1

DESIGN GROUND MOTIONS NEAR ACTIVE FAULTS

Jonathan D. Bray,1 Adrian Rodriguez-Marek2

and Joanne L. Gillie3

SUMMARY
Forward-Directivity (FD) in the near-fault region can produce intense, pulse-type motions that differ
significantly from ordinary ground motions that occur further from the ruptured fault. Near-fault FD
motions typically govern the design of structures built close to active faults so the selection of design
ground motions is critical for achieving effective performance without costly over-design. Updated
empirical relationships are provided for estimating the peak ground velocity (PGV) and period of the
velocity pulse (Tv) of near-fault FD motions. PGV varies significantly with magnitude, distance, and site
effects. Tv is a function of magnitude and site conditions with most of the energy being concentrated
within a narrow-period band centred on the pulse period. Lower magnitude events, which produce lower
pulse periods, might produce more damaging ground motions for the stiff structures more common in
urban areas. As the number of near-fault recordings is still limited, fully nonlinear bi-directional shaking
simulations are employed to gain additional insight. It is shown that site effects generally cause Tv to
increase. Although the amplification of PGV at soil sites depends on site properties, amplification is
generally observed even for very intense rock motions. At soft soil sites, seismic site response can be
limited by the yield strength of the soil, but then seismic instability may be a concern.

FORWARD-DIRECTIVITY formed by the accumulation of the shear waves travelling

Near-fault ground motions are significantly influenced by the ahead of the rupture front. When a site is located at one end of
rupture mechanism and slip direction relative to the site and the fault and rupture initiates at the other end of the fault and
by the permanent ground displacement at the site resulting travels towards the site, the arrival of the wave front is seen as
from tectonic movement. When the rupture and slip direction a large pulse of motion that occurs near the beginning of the
relative to a site coincide, and a significant portion of the fault record. Thus, FD motions typically occur at sites near the end
ruptures towards the site, the ground motion can exhibit the of a strike-slip fault when the rupture moves towards the site,
effects of forward-directivity (FD) [1]. Most of the energy in and at sites located in the up-dip projection of a ruptured dip-
FD motions is concentrated in a narrow frequency band and is slip fault (i.e., reverse or normal fault). The radiation pattern
expressed as one or more high intensity velocity pulses of the shear dislocation on the fault causes this large pulse of
oriented in the fault-normal direction. These intense velocity motion to be oriented in a direction perpendicular to the fault
pulses can lead to severe structural damage. plane [1]. These effects are typically long-period in nature and
are best observed in the velocity-time history. FD conditions
Ground motions close to the surface rupture may also contain produce ground motions that have large amplitude and short
a significant permanent displacement, which is called fling- durations. However, if a site is located at one end of the fault
step, and this may lead to a high intensity velocity pulse in the and rupture propagates away from the site, the opposite effect
direction of the fault displacement. Pulses from fling-step have is observed (i.e., backward-directivity), and the motion is
different characteristics than FD pulses. Whereas FD is a characterized by longer duration and lower amplitude ground
dynamic phenomenon that produces no permanent ground motions.
displacement and hence two-sided velocity pulses, fling-step
is a result of a permanent ground displacement that generates Some examples of near-fault FD motions are shown in Figure
one sided velocity pulses. The development of design ground 1. The use of the velocity-trace plot shown at the right is
motions for a project site close to an active fault should useful, because FD motions typically exhibit a systematic
account for these special aspects of near-fault ground motions. difference between the fault-normal and fault-parallel
Fling-step considerations are discussed in [2]. In this paper, components of motion. The fault-normal component is
near-fault forward-directivity effects are addressed. systematically more intense than the fault-parallel component
of motion. It is important to remember that the average of the
The effects of forward-directivity are generated because the two components of motion is systematically more intense at
velocity of the fault rupture front is only slightly less than the long periods than ordinary motions as well. Hence, near-fault
shear wave propagation velocity [1]. As the rupture front FD fault-normal components of motion are especially severe
propagates from the focus of the event, a shear wave front is and potentially destructive.

1
Professor, Dept. of Civil and Environmental Engineering, University of California, Berkeley, CA 94720-1710, USA, email:
[email protected]
2
Associate Professor, Dept. of Civil and Environmental Engineering, Washington State University, PO Box 642910, Pullman, WA
99164-2910, USA
3
HWA Geosciences, 19730 64th Avenue West, Suite 200, Lynnwood, WA 98036-5957, USA.

BULLETIN OF THE NEW ZEALAND SOCIETY FOR EARTHQUAKE ENGINEERING, Vol. 42, No. 1, March 2009
2

Fault Zone 9, 2004 Parkfield EQ.

50 50
Thick: Fault Normal

FP Velocity (cm/s)
Thin: Fault Parallel

Velocity (cm/s)
0 0

-50 -50
0 5 10 15 20 -50 0 50

Arcelik, 1999 Kocaeli EQ.

50 50

FP Velocity (cm/s)
Velocity (cm/s)

0 0

-50 -50
0 5 10 15 20 -50 0 50
Time (s) FN Velocity (cm/s)

Figure 1: Fault normal (FN) and fault parallel (FP) horizontal velocity-time histories and velocity-traces for
near-fault records from the 2004 Parkfield (Mw = 6.0) and the Kocaeli (Mw = 7.5) earthquakes.

Pulse-type motions are critical in the design of structures in earthquakes, and the relationships of Bray and Rodriguez-
the near-fault zone. Two approaches have been used to Marek [8] are updated in this paper.
account for near-fault ground motions in design. The
frequency-domain approach uses empirical factors to modify Near-fault FD records were selected from the strong motion
acceleration response spectra for sites that are affected by database of the Pacific Earthquake Engineering Research
forward-directivity effects [1, 3]. However, advanced dynamic Center (http://peer.berkeley.edu/). Records with geometric
analyses indicate that the amplitude, period, and number of conditions leading to FD were used. Records were selected if
significant pulses in the velocity-time history primarily control the ratio of fault-normal to fault-parallel spectral acceleration
the performance of structures (e.g. [4, 5]). The alternative at a period of three seconds predicted by [1] was greater than
time-domain approach characterizes the FD motion through its one. Recordings not possessing at least some features of FD
velocity-time history (Figure 2), with its peak ground velocity characteristics were excluded from the analysis. FD
(PGV), predominant pulse period (Tv), and number of characteristics are positive fault-normal to fault-parallel
significant velocity pulses (Nc). In this preferred approach, it is response spectral ratios for long periods, and a reasonably
crucial that reliable estimates of PGV and Tv be obtained for well-defined velocity pulse in the fault-normal direction. The
near-fault FD motions. records are from shallow earthquakes (Mw ≥ 6) in active
tectonic regions at rupture distances (R = closest distance to
the fault plane) less than 20 km. Fourteen near-fault records
PGV from four earthquakes (i.e., the 1986 Palm Springs, 2002
Denali, 2003 Bam, and 2004 Parkfield earthquakes) were
added to the database of Bray and Rodriguez-Marek [8]. Three
records from the 1999 Chi-Chi, Taiwan earthquake were
Velocity (cm/s)

processed to remove the fling-step present in the records by

fitting a hyperbolic tangent function to the displacement-time
Tv time (s) history of the record and subsequently subtracting this motion
from the time history (Rathje, pers. Comm. 2000). Ground
motion sites were classified as either rock/shallow stiff soil
(i.e., only 0 – 20 m of soil or weathered rock over competent
rock) or soil (i.e., mostly stiff soil with shear wave velocity, Vs
1st cycle 2nd cycle 3rd cycle > 180 m/s). Soft soil and liquefiable sites were excluded.
Additional details about the earthquakes, records, and fault-
Figure 2: Simplified Velocity-Time History Showing Peak normal component orientation used are provided in [8] and
Ground Velocity (PGV), Period of the Velocity- [9].
Pulse (Tv), and number of significant cycles of The predictive relationships for PGV and Tv in this study (as
motion (Nc). well as the previous study by Bray and Rodriguez-Marek, [8])
include the influence of site conditions (i.e., “rock” or “soil”)
EMPIRICAL RELATIONSHIPS as well as earthquake moment magnitude (Mw) and the closest
distance from the site to the ruptured fault (R). The empirical
Empirical ground motion relationships may be used to develop evidence clearly points to a systematic difference between
reasonable estimates of PGV and Tv for FD motions. A near-fault FD motions recorded on rock and on soil sites.
number of researchers have developed empirically based Ground motions recorded in soil tend to have longer pulse
predictive relationships for these near-fault ground motion periods and larger PGVs than those recorded at rock sites [8].
parameters (e.g. [6, 4, 7]). Most recently, Bray and Rodriguez- An example of this is the set of ground motions recorded in
Marek [8] used a comprehensive database of FD ground Gilroy during the 1989 Loma Prieta earthquake as shown in
motions to develop empirical relationships for PGV and Tv. Figure 3.
This database was enhanced with FD records from recent
 3

Gilroy #1: Rock

Pulse Period
Tv = 1.2 s; PGV = 39 cm/s
Somerville [6] provides justification for using self-similar
Gilroy Gavilan College: Rock scaling relationships to constrain fault parameters. The use of
Tv = 1.2 s; PGV = 31 cm/s this scaling relationship indicates that the pulse period is about
Velocity (cm/s)

Gilroy #2: Soil

two times larger than the rise time of slip on a fault, which
Tv = 1.4 s; PGV = 46 cm/s measures the duration of slip at a single point in the fault.
From the mechanics of fault rupture, the rise time can be
Gilroy #3: Soil established as a lower bound for pulse period [6]. Because the
Tv = 1.5 s; PGV = 49 cm/s
logarithm of rise time is a linear function of moment
magnitude, the use of a linear relationship between logarithm
20 cm/s
of rise time and magnitude is justified. Thus, the relationship
used by [8] for pulse period is:
0 5 10 15
Time (s)
ln(Tv)ij = f + hMw + ηi + εij (3)
200

180
Gilroy #1: Rock where (Tv)ij is the pulse period of the j recording from the ith
th
Gilroy Gavilan College: Rock
Gilroy #2: Soil
event; Mw is the moment magnitude of event i; f and h are
160
regression parameters determined by the data through the
Spectral Velocity (cm/s)

Gilroy #4: Soil

140
random effects model; ηi is the inter-event term; and εij
120 represents the intra-event variations. The updated parameters
100 for Equation (3) based on the regression analysis of the larger
80
data included in this paper are provided in Table 2.
60

40 Table 2. Parameters for the relationship for velocity pulse

20 period (Equation 3).
0
10
-2
10
-1
Period (s) 10
0
10
1 Data Set f h σ τ σtotal

Figure 3: Recorded fault-normal motions during the 1989 All

-4.42 0.75 0.41 0.381 0.56
Loma Prieta earthquake. Gilroy #1, Gavilan Motions
Coll., Gilroy #2, and Gilroy #3 have rupture Rock -6.37 1.00 0.46 0.29 0.55
distances of 11, 12, 13, and 14 km, resp.
(modified from [8]). Soil -3.71 0.65 0.35 0.37 0.51

Peak Ground Velocity

With the data for near-fault ground motions being restricted to Number of Significant Cycles
relatively small source-to-site distances, the functional form of
The number of significant cycles of motion is defined as half
the model for estimating PGV can be simplified to:
the number of half-cycle (one-sided) velocity pulses that have
ln(PGVij) = a + b Mw + c ln (R2 + d2) + ηi + εij (1) an amplitude at least 50% of the peak ground velocity of the
ground motion. Due to chaotic nature of fault rupture and the
where PGVij is the peak ground velocity in units of cm/s of the uncertainty in characterizing the details in the rupture process
jth recording from the ith event; Mw is moment magnitude of that determine the number of significant cycles, it is not
event i; R is rupture distance in km; and a, b, c, and d are possible at this time to develop a robust relationship for
regression parameters; and ηi and εij represent the inter- and predicting this important parameter. However, an examination
intra-event variations, respectively, obtained using the random of FD records does provide useful insights.
effects model [10]. The inter-event and the intra-event error
terms are assumed to be independent normally distributed Some earthquake events have a well-defined pulse sequence
random variables with variances τ2 and σ2, respectively. The for nearly all of its near-fault motions. This might be expected
standard error associated with the estimate of PGV is then for faults that have a relatively uniform slip distribution or
earthquakes where slip is concentrated over a single zone. For
σtotal2 = τ2 + σ2 (2) these events, stations that are close to each other will likely be
equidistant to regions of high slip. Moreover, path effects are
The functional form of Equation (1) for PGV as a function of
minimized for stations in the near-fault region. However, for
distance results in a nearly zero slope at close distances to the
an earthquake with highly non-uniform slip, such as the 1994
fault, and it decreases linearly with the logarithm of distance at
Northridge earthquake, the type of pulse sequence observed
larger distances. The statistical analysis was performed on the
depends on the instrument’s distance relative to the asperities.
entire dataset and then separately on the rock and soil motions.
Somerville [6] suggests that the number of half-sine pulses in
The parameters for Equation (1) are provided in Table 1.
the velocity time-history might be associated with the number
of asperities in a fault rupture. Thus, details in the rupture
Table1. Parameters for the PGV relationship (Equation 1). process, such as the number of asperities of the fault and the
slip distribution on the causative fault, determine the number
Data
a b c d σ τ σtotal of significant pulses in a FD motion.
Set
All Examination of the near-fault FD velocity-time histories
2.05 0.55 -0.39 5.00 0.37 0.24 0.44 included in this study does indicate that it is unlikely that a
Motions
near-fault FD record will have more than two significant full
Rock* 1.86 0.55 -0.39 5.00 - - 0.40 cycles of motion. More than half of the FD records contained
Soil 2.11 0.55 -0.39 5.00 0.33 0.30 0.44 only one significant full cycle of motion (i.e., one two-sided
* velocity pulse). Hence, in developing design ground motions
Parameters had to be obtained using maximum likelihood. for use in projects, it is reasonable to select records that have
4

only one or two significant cycles of motion. It would be elastic half-space [15]. The implementation of the constitutive
unnecessarily conservative to use simulated or modified model in the finite element code GeoFeap [16] is discussed in
ground motions records that have a large number of significant [17]. This seismic site response analysis procedure has been
cycles of motion in their velocity-time history. validated using field downhole array recordings and laboratory
shaking table measurements [17].
Although near-fault FD motions are more intense than
ordinary records, they are shorter in duration. The seismic Eight parameters are required to define the soil model. Elastic
energy is compact, which leads to the high intensities, but also soil response is determined by the shear wave velocity (Vs)
requires that the duration of significant shaking be short. and Poisson’s ratio (v), which is assumed to be 0.49 to
Hence, it would also be unnecessarily conservative to use high approach a fully undrained behavior. The exponential
intensity motions with long durations of strong shaking. It is interpolation function of Borja and Amies’ model is defined
more likely for near-fault FD motions to be at or below the by two model parameters and the kinematic hardening
median minus one standard deviation for significant duration. parameter of the bounding surface. Soil strength is defined by
the radius of the bounding surface, R, which is given by 1.6 Su,
where Su is the soil’s undrained strength in triaxial
Application into Probabilistic Seismic Hazard Analyses compression. When the soil will not reach shear failure, the
and Performance Based Design parameter R can be used as a curve fitting parameter. Energy
Estimates of seismic hazard are usually made using dissipation is naturally produced by the constitutive model
Probabilistic Seismic Hazard Analyses (PSHA). PSHA through hysteretic damping. At small strain levels, damping is
predicts the mean annual rate of exceedance of a ground incorporated through Rayleigh damping, which is fully
motion parameter (e.g., an Intensity Measure). These intensity defined by the equivalent damping ratio at small strains, ξ, and
measures, in turn, can be used to predict structural response in a frequency band where ξ is matched.
what is termed Performance Based Design (PBD). Equations Nonlinear site response is performed for generalized soil
(1) and (3) can be used in PSHA or in PBD for near-fault sites profiles subject to simplified pulse-type input motions so that
provided that the probability of occurrence of the pulse is insights can be made regarding the effects of site conditions in
known, and the cross correlation of PGV and Tv is also known. the near-fault region. Studies by structural engineers have
Thotong et al. [11] present a preliminary model for the shown that these simplified representations are capable of
probability of occurrence of a pulse. The correlation capturing the salient response features of structures subjected
coefficient between ln(Tv) and ln(PGV) for the dataset used in to near-fault ground motions (e.g., [4, 5]). Bray and
this study is 0.24. The positive correlation coefficient implies Rodriguez-Marek [8] developed a simplified representation of
that Tv increases as PGV increases, which is an expected FD velocity-time histories using sine pulses in both the fault-
outcome. However, the residuals of Tv and PGV (i.e., the normal and fault-parallel directions. Ground motions are fully
difference between measured values their estimated values defined by the period of each cycle, their corresponding
using Equations 1 or 3) are uncorrelated. amplitude, and the number of significant pulses (see Figure 2).
The PGV is the largest amplitude of all cycles and the pulse
SEISMIC SITE RESPONSE FOR FD MOTIONS period of the record, Tv, is defined as the period of the cycle
with the largest amplitude. A series of input velocity-time
histories were created by parsing together sequences of sine
Analytical Framework pulses. The amplitude and period of these ground motions
were varied systematically. Pulse periods were varied from
Site conditions were found to be potentially of great 0.6 s to 4.0 s, and pulse amplitudes from 75 to 300 cm/s.
importance in discerning the characteristics of near-fault FD
motions. It should not be surprising that site effects have the
potential for significantly modifying the ground motion at a Generalized Site Profiles
deep or soft soil site compared to that which occurs at depth in
the bedrock. The importance of local site conditions has been Three generalized site profiles were created to represent
highlighted in a large number of empirical and analytical common site classes used in building codes (Very Stiff Soil,
studies and is reflected in most building codes (e.g., the 2006 Stiff Soil, and Soft Soil; corresponding to 2006 IBC Site
International Building Code [12]). classes C, D, and E, respectively). The selected shear wave
velocity profiles are obtained from an extensive database of
There are not a sufficient number of rock and soil recordings shear wave velocity logs of sites located largely within
in close proximity to each other that contain near-fault FD California (Silva, pers. comm. 2000). The shear wave
characteristics to allow a detailed empirical study of site velocities for all three generalized site profiles are shown in
effects. Instead, numerical simulations are utilized. In a study Figure 4. Shear wave velocities at depth (where Vs data is
by Rodriguez-Marek and Bray [13], seismic site response is scarce) for the Site D profile were obtained assuming that
modelled by means of a time-domain finite element analysis shear wave velocities are proportional to (σ'm(z))n, where σ'm
using the fully nonlinear multi-axial total stress soil is the mean effective stress at depth z and n = 0.25 [18]. The
constitutive model of Borja and Amies [14]. Bi-directional soft clay profile (Site E) represents typical Bay Mud sites from
shaking is imposed to explore the combined effects of the the San Francisco Bay region. The density of the stiff soils
more intense fault-normal component and the less intense, but was about 1.9 Mg/m3; whereas the density of Holocene clay
still important, fault-parallel component of motion. One- was 1.6 Mg/m3. The profiles were placed on 3 m of weathered
dimensional propagation of horizontal shear waves is rock that in turn overlies a rock elastic half-space with a shear
modelled by a column of 8-node tri-linear brick elements, wave velocity of 1200 m/s and a density of 2.4 Mg/m3. For
where each node is allowed to move in two horizontal Site D, the depth of the profile was varied from 30 to 200 m to
directions. Stress-free boundary conditions are imposed at the study the effects of variations of depth to bedrock on site
top of the column, and viscous dashpots are placed at the base response.
of the soil column to model the energy absorption of the
 5

Vs (m/s) Vs (m/s) Vs (m/s) Su (kPa)

0 500 1000 0 500 1000 0 500 1000 1500 0 200 400 600
0 0 0 0
Fill
10 10

Soft
Clay
50
10 20 20

Depth (m)
Depth (m)

Depth (m)
100 30 30

Stiff Clay
20
40 40
150
50 50

30
200 60 60
Rock
Selected Profile 70 70
Average and ± one standard deviation

(a) Site C (b) Site D (c) Site E

Figure 4: Three generalized site profiles used in the dynamic analyses and representative values from a set of recorded
Vs profiles in California (Silva, pers. Comm.) (from [13]).

The strength profile of the clayey soils was developed using parallel component velocities lead to larger earthquake-
Su/σv' = 0.8 for the stiff soils and Su/σv' = 0.3 for the soft clay induced shear strains, a softer response, and hence, a larger
in the Site E profile. A lower bound of Su = 150 kPa and “degraded site period.” Different shapes of the input pulse
Su = 25 kPa were used for the stiff and soft clays, respectively. period also affect shear strain levels in the soil. Depending on
The static shear strength was multiplied by a factor of 1.4 to the coincidence of fault-normal and fault-parallel peaks in
account for rate effects during the one primary cycle of rapid velocity, motions can induce larger strains and result in higher
earthquake loading [19]. The nonlinear properties of the soil degraded site periods [17].
were obtained by matching the Borja and Amies [14] model to
Representative results are shown in Figure 5 for a set of bi-
widely used strain-dependent shear modulus reduction and
directional seismic site response analyses performed for a
material damping relationships. The PI = 15 and PI = 30
deep stiff soil site undergoing near-fault FD simplified half-
curves of Vucetic and Dobry [20] were used for the stiff clay
cycle motions. For the relatively low pulse period input
soils, and the curves of Isenhower and Stokoe [21] were used
motion (Figure 5b), there is significant elongation of the pulse
for the soft clays. Additional details are provided in [17].
period due to the soil, but there is not amplification of the
PGV as the input pulse period is not near the degraded period
Discussion of Results of the soil deposit. However, for the 2 s pulse period input
motion (Figure 5a), there is significant amplification of the
The concept of an equivalent-linear “degraded site period” is PGV, because it more closely coincides with the “degraded
still useful for interpreting the results of these nonlinear site period” of the deep, stiff soil deposit.
response analyses to intense near-fault FD motions. The
degraded site period (Ts') is calculated as: Ts' = 4H/Vs', where The relationship between output (PGVsoil) and input (PGVrock)
H is the soil depth and Vs' is the average effective shear wave intensities for all site response analyses for the Stiff Soil
velocity of the soil deposit using a shear modulus that is profile (IBC Site D with soil depths ranging from 30 to 200 m)
consistent with the effective shear strain induced in each layer are shown in Figure 6. The ratio of PGVsoil to PGVrock is
of soil (i.e., γeff = (n)γmax, where n is nearly 1.0 for pulse-type generally between one and two. As a comparison to these
motions). The “degraded site period” increases with increasing analytical results, the results from an empirical study by Silva
soil depth, decreasing soil stiffness, and increasing soil (pers. comm. 1998) exhibit a trend that is consistent with the
nonlinearity resulting from more intense rock motions. results from these analyses. The computed amplification of
PGV also agrees fairly well with the mid-period amplification
The largest amplification of PGV through a soil site occurs factor for Site D in the 2006 IBC, suggesting that spectral
near its “degraded site period.” The pulse period of the soil amplification factors in the mid-period range (T ~ 1 s) are
motion tends to approach the “degraded site period” when it is consistent with these PGV amplification factors. As indicated
initially lower than the “degraded site period.” Hence, input by the results shown in Figure 5, significant amplification still
pulses that have lower periods undergo more elongation than occurs for intense rock motions. Hence, the amount of
those with higher periods. When the input velocity pulse nonlinearity in the amplification of PGV (or mid-period
period is higher than the “degraded site period,” the soil spectral acceleration amplification) is fairly minor for stiff soil
deposit has a pseudo-rigid body response and pulse period is profiles. Closer examination of the results in Figure 6 indicate
not affected. Although the stronger fault-normal component of that generally more amplification of PGV occurs for rock
near-fault FD motion is more critical, the fault-parallel input motions with higher pulse periods and less amplification
component can also affect the response of a site. Larger fault- occurs for input motions with lower pulse periods.
6

a) b)
Figure 5: Responses of a deep, stiff soil site to a half-sine-pulse input rock motion with PGV = 120 cm/s: (a) results
for an input pulse period of 2.0 s, and (b) results for an input pulse period of 0.6 s.
600

500

400
2 1
PGVsoil (cm/s)

1 1
300

200

Pulse Motions
Tv = 0.6 s
100 Tv = 1.0 s
Tv = 2.0 s
Tv = 4.0 s
0
0 100 200 300 400 500 600
PGVrock (cm/s)

× Pacoima Dam, San Fernando EQ.

○ Gilroy Gavilan College, Loma Prieta EQ

□ Pacoima Dam Downstream, Northridge EQ

∆ KJMA, Kobe EQ
Average for near-fault motions:
♦
M = 6 – 7, R < 10 km (Silva, pers. comm. 1998) comm.)
Figure 6: Calculated PGV amplification at Stiff Clay profiles (see [17] for more details).

Seismic site response analyses of the Soft Soil site (Site E) do not capture soil failure. Although seismic site response can
indicate that earthquake-induced shear stresses fully mobilized be limited by the yield strength of soft soil, seismic instability
the dynamic strength of the soft clay for even relatively effects must now be evaluated.
moderate near-fault motions. Thus, the soft soil deposit’s low
strength limits the site’s peak seismic response. Significant Seismic site response also affects the pulse period of the near-
variations in the shape of the soil’s strain-dependent shear fault FD motion. Deep soil deposits with long degraded site
modulus reduction and material damping curves at periods tend to lengthen the pulse period of the input rock
intermediate strain levels do not affect the calculated PGV or motion for all cases except when the rock pulse period is much
Tv of the surface motion. Soil yielding leads to a significantly greater than the degraded site period. Representative results
higher “degraded site period” with a significant amount of for IBC 2006 Site D profiles are shown in Figure 8. The ratio
energy dissipation through plastic yielding. This leads to of soil to rock pulse period may be as high as 2 for short pulse
relatively more velocity period elongation at soft soil sites and periods (Tv < 1 s), and this ratio approaches one as the input
greater attenuation of input rock PGV for intense rock motions rock pulse period exceeds a few seconds. A greater amount of
than at stiff soil sites (Figure 7). This attenuation of pulse pulse period elongation occurs for very intense FD motions
amplitude is not reflected in mid-period amplification factors because of the greater nonlinearity in the seismic response of
in the 2006 IBC, because these factors are largely obtained the soil at these intense levels of shaking. Hence, site effects
from extrapolation of empirical site amplification factors for are an important consideration in characterizing near-fault FD
less intense motions and from equivalent-linear analyses that ground motions for use in design.
 7

Figure 7: Results of seismic analyses of the Soft Clay and Stiff Clay profiles with soil depths of 60 m. Results are shown
for various pulse shapes (see [17] for more details). Also shown for comparison are the mid-period
amplification factors (Fv) from the 2006 IBC, and the predictions using the empirical model in Equation (1).

Findings from Analytical Study unique nature of FD motions. Forward-directivity motions are
often intense, pulse-type motions, which are significantly
The characteristics of the near-fault FD motions at the surface different from ordinary ground motions. These motions are
of a soil deposit are primarily influenced by the characteristics best described by their velocity-time history, which requires
of the input rock motion (i.e., its intensity, pulse period, and estimation of its peak ground velocity (PGV), predominant
number of significant cycles) and the characteristics of the soil pulse period (Tv), and number of significant velocity pulses
profile (i.e., soil type, stiffness, and depth to bedrock). (Nc).
Importantly, the PGV of the motion computed at the top of the
soil is generally larger than the PGV of the rock input motion, Using recent FD motions, empirical relationships have been
with the exception of input motions with large intensities and updated for estimating the PGV and Tv of near-fault FD
short pulse periods. The Tv of the soil motion also motions. PGV varies significantly with magnitude, distance,
systematically increases with increasing soil depth or and site effects. Tv is a function of magnitude and site
increasing rock input motion intensity. Soil stiffness also conditions with most of the energy being concentrated within
affects the amplitude and period of input pulses. The largest a narrow-period band centred on the pulse period. As lower
amplification of PGV for the Very Stiff Soil profile (Site C) magnitude events produce lower pulse periods, which better
occurs for input motions with short pulse periods, whereas the matches the low natural period of common buildings in urban
largest amplification of PGV for the Stiff Soil profile (Site D) areas, FD ground motions from these events have the potential
occurs at intermediate periods. to produce more damage than higher magnitude earthquakes
in the near-fault region. Empirical relationships cannot be used
Site conditions play an important role in shaping the at this time to predict Nc, because it depends on details of the
characteristics of near-fault FD motions at soil sites. Thus, the rupture mechanism that cannot be known a priori. However, it
important influence of local soil conditions on FD motions is most likely that near-fault FD motions have only one or two
should be considered when designing structures in the near- significant cycles of motion. The compact FD wave form
fault region. Site conditions affect the amplitude of the surface produces intense motions that are of short duration. Design
motion (i.e., its PGV) and its frequency content (i.e., its Tv). near-fault FD velocity-time histories should not have a large
Fully nonlinear site-specific response analysis is required to number of significant cycles of motion.
capture the nonlinear response of soil deposits under the
intense levels of shaking of FD motions. As guidance for this Fully nonlinear bi-directional shaking simulations confirm
site-specific analysis, the likely range of site and intensity indications from empirical evidence that site effects are
dependent amplification factors for PGV can be estimated important to consider in the near-fault region. It is shown that
using Figures 6 and 7, and the amount of pulse period site effects generally cause Tv to increase, and that
elongation can be estimated using Figure 8. amplification of PGV depends on site properties, but
amplification is generally observed even for very intense rock
motions. At soft soil sites, seismic site response can be limited
CONCLUSIONS by the yield strength of the soil. In these cases, the seismic
Near-fault forward-directivity motions typically govern the stability of the site and the building’s foundation elements
design of structures built close to active faults. Hence, ground should be evaluated in terms of seismically induced permanent
motions for use in evaluating designs in the near-fault region deformations.
should be selected carefully to represent satisfactorily the
8

Figure 8: Pulse period of soil motion normalized by pulse period of rock motion vs. rock motion pulse period. Results
are from seismic site response analyses for IBC 2006 Site C and D profiles. The heavy line represents the
results from the Bray and Rodriguez-Marek (2004) regression of empirical records (modified from [13]).

9 Gillie J.L. (2005) Nonlinear Response Spectra of

ACKNOWLEDGEMENTS Forward-Directivity Ground Motions. MSc Thesis,
Financial support was provided by the Pacific Earthquake Washington State University, Pullman, WA.
Engineering Research Center through Award 2011999 and by 10 Abrahamson, N.A., and Youngs, R.R. (1992) “A stable
the David and Lucile Packard Foundation. The authors wish to algorithm for regression analyses using the random
thank Dr. Walter Silva of Pacific Engineering and Analysis for effects model”. Bulletin of the. Seismological Society of
sharing his ground motion database and soil profile America, 82(1), 505-510.
information. Professor Pestana of the University of California
provided useful comments on the soil modelling. 11 Tothong P., Cornell C.A., and Baker J.W. (2007)
“Explicit directivity-pulse inclusion in probabilistic
seismic hazard analysis”, Earthquake Spectra, 23 (4),
REFERENCES 867-891.
1 Somerville, P.G., Smith, N.F., Graves, R.W., and 12 International Code Council (2006) International Building
Abrahamson, N.A. (1997) “Modification of empirical Code, Country Club Hills, IL.
strong ground motion attenuation relations to include the
amplitude and duration effects of rupture directivity”. 13 Rodriguez-Marek, A. and Bray, J.D. (2006) “Seismic site
Seismological Research Letters, 68(1), 199-222. effects for near-fault forward directivity ground motions”,
Journal of Geotechnical and Geoenvironmental
2 Stewart, J.P., Chiou, S-J, Bray, J.D., Graves, R.W., Engineering, ASCE, 132(12), 1611-1620.
Somerville, P.G., Abrahamson, N.A. (2001) “Ground
Motion Evaluation Procedures For Performance-Based 14 Borja, R.I., and Amies, A.P. (1994) “Multiaxial cyclic
Design, PEER-2001/09”. Pacific EQ Engrg. Research plasticity model for clays”. Journal of Geotechnical
Center, Univ. of Calif., Berkeley, Sep., 229 pages. Engineering, ASCE, 120(6), 1051-1070.

3 Abrahamson, N.A. (2000) “Effects of rupture directivity 15 Lysmer J.M., and Kuhlmeyer, A.M. (1969) “Finite
on probabilistic seismic hazard analysis”. Proceedings, dynamic model for infinite media”. Journal of the
Sixth International Conference on Seismic Zonation, Engineering Mechanics Division, ASCE, 95(4) 859-877.
Palm Springs, CA, Nov. 12-15. 16 Espinoza R.D., Bray, J.D., Soga, K., and Taylor, R.L.
4 Alavi, B., and Krawinkler, H. (2000) “Consideration of (1995) GeoFEAP: Geotechnical Finite Element Analysis
near-fault ground motion effects in seismic design”. Program, Report UCB/GT/95-05, Dept. Civil Eng., Univ.
Proceedings, 12th World Conf. on Earthquake of California, Berkeley.
Engineering., Auckland, New Zealand. 17 Rodriguez-Marek, A. (2000) Near-Fault Seismic Site
5 Sasani, M. and Bertero, V.V. (2000) “Importance of Response. Ph. D. Dissertation. Univ. of Calif., Berkeley.
severe pulse-type ground motions in performance-based 18 Hardin, B. O., and Drnevich, V. P. (1972) “Shear
engineering: historical and critical review”. Proc., 12th modulus and damping in soils: measurements and
World Conf. on EQ Engrg., Auckland, New Zealand. parameter effects”. Journal of the Soil Mechanics and
6 Somerville, P.G. (1998) “Development of an improved Foundation. Engineering, ASCE, 98(SM6), 603-624.
ground motion representation for near-fault ground 19 Lefebvre, G., and LeBoeuf, D., (1987) “Rate effects and
motions.” SMIP 98, Seminar on Utilization of Strong cyclic loading of sensitive clays.” Journal of.
Motion Data: Oakland, CA. Geotechnical Engineering, ASCE, 113(5), 476-489.
7 Mavroeidis, G.P., and Papageorgiou A.S. (2003) “A 20 Vucetic, M., and Dobry, R. (1991) “Effect of Soil
mathematical representation of near-fault ground Plasticity on Cyclic Response”. Journal of. Geotechnical
motions.” Bulletin of the Seismological Society of Engineering, ASCE, 117(1), 89-107.
America, 93(3), 1999-1131.
21 Isenhower, W. M., and Stokoe, K. H. II, (1981) “Strain-
8 Bray, J.D., and Rodriguez-Marek, A., (2004) “Charac- rate dependent shear modulus of San Francisco Bay
terization of forward-directivity ground motions in the mud”. Inter. Conf. on Recent Advances in Geotechnical
near-fault region.” Soil Dynamics and Earthquake EQ Engineering., Univ. of Missouri, Rolla, 597-602.
Engineering, 24, 815-828.