Soils and Foundations 2012;52(2):299311

The Japanese Geotechnical Society

Soils and Foundations

www.sciencedirect.com
journal homepage: www.elsevier.com/locate/sandf

Coefcient of earth pressure at rest for normally and overconsolidated

peat ground in Hokkaido area
Hirochika Hayashia,n, Nobutaka Yamazoeb, Toshiyuki Mitachic,
Hiroyuki Tanakad, Satoshi Nishimotoa
a

Civil Engineering Research Institute for Cold Region, 1-3 Hiragishi, Sapporo, Japan
b
C-way Engineering Co., Ltd., 2-5 Higashi-Sapporo, Sapporo, Japan
c
College of Industrial Technology, Nihon University, 1-2-1 Izumi-cho, Narashino, Japan
d
Graduate School of Engineering, Hokkaido University, West8 North13, Sapporo, Japan
Available online 29 March 2012

Abstract
Peat, which is widely distributed in Hokkaido, is a very soft and problematic soil. To perform an elasto-plastic Finite Element (FE)
analysis, it is important to accurately determine the initial stress conditions, and among them, the value of the coefcient of earth
pressure at rest (K0 value) is particularly important. A K0-consolidation test using triaxial testing apparatus and a at dilatometer were
performed to investigate the K0 value for peat ground in Hokkaido, Japan. It was found that the K0 value for normally consolidated
peat and organic clay (K0NC) decreases with an increase in the ignition loss. The K0 value for overconsolidated peat and organic clay
(K0OC) is more strongly dependent on the over consolidation ratio (OCR) than that of usual inorganic clay. That is, it is known that
K0OC is empirically related to K0NC, as expressed by K0OC K0NCOCRm, and in peat and organic clay the power of m increases with their
ignition loss. An experimental equation to estimate K0 using a at dilatometer for peat ground is also proposed.
& 2012 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.
Keywords: K0 value; Flat dilatometer; K0-consolidation test; Normal consolidation; OCR; Organic clay; Over consolidation; Peat (IGC: D5/D6)

1. Introduction
Peat ground is widely distributed throughout the Hokkaido
region of Japan. Peat contains a large amount of organic
matter and has a very high natural water content. Its
unsuitability as a foundation material has brought about
n

Corresponding author.
E-mail address: hayashi@ceri.go.jp (H. Hayashi).

0038-0806 & 2012 The Japanese Geotechnical Society. Production and

hosting by Elsevier B.V. All rights reserved.
Peer review under responsibility of The Japanese Geotechnical Society
http://dx.doi.org/10.1016/j.sandf.2012.02.007

Production and hosting by Elsevier

serious problems for construction works. In Hokkaido, the

peat layer often overlies thick layers of organic clay and
Holocene inorganic clay. Embankment constructions tend
to deform the peat ground, not only due to consolidation
but also shear. Hayashi et al. (1994) have demonstrated that
a Finite Element (FE) analysis based on Cam-clay model
(Schoeld and Wroth, 1968) is quite effective in predicting
the deformation of peat ground, since this model is capable
of taking deformation caused by consolidation as well as
shear into account. In order to perform an elasto-plastic FE
analysis including the Cam-clay model, accurate initial
stress conditions need to be determined. The value of the
coefcient of earth pressure at rest (K0 value) is particularly
important. In this paper, we propose a method for determining the K0 value for peat ground.
Remarkably little research has been done on the K0 value
for peat and organic clay compared to usual inorganic clay, so
very little data is available in the literature. Miyakawa et al.

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

(1974), Edil and Dhowian (1981), Fujikawa et al. (1983),

Kawano et al. (1986), and Mesri and Ajlouni (2007) reported
the K0 values for peat ground measured with one-dimensional
oedometer test, with stress conditions restricted to the normally consolidated state. As will be explained later, peat
ground is usually overconsolidated to some degree due to
the in situ stresses which vary with seasonal changes in the
groundwater table. This factor is signicant since, combined
with the extremely low density of peat, it results in high OCR
values. As such, it is very important for determining the K0
value of overconsolidated peat ground for numerical solutions.
It should be kept in mind that the engineering properties
of peat ground are extremely heterogeneous. Therefore, the
soil parameters for practical designs should be obtained
from a large number of simple and robust tests, rather than
from a few complicated and sophisticated tests. It is also
useful to determine the soil parameters using simple in-situ
tests that can measure soil parameters continuously to the
required depth. In this study, methods for determining the
K0 of peat ground are established by performing laboratory
K0-consolidation tests using triaxial testing apparatus on
undisturbed peat and organic clay, as well as the use of
in-situ tests with a at dilatometer (DMT).
2. Importance of K0 value at various OCR
2.1. OCR for peat ground
Peat is usually found directly from the ground surface.
Since the wet density of peat is very low and the groundwater level in the peat ground is very close to the ground
surface, the effective overburden pressure is very low unless
articial loading is applied. Furthermore, the groundwater
level in the peat ground changes seasonally, so peat ground
tends to be overconsolidated, even with no increase in an
articial stress. Futatsukawa and Kikuchi (1988) performed
a detailed survey of changes in the groundwater levels of
peat ground from 1985 to 1987 in the suburbs of Sapporo,
Hokkaido. They found that the groundwater level was the
highest from March to April and at its lowest during
summer. The change in the groundwater level between these
two seasons varies from 50 cm to 1 m. Let us consider the
OCR for an imaginary ground consisting of a top usual soil
layer with 0.5 m thickness and a peat layer with 4.5 m
thickness, whose gt are 15.0 kN/m3 and 10.3 kN/m3, respectively (Fig. 1). For this ground condition, the change in the
OCR will be calculated when the groundwater level drops by
1 m. These adopted numbers are typical values for the peat
ground in Hokkaido (CERI, 2002). Two cases of the initial
groundwater level are examined: the ground surface and the
boundary between the top soil and the peat layer. The
calculated OCR is shown in Fig. 1, indicating that the OCR
increases between 2 and 5 with a 1 m lowering of the
groundwater level.
In addition to these seasonal uctuations, the groundwater levels in the peat ground are also widely inuenced
by drainage, ranging in an order estimated between several

Soil type

Wet density
3
(kN/m )

OCR
0

Surface
soil

15.0
1

Peat

Depth (m)

300

10.3

Drop by 1m from WH
(GL-0m)
Drop by 1m from WH
(GL-0.5m)

WH: groundwater level

Fig. 1. Calculated OCR under the assumption of typical peat ground.

:Settlement plate
:Pore water pressure meter
:Stakes

30.0
22.0

4.0
4.25
0.75
GL-4.80
GL-7.95
GL-12.00

Sm
Ap

Bk

Aco
As

GL-2.8
GL-5.8
GL-10.4

Ac

GL-17.8

4.0
1.5 9.0

3.5
0.42

GL-22.80
Borehole inclinometer

unit: (m)

Fig. 2. Cross-section of the embankment and the subsoil at the Toubetsu

site: Bk: embankment, Sm: sand mat.

dozen centimeters to about 1 m (Hokkaido branch of JGS,

2002). As will be shown later in Figs. 710, which describe
the soil proles of the four investigation sites, peat soils are
slightly overconsolidated. Therefore, it is necessary to
carefully examine the relationship between the OCR and
K0 under in-situ stress to determine the K0 value for peat
ground.
2.2. Numerical evaluation for effects of K0 value on lateral
deformation
To evaluate the inuence of the K0 value on calculated
deformation, an FE analysis was conducted on a test
embankment constructed in Toubetsu, Hokkaido. Fig. 2
shows a cross-section of the test embankment and soil
layers. Fig. 3 shows the geotechnical properties at the site.
The thickness of the peat layer (Ap) was 4.8 m. Its water
content and ignition loss range were from 500800%
and 6090%, respectively. Organic clay (Aco), alluvial
sand (As) and clay layers (Ac) were found below the peat
layer. The permeability of the sand layer located under the

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

Soil type

Natural water content

wn (%)

qc
(MN/m2)

groundwater
level
GL-0.42m

0.1

1.0

10.0

500

1000

Ignition loss
Li (%)

50

301
Yield stress
Pc (kN/m2)

Compression index
Cc

100

10

50

100 150

OCR

1.0

2.0

3.0

4.0

0
Peat (Ap)
5

Organic clay
(Aco)

10
Depth (m)

sand
(As)

15
Clay
(Ac)
20
effective overburden
stress
25

Fig. 3. Geotechnical proles at the Toubetsu site.

Aco layer is high enough to provide drainage. The height

of the embankment was 5 m, and its construction process
is illustrated in Fig. 4.
A soil/water-coupled elasto-plastic FE analysis was
conducted based on the nite deformation theory, which
is considered a suitable method for peat ground with
extremely large deformation (Yamazoe and Mitachi,
2007). In this analysis, the modied Cam-clay model was
used to describe the dilatancy characteristics of peat,
organic clay and clay layers. The relationship between
the specic volume (v) and effective mean stress (p0 ) was
assumed to be linear in ln vln p0 plot. The purpose of the
FE analysis in this study is to clarify the importance of K0
value for peat by demonstrating the inuence of differences
in the K0 value on analytical results. It is known that
lateral displacement calculated through FE analysis is
greatly affected by differences in constitutive models.
Mitachi et al. (2010) studied the extent to which several
constitutive models with different yield surface shapes,
including the original Cam-clay model (Roscoe et al.,
1963), the modied Cam-clay model (Roscoe and
Burland, 1968), the SekiguchiOhta model (Sekiguchi
and Ohta, 1977) and another model developed by introducing the anisotropy parameter Zn (Sekiguchi and Ohta,
1977) into the modied Cam-clay model (Hashiguchi and
Chen, 1998; Asaoka et al., 2000; Mizuno et al., 2007), were
applicable to the deformation analysis of peat ground. The
results showed that models with lower levels of strength and
deformation resistance in the extension mode tended to
result in the overestimation of lateral displacement. Examination of analysis technique differences between innitesimal deformation theory and nite deformation theory also
revealed that analysis based on the latter (Yamazoe and
Mitachi, 2007) was more applicable to peat ground. The

models and analysis method used in this study were based on

the results obtained by Mitachi et al. (2010). The sand and
embankment were assumed to be elastic materials.
As shown in Table 1, the K0OC value of the peat layer
was changed from 0.2 to 1.0 to examine its inuence.
Table 2 lists other soil parameters used for the analysis,
based on the results from oedometer, triaxial compression
(CU ) and physical index tests. However, the effective
angle of shear resistance (f0 ) of the peat (Ap) layer was
found using the estimation equation (Eq. (1)) presented by
Hayashi et al. (2007); where Li is ignition loss (%):
f0 0:19Li 32

The soil parameters were determined in line with the

procedure proposed by Hayashi et al. (2007) as follows.
The compression index ln in relationships between ln v and
ln p0 was determined by the oedometer test. The swell index
kn was assumed to be 0.1 times of the ln. The initial
volume ratio v0 was calculated using the natural water
content and the soil particle density. The critical state
parameter M was determined using the f0 given above.
The OCR was the ratio of the consolidation yield stress
obtained from the oedometer test to the effective overburden pressure calculated by the wet density of each
soil layer.
Hayashi et al. (2008) reported that the coefcient of
permeability (kf) in peat layers measured by the eld test is
greater than that measured by the oedometer test (kl), and
kf is 1030 times as high as kl. Based on these results, k for
the FE analysis was assumed to be 10 times larger than
that measured by the oedometer test. The change in the
coefcient of permeability lnk was determined from the
relationship between the void ratio and the coefcient in a
logarithmic expression (the e-log k relationship) under

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

302

Thickness of embankment (m)

6.0
Actual
Modeling

5.0
4.0
3.0
2.0
1.0
0.0

100

100

200
300
Elapsed time (days)
200

300

400

500

400

500

0.0
Measured

K0OC = 0.2

K0OC = 0.4

K0OC = 0.6

K0OC = 0.8

K0OC = 1.0

Surface settlement (m)

-1.0

between 0 and 86 days), but the settlement caused by the

embankment construction (86 days later) can be well
predicted. Although calculated settlements decreased
slightly with increase in the K0OC value, no signicant
difference was observed, considering that practical estimates in peat ground tend to be out by 10 cm or more.
Fig. 5 shows the observed lateral displacement of the
ground under the toe of the embankment at the completion of the embankment (236 days later). It was as large as
60 cm at a depth of 4 m, which is about 25% of the
settlement measured at the center of the embankment. As
shown in the gure, calculated lateral displacements are
very sensitive to the order of the K0OC value, and decrease
with increase in the K0OC value. The difference in the
maximum lateral displacements with K0OC values from 0.2
to 1.0 is 22 cm. It should be noted that, in practice, the
required accuracy is sometimes as small as 1 cm for
existing facilities near the embankment, so that the
importance of the K0OC value in the analysis needs to be
realized.

3. Investigation sites
-2.0

-3.0

-4.0
Fig. 4. Time history of the embankment construction and the settlement
of the ground surface at the center of the embankment.

Table 1
K0oc of peat layer used.
Case

K0OC

1
2
3
4
5

0.2
0.4
0.6
0.8
1.0

normal consolidation conditions obtained from the

oedometer test.
Fig. 4 shows the process of the embankment construction and the observed settlement on the ground surface at
the center of the embankment. The embankment construction commenced 86 day after sand mat construction. Large
settlement (more than 3 m) due to the embankment load
was observed. The analysis slightly overestimated the
settlement due to the sand mat construction (elapsed time

Investigations of the K0 values of peat ground were

performed at four sites: Mihara and Shinotsu in Ebetsu
City; Tsuruno in Kushiro City; and Riyamunai in Kyouwa
town (Fig. 6). In the laboratory, the following tests were
conducted: a K0-consolidation test using a triaxal testing
apparatus, a consolidated undrained triaxial compression
test, an oedometer test, and several other tests for physical
properties. The in-situ tests performed were a dilatometer
(DMT) and an electric cone penetration test (CPT).
Figs. 710 show the geotechnical properties of the
investigated site in this study. The yield stress pc in the
gures was determined using the e-log p relationship
obtained from the oedometer test. As the pc value for peat
was extremely small, the consolidation pressure for the rst
step in the oedometer test was set at 5 kN/m2.
The top layer at the Mihara site consists of peat with
400500% natural water content. The underlying soils are
organic clay, sandy silt and clay, which is a typical
sequence of the ground covered by peat at the ground
surface found in Hokkaido. The OCR of the peat is in the
range of 12, indicating it is slightly overconsolidated.
The Shinotsu site is located near the Mihara site.
Therefore, the ground composition at the Shinotsu is
similar to that at Mihara, except for the relatively large
OCR for the peat ground.
Fig. 9 illustrates the engineering properties of the ground
at the Tsuruno site. The soft deposits consisting of
alternating layers of peat, organic clay and clay layers at
the Tsuruno site are thinner than that at the other two
sites. The OCR of the peat layer is less than 2, which is
lower than the OCRs of the organic clay and clay layers.
The natural water content of the peat at the Riyamunai
site is 500850% and its OCR ranges between 2 and 3.

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

303

Table 2
Soil parameters used.

Bk
Sm
Ap-u
Ap-l
Aco
As
Ac-u
Ac-l

Depth (m)

Model

rt (g/cm3)

E (kN/m2)

ln

kn

v0

OCR

K0NC

K0OC

k0 (m/d)

lnk

5.000.75
0.750.00
0.00 to  2.40
 2.40 to 4.80
 4.80 to 7.95
 7.95 to 12.00
 12.00 to 17.40
 17.40 to 22.80

B
B
A
A
A
B
A
A

1.70
1.70
1.09
1.06
1.41
1.80
1.73
1.66

5600
5600

9100

0.33
0.33
0.22
0.22
0.30
0.34
0.32
0.36

0.305
0.356
0.130

0.099
0.120

0.031
0.036
0.013

0.010
0.012

8.8
14.8
3.35

1.90
1.90
1.38

1.30
1.03

3.20
2.27
1.26

1.31
1.00

0.28
0.28
0.44
0.52
0.47
0.56

0.73
0.55
0.50

0.54
0.56

2.0E 02
3.0E 02
3.0E 03
1.4E 00
4.0E 04
3.0E 04

0.201
0.196
0.109

0.103
0.129

2.18
2.43

Model-A: Modified Cam-clay model, Model-B: Elastic material, rt: wet density, E: modulus of elasticity, n: Poissons ratio, ln, kn: compression index, swell
index in relationships between ln v and ln p0 , v0: initial volume ratio, M: critical state parameter, OCR: over consolidation ratio, K0NC: coefcient of earth
pressure at rest in normally consolidated condition, K0OC: coefcient of earth pressure at rest in overconsolidated condition, k0: initial coefcient of
permeability, lnk: change in permeability coefcient in relationships between ln v and ln p0 .

Lateral displacement (m)

inside of embankment
-0.5
0

0.0

outside of embankment
0.5

1.0

Depth (m)

10

15
Measured

K0OC = 0.2

K0OC = 0.4

K0OC = 0.6

K0OC = 0.8

K0OC = 1.0

20
Fig. 5. Lateral displacement of the ground under the toe of the
embankment slope at the completion of the embankment construction.

4. Testing procedures
4.1. K0-consolidation test
A K0-consolidation test using triaxial testing apparatus
(Fig. 11) was conducted in accordance with Japanese
Geotechnical Society standards (JGS 0525-2000) on undisturbed samples that were collected using a thin-walled tube
sampler with a xed piston and an inside diameter of
75 mm. To avoid unnecessary disturbance, the diameter of

the peat specimen was the same as the sample retrieved

from the sampling tube, i.e., 75 mm, and the height of the
specimen was 75 mm. Otherwise, specimens of the other
soil types were trimmed to 50 mm in diameter and 100 mm
in height. To accelerate the dissipation of the excess pore
water pressure, four pieces of lter paper were placed on
the circumference of the test specimens. Furthermore, a
nylon mesh was placed on both ends of the test specimens.
At the initial state of the consolidation, isotropic consolidation pressure was applied. The intensity was low
enough not to cause an excess lateral displacement, i.e.,
2 kN/m2 for peat and 5 kN/m2 for the organic clay test
specimens. Then, the lateral pressure in the test was
automatically controlled so that the lateral strain of the
specimen should be zero, calculated from the axial displacement and the volume change.
To avoid any decrease in effective load by the generation of
excess pore water pressure, axial loading was applied at the
slow rate of 0.2 kN/m2/min. In this study, the pore water
pressure was not measured because the top and bottom ends
of the test specimens were kept drained. However, the
generated excess pore water pressure in the specimen was
considered small enough to accurately measure the K0 value
during consolidation for the following reason. Oda and
Mitachi (1992) performed a K0-consolidation test in a triaxial
cell on clay samples with a coefcient of consolidation (cv) of
1013 cm2/day. They reported that the measured excess pore
water pressure at the bottom of the specimen was less than
10% of the vertical stress under an axial loading rate of
0.9 kN/m2/min or lower. The axial loading rate adopted in
this study is slower than this rate and the cv of peat and
organic clay is similar to or approximately 10 times the values
of the clay they used. Therefore, it would follow that almost
no excess pore water pressure was generated under the axial
loading rate of 0.2 kN/m2/min in this experiment.
It was observed that the lateral strain of the test
specimens during the K0-consolidation was about 0.01
0.03%, which is sufciently below the value of 0.05%
specied by the Japanese Geotechnical Society (JGS 05252000). A comparison of the specimen before and after the

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

304

Fig. 6. Location of the investigation sites in this study (distribution of peaty soft ground by Noto, 1991).

Depth
(m)

Soil type

2.2

Peat

4.2

Organic clay

groundwater
level
GL-1.2m

qt (CPT)
0

(MN/m2)
5

10

Natural water content

Wn (%)
0
300
600

Compression index
Cc
0.0
3.0
6.0

Yield stress
Pc (kN/m2)
0

100

OCR
200

0.0

1.0

2.0

Sandy silt
Depth (m)

10.8

Clay

10

15

20

20.5

effective
overburden stress
25

Fig. 7. Geotechnical proles at the Mihara site.

Depth
(m)

Soil type

groundwater
level
GL-0.2m

q t (CPT)

0
2.7
4.2
5.3

(MN/m2)
5
10

Natural water content

W n (%)
0

300

600

Yield stress
P c (kN/m2)

Compression index
Cc
0.0

3.0

6.0

60

OCR
120

Peat
Organic clay
Peat

Fine sand

Silt

20.3
22.0

Depth (m)

8.8
10

15

20
effective
overburden stress

Silt / Fine sand

25

Fig. 8. Geotechnical proles at the Shinotsu site.

0.0

3.0

6.0

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

qt(CPT)
(MN/m2)

Soil type

2.1

Peat

2.6

Organic clay

3.4

Clay

3.9
4.3
4.7
5.2

Fine sand
Sandy clay
Clay
Organic clay

6.2

Peat

6.9

Organic clay

7.5

Fine sand

groundwater
level
GL-1.2m

Compression
index
Cc

Natural water
conetnt
Wn (%)
0

10

300

600

0.0

4.0

OCR

Yield stress
Pc (kN/m2)
0

8.0

50

0.0

100

2.0

4.0

0
1
2
3

Depth (m)

Depth
(m)

305

4
5
6
7

effective
overburden stress

Fig. 9. Geotechnical proles at the Tsuruno site.

Depth
(m)

Soil type

4.8

Peat

qt (CPT)
(M N/m2)

groundwater
level
GL-0.4m

10

Natural water
content
Wn (%)
0
500
1000

Compression
index
CC
0.0
6.0
12.0

OCR

Yield stress
Pc (kN/m2)
0

20

40

0.0

2.0

4.0

Depth (m)

12.0

Clay

10

12
13.0

Fine sand

Fig. 10. Geotechnical proles at the Riyamunai site.

test (see Fig. 12) indicated no noticeable large lateral

deformations.
4.2. Consolidated undrained triaxial compression test
A series of isotropically consolidated undrained triaxial
compression test was performed for sampled specimens,
following the Japanese Geotechnical Society specications
(JGS 0523-2000). The peat specimens were 75 mm in diameter
and 150 mm in height, while the organic clay and clay
specimens were 50 mm in diameter and 100 mm in height.
To obtain strength parameters under the normal consolidated state, the order of the consolidation pressure was
large enough to exceed the consolidation yield stress.

Consolidation was terminated when the volume change

versus logarithm of time (t) curve reaches 3t-line, as
specied by the JGS. After that, the specimen was
subjected to undrained shear at an axial strain rate of
0.1%/min. Although most of the peat exhibited no peak in
axial strain, the effective angle of shear resistance (f0 ) was
dened at an axial strain of 15%.
4.3. Ignition loss test
The specimen was heated at 700800 1C according to
Japanese Industrial Standards (JIS A 1226) with the
exception of the ignition duration: i.e., 4 h, which is 3 h
longer than speicied. This is because Nakayama et al.

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

306

(1989) reported that when ignition time is short, the

organic matter in peat does not completely ignite and an
hour testing underestimates 30% of the real ignition loss
for highly organic soils. In addition to the longer ignition
time, the specimen was prepared for the test at 90 1C rather
than 110 1C since some of the organic matter may have
been burned and be lost at 110 1C. Considering the
heterogeneity of the peat ground, three specimens were
tested and Li in this paper is the average of the measured
values for these specimens.
4.4. Flat Dilatometer Test (DMT)
DMT is an in-situ horizontal loading test, penetrating a
thin blade (93 mm wide  16 mm thick) into the ground
and expanding the center of a metallic membrane in the
blade by gas pressure. From the DMT, three indices
are calculated. Only the horizontal stress index, KD
was discussed in this study for evaluating the K0. KD can
be expressed by Eq. (2), where P0 is the measured gas
pressure at 0.1 mm expansion of the membrane, u0 is the
hydrostatic pressure, and s0v is the effective overburden

pressure.
0

KD P0 2u0 =sv

5. K0 value from laboratory and eld tests

5.1. K0 at normal consolidated state
Fig. 13 plots a typical relationship between the axial
consolidation pressure (s0a) and the K0 value for peat retrieved
from the Tsuruno site at depths of 1.391.56 m. When the
axial consolidation pressure exceeds the consolidation yield
stress and enters the normal consolidation stage, the K0 value
becomes almost constant. The constant K0 value under the
normally consolidated state has been also conrmed by
several researchers, such as Miyakawa et al. (1974),
Fujikawa et al. (1983), Kawano et al. (1986), Edil and
Dhowian (1981), and Mesri and Ajlouni (2007). In this paper,
the K0 value in the normal consolidation is denoted as K0NC.
Fig. 14 shows the relationship between Li and K0NC at
the four investigation sites. The K0NC decreases linearly
with increases in Li (i.e., increases in organic matter
1.5
Peat
1.0
K0

Removaling
K0NC = 0.38

0.5
Yeild stress Pc

Loading

0.0
0

20

40
'

Fig. 11. Set up of K0-consolidation test used.

before testing

60

80

(kN/m2)

Fig. 13. The typical relationship between axial consolidation pressure s0 a

and K0 value (Tsuruno site at depths of 1.391.56 m).

after testing

Fig. 12. Peat specimen tested (Tsuruno, depth: 1.391.56 m, initial isotropic consolidation pressure: 2 kN/m2): (a) before testing and (b) after testing (cut
to two pieces).

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

0.8

Peat
Organic clay
Clay
Mitachi and Fujiwara (1986)

0.8

K0NC = 1-sin
'
0.6

K0NC

0.6

K0NC

307

0.4

0.2

0.4

Peat
Organic clay
Clay

0.2
K0NC = 0.47-0.0025 Li (%)

0.0
0

50
Ignition loss Li (%)

0.0
0.2

100

Fig. 14. The relationship between ignition loss and K0NC.

1.0

Peat

K0

10

Mitachi and Fujiwara (1986) also measured the K0NC of

undisturbed peat collected in Hokkaido, by a similar K0consolidation test using the triaxial test apparatus. Their
measured point is also well tted by this equation.
For the relationship between K0NC and f0 , the following
equation by Jaky (1948) is well known:
K0NC 12sin f0

0.8

Fig. 15. The relationship between K0NC and sin f0 .

content). For organic clay and peat with relatively low Li,
the K0NC is approximately 0.4, which is around 0.1 smaller
than that of inorganic clay. Samples with high Li (Li 85%
and 91%) indicate K0NC values of 0.21 and 0.32, which is
very low compared to that of inorganic clay.
The relationship between Li and K0NC can be approximated as the linear line shown in Fig. 14, including
inorganic clay as well as organic clay. This relation is
approximated as follows, where Li is ignition loss in %.
K0NC 0:4720:0025Li

0.6
sin
'

0.4

Watabe et al. (2003) conrmed that Eq. (4) gives

accurate results for mineral clay. The relationship between
K0NC and sin f0 obtained from the isotropically consolidated undrained triaxial compression test is shown in
Fig. 15. It can be seen that Jaky relation can be also
applied to peat; in other words, the small K0 value for peat
is due to large f0 of peat.

K0OC = 0.41

OCR = 1.37
0.1
10

1
OCR

Fig. 16. The typical relationship between OCR and K0OC (Tsuruno site at
depths of 1.391.56 m).

1.0
m = 0.005 Li (%) + 0.45
0.8

5.2. K0 at the overconsolidated state

As discussed in the previous chapter, the peat layer is in
general somewhat overconsolidated and the K0OC value for
peat is strongly affected by the calculated lateral displacements by FE analysis. After applying an axial consolidation
pressure higher than the consolidation yield stress, unloading
was performed in the K0-consolidation test, as shown in
Fig. 13. It can be seen that the K0 value during the unloading
process is greater than that at the normal consolidated state.
Fig. 16 shows a typical relationship between OCR and K0OC
obtained from the results shown in Fig. 13. At the depth

0.6

Peat
Organic clay
Clay

0.4

0.2
0

50
Ignition loss Li (%)

100

Fig. 17. The relationship between m value and ignition loss.

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

308
Table 3
Summary of test results in this study.

Li (%)

OCR

f0 (deg)

K0NC

K0OC

82
64
57

17

1.9
1.0
1.8

45.3
30.4
31.9

0.57
0.52
0.53

0.49
0.53
0.52

0.77
0.52
0.75

Peat
Organic clay
Organic clay
Peat
Peat
Clay
Clay

407
325
168
333
259
68
67

45
27
17
51
42
6
6

4.2
3.3
3.3
4.0
4.0
1.4
1.2

36.9
36.5
36.5
41.3
41.3
30.4
31.9

0.29
0.30
0.39
0.30
0.38
0.53
0.53

0.62
0.57
0.61
0.67
0.65
0.51
0.47

0.69
0.56
0.82
0.67
0.80
0.61
0.55

1.391.56
1.902.07
2.933.11
4.474.64
4.774.95
5.295.47
6.476.65

Peat
Peat
Clay
Clay
Organic clay
Peat
Organic clay

423
421
81
96
91
540
176

52
34

17
85
14

1.4
1.4
3.2
3.0
1.5
1.3
2.0

42.5
42.5
32.1
32.1

46.4

0.38
0.41
0.45
0.45
0.43
0.21
0.43

0.64
0.69
0.54
0.49
0.48
0.83
0.53

0.41
0.47
0.87
0.70
0.50
0.23
0.58

2.412.59
4.574.74
7.477.60

Peat
Peat
Clay

913
628
85

91
59
11

1.8
2.3
4.2

48.1
50.7
30.0

0.32
0.35
0.48

0.90
0.61
0.48

0.53
0.55
0.87

wn (%)

Site

Depth(m)

Soil type

Mihara

3.003.86
12.5013.36
16.5017.36

Organic clay
Clay
Clay

Shinotsu

2.142.31
2.993.17
3.834.00
4.184.35
5.025.19
12.4312.55
16.4516.59

Tsuruno

Riyamunai

1.0

1
peat
organic clay
clay

1:1.3

1:0.8

K0

Marchetti (1986)
K0OC = (KD/1.5)0.47-0.6

Iwasaki (1999)
K0OC = 0.29 KD0.57

K0DMT = 0.5 KD0.22

0.8

K0OC = 0.5 KD0.22

0.1
0.01

1:1

0.6

0.4
peat
organic clay
clay

0.2

0.1

10

KD

0.0
0.0

Fig. 18. The relationship between KD and K0 obtained from K0consolidation tests.

0.2

0.4

0.6

0.8

1.0

K0
Fig. 19. The relationship between K0DMT estimated by the proposed
equation and K0 value from K0-consolidation tests.

where this sample was retrieved, the OCR was 1.37, and the
K0OC at the in situ was estimated as 0.41.
It is well known that the K0OC for inorganic soils
increases linearly with the OCR in the loglog plot.
Schmidt (1966) proposed the following equation for the
relationship between OCR and K0OC:
K0OC K0NC OCRm

where m is the gradient when the K0OC and OCR are

plotted in loglog scale.
In this study, the relationship between OCR and K0OC
for peat as well as organic clay exhibited the same relation
as inorganic soils, as shown in Fig. 16. That is, the K0 of

peat and organic clay are dependent on stress history and

can be expressed by a function of OCR.
Fig. 17 shows the relationship between the m value and Li.
The m value for inorganic clay obtained by this study ranged
between 0.46 and 0.52. Watabe et al. (2003) reported that m
values measured from a large number of clayey sites around
the world have similar m values. Hamouche et al. (1995)
reported that clays from sites in North America present the m
value of 0.40.5 and some sensitive clays show relatively high
m values. These researchers suggest that most inorganic clays
have the m value of approximately 0.5, except for some very
sensitive clays. In contrast, organic clay and peat have very
high m values, particularly peat. For the sample with high

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

ignition loss (Li 91%), the m value was as large as 0.90.

This indicates that the K0OC of peat is more strongly affected
by OCR than that of inorganic clay. Furthermore, it was
found that the m value increased linearly with increases in Li
(see Fig. 17), and this relationship can be approximated as
follows:
m 0:005Li 0:45

results in this study are summarized in Table 3. The Wn, Li

and f0 values for peat and organic clay shown in
Table 3 are within the range of previous data obtained in
Hokkaido region (JGS, 1990).
5.3. K0 obtained from DMT

For heterogeneous materials such as peat, a simple

in situ test would be the most preferable method since it
does not require soil samples whose properties are signicantly changed vertically and horizontally. The DMT
seems an effective method for estimating the K0 value of
peat ground because of a simple and convenient in situ test
method. In this study, a method for estimating the K0
value from DMT was examined. As mentioned earlier,

Combined with Eqs. (3), (5) and (6), K0OC can be nally
expressed as follows, in terms of Li and OCR:
K0OC 0:4720:0025Li OCR0:005Li 0:45

Considering heterogeneity in the peat layer, Eq. (7) is

very useful in practice because it consists of relatively
simple and robust parameters of Li and OCR. The test

Depth
(m)

K0

KD

Soil type
0

Peat
2.7
4.2 Organic clay
5.3
Peat

309

10

0.0

0.5

1.0

1.5

Fine sand

Depth (m)

8.8

Clay

10

15

20.3

20
Iwasaki (1999)
Marchetti (1980)
Proposed Eq.(11)
Lab test

22.0 Silt/Fine sand

25

Depth
(m)

Soil type
0
Peat

6.2

Clay

Peat

6.9 Organic clay

7.5

Fine sand

0.0

0.5

1.0

1.5

3
Depth (m)

3.9
4.3
4.7
5.2

K0
10

2.6 Organic clay

Fine sand
Sandy clay
Clay
Organic clay

KD
5

2.1

3.4

4
5
6
7
8

Iwasaki (1999)
Marchetti (1980)
Proposed Eq. (11)
Lab test

Fig. 20. The depth distribution of K0 and K0DMT: (a) Shinotsu site and (b) Tsuruno site.

310

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

KD was an index obtained by DMT, and using KD, several

equations have been proposed to estimate K0, as follows
(Eq. (8)Marchetti, 1980; Eq. (9)Iwasaki, 1999):
K0DMT KD =1:50:47 0:6

K0DMT 0:29KD0:57

where K0DMT is estimated K0 value using DMT.

Fig. 18 shows the relationship between KD and K0 obtained
from the K0-consolidation tests in a loglog plot. The
measured trend in the relationship between KD and K0 for
peat layers is very different from the lines obtained using
Marchettis and Iwasakis equations, especially in small KD.
The best tted line for the peat layer is the following equation:
K0DMT 0:5KD0:22

10

To verify the applicability of Eq. (10), the relationship

between the K0DMT estimated by Eq. (10) and the K0 values
from K0-consolidation tests is shown in Fig. 19. The ratios of
K0 to K0DMT are approximately in the range of 0.81.3.
Hayashi and Nishimoto (2006) reported that the ratios of K0
to K0DMT estimated by Marchettis and Iwasakis equations
for clay layers in Hokkaido are approximately 0.61.2. These
results show that the applicability of Eq. (10) for peat and
organic clay layers is similar to that of Marchettis and
Iwasakis equations for clay.
Fig. 20 shows a comparison of the estimated K0 values by
DMT using different equations including the proposed one
in this paper, with obtained K0 values from K0-consolidation tests, at the Shinotsu and the Tsuruno sites. The
applicability in K0 estimated by Marchettis and Iwasakis
equations for inorganic clay is relatively good. However,
large scatters are observed in the relationship between
estimated K0DMT and K0 for the peat and organic clay
layers. Especially when the KD is very low (KD 0.070.18),
the estimates become increasingly inaccurate. This is particularly true where the points show the data at the Tsuruno
site: the K0DMT estimated by Marchettis equation presents
a negative value. For peat and organic clay layers, Eq. (10)
provides much more accurate K0 values.
6. Conclusions
The value of K0 for peat and organic clay was obtained
using triaxial testing apparatus and a method for estimating K0 was proposed in terms of simple soil parameters:
OCR and the ignition loss. A at dilatometer test was also
performed and a new correlation of K0 with KD was
established. The main results of this study are summarized
below:
(1) The K0NC for peat and organic clay under the normally
consolidated state are smaller than that for usual
inorganic soils, and the K0NC decreases linearly with
increases in the ignition loss (Li). This relationship is
described by K0NC 0.470.0025Li (%).

(2) The peat layer has a tendency to be overconsolidated due

to the seasonal changes in groundwater level. Furthermore, the K0OC value for peat is strongly affected by the
calculated lateral displacements by elasto-plastic FE
analysis. Therefore, it is necessary to carefully examine
the relationship between OCR and K0 value under in-situ
stress to determine the K0 value for peat ground.
(3) The K0 of peat and organic clay are dependent on
stress history and can be expressed by a function of
OCR, as indicated in K0OC K0NCOCRm.
(4) The K0 value for peat and organic clay layers depends
more strongly on OCR than that of inorganic clay. The
m value, which is the power on OCR, increases linearly
with increases in Li. This relationship is described by
m 0.005Li (%) 0.45.
(5) When using DMT to estimate the K0 of peat and
organic clay layers, K0DMT 0.5K0.22
shows higher
D
applicability than the conventional equations.

References
Asaoka, A., Nakano, M., Noda, T., 2000. Superloading yield surface
concept for highly saturated soil behavior. Soils and Foundations
40 (2), 99110.
Civil Engineering Research Institute of Hokkaido, 2002. Manual on
Countermeasures for Peat Ground, pp. 45 (in Japanese).
Edil, J.B., Dhowian, A.W., 1981. At-rest lateral pressure of peat soils.
Journal of the Geotechnical Engineering Division, ASCE 107 (GT2),
201220.
Fujikawa, T., Ogawa, S., Fukuda, M., 1983. Pore water pressure and K0
of clay and peat during consolidation state. In: Proceedings of the
43rd Annual Conference, vol. 3, JSCE, pp. 515516 (in Japanese).
Futatsukawa, K., Kikuchi, J., 1988. Countermeasure and settlement
properties of peat land in Yonesato of Sapporo. Technical Report
of the Annual Meeting of the Hokkaido Branch, vol. 28, JGS,
pp. 191200 (in Japanese).
Hayashi, H., Nishikawa, J., Odajima, H., Mitachi, T., Fukuda, F., 1994.
Deformation analysis of peat ground with Cam clay model. In:
Proceedings of the International Symposium on Pre-failure Deformation Characteristics of Geomaterials, Sapporo, vol. 1, pp. 575581.
Hayashi, H., Nishimoto, S., 2006. Estimation of K0 for peaty soft ground
using at dilatometer. In: Proceedings of the 61st Annual Conference,
vol. 3, JSCE, pp. 347348 (in Japanese).
Hayashi, H., Mitachi, T., Tanaka, H., Nishimoto, S., 2007. Determination procedure of soil parameters for elasto-plastic FE analysis of peat
ground. In: Proceedings of 13th Asian Regional Conference on
SMGE, vol. 1, pp. 145148.
Hayashi, H., Mitachi, T., Nishimoto, S., 2008. Evaluation on permeability
of peat using in-situ permeability test and oedmeter test. Journal of
Geotechnical Engineering, JSCE 64 (3), 495504 in Japanese.
Hamouche, K.K., Leroueil, S., Roy, M., Lutenegger, A.J., 1995. In-situ
evaluation of K0 in eastern Canada clays. Canadian Geotechnical
Journal 32, 677689.
Hashiguchi, K., Chen, Z.P., 1998. Elastoplastic constitutive equation of
soils with the subloading surface and the rotational hardening.
International Journal for Numerical and Analytical Method in
Geomechanics 22, 197227.
Hokkaido Branch of JGS, 2002. Settlement of Peat Ground in Hokkaido
and Its Countermeasures, pp. 45 (in Japanese).
Iwasaki, K., 1999. Evaluation of lateral earth pressure at rest of cohesive soils
using at dilatometer. In: Proceedings of the 30th Japan National
Conference on Geotechnical Engineering, JGS, pp. 271272 (in Japanese).

H. Hayashi et al. / Soils and Foundations 52 (2012) 299311

Jaky, J., 1948. Pressure in silos. In: Proceedings of the 2nd International
Conference on SMGE, vol. 1, pp. 103109.
Japanese Geotechnical Society, 1990. Engineering of Highly Organic Soil.
pp. 158162 (in Japanese).
Japanese Geotechnical Society, 2000. Japanese Standards for Test
MethodsStandards and Explanations (in Japanese).
Kawano, K., Yamaguchi H., Ohira, Y., 1986. K0 of peat. In: Proceedings
of the 41st Annual Conference, vol. 3, JSCE, pp. 477478 (in
Japanese).
Marchetti, S., 1980. In-situ tests by at dilatometer. Journal of the
Geotechnical Engineering Division, ASCE 106 (GT3), 229321.
Mesri, G., Ajlouni, M., 2007. Engineering properties of brous peats.
Journal of the Geotechnical and Geoenvironmental Engineering,
ASCE 133 (7), 850866.
Mitachi, T., Fujiwara, Y., 1986. Consolidated undrained triaxial shear
behavior of peat. In: Bulletin of the Faculty of Engineering, Hokkaido
University, vol. 129, pp. 114 (in Japanese).
Mitachi, T., Yamazoe, N., Hayashi, H., Ogino, T., 2010. Applicability of
several constitutive models and methods of deformation analysis on
peaty soft ground. Journal of Geotechnical Engineering, JSCE 66 (1),
120 (in Japanese).
Miyakawa, I., Shibata, T., Takano, T., 1974. The strength characteristics
of organic soils using K0-consolidation. In: Proceedings of Annual
Meeting of Tohoku Branch, JSCE, pp. 4446 (in Japanese).
Mizuno, K., Tsuchida, T., Kobayashi, M., Watabe, Y., 2007.
New constitutive equation of clay and the prediction of horizontal

311

deformations of large scale embankments. Journal of Geotechnical

Engineering, JSCE 63 (4), 936953 (in Japanese).
Nakayama, M., Yamaguchi, H., Ohira, Y., 1989. Consideration on test
method for ignition loss of soils (part 1). In: Proceedings of the 44th
Annual Conference, vol. 3, JSCE, pp. 864865 (in Japanese).
Noto, S., 1991. Peat Ground Engineering. Gihodo-shuppan Co., Ltd.,
Tokyo, pp. 45 (in Japanese).
Oda, Y., Mitachi, T., 1992. Triaxial K0-consolidation test on saturated
clay. Journal of Geotechnical Engineering, JSCE 448/III-19, 4552.
Roscoe, K.H., Schoeld, A.N., Thurairajah, A., 1963. Yielding of clays in
states wetter than critical. Geotechnique 13 (3), 221240.
Roscoe, K.H., Burland, J.B., 1968. On the Generalized StressStrain
Behavior of wet Clay, Engineering Plasticity. Cambridge University
Press, pp. 535609.
Schmidt, B., 1966. Discussion of earth pressures at rest related to stress
history. Canadian Geotechnical Journal 3, 239242.
Schoeld, A.N., Wroth, C.P., 1968. Critical State Soil Mechanics.
McGraw-Hill, London.
Sekiguchi, H., Ohta, H., 1977. Induced anisotropy and time dependency
of clay. In: Proceedings of 9th International Conference on SMFE,
vol. 1, pp. 229239.
Watabe, Y., Tanaka, M., Tanaka, H., Tsuchida, T., 2003. K0-consolidation in a triaxial cell and evaluation of in-situ K0 for marine clays with
various characteristics. Soils and Foundations 43 (1), 120.
Yamazoe, N., Mitachi, T., 2007. Finite deformation elasto-plastic FE
analysis on the peat ground under vacuum surcharge preloading. Journal
of Geotechnical Engineering, JSCE 63 (4), 11131131 (in Japanese).