Geosynthetics International, 2003, 10, No.

Technical Note
Analysis of geotextile tube behaviour by largescale eld model tests
E. C. Shin1 and Y. I. Oh2
1
Associate Professor, Department of Civil & Environmental System Engineering, University of
Incheon, 177 Dowha-Dong, Nam-Gu, Incheon, Republic of Korea, Telephone: +82 32 770 8466,
Telefax: +82 32 770 846, E-mail: ecshin@incheon.ac.kr
2
Post-doctoral Fellow, Department of Civil Engineering, California State University at Sacramento,
6000 J Street, Sacramento, California 95819 6023, USA, Telephone: +1 916 564 3836, E-mail:
sac52888@saclink.csus.edu

Received 19 August 2002, revised 1 August 2003, accepted 8 August 2003

ABSTRACT: This paper presents the geotechnical behaviour of geotextile tubes based on the results
of two eld model tests. The tests were performed on two types of dredged material (sand and silty
clay) at the Songdo land reclamation site in the Bay of Incheon, Republic of Korea. The dredged
materials were hydraulically pumped into the geotextile tubes by dredging ship. The shape variations
of geotextile tubes with the two dierent dredged materials are also presented. The eld test results
show that, when dredged sand is pumped in to ll the geotextile tube, the tube drops about 40% in
height within 2 days after completion of drainage. By contrast, if the clayey lling material is used,
after drainage the height of the geotextile tube reduces by approximately 50% within a month.
KEYWORDS: Geosynthetics, Dredged material, Field model test, Geotextile tube, Setting velocity,
Tube shape
REFERENCE: Shin, E. C. & Oh, Y. I. (2003). Analysis of geotextile tube behaviour by large-scale
eld model tests. Geosynthetics International, 10, No. 4, 134141

1. INTRODUCTION
Geotextile tubes have been used in the past for many
projects, but they are used mainly for ood and water
control. They are also used as prevention against beach
erosion, for shore protection, and for containment of
contaminated material. The geotextile sheets are permeable, yet soil-tight: hence excess water is drained from
the geotextile tube. Koerner and Welsh (1980) and
Pilarczyk (1990) provide overviews of the many applicationsprimarily erosion controlusing the various
types of geotextile container. Though construction of
erosion control structures using dredged materials in
geotextile containers is quite straightforward, underwater installation is often required, and is thoroughly
discussed by de Groot et al. (1994). Environmental
dredging and backll technology using geotextile tubes
were reported by Fowler and McLellan (1997), Pilarczyk
(1996) and Mori et al. (2002).
The most attractive features of geotextile tube technology are that it can use in-situ lling materials by
hydraulic pumping, it can be established economically,
and the construction process is faster than other tech134

nologies. The properties of the lling materials are very

important in the design and analysis of the behaviour of
geotextile tubes. Although coarse-grained material such
as sand and silty sand is a good material for tube
construction, ne-grained materials are frequently used
in eld conditions. At present most of the geotechnical
issues are focused on how to accurately predict the nal
tube height during and after construction of the tube,
especially in the case of tubes lled with ne-grained
material. This paper presents eld model test results of
geotextile tubes with dierent lling materials (dredged
sand, silty clay), which are compared with theoretical
analysis of shape variation.

2. METHODS FOR GEOTEXTILE TUBE

SHAPE VARIATION
2.1. Volume reduction approaches
Leshchinsky et al. (1996) proposed formulations for the
change of tube height caused by the consolidation
process, along with use of the computer program
GeoCoPS (ADAMA Engineering 1996). The governing
1072-6349 # 2003 Thomas Telford Ltd

135

Analysis of geotextile tube behaviour by large-scale eld model tests

mechanism of this approximation is that only the tubes
height changed during the consolidation process. Also,
the approximation option of GeoCoPS provides two
shape types (ellipse and rectangle), which are selected by
the magnitude of height drop, desired density, and nal
tube shape. Using the basic weightvolume relationships,
one obtains
Dh
De
e0
ef Gs w0
wf

h0
1 e0
1 e0
1 w0 Gs

where Dh and h0 are the decrease in height and the initial

height of the tube; Gs is the specic gravity of the solid;
e0 is the initial void ratio; De is the change in void ratio;
and w0 and wf are the initial and nal water content of
the ll material. It can be shown that


Gs
gslurry gw



w0 
2a
Gs gslurry gw
1


Gs
gsoil gw



wf 
Gs gsoil gw
1

2b

where gslurry and gsoil are the unit weight of the slurry and
the consolidated soil, respectively. By combining Equations 1 and 2 one can estimate the decrease of tube height
and the unit weight of materials in the tube.
2.2. Settling velocity approaches
The geotextile tubes can be lled with any material
capable of being transported hydraulically. As the soil
water mixture is pumped into a geotextile tube, there are
oating and suspension conditions due to the high
pumping pressure and the ow of the mixture. Generally, beach or river sand is the perfect material for
geotextile tube structures because of its immediate
settlement and rapid drainage. However, for ne-grained
material lls such as clayey and silty soils, suspended
solids settle by gravity: hence the concept of the settling
velocity approach, based on the settling and self-weight
consolidation mechanism. The US Army Corps of
Engineers have studied the dredged materials settling
process. They have proposed the typical settling process
diagram shown in Figure 1.

Zone settling

3. EXPERIMENTAL CONDITIONS
3.1. Materials
The geotextiles normally used to construct geotextile
tubes are either woven geotextiles or composite geotextiles (i.e. an external layer of woven geotextile and an
internal layer of nonwoven geotextile). For the present
study only one woven geotextile type was used. Its
physical properties are given in Table 1. The soils used
for this study were dredged sand and dredged silty clay.
Laboratory tests were conducted on dredged soils
collected from the Songdo land reclamation area. The
physical properties are given in Tables 2 and 3. The
dredged sand and silty clay were classied as SP, and OL
or ML respectively, according to the Unied Soil
Classication System (USCS). The particle-size distributions of the soils are shown in Figure 2.
3.2. Field model tests
The eld model tests were conducted for two dierent
types of ll material at the Songdo land reclamation site
in the Bay of Incheon, Korea. A single layer of geotextile
(Table 1) was used for fabricating the geotextile tubes.
Each tube had a circumference of 8.0 m, and was 25.0 m
long. It was lled with on-site material. The soil used to
ll the geotextile tube was hydraulically pumped from a
Table 1. Properties of geotextile tube
Property
Mass per unit area
Tensile strength
Elongation
Coecient of permeability
Eective opening size
(U.S. sieve)
Material

Test method

Unit

Geotextile

5.9
196
1050
10721074
100

ASTM D-5261 N/m

ASTM D-4595 kN/m
ASTM D-4595
%
ASTM D-4491 cm/s
ASTM D-4751

PET (polyester)

Table 2. Physical properties of dredged sand

Item

Quantity

Eective size, D10 (mm)

Specic gravity, Gs
Coecient of uniformity, Cu (%)
Coecient of curvature, Cc (%)
Max. dry unit weight, gdmax (kN/m3)
Optimum moisture content, wopt (%)
USCS

0.09
2.65
4.67
1.06
15.3
16.2
OL or ML

Depth to interface

Slope = zone settling velocity

Table 3. Physical properties of dredged silty clay

Item
Compression
settling

Natural water content, wn (%)

Specic gravity, Gs
Liquid limit, LL (%)
Plastic limit, PL (%)
Plastic index, PI
Passing 0.075 mm sieve (%)

Quantity
68.0
2.7
38.0
32.0
6.0
82.0

Time

Figure 1. Typical zone settling process

USCS

OL or ML

136

Shin and Oh
4.0 m intervals immediately below the nonwoven geotextile layer. A nonwoven geotextile was placed below
the tube to prevent erosion of the surrounding soil
during dissipation of water from inside. Starting with the
slurry pumping, observations of various features of the
lled tubes were made at various time intervals up to 3
months for the sand-lled tubes and 11 months for the
clay-lled tubes.

100
Dredged sand
Dredged silty clay

Percent finer (%)

80

60

40

20

4. EXPERIMENTAL OBSERVATIONS
0
10

0.1

0.01

4.1. Shape of geotextile tube

0.001

Figure 4 shows the variations of pumping time and

staged pumping steps with the heights of the geotextile
tubes. The pumping time of the tube lled with dredged
sand was less than 1 h, and the desired nal height was
attained after only one pumping process. However, for
the silty clay, the pumping time was more than 10 h,
because the process of soil particle settling and drainage
occurred slowly. Also, to obtain the desired height, three
stages of additional slurry pumping were performed.

Particle size (mm)

Figure 2. Particle-size distribution of dredged soil

dredging ship. During lling of the tube, the pumping

pressure and speed were varied to determine the
optimum values. A schematic diagram of the eld
model test and the placement of the pressure cells is
shown in Figure 3. The pressure cells were installed at
Inlet

Outlet

19 m
Dredging ship

Pressure cell

4m

4m

4m

4m

Geotextile (nonwoven)

0.5 m
3m
1.22 m

Geotextile (nonwoven)

2m

0.9 m

Figure 3. Schematic diagram of eld model test

1.6
Pumping complete: dredged sand
(last pumping height: 1.22 m)

Geotextile tube height (m)

1.4

1.2

1.0

Pumping complete: dredged silty clay

(last pumping height:1.2 m)

3rd pumping complete

Dredged sand
Dredged silty clay

2nd pumping complete

1st pumping complete

0.8

0.6

0.4
3rd drainage complete

0.2
1st drainage complete 2nd drainage complete

10!1

10!0

101

102

Elapsed time (h)

Figure 4. Hydraulic pumping steps with elapsed time

103

104

Analysis of geotextile tube behaviour by large-scale eld model tests

137

Pumping complete: 122 cm

After 2 h: 92 cm

After 24 h: 74 cm

Stabilized height: 68 cm

(a)

Pumping complete:120 cm
After 24 h: 98 cm
After 7 days: 66 cm

After 5 weeks: 61.5 cm

Stabilized height: 59.5 cm

(b)

Figure 5. Shape variation of geotextile tube with elapsed time: (a) dredged sand; (b) dredged silty clay

Figure 5 shows the variation of the general shape of the

geotextile tube from the time of completion of hydraulic
lling. The sand-lled tube stabilised after one day, but
the clay-lled tube took more than a week. As drainage
occurred, the height of the tube decreased, accompanied
by an increase in its width and in the area of contact
between the tube and the foundation. The increase in
width is due to spreading of the soil contained in the tube.
Figure 6 shows a plot of the decrease in height with nal
density ratio for geotextile tubes lled with sand and with
dredged silty clay. For the clay-lled tube (Figure 6b)
there are three experimental curves: one for the upper
portion, one for the lower portion, and one for the
average values. The theoretical curve of Dh/h0 against
gsoil/gw has also been plotted in this gure, using the initial
water contents and soil parameters. A comparison
between the theoretical and experimental plots shows
that, at high densities, the experimentally observed value
of Dh/h0 is signicantly less than that found from the
volume reduction approach, owing to deviation between
the theory and the real behaviour of the tube. Hence,
when the tube is lled with ne-grained material, a
dierent approach is needed. Also, the settling velocity
methods may have to consider the real settling and selfweight consolidation behaviour of the tube. Figure 7
shows a comparison of the decrease velocity of geotextile
tube height for dierent lling materials using the settling
velocity approach. Both sets of test results are presented
as a zone-settling process. The settling process of the
geotextile tube was divided into three stages, as illustrated

in Figure 7. The decrease in velocity of the sand-lled

geotextile tube is about twice that of the clay-lled tube,
indicating that the coarse-grained material settles faster
than the ne-grained material.
4.2 Settling characteristics of dredged silty clay
In this study, settling tests were performed to determine
the settling characteristics of silty clay with various
initial water contents. The tests were conducted by
settling column, as proposed by the US Army Corps of
Engineers (USACE 1987). The soil samples were
prepared with initial water contents of 233%, 400%
and 900%, respectively. The results were compared with
those obtained by large-scale eld model tests using the
settling velocity approach. Figure 8 shows the plot of
sedimentation height against logarithmic elapsed time
for each initial water content soil sample. It can be seen
that the higher concentration samples are more
representative of the zone-settling process. After the
zone-settling process, because of self-weight consolidation settling, the sediment height will increase slightly.
The settling velocity approaches are shown in Figure 9.
The settling velocity as used herein is dened as the
tangent intersection between the initial and at,
straighter portion of the depth to interface variation
curve with linear settling time. It can be seen from
Figure 9 that the initial settling time and velocity
decreases with increases in the initial water content. The
calculated settling velocities are shown in Table 4 for
settling tests and eld model tests. The settling velocities

138

Shin and Oh
100

Theoretical
Experimental

h/h0 (%)

80

60

40

20

0
1.2

1.3

1.4

1.5
1.6
(soil/w)final

1.7

1.9

1.8

(a)

100

h/h0 (%)

80

60

40

Theoretical
Experimental (upper)
Experimental (lower)
Experimental (average)

20

0
1.0

1.1

1.2

1.3

1.4
1.5
(soil/w)final

1.6

1.7

1.8

1.9

(b)

Figure 6. Drop in height of geotextile tube with density: (a) dredged sand; (b) dredged silty clay

from the eld model tests agreed well with those of the
settling tests.
The settling tests show that the settling time and
velocity, as important settling characteristics of negrained soils, are signicantly aected by the concentration of the slurry and clay content. Also, the settling time
and shape variation of geotextile tubes after pumping is
completed can be predicted by settling tests using various
initial water contents and other inuence factors (slurry
concentration and clay content).
Table 4. Settling velocity
Settling tests
Initial water content, w0 (%) 233
400
900
Settling velocity, vs (cm/min) 1.207 1.084 0.484

Field model test

650.6
0.671

5. CONCLUSIONS
This paper has presented various issues related to the
consolidation of geotextile tubes lled with dredged sand
and silty clay: the prediction of tube shape, and the
concept of a settling and self-weight consolidation. The
duration of the dewatering and consolidation period is
dependent on the type of geotextile used and also on the
type of ll material that is pumped into the tube. The
drop in height and shape variation of geotextile tubes is a
complex phenomenon and is dependent on many factors,
such as the density, initial water content, tensile strength
of geotextile, interaction of soil and geotextile. Therefore
the comprehensive volume reduction method deviates
widely between the theoretical prediction and the
observed behaviour of the geotextile tube. The shape
variation and consolidation behaviour of geotextile

139

Analysis of geotextile tube behaviour by large-scale eld model tests

0.1
Initial water content = 171%
0

Decrease of tube height (m)

0.1
0.2

Y=(0.0188)+0.79386*log10(X)
VD(decrease velocity)=0.79386 (cm/min)

0.3
0.4
0.5
0.6

Flocculation stage
about 1 h

0.7
0.01

Settling stage Consolidation stage

about 3 h

0.1

10

100

1000

10 000

Elasped time (h)

(a)

0
Initial water content=425.6%
0.1

Decrease of tube height (m)

0.2
Y=(0.25098)+0.40284*log10(X)
VD(decrease velocity)=0.40284 (cm/min)

0.3
0.4
0.5
0.6
0.7
0.8
0.01

Flocculation stage Settling stage Consolidation stage

about 4 days
about 7 h
0.1

10

100

1000

10 000

100 000

Elapsed time (h)

(b)

Figure 7. Decrease velocity of geotextile tube height: (a) dredged sand; (b) dredged silty clay

Sedimentation height (mm)

40

80

120

w0=233%
w0=400%
w0=900%

160

200
0.1

10

100
Time (min)

1000

10 000

Figure 8. Settling curve of dredged silty clay with elapsed time and initial water content

100 000

140

Shin and Oh
0

Depth to interface (cm)

2
4

Y=0.10105+1.20681*X
vs(settling velocity) = 1.20681 (cm/min)

6
8
10
12
14
16

18 Settling stage Consolidation stage

20

10

15

20

25

30

35

40

45

50

Settling time (min)

(a)

2
Y=0.19826+1.08452*X
vs(settling velocity) = 1.08452 (cm/min)

Depth to interface (cm)

Depth to interface (cm)

6
8
10

Settling stage Consolidation stage

Y=0.40217+0.48417*X
vs(settling velocity) = 0.48417 (cm/min)

12

Settling stage Consolidation stage

14
0

10

15

20

25

30

35

40

5
0

10

15

20

Settling time (min)

Settling time (min)

(b)

(c)

25

30

35

Figure 9. Settling velocity of dredged silty clay. Initial water content: (a) 233%; (b) 400%; (c) 900%

tubes lled with ne-grained material are highly aected

by the settling and self-weight consolidation processes.
Based on the comparison of eld and laboratory test
results, there is quite good agreement between laboratory test data and the observed consolidation behaviour.
Therefore the results of advanced settling tests can
roughly predict the behaviour of geotextile tubes.

NOTATIONS
Basic SI units are given in parentheses
Cc
Cu
D10, D30 and D60
e0
ef
Gs
h0
vD
vs
w0
wf

coecient of curvature
D230 =D10  D60 (dimensionless)
coecient of uniformity D60 =D10
(dimensionless)
equivalent particle diameters at
10%, 30% and 60% passing,
respectively
initial void ratio (dimensionless)
nal void ratio (dimensionless)
specic gravity of solid
(dimensionless)
initial height of tube (m)
decrease velocity (m/s)
settling velocity (m/s)
initial water content of ll material
(%)
nal water content of ll material
(%)

gslurry
gsoil
De
Dh

unit weight of slurry (N/m3)

unit weight of slurry and consolidated soil (N/m3)
change in void ratio (dimensionless)
decrease in height of tube (m)

REFERENCES
ADAMA Engineering (1996). GeoCoPS: Geotextile Tubes Contained
Pressurized Slurry, computer program, ADAMA Engineering,
Newark, Delaware.
ASTM D 4491. Standard Test Method for Water Permeability of
Geotextiles by Permittivity, American Society for Testing and
Materials, West Conshohocken, Pennsylvania, USA.
ASTM D 4595. Standard Test Method for Tensile Properties of
Geotextiles by the Wide Width Strip Method, American Society for
Testing and Materials, West Conshohocken, Pennsylvania, USA.
ASTM D 4751. Standard Test Method for Determining Apparent
Opening Size of a Geotextile, American Society for Testing and
Materials, West Conshohocken, Pennsylvania, USA.
ASTM D 5261. Standard Test Method for Mass Per Unit Area of
Geotextile, American Society for Testing and Materials, West
Conshohocken, Pennsylvania, USA.
Fowler, J. & McLellan, N. (1997). Geotextile tube development and
lling technique. Proceedings of the National Workshop on Geotextile
Tube Applications, Wetland Research Program Technical Report,
US Army Corps of Engineers, WRP-RE, 17, pp. 18.
de Groot, M. B., Bakker, K. J. & Verheij, H. J. (1994). Design
relationship for lters in bed protection. ASCE Journal of Hydraulic
Engineering, 120, No. 9, 10821087.
Koerner, R. M. & Welch, J. P. (1980). Construction and Geotechnical
Engineering Using Synthetic Fabrics, Wiley, New York, pp. 160229.
Leshchinsky, D., Leshchinsky, O., Ling, H. I. & Gilbert, P. A. (1996).
Geosynthetic tube for conning pressurized slurry: some design

Analysis of geotextile tube behaviour by large-scale eld model tests

aspects. ASCE Journal of Geotechnical Engineering, 122, No. 8, 682
690.
Mori, H., Miki, H. & Tsuneoka, N. (2002). The use of geo-tube method
to retard the migration of contaminants in dredged soil. Proceedings
of the 7th International Conference on Geosynthetics, Nice, France, 3,
pp. 10171020.
Pilarczyk, K.W. (1990). Coastal Protection, A. A. Balkema, Rotterdam.

141

Pilarczyk, K. W. (1996). Geosystems in hydraulic and coastal engineering: an overview. Proceedings of the 1st European Geosynthetics
Conference, Maastricht, the Netherlands, pp. 899906.
USACE (1987). Conned Disposal of Dredged Material, EM 11102
5027, Dept. of the Army, US Army Corps of Engineers,
Washington, DC.

The Editors welcome discussion on all papers published in Geosynthetics International. Please email your contribution to
discussion@geosynthetics-international.com by 15 August 2004.