 
7.5  0.4340  0.3290  0.3641  0.4467  0.7000 
10  0.3320  0.4040  0.3495  0.3973  0.7000 

48
4 Evaluation of Updated Model

4.1 BETWEEN-EVENT RESIDUALS

Figure 4.1 shows the between-event residuals (event term), for 4 spectral periods of T=0.01
(PGA), 0.2, 1, and 3 sec. In the range 3.5 ≤ M ≤ 8, they do not exhibit a significant trend with M
or a large offset from 0. Our updated model has the tendency to under predict T=3 sec data in
the range of M < 3.5, as evidenced by the positive event terms. There are outliers at large
magnitude exhibiting large (> 2 ) absolute event terms at different periods. They are the 1999
M7.6 ChiChi earthquake (for PGA and T=0.2 sec), the 2000 M6.6 Tottori earthquake (for PGA,
T=0.2 and 3 sec), and the 2008 M7.9 Wenchuan, China, earthquake (for T=1 and 3 sec). All
three earthquakes occurred outside California.

49
 Figure 4.1 Between-event residuals (event terms) for spectral periods of 0.01
(PGA), 0.2, 1, and 3 sec.

4.2 WITHIN-EVENT RESIDUALS

Figures 4.2–4.5 show the within-event residuals plotted versus M, RRUP, VS30, and ΔZ1.0 for
spectral periods of 0.01 (PGA), 0.2, 1 and 3 sec, respectively. In general, these residuals do not
exhibit significant trends within the body of a predictor, but several trends are noted near the
ends of predictor domain. We assumed no site response relative to yref for VS30 greater than 1130
m/sec. Although a limited number of such data are in the PEER-NGA database, their residuals, if
anything, show a slight downward trend for 0.01 and 0.2 sec, and a upward trend for T=3 sec.
Non-California data are responsible for the downward trend for 0.2 sec at VS30 < 180 m/sec. It
suggests there may be a regional difference in VS30 scaling, possibly in the nonlinear part of the
soil response.

50
Figure 4.2 Within-event residuals for spectral period of 0.01 sec (PGA) plotted
against M, RRUP, VS30, and Z1.0.

51
Figure 4.3 Within-event residuals for spectral period of 0.2 sec plotted against
M, RRUP, VS30, and Z1.0.

52
Figure 4.4 Within-event residuals for spectral period of 1 sec plotted against
M, RRUP, VS30, and Z1.0.

53
 Figure 4.5 Within-event residuals for spectral period of 3 sec plotted against
M, RRUP, VS30, and Z1.0.

4.3 SOIL NONLINEARITY

As discussed in Section 3.2.3, we retained both the formulation and model coefficients (2,  3,
and  4) for nonlinear soil response. To provide a visual validation of this decision, we present
Figure 4.6. Residuals in the figure are within-event residuals for T=0.2 sec computed for our
updated model without the effects of VS30 (i.e., for a VS30 of 1130 m/sec). Residuals computed
this way can be regarded as soil amplification relative to the event-specific median motion on
reference rock condition. The residuals are grouped by the level of reference motion
( y ref exp( ) ) and plotted against VS30 in Figure 4.6 to show how VS30 trend varies with the input
reference motion. The soil amplification predicted by our updated model is also shown in the
figure as the thick solid (dark red) curve. The predicted curve tracks closely the VS30 trend of the
residuals (shown as the solid orange curve) for VS30 > 200 m/sec, confirming the general validity
of our nonlinear soil response model. For VS30 < 200 m/sec, our model over estimates the
empirical amplification; in other words, the nonlinearity in the data on very soft soil is stronger
than what is predicted by our model.

54
Figure 4.6(a) Intra-event residuals plotted as a function of VS30. These residuals
are computed for the updated model for the VS30=1130 m/sec
condition. Data from M > 6 earthquake are shown in black; data
from M < 6 are in blue. The smoothed VS30 trend of these residuals is
shown as the thick orange curve. The thick solid red curve is the
predicted nonlinear soil amplification by our updated model and the
thin read solid line is the predicted linear amplification. The range
of event-specific median motion on reference condition is shown in
the lower left corner of each plot.

55
 Figure 4.6(b) (continued).

4.4 COMPARISON WITH THE 2008 CHOIU AND YOUNGS GMPE

In 2008, we used Sadigh et al. [1997] as the baseline for comparison. In this update, our 2008
model (CY2008) becomes the new baseline to which we now compare the updated model
(CY2013).

56
4.4.1 Comparison of Median Motion

Comparisons of median prediction between CY2008 and CY2013 are shown in Figures 4.7 and
4.8. For comparison plots, we use VS30 = 760 m/sec and DPP = 0, for average directivity. We
set ZTOR and Z1.0 to 0 for predictions by CY2013. Equivalently, for predictions by CY2008 we
use average ZTOR for the given M (Equation (2.4) if reverse earthquake, Equation (2.5) if strike-
slip earthquake) and average Z1.0 for the given VS30 [Equation (2.1)].
As discussed in Chapter 3, we made modifications/additions to the 2008 formulation and
revised a large number of model coefficients. The unmodified coefficients were found to be still
valid, as discussed previously. Many of these model changes affect only ground motion
prediction for SMM earthquakes (M < 5.5). As a result, large discrepancy in predictions between
CY2008 and CY2013 occurs mostly at M < 5.5 (Figure 4.7 and Figure 4.8). At M > 6.0, the
typical discrepancy is smaller than 20%.
Figure 4.9 shows comparisons of predicted HW motions for a reverse earthquake of 45⁰
dip. For this example, we set ZTOR to zero for both models to show the HW effect of a surface
rupture event. Predicted HW amplification by CY2013 is smaller and shows stronger M scaling
than what were predicted by CY2008, as discussed in Section 3.1.4.
Figure 4.10 and Figure 4.11 show comparisons of predicted median response spectra for
VS30 =760 m/sec and VS30 =310 m/sec, respectively.

57
Figure 4.7(a) Predicted median amplitude vs. magnitude for vertical strike slip
earthquake (ZTOR=0, Z1.0= 0, and DPP=0).

58
Figure 4.7(b) (continued).

59
Figure 4.8(a) Predicted median amplitude vs. distance for vertical strike slip
earthquake (ZTOR=0, Z1.0= 0, and DPP=0).

60
Figure 4.8(b) (continued).

61
Figure 4.9(a) Predicted median amplitude vs. distance for reverse slip
earthquake of 45 dip. Note that this comparison is for surface
rupture event, ZTOR=0. Z1.0 and DPP are equal to 0.

62
Figure 4.9(b) (continued).

63
Figure 4.10 Median response spectra predicted by the 2008 Chiou and Youngs
NGA model (thin lines) and the updated model (thick lines).
Predictions are made for vertical strike-slip earthquakes and
NEHRP B-C boundary (VS30=760 m/sec).

64
Figure 4.11 Median response spectra predicted by the 2008 Chiou and Youngs
NGA model thin lines) and the updated model (thick lines).
Predictions are made for vertical strike-slip earthquakes and firm
soil condition (VS30=310 m/sec).

65
4.4.2 Comparison of Aleatory Variability

Figure 4.12 compares the total standard deviations for the updated model under linear soil
response with those for CY2008 model. Figure 4.13 shows the effect of soil nonlinearity on the
total standard deviations of PGA.

Figure 4.12 Total standard deviations for the updated model with effects for
linear soil response.

66
Figure 4.13 Total standard deviations for the updated model with effects of
nonlinear soil response.

67
68
5 Example Calculations

The updated GMPE is implemented in the FORTRAN routine CY13.FOR. This program is
included in the companion package. Also in the package are the inputs and outputs for four
example scenarios: M5 and M7 strike-slip earthquakes and M5 and M7 reverse-faulting
earthquakes. The required input variables are indicated by the header record in the example input
files. The routine accepts its main input and writes the output to the console. After invoking at
the command prompt, the routine loops over prompts for the input and output files.

69
70
6 Model Applicability

The GMPE update developed in this study is considered to be applicable for estimating pseudo-
spectral accelerations (5% of critical damping) and peak motions for earthquakes in active
tectonic regions in which the following conditions apply:
 3.5 ≤ M ≤ 8.5 for strike-slip earthquakes
 3.5 ≤ M ≤ 8.0 for reverse and normal faulting earthquakes
 ZTOR ≤ 20km
 0 ≤ RRUP ≤ 300 km
 180 m/sec ≤ VS30 ≤ 1500 m/sec
Compared to Chiou and Youngs [2008a], the lower bound of the applicable magnitude
range was decreased to 3.5 because of the large number of SMM earthquakes in our regression
dataset. However, because all SMM data were from California, our GMPE may not be applicable
to SMM earthquakes in other active tectonic regions.
The upper bound of the applicable distance was increased to 300 km because of the use
of extensive data at distances from 200 to 300 km (Figure 2.1). For application in other active
tectonic regions where earthquakes at distances greater than about 50 km are a major contributor
to the hazard, adjustments to the γ(M) coefficients cγ1 and cγ2 may be warranted. These
adjustments can be made using the hybrid approach developed by Campbell [2003]. In making
such adjustments, we stress the need for the user to obtain estimates of Q for the two regions that
are based on geometric spreading models at large distances that are consistent with the one used
in this study.
The site response portion of the ground motion model was constrained such that all
ground motion amplification factors are 1 for VS30 greater than 1130 m/sec. As the rock velocity
increases we expect shallow crustal damping (i.e., “kappa”) to decrease, resulting in increases in
high-frequency motion. Data for such sites are not sampled in the NGA-West2 database in
sufficient quantity to estimate this effect, and it is not captured in our model. Such effects should
be considered if the model is to be applied to sites with VS30 greater than 1500 m/sec.
We increased the lower bound of the VS30 applicable range because of the residual trend
observed on Figure 4.3.
The updated model was developed using recordings from earthquakes with a maximum
ZTOR of 20 km. Furthermore, the ZTOR – M data shown in Figure 2.5 suggest that the applicable

71
range of ZTOR should be decreasing with M. We do not recommend using large ZTOR for M > 7
as such events are not well presented in the NGA-West2 database.
The ground motion model presented here is sensitive to the value of Z1.0. Majority of
data used in our updated model were from the southern California, the San Francisco Bay area,
and Japan. When applying our model to these regions, the same 3D velocity models should be
used to obtain site Z1.0. For application to a site not covered by these velocity models and there
are no other information to determine the site Z1.0, it is suggested that the user use Z1.0 = 0 to
predict the median amplitude for sites of the same VS30. When applying our GMPE to a site
whose Z1.0 is much smaller than the average Z1.0 (a large negative Z1.0), the prediction should be
checked to ensure that the predicted motion (particularly for long periods) is not lower than the
predicted median for reference rock condition (VS30=1130).

72
 REFERENCES

Abrahamson N.A., Silva W.J. (2008). Summary of the Abrahamson and Silva NGA ground-motion relations,
Earthq. Spectra 24: 67–97.
Ancheta T.D., Darragh R.B., Stewart J.P., Seyhan E., Silva W.J., Chiou B.S-J., Wooddell K.E., Graves R.W., Kottke
A.R., Boore D.M., Kishida T., Donahue J.L. (2013). PEER NGA-West2 database, PEER Report 2013/03,
Pacific Earthquake Engineering Research Center, University of California, Berkeley, CA
Boatwright J. (2007). The persistence of directivity in small earthquakes, Bull. Seismol. Soc. Am. 97(6): 1850–1861,
doi: 10.1785/0120050228.
Boore D.M. (2010). Orientation-independent, nongeometric-mean measures of seismic intensity from two horizontal
components of motion, Bull. Seismol. Soc. Am., 100(4): 1830–1835.
Bragato P.L. (2004). Regression analysis with truncated samples and its application to ground-motion attenuation
studies, Bull. Seismol. Soc. Am., 94: 1369–1378.
Campbell K.W. (2003). Prediction of strong ground motion using the hybrid empirical method and its use in the
development of ground motion (attenuation) relations in eastern North America, Bull. Seismol. Soc. Am., 93:
1012–1033.
Chiou B.S.-J., Spudich P. (2013). The Chiou and Spudich NGA-West2 directivity predictor DPP, PEER Report
2013/09, Pacific Earthquake Engineering Research Center, University of California, Berkeley, CA
Chiou, B.S.-J., Youngs R.R. (2008a). An NGA model for the average horizontal component of peak ground motion
and response spectra, Earthq. Spectra, 24: 173–216.
Chiou B.S.-J., Youngs R.R. (2008b). NGA model for the average horizontal component of peak ground motion and
response spectra, PEER Report 2008/09, Pacific Earthquake Engineering Research Center, University of
California, Berkeley, CA
Chiou B.S.-J., Youngs R.R. (2012). Updating the Chiou and Youngs NGA model: regionalization of anelastic
attenuation, Proceedings, 15th World Conference on Earthquake Engineering, Lisbon, Portugal.
Chiou B.S.-J., Youngs R.R., Abrahamson N.A, Addo K. (2010). Ground-motion attenuation model for small-to-
moderate shallow crustal earthquakes in California and its implications on regionalization of ground-motion
prediction models, Earthq. Spectra, 26: 907–926.
Donahue J.L., Abrahamson N.A. (2013). Hanging wall scaling using finite-fault simulations, PEER Report 2013/14,
Pacific Earthquake Engineering Research Center, University of California, Berkeley, CA.
Geomatrix Consultants (1995). Seismic Design Mapping State of Oregon, report prepared for the Oregon
Department of Transportation, Personal Services Contract 11688, Geomatrix Consultants Project No. 2442,
January.
Kaklamanos J., Baise L.G., Boore D.M. (2011). Estimating unknown input parameters when implementing the NGA
ground-motion prediction equations in engineering practice, Earthq. Spectra 27: 1219–1235.
Oppenheimer D., Beroza G., Carver G., Dengler L., Eaton J., Gee L., Gonzalez F., Jayko A., Li W.H., Lisowski M.,
Magee M., Marshall G., Murray M., McPherson R., Romanowicz B., Satake K., Simpson R., Somerville P.,
Stein R., Valentine, D. (1993). The Cape Mendocino, California earthquakes of April 1992: Subduction at the
Triple Junction, Science, 261: 433–438.
R Development Core Team (2009). R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria, ISBN 3-900051-07-0, URL http://www.R-project.org.
Sadigh K., Chang C.-Y., Egan J.A., Makdisi F.I., Youngs, R.R. (1997). Attenuation relationships for shallow crustal
earthquakes based on California strong motion data, Seismol. Res. Lett. 68: 180–189.

73
Spudich P., Bayless J.R., Baker J.W., Chiou B.S-J., Rowshandel B., Shahi S., Somerville P. (2013). Final report of
the NGA-West2 directivity working group, PEER report 2013/09, Pacific Earthquake Engineering Research
Center, University of California, Berkeley, CA.
Spudich P., Chiou, B.S.-J. (2008). Directivity in NGA earthquake ground motions: analysis using isochrone theory,
Earthq. Spectra, 24: 279–298.
Toro G.R. (1981). Biases in Seismic Ground Motion Prediction, Ph.D. Thesis, Dept. of Civil Engineering,
Massachusetts Institute of Technology, 133 pgs.
Wooddell K.E., Abrahamson N.A. (2012). New earthquake classification scheme for mainshocks and aftershocks in
the NGA-West2 ground motion prediction equations, Proceedings, 15th World Conference on Earthquake
Engineering, Lisbon, Portugal.

74
Appendix A: Model Coefficients for Non-
California Regions

In this appendix we provide the estimated model coefficients obtained as part of the model
update to account for known regional difference in anelastic attenuation and site effects. These
coefficients are:
Jp-It :  adjustment factor for Japanese and Italian data.
Note that, since the Japanese and Italian events used in our update are of 6.0 < M < 6.9,
this factor is applicable only for that M range.
Wn :  adjustment factor for the M 7.9, 2008 Wenchuan earthquake data.
1Jp : 1 for Japanese data.
5Jp : 5 for Japanese data.
6Jp : 6 for Japanese data.

75
 Table A.1 Model coefficients for non-California regions.

Period  
(sec)  Jp-It  Wn  1Jp  5JP  6JP 
0.01  1.5817  0.7594  ‐0.6846  0.4590  800 
0.02  1.5740  0.7606  ‐0.6681  0.4580  800 
0.03  1.5544  0.7642  ‐0.6314  0.4620  800 
0.04  1.5502  0.7676  ‐0.5855  0.4530  800 
0.05  1.5391  0.7739  ‐0.5457  0.4360  800 
0.075  1.4804  0.7956  ‐0.4685  0.3830  800 
0.1  1.4094  0.7932  ‐0.4985  0.3750  800 
0.12  1.3682  0.7768  ‐0.5603  0.3770  800 
0.15  1.3241  0.7437  ‐0.6451  0.3790  800 
0.17  1.3071  0.7219  ‐0.6981  0.3800  800 
0.2  1.2931  0.6922  ‐0.7653  0.3840  800 
0.25  1.3150  0.6579  ‐0.8469  0.3930  800 
0.3  1.3514  0.6362  ‐0.8999  0.4080  800 
0.4  1.4051  0.6049  ‐0.9618  0.4620  800 
0.5  1.4402  0.5507  ‐0.9945  0.5240  800 
0.75  1.5280  0.3582  ‐1.0225  0.6580  800 
1  1.6523  0.2003  ‐1.0002  0.7800  800 
1.5  1.8872  0.0356  ‐0.9245  0.9600  800 
2  2.1348  0.0000  ‐0.8626  1.1100  800 
3  3.5752  0.0000  ‐0.7882  1.2910  800 
4  3.8646  0.0000  ‐0.7195  1.3870  800 
5  3.7292  0.0000  ‐0.6560  1.4330  800 
7.5  2.3763  0.0000  ‐0.5202  1.4600  800 
10  1.7679  0.0031  ‐0.4068  1.4640  800 

76
77
78
 PEER REPORTS

PEER reports are available as a free PDF download from http://peer.berkeley.edu/publications/peer_reports_complete.html. Printed
hard copies of PEER reports can be ordered directly from our printer by following the instructions at
http://peer.berkeley.edu/publications/peer_reports.html. For other related questions about the PEER Report Series, contact the
Pacific Earthquake Engineering Research Center, 325 Davis Hall mail code 1792, Berkeley, CA 94720. Tel.: (510) 642-3437; Fax:
(510) 665-1655; Email: [email protected]

PEER 2013/07 Update of the Chiou and Youngs NGA Ground Motion Model for Average Horizontal Component of Peak Ground
Motion and Response Spectra. Brian S.J. Chiou and Robert R. Youngs. May 2013.

PEER 2013/06 NGA-West2 Campbell-Bozorgnia Ground Motion Model for the Horizontal Components of PGA, PGV, and 5%-
Damped Elastic Pseudo-Acceleration Response Spectra for Periods Ranging from 0.01 to 10 sec. Kenneth W.
Campbell and Yousef Bozorgnia. May 2013.

PEER 2013/05 NGA-West2 Equations for Predicting Response Spectral Accelerations for Shallow Crustal Earthquakes. David M.
Boore, Jonathan P. Stewart, Emel Seyhan, Gail M. Atkinson. May 2013.

PEER 2013/04 Update of the AS08 Ground-Motion Prediction Equations Based on the NGA-West2 Data Set. Norman SA.
Abrahamson, Walter J. Silva, and Ronnie Kamai. May 2013.

PEER 2013/03 PEER NGA-West2 Database. Timothy D. Ancheta, Robert B. Darragh, Jonathan P. Stewart, Emel Seyhan, Walter
J. Silva, Brian S.J. Chiou, Katie E. Wooddell, Robert W. Graves, Albert R. Kottke, David M. Boore, Tadahiro
Kishida, and Jennifer L. Donahue. May 2013.

PEER 2013/02 Hybrid Simulation of the Seismic Response of Squat Reinforced Concrete Shear Walls. Catherine A. Whyte and
Bozidar Stojadinovic. May 2013.

PEER 2013/01 Housing Recovery in Chile: A Qualitative Mid-program Review. Mary C. Comerio. February 2013.

PEER 2012/08 Guidelines for Estimation of Shear Wave Velocity. Bernard R. Wair, Jason T. DeJong, and Thomas Shantz.
December 2012.

PEER 2012/07 Earthquake Engineering for Resilient Communities: 2012 PEER Internship Program Research Report Collection.
Heidi Tremayne (Editor), Stephen A. Mahin (Editor), Collin Anderson, Dustin Cook, Michael Erceg, Carlos
Esparza, Jose Jimenez, Dorian Krausz, Andrew Lo, Stephanie Lopez, Nicole McCurdy, Paul Shipman, Alexander
Strum, Eduardo Vega. December 2012.

PEER 2012/06 Fragilities for Precarious Rocks at Yucca Mountain. Matthew D. Purvance, Rasool Anooshehpoor, and James N.
Brune. December 2012.

PEER 2012/05 Development of Simplified Analysis Procedure for Piles in Laterally Spreading Layered Soils. Christopher R.
McGann, Pedro Arduino, and Peter Mackenzie–Helnwein. December 2012.

PEER 2012/04 Unbonded Pre-Tensioned Columns for Bridges in Seismic Regions. Phillip M. Davis, Todd M. Janes, Marc O.
Eberhard, and John F. Stanton. December 2012.

PEER 2012/03 Experimental and Analytical Studies on Reinforced Concrete Buildings with Seismically Vulnerable Beam-Column
Joints. Sangjoon Park and Khalid M. Mosalam. October 2012.

PEER 2012/02 Seismic Performance of Reinforced Concrete Bridges Allowed to Uplift during Multi-Directional Excitation. Andres
Oscar Espinoza and Stephen A. Mahin. July 2012.

PEER 2012/01 Spectral Damping Scaling Factors for Shallow Crustal Earthquakes in Active Tectonic Regions. Sanaz Rezaeian,
Yousef Bozorgnia, I. M. Idriss, Kenneth Campbell, Norman Abrahamson, and Walter Silva. July 2012.

PEER 2011/10 Earthquake Engineering for Resilient Communities: 2011 PEER Internship Program Research Report Collection.
Eds. Heidi Faison and Stephen A. Mahin. December 2011.

PEER 2011/09 Calibration of Semi-Stochastic Procedure for Simulating High-Frequency Ground Motions. Jonathan P. Stewart,
Emel Seyhan, and Robert W. Graves. December 2011.

PEER 2011/08 Water Supply in regard to Fire Following Earthquake. Charles Scawthorn. November 2011.

PEER 2011/07 Seismic Risk Management in Urban Areas. Proceedings of a U.S.-Iran-Turkey Seismic Workshop. September
2011.

PEER 2011/06 The Use of Base Isolation Systems to Achieve Complex Seismic Performance Objectives. Troy A. Morgan and
Stephen A. Mahin. July 2011.
PEER 2011/05 Case Studies of the Seismic Performance of Tall Buildings Designed by Alternative Means. Task 12 Report for
the Tall Buildings Initiative. Jack Moehle, Yousef Bozorgnia, Nirmal Jayaram, Pierson Jones, Mohsen Rahnama,
Nilesh Shome, Zeynep Tuna, John Wallace, Tony Yang, and Farzin Zareian. July 2011.

PEER 2011/04 Recommended Design Practice for Pile Foundations in Laterally Spreading Ground. Scott A. Ashford, Ross W.
Boulanger, and Scott J. Brandenberg. June 2011.

PEER 2011/03 New Ground Motion Selection Procedures and Selected Motions for the PEER Transportation Research Program.
Jack W. Baker, Ting Lin, Shrey K. Shahi, and Nirmal Jayaram. March 2011.

PEER 2011/02 A Bayesian Network Methodology for Infrastructure Seismic Risk Assessment and Decision Support. Michelle T.
Bensi, Armen Der Kiureghian, and Daniel Straub. March 2011.

PEER 2011/01 Demand Fragility Surfaces for Bridges in Liquefied and Laterally Spreading Ground. Scott J. Brandenberg, Jian
Zhang, Pirooz Kashighandi, Yili Huo, and Minxing Zhao. March 2011.

PEER 2010/05 Guidelines for Performance-Based Seismic Design of Tall Buildings. Developed by the Tall Buildings Initiative.
November 2010.
PEER 2010/04 Application Guide for the Design of Flexible and Rigid Bus Connections between Substation Equipment Subjected
to Earthquakes. Jean-Bernard Dastous and Armen Der Kiureghian. September 2010.
PEER 2010/03 Shear Wave Velocity as a Statistical Function of Standard Penetration Test Resistance and Vertical Effective
Stress at Caltrans Bridge Sites. Scott J. Brandenberg, Naresh Bellana, and Thomas Shantz. June 2010.
PEER 2010/02 Stochastic Modeling and Simulation of Ground Motions for Performance-Based Earthquake Engineering. Sanaz
Rezaeian and Armen Der Kiureghian. June 2010.
PEER 2010/01 Structural Response and Cost Characterization of Bridge Construction Using Seismic Performance Enhancement
Strategies. Ady Aviram, Božidar Stojadinović, Gustavo J. Parra-Montesinos, and Kevin R. Mackie. March 2010.
PEER 2009/03 The Integration of Experimental and Simulation Data in the Study of Reinforced Concrete Bridge Systems
Including Soil-Foundation-Structure Interaction. Matthew Dryden and Gregory L. Fenves. November 2009.

PEER 2009/02 Improving Earthquake Mitigation through Innovations and Applications in Seismic Science, Engineering,
Communication, and Response. Proceedings of a U.S.-Iran Seismic Workshop. October 2009.

PEER 2009/01 Evaluation of Ground Motion Selection and Modification Methods: Predicting Median Interstory Drift Response of
Buildings. Curt B. Haselton, Ed. June 2009.

PEER 2008/10 Technical Manual for Strata. Albert R. Kottke and Ellen M. Rathje. February 2009.

PEER 2008/09 NGA Model for Average Horizontal Component of Peak Ground Motion and Response Spectra. Brian S.-J. Chiou
and Robert R. Youngs. November 2008.

PEER 2008/08 Toward Earthquake-Resistant Design of Concentrically Braced Steel Structures. Patxi Uriz and Stephen A. Mahin.
November 2008.

PEER 2008/07 Using OpenSees for Performance-Based Evaluation of Bridges on Liquefiable Soils. Stephen L. Kramer, Pedro
Arduino, and HyungSuk Shin. November 2008.

PEER 2008/06 Shaking Table Tests and Numerical Investigation of Self-Centering Reinforced Concrete Bridge Columns. Hyung
IL Jeong, Junichi Sakai, and Stephen A. Mahin. September 2008.

PEER 2008/05 Performance-Based Earthquake Engineering Design Evaluation Procedure for Bridge Foundations Undergoing
Liquefaction-Induced Lateral Ground Displacement. Christian A. Ledezma and Jonathan D. Bray. August 2008.

PEER 2008/04 Benchmarking of Nonlinear Geotechnical Ground Response Analysis Procedures. Jonathan P. Stewart, Annie
On-Lei Kwok, Yousseff M. A. Hashash, Neven Matasovic, Robert Pyke, Zhiliang Wang, and Zhaohui Yang.
August 2008.

PEER 2008/03 Guidelines for Nonlinear Analysis of Bridge Structures in California. Ady Aviram, Kevin R. Mackie, and Božidar
Stojadinović. August 2008.

PEER 2008/02 Treatment of Uncertainties in Seismic-Risk Analysis of Transportation Systems. Evangelos Stergiou and Anne S.
Kiremidjian. July 2008.

PEER 2008/01 Seismic Performance Objectives for Tall Buildings. William T. Holmes, Charles Kircher, William Petak, and Nabih
Youssef. August 2008.

PEER 2007/12 An Assessment to Benchmark the Seismic Performance of a Code-Conforming Reinforced Concrete Moment-
Frame Building. Curt Haselton, Christine A. Goulet, Judith Mitrani-Reiser, James L. Beck, Gregory G. Deierlein,
Keith A. Porter, Jonathan P. Stewart, and Ertugrul Taciroglu. August 2008.

PEER 2007/11 Bar Buckling in Reinforced Concrete Bridge Columns. Wayne A. Brown, Dawn E. Lehman, and John F. Stanton.
February 2008.
PEER 2007/10 Computational Modeling of Progressive Collapse in Reinforced Concrete Frame Structures. Mohamed M. Talaat
and Khalid M. Mosalam. May 2008.

PEER 2007/09 Integrated Probabilistic Performance-Based Evaluation of Benchmark Reinforced Concrete Bridges. Kevin R.
Mackie, John-Michael Wong, and Božidar Stojadinović. January 2008.

PEER 2007/08 Assessing Seismic Collapse Safety of Modern Reinforced Concrete Moment-Frame Buildings. Curt B. Haselton
and Gregory G. Deierlein. February 2008.

PEER 2007/07 Performance Modeling Strategies for Modern Reinforced Concrete Bridge Columns. Michael P. Berry and Marc
O. Eberhard. April 2008.

PEER 2007/06 Development of Improved Procedures for Seismic Design of Buried and Partially Buried Structures. Linda Al Atik
and Nicholas Sitar. June 2007.

PEER 2007/05 Uncertainty and Correlation in Seismic Risk Assessment of Transportation Systems. Renee G. Lee and Anne S.
Kiremidjian. July 2007.

PEER 2007/04 Numerical Models for Analysis and Performance-Based Design of Shallow Foundations Subjected to Seismic
Loading. Sivapalan Gajan, Tara C. Hutchinson, Bruce L. Kutter, Prishati Raychowdhury, José A. Ugalde, and
Jonathan P. Stewart. May 2008.

PEER 2007/03 Beam-Column Element Model Calibrated for Predicting Flexural Response Leading to Global Collapse of RC
Frame Buildings. Curt B. Haselton, Abbie B. Liel, Sarah Taylor Lange, and Gregory G. Deierlein. May 2008.

PEER 2007/02 Campbell-Bozorgnia NGA Ground Motion Relations for the Geometric Mean Horizontal Component of Peak and
Spectral Ground Motion Parameters. Kenneth W. Campbell and Yousef Bozorgnia. May 2007.

PEER 2007/01 Boore-Atkinson NGA Ground Motion Relations for the Geometric Mean Horizontal Component of Peak and
Spectral Ground Motion Parameters. David M. Boore and Gail M. Atkinson. May. May 2007.

PEER 2006/12 Societal Implications of Performance-Based Earthquake Engineering. Peter J. May. May 2007.

PEER 2006/11 Probabilistic Seismic Demand Analysis Using Advanced Ground Motion Intensity Measures, Attenuation
Relationships, and Near-Fault Effects. Polsak Tothong and C. Allin Cornell. March 2007.

PEER 2006/10 Application of the PEER PBEE Methodology to the I-880 Viaduct. Sashi Kunnath. February 2007.

PEER 2006/09 Quantifying Economic Losses from Travel Forgone Following a Large Metropolitan Earthquake. James Moore,
Sungbin Cho, Yue Yue Fan, and Stuart Werner. November 2006.

PEER 2006/08 Vector-Valued Ground Motion Intensity Measures for Probabilistic Seismic Demand Analysis. Jack W. Baker and
C. Allin Cornell. October 2006.

PEER 2006/07 Analytical Modeling of Reinforced Concrete Walls for Predicting Flexural and Coupled–Shear-
Flexural Responses. Kutay Orakcal, Leonardo M. Massone, and John W. Wallace. October 2006.

PEER 2006/06 Nonlinear Analysis of a Soil-Drilled Pier System under Static and Dynamic Axial Loading. Gang Wang and
Nicholas Sitar. November 2006.

PEER 2006/05 Advanced Seismic Assessment Guidelines. Paolo Bazzurro, C. Allin Cornell, Charles Menun, Maziar Motahari,
and Nicolas Luco. September 2006.

PEER 2006/04 Probabilistic Seismic Evaluation of Reinforced Concrete Structural Components and Systems. Tae Hyung Lee
and Khalid M. Mosalam. August 2006.

PEER 2006/03 Performance of Lifelines Subjected to Lateral Spreading. Scott A. Ashford and Teerawut Juirnarongrit. July 2006.

PEER 2006/02 Pacific Earthquake Engineering Research Center Highway Demonstration Project. Anne Kiremidjian, James
Moore, Yue Yue Fan, Nesrin Basoz, Ozgur Yazali, and Meredith Williams. April 2006.

PEER 2006/01 Bracing Berkeley. A Guide to Seismic Safety on the UC Berkeley Campus. Mary C. Comerio, Stephen Tobriner,
and Ariane Fehrenkamp. January 2006.

PEER 2005/16 Seismic Response and Reliability of Electrical Substation Equipment and Systems. Junho Song, Armen Der
Kiureghian, and Jerome L. Sackman. April 2006.

PEER 2005/15 CPT-Based Probabilistic Assessment of Seismic Soil Liquefaction Initiation. R. E. S. Moss, R. B. Seed, R. E.
Kayen, J. P. Stewart, and A. Der Kiureghian. April 2006.

PEER 2005/14 Workshop on Modeling of Nonlinear Cyclic Load-Deformation Behavior of Shallow Foundations. Bruce L. Kutter,
Geoffrey Martin, Tara Hutchinson, Chad Harden, Sivapalan Gajan, and Justin Phalen. March 2006.

PEER 2005/13 Stochastic Characterization and Decision Bases under Time-Dependent Aftershock Risk in Performance-Based
Earthquake Engineering. Gee Liek Yeo and C. Allin Cornell. July 2005.
PEER 2005/12 PEER Testbed Study on a Laboratory Building: Exercising Seismic Performance Assessment. Mary C. Comerio,
editor. November 2005.

PEER 2005/11 Van Nuys Hotel Building Testbed Report: Exercising Seismic Performance Assessment. Helmut Krawinkler,
editor. October 2005.

PEER 2005/10 First NEES/E-Defense Workshop on Collapse Simulation of Reinforced Concrete Building Structures. September
2005.

PEER 2005/09 Test Applications of Advanced Seismic Assessment Guidelines. Joe Maffei, Karl Telleen, Danya Mohr, William
Holmes, and Yuki Nakayama. August 2006.

PEER 2005/08 Damage Accumulation in Lightly Confined Reinforced Concrete Bridge Columns. R. Tyler Ranf, Jared M. Nelson,
Zach Price, Marc O. Eberhard, and John F. Stanton. April 2006.

PEER 2005/07 Experimental and Analytical Studies on the Seismic Response of Freestanding and Anchored Laboratory
Equipment. Dimitrios Konstantinidis and Nicos Makris. January 2005.

PEER 2005/06 Global Collapse of Frame Structures under Seismic Excitations. Luis F. Ibarra and Helmut Krawinkler. September
2005.

PEER 2005//05 Performance Characterization of Bench- and Shelf-Mounted Equipment. Samit Ray Chaudhuri and Tara C.
Hutchinson. May 2006.

PEER 2005/04 Numerical Modeling of the Nonlinear Cyclic Response of Shallow Foundations. Chad Harden, Tara Hutchinson,
Geoffrey R. Martin, and Bruce L. Kutter. August 2005.

PEER 2005/03 A Taxonomy of Building Components for Performance-Based Earthquake Engineering. Keith A. Porter.
September 2005.

PEER 2005/02 Fragility Basis for California Highway Overpass Bridge Seismic Decision Making. Kevin R. Mackie and Božidar
Stojadinović. June 2005.

PEER 2005/01 Empirical Characterization of Site Conditions on Strong Ground Motion. Jonathan P. Stewart, Yoojoong Choi,
and Robert W. Graves. June 2005.

PEER 2004/09 Electrical Substation Equipment Interaction: Experimental Rigid Conductor Studies. Christopher Stearns and
André Filiatrault. February 2005.

PEER 2004/08 Seismic Qualification and Fragility Testing of Line Break 550-kV Disconnect Switches. Shakhzod M. Takhirov,
Gregory L. Fenves, and Eric Fujisaki. January 2005.

PEER 2004/07 Ground Motions for Earthquake Simulator Qualification of Electrical Substation Equipment. Shakhzod M.
Takhirov, Gregory L. Fenves, Eric Fujisaki, and Don Clyde. January 2005.

PEER 2004/06 Performance-Based Regulation and Regulatory Regimes. Peter J. May and Chris Koski. September 2004.

PEER 2004/05 Performance-Based Seismic Design Concepts and Implementation: Proceedings of an International Workshop.
Peter Fajfar and Helmut Krawinkler, editors. September 2004.

PEER 2004/04 Seismic Performance of an Instrumented Tilt-up Wall Building. James C. Anderson and Vitelmo V. Bertero. July
2004.

PEER 2004/03 Evaluation and Application of Concrete Tilt-up Assessment Methodologies. Timothy Graf and James O. Malley.
October 2004.

PEER 2004/02 Analytical Investigations of New Methods for Reducing Residual Displacements of Reinforced Concrete Bridge
Columns. Junichi Sakai and Stephen A. Mahin. August 2004.

PEER 2004/01 Seismic Performance of Masonry Buildings and Design Implications. Kerri Anne Taeko Tokoro, James C.
Anderson, and Vitelmo V. Bertero. February 2004.

PEER 2003/18 Performance Models for Flexural Damage in Reinforced Concrete Columns. Michael Berry and Marc Eberhard.
August 2003.

PEER 2003/17 Predicting Earthquake Damage in Older Reinforced Concrete Beam-Column Joints. Catherine Pagni and Laura
Lowes. October 2004.

PEER 2003/16 Seismic Demands for Performance-Based Design of Bridges. Kevin Mackie and Božidar Stojadinović. August
2003.

PEER 2003/15 Seismic Demands for Nondeteriorating Frame Structures and Their Dependence on Ground Motions. Ricardo
Antonio Medina and Helmut Krawinkler. May 2004.

PEER 2003/14 Finite Element Reliability and Sensitivity Methods for Performance-Based Earthquake Engineering. Terje
Haukaas and Armen Der Kiureghian. April 2004.
PEER 2003/13 Effects of Connection Hysteretic Degradation on the Seismic Behavior of Steel Moment-Resisting Frames. Janise
E. Rodgers and Stephen A. Mahin. March 2004.

PEER 2003/12 Implementation Manual for the Seismic Protection of Laboratory Contents: Format and Case Studies. William T.
Holmes and Mary C. Comerio. October 2003.

PEER 2003/11 Fifth U.S.-Japan Workshop on Performance-Based Earthquake Engineering Methodology for Reinforced
Concrete Building Structures. February 2004.

PEER 2003/10 A Beam-Column Joint Model for Simulating the Earthquake Response of Reinforced Concrete Frames. Laura N.
Lowes, Nilanjan Mitra, and Arash Altoontash. February 2004.

PEER 2003/09 Sequencing Repairs after an Earthquake: An Economic Approach. Marco Casari and Simon J. Wilkie. April 2004.

PEER 2003/08 A Technical Framework for Probability-Based Demand and Capacity Factor Design (DCFD) Seismic Formats.
Fatemeh Jalayer and C. Allin Cornell. November 2003.

PEER 2003/07 Uncertainty Specification and Propagation for Loss Estimation Using FOSM Methods. Jack W. Baker and C. Allin
Cornell. September 2003.

PEER 2003/06 Performance of Circular Reinforced Concrete Bridge Columns under Bidirectional Earthquake Loading. Mahmoud
M. Hachem, Stephen A. Mahin, and Jack P. Moehle. February 2003.

PEER 2003/05 Response Assessment for Building-Specific Loss Estimation. Eduardo Miranda and Shahram Taghavi.
September 2003.

PEER 2003/04 Experimental Assessment of Columns with Short Lap Splices Subjected to Cyclic Loads. Murat Melek, John W.
Wallace, and Joel Conte. April 2003.

PEER 2003/03 Probabilistic Response Assessment for Building-Specific Loss Estimation. Eduardo Miranda and Hesameddin
Aslani. September 2003.

PEER 2003/02 Software Framework for Collaborative Development of Nonlinear Dynamic Analysis Program. Jun Peng and
Kincho H. Law. September 2003.

PEER 2003/01 Shake Table Tests and Analytical Studies on the Gravity Load Collapse of Reinforced Concrete Frames. Kenneth
John Elwood and Jack P. Moehle. November 2003.

PEER 2002/24 Performance of Beam to Column Bridge Joints Subjected to a Large Velocity Pulse. Natalie Gibson, André
Filiatrault, and Scott A. Ashford. April 2002.

PEER 2002/23 Effects of Large Velocity Pulses on Reinforced Concrete Bridge Columns. Greg L. Orozco and Scott A. Ashford.
April 2002.

PEER 2002/22 Characterization of Large Velocity Pulses for Laboratory Testing. Kenneth E. Cox and Scott A. Ashford. April
2002.

PEER 2002/21 Fourth U.S.-Japan Workshop on Performance-Based Earthquake Engineering Methodology for Reinforced
Concrete Building Structures. December 2002.

PEER 2002/20 Barriers to Adoption and Implementation of PBEE Innovations. Peter J. May. August 2002.

PEER 2002/19 Economic-Engineered Integrated Models for Earthquakes: Socioeconomic Impacts. Peter Gordon, James E.
Moore II, and Harry W. Richardson. July 2002.

PEER 2002/18 Assessment of Reinforced Concrete Building Exterior Joints with Substandard Details. Chris P. Pantelides, Jon
Hansen, Justin Nadauld, and Lawrence D. Reaveley. May 2002.

PEER 2002/17 Structural Characterization and Seismic Response Analysis of a Highway Overcrossing Equipped with
Elastomeric Bearings and Fluid Dampers: A Case Study. Nicos Makris and Jian Zhang. November 2002.

PEER 2002/16 Estimation of Uncertainty in Geotechnical Properties for Performance-Based Earthquake Engineering. Allen L.
Jones, Steven L. Kramer, and Pedro Arduino. December 2002.

PEER 2002/15 Seismic Behavior of Bridge Columns Subjected to Various Loading Patterns. Asadollah Esmaeily-Gh. and Yan
Xiao. December 2002.

PEER 2002/14 Inelastic Seismic Response of Extended Pile Shaft Supported Bridge Structures. T.C. Hutchinson, R.W.
Boulanger, Y.H. Chai, and I.M. Idriss. December 2002.

PEER 2002/13 Probabilistic Models and Fragility Estimates for Bridge Components and Systems. Paolo Gardoni, Armen Der
Kiureghian, and Khalid M. Mosalam. June 2002.

PEER 2002/12 Effects of Fault Dip and Slip Rake on Near-Source Ground Motions: Why Chi-Chi Was a Relatively Mild M7.6
Earthquake. Brad T. Aagaard, John F. Hall, and Thomas H. Heaton. December 2002.
PEER 2002/11 Analytical and Experimental Study of Fiber-Reinforced Strip Isolators. James M. Kelly and Shakhzod M. Takhirov.
September 2002.

PEER 2002/10 Centrifuge Modeling of Settlement and Lateral Spreading with Comparisons to Numerical Analyses. Sivapalan
Gajan and Bruce L. Kutter. January 2003.

PEER 2002/09 Documentation and Analysis of Field Case Histories of Seismic Compression during the 1994 Northridge,
California, Earthquake. Jonathan P. Stewart, Patrick M. Smith, Daniel H. Whang, and Jonathan D. Bray. October
2002.
TM
PEER 2002/08 Component Testing, Stability Analysis and Characterization of Buckling-Restrained Unbonded Braces .
Cameron Black, Nicos Makris, and Ian Aiken. September 2002.

PEER 2002/07 Seismic Performance of Pile-Wharf Connections. Charles W. Roeder, Robert Graff, Jennifer Soderstrom, and Jun
Han Yoo. December 2001.

PEER 2002/06 The Use of Benefit-Cost Analysis for Evaluation of Performance-Based Earthquake Engineering Decisions.
Richard O. Zerbe and Anthony Falit-Baiamonte. September 2001.

PEER 2002/05 Guidelines, Specifications, and Seismic Performance Characterization of Nonstructural Building Components and
Equipment. André Filiatrault, Constantin Christopoulos, and Christopher Stearns. September 2001.

PEER 2002/04 Consortium of Organizations for Strong-Motion Observation Systems and the Pacific Earthquake Engineering
Research Center Lifelines Program: Invited Workshop on Archiving and Web Dissemination of Geotechnical
Data, 4–5 October 2001. September 2002.

PEER 2002/03 Investigation of Sensitivity of Building Loss Estimates to Major Uncertain Variables for the Van Nuys Testbed.
Keith A. Porter, James L. Beck, and Rustem V. Shaikhutdinov. August 2002.

PEER 2002/02 The Third U.S.-Japan Workshop on Performance-Based Earthquake Engineering Methodology for Reinforced
Concrete Building Structures. July 2002.

PEER 2002/01 Nonstructural Loss Estimation: The UC Berkeley Case Study. Mary C. Comerio and John C. Stallmeyer.
December 2001.

PEER 2001/16 Statistics of SDF-System Estimate of Roof Displacement for Pushover Analysis of Buildings. Anil K. Chopra,
Rakesh K. Goel, and Chatpan Chintanapakdee. December 2001.

PEER 2001/15 Damage to Bridges during the 2001 Nisqually Earthquake. R. Tyler Ranf, Marc O. Eberhard, and Michael P.
Berry. November 2001.

PEER 2001/14 Rocking Response of Equipment Anchored to a Base Foundation. Nicos Makris and Cameron J. Black.
September 2001.

PEER 2001/13 Modeling Soil Liquefaction Hazards for Performance-Based Earthquake Engineering. Steven L. Kramer and
Ahmed-W. Elgamal. February 2001.

PEER 2001/12 Development of Geotechnical Capabilities in OpenSees. Boris Jeremić. September 2001.

PEER 2001/11 Analytical and Experimental Study of Fiber-Reinforced Elastomeric Isolators. James M. Kelly and Shakhzod M.
Takhirov. September 2001.

PEER 2001/10 Amplification Factors for Spectral Acceleration in Active Regions. Jonathan P. Stewart, Andrew H. Liu, Yoojoong
Choi, and Mehmet B. Baturay. December 2001.

PEER 2001/09 Ground Motion Evaluation Procedures for Performance-Based Design. Jonathan P. Stewart, Shyh-Jeng Chiou,
Jonathan D. Bray, Robert W. Graves, Paul G. Somerville, and Norman A. Abrahamson. September 2001.

PEER 2001/08 Experimental and Computational Evaluation of Reinforced Concrete Bridge Beam-Column Connections for
Seismic Performance. Clay J. Naito, Jack P. Moehle, and Khalid M. Mosalam. November 2001.

PEER 2001/07 The Rocking Spectrum and the Shortcomings of Design Guidelines. Nicos Makris and Dimitrios Konstantinidis.
August 2001.

PEER 2001/06 Development of an Electrical Substation Equipment Performance Database for Evaluation of Equipment
Fragilities. Thalia Agnanos. April 1999.

PEER 2001/05 Stiffness Analysis of Fiber-Reinforced Elastomeric Isolators. Hsiang-Chuan Tsai and James M. Kelly. May 2001.

PEER 2001/04 Organizational and Societal Considerations for Performance-Based Earthquake Engineering. Peter J. May. April
2001.

PEER 2001/03 A Modal Pushover Analysis Procedure to Estimate Seismic Demands for Buildings: Theory and Preliminary
Evaluation. Anil K. Chopra and Rakesh K. Goel. January 2001.
PEER 2001/02 Seismic Response Analysis of Highway Overcrossings Including Soil-Structure Interaction. Jian Zhang and Nicos
Makris. March 2001.

PEER 2001/01 Experimental Study of Large Seismic Steel Beam-to-Column Connections. Egor P. Popov and Shakhzod M.
Takhirov. November 2000.

PEER 2000/10 The Second U.S.-Japan Workshop on Performance-Based Earthquake Engineering Methodology for Reinforced
Concrete Building Structures. March 2000.

PEER 2000/09 Structural Engineering Reconnaissance of the August 17, 1999 Earthquake: Kocaeli (Izmit), Turkey. Halil Sezen,
Kenneth J. Elwood, Andrew S. Whittaker, Khalid Mosalam, John J. Wallace, and John F. Stanton. December
2000.

PEER 2000/08 Behavior of Reinforced Concrete Bridge Columns Having Varying Aspect Ratios and Varying Lengths of
Confinement. Anthony J. Calderone, Dawn E. Lehman, and Jack P. Moehle. January 2001.

PEER 2000/07 Cover-Plate and Flange-Plate Reinforced Steel Moment-Resisting Connections. Taejin Kim, Andrew S. Whittaker,
Amir S. Gilani, Vitelmo V. Bertero, and Shakhzod M. Takhirov. September 2000.

PEER 2000/06 Seismic Evaluation and Analysis of 230-kV Disconnect Switches. Amir S. J. Gilani, Andrew S. Whittaker, Gregory
L. Fenves, Chun-Hao Chen, Henry Ho, and Eric Fujisaki. July 2000.

PEER 2000/05 Performance-Based Evaluation of Exterior Reinforced Concrete Building Joints for Seismic Excitation. Chandra
Clyde, Chris P. Pantelides, and Lawrence D. Reaveley. July 2000.

PEER 2000/04 An Evaluation of Seismic Energy Demand: An Attenuation Approach. Chung-Che Chou and Chia-Ming Uang. July
1999.

PEER 2000/03 Framing Earthquake Retrofitting Decisions: The Case of Hillside Homes in Los Angeles. Detlof von Winterfeldt,
Nels Roselund, and Alicia Kitsuse. March 2000.

PEER 2000/02 U.S.-Japan Workshop on the Effects of Near-Field Earthquake Shaking. Andrew Whittaker, ed. July 2000.

PEER 2000/01 Further Studies on Seismic Interaction in Interconnected Electrical Substation Equipment. Armen Der Kiureghian,
Kee-Jeung Hong, and Jerome L. Sackman. November 1999.

PEER 1999/14 Seismic Evaluation and Retrofit of 230-kV Porcelain Transformer Bushings. Amir S. Gilani, Andrew S. Whittaker,
Gregory L. Fenves, and Eric Fujisaki. December 1999.

PEER 1999/13 Building Vulnerability Studies: Modeling and Evaluation of Tilt-up and Steel Reinforced Concrete Buildings. John
W. Wallace, Jonathan P. Stewart, and Andrew S. Whittaker, editors. December 1999.

PEER 1999/12 Rehabilitation of Nonductile RC Frame Building Using Encasement Plates and Energy-Dissipating Devices.
Mehrdad Sasani, Vitelmo V. Bertero, James C. Anderson. December 1999.

PEER 1999/11 Performance Evaluation Database for Concrete Bridge Components and Systems under Simulated Seismic
Loads. Yael D. Hose and Frieder Seible. November 1999.

PEER 1999/10 U.S.-Japan Workshop on Performance-Based Earthquake Engineering Methodology for Reinforced Concrete
Building Structures. December 1999.

PEER 1999/09 Performance Improvement of Long Period Building Structures Subjected to Severe Pulse-Type Ground Motions.
James C. Anderson, Vitelmo V. Bertero, and Raul Bertero. October 1999.

PEER 1999/08 Envelopes for Seismic Response Vectors. Charles Menun and Armen Der Kiureghian. July 1999.

PEER 1999/07 Documentation of Strengths and Weaknesses of Current Computer Analysis Methods for Seismic Performance of
Reinforced Concrete Members. William F. Cofer. November 1999.

PEER 1999/06 Rocking Response and Overturning of Anchored Equipment under Seismic Excitations. Nicos Makris and Jian
Zhang. November 1999.

PEER 1999/05 Seismic Evaluation of 550 kV Porcelain Transformer Bushings. Amir S. Gilani, Andrew S. Whittaker, Gregory L.
Fenves, and Eric Fujisaki. October 1999.

PEER 1999/04 Adoption and Enforcement of Earthquake Risk-Reduction Measures. Peter J. May, Raymond J. Burby, T. Jens
Feeley, and Robert Wood.

PEER 1999/03 Task 3 Characterization of Site Response General Site Categories. Adrian Rodriguez-Marek, Jonathan D. Bray,
and Norman Abrahamson. February 1999.

PEER 1999/02 Capacity-Demand-Diagram Methods for Estimating Seismic Deformation of Inelastic Structures: SDF Systems.
Anil K. Chopra and Rakesh Goel. April 1999.

PEER 1999/01 Interaction in Interconnected Electrical Substation Equipment Subjected to Earthquake Ground Motions. Armen
Der Kiureghian, Jerome L. Sackman, and Kee-Jeung Hong. February 1999.
PEER 1998/08 Behavior and Failure Analysis of a Multiple-Frame Highway Bridge in the 1994 Northridge Earthquake. Gregory L.
Fenves and Michael Ellery. December 1998.

PEER 1998/07 Empirical Evaluation of Inertial Soil-Structure Interaction Effects. Jonathan P. Stewart, Raymond B. Seed, and
Gregory L. Fenves. November 1998.

PEER 1998/06 Effect of Damping Mechanisms on the Response of Seismic Isolated Structures. Nicos Makris and Shih-Po
Chang. November 1998.

PEER 1998/05 Rocking Response and Overturning of Equipment under Horizontal Pulse-Type Motions. Nicos Makris and
Yiannis Roussos. October 1998.

PEER 1998/04 Pacific Earthquake Engineering Research Invitational Workshop Proceedings, May 14–15, 1998: Defining the
Links between Planning, Policy Analysis, Economics and Earthquake Engineering. Mary Comerio and Peter
Gordon. September 1998.

PEER 1998/03 Repair/Upgrade Procedures for Welded Beam to Column Connections. James C. Anderson and Xiaojing Duan.
May 1998.

PEER 1998/02 Seismic Evaluation of 196 kV Porcelain Transformer Bushings. Amir S. Gilani, Juan W. Chavez, Gregory L.
Fenves, and Andrew S. Whittaker. May 1998.

PEER 1998/01 Seismic Performance of Well-Confined Concrete Bridge Columns. Dawn E. Lehman and Jack P. Moehle.
December 2000.
 ONLINE PEER REPORTS

The following PEER reports are available by Internet only at http://peer.berkeley.edu/publications/peer_reports_complete.html.

PEER 2012/103 Performance-Based Seismic Demand Assessment of Concentrically Braced Steel Frame Buildings. Chui-Hsin
Chen and Stephen A. Mahin. December 2012.
PEER 2012/102 Procedure to Restart an Interrupted Hybrid Simulation: Addendum to PEER Report 2010/103. Vesna Terzic and
Bozidar Stojadinovic. October 2012.
PEER 2012/101 Mechanics of Fiber Reinforced Bearings. James M. Kelly and Andrea Calabrese. February 2012.
PEER 2011/107 Nonlinear Site Response and Seismic Compression at Vertical Array Strongly Shaken by 2007 Niigata-ken
Chuetsu-oki Earthquake. Eric Yee, Jonathan P. Stewart, and Kohji Tokimatsu. December 2011.
PEER 2011/106 Self Compacting Hybrid Fiber Reinforced Concrete Composites for Bridge Columns. Pardeep Kumar, Gabriel Jen,
William Trono, Marios Panagiotou, and Claudia Ostertag. September 2011.
PEER 2011/105 Stochastic Dynamic Analysis of Bridges Subjected to Spacially Varying Ground Motions. Katerina Konakli and
Armen Der Kiureghian. August 2011.
PEER 2011/104 Design and Instrumentation of the 2010 E-Defense Four-Story Reinforced Concrete and Post-Tensioned
Concrete Buildings. Takuya Nagae, Kenichi Tahara, Taizo Matsumori, Hitoshi Shiohara, Toshimi Kabeyasawa,
Susumu Kono, Minehiro Nishiyama (Japanese Research Team) and John Wallace, Wassim Ghannoum, Jack
Moehle, Richard Sause, Wesley Keller, Zeynep Tuna (U.S. Research Team). June 2011.
PEER 2011/103 In-Situ Monitoring of the Force Output of Fluid Dampers: Experimental Investigation. Dimitrios Konstantinidis,
James M. Kelly, and Nicos Makris. April 2011.
PEER 2011/102 Ground-motion prediction equations 1964 - 2010. John Douglas. April 2011.
PEER 2011/101 Report of the Eighth Planning Meeting of NEES/E-Defense Collaborative Research on Earthquake Engineering.
Convened by the Hyogo Earthquake Engineering Research Center (NIED), NEES Consortium, Inc. February
2011.
PEER 2010/111 Modeling and Acceptance Criteria for Seismic Design and Analysis of Tall Buildings. Task 7 Report for the Tall
Buildings Initiative - Published jointly by the Applied Technology Council. October 2010.
PEER 2010/110 Seismic Performance Assessment and Probabilistic Repair Cost Analysis of Precast Concrete Cladding Systems
for Multistory Buildlings. Jeffrey P. Hunt and Božidar Stojadinovic. November 2010.
PEER 2010/109 Report of the Seventh Joint Planning Meeting of NEES/E-Defense Collaboration on Earthquake Engineering.
Held at the E-Defense, Miki, and Shin-Kobe, Japan, September 18–19, 2009. August 2010.
PEER 2010/108 Probabilistic Tsunami Hazard in California. Hong Kie Thio, Paul Somerville, and Jascha Polet, preparers. October
2010.
PEER 2010/107 Performance and Reliability of Exposed Column Base Plate Connections for Steel Moment-Resisting Frames.
Ady Aviram, Božidar Stojadinovic, and Armen Der Kiureghian. August 2010.
PEER 2010/106 Verification of Probabilistic Seismic Hazard Analysis Computer Programs. Patricia Thomas, Ivan Wong, and
Norman Abrahamson. May 2010.
PEER 2010/105 Structural Engineering Reconnaissance of the April 6, 2009, Abruzzo, Italy, Earthquake, and Lessons Learned. M.
Selim Günay and Khalid M. Mosalam. April 2010.
PEER 2010/104 Simulating the Inelastic Seismic Behavior of Steel Braced Frames, Including the Effects of Low-Cycle Fatigue.
Yuli Huang and Stephen A. Mahin. April 2010.

PEER 2010/103 Post-Earthquake Traffic Capacity of Modern Bridges in California. Vesna Terzic and Božidar Stojadinović. March
2010.
PEER 2010/102 Analysis of Cumulative Absolute Velocity (CAV) and JMA Instrumental Seismic Intensity (IJMA) Using the PEER–
NGA Strong Motion Database. Kenneth W. Campbell and Yousef Bozorgnia. February 2010.

PEER 2010/101 Rocking Response of Bridges on Shallow Foundations. Jose A. Ugalde, Bruce L. Kutter, and Boris Jeremic. April
2010.
PEER 2009/109 Simulation and Performance-Based Earthquake Engineering Assessment of Self-Centering Post-Tensioned
Concrete Bridge Systems. Won K. Lee and Sarah L. Billington. December 2009.
PEER 2009/108 PEER Lifelines Geotechnical Virtual Data Center. J. Carl Stepp, Daniel J. Ponti, Loren L. Turner, Jennifer N. Swift,
Sean Devlin, Yang Zhu, Jean Benoit, and John Bobbitt. September 2009.
PEER 2009/107 Experimental and Computational Evaluation of Current and Innovative In-Span Hinge Details in Reinforced
Concrete Box-Girder Bridges: Part 2: Post-Test Analysis and Design Recommendations. Matias A. Hube and
Khalid M. Mosalam. December 2009.
PEER 2009/106 Shear Strength Models of Exterior Beam-Column Joints without Transverse Reinforcement. Sangjoon Park and
Khalid M. Mosalam. November 2009.

PEER 2009/105 Reduced Uncertainty of Ground Motion Prediction Equations through Bayesian Variance Analysis. Robb Eric S.
Moss. November 2009.

PEER 2009/104 Advanced Implementation of Hybrid Simulation. Andreas H. Schellenberg, Stephen A. Mahin, Gregory L. Fenves.
November 2009.

PEER 2009/103 Performance Evaluation of Innovative Steel Braced Frames. T. Y. Yang, Jack P. Moehle, and Božidar
Stojadinovic. August 2009.

PEER 2009/102 Reinvestigation of Liquefaction and Nonliquefaction Case Histories from the 1976 Tangshan Earthquake. Robb
Eric Moss, Robert E. Kayen, Liyuan Tong, Songyu Liu, Guojun Cai, and Jiaer Wu. August 2009.

PEER 2009/101 Report of the First Joint Planning Meeting for the Second Phase of NEES/E-Defense Collaborative Research on
Earthquake Engineering. Stephen A. Mahin et al. July 2009.

PEER 2008/104 Experimental and Analytical Study of the Seismic Performance of Retaining Structures. Linda Al Atik and Nicholas
Sitar. January 2009.

PEER 2008/103 Experimental and Computational Evaluation of Current and Innovative In-Span Hinge Details in Reinforced
Concrete Box-Girder Bridges. Part 1: Experimental Findings and Pre-Test Analysis. Matias A. Hube and Khalid M.
Mosalam. January 2009.

PEER 2008/102 Modeling of Unreinforced Masonry Infill Walls Considering In-Plane and Out-of-Plane Interaction. Stephen
Kadysiewski and Khalid M. Mosalam. January 2009.

PEER 2008/101 Seismic Performance Objectives for Tall Buildings. William T. Holmes, Charles Kircher, William Petak, and Nabih
Youssef. August 2008.

PEER 2007/101 Generalized Hybrid Simulation Framework for Structural Systems Subjected to Seismic Loading. Tarek Elkhoraibi
and Khalid M. Mosalam. July 2007.

PEER 2007/100 Seismic Evaluation of Reinforced Concrete Buildings Including Effects of Masonry Infill Walls. Alidad Hashemi
and Khalid M. Mosalam. July 2007.
The Pacific Earthquake Engineering Research Center (PEER) is a multi-institutional research and
education center with headquarters at the University of California, Berkeley. Investigators from over 20
universities, several consulting companies, and researchers at various state and federal government
agencies contribute to research programs focused on performance-based earthquake engineering.

These research programs aim to identify and reduce the risks from major earthquakes to life safety and
to the economy by including research in a wide variety of disciplines including structural and geotechnical
engineering, geology/seismology, lifelines, transportation, architecture, economics, risk management, and
public policy.

PEER is supported by federal, state, local, and regional agencies, together with industry partners.

PEER Core Institutions:

University of California, Berkeley (Lead Institution)
California Institute of Technology
Oregon State University
Stanford University
University of California, Davis
University of California, Irvine
University of California, Los Angeles
University of California, San Diego
University of Southern California
University of Washington

PEER reports can be ordered at http://peer.berkeley.edu/publications/peer_reports.html or by contacting

Pacific Earthquake Engineering Research Center

University of California, Berkeley
325 Davis Hall, mail code 1792
Berkeley, CA 94720-1792
Tel: 510-642-3437
Fax: 510-642-1655
Email: [email protected]

ISSN 1547-0587X