CHAPTER 9

GEOTECHNICAL DESIGN OF
SHALLOW FOUNDATIONS

9.1 INTRODUCTION

The geotechnical design of shallow foundations requires specific computa-

tions for the following two purposes: (1) assurance that the foundation does
not collapse by plunging into the soil or by rotating excessively and (2)
assurance that the short-term and long-term total and differential settlement
will be tolerable for the particular structure. The most frequent approach is
to use the bearing-capacity equations in Chapter 7, or a similar set of equa-
tions, to compute the ultimate bearing stress qd that will cause collapse; to
apply a factor of safety to the ultimate bearing stress; to obtain the allowable
soil bearing stress qa; and to compute the short-term or long-term settlement
with the value of qa. These procedures are presented in detail in the following
sections along with other features related to the design of shallow foundations.

9.2 PROBLEMS WITH SUBSIDENCE

Subsidence is the general settlement of the ground surface due principally to

the extraction of water or perhaps oil. Subsidence can also result from the
collapse of underground cavities (Gray and Meyers, 1970) or the vibration of
deposits of sand (Brumund and Leonards, 1972). Lowering of the water table
increases the vertical stress on water-bearing soil and on all lower strata. In
simple terms, lowering of the water table means that vertical stress is in-
creased by the total unit weight less the buoyant unit weight times the distance
of the lowering. If the affected soil is compressible and fine-grained, the
resulting settlement will be time-dependent and perhaps large.

235
236 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

Mexico City has been especially affected by subsidence (Zeevaert, 1980).

Figure 9.1 shows a group of attendees at a conference in Mexico City standing
next to a well casing whose top had originally been near the ground surface.
The result of pumping water from strata below the city is graphically evident
by comparing the exposed length of the casing to the 6-ft height of Dr. Wil-
liam R. Cox. If the water table is lowered uniformly and if the soil is uniform,
subsidence could be uniform, but uniformity seldom exists. As noted earlier
the Palace of Fine Arts has settled significantly but remains in heavy use.
Some design guidelines may show that such a structure is intolerant of dif-

Figure 9.1 Photograph from Mexico City showing conference attendees standing
next to a well casing showing subsidence over many years.
 9.3 DESIGNS TO ACCOMMODATE CONSTRUCTION 237

ferential settlement. The authors are reluctant to present such guidelines be-
cause of special circumstances that apply to many such structures.
Subsidence due to lowering of the water table has affected other places,
including Ottawa, Canada (Bozuzuk and Penner, 1971), the Gulf Coast of
Texas (Dawson, 1963), Venice (Berghinz, 1971), and London (Wilson and
Grace, 1939). Surface and near-surface facilities such as roads, streets, and
pipelines have been adversely affected (Zeevaert, 1980). In addition, fractures
have been observed near the edges of the area of subsidence where the change
in ground surface elevation was severe. Fracturing due to subsidence from
the lowering of the water table is difficult and perhaps impossible to predict,
but the geotechnical engineer must be aware of the chance of such occur-
rences.
Subsidence is a potential problem in areas of abandoned mines (Gray and
Meyers, 1970) and in areas of karstic geology. Location of underground cav-
ities emphasizes the need for a proper investigation of the soil at a site and
suggests the possible use of geophysical techniques.

9.3 DESIGNS TO ACCOMMODATE CONSTRUCTION

9.3.1 Dewatering During Construction

Placing shallow foundations below the water table presents a series of chal-
lenges to the geotechnical engineer. Pumping from a sump in the excavation
is frequently unacceptable because of the danger of collapse of the bottom of
the excavation as a result of the lowered effective stress due to the rising
water. The use of well points with good controls is usually acceptable. Fur-
thermore, care must be taken that a nearby building is not affected by lowering
the water table beneath the building.
Dr. Leonardo Zeevaert (1975) spoke of dewatering during construction of
the Latino Americano Tower. Water that was removed from beneath the ex-
cavation was injected behind a sheet-pile wall to keep the water table at
almost a constant elevation with respect to nearby structures. Also, records
were made of a crack survey of the nearby buildings so that any new damage
could be detected. No additional damage was found.

9.3.2 Dealing with Nearby Structures

Installation of foundations close to an existing structure can affect that struc-
ture. Most problems occur with the installation of deep foundations, where
driving of piles or excavation for drilled shafts could cause vibrations or even
the lateral movement of soil. However, an excavation with a substantial depth
for a mat, for example, could create a severe problem. Extraordinary measures
238 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

must sometimes be taken, including possible underpinning of the foundations

of adjacent structures.

9.4 SHALLOW FOUNDATIONS ON SAND

9.4.1 Introduction
As noted in Chapter 6, designers of footings on sand must be aware of the
possible settlement due to dynamic loads. Many investigators have studied
the problem, with most concern focusing on the design of compaction equip-
ment. Brumund and Leonards (1972) performed tests in the laboratory with
a tank of Ottawa sand at a relative density of 70%. A plate with a diameter
of 4 in. was vibrated and a settlement of approaching 1 in. was measured.
The relative density of sand deposits and all relevant properties of the sand
must be considered when designing the foundations for vibrating machinery.
Data provided by the machine’s manufacturer must be accessed in making
such a design.
The discussion in Chapter 4 indicated that sampling deposits of sand with-
out binder below the water table is not possible without extraordinary mea-
sures, such as freezing, and in situ techniques for investigating the sand are
normally required. The two principal tools for obtaining data on the charac-
teristics of sand are the SPT and the static cone, sometimes termed the Dutch
cone. The cone consists of a 60⬚ hardened steel point, a projected end area
of 10 cm3, and a rate of advancement of 2 cm/sec. Data from the static cone,
if obtained in the recommended manner, are reproducible because the resis-
tance along the push rods is eliminated, the force of penetration is measured
with a calibrated load cell, and the rate of penetration is standard.
The SPT has been used widely in the United States for many years and is
the usual technique for determining the in situ characteristics of sand. Testing
with the static cone is popular in Europe but has been slow in gaining pop-
ularity in the United States; however, many U.S. geotechnical firms can now
perform the cone test.
Data from the SPT, described in Chapter 4, can vary widely from one
operator to another and during the performance of a particular test. Robertson
and Campanella (1988) have referenced other authors and have noted that the
energy delivered to the driving rods during an SPT can vary from about 20%
to 90% of the theoretical maximum, with the variation related to the number
of turns of rope around the cathead, the height of fall, the type of drill rig,
and the operator. They concluded that an energy ratio of 55% to 60% is the
average energy level employed in the field, suggesting that SPT values have
some degree of uncertainty.
 9.4 SHALLOW FOUNDATIONS ON SAND 239

In the two sections that follow, two methods are presented for obtaining
the immediate settlement of shallow foundations on sand. The first is based
on the static-cone test and the second is based on the SPT.

9.4.2 Immediate Settlement of Shallow Foundations on Sand

Schmertmann Method A method for estimating the settlement of shallow
foundations on sand has been proposed by Schmertmann (1970) and col-
leagues (1978). The resistance to penetration is termed the static-cone bearing
capacity and is given by the symbol qc. Schmertmann noted that the values
of qc can vary widely with depth, and constant values, termed qc, are used
over a particular depth. Other static-cone systems can be used for measuring
the resistance of penetration if correlations are obtained with the Dutch cone.
Schmertmann’s concept for the settlement of foundations on sand was to
integrate the strain of the sand beneath the loaded foundation where the unit
load is substantially less than the failure load. FEM studies were performed,
and the results revealed that the maximum strain did not occur at the base of
the foundation but at some distance below the base. Further, after a certain
distance below the loaded area, the amount of strain could be considered
negligible. Schmertmann and his co-workers adopted two patterns for the
influence factor for strain due to rigid footings on sand, as shown in Figure
9.2, one for axisymmetrical loading and the other for plane-strain loading, as
from a continuous footing. Schmertmann includes recommendations for the
stiffness of the sand Es as a function of the value of qc or qc. The primary
computation, based on experimental values of qc and on the location and
magnitude of the applied load, is modified to account for strain relief due to
embedment and for time-related creep. With regard to time-related properties
of sand, the phenomena related to creep or stiffening are not as evident as is
the drainage of water in the settlement of soft clay, but empirical evidence is
strong (Mitchell and Solymar, 1984; Mesri et al., 1990). In his original paper
(1970), Schmertmann showed good to excellent agreement between computed
values and values obtained from a significant number of foundations where
settlement was measured.
With respect to the examples that were analyzed and the presentation that
follows, it is assumed that the net load imposed by the shallow foundation
does not approach the value causing a collapse. Section 9.4.3 shows the com-
putation of the magnitude of the pressure from the shallow foundation that
causes a bearing-capacity failure in the cases where settlement has been
computed.
Schmertmann noted that strain ␧z at any depth z beneath a loaded area may
be computed as follows:

p
␧z  (1  ␯)[(1  2v)A  F] (9.1)
E
240 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

RIGID FOOTING VERTICAL STRAIN INFLUENCE FACTOR = Iz

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
0

B/2 FOR AXISYMMETRICAL

RELATIVE DEPTH BELOW FOOTING LEVEL

USE Es = 2.5 qc
IIP = 0.5 + 0.1 Δp .....(1)
σVP
B

2B FOR PLANE STRAIN

USE Es = 3.5 qc

PLANE STRAIN
L/B > 10
3B
(a) Modified strain influence
factor distributions

4B
B
Δp = p – p´o

p
p´o
B/2 (oxisym)
B (pl. str.) σ´vp
DEPTH TO IIP

Figure 9.2 Recommended modified values for strain influence factor diagrams and
matching sand moduli (from Schmertmann et al., 1978).

where

p  intensity of load on a homogeneous, isotropic, elastic half

space,
E  modulus of elasticity,
␯  Poisson’s ratio, and
A and F  dimensionless factors that depend on the location of the point
considered.

If p and E remain constant, the vertical strain is dependent on the vertical

strain factor, Iz:

Iz  (1  ␯)[(1  2␯)A  F] (9.2)

The Schmertmann method for settlement is implemented by use of the fol-

lowing equations:
 9.4 SHALLOW FOUNDATIONS ON SAND 241

冕 ␧ dz 艑 冕 冉EI 冊dz 艑 C C ⌬p 冘 冉EI 冊⌬z

 ƒ(B)
ƒ(B)
SS  z
z
1 2
z
(9.3)
s 0 s
0 0

C1  1  0.5 冉冊p0
⌬p
(9.4)

C2  1  0.2 log 冉 冊tyr

0.1
(9.5)

where

SS  settlement according to the Schmertmann method,

ƒ(B)  2B for the axisymmetric case and 4B for the plane-strain case (see
Figure 9.2),
Es  modulus of the sand,
C1  coefficient reflecting arching-compression relief,
C2  coefficient reflecting creep with time, and
tyr  time in years.

Schmertmann and colleagues (1978) recommend that, for intermediate cases

of footing shape, both diagrams in Figure 9.2 should be employed and inter-
polation employed for the appropriate solution.
The Schmertmann method is best demonstrated by the computation of an
example that was presented in 1970 but is modified here to reflect changes
presented in 1978. The footing in the example has a width of 2.60 m and a
length of 23 m; therefore, the influence-factor diagram for the plane-strain
case will be used. The base of the footing is 2 m below the ground surface
with a unit load p of 180.4 kPa, and the overburden pressure p0 is 31.4 kPa,
yielding the net load ⌬p at the base of the footing of 149 kPa.
Schmertmann (1970) assumed that the cone-penetration test was performed
and that values of qc had been obtained for layers of sand of various thick-
nesses, starting at the base of the footing. The unit weight of the soil was
assumed to be 15.7 kN/m3 above the water table and 5.9 kN/m3 below the
water table, with the water table at the base of the footing. The diagram for
plane strain may be used, but the value of Izp must be increased as shown by
the equation in the Figure 9.2; therefore, ␴⬘vp  (2.0)(15.7)(2.0)(5.9)  43.2
kN/m3.

Izp  0.5  0.1 冉 冊

149
43.2
0.5

 0.5  0.1(1.85)  0.69

242 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

The values of Iz, shown in Figure 9.2, must be modified by increasing the
value of Izp from 0.5 to 0.69. The revised diagram for Iz is shown in Figure
9.3 along with assumed values of qc from the cone test, following the pattern
of values employed by Schmertmann in the original example.
The results from the computations for the revised example are shown in
Table 9.1. The stratum was divided into layers, as shown in Figure 9.3, de-
pending on values of qc, except where a layer was adjusted to fit the peak
value of qc. The table extends below the footing to a depth of 4B or (4)(2.60)
 10.4 m. Eleven layers are shown in Table 9.1 with pertinent values of the
relevant parameters. For a layer of a given thickness and with the correspond-
ing value of qc, an average value of Iz was obtained by taking values at the
top and bottom of each layer. The value of Es for the plane-strain case was
taken as 3.5 times qc. The values of C1 and C2, for a period of 5 years, were
computed:

C1  1  0.5 冉 冊
31.4
149
 1.11

C2  1  0.2 log冉 冊 5
0.1
 1.34

The computations of settlement, layer by layer, shown in Table 9.1 yielded

a total settlement of 54.1 mm, or 2.1 in. after 5 years. The immediate settle-

Figure 9.3 Modified strain-influence factor and assumed values of qc from the cone
test.
 9.4 SHALLOW FOUNDATIONS ON SAND 243

TABLE 9.1
⌬Z qc Es ZL ⌬p ⌬z
Layer (m) (MPa) (MPa) (m) Iz C1 C2 (MPa) (m 103)
1 1.0 2.50 8.75 0.50 0.295 1.11 1.34 0.149 7.5
2 1.6 3.50 12.25 1.80 0.540 1.11 1.34 0.149 15.6
3 0.4 3.50 12.25 2.80 0.673 1.11 1.34 0.149 4.9
4 0.5 7.00 24.50 3.25 0.633 1.11 1.34 0.149 2.9
5 1.0 3.00 10.50 4.00 0.568 1.11 1.34 0.149 12.0
6 0.5 8.50 29.75 4.75 0.500 1.11 1.34 0.149 1.9
7 1.5 17.00 59.50 5.75 0.410 1.11 1.34 0.149 2.3
8 1.0 6.00 21.00 7.00 0.300 1.11 1.34 0.149 3.2
9 1.0 10.00 35.00 8.00 0.213 1.11 1.34 0.149 1.3
10 1.5 4.00 14.00 9.25 0.103 1.11 1.34 0.149 2.4
11 0.4 6.50 22.75 10.20 0.018 1.11 1.34 0.149 0.1
Sum 
54.1 mm

ment was computed to be (54.1)/(1.34)  40.4 mm or 1.6 in. Schmertmann

and his co-workers present a formal, step-by-step procedure for estimating
the immediate and time-related settlement of shallow foundations on sand.
The method is largely empirical but is founded on sound principles of me-
chanics and is validated by yielding reasonable comparisons with experimen-
tal values of settlement.
With regard to the magnitude of the settlement after 5 years, the geotech-
nical and structural engineers must decide what settlement is tolerable for the
structure being designed. The informal guideline for the design of shallow
foundations on sand is that settlement will control the allowable load. Fre-
quently, a settlement of 1 in. (25.4 mm) is selected as the allowable settlement.
The authors believe that the allowable settlement and the factor of safety to
prevent a bearing-capacity collapse are matters to be considered for each
structure, except as dictated by building codes.
An example of the bearing capacity of a shallow foundation on sand is
presented in Section 9.4.3.

Peck, Hanson, and Thornburn Method The following equations present

a method based on use of the SPT (Peck et al., 1974; Dunn et al., 1980). As
noted, appropriate modifications are included to account for the position of
water table and overburden pressure.

qa  Cw(0.41)NSPT S (9.6)
Dw
Cw  0.5  0.5 (9.7)
Dƒ  B

where
244 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

qa  net bearing pressure (kPa) causing a settlement S in millimeters,

Cw  term to adjust for the position of the water table,
NSPT  corrected blow count from the SPT,
Dw  depth to water table from ground surface,
Dƒ  depth of overburden, and
B  width of footing (Dw, Dƒ, and B are in consistent units).

The authors included Figure 9.4 to correct NSPT to account for pressure from
the overburden, ␴⬘0 (the dependence on units should be noted).
The example of a footing on sand is shown in Figure 9.5. The footing is
3 ft by 3 ft2 and placed l.2 m below the ground surface. The unit weights of
the sand above and below the water table are shown. The corrected value of
NSPT is given as 20. The above equations are used to find the value of qa
where the value of S is 25 mm.

1.2
Cw  0.5  0.5  0.64
1.2  3
qa  0.64(0.41)(20)(25)  131 kPa

The bearing capacity of the footing in Figure 9.5 will be studied in the next
section as a means of evaluating the answer shown above.

9.4.3 Bearing Capacity of Footings on Sand

The procedures presented in Chapter 7 may be used to compute the bearing
capacity of the footing shown in Figure 9.5, where the corrected value of

Figure 9.4 Factor used to multiply by NSPT to obtain the corrected value accounting
for pressure from overburden (from Peck et al., 1974).
 9.4 SHALLOW FOUNDATIONS ON SAND 245

Figure 9.5 Example of a computing settlement where the value of NSPT is available.

NSPT is given as 20. Figure 9.6 shows a correlation between the corrected
value of NSPT and ␾⬘, where ␾⬘ is the friction angle from effective stress
analysis. If data are available from the cone test, Figure 9.7 shows the cor-
relation between the value of qc from the cone test and ␾⬘ for uncemented
quartz sand, taking the vertical effective stress into account.
For a corrected value of NSPT of 20, Figure 9.6 shows the value of the
friction angle ␾⬘ to be 33⬚. Table 7.1 shows the value of Nq to be 26.1 and

Figure 9.6 Correlation between corrected NSPT and friction angle from effective stress
(from Peck et al., 1974).
246 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

Figure 9.7 Suggested correlation between qc from the cone test and peak friction
angle ⌽⬘ for uncemented quartz sands considering vertical effective stress (from Rob-
ertson and Campanella, 1983).

the value of N␥ to be 24.4. The requisite formula is Eq. 7.7 for the case in
Figure 9.5:

Qd 1
 qd  ␥BN␥ s␥ d␥  ␥DƒNqsqdq
A 2

where s␥  0.6, sq  1.54, d␥  1, and dq  1.37. The following values are

computed:

1
qd  (7.5)(3)(24.4)(0.6)(1)  (17.3)(1.2)(26.1)(1.54)(1.37)
2
 165  1143  1308 kPa
 9.5 SHALLOW FOUNDATIONS ON CLAY 247

If a factor of safety of 3 is employed, the value of qa is 436 kPa. Employing

the correlation shown in the previous section, the settlement for a qa of 436
kPa would be 83 mm, a value that is too large for most structures.
This brief exercise shows that settlement will control the design of many
shallow foundations on sand. However, the authors recommend that each case
be considered in detail and that the decision on the value of qa be reserved
until all factors are considered.
If the cone test had been performed for uncemented quartz sand, the value
of ␾⬘ could have been obtained from Figure 9.7. The bearing capacity qd
could have been computed by using Eq. 7.7, as shown above.

9.4.4 Design of Rafts on Sand

The procedures for computation of bearing capacity and settlement shown
earlier may be extended to the design of a raft supported by sand. The prin-
cipal problems facing the geotechnical engineer are (1) characterizing the soil
below the raft and (2) selecting the properties to be used in the design. The
geotechnical engineer will evaluate all of the available data gathered from
field trips, geologic studies, and soil borings. The depth of the relevant soil
beneath the footing may be selected by the following rule of thumb: use the
net load at the base of the raft, use the theory of stress distribution with depth
(Chapter 7), and find the depth where the stress increase under the area of
heaviest load is equal to 10% or less of the net load.
Use of the rule of thumb will normally result in a depth encompassing
various kinds of soil. With specific regard to a raft on sand, the properties of
the sand may vary widely, perhaps as shown by the value of qc in Figure 9.4.
For the computation of settlement the sand may be considered layer by layer,
but for bearing capacity, the engineer must select parameters that reflect the
overall behavior of the stratum of sand. Modifying the equations for bearing
capacity is not feasible, and no rule of thumb exists; instead, the engineer
uses judgment, taking into account properties of the sand in the zones of
maximum stress and adopting conservative values of soil properties.

9.5 SHALLOW FOUNDATIONS ON CLAY

9.5.1 Settlement from Consolidation

The procedure for predicting settlement of a shallow foundation on saturated
clay was presented in Chapter 7. The theories for total settlement and for
time rate of settlement are one-dimensional; that is, the flow of water from
the soil is vertical, with no lateral flow or lateral strain. The theories are
undoubtedly deficient for a number of reasons. Some of the deficiencies are
as follows: excess porewater pressure can be dissipated by the lateral flow of
248 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

water; the stratum of clay will likely have a degree of nonuniformity, and
average values must be used in the analyses; the stiffness of a structure will
affect the distribution of stress and the differential settlement; and settlement
can be shown to depend to some extent on the value of the pore-pressure
coefficient A (Skempton and Bjerrum, 1957). In spite of the limitations noted
above, the equations for total settlement and time rate of settlement of shallow
foundations on saturated clay can be used with some confidence.
Skempton and Bjerrum (1957) include data on the final settlement of four
structures on normally consolidated clay where settlement was significant and
where the computed settlement can be compared with the observed settlement
(Table 9.2). The data are undoubtedly for the portion of the structure where
the settlement was largest. The computed settlement includes an estimate of
the initial settlement, as discussed in the following section.
The data in Figure 9.8 are remarkable where the time rate of settlement is
shown for three of the structures listed in Table 9.2 and where measurements
were made over many years. Excellent agreement is indicated between ob-
served and computed settlements over much of the life of each structure.
The procedures presented in Chapter 7 for the computations of total set-
tlement and time rate of settlement can be used with confidence in many
instances. Limitations on the theory are noted above, and implementation
of the procedures can be expensive and time-consuming. Foundation investi-
gation and laboratory studies must be of high quality, and even then the
geotechnical engineer is faced with difficult decisions about relevant soil
properties.
The settlement due to consolidation of clay will be illustrated using Figure
7.9a. The following assumptions are made with regard to depths and soil
properties: the top of the clay layer is at 8 ft below the bottom of the footings,
the clay layer has a thickness of 4 ft, the footings are embedded 2 ft into
the upper sand, the upper 2 ft of the sand are saturated by capillarity, the
water table is at the base of the footings, the total unit weight of the sand
below the water table is 120 lb/ft3, and the total unit weight of the clay is
110 lb/ft3.
The effective stress at the mid-height of the clay layer prior to construc-
tion the footing may be computed as follows:

TABLE 9.2 Comparison of Observed and Computed Settlement for Four

Structures
Structure Observed Settlement (in.) Computed Settlement (in.)
Oil tank, Isle of Grain 21 21.5
Masonic Temple, Chicago 10 12
Monadnock Block, Chicago 22 22
Auditorium tower, Chicago 24 28.5
 9.5 SHALLOW FOUNDATIONS ON CLAY 249

Time – years
0 5 10 15 20 25 30

end of
construction
MASONIC TEMPLE
Net Settlement – inches

calc.

10
obs.

15
Time – years
0 10 20 30 40 50 60

end of
construction
MONADNOCK BLOCK
Net Settlement – inches

10

calc.

20
obs.

35
Time – years
10 20 30 40 50 60
0
end of
construction
AUDITORIUM TOWER
Net Settlement – inches

10

calc.
20

obs.

30
Figure 9.8 Observed and calculated settlement of buildings on Chicago clay (from
Skemptom and Bjerrum, 1957).
250 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

␴0⬘  (2)(120)  (8)(120  62.4)  (2)(110  62.4)  796 lb/ft2

 0.796 k/ft2

Figure 7.12 shows that the value of ␴v was 0.580 k/ft2 at the top of the clay
layer due to the imposition of the footing loads. A similar exercise used the
Newmark chart, and the value of ␴v at the bottom of the clay layer was 0.423
k/ft2. The value of ␴v at the mid-height of the clay layer was 0.502 k/ft2.
It is assumed that consolidation tests have been performed for the clay in
the layer between the sand strata. The clay was found to be normally con-
solidated, and the following values were obtained: e0  0.7, cc  0.36, and
cr  0.04 ft2 /day.
From Eq. 7.36,

S
4
1  0.7冋0.36 log
0.796  0.502
0.796 册
 0.286 ft

Referring to Figure 7.14, data can be obtained for preparing a table showing
settlement as a function of time, as shown below. The time for a specific
amount of settlement to occur may be obtained from Eq. 7.44, as shown for
time for 50% of the settlement to occur.

cv 0.04 ft2 /day

Tv  t 0.197  t t  19.7 days
H2 4 ft2

U, % Tv S, ft t, days
11 0.01 0.032 1.0
20 0.03 0.057 3.0
30 0.066 0.086 6.6
40 0.12 0.114 12.0
50 0.197 0.143 19.7
60 0.28 0.172 28.0
70 0.38 0.200 38.0
80 0.48 0.229 48.0
90 0.85 0.243 85.0
100 0.289 Infinite

Information such as that in the table above can be of great value to the
engineer and other professionals in planning the design of a structure of sat-
urated clay.

9.5.2 Immediate Settlement of Shallow Foundations on Clay

Skempton (1951) developed an equation using of some approximations of the
theory of elasticity and showed the application of the methods to the imme-
 9.5 SHALLOW FOUNDATIONS ON CLAY 251

diate settlement of shallow foundations. The theory of elasticity yields the

following expression for the mean settlement ␳ of a foundation of width B
on the surface of a semi-infinite solid:

1  ␯2
␳  qBI␳ (9.8)
E

where

q  bearing pressure on the foundation,

Ip  influence value, depending on the shape and rigidity of the
foundation,
␯  Poisson’s ratio of the soil, and
E  modulus of elasticity of the soil.

The above equation may be rewritten in the following convenient form:

␳ q qd 1  ␯ 2
 I (9.9)
B qd su ␳ E/su

where

qd  ultimate bearing capacity, ␯ for undrained condition of clay  1⁄2,

I␳  ␲ /4 for a rigid circular footing on the surface of the clay, and
qd /su  6.8 for the circular footing on the surface of clay as found by
Skempton, where su is the undrained shear strength of the clay.

Making the substitutions, the following equation was obtained for a circular
footing on the surface of saturated clay:

␳1 4 q
 (9.10)
B E/su qd

Skempton noted that for footings at some depth below the ground surface,
the influence value of I␳ decreases but the bearing capacity factor Nc increases;
therefore, to a first approximation, the value of I␳ Nc is constant, and Eq, (9.10)
applies to all circular footings. In the axial compression test for undrained
specimens of clay, the axial strain for the deviator stress of (␴1  ␴3) is given
by the following expression:

(␴1  ␴3)
␧ (9.11)
E
252 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

where E is the secant Young’s modulus at the stress (␴1  ␴3). The above
equation may be written more conveniently as follows:

␴1  ␴3 (␴1  ␴3)t 1
␧ (9.12)
(␴1  ␴3)t su E/su

But in saturated clay with no change in water content under the applied stress,
(␴1  ␴3)t
 2.0; thus, the following equation may be written:
su

2 (␴1  ␴3)
␧ (9.13)
E/su (␴1  ␴3)t

Comparing Eq. 9.10 with Eq. 9.13 shows that, for the same ratio of applied
stress to ultimate stress, the strain in the loading test is related to that in the
compression test by the following equation:

␳1
 2␧ (9.14)
B

The equation states that the settlement of a circular footing resting on the
surface of a stratum of saturated clay can be found from the stress-strain
curve of the clay. Skempton performed some tests with remolded London
clay and found that Eq. 9.14 could be used to predict the results of the footing
q
tests except for high values of . He further noted that the values of I␳
qd
increased from 0.73 for a circle to 1.26 for a footing with a ratio of length
to width of 10⬊1, while the bearing capacity factor Nc decreased from 6.2 to
5.3 for the same range of dimensions. Skempton concluded that the proposed
formulation could be used with a degree of approximation of 20% for any
shape of footing at any depth.

9.5.3 Design of Shallow Foundations on Clay

The equations presented in Chapter 7 for computing the bearing capacity for
footings of a variety of shapes, depths below ground surface, at some slope
with the ground surface and in sloping soil, may be used in design. The
proposal by Skempton (1951) shown above may be used for computing the
immediate settlement of a shallow foundation. The applicability of the Skemp-
ton procedure will be demonstrated by analyzing experiments performed by
O’Neill (1970).
O’Neill placed instruments along the length of four drilled shafts to mea-
sure the magnitude of axial load in the shafts during the performance of axial-
load tests. The downward movement along each of the shafts was found by
subtracting computed values of the shortening of points along the shafts from
 9.5 SHALLOW FOUNDATIONS ON CLAY 253

the observed movement at the top of the shafts. The instrumentation with
interpretation allowed load-settlement curves to be produced for each of the
four tests.
Beaumont clay was deposited by the alluvial process in the first Wisconsin
Ice Age and is somewhat heterogeneous, with inclusions of sand and silt. The
liquid limit is around 70 and the plastic limit is around 20, although wide
variations in both indices are common. The formation was preconsolidated
by desiccation, with the indicated preconsolidation pressure at about 4 tons/
ft2. Numerous cycles of wetting and drying produced a network of randomly
oriented and closely spaced fissures. The fissures led to considerable varia-
tions in measured shear strength, as indicated by the shear-strength profile
shown in Figure 9.9. Confining pressure in the performance of the tests was
equal to the overburden pressure. The scatter in the test results is not unusual
and indicates practical problems in sampling and testing of many overcon-
solidated clays. The solid line in the figure was used in interpreting the results
of testing of the foundation.
A stress-strain curve is shown in Figure 9.10 for the upper stratum at
O’Neill’s site. The shear strength was the average of two diameters below the
base of each of the shafts. Significant data for the four test shafts are shown
in Table 9.3. Computed values of the bearing capacity factor for three of the
tests are very close to the Hansen value of 9.0. The overburden pressure and

Figure 9.9 Shear strength profile from triaxial testing (from O’Neill, 1970).
254 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

Figure 9.10 Stress-strain curves for beaumont clay (from O’Neill, 1970).

the pressure from the weight of the concrete in the drilled shaft were assumed
to be equal in computing the bearing capacity.
Load-settlement curves from the experiments performed by O’Neill are
plotted in Figures 9.11 through 9.14. The stress-strain curve in Figure 9.10
was used in making the Skempton computations. The lack of much better
agreement between the measured and computed values is not surprising in
view of the nature of the soil at the test site. The presence of randomly
oriented and closely spaced fissures means that the soil in the triaxial tests
was less stiff than the soil at the base of the test shafts during the testing.
As shown in Figures 9.11 to 9.14, the values for pile settlement are indi-
cated for a factor of safety of 3 based on the data from the stress-strain curves.
Assuming that the data from load tests were not available, the stress-strain
curves yielded an average settlement value of 0.24 in. for the shafts with a
base diameter of 30 in. and 0.27 in. for the shaft with a base diameter of 90
in. The numbers should not have caused the designer much concern, even

TABLE 9.3 Values for Each Shaft Tested by O’Neill

Ultimate
Diameter Avg. Su at Load on Computed
Shaft Penetration of Base 2D Below Base Value of
Designation (ft) (in.) Base (tsf) (tons) Nc
S1 23 30 1.16 52 8.91
S2 23 90 1.16 445 8.68
S3 24 30 1.05 47 9.12
S4 45 30 2.28 142 12.68
 9.6 SHALLOW FOUNDATIONS SUBJECTED TO VIBRATORY LOADING 255

60

Qexp = 52 T
50

N c = 8.91 Q ana = 52.5 T

40
Load (tons)

30

20
FS = 3, ρ exp = 0.18, ρ ana = 0.22 in.

10
O'Neill
Skempton
0
0 0.5 1 1.5 2 2.5 3
Settlement (in.)

Figure 9.11 Load-settlement curves for test shaft S1.

though the corresponding numbers from the load tests were 0.18 in., 0.05 in.,
and 0.09 in. (shafts 1, 3 and 4), average of 0.11 in. and 0.21 in.

9.5.4 Design of Rafts

The comments on the design of rafts on sand apply as well to the design of
rafts on clay. The equations for bearing capacity and settlement of shallow
foundations presented in Chapter 7 may be used. Emphasis is placed on a
soil investigation and a laboratory study of high quality. The engineer must
exercise extra care in selecting parameters for use in design. If differential
settlement is revealed to be a problem, the geotechnical engineer may work
closely with the architect and the structural engineer in placing and sizing the
shallow foundations to achieve an optimal solution.

9.6 SHALLOW FOUNDATIONS SUBJECTED TO

VIBRATORY LOADING

The response of foundations under dynamic loads is significantly different

from that under static loads. The analysis and design of foundations subjected
256 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

500

Qana = 461 T
450 Q exp = 445 T

400
N c = 8.68
350

300
Load (tons)

250

200

FS = 3, ρ exp = 0.21 in., ρ ana = 0.27 in.

150

100

50 O'Neill
Skempton
0
0 1 2 3 4 5 6 7 8
Settlement (in.)

Figure 9.12 Load-settlement curves for test shaft S2.

to machine vibration or impact from earthquake is a difficult problem because

of the complex interaction between the structural system and supporting soil/
rock stratum. The proper solution of the problem requires an accurate pre-
diction of the foundation response to these loads. A complete and rigorous
analysis must account for the following: the three-dimensional nature of the
problem, material and radiation damping of the soil, variation of soil prop-
erties with depth, embedment effects, and interaction effects between multiple
foundations through the soil. In the last 30 years a number of solutions have
been developed for design of machine foundations. Extensive reviews of these
developments were presented by Richart et al (1970), Lysmer (1978), Roessett
(1980), Gazetas (1983, 1991), and Novak (1987). These are beyond the cur-
rent scope of this book.
The topic is discussed here to emphasize the fact that the vibration of sand
can cause densification of the sand with consequent settlement of the foun-
dation. Therefore, the geotechnical engineer must consider the relative density
of the sand at the site. If it is close to unity, vibration is likely not to be a
problem. Otherwise, settlement will occur and provisions must be made to
deal with it.
 9.7 DESIGNS IN SPECIAL CIRCUMSTANCES 257

50
Qana = 46.4 T
45

Q exp = 47 T
40
Nc = 9.12
35

30
Load (tons)

25

20

FS = 3, ρ exp = 0.05 in., ρ ana = 0.22 in.

15

10

5 O'Neill
Skempton
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Settlement (in.)
Figure 9.13 Load-settlement curves for test shaft S3.

Two procedures may be considered: (1) the use of deep foundations driven
into stable soil with respect to vibration and (2) the use of soil-improvement
methods to make the sand resistant to vibration.

9.7 DESIGNS IN SPECIAL CIRCUMSTANCES

9.7.1 Freezing Weather

Frost Action The depth of freezing of the soil in regions of the United
States and areas to the south of Canada is shown in Figure 9.15. Ice lenses
may form if the frozen surface soils include significant moisture, and the
expansion due to the ice could cause some heave of the foundation. Engineers
in cold climates must consult local building codes on the required depth of
shallow foundations, taking into account the nature of the soils. A particular
problem occurs if the soil at and above the water table consists of silt and
fine sand, as shown in Figure 9.16.
The capillary rise in the granular soil between the water table and the
shallow foundation allows water to accumulate in quantities, creating ice
258 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

160

Qexp = 142 T
140

120

100 Q ana = 100.9 T

Load (tons)

N c = 12.68
80

60

40
FS = 3, ρ exp = 0.09 in., ρ ana = 0.29 in.

20
O'Neill
Skempton
0
0 0.5 1 1.5 2 2.5 3
Settlement (in.)

Figure 9.14 Load-settlement curves for test shaft S4.

lenses. With time, more water rises and additional ice lenses are formed. On
some occasions, a large quantity of ice can collect in a restricted area. The
increased thickness of the ice causes a heave of the shallow foundation, as
shown in Figure 9.16. When warm weather occurs the ice in the soil melts,
leaving the soil with high water content and a low value of bearing capacity.
The conditions shown in Figure 9.15 can lead to failures of roadway pavement
that can be dangerous to vehicular traffic. A remedy to the problem illustrated
in Figure 9.16 is to place a layer of coarse granular soil in the zone of
capillary rise to prevent the upper flow of capillary water.
In northern climates, the soil may remain permanently frozen, a condition
termed permafrost. The construction of light structures and other facilities in
such regions is aimed at preventing the thawing of the permafrost. Structures
are supported on posts or short piles so that air circulation beneath the struc-
ture and above the frozen soil is allowed to continue.

Placing Concrete Freezing weather increases the cost of construction be-

cause the concrete must be protected. The protection against freezing must
last until the concrete ‘‘takes a set’’ or hardens so that free water is not present.
In some climates where freezing weather continues for many months, foun-
 9.7 DESIGNS IN SPECIAL CIRCUMSTANCES 259

Figure 9.15 Maximum anticipated depths of freezing as inferred from city building
codes. Actual depths may vary considerably, depending on cover, soil, soil moisture,
topography, and weather (from Spangler and Handy, 1982).

Figure 9.16 Sketches of frost action.

260 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

dation elements have been prefabricated in workshops and placed in prepared

areas at the job site. Short-term settlement is more likely to occur with pre-
fabricated foundations because the contact between the soil and the founda-
tion is less than perfect.

9.7.2 Design of Shallow Foundations on Collapsible Soil

Collapsible soils were described briefly in Chapter 2. Such soils consist of
thick strata of windblown fine grains, deposited over long periods of time,
and reinforced by remains of vegetation or by cementation. The soil will
remain stable under moderate loads as determined by the appropriate testing
procedures; however, on becoming saturated, the soils will collapse. Prevent-
ing a rising water table or saturation from precipitation is essential. Therefore,
if drainage can be assured or a rising water table can be prevented, shallow
foundations can be installed and will remain stable. If saturation of the col-
lapsible soil is judged to be possible, some form of deep foundation will be
necessary.

9.7.3 Design of Shallow Foundations on Expansive Clay

Expansive clay is an extremely troublesome foundation material and exists
widely in the United States, as shown in Figure 9.17. Chapters 2 and 6 present
some of the factors related to expansive clay. Owners of homes and small
businesses must pay a premium for building or repairing the foundation. Oth-
erwise, they will be left with a substandard and unsightly structure.
The soil at a particular site may appear to be firm and substantial, and
potential foundation problems may not be evident. Unfortunately, the local
building codes in many cities may not address the requirements for identifying
expansive clay and then specifying appropriate procedures for constructing a
foundation if expansive clay is present. Furthermore, years may pass before
the damage from an improper foundation becomes evident.
Expansive clay is readily seen if a site is visited during dry weather. Cracks
in the surface soil are visible and may extend several feet below ground
surface. Geotechnical investigations to identify the problem may involve sev-
eral kinds of laboratory tests. These tests include liquid limit, plastic limit,
shrinkage limit, swelling pressure under confinement, free-swell test, and
identification of clay minerals. Tables 9.4 and 9.5 give recommendations for
indicating the swelling potential of clay on the basis of laboratory tests.
The value of the plasticity index is used by many engineers to evaluate the
character of a clay soil. Tables 9.4 and 9.5 suggest that a plasticity index of
15 or less means that the clay is expected to swell very little, if at all.
If a clay soil is judged to be expansive, the thickness of the zone is con-
sidered. If it is shallow, the troublesome soil may be removed. Alternatively,
the expansive clay may be treated with chemicals such as lime. The lime must
be mixed with the soil by tilling, not by the injection of a slurry of lime and
261

Figure 9.17 Distribution of potentially expansive soils in the United States (from Nelson and Miller, 1992).
262 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

TABLE 9.4 Swelling Potential from Laboratory Tests

Percentage of Volume
Percentage of Sample Expansion from Dry
Less Than 1 ␮m Plasticity Index Shrinkage Limit to Saturation
28 35 11 30
20–31 25–41 7–12 20–30
13–23 15–28 10–16 10–20
15 18 15 10
Source: Holtz and Gibbs (1954).

water. The permeability of clay is so low that the slurry will not readily
penetrate it.
If the stratum of clay is so thick that it cannot be removed or treated, then
the engineer usually considers the use of a stiffened slab on grade, such as
the BRAB slab (Building Research Advisory Board, 1968). The site is graded
with deepened sections at relatively close spacing where reinforcing steel is
placed to form beams into the foundation. The depth of the beams and the
spacing depend on the judgment about shrinkage potential. The purpose of
the beams is to make the slab so stiff that differential settlement is minimal.
Over time, moisture will collect under the center of the slab because of re-
stricted evaporation, but the edges of the slab will respond to changes in
weather. Some homeowners have used plantings or other means to prevent
shrinkage of the soil around a stiffened slab.
In some cases, a founding stratum may occur below the expansive clay.
Drilled shafts can be installed with bells and designed to sustain the uplift
from the expanding clay. If drilled shafts are used, a beam system should be
employed to provide a space above the surface of the clay to allow swelling
to occur without contact with the beams and floor system. Further, the bond
between the expanding clay and the shaft should be eliminated or the shaft
should be heavily reinforced to sustain uplift forces.

9.7.4 Design of Shallow Foundations on Layered Soil

Rarely does the geotechnical engineer encounter a design where the founding
stratum is uniform and homogeneous. The reverse is usually the case; either

TABLE 9.5 Swelling Potential from the Plasticity

Index
Swelling Potential Plasticity Index
Low 0–15
Medium 10–15
High 35–55
Very high ⱖ55
Source: Chen (1988).
 9.7 DESIGNS IN SPECIAL CIRCUMSTANCES 263

the soil is of the same sort but with widely varying properties, or two or more
layers exist in the zone beneath the foundation. Settlement may be computed
with procedures that have been presented, but questions arise when selecting
the bearing capacity. If the additional cost is minor, the bearing capacity may
be based on the weakest stratum in the founding stratum. The second possi-
bility is that settlement will control the design, as shown earlier in this chapter,
rather than bearing capacity.
In some studies, the shapes of the failure surfaces have been modified to
reflect the presence of layers with different characteristics (Meyerhof, 1978;
Meyerhof and Hanna, 1974). A more favorable approach is to employ FEM,
as demonstrated by the material in the following section. A number of codes
are available. Such codes will continue to be important in the analytical work
of engineering offices.

9.7.5 Analysis of Response of a Strip Footing

by Finite Element Method
As noted in the previous section, the FEM may be used to a considerable
advantage in the analysis of shallow foundations. The application to the fol-
lowing example is an elementary use of FEM by serves to indicate the power
of the method. A strip footing with a width, B, of 120 in. is resting on the
surface of a soil, Figure 9.18, with the following characteristics: E  30,000
psi, v  0.3, c  10 psi, and ␾  20⬚. The unit weight of the soil was

Figure 9.18 Model of a strip footing by FEM.

264 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

assumed to be 120 pcf. The Mohr-Coulomb model was used to represent the
nonlinear behavior of the soil.
The result from the FEM computation, where a series of deformations of
the footing was used as input, is shown in Figure 9.19. The analysis assumed
a rough footing because the soil was not allowed to spread at the base of the
footing. As may be seen, the unit value of the ultimate bearing capacity (qu)
of the footing was predicted as 185 psi. The Terzaghi equation for bearing
capacity, Eq. 7.3, is re-written to obtain the unit value of ultimate bearing
capacity and shown in Eq. 9.15.

1
qu  ␥BN␥  ␥DƒNq  cNc (9.15)
2

The footing is resting on the soil surface, so the value of Dƒ is 0. The prop-
erties of the soil show above were employed, along with the following values
of the bearing capacity factors (see Chap 7): N␥  3.64, Nc  17.69, and the
value of qu was computed as follows

1
qu  (0.0694)(120)(3.64)  (10)(17.69)  15.2  176.9  192.1 psi.
2

The FEM yielded a value quite close to the result from the Terzaghi equations.
The methods presented herein do not allow a check to be made of the initial
slope of the load-settlement curve in Fig. 9.19. The deformation field of soil
in the vicinity of the strip footing generated by the FEM output is presented
in Fig. 9.20 which is consistent with the theory introduced in Chapter 7.
The excellent comparison between results from the FEM for a solution of
a straightforward problem indicates that the method can be applied to complex
problems in bearing capacity including response to eccentric loading and be-
havior of shallow foundations on layered soils. However, in the solution of

Figure 9.19 Load-settlement curve for a strip footing from analysis by FEM.
 PROBLEMS 265

Figure 9.20 Deformation field for soil near a strip footing from analysis by FEM.

the example, the engineer faced the problem of selecting a constitutive model
for the soil, the size and number of the elements, the distance to each of the
boundaries, and appropriate restraint at the boundaries. Experience in using
the method is valuable, but different trials by varying the parameters in the
solution may be necessary. Many engineering offices are routinely employing
codes for the FEM and more use is expected in the future.

PROBLEMS

P9.1. Cummings (1947) described the nature of the clay in the Valley of
Mexico and showed the results of consolidation tests performed at
Harvard, where the void ratios were extremely high and unusual.
The e-log-p plot in Figure 9.21 is patterned after the test results
reported by Cummings. Using that plot and estimating the settlement
at a location in Mexico City from the photograph in Figure 9.1, com-
pute the amount the water table must be lowered to cause the settle-
ment indicated. Assume the clay stratum to be 150 ft thick, the initial
void ratio to be 6.60, the initial compressive stress to be 0.9 tsf, and
that the clay remains saturated by capillarity as the water table is
lowered. The answer may be surprising.
P9.2. The plot in Figure 9.22 shows values of qc from the cone penetration
test at a site where a footing is to be constructed. Use the Schmert-
266 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

5
Void Ratio

0
0.01 0.1 1 10

Pressure, tsf
Figure 9.21 Plot of e versus log-p for soil similar to that in the Valley of Mexico.

mann method to compute the settlement of a square footing, 2 by 2

m in plan, with its base at 1 m below the ground surface. Assume
that the soil is sand with an average grain size of 0.5 mm and a
submerged unit weight of 9.35 kN/m3. The load at the base of the
footing is 80 kPa. Use three layers below the base of the footing in
making settlement computations for 1 year and 10 years.
 PROBLEMS 267

Figure 9.22 Example of the result of the cone penetration test (from Robertson and
Campanella, 1988).

P9.3. Use the data from Problem P9.2 to compute the bearing capacity for
the footing. Use Figure 9.23 in selecting a value of ␾ for use in
bearing-capacity computations, and consider the low value qc at the
base of the footing and the significant increase with depth below the
base. Assume a factor of safety of 3 and compute the value of the
safe bearing capacity. Compare the safe bearing capacity with the unit
load used in Problem P9.2. Discuss the allowable load or safe load
you would select. (Check construction on figure and modify as
necessary.)
P9.4. Estimate the settlement of a footing 10 ft2 resting at a depth of 2 ft
below the surface of a coarse sand and subjected to a total load of
400 kips. The water table is at a depth of 2 ft, and the corrected value
of NSPT is 30.
P9.5. Compute the unit bearing capacity in kips per square foot of the foot-
ing in Problem P9.4, assuming that the total unit weight of the sand
is 122 pcf. Compute the factor of safety for the load of 400 kips.
268 GEOTECHNICAL DESIGN OF SHALLOW FOUNDATIONS

Figure 9.23 Suggested correlation between qc from the cone test and peak friction
angle ⌽⬘ for uncemented quartz sands considering vertical effective stress (from
Robertson and Campanella, 1983).

P9.6. The unconfined compressive strength of a clay is 1.2 kips/ft2, the

water table is 4 ft below the ground surface, the total unit weight of
the clay ␥ is 118 lb/ft3, and a drilled shaft with a diameter of 4.25 ft
placed at a depth of 10 ft supports a column load by end bearing only.
(a) Use a factor of safety of 2.8 and compute the safe column load
on the foundation. (b) An excavation near the drilled shaft removes
the soil down to the water table. What is the effect on the factor of
safety?
P9.7. A water tank with a height of 25 ft and a diameter of 50 ft rests on
a stratum of sand with a thickness of 12 ft. A stratum of clay with a
thickness of 20 ft exists below the sand, and a thin layer of gravel
exists below the clay. The water table is at the surface of the clay.
Plot a curve showing the stress increase in the clay as a function of
depth for the center and edge of the tank. Assume the tank to be filled
with water to a depth of 23 ft. Ignore the weight of the steel in the
tank.
 PROBLEMS 269

P9.8. Assume that consolidation tests were performed on the clay in Prob-
lem P9.7, that the soil is normally consolidated, and that the following
values were obtained for the clay: ␥sat  120 pcf; e0  1.10; cc 
0.40; cv  0.025 ft2 /day; sand, ␥  125 pcf. Compute the settlement
in inches at the center and edge of the tank.
P9.9. Plot curves showing the settlement of the center of the tank as a
function of time.
P9.10. Plot a curve showing the distribution of excess porewater pressure in
the clay stratum for consolidation of about 50% (use Tv  0.2).
P9.11. If the differential settlement of the tank is judged to be a problem,
list three steps that may be taken to reduce it.