Japanese Geotechnical Society Special Publication The 15th Asian Regional Conference on

Soil Mechanics and Geotechnical Engineering

Land reclamation & soil improvement works for a coal-fired power plant in Malaysia

Yoon C. Lam i), Dennis Ganendra ii) and Krishna Prasad iii)

i) Technical Director, Minconsult Sdn Bhd, Lot 6, Jalan 51A/223, 46100 Petaling Jaya, Selangor, Malaysia.
ii) Director, Minconsult Sdn Bhd, Lot 6, Jalan 51A/223, 46100 Petaling Jaya, Selangor, Malaysia.
iii) Manager, Minconsult Sdn Bhd, Lot 6, Jalan 51A/223, 46100 Petaling Jaya, Selangor, Malaysia.

ABSTRACT

The project site is underlain by superficial deposits of recent to sub-recent age alluvium with the upper soil
stratum comprising very soft to soft marine clays (Halocene clay) of up to about 25m thick. The reclaimed area on
swampy coastal land was pre-loaded with hydraulic sand fill and aided with pre-fabricated vertical drains (PVD) to
accelerate the consolidation settlement. This paper presents the design considerations including the important factors
for realistic prediction of consolidation settlement and fill embankment slope stability. Emphasis is placed on the
monitoring data and their comparison with the design predictions.

Keywords: reclamation, improvement, consolidation, settlement, back-analysis, instrumentation

1 INTRODUCTION
Three coal-fired generating units with an aggregate
net capacity of 2100 MW were proposed on a fast-track
schedule. The project site is located on low-lying
swampy coastal land near the southern tip of west coast
MALAYSIA
Peninsular Malaysia as shown in Figure 1 & 2.
The project site is underlain by superficial deposits
of recent to sub-recent age alluvium. The upper soil
stratum comprises very soft to soft marine clays
(Halocene clay) of about 25m thick. A comprehensive
programme of ground investigation and laboratory tests
were carried out to obtain soil parameters for detailed
engineering design of the proposed reclamation works. INDONESIA
The adopted design is based on well-established Project Site
empirical relationships for soft ground engineering and
advanced finite element modeling. The site was
extensively instrumented and monitored using
settlement plates, surface markers, inclinometers, Fig. 1. Key Plan of Project Site
piezometers and vane shear tests for stability analysis.
The Observation Method was implemented for the
reclamation construction. Monitoring data were PORT OF
TANJUNG
continuously reviewed and compared with the initial
PEL EPAS
theoretical predictions to confirm the validity of the
adopted parameters and assumptions used in the design
calculations and advanced numerical analyses. PROJECT
SITE
This paper presents the design considerations
including the important factors for realistic prediction
of consolidation settlement and fill slope stability.
Emphasis is placed on the instrument monitoring data
SINGAPORE
and their comparisons with the design predictions.
This paper will serve as an informative case history for
land reclamation construction over soft marine clay. Fig. 2. Location of Project Site beside a major shipping channel.

http://doi.org/10.3208/jgssp.SEA-12 1767
1.1 Technical Requirements
The scope of works under the contract requires EPC
contractor to carry out detailed engineering design and
implementation of soil improvement works to attain the
specified key performance objectives for settlement and
bearing capacity for respective areas before the final
Acceptance dates. Tight construction schedule and key
geotechnical performance objectives were imposed.
This include the achievement of target settlements
equivalent to at least 90% of primary consolidation
settlement at a load attributable to the finished platform
level plus 14m high coal stacks to be stocked piled at
the proposed storage yard. The contract required that
the specified performance criteria including differential
settlement and bearing capacity criteria should be
achieved within the agreed final acceptance dates.
2 CHARACTERIZATION OF PROJECT SITE
2.1 The Site
The reclamation site is located at Tg. Bin, Johor. It
is adjacent to a major navigation channel for Port of
Tanjung Pelepas (PTP) and overlooking Singapore to
the south. The coastal area is characterized by
mangrove swamp and mud flats. The site is at an
average ground level of +0.75m above MLSD. The
total area of reclamation is approximately 70 hectares
and for functional and contractual purposes, it was
sub-divided into four areas, namely the Power island,
coal storage yard, switch-yard and construction
lay-down areas. An indicative footprint of the land
reclamation works is shown in Figure 3.
SWITCH YARD
LAYDOWN
POWER ISLAND
COAL STORAGE
YARD
Fig. 3. Footprint of reclamation
2.2 Geology
The site is underlain by superficial deposits of recent
to sub-recent age alluvium (Geological Survey of
Malaysia, 1973). This Holocene deposit is
predominantly unconsolidated sand and clay of both
fluviatile and shallow-marine origin. This is further
underlain by older alluvium of Pleistocene age
comprising semi-consolidated gravel, sand, clay and
boulder beds. The bedrock is comprised of sedimentary
and volcanic rock formation similar to the Jurong
1768
Formation in Singapore, which is sub-divided into two
members, viz (i) Bukit Resam clastic member which
was deposited in the Upper Triassic to Middle Jurassic
age and (ii) Gunung Pulai volcanic member which was
deposited in Lower Triassic or older age.
The Bukit Resam clastic member comprises shale,
siltstone, sandstone and conglomerates. It is believed
that most rocks in this sequence had undergone some
low-grade metamorphism which had converted the shale
and sandstone to schist and meta-sandstone. The
Gunung Pulai volcanic member on the other hand is
comprised of mainly rhyodacite tuff with minor
inter-beds of other acids to intermediate tuffs and lavas.
The depth to bedrock varies from 23m to greater
than 38m below the prevailing ground levels. Overlying
the bedrocks are weathered soil zones and overburden
layers. The zone of weathered soil material is expected
to vary from 12m to 28m below the ground level.
Generally, the thickness of medium stiff to very stiff soil
material varies between 2m to 10m and is further
underlain by hard to very hard soil stratum. It can be
noted that the superficial deposits of marine clay with
occasional inter-bedded sand lenses extends throughout
the project site, up to a depth of about 28m below the
existing ground level.
Considering the low-lying coastal site, the
prevailing ground water table is generally high; varying
from above ground surface to approximately 1m to 3m
beneath the existing ground surface.
2.3 Subsoil Conditions
The subsoil condition throughout the site is generally
quite consistent, with two distinct subsoil strata. The
upper soil stratum is the alluvial deposits consisting
predominantly very soft to soft marine clays (with
occasional inter-beds of sand lenses) approximately 22m
to 25m thick at the coal yard. SPT N-value within the
marine clay is generally 0 blows/30cm at the top upper
stratum of up to 20m. In-situ penetration vane shear
tests carried out within the soft marine clay layer at the
coal storage yard indicates undrained in-situ and
remoulded shear strength, varying increasingly with
depth from 4 to 53.kPa and 2 to 20.kPa, respectively.
The plot of undrained shear strength versus depth is
presented in Fig. 4. The proposed trend-line for
undrained shear strength versus depth adopted for
design is also indicated in the said graph (Cu/v=0.3).
The sensitivity of the clay ranges from 2 to 7, with an
average value of approximately 3.
Cu of the marine clays as determined by
unconsolidated-undrained.(UU) triaxial tests with
confining pressures near or slightly above the effective
overburden pressure, ranges between 6 to 42.kPa. These
values are observed to be comparatively lower than
those obtained from the in-situ Geonor vane shear tests.
It is postulated that some disturbance of the soil samples
may have attributed to lower values of Cu as obtained
from UU tests.
 The marine clay is classified as having high to LL, PL, MC (%)
extremely high plasticity (CV-CE) as per the British Soil 0 20 40 60 80 100 120
Classification System (BS5930). A plot of Atterberg 0

limits on the plasticity chart is given in Fig 5.

DEPTH BELOW GROUND LEVEL (m)

Although the general distinction between Clay and Silt 5
is often taken to be the A-line on the plasticity chart,
with Clays plotting above and Silts below; however, the
reliability of the A-line in this regard is poor especially 10
for borderline cases. The variations of subsoil
engineering parameters (e.g. moisture content, Atterberg
15
limits, compression index, coefficient of consolidation,
pre-consolidation pressure, etc.) with depth are shown
in Fig 6 and 7. Over-consolidation ratios derived from 20
1-D oedometer tests are shown in Fig. 8.
UNDRAINED SHEAR STRENGTH (kPa)
25
0 10 20 30 40 50
LL
0
PL
Vane Shear - Peak
MC
Vane Shear - Remoulded 30
DEPTH BELOW GROUND LEVEL(m)

UU Triaxial Test
5 Fig. 6. Atterberg Limits vs. Depth

Cc / ( 1+eo ) Cv (m2 / year) at Pc Pc (kPa)

10 0.0 0.2 0.4 0.6 0 2 4 6 0 50 100
0 0 0
DEPTH BELOW GROUND LEVEL (m)

15 5 5 5

Pc
10 10 10
20

Su /v' = 0.3 15 15 15

25
Fig. 4. Undrained Shear Strength vs. Depth 20 20 20
'Vo
Adopted
UPPER PLASTICITY RANGE (U) Trend Line
100
INTER- VERY HIGH EXTREMELY HIGH
LOW (L)
MEDIATE
HIGH (H)
(V) PLASTICITY (E)
25 25 25
90
M = SILT (below A-Line) Fig. 7. Compressibility of Marine Clay
80 C = CLAY (above A-Line)
(Ref:BS5930:1999, Fig.18)
PLASTICITY INDEX(%)

70
CE
CV OVER-CONSOLIDATION RATIO (OCR)
60
CH A-Line 0 1 2 3 4 5
50
0
CI ME
40
DEPTH BELOW GROUND LEVEL (m)

CL
30

MV
20 5

10
6 MI MH
ML
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 10
LIQUID LIMIT (%)

Fig. 5. Plot of Atterberg Limits on Plasticity Chart

15
Immediately underlying the marine clay is the
sedimentary residual soil of the Bukit Resam/Gunung
Pulai formation. The thickness of residual soils ranges 20
from 2m to 10m and generally comprise of firm to stiff 1-D Oedometer Test
clayey silt/silty clay with SPT-N values ranging from 5 Trendline of OCR
to greater than 50 blows/30cm. At deeper levels, highly to 25

moderately weathered form of the parent rocks (SPT-N Fig. 8. Over-consolidation Ratio of Marine Clay
>100 blows/30 cm) namely, siltstones, sandstones and
volcanic rocks are encountered.

1769
 The available CPTU data confirm the subsoil 3.1 Estimation of Total Settlement
stratification as generalised from the borehole results. The estimated total primary settlement of 5.25 m is
In particular, the soft marine clay strata can be easily computed for a surcharge load equivalent to 15.4 m
identified from the deep sounding profiles based on the thick sand fill, overlying the highly compressible clay
very low cone resistance, qc and values of friction ratio. stratum of about 24.5 m. The adopted parameters are
at the coal storage yard is plotted in Fig. 9. It can be shown in Table 1.
deduced that the thickness of the highly compressible
marine clay varies from 22m to 27m. This compressible Table 1. Parameters for estimation of Primary Settlement.
subsoil layer is expected to consolidate significantly Soft clay thickness 24.5 meters
under the fill embankment and surcharge load. The OCR Varying with depth as per Fig. 8
geotechnical aspects are addressed in the detailed design cc/1+eo 0.225 at top, varying as per Fig 7
for proposed soil improvement works at the Coal Storage cr/1+eo 0.2*[cc/1+eo]
yard. unsat for hydraulic sand 18.0 kN/m3
Qc (kPa) Friction Ratio (%) Pore Pressure (kPa) sat for hydraulic sand 19.0 kN/m3
0 1 2 3 0 2 4 6 8 0 400 800 unsat for dry filled sand 17.5 kN/m3
0 0 0
sat for dry filled sand 17.5 kN/m3
sat for soft clay 14.5 kN/m3
(m) (m)

5 5 5
Water table At ground level, RL+0.75m MLSD
LEVEL
Depth

10 10 10 For over-consolidated soils, the primary settlement

DEPTH BELOW GROUND

is computed as follows:
15 15 15
If vo + < c :
Sult = cr/1+eo H log ([vo + ]/c) (1)
20 20 20

25 25
30 30
Fig. 9. Typical Subsoil Profile from CPTU
3 PERFORMANCE
The most stringent requirements are imposed on the
works for the proposed 20 hectares coal storage yard.
This includes the need to achieve a target settlement
equivalent to at least 90% of primary consolidation
settlement at a load attributable to the finished platform
level (RL +4.0m MLSD) plus the 14m high coal stacks.
Considering a specific gravity of coal at 9.5 kN/m3, the
minimum equivalent sand fill thickness at full
surcharge load required for the works is 15.4 m. In
addition, it is specified that the differential settlements
along the center-line of the 500 m long corridor of
stacker reclaimers shall not exceed 125 mm over 50 m
distance at 12 months after hand-over the agreed date.
For economic consideration, there is also a challenge to
limit the total volume of hydraulic sandfill to be
imported. This involved sub-dividing the coal-yard into
3 zones to allow the roll-over of surcharge fill while
having to meet the tight construction schedule.
The specified method for confirming compliance is
the Asoaka construction (Asaoka, 1978) using data
from 20 settlement plates installed at approximately a
hundred metres in grid spacing.
Uo If vo < c < vo + :
25 Sult = (cr/1+eo) H log (c/vo) +
(cc/1+eo) H log ([vo+]/c) (2)
where:
30
Sult = ultimate primary settlement
vo = initial vertical stress
c = pre-consolidation pressure
H = thickness of compressible layer
= increase in vertical stress on the clay layer.
4 SEQUENCE OF CONSTRUCTION
The stability of the fill embankment is of paramount
concern in view of the prevailing very soft ground. The
sequence of staged construction was determined based
on the need to ensure a minimum factor safety of 1.10
against slope instability and the short time period
available for construction, commencing from the
placement of first hydraulic sand fill layer to the
removal of excess surcharge to designed finished
platform level. PVDs were installed at triangular grid
spacing of 1.0 m c/c and to an average termination
depth of about 27m, from the top of the first lift.
The typical sequence of staged construction is
presented in Table 2. The rest period in between each
lift is necessary to allow a sufficient gain in shear
strength of the marine clay before the placement of the
next lift. The fill slope geometry, sequence of staged
construction and subsoil parameters are numerically
modeled using the finite element software, PLAXIS,
Professional Version 8.2.

1770
Table 2. Sequence of Staged Construction Table 3. Input Parameters for Plaxis
Lift No. Cummulative Construction Sequence Parameters Sand Fill Marine Clay Sandy Silt
Thickness (m) Material Model Mohr Coulomb Soft Soil Mohr Coulomb
1 3.0 Placement of hydraulic sand fill Material Type Drained Undrained Undrained
Installation of PVDs unsat (kN/m3) 18 14 17
Rest Period sat (kN/m3) 19 14.5 18
2 5.5 Placement of hydraulic sand fill kx (m/day) 8.64 1.0E-5 8.64E-3
Rest Period ky (m/day) 8.64 2.2E-5 8.64E-3
3 8.0 Placement of hydraulic sand fill eint 1.0 2.1 1.0
Rest Period Eref (kPa) 7,500 - 15,000
4 10.5 Placement of dry sand fill 0.3 - 0.3
Rest Period * + - 0.14 -
5 13.0 Placement of dry sand fill * + - 0.04 -
Rest Period cref (kPa) 1.0 2.0 10.0
6 15.4 Placement of dry sand fill 30 23 30
Rest Period ur - 0.15 -
# RL+4.0 Removal of excess surcharge fill +
In Plaxis, * = (cc/1+eo)2.3 and * = (cr/1+eo) (2/2.3).
Note: # Removal of surcharge load to the finished platform level
shall depend on settlement monitoring results and degree of Time (Days)
consolidation achieved. 0 100 200 300 400 500 600
0.0
Ultimate Settlement Prediction
Removal of surcharge
5 NUMERICAL MODELLING -1.0

The adopted mesh for numerical analyses using

Settlement (m)

-2.0
Plaxis is shown in Fig 10. The analyses were carried
out using a plane strain model with 6-node elements for -3.0
Settlement at Removal
a soft clay thickness of 24.5m. The Plaxis function, of Surcharge = 4.70 m
-4.0
updated mesh and pore water pressure, is switched on At Hand-over, Settlement = 4.51 m

to take into account the large deformation of the mesh. -5.0

The drain function in Plaxis is utilised to model the Ultimate Settlement = 5.17m

PVD. On the vertical face of PVD, the excess pore -6.0

water pressure (epwp) is automatically set to zero. The
left and right boundaries of the mesh were specified as
closed consolidation boundary. A roller displacement
boundary condition was specified at the left and right
boundaries of the mesh. At the bottom of mesh,
displacement boundary condition is fixed in both
horizontal and vertical directions. Ground water level
is specified at the top of marine clay layer.
The input parameters adopted for the sand fill, soft
marine clay and sandy silt stratum are summarised in
Table 3. The predicted total primary settlement is
5.17m and the results are presented in Fig 11. It is in
good agreement with the estimated 5.24m as obtained
hand calculations with equations (1) & (2). The
estimated settlement upon removal of surcharge at Day
281 is 4.70 m, which is equivalent to a degree of
consolidation of about 91%, a required in the Contract.
Fig. 11. Predicted Settlement at top of Clay Layer
6 STABILITY OF FILL EMBANKMENT
Prior to the PLAXIS analysis, the stability of
proposed embankment at each stage of filling were
analysed using SLOPE/W. The initial undrained shear
strength, cu of the soft clay (before placement of sand
fill) is taken as per Fig. 4. Some gain in shear strength
of the soft marine clay is expected due to the
consolidation process as the thickness of sand fill
increases. A conservative estimation for the gain in
shear strength is made at each relevant section of
subsoil underneath the fill embankment in relation to
the installed PVD and geometry of the slopes as
determined using the following equation:
cu =cu/v *U (3)
where
cu/v = 0.3 (see Fig.4)
U = predicted degree of consolidation at each Lift
= surcharge load (kPa), taking into account the
buoyant unit weight of fill that settles below
the groundwater table.

The rest period in between each sand fill lift is adjusted

to ensure a minimum safety factor of 1.10 is obtained at
all times. Morgenstern and Price limit equilibrium
Fig. 10. Mesh of numerical modelling method is used in the stability analyses computation.

1771
Prior to the placement of each subsequent lift of sand
fill, a review of geotechnical instrumentation results
and in-situ field tests were carried out to verify the gain
in shear strength of the underlying soft clay. This is to
ensure that the fill slopes are stable at all times,
especially immediately after placement of a new lift.
7 SETTLEMENT-TIME BEHAVIOUR
For monitoring purposes, the site is divided into
approximately 100m x 100m square panels with a
settlement plate installed at its center. The settlement
time behaviour of the ground were closely monitored
throughout the construction period and a typical
settlement-time plot is presented in Fig. 12. As per
contract requirements, Asaokas method is used to
predict ultimate settlement and hence, the degree of
consolidation achieved prior to the removal of excess
surcharge fill.
Time [days]
0 100
0.0
-1.0
Settlement [m]
-2.0
-3.0
-4.0
-5.0
18
16
14
Fill Thickness [m]
12
10
8
6
4 Construction is acknowledged.
2
0 REFERENCES
0 100
Fig.12. Typical measured time-settlement behaviour
Back-analysis for the coefficient of horizontal
consolidation (ch) were carried out from the slope of
best-fit line of Asaokas plot (1) using Barrons (1948)
theory of radial consolidation expressed by:
-8ch t
1 = exp
de2 F(n)
where:
t = time increment adopted in Asaokas plot
de = equivalent soil cylinder diameter
= 1.05 PVD spacing installed on triangular grid,
F(n) ln(n)0.75, where n= de/dw
dw = equivalent PVD diameter = 2(a+b)/
a and b are width (100 mm) and thickness (4 mm) of
PVD drain, respectively.
It must be noted that the F(n) function in equation
(4) does not specifically account for smear effect
caused by the PVD installation and the back-analysed
ch is therefore a lumped value which also reflects the
smear effect embodied in the slope (1) of the best-fit
Asaoka line. The back-analysed ch values range from
1.38 to 2.55 compared to the adopted design value of
1.0.
Back-calculations were also carried out to estimate
the compression ratio of the underlying soft clay using
equation (2) and assuming that cc=5cr. It was found that
the back-analysed average compression ratio (cr/1+eo)
ranges from 0.26 to 0.32, which are generally within
the adopted values in the design as shown in Fig. 7.
8 CONCLUSIONS
Significant advances have been made in recent
decades on the understanding of soft clay behaviour
although it remained a challenge having to rapidly
accelerate the construction works for proposed Power
Plants under such ground conditions. In the quest to
200 300 400
rapidly surcharge the fill embankment, it is very
important to ensure its stability at all times with the
appropriate prediction on the gain of shear strength for
the soft clay under surcharge loads.
ACKNOWLEDGEMENTS
The data for this work has been collected by IMW
Dredging Sdn Bhd who is the appointed EPC
contractor. It would not have been possible to prepare
this paper without the valuable source data. The authors
wish to thank Malakoff Berhad and IMWD for
permission to publish this paper and bring this
important project to a wider audience. The support of
Rentak Jitu Project Management Sdn Bhd and Zelan
200 300 400
1) Asaoka, A. (1978): Observational procedure of settlement
prediction, Soils and Foundations 18(4):87-101.
2) Barry, T., Krishnan, S. and Van der Bend, L (2004),
Reclamation for a Power Plant in Johor, Malaysia,
Proceedings15th Southeast Asian Geotechnical Society
Conf, 22-26 Nov 2004, Bangkok, Thailand, pp. 527-532.
3) Barron, R.A. (1948), Consolidation of fine-grained soils by
drain wells. Transactions of ASCE, Vol. 113, 718-742.
4) Bergado, D.T. and Long, P.V. (1994), Numerical Analysis
(4) of Embankment on Subsiding Ground Improved by Vertical
Drains, Proceedings Intl. Conf. Soil Mechanics and Soil
Mechanics and Found. Engr., New Delhi, pp. 1361-1366.
5) Bo, M.W., Chu, J., Low, B.K., Choa, V (2003): Soil
Improvement Prefabricated Vertical Drain Techniques.
Thomson Press, Singapore. 359 pp.
6) Plaxis (2002): Plaxis 2D-Version 8.2, Materials Models
Manual, A.A. Balkema.
7) Tan, S.A (1995): Validation of hyperbolic method for
settlement in clays with vertical drains. Soils and
Foundations 35(1): 101-103.
1772