See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/245407970

Reclamation of a slurry pond in Singapore

Article in ICE Proceedings Geotechnical Engineering · January 2009

DOI: 10.1680/geng.2009.162.1.13

CITATIONS READS

42 938

3 authors:

Arul Arulrajah Myint Win Bo

Swinburne University of Technology Bo & Associates Ltd
428 PUBLICATIONS 11,757 CITATIONS 127 PUBLICATIONS 3,843 CITATIONS

SEE PROFILE SEE PROFILE

Jian Chu
Nanyang Technological University
366 PUBLICATIONS 9,174 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

A review of the recent studies on innovative bricks View project

Applicability of Construction and demolition Waste (C&D) and plastic (PET) waste in Road and Railway applications View project

All content following this page was uploaded by Arul Arulrajah on 23 March 2015.

The user has requested enhancement of the downloaded file.

Proceedings of the Institution of
Civil Engineers
Geotechnical Engineering 162
February 2009 Issue GE1
Pages 13–20
doi: 10.1680/geng.2009.162.1.13

Paper 800054
Received 02/07/2007
Accepted 10/10/2008
Jian Chu Myint Win Bo Arul Arulrajah
Keywords: geotechnical Associate Professor, School of Director (Geo-services), DST Senior Lecturer, Swinburne
engineering/land reclamation Civil and Environmental Consulting Engineers Inc., Thunder University of Technology,
Engineering, Nanyang Technological Bay, Ontario, Canada Melbourne, Australia
University, Singapore

Reclamation of a slurry pond in Singapore

J. Chu PhD, M. W. Bo DUC, MSc, PhD, CGeol, CSci, CEng, CEnv, EurGeol, EurEng, PGeo, FICE, FGS and
A. Arulrajah PhD, CPEng, FIEAust

A case study for the reclamation of a slurry pond as part

of an offshore reclamation project in Singapore is
presented in this paper. The slurry pond covered an area
of 180 ha. The slurry in the pond was recently deposited Slurry pond
ultra-soft high-plasticity clay. The water content of the
slurry was more than 120% and the undrained shear
strength was less than 8 kPa. The reclamation was first
carried out by spreading sand fill in thin layers 20 cm
thick using a specially designed sand spreader. The filling Proposed runway
speed was carefully controlled to allow the slurry to be and taxiway
W

consolidated before more fill could be placed. Despite

S

Outflow containment bund

E

the precautions a failure occurred, in the form of mud

bursting. As a remedial measure, geotextile sheets were
Figure 1. Location of slurry pond and site plan of Changi East
used to cover a total area of 630 000 m2 before more
reclamation project
sand fill was placed. After the completion of fill
placement, fill surcharge and prefabricated vertical
drains (PVDs) were used to improve and accelerate the
consolidation of the slurry. As the performance of PVDs
would deteriorate after they had undergone large
deformation, they were installed in two passes. In the
first pass PVDs were inserted with a square grid spacing
of 2.0 m. After nearly 1.5 m of settlement had taken
place, the second pass of PVDs with the same spacing
was installed at the centre of the square grid of the
PVDs installed in the first pass. After nearly 4 years of
consolidation, the top of the slurry had settled more
than 3 m. The undrained shear strength had also
increased substantially. Therefore the use of PVDs for
the improvement of the ultra-soft slurry was successful
in this project.
Figure 2. Picture of slurry pond before reclamation
1. INTRODUCTION
The Changi East reclamation project was carried out between
1991 and 2005 to create 2000 ha of land offshore for the
extension of the Changi International Airport and other borrow pit. A containment sand bund was constructed around
infrastructure developments in Singapore. 1,2 This project this borrow pit in 1986 to the crest level of about +5 mCD.
included the reclamation of a 180 ha slurry pond. The Subsequently silt and clay washings from other sand-
location of the slurry pond and the site plan of the Changi quarrying activities in the eastern part of Singapore were
East reclamation project are shown in Figure 1. A picture of transported through pipelines with water and discharged into
the slurry pond, taken before reclamation, is shown in Figure this contained area to form a pond. Therefore the slurry inside
2. The slurry pond was trapezoidal in shape, approximately the pond consisted mainly of clay and silt. 3,4 The water level
2000 m long and 750–1050 m wide, as shown in Figure 3. in the pond was at +3 mCD. During subsequent years
This slurry pond was created by dredging the original seabed suspension in the slurry had settled in the pond, and the
to an elevation of
22 mCD (Admiralty chart datum, where majority of the slurry was undergoing self-weight
mean sea level is at +1.6 mCD) between 1975 and 1978 as a consolidation. 4

Geotechnical Engineering 162 Issue GE1 Reclamation of a slurry pond in Singapore Chu et al. 13
 X1600

X1800

X2000

X2200

X2400

X2600

X2800

X3000

X3200
Atterberg limits and

X3400
water content: % Vane shear strength: kPa

Y2140 0 100 200 300 400 0 6 12 18

0 0
⫺1
⫺ Y1955
2
⫺1
⫺4 5 5

4
⫺
⫺4
Y1775

⫺
3
Y1595

⫺3
⫺

10 10
5

Depth: m

Depth: m
⫺2 Y1415
⫺1 ⫺3
Y1235 15 15

20 20
Figure 3. Dimensions of slurry pond, and isolines indicating Plastic limit
elevation of top of slurry in the pond (mCD). The water table Water content
was at +3 mCD Liquid limit
25 25
(a) (b)

2. SITE CONDITIONS AND PROPERTIES OF SLURRY Figure 5. Variation of basic slurry properties with depth:
(a) liquid limit, plastic limit and water content; (b) vane shear
Site investigation for the slurry pond was carried out in 1992. It
strength
included 82 gamma-ray density probes and 20 boreholes with
sampling and in situ vane shear tests. The sampling was carried
out using twist samplers. The twist sampler provided a closing
100
door on a twisting mechanism, and thus allowed very soft core Percentage passing
90
samples to be taken continuously. A cross-section of the soil 80
profile in the slurry pond along chainage X2600 is shown in 70
60
Figure 4. The elevation at the top of the slurry varied from -1 to 50
-5 mCD as shown by the isolines in Figure 3. There was no clear 40
30
boundary between water and slurry, as the transition was rather
20
gradual. The top of the slurry was taken as the elevation at 10
which the density was greater than 1.1 Mg/m3 . The bottom of 0
0·001 0·010 0·100 1·000
the slurry was in the range 0 to
22 mCD. The thickness of the Particle size: mm
slurry ranged between 3 and 20 m.
Figure 6. Grain size distributions of the slurry
The slurry material consisted mainly of high-plasticity clay.
The variation of liquid limit, plastic limit and moisture content
distribution with depth is shown in Figure 5a. The water Property Value
content of the slurry varied from 60% to 300%, but was mainly
in the range of 120–180%. The bulk density was low, with Liquid limit: % 65–115
values ranging from 1.2 to 1.4 Mg/m3 . The slurry was still Plastic limit: % 22–45
undergoing self-weight consolidation. The undrained shear Water content: % 75–180
strength profile of the slurry as determined by field vane shear Bulk unit weight: kN/m3 11.0–16.0
Specific gravity 2.66–2.70
testing is shown in Figure 5b. The majority of the data points Void ratio 2.0–4.5
in Figure 5b show an increasing trend with depth at a rate of Fines content: % 70–90
0.2 kPa/m depth. Even so, the values were less than 8 kPa for
soils at 20 m depth. The grain size distribution of the slurry is Table 1. Properties of soil in the slurry pond
shown in Figure 6. It can be seen that the upper bound of D50
was 0.024 mm, but mostly in the range smaller than 0.001, and
D85 was in the range 0.004–0.02 mm. The fines content was in the range 70–93%. The other physical properties of the soil in
the slurry pond are shown in Table 1.

6·0 3. RECLAMATION AND SOIL IMPROVEMENT WORKS

Y1250 Y1450 Y1650 Y1850 Y2050
0 3.1. Sand placement by spreading
Elevation: mCD

Sand
As the top surface of the slurry had little strength, direct
⫺6·0
hydraulic placement of sand fill on top of the slurry would
Slurry
⫺12·0 have caused the slurry to be displaced by the penetration of the
sand fill. The sand fill therefore had to be spread in thin layers
⫺18·0 using a specially designed sand spreader, as shown in Figure 7.
Original dredged profile
⫺22·0 The sand fill, with a high water to sand ratio, was pumped into
the spreader using a suction dredger. The sand spreading was
Figure 4. Cross-section of slurry pond along chainage X2600
and sand fill placed before failure carried out by moving the spreader repeatedly from left to right
along the x-direction (see Figure 3). Small lifts of 20 cm were

14 Geotechnical Engineering 162 Issue GE1 Reclamation of a slurry pond in Singapore Chu et al.
 Generator
Dredging pipe No. 2

Floater Dia. 900 mm

Control
room

Pontoon No. 1 Pontoon

Generator

No. 3 anchor No. 2 anchor

12 m ⫻ 4 ⫽ 48 m

Pontoon

Sand fill

Figure 7. Sand spreading system used for slurry pond

used in the first phase of spreading to ensure stability of the

fill. A waiting time was given between each 20 cm lift to allow 6·0
the slurry to gain strength under the small surcharge of the
0
sand fill. A hydrographic survey carried out along chainage
Elevation: mCD

X2600 when the sand layer reached 0 mCD is shown in Figure ⫺6·0
4. When the fill reached an elevation between 0 and +2 mCD, a
⫺12·0 Clay
failure in the form of slurry bursting occurred at the location Sand
shown in Figure 8. A cross-section across the failure area along ⫺18·0 Clay
chainage Y1700 is shown in Figure 9. The boundary between
⫺24·0 Sand
the sand fill and the slurry was established by observation of
the borehole samples. It can be seen from Figure 9 that, at the X1800 X2200 X2600 X3000 X3400
X2000 X2400 X2800 X3200
burst location, the sand fill was displaced to two sides and the
mud shot out to an elevation of +2 mCD.
Figure 9. Cross-section of slurry pond along chainage Y1700
after failure
The failure was attributed to uneven settlements and uneven
spreading of the sand fill. It can be seen from Figures 4 and 9

that the thickness of the slurry layer varied considerably

X1600

X1800

X2000

X2200

X2400

X2600

X2800

X3000

X3200

X3400

across both the x and y directions. The amount of

compressibility of the slurry layer was thus different across
both the x and y directions, even when the load applied was
Y2000
the same. The pore water pressure dissipation rates would also
Failure Y1800 be different. The profile of the sand fill before failure along
area
X2600 is shown in Figure 4. The profile of the sand fill along
Y1600
Y1700 can also be seen from the stable sections in Figure 9.
Y1400 Although the sand fill was spread carefully, a 1 m or so
difference in the sand layer thickness could have resulted. The
Y1200 difference in loads, coupled with the different pore pressure
dissipation rates in the soil, led to upheaval of the slurry at
some locations. This in turn displaced the sand above the
Figure 8. Location of failure area where bursting of slurry
occurred upheaved slurry to two sides and subsequently caused the
overburden pressure provided by the sand to be reduced.

Geotechnical Engineering 162 Issue GE1 Reclamation of a slurry pond in Singapore Chu et al. 15
 When the pore pressure became higher than the overburden used was 320 t. To overcome this problem, the geotextile sheet
pressure a mud burst occurred. was folded into zigzag strips 10 m wide, as shown in Figure 11.
This made unfolding of the geotextile sheet much easier. The
3.2. Remedial measures using geotextile sheet was anchored at one end by placing sand on top of it. The
As a remedial measure, geotextile fabric was used to cover the geotextile was then folded back to cover the sand, which would
failure area. Geotextile has also been used for other, similar be covered again by sand fill placed subsequently. The amount
projects.5,6 Before this, the burst mud was partially removed of sand used was 3–6 m3 per metre run. The other end of the
using a high-capacity submersible mud pump. The tensile geotextile sheet was connected to pipelines floating on water.
strength of the geotextile required was estimated from the The pipelines were pulled slowly towards the other side of the
equation 7 pond using seven bulldozers. Plastic buoys were also used for
the inner area of the geotextile sheet to keep it floating on top


p
Tg sin
¼ c u Nc þ ª9D þ Tg sin
Nq of the water. When the geotextile sheet reached the other side
1 of the pond, it was fixed in place by placing sand fill on top in
þ 0:5ª2 Nª similar fashion as for the other end. For the area where double
layers of geotextile were used, the second layer was placed in
where p is the vertical stress due to sand fill; Tg is the tensile the same way after the first layer had been installed. After the
strength of the geotextile;
is the angle of curvature of the geotextile sheet was properly anchored at both ends, the plastic
geotextile due to arching, determined based on the length and buoys were removed to allow the geotextile sheet to sink. At the
width of the geotextile sheet used and the predicted maximum interface of the two areas where the two different geotextiles
settlement of the geotextile along the centreline; cu is the were used, an overlapping length of 50 m was provided. A sand
undrained shear strength of the slurry; ª9 is the effective unit layer 50 cm thick was placed on the overlapping area for
weight of underlying soil; D is depth and Nc , Nq and Nª are anchoring.
Terzaghi’s bearing capacity coefficients.

Based on the calculation, a tensile strength of 150 kN/m in 3.3. Second phase of sand spreading
both the warp and weft directions was determined. Two types After placement of the geotextile, sand fill spreading resumed,
of woven geotextile were used. The first type, HS150/150, had using the same sand spreader, until the ground elevation of
a tensile strength of 150 kN/m in both directions. It was placed +4 mCD was reached. In order to achieve this level, the
in a single layer to cover an area 700 m by 300 m, shown in containment bund surrounding the slurry pond was elevated to
Figure 10 as area 2 on the left. Owing to the shortage of supply +6 mCD and the water level to + 5.5 mCD by pumping in
of this type of geotextile, a second type, HS100/50, with tensile seawater. Spreading was still carried out in stages, with the
strengths of 100 kN/m and 50 kN/m in the warp and weft thickness of each layer controlled within 50 cm. Mini-cone
directions respectively, was also used. By overlaying two layers penetration tests (MCPTs) were performed to determine the soil
with different warp and weft directions, a combined 150 kN/m profile of the slurry pond after fill placement. The MCPT test
tensile strength was achieved. The second type of geotextile used a 308 apex cone with a cone tip area of 2 cm2 . Based on the
was placed in two layers over an area 700 m by 600 m, shown MCPT results, the soil profile along chainage Y1700 is shown in
in Figure 10 as area 3 on the right. Therefore the total area Figure 12. It can be seen that the sand fill was successfully
covered by the geotextile was 700 m 3 900 m, or 630 000 m2 . placed on top of the slurry, except for the existence of some
small sand pockets. The results of MCPT test 47 are given in
The installation was carried out in the following sequence. The Figure 13. Based on the tip resistance and friction ratio shown in
geotextile as supplied was in rolls 5 m wide and 90 m long. Figure 13, it can be seen that at this location the sand layer
They were sewn together using portable sewing machines to ranged from zero to about 8 m deep, the clay layer started from
form a geotextile sheet either 700 m 3 600 m or 700 m 3 11 m, and a mixed clay and sand layer occurred in between. The
900 m. When the geotextile was sewn together to form one
piece, it was extremely heavy and required considerable force to
be pulled during installation. The total weight of the geotextile

Y2000

Y1800

Y1600
1
Y1400

Y1200
10 m
2 3

Figure 10. Layout of geotextile sheets used to cover failure

area of slurry pond: 1, failure area; 2, single layer of
geotextile, 700 m 3 300 m; 3, double layers of geotextile,
700 m 3 600 m Figure 11. Folding of geotextile sheet for installation

16 Geotechnical Engineering 162 Issue GE1 Reclamation of a slurry pond in Singapore Chu et al.
 6·0 large ground settlement would cause the drains to buckle and
thus affect their performance. The deterioration in the
0 performance of the vertical drains installed in the first pass was
indicated by the fact that the rate of settlement had reduced to
Elevation: mCD

Sand Sand
pocket 20 cm/month, although the pore water pressure remained very
⫺6·0

Slurry
high in the soil. 4,9 The installation of PVDs itself assisted in the
87
⫺12·0 MCPT34 dissipation of the pore water pressure, as mud was seen to come
40 43 97
60 out through the annulus of the mandrel. 10 This is another
93
⫺18·0 advantage of the two-stage PVD installation method. The final
33 84 surcharge level of + 9 mCD was placed after installation of the
20 47
⫺24·0 30 52 57 7277 second round of vertical drains.
62 67 82 92
X1500 X2000 X2500 X3000 X3500
The settlement and pore water pressure dissipation within the
Figure 12. Soil profile along chainage Y1700 after reclamation first 480 days are shown in Figure 14. The positions of the
as identified by mini-CPT tests. MCPT borehole numbers are settlement and pore pressure measuring devices are shown in
shown. For each borehole, darkened area indicates presence
of clay; the rest is sand Figure 15. It can be seen from Figure 14b that, within the first
480 days, there was 2.7 m of ground settlement, but the pore
pressure dissipations (Figure 14c) were slow. Using Figure 14c,

Cone resistance/MPa-friction ratio: % 7

0 1 2 3 4 5 6 Second PVD
installation

Surcharge elevation: mCD

0

2 5

6 3
First PVD
8 installation
Depth: m

10
1
0 90 180 270 360 450 540
12 Time: days
(a)
14

16 Time: days
Cone resistance 0 90 180 270 360 450 540
18 0
Friction ratio
0·5
20
Settlement: m

1·0
22
1·5
Figure 13. Results of mini-CPT test MCPT47
2·0

2·5 SP-506 DS-503

DS-504 DS-505
DS-511 DS-507
low tip resistance for the clay layer indicates that the shear 3·0
(b)
strength of the clay slurry was still low after the sand fill had
been placed. Therefore further soil improvement was carried out 150
by preloading with PVDs after the sand placement.
Excess pore pressure: kPa

125

3.4. Installation of vertical drains 100

Before installation of the PVDs at the +4 mCD level, the water 75

level in the slurry pond was lowered to +3 mCD. Colbond drains PZ-47
50 PZ-48
CX1000 were installed with 2 m 3 2 m square spacing as a first PZ-49
pass. The specifications for the PVDs are given in Reference 2. 25
The procedure discussed by Chu et al. 8 was followed in selecting
0
the type of drain and drain spacing. The surcharge in the first 0 90 180 270 360 450 540
stage was placed to +6 mCD. The settlement of the fill was Time: days
(c)
monitored using an instrumentation scheme as detailed by Bo et
al.9 After approximately 1.5 m of settlement had taken place, a
second pass of vertical drains was installed with the same square Figure 14. Monitoring data measured against time during
reclamation of slurry pond: (a) surcharge variation;
grid of 2.0 m in the centre of the square grid of PVDs installed in
(b) settlement; (c) excess pore pressure. SP, settlement plate;
the first round. The combined effective drain spacing is 1.4 m 3 DS, deep settlement gauge; P2, vibrating wire piezometer
1.4 m. It was necessary to install the PVDs in two passes, as the

Geotechnical Engineering 162 Issue GE1 Reclamation of a slurry pond in Singapore Chu et al. 17
 SP 506 WS 110
⫹6·0 mCD

⫹4·0 mCD
⫹3·50

PP 467 Sand
DS 504 PZ 049
DS 504 PP 468
SP 508 PZ 049 PP – Pneumatic piezometer
DS 503 PZ – Vibrating-wire piezometer
PP 466
PZ 049 DS – Deep settlement gauge
DS 507 DS 511 WS 110 SP – Settlement plate ⫺5·30 DS 503
WS – Water standpipe Very soft slurry soil

PBS 5 ⫺8·00 DS 504 PP 466 PZ 047

⫺10·00 DS 505 PP 467 PZ 048

⫺12·00 DS 511 PP 468 PZ 049

3m ⫺13·0 mCD

3m
⫺14·00 DS 507
(a) Sand

(b)

Figure 15. Arrangement of instrumentation: (a) plan; (b) elevation

the pore pressure distribution in 480 days is plotted in Figure Field vane shear (FVT) and piezocone (CPTu) tests were
16. Based in Figure 16, the average degree of consolidation can conducted 14 months after application of the surcharge. A
be estimated as 42%. Using this value, and taking the clay comparison of the undrained shear strength profiles obtained
slurry layer as 7.8 m thick and double-drained, ch can be back- from the FVT tests conducted before surcharge and 14 months
calculated as 0.27 m2 /year when cv is assumed to be 0.1 m2 / after surcharge, as well as from the CPTu tests conducted 14
year. This value is of the same order of magnitude as that months after surcharge, is shown in Figure 17. Note that the
reported by Bo et al.4 Applying Asaoka’s method 11 to the ground had settled by more than 2 m within 14 months, as
settlement data in Figure 14b, the average degree of shown in Figure 14a. This explains why the starting-points of
consolidation is calculated as 91%, which seems too high. the in situ tests conducted before and after the surcharge are
Therefore Asaoka’s method is not applicable in this case, where different. The undrained shear strength determined by CPTu
large deformation of sediment was encountered. was based on a cone factor of 12. 3 In Figure 17, the undrained
shear strength profile estimated by assuming a uniform degree
of consolidation of 90% is also plotted for comparison. It can
Pore water pressure: kPa be seen that the slurry had little shear strength initially.
0 100 200 300 However, after consolidation, a substantial increase in the
⫺7
undrained shear strength had been achieved 14 months after
the surcharge placement. As mentioned, a silty sand was
⫺8
present at an elevation of
12.5 mCD and a sand blanket was
placed on the top surface. Therefore there was combined
vertical and horizontal drainage near the top and bottom
⫺9 boundary. This explains why the shear strength increment was
greater at both the top and the bottom, where the undrained
Elevation: m

shear strength had reached a value equivalent to 90% of pore

⫺10
pressure dissipation. The consolidation in the middle of the
clay layer, ranging from
9.5 to
12.0 m, was contributed
mainly by horizontal drainage to the PVDs, as indicated in the
⫺11
analysis by Chu et al. 10 The increase in the undrained shear
strength in this section was thus relatively uniform, as
Hydrostatic
⫺12 indicated by the dashed lines. Using the undrained shear
Surcharge
strength profile shown in Figure 17, the average degree of
At 480 days
consolidation can be estimated as 45%. This agrees with that
⫺13 estimated using the field pore pressures shown in Figure 16.
Figure 16. Pore water pressure profile 16 months after
surcharge The much delayed dissipation of pore pressure in the slurry is
typical of consolidation of ultra-soft soil, as the slurry may

18 Geotechnical Engineering 162 Issue GE1 Reclamation of a slurry pond in Singapore Chu et al.
 Shear strength: kPa
at other locations were also monitored, and settlements greater
0 10 20 30 40 50 60 70 80 90 100 than 3.5 m were observed. 9 It was planned to remove the
⫺7
surcharge only once the land was used in the future. So far,
⫺8 only the eastern boundary of the slurry pond (see Figure 1),
FVT (prior to
where the slurry layer was thin, has been used. The surcharge
Elevation: mCD

⫺9 reclamation)
FVT in most of the areas has not yet been removed. The long-term
⫺10 settlement of the slurry would be an interesting topic to be
CPTu
reported separately.
⫺11 90% DOC

⫺12 4. CONCLUSIONS
A case study for the reclamation of a slurry pond as part of the
⫺13
offshore Changi East reclamation project in Singapore is
Figure 17. Comparison of undrained shear strength profiles presented in this paper. As the slurry was ultra-soft, direct
measured before and after consolidation for 14 months placement of sand fill onto the slurry caused the mud to burst,
even though the fill was placed in thin layers using a specially
designed sand spreader. As a remedial measure, geotextile
have to undergo sedimentation and self-weight consolidation sheets were used to cover a total area of 630 000 m2 (700 m 3
stages before it can be transformed into ‘soil’. 10,12 Excessive 900 m) before further fill placement. The slurry was treated
buckling of the PVDs, possible clogging of the filter of the using two passes of PVDs and fill surcharge. The following
PVDs, the Mandel–Cryer effect and non-uniform consolidation conclusions can be drawn from this case study.
of soil around the PVDs are possible other reasons. 10,12
(a) For reclamation over ultra-soft, high water content slurry,
Owing to the slow dissipation in pore pressure, it took several direct placement of fill can be difficult. An attempt was
years to complete the primary consolidation despite the use of made to place the fill in successive thin, 20 cm layers using
PVDs at an effective spacing of 1.4 m 3 1.4 m, achieved by a specially designed sand spreader. However, this was not
combining both passes of PVD installation. The complete entirely successful as a failure in the form of a mud burst
loading history and surface settlement curves at the same occurred around the centre of the pond. Uneven settlement
location as shown in Figure 14 are presented in Figure 18. As and uneven spreading of the sand fill were considered to be
can be seen in Figure 18a, one more surcharge increment of the main reasons for the failure. The uneven settlement was
5.5 m was applied after 2.3 years. The additional settlement due mainly to the differing compressibility of the slurry
induced by this surcharge increment was only about 0.5 m, layer with varying thickness across the cross-section. As a
which was much smaller than the nearly 3 m of settlement remedial measure, geotextile was used as a reinforcement
developed in the preceding stage. This indicated that the soil to cover the failed area of 630 000 m2 before fill placement.
had been much improved over the preceding 2.3 years. The (b) When PVDs are used for the improvement of ultra-soft,
total settlement 4 years after preloading was 3.5 m, and the high water content soil, their performance may deteriorate
degree of consolidation achieved was 85%. Ground settlements after they have experienced large deformation. This
problem was overcome in this project by installing PVDs in
two passes, one at the start and another after some
10
settlement had occurred. The second pass of PVDs was
Surcharge elevation: mCD

installed in the centre of the grid of the PVDs installed in

8 Second PVD installation
the first round. This method was proven to be successful. In
6
4 years, the top of the slurry had settled by more than 3 m.
The undrained shear strength of the slurry soil had also
4 increased substantially.
2
First PVD installation (c) Even with the use of PVDs at a close spacing, the
consolidation of ultra-soft soil is a slow process, and the
0 dissipation of pore water pressure often lags behind the
0 1·5 3·0 4·5
Time: years settlement. In this case, the degree of consolidation can be
(a) considerably overestimated using the settlement data. It is
therefore necessary to estimate the degree of consolidation
Time: years
0 1·5 3·0 4·5 using pore water pressure or undrained shear strength
0
profiles, as suggested in this paper.
1
Settlement: m

ACKNOWLEDGEMENT
2 The authors would like to thank Mr Wei Guo for his help in the
preparation of some of the figures for this paper.
3
REFERENCES
4 1. BO M. W., CHU J. and CHOA V. The Changi East reclamation
(b)
project in Singapore. In Ground Improvement: Case
Figure 18. Surcharge and settlement against time Histories (INDRARATNA B. and CHU J. (eds)). Elsevier, Oxford,
2005, pp. 247–276.

Geotechnical Engineering 162 Issue GE1 Reclamation of a slurry pond in Singapore Chu et al. 19
 2. CHU J., BO M. W. and ARULRAJAH A. Soil improvement 7. NA Y. M., CHOA V., BO M. W. and ARULRAJAH A. Use of
works for an offshore land reclamation. Proceedings of the geosynthetics for reclamation on slurry like soil
Institution of Civil Engineers—Geotechnical Engineering, foundation. In Problematic Soils (YONAGISAWA E., MOROTO
2009, 162, No. 1, 21–32. N. and MITACHI T. (eds)). Balkema, Rotterdam, 1998, pp.
3. BO M. W., ARULRAJAH A., CHOA V. and CHANG M. F. Site 767–771.
characterization for a land reclamation project at Changi 8. CHU J., BO M. W. and CHOA V. Practical consideration
in Singapore, Proceedings of the 1st International. for using vertical drains in soil improvement projects.
Conference on Site Characterization, Atlanta, GA, 1998, Geotextiles and Geomembranes, 2004, 22, No. 3, 101–117.
333–338. 9. BO M. W., CHOA V. and WONG K. S. Reclamation and soil
4. BO M. W., ARULRAJAH A., CHOA V. and NA Y. M. Land improvement on ultra-soft soil. Ground Improvement,
reclamation on slurry-like soil foundation. In Problematic 2005, 9, No. 1, 23–31.
Soils (YONAGISAWA E., MOROTO N. and MITACHI T. (eds)). 10. CHU J., BO M. W. and CHOA V. Improvement of ultra-soft
Balkema, Rotterdam, 1998, pp. 763–766. soil using prefabricated vertical drains. Geotextiles and
5. BROMS B. B. Stabilization of very soft clay using geofabric. Geomembranes, 2006, 24, No. 6, 339–348.
Geotextiles and Geomembranes, 1987, 5, No. 1, 17–28. 11. ASAOKA A. Observational procedure of settlement
6. TERASHI M. and KATAGIRI M. Key issues in the application of prediction. Soils and Foundations, 1978, 18, No. 4, 87–101.
vertical drains to a sea reclamation by extremely soft clay 12. CHU J., GOI M. H. and LIM T. T. Consolidation of cement-
slurry. In Ground Improvement: Case Histories (INDRARATNA treated sewage sludge using vertical drains. Canadian
B. and CHU J. (eds)). Elsevier, 2005, pp. 145–158. Geotechnical Journal, 2005, 42, No. 2, 528–540.

What do you think?

To comment on this paper, please email up to 500 words to the editor at [email protected]
Proceedings journals rely entirely on contributions sent in by civil engineers and related professionals, academics and students. Papers
should be 2000–5000 words long, with adequate illustrations and references. Please visit www.thomastelford.com/journals for author
guidelines and further details.

20 Geotechnical Engineering 162 Issue GE1 Reclamation of a slurry pond in Singapore Chu et al.

View publication stats