Ground Improvement (2007) 11, No.

2, 6776 67

Preloading and prefabricated vertical drains

design for foreshore land reclamation projects:
a case study
M. W. BO*, A. ARULRAJAH y and H. NIKRAZ{
*Faber Maunsell Ltd, Bradford, UK; y Swinburne University of Technology, Hawthorn,
Australia; { Curtin University of Technology, Perth, Australia

The Changi East Reclamation Project in Singapore com- Le projet Changi East Reclamation a` Singapour porte sur
prises land reclamation in foreshore conditions. As the la rehabilitation de terrain dans les conditions destran.
foreshore area was underlain by a thick layer of soft Comme la zone destran repose sur une couche epaisse
marine clay, a high magnitude of primary and secondary dargile marine molle, un tassement des consolidations
consolidation settlement was expected, owing to fill and primaire et secondaire de grande amplitude etait presume,
future live load. An average fill thickness of 10 m was en raison de remplissage et de charge utile future. On
predicted to contribute to a magnitude of settlement great- estimait quune epaisseur de remplissage de 10 m pourrait
er than 2 m over a period of several decades. In order to contribuer a` une amplitude de tassement de plus de 2 m
minimise this expected future settlement, it was necessary sur une periode de plusieurs decennies. Afin de minimiser
to accelerate the consolidation process to complete the letendue du tassement prevue, il etait necessaire dacce-
majority of the settlement during construction stage. A lerer le processus de consolidation pour que la majorite du
combination of prefabricated vertical drains (PVDs) with tassement se termine pendant la phase de construction. Ce
preloading ground improvement was successfully applied projet combine avec succe`s lutilisation de drains verticaux
in this project. The technique comprises the installation of prefabriques (DVP) avec amelioration des sols par prechar-
prefabricated vertical drains and the subsequent place- gement. Le technique adoptee consiste a` installer des
ment of surcharge to accelerate the consolidation of the drains verticaux prefabriques puis a` placer une surcharge
underlying marine clay. This paper discusses the theories pour accelerer la consolidation de la couche dargile
of radial drainage and the preloading technique, and marine sous-jacente. Cet article presente une discussion
considerations and design methodologies for the ground sur les theories de drainage radial et la technique de
treatment of marine clay with prefabricated vertical drains prechargement, ainsi que des considerations et des meth-
in such foreshore land reclamation projects. In addition, odologies de conception geotechnique pour la consolida-
predictions of magnitude, time rate of settlement and tion de largile marine dans le cadre dun tel projet de
performance assessment of prefabricated vertical drains rehabilitation de terrain dans les conditions dun estran. Il
are also discussed. The design predictions at a case study examine egalement les predictions concernant lamplitude
site were compared with the field instrumentation results et le taux de tassement en fonction du temps, et les
in order to verify the design approach used. It is found performances des drains verticaux prefabriques. Les previ-
that PVDs with preloading are the most effective method sions de conception sur un site detude de cas sont
for improving soft clay under land reclamation, based on comparees avec les resultats dexperimentation de terrain
the soil conditions present at the case study site. afin de verifier ladequation de lapproche technique
utilisee. En conclusion, la methode de PVD avec prechar-
gement se montre la plus efficace pour ameliorer largile
molle dans le cadre de rehabilitation de terrain, lorsque
Keywords : land reclamation; prefabricated vertical lon se base sur les conditions de sol presentes sur le site
drains; preloading; soft clay; soil improvement detude de cas.

Notation Cv n
de
coefficient of consolidation for vertical flow for layer n
diameter of equivalent soil cylinder
Ch coefficient of consolidation for horizontal flow ds diameter of smeared zone
Ckv permeability change index dw equivalent diameter of drain
Cv coefficient of consolidation for vertical flow e0 natural void ratio
Cvi assumed average value for coefficient of consolidation for F (n) drain spacing ratio function
vertical flow Fs (n) drain spacing ratio function with smear effect
H91 equivalent thickness of layer 1
Hdri equivalent drainage length
H9Ti total equivalent thickness of all layers
(GI 5238) Paper received 24 March 2005; last revised 6 October 2005;
kh horizontal hydraulic conductivity of soil
accepted 21 July 2006

www.groundimprovement.com 1751-7621 (Online) 1365-781X (Print) # 2007 Thomas Telford Ltd

M. W. Bo et al.

kr coefficient of permeability of remoulded soil well resistance required in order to eliminate future settlement caused by
coefficient fill and future loads. The marine clay characteristics are
kw coefficient of longitudinal permeability of drain briefly described in the next section. For further details see
n drain spacing ratio Bo et al. (1998a, 2003), Arulrajah et al. (2004a, 2004b) and
s smear zone ratio Arulrajah (2005).
St settlement at time t
Sult ultimate settlement
Tr time factor for consolidation by horizontal drainage
Tv time factor for consolidation by vertical drainage
t time in years Ground condition
Ur average degree of consolidation with respect to radial flow
Uv average degree of consolidation with respect to vertical flow The area is underlain by two layers of highly compressible
Uvr average degree of consolidation with respect to both vertical marine clay known locally as upper and lower marine clay.
and radial flows These two layers are separated by a thin intermediate layer
of desiccated lower marine clay. The upper marine clay is
usually about 10 m thick, and the thickness of the lower
marine clay can be as much as 30 m. The intermediate clay
Introduction layer is usually about 12 m thick. The lower marine clay is
underlain by an old alluvium silty sand layer. The upper
The site for the study is located in the Changi East marine clay has a moisture content of 6080%, whereas that
Reclamation project in the Republic of Singapore (Fig. 1). of the lower marine clay is 4060%. The intermediate layer
The project comprises land reclamation and ground im- has a low moisture content, of 2030%. Both the upper and
provement works to allow for the future expansion of lower marine clay layers are highly plastic, with plasticity
Changi International Airport, comprising the runway, taxi- indices ranging between 40% and 60%. Both layers are
ways, turn-offs and associated airport facilities. The area is lightly consolidated, with compression indices ranging be-
located on the foreshore, east of Singapore, with seabed tween 0.6 and 1. The upper marine clay layer has a low
elevation varying from
2 mCD to
8 mCD. (mCD stands coefficient of consolidation of 0.50.6 m2 /year, whereas that
for metres chart datum; mean sea level is +1.6 mCD.) of the lower marine clay is 0.81.5 m2 /year.
Vertical drains were designed for installation in the project The permeability of Singapore marine clay ranges between
site at the platform level of +4 mCD at 1.5 m square spacing. 10
10 and 10
9 m/s due to vertical flow and 10
9 m/s due to
A surcharge placement 6 m in height (+10 mCD) was horizontal flow. The permeability anisotropy is about 2.
designed to be carried out. The area is underlain by a thick Permeability is generally reduced with a decrease in void
layer of Singapore marine clay, which is highly compressible ratio. The permeability change index (Ckv ) is 0.3 of the
and has low permeability. Soil improvement work is natural void ratio e0 . Further details of the characteristics of

REPUBLIC OF SINGAPORE

Safra
Country Club EX
T
D GE
V R
AI NV
EN
TG N N T.M.C.C. LOCATION
EX
DR
AI Singapore Changi Airport

Changi Coastal Road

Project Area: ha Length of
vertical
Phase 1C
drain: Mm
Phase 1A 501
Phase 1B 520 28
Phase 1C Phase 1B Phase 1C 451 49
A A Area A North 91 13
Area A South 450 50
Phase 1A Total 2086 140
B B
Area A
Area A South
North

Sea

0 500 1000 2000 3000 m

Fig. 1. Site plan

68
 Preloading and prefabricated vertical drains for foreshore land reclamation

Prior to reclamation Water Clay Field vane shear Compression Preconsolidation

borehole PB-39 content: % fraction: % strength: kN/m2 index pressure: kN/m2

Seabed 2329 mCD 0 50 100 0 20 40 60 0 50 100 0 05 10 15 20 0 100 200 300

Fine to medium sand

25

Very soft marine

210 clay with some seashell
fragments

215 Soft silty clay

Firm silty clay

Elevation: mCD

220

Soft marine clay

225

230 Soft to firm silty clay

with traces of organic
matter

Clayey silty sand

235 Stiff clayey sand
Dense silty sand Effective
overburden
Dense clayey sand pressure
240
PL M/C LL

Fig. 2. Typical soil characterisation of study area

Singapore clay (consolidation, compressibility and hydraulic kv and kh: m/s

conductivity) can be found in Bo et al. (1998, 2003) and Chu 100 3 10211 100 3 10210 100 3 1029 100 3 1028 100 3 1027
0
et al. (2002). Fig. 2 presents a typical profile of geotechnical
characteristics, and Fig. 3 shows the permeability profile of Location 1 LAB. kv

the case study area. Location 2 LAB. kv

Location 3 LAB. kv
5
Location 1 LAB kh

Improving soft ground Location 2 LAB kh

Range of kh
Range of kv

Location 3 LAB kh
Soft ground can be improved by several methods, one of 10
the simplest of which is preloading. However, preloading
generally provides an average degree of consolidation rather
Depth: m

than a homogeneous improvement throughout the entire soil

layer. In addition, there is no acceleration process involved 15
when preloading alone is used. An alternative to preloading
is to improve the soft soil with prefabricated vertical drains
supplemented by preloading. This combination of vertical
20
drains and preloading will enable prestressing and accelerat-
ing consolidation of the soft soil and ensure a more
homogeneous improvement.
Other suitable methods of improving soft soil include
25
vacuum preloading, lowering of groundwater level, and the
electro-osmosis process. Details of these ground improve-
ment methods have been discussed extensively by Bjerrum
et al. (1967), Esrig (1968), Wade (1976), Mitchell (1981), 30
Hausmann (1990) and Bo and Choa (2004). Each of these
methods has its advantages and disadvantages, but the Fig. 3. Horizontal and vertical permeability profiles of the study area

69
M. W. Bo et al.

Time
method of preloading in combination with prefabricated t1 t2
vertical drains is well accepted by practising engineers.

Prefabricated vertical drains and

Settlement
preloading
90% Degree of consolidation
with 100 kPa load
The use of prefabricated vertical drains (PVDs) with
70%
preloading was considered the most feasible method, and
was used in this project. The objective of the technique is to Degree of consolidation
with 130 kPa load
accelerate the rate of consolidation and to minimise future
settlement of the treated area under future dead and live
loads. The soil improvement works are carried out in such a
way that a specified degree of primary consolidation is Fig. 4. Concept of effectiveness in application of surcharge
attained within the desired time frame by improving the soil
drainage system. The primary use of PVDs is to accelerate
consolidation to greatly decrease the duration of the con-
solidation process caused by an embankment built over soft pression can be minimised only by preloading the soil
soils. This will ensure that the final construction can be beyond its expected future load, because secondary com-
completed in a reasonable time and with minimal post- pression decreases with effective stress.
construction settlement. Preloading increases the effective
stress and reduces the compressibility of weak ground by
forcing soft soils to consolidate. By doing so, the consolida- Prediction of settlement
tion process also improves the strength of in situ soft soils.
Settlement was predicted using laboratory results from
samples collected from offshore investigation boreholes. The
Design of prefabricated vertical drains compression parameters and stress history were carefully
The main variables of the design process are the surcharge obtained from the results of primary consolidation tests
magnitude, the spacing of vertical drains, the duration of carried out on the undisturbed piston samples. The moisture
preloading, the degree of consolidation, and the coefficient content and void ratio of soils, which are important
of consolidation for horizontal flow (Ch ). Conventional parameters for predicting the magnitude of settlement, are
methods (Terzaghi, 1925; Carrillo, 1942) were used to predict obtained after the necessary salt correction. Because of the
the magnitude and time rate of consolidation settlement. salt content in the marine deposit, both the moisture content
and the void ratio determined by oven-drying in the
laboratory are underestimated values (Bo et al., 1998b, 2003;
Design of surcharge level Bo and Choa, 2004). This shifts the curve of void ratio
The aim of the surcharge placement works is to preload against log of effective stress to the left, and underestimates
the foundation soil to attain an effective stress that exceeds the yield stress. The magnitude of settlement was calculated
the pressure due to the design load. Determination of on several subdivided layers in order to be able to predict
surcharge level is dependent upon the expected settlement the ultimate settlement accurately. The use of an insufficient
and the future load. In a reclamation project where future number of subdivisions leads to underestimation of the
loading is not required, surcharge is still required to ultimate settlement (Bo and Choa, 2004). The reduction of
compensate for future settlement due to the fill load. A more load due to submergence of the fill below groundwater
acceptable way of surcharging for such a project is to place a level, and lowering of the groundwater level after stabilisa-
thickness equivalent to a multiplier of the predicted settle- tion, were also taken into considered in the additional load.
ment to compensate for the shortfall resulting from the
remaining degree of consolidation (Bo and Choa, 2004).
Generally, 90% degree of consolidation is specified. This is Determination of coefficients of consolidation
because it takes far longer to complete the final 10% than the The consolidation process with PVDs involves consolida-
first 90%. In this case, a multiplier of 1.1 times the predicted tion due to both vertical flow (Cv ) and horizontal flow (Ch ),
settlement is adopted to compensate for the shortfall result- although the former is relatively less important. Since the
ing from the 10% residual consolidation settlement. Where quaternary marine formation has been deposited under
future loading is required, a preload magnitude equivalent quiet marine conditions, there is generally no significant
to the future load and 1.1 times the expected settlement with stratification, though lamination and foliation may be pre-
fill and surcharge load is advised. (That is, 100% of 1 m sent at some locations. Hence the degree of anisotropy for
settlement is 1.0 m whereas 90% of 1.1 m settlement is hydraulic flow is considered to be low. Several in situ
0.99 m. Therefore this contributes the same magnitude of dissipation tests using cone penetrometer, dilatometer and
settlement.) In some cases, even with the closest practicable self-boring pressuremeter were carried out to determine the
spacing of PVDs, it may be difficult to reduce the consolida- coefficient of consolidation due to horizontal flow and the
tion time to the desired duration. In such cases, an horizontal hydraulic conductivity of the marine clay. Specia-
alternative method is to increase the preloading magnitude lised consolidation tests using a hydraulic Rowe cell have
to 2030% higher than the net future load and aim for a also been carried out to obtain the laboratory values. Details
lower degree of consolidation. This concept is illustrated in of the hydraulic conductivity and consolidation properties of
Fig. 4. the Singapore marine clay at Changi have been discussed by
It is known that improvement with PVDs can eliminate Bo et al. (1998b) and Chu et al. (2002). It has been reported
primary consolidation settlement. However, secondary com- that the ratio Ch /Cv for Singapore marine clay is as low as

70
 Preloading and prefabricated vertical drains for foreshore land reclamation

1.5. However, it is common practice for vertical drain where de is the diameter of the equivalent soil cylinder.
installation in marine clays to assume a Ch value of twice Cv 
1:128 3 spacing (square pattern)
(i.e. Ch 2Cv ), because there could have been some de (8)
1:05 3 spacing (triangular pattern)
lamination and foliation in the marine clay formation. Cv of
the marine clay was also determined from the laboratory Onoue (1988) suggested the simplified formula given
consolidation tests. The design prediction analysis was below for the average degree of consolidation with respect
carried out for various Ch /Cv ratios. to radial flow, Ur . The equation uses the Yoshikuni and
Nakanodo (1974) well resistance coefficient L, defined later.

Equivalent thickness to calculate consolidation
8Tr

Ur 1
exp (9)
caused by vertical flow F n 0:8L
As the marine clay consists of several layers, the coeffi- n2 3n2
1
cients of consolidation due to vertical flow vary along the F n log e n
(10)
n2
1 4n2
profile. As the conventional method allows only a single
value of Cv , the equivalent thickness of the marine clay was where n is the drain spacing ratio and
calculated applying the following equations to enable the
equivalent thickness, equivalent drainage length and as- de
n (11)
sumed coefficient of vertical consolidation to be used as dw
input values (Bo et al., 2003; Bo and Choa, 2004; Arulrajah, where dw is the diameter of the vertical drain.
2005). Carrillo (1942) derived an expression for the average
The equivalent thickness of layer 1, H91 , is given by degree of consolidation for combined vertical and radial

0:5 water flow, Uvr :
Cvi
H91 H1 (1)
Cv1 1
Uvr 1
Uv 1
Ur (12)
The time rate of settlement St for vertical drains can be
where Cvi is an assumed average value, and Cv n is Cv for calculated at any particular time for the various surcharge
layer n. heights by
The total equivalent thickness of all layers, H9Ti , is given
by St Sult Uvr (13)

H9Ti H91 H92 H93 . . . H9n (2)

Well resistance
The equivalent drainage length, Hdri , for double drainage The relevant features for the design and performance of
is given by vertical drains are their hydraulic properties: the discharge
H9i capacity and their filter permeability. If, during the consoli-
Hdri (3) dation period, the discharge capacity of the drain is reached,
2
the overall consolidation process is retarded. In such cases
and for single drainage is given by the drains present resistance to the water flowing in them.
Hdri H9i (4) In the calculations, the well resistance parameter L, as
developed by Yoshikuni and Nakanodo (1974), has been
used.
Vertical and radial consolidation It is given by
As the consolidation process with PVDs involves both


2
vertical and radial flows, each process was first calculated 32 kh Hdri
L (14)
separately. The equations applicable for the calculations of 2 kw dw
the time rate of settlement due to vertical flow are as
follows. where kh is the horizontal hydraulic conductivity of the soil,
The time factor for consolidation by vertical drainage, Tv , and kw is the coefficient of longitudinal permeability of the
is given by drain.
Cvi t
Tv (5)
H2dri Smear effect
It is often assumed that the installation of a drain does not
The average degree of consolidation with respect to change the properties of the surrounding soil. In reality,
vertical flow, Uv , is given by however, drain installation disturbs the soil to a large extent,
: depending on the soils sensitivity and macro-fabric (Rowe,
4Tv =0 5
Uv h i: (6) 1972). Barron (1948) and Hansbo (1979) analysed the effect
: 0 179
1 4Tv =2 8 of this soil disturbance by assuming an annulus of smeared
clay around the drain. Within this annulus, of diameter ds ,
The solution for radial water flow towards the central the remoulded soil has a lower coefficient of permeability,
drain goes back to Rendulic (1936). The result is generally kr , than the kh of the undisturbed clay. This leads to a new
expressed in terms of the average consolidation ratio for boundary condition between the undisturbed zone and the
radial drainage, Ur . smeared annulus, and this affects the solution by changing
The time factor for consolidation by horizontal drainage, the drain factor F (n) defined earlier as follows.
Tr , is given by The drain factor with smear effect, Fs (n) is given by



Ch t n kh
Tr (7) Fs n loge
0:75 loge s (15)
d2e s kr
71
M. W. Bo et al.

Time: months
where kh /kr is assumed to be 2; s is the smear zone ratio 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
ds /dw ; and ds is the diameter of the smeared zone. 0
Ch 5 2 Cv
10 Ch 5 3 Cv
Ch 5 4 Cv
20

Degree of consolidation, U: %
30
Predictions of magnitude and time rate of 40
settlement with PVDs
50
Land was reclaimed 1 m above high tide level to +4 mCD
60
by hydraulic filling. PVDs were installed at +4 mCD level.
70 Surcharge
After installation of the PVDs, surcharge was placed to period 5 20 months
+10 mCD by hydraulic filling. Fig. 5 shows the design 80
construction sequence at the case study area (vertical drain
90
area).
The increase in groundwater level after placement of the 100
surcharge over the area is expected, owing to a high
infiltration rate and the volume of water released from the Fig. 7. Design curves for vertical drain area
underlying soil during consolidation.
The predictions were made in advance of the field
installation of PVDs and the surcharge placement works by
applying Terzaghi (1925) and Carrillo (1942). Curves were Performance monitoring and
plotted for various Ch /Cv ratios. For the design predictions, assessment
the coefficient due to horizontal flow, Ch , was assumed to be
twice that of Cv : that is, Ch 1 and 2 m2 /year for the upper In order to monitor the performance of ground improve-
and lower marine clay respectively. Figs 6 and 7 show ment and to validate the efficiency of the prefabricated
predicted curves for settlement and degree of consolidation vertical drain system several geotechnical instruments were
generated from the design of vertical drains with preloading installed to monitor the degree of consolidation, both in the
for the vertical drain area. area with PVDs and in the area without PVDs, as a control
area. Settlement gauges, including deep settlement gauges,
were installed at the top of each sublayer, whereas piezo-
meters were installed at the centre of each compressible
12 layer in order to monitor the settlement and pore pressure
Surcharge period 5 0 months Surcharge period 5 20 months
dissipation. Details of the arrangement of the instruments
10 are shown in Fig. 8.
8
Settlement and pore pressure were monitored at close
intervals in the first three months, and at wider intervals
Elevation: mCD

Install vertical
6 drains
during the later part of monitoring. Typical monitoring data
are shown, together with the construction stages, in Fig. 9.
4
The degree of consolidation in terms of pore pressure was
Assumed groundwater level
2 also determined from the isochrones of the pore pressure
dissipation data, as shown in Fig. 9.
0 The more accurate ultimate settlements were predicted
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
Time: months using the field settlement results by applying the Asaoka
22
(1978) and hyperbolic methods (Tan, 1993). Fig. 10 presents
24 the results of the analysis by applying these methods, from
which the degree of consolidation in terms of settlement was
Fig. 5. Design construction sequence at vertical drain location A2S-6 determined. Both methods can predict the ultimate settle-
ment more accurately when the degree of consolidation
becomes greater than 60%.
Time: months
The Asaoka method generally plots the settlement at time
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 t against the settlement at time t
1 with the same scales on
0
C h 5 2 Cv the x and y axes. When the settlement at time t is equal to
02 C h 5 3 Cv
04 C h 5 4 Cv
that at time t
1, it is considered that the ultimate
06 settlement has been achieved. This can be predicted by
08 determining the intercept of the plotted trend line with a 458
10
line by extrapolating.
12
Settlement: m

14 The hyperbolic method plots the settlement against

16 the time divided by the settlement. The reciprocal of the
18 gradient of the linear best-match line is defined as the
20 Surcharge
period 5 20 months ultimate settlement. Details of these methods can be found
22
24 in Bo and Choa (2004).
26 Table 1 presents a comparison of the design prediction
28 results with that determined from the field instrumentation
30
data at the vertical drain area, 20 months after surcharge.
32
The field instrumentation results and assessment of the
Fig. 6. Timesettlement curves predicted for different Ch for vertical drain vertical drain area has been discussed previously by
area Arulrajah (2005) and Arulrajah et al. (2004a). The results of

72
 Preloading and prefabricated vertical drains for foreshore land reclamation

Control area Vertical drain area

FT22: prior to reclamation

FT29: after ground improvement FT28 (A2S26): after ground improvement
PB288
PB239
PB287
DS SP204 SP 095 WS220
PP WS204 DS PP
110 mCD
Surcharge
14 mCD

Sandfill

2410 DS 093
2630 PP 076
2744 PP 20
Upper marine clay
21080 DS 106 2950 PP 077
21144 PP 21
21444 DS 012 21270 PP 078
21480 DS 107
21724 DS 013 21594 PP 22
21944 DS 014
21844 PP 23 21740 PP 079

22200 DS 108
22244 PP 59 Intermediate 22244 PP 080
stiff clay
22874 DS 015 22644 PP 24 22710 DS 109
22750 PP 081

Lower marine clay

23384 DS 016 23144 PP 25 23200 PP 082
23410 DS 110

Old alluvium

Piezometers
Deep settlement gauges

Fig. 8. Cross-sectional soil profile showing instrument elevations

12 Vertical A2S-6
Borehole PB- Instrument 10 mCD
Fill elevation: m CD

drain Excess pore pressure-time A2S-5

Original 8 installation excess pore pressure: Kpa Settlement: m
E.L (m CD)
seabed 329m CD
Water level 0 40 80 120 160 200 0 05 1 15 2 25
Fine to medium DS-93 4
5 41m CD 5
Total additional load

sand
PP-76
0
Very soft
PP-77
Submerged additional load

marine clay
991m CD 4
10 with some 10
sea shell
DS-106 0 04 08 12 16 2 24
fragments
109m CD
DS-107 Time: Yr
15 148m CD A2S-6: Construction sequence 15
Elevation: m CD

Soft silty clay

PP-79
Firm silty clay 178m CD 30
Piezometric elevation:

DS-108 PP-076
20 20m CD 25 PP-077
PP-079
20
PP-80
15 m 15 m drains PP-080
Soft marine 229m CD
20 PP-081
PP-082
m CD

clay Water level

25 DS-109
15 25
271m CD
10
PP-81
278m CD 5
30 Soft to firm 30 7-Apr-95 7-Apr-95
silty clay PP-82 0 7-Oct-95 7-Oct-95
3208m CD 0 04 08 12 16 2 24 21-Sep-96 21-Sep-96
Clayey sand
35 Silty sand
DS-110 Time: Yr 35
341m CD Date of surcharge: 30th Dec 1994 Date of surcharge: 30th Dec 1994
Dense silty sand Date of Analysis: 16th Sep 1996 Date of Analysis: 23rd Sep 1996
Dense clayey Time vs. piezometric level
sand
40 40

Fig. 9. Instrument monitoring data at vertical drain area: (a) construction sequence; (b) time against piezometric level; (c) excess pore pressure against time;
(d) settlement

the design predictions are found to be in good agreement applying the hyperbolic method and 80.0% using piezometer
with that of the field instrumentation results. The degree of data.
consolidation, based on the ultimate settlement predicted by In order to be able to assess the improvement of soil in
applying the Terzaghi theory, is found to be in good terms of effective stress gain and strength gain, post-
agreement with that determined from the field instrumenta- improvement borehole and in situ tests were carried out in
tion data. both the soil improvement area and the control area when
For the vertical drain area, a degree of consolidation of the geotechnical instruments indicated that the required
83.3% was obtained based on the design predicted ultimate degree of consolidation had been achieved. Post-improve-
settlement as compared with 80.1% based on that predicted ment tests were carried out more or less at the same location
applying the Asaoka method, 80.0% based on that predicted where, prior to reclamation, borehole and in situ tests were

73
M. W. Bo et al.

Borehole PB239 Instrument 12 Vertical drain 110 mCD

Fill elevation: mCD

installation
Original 8
seabed 2329 Water level
4
Fine to DS293
25 medium sand 241 mCD 0

24
Very soft 0 05 10 15 20 25
marine clay Time: years
210 with some DS2106 (a)
seashell
2109 mCD
fragments
DS2107 Time: years
215 Soft silty 0 05 10 15 20 25
2148 mCD 000
clay DS2110
050 DS2109

Settlement: m
DS2108
Firm silty clay DS2107
100
Elevation: mCD

DS2108 DS2106
220 150
220 mCD 200
DS293
250 SP295
Soft marine 300
clay (b)
225
DS2109
2271 mCD

230 Soft to firm

silty clay
2500 DS2110
Time/settlement:

DS2110
Clayey sand 2000 DS2109
235 2341 mCD
days/m

Silty sand 1500

DS2108
Dense silty sand 1000 DS2107
Dense clayey 500
DS2106
DS293
sand SP295
240 0
0 100 200 300 400 500 600 700 800 900
Time: days
(c)

35 35 35 35
DS293 DS2106 DS2107 30
30 30 30 DS2108
Settlement at t: m

Settlement at t: m
Settlement at t: m

Settlement at t: m

25 25 25 25
20 20 20 20
15 15 15 15
10 10 10 10
05 05 05 05
0 0 0 0
0 05101520253035 0 05 10 15 20 25 30 35 0 05 10 15 20 25 30 35 0 05 10 15 20 25 30 35
Settlement at (t 2 1): m Settlement at (t 2 1): m Settlement at (t 2 1): m Settlement at (t 2 1): m
Date of surcharge: 30 Dec 1994
Date of analysis: 16 Sep 1996
(d)

Fig. 10. Ultimate settlement prediction by hyperbolic and Asaoka methods at vertical drain location (A2S-6): (a) construction sequence; (b) settlement
against time; (c) hyperbolic method; (d) Asaoka method

Table 1. Comparison between design (Ch 2Cv ) and Asaoka, hyperbolic and piezometer methods at vertical drain area, 20 months after surcharge

Sub-area Comparison Design Asaoka Hyperbolic Piezometer

Vertical drain Ultimate settlement: m 3.005 3.000 3.005

1.5 3 1.5 m
Settlement to date: m 2.504 2.404 2.404

Degree of consolidation, U: % 83.3 80.1 80.0 80.0

carried out. Note that, in the figures, PB39 is the reference tion tests, self-boring pressuremeter tests and dilatometer
borehole for the vertical drain area (A2S-6) and was bored tests. Collected samples were tested for physical strength,
prior to reclamation, whereas PB-87 is the reference borehole consolidation characteristics, and preconsolidation pressure
for the same area (A2S-6), bored out after reclamation and measurements.
ground improvement. The locations of these boreholes are Figure 11 presents a comparison of the undrained shear
shown in Fig. 8. strengths prior to and after ground improvement based on
Post-improvement in situ tests consist of boring and various in situ test methods and laboratory tests in the case
undisturbed sampling, field vane shear tests, cone penetra- study area. Fig. 12 presents the degree of consolidation

74
 Preloading and prefabricated vertical drains for foreshore land reclamation

After soil improvement Field vane: Cone penetraton test: Self boring Dilatometer:
Prior to reclamation pressure/meter: kN/m2
with vertical drain kN/m2 kN/m2 kN/m2
PB-39 (Seabed 329 m CD) 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
Fine to medium
5 sand
PB-87 (671 m CD)
Very soft
marine clay
with some Soft to firm
10 sea shell marine clay
fragments

Firm to stiff
15 Soft silty clay marine clay
Elevation: m CD

Firm silty clay

20
Firm
marine clay
Soft marine
clay
25

Soft to firm
30 silty clay
Firm to stiff
with traces of
marine clay
organic matters
Clayey silty sand Date of Surcharge:
35 Silty sand Silty sand 30th Dec 1994
Dense silty sand Dense silty sand
Dense clayey Dense clayey Date of Assessment Test: Date of Assessment Test Date of Assessment Test Date of Assessment Test
sand sand
40 4 Jun 1996 6th Aug 1996 9th Nov 1996 12th Nov 1996

Fig. 11. Comparisons of pre- and post-improvement shear strengths from various in situ tests at vertical drain location

interpreted from instrument monitoring data and in situ different magnitudes (Bo et al., 1997). Different preconsolida-
tests for various sub-layers in the vertical drain area. Details tion pressures also result from variation in the scale used in
of the methods for determining undrained shear strength, the elog p9 curve, even when the same method is applied
effective stress gain, degree of consolidation and overconso- (Mikasa, 1995). These effects are discussed in detail in Bo et
lidation ratio from laboratory and in situ tests have been al. (2003) and Bo and Choa (2004).
discussed by Bo et al. (1997, 1998, 2003) and Bo and Choa
(2004), and are not repeated here.
However, the preconsolidation pressure and undrained Conclusions
shear strength from laboratory tests were found to be
underestimated, owing to limitations on the test results. It is This paper discusses briefly the various theories of radial
well accepted that the determination of preconsolidation and vertical drainage, considerations and design method-
pressure is affected by the duration of loading, rate of ologies for the ground treatment of marine clay with
loading, rate of strain and temperature applied during the prefabricated vertical drains in an offshore reclamation
test (Leroueil et al., 1985). The effect of salt content on the e project. The design approach used has been verified and
log p9 curve has also been explained in an earlier section. It substantiated in comparison with field instrumentation and
has also been reported that the preconsolidation pressure in situ tests results at a case study site. The results of the
determined by applying different methods gives slightly design predictions are found to be in good agreement with

Borehole PB-39 Instrument Piezometer Deep settlement gauge Deep settlement gauge CPT holding
(Hyperbolic) (Asaoka)
Original
E.L. (m CD)
Seabed 329 100% 80% 60% 40% 20% 0% 100% 80% 60% 40% 20% 0% 100% 80% 60% 40% 20% 0% 100% 80% 60% 40% 20% 0%
Fine to medium DS-93
5 sand 41 m CD 5
PP-76
Very soft 692 m CD
marine clay PP-77
10 with some 991 m CD 10
sea shell DS-106
fragments 109 m CD
DS-107
15 Soft silty clay 148 m CD 15
PP-79
Elevation: m CD

Firm silty clay 178 m CD

20 20
PP-80
Soft marine
229 m CD
clay
25 DS-109 25
271 m CD
PP-81
278 m CD
30 Soft to firm 30
silty clay PP-82
3208 m CD
Clayey sand Date of Surcharge:
35 Silty sand 35 30th Dec 1994
Dense silty sand
Date of Analysis: Date of Analysis: Date of Test:
Dense clayey
sand Date of Analysis: 16th Aug 1996 16th Sep 1996 30th Nov 1996
40 40
21st Aug 1996

Fig. 12. Comparison of degree of consolidation from different instruments at vertical drain area

75
M. W. Bo et al.

that of the field instrumentation results. The degree of theory of consolidation of soils. Journal of Mathematics and
consolidation obtained by the design predictions is found to Physics, 21, No. 1, 15.
be only slightly higher than that of the field instrumentation Chu J., Bo M. W., Chang M. F. and Choa V. (2002) The
consolidation and permeability properties of Singapore marine
results. This indicates that the design approach used is valid
clay. Geotechnical and Geoenvironmental Engineering Journal,
for the design of prefabricated vertical drains installed in ASCE, 128, No. 9, 724732.
soft marine clay. Esrig M. I. (1968) Pore pressure, consolidation and electronics.
Journal of the Soil Mechanics and Foundation Division, ASCE, 94,
No. SM4, 899921.
Hansbo S. (1979) Consolidation of clay by band-shaped prefabri-
References cated drains. Ground Engineering, 12, No. 5, 1625.
Hausmann M. R. (1990) Engineering Principles of Ground Modification.
Arulrajah A. (2005) Field Measurements and Back-Analysis of Marine
McGraw-Hill, Singapore, pp. 251265.
Clay Geotechnical Characteristics under Reclamation Fills. PhD
Leroueil S., Kabbaj M., Tavenas F. and Bouchard R. (1985)
thesis, Curtin University of Technology, Perth, Australia.
Stressstrain rate relation for the compressibility of sensitive
Arulrajah A., Nikraz H. and Bo M. W. (2004a) Field instrumenta-
clays. Geotechnique, 35, No. 2, 159180.
tion assessment of offshore land reclamation works. Journal of
Mikasa M. (1995) Four further basic questions on consolidation. In
the International Society of Offshore and Polar Engineering, 14, No.
Compression and Consolidation of Clayey Soils (eds H. Yoshikuni
4, 315320.
and O. Kusakabe). Balkema, Rotterdam, pp. 10971098.
Arulrajah A., Nikraz H. and Bo M. W. (2004b) Observational
Mitchell J. K. (1981) Soil improvement: state of the art report.
method of assessing improvement of marine clay. Ground
Proceedings of the 10th International Conference on Soil Mechanics
Improvement, 8, No. 4, 151169.
and Foundation Engineering, Stockholm, 4, 509565.
Asaoka A. (1978) Observational procedure of settlement prediction.
Onoue A. (1988) Consolidation by vertical drains taking well
Soil and Foundations, 18, No. 4, 87101.
resistance and smear into consideration. Soil and Foundations, 28,
Barron P. (1948) Consolidation of fine grained soils by drain wells.
No. 4, 165174.
Transactions of the ASCE, 113, No. 2346, 718754.
Rendulic L. (1936) Porenziffer und Porenwasserdruck in Tonen.
Bjerrum L., Moum J. and Eide O. (1967) Application of electro-
Bauingenieur, 17, No. 51/52, 559564.
osmosis to a foundation problem in a Norwegian quick clay.
Rowe P. W. (1972) The relevance of soil fabric to site investigation
Geotechnique, 17, No. 3, 214235.
practice. Geotechnique, 22, No. 2, 195300.
Bo M. W., Arulrajah, A. and Choa V. (1997) Assessment of degree
Tan S. A. (1993) Ultimate settlement by hyperbolic plot for clays
of consolidation in soil improvement project Proceedings of
with vertical drains. Journal of Geotechnical Engineering, ASCE,
the International Conference on Ground Improvement Techniques,
119, No. 5, 950956.
Macau, pp. 7180
Terzaghi K. (1925) Der grundbruch an stauwerken and seine
Bo M. W., Arulrajah A., Choa V. and Chang M. F. (1998a) Site
verhutung. Die Wasserkraft, 17, No. 24, 445449: reprinted in
characterization for a land reclamation project at Changi in
From Theory to Practice in Soil Mechanics, John Wiley & Sons,
Singapore. In Geotechnical Site Characterisation (eds P. K. Robert-
New York, pp. 114118.
son and P. W. Mayne). Balkema, Rotterdam, pp. 333340.
Wade M. H. (1976) Slope stability by electro-osmosis. Proceedings of
Bo M. W., Cao L. F., Chu J. and Choa V. (1998b) One-dimensional
the 29th Canadian Geotechnical Conference, Vancouver, Section 10,
consolidation tests on Singapore marine clay at Changi. Proceed-
4466.
ings of the 13th Southeast Asian Geotechnical Conference, Taipei,
Yoshikuni H. and Nakanodo H. (1974) Consolidation of fine-
199206.
grained soils by drain well with filter permeability. Soils and
Bo M. W., Chu J., Low B. K. and Choa V. (2003) Soil Improvement:
Foundations, 14, No. 2, 3546.
Prefabricated Vertical Drain Techniques. Thomson Learning, Singa-
pore.
Bo M. W. and Choa V. (2004) Reclamation and Ground Improvement.
Thomson Learning, Singapore. Discussion contributions on this paper should reach the
Carrillo N. (1942) Simple two and three dimensional cases in the editor by 1 October 2007

76