Ch07_DAY 10/25/01 3:00 PM Page 7.

CHAPTER 7
EARTHQUAKE-INDUCED
SETTLEMENT

The following notation is used in this chapter:

SYMBOL DEFINITION

amax, ap Maximum horizontal acceleration at ground surface (also known as peak ground accelera-
tion)
CSR Cyclic stress ratio
Dr Relative density
Em Hammer efficiency
F Lateral force reacting to earthquake-induced base shear
FS, FSL Factor of safety against liquefaction
g Acceleration of gravity
Geff Effective shear modulus at induced strain level
Gmax Shear modulus at a low strain level
H Initial thickness of soil layer
H1 Thickness of surface layer that does not liquefy
H2 Thickness of soil layer that will liquefy during earthquake
H Change in height of soil layer
k0 Coefficient of earth pressure at rest
Ncorr Value added to (N1 )60 to account for fines in soil
N Uncorrected SPT blow count (blows per foot)
N1 Japanese standard penetration test value for Fig. 7.1
(N1)60 N value corrected for field testing procedures and overburden pressure
OCR ′ /v′0
Overconsolidation ratio  vm
qc1 Cone resistance corrected for overburden pressure
rd Depth reduction factor
ue Excess pore water pressure
V Base shear induced by earthquake
 Earthquake-induced maximum differential settlement of foundation
εv Volumetric strain
eff Effective shear strain

7.1
Ch07_DAY 10/25/01 3:00 PM Page 7.2

7.2 CHAPTER SEVEN

max Maximum shear strain

t Total unit weight of soil
max Earthquake-induced total settlement of foundation
v 0 Total vertical stress
m′ Mean principal effective stress
vm
′ Maximum past pressure, also known as preconsolidation pressure
v′0 Vertical effective stress
1′ Major principal effective stress
2′ Intermediate principal effective stress
3′ Minor principal effective stress
v Increase in foundation pressure due to earthquake
cyc Uniform cyclic shear stress amplitude of earthquake
eff Effective shear stress induced by earthquake
max Maximum shear stress induced by earthquake

7.1 INTRODUCTION

As discussed in Sec. 4.2, those buildings founded on solid rock are least likely to experi-
ence earthquake-induced differential settlement. However, buildings on soil could be sub-
jected to many different types of earthquake-induced settlement. This chapter deals with
only settlement of soil for a level-ground surface condition. The types of earthquake-
induced settlement discussed in this chapter are as follows:
● Settlement versus the factor of safety against liquefaction (Sec. 7.2): This section dis-
cusses two methods that can be used to estimate the ground surface settlement for vari-
ous values of the factor of safety against liquefaction (FS). If FS is less than or equal to
1.0, then liquefaction will occur, and the settlement occurs as water flows from the soil
in response to the earthquake-induced excess pore water pressures. Even for FS greater
than 1.0, there could still be the generation of excess pore water pressures and hence set-
tlement of the soil. However, the amount of settlement will be much greater for the liq-
uefaction condition compared to the nonliquefied state.
● Liquefaction-induced ground damage (Sec. 7.3): There could also be liquefaction-
induced ground damage that causes settlement of structures. For example, there could be
liquefaction-induced ground loss below the structure, such as the loss of soil through the
development of ground surface sand boils. The liquefied soil could also cause the devel-
opment of ground surface fissures that cause settlement of structures.
● Volumetric compression (Sec. 7.4): Volumetric compression is also known as soil den-
sification. This type of settlement is due to ground shaking that causes the soil to com-
press together, such as dry and loose sands that densify during the earthquake.
● Settlement due to dynamic loads caused by rocking (Sec. 7.5): This type of settlement
is due to dynamic structural loads that momentarily increase the foundation pressure act-
ing on the soil. The soil will deform in response to the dynamic structural load, resulting
in settlement of the building. This settlement due to dynamic loads is often a result of the
structure rocking back and forth.
Ch07_DAY 10/25/01 3:00 PM Page 7.3

EARTHQUAKE-INDUCED SETTLEMENT 7.3

The usual approach for settlement analyses is to first estimate the amount of earthquake-
induced total settlement max of the structure. Because of variable soil conditions and struc-
tural loads, the earthquake-induced settlement is rarely uniform. A common assumption is
that the maximum differential settlement  of the foundation will be equal to 50 to 75 per-
cent of max (that is, 0.5max    0.75max). If the anticipated total settlement max and/or
the maximum differential settlement  is deemed to be unacceptable, then soil improve-
ment or the construction of a deep foundation may be needed. Chapters 12 and 13 deal with
mitigation measures such as soil improvement or the construction of deep foundations.

7.2 SETTLEMENT VERSUS FACTOR OF SAFETY

AGAINST LIQUEFACTION

7.2.1 Introduction

This section discusses two methods that can be used to estimate the ground surface settle-
ment for various values of the factor of safety against liquefaction. A liquefaction analysis
(Chap. 6) is first performed to determine the factor of safety against liquefaction. If FS is
less than or equal to 1.0, then liquefaction will occur, and the settlement occurs as water
flows from the soil in response to the earthquake-induced excess pore water pressures.
Even for FS greater than 1.0, there could still be the generation of excess pore water pres-
sures and hence settlement of the soil. However, the amount of settlement will be much
greater for the liquefaction condition compared to the nonliquefied state.
This section is solely devoted to an estimation of ground surface settlement for various
values of the factor of safety. Other types of liquefaction-induced movement, such as bear-
ing capacity failures, flow slides, and lateral spreading, are discussed in Chaps. 8 and 9.

7.2.2 Methods of Analysis

Method by Ishihara and Yoshimine (1992). Figure 7.1 shows a chart developed by
Ishihara and Yoshimine (1992) that can be used to estimate the ground surface settlement
of saturated clean sands for a given factor of safety against liquefaction. The procedure for
using Fig. 7.1 is as follows:

1. Calculate the factor of safety against liquefaction FSL: The first step is to calculate
the factor of safety against liquefaction, using the procedure outlined in Chap. 6 [i.e., Eq.
(6.8)].
2. Soil properties: The second step is to determine one of the following properties:
relative density Dr of the in situ soil, maximum shear strain to be induced by the design
earthquake max, corrected cone penetration resistance qc1 kg/cm2, or Japanese standard
penetration test N1 value.
Kramer (1996) indicates that the Japanese standard penetration test typically transmits
about 20 percent more energy to the SPT sampler, and the equation N1  0.83(N1)60 can be
used to convert the (N1)60 value to the Japanese N1 value. However, R. B. Seed (1991) states
that Japanese SPT results require corrections for blow frequency effects and hammer
release, and that these corrections are equivalent to an overall effective energy ratio Em of
0.55 (versus Em  0.60 for U.S. safety hammer). Thus R. B. Seed (1991) states that the
(N1)60 values should be increased by about 10 percent (that is, 0.6/0.55) when using Fig. 7.1
to estimate volumetric compression, or N1  1.10(N1)60. As a practical matter, it can be
Ch07_DAY 10/25/01 3:00 PM Page 7.4

7.4 CHAPTER SEVEN

FIGURE 7.1 Chart for estimating the ground surface settlement of clean sand as a function
of the factor of safety against liquefaction FSL. To use this figure, one of the following proper-
ties must be determined: relative density Dr of the in situ soil, maximum shear strain to be
induced by the design earthquake max, corrected cone penetration resistance qc1 (kg/cm2), or
Japanese standard penetration test N1 value. For practical purposes, assume the Japanese stan-
dard penetration test N1 value is equal to the (N1)60 value from Eq. (5.2). (Reproduced from
Kramer 1996, originally developed by Ishihara and Yoshimine 1992.)

assumed that the Japanese N1 value is approximately equivalent to the (N1)60 value calcu-
lated from Eq. (5.2) (Sec. 5.4.3).
3. Volumetric strain: In Fig. 7.1, enter the vertical axis with the factor of safety against
liquefaction, intersect the appropriate curve corresponding to the Japanese N1 value [assume
Japanese N1  (N1)60 from Eq. (5.2)], and then determine the volumetric strain εv from the
horizontal axis. Note in Fig. 7.1 that each N1 curve can be extended straight downward to
obtain the volumetric strain for very low values of the factor of safety against liquefaction.
4. Settlement: The settlement of the soil is calculated as the volumetric strain,
expressed as a decimal, times the thickness of the liquefied soil layer.
Ch07_DAY 10/25/01 3:00 PM Page 7.5

EARTHQUAKE-INDUCED SETTLEMENT 7.5

Note in Fig. 7.1 that the volumetric strain can also be calculated for clean sand that has
a factor of safety against liquefaction in excess of 1.0. For FSL greater than 1.0 but less than
2.0, the contraction of the soil structure during the earthquake shaking results in excess pore
water pressures that will dissipate and cause a smaller amount of settlement. At FSL equal
to or greater than 2.0, Fig. 7.1 indicates that the volumetric strain will be essentially equal
to zero. This is because for FSL higher than 2.0, only small values of excess pore water pres-
sures ue will be generated during the earthquake shaking (i.e., see Fig. 5.15).

Method by Tokimatsu and Seed (1984, 1987). Figure 7.2 shows a chart developed by
Tokimatsu and Seed (1984, 1987) that can be used to estimate the ground surface settle-
ment of saturated clean sands. The solid lines in Fig. 7.2 represent the volumetric strain for
liquefied soil (i.e., factor of safety against liquefaction less than or equal to 1.0). Note that
the solid line labeled 1 percent volumetric strain in Fig. 7.2 is similar to the dividing line in
Fig. 6.6 between liquefiable and nonliquefiable clean sand.
The dashed lines in Fig. 7.2 represent the volumetric strain for a condition where excess
pore water pressures are generated during the earthquake, but the ground shaking is not suf-
ficient to cause liquefaction (that is, FS 1.0). This is similar to the data in Fig. 7.1, in that

FIGURE 7.2 Chart for estimating the ground surface settlement of clean sand for fac-
tor of safety against liquefaction less than or equal to 1.0 (solid lines) and greater than
1.0 (dashed lines). To use this figure, the cyclic stress ratio from Eq. (6.6) and the (N1)60
value from Eq. (5.2) must be determined. (Reproduced from Kramer 1996, originally
developed by Tokimatsu and Seed 1984.)
Ch07_DAY 10/25/01 3:00 PM Page 7.6

7.6 CHAPTER SEVEN

the contraction of the soil structure during the earthquake shaking could cause excess pore
water pressures that will dissipate and result in smaller amounts of settlement. Thus by
using the dashed lines in Fig. 7.2, the settlement of clean sands having a factor of safety
against liquefaction in excess of 1.0 can also be calculated.
The procedure for using Fig. 7.2 is as follows:

1. Calculate the cyclic stress ratio: The first step is to calculate the cyclic stress ratio
(CSR) by using Eq. (6.6). Usually a liquefaction analysis (Chap. 6) is first performed, and
thus the value of CSR should have already been calculated.
2. Adjusted CSR value: Figure 7.2 was developed for a magnitude 7.5 earthquake.
Tokimatsu and Seed (1987) suggest that the cyclic stress ratio calculated from Eq. (6.6) be
adjusted if the magnitude of the anticipated earthquake is different from 7.5. The corrected
CSR value is obtained by dividing the CSR value from Eq. (6.6) by the magnitude scaling
factor from Table 6.2. The chart in Fig. 7.2 is entered on the vertical axis by using this cor-
rected CSR value.
As will be illustrated by the following example problem, applying an earthquake mag-
nitude correction factor to the cyclic stress ratio is usually unnecessary. The reason is that
once liquefication has occurred, a higher magnitude earthquake will not result in any addi-
tional settlement of the liquefied soil. Thus as a practical matter, the chart in Fig. 7.2 can
be entered on the vertical axis with the CSR value from Eq. (6.6).
3. (N1)60 value: Now calculate the (N1)60 value [Eq. (5.2), see Sec. 5.4.3]. Usually a
liquefaction analysis (Chap. 6) is first performed, and thus the value of (N1)60 should have
already been calculated.
4. Volumetric strain: In Fig. 7.2, the volumetric strain is determined by entering the
vertical axis with the CSR from Eq. (6.6) and entering the horizontal axis with the (N1)60
value from Eq. (5.2).
5. Settlement: The settlement of the soil is calculated as the volumetric strain,
expressed as a decimal, times the thickness of the liquefied soil layer.

Example Problem. This example problem illustrates the procedure used to determine the
ground surface settlement of soil using Figs. 7.1 and 7.2.
Use the data from the example problem in Sec. 6.4.5. Assume that the liquefied soil
layer is 1.0 m thick. As indicated in Sec. 6.4.5, the factor of safety against liquefaction is
0.26, and the calculated value of (N1)60 determined at a depth of 3 m below ground surface
is equal to 7.7.
● Solution using Fig. 7.1: For Fig. 7.1, assume that the Japanese N1 value is approxi-
mately equal to the (N1)60 value from Eq. (5.2), or use Japanese N1  7.7. The Japanese
N1 curves labeled 6 and 10 are extended straight downward to FS  0.26, and then by
extrapolating between the curves for an N1 value of 7.7, the volumetric strain is equal to
4.1 percent. Since the in situ liquefied soil layer is 1.0 m thick, the ground surface settle-
ment of the liquefied soil is equal to 1.0 m times 0.041, or a settlement of 4.1 cm.
● Solution using Fig. 7.2: Per the example problem in Sec. 6.4.5, the cyclic stress ratio
from Eq. (6.6) is equal to 0.34, and the calculated value of (N1)60 determined at a depth
of 3 m below ground surface is equal to 7.7. Entering Fig. 7.2 with CSR  0.34 and (N1)60
 7.7, the volumetric strain is equal to 3.0 percent. Since the in situ liquefied soil layer
is 1.0 m thick, the ground surface settlement of the liquefied soil is equal to 1.0 m times
0.030, or a settlement of 3.0 cm.

Suppose instead of assuming the earthquake will have a magnitude of 7.5, the exam-
ple problem is repeated for a magnitude 51
4 earthquake. As indicated in Table 6.2, the
Ch07_DAY 10/25/01 3:00 PM Page 7.7

EARTHQUAKE-INDUCED SETTLEMENT 7.7

magnitude scaling factor  1.5, and thus the corrected CSR is equal to 0.34 divided by
1.5, or 0.23. Entering Fig. 7.2 with the modified CSR  0.23 and (N1)60  7.7, the volu-
metric strain is still equal to 3.0 percent. Thus, provided the sand liquefies for both the
magnitude 51⁄4 and magnitude 7.5 earthquakes, the settlement of the liquefied soil is the
same.
● Summary of values: Based on the two methods, the ground surface settlement of the
1.0-m-thick liquefied sand layer is expected to be on the order of 3 to 4 cm.

Silty Soils. Figures 7.1 and 7.2 were developed for clean sand deposits (fines  5 per-
cent). For silty soils, R. B. Seed (1991) suggests that the most appropriate adjustment is to
increase the (N1)60 values by adding the values of Ncorr indicated below:

Percent fines Ncorr

5 0
10 1
25 2
50 4
75 5

7.2.3 Limitations

The methods presented in Figs. 7.1 and 7.2 can only be used for the following cases:
● Lightweight structures: Settlement of lightweight structures, such as wood-frame
buildings bearing on shallow foundations
● Low net bearing stress: Settlement of any other type of structure that imparts a low net
bearing pressure onto the soil
● Floating foundation: Settlement of floating foundations, provided the zone of lique-
faction is below the bottom of the foundation and the floating foundation does not impart
a significant net stress upon the soil
● Heavy structures with deep liquefaction: Settlement of heavy structures, such as mas-
sive buildings founded on shallow foundations, provided the zone of liquefaction is deep
enough that the stress increase caused by the structural load is relatively low
● Differential settlement: Differential movement between a structure and adjacent appur-
tenances, where the structure contains a deep foundation that is supported by strata below
the zone of liquefaction
The methods presented in Figs. 7.1 and 7.2 cannot be used for the following cases:
● Foundations bearing on liquefiable soil: Do not use Figs. 7.1 and 7.2 when the foun-
dation is bearing on soil that will liquefy during the design earthquake. Even lightly
loaded foundations will sink into the liquefied soil.
● Heavy buildings with underlying liquefiable soil: Do not use Figs. 7.1 and 7.2 when the
liquefied soil is close to the bottom of the foundation and the foundation applies a large
net load onto the soil. In this case, once the soil has liquefied, the foundation load will
cause it to punch or sink into the liquefied soil. There could even be a bearing capacity
type of failure. Obviously these cases will lead to settlement well in excess of the values
obtained from Figs. 7.1 and 7.2. It is usually very difficult to determine the settlement for
these conditions, and the best engineering solution is to provide a sufficiently high static
Ch07_DAY 10/25/01 3:00 PM Page 7.8

7.8 CHAPTER SEVEN

factor of safety so that there is ample resistance against a bearing capacity failure. This is
discussed further in Chap. 8.
● Buoyancy effects: Consider possible buoyancy effects. Examples include buried stor-
age tanks or large pipelines that are within the zone of liquefied soil. Instead of settling,
the buried storage tanks and pipelines may actually float to the surface when the ground
liquefies.
● Sloping ground condition: Do not use Figs. 7.1 and 7.2 when there is a sloping ground
condition. If the site is susceptible to liquefaction-induced flow slide or lateral spreading,
the settlement of the building could be well in excess of the values obtained from Figs.
7.1 and 7.2. This is discussed further in Chap. 9.
● Liquefaction-induced ground damage: The calculations using Figs. 7.1 and 7.2 do not
include settlement that is related to the loss of soil through the development of ground
surface sand boils or the settlement of shallow foundations caused by the development
of ground surface fissures. These types of settlement are discussed in the next section.

7.3 LIQUEFACTION-INDUCED GROUND DAMAGE

7.3.1 Types of Damage

As previously mentioned, there could also be liquefaction-induced ground damage that

causes settlement of structures. This liquefaction-induced ground damage is illustrated in
Fig. 7.3. As shown, there are two main aspects to the ground surface damage:

1. Sand boils: There could be liquefaction-induced ground loss below the structure, such
as the loss of soil through the development of ground surface sand boils. Often a line of
sand boils, such as shown in Fig. 7.4, is observed at ground surface. A row of sand boils
often develops at the location of cracks or fissures in the ground.
2. Surface fissures: The liquefied soil could also cause the development of ground sur-
face fissures which break the overlying soil into blocks that open and close during the
earthquake. Figure 7.5 shows the development of one such fissure. Note in Fig. 7.5 that
liquefied soil actually flowed out of the fissure.

The liquefaction-induced ground conditions illustrated in Fig. 7.3 can damage all types
of structures, such as buildings supported on shallow foundations, pavements, flatwork,
and utilities. In terms of the main factor influencing the liquefaction-induced ground dam-

FIGURE 7.3 Ground damage caused by the liquefaction of an underlying soil layer. (Reproduced from
Kramer 1996, originally developed by Youd 1984.)
Ch07_DAY 10/25/01 3:00 PM Page 7.9

EARTHQUAKE-INDUCED SETTLEMENT 7.9

FIGURE 7.4 Line of sand boils caused by liquefaction during the

Niigata (Japan) earthquake of June 16, 1964. (Photograph from the
Steinbrugge Collection, EERC, University of California, Berkeley.)

age, Ishihara (1985) states:

One of the factors influencing the surface manifestation of liquefaction would be the thick-
ness of a mantle of unliquefied soils overlying the deposit of sand which is prone to liquefac-
tion. Should the mantle near the ground surface be thin, the pore water pressure from the
underlying liquefied sand deposit will be able to easily break through the surface soil layer,
thereby bringing about the ground rupture such as sand boiling and fissuring. On the other
hand, if the mantle of the subsurface soil is sufficiently thick, the uplift force due to the excess
water pressure will not be strong enough to cause a breach in the surface layer, and hence, there
will be no surface manifestation of liquefaction even if it occurs deep in the deposit.

7.3.2 Method of Analysis

Based on numerous case studies, Ishihara (1985) developed a chart (Fig. 7.6a) that can be
used to determine the thickness of the unliquefiable soil surface layer H1 in order to prevent
damage due to sand boils and surface fissuring. Three different situations were used by
Ch07_DAY 10/25/01 3:00 PM Page 7.10

7.10 CHAPTER SEVEN

FIGURE 7.5 Surface fissure caused by the Izmit earthquake in

Turkey on August 17, 1999. Note that liquefied soil flowed out of
the fissure. (Photograph from the Izmit Collection, EERC,
University of California, Berkeley.)

Ishihara (1985) in the development of the chart, and they are shown in Fig. 7.6b.
Since it is very difficult to determine the amount of settlement due to liquefaction-induced
ground damage (Fig. 7.3), one approach is to ensure that the site has an adequate surface layer
of unliquefiable soil by using Fig. 7.6. If the site has an inadequate surface layer of unlique-
fiable soil, then mitigation measures such as the placement of fill at ground surface, soil
improvement, or the construction of deep foundations may be needed (Chaps. 12 and 13).
To use Fig. 7.6, the thickness of layers H1 and H2 must be determined. Guidelines are as
follows:

1. Thickness of the unliquefiable soil layer H1: For two of the three situations in Fig.
7.6b, the unliquefiable soil layer is defined as that thickness of soil located above the
groundwater table. As previously mentioned in Sec. 6.3, soil located above the groundwa-
ter table will not liquefy.
One situation in Fig. 7.6b is for a portion of the unliquefiable soil below the groundwa-
ter table. Based on the case studies, this soil was identified as unliquefiable cohesive soil
(Ishihara 1985). As a practical matter, it would seem the “unliquefiable soil” below the
Ch07_DAY 10/25/01 3:00 PM Page 7.11

EARTHQUAKE-INDUCED SETTLEMENT 7.11

FIGURE 7.6 (a) Chart that can be used to evaluate the possibility of liquefaction-induced ground damage
based on H1, H2, and the peak ground acceleration amax. (b) Three situations used for the development of the
chart, where H1  thickness of the surface layer that will not liquefy during the earthquake and H2  thick-
ness of the liquefiable soil layer. (Reproduced from Kramer 1996, originally developed by Ishihara 1985.)

groundwater table that is used to define the layer thickness H1 would be applicable for any
soil that has a factor of safety against liquefaction in excess of 1.0. However, if the factor
of safety against liquefaction is only slightly in excess of 1.0, it could still liquefy due to
the upward flow of water from layer H2. Considerable experience and judgment are
required in determining the thickness H1 of the unliquefiable soil when a portion of this
layer is below the groundwater table.
2. Thickness of the liquefied soil layer H2: Note in Fig. 7.6b that for all three sit-
uations, the liquefied sand layer H2 has an uncorrected N value that is less than or
equal to 10. These N value data were applicable for the case studies evaluated by
Ishihara (1985). It would seem that irrespective of the N value, H2 could be the thick-
ness of the soil layer which has a factor of safety against liquefaction that is less than
or equal to 1.0.

7.3.3 Example Problem

This example problem illustrates the use of Fig. 7.6. Use the data from Prob. 6.15, which
deals with the subsurface conditions shown in Fig. 6.15 for the sewage disposal site. Based
on the standard penetration test data, the zone of liquefaction extends from a depth of 1.2
to 6.7 m below ground surface. Assume the surface soil (upper 1.2 m) shown in Fig. 6.15
consists of an unliquefiable soil. Using a peak ground acceleration amax of 0.20g, will there
be liquefaction-induced ground damage at this site?
Ch07_DAY 10/25/01 3:00 PM Page 7.12

7.12 CHAPTER SEVEN

Solution. Since the zone of liquefaction extends from a depth of 1.2 to 6.7 m, the thick-
ness of the liquefiable sand layer H2 is equal to 5.5 m. By entering Fig. 7.6 with H2  5.5
m and intersecting the amax  0.2g curve, the minimum thickness of the surface layer H1
needed to prevent surface damage is 3 m. Since the surface layer of unliquefiable soil is
only 1.2 m thick, there will be liquefaction-induced ground damage.
Some appropriate solutions would be as follows: (1) At ground surface, add a fill layer
that is at least 1.8 m thick, (2) densify the sand and hence improve the liquefaction resis-
tance of the upper portion of the liquefiable layer, or (3) use a deep foundation supported
by soil below the zone of liquefaction.

7.4 VOLUMETRIC COMPRESSION

7.4.1 Main Factors Causing Volumetric Compression

Volumetric compression is also known as soil densification. This type of settlement is due to
earthquake-induced ground shaking that causes the soil particles to compress together.
Noncemented cohesionless soils, such as dry and loose sands or gravels, are susceptible to this
type of settlement. Volumetric compression can result in a large amount of ground surface set-
tlement. For example, Grantz et al. (1964) describe an interesting case of ground vibrations
from the 1964 Alaskan earthquake that caused 0.8 m (2.6 ft) of alluvium settlement.
Silver and Seed (1971) state that the earthquake-induced settlement of dry cohesionless
soil depends on three main factors:

1. Relative density Dr of the soil: The looser the soil, the more susceptible it is to volu-
metric compression. Those cohesionless soils that have the lowest relative densities will
be most susceptible to soil densification. Often the standard penetration test is used to
assess the density condition of the soil.
2. Maximum shear strain max induced by the design earthquake: The larger the shear
strain induced by the earthquake, the greater the tendency for a loose cohesionless soil
to compress. The amount of shear strain will depend on the peak ground acceleration
amax. A higher value of amax will lead to a greater shear strain of the soil.
3. Number of shear strain cycles: The more cycles of shear strain, the greater the ten-
dency for the loose soil structure to compress. For example, it is often observed that the
longer a loose sand is vibrated, the greater the settlement. The number of shear strain
cycles can be related to the earthquake magnitude. As indicated in Table 2.2, the higher
the earthquake magnitude, the longer the duration of ground shaking.

In summary, the three main factors that govern the settlement of loose and dry cohe-
sionless soil are the relative density, amount of shear strain, and number of shear strain
cycles. These three factors can be accounted for by using the standard penetration test, peak
ground acceleration, and earthquake magnitude.

7.4.2 Simple Settlement Chart

Figure 7.7 presents a simple chart that can be used to estimate the settlement of dry sand
(Krinitzsky et al. 1993). The figure uses the standard penetration test N value and the peak
ground acceleration ap to calculate the earthquake-induced volumetric strain (that is, H/H,
expressed as a percentage). Figure 7.7 accounts for two of the three main factors causing
Ch07_DAY 10/25/01 3:00 PM Page 7.13

EARTHQUAKE-INDUCED SETTLEMENT 7.13

volumetric compression: the looseness of the soil based on the standard penetration test and
the amount of shear strain based on the peak ground acceleration ap.
Note in Fig. 7.7 that the curves are labeled in terms of the uncorrected N values. As a
practical matter, the curves should be in terms of the standard penetration test (N1)60 values
[i.e., Eq. (5.2), Sec. 5.4.3]. This is because the (N1)60 value more accurately represents the
density condition of the sand. For example, given two sand layers having the same uncor-
rected N value, the near-surface sand layer will be in a much denser state than the sand layer
located at a great depth.
To use Fig. 7.7, both the (N1)60 value of the sand and the peak ground acceleration ap
must be known. Then by entering the chart with the ap /g value and intersecting the desired
(N1)60 curve, the volumetric strain (H/H, expressed as a percentage) can be determined.
The volumetric compression (i.e., settlement) is then calculated by multiplying the volu-
metric strain, expressed as a decimal, by the thickness of the soil layer H.

7.4.3 Method by Tokimatsu and Seed

A much more complicated method for estimating the settlement of dry sand has been pro-
posed by Tokimatsu and Seed (1987), based on the prior work by Seed and Silver (1972)
and Pyke et al. (1975). The steps in using this method are as follows:

1. Determine the earthquake-induced effective shear strain eff. The first step is to
determine the shear stress induced by the earthquake and then to convert this shear stress
to an effective shear strain eff . Using Eq. (6.6) and deleting the vertical effective stress v0
′
from both sides of the equation gives

cyc  0.65rd v0 (amax/g) (7.1)

FIGURE 7.7 Simple chart that can be used to determine the settlement of dry sand. In this figure, use the
peak ground acceleration ap and assume that N refers to (N1)60 values from Eq. (5.2). (Reproduced from
Krinitzsky et al. 1993, with permission from John Wiley & Sons.)
Ch07_DAY 10/25/01 3:00 PM Page 7.14

7.14 CHAPTER SEVEN

where cyc  uniform cyclic shear stress amplitude of the earthquake

rd  depth reduction factor, also known as stress reduction coefficient (dimen-
sionless). Equation (6.7) or Fig. 6.5 can be used to obtain the value of rd.
v 0  total vertical stress at a particular depth where the settlement analysis is being
performed, lb/ft2 or kPa. To calculate total vertical stress, total unit weight t
of soil layer (s) must be known.
amax  maximum horizontal acceleration at ground surface that is induced by the
earthquake, ft/s2 or m/s2, which is also commonly referred to as the peak
ground acceleration (see Sec. 5.6)
g  acceleration of gravity (32.2 ft/s2 or 9.81 m/s2)

As discussed in Chap. 6, Eq. (7.1) was developed by converting the typical irregular
earthquake record to an equivalent series of uniform stress cycles by assuming that cyc 
0.65max, where max is equal to the maximum earthquake-induced shear stress. Thus cyc is
the amplitude of the uniform stress cycles and is considered to be the effective shear stress
induced by the earthquake (that is, eff  cyc). To determine the earthquake-induced effec-
tive shear strain, the relationship between shear stress and shear strain can be utilized:

cyc  eff  effGeff (7.2)

where eff  effective shear stress induced by the earthquake, which is considered to be
equal to the amplitude of uniform stress cycles used to model earthquake
motion (cyc  eff), lb/ft2 or kPa
eff  effective shear strain that occurs in response to the effective shear stress
(dimensionless)
Geff  effective shear modulus at induced strain level, lb/ft2 or kPa

Substituting Eq. (7.2) into (7.1) gives

eff Geff  0.65rd v0 (amax/g) (7.3)

And finally, dividing both sides of the equation by Gmax, which is defined as the shear
modulus at a low strain level, we get as the final result

  v0
  
g 
Geff amax
eff   0.65rd  (7.4)
Gmax Gmax

Similar to the liquefaction analysis in Chap. 6, all the parameters on the right side of the
equation can be determined except for Gmax. Based on the work by Ohta and Goto (1976)
and Seed et al. (1984, 1986), Tokimatsu and Seed (1987) recommend that the following
equation be used to determine Gmax:

Gmax  20,000 [(N1)60] 0.333 (m′ )0.50 (7.5)

where Gmax  shear modulus at a low strain level, lb/ft2
(N1)60  standard penetration test N value corrected for field testing procedures and
overburden pressure [i.e., Eq. (5.2)]
m′  mean principal effective stress, defined as the average of the sum of the three
principal effective stresses, or (1′  2′  3′)/3. For a geostatic condition and
a sand deposit that has not been preloaded (i.e., OCR  1.0), the coefficient
Ch07_DAY 10/25/01 3:00 PM Page 7.15

EARTHQUAKE-INDUCED SETTLEMENT 7.15

of earth pressure at rest k0  0.5. Thus the value of m′  0.67v0′ . Note in
Eq. (7.5) that the value of m′ must be in terms of pounds per square foot.
After the value of Gmax has been determined from Eq. (7.5), the value of eff (Geff /Gmax)
can be calculated by using Eq. (7.4). To determine the effective shear strain eff of the soil,
Fig. 7.8 is entered with the value of eff (Geff /Gmax) and upon intersecting the appropriate
value of mean principal effective stress (m′ in ton/ft2), the effective shear strain eff is
obtained from the vertical axis.
2. Determine the volumetric strain εv. Figure 7.9 can be used to determine the volu-
metric strain εv of the soil. This figure was developed for cases involving 15 equivalent uni-
form strain cycles, which is representative of a magnitude 7.5 earthquake. In Fig. 7.9, the
cyclic shear strain cyc is equivalent to the effective shear strain eff calculated from step 1,
except that the cyclic shear strain cyc is expressed as a percentage (%cyc  100 eff ). To
determine the volumetric strain ε v in percent, either the relative density Dr of the in situ soil
or data from the standard penetration test must be known. For Fig. 7.9, assume the N1 in the
figure refers to (N1)60 values from Eq. (5.2).
To use Fig. 7.9, first convert eff from step 1 to percent cyclic shear strain (%cyc 
100eff ). Then enter the horizontal axis with percent cyc, and upon intersecting the relative

FIGURE 7.8 Plot that is used to estimate the effective shear strain eff
for values of eff (Geff /Gmax) from Eq. (7.4) and the mean principal
effective stress m′ . (Reproduced from Tokimatsu and Seed 1987, with
permission from the American Society of Civil Engineers.)
Ch07_DAY 10/25/01 3:00 PM Page 7.16

7.16 CHAPTER SEVEN

density Dr curve or the (N1)60 curve, the value of the volumetric strain εv is obtained from
the vertical axis.
3. Multidirectional shear: The development of Fig. 7.9 was based on unidirectional
simple shear conditions, or in other words, shear strain in only one direction. However,
actual earthquake shaking conditions are multidirectional, where the soil is strained back
and forth. Based on unidirectional and multidirectional tests, Pyke et al. (1975) conclude
that “the settlements caused by combined horizontal motions are about equal to the sum of
the settlements caused by the components acting alone.” Therefore, the unidirectional vol-
umetric strains determined from Fig. 7.9 must be doubled to account for the multidirec-
tional shaking effects of the earthquake.
4. Magnitude of the earthquake: Figure 7.9 was developed for a magnitude 7.5 earth-
quake (that is, 15 cycles at 0.65max). Table 7.1 presents the volumetric strain ratio that can
be used to determine the volumetric strain εv for different-magnitude earthquakes. The pro-
cedure is to multiply the volumetric strain εv from step 3 by the volumetric strain ratio VSR
from Table 7.1.
Note that the volumetric strain ratio is similar in concept to the magnitude scaling fac-
tor (MSF) in Table 6.2. It would seem that the volumetric strain ratio in Table 7.1 should be
equal to the inverse of the magnitude scaling factors in Table 6.2 (that is, VSR  1.0/MSF).
However, they do not equate because the correction in Table 7.1 is made for volumetric
strain, while the correction in Table 6.2 is made for shear stress.
5. Settlement: Because of the variations in soil properties with depth, the soil profile
should be divided into several different layers. The volumetric strain from step 4 is then cal-
culated for each layer. The settlement for each layer is the volumetric strain, expressed as
a decimal, times the thickness of the layer. The total settlement is calculated as the sum of
the settlement calculated for each soil layer.
Section 7.4.4 presents an example problem illustrating the various steps outlined above.
This method proposed by Tokimatsu and Seed (1987) is most applicable for dry sands that

FIGURE 7.9 Plots that can be used to estimate the volumetric strain εv based on the cyclic shear strain cyc
and relative density Dr or N1 value. Assume that N1 in this figure refers to the (N1)60 values from Eq. (5.2).
(Reproduced from Tokimatsu and Seed 1987, with permission from the American Society of Civil Engineers.)
Ch07_DAY 10/25/01 3:00 PM Page 7.17

EARTHQUAKE-INDUCED SETTLEMENT 7.17

TABLE 7.1 Earthquake Magnitude versus Volumetric Strain Ratio for Dry Sands

Number of representative
Earthquake magnitude cycles at 0.65max Volumetric strain ratio

81⁄ 2 26 1.25
71⁄ 2 15 1.00
63⁄4 10 0.85
6 5 0.60
51⁄4 2–3 0.40

Notes: To account for the earthquake magnitude, multiply the volumetric strain εv from
Fig. 7.9 by the VSR. Data were obtained from Tokimatsu and Seed (1987).

have 5 percent or less fines. For dry sands (i.e., water content  0 percent), capillary action
does not exist between the soil particles. As the water content of the sand increases, capil-
lary action produces a surface tension that holds together the soil particles and increases
their resistance to earthquake-induced volumetric settlement. As a practical matter, clean
sands typically have low capillarity and thus the method by Tokimatsu and Seed (1987)
could also be performed for damp and moist sands.
For silty soils, R. B. Seed (1991) suggests that the most appropriate adjustment is to
increase the (N1)60 values by adding the values of Ncorr indicated in Sec. 7.2.2.

7.4.4 Example Problem

Silver and Seed (1972) investigated a 50-ft- (15-m-) thick deposit of dry sand that experi-
enced about 21⁄ 2 in (6 cm) of volumetric compression caused by the San Fernando earth-
quake of 1971. They indicated that the magnitude 6.6 San Fernando earthquake subjected
the site to a peak ground acceleration amax of 0.45g. The sand deposit has a total unit weight
t  95 lb/ft3 (15 kN/m3) and an average (N1)60  9. Estimate the settlement of this 50-ft-
(15-m-) thick sand deposit using the methods outlined in Secs. 7.4.2 and 7.4.3.

Solution Using Fig. 7.7. As shown in Fig. 7.7, the volumetric compression rapidly
increases as the (N1)60 value decreases. Since the peak ground acceleration ap  0.45g, the
horizontal axis is entered at 0.45. For an (N1)60 value of 9, the volumetric strain H/H is
about equal to 0.35 percent. The ground surface settlement is obtained by multiplying the
volumetric strain, expressed as a decimal, by the thickness of the sand layer, or 0.0035
50 ft  0.18 ft or 2.1 in (5.3 cm).

Solution Using the Tokimatsu and Seed (1987) Method. Table 7.2 presents the solution
using the Tokimatsu and Seed (1987) method as outlined in Sec. 7.4.3. The steps are as fol-
lows:

1. Layers: The soil was divided into six layers.

2. Thickness of the layers: The upper two layers are 5.0 ft (1.5 m) thick, and the lower
four layers are 10 ft (3.0 m) thick.
3. Vertical effective stress: For dry sand, the pore water pressures are zero and the ver-
tical effective stress v0
′ is equal to the vertical total stress v . This stress was calcu-
lated by multiplying the total unit weight (t  95 lb/ft3) by the depth to the center of
each layer.
 Ch07_DAY
10/25/01
3:00 PM
TABLE 7.2 Settlement Calculations Using the Tokimatsu and Seed (1987) Method

Page 7.18
Multi-
Layer Gmax directional
Layer thickness, v′ 0  v , [Eq. (7.5)] eff(Geff/Gmax) eff %cyc  εv shear  Multiply Settlement,
number ft lb/ft2 (N1)60 kip/ft2 [Eq. (7.4)] (Fig. 7.8) 100eff (Fig. 7.9) 2εv by VSR in

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

1 5 238 9 517 1.3 104 5 104 5 102 0.14 0.28 0.22 0.13
2 5 713 9 896 2.3 104 1.0 103 1.0 101 0.29 0.58 0.46 0.28
3 10 1425 9 1270 3.1 104 1.3 103 1.3 101 0.40 0.80 0.64 0.77
4 10 2375 9 1630 3.9 104 1.4 103 1.4 101 0.43 0.86 0.69 0.83
5 10 3325 9 1930 4.4 104 1.3 103 1.3 101 0.40 0.80 0.64 0.77
6 10 4275 9 2190 4.8 104 1.3 103 1.3 101 0.40 0.80 0.64 0.77

7.18
Total  3.5 in
Ch07_DAY 10/25/01 3:00 PM Page 7.19

EARTHQUAKE-INDUCED SETTLEMENT 7.19

4. (N1)60 values: As previously mentioned, the average (N1)60 value for the sand deposit
was determined to be 9.
5. Gmax: Equation (7.5) was used to calculate the value of Gmax. It was assumed that the
mean principal effective stress m′ was equal to 0.65v′ 0 . Note that Gmax is expressed in
terms of kips per square foot (ksf) in Table 7.2.
6. Equation (7.4): The value of eff (Geff /Gmax) was calculated by using Eq. (7.4). A peak
ground acceleration amax of 0.45g and a value of rd from Eq. (6.7) were used in the
analysis.
7. Effective shear strain eff : Based on the values of eff(Geff /Gmax) and the mean princi-
pal effective stress (m′ in ton/ft2), Fig. 7.8 was used to obtain the effective shear strain.
8. Percent cyclic shear strain %cyc: The percent cyclic shear strain was calculated as
eff times 100.
9. Volumetric strain εv: Entering Fig. 7.9 with the percent cyclic shear strain and using
(N1)60  9, the percent volumetric strain εv was obtained from the vertical axis.
10. Multidirectional shear: The values of percent volumetric strain εv from step 9 were
doubled to account for the multidirectional shear.
11. Earthquake magnitude: The earthquake magnitude is equal to 6.6. Using Table 7.1,
the volumetric strain ratio is approximately equal to 0.8. To account for the earthquake
magnitude, the percent volumetric strain εv from step 10 was multiplied by the VSR.
12. Settlement: The final step was to multiply the volumetric strain εv from step 11,
expressed as a decimal, by the layer thickness. The total settlement was calculated as
the sum of the settlement from all six layers (i.e., total settlement  3.5 in).

Summary of Values. Based on the two methods, the ground surface settlement of the 50-
ft- (15-m-) thick sand layer is expected to be on the order of 2 to 31⁄ 2 in (5 to 9 cm). As pre-
viously mentioned, the actual settlement as reported by Seed and Silver (1972) was about
21⁄ 2 in (6 cm).

7.4.5 Limitations

The methods for the calculation of volumetric compression as presented in Sec. 7.4 can
only be used for the following cases:
● Lightweight structures: Settlement of lightweight structures, such as wood-frame
buildings bearing on shallow foundations
● Low net bearing stress: Settlement of any other type of structure that imparts a low net
bearing pressure onto the soil
● Floating foundation: Settlement of floating foundations, provided the floating founda-
tion does not impart a significant net stress upon the soil
● Heavy structures with deep settlement: Settlement of heavy structures, such as massive
buildings founded on shallow foundations, provided the zone of settlement is deep
enough that the stress increase caused by the structural load is relatively low
● Differential settlement: Differential movement between a structure and adjacent appur-
tenances, where the structure contains a deep foundation that is supported by strata below
the zone of volumetric compression
Ch07_DAY 10/25/01 3:00 PM Page 7.20

7.20 CHAPTER SEVEN

The methods for the calculation of volumetric compression as presented in Sec. 7.4 can-
not be used for the following cases:
● Heavy buildings bearing on loose soil: Do not use the methods when the foundation
applies a large net load onto the loose soil. In this case, the heavy foundation will punch
downward into the loose soil during the earthquake. It is usually very difficult to deter-
mine the settlement for these conditions, and the best engineering solution is to provide
a sufficiently high static factor of safety so that there is ample resistance against a bear-
ing capacity failure. This is further discussed in Chap. 8.
● Sloping ground condition: These methods will underestimate the settlement for a slop-
ing ground condition. The loose sand may deform laterally during the earthquake, and the
settlement of the building could be well in excess of the calculated values.

7.5 SETTLEMENT DUE TO DYNAMIC LOADS

CAUSED BY ROCKING

Details on this type of settlement are as follows:

● Settlement mechanism: This type of settlement is caused by dynamic structural loads
that momentarily increase the foundation pressure acting on the soil, such as illustrated
in Fig. 7.10. The soil will deform in response to the dynamic structural load, resulting in
settlement of the building. This settlement due to dynamic loads is often a result of the
structure rocking back and forth.
● Vulnerable soil types: Both cohesionless soil and cohesive soil are susceptible to rock-
ing settlement. For cohesionless soils, loose sands and gravels are prone to rocking set-
tlement. In addition, rocking settlement and volumetric compression (Sec. 7.4) often
work in combination to cause settlement of the structure.
Cohesive soils can also be susceptible to rocking settlement. The types of cohesive
soils most vulnerable are normally consolidated soils (OCR  1.0), such as soft clays and
organic soils. There can be significant settlement of foundations on soft saturated clays
and organic soils because of undrained plastic flow when the foundations are overloaded
during the seismic shaking. Large settlement can also occur if the existing vertical effec-
tive stress v′ 0 plus the dynamic load v exceeds the maximum past pressure vm ′ of the
cohesive soil, or v′ 0  v vm′ .
Another type of cohesive soil that can be especially vulnerable to rocking settlement
is sensitive clays. These soils can lose a portion of their shear strength during the cyclic
loading. The higher the sensitivity, the greater the loss of shear strength for a given shear
strain.
● Susceptible structures: Lightly loaded structures would be least susceptible to rocking
settlement. On the other hand, tall and heavy buildings that have shallow foundations
bearing on vulnerable soils would be most susceptible to this type of settlement.
● Example: Figure 7.11 presents an example of damage caused by rocking settlement.
The rocking settlement occurred to a tall building located in Mexico City. The rocking
settlement was caused by the September 19, 1985, Michoacan earthquake, which is
described in Sec. 4.6.1.

In terms of the analysis for rocking settlement, R. B. Seed (1991) states:

Ch07_DAY 10/25/01 3:00 PM Page 7.21

EARTHQUAKE-INDUCED SETTLEMENT 7.21

FIGURE 7.10 Diagram illustrating lateral forces F in response to the base shear V caused by the earth-
quake. Note that the uniform static bearing pressure is altered by the earthquake such that the pressure is
increased along one side of the foundation. (Reproduced from Krinitzsky et al. 1993, with permission from
John Wiley & Sons.)

Vertical accelerations during earthquake seldom produce sufficient vertical thrust to cause
significant foundation settlements. Horizontal accelerations, on the other hand, can cause
“rocking” of a structure, and the resulting structural overturning moments can produce signif-
icant cyclic vertical thrusts on the foundation elements. These can, in turn, result in cumulative
settlements, with or without soil liquefaction or other strength loss. This is generally a poten-
tially serious concern only for massive, relatively tall structures. Structures on deep founda-
tions are not necessarily immune to this hazard; structures founded on “friction piles” (as
opposed to more solidly-based end-bearing piles) may undergo settlements of up to several
inches or more in some cases. It should be noted that the best engineering solution is generally
simply to provide a sufficiently high static factor of safety in bearing in order to allow for ample
resistance to potential transient seismic loading.

As indicated above, the best engineering solution is to provide a sufficiently high factor
of safety against a bearing capacity failure, which is discussed in Chap. 8.

7.6 PROBLEMS

The problems have been divided into basic categories as indicated below:

Liquefaction-Induced Settlement

7.1 Use the data from the example problem in Sec. 7.2.2, but assume that amax/g  0.1
and the sand contains 15 percent nonplastic fines. Calculate the settlement, using Figs. 7.1
and 7.2. Answer: See Table 7.3.
Ch07_DAY 10/25/01 3:00 PM Page 7.22

7.22 CHAPTER SEVEN

FIGURE 7.11 Settlement caused by the building rocking back

and forth during the Michoacan earthquake in Mexico on September
19, 1985. (Photograph from the Steinbrugge Collection, EERC,
University of California, Berkeley.)

7.2 Use the data from the example problem in Sec. 7.2.2, but assume that amax/g  0.2
and the earthquake magnitude M  51⁄4. Calculate the liquefaction-induced settlement,
using Figs. 7.1 and 7.2. Answer: See Table 7.3.
7.3 Use the data from the example problem in Sec. 7.2.2, but assume at a depth of 3 m
that qc  3.9 MPa. Calculate the liquefaction-induced settlement, using Figs. 7.1 and 7.2.
Answer: See Table 7.3.
7.4 Use the data from the example problem in Sec. 7.2.2, but assume that the shear
wave velocity Vs  150 m/s. Calculate the liquefaction-induced settlement, using Figs. 7.1
and 7.2. Answer: See Table 7.3.
7.5 Use the data from the example problem in Sec. 7.2.2, but assume that the soil type
is crushed limestone (i.e., soil type 1, see Fig. 6.12) and at a depth of 3 m, qc1  5.0 MPa.
Calculate the liquefaction-induced settlement, using Figs. 7.1 and 7.2. Answer: See Table 7.3.
7.6 Use the data from the example problem in Sec. 7.2.2, but assume that the soil type
is silty gravel (i.e., soil type 2, see Fig. 6.12) and at a depth of 3 m, qc1  7.5 MPa. Calculate
the liquefaction-induced settlement, using Figs. 7.1 and 7.2. Answer: See Table 7.3.
 Ch07_DAY
10/25/01
3:00 PM
TABLE 7.3 Summary of Answers for Probs. 7.1 to 7.9

(N1)60 bl./ft Cyclic Settlement, Settlement,

FS 

Page 7.23
Earthquake qc1, MPa Cyclic stress resistance cm cm
Problem no. Soil type amax/g magnitude Vs1, m/s ratio (CSR) ratio (CRR) CRR / CSR (Fig. 7.1) (Fig. 7.2)

Section 7.2.2 Clean sand 0.40 71

2 7.7 blows/ft 0.34 0.09 0.26 4.1 3.0
Problem 7.1 Sand—15%
fines 0.10 71
2 7.7 blows/ft 0.084 0.14 1.67 0.15 0.15
Problem 7.2 Clean sand 0.20 51
4 7.7 blows/ft 0.17 0.14 0.82 4.1 2.9
Problem 7.3 Clean sand 0.40 71
2 5.8 MPa 0.34 0.09 0.26 3.6 3.0
Problem 7.4 Clean sand 0.40 71
2 185 m/s 0.34 0.16 0.47 2.8 2.1
Problem 7.5 Crushed
limestone 0.40 71
2 5.0 MPa 0.34 0.18 0.53 4.2 3.1
Problem 7.6 Silty gravel 0.40 71
2 7.5 MPa 0.34 0.27 0.79 3.0 2.2
Problem 7.7 Gravelly
sand 0.40 71
2 14 MPa 0.34 0.44 1.29 0.3 1.2
Problem 7.8 Eolian
sand 0.40 71
2 7.7 blows/ft 0.34 0.09 0.26 4.1 3.0
Problem 7.9 Loess 0.40 71
2 7.7 blows/ft 0.34 0.18 0.53 3.0 2.3

7.23
Note: See App. E for solutions.
Ch07_DAY 10/25/01 3:00 PM Page 7.24

7.24 CHAPTER SEVEN

7.7 Use the data from the example problem in Sec. 7.2.2, but assume that the soil type
is gravelly sand (i.e., soil type 3, see Fig. 6.12) and at a depth of 3 m, qc1  14 MPa.
Calculate the settlement, using Figs. 7.1 and 7.2. Answer: See Table 7.3.
7.8 Use the data from the example problem in Sec. 7.2.2, but assume that the soil type
is eolian sand (i.e., soil type 4, see Fig. 6.12). Calculate the liquefaction-induced settlement,
using Figs. 7.1 and 7.2. Answer: See Table 7.3.
7.9 Use the data from the example problem in Sec. 7.2.2, but assume that the soil type
is noncemented loess (i.e., soil type 7, see Fig. 6.12). Calculate the liquefaction-induced
settlement, using Figs. 7.1 and 7.2. Answer: See Table 7.3.
7.10 Assume a site has clean sand and a groundwater table near ground surface. The
following data are determined for the site:

Layer depth, m Cyclic stress ratio (N1)60

2–3 0.18 10
3–5 0.20 5
5–7 0.22 7

Using Figs. 7.1 and 7.2, calculate the total liquefaction-induced settlement of these layers
caused by a magnitude 7.5 earthquake. Answer: Per Fig. 7.1, 22 cm; per Fig. 7.2, 17 cm.

Liquefaction-Induced Settlement, Subsoil Profiles

7.11 Use the data from Prob. 6.12 and the subsoil profile shown in Fig. 6.13. Ignore any
possible settlement of the soil above the groundwater table (i.e., ignore settlement from ground
surface to a depth of 1.5 m). Also ignore any possible settlement of the soil located below a
depth of 21 m. Using Figs. 7.1 and 7.2, calculate the earthquake-induced settlement of the sand
located below the groundwater table. Answer: Per Fig. 7.1, 61 cm; per Fig. 7.2, 53 cm.
7.12 Use the data from Prob. 6.15 and the subsoil profile shown in Fig. 6.15. Ignore
any possible settlement of the surface soil (i.e., ignore settlement from ground surface to a
depth of 1.2 m). Also ignore any possible settlement of soil located below a depth of 20 m.
Using Figs. 7.1 and 7.2, calculate the earthquake-induced settlement of the sand located
below the groundwater table. Answer: Per Fig. 7.1, 22 cm; per Fig. 7.2, 17 cm.
7.13 Figure 7.12 shows the subsoil profile at the Agano River site in Niigata. Assume
a level-ground site with the groundwater table at a depth of 0.85 m below ground surface.
The medium sand, medium to coarse sand, and coarse sand layers have less than 5 percent
fines. The fine to medium sand layers have an average of 15 percent fines. The total unit
weight t of the soil above the groundwater table is 18.5 kN/m3, and the buoyant unit weight
b of the soil below the groundwater table is 9.8 kN/m3.
The standard penetration data shown in Fig. 7.12 are uncorrected N values. Assume a
hammer efficiency Em of 0.6 and a boring diameter of 100 mm; and the length of drill rods
is equal to the depth of the SPT below ground surface. The design earthquake conditions
are a peak ground acceleration amax of 0.20g and magnitude of 7.5. Based on the standard
penetration test data and using Figs. 7.1 and 7.2, calculate the earthquake-induced settle-
ment of the soil located at a depth of 0.85 to 15.5 m below ground surface. Answer: Per
Fig. 7.1, 30 cm; per Fig. 7.2, 24 cm.
7.14 Figure 7.13 shows the subsoil profile at a road site in Niigata. Assume a level-
ground site with the groundwater table at a depth of 2.5 m below ground surface. Also
assume that all the soil types located below the groundwater table meet the criteria for
potentially liquefiable soil. The medium sand layers have less than 5 percent fines, the
Ch07_DAY 10/25/01 3:00 PM Page 7.25

EARTHQUAKE-INDUCED SETTLEMENT 7.25

FIGURE 7.12 Subsoil profile, Agano River site, Niigata. (Reproduced from Ishihara, 1985.)

sandy silt layer has 50 percent fines, and the silt layers have 75 percent fines. The total unit
weight t of the soil above the groundwater table is 18.5 kN/m3, and the buoyant unit weight
b of the soil below the groundwater table is 9.8 kN/m3.
The standard penetration data shown in Fig. 7.13 are uncorrected N values. Assume a
hammer efficiency Em of 0.6 and a boring diameter of 100 mm; and the length of drill rods
is equal to the depth of the SPT below ground surface. The design earthquake conditions
are a peak ground acceleration amax of 0.20g and magnitude of 7.5. Based on the standard
penetration test data and using Figs. 7.1 and 7.2, calculate the earthquake-induced settle-
ment of the soil located at a depth of 2.5 to 15 m below ground surface. Answer: Per
Fig. 7.1, 34 cm; per Fig. 7.2, 27 cm.
7.15 Use the data from Prob. 6.18 and Fig. 6.11. Based on Figs. 7.1 and 7.2, calculate
the earthquake-induced settlement of the soil located at a depth of 0.5 to 16 m below ground
surface for the before-improvement and after-improvement conditions. Answers: Before
Ch07_DAY 10/25/01 3:01 PM Page 7.26

7.26 CHAPTER SEVEN

FIGURE 7.13 Subsoil profile, road site, Niigata. (Reproduced from Ishihara, 1985.)

improvement: per Fig. 7.1, 45 cm; per Fig. 7.2, 35 cm. After improvement: per Fig. 7.1, 0.3
cm; and per Fig. 7.2, 2.7 cm.
7.16 Use the data from Prob. 6.12 and Fig. 6.13. Assume that there has been soil
improvement from ground surface to a depth of 15 m, and for the zone of soil having soil
improvement (0 to 15-m depth), the factor of safety against liquefaction is greater than 2.0.
A mat foundation for a heavy building will be constructed such that the bottom of the mat
is at a depth of 1.0 m. The mat foundation is 20 m long and 10 m wide, and according to
the structural engineer, the foundation will impose a net stress of 50 kPa onto the soil (the
50 kPa includes earthquake-related seismic load). Calculate the earthquake-induced settle-
ment of the heavy building, using Figs. 7.1 and 7.2. Answer: Per Fig. 7.1, 17 cm; per
Fig. 7.2, 19 cm.
7.17 Use the data from Prob. 6.15 and Fig. 6.15. A sewage disposal tank will be
installed at a depth of 2 to 4 m below ground surface. Assuming the tank is empty at the
Ch07_DAY 10/25/01 3:01 PM Page 7.27

EARTHQUAKE-INDUCED SETTLEMENT 7.27

time of the design earthquake, calculate the liquefaction-induced settlement of the tank.
Answer: Since the tank is in the middle of a liquefied soil layer, it is expected that the empty
tank will not settle, but rather will float to the ground surface.

Liquefaction-Induced Ground Damage

7.18 A soil deposit has a 6-m-thick surface layer of unliquefiable soil underlain by a
4-m-thick layer that is expected to liquefy during the design earthquake. The design earth-
quake has a peak ground acceleration amax equal to 0.40g. Will there be liquefaction-
induced ground damage for this site? Answer: Based on Fig. 7.6, liquefaction-induced
ground damage is expected for this site.
7.19 Use the data from Prob. 6.12 and Fig. 6.13. Assume that the groundwater table
is unlikely to rise above its present level. Using a peak ground acceleration amax equal to
0.20g and the standard penetration test data, will there be liquefaction-induced ground
damage for this site? Answer: Based on Fig. 7.6, liquefaction-induced ground damage is
expected for this site.
7.20 Use the data from Prob. 7.13 and Fig. 7.12. Assume that the groundwater table
is unlikely to rise above its present level. Using a peak ground acceleration amax equal to
0.20g and the standard penetration test data, determine the minimum thickness of a fill
layer that must be placed at the site in order to prevent liquefaction-induced ground dam-
age for this site. Answer: Based on Fig. 7.6, minimum thickness of fill layer  2.2 m.
7.21 Use the data from Prob. 7.14 and Fig. 7.13. Assume that the groundwater table
is unlikely to rise above its present level. Using a peak ground acceleration amax equal to
0.20g and the standard penetration test data, will there be liquefaction-induced ground
damage for this site? Answer: The solution depends on the zone of assumed liquefaction
(see App. E).

Volumetric Compression
7.22 Solve the example problem in Sec. 7.4.4, using the Tokimatsu and Seed (1987)
method and assuming that the 50-ft-thick deposit of sand has (N1)60  5. Answer: 11 in
(28 cm).
7.23 Solve the example problem in Sec. 7.4.4, using the Tokimatsu and Seed (1987)
method and the chart shown in Fig. 7.7, assuming that the 50-ft-thick deposit of sand has
(N1)60  15. Answer: Using the Tokimatsu and Seed (1987) method, settlement  1.3 in
(3.3 cm). Using the chart shown in Fig. 7.7, settlement  0.9 in (2 cm).
7.24 Solve the example problem in Sec. 7.4.4, using the Tokimatsu and Seed (1987)
method and the chart shown in Fig. 7.7, assuming that the 50-ft-thick deposit of sand will
be subjected to a peak ground acceleration of 0.20g and the earthquake magnitude  7.5.
Answer: Using the Tokimatsu and Seed (1987) method, settlement  0.9 in (2.3 cm). Using
the chart shown in Fig. 7.7, settlement  0.6 in (1.5 cm).
7.25 Solve the example problem in Sec. 7.4.4, using the Tokimatsu and Seed (1987)
method and assuming that the 50-ft-thick deposit of sand has (N1)60  5, a peak ground accel-
eration of 0.20g, and the earthquake magnitude  7.5. Answer: Settlement  2 in (5 cm).