Conference Paper, Published Version

Oberhagemann, Knut; Stevens, M. A.; Haque, S. M. S.; Faisal, M. A.

Geobags for Riverbank Protection
Third International Conference on Scour and Erosion, ICSE 3. 01.-03. November 2006 in
Amsterdam, The Netherlands

Verfügbar unter/Available at: https://hdl.handle.net/20.500.11970/100057

Vorgeschlagene Zitierweise/Suggested citation:

Oberhagemann, Knut; Stevens, M. A.; Haque, S. M. S.; Faisal, M. A. (2006-11): Geobags for
Riverbank Protection. In: Verheij, H.J.; Hoffmans, Gijs J. (Hrsg.): Proceedings 3rd
International Conference on Scour and Erosion (ICSE-3). November 1-3, 2006, Amsterdam,
The Netherlands. Gouda (NL): CURNET. S. 494-501.

Standardnutzungsbedingungen/Terms of Use:

Die Dokumente in HENRY stehen unter der Creative Commons Lizenz CC BY 4.0, sofern keine abweichenden
Nutzungsbedingungen getroffen wurden. Damit ist sowohl die kommerzielle Nutzung als auch das Teilen, die
Weiterbearbeitung und Speicherung erlaubt. Das Verwenden und das Bearbeiten stehen unter der Bedingung der
Namensnennung.
Im Einzelfall kann eine restriktivere Lizenz gelten; dann gelten abweichend von den obigen Nutzungsbe-
dingungen die in der dort genannten Lizenz gewährten Nutzungsrechte.

Documents in HENRY are made available under the Creative Commons License CC BY 4.0, if no other license is
applicable. Under CC BY 4.0 commercial use and sharing, remixing, transforming, and building upon the material
of the work is permitted.
In some cases a different, more restrictive license may apply; if applicable the terms of the restrictive license will
be binding.

Powered by TCPDF (www.tcpdf.org)

 Geobags for Riverbank Protection
K. Oberhagemann *, M.A. Stevens**, S.M.S. Haque***, and M.A. Faisal***
* Northwest Hydraulic Consultants Ltd., Edmonton, Canada
** River Consultant, Boulder, USA
*** Bangladesh Water Development Board, Dhaka, Bangladesh

moving the embankment to stopping the erosion. Moving

INTRODUCTION the embankment again was not socially or politically
The large rivers in the deltaic country of Bangladesh acceptable.
are up to 30 m deep in places, and scour even deeper.
Strong river currents erode the fine sand from the toe of
the riverbank, steepening its slope. The upper bank then
JAMUNA
fails as a wedge slide or in some cases as a flow slide.
The river removes the slide material from the toe of the RIVER
slope and the erosion process repeats. In the Jamuna Western
Channel
River, banks have receded locally by more than 1 km in a
single year. In addition, there are numerous small rivers
with the same sandy banks. In this nation, bank erosion is
pandemic.
The traditional bank erosion protection is too expensive
for almost all applications in the large rivers. Costs
ranged from USD 29 M to 6 M per kilometer of bank
protected with revetments [1]. There are shortages of
local aggregate for concrete; no suitable rock for riprap;
no heavy marine equipment for construction; troublesome
river currents; and great depths to protect.
Instead, the resources of the country are sand, labor,
and experience with simple floating equipment. Geobags
– geotextile bags filled with fine river sand – serve to Hurashagar
reduce costs to feasible level for protection. About half River
the cost for the slope protected with geobags is the
purchase of the geotextile material. Initially, the
geotextile fabric was imported in large rolls. Now, it is
also produced locally.
Here, the ongoing efforts to protect the Padna Irrigation
and Rural Development Project (PIRDP) with geobags
adjacent to the Jamuna River are described. Another
project at the confluence of the Padma and Upper Meghna
Rivers was protected in the same way at the same time,
but is not reported here.
THE SITE AT THE PIRDP
The PIRDP is a Flood Control, Drainage, and Irrigation
(FCDI) Project protected from Brahmaputra/Jamuna River
flooding by earthen embankments, and is drained and
irrigated by sluice gates and aided by pumping. There are
35,000 ha and 250,000 people inside the embankments. JAMUNA
The Project was put into operation in the late 1980’s and RIVER
became threatened by riverbank erosion in the late 1990’s.
The site is approximately 25 km upstream from the Western Eastern
Jamuna-Ganges confluence. The rates of bankline erosion Channel Channel
are in the order of 100 m per year and vary along a 14-km
reach starting upstream where the small Hurashagar River
joins the Jamuna from the west (Fig. 1). A section of
embankment was retired from the immediate riverbank in FIGURE 1
1997/98 just before it would have disappeared into the RIVER SITUATION ALONGSIDE THE PIRDP. NOTE THAT PROTECTIVE
river. Since erosion was continuing, efforts changed from WORKS HAS BEEN BUILT FROM KAITALA TO THE HURASHAGAR RIVER.
 By the time a feasibility study had been completed a 4. Modified designs implemented since 2004.
crises had been reached. An emergency placing of
geobags was begun to save the Project and a disaster plan a. Initial Experimentaiton during the
was conceived in case the bank protection was not mid 1990
adequate. Several projects used geobags as alternatives for mostly
temporary works as emergency protection or as falling
RIVER MORPHOLOGY aprons. Little systematic reporting on these developments
The Jamuna River is braided with generally two major exist even though in some places work has withstood all
channels on the east and west sides, islands in the middle loads since 10 years now.
and smaller channels cross-connecting the east and west b. Emergency Proetction in 2000
sides. The Hurashagar River adds a complexity to the
west bank hydrodynamics, creating a small but influential The DHV Consultants [1] proposed to fill geotextile
confluence with its mammoth neighbor. The bankfull bags (geobags) with local sand and place them on the
depth in the area of deep scour is approximately 30 m at eroding bank as a feasible emergency measure. The case
times and is marked on the surface during low flow by an for this solution is that sand and labor are in plentiful
upwelling of brown water in the clearer current. supply and inexpensive. The bagsize was 250 kg.
Flow along the eroding bank is essential north-south, Initial placement of geobags in 2001 was done along
and parallel to the bankline during the flood season but the riverbank as an emergency measure – it was a case of
angles towards the bank during the dry season. The "do it now or lose the project." At the same time, a
lowest scoured bed level can be during low-flow. The disaster plan was prepared in case the river should breach
morphology is now well understood. The bankline the flood-control embankment. To the extent that the
movement at the Project is the result of bankline erosion Project is still intact, it can be claimed that protection with
upstream of the Hurashagar delivering an extra large load geobags has been a success.
of sediment to the sandbars next to the Project (Fig. 1). c. Feasibility Level Design in 2002
The Jamuna River carries as much as 85,000 m3/s Halcrow and Associates [2, 3, 4] conducted the
during large floods but that discharge is spread on average feasibility study of geobag protection and recommended
across a 12.5-km wide expanse of major and minor that bank-slope revetment of geobags was viable.
channels and island floodplain. It is the water that travels However, there were questions about the technical
immediately adjacent to the Project bankline at speeds up feasibility. Certain basic assumptions about geobags could
to 3 m/s that define the protection works underwater. not be proven either theoretically or by experience, in-
Eroding bank slopes are on average 1 vertical to 2 country or elsewhere. The proposed geobags design was
horizontal (1V:2H). At the surface strong winds cause 1- essentially experimental, based on knowledge of “trench-
m high waves that erode the exposed bankline so filled” revetment behavior in the USA (Fig. 2). Plans
protection is needed for that as well. were made to accommodate changes for the geobag
DEVELOPMENT OF GEOBAG PROTECTION revetment if they appeared necessary or advisable. This
need for change became known as the “adaptive
The riverbank protection concept developed in phases: approach.” Adaptation would be necessary because the
1. Initial experimentation during the mid-1990 behavior of the bags on the eroding bank was not certain
2. Suggested for emergency protection alongside and the river is prone to change in ways that are not
the PIRDP in 2000 entirely predictable, even in the short term.
3. -Feasibility level designs in 2002

FIGURE 2
FEASIBILITY DESIGN CONCEPT: A HEAP OF BAGS DUMPED FROM THE BANKLINE LAUNCHES DOWN THE SLOPE
PROVIDING AN AVERAGE 0.9 M THICK PROTECTIVE LAYER ON A SLOPE OF 1V:2H
 FIGURE 3
GEOBAGS. THE GRID ON THE THREE LARGEST BAGS IS 10 CM BY 10 CM, AND 5 CM BY 5 CM ON THE SMALLEST. THE
126-KG BAG IS APPROXIMATELY ONE METER HIGH.

In the feasibility assessment [3] a gradation of geobags The original feasibility specifications called for non-
was proposed based on that had been recommended for plastic, non-saline sand free from silt, clay, roots, and
quarry rock [5] in the USA. The proposed sizes were later other organic materials. The minimum grain size was
modified after field experience in 2002 [4]. The adopted 0.074 mm, meaning no silt. Experience has indicated that
masses of the bags became 11, 36, 78, and 126 kg when such sand does not leak from the geobags. Moreover, no
filled with dry sand (Fig. 3). All sizes were combined into damage to properly seamed geobags has been experienced
mix each comprising 25 percent of the total mass. The when bags are dropped from the water surface.
density of fill sand was taken as 1500 kg/m3. Diving investigations on the first implemented works
d. Modified Designs Implemented since indicated that geobags launch down the slope and protect
the bank from further erosion. The launching, however,
2004
does not result in a multiple layer coverage as assumed
The first major adaptation based on field experience during the feasibility study but mostly in a one-layer thick
was to eliminate the two smallest sizes from any more protection. Consequently, the implementation concept
consideration (Tab. I). Now, either one of the larger sizes was modified to arrive at a stable multiple layer coverage.
is judged adequate. This finding was later reconfirmed by Life forms use the geobags as substrate on which to live
hydraulic model tests [6]. For stability against currents, and grow. Some small tubes built from mica flakes and
the larger bags are better. As yet, there is no evidence that inhabited by worms about 1 cm long are attached to the
sand-filled geobags for revetment protection in surface of geobags. At some places, small snails are
Bangladesh should be heavier than 126 kg. attached to the bags; at others, there are fungi on the bag
A bag filled to 80% of capacity (flatter shape) covers surface.
80% of its unfilled area (length x width), whereas when When the opportunity arises, a 126-kg geobag will be
filled to 100 % (rounded shape) it covers only 75 % of its weighed field dry, and then submerged into water. The
unfilled area. Based on observations, the latest weight will be monitored until such time as it becomes
specifications call for 100% filling with dredged sand. constant. This will give an indication of how fast the air is
Underwater consolidation reduces the volume to expelled. The final weight is the submerged weight of the
approximately 85 %. There is discussion about the shape bag and its sand.
still. At this time, there is no compelling reason to change
the empty-bag shape. Geobags are manufactured from polypropylene or
polyester textile fabric, which is non-woven and needle-
TABLE 1. punched and not solely thermally bonded. The textile has
MASS AND SIZE OF EMPTY GEOBAGS a density of about 400 g/m2 and a tensile strength of more
than 20 kN/m. It is UV stabilized to ensure retention of at
Designation Dry mass Length Width Length/
Type (kg) (m) (m) Width least 70% of its original tensile strength before exposure.
The porosity (ratio of the volume of voids to the total
A 126 1.03 0.70 1.47
volume of fabric) of the geotextile is required to be at least
B 78 0.83 0.60 1.38 80%. After observing the unraveling of certain used
geotextile materials, abrasion tests have been specified to
C 36 0.68 0.45 1.51 assure the long-term stability (Fig. 4). There are tests
D 11 0.40 0.20 1.33 specified for other properties of the geobag material, the
bag, and its seams.
 downstream. The gravity force has an against-slope
component tending to keep the bag in place and a down-
slope component tending to move it down the slope.
Adjacent bags can affect stability depending on their own
stability and orientation and on the points or areas of
contact.
b. Physical Hydrulic Model Tests
The hydrodynamic behavior of geobags was
investigated by hydraulic model testing at a geometric
scale of 20 to 1 (prototype to model) in the laboratory of
nhc in Vancouver, Canada [6]. Model geobags,
consisting of permeable cloth fabric filled with fine sand,
were placed on banks formed in crushed walnut-shells at
slopes of 1V:1.5H and 1V:2H. They were displaced at the
incipient motion velocities shown in Table 2. The
tabulated bag masses are sand-filled dry scaled-up
prototype values, and the "bank" velocities represent
scaled-up depth-averaged values at a point one-third of the
slope length inshore from the initial toe of the slope.

TABLE 2.
FIGURE 4 INCIPIENT MOTION VELOCITIES (PROTOTYPE VALUES)
DIFFERENT TYPES OF GEOBAGS PLACED AS TEMPORARY Mass of Geobag Bank Velocity
WAVE PROTECTION. NOTE THE ABRASION OF THE BAG (kg) (m/s)
SURFACE OF THE BAG ON TOP Slope 1V:1.5H

KEY ELEMENTS OF GEOBAG REVETMENTS AT THE PIRDP 126 2.6

a. General Considerations 90 2.4

In the geobags revetment, the geobags form a thin layer 38 1.7

over the natural (unprepared) bank slope. No filter is
Slope 1V:2H
required. The design calls for placing geobags only below
low-water level. Above low water, concrete blocks or 126 2.9
other hard material are used to provide the additional
stability to resist wave attack and to guard against
vandalism. Waves produce a significant pounding action Incipient motion velocities for angular rock (22 kg),
on the bags [ 7] and have moved heavy geobags. When rounded rock (50kg) and concrete blocks (65 kg) were
left unchecked, waves cut a vertical notch in the more determined in the model as well. Incipient motion
cohesive top bank. Also, geobags above water have been velocities were practically the same for all three and for
sliced, drained of sand and the fabric taken was to serve the 126-kg geobags.
such functions as drapery for doors. The model information was analyzed in the context of
On the revetment, the geobags are subjected to the U.S. Army Corps of Engineers’ design equation [5,9]
fluctuating hydrodynamic forces of pressure and shear for rock riprap used as bank protection:
caused by the water flowing over them. Gravity is the 1.25
D30 ⎛ V2 γ ⎞
main stabilizing force acting to hold the geobags in place = S f C S CV CT ⎜⎜ ⎟
γ s − γ ⎟⎠
(1)
against hydrodynamic forces - provided that the bank is
not too steep, in which case gravity can become a de-
Y ⎝ K 1 gY
stabilizing force. where
Beneath the thin layer of geobags is the natural bank V = local vertically-averaged velocity
material consisting of sand, silt, or clay, which also must Sf = safety factor, minimum recommended value for
be stable in a geotechnical sense. riprap design = 1.1
Riverbed scour at the toe of the bank is a most Cs = 0.30 for angular rock and 0.36 for rounded
important factor affecting bank slope stability. The river CV = coefficient for vertical velocity distribution,
bed at the Project consists of fine and very fine sand, with range 1.0 to 1.28 for straight channels to abrupt
median diameters of 0.1 to 0.3 mm. Almost any river flow bends
can disturb this material, and floods move great masses of CT = coefficient for riprap layer thickness, 1.0 or less
bed material eroded from the bank and picked up from the with increasing thickness
bed and sandbars. A local rate of scour deepening of 5 K1 = side slope correction factor
m/day has been measured at the lead spur for protection D30 = size of stone for which 30 % by weight is finer
works in the Jamuna River at Sirajganj [8]. Y = depth of flow
In hydrodynamic stability assessment, it is assumed that γ = specific weight of water
the bank slope behind the geobag skin is stable and that γs = specific weight of stone
the bags do not slide on the bank material, only on each
other. The hydrodynamic forces tend to move the bag
 Key values recommended for the side slope correction c. Geotechnical Studies
factor K1 are as follows (Table 3): Geotechnical aspects discussed in this report include
movements in the riverbank soil and on the interface
TABLE 3.
CORRECTION FACTORS FOR SIDE SLOPES between geobags and bank material. This includes
geobags sliding on sand, and bags and sand bonded so that
Slope angle Correction Factor bags plus underlying sand slide on sand.
Slope
(degrees) K1
Before the installation of geobag protection, eroding
1V:1.5H 33.5 0.71 banks at the Project were characterized by erosion of sand
1V:2.0H 26.5 0.88 from the bank slope and river bed by currents, followed by
geotechnical failure of the more cohesive upper layer,
1V:3.0H 18.0 1.00 extending from the top of the bank down to low-water.
Flatter than
- 1.00 Wave erosion was also an important factor.
1V:3.0 H
In the feasibility study [2,3], it was recommended that
the upper slope of the bank be dressed to 1V:2H before
From the experimental results it was determined that a placing the geotextile filter and concrete blocks. The
value of 0.77 should be used for the shape factor Cs for the launching heap would consist of a graded mixture of
two largest model geobags bags on 1V:2H and 1V:1.5H geobags dumped below low water. The width of the heap,
slopes. The diameter of a geobag was taken as the cube normal to the bank, would be based on the design scour
root of the volume. depth. The riverward slope of the heap was set at
It has been reported [10] that when the flow velocity 1V:1.25H (approximately 39 degrees to the horizontal).
exceeds 1.5 m/s or so, sand can move inside a bag from The angle of repose for model geobags is 52 degrees, but
the upstream to the downstream side. In special cases a the bags could be heaped at a steeper slope by piling them.
bag could roll because of this movement. Such sand The launched slope was considered stable.
movement could not occur in the model geobags, and has Model tests at BUET [3] indicated that if a slope failure
not been observed to date in the field. Velocities at or occurs for one reason or another - for example, the loose
adjacent to the bags on the revetment are generally not this state of the soil - the geobags have no chance to launch.
high. Rather, they fail as a whole, the heap sliding down the
Model geobags slide over each other on bank sloes of slope as a unit.
approximate 52 degrees. Prototype bags are just slightly Considerable geotechnical investigation has been done
less stable, sliding over each other on slopes of 47 degrees for the Project, including drilling and logging boreholes,
Model 90-kg geobags were dropped into a 10-m depth and estimating soil strength properties. At the Project, the
of water flowing at 1.7 and 3.3 m/s (scaled-up prototype slopes are generally stable. For inclinations steeper than
values). It was difficult to achieve complete coverage 1V:2H (Fig. 5), slope stability where it exists results
even when the bank was visible and bags were dropped to from“hidden stabilizing influences,” that is, geotechnical
cover an observed bare spot. Bags tended to cluster in factors that are not taken into consideration in common
random piles surrounded by bare patches. Mixtures of practice. The steep parts of the natural slopes are
bags achieved even more precarious coverage. considered to be at the “ultimate limit of stability.”

FIGURE 5
BANK PROFILES TAKEN AT THE ERODING SECTION IN APRIL 2004 WHEN THE WATER LEVEL
WAS APPROXIMATELY +5.0 M PWD.
 There is an upper layer, 5 to 6 m thick, of floodplain protection. Divers’ observations clearly showed that this
soil consisting of clay and clay-silt with low plasticity. did not happen. The coverage was either by single bags or
The clay layer reaches down from floodplain level to sometimes lumpy with bare patches. The smallest bags
approximately low-water level. Below this upper clay disappeared. Clearly adaptations were needed.
layer, the bank consists of fine-grained and poorly graded For predominantly construction purposes, single-size
sand of medium compactness. The specific weight of the geobags are favorable so only 126 kg bags will be used for
sand grains is taken as 26.5 kN/m3. The effective shear future work in Jamuna River. The smaller size of 78 kg is
strength of the clay allows for vertical faces up to 4 to 5 m proposed for smaller rivers. If there is to be a mix, it will
high. be with the 78- and 126-kg bags.
Sand dominates the overall stability. It can fail below The protective system was to remain geobag revetment
the clay, and then the clay collapses afterwards. Sand protection below low water level and concrete blocks or
usually fails more or less on a flat plane, the movement interconnected systems such as grout-filled mattresses
being in the form of a wedge translation. Slip circle above low water.
failures are uncommon. A multi-step implementation system combining a fast
For the flatter slopes of 1V:2H to 1V:2.5 H, the angle response to erosion threat and an optimized use of bags
of internal friction is 28 to 30 degrees. For slopes of has been developed and implemented that has provided
1V:1.5H and steeper, the angle of internal friction is 32 to satisfactory protection.
35 degrees. Immediate Protection: Imminent river erosion
There are three geotechnical ways the slope can fail. requires a fast response. This is provided through mass
They are: geo-mechanical; flow slide of sand; and dumping of bags along the eroding bank, allowing the
liquefaction due to dredging or earthquakes. Scour bags to launch down the slope (Fig. 6). The result is a
(erosion of bed and lower bank material by flowing water) commonly one-bag thick cover layer, which substantially
is an important factor triggering the slope failure process. reduces erosion rates but is not stable in the long run.
Scour is considered a hydrodynamic issue. During this initial stage only temporary wave protection
The slopes are prevented from such failures by above low water level, consisting of geobags, can be
protecting the toe and lower bank erosion that would provided.
steepen the slope, preventing rapid changes in soil stress First level protection: A three-bag layer is placed over
levels at the toe of the bank, and stabilizing the top the launched bags making, on the average, a four-layer
bankline. thickness on the slope after completion of this first level
d. The Adaptive Approach protection (Fig. 7). In addition, a thin and wide falling
apron for the expected future scour is placed at the toe of
In the feasibility study it was conceived that a heap of this protection. Lately, 12 to 15 m wide falling aprons are
geobags of different sizes placed along the bank just built consisting of three layers of bags. This falling apron
below low water would launch when undercut by erosion can cover up to 15 m scour depth.
and cover the eroding area with a 0.9 m thick layer of

Stockpiled Filled Bags

Launching Heap

FIGURE 6
IMMEDIATE PROTECTION IMPLEMENTED AT THE PIRDP IN 2002.
 Anchor Line

FIGURE 7
1ST LEVEL PROTECTION DUMPED FROM POSITIONED BARGES

Adaptation: The river response to protection uniform surface. It is planned to place 1.5 layers of bags
commonly results in toe scour along revetments. For this after reaching deepest scour depths.
purpose falling aprons are placed along the toe. These Maintenance: Regular maintenance is a long-term
falling aprons may have to be upgraded to first level operation during the lifetime of the protective system.
protection after their deployment. In addition to depth The normal maintenance is expected to start about 5 to 10
changes, the erosion can shift in upstream or downstream years after implementation and after completion of second
direction. In these cases, immediate and first level level protection to deeper scour levels.
protection will be repeated for areas under new attack. Phased Construction and Monitoring: To obtain
Second level protection: The river reacts to the bank sustainable bank protection several phases of construction
protection during the initial years and there are changes to in the same area are required. A period of monitoring and
the overall morphology. Settlements and adjustments of adaptation follows the initial immediate and first-level
the unprepared uneven bank will occur. Scour might protection construction. The phased implementation
reach deeper levels and the falling apron at the toe starts concept requires plenty of resources for monitoring and
deploying. The second level protection is designed to supervision.
improve the protective layer of first level protection and
subsequent adaptation works and to arrive at a more

FIGURE 8
CONSTRUCTION OF PERMANENT WAVE PROTECTION ABOVE LOW WATER LEVEL DURING THE ADAPTATION PHASE
 [2] Halcrow and Associates, 2002, Feasibility Study - Final Report,
SUMMARY Jamuna-Meghna River Erosion Mitigation Project, Volume 1 (Phase II).
Main Report prepared for Bangladesh Water Development Board and
The hastily designed and constructed emergency Asian Development Bank, May.
geobag revetment prevented the PIRDP from becoming a
[3] Halcrow and Associates, 2002, Feasibility Study - Final Report,
disastrous failure. It was a success! To date, more than Jamuna-Meghna River Erosion Mitigation Project, Volume 2 (Phase II).
five million geobags have been placed here and at the Report prepared for Bangladesh Water Development Board and Asian
Padma-Upper Meghna confluence. Nearly all phases of Development Bank, May.
the initial concept have been modified based on field [4] Halcrow and Associates, 2002, Feasibility Study - Final Report,
(most importantly, diving inspections) and laboratory Addendum, Jamuna-Meghna River Erosion Mitigation Project, Volume
experience. This adaptation method is used to adjust the 2 (Phase II). Report prepared for Bangladesh Water Development Board
works to suit conditions that cannot be predicted and to and Asian Development Bank, May.
make improvement to any aspect of the works deemed [5] US Army Corps of Engineers, 1991, Hydraulic Design of Flood
Control Channels, EM 1110-02-1601.
deficient, whether it is in design, construction,
[6] [6] nhc (northwest hydraulics consultants), 2006, Jamuna-Meghna
management, scheduling, or other issue. The revetment River Erosion Mitigation Project, Physical Model Study (Vancouver,
derived at is the most cost effective solution at estimated Canada). Final report prepared for Bangladesh Water Development
cost of around USD 2 M per kilometer on average. Board, Jan.
[7] Individual Consultants, 2003, Geobags Protection, Project Concept,
Experience, Future, Jamuna-Meghna Erosion Mitigation Project, Part A.
ACKNOWLEDGMENT Report prepared for Bangladesh Water Development Board and Asian
Development Bank, June.
The Jamuna-Meghna River Erosion Mitigation Project [8] Oberhagemann, K. and Noor, Abdun, 2004, Scour and Erosion
under which the described work is being developed is alongside Bank Protection work: Case Studies from Bangladesh. Paper
supported through Asian Development Bank Loan No presented at Singapore meeting of ICSE.
1941 BAN (SF). [9} US Army Corps of Engineers, 1994, Channel Stability Assessment
for Flood Control Projects, EM 1110-2-1418.
REFERENCES [10] Pilarczyk, K.W., 2000, Geosynthetics and Geosystems in
[1] DHV Consultants DV, 2000, River Erosion Prevention and Hydraulic and Coastal Engineering, A.A. Balkema, Rotterdam.
Morphology Study of Pabna Irrigation and Rural Development Project
(PIRDP). Final report prepared for the Bangladesh Water Development
Board and Asian Development Bank, Nov.