Geotextiles and Geomembranes 29 (2011) 425e439

Contents lists available at ScienceDirect

Geotextiles and Geomembranes

journal homepage: www.elsevier.com/locate/geotexmem

Geotextile sand container shoreline protection systems: Design and application

W.P. Hornsey a, *, J.T. Carley b, I.R. Coghlan b, R.J. Cox b
a
Geofabrics Australasia (Pty) Ltd, Gold Coast, Australia
b
Water Research Laboratory, School of Civil and Environmental Engineering, University of New South Wales, Manly Vale, Australia

a r t i c l e i n f o a b s t r a c t

Article history: The use geotextile sand containers (GSCs) as shoreline protection systems, has grown moderately since
Received 4 January 2010 the first applications in the 1970s. This slow growth can be attributed to two factors; firstly, the lack of
Received in revised form understanding of coastal processes and design fundamentals by the larger geosyntheticcommunity in
24 October 2010
order to provide coastal engineers with suitable solutions, and secondly; there has been very little
Accepted 7 November 2010
Available online 4 March 2011
rigorous scientific wave flume testing with which to analyse the wave stability of geotextile sand
containers.
The application of geotextile containers in coastal protection works can be traced back to early works
Keywords:
Geotextile container
carried out in 1970s. The application of these types of structures was somewhat haphazard as very little
Coastal was understood about the wave stability and durability of the structures. Early wave stability work was
Wave stability carried out Ray (1977) and Jacobs (1983) with small containers, however, the testing programs were
Geotextile durability limited and did not provide sufficient confidence in the product to carry out exhaustive engineering
Shoreline protection design. As a result, the technology until recently has relied on manufacturers’ design suggestions based
on monitoring of actual structures. Over the past five years, coastal population pressure, extreme events
and concerns over climate change and sea level rise have resulted in more emphasis being placed on
shoreline protection systems. Geotextile manufacturers have responded to the challenges put forward by
design engineers and intensive research has been carried out in the field.
This paper outlines the current “state of the art” in terms of the design and specification of geotextile
sand containers (GSC). This paper covers the key issues which will ensure the long term integrity of
a geotextile shoreline protection system is maintained, these issues include:

Container stability;

Detailed analysis of recent large scale wave flume testing which assess filling capacity, size of
container, structure slope and scour protection etc.;

Container/geotextile durability;

Methods and specifications used to limit the effects of the fundamental factors affecting the life
span of geotextile containers such as vandal resistance, UV degradation and abrasion resistance etc.
Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.

1. Introduction The Australian trend in population movement is away from

The appeal of living or playing alongside the water, whether it is
the beach, river or canal has placed these amenities under
considerable pressure, which may be exacerbated by sea level rise,
climate change and extreme events. Issues such as maintaining
property boundaries while providing safe public access have
become an important issue for many stakeholders.
* Corresponding author.
E-mail addresses: w.hornsey@geofabrics.com.au (W.P. Hornsey), james.carley@
unsw.edu.au (J.T. Carley), i.coghlan@wrl.unsw.edu.au (I.R. Coghlan), r.cox@unsw.
edu.au (R.J. Cox).
inland towns or large cities towards coastal communities. The rate
of growth in coastal areas is 60% higher than the national average.
This phenomena known locally as ‘sea change’ has placed consid-
erable pressure on the valuable foreshore environment and
amenity. Where beaches were previously allowed to erode or
accrete naturally depending on the natural coastal processes,
ongoing development has placed artificial boundaries on the extent
to which the erosion can take place, resulting in the construction of
revetments (sea walls), groynes and other structures to protect this
development. Traditionally these structures have been constructed
using rock, concrete or wood, however, a combination of limited
access to suitable natural materials to construct these features
and the beach users’ demand for more user friendly materials has

0266-1144/$ e see front matter Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.geotexmem.2011.01.009
426 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439
led to the development of geotextile sand containers (GSC) for
shoreline protection.
The choice of which erosion control system to adopt, has
traditionally revolved around the use of rock and concrete, which
while structurally effective, is not currently considered environ-
mentally or user friendly. The demand for alternative solutions has
lead to the development of a number of innovative products, one of
which is sand filled geotextile containers. The use of geotextiles in
erosion control structures is not new, however, their use as the
primary defence against erosion has resulted in the development of
specialised materials, which can withstand the harsh conditions
experienced in exposed conditions. The use of sand filled geotextile
containers provides the designer with an alternative “soft” solution,
which provides effective erosion control whilst maintaining a user
friendly amenity.
This paper outlines the “state of the art” in terms of the design
and specification of geotextile sand containers (GSCs) and reviews
a number of significant Australian projects. The first large scale
projects using geotextile sand containers were carried out in
Australia in the late 1980s and early 1990s. This experience con-
sisted mainly of exposed structures constructed using small
diameter (1.2 m) dredge filled tubes. Projects such as Kirra Groyne
(1985) and Russell Heads Groyne (1993) (Restall et al., 2002) among
others provided valuable data on the strengths and limitations of
these types of structures. In the late 1990s and early 2000s the
method of manufacture of the geotextile containers was changed in
order to overcome the limitations which became apparent after
some years of monitoring of the aforementioned structures. The
tubes were replaced by dry filled smaller 0.75 m3 and 2.5 m3
Fig. 1. Typical sand filled geotextile container projects.
individually stacked containers, as the preferred method of
construction, which overcame issues associated with exposed
geotextile tube structures (Fig. 1).
2. Geotextile sand container (GSC) options
The early geotextile containers consisted predominantly of tubes
of varying length and circumference manufactured predominantly
from woven geotextiles (depending on the manufacturer). Tubes for
the purpose of this paper are defined as: hydraulically filled geo-
textile form which when filled have an elliptical cross section with
a final height greater than 1 m and length of 20 m or greater. The
tubes allowed rapid deployment over large areas while the woven
geotextile provided relatively high strength at reasonably low cost.
Individual containers on the other hand are smaller units manu-
factured predominantly from non-woven geotextiles, designed for
dry filling. Individual containers are defined as: a sand filled geo-
textile form which when filled will have an approximately rectan-
gular cross section with a final fill height of less than 1 m and length
of less than 3 m. Table 1 compares the two alternatives and rates
their performance characteristics.
3. Design
The design of geotextile shoreline protection systems can be
divided into two distinct sections:

Container stability (global/wave stability); and,

Geotextile durability.
 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439 427

Table 1 be noted that there will always be some variation on the final
Container type and geotextile comparison. dimension of the containers, however, the aim should be to fill the
Characteristics Individual containers Large tubes container to capacity. Reasons for variations in dimensions are
(up to 2.5 m3) (>8 m circumference) listed below:
Stability
Global (wave) stability þ þþ
Geotextile elongation e The inherent elongation (stretch)
Toe stability þþ þ
associated with staple fibre geotextiles means that if the
Interface friction þþ/ þþ/
(non-woven/woven) container is handled repeatedly during installation the geo-
Construction textile will stretch resulting in different dimensions to other
Sand pumping equipment þ þþ containers which have been handled less.
available
Fill material e The grading of the fill material will have an
<30 Tonne excavator þþ e
available
influence on the insitu density and final dimensions of the
Location container. A fine grained material will behave differently to
High pedestrian traffic þþ e a coarse grained fill material during filling and placement,
(damage potential) which will also depend on the moisture content. The fill
Back of beach þþ þþ
grading and moisture content will affect the dimensions.
(covered structure)
Groynes (public safety) þþ e
Filling and placement equipment e The filling and placement
Durability equipment used undoubtedly has the largest impact on the
UV stability (NW/W) þþ/þ þþ/þ final dimensions as the equipment controls the extent to which
Ease of repair (NW/W) þþ/ þþ/þ the container can be filled and the stress which can be exerted
Abrasion resistance (NW/W) þþ/ þþ/
Fines retention (NW/W) þþ/þ þþ/þ
on the geotextile.
Cost
Nonwoven (NW) geotextile þ þþ A detailed analysis of the full scale containers (0.75 m3 and
Woven (W) geotextile þþ þþþ 2.5 m3) was carried out by Blacka et al. (2006, 2007) (Fig. 2) details of
which are summarised in Table 2 and represented graphically in
Fig. 3.
It has long been recognized that Hudson’s formula (Hudson,
1953) was not an appropriate method to assess the stability of
geotextile container structures and that the absence of rigorous
testing had limited the extent to which these systems could be
applied. The latest research has concentrated on the development
of design curves for small (0.75 m3 and 2.5 m3) sand filled geo-
textile containers manufactured from staple fibre geotextile.

3.1. Global/Wave stability

To date the global stability information has been limited

primarily due to the significant costs associated with carrying out
large scale wave flume modelling and secondly a lack of coopera-
tion between installers and academics. Without accurate
measurement of full scale containers the data obtained from the
wave flume testing will be compromised. Detailed dimensioning of
containers as filled on site is required to ensure accurate scaling of
the containers and ultimately accurate forecasting of the perfor-
mance of the system.
In 2008 the Water Research Laboratory of the University of New
South Wales, based in Sydney, Australia, carried out detailed
analysis of the wave stability of two small container options,
nominal size 0.75 m3 and 2.5 m3. These specific containers sizes
were chosen as they represented the normal state of practice for
small geotextile containers used as shoreline protection systems in Fig. 2. Full scale dimensioning.
Australia over the past 10 years. The aim of the research was to
provide engineers with proven data which will allow designs to be Table 2
Field measured parameters.
carried with confidence. The research consisted of four key
components, namely: scaling, preliminary wave flume testing, Property (mean) 2.5 m3 (Nominal) 0.75 m3 (Nominal)
comprehensive flume testing, and development of design curves Container volume 2.76 m3 0.87 m3
and design methodology. Saturated GSC weight 4616 kg 1507 kg
Inferred dry weight 3585 kg 1154 kg
Saturated GSC bulk 1674 kg/m3 1729 kg/m3
3.1.1. Scaling
density
Scaling and accurate dimensioning of the containers was Inferred dry bulk 1302 kg/m3 1327 kg/m3
viewed as the key to providing representative results, and this was density
proven during the test program. In previous wave flume testing Measured dimensions 2.6 m (L)  1.9 m (W) 1.8 m (L)  1.5 m
(typical assumed  0.58 m (D) (W)  0.42 m (D)
there has been limited emphasis placed determining the fill
dimensions) (2.4 m  1.8 m  0.65 m) (1.6 m  1.2 m  0.4 m)
capacity and final dimensions of the full size containers. It should
428 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439

Table 4
Test programme variables.

Test condition Variable

GSC Size 0.75 m3 model
Revetment Structure Slope 1V:1.0H, 1V:1.5H, 1V:2H
GSC packing technique Stretcher bond, long axis aligned
with the shoreline
Number of GSC layers Single and double layer
Foreshore slope 1V:5H, 1V:10H, 1V:20H
Still water level 0.0 m MSL, þ1.5 m MSL, þ3.0 m MSL
(SWL rel Mean Sea Level)
Peak spectral wave period, Tp 5 s, 10 s, 15 s
Significant wave height 0.5 me2.0 m
at the structure, Hs

Fig. 3. Graphical representation of the container dimensions.

From this data the following scaling was accepted for the large
scale wave flume testing.
3.1.2. Initial wave flume testing
Scaling of the containers provides two important functions,
ensuring that the scale model containers are representative of full
size geosynthetic sand container, and that testing to failure of the
system is possible using the available equipment.
Various aspects need to be scaled to ensure representative
results, including dimensions, mass, density and frictional inter-
action. The first part of the research and development program
measured field representative geosynthetic sand containers Blacka
et al. (2006, 2007). The second stage was designing a container that
would fail when subjected to wave action in the two dimensional,
3 m wide wave flume. The wave flume testing was divided into
three phases, namely:

Phase 1 (Tests 1e11): Carried out to determine size of
containers required which would result in failure of the
structure. Testing using monochromatic waves was carried out
on model 0.75 m3 containers firstly at 1:4 geometric scale, then
1:8 geometric scale. Containers were placed with their long
axis perpendicular to wave attack, placed in a “stretcher bond”
fashion in a single layer up the slope.

Phase 2 (Tests 12e74): The second phase concentrated on the
characteristics of the geotextile and also the sand fill used
inside the containers to try and match prototype container
characteristics. In this phase the double layer structure was
introduced, it was noticed that at a 1:8 geometric scale failure
was not always induced, thus the final scale was reduced to
1:10 (for the 0.75 m3 containers) and 1:13 (for the 2.5 m3
containers). Model scaling was undertaken according to Frou-
dian similitude. See Table 3 for actual geosynthetic sand
container properties and model comparisons.

Phase 3 (Tests 75e101): The final phase of testing was carried
out using irregular waves with the various test program vari-
ables as defined in Table 4. Detailed video footage of all testing
was undertaken in order to accurately review the mode of
failure. The containers were also numbered to allow tracking of
movement of individual containers.
3.1.3. Test regime
As each site is different, long term structural performance
requires site specific conditions to be accounted for in the design.
To prevent limiting the design envelope produced in the program
to a small range of conditions or sites, a range of variables was
tested, including different wave conditions, GSC layers, foreshore
slopes and revetment slopes. This wide design envelope was
included in the test program to encompass most conditions likely
to be encountered worldwide.
Table 4 indicates the tested conditions on the 1:10 scale 0.75 m3
geosynthetic sand containers. Results are converted to 1:13 scale
for the 2.5 m3 container.

Table 3
Field and model comparisons.

Property 2.5 m3 Prototype 2.5 m3 Model (1:13 scale) 0.75 m3 Prototype 0.75 m3 Model (1:10 scale)
Dimensions (full scale units)
Length 2.6 m 2.15 m 1.8 m 1.65 m
Width 1.9 m 1.82 m 1.5 m 1.4 m
Depth 0.58 m 0.56 m 0.42 m 0.43 m
Volume 2.76 m3 2.22 m3 0.87 m3 1.01 m3
Mass
Saturated GSC weight 4616 kg 5577 kg 1507 kg 1628 kg
Dry weight in air after filling 3585 kg 2746 kg 1154 kg 1250 kg
Saturated GSC bulk density 1674 kg/m3 1612 kg/m3 1729 kg/m3 1612 kg/m3
Geotextile (model units
for model column)
Manufacturing process Nonwoven staple Nonwoven staple Nonwoven staple Nonwoven staple
fibre fibre fibre fibre
Thickness >5.9 mm >2.5 mm >5.0 mm >2.5 mm
Strength >30 kN/m 4 kN/m 18.5 kN/m 4 kN/m
Elongation at failure >70% >70% >70% >70%
Hydraulic conductivity 27 L/m2/s 300 L/m2/s 62 L/m2/s 300 L/m2/s
Friction coefficient 0.5e0.7 0.3 0.5e0.7 0.3
Fill material
Grain size (d50) 0.3 mm 0.22 mm 0.3 mm 0.22 mm
 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439 429

Fig. 4. Test configuration.

Another important aspect that needs to be identified is the

structural arrangement and construction methodology of the
model revetment. In addition to those aspects identified above, Fig. 5. 0.75 m3 container stability chart.
important design considerations should be representative in the
model including; toe scour protection, particularly important in the
dynamic coastal environment, and reducing as much as possible,
boundary effects from the testing apparatus. See Fig. 4 for the
typical arrangement, where AHD is Australian Height Datum,
which is approximately mean sea level.
Once the variables and structural profile were identified,
a failure classification of the structure needed to be identified. To be
comparable with existing systems, a similar failure classification to
interlocking concrete armour units (U.S. Army Corps of Engineers,
2006) was selected. This identifies specific failure modes depend-
ing on displacement or loss of armour units or in this case geo-
synthetic sand containers. See Table 5 for failure or structural
damage criteria.
The test program undertaken on each structure, with different
characteristics as identified in Table 4, consisted of both mono-
chromatic and irregular waves totalling 1100 per test. The test
duration of 1100 waves gave a probability of 99% that deep water
Fig. 6. 2.5 m3 container stability chart.
maximum wave height (Hmax) was approximately 1.87 times the
significant wave height at structure (Hs), thus giving the highest
confidence that the structure has been exposed to the highest
possible waves within the spectrum. It should be noted that these curves only apply to containers
The change in still water level, wave height and period enabled manufactured from the same geotextile, filled using the same
different types of waves to be impacted on the revetment, method, and which achieve similar fill capacities. The research
unbroken, broken and breaking waves, exposing the structure to all carried out was limited to a non-woven geotextile and the
types of conditions. container modelling was based on Australian practices.
An interesting outcome of the research was that unlike
3.1.4. Design curves and design methodology conventional rock revetments where the shallower the slope
The results from the testing program (77 tests on the final angle the greater the stability of the structure. Monitoring of the
structure) were used to generate a design guideline using the failure mechanisms during the testing confirmed the hypothesis
geosynthetic sand containers. The results are based on the “Initial put forward by Oumeraci et al. (2003) that failure is due to forward
Damage” criteria as discussed in the previous section (i.e. 0e2% on displacement of individual containers, this phenomenon is
the double layer system). controlled by contact surface area between containers, interface
The following charts (Figs. 5 and 6) represent the results, design friction of geotextiles and the vertical load applied i.e. the steaper
charts, from the irregular wave tests (1100 waves) on the structure the slope the higher the confining pressure. Similarly the signifi-
conducted with different revetment slopes, foreshore slopes, still cance of fill capacity of the container can be shown in Table 6 below
water levels and wave periods. The charts are based on the more where the Water Research Laboratory results Coghlan et al. (2008,
stable double layer geosynthetic sand container revetment 2009) and Oumeraci et al. (2003) results are plotted on the same
structure. axis. Oumeraci et al. (2003) used a nominal 80% fill value for the
container (based on German construction practice) rather than the
fill to capacity applied in the WRL results (based on Australian
construction practice).
Table 5
Structural damage classification (proportion of units displaced). When comparing the results of this test regime with the
Oumeraci et al. results it becomes clear that as the wave height and
Armour layers Slope No damage Initial Intermediate Failure
period increase, the stability of the 80% filled containers drops off
damage damage
significantly when compared to the containers filled to capacity.
Single 1:1e1:2 0% 0e1 % 1e10 % >10 %
Double 1:1e1:2 0% 0e2 % 2e15 % >15 %
Additional work examining the overall hydraulic permeability of
structures composed of sand filled geotextile containers were also
430 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439

Table 6
80% Fill capacity Oumeraci et al. (2003) vs. fill to capacity.

Revetment Peak wave Design significant wave height, Hs (m)

structure slope period Tp (s) 0.75m3 containers, 1:1 slope

Oumeraci et al. (2003) WRL Physical

stability equation
1V:1.0H 5 1.73
10 1.09
15 0.83
1V:1.5H 5 1.56
10 0.98
15 0.75
1V:2.0H 5 1.29
10 0.81
15 0.62
undertaken by Recio and Oumeraci (2008). This study examined
how the mode of placement of individual containers influenced the
permeability of an entire GSC structure. Three different packing
techniques were tested during their study, but these were not
directly comparable with those modelled by Coghlan et al. (2008,
2009). Recio and Oumeraci (2008) concluded that although
permeability and interlocking both contribute to the hydraulic
stability of GSC structures under wave attack, interlocking plays
a more important role than permeability. This observation, that
interlocking of individual containers is critical for stability, is
consistent with the findings of the Coghlan et al. (2008, 2009).
3.1.5. Sand retention
As shown above, the fill capacity of the container has a signifi-
cant impact on its stability, it is therefore critical that this fill
capacity be maintained for the life of the structure. One area often
overlooked by designers is the sand retention capacity of the geo-
textile which will influence the long term fill capacity of the
geosynthetic sand container.
Geosynthetic containers and the sand contained within are
exposed to the most aggressive flow conditions a geosynthetic filter
system is likely to encounter. The geotextile is likely to undergo
some deformation due to either installation or sand movement
during wave attack, which with some woven geotextiles alters the
pore size and retention capacity of the container. The high pore
water pressure loads exerted under wave impact combined with
sand movement within the container, which depends on the
direction of wave attack means that the traditional filter zone is
never developed. The entire body of sand within the container
remains effectively mobile during the life of the structure and this
sand is subjected to large suction forces during wave and tidal
drawdown.
In order to accurately assess the long term sand retention
capacity of the geotextile it is recommended that 95% retention be
achieved when tested according to the Hydrodynamic Test NFG
38-017-1989. The test method forces water though the geotextile
and sand in two directions which does not allow a stable filter zone
to form within the sand. Experience has shown that this test best
mimics the actual field conditions to which the container is likely to
subject to.
model Results
1.7
1.5
1.3
1.5
1.3 Fig. 7. Flexible toe container detail.
1.1
1.3
1.1 overcome this issue it is recommended that an additional scour
0.9 container be incorporated in the design which will prevent
undermining of the structure (refer Fig. 7).
The size of this toe container should be compatible with the wave
conditions expected at the toe of the structure. Weerakoon et al.
(2003) showed that using a toe protection container which was
too small led to damage and failure of the structure.
3.2. Durability
Geotextile sand containers are exposed to the elements for long
periods of time and unlike applications such as separation, where
they are expected to perform the filtration and reinforcing function
only until consolidation of the subgrade has occurred, they are
expected to perform the design functions for a number of years.
The key factors influencing the long term durability of the geo-
textile sand containers are discussed in the following sections.
3.2.1. Damage resistance
Exposed to the elements to a far higher degree than most other
geotextile applications, such as drainage and separation, means
damage to sand filled geotextile containers is unavoidable and
without question the key factor limiting the life of a geotextile
container structure. The type (cause) of damage which is likely to
occur can be categorised into two broad segments, namely:

Incidental (or accidental) damage (from driftwood or boat
impact) (Figs. 8 and 9);

Vandalism (knife cuts and punctures).
In both these cases limiting the extent and the amount of this
damage has a significant impact on the durability of the structure as
a whole. Experience has shown that a structure consisting of small
containers is far less likely to sustain a single catastrophic damage

3.1.6. Scour protection

As with all coastal structures, toe stability is critical in ensuring
the survival of a geotextile sand container structure. The toe of the
structure should be located at a level where it is unlikely to be
undermined. In most cases if the base of the structure is founded at
0 m LAT the structure will perform adequately, however, site
specific assessment is required. In extreme storm events large scale
erosion can occur at the toe of the structure and in order to Fig. 8. Potential damage from drift wood.
 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439 431

Fig. 10. Vandalised geotextile tubes.

Fig. 9. Potential damage from boats.

event and that the choice of geotextile used for the manufacture of
the containers has a marked influence over the long term surviv-
ability of the structure. Repairs can be effected to the containers,
however, experience has shown that clients are less likely to apply
complex patch methods. Two simple repair methods have been
developed for wet or dry applications to ensure the long term
durability of the structure.

3.2.1.1. Incidental damage. The relevance of incidental damage will

vary from site to site, but is a key factor when structures are con-
structed on river banks or adjacent to boat launching facilities. In this
application, limiting the amount of the damage is critical. This can be
achieved by choosing a geotextile which is less susceptible to
damage.
In both Australia and America, the survivability of a geotextile is
based on the % elongation i.e. the higher the elongation, the lower
the damage potential. In order to accurately compare the damage
resistance, Koerner (2005) suggests the use of impact energy as an
effective measure of the relative damage of various geotextiles.
Impact energy takes into account both strength (force) and elon-
gation (distance) which can be calculated using the CBR test method
which measures the force (N) and the elongation (mm) for a given
geotextile. Examples of impact energy of three common generic
geotextiles of the same CBR strength are provided in Table 7.
This methodology allows designers to carry out simple
comparisons of the relative damage resistance of the geotextiles
proposed for the manufacture of geotextile sand containers.
3.2.1.2. Vandalism. In Australia, vandalism (primarily knife cuts) is
the greatest threat to geosynthetic sand containers. Standard geo-
textiles (woven or non-woven) provide limited resistance to knife
attack and a number of high profile projects such as Kirra Groyne
(Gold Coast, Australia), and St Clair revetment (Dunedin, New
Zealand) (Fig. 10) and Rockingham (Western Australia) shown in
Fig. 11 have ultimately failed due to vandalism. Various coatings
have been trialled since the late 1990s to improve the vandal
resistance of the geotextiles, including polyurethane spray on
Table 7
Impact energy comparisons.
Geotextile CBR strength (N) Elongation (m) Impact energy (J)
Staple fibre 3800 0.070 266
Continuous filament 3800 0.055 209
Woven 3800 0.035 133
Fig. 11. Vandalised geotextile container.
coating, which provided limited improvement with considerable
cost implications.
In 2001 a composite geotextile (Fig. 12) was developed which
consisted of a standard inner geotextile with a coarse fibre geo-
textile bonded to the outer surface. The coarse outer layer allows
sand to be trapped within the geotextile and this trapped sand
provides further protection from knife cuts. It is important that the
outer layer is correctly bonded to the inner layer as poor bonding
will allow stripping of the outer layer and loss of protection to the
container.
At present there are no indicator tests available which model
puncturing or cutting of the containers with a knife or sharp
instrument, hence there is limited information available to engi-
neers to base a vandal resistance specification. One solution
proposed is to modify the current ASTM D4833-00 Rod Puncture
test to create a sharp point thereby mimicking a puncture with
a sharp instrument (Fig. 13). Testing carried out using this method
has shown that adding more geotextile mass increases the punc-
ture resistance moderately, however, the inclusion of sand in the
outer layer provides significant improvement, see Table 8.
The quantity of sand contained within the outer geotextile has
an impact on the ultimate puncture resistance of the geotextile.
That is, lower outer geotextile mass results in less sand capture
capacity and lower modified puncture resistance. Trials with
a range of outer geotextile masses have shown that a 900 g/m2
432 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439

Fig. 12. Composite Geotextile.

outer layer provides the best solution in terms of sand capture,

Table 8
protection and cost.
Sharp instrument puncture resistance.

3.2.2. UV degradation Product: Units Test show Geotextile Puncture

staple fibre method: modified mass increase resistance
Unlike other geotextile applications such as separation and
geotextile ASTM D4355 increase
drainage where the geotextile is buried and UV exposure is limited
600 g/m2 N 166 e e
to a maximum of 2e3 months, geotextile sand containers are 800 g/m2 N 215 33% 29%
expected to withstand the effects of UV Radiation for many years. 1200 g/m2 N 328 50% 53%
Projects such as the Russell Heads groyne (1993) and Stockton 2100 g/m2 N 419 75% 27%
2100 g/m2 þ N 718 75% 119%
3000 g/m2 sand

Beach revetment (1996) have proven the medium term perfor-

mance of the containers. The ultimate life of these containers in
terms of UV exposure is yet to be determined and improvements in
polymer technology are expected to continue to improve the
durability of the containers.
At present a designer must rely on accelerated UV degradation
to assess the long term performance of a geotextile sand container
structure. The current ASTM D4355 defines long term testing as the
strength retained after 500 h exposure, however, this definition is
based on the exposure for standard separation applications
described above. Extended accelerated testing has been carried out
on a range of staple fibre geotextiles used in Australia for the

Long Term UV Resistance (ASTM D4355)

100
Strength Retained (%)

80

60

40

20

0
0 1000 2000 3000 4000 5000 6000 7000
Exposure Duration (hrs)

Fig. 13. Modified rod puncture apparatus. Fig. 14. Long term accelerated UV exposure testing.
 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439 433

manufacture of geotextile sand containers, results of which are Minimum strength retention of 70% after 80,000 cycles is recom-
shown in Fig. 14. mended for geotextile sand containers exposed coastal
For geotextile sand containers which will be expected to be applications.
exposed and perform their design function for a number of years,
extended testing beyond the maximum 500 h to a minimum of 4. Applications
2000 h is recommended.
4.1. Stockton beach revetment
3.2.3. Abrasion resistance
Abrasion due to water borne sand, shell and coral fragments is 4.1.1. Project location
a threat unique to geotextile sand containers. This abrasion can be Stockton Beach is located to the north of the Hunter River
a significant factor for containers located in the nearshore surf trained entrance in Newcastle, Australia (Fig. 17).
zone where the movement of water and coarse material is
continually occurring, and is particularly noticeable in containers 4.1.2. Date constructed
located at sea bed level. Underestimating the effect of the abrasion 1996.
forces in coastal applications can limit the life of the individual
containers and ultimately the structure as a whole (Figs. 15 4.1.3. Principal
and 16). Newcastle City Council.
A number of textile abrasion test methods are available,
however, these are based on a modified Stoll abrasion test which is 4.1.4. Description
used to determine the abrasion resistance of women’s woven and 48 m long by 4.5 m high double layer sand container revetment,
knitted stockings. The test method which best mimics actual field at a 1.5H:1V slope (Fig. 18).
abrasion is the German Rotating Drum test method BAW Federal
Waterways Engineering and Research Institute (1994), which was 4.1.5. Project objectives
developed specifically for geotextiles used in coastal and waterway To provide temporary erosion protection to the surf life saving
applications. This test subjects the geotextile to 80,000 abrasion club (SLSC). Severe erosion to the beachfront at Stockton beach had
cycles with a mixture of water and fine gravel and measures the % placed the Stockton Beach SLSC in danger of collapse. Due to state
strength retained on completion of the test. government regulatory requirements, an interim measure was the
The relative abrasion resistance of various generic geotextile only rapid solution while a coastline management plan was
classes i.e. staple fibre, continuous filament or woven is noted. completed.

4.1.6. Site conditions

Exposed ocean beach with offshore significant wave heights of
up to 9 m during major storms. Depth limited waves influenced by
wave setup at the toe of the structure.

4.1.7. Requirements and constraints

The client required a temporary structure which would have
a minimum life span of 6 months. The material was required to be

Fig. 15. Abrasion at the toe of a near shore structure.

Fig. 17. Satellite view of Stockton beach.

Fig. 16. Abrasion at the base of a submerged structure. Fig. 18. Structure X section.
434 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439
removed easily with no debris remaining after removal. A user
friendly aesthetically pleasing structure was desirable as the
structure was used to protect a busy life saving club and would be
subjected to high pedestrian traffic.
4.1.8. Construction techniques
The empty geotextile containers were placed in a filling frame
and filled to capacity using an excavator, the container was then
sewn closed using a hand held sewing machine. The containers
were then lifted and placed using a modified rock grab. The rock
grab was modified in such a way as to limit the stress on the geo-
textile during the lifting operation. New equipment has since been
developed to simplify installation and reduce the risk of container
damage during construction.
The full containers measured 1.7 m  1.4 m  0.4 m when placed
into the structure and were laid in a stretcher bond format to
ensure maximum interlock at a 1.5H: 1.0V slope. The revetment
was constructed using the double layer configuration and incor-
porated a self-healing “Dutch Toe” to prevent undermining of the
toe of the structure during large storm events. The flexible toe
container matched the dimensions of the containers used in
the main revetment structure because this container would be the
most exposed to wave attack and hence needed to be designed to
withstand the design wave climate.
Table 9
Geotextile container characteristics.
Thickness CBR burst Tensile strength Seam strength
5.3 mm 7.0 kN @ 60% 42 kN/m XD Min. 80% of base fabric
27 kN/m MD
Fig. 19. Stockton beach revetment time line.
4.1.9. Container description
Nominal 0.75 m3 sand filled geotextile containers and Heavy
duty UV stabilised non-woven staple fibre needle punched geo-
textile (Table 9).
4.1.10. Evaluation and comment
This was the first of the engineered sandbag revetments con-
structed to protect oceanfront properties in Australia. The option to
construct in this manner was largely influenced by the immediate
need for protection of the site. The revetment was built as a tempo-
rary/removable structure, as approvals for a permanent structure
would not have been issued without a full environmental assess-
ment into the impact of the long term/permanent proposal, by
which time this important community facility may have been lost.
The time taken to prepare the documentation and receive approvals
for a permanent solution (i.e. a more conventional rock wall
approach) would have delayed the project for some months and may
have led to the loss of the valuable surf life saving club building. The
loss of the structure was unacceptable to both the Newcastle City
Council (owners of the property) and the general public.
Despite the “temporary” nature of the structure, the non-woven
geotextile containers have withstood a number of storm cycles over
14 years of service (Fig. 19). This installation has outlived the
original design requirements and met the objectives of protecting
the surf club while complying with providing a ‘soft’ interim
solution to the total coastline management problem at this site. The
“soft” solution has also proven popular with beach goers, who find
the structure a user friendly option when compared with conven-
tional rock and concrete structures. To date no “permanent” works
have been carried out.
The colour of the geotextile was important to the client who
wanted the structure to blend in with the natural surroundings to
 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439 435

make the structure as unobtrusive as possible. The advantage on

the non-woven geotextile in regard to this requirement is that the
open structure traps the local soil/sand and therefore takes on the
colour of the surrounding area.

4.2. Maroochydore beach revetment

4.2.1. Project location

This beach is located at the southern side of the mouth of the
Maroochy River in South-East Queensland, Australia, and includes
both river and ocean frontage.

4.2.2. Date constructed

2000e2001.

4.2.3. Principal
Maroochy Shire Council. Fig. 21. Self healing characteristics of sand filled geotextile container wall.

4.2.4. Description
of such structures was higher than expected. Some undermining
200 m long by 2.5 m high single layer sand container revetment,
of the toe occurred during king tides and 3 m swell in January
constructed at an angle of 75 from horizontal (1V:0.27H).
2002 (Fig. 20), which resulted in the settlement of approximately
35 m of the wall. Such conditions are likely to have resulted in
4.2.5. Project objectives
failure of a rubble wall with the same design (2.5 m high 1.1 m
Interim protection measures to stabilise the foreshore.
thick @ 15 degrees off vertical). Toe protection of the structure was
identified as a very important feature of any wall. Correct depth of
4.2.6. Site conditions
base and a self-healing toe (Fig. 7) will ensure the durability of the
The mouth, continuing a trend identified in the late 1980s and
structure.
early 1990s, reverted to a more southerly discharge location. During
The self-healing qualities of high elongation flexible sand filled
November 2000, the erosion problem on Maroochydore Beach had
staple fibre geotextile containers have been proven. Some
propagated to such an extent that the foreshore and caravan park
containers have been damaged but the integrity of the structure
were likely to be threatened during the imminent cyclonic season
has not been compromised, as the containers have been able to
and king tides.
mould themselves into the void left by damaged containers
(Fig. 21).
4.2.7. Construction techniques
Although not the first application of its kind the success of the
Utilising two small excavators (5 tonne and 8 tonne), 3000 non-
project (withstanding severe storm attack) has led to the
woven geotextile units were filled and placed as a defence barrier in
construction of a groyne as part of the works using innovative
the dunal system.
products and construction techniques.
4.2.8. Geosynthetics used
Nominal 0.75 m3 sand filled geotextile containers and UV sta-
bilised non-woven staple fibre needle punched geotextile. 4.3. Maroochydore beach groynes

4.2.9. Evaluation and Comment 4.3.1. Project location

During the early months of 2001, king tides repeatedly tested Maroochydore Main Beach, Sunshine Coast, Queensland.
the interim defence barrier. Observations of direct overtopping
during consistent 2 m high wave attack, proved that the stability 4.3.2. Date constructed
November 2001.

4.3.3. Principal
Maroochy Shire Council.

4.3.4. Description
100 m long  2.5 m high sand-filled groyne.

4.3.5. Project objectives

To stabilise the Maroochydore main beach that had eroded by
approximately 150 m in 2 years due to changes in the location of
the river mouth. The structure had to be easily removable should it
be detrimental to the beaches north of the groyne.

4.3.6. Site conditions

Easterly facing open surf beach with an offshore significant
wave height of up to 8 m during storms. Depth limited maximum
Fig. 20. King tides and large swell e January 2002. wave height at head of structure of 3 m (Fig. 22).
436 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439

Fig. 22. Satellite view of Maroochydore beach groynes.

Fig. 24. Placement using specialised equipment.

4.3.7. Community requirements and constraints
As the existing sand filled geotextile container revetment wall
was proving to be popular with fishers and the general public,
a similar groyne structure was envisioned. This structure would
blend in with the natural sandy beach and would not form a hazard
to beach goers.
4.3.8. Construction techniques
The groyne consisted of 4 layers of containers stacked into
a pyramid shape as show in Fig. 23. An 8 tonne excavator was used
to fill the containers while a 35 tonne excavator was used to place
the containers.
4.3.9. Geosynthetics used
2.5 m3 sand filled geotextile containers manufactured from
a composite UV stabilised non-woven staple fibre needle punched
geotextile with high tenacity polyester thread in all seams (Table 10).
4.3.10. Evaluation and comments
The geotextile sand containers and construction methods
used in this project are a result of the years of development into
geosynthetic containers as described in the previous case
studies. For an exposed structure of this nature it was thought
that the 0.75 m3 containers used on the revetment wall would
be too small and that the container size would have to be
increased substantially to ensure stability. The weight/size of
the container was limited, however, to the capabilities of
a 35 tonne excavator as this type of excavator is accessible to
most contractors. The containers were installed using a specially
developed filling and placing apparatus designed to ensure
improved filling and simple handling of such large containers.
The method of lifting and placing containers with a modified
rock grab, as with the 0.75 m3 containers, would have over
stressed the geotextile and seams. The containers were filled in
a lifting/placement cradle and placed using the 35 tonne exca-
vator. No double handling of the container was necessary. This
method of placement is thought to be the first of its kind in the
world, which allows very accurate positioning with low stress
on the geotextile and seams (Fig. 24).
The composite geotextile developed for the Narrowneck reef
project was used for all containers in the groyne minimising the
risk due to vandalism (Fig. 12). After 7 months of installation, only
one incident of vandalism had occurred and was patched using the
screw down patch method. Specialised sewing equipment was
developed for this project to ensure a high strength seam on this
thick puncture resistant geotextile. The groyne has performed
exceptionally well and has withstood wave heights of greater than
the 3 m. Based on the performance of this groyne the council
subsequently constructed a further three groynes to the north of
the structure in 2003 (Fig. 25).
In December 2009 two containers were removed from Groynes
one (constructed in 2001) and two (constructed in 2003) to allow
measurement of real time performance of the geotextiles and
seams used on this project. The containers selected were crest
containers located within 10 m of the head of the structure and had
therefore been exposed to both UV degradation and wave action for
the full period of installation (6e8 years). Testing showed that the
geotextile strength had reduced by less than 15% when compared
to original MARV values.

4.4. Limeburners breakwater

Fig. 23. Section through groyne. 4.4.1. Project location

Geelong, Victoria, Australia.
Table 10
Geotextile container characteristics. 4.4.2. Date constructed
September 2004.
Thickness CBR burst Tensile strength Seam strength
11 mm 12 kN @ 60% 75 kN/m XD Min. 80% of base fabric
45 kN/m MD
4.4.3. Principal
City of Greater Geelong.
 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439 437

Fig. 25. Maroochydore groyne 1 time line.

4.4.4. Description 4.4.6. Site conditions

80 m long  5.5 m high emergent breakwater. Structure founded on existing rock sea bed located at 4.0 m
LAT and located 40 m off the end of the harbour entrance (Fig. 26).
4.4.5. Project objectives
To reduce wave transmission into the 20 m wide Limeburners 4.4.7. Community requirements and constraints
Point boat harbour. The harbour was subjected wind waves of up to 0.5 m, which
entered the harbour and caused considerable damage to the
floating pontoons and made launching/retrieving recreational
vessels difficult and dangerous. The aim of the breakwater was to
prevent transmission of the waves thereby providing a safer
amenity.

4.4.8. Construction techniques

Accurate placement on the containers was of the utmost
importance to ensure the final height achieved the specified design
height. In order to achieve this, anchor blocks were placed at the
end of the structure and along the length to hold the containers in
position while they were being filled. Deployment of the mega
containers was carried out off a 3 m  3 m floating pontoon which
also acted as the support platform for the dredge pipe. The
containers were filled using a 10 inch (250 mm) suction cutter
dredge. Due to the very fine nature of the dredge material the
containers took 5e8 h to fill, compared with the normal 2e3 h
which can be achieved when using good quality sand Hornsey et al.
(2009).

4.4.9. Geosynthetics used

The 40 m long by 12 m circumference geotextile containers
were manufactured from a combination of a standard heavy duty
Fig. 26. Satellite view of lime burners point harbour. staple fibre geotextile on the base and sides, and the composite
 Fig. 27. Lime burners breakwater time line.
staple fibre geotextile on the exposed surface where significant
exposure to UV radiation was expected. High tenacity polyester
thread was used on all seams.
4.4.10. Evaluation and comments
To date the project has provided the required protection to the
harbour reducing the maximum wave heights at the pontoons from
0.35 m to 0.12 m (Cardno Lawson and Treloar, 2008) (Fig. 27). The
structure poses limited threat to the small craft using the harbour,
and in the unlikely event that they should collide with the break-
water the damage to the small craft is expected to be far less than if
it were to impact with a rock/concrete structure.
Using a floating pontoon as a platform to deliver the dredge
material to the containers limited the time available to fill the
containers, as work had to be aborted when winds exceeded 15
knots and wave heights exceeded 0.5 m. A jack up pontoon may
have overcome this problem and allowed faster construction times.
Evaluation of the fill material available prior to starting the
project will allow better estimation of construction times. Despite
the poor quality fill material there has been minimal settlement of
the structure (<300 mm) and the crest height is within the design
limits (Fig. 27).
5. Conclusion
Coastal protection has become a topic of international impor-
tance due to population pressures, extreme events, and concerns
regarding climate change and sea level rise. This has prompted the
search for new and improved systems to protect valuable assets
located on the coast. Experience has shown that the coastal envi-
ronment is a very harsh environment with multiple factors such as
wave attack, abrasion and extreme weathering limiting the effec-
tiveness of many structures. Structures built in these areas,
especially those designed as protection structures should be con-
structed with materials that are not only cost effective, but also
resistant to the many forces discussed in this paper, to ensure a long
term, stable structure. For centuries humans have developed
systems to provide protection to the coastline, and over the years as
technology has improved, the number of systems available has
increased.
Sand filled geotextile containers are a more recent development
designed primarily as a user friendly alternative to traditional rock
rubble structures or in applications where rock is not readily
available. The use of geosynthetic sand containers in coastal
structures is increasing worldwide, in a wide range of applications
such as revetments, coastal protection reefs, surfing reefs, break-
waters etc. Due to the nature of the coastal environment, designers
of geosynthetic structures must allow for a range of factors which
have an impact on the global stability of the structure and the
durability of the individual containers.
As a minimum the following checks should be carried out on the
proposed structure in order to assess the suitability/durability of
the geosynthetic solution as a whole for the application:
Stability
Global/Wave stability e based on actual testing and
container fill volumes.
 W.P. Hornsey et al. / Geotextiles and Geomembranes 29 (2011) 425e439
Scour protection e providing adequate protection to the toe
of the structure.
Sand retention e based on high impact two way flow
conditions.
Durability
Damage resistance e both incidental and vandalism
UV degradation e long term accelerated testing & real time
analysis
Abrasion resistance e based on accelerated testing & real
time analysis
This paper outlines recent advances in geosynthetic technology,
continual laboratory testing and site monitoring of installations
which has led to the development of a robust design methodology.
Issues such as container filling, geotextile selection and extended
durability testing are key factors influencing the design of a geo-
textile sand container system. The information provided allows
engineers to design geotextile sand containers with more confi-
dence in the long term performance. It should be noted however
that this geosynthetic application can still be considered in its
infancy and that further testing and monitoring should be carried
out in order to increase understanding of the durability and
stability of sand filled geotextile containers.
References
BAW Federal Waterways Engineering & Research Institute., 1994. Guidelines for
Testing Geotextiles for Navigable Waterways (Germany).
Blacka, M.J., Carley, J.T., Cox, R.J., 2006. “Field Measurement of ELCOMAX ELCOROCK
Geocontainers at Clifton Springs Boat Harbour, Victoria”. WRL Technical Report,
June 2006. Water Research Laboratory, Sydney Australia.
439
Blacka, M.J., Carley, J.T., Cox, R.J., Hornsey, W.P., Restall, S.J., 2007. “Field measure-
ments of full Sized Geocontainers”. In: Proceedings Coasts and Ports Conference
2007. The Institution of Engineers, Australia.
Cardno Lawson and Treloar., 2008. “Wave Climate at Limeburners Point Boat
Harbour” Report RM2166/LJ5568 Ver 1, May 2008.
Coghlan, I.R., Carley, J.T., Cox, R.J., Mariani, A., 2008. “Research and Development
Studies for ELCOMAX ELCOROCK Geotextile Sand Containers”. WRL Technical
Report, July 2008. Water Research Laboratory, Sydney Australia.
Coghlan, I.R., Carley, J.T., Cox, R.J., Blacka, M.J., Mariani, A., Restall, S.J., Hornsey, W.P.,
Sheldrick, S.M., 2009. “Two-Dimensional Physical modelling of sand filled
Geocontainers for coastal protection”. In: Proceedings of Australasian Coasts
and Ports Conference 2009. The Institution of Engineers Australia,
Wellington NZ.
Hornsey, W.P., Sheldrick, S.M., Carley, J.T., 2009. “Geotextile sand containers:
innovative shoreline protection systems”. In: Proceedings GeoAfrica 2009
Conference (Cape Town South Africa).
Hudson, R.Y., 1953. “Wave Forces on Breakwaters”, Transactions of the American
Society of Civil Engineers, Vol. 118. ASCE, p. 653.
Jacobs, B.K., 1983. “Sandbag Stability and Wave Runup on Beach Slopes”. University
of Delaware. Reseach Report No. CE 83.
Koerner, R.M., 2005. Designing with Geosynthetics Fifth Edition. Pearson Prentice
Hall, NJ, USA.
Oumeraci, H., Hinze, M., Bleck, M., dan Kortenhaus, A., 2003. “Sand-Filled geotextile
containers for Shore protection”. In: Proceedings COPEDEC VI, 2003 (Colombo,
Sri Lanka).
Ray, R., 1977. “A Laboratory Study of the Stability of Sand-filled Nylon Bag Break-
water”. Coastal Engineering Research Centre, Vicksburg, USA.
Recio, J., Oumeraci, H., 2008. “Hydraulic permeability of structures made of geo-
textile sand containers: laboratory tests and conceptual model”. Geotextiles and
Geomembranes Vol. 26, 473e487.
Restall, S.J., Jackson, L.A., Heerten, G., Hornsey, W.P., 2002. “Case studies
showing the growth and development of geotextile sand containers: an
Australian perspective”. Geotextiles and Geomembranes Vol. 20, 321e342.
U.S. Army Corps of Engineers., 2006. Coastal Engineering Manual, EM 1110-2-1100
Part II: (Chapter 2),p. 41and Part IV: (Chapter 5), pp. 55e61.
Weerakoon, S., Mocke, G.P., Smit, F., Al Zahed, K., 2003. “Cost effective coastal
protection works using sand filled geotextile containers”. In: Proceedings
COPEDEC VI, 2003 (Colombo, Sri Lanka).