Engineering Geology 122 (2011) 146159

Contents lists available at ScienceDirect

Engineering Geology
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e n g g e o

Changes in late PleistoceneHolocene sedimentary facies of the Mekong River Delta

and the inuence of sedimentary environment on geotechnical
engineering properties
Minh Hoang Truong a,, Van Lap Nguyen b, Thi Kim Oanh Ta b, Jiro Takemura c
a
b
c

Department of Engineering Geology, Ho Chi Minh University of Natural Science, 227 Nguyen Van Cu Str., Dist. 5, Ho Chi Minh City, Viet Nam
Vietnam Academy of Science and Technology, HCMC Institute of Resources Geography 1Mac Dinh Chi Str., Dist. 1, Ho Chi Minh City, Viet Nam
Department of Civil Engineering, Tokyo Institute of Technology, 2-12-1 O-okayama, Meguro-ku, Tokyo 152-8552, Japan

a r t i c l e

i n f o

Article history:
Received 30 March 2010
Received in revised form 28 May 2011
Accepted 31 May 2011
Available online 31 July 2011
Keywords:
Holocene
Mekong River Delta
Facies
Geotechnical properties
Sedimentary environment

a b s t r a c t
The aim of the study was to characterize a variety of sedimentary facies from the Mekong River delta (MRD)
and to determine their geotechnical properties. The paper furthermore discusses the inuences of
sedimentary environment and conditions on geotechnical properties of the sedimentary facies. A core at
the Vinhlong province, MRD, sufciently presented the sedimentary facies of the area. Eight successive facies
were identied based upon sedimentary properties. Characteristics of the unit showed development of
sedimentary facies. Each sedimentary facies was formed under a different environment and revealed typical
geotechnical properties. Estuarine channel, estuarine marine, delta front-mouth bar, and sub- to inter-tidal
at facies were formed under strong hydrodynamic conditions. The sediments are from clay to coarse sand
and even pebbles, and the sedimentary structures are plentiful, such as intercalated clay, silt beddings and ne
to coarse sand beddings with very different thicknesses. Strengths of these sedimentary facies vary
signicantly, which can be observed by cone penetration test (CPTU) results, but not by standard penetration
test (SPT). The normalized values from CPTU show sawtooth graphs with large variations, especially delta
front-mouth bar and sub- to inter-tidal at facies at shallower depth. The soil-behavior-types determined
from the normalized values are plentiful from cohesionless to cohesive soils. As an example, an estuarine
channel facies experiencing a large consolidation pressure, in relatively long-time (over 9920 yr BP) has
geotechnical properties with heavily overconsolidated conditions and high strength. On the other hand,
marsh, open bay, pro-delta and March/ood plain facies, which formed under relatively low hydrodynamic
conditions, have simple and homogenous sedimentary properties. The sediments are commonly silt and clay,
and sedimentary structures are very faint interbedded clay and silt laminae, and rarely very ne sand laminae.
Strengths of these sedimentary facies increase linearly with depth, the normalized values are all rather
constant with depth, and their soil behavior-type is only normally consolidated clays from CPTU results. It
could be said that the particular sedimentary conditions in the late PleistoceneHolocene deposits of MRD can
be reasonably estimated by the CPTU test.
2011 Elsevier B.V. All rights reserved.

Contents
1.
2.

3.

Introduction . . . . . . . . . . . . . . . . . . . . . .
Investigation program . . . . . . . . . . . . . . . . .
2.1.
Investigation site. . . . . . . . . . . . . . . . .
2.2.
Boring and sampling, and geological investigations.
2.3.
In situ tests . . . . . . . . . . . . . . . . . . .
2.4.
Geotechnical laboratory tests . . . . . . . . . . .
Results . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.
Lithostratigraphy and inferred depositional facies .
3.2.
Results of geotechnical engineering tests . . . . .

Corresponding author. Tel.: + 84 8 3825 8156; fax: + 84 8 3825 8156.

E-mail address: tmhoang@hcmuns.edu.vn (M.H. Truong).
0013-7952/$ see front matter 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.enggeo.2011.05.012

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

147
148
148
148
149
149
150
150
152

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

3.2.1.
In-situ tests . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.2.
Laboratory tests. . . . . . . . . . . . . . . . . . . . . . . . . .
4.
Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.
Sedimentary facies changes and late PleistoceneHolocene development of the
4.2.
Relationship between the sedimentary facies and geotechnical properties. . .
4.2.1.
Transgressive incised-valley ll sediments . . . . . . . . . . . . .
4.2.2.
Holocene delta sediments . . . . . . . . . . . . . . . . . . . . .
5.
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction
The Mekong River delta (MRD), the largest delta in Vietnam, is
located in southern Vietnam between 8 30 and 11 00N and 104 30
and 106 50E (Figure 1). A great at plain of 62,520 km2 has been
created by the progradation of the delta during high stand and
regression over the last 4550 yr BP (Nguyen et al., 2000). During this
time period, the landforms, sedimentary structures, thickness and
material composition of sedimentary facies in the late Pleistocene
Holocene deltaic sediments have undergone complex changes (Ta et al.,
2002a, 2002b; Umitsu et al., 2003).
Several studies have been conducted about specic geotechnical
engineering problems in the MRD. Dornbusch et al. (1969) studied the
distribution of construction materials according to grain size and
Atterberg limit from the ground surface to a depth of 10 m.
Takemura et al. (2007) studied the characterization of alluvial
deposits in the region.
The structures of post-glacial natural soils depend on both the
depositional conditions and post-depositional processes. Research has
also been carried out on normally consolidated natural argillaceous

. . .
. . .
. . .
delta
. . .
. . .
. . .
. . .
. . .
. . .

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

147

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

152
153
154
154
156
156
157
158
159
159

sediments deposited at many places in the world during the postglacial period. Studying the compression curves of various natural
clays, Burland (1990) proposed a method of describing the level of
structure of natural clay using the sedimentation compression line
(SCL) and the intrinsic compression line (ICL). The micro- and macrostructure of natural soils causes them to differ from reconstituted soils
in a number of important ways. In this respect, following Mitchell
(1976), the term structure implies a combination of fabric
(arrangement of particles) and interparticle bonding. Studying
sedimentation compression curves, Burland (1990) found a difference
between British post-glacial clays present at Shellhaven (Skempton
and Henkel, 1953) and Gosport (Skempton, 1970) and showed that
their structures depend both on the depositional conditions and on
the post-depositional processes. Wu (1958) also found that the
sedimentary materials and post-depositional processes affect the
geotechnical properties of sediments. The late PleistoceneHolocene
sediments that have accumulated during the post-glacial period
consist of several sedimentary facies, each of which was formed in a
different sedimentary environment and has typical sedimentary
structures and materials. Given the importance of sedimentary

Fig. 1. Environmental sedimentary map of the Mekong River Delta (Nguyen et al. 2000), location of the VLM1 borehole (lled circle) and the VL1 (Ta et al., 2002b) and BT2 boreholes
(Ta et al., 2001, 2002a) (open circle and square, respectively): 1. Channel bar, 2. Point bar, 3. Bank, 4. Flood basin, 5. Back swamp, 6. Swamp, 7. Flood plain, 8. Abandoned channel, 9.
Alluvial apron, 10. Coastal plain, 11. Marsh, 12. Salt marsh, 13. Mangrove marsh, 14. Relict beach ridge or sand dune, 15. Sand spit, 16. Tidal at, 17. Undivided deposits of late
Pleistocene age, 18. Weathered land, 19. Basement rock, 20. Line of prole, 21. VLM1 borehole, 22. VL1 borehole, 23. BT2 borehole.

148

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

Fig. 2. Location of Vinhlong investigation site on the map of the MRD and Vinhlong province simplied after Ma et al. (2003) and plan layout of VLM1 borehole, CPTU1, CPTU2, and
SPT tests at the site.

environments on the properties of individual facies, each sedimentary

facies would perhaps be expected to possess unique geotechnical
properties. However, with respect to the MRD, no research has been
conducted to conrm this matter. Therefore, a study of the changes in
the late Pleistocene-Holocene sedimentary facies and the sedimentary
facies' geotechnical properties at the site of Vinhlong in the MRD was
conducted.
2. Investigation program
2.1. Investigation site
The investigation was conducted in Vinhlong province, center of
the MRD (Figures 1 and 2). The borehole (designated VLM1) was
located at latitude 10 14 2 N, longitude 105 59 8 E at an altitude of

(a)

(b)

z = + 2 m above the present mean sea level and reached to z =

46.05 m. The ground water level was 0.65 m below the ground
surface.
2.2. Boring and sampling, and geological investigations
A hydraulic type thin-walled tube sampler with a xed piston was
used to obtain soil samples. A stainless steel sampling tube with 2 mm
thickness, 85 mm inside diameter and 710 mm length was pushed
into the ground by water pressure. The core was divided into several
samples, each 100 mm in height, which were individually inspected
and photographed. From the appearance of the soil surface, detailed
soil proles were determined. Carbon isotope ( 14C) dating of the
organic material in some soil samples was performed by the Beta
Analytic Radiocarbon Dating Lab, Japan. Clay mineral analyses based

(c)

Fig. 3. Piezo-cone enetrometer (CPTU), (a) details of CPTC, (b) cone tip, (c) penetrating rods.

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

on X-ray diffraction methods were carried out using the D8 ADANCE

automatic system at the Analysis Lab Center, Vietnam Petroleum
Institute.
2.3. In situ tests
Two Cone Penetration Tests (CPTU) were carried out at the site
using a piezo-cone penetrometer with friction sleeve with the
specication shown in Fig. 3. The cone penetrometer was pushed
into ground using hydraulic type penetration machine. One CPTU,
designated CPTU1, was conducted to a depth of z = 47.8 m and the
other, designated CPTU2, to z = 21.7 m. A Standard Penetration Test
(SPT) was also carried out every 2 m to a depth of z = 60 m. The
plan layout of the borehole, VLM1, and the in situ tests are shown in
Fig. 2.

weight, sat, specic gravity, and Gs (Head, 1985a), were obtained

every 0.1 m. The liquidity index, LI, was estimated from wn, wp, and
wL. On the other hand, vertical effective stress v0 was estimated
from sat. Unconned compressive tests were conducted both for
undisturbed soils and for remolded soil to obtain sensitivity, St. For
evaluating one-dimensional consolidated properties, incremental
loading oedometer tests (IL) (Head, 1985b), were mainly conducted
at the Engineering Geology Laboratory of Ho Chi Minh City University
of Natural Science, Vietnam (HCMCUS). Constant rate of strain
consolidation tests (CRS) (JGS, 2000) were also conducted on typical
soil samples taken from each facies at Geomechanics Lab of Tokyo
Institute of Technology. The yield stresses y IL and y CRS were
estimated from the IL and CRS results. Void indices Iv0 for in situ void
ratio e0 were estimated using Eq. (1) (Burland, 1990),

Iv0 =

2.4. Geotechnical laboratory tests

Various basic geotechnical properties, such as grain size distribution, natural water content, wn, plastic limit, wp, liquid limit, wL, unit

149

e0 e100
e
1000 e
100

where e *100 and e*1000 are the void ratios of the intrinsic compression
curve at v = 100 kPa and 1000 kPa, respectively. The intrinsic

Fig. 4. Geological column of the VLM1 core and its correlation with lithostratigraphic units.

150

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

compression curve can be obtained from samples reconstituted at a

water content of between wL and 1.5wL.
3. Results
3.1. Lithostratigraphy and inferred depositional facies
The sedimentary environments and stratigraphy of the VLM1 core
were reconstructed based on the characteristics of grain size, color,
sedimentary structure, clay minerals, fossils and carbon isotope ( 14C)
ages of the sediments. The sediments of the VLM1 core can be divided
into eight lithostratigraphic units. Then, eight depositional facies are
inferred based on the characteristics of the units (Figures 4 and 5) and

grain size fractions (Figure 6c). The characteristics of these units,

corresponding depositional facies are presented below in ascending
order.
(1) Unit 1 (z = 46.05 to 41.5 m)/estuarine channel facies
This unit has inhomogeneous sediments including darkish and
greenish-gray silty clay to clayey silt (Figure 6c) and discontinuous ne sand laminae of 2 mm thickness (Figure 5a). It also
contains scattered laterite (510 mm in diameter) and angular
quartz (maximum 15 mm in length) pebbles (Figure 5a).
Parallel and lenticular beddings are common. In comparison to
other facies, the smectite mineral content of this layer is the
highest, while the chlorite and illite contents are the lowest
(Table 1). These characteristics indicate that the sediments

Fig. 5. Selected photographs of sedimentary structures from the VLM1 core: a) (at depth 44.5 m) discontinuous ne sand laminae of 2 mm thick, angular quartz pebbles; clay mass,
b) (40.95 m) inter-bedded, brownish gray silty clay and clayey to sandy silt, faint bedding, c) (32.6 m) mollusca, d) (30.1 m) discontinuous parallel laminae, wavy and aser,
e) (27.35 m) parallel laminae, wavy bedding, f) (25.0 m) lenticular and aser bedding, g) (22.15 m) clayey silt and silty clay mud with discontinuous, very thin sandy bedding and
calcareous incipient, h) (21.21 m) discontinuous sandy laminae, i) (19.54 m) many parallel laminated very ne sand seams and calcareous nodules, j) (13.75 m) ne to coarse sand
layers with parallel clayey laminae, k) (7.70 m) parallel, discontinuous, lenticular and wavy bedding, l) (2.6 m) intercalated sand laminae and mud (tidal rhythmites), humus matter
and burrow.

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

151

Fig. 6. Summary of laboratory test results: (a) Geological column of the VLM1 core, (b) Description of the material, (c) Grain size distribution, (d) Saturated unit weight sat and
specic gravity Gs, (e) Natural water content wn, plastic limit wp, liquid limit wL, (f) Liquidity index LI, (g) Sensitivity, ratio of compression strength for undisturbed sample qu to
remold sample qru, (h) Yield stress from IL oedometer test y IL and CRS test y CRS, with vertical effective overburden stress v0.

environment, and at 41.5 m is dated at 9910 50 14C yr BP and

at 30.75 m is dated at 9090 40 14C yr BP (Table 2). These are
all features of a marsh facies.
(3) Unit 3 (z= 27.8 to 25 m)/estuarine marine facies
Unit 3 can be divided into two parts. The lower part of the unit
(27.8 to 26.4 m), consists of intercalated darkish gray silty
sand to medium sand and clayey silt to silty clay (Figure 5e),
coarsening in upward succession. The upper part (26.4 to
25 m), consists of greenish-gray silty sand and greenish-gray
clayey silt (Figures 5f and 6c). In general, parallel laminae, wavy
bedding (Figure 5e), lenticular bedding, and aser bedding
(Figure 5f) characterize the unit. Shell fragments and humus
material are scattered throughout the unit. This unit is relatively
inhomogeneous. The textural characteristics and the marine
bivalve fossils such as mollusca indicate that the sand and sandy
silt were deposited in an estuarine environment. In addition, they
indicate a transgressive lag deposit or estuarine channel deposit
affected by strong tidal currents.
(4) Unit 4 (z= 25 to 22.0 m)/open bay mud facies
This unit consists of darkish-gray clayey silt with sets of
interbedded gray clay (2555 mm thickness) and dark clay (2

were deposited under dynamic hydrological conditions, which

might be caused by the presence of an estuarine channel or a
tidal river.
(2) Unit 2 (z= 41.5 to 27.8 m)/salt marsh facies
Unit 2 consists of two parts. The lower part (41.5 to 37 m) is
characterized by inter-bedded, brownish gray silty clay and
clayey to sandy silt. Faint bedding exists in the brownish-gray silt
(Figures 5b and 6c). The upper part of Unit 2 (37 to 27.8 m)
contains discontinuous parallel laminae, wavy, aser (Figure 5d),
parallel and lenticular bedding, humus material, bioturbation,
calcareous concretions, and mollusca (Figure 5c). Mica akes are
scattered throughout the unit. The clay mineral contents of three
specimens in this facies are almost equal and the illite mineral
content is the highest in comparison among all the facies
(Table 1). They show high homogeneity levels. These relatively
homogeneous conditions with higher content of ne material
imply that these sediments were deposited under relatively quiet
hydrological conditions so that concretions of soft calcareous
material produced calcareous pebbles in the sandy and clayey
silts. Mollusca fossils indicate a marine-brackish water habitat.
The sediment accumulated in a muddy tidal at/salt marsh

Table 1
Results of clay mineral content analysis of argillaceous sediments of the facies from the VLM1 core.
Altitude
(m)

Unit

Sedimentary facies

Kaolinite
(%)

Chlorite (%)

Illite
(%)

Smectite
(%)

Mixture of illite and smectite

(%)

2.15
5.18
18.1
20.04
24.88
27.2
32.8
36.4
37.58
45.85

7
7
6
5
4
3
2
2
2
1

Sub- to inter-tidal at

17.4
17.1
21.8
15.9
17.7
16.6
15.6
15.0
15.1
16.6

14.4
13.4
18.6
12.3
15.2
13.4
13.6
15.3
14.5
10.7

54.6
57.0
56.7
58.8
54.2
62.6
63.8
61.1
60.7
44.7

11.7
10.7
1.4
11.6
11.4
6.0
5.3
6.9
8.4
26.4

1.8
1.8
1.4
1.3
1.6
1.4
1.7
1.8
1.4
1.7

Delta front
Pro-delta
Open-bay
Estuarine marine
Marsh

Estuarine channel

152

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

Table 2
List of 14C ages from the VLM1 core.
Altitude
(m)

Materials

Delta 13C
(permil)

Conventional
age (yr BP)

14

Organic

25.5

3810 40

18.5

Organic

26

4560 40

21.95

Organic

25.4

6430 40

24.88

Organic

26.1

6490 40

30.75

Organic

27

9090 40

41.5

Organic

25.6

9910 50

14

Estimated depositional
rate (mm/year)
6
1.8
48.8
23.3
13

BETA:

14

C dating in beta analytic.

(7) Unit 7 (z= 5.5 to 1.5 m)/sub- to inter-tidal at facies

Unit 7 consists of greenish- and darkish-gray clayey silt, sandy silt
and very ne to medium sand, showing parallel laminae,
discontinuous parallel laminae, and lenticular and wavy bedding.
The deposits from 2.4 to 3 m consist of intercalated sand and
mud which resemble tidal rhythmites (Figures 5l and 6c). Humus
matter, bioturbation and burrows (Figure 5l) are present. These
features characterize sub- to inter-tidal at facies.
(8) Unit 8 (z = 1.5 to +1 m)/ood plain/marsh facies
This unit contains greenish-darkish gray clayey silt mud and
silty clay mud (Figure 6c) with small yellowish-greenish spots
from 1.5 to +0.5 m and greenish- and darkish-gray medium
to ne sandy mud, rich in organic materials, from +0.5 to
+1.0 m (Figures 4 and 6c). These sediments show that this
facies was formed by ooding.
3.2. Results of geotechnical engineering tests

5 mm thickness), and greenish-gray mud (Figures 5g and 6c)

with discontinuous, very thin sandy bedding. Some sets of faintly
parallel laminae are also seen, which might be formed by
seasonal uctuations in suspended sediment load or variations of
the amount and kinds of supplied organic materials, or,
alternatively, by tidal inuences. Although the grain size
distribution shows a ning-upward succession, the unit is
relatively homogeneous with the highest mud content among
all the facies (Figure 6c). Humus matter, mica akes, shell
fragments, calcareous incipience are scattered throughout the
unit. It ended at 6490 40 14C yr BP at 24.88 m (Table 2). From
above observations, this unit is inferred as open bay facies.
(5) Unit 5 (z= 22.0 to 18.5 m)/pro-delta mud facies
Unit 5 can be divided into two parts. The sediments of lower
part (22 to 20 m), includes gray silty clay and very ne
sand. The sediments commonly present interbedded greenishgray clayey silt (2530 mm thickness), silty clay (25 mm
thickness) (Figure 6c) and discontinuous parallel laminated
mud (Figure 5h). They are dated at 6430 40 14C yr BP at
21.95 m and ended at 4560 40 14C yr BP at 18.5 m
(Table 2). The upper part, (20 to 18.5 m), contains many
parallel laminated very ne sand seams (Figure 5i). In general,
mica akes, calcareous nodules (Figure 5i) and calcareous
incipient are scattered throughout the unit. There are signicant
variations in the lithology and sedimentary structures of this
layer, which could be caused by the effects of higher sediment
supply and/or more active hydrodynamics than previous time
when depositional environment transit from prodelta to delta
front. However, this unit is considered relatively homogeneous
in comparison with the overlying unit (Unit 6).
(6) Unit 6 (z=18.5 to 5.5 m)/delta front-mouth bar facies
Unit 6 contains intercalated greenish-gray clay and silt, greenishgray sandy silt and ne to coarse sand, which is highly
inhomogeneous, with several sandy layers of various thicknesses.
Some of the sandy layers are relatively thick with parallel clayey
laminae (Figures 5j and 6c). In general, medium to coarse sands are
common soil types of the unit (Figure 4). Lenticular, aser, and
wavy bedding (Figure 5k), burrow, bioturbation, plant fragments
and mica akes are also scattered throughout the sediment. The
kaolinite and chlorite mineral contents of the argillaceous
sediments are the highest, and the smectite mineral content is
the lowest among all the facies (Table 1). These various
sedimentary structures may indicate strong hydrodynamic conditions mainly resulting from tidal currents and ooding causing
high depositional rates. It is inferred that the sedimentary process
forming this facies occurred in a river-mouth environment that
formed a silty sand mouth bar in the delta front. The measured 14C
age at 14 m is 381040 14C yr BP (Table 2).

3.2.1. In-situ tests

Measured quantities of the cone penetration tests (CPTU1 &
CPTU2) named cone resistance, qt, sleeve friction, fs, and pore water
pressure measured behind the cone tip, u2, are shown in Figs. 7 and
8 together with the columnar section of VLM1. All of these values
show an increasing trend with depth. The sudden increases in qt and fs
with a corresponding drop in u2 indicate the presence of a sand layer
or of intermediate soils with high permeability (i.e., cohesionless soil).
As the cone penetrates into cohesionless soil, the pore water pressure
immediately dissipates and reduces to the hydrostatic water pressure,
u0. The value of pore water pressure drops further, becoming lower
than the hydrostatic water pressure when the cone encounters a pure
sand soil. This observation helps us to identify the different types of
cohesionless soils. The alternating of cohesionless and cohesive soil
layers results in a saw-tooth shape in the overall plot of qt, fs and u2 as
a function of depth. The change in the amplitude of the saw-tooth
depends on the arrangement of the soil layers and the mixed levels of
sand, silt and clay. If the thickness of the cohesionless or cohesive soil
layer is large, the uctuation in amplitude will also be large, and the
soil layers are inhomogeneous. A typical soil prole can be estimated
by soil-behavior-type classication using the following normalized
values (Robertson, 1990, 1991):
Normalized cone resistance:
Qt =

qt vo

vo

Normalized friction ratio:

FR =

fs
100%
qt vo

Normalized pore pressure ratio:

Bq =

uuo
qt vo

where v0 and v0 are total and effective vertical stress.

The depth variations in the normalized values obtained from
CPTU1 and CPTU2 (CPTUs) are shown in Fig. 9. The soil-behavior types
estimated from the relationship between Qt and FR (Robertson, 1990,
1991) are shown in Figs. 7 and 8. The soil-behavior types estimated
from the relationship between Qt and Bq (Robertson, 1991) were
almost the same as those estimated from the QtFR relationship. In the
cohesive soil layers with homogeneous material properties, Qt, FR and
Bq are all rather constant with depth, in Unit 2 (41.5 to 27.8 m),
Unit 4 ( 25 to 22.0 m) and Unit 5 (22.0 to 18.5 m). The
common soil-behavior-type of these units is normally consolidated
claysclay to silty clay (Figure 7b).

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

153

Fig. 7. Results of the in situ tests at the Vinhlong site: (a) columnar section of the VLM1 core, (b) soil-behavior-type classication by Qt and FR obtained from CPTU1, (c)(e) cone
resistance, qt, pore water pressure, u2, and sleeve friction, fs, of CPTU1, (f) N value from the SPT.

The inhomogeneous characteristics of the soil layers were

conrmed by both CPTU1 and CPTU2 results (Figures 79). At the
deeper unit (Unit 1, 47.8 to 41.5 m), the normalized values show
large variations and the various soil behavior-types are estimated as
slightly overconsolidated to overconsolidated clays (clay to silty clay),
silt mixtures (clayey silt to silty clay), and sand mixtures (silty sand to
sandy silt) (Figure 7b). At 47.8 m, the qt value increased greatly
(Figure 7c), although the penetration machine used for the investigation could not penetrate through this layer since its capacity limit
was already reached. In Unit 3 (27.8 to 25 m), the normalized
values vary considerably and the estimated soil behavior-types are
normally consolidated to slightly overconsolidated clay to silty clay,
slightly sensitive silt mixtures of clayey silt to silty clay, and slightly
sensitive sand mixtures of silty sand to sandy silt. The variation of the
normalized values at shallower depths (less than 18.5 m) are the
most signicant with very high frequency variations with depth,
especially in Unit 6 ( 18.5 to 5.5 m) as shown in Figs. 7 and 8. As a
result, a variety of soil behavior-types are obtained with very thin
thickness, including sensitive ne-grained, normally consolidated to

slightly over-consolidated clay, slightly sensitive to sensitive and

normally consolidated to slightly overconsolidated silt mixtures,
slightly sensitive to sensitive sand mixtures, and slightly sensitive
sands. From the comparison of the results of CPTU1 and CPTU2 shown
in Figs. 79, some differences can be conrmed in the two tests,
implying that soil variations may exist horizontally even in the
distance of 10 m (Figure 2).
The SPT results (Figure 7f), form 1 to 38 m, the N values are
very small and do not increase, or increase with a little quantity in the
range of 38 to 45 m with depth; simultaneously, N values on
homogeneous and inhomogeneous soil layers are not clearly different.
Hence, the N values cannot represent the small variation of the
penetration resistance precisely, accurately, and continuously observed by the CPTU on the late PleistoceneHolocene sediments in the
MRD.
3.2.2. Laboratory tests
Basic soil properties, such as grain size distribution, natural water
content, wn, plastic limit, wp, liquid limit, wL, unit weight, sat, specic

154

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

Fig. 8. Results of the in situ tests at the Vinhlong site: (a) columnar section of the VLM1 core, (b) soil-behavior-type classication by Qt and FR obtained from CPTU2, (c)(e) cone
resistance, qt, pore water pressure, u2, and sleeve friction, fs, of CPTU2 compared with those of CPTU1.

gravity, Gs, are plotted against elevation in Fig. 6ce. In the gure, also
shown are liquidity index, IL, sensitivity, St, evaluated from the unconned
compression strength, and consolidation yield stresses y IL and y CRS
(Figure 6fh). Measured liquidity indices are all greater than one for the
soils at shallow depth (Units 6 to 8) with high sensitivity especially in Unit
6. The measured plasticity index, PI, and wL of the clayey soils for the
deeper units (Unit 1 to Unit 4) and the shallower units (Unit 5 to Unit 8)
are separately plotted in the plasticity chart (Figure 10). All soils are
classied inorganic soils with different plasticity.
Most values of y IL and y CRS lie on or above the line v0,
indicating that argillaceous soils are typically normally consolidated
or somewhat overconsolidated. Man (2003) reported that many soil
investigations done in MRD showed over-consolidation ratio less than
one from y IL a smaller than 0v and suggesting potential problems
in the testing procedure of consolidation tests commonly done in
Vietnam. However, it can be conrmed from the consolidation test
data that the consolidation of soils in the site has been completed, even
for the quite rapid depositional condition discussed in the subsequent
chapter. The vertical yield stresses from the CRS test are often more
than from the IL test. The CRS test compression curves e-log v is
sharper, and identies vertical yield stress,vy, more easily and
precisely (Figure 11).
4. Discussions
4.1. Sedimentary facies changes and late PleistoceneHolocene
development of the delta
The changes in late PleistoceneHolocene sedimentary facies at the
VLM1 core are inherently a continuous natural process that progresses
with different speeds in the delta. They also follow normal sedimentary
laws; i.e., in the marine transgression, the trend is ning upwards; and in
the marine regression, the trend is coarsening upwards. The sedimentary environments that created each sedimentary facies are different,
but sometimes reciprocally affected to some extent, especially at the

beginning and the ending of each sedimentary environment. As a result,

the sedimentary facies formed often display two somewhat different
parts but they still carry the sedimentary facies' main characteristics, like
the marsh and pro-delta facies. If sedimentary environment changed
rapidly, sedimentary facies formed will be entirely different from the
previous one, like estuarine marine facies (Unit 3) and delta frontmouth bar facies (Unit 6).
In the estuarine channel facies, the laterite pebbles are splintered and
perfectly round in shape. These characteristics are reliable indicators of
their origin. These laterite pebbles were formed in the late Pleistocene
undifferentiated sediments, which were affected by high hydrodynamic
activity. They then moved into the above estuarine channel sediments.
These facts indicate that the estuarine channel facies unconformably
overlay the late Pleistocene undifferentiated sediments. The fact that the
shapes of the plant fragments are unchanged in the estuarine sediments
conrms that the water level rose rapidly with plentiful sedimentary
supply in this specic type of environment. In Table 3, VLM1 core data
were compared with data from the BT2 core (Ta et al., 2001, 2002a) in the
incised valley and with data from the VL1 core (Ta et al., 2002b) in the
interuvial zone (Figure 1). The VLM1 and BT2 cores began with
estuarine channel sediment but in the VL1 core, the sediments began in
an open bay environment. Hence, it is likely that the VLM1 core was
located in an incised valley. According to a mechanism of incision
proposed by Woodroffe (1993) and Hanebuth et al. (2000), following the
last glacial stage, sea level rose rapidly and simultaneously created an
estuarine channel environment. This was an early stage of water level rise
at the VLM1 core site.
The river mouth shifted gradually landwards. The rapid transgression
of the channel led to speedy horizontal translation of the shoreline that
reached several tens of meters per year in Southeast Asia (Woodroffe,
1993). Because of different topographical conditions, the altitudes of the
appearance of estuary channel and marsh facies at three core positions
were quite different (Table 3). The shoreline in the MRD site probably
created a huge marsh environment, which was conrmed by the VLM1
core. It can be inferred that the sea level was over 30.75 m at about

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

Qt

(a) Normalized cone

resistance, Qt

FR (%)

(b) Normalized
friction ratio, FR (%)

155

Bq

(c) Pore pressure

ratio, Bq

Fig. 9. Normalized cone resistance, sleeve friction and pore water pressure obtained from CPTU1 and CPTU2.

9090 40 yr BP and the lowest part of the salt marsh facies in the VLM1
(41.5 m) yielded an age of 991050 yr BP. From these data, it can be
estimated that a salt marsh environment existed at the VLM1 site
between 9910 and 8800 yr BP, the salt marsh facies was formed with a
sedimentation rate of 13 mm/year (Table 2). Subsequently, sea level
continued to increase, and the transgression advanced landward,
resulting in the simultaneous formation of an estuarine marine
environment. The estuarine marine facies was created with a sedimentation rate of 1.2 mm/year. The rate of transgression for the VLM1 core
site was so rapid that it was converted into an open bay environment. An
open bay mud facies with ning upwards succession was created with a
rate of 48 mm/year. This rate is the rapidest among all the sedimentary
facies, which indicates that a maximum transgression, dated at 6430 yr
BP, occurred at the time of this open bay facies. These data coincide with
the maximum Holocene transgression at around 6000 yr BP in the MRD
(Nguyen et al., 2000). From the 14C ages of 11,340 115 yr BP at
60.87 m in the BT2 core site (Ta et al., 2001) and 6430 yr BP at 21.95 m
in the VLM1 core site, it might be said that a marine transgression
succession might have occurred during the 11,5006400 yr BP. The
incised-valley was lled by estuarine channel, salt marsh, estuarine
marine and open bay mud sediments, which can be conrmed from the
evidence that altitude differences of the bay mud facies become small
compared to the previous one (Table 3).
Subsequently, a marine regression caused by the combined effects of
sea level fall and high sediment supply occurred. The regression began

and a coarsening-upward succession in the pro-delta facies was

gradually formed with very ne sand to ne laminae with a depositional
rate of 1.84 mm/yr. The coarsening-upward succession can be seen
most clearly in the delta front with the coarse sand materials. The coarse
sand layers are so abundant that their thicknesses increase from 1 to
2 mm (at 14.8 m) to 1.3 m (at 12 m) (Figure 4). The lower part of
the delta front facies was formed with a rate of 6 mm/yr. The delta front
facies in the VLM1 core with ages of around 3810 yr BP at 14 m, and in
the VL1 core with ages of around 3630 yr BP at 10.0 m (Ta et al.,
2002b), are consistent with the unconrmed coastlines in the idealized
model of coastal evolution during 45003000 yr BP (Nguyen et al.,
2000). These data coincide with the evidence of coastal evolution in the
marine regression during the 4000-3000 yr BP relative to the positions
of the VLM1 and VL1 cores (Figure 1). Therefore, the data of the
regression stage in this VLM1 core supports the model by Nguyen et al.
(2000). The age of sediments of the regression stage is estimated as
64002400 14C yr BP.
After 2400 yr BP, the marine regression was completed and a
sandy silt facies was formed in a sub- to inter-tidal at sedimentary
environment. The last phase is the development of the ood plain/
marsh environments. After the marine regression was completed, the
topography of the MRD was not entirely a at terrain likely to form
marshes. The VLM1 core site is located in the ood basins (Nguyen et
al., 2000), 5.5 km from the Mekong Rivers (Figure 2). Long-continued
accumulation of silt and clay was made from over-bank ows and the

156

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

100

Low plasticity

(a)

90

High
plasticity

Medium
plasticity

Very high
plasticity

Unit 1

Extremely high
plasticity

Plasticity Index, PI, (%)

80

Unit 2

(CE)

70

Unit 3

60

(CV)
50

(CH)

Unit 4

(ME)

40
30

(MV)

(CI)

Tanan
(Takemura
et al., 2007

20

(MH)
10

(CL)

0
0

10

20

Cantho
(Takemura
et al., 2007)

(MI)

(ML)
30

40

50

60

70

Liquid limit
100

Low plasticity

(b)

90

High
plasticity

Medium
plasticity

80

90

Very high
plasticity

110

120

130

Extremely high
plasticity

80

Plasticity Index, PI, (%)

100

WL (%)

Unit 5

(CE)

Unit 6

70
60

Unit 7

(CV)

50

(CH)

(ME)

40
30

Unit 8

(MV)
Tanan
(Takemura
et al., 2007)

(CI)

20

(MH)
(CL)

10

(MI)

(ML)

0
0

10

20

30

40

50

60

70

Liquid limit

80

90

100

110

120

130

Cantho
(Takemura
et al., 2007)

WL (%)

Fig. 10. Plasticity chart of soils in the VLM1 core and Tanan and Cantho cores (Takemura et al., 2007): (a) deeper deposits from Unit 1 to Unit 4 and (b) shallower deposits from Unit 5
to Unit 8.

extremely high hydrodynamic activity during the ood seasons

resulted in the deposition of coarse materials, which ended up with
the top of the deposition at +1.0 m altitude.

4.2.1. Transgressive incised-valley ll sediments

obtained are similar to the characteristics of the sedimentary

properties. From the soil layer in the range of 49 to 47 m, it can
be seen that the lower soil layer, where the materials were
cohesionless soil, was formed rather early under a large effective
overburden pressure. The results also showed that the soil layer is
heavily overconsolidated state and that its strength increased highly.
From 47 to 45 m, the soil-behavior-type is overconsolidated clays
and the natural water contents, wn, are nearly equal to the plastic
limit, wp, (Figure 6e). These wn values are appropriate to rm state
with liquidity index LI around zero (Figure 6f). This is probably a
result of the sediments containing numerous plant fragments and
plentiful organic materials. Especially, in the range of 44 to
41.35 m, the wn is approximately 60% higher than the wp (Figure
6e) and the soil-behavior-type is normally consolidated clays (Figure
6h and 7b). From 49.6 to 60 m, N value suddenly increased to
over 50 (Figure 7f), implying the late Pleistocene undifferentiated
sediments with the age over 40,000 yr BP (Ta et al., 2002a, 2002b).

4.2.1.1. Estuarine channel/tidal river sandy silt facies (Unit 1). The
results of lab tests (Figure 6) and of CPT1 on the facies showed that the
soil-behavior-types vary and that their formation sequences are
similar to sedimentary structures (Figure 7a and b). All the properties

4.2.1.2. Salt marsh facies (Unit 2). The CPTU1 results indicate highly
homogeneous structure and material. The materials are primarily
medium plasticity clay CI, a little high plasticity clay CH, and rarely
medium plasticity silt MI (Figure 10). The clay mineral contents of the

4.2. Relationship between the sedimentary facies and geotechnical

properties
As discussed in 4.1, the VLM1 core site is located in an incised
valley and includes two sedimentary successions, i.e., (1) transgressive incised-valley ll sediments deposited between 11,500 and
6400 yr BP, which include estuarine channel, salt marsh, estuarine
marine and open bay mud facies; (2) regressive Holocene delta
sediments deposited between 6400 and 2400 yr BP, which include
pro-delta, delta front-mouth bar, sub- to inter-tidal at, and ood
plain/marsh facies.

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

2.0

(a)

z= -22.12m (IL)

1.8

z= -28.04m (IL)
z= -30.5m (IL)

1.6

157

4.2.1.4. Open bay mud facies (Unit 4). The CPTU1 results of this facies
show a main soil-behavior-type of normally consolidated clay to silty
clay (Figure 7b). The materials are only very high plasticity clay CV
(Figure 10). This shows high homogeneity levels and correlates with
the sedimentary properties.

z= -30.81m (IL)

1.4

z= -32.61m (IL)

1.2

z= -32.64m (IL)
z= -36.03m (IL)

1.0

z= -41.51m (IL)

0.8

z= -41.52m
reconstituted

0.6

z= -36.4m
(CRS)
z= -30.88m
(CRS)

0.4

z= -30.88m
reconstituted

0.2
1

10

100

1000

10000

(kPa)
z= -1.42m (IL)

2.0

(b)

z= -2.84m (IL)

1.8
z= -6.35m (IL)

1.6

z= -6.38m (IL)
z= -7.9m (IL)

1.4

z= -9.59m (IL)

1.2
z= -9.62m (IL)

1.0

z= -9.74m
(CRS)
z= -9.76m
(CRS)

0.8

z= -14.5m (IL)

0.6
z= -9.735m
reconstituted

0.4

z=-9.76m
reconstituted

0.2
1

10

100

1000

10000

z=-7.9m
reconstituted

(kPa)
Fig. 11. Compression curves, elog v relationship, obtained from IL loading oedometer
tests and CRS tests for Vinhlong cohesive soils: (a) deeper deposits from Unit 1 to Unit 4
and (b) shallower deposits from Unit 5 to Unit 8.

three specimens from this facies are almost equal and the illite
mineral content is highest in comparison with all the facies (Table 1).

4.2.1.3. Estuarine marine sand and sandy silt facies (Unit 3). The CPTU1
results of this facies from 27.8 to 26.4 m revealed that Qt
increased well beyond the usual increase with depth (Figure 9a).
Soil-behavior-types are clay to silty clay and silty sand to sandy silt,
and the formation sequences are similar to those of sedimentary
structures (Figure 7a). At the upper part of this facies, from 26.4 to
25 m, the Qt trend is smaller than that of the lower facies (Figure 9a).
Soil-behavior-types are most likely clay to silty clay and clayey silt to
silty clay (Figure 7b) and correlate with the sedimentary properties in
the upper part of the facies. The cohesive soils are only low plasticity
clay CL (Figure 10).

4.2.2. Holocene delta sediments

4.2.2.1. Pro-delta mud facies (Unit 5). The CPTUs results of this facies
were divided into two parts (Figs. 7 & 8). The lower part, from 22 to
20 m, revealed that Qt, FR, and Bq are rather constant with depth and
that the soil-behavior-type is only normally consolidated clay to silty
clay. However, the upper part shows small variations in these values,
with characteristics correlative with the sedimentary properties and
the soil-behavior-types of normally consolidated clay to silty clay with
intercalated silt mixtures (Figures 7b and 8b). The cohesive soils are
mostly medium plasticity clay CI (Figure 10) and this differentiates
the pro-delta mud facies from the open bay mud facies below, where
cohesive soils are very high plasticity clay CV.
4.2.2.2. Delta front-mouth bar sand, silty sand facies (Unit 6). In the
CPTUs results of this facies, Qt, FR, and Bq show saw-tooth graphs with
large variations; Qt, FR, and Bq are the largest among all the facies
(Figure 9). Soil-behavior-types classication is varied (Figures 7b and
8b). In general, soil-behavior-type mainly tends to be cohesionless
soils. However, cohesive soils are also plentiful (Figures 6b and 10).
These results are correlative with the sedimentary properties. As
pointed out in Section 3.2.1, although the distance between the CPTU1
and CPTU2 positions is only 10 m (Figure 2), the soil behavior type at
the same altitude in this facies appeared quite different. The marked
variation of soil proles is a typical characteristic of delta front-mouth
bar facies, and deposits near the main Mekong River.
As shown in Fig. 11, the shape of the compression curves, initial
void ratios and yield stresses change signicantly with depths even in
the same facies. The void indices, Iv0, are plotted against the logarithm
of in situ effective overburden stress in Fig. 12, together with data on
Cantho cohesive and Tanan cohesive soils (Takemura et al., 2007). The
Ivo values are close to or below the Sedimentary Compression Line
(SCL) proposed by Burland (1990) for the most of the plots. The
calcareous incipience in soil specimens like Unit 4 of the IL test made
them become lightly overconsolidated. The intercalated clay bedding
in the sand layers in Unit 6 are somewhat overconsolidated, though
their ages are less than 4560 yr BP. These matters show the level of
microstructure development during the post-depositional processes
depends on sedimentary structures and materials. It is possible to
measure the sedimentary conditions of natural cohesive soils in MRD
using Ivo values, as long as the test specimen is homogeneous, as
shown in Fig. 13a, c, and d. Judging from Fig. 12, it can be said that the
microstructure or fabric of Mekong delta soils is typical of that of
relatively young natural deposits. However, some of the observed Iv0
values are well above SCL for some samples in the delta front-mouth
bar facies; this is attributed to inhomogeneity of specimens with a
number of thin sand seams (Figure 13b). The sand in the thin layers
becomes mixed with clay when the reconstituted sample is prepared,
resulting in well-sorted material that may not represent intrinsic
properties of the sample. This should be noted in the application of Iv0,
as discussed by Takemura et al. (2007) for Cantho and Tanan cohesive
soils, similarly in the case of Atterberg limits. These cases often occur
in the delta front facies.
4.2.2.3. Sub- to inter-tidal at sandy silt facies (Unit 7). The results of
CPTUs of this facies showed that Qt, FR, and Bq have saw-tooth graphs
with large amplitudes. The soil-behavior-types classication varies;
especially, the clean sand to silty sand layers appear many times and
are rather thick in comparison with other facies (Figures 7b and 8b)
and their formatted sequence is approximate to the sedimentary

158

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

Table 3
Comparison of thickness, altitude of appearance of sedimentary facies, 14C age, and altitude difference between maximum and minimum appearances of the VLM1 core with BT2 (Ta
et al., 2001) and VL1 cores (Ta et al., 2002b).
Core

Estuary
channel

Marsh

Estuary

Bay

Pro-delta

Delta
front

Marsh/ood Facies
Sub- to
inter-tidal plain
at

VLM1

41.5
N4.5
9910 ( 41.5)
62.3
6.7
35.0a

27.8
13.7
9090 ( 30.8)
54.5
7.8
13,258 (60.9)
35.0a

25
2.8
6490 ( 24.9)
35.95
18.55
8021 ( 52.38)
35.0a

26.7

10.95

18.5
3.5
4560 ( 18.5)
17
3

14.5
10
4826 ( 18.6)
4

5.5
13
3810 ( 14)
8
9
5309 ( 15.4)
4.5
10
3622 ( 9,96)
3.5

1.5
4

2
6

0
4.5

27.8

22
3
6430 ( 22)
20
15.95
5578 ( 32.5)
24.5
11
6111 ( 33.7)
4.5

BT2

VL1

1
2.5

2
4

2
2
3425 (0.45)
1

Altitude of appearance (m)

Thickness of facies (m)
14
C age (year BP) (at altitude (m))
Altitude of appearance (m)
Thickness of facies (m)
14
C age (year BP) (at altitude (m))
Altitude of appearance (m)
Thickness of facies (m)
14
C age (year BP) (at altitude (m))
Altitude difference between maximum
and minimum appearances

At VL1 undifferentiated Pleistocene sediment exists directly below Bay facies.

sensitive clayey silt to silty clay, approximating silty clay and clayey
silt mud in its sedimentary properties (Figures 7b and 8b). The
cohesive soils are medium and high plasticity clays CI, CH (Figures 6b
and 10). At a depth of + 0.5 to +1 m, soil-behavior-types are gravelly
sand to sand and clean sand to silty sand, corresponding to the nemedium sand mud formed by the ood. The sequence of the
mechanical behavior observed by CPTUs is correlative with the
sedimentary properties. These data characterize the ood plain/marsh
facies.
5. Conclusions

Fig. 12. Relationship between void indices Iv0 and effective overburden stress vo for the
Vinhlong cohesive soils, with data of Cantho and Tanan cohesive soils (Takemura et al.,
2007).

structure with intercalated sandy and silty mud. The cohesive soils are
commonly medium plasticity clay CI, and a little silt MI (Figures 6b
and 10). These features characterize sub- to inter-tidal at facies.
4.2.2.4. Flood plain/marsh facies (Unit 8). The CPTUs results of this
facies showed that the main soil-behavior-type is slightly sensitive to

Changes in late PleistoceneHolocene sedimentary facies in the

VLM1 core, MRD, are sedimentary successions with trends of ning
upwards in the transgression and coarsening upwards in the
regression. The estuarine channel, estuarine marine, delta frontmouth bar facies, which were formed under strong hydrodynamic
conditions have various sedimentary structures and material; their
soil-behavior-types vary largely and their formed sequences are
similar to the characteristics of sedimentary structures with inhomogeneity. The estuarine channel facies experienced post-depositional
processes with large consolidation pressure and long time resulting in
geotechnical properties related to overconsolidated conditions, with
high strength. The delta front-mouth bar facies, was formed under
extremely strong hydrodynamic conditions, and as a result has
sedimentary properties with high inhomogeneity. Its' mechanical
behavior shows great variability, as shown by CPTU results. The
marsh, bay, and pro-delta facies were formed under low hydrodynamic conditions. They are highly homogeneous and their strengths

Fig. 13. Observed cross section of test specimens for oedometer and CRS tests.

M.H. Truong et al. / Engineering Geology 122 (2011) 146159

increase linearly with depth. In this case, may combine with the
plasticity chart in identifying exactly characteristics of sedimentary
facies because each sedimentary facies has typical components of clay
mineral species. The results indicate that there is a close relationship
between the sedimentary facies and geotechnical properties. Therefore, it can be concluded that the sedimentary environments
signicantly affect the formation of materials, macro- and microstructure, and orientation of properties each sedimentary facies.
However, the illite mineral content is high (over 44%) in all the facies.
The results of this study allow the identication of soil properties and
estimation of the general tendencies of the mechanical behavior of soils
of the MRD region in the context of geotechnical engineering. From
previous sedimentary proles, we can guess the mechanical behavior.
The particular sedimentary conditions in the late PleistoceneHolocene
deposits can be reasonably estimated by the CPTU.
Acknowledgements
The authors would like to thank the Japan Society for the Promotion
of Science for its encouragement and nance. This work was supported,
in part, by the NAFOSTED of Vietnam, the projects no. 105.01-2010.02
and 105.09-2010.05. The Civil Engineering Department-Tokyo Tech
(TIT), the Port and Airport Research Institute at Yokosuka, Japan, and the
Engineering Geology lab, HCMCUS, created favorable conditions in
which to conduct the lab tests and collect documents. Mr. Y. Yuasa, TIT,
collaborated in conducting the tests, Mr. C.C. Truong, TEDI South,
assisted in eld tests. We thank two reviewers who have kindly given
their time and expertise to comment on the paper. We also express our
gratitude to Editor's comments and encouragement.
References
Burland, J.B., 1990. On the compressibility and shear strength of natural clays.
Geotechnique 40 (3), 329378.
Dornbusch, W.K., May, J.R., Covey, W.P., 1969. Distribution of coarse-grained
construction materials and potential construction sites in the Mekong Delta,
South Vietnam. U.S. Army Engineer Waterways Experiment Station Corps of
Engineers, Technical report S-69-7, vol. II.
Hanebuth, T., Stattegger, K., Grootes, P.M., 2000. Rapid ooding of the Sunda Shelf: a
late- Glacial sea-level record. Science 288, 10331035.

159

Head, K.H., 1985a. Soil Classication and Compaction Tests, vol. 1. Pentech Press,
London.
Head, K.H., 1985b. Permeability, Shear Strength and Compressibility Tests, Vol. 2.
Pentech Press, London, pp. 581729.
Japan Geotechnical Society (JGS), 2000. Test method for one-dimentional consolidation
properties of soils using constant rate of strain loading. JGS 0412, Laboratory Soil
Testing Methods. JGS, pp. 389414.
Ma, C.C., Nguyen, H.D., Le, D.N., 2003. Map of mineral geology of the Mekong River
Delta. Geological map league published (in Vietnamese).
Man, B.T., 2003. Initial estimation consolidation characteristics of soft Mekong Delta
clay for engineering practice. Proc. Intn. Workshop of Hanoi Geoengineering 2003
and 2004 3749.
Mitchell, J.K., 1976. Fundamentals of Soil Behavior. University of California, Berkeley.
422 pp.
Nguyen, V.L., Ta, T.K.O., Tateishi, M., 2000. Late Holocene depositional environments
and coastal evolution of the Mekong River Delta, Southern Vietnam. Journal of
Asian Earth Sciences 18, 427439.
Robertson, P.K., 1990. Soil classication using the cone penetration test. Canadian
Geotechnical Journal 27, 151158.
Robertson, P.K., 1991. Soil classication using the cone penetration test: reply. Canadian
Geotechnical Journal 28, 176178.
Skempton, A.W., 1970. The consolidation of clays by gravitational compaction.
Quarterly Journal of the Geological Society 125, 373411.
Skempton, A.W., Henkel, D.J., 1953. The post-glacial clays of the Thames Estuary at
Tilbury and Shellhaven. Prof. 3 rd Int. Conf. Soil Mech. and Foundation Engineering,
vol. 1, pp. 302308.
Ta, T.K.O., Nguyen, V.L., Tateishi, M., Kobayahi, I., Saito, Y., 2001. Sedimentary facies,
diatom and foraminifer assemblages in a late Pleistocene-Holocene incised-valley
sequence from the Mekong River Delta, Bentre Province, Southern Vietnam: the
BT2 core. Journal of Asian Earth Sciences 20, 8394.
Ta, T.K.O., Nguyen, V.L., Tateishi, M., Kobayahi, I., Saito, Y., Nakamura, T., 2002a.
Sediment facies and Late Holocene progradation of the Mekong River Delta in
Bentre Province, southern Vietnam: an example of evolution a tide-dominated to a
tide- and wave-dominated delta. Sedimentary Geology 152, 313325.
Ta, T.K.O., Nguyen, V.L., Tateishi, M., Kobayahi, I., Tanabe, S., Saito, Y., 2002b. Holocene
delta evolution and sediment discharge of the Mekong River, Southern Vietnam.
Quaternary Science Reviews 21, 18071819.
Takemura, J., Watabe, Y., Tanaka, M., 2007. Characterization of alluvial deposits in
Mekong Delta. Proc. Int. Workshop Characterization and Engineering Properties of
Natural Soils, vol. 3, pp. 18051829.
Umitsu, M., Nguyen, V.L., Ta, T.K.O., 2003. Late Holocene landform evolution of the
Mekong River Delta, Vietnam. GI2S Coast, Research Publication, 4 (Abstract,
Otranto/Taranto Puglia (Italy)).
Woodroffe, C.D., 1993. Late Quaternary evolution of coastal and lowland riverine plains
of southeast Asia and northern Australia: an overview. Sedimentary Geology 83,
163175.
Wu, T.H., 1958. Geotechnical properties of glacial lake clays. Proc. Am. Soc. Civ. Engrs,
84, p. SM3 (paper 1732).