on the homogeneity of undisturbed clay samples...

57ième CONGRÈS CANADIEN DE GÉOTECHNIQUE 57TH CANADIAN GEOTECHNICAL CONFERENCE5ième CONGRÈS CONJOINT SCG/AIH-CNN 5TH JOINT CGS/IAH-CNC CONFERENCE

ON THE HOMOGENEITY OF UNDISTURBED CLAY SAMPLES RETRIEVED

FROM OSAKA BAY Yoichi Watabe, Port and Airport Research Institute, Yokosuka, JapanTomoko Takemura, Port and Airport Research Institute, Yokosuka, JapanYasunori Shiraishi, Nippon Koei Co., Ltd., Tokyo, JapanTomohide Murakami, Saeki Kensetsu Kogyo Co., Ltd., Tokyo, Japan

ABSTRACTSamples collected from both Holocene and Pleistocene layers in Osaka Bay were examined. The retrieved sample with alength of about 1 m was divided into every 25 mm in order to trim consolidation specimens, and the followings wereexamined: i) natural water content; ii) Atterberg’s limits; iii) grain size fractions; iv) bulk density; v) particle density; vi) voidratio; vii) consolidation yield stress; viii) compression index; ix) coefficient of consolidation; and x) SEM photographing. Theobjective of this study is to evaluate variations of soil parameters obtained from laboratory tests. It is concluded that the variation of physical and mechanical properties is mainly influenced by macro-scale heterogeneity or sample disturbance,since specimen size for laboratory tests is sufficiently large on microscopic heterogeneity.

RÉSUMÉLes échantillons d'argile se sont rassemblés des couches holocène et pléistocène dans la baie d'Osaka ont été examinées.Les échantillons récupéré en une longueur d'environ 1 m a été divisé en tous les 25 millimètres afin d'couper des spécimensde consolidation, et les paramètres suivants ont été examinée: i) teneur en eau naturelle; ii) limites des liquidité et plasticité;iii ) fractions de granulométrie; iv ) densité en humide; v ) densité des grains; vi ) indice des vides; vii ) pression depréconsolidation; viii) indice de compression; ix) coefficient de consolidation; et x) photographie de microscope électroniquede balayage. L'objectif de cette étude est d'évaluer des variations des paramètres d’argile à la baie d’Osaka.

1. INTRODUCTION

Table 1. List of tested samples.Osaka Bay clay retrieved from the seabed was investigatedfor the Kansai International Airport project, which isconstructed on an artificial island in Osaka Bay 5 km off the Senshu area 35 km southwest of Osaka City. The first phase of the Airport with only one runway was in operationfrom September, 1994, and the second phase is underconstruction for inauguration with two parallel runways in2007. Consolidation parameters obtained from consolidationtests, carried out every 1—2 m depth(s), show a significantvariation (Tanaka et al. 2003a). It has not been cleared yetwhether the variation was caused by sample disturbance or derived from sample heterogeneity, even in a precise conepenetration test applied to great depths (Tanaka et al. 2003b).

ElevationEL. (–m)

Atter-berglimits

Grainsize

CRS SEMElevationEL. (–m)

Atter-berglimits

Grainsize

CRS SEM

37.500–37.525 125.500–125.52537.525–37.550 125.525–125.55037.550–37.575 125.550–125.57537.575–37.600 125.575–125.60037.600–37.625 125.600–125.62537.625–37.650 125.625–125.65037.650–37.675 125.650–125.67537.675–37.700 125.675–125.70037.700–37.725 125.700–125.72537.725–37.750 125.725–125.75037.750–37.775 125.750–125.77537.775–37.800 125.775–125.80037.800–37.825 125.800–125.82537.825–37.850 125.825–125.85037.850–37.875 125.850–125.87537.875–37.900 125.875–125.90037.900–37.925 125.900–125.92537.925–37.950 125.925–125.95037.950–37.975 125.950–125.97537.975–38.000 125.975–126.00038.000–38.025 126.000–126.02538.025–38.050 126.025–126.05038.050–38.075 126.050–126.07538.075–38.100 126.075–126.10038.100–38.125 126.100–126.12538.125–38.150 126.125–126.15038.150–38.175 126.150–126.17538.175–38.200 126.175–126.20038.200–38.225 126.200–126.22538.225–38.250 126.225–126.25038.250–38.275 126.250–126.27538.275–38.300 126.275–126.30038.300–38.325 126.300–126.32538.325–38.350 126.325–126.35038.350–38.375 126.350–126.37538.375–38.400 126.375–126.40038.400–38.425 126.400–126.42538.425–38.450 126.425–126.45038.450–38.475 126.450–126.47538.475–38.500 126.475–126.500

In this study, samples collected from both Holocene andPleistocene layers were examined. The retrieved samplewith a length of about 1 m was divided into every 25 mm inorder to trim consolidation specimens, and the followingparameters were examined: i) natural water content (wn); ii)Atterberg’s limits (wL and wp); iii ) grain size fractions (clayfraction, silt fraction, and sand fraction); iv) bulk density ( t);v ) particle density ( s); vi ) void ratio (e); vii ) consolidationyield stress (p'c); viii) compression index (Cc); ix) coefficient of consolidation (cv); and x ) scanning electron microscopic(SEM) photographing. The objective of this study is toevaluate variations of soil parameters obtained fromlaboratory tests. The test program corresponding to thedivided samples is listed in Table 1.

Session 7E

1.4 1.5 1.6

Bulk density t (Mg/m3)

(c)

t=1.550±0.010 Mg/m3

2.5 2.6 2.7 2.8

Particle density s (Mg/m3)

(d)

s=2.644±0.028 Mg/m3

1.8 1.9 2.0 2.1 2.2

Void ratio e

(e)

e =1.954±0.058 %

0 25 50 75 100

Fraction (%)

0.005mm0.075mm

clay silt

(b)

-38.3

-38.2

-38.1

-38.0

-37.9

-37.8

-37.7

-37.6

-37.5

0 25 50 75 100

Wataer content w (%)

Ele

vatio

n E

L. (

m)

p n L(a)

w n=73.09±2.06 %

1.50

-1.5

11.

51-1

.52

1.52

-1.5

31.

53-1

.54

1.54

-1.5

51.

55-1

.56

1.56

-1.5

71.

57-1

.58

1.58

-1.5

91.

59-1

.60

t

2.60

-2.6

12.

61-2

.62

2.62

-2.6

32.

63-2

.64

2.64

-2.6

52.

65-2

.66

2.66

-2.6

72.

67-2

.68

2.68

-2.6

92.

69-2

.70

s

64-6

666

-68

68-7

070

-72

72-7

474

-76

76-7

878

-80

80-8

282

-84

w n

1.86

-1.8

81.

88-1

.90

1.90

-1.9

21.

92-1

.94

1.94

-1.9

61.

96-1

.98

1.98

-2.0

02.

00-2

.02

2.02

-2.0

42.

04-2

.06

e

Figure 1. Profiles of physical properties with depth in the Holocene clay.

1.6 1.7 1.8

Bulk density t (Mg/m3)

(c)

t=1.726±0.030 Mg/m3

2.5 2.6 2.7 2.8

Particle density s (Mg/m3)

(d)

s=2.695±0.013 Mg/m3

-126.3

-126.2

-126.1

-126.0

-125.9

-125.8

-125.7

-125.6

-125.5

0 25 50 75 100Water content w (%)

Ele

vatio

n E

L. (

m)

p n L(a)

w n=41.76±2.19 %

1.0 1.1 1.2 1.3 1.4

Void ratio e

(e)

e =1.215±0.068 %

0 25 50 75 100

Fraction (%)

0.005mm0.075mm(b)

clay silt

1.68

-1.6

91.

69-1

.70

1.70

-1.7

11.

71-1

.72

1.72

-1.7

31.

73-1

.74

1.74

-1.7

51.

75-1

.76

1.76

-1.7

71.

77-1

.78

t

2.65

-2.6

62.

66-2

.67

2.67

-2.6

82.

68-2

.69

2.69

-2.7

02.

70-2

.71

2.71

-2.7

22.

72-2

.73

2.73

-2.7

42.

74-2

.75

s

30-3

232

-34

34-3

636

-38

38-4

040

-42

42-4

444

-46

46-4

848

-50

w n

1.10

-1.1

21.

12-1

.14

1.14

-1.1

61.

16-1

.18

1.18

-1.2

01.

20-1

.22

1.22

-1.2

41.

24-1

.26

1.26

-1.2

81.

28-1

.30

e

Figure 2. Profiles of physical properties with depth in the Pleistocene clay.

2. SAMPLE

Osaka Bay clay examined in this study was sampled duringthe ground survey for the second phase construction ofKansai International Airport. The deep sampling waseffectively carried out by PARI (formerly PHRI) wire line method (Horie et al. 1984; Kanda et al. 1991) as illustratedin Watabe et al. 2002. At the sampling site, the water depthwas 19 m, the thickness of Holocene clay layer was 25 m,

and the Pleistocene clay deposited more than 350 m thick alternately with some sandy layers. The clay at G.L. –120deposited 270,000 years ago. The clay layer shows slightoverconsolidation (OCR = 1.1—1.5), but it is, generally, innormal consolidation (Akai et al. 1995; Akai, 2000; Watabeet al. 2002). Tanaka and Locat (1999) indicated that theconsolidation behavior is possibly characterized by thediatoms which are abundantly contained in the Osaka Bayclay. In this study, both samples retrieved from a Holocene

Session 7E

-38.3

-38.2

-38.1

-38.0

-37.9

-37.8

-37.7

-37.6

-37.5

70 90 110 130 150 170Yield stress p 'c (kPa)

Ele

vatio

nE

L. (

m)

(a)

p 'c=133±10 kPa

'v0

0.4 0.7 1.0 1.3 1.6 1.9

Compression index C c, C c*

Cc* Cc(b)

C c*=0.601±0.063 C c=1.525±0.158

50 60 70 80 90

Coefficient of consolidation c v (cm2/day)

(c)

c v=71.0±5.1 cm2/day

120-

123

123-

126

126-

129

129-

132

132-

135

135-

138

138-

141

141-

144

144-

147

147-

150

p 'c

0.50

-0.5

30.

53-0

.56

0.56

-0.5

90.

59-0

.62

0.62

-0.6

50.

65-0

.68

0.68

-0.7

10.

71-0

.74

0.74

-0.7

70.

77-0

.80

C c*

1.30

-1.3

51.

35-1

.40

1.40

-1.4

51.

45-1

.50

1.50

-1.5

51.

55-1

.60

1.60

-1.6

51.

65-1

.70

1.70

-1.7

51.

75-1

.80

C c

50-5

454

-58

58-6

262

-66

66-7

070

-74

74-7

878

-82

82-8

686

-90

c v

Figure 4. Profiles of consolidation properties with depth in theHolocene clay.

0.0

0.5

1.0

1.5

2.0

2.5

10 100 1000 10000log p ' (kPa)

Vo

id r

atio

e

(a) Holocene clay EL. –37.4—38.4m

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

10 100 1000 10000log p ' (kPa)

Vo

id r

atio

e

(b) Pleistocene clay EL. –125.4—126.4m

Figure 3. Relationships between void ratio and consolidationpressure.

-126.3

-126.2

-126.1

-126.0

-125.9

-125.8

-125.7

-125.6

-125.5

700 900 1100 1300

Yield stress p 'c (kPa)

Ele

vatio

n E

L. (

m)

(a)p 'c=1028±79 kPa

'v0

0.4 0.6 0.8 1.0 1.2

Compression index C c, C c*

Cc* Cc(b)C c*=0.569±0.048 C c=0.921±0.108

90 100 110 120 130 140 150

Coefficient of consolidation c v (cm2/day)

(c)c v=117.6±12.8 cm2/day

800-

840

840-

880

880-

920

920-

960

960-

1000

1000

-104

010

40-1

080

1080

-112

011

20-1

160

1160

-120

0

p 'c

0.44

-0.4

60.

46-0

.48

0.48

-0.5

00.

50-0

.52

0.52

-0.5

40.

54-0

.56

0.56

-0.5

80.

58-0

.60

0.60

-0.6

20.

62-0

.64

C c*

0.60

-0.6

60.

66-0

.72

0.72

-0.7

80.

78-0

.84

0.84

-0.9

00.

90-0

.96

0.96

-1.0

21.

02-1

.08

1.08

-1.1

41.

14-1

.20

C c

90-9

696

-102

102-

108

108-

114

114-

120

120-

126

126-

132

132-

138

138-

144

144-

160

c v

layer (EL.–126 / G.L.–107 m) and a Pleistocene layer (EL. –38 / G.L. –19 m) were examined.

3. TEST PROGRAM

Atterberg’s limits (wL and wp), soil particle density ( s), andgrain size distribution were examined before carrying out theseries of consolidation tests; then the constant rate strainconsolidation tests (CRS tests) were carried out. Naturalwater content (wn), bulk density ( t), void ratio (e),consolidation yield stress (p'c), compression index (Cc),compression index at a consolidation pressure of greaterthan 5,000 kPa (Cc*), and coefficient of consolidation (cv)were obtained from the CRS tests. The specimens, trimmed into a size of 60 mm in diameter and 20 mm in height, wereexamined with a back pressure of 98.1 kPa and in a strainrate of 0.02%/min.

Using soil chips from specimen trimming, microfabric wasSEM-photographed. In the specimen preparation for SEM,freeze-cut-drying method (Kang et al. 2003) was adopted tominimize the sample disturbance and to obtain a flatobservation surface. Four positions every 0.3 mm on eachspecimen were photographed by 4 different magnifications

Figure 5. Profiles of consolidation properties with depth in thePleistocene clay.

Session 7E

as 5000, 2500, 1000, and 500 times. The SEM photographswere image-analyzed by using the free-software “ScionImage”, which can transform the photographed image ingray scale (0—256) into a binarized image with a thresholdvalue. In this study, a threshold value of 200 was adopted.The black percentage in the binarized black and whiteimage was calculated.

4. TEST RESULTS

4.1 Physical and consolidation properties

Physical properties of the Holocene and the Pleistoceneclays are shown in Figures 1 and 2, respectively.Frequencies of wn, t, s, and e, the mean value (MV), and the standard deviation (SD) are also shown in the figures.Both clays have essentially common properties as wL = 80%, wp = 30%, plasticity index (Ip) of 50, clay fraction (<5 m) of 30—40%, sand fraction (>75 and <2 m) of 0—20%. The wn

of the Holocene clay (73.09±2.06%) is slightly less than wL,and wn of the Pleistocene clay (41.76±2.19%) is slightly

larger than wp. For the both clays, SD of wn is only about 2%.The profile of t of the Holocene clay (1.550±0.010 Mg/m3)is very homogeneous; however, that of the Pleistocene clay(1.726±0.030 Mg/m3) shows slightly larger variation causedby a high t value close to 1.8 Mg/m3 at around EL. –126.05m (wn shows a slightly smaller value).

-38.4

-38.3

-38.2

-38.1

-38.0

-37.9

-37.8

-37.7

-37.6

-37.5

0.6 0.8 1.0 1.2 1.4

Normalized e

Ele

vatio

n E

L.

(m)

(a) e S.D.=0.024

0.6 0.8 1.0 1.2 1.4

Normalized p 'c

S.D.=0.075(b) p 'c

0.6 0.8 1.0 1.2 1.4

Normalized C c

S.D.=0.104(c) C c

0.6 0.8 1.0 1.2 1.4

Normalized c v

(d) c v S.D.=0.072

Figure 6. Profiles of normalized consolidation parameterswith depth in the Holocene clay.

The profile of s of the Holocene clay (=2.644±0.028 Mg/m3)is slightly variable, corresponding to the profile of e

(1.954±0.058). The s shows high values of 2.69—2.70Mg/m3 at round EL. –37.75 m and –38.15 m resulting highere values. This fact means that, if MV of 2.644 Mg/m3 wasadopted as s to calculate e for all depths, e could show avery homogeneous profile. It cannot be judged between theprofile of s and the constant value of s. On the other hand,the profile of s of the Pleistocene clay (=2.695±0.013Mg/m3) is very homogeneous. Thus, the profile of e

corresponds to the profile of t. The profiles at EL. –125.05 m indicate that this depth is silty; however, this cannot beclearly confirmed from grain size fractions and SEM images(Figure 11).

Relationships between void ratio and consolidation stress (e–log p') for (a) Holocene clay and (b) Pleistocene clay areshown in Figure 3. All curves for both clays converge acurve under high p', though initial e values show somevariation. The inversed S-shaped curves, even Holoceneclay, are characteristically observed as aged clay. The curve without a clear yielding point in Figure 3a is EL. –37.59 m,and the three curves from smaller e values in Figure 3b arearound EL. –126.05 m.

Profiles of p'c, Cc, Cc*, and cv with depth for the Holoceneand the Pleistocene clays are shown in Figures 4 and 5,respectively.

In the Holocene clay, p'c (=133±10 kPa) slightly increaseswith depth. The SD of Cc is 0.158, but the SD of Cc* is only0.063. However coefficient of variation (CV) defined by theratio of SD to MV (CV=SD/MV) is the same as 10% for bothCc and Cc*. In addition, the specimen from EL. –37.59 m, which shows smaller p'c and Cc possibly caused by sampledisturbance, is not identified in the profile of Cc*.

-126.4

-126.3

-126.2

-126.1

-126.0

-125.9

-125.8

-125.7

-125.6

-125.5

0.6 0.8 1.0 1.2 1.4

Normalized e

Ele

vatio

n E

L.

(m)

(a) e S.D.=0.056

0.6 0.8 1.0 1.2 1.4

Normalized p 'c

S.D.=0.077(b) p 'c

0.6 0.8 1.0 1.2 1.4

Normalized C c

S.D.=0.117(c) C c

0.6 0.8 1.0 1.2 1.4

Normalized c v

(d) c v S.D.=0.109

Figure 7. Profiles of normalized consolidation parameters with depth in the Pleistocene clay.

In the Pleistocene clay, p'c is evaluated as 1028±79 kPa, butit shows a larger p'c at around EL. –126.05 m. This larger p'cis caused from the unclear p'c on the sloping shoulder of e–log p' curves as shown in Figure 3b. Cc=0.921±0.108 andCc*=0.569±0.048 are about 60% and 95% of those in theHolocene clay, respectively, indicating that both clays areessentially the same consolidation properties in normalconsolidation range. The Cc/Cc* is calculated as 1.6, whichis smaller than that of 2.5 calculated in the Holocene clay.

In order to evaluate the variation of e, p'c, Cc, and cv, these parameters are normalized by their mean values,respectively, and profiles of the normalized parameters ofthe Holocene and Pleistocene clays with depth are shown inFigures 6 and 7, respectively.

In the Holocene clay, SD values of normalized e, p'c, Cc,and cv are 0.024, 0.075, 0.104, and 0.072, respectively. SD

Session 7E

(a) x5000(a) x5000

5 m

(b) x2500(b) x2500

10 m

(c) x1000(c) x1000

10 m

(d) x500(d) x500

50 m

B/(B+W)=15.82%(e) x5000(e) x5000

5 m

B/(B+W)=20.79%(f) x2500(f) x2500

10 m

B/(B+W)=22.07%(g) x1000(g) x1000

10 m

B/(B+W)=22.33%(h) x500(h) x500

50 m

Figure 8. SEM and binarized images of the Holocene clay at EL. –37.96 m.

(a) x5000(a) x5000

5 m

(b) x2500(b) x2500

10 m

(c) x1000(c) x1000

10 m

(d) x500(d) x500

50 m

B/(B+W)=14.06%(e) x5000(e) x5000

5 m

B/(B+W)=17.51%(f) x2500(f) x2500

10 m

B/(B+W)=24.07%(g) x1000(g) x1000

10 m

B/(B+W)=17.07%(h) x500(h) x500

50 m

Figure 9. SEM and binarized images of the Holocene clay at EL. –38.19 m.

of e is the smallest and that of Cc is the largest among them.In the Pleistocene clay, SD of normalized e of 0.056 is abouttwice of that in the Holocene clay, caused from the smaller evalues at around EL. –126.05 m. SD of normalized p'c is0.077, which is almost the same as that in the Holoceneclay. SD of normalized Cc of 0.117 is also the same level asthat in the Holocene clay. However, SD of normalized cv of 0.109 is 50% larger than that in the Holocene clay.

Examining the clay samples, divided into every 25 mm, retrieved from both Holocene and Pleistocene layers inOsaka Bay resulted that the clay layers are relativelyhomogeneous with SD values of the normalized parameters(e, p'c, Cc, and cv) in a range of 2.5—12.0%. SD of normalized e is only 2.5% in the Holocene clay and 5.6% inthe Pleistocene clay, but SD values of the consolidation

parameters (p'c, Cc, and cv) obtained from CRS tests aregreater than these. Since, in an ordinary ground survey fordesign, consolidation parameters are examined every oneor a few meter(s), it is difficult to evaluate the variation ofobtained parameters; thus, it cannot be determined if the tested specimen was representative of the examining layer.Accordingly, it is important to select representativespecimens in the layer considering the profiles of physicalproperties.

4.2 SEM observation

Observed SEM images and those binarized images for theHolocene clay at EL. –37.96 and –38.19 m and the Pleistocene clay at EL. –125.54 and –126.01 m are shownin Figures 8, 9, 10, and 11, respectively. In Figures 8 and 11,

Session 7E

(a) x5000(a) x5000

5 m

(b) x2500(b) x2500

10 m

(c) x1000(c) x1000

10 m

(d) x500(d) x500

50 m

B/(B+W)=15.08%(e) x5000(e) x5000

5 m

B/(B+W)=30.56%(f) x2500(f) x2500

10 m

B/(B+W)=31.88%(g) x1000(g) x1000

10 m

B/(B+W)=30.17%(h) x500(h) x500

50 m

Figure 10. SEM and binarized images of the Pleistocene clay at EL. –125.54 m.

(a) x5000(a) x5000

5 m

(b) x2500(b) x2500

10 m

(c) x1000(c) x1000

10 m

(d) x500(d) x500

50 m

B/(B+W)=39.74%(e) x5000(e) x5000

5 m

B/(B+W)=31.12%(f) x2500(f) x2500

10 m

B/(B+W)=27.01%(g) x1000(g) x1000

10 m

B/(B+W)=29.16%(h) x500(h) x500

50 m

Figure 11. SEM and binarized images of the Pleistocene clay at EL. –126.01 m.

the images in magnification of both 5000 and 2500 timesshow enlarged structure of soil fabric. On the other hand, inFigures 9 and 10, the images in magnification of both 5000and 2500 times show particular parts on pyrite (Figure 9)and diatom (Figure 10), not representing the soil fabric.

Profiles with depth (a—d) and frequencies (e—h) on blackpercentage in the binarized black and white images for theHolocene and the Pleistocene clays are shown in Figures12 and 13, respectively. Width shown by broken line showsa range of MV±SD. The width tends to decrease when themagnification decreases.

In the Holocene clay (Figure 12), SD decreases when themagnification decreases as 5000, 2500, and 1000 times,indicating that the impression of the SEM images could

change. However, SD values of both magnifications of 1000and 500 times are at the same level. This fact can bevisually impressive in Figure 8.

In the Pleistocene clay (Figure 13), SD tends to decreasewhen the magnification decreases. SD of the magnificationof 2500 times is about half of that of 5000 times, but SDvalues of 2500, 1000, and 500 times are at the same levelof around 6%. This fact can be visually impressive in Figure11.

In the Holocene clay, SD values of the magnification of 2500times and both 1000 and 500 times are 72% and 48%,respectively, of that of 5000 times. This indicates that everySEM image with a high magnification gives differentimpression, but that with a low magnification of either 1000

Session 7E

x500

-38.3

-38.2

-38.1

-38.0

-37.9

-37.8

-37.7

-37.6

-37.5

0 10 20 30 40 50

Black/(Black+White) (%)

Ele

vatio

nC

.D.L

. (m

)

(d)

x500

0-5

5-1

0

10-1

5

15-2

0

20-2

5

25-3

0

30-3

5

35-4

0

40-4

5

45-5

0


Fre

quency

23.5±4.3 %

(h)

x1000

-38.3

-38.2

-38.1

-38.0

-37.9

-37.8

-37.7

-37.6

-37.5

0 10 20 30 40 50


Ele

vatio

n C

.D.L

. (m

)

(c)x2500

-38.3

-38.2

-38.1

-38.0

-37.9

-37.8

-37.7

-37.6

-37.5

0 10 20 30 40 50


Ele

vatio

n C

.D.L

.(m

)

(b)x5000

-38.3

-38.2

-38.1

-38.0

-37.9

-37.8

-37.7

-37.6

-37.5

0 10 20 30 40 50


Ele

vatio

nE

L.

(m)

(a)

x1000

0-5

5-1

0

10-1

5

15-2

0

20-2

5

25-3

0

30-3

5

35-4

0

40-4

5

45-5

0


Fre

quency

23.2±4.3 %

(g)x2500

0-5

5-1

0

10-1

5

15-2

0

20-2

5

25-3

0

30-3

5

35-4

0

40-4

5

45-5

0


Fre

quency

21.7±6.5 %

(f)x5000

0-5

5-1

0

10-1

5

15-2

0

20-2

5

25-3

0

30-3

5

35-4

0

40-4

5

45-5

0


Fre

que

ncy

21.4±9.0 %

(e)

Figure 12. Profiles of black percentage with depth and frequencies of the binarized images in the Holocene clay.

x500

-126.3

-126.2

-126.1

-126.0

-125.9

-125.8

-125.7

-125.6

-125.5

0 10 20 30 40 50 60 70


Ele

vatio

n C

.D.L

. (m

)

(d)

x500

0-5

5-10

10-

15

15-

20

20-

25

25-

30

30-

35

35-

40

40-

45

45-

50

50-

55

55-

60

60-

65

65-

70


Fre

quen

cy 30.2±5.7 %

(h)

x1000

-126.3

-126.2

-126.1

-126.0

-125.9

-125.8

-125.7

-125.6

-125.5

0 10 20 30 40 50 60 70


Ele

vatio

n C

.D.L

. (m

)

(c)x2500

-126.3

-126.2

-126.1

-126.0

-125.9

-125.8

-125.7

-125.6

-125.5

0 10 20 30 40 50 60 70


Ele

vatio

nC

.D.L

. (

m)

(b)x5000

-126.3

-126.2

-126.1

-126.0

-125.9

-125.8

-125.7

-125.6

-125.5

0 10 20 30 40 50 60 70


Ele

vatio

n E

L. (

m)

(a)

x1000

0-5

5-10

10-

15

15-

20

20-

25

25-

30

30-

35

35-

40

40-

45

45-

50

50-

55

55-

60

60-

65

65-

70


Fre

quenc

y

28.4±6.3 %

(g)x2500

0-5

5-10

10-

15

15-

20

20-

25

25-

30

30-

35

35-

40

40-

45

45-

50

50-

55

55-

60

60-

65

65-

70


Fre

que

ncy

28.1±6.3 %

(f)x5000

0-5

5-10

10-

15

15-

20

20-

25

25-

30

30-

35

35-

40

40-

45

45-

50

50-

55

55-

60

60-

65

65-

70

Black/(White+Black) (%)

Fre

quenc

y 25.9±12.2 %

(e)

Figure 13. Profiles of black percentage with depth and frequencies of the binarized images in the Pleistocene clay..

or 500 times gives the similar impression. In the Pleistoceneclay, SD value of the magnification of 2500 times is 52% ofthat of 5000 times, and that of either 2500, 1000, or 500times is at the same level.

Figure 14 shows relationships (a) MV of black percentageand (b) CV of black percentage versus the magnification. InFigure 14a, MV slightly decreases when the magnificationincreases, but essentially constant. From CV in Figure 14b,

it can be said that the variation of the SEM images becomessignificant when observation magnification is greater than1000 and 2500 times for the Holocene and Pleistoceneclays, respectively. Therefore, a representative structure inthe microfabric could be observed in an area of 150×120 mwith the magnification of 1000 times for the Holocene clayand 50×40 m with the magnification of 2500 times for the Pleistocene clay. Since specimen size for laboratory tests is sufficiently large on microscopic heterogeneity, the variation

Session 7E

of physical and mechanical properties is mainly influencedby macro-scale heterogeneity or sample disturbance.

0

5

10

15

20

25

30

35

0 2000 4000 6000

Magnification

Ava

rage

of

Bla

ck/(

Bla

ck+

Wh

ite)

(%

)

Pleistocene

Holocene

(a)

0.0

0.1

0.2

0.3

0.4

0.5

0 2000 4000 6000

Magnification

Coe

ffici

ent o

f va

riatio

n (%

)

Pleistocene

Holocene

(b)

Figure 14. Relationships of (a) average of black percentageand (b) coefficient of variation versus the observationmagnification.

5. CONCLUSIONS

In this study, clay sample in 1 m length eac

h retrieved from either Holocene or Pleistocene layer inOsaka Bay was divided into every 25 mm in order to exam the variation of both physical and consolidationproperties. The followings are derived as conclusions: theclay layers are relatively homogeneous with SD values ofthe normalized parameters (e, p'c, Cc, and cv) in a range of 2.5—12.0%. SD of normalized e is only 2.5% in theHolocene clay and 5.6% in the Pleistocene clay, but SD values of consolidation parameters (p'c, Cc, and cv)obtained from CRS tests are greater than these. It can besaid that the variation of the SEM images becomessignificant when observation magnification is greater than1000 and 2500 times for the Holocene and Pleistoceneclays, respectively. Since specimen size for laboratorytests is sufficiently large on microscopic heterogeneity,the variation of physical and mechanical properties ismainly influenced by macro-scale heterogeneity orsample disturbance.

REFERENCES

Akai, K. 2000. Insidious settlement of super-reclaimedoffshore seabed, Coastal Geotechnical Engineering inPractice, Balkema, pp. 243-248.

Akai, K., Nakaseko, K., Matsui, T., Kamon, M, Sugano, K., Tanaka, Y., and Suwa, S. 1995. Geotechnical andgeological studies on seabed in Osaka Bay,Proceedings of the 11th European Conference on SoilMechanics and Foundation Engineering, Vol. 8, pp. 1-6.

Horie, H., Zen, K., Ishii, I., and Matsumoto, K. 1984.Engineering properties of marine clay in Osaka Bay,(Part-1) Boring and sampling, Technical Note of the Port and Harbour Research Institute, Ministry of Transport,Japan, No. 498, 5-45. (in Japanese)

Kanda, K., Suzuki, S., and Yamagata, N. 1991. Offshore soil investigation at the Kansai International Airport,Proceedings of the International Conference onGeotechnical Engineering for Coastal Development,Geo-Coast '91, Vol. 1, pp. 33-38.

Kang, M.-S., Watabe, Y., and Tsuchida, T. 2003. Effect of drying process on the evaluation of microstructure of clays using scanning electron microscope (SEM) andmercury intrusion porosimetry (MIP), Proceedings of the13th International Offshore and Polar EngineeringConference, pp. 385-392.

Tanaka, H. and Locat, J. 1999. A microstructuralinvestigation of Osaka Bay clay: the impact ofmicrofossils on its mechanical behaviour, CanadianGeotechnical Journal, Vol. 36, pp. 493-508.

Tanaka, H., Ritoh, F., and Omukai, N. 2003a.Geotechnical properties of clay deposits of the OsakaBasin, Characterisation and Engineering Properties ofNatural Soils, Swets & Zeitlinger, pp. 455-474.

Tanaka, H., Tanaka, M., Suzuki, S., and Sakagami, T.2003b. Development of a new cone penetrometer andits application to great depths of Pleistocene clays, Soilsand Foundations, Vol. 43, No. 6, pp. 51-61.

Watabe, Y., Tsuchida, T., and Adachi, K. 2002.Undrained shear strength of Pleistocene clay in OsakaBay, Journal of Geotechnical and GeoenvironmentalEngineering, ASCE, Vol. 128, pp. 216-226.

Session 7E

on the homogeneity of undisturbed clay samples...

Documents