Civil Engineering Forum Volume XX/1 - September 2011

STABILITY OF UNDERWATER STRUCTURE UNDER WAVE ATTACK

C. Paotonan
Engineering Faculty, Hasanuddin University, Makassar, INDONESIA
E-mail: patotonan_ch@yahoo.com
D. B. P. Allo
Civil Engineering Department, Cenderawasih University, Jayapura, INDONESIA
E-mail: daniel_bpa@yahoo.com

ABSTRACT

Geotube is, among others, a type of coastal structure that is increasingly accepted for coastal
protection especially underwater breakwater. Besides its relatively low cost, it has other advantages
such as flexibility, ease of construction and the fact that it can be filled with local sand material.
Similar to all other coastal structures, it should also be stable under wave attack. A simple theoretical
approach based on linear wave was adopted to estimate the stability of such structure. The theoretical
solution was then compared with an experimental study. The experimental study was conducted at the
Hydraulics and Hydrology Laboratory of Universitas Gadjah Mada. However, instead of a real
geotube, PVC pipe was used where the weight of the PVC was varied by adjusting the volume of sand
in the pipe. The result indicated that the agreement between the theoretical solution and the
experiment was encouraging. The analytical solution may be utilized to predict underwater pipe
stability under wave attack with certain degree of accuracy.
Keywords: Geotube, initial movement, submerged structure.

INTRODUCTION When used as a breakwater it has to be strong

Geotube is a type of coastal structure which can
be used as coastal protection. Basically geotube is
a geosyntetic type of material which is stitched to
form a tube when filled with sand or cement
material.
Geotube can be used as groin, jetty, or even
breakwater as long as the material is strong
enough against debris or sunlight. The advantage
of geotube is its flexibility size as normally the
size of geotextile is almost unlimited when
stitched to one another. It is also lightweight
which makes transportation easy. There are other
economic advantages that make the structure
increasingly acceptable.
enough to withstand wave force and current
(Pilarczyk, 1998, 2000). For this reason, the
paper discusses the stability of geotube
underwater breakwater.
Researches related to geotube as a coastal
structure have been carried out by many such as
Shin E.C, and Oh Y.I. (2007) who studied the
stability of Geotube based on two-dimensional
physical model. Paotonan et al (2011) analyzed
the geotube stability under sinusoidal wave
attack.
The use of geotube as coastal protection has been
realized in many parts of Indonesia and other
countries. El Dorado Royale Resor, Mexico is
one of the examples.

1219
Civil Engineering Forum Volume XX/1
X - Septem
mber 2011

Geo otube can also

a be fouund in Yucatan coast, B. Liftt Force
Meexico to prottect the erodded beach. The
T geotube When wave propaagates abovee the geotubbe, there
in El
E Dorado Royale
R Resoort, Mexico is a typical shouldd be differennt velocities from the topp to the
detached break kwater. Thee length of the coastal bottom
m of the geottube. In factt, the velocitty at the
linee protected by the sttructure is 4 km. In bottom
m of geotubee may be reg garded as zerro or no
Inddonesia, geootubes werre installed in many flow. The flow above the geotube may m be
locations, for example at a Lombangg coast of assumeed to follow
w the linear wave
w theory (Figure
Tirrtamaya resoort area near Indramayu, West Java. 1). Duue to such diifferential floow, there is vertical
Thee cross secttion of geotuube may bee shaped as force (lift
( force) accting on the geotube.
g
reqquired, thouggh it is mosttly rounded when fully
filleed with sandd. On the otther hand, thhe shape of
the geotube willl be close too ellipse when the sand
fill is not fulll. In this paaper, the geeotube was
assumed to be fully filled with
w sand an nd the shape
wass nearly rounnd.

THHEORETICA AL APPROA ACH

Thee forces actting on a geeotube are drag force,
inertia force, lift force, gravitatioonal force,
buooyancy forcee that is desccribed below
w.
Figure 1. Lift force on thhe geo tube
A. Gravitationaal force
The lifft force was given by Deean (1992) aas in this
Thee weight off the structu
ure in air (W
Wa) may be
equatioon
wriitten as:
D 2
Wa   s gVs (1) FL  C L  u (6)
2
Wh here ρs is geeotube speciific mass, Vs is volume Wheree u is velocitty at the topp of the struucture in
perr meter lenngth of geootube, and g is earth m/s, annd CL is a llift force coefficient. Liift force
graavitational accceleration. The unit off ρs, Vs, and coefficcient on cyliindrical boddy is 4.493 (Wilson
(
a kg/m3, m3/m, and m/s
g are m 2 respectivvely. If the and Reeid 1963). Inn shallow water,
w velocitty under
cro
oss section iss round with h diameter D in meter, linear wave
w is simpply
the weight of geotube
g is
D2 (2) (7)
Wa   s g  g
4 ua
hs
Thee buoyancy force (FB) that is respponsible for Wheree hs is the deepth at the tooe of the
the reduced weeight in the water
w can be written as: geotubbe and a is w
wave amplituude.

FB  gVs (3) Substittuting u in Equation 7, using Equuation 8

yields

alteernatively, foor cylindricaal geotube it reads CL

FL  ga 2 D / hs (8)
D2 2
FB  g (4)
4 or

Thee net force on

o geotube in the water is given by H2
Equ
uation 5. FL  C L g D / hs (9)
8
D2
W R   s   g (5) where H is wave hhigh.
4
Civil Engineering Forum Volume XX/1
X - Septem
mber 2011

C. Drag
D force In thiss case, the ggeotube is not
n stable when
w the
Thee drag force takes the form of total foorce due to bbuoyancy annd lift force is
i larger
than thhe weight of the structuree. The seconnd mode
D 2 is duee to the com mbined forcces of vertical and
FD  CD  u (10) horizonntal forces (drag forrce) as givven by
2
Equation 14.

Wh here C D is drag
d force cooefficient, which
w varies  D2 H2  (14)
from
m 1 to apprroximately 2.2
2 and is a function of f   s   g  C L g D / hs 
 4 8 
Keuulegan Carppenter (Dean and Dalrym mple, 1986).
2
Equuation 10, drag
d force is
i function of velocity H
 C D g D / hs
andd geotube diaameter. 8
Forr the case of submerg ged geotube, velocity
The cooefficient f is
i introduced d in Equatioon 14 as
disttribution caan be simpliified and drrawn as in
the friction factorr between thhe geotube and the
Fig
gure 2.
bottom m. The frictiion coefficieent f ranges from as
little as 0.0002 (onn steel) to ass high as 0.44 (loose
sand). In this experriment, the model
m was placed
p at
the boottom of thhe flume, whichw was slightly
corrodded steel. Therefore, in thhis matter thhe value
of f will
w be varieed to find out the corrrelation
betweeen the experimental dataa and the theeoretical
result. Equation 144, can be simmplified as:

  H2 H2
Figure 2. Sim
mplification of velocity
v distrib
bution on f  s  1D  fC
CL / hs  C D / hs (15)
submerged geotube
g    2 2

On Figure 2, diistribution of
o velocity inn triangular. or
Forr simplificatiion, averagee velocity is considered.
Theerefore, struucture is loccated in shaallow water   ffC L  C D 2
then equation 101 can be wrritten as: f  s D  H / hs (16)
   2
CD
FD  ga 2 D / hs (11) Since the
t drag force coincidess with the lifft force,
2 the seccond mode is more lik kely to occur. The
wave height
h requiired to destabilize the geotube
Or may bee written as:

H2 (12)    
FD  C D  g D / hs 2 f   s  1D hs
8
H     (17)
fC L  C D
D. Geotube inittial motion
Wh hen an objeect (geotube) is submerrged in the In nonn-dimensionaal parameterr form, Equaation 15
watter, there are
a two possible modess of initial can be written as:
mootion. The first
f is duee to verticaal force as
desscribed by Eq
quation 13.    
2 f   s  1 
D2 H2 H
     hs (18)
W R   s   g  C L g D / hs (13) D fC L  C D D
4 8
Civil Engineering Forum
Therefore, the non-dimensional parameters
relevant to the initial motion of the geotube are
 h
H/D, s  1 and s Equation 16 will be used to
 D
calculate minimum wave height needed for the
structure to start moving.
E. Physical Model Simulation
The experimental study was conducted in a
regular wave flume of 30 cm wide. PVC pipes of
different sizes that were closed at both ends were
used as geotube models. In order to vary the
weight of the model and hence its overall density,
the pipes were filled with water and sand of
different volumes. The diameter of the pipes was
13.02 cm (model I), 11.02 cm (model II) and
8.374 cm (Model III). The models are shown in
Figure 3.
Model I Model II Model III
Model I Model II Model III
Figure 3. Model of geotubes
Each model was tested under sinusoidal waves
attack. The test was initiated using small wave
amplitudes and was increased at small increment
to observe the initial motion of the geotube
models that indicate instability. The wave heights
were measured using wave probes. The geotube
model during the test is shown in Figure 4.
1222
Volume XX/1 - September 2011
Model
Figure 4. Testing models in the wave flume
The models were run for all (three geotubes)
models. The results, together with the analytical
approach, were given in the next section.
RESULTS AND DISCUSSION
The results of the experimental work are given in
Table 1 and Figure 5. It can be seen that the
agreement between the theoretical and the
experimental data is encouraging. However, the
real friction coefficient has to be determined for
better comparison. In this study, the friction
coefficients was tried and the best fit was chosen.
Hence, with such value of f (the best fit) only a
trend comparison between the experiment and the
theoretical solution can be made. The final trial
indicated that f = 0.00546 produced the best fit.
0.025
Data
Experiment
0.020
H (theory)
0.015
0.010
0.005
0.000
0.000 0.010 0.020 0.030
H (experiment wave height, m)
Figure 5. Theoretical vs Experimental wave height required
for geotube models initial motion
( f = 0.00546; CD = 2.200 and CL = 4.489)
Civil Engineering Forum Volume XX/1 - September 2011

Table 1. Experimental and theoretical minimum wave height for geotubes model stabilization
H
Model m V ρs D hs H (theoretical, m)
Specification (kg) (m3) (kg/m3) (m) (m) (experiment, m) with f = 0.00546
(-)
4.516 0.004 1.2600 0.130 0.250 0.0110 0.0110
100% of 4.516 0.004 1.2600 0.130 0.250 0.0110 0.0110
I
water 4.516 0.004 1.2600 0.130 0.250 0.0110 0.0110
4.516 0.004 1.2600 0.130 0.250 0.0110 0.0110
2.965 0.003 1.2600 0.110 0.250 0.0110 0.0110
100% of 2.965 0.003 1.2600 0.110 0.250 0.0110 0.0110
water 2.965 0.003 1.2600 0.110 0.250 0.0100 0.0110
2.965 0.003 1.2600 0.110 0.250 0.0100 0.0110
3.481 0.003 1.3660 0.110 0.250 0.0130 0.0120
75% of 3.481 0.003 1.3660 0.110 0.250 0.0120 0.0120
dry sand 3.481 0.003 1.3660 0.110 0.250 0.0120 0.0120
3.481 0.003 1.3660 0.110 0.250 0.0120 0.0120
II
4.627 0.003 1.8170 0.110 0.250 0.0180 0.0190
100% of 4.627 0.003 1.8170 0.110 0.250 0.0190 0.0190
dry sand 4.627 0.003 1.8170 0.110 0.250 0.0180 0.0190
4.627 0.003 1.8170 0.110 0.250 0.0190 0.0190
5.605 0.003 2.2000 0.110 0.250 0.0220 0.0230
100% of 5.605 0.003 2.2000 0.110 0.250 0.0220 0.0230
wet sand 5.605 0.003 2.2000 0.110 0.250 0.0220 0.0230
5.605 0.003 2.2000 0.110 0.250 0.0230 0.0230
1.845 0.001 1.2600 0.0840 0.250 0.0090 0.0090
100% of 1.845 0.001 1.2600 0.0840 0.250 0.0090 0.0090
III
water 1.845 0.001 1.2600 0.0840 0.250 0.0090 0.0090
1.845 0.001 1.2600 0.0840 0.250 0.0090 0.0090

Figure 5 shows that the trend of the experimental effect of water depth and geotube diameter ratio
wave heights that agree with the theoretical on H/D can be obtained. The influence of hs/D on
solution. Using regression technique a fitting H/D for a constant ρs/ρ can be seen in Figure 7.
curve as in Figure 5 indicates a slope of unity,
showing the best fit of the f value and indicates
the agreement of the experimental and theoretical 0.23
data trend. The parameter for the best agreement
was 0.00546; 2.200, and 4.493 for f, CD, and CL 0.20
respectively at the initial movement of a 0.17
submerged coastal structure. In terms of non
dimensional parameter, the results of the 0.14
H/D

Theorectical
experiment may be compared with the theoretical 0.11 Experiment Data
solution in Figure 6 and Figure 7.
0.08
Figure 6 shows that increasing value of (ρs/ρ-1) 0.05
causes the increasing value of H/D, which means
0.200 0.700 1.200
that when the value of specific gravity of
ρs/ρ-1
structure increases, the wave height required to
stir the structure is greater. If the geotube specific Figure 6. Relationship between ρs/ρ-1 and H/D
gravity and the water depth are constant, the

1223
Civil Engineering Forum Volume XX/1 - September 2011

0.12 0.60
moving
0.11 0.50
0.10 0.40
0.09 0.30
unmoving
H/D

H/D
0.08 0.20
Theorectical
0.07 Experiment Data 0.10
0.06 0.00
1.50 2.00 2.50 3.00 3.50 0.00 5.00 10.00 15.00 20.00
hs/D ((f(ρs/ρ-1)(hs/D)/(fCl+CD))0.5

Figure 7. Relationship between hs/D and H/D Figure 9. Geotube Initial movement graph

Again, Figure 7 shows that the trend between the Using Figure 9, the initial movement of
measurements in the laboratory and the submerged structure can be predicted. A situation
theoretical calculation is similar. The increasing that is represented by a point located above of the
value of hs/D causes the increasing H/D value. curve of Figure 9 is unstable and vice versa.
Using Equation16 the combination of non Suppose a geotube under wave attack of (ρs/ρ-
dimensional parameter that affects H/D is 1)(hs/D) = 5.0 and H/D = 0.4 it indicates that the
 s h structure is theoretically unstable while a geotube
  1 s where the correlation is indicated in
  D of (ρs/ρ-1)(hs/D) = 5.0 and H/D = 0.1 is
Figure 8. The line and scatter point in the Figure theoretically settled.
are theoretical line and data experiment,
respectively. CONCLUSIONS
Based on the explanation above, it can be
0.25 concluded into some points below. Firstly, the
stability of a submerged structure is influenced by
0.20 wave parameters (height and water depth) and
structure parameters; those are the specific
0.15
gravity and the diameter of structure, D.
H/D

0.10 Secondly, the wave height trend which is able to

move the structure obtained from experimental
0.05 Theoretical
data and calculation with Equation 15 are equal.
Experiment Data
Thirdly, as the value of structure specific gravity
0.00
increases, the wave height which is needed to
0.00 1.00 2.00 3.00
move the structure is increasing too. Or in other
(ρs/ρ-1)(hs/D) words as the specific gravity of structure
Figure 8. The combinative effects of ρs/ρ-1 and hs/D increases, the structure is more stable as well.
on H/D Fourthly, if the value of water depth and structure
diameter is getting greater, the effect on wave
Figure 8 shows the increasing value of (ρs/ρ- height and structure diameter ratio is getting
1)(hs/D) causes the increasing values of H/D. greater and follows the logarithmic trend.
Theoretical value of H/D on Figures 5 to 8, were Combination of non-dimensional parameter
calculated by assuming that f, CD, and CL values which influences the ratio of wave height value
were 0.00546; 2.200 and 4.498 respectively. and structure diameter is (ρs/ρ-1)(hs/D). As the
Using Equation 17, H/D values as a function of value of (ρs/ρ-1)(hs/D) increases, the value of H/D
geotube specific gravity and ratio of depth water increases too. The initial movement of geotube
to geotube diameter can be calculated where the can be predicted by using Equation 17 or Figure
result is presented in Figure 9. 9.
1224
Civil Engineering Forum Volume XX/1 - September 2011

ACKNOWLEDGMENTS Pilarczyk, K.W. (1999). Geosynhtetics and

The authors would like to thank Directorate
General of Higher Education (DIKTI) and
Hasanuddin University for financial support to do
the research. Invaluable comments, suggestions
and corrections by Prof. R. Triatmadja especially
with respect to the stability of geotube of the
original paper is highly appreciated. The authors
also express their sincere gratitude to the Head of
the Centre of Engineering Science Studies and
the Head of Hydraulics and Hydrology
Laboratory, Universitas Gadjah Mada, for the
facilities to support this research.
Geosystem in Hydraulic and Coastal
engineering, A. A Balkema, Rotterdam
(balkema@balkema.nl; www.balkema.nl).
Pilarczyk, K.W. (1998). Stability criteria for
geosystems—an overview. Proceedings of
the Sixth International Conference on
Geosynthetics, Atlanta, USA, vol. 2. pp.
1165–1172.
Pilarczyk, K.W. (2000). Geosynthetics and
Geosystems in Hydraulic and Coastal
Engineering. A.A. Balkema, Rotterdam.

REFERENCES
Dean, R.G. and Dalrymple, R.A. (1993). Water Shin E.C. and Oh Y.I. (2007). Coastal erosion
Wave Mechanic for Engineer and Scientist. prevention by geotextile tube technology.
World Scientific Publishing. Singapore. Geotextiles and Geomembranes 25 (2007)
264-277.
Koerner, G.R., Koerner, R.M. (2006). Geotextile
tube assessment using a hanging bag test. Sprague, C.J. (1995). P.E.T. Geotextile tube and
Geotextiles and Geomembranes 24 (2), containers for beneficial use of dredged
129–137. material. Contract Report for Bradley
Industrial Textiles, Inc., Valparaiso, FL,
Muthukumaran, A.E., Ilamparuthi, K. (2006). and Hoechst Celanese Corporation,
Laboratory studies on geotextile filters as Spundbond Business Group, Spatan.
used in geotextile tube dewatering.
Geotextiles and Geomembranes 24 (4),
210–219.
Oh, Y.I., and Shin, E,C. (2006). Using submerged
geotextile tubes in the protection of the E.
Korean shore. Coastal Engineering 53
(2006) 879–895, accepted 1 June 2006,
www.Sciencedirect.com.
Paotonan. C, Nur Yuwono, Radianta Triatmadja,
and Bambang Triatmodjo (2011).
Theoretical Approach of Geotextile Tube
Stability as a Submerged Coastal Structure.
Proceeding of International Seminar on
Water Related Risk Management, Jakarta.
Pilarczyk, K. W. (1995). Novel systems in coastal
engineering: Geotextile systems and other
methods, an overview. HYDRO pil Report,
Road and Hydraulic Engineering Division
of the Rijkswaterstaat, Delft, The
Netherlands.

1225
Civil Engineering Forum Volume XX/1 - September 2011

[this page intentionally left blank]

1226