Estimating Liquefaction-Induced Lateral Displacements

Using the Standard Penetration Test

or Cone Penetration Test
G. Zhang1; P. K. Robertson, M.ASCE2; and R. W. I. Brachman3

Abstract: A semiempirical approach to estimate liquefaction-induced lateral displacements using standard penetration test 共SPT兲 or cone
penetration test 共CPT兲 data is presented. The approach combines available SPT- and CPT-based methods to evaluate liquefaction potential
with laboratory test results for clean sands to estimate the potential maximum cyclic shear strains for saturated sandy soils under seismic
loading. A lateral displacement index is then introduced, which is obtained by integrating the maximum cyclic shear strains with depth.
Empirical correlations from case history data are proposed between actual lateral displacement, the lateral displacement index, and
geometric parameters characterizing ground geometry for gently sloping ground without a free face, level ground with a free face, and
gently sloping ground with a free face. The proposed approach can be applied to obtain preliminary estimates of the magnitude of lateral
displacements associated with a liquefaction-induced lateral spread.
DOI: 10.1061/共ASCE兲1090-0241共2004兲130:8共861兲
CE Database subject headings: Liquefaction; Lateral displacement; Earthquakes; Sand; In situ tests.

Introduction direct methods to estimate liquefaction-induced ground deforma-

tions for low-to medium-risk projects and to provide preliminary
Earthquake shaking may trigger the liquefaction of a saturated estimates for high-risk projects.
sandy soil in the ground. During past major earthquakes, enor- Methods using standard penetration test 共SPT兲 data are avail-
mous damage to engineered structures and lifelines has been able for estimating lateral displacements in a liquefaction-induced
caused by liquefaction-induced ground failures 共e.g., Hamada and lateral spread 共Rauch and Martin 2000; Bardet et al. 2002; Youd
O’Rourke 1992兲. Generally, liquefaction-induced ground failures et al. 2002兲. These methods are empirical and do not incorporate
include flow slides, lateral spreads, ground settlements, ground the extensive knowledge gained from laboratory studies of lique-
oscillation, and sand boils. Lateral spreads are the pervasive types faction. Further, even though the cone penetration test 共CPT兲 has
of liquefaction-induced ground failures for gentle slopes or for greater repeatability and reliability, and provides a continuous
nearly level 共or gently inclined兲 ground with a free face 共e.g., profile compared with other field tests, no CPT-based method to
river banks, road cuts兲. This paper focuses on the estimation of estimate liquefaction-induced lateral displacements is currently
available.
lateral displacements associated with liquefaction-induced lateral
The objective of this paper is to present an approach for esti-
spreads.
mating liquefaction-induced lateral displacements using SPT or
Several methods have been proposed to estimate liquefaction-
CPT data. The approach combines available SPT- or CPT-based
induced lateral ground displacements including numerical mod-
methods to estimate liquefaction potential with laboratory test
els, laboratory tests, and field-test-based methods. Challenges as-
results for clean sand to estimate the potential maximum cyclic
sociated with sampling loose sandy soils limit the applications of
shear strains for saturated sandy soils under seismic loading. Case
numerical and laboratory testing approaches in routine practice.
history data are used to develop empirical correlations for lateral
Field-test-based methods are likely best suited to provide simple
displacement for: 共1兲 gently sloping ground without a free face,
共2兲 level ground with a free face, and 共3兲 gently sloping ground
1
Project Engineer, EBA Engineering Consultants Ltd., 14940-123 with a free face.
Ave., Edmonton AB, Canada T5V 1B4.
2
Associate Vice President 共Research/Industry兲 and Professor, Dept. of
Geotechnical Engineering, Univ. of Alberta, Edmonton AB, Canada T6G Mechanism of Liquefaction-Induced Lateral Spreads
2G7.
One-g shake table tests 共e.g., Sasaki et al. 1991; Yasuda et al.
3
Assistant Professor, GeoEngineering Centre at Queen’s-RMC,
Queen’s Univ., Kingston ON, Canada K7L 3N6 共corresponding author兲. 1992兲 and centrifuge model tests 共e.g., Abdoun 1997; Taboada-
E-mail: brachman@civil.queensu.ca Urtuzuastegui and Dobry 1998兲 have been conducted to investi-
Note. Discussion open until January 1, 2005. Separate discussions
gate the mechanisms of liquefaction-induced ground lateral
must be submitted for individual papers. To extend the closing date by
spreads. These tests generally support the hypothesis that lateral
one month, a written request must be filed with the ASCE Managing
Editor. The manuscript for this paper was submitted for review and pos- spreads result from distributed residual shear strains throughout
sible publication on December 19, 2001; approved on December 21, the liquefied layers. The residual shear strains in liquefied layers
2003. This paper is part of the Journal of Geotechnical and Geoenvi- are primarily a function of: 共1兲 maximum cyclic shear strains
ronmental Engineering, Vol. 130, No. 8, August 1, 2004. ©ASCE, ISSN ␥ max , and 共2兲 biased in situ static shear stresses. In this paper,
1090-0241/2004/8-861– 871/$18.00. ␥ max refers to the maximum amplitude of cyclic shear strains that

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004 / 861

are induced during undrained cyclic loading for a saturated sandy
soil without biased static shear stresses in the direction of cyclic
loading. Biased in situ static shear stresses are mainly controlled
by ground geometry at the site 共e.g., ground slope, free face
height, and the distance to a free face兲. The thickness of liquefied
layers will also influence the magnitude of lateral displacements,
with greater lateral displacements for thicker liquefied layers.
Both ␥ max and the thickness of liquefied layers are affected by soil
properties and earthquake characteristics.

Estimation of Maximum Shear Strains from

Standard Penetration Test or Cone Penetration Test
Data

Maximum Cyclic Shear Strains of Clean Sands from

Laboratory Tests Fig. 1. Relationship between maximum cyclic shear strain and factor
of safety for different relative densities D r for clean sands 关based on
Nagase and Ishihara 共1988兲 conducted cyclic simple shear tests data from Ishihara and Yoshimine 共1992兲 and Seed 共1979兲兴
on saturated loose, medium-dense, and dense samples of clean
Fuji River sand under loading conditions without biased static hawy and Mayne 1991兲 for a given SPT N value and is applicable
shear stresses. Based mainly on these laboratory tests, Ishihara for relatively young 共⬍5,000 years兲 or normally consolidated
and Yoshimine 共1992兲 established the relationship between the sand.
␥ max and the factor of safety 共FS兲 against liquefaction for different The correlation between D r and cone tip resistance (q c ) of
relative densities (D r ) of clean sands. Modification of this rela- Tatsuoka et al. 共1990兲 was used
tionship is required to account for dilative response of the soil that
D r ⫽⫺85⫹76 log共 q c1N 兲 共 q c1N ⭐200兲 (2)
may restrict the development of shear strains. Evidence of soil
dilative response at large shear strains has been observed in und- where q c1N ⫽normalized CPT tip resistance corrected for effec-
rained cyclic laboratory tests, one-g shake table studies, centri- tive overburden stresses corresponding to 100 kPa 共Robertson and
fuge experiments, and in situ seismic responses 共Elgamal et al. Wride 1998兲. This correlation provides slightly smaller and more
1998兲. Seed 共1979兲 postulated that only a limited amount of shear conservative estimates of relative density than the correlation by
strain could be developed for sand at any given relative density, Jamiolkowski et al. 共1985兲 when q c1N is less than about 100.
regardless of the number of stress cycles applied, and that further
increases in strain could be difficult to achieve unless the full Correction for Grain Characteristics
undrained resistance of the soil was exceeded. In this study, the
relationship between ␥ max and FS developed by Ishihara and The relationship between ␥ max and FS proposed by Ishihara and
Yoshimine 共1992兲 were modified by limiting the maximum shear Yoshimine 共1992兲 was developed based on laboratory test results
strains as proposed by Seed 共1979兲. on clean sands. The equivalent clean sand normalized SPT N
value, (N 1 ) 60cs , and the equivalent clean sand normalized CPT
penetration resistance, (q c1N ) cs , is used to account for the effect
Factor of Safety and Relative Density from the of grain characteristics 共or fines content兲 on SPT N values or CPT
Standard Penetration Test and Cone Penetration Test soundings in evaluating liquefaction potential of silty sands 共Youd
Both the factor of safety against liquefaction and relative density et al. 2001兲. In this work, (N 1 ) 60cs or (q c1N ) cs , was also applied
(D r ) are needed to estimate ␥ max of a clean sand for a given to quantify the effect of grain characteristics 共or fines content兲 on
earthquake. FS can be evaluated from liquefaction potential SPT N values or CPT soundings in estimating ␥ max . The param-
analysis using the SPT- or CPT-based methods summarized by eter, (N 1 ) 60cs or (q c1N ) cs for a silty sand, can then be treated as the
Youd et al. 共2001兲, hereafter referred to as the NCEER 共National SPT N value or CPT cone tip resistance for a clean sand and be
Center for Earthquake Engineering Research兲 SPT-based method used directly to estimate ␥ max . This assumes that the effect of
or the NCEER CPT-based method. grain characteristics or fines content on lateral spreading is similar
Relative densities may be estimated from correlations with to its effect on liquefaction triggering. This assumption will be
either SPT or CPT results. A modified version of Meyerhof’s discussed further later in this paper.
共1957兲 correlation was used to estimate relative densities of a Fig. 1 shows the relationships used in this study between ␥ max
clean sand from SPT blow counts and FS for different D r . With either SPT or CPT data and the
parameters for a given earthquake, (N 1 ) 60cs or (q c1N ) cs and FS for
D r ⫽16• 冑共 N 1 兲 78⫽14• 冑共 N 1 兲 60 关共 N 1 兲 60⭐42兴 (1) sandy soils can be estimated from the liquefaction potential analy-
where D r ⫽relative density of a clean sand as a percentage; sis using the NCEER SPT- or CPT-based method. An estimate of
(N 1 ) 60⫽normalized SPT N value corrected for the rod energy ␥ max can then be obtained from Fig. 1 for every reading in the
ratio 共60% reference energy兲, overburden effective stress 共100 SPT or CPT results.
kPa reference effective stress兲, rod length, borehole diameter, and
sampling method, as discussed in Youd et al. 共2001兲; and (N 1 ) 78
is equal to (N 1 ) 60/1.3. Generally, the calculated relative density Lateral Displacement Index
using Eq. 共1兲 is reasonably consistent with those calculated using Integrating the calculated ␥ max values with depth will produce a
other available correlations 共Seed 1979; Skempton 1986; Kul- value that is defined as the lateral displacement index 共LDI兲

862 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004

 Fig. 2. Example plots illustrating the major steps for calculating the lateral displacement index 共LDI兲 using the CPT-based approach

LDI⫽ 冕 0
Z max
␥ maxdz (3)
Case Histories

A total of 13 available case histories associated with 12 past

where Z max⫽maximum depth below all the potential liquefiable major earthquakes were studied as listed in Table 1. A summary
layers with a calculated FS ⬍2.0. However, evaluation of lique- of each case history and the rational behind the selection of data
faction at depths greater than 23 m is beyond the range where the for use in this paper can be found in Zhang 共2001兲. Several case
NCEER CPT and SPT methods have been verified and where histories involving lateral spreads that were impeded by shear
routine applications should be applied 共Youd et al. 2001兲. forces along the margins of the failure zone or by engineered
Figs. 2共a–f兲 illustrate the major steps to calculate LDI using structures 共e.g., retaining walls, quay walls兲 were not used in this
CPT data. Figs. 2共a and b兲 show the CPT tip resistance (q c ) and study. For example, data from the 1995 Hyogoken-Nambu 共Kobe兲
sleeve friction ratio (R f ), which can be calculated directly from earthquake were not used in this study since lateral spreading in
CPT soundings. Based mainly on q c and R f , the equivalent clean the Port area was believed to be greatly restricted by caisson-type
sand normalized tip resistance (q c1N ) cs can be calculated using quay walls 共Zhang 2001兲. Lateral spreads that were possibly as-
sociated with multiple failure mechanisms 共e.g., several case sites
the NCEER CPT-based method, Fig. 2共c兲. FS can be estimated
from the 1994 Northridge Earthquake兲 were also not included in
using the NCEER CPT-based method with (q c1N ) cs and the de-
the database either. The possibility of local slump failure or flow
sign earthquake parameters 关Fig. 2共d兲兴. Note that in Fig. 2共d兲, a
failure increases in the zone close to a free face with a steep slope
soil layer below 8 m is estimated to be a nonliquefiable clayey
due to the presence of high static shear stresses. Further, the
soil based on CPT data and is assigned a large value of FS mechanism associated with a local slump failure or flow failure is
共⬎2.0兲. Fig. 2共e兲 shows ␥ max as estimated from the relationships fundamentally different from that for lateral spreading. Therefore
in Fig. 1. Finally, the lateral displacement index 共LDI兲 can be some case sites 共e.g., Sandholdt Road at Moss Landing兲 where the
calculated by integrating the ␥ max with depth 关Fig. 2共f兲兴. LDI is horizontal distance from the free face toe is smaller than four
the integrated value at the ground surface. The major procedures times the free face height were excluded from the database. Se-
to calculate the LDI using SPT data are similar to those described lection of cases also required that SPT or CPT data were available
above, although the profile may be less detailed than that pro- at locations within 100 m of where liquefaction-induced lateral
duced from CPT data. displacement was measured. Case histories were divided into
Although LDI has the units of displacement, it is intended only those with: 共1兲 gently sloping ground without a free face, 共2兲
to provide an index to quantify potential lateral displacements for nearly level ground with a free face, and 共3兲 gently sloping
a given soil profile, soil properties, and earthquake characteristics. ground with a free face.
The actual magnitude of lateral displacement depends on both A total of 291 measured values of liquefaction-induced lateral
LDI and geometric parameters characterizing ground geometry. spreading displacements from the 13 case histories were used in
In the following sections, lateral spread case histories are used to this study. There were usually more than one distinct lateral
account for the influence of ground geometry on lateral displace- spread features/sites for each case history studied, as listed in
ments. Tables 2– 4. Multiple lateral displacements with different mea-

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004 / 863

Table 1. Cases Selected for Database on Liquefaction Lateral Spreads
Liquefied soil
Case References Liquefied soil description thickness 共m兲
Hokkaido 1993 Isoyama 共1994兲 Silty sand with about 10% fines 4.5–5.5
Dagupan 1990 Wakamatsu et al. 共1992兲; Ishihara et al. 共1993兲; Tokimatsu et al. 共1994兲 Fine clean sand and silty sand 5.1– 8.6
Moss Landing 1989 O’Rourke and Pease 共1992兲; Boulanger et al. 共1995, 1997兲; Mejia 共1998兲 Clean sand and silty sand 0.8 –5.2
Wildlife 1987 Bennett et al. 共1984兲; Holzer et al. 共1989兲; Youd and Bartlett 共1988兲; Silty sand or sandy silt with 0.5–2.7
Dobry et al. 共1992兲 17–50% fines
Noshiro 1983 Hamada et al. 共1986兲; Bartlett 共1991兲; Hamada 共1992b兲 Clean dune sand and alluvial sand 1.0–7.0
Heber Road 1979 Bennett et al. 共1981兲; Norton 共1983兲; Youd and Bennett 共1983兲; Channel sand with 14 –37% fines 0.0–5.3
Dobry et al. 共1992兲
Juvenile Hall 1971 Bennett 共1989兲; O’Rourke et al. 共1992b兲 Alluvium with 50– 80% fines 0.9–3.0
Jensen Plant 1971 O’Rourke et al. 共1992b兲 Alluvium with 32– 62% fines 0.0–7.7
Niigata 1964 Bartlett 共1991兲; Hamada 共1992a兲; Bartlett and Youd 共1995兲; Mainly clean, fine or medium 0.6 –18.1
Hamada et al. 共1986兲 sands
Alaska 1964 Bartlett 共1991兲; Bartlett and Youd 共1995兲; McCulloch and Bonilla 共1970兲 Sand and gravel, silty sand, sandy 4.0–9.4
silt
Fukui 1948 Hamada et al. 共1992b兲; Rauch 共1997兲 Silt, silty sand, sand, sandy gravel 1.7–5.7
Kanto 1923 Hamada et al. 共1992a兲; Rauch 共1997兲 Fine sand with about 10% fines 8.0– 8.8
San Francisco 1906 Pease and O’Rourke 共1993, 1998兲; O’Rourke et al. 共1992a兲; Relatively clean dune sand 1.3– 8.5
Youd and Hoose 共1976, 1978兲

sured values at a given case site were collected for most of the Correlation between Actual Lateral Displacement,
case histories. These points corresponded to either different local Lateral Displacement Index, and Geometric
ground geometric parameters or soil profile/properties. The mini- Parameters
mum spacing of any two points collected for this study was about
20 m or more. Measurements of lateral displacement that were
deemed to be too close to the boundaries of a lateral spread were Gently Sloping Ground without a Free Face
not included in the database.
In situ test results from 150 SPTs and 41 CPTs associated with Cases with Standard Penetration Test Data
the studied case histories were collected. Of the 13 case histories, A total of five case histories with SPT data were studied for the
SPT data were available for nine of the cases and CPT data were case of gently sloping ground without a free face, as listed in
available only for five of the cases. The case history data were Table 2. The corresponding measured lateral displacement values
collected from the available publications as listed in Table 1. Par- or ranges are also listed and are labeled LD. For each case, the
tial data, especially for the 1964 Niigata earthquake, were ob- NCEER SPT-based method was applied to evaluate liquefaction
tained from the database compiled by Bartlett 共1991兲. potential using SPT data. The lateral displacement index 共LDI兲
In most cases the magnitude of lateral displacement were ob- was then calculated based on the results of the liquefaction po-
tained from aerial photos that were taken before and after the tential analysis, the relationship in Fig. 1, and using Eq. 共3兲. An
earthquake or from reports of dislocated or offset buildings, average of the calculated values of LDI was used as a represen-
bridge components, fences, canals, etc. The estimated values of tative LDI value if more than one SPT profile was conducted
accuracy 共or potential error兲 associated with the measured dis- close to where LD was measured. A total of 132 data sets were
placements are also given in Tables 2– 4. obtained from the five case histories with SPT data.

Table 2. Cases with Gently Sloping Ground without a Free Face

Noshiro Juvenile Hall Niigata Fukui Kanto San Francisco
Case history 1983 1971 1964 1948 1923 1906
Number of sites 3 1 3 2 1 2
Number of LD data 23 1 103 4 1 8
LD 共cm兲 65–298 168 80– 460 100–350 250 60–210
Accuracy of LD 共⫾cm兲 17 2 72 190 ⬎50 10–50
LDI 共cm兲 17–164 60 SPT 37–538 19–250 184 –232 33–173
18 CPT
S (%) 0.2–7.5 1.2 0.2–0.9 0.4 –3.4 1.6 1.0–2.3
Mw 7.7 6.4 7.5 7.0 7.9 7.9
a max (g) 0.25 0.55 0.19 0.25 0.25 0.6
Number of SPT and CPT 21 SPT 5 SPT 27 SPT 3 SPT 2 SPT 17 CPT
5 CPT
Note: LD⫽lateral displacement; LDI⫽lateral displacement index; SPT⫽standard penetration test; and CPT⫽cone penetration test.

864 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004

Table 3. Cases with Nearly Level Ground and a Free Face
Moss Jensen
Dagupan Landing Plant Niigata Alaska
Case history 1990 1989 1971 1964 1964
Number of sites 3 2 1 2 1
Number of LD data 7 6 13 66 1
LD 共cm兲 50– 600 30–125 2–100 41–1015 157
Accuracy of LD ⬎50 ⬎10 47 72 10–50
共⫾cm兲
LDI 共cm兲 79–220 28 –114 5–20 246 – 637 173
H (m) 4 –11.5 1.9–2.4 10.4 –17.2 4.9–5.2 4.9
L/H 3.8 –27.3 6.3–23.5 8.7–30.5 5.1–36.2 6.2
Mw 7.6 7.0 6.4 7.5 9.2
a max (g) 0.2 0.25 0.55 0.19 0.31
Number of SPT 3 7 20 47 3
and CPT 共SPT兲 共CPT兲 共SPT兲 共SPT兲 共SPT兲
Note: LD⫽lateral displacement; LDI⫽lateral displacement index; SPT
⫽standard penetration test; and CPT⫽cone penetration test.

Fig. 3共a兲 is a plot of LD/LDI versus ground slope S for the five
case histories. The ground slope is defined as the average gradient
over relatively long 共20 m or more兲 ground with a similar general
gradient. A general trend of increasing LD/LDI with increasing
ground slope can be seen from Fig. 3共a兲 that can be expressed as
LD
⫽S⫹0.2 共 for 0.2%⬍S⬍3.5% 兲 (4)
LDI
where S⫽ground slope as a percentage.
A dominant portion 共95%兲 of the data in Fig. 3共a兲 was col-
lected from two Japanese case histories 共Niigata and Noshiro兲.
The liquefied soils at these sites were generally clean sands. Be-
cause the ground slopes at all the sites for the Niigata case history
were between 0.2 and 0.9% and with an average of about 0.5%,
the data for this case history dominates the trend line in Fig. 3共a兲 Fig. 3. Ratio of measured lateral displacement lateral displacement
for a ground slope less than 1%. The data for the Noshiro case 共LD兲 to lateral displacement index lateral displacement index 共LDI兲
versus ground slope S for case histories with gently sloping ground
history dominates the relationship in Fig. 3共a兲 for a ground slope
without a free face: 共a兲 standard penetration test-based data and 共b兲
greater than 1%, especially for a ground slope greater than 3.5%.
cone penetration test-based data
The data from the other three case histories generally fits well
with the relationship in Fig. 3共a兲 for ground slopes ranging from
about 0.5 to 3.5%. Given the exclusive reliance on the Noshiro Cases with Cone Penetration Test Data
case history for ground slopes greater than 4%, the recommended San Francisco and Juvenile Hall are two cases with CPT data and
range of ground slope for Eq. 共4兲 is between 0.2 and 3.5%. gently sloping ground without a free face. Eight data points were
obtained for the San Francisco case history. Only one set of data
was obtained for the Juvenile Hall case history because of a uni-
Table 4. Cases with Gently Sloping Ground and a Free Face form measured lateral displacement of 1.68 m and a ground slope
Heber of 1.2% at the main part of the lateral spreading section where
Hokkaido Wildlife Road Niigata most of the penetration tests were conducted. Fig. 3共b兲 is a plot of
Case history 1993 1987 1979 1964 LD/LDI versus ground slope S for the 9 CPT-based data sets as
well as the trend line that was developed above based on SPT-
Number of sites 1 1 1 1
Number of LD data 6 3 17 32
based data.
LD 共cm兲 96 –286 11–18 30– 424 67– 617 The point from the Juvenile Hall case history associated with
Accuracy of LD 共⫾cm兲 22 2 10–50 72 the 1971 San Fernando earthquake is far removed from the trend
LDI 共cm兲 130–210 24 –35 82–192 78 – 402 line in Fig. 3共b兲. It is believed that the main reason for this in-
S (%) 0.7, 0.8 ⫺0.47 1.5 ⫺0.34 –0.48 consistency is because the NCEER CPT-based method generally
H (m) 1.9–2.5 2.4 1.6 3.4 –5.2 treats soils with a soil behavior type index I c greater than 2.6 as
L/H 17–36 6.9–10.2 7.6 –25 4.9–37.3 nonliquefiable. The soil behavior type index is calculated from
Mw 7.7 6.5 6.5 7.5 CPT data and provides an index of the soil grain characteristics
a max (g) 0.25 0.21 0.6 0.19 共Robertson and Wride 1998兲. An I c equal to 2.6 corresponds to a
Number of SPT and CPT 4 4 8 15 calculated fines content of about 50– 60% 共Zhang 2001兲. Since
共SPT兲 共CPT兲 共CPT兲 共SPT兲 Bennett 共1989兲 reported that the fines contents for the liquefied
Note: LD⫽lateral displacement; LDI⫽lateral displacement index; SPT soils at the Juvenile Hall case site ranged from 50 to 80% with an
⫽standard penetration test; and CPT⫽cone penetration test. average of 65%, this implies that the majority of the liquefied

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004 / 865

 Level Ground with a Free Face
In this study, terrain with a slope of less than 0.15% is defined as
‘‘level’’ ground. Two major geometric parameters characterizing
ground geometry for level ground with a free face are free face
height H, which is the elevation difference between the level
ground surface and the toe of a free face, and the horizontal
distance L from the toe of a free face 共see the inset of Fig. 4兲.
Bartlett 共1991兲 observed from case histories that lateral displace-
ments were greater closer to a free face and for higher free faces.
Similar trends were observed at the Kobe Port area during the
1995 Hyogoken-Nambu earthquake 共Ishihara et al. 1996兲 even
though these lateral displacements are believed to have been re-
stricted by quay walls. The ratio of L/H was adopted to establish
a correlation between LD and LDI for level ground with a free
face.

Cases with Standard Penetration Test Data

Four case histories with SPT data for level ground conditions with
a free face are available, as listed in Table 3. A total of 87 data
points from the four case histories associated with four different
earthquakes are plotted in Fig. 4共a兲. The data in Fig. 4共a兲 is pre-
dominantly from the 1964 Niigata case history, however, the re-
sults from the other case histories generally follow the same
trend. Although there is considerable scatter, Fig. 4共a兲 shows a
trend of decreasing LD/LDI with increasing L/H that may be
characterized by
LD
LDI
⫽6•
L
H 冉冊 ⫺0.8
共 for 4⬍L/H⬍40兲 (5)

This general trend is applicable for the range of earthquake and

ground conditions listed in Table 3.

Case with Cone Penetration Test Data

Fig. 4. Ratio of measured lateral displacement lateral displacement Fig. 4共b兲 is a plot of LD/LDI versus L/H based on the data from
共LD兲 to lateral displacement index lateral displacement index 共LDI兲 the Moss Landing case history, which is the only case with CPT
versus L/H for case histories with level ground and a free face: 共a兲 data and level ground conditions with a free face that qualified for
standard penetration test-based data and 共b兲 cone penetration test- this study. The data in Fig. 4共b兲 suggests that the trend line de-
based data veloped based on the SPT-based data reasonably fits the CPT-
based data from the Moss Landing case history.
The intent of the empirical relationship of Eq. 共5兲 is to account
for the influence of the height and location of the free face on
soils in the Juvenile Hall area may have a calculated I c greater lateral displacement. The good agreement between the SPT-based
than 2.6 and thus would be evaluated as nonliquefiable soils by and CPT-based data may suggest that Eq. 共5兲 may be independent
the NCEER CPT-based method. This would result in a smaller of either using SPT or CPT data 共provided that NCEER proce-
calculated value of LDI and thus a higher value of LD/LDI, as dures are used for liquefaction potential evaluation兲. It is possible
shown in Fig. 3共b兲. As suggested by Robertson and Wride 共1998兲, that this relationship solely captures the influence of H and L on
samples of soil should be obtained for I c greater than 2.6 and lateral displacements. This preliminary observation is encourag-
liquefaction evaluated using the other criteria, such as the Chinese ing, however, additional data, especially CPT-based data from
criteria 共NCEER 1997兲. Therefore the point in Fig. 3共b兲 for the new case histories, is required to further evaluate this observation.
Juvenile Hall site may be ignored in qualifying the correlation
between LD/LDI and ground slope using CPT data, and illustrates Gently Sloping Ground with a Free Face
the caution required with assessing liquefaction potential for soils
with I c greater than 2.6. Four cases were available for gently sloping ground with a free
Fig. 3共b兲 shows that the data for the San Francisco case history face. In addition to the free face height and the distance from the
associated with the 1906 San Francisco earthquake reasonably fit free face, ground slope S is required to characterize the geometry
the trend line obtained from SPT data with some scatter. This of the site 共Fig. 5兲. A positive ground slope is inclined towards the
agreement may suggest that the relationship between LD/LDI and free face 共e.g., a channel兲 whereas a negative ground slope is
ground slope is independent of either using SPT or CPT data and inclined away from the free face.
possibly that it solely captures the influence of ground slope on Table 4 provides a summary of the main parameters for these
lateral displacements. This is encouraging, however, additional cases. SPT results were available for two cases, and CPT results
CPT-based data from new case histories are required to further were available for the other two cases. Since the relationship be-
verify this observation. tween LD/LDI and geometric parameters appears to be indepen-

866 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004

 the results from the Wildlife case (S⫽⫺0.5%) lie toward the
lower range of the scatter. Larger positive slopes tend to increase
LD/LDI, as the results for the Heber Road (S⫽1.5%) and Hok-
kaido (S⫽0.7 to 0.8%兲 cases lie toward the upper range of the
scatter. Thus the ground slope appears to influence the magnitude
of lateral displacements, however, there is insufficient data at
present to quantify this effect. Caution is required to estimate
liquefaction-induced lateral displacements for gently sloping
ground with slopes greater than 0.5%.

Proposed Approach to Estimate Lateral

Displacements Using Standard Penetration
Test or Cone Penetration Test Data

The following method may be used to estimate liquefaction in-

duced lateral displacements.
Fig. 5. Ratio of measured lateral displacement 共LD兲 to lateral Step 1: Assess the liquefaction potential using either the
displacement index 共LDI兲 versus L/H for case histories for level and NCEER SPT- or CPT-based methods.
gently sloping ground with a free face Step 2: Calculate the lateral displacement index 共LDI兲 using
the relationship plotted in Fig. 1 and Eq. 共3兲.
Step 3: Knowing ground slope 共S兲 or/and free face height 共H兲
dent of either using SPT-based or CPT-based data, SPT and CPT and the distance to a free face 共L兲, estimate the lateral displace-
cases are combined together to provide a larger data set. ment 共LD兲 using either
Fig. 5 illustrates the relationship between LD/LDI and L/H LD⫽ 共 S⫹0.2兲 •LDI 共 for 0.2%⬍S⬍3.5% 兲 (6)
using both the SPT- and CPT-based data for the four available for gently sloping ground without a free face, or
case histories for gently sloping ground with a free face. The
LD⫽6• 共 L/H 兲 ⫺0.8•LDI 共 for 4⬍L/H⬍40兲 (7)
datapoints from Figs. 4共a and b兲 are also included in Fig. 5 to
provide a comparison with nearly level ground and a free face. for level ground with a free face.
The data points for gently sloping ground and a free face lie The proposed approach is recommended for use within the
generally within the scatter of results for nearly level ground with ranges of earthquake properties and ground conditions listed in
a free face. It appears that a negative slope decreases LD/LDI as Tables 2– 4, namely moment magnitude of earthquake between

Fig. 6. Comparison of measured and calculated lateral displacements for the available case histories for: 共a兲 gently sloping ground without a free
face; 共b兲 level ground with a free face; and 共c兲 gently sloping ground with a free face

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004 / 867

Fig. 7. Comparison of measured lateral displacements less than 1 m and calculated lateral displacements for the available case histories for: 共a兲
gently sloping ground without a free face; 共b兲 level ground with a free face; and 共c兲 gently sloping ground with a free face

6.4 and 9.2, peak surface acceleration between 0.19g and 0.6g, histories because of the relatively poor accuracy 共⫾0.72 m兲 of the
and free face heights less than 18 m. The case history data used measured displacements and relatively flat ground slopes 共0.2–
for developing the proposed approach, especially for gently slop- 0.9%兲, where local topography variations and/or the presence of
ing ground without a free face, were dominantly from two Japa- buildings may have more significant effects on lateral displace-
nese case histories associated with the 1964 Niigata and 1983 ments than those for steeper slopes.
Nihonkai-Chubu earthquakes, where the liquefied soils were Given the complexity of liquefaction-induced lateral spreads,
mainly clean sand only. The values for the geometric parameters considerable variations in magnitude and distribution of lateral
used in developing the proposed approach were within limited displacements at a given site are expected. Considering the ac-
ranges, as specified in Eqs. 共6兲 and 共7兲. It is recommended that the cepted ability of current calculations of ground settlements in
approach not be used when the values of the geometric param- sand for the simple case of static vertical loading, the accuracy of
eters go beyond the specified ranges. the proposed approach may be reasonable and acceptable at the
Caution should also be exercised when a substantial zone of current stage for low to medium risk projects.
soil with a very low value penetration resistance 关i.e., (N 1 ) 60cs
⬍10 or (q c1N ) cs⬍50] is encountered during liquefaction potential
analysis. For such cases, more extensive investigation should be Small Lateral Displacements
made and other approaches should be taken to evaluate the po-
tential for flow failure of the soil, especially when the static shear Often of practical interest is whether lateral displacements at a
stresses in the ground are relatively high. Deformations caused by site will be sufficiently small such that they will not have a det-
flow failures can be much larger than those by lateral spreads, and rimental effect on a structure. Fig. 7 is an enlarged portion of Fig.
their estimation is beyond the scope of this paper. 6 to better show the variability of the proposed approach when
measured lateral displacements are less than 1 m. In all cases it is
not possible to make definitive conclusions given the small num-
Discussion ber of measured displacements less than 1 m and there is no
information for gently sloping ground with measured displace-
Variability of the Proposed Approach ments less than 0.5 m. However, apart from nine observations
Measured lateral displacements from the available case histories associated with Niigata, the predicted lateral displacements are
are compared with the calculated lateral displacements using the generally within 50–200% of the measured values. Thus if the
proposed approach in Fig. 6. Generally, about 90% of the calcu- calculated lateral displacement using the proposed approach was
lated lateral displacements using the proposed approach showed 1 m, it would not be possible to conclude that the displacement
variations between 50 and 200% of measured values for the case would be 1 m, but rather may be expected to vary between 0.5
histories studied. The proposed approach could underestimate or and 2 m. Further, given the accuracy of the measured lateral dis-
overestimate liquefaction-induced lateral displacements by up to a placements for most case histories ranges from ⫾0.1 to ⫾1.92 m
factor of 2. The accuracy of the calculated displacements for the 共see Tables 2– 4兲, it is therefore unrealistic to expect the accuracy
Niigata case history is slightly lower than that for other case of calculated lateral displacements to be less than ⫾0.1 m.

868 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004

Comparison of Proposed Approach with Youd et al. and 4. Future research is needed to further verify this assumption
(2002) with new data as it becomes available.
The proposed approach was developed using the limited data
Several SPT-based models 共Rauch and Martin 2000; Youd et al. from available case histories and as such is applicable only for
2002兲 are currently available to estimate liquefaction-induced lat- similar earthquakes and ground conditions. In situ CPT data were
eral displacements. Rauch and Martin’s empirical method was available only for five of all the case histories studied in this
developed using case history data from a multiple linear regres- paper. Therefore additional case history data, especially with CPT
sion analysis. However, it can only be used to estimate an average data, are required to evaluate the proposed approach. In addition,
value of lateral displacements on a potential lateral spread. The the proposed approach was developed based on data with limited
model of Youd et al. 共2002兲 was developed using multiple linear ranges. More data from new case histories are needed to update
regression techniques and case history data including seismologi- the proposed approach.
cal, topographical, geological, and geotechnical parameters. Al- Several factors including boundaries and three-dimensional
though their method is very simple and straightforward to use, the distribution of liquefied layers, redistribution and drainage of ex-
method is purely empirical, does not build upon liquefaction po- cess pore pressures in a lateral spread, and isolation of accelera-
tential analysis, and relies on using SPT-based average values for tion due to liquefaction of underlying soil layers may also influ-
some input data. ence liquefaction-induced lateral displacements 共Zhang 2001兲.
Preliminary comparisons between the proposed approach and However, they were not quantified in the proposed approach and
the model of Youd et al. 共2002兲 indicated that the accuracy of would need to be evaluated on a case-by-case basis.
these two methods were generally similar for estimating lateral
displacements for gently sloping ground without a free face or
level ground with a free face for the Niigata case history. Further Conclusions
evaluations are required using data from new case histories as
they become available before a general conclusion can be made. An approach to estimate liquefaction-induced lateral spreading
displacements was presented. The proposed approach uses either
SPT or CPT data and combines results from laboratory tests with
Assumptions, Limitations and Complicating Factors case history data from previous earthquakes. The method captures
the mechanisms of liquefaction-induced lateral spreads and char-
Multiple displacements with different measured values at a given acterizes the major factors controlling lateral displacements. Ap-
case site were collected for most of the case histories. These plication of the proposed method is quite simple and can be ap-
points corresponded to either different local ground geometric plied with only a few additional calculations following SPT- or
parameters or soil profile/properties and were included to evaluate CPT-based liquefaction-potential analysis. The proposed ap-
the effects of local geometry and soil profile/properties under a proach may be suitable to estimate the magnitude of lateral dis-
given earthquake. However, their inclusion may bias the database placements associated with liquefaction-induced lateral spread for
to certain particular sites. The effect of this bias depends on the gently sloping 共or level兲 ground with or without a free face for
spread of the data and the method of analysis used to quantify the low to medium-risk projects, or to provide preliminary estimates
data. Approaches using multiple regression analysis would be bi- for higher risk projects.
ased since some observations may not necessarily be statistically Given the complexity of liquefaction-induced lateral spreads,
independent. The approach adopted in this paper is to plot all of considerable variations in magnitude and distribution of lateral
the available observations that met the screening criterion to show displacements are expected. Generally, the calculated lateral dis-
the trends and variability of the data. Relationships presented to placements using the proposed approach for the available case
quantify the trend in the database in this paper were fit by-eye histories showed variations between 50 and 200% of measured
rather than using regression analysis. values. The accuracy of ‘‘measured’’ lateral displacements for
Moment magnitude of an earthquake (M w ) and peak surface most case histories is about ⫾0.1 to ⫾1.92 m. Therefore it is
acceleration (a max) at a given site are used to characterize the unrealistic to expect the accuracy of estimated lateral displace-
earthquake size and the intensity of strong ground motion in this ments be within ⫾0.1 m. The reliability of the proposed approach
approach. Other earthquake characteristics 共e.g., the duration of can be fully evaluated only over time with more available case
strong motion and the distance to a seismic energy source or fault histories.
rupture兲 are believed to inherently correlate with M w and a max . The proposed approach was developed using case history data
Generally, the higher values of M w and a max will result in lower with limited ranges of earthquake parameters, soil properties, and
values of FS and thicker zones of liquefied soil, resulting in geometric parameters. Therefore it is not recommended that the
higher values of LDI. It is implicitly assumed that the effects of approach be applied for values of input parameters beyond the
earthquake characteristics on lateral spreading displacements can specified ranges. Engineering judgement and caution should be
be quantified by their effects on FS and thickness of liquefied always exercised in applying the proposed approach and in inter-
soils or LDI. This assumption is tenable based on the available preting the results. Additional new data are required to further
data as no biased trends were identified from the case history data evaluate and update the proposed approach.
shown in Figs. 3 and 4 for different earthquakes with wide ranges
of properties (M w from 6.4 to 7.7 and a max from 0.19g to 0.55g).
However, given the limited number of cases, further evaluation of Acknowledgments
the influence of earthquake characteristics is needed once addi-
tional data becomes available. This research was funded by the Natural Science and Engineering
It is also assumed in this paper that the effect of fines content Research Council of Canada 共NSERC兲. Financial support was
on lateral spreading is similar to that on liquefaction triggering. also provided by the University of Alberta to the first writer
No discernable trend between fines content and lateral displace- through FS Chia PhD and Coal Mining Research Company
ment was observed based on the available data shown in Figs. 3 Scholarships. The lateral spread database compiled by Dr. S. F.

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004 / 869

Bartlett and later then published by Dr. T. L. Youd of Brigham thesis, the Rensselaer Polytechnic Institute, Troy, N.Y.
Young University on the world wide web for researchers to down- Bardet, J. P., Tobita, T., Mace, N., and Hu, J. 共2002兲. ‘‘Regional modeling
load was a highly valuable source of case history data. of liquefaction-induced ground deformation.’’ Earthquake Spectra,
18共1兲, 19– 46.
Bartlett, S. F. 共1991兲. ‘‘Empirical analysis of horizontal ground displace-
Appendix. Relationship between Maximum Cyclic ment generated by liquefaction-induced lateral spreads.’’ PhD thesis,
Shear Strain and Factor of Safety Brigham Young Univ., Provo, Utah.
The mathematic expressions for the curves in Fig. 1 are Bartlett, S. F., and Youd, T. L. 共1995兲. ‘‘Empirical prediction of
liquefaction-induced lateral spread.’’ J. Geotech. Eng., 121共4兲, 316 –
if D r ⫽90%, ␥ max⫽3.26共 FS兲 ⫺1.80 for 0.7⭐FS⭐2.0 (8) 329.
Bennett, M. J. 共1989兲. ‘‘Liquefaction analysis of the 1971 ground failure
if D r ⫽90%, ␥ max⫽6.2 for FS⭐0.7 (9)
at the San Fernando Valley Juvenile Hall, California.’’ Bull. Assoc.
⫺2.08
if D r ⫽80%, ␥ max⫽3.22• 共 FS兲 for 0.56⭐FS⭐2.0 Eng. Geologists, XXVI共2兲, 209–226.
(10) Bennett, M. J., McLaughlin, P. V., Sarmiento, J. S., and Youd, T. L.
共1984兲. ‘‘Geotechnical investigation of liquefaction sites, Imperial
if D r ⫽80%, ␥ max⫽10 for FS⭐0.56 (11) Valley, California.’’ Open-File Rep. 84-252, U.S. Geological Survey,
Menlo Park, Calif., 103.
if D r ⫽70%, ␥ max⫽3.20• 共 FS兲 ⫺2.89 for 0.59⭐FS⭐2.0 Bennett, M. J., Youd, T. L., Harp, E. L., and Wieczorek, G. F. 共1981兲.
(12) ‘‘Subsurface investigation of liquefaction, Imperial Valley Earth-
if D r ⫽70%, ␥ max⫽14.5 for FS⭐0.59 (13) quake, California, October 15, 1979.’’ Open-File Rep. 81-502, U.S.
Geological Survey, Menlo Park, Calif., 83.
if D r ⫽60%, ␥ max⫽3.58• 共 FS兲 ⫺4.42 for 0.66⭐FS⭐2.0 Boulanger, R. W., Idriss, I. M., and Mejia, L. H. 共1995兲. ‘‘Investigation
(14) and evaluation of liquefaction related ground displacements at Moss
Landing during the 1989 Loma Prieta earthquake.’’ Rep. No. UCD/
if D r ⫽60%, ␥ max⫽22.7 for FS⭐0.66 (15) CGM-95/02, Dept. of Civil & Environmental Engineering, Univ. of
⫺6.39 California at Davis, Davis, Calif.
if D r ⫽50%, ␥ max⫽4.22• 共 FS兲 for 0.72⭐FS⭐2.0
(16) Boulanger, R. W., Mejia, L. H., and Idriss, I. M. 共1997兲. ‘‘Liquefaction at
Moss Landing during Loma Prieta earthquake.’’ J. Geotech. Geoenvi-
if D r ⫽50%, ␥ max⫽34.1 for FS⭐0.72 (17) ron. Eng., 123共5兲, 453– 467.
Dobry, R., Baziar, M. H., O’Rourke, T. D., Roth, B. L., and Youd, T. L.
⫺7.97
if D r ⫽40%, ␥ max⫽3.31• 共 FS兲 for 1.0⭐FS⭐2.0 共1992兲. ‘‘Liquefaction and ground failure in the Imperial Valley,
(18) Southern California, during the 1979, 1981 and 1987 earthquakes.
if D r ⫽40%, ␥ max⫽250• 共 1.0⫺FS兲 ⫹3.5 for 0.81⭐FS⭐1.0 Case Studies of Liquefaction and Lifeline Performance during Past
Earthquakes, Vol. 2.’’ Technical Rep. NCEER-92-0002, 4-1– 4-84.
(19)
Elgamal, A., Dobry, R., Parra, E., and Yang, Z. 共1998兲. ‘‘Soil dilation and
if D r ⫽40%, ␥ max⫽51.2 for FS⭐0.81 (20) shear deformations during liquefaction.’’ Proc., Fourth Int. Conf. on
Case Histories in Geotechnical Engineering, St. Louis, 1238 –1259.
Hamada, M. 共1992a兲. ‘‘Large ground deformations and their effects on
lifelines: 1964 Niigata earthquake. Case Studies of Liquefaction and
Notation
Lifeline Performance During Past Earthquake, Vol. 1.’’ Technical Rep.
The following symbols are used in this paper: NCEER-92-0001, National Center for Earthquake Engineering Re-
a max ⫽ peak surface acceleration; search, Buffalo, N.Y., 3-1–3-123.
D r ⫽ relative density of a clean sand as a percentage; Hamada, M. 共1992b兲. ‘‘Large ground deformations and their effects on
FS ⫽ factor of safety against liquefaction; lifelines: 1983 Nihonkai-Chubu earthquake. Case Studies of Liquefac-
tion and Lifeline Performance During Past Earthquake, Vol. 1.’’ Tech-
H ⫽ free face height;
nical Rep. NCEER-92-0001, National Center for Earthquake Engi-
I c ⫽ soil behavior type index;
neering Research, Buffalo, N.Y., 4-1– 4-85.
L ⫽ horizontal distance from the toe of a free face; Hamada, M., and O’Rourke, T. D. 共1992兲. ‘‘Case studies of liquefaction
LD ⫽ actual lateral displacement; and lifeline performance during past earthquakes. Volume 1, Japanese
LDI ⫽ lateral displacement index; Case Studies.’’ Technical Rep. NCEER-92-0001, National Center for
M w ⫽ moment magnitude of earthquake; Earthquake Engineering Research, Buffalo, N.Y.
(N 1 ) 60 ⫽ corrected normalized SPT N value; Hamada, M., Wakamatsu, K., and Yasuda, S. 共1992a兲. ‘‘Liquefaction-
(N 1 ) 60cs ⫽ equivalent clean sand normalized SPT N value; induced ground deformation during the 1923 Kanto earthquake. Case
q c ⫽ CPT tip resistance; studies of liquefaction and lifeline performance during past earth-
q c1N ⫽ corrected normalized CPT tip resistance; quakes, Vol. 1.’’ Technical Rep. NCEER-92-0001, National Center for
(q c1N ) cs ⫽ equivalent clean sand normalized CPT penetration Earthquake Engineering Research, Buffalo, N.Y.
Hamada, M., Yasuda, S., Isoyama, R., and Emoto, K. 共1986兲. Study on
resistance;
liquefaction induced permanent ground displacements, Association
R f ⫽ CPT sleeve friction ratio; for the Department of Earthquake Prediction in Japan, Tokyo, 87.
S ⫽ ground slope; Hamada, M., Yasuda, S., and Wakamatsu, K. 共1992b兲. ‘‘Large ground
Z max ⫽ maximum depth of potential liquefiable layers deformations and their effects on lifelines: 1948 Fukui earthquake.
with a calculated FS⬍2.0 at a studied site; Case Studies of Liquefaction and Lifeline Performance During Past
z ⫽ depth below ground surface; and Earthquakes, Vol. 1.’’ Technical Rep. NCEER-92-0001, National Cen-
␥ max ⫽ maximum amplitude of cyclic shear strain. ter for Earthquake Engineering Research, Buffalo, N.Y.
Holzer, T. L., Youd, T. L., and Bennett, M. J. 共1989兲. ‘‘In situ measure-
References ment of pore pressure build up during liquefaction.’’ Proc., 20th Joint
Meeting of the U.S.—Japan Cooperative Program in the National
Abdoun, T. H. 共1997兲. ‘‘Modeling of seismically induced lateral spread- Resources, Panel on Wind and Seismic Effects, U.S. Department of
ing of multi-layered soil and its effect on pile foundations.’’ PhD Commerce.

870 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004

Ishihara, K., Acacio, A., and Towhata, I. 共1993兲. ‘‘Liquefaction-induced No. 14-08-0001-G2128, Cornell Univ., Ithaca, N.Y.
ground damage in Dagupan in the July 16, 1990 Luzon earthquake.’’ Pease, J. W., and O’Rourke, T. D. 共1998兲. ‘‘Liquefaction hazards in the
Soils Found., 33共1兲, 133–154. Mission District and South of Market area, San Francisco. The Loma
Ishihara, K., Yoshida, K., and Kato, M. 共1996兲. ‘‘Ground deformation Prieta, California, Earthquake of October 17, 1989—Liquefaction.’’
characteristics caused by lateral spreading during the 1995 Hanshin- U.S. Geological Survey Professional Paper 1551-B, B25–B60.
Awaji Earthquake.’’ Proc., 6th Japan-U.S. Workshop on Earthquake Rauch, A. F. 共1997兲. ‘‘EPOLLS: An empirical method for predicting sur-
Resistant Design of Lifeline Facilities and Countermeasures for Soil face displacements due to liquefaction-induced lateral spreading in
Liquefaction, Tech. Rep. NCEER-96-0012, 221–242. earthquakes.’’ PhD thesis, Virginia Polytechnic Institute and State
Ishihara, K., and Yoshimine, M. 共1992兲. ‘‘Evaluation of settlements in Univ., Blacksburg, Va.
sand deposits following liquefaction during earthquakes.’’ Soils Rauch, A. F., and Martin, J. R., II. 共2000兲. ‘‘EPOLLS model for predict-
Found., 32共1兲, 173–188. ing average displacements on lateral spreads.’’ J. Geotech. Geoenvi-
Isoyama, R. 共1994兲. ‘‘Liquefaction-induced ground failures and displace- ron. Eng., 126共4兲, 360–371.
ments along the Shiribeshi-toshibetsu River caused by the 1993 Robertson, P. K., and Wride, C. E. 共1998兲. ‘‘Evaluating cyclic liquefac-
Hokkaido-nansei-oki earthquake.’’ Proc., 5th U.S.-Japan Workshop tion potential using the CPT.’’ Can. Geotech. J., 35共3兲, 442– 459.
on Earthquake Resistant Design of Lifeline Facilities and Counter- Sasaki, Y., Tokida, K., Matsumoto, H., and Saya, Y. 共1991兲. ‘‘Shake table
measures Against Soil Liquefaction, Snowbird, Utah, Technical Rep. tests on lateral ground flow induced by soil liquefaction.’’ Proc., 3rd
NCEER-94-0026. Japan-U.S. Workshop on Earthquake Resistant Design of Lifeline Fa-
Jamiolkowski, M., Ladd, C. C., Germaine, J. T., and Lancellotta, R. cilities and Countermeasures for Soil Liquefaction, San Francisco,
共1985兲. ‘‘New developments in field and laboratory testing of soils.’’ CA, T. D. O’Rourke and M. Hamada, eds., National Center for Earth-
Proc., 11th Int. Conf. on Soil Mechanics and Foundation Engineering, quake Engineering Research, SUNY-Buffalo, Buffalo, N.Y., 371–385.
San Francisco, Vol. 1, 57–153. Seed, H. B. 共1979兲. ‘‘Soil liquefaction and cyclic mobility evaluation for
Kulhawy, F. H., and Mayne, P. W. 共1991兲. ‘‘Relative density, SPT and level ground during earthquakes.’’ J. Geotech. Eng. Div., Am. Soc.
Civ. Eng., 105共2兲, 201–255.
CPT interrelations.’’ Proc., 1st Int. Symp. on Calibration Chamber
Skempton, A. K. 共1986兲. ‘‘Standard penetration test procedures and the
Testing, ISCCCT1, Potsdam, N.Y., 197–211.
effects in sands of overburden pressure, relative density, particle size,
Mejia, M. J. 共1998兲. ‘‘Liquefaction at Moss Landing. The Loma Prieta,
aging and overconsolidation.’’ Geotechnique, 36共3兲, 425– 447.
California, Earthquake of October 17, 1989—Liquefaction.’’ U.S.
Taboada-Urtuzuastegui, V. M., and Dobry, R. 共1998兲. ‘‘Centrifuge mod-
Geological Survey Professional Paper 1551-B, B129–B150.
eling of earthquake-induced lateral spreading in sand.’’ J. Geotech.
Meyerhof, G. G. 共1957兲. ‘‘Discussion on research on determining the
Geoenviron. Eng., 124共12兲, 1195–1206.
density of sands.’’ Proc., 4th Int. Conf. of Soil Mechanics and Foun-
Tatsuoka, F., Zhou, S., Sato, T., and Shibuya, S. 共1990兲. ‘‘Method of
dation Engineering, London, Vol. 3, 110.
evaluating liquefaction potential and its application.’’ Rep. on Seismic
McCulloch, D. S., and Bonilla, M. G. 共1970兲. ‘‘Effects of the earthquake
hazards in the soil deposits in urban areas, Ministry of Education of
of March 27, 1964, on the Alaska railroad.’’ U.S. Geological Survey Japan, 75–109 共in Japanese兲.
Professional Paper 545-D. Tokimatsu, K., Kojima, H., Kuwayama, S., Abe, A., and Midorikawa, S.
Nagase, H., and Ishihara, K. 共1988兲. ‘‘Liquefaction-induced compaction 共1994兲. ‘‘Liquefaction-induced damage to buildings in 1990 Luzon
and settlement of sand during earthquakes.’’ Soils Found., 28共1兲, 65– earthquake.’’ J. Geotech. Eng., 120共2兲, 290–307.
76. Wakamatsu, K., Yoshida, N., Suzuki, N., and Tazoh, T. 共1992兲.
National Center for Earthquake Engineering Research 共NCEER兲. 共1997兲. ‘‘Liquefaction-induced large ground deformations and their effects on
Proc., NCEER Workshop on Evaluation of Liquefaction Resistance of lifelines during the 1990 Luzon, Philippines earthquake. Case Studies
Soils, T. L. Yound and I. M. Idriss, eds., Technical Rep. NCEER-97- of Liquefaction and Lifeline Performance During Past Earthquakes,
0022, Salt Lake City. Vol. 1.’’ Technical Rep. NCEER-92-0001, National Center for Earth-
Norton, W. E. 共1983兲. ‘‘In situ determination of liquefaction potential quake Engineering Research, Buffalo, N.Y.
using the PQS probe.’’ Technical Rep. GL-83-15, Geotechnical Labo- Yasuda, S., Nagase, H., Kiku, H., and Uchida, Y. 共1992兲. ‘‘The mecha-
ratory, U.S. Army Engineer Waterways Experiment Station, Vicks- nism and a simplified procedure for the analysis of permanent ground
burg, Miss. displacement due to liquefaction.’’ Soils Found., 32共1兲, 149–160.
O’Rourke, T. D., Beaujon, P. A., and Scawthorn, C. R. 共1992a兲. ‘‘Large Youd, T. L., et al. 共2001兲. ‘‘Liquefaction resistance of soils: summary
ground deformations and their effects on lifeline facilities: 1906 San report from the 1996 NCEER and 1998 NCEER/NSF Workshops on
Francisco earthquake. Case studies of liquefaction and lifeline perfor- evaluation of liquefaction resistance of soils.’’ J. Geotech. Geoenvi-
mance during past earthquakes, Vol. 2.’’ Technical Rep. NCEER-92- ron. Eng., 127共10兲, 817– 833.
0002, National Center for Earthquake Engineering Research, Buffalo, Youd, T. L., and Bartlett, S. F. 共1988兲. ‘‘US case histories of liquefaction-
N.Y. induced ground displacement.’’ 1st Japan-United States Workshop on
O’Rourke, T. D., and Pease, J. W. 共1992兲. ‘‘Large ground deformations Liquefaction, Large Ground Deformation and Their Effects on Life-
and their effects on lifeline facilities: 1989 Loma Prieta Earthquake. line Facilities, Tokyo, 22–31.
Case studies of liquefaction and lifeline performance during past Youd, T. L., and Bennett, M. J. 共1983兲. ‘‘Liquefaction sites, Imperial
earthquakes, Vol. 2: United States case studies.’’ Technical Rep. Valley, California.’’ J. Geotech. Eng., 109共3兲, 440– 457.
NCEER-92-0002, 5-1–5-79. Youd, T. L., Hansen, C. M., and Bartlett, S. F. 共2002兲. ‘‘Revised multi-
O’Rourke, T. D., Roth, B. L., and Hamada, M. 共1992b兲. ‘‘Large ground linear regression equations for prediction of lateral spread displace-
deformations and their effects on lifeline facilities: 1971 San ment.’’ J. Geotech. Geoenviron. Eng., 128共12兲, 1007–1017.
Fernando earthquake. Case studies of liquefaction and lifeline perfor- Youd, T. L., and Hoose, S. N. 共1976兲. ‘‘Liquefaction during 1906 San
mance during past earthquakes, Vol. 2.’’ Technical Rep. NCEER-92- Francisco Earthquake.’’ J. Geotech. Eng., 102共5兲, 425– 439.
0002, National Center for Earthquake Engineering Research, Buffalo, Youd, T. L., and Hoose, S. N. 共1978兲. ‘‘Historic ground failures in North-
N.Y. ern California triggered by earthquakes.’’ U.S. Geological Survey Pro-
Pease, J. W., and O’Rourke, T. D. 共1993兲. ‘‘Liquefaction hazards in the fessional paper 993, U.S. Geological Survey.
Mission District and South of Market areas, San Francisco, California. Zhang, G. 共2001兲. ‘‘Estimation of liquefaction-induced ground deforma-
Liquefaction and ground failure in San Francisco: Site investigation, tions by CPT&SPT-based approaches.’’ PhD thesis, Univ. of Alberta,
modeling, and hazard assessment for the urban environment.’’ Grant Edmonton, Alta., Canada.

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2004 / 871