Bank erosion Mekong Delta and Red River

Full text

Bank Erosion in Mekong Delta and Red River
Vietnam
Mission report – 23 November-6 December 2003
Delft, March 2004

2

3
Contents
Main report
Foreword
Executive summary
1. Introduction
1.1 Background
1.2 Terms of reference
1.3 Approach
2. Bank erosion along the Mekong and the Red River (pre-information)
2.1 General
2.2 Mekong River
2.3 Mekong Delta
2.4 Red River Delta
3. Previous study on Mekong bank erosion
4. Mission to Mekong Delta and Red River
4.1 Approach (field visits and meetings)/reference to Appendices
4.2 River hydrodynamic, morphology and bank erosion aspects (G. Klaassen)
4.3 Prediction of bank erosion (G. Klaassen)
4.4 Geotechnical aspects (J. Lindenberg)
4.5 Bank protection (M.van der Wal)
4.6 Summary of personal observations (K. Pilarczyk)
5. Analyse, final conclusions and recommendations of bank erosion mission
5.1 General considerations on Mekong and Red River delta’s
5.2 Measures to cope with or counter bank erosion
5.3 Legal and socio-economic aspects
5.4 Integrated River Basin Planning and Management
5.5 Capacity building
5.6 data and Information management
5.7 Conclusions and recommendations in short (as a reminder)
5.8 Proposed Action Plan 2004-2006
Appendices
I Mission participants
II Contacts in Vietnam
III Mission program and schedule
IV Field visits Mekong River
V Meetings Mekong River
VI Field visits Red River
VII Meetings Red River (incl. Final meeting)
VIII Damage overview (collected data)
IX Supplementary informations

4
Supplement: Short review on bankerosion and protection
1. Causes of erosion and failure a
2. Cliff erosion
3. Bank erosion in stable river systems
4. Bank erosion and planform changes
5. Survey and data collection
6. Types of bank protection
7. Techniques of bank protection
8. References

5
Main report
Foreword
The Department of Dike Management and Flood Control (DDMFC) of the Ministry of
Agriculture and Rural Development (MARD) requested the Rijkswaterstaat (Dutch Public
Works Dpt.) to review the problem of bank erosion in Vietnam and to assist in the preparation of
a plan for tackling this problem. The mission took place in the period 23 November-6 December
2003.
The results of this mission, conform to its terms of reference, are presented in this report.
It should be stressed that it was a challenging task, which would have been impossible to realize
without the close cooperation of Vietnamese experts and counterpart staff. We wish to express
our deep appreciation for the friendship and dedication of all persons who have been involved in
this complex effort.
Executive summary
The mission and this (preliminary) study report were performed to support MARD/DDMFC in
their approach to the bank erosion problems, especially concerning the improvement of the
organizational and technical measures, and possibly, to obtain an international help for bank
erosion problems in Mekong Delta and Red River Delta in Vietnam.
The mission spent one week in the Mekong Delta and 4 days in the Red River area.
To obtain the general picture of the problems the Mission has visited a number of representative
sites in the upper part of Mekong Delta (non-tidal area) and in the lower part of the Red River
area. Special attention and time was allocated for discussion with persons working for a long
time in the visited areas (praciticians) or doing studies for these areas (researchers).
The Mission recognizes the large scale of erosion problems in Vietnam, which are associated
with many social and economic implications and consequences. The financial funds for tackling
these problems are extremely limited. It is surprisingly to see that even under such difficult
conditions (within such strict constraints), the DDMFC, provinces and local agencies are able to
generate very good results. However, it can be defined more as anemergency management than
a planned management. This approach is not transparent enough (especially for presenting the
urgency of problems) and it is from be optimal. Besides the financial and technical matters the
organizational matters and cooperation (optimal use op human potential and facilities) matters
needs further improvement.
The limited time allocated for this mission did not allow for making real feasibility studies for
the problem considered. This report provides the results of the findings of the mission; these are
more factual findings than an in-depth analyse. In this way the information collected or
impression of individual members with different backgrounds can be preserved for possible use
in the follow-up studies. The main recommendations concern improvement of prediction and
monitoring techniques, setting up of data base, Strategic plan and capacity building.
The Mission hopes, however, that the presented analyse of the problem and conclusions and
recommendations in Chapter 5 will be of value for formulation of more systematic approach. An
important part of that approach should be formulation of National Strategy on Bank Erosion
Mitigation and Capacity Building needed for implementation of this new approach.

6

7
1. Introduction
1.1 Background
The Mekong Delta in Vietnam is the most downstream part of the Lower Mekong River Basin
and it is of great importance to the Vietnamese community and economy. It is potentially an
area of great productive capacity and its development is of crucial importance to the nation’s
economic prosperity and food balance. At the same time the Delta is a difficult area, with both
considerable physical resources and environmental constraints: great annual variety in the
Mekong’s hydrological regime, large tracts of lands with acid sulphate soils and vulnerable
wetlands. It is also an area, which is heavily and frequently affected by flooding and bank
erosion resulting in loss of live and high economical damage. The similar problems are noticed
in the red River area. The Vietnamese Government has recognized this problem in recent years
and it has been decided to undertake the necessary remedial actions.
The Department of Dike Management and Flood Control (DDMFC) of the Ministry of
Agriculture and Rural Development (MARD) requested the Rijkswaterstaat (Dutch Public
Works Dpt.) to assist in the preparation of a plan for tackling the problem of bank erosion. This
mission is the first step in this direction. The results of the mission, conform to its terms of
reference, are presented in this report.
1.2 Terms of reference
The mission aims at providing a framework for tackling the problem of bank erosion in the
Mekong Delta and along the Red River, with reference to the following elements:
To counteract present problems
1. Assessment through field visits of bank erosion problems in the Mekong Delta and along the
Red River, especially during or just after floods
2. General analysis of the bank erosion problems in an attempt to determine mechanisms,
causes and other factors which have a crucial impact
3. Assessment for the 6 Mekong problem locations at Thuong Phuoc, Tan Chau, Hong Ngu,
Sadec, Long Xuyen and Can Tho, and for some locations in the Red River, whether
technical solutions are applicable and what are their limits and implications
To counteract future problems
4. Indication of which banks have a high risk on future bank erosion
5. Analysis of the used prediction method and suggestions for improvement
6. Proposal for a straightforward monitoring and data collection system to enhance insight in
bank erosion problems
7. Analysis of existing protection structures and techniques and suggestions for improvement
8. Recommendations for new protection structures and flood fighting strategies, which also
involve the local population
9. Definition of a suitable approach to bank erosion problems and assessment of the need for a
master plan in which the local population participates and which involves both bank
protection and planform stabilization of river branches

8
1.3 Approach
To comply with the terms of reference the following approach was taken:
1. Literature study about the general physical aspects as well as bank erosion along the Mekong
and the Red River;
2. Study of the available research data about bank erosion along the the Mekong River;
3. Field visits to existing revetments and bank erosion locations along the Mekong and the Red
River;
4. Meetings with the authorities of the DDMFC (Department of Dike Management and Flood
Control), the WRRI (Water Resources Research Institute), the HWRU (Hanoi Water
Resources University), the SARD (Service for Agriculture and Rural Development) and the
Southern PI.

9

10
2. Bank erosion along the Mekong and the Red River
(pre-information collected before mission)
2.1 General
Vietnam is situated in the tropical monsoon area of South East Asia with an average rainfall of
1800 to 2500 mm/year and is a typhoon-prone country.
A large number of people who are mainly involved in the agricultural and fishery sectors live on
the low lying river floodplains, deltas and coastal margins. The most important ports are located
along the coast. The potential for disaster in these areas is high, as protective river and sea dikes
are frequently overtopped or breached, which results in flooding. Otherwise severe bank erosion
occurs in the deltas of the Red River and the Mekong. Flooding and bank erosion cause loss of
life and damage to agricultural land and infrastructure.
Figure 1: Flooding in Vietnam
2.2 Mekong River
The Mekong River is one of the 12 great rivers of the world and is approximately 4,220
kilometres long. From its source in the Xizang plateau, the river flows through the Xizang and
Yunnan regions of China, forms the boundary between Laos and Burma and after between Laos
and Thailand. Below Phnom Penh, it divides into two branches, the Song Han Giang and Song
Tien Giang, and continues through Cambodia and the Mekong basin before draining into the
South China Sea through nine mouths, which in Vietnam are referred to as nine ‘dragons’. The
river is heavily silted and is navigable by seagoing craft of shallow draft as far as Kompong
Cham in Cambodia.
A tributary entering the river at Phnom Penh drains the Tonle Sap Lake, a shallow fresh- water
lake that acts as a natural reservoir to stabilize the flow of water through the lower Mekong.

11
When the river is in the flood stage, its silted delta outlets are unable to carry off the high
volume of water. Floodwaters back up into the Tonle Sap, causing the lake to inundate as much
as 10,000 square kilometres. When the flood subsides the flow of water reverses and proceeds
from the lake to the sea. This effect significantly reduces the danger of devastating floods in the
Mekong Delta, where the river floods the surrounding fields each year to a level of one to two
metres. Habitation of the delta remained restricted by these stagnant waters until canals could be
constructed, at the end of the 19th century (Mburu, 2001).
2.3 Mekong Delta
The Vietnamese part of the Mekong river has a length of 230 km and is called Cuu Long. It has
two main branches, the Tien (North branch) and the Hau (South branch). The Mekong Delta is
an important region in Vietnam; with its area of 39,000km2, it covers about 12% of the country’s
total area and provides about 50% of the national agricultural production. About 15 million
people live within the delta with 3.5 million in the urban centres.
Figure 2: Mekong Delta location, provinces and regions
The Mekong Delta is a low-level plain not more than three metres above sea level at any point
and criss-crossed by a maze of canals and rivers for transport, irrigation, drainage and flood
control. So much sediment is carried by the Mekong's various branches and tributaries that the
delta advances sixty to eighty metres into the sea every year. The estimated amount of sediment
deposited annually is about 1 billion cubic metres. About 10,000 square kilometres of the delta
are under rice cultivation, making the area one of the major rice-growing regions of the world.

12
To prevent detrimental effects to any of the riparian countries all developments on the main
stream as well as the tributaries are co-ordinated by an inter-riparian Mekong Committee.
The Mekong River has created a variety of natural landscapes, ranging from tidal flats, sandy
ridges and tidal backswamps in the coastal plain, estuaries at river mouths, to river floodplains,
broad depressions, peat swamps, alluvial levees and terraces further inland.
Wetlands created by seasonal or permanent inundation have an important function in the Delta.
They form a buffer between sea and land, trap river borne sediment brought with floods, play an
important role in soil conservation and coastal protection, provide a habitat for wildlife, and
serve as spawning and nursing grounds for fish. They are extremely fragile and could easily and
irreversibly be affected by improper management.

13
Figure 3: Hydrology of the Mekong Delta

14
Water regime and tides
The most determinant features of the natural water regime of the Delta are depicted in the figure
3. Both rainfall and river flow in the Delta have a pronounced seasonal pattern with very high
and very low rainfall and discharge values in respectively the wet season and the dry season
which each last for roughly half a year. Periods of water excess alternate with periods of water
shortage and all the necessary water control measures essentially originate from this regime
feature.
Mean annual precipitation ranges from about 2,400mm in the western part of the Delta to
1,300mm in the central part and 1,600mm in the eastern part. The duration of the rainy season is
from April to November in the western part and from May to November in the rest of the Delta.
Another important feature of the water regimes of the Delta are the tides of the surrounding seas.
The tide of the South China Sea is predominantly semi-diurnal with an amplitude of some 2.5-
3.0m. The tide of the Gulf of Thailand, however, is mostly of the diurnal type, while its
amplitude is only some 0.4-1.2m. The tides have a significant influence on the river and
connected canals in the coastal zone and also in the area adjoining the main Mekong river
branches all the way into Cambodia.
On the long-term, also the impact of the sea level rise on the Delta will be considerable, given
the extremely flat topography and the tidal influence throughout the Delta (0.3m is often
mentioned as the most probable sea level rise for this area).
Flooding and saline intrusion
The average elevation of the Vietnamese part of the Delta equals about 0.8m+MSL. During the
period of high discharge, the banks of the Mekong in the north of the Delta, below Kampong
Cham and above Can Tho, are overtopped and the land is inundated, up to depths of 4.5m. The
inundation usually starts in July/August and ends in November/December. A positive effect of
the flooding is the deposition of sediments in the flood plains.
As the capacity of the river system increases downstream, there is a considerable attenuation of
the water levels and less flooding. At the coast, the combined action of river deposition and the
sea has created a slightly higher coastal belt which further reduces flooding.
The flooding problem in the North is aggravated by high rainfall. In the South, excess rain water
also leads to large scale inundation of the land outside the river flooding zone. This occurs
especially in the South Western part of the Delta. In the poorly drained depression areas, the
inundation may last as long as 6 months.
During the March-May period Mekong discharges are low and for an important part required to
prevent deep saline intrusion. Higher rates of abstraction would increase salinity intrusion which
is already affecting large areas.
Flood protection
The Mekong Delta is largely unprotected and therefore characterized by widespread,
uncontrolled and prolonged floods. A system of drainage channels and pumping stations is used
to make agriculture possible. Houses are situated on high places such as along the canal banks,
roads or sandy ridges but during the high floods they are usually still flooded. Alternatively, the

15
houses are built on stilts or raised foundations above the flood level. Boats are used for
communication during floods.
Flood control usually consists of low embankments along the primary and secondary canals,
while the secondary-tertiary canal connections are provided with simple sluices or temporary
earth dam closures. These means provide partial protection to agricultural production during the
early part of the rainy season. In the deeply flooded areas, embankments are overtopped later in
the flood season during a normal flood year but the embankments may give year round
protection in a low flood year. In the coastal area some flood protection schemes also prevent
salt water intrusion. Quite a few large sluices in primary-secondary connections were built in
that area and many more are under construction or planned.
The main canal systems are generally planned, designed, constructed and operated by the
national and provincial water resources development organizations (primary and secondary
units). The tertiary canals/water control system and on-farm developments, on the other hand,
are undertaken at the district/village/farmers group level. Flood control is generally practised at
the level of provincial authorities (secondary unit).
Development and flood protection strategies
The need for economic growth and diversification of the economy in an environmentally sound
and sustainable manner will govern the scope and pace of the development of the Delta’s
resources.
The main thrust of water resources development would be on-farm development and canal
improvement to bring more irrigation water to the already irrigated areas and to improve
drainage conditions and promote flushing of acid water. The development would also include
embankment improvement in the deeply flooded areas, for the time being to prevent flooding till
the end of August only and full year round protection in the shallowly flooded, already more
developed areas.
On good soils forest cannot compete with crop production or aquaculture in terms of income or
employment generation. Its development potential lies in areas with low graded, acid sulphate
soils and along the coast. Inland swamps (Melaleuca) and mangrove forests are essential for
biodiversity conservation and to save the few natural reserves that have been left. The
sustainability of shrimp culture and fisheries depends on their existence. In addition, they
provide coastal protection.
Because of the specific situation of the Mekong Delta it is neither economically justified nor
environmentally sound to provide complete flood protection. Controlled flooding would still
allow for acidity flushing, would maintain the natural fertilizing effect of sediment, and would
minimize the disruption of fish migration and spawning. The actual policy recommendations
include:
1. Low embankments in the deeply flooded parts to protect against early floods;
2. Full embankments in shallow agricultural areas to protect against 10-year floods;
3. No embankments on land that has potentially serious acid-sulphate problems;
4. Adequate forecasting and warning systems;
5. Adequate evacuation plans and an adequate number of escape/rescue facilities;
6. Maintenance of natural flooding regimes in sanctuaries and swamps and mangrove forests.

16
Figure 4: Flood map 1984
Channel migration
The high sediment load of the Mekong River system, estimated at 160 million tons per year,
results in an inherently dynamic channel system with rapid rates of change. Commonly, such
changes are associated with channel migration, whereby deposition along a riverbank is
countered by erosion of the opposite bank. Susceptibility to channel migration and the type of
mechanism responsible vary according to the location within the deltaic system. The upper delta
experiences very rapid rates of channel migration (with banks erosion rates commonly up to 20
m/year), caused by the lateral accretion of point-bars and mid-channel bars / islands, and the
downstream migration of mid-channel bars.
Mid- and lower delta channels are more stable (bank erosion 5-10 m/year), and channel change
here is mainly caused by the slow accretion of elongated point-bars and mid-channel bars. The

17
slower current velocities and cohesive bank material, as well as the protection afforded by
mangroves and nypa palms (Nypa fruticans) in saline reaches, are the principal reasons for the
relative channel stability here. Near the mouths of the main distributaries, channel changes are
common and result from the formation and shifting of distributary-mouth bars.
Another group of channel change involves the abandonment of channel segments, which
generally leads to their progressive siltation. At a small scale, channels separating a mid-
channel or channel or distributary-mouth bar from the river bank may infill with sediment to
eventually result in the coalescence of the bar with the bank.
At a larger scale, individual distributaries may also become abandoned. The progressive
sediment accumulation within the Ba Lai sub-branch of the Mekong is a manifestation of this.
Also, many of the smaller rach-type channels along the South China Sea coast (i.e. Ca Mau
peninsula and the area about the mouths of the Saigon and Vaico rivers) are prone to change in
position and abandonment, as the large tidal range along this coast results in the progressive
inward transport of sediment from the sea and eventual channel infilling. Mangroves are likely
to assist in sediment accumulation within these channels.
Sedimentation and erosion processes in the Mekong Delta are highly seasonal given the large
annual fluctuation in both the river discharge and sediment load. Suspended sediment load of the
river inflow varies from less than 100 mg/l during the dry season to 600 mg/l during the peak
flood season.
During the flood season, most bedload, consisting predominantly of sandy material, is
transported and deposited on the channel bed and in bars. The finer suspended load is either
deposited on the delta plain through overbank flooding or flushed out into the ocean.
During the low-flow period, suspended sediments also get deposited in-channel. In the seaward
parts of the channels, this deposition is aided by saline intrusion, which causes sediment flushed
to sea during the flood season to be re-imported into the delta. In the larger channels, much of
the dry-season deposition is ephemeral, as the fine sediment is reworked during the following
flood season.
In the smaller channels, tidal creeks and canals, mud deposition is more likely to be cumulative
over successive dry seasons.
Bank erosion
Bank erosion is considered a serious socio-economic problem in the upper delta provinces of An
Giang and Dong Thap provinces. Problems are especially severe at Tan Chau on the Mekong
branch in An Giang, where erosion rates attain 30 m/year, and approximately 400 households
have had to be relocated recently due to destruction of their dwellings through bank collapse.
Bank erosion has resulted in major disruptions to local livelihoods, and financial burden on the
provincial government by necessitating the relocation of inhabitants and localised bank
protection works (e.g. Truong Dang Quang). Losses due to bank erosion appear to have
increased in the last decade, probably due to the growing urban population and the resultant
concentration of activity and capital along the waterfront (e.g. Truong Dang Quang).
The severity of erosion at Tan Chau is largely attributable to the sharp meander-bend
morphology, which focuses the river flow energy onto the concave bank (where the town is
situated). The gradual downstream rotation of the point-bar on the opposite bank has resulted in
a progressive downstream shift in the zone of erosion; stretches of river bank upstream of Tan
Chau, which formerly experienced severe erosion are now experiencing bank accretion (e.g.
Truong Dang Quang).

18
Other erosion hotspots further downstream within An Giang (e.g. at Long Xuyen) are mostly
associated with the downstream migration of mid-channel bars, which creates a shifting zone of
erosion downstream and to the sides of the bar, and a zone of accretion to its upstream.
Sand mining
Sedimentation on the opposite bank, which accompanies bank erosion, also represents an
economic cost in places, through the shoaling of navigation channels, the stranding of wharves,
docks and other water transport infrastructure, and the blocking of entrances to canals. However,
sedimentation in the main distributary channels is regarded by many as an economic benefit,
given the predominantly sandy nature of channel sediments, and the increasing demand for
construction sand driven by urban expansion.
Numerous sand dredging operations exist along most of the length of both the Mekong and the
Bassac branches. An individual operation may extract volumes in the order of 10.000 m3/year
from the bed of the channels (e.g. Ky Quang Vinh). Local over-exploitation of sand is also
blamed for the frequent occurrence of bank erosion in the Mekong Delta.
2.4 Red River Delta
The two major rivers systems in the Red River Delta in the north of Vietnam are the Red River
system and the Thai Binh River system. The Red River system consists of the confluents Da
River, Thao River and Lo River and five branches, these being the Duong River, Luoc River,
Tra Ly River, Dao River and Ninh Co River. The Red River is so named because of the high
amounts of red sediment it carries. The Red River carries about 200 million tonnes of sediment
each year.
In the Red River Delta in, people have built 3000km of river dikes and 1500km of sea and
estuary dikes to protect against flooding. Many of these dikes are old and were built by using
inadequate manual construction technology and poor materials. Dike foundation conditions and
stability have not always been properly evaluated before construction or improvement. River
dikes often suffer damage from under-seepage and piping, slides or local collapse during high
flood stages. Moreover, the construction of dikes has gradually reduced the areas of the flood
plains that are available to accommodate excess flood flows, with the result that river-flood
levels have become increasingly higher.
Bank erosion
Due to instability of the river channel the Red River is affected by siltation of the river channel
below Son Tay as well as a general increase in bank erosion, threatening dikes at numerous
locations. The most likely causes for this instability are an increase in slash-and-burn land
clearing practices in the highlands of Vietnam and China and the release of high energy,
sediment-poor water from reservoirs such as Hoa Binh.
Slash-and-burn land clearing increase both runoff and sediment load, resulting in sedimentation
below Son Tay, where the bed gradients are lower and the river can no longer cope with the
sediment. This sedimentation may cause widening and meandering of the river, resulting in local
bank erosion.
The release of high energy, sediment-poor water from reservoirs leads to scour and bank erosion
for some distance below the dam, and sedimentation beyond this area, when the bed gradient

19
becomes smaller. Furthermore flow regulation by reservoirs implies an increase in the mid-bank
to full-bank flow duration and therefore in bank saturation and bank erosion.
Bank erosion and other negative impacts of channel instability are counteracted by structures,
which should stabilize the riverbank and channel and protect the dikes. Usually, the following
structures are applied:
1. Rock filled wire baskets (gabion mattresses), underlain with geotextile filter cloth to prevent
erosion. A concern with these gabions is the disintegration of the wire casings by corrosion,
which eventually will lead to flowing away of the relatively small stones and failure of the
structure;
2. Rock and concrete blocks and mats;
3. 2 to 3 meter thick bamboo platings, spaced 5 to 10 meters apart at the toe of the dike;
4. Groins or hard points.
The dikes are protected usually by hard revetments, e.g. of the following types:
1. Interlocking rectangular blocks with raised rectangular surface;
2. Six-sided interlocking blocks with a raised triangular surface.
3. Riprap
The function of the raised surfaces is the dissipation of wave energy. The revetments are usually
underlain with a granular filter, or, in more recent structures, with a geotextile filter cloth to
prevent erosion.
3. Previous study on Mekong bank erosion
In 2001 the Water Resources Research Institute of the Ministry of Agriculture and Rural
Development of Vietnam published the report “Study on forecasting and preventing bank
erosion for the Cuu Long River”, dealing with past and future bank erosion in the Mekong
Delta. This chapter gives an overview of the results of this study.
Damage between 1990 and 2000
In 2000 a flood occurred with an estimated return period of 70 years. Bank erosion due to this
and other floods between 1990 and 2000 caused 32 casualties and led to the destruction of 5
roads and several villages, comprising some 2,200 houses.
Erosion locations
There are quite a few maps which show the displacement of the Mekong river bed since 1890.
There are some 60 to 70 locations with an average bank erosion up to 10 m per year and more.
The historic flood of 2000 at one place resulted in 70 m of erosion.
Most erosion takes place in the 'upstream' half of the Vietnamese Mekong (comprising the first
130 km from the border with Cambodia, the Tien and Hau branches have not yet further divided
in this stretch), and mostly along the Tien branch.
The 6 most threatened locations are Thuong Phuoc (30 to 50 m of erosion per year), Tan Chau,
Hong Ngu, Sa Dec (all on the upstream reach of the Tien), Long Xuyen en Can Tho (both on the
upstream reach of the Hau).
Erosion mechanisms
The dominant erosion mechanism on the upstream reach of the river is toe scour with
consequent collapsing of the bank. On the downstream reach the tide causes an alternate inflow

20
and outflow of groundwater, which washes out minerals and fine soil particles from the banks,
after which sheet erosion occurs.
Predicted erosion until 2010
From the Ibadzade and Popov equations a specific equation was derived to calculate future
erosion along the Vietnamese Mekong. For the problem locations of Thuong Phuoc and Sa Dec
this equation yielded a land loss of respectively 120 and 140 ha. by 2010. For the 4 remaining
problem locations a land loss between 10 and 25 ha. were predicted.
Attempts to simulate the erosion at Sadec with a 1-D numerical river morphology model failed.
2-D or 3-D modelling is only expected to yield more results if based on a better understanding
of the erosion mechanisms.
Protection structures
To stabilize the riverbanks and avoid erosion at Tan Chau, Hong Ngu, Sa Dec and Long Xuyen,
revetments, gabions, groins and floating breakwaters screens are applied or proposed.
Conclusions and recommendations1. Continue to investigate and evaluate the human impacts on
the riverbank erosion on the Cuu Long River system.
2. A 1-D numerical river morphology modelling have been developed and applied to calculate
the erosion rate on the riverbank on the Tien River nearby the Sa Dec town. But the outputs of
the modelling are not reasonable and do not coincide with reality. The 1D river morphology
could not be able to compute the horizontal erosion. At least 2D numerical river morphology
modelling needs to develop and apply for prediction erosion on the Cuu Long River system in
next phase of the study. The erosion mechanisms are not clear and needs to clarify before
develop and apply any numerical modelling.
3. Continue to develop and perfect the technology for forecasting, warning, protecting the
erosion on riverbank on the Cuu Long river system.
Follow-up/ongoing study:
Program KC. 08: Enviroment and Natural Disaster Prevention ( National Program); project
kc.08.15Project title: Research on the causes and the solutions to prevent river bank erosion
and deposition for the Lower Mekong Delta River System (LMDRS). Project Duration:
October 2001 – September 2004. Cooperation Institutes (main):
- Sub-Institute of Geography – Center for National Science and Technology
- Sub-Institute of Water Resources Research Institute– Ho Chi Minh City
- Ho Chi Minh City University of Technology
- Vietnam National University - Ho Chi Minh City
- Southern Hydro-Meteology Station
- Offices of Science Technology & Environment, Agricultural & Rural Development in the
Lower Mekong Delta.

21
4. Mission to Mekong Delta and Red River
4.1 Approach (field visits and meetings)/reference to Appendices
The mission spent one week in the Mekong Delta and 4 days in the Red River area. The mission
participants from Netherlands and Vietnam are listed in Appendix I. The institutions visited and
persons met during this mission, and the time schedule of visits are mentioned in Appendices II
and III, respectively. The reports from field visits and discussion with local agencies are
included in Appendices IV to VI. Some facts on damage collected during the mission, hovever
not complete/not fully representative, are presented in Appendix VII.
The mission was not intended to make a professional consultancy report on the bank erosion
problems in Vietnam. The goal of the mission was to get a more or less reprentative picture of
these problems and to make (general) recommendation on the future tackling of these problems.
Due to the limited time it was not possible to make a full inventory of problems on bank erosion
in Vietnam. Therefore, to get a general (hopely, representative) picture of the problem it was
decided to visit a number of representative sites in the upper part of Mekong Delta (non-tidal
area) and in the lower part of the Red River area. Special attention and time was allocated for
discussion with persons working for a long time in the visited areas (praciticians) or doing
studies for these areas (researchers). The guids from Dike Deparment (DDMFC) provided
useful information on legal, organizational and planning aspects.
The direct findings of the mission members are reported in this Chapter. They are listed based
on disciplinary expertise of members (hydrodynamic, morphology, erosion, geotechnics, bank
protection), but also including their personal remarks/observations. he general observations by
the mission leader are also included in this Chapter. We have included all these personal
observations to avoid that we will miss some valuable information and also to show how
persons look to the problem from their own (disciplinary) perception.
The final analyse and findings are reported in Chapter 5.
The mission likes to stress once again that such short visit can not give the full overview of the
erosional problems in Vietnam (it can not replace the fundamental study on this problem).
However, it can be seen as a second opinion on this problem provided by an independent group
from abroad. It has provided the opportunity to confront the actual Vietnamese approach to this
problem with some foreign approaches/expertise, and has made possible to draw some
conclusions and recommendations in this respect. The mission hopes that their findings will be
of use for further improvements of the Vietnamese approach.

22
4.2 River hydrodynamic, morphology and bank erosion aspects
4.2.1 Introduction
The Mission has spent only a limited time along the Mekong and the Red Rivers. This did not
allow for a detailed study of their characteristics. Moreover, no studies were (made) available in
which the morphological features of the two rivers are described extensively. The Mission feels
however that on the long run river training measures can be successful only when a good
understanding of the behaviour of the considered river is available.
Therefore a summary of the characteristics of both rivers is given in this Chapter, based on the
limited data available to the Mission. Additionally some observations made during the various
visits are included in this Chapter as well. This is supplemented by a number of
recommendations on data collection and the establishment of data base. Finally the question is
addressed whether bank erosion along the two main rivers in Vietnam has increased in the last
decade, and if so, what could be the most probable cause(s).
4.2.2 Morphology of Mekong and Red River, similarities and differences
Some morphological features of the Mekong and Red River are discussed hereafter, based on
what was easily available (either as map or report, or via the Internet). First per river and in the
last part of this Section a comparison between the two rivers is made.
Mekong River
For information on the Mekong River use was made of in particular a report by Le Manh Hung
& Dinh Cong San (2002) and a paper by Ngaonh & Akira (2003), although also some other
sources of information were used.
The catchment area of the Mekong is about 800,000 km2, of which about 65,000 km2 is in
Vietnam. The total length of the Mekong River is about 4200 km, of which about 200 km is in
Vietnam.
The planform of the Mekong River is characterised by two separate branches plus a number of
lateral connections, mostly man-made. The river having two parallel branches is quite unique,
and is probably is due to the fact that in geological times, the Mekong was not yet connected to
the Tongle Sap lake and the Bassac river. In this assumption the Tien River is the remnant of the
”original” Mekong River, whereas the Hau river is the original drainage channel of the Tongle
Sap system.
The Tien river is slightly meandering, whereas the Hau river is virtually straight. Probably there
is a strong influence of neotectonics and active faults on the river planform. This is shown in
Figure 5. The Hau River and the lower reach of the Tien Rivers coincide with active faults. Near
Tan Chau an active fault crosses the Tien River, and this may be the cause of the almost square
angle the river channel is making at that location (see also Section 2.3). In the other reaches of
the Tien River the curvature of the river bends is much larger and appear to correspond to a
“normal” alluvial river (see also Section 4.2.3).
The average discharge of the Mekong River is about 15,000 m3/s. Flood discharges of the two
branches combined vary over the years between 20,000 and 35,000 m3/s. Hence the ratio
between the flood and average discharges is small. The difference in flood levels in Vietnam is
not excessive. The maximum water levels near the Cambodian border are about 4 m above

23
Mean Sea Level (MSL) (see the Figures 6 and 7), and the lowest water level is almost MSL.
This implies a flood-low water range of 4 m decreasing in downstream direction. As the length
of the Tien and Hau Rivers in Vietnam is about 200 km each, water level slopes are very gentle,
and are even gentler during the low flow season.
Figure 5. Faults and their possible impact on the planform of the Lower Mekong River
Figure 6. Water levels in the Lower Mekong during the year 1982

24
Figure 7. Flood levels Lower Mekong River
The bed material of the Mekong River is fine sand. Le Manh Hung & Dinh Cong San (2002)
show (in their Figure III.a) that the D50 of the Tien channel reduces in Vietnam from 0.25 to 0.1
mm. The sediment load of the Mekong River is low. According to data provided in Jansen
(1979), the sediment load of the river at its mouth is about 80 million ton per year. In view of the
catchment area of about 0.8 million km2, this corresponds to an average denudation rate of
(only) 0.07 mm/year. Some other rivers in South-East Asia have much higher denudation rates
(Yangtse Kiang 0.2 mm/yr, Yellow River almost 2 mm/year), but the Chao Phrya has a
comparable denudation rate (0.05 mm/yr). Average sediment concentrations in the Mekong are
about 200 ppm. This all suggests a river, which is morphologically not extremely active.
In the Tien River some large islands are present where the river bifurcates. These islands appear
to be quite stable. They are characterized by dense and old vegetation and many settlements.
These islands, the areas in between the two Mekong channels and the floodplain on both sides of
the river are flooded yearly. Flood protection is virtually absent, although the Mission feels that
the road building which is going on over the last decade, is creating obstacles to the flow. The
population is apparently adjusted to the regular flooding pattern.
Local bank erosion was observed by the Mission at a number of places, mostly along the Tien
River. Also at the upstream part of some stable islands bank erosion was observed. Typical bank
erosion rates are in the order of 10 m/yr and apparently do not exceed 40 m/yr (see Appendix
VIII, which is based on information provided by the provinces visisted by the mission). The
alluvial reach of the Mekong is composed of sand with mud and clay (see Figure 8). Erosion
depth of up to 30 m have been observed in front of eroding banks.

25
Figure 8. Some information on the composition of the Mekong Delta
During their stay the Mission visited some locations where bank protection works had been
carried out (Tan Chau, Long Vinh) and also one location where bank protection works were
under construction (Sa Dec). The bank protection works are constructed where important areas
are in danger of being eroded. At other less important locations population is removed from the
eroding areas and re-located more inland.
Red River
Some data on the Red River were found in Experco (1994) and Nguyen Tuan Anh & Tran Xuan
Thai (2000), and these are summarized hereafter. The total catchment area of the Red River
(locally called the Hong River) is about 169,000 km2, of which about half is located within
Vietnam. The total length of the Red River is about 1150 km, of which about 500 km is in
Vietnam. The main tributary is the Da River with a watershed area of about 53,000 km2. Most of
the river basin is mountaneous and of a high altitude (70% over 500 m). The forest cover
continues to decrease. Some data for the Da River show that the forest cover has decreased from
77% in 1943 to 9% in 1981. This will have had an enormous impact on peak flows and sediment
load of the river. Only in the lower part of the Red River the river is flowing through an alluvial
valley.
The Red River is flowing through a seismically active area. The general direction of the river
seems to coincide with a fault like is the case with the Hau River as part of the Mekong. The
Red River zone is characterised by a substantial earthquake risk, high levels risks being incurred
by the various structures built in the Red River area (see Experco (1994). In the alluvial reach is
made up of quaternary sediments and comprise of (in the Hanoi area, see Experco (1994)) a 1 to
6 m deep layer of clayey silts overlying fine sands and at some places a layer of impervious
clay. During floods the floodplains are continuously supplied with sediments rich in organic
matter. Experco (1994) mention representative particle sizes of 0.2 mm for the silty snad and 0.5
mm for the sand.
The average annual discharge of the Red River is about 4300 m3/s, whereas the maximum
recorded flood in the last decades is about 34,000 m3/s and the minimum (after construction of
the Hoa Binh dam on the Da River) about 1200 m3/s. The most striking aspect of the river is the
great difference between the flood and the low-flow discharges (about 20:1). The water level

26
variation is between 1 to 3 m above the floodplain level during flood to 5 to 6 m below
floodplain levels during the low flow season. The flow width during low flow varies between
200 and 600 m, while during flood the width of the river reaches 2 to 3 km (Experco, 1994).
During flood the Red River is characterised by a single channel, which is essentially
meandering. During low flow a transition to braided is noticeable, and islands can be observed.
The location of these islands and of the main flow channel is changing over the years, and this
contributes to changes in bank erosion location along the Red River. The sandy islands are quite
unstable and hence they allow only for some minor vegetation.
The slope of the Red River is slightly affected by the discharge of the river. Only in the lowest
reach of the Red River is the slope affected by the effect of the sea level. During low flow the
slopes of the Red River vary between 3 and 6 cm/km. During flood the slope of the reach
upstream of Hanoi increases to about 10 cm/km. Sediment concentrations of about 2,000 ppm
have been observed, indicating that at present the Red River is morphologically much more
active than the Mekong River. According to Nguyen Tuan Anh & Tran Xuan Thai (2000) the
Red River is aggrading with a rate of 1.5 cm/year.
Bank erosion along the Red River is in the order of several tens of meter per year at those
locations where bank erosion is actually taking place (of course at other locations bank growth is
taking place or islands are increasing in width). According to Experco (1994) the bank erosion is
governed by two aspects notably high velocities along the bank and the orientation of the
velocity vector towards the bank (and it is expressed in these two parameters because Experco
was using 2D mathematical flow computations; see also hereafter and Chapter 4 for other
parameters of importance).
Nowadays a continuous dike system is present along the Red River. In some cases a retired
embankment is present as well. At many places bank protection works are present. The Mission
did not visit all locations where bank erosion was or is active and moreover the river was
approached from one side only. Also the maps provided were for one province (and hence for
one bank of the river) only. It is amazing that no maps were available where the bank protection
on both banks was marked.
Comparison of the Mekong and the Red River
In Table 1 a comparison between the Mekong River and the Red River is given. The differences
between the two rivers should and will have an impact on the long term strategy for coping with
the river and for the river training works. See Chapter 5.
Aspect Mekong River Red River
Catchment area (km2) and length of
river (km)
Catchment area 800,000 km2
Total length of river 4200 km of
which about 200 km in Vietnam
Catchment area 143,600 km2
Total length of river 1150 km, of
which about 500 km in Vietnam
Number of channels Two separate branches plus a
number of lateral connections
Single channel
Planform Tien river slightly meandering; Hau
river straight
Meandering, transition to braided
Geological setting Alluvial plain with strong effect of
fauls and neotectonics
Alluvial plain with possibly some
impact of geology
Discharge (m3/s) Average 15,000 m3/s; flood
between 20,000 and 35,000 m3/s
Average 4,300 m3/s; flood
discharge about 35,000 m3/s

27
Difference between flood level and
low water
Maximum about 3 m and reducing
towards the sea; tidal motion
appreciable even near Cambodian
border during low flows
Difference much larger; and tidal
motion not appreciable in more
upstream reaches
Sediment load Low (0.07 mm/year = 80 million
tons/yr); average concentration 170
ppm
Higher (about 0.6 mm/yr = 114
million tons year); average (flood?)
concentration 2,000 ppm
Islands and bars Almost stable islands with dense
vegetation and many settlements
Unstable sand islands with some
vegetation; nevertheless about
600,000 people living on islands (to
be checked)
Average bed level Stable Aggrading (1.5 cm/yr?)
Flood protection Virtually absent (roads) Continuous dike system
Bank protection Some local bank protection works Along all (?) outer bends
4.2.3 Morphological causes of bank erosion along Mekong and Red Rivers
During their stay in Vietnam the Mission discussed with several institutes their understanding of
bank erosion along the Mekong and the Red River and the ongoing studies into bank erosion
along these rivers. Based on these discussions the Mission holds the view that it is urgently
needed that studies are carried out to better understand the cause of the different types of river
bank erosion in order to come up with sustainable solutions to cope with this bank erosion.
The causes of bank erosion along rivers can be discussed in different ways. Experco (1994)
attempts to explain bank erosion from high velocities (in excess of the critical velocities) and the
direction of the velocity towards the bank. According to many handbooks the immediate cause
of bank erosion along rivers is the peeling off of sediments (banks of loose material) or the
collapse of a bank due to a bank slide (cohesive banks). Along the Red River bank slides appear
to be the major cause of the bank erosion. Along the Mekong River this cause is less clear (see
also Section 4.2.4).
With a slightly different perspective, bank erosion of cohesive banks can also be explained by
the occurrence of deep scour holes in front of the bank, which might cause deep-seated slides.
This might be enhanced by over-pressure of the groundwater after the recession of the floods.
The cause of deep scour holes (in the Mekong and in the Red River scour holes with depths in
the order of 30 m have been observed) can be either outer bend erosion, confluence scour or
other (see e.g. the overview in Hoffmans and Verhey (1997).
One level higher the cause of bank erosion can also be explained in terms of the planform
development of the river. The development of a curved channel will induce outer bend scour,
which in turn will cause deep scour holes in front of the bank which will result in a collapse of
the bank by a bank slide. The ultimate cause of the bank erosion in this reasoning is thus the
overall planform development of the river, including the occurrence of islands. These islands
become apparent during the low flow period, but probably they are also remain to be present
during the flood period as well.
Hereafter the cause of bank erosion is discussed in terms of the overall planform development.
For the Mekong River bank erosion occurs mainly along the Tien channel. The normal condition
is bank erosion along the slightly meandering course. In Figure 9(a) this shown for the Tien
channel near Sa Dec. The bank erosion is located in the beginning of the bend, and it continues
unless checked by bank erosion works.

28
Another example of bank erosion along the Tien channel of the Mekong is shown in Figure 9(b).
Here bank erosion is mainly along the upstream part of the island, but there is also some bank
erosion of the bank downstream of the island. This can be interpreted as a slow movement of the
island in downstream direction. In this case two islands are present and the channel between
them appears to scour.
The final example of bank erosion along the Tien channel of the Mekong River is given in
Figure 9(c). As is shown in Figure 9 an active fault is present which crosses the river. As
hypothesis it is suggested here that the fault causes a horizontal movement, and this would imply
that the sharp bends (90 degrees, well in excess of the other bends of the Tien channel) and the
corresponding deep scour holes and bank erosion at Tang Chau and at Nong Ngir are in the end
caused by neotectonic movements.
In the Red River also different types of bank erosion can be identified. The most common are
linked to bank erosion along a meandering river. In the Red River the amplitude of the meanders
is more developed and the radius of curvature is smaller than in the Mekong River (see Figure
10). The location of the strongest bank erosion can be concluded from the location of the bank
protection works, which are also shown in Figure 10. Near the confluence of the Red River with
the Da River also confluence scour may play a role. Downstream of the Hoa Binh dam
degradation is taking place, which induces collapse of the riverbanks.
The Mission observed that within Vietnam limited understanding is available on the ultimate
course of river bank erosion. The Mission strongly advocates that detailed studies should be
carried
Note: Geotechnical aspects of bank erosion are discussed in Chapter 4.4.

29
(a) Bank erosion along the Mekong (Tien channel) near Sa Dec
(b) Bank erosion of upstream part of island in Tein channel near Cao Lanh
(c) Planform changes and bank erosion (in red) in the Mekong (Tien channel) near Tan Chau and Nong Ngir
Figure 9. Different types of bank erosion along the Mekong River in relation to planform
development

30
Figure 10. River planform and bank protection works along the Red River
4.2.4 Monitoring of bank erosion
Monitoring of bank erosion is required for taking timely action in terms of the implementation
of bank protection measures or to evacuate people from endangered areas. Monitoring of bank
erosion however is also needed to get a better understanding of the causes of bank erosion. Once
these causes are better understood also better prediction methods can be developed.
The mission is not fully aware of where and how frequent bank erosion along the Mekong and
the Red River are measured. Still the Mission proposes, if not done already, to measure bank
erosion on a yearly basis. This can be done by carrying out bank line surveys. Two modern
techniques can be used in this respect, notably:
• hand-held DGPS
• satellite imagery
Hand-held DGPS provides nowadays a very elegant and cheap method to determine bank lines.
By walking along the bank and regularly determining the location, bank lines can be established
easily and very accurate. Possibly even normal GPS might suffice, but this should be checked. A
disadvantage of this method is that only bank lines are determined, whereas for an improved
understanding of bank erosion processes also the overall planform, inclusive the location of
islands and the main channel is required. In this respect satellite imagery is much more usefull.
Satellite imagery has another advantage: also historical data (from 1973) are available which
implies that already on the basis of these historical data a prediction method can be developed.
However, the oldest satellite imagery are not very accurate (pixel size of 200 m compared to 10
m for the present generation of Landsat satellite (sensors) and the most recent (and accurate)
images are expensive. By using an intelligent mix of these two methods, yearly bank lines and

31
hence yearly bank erosion rates can be obtained in addition to more general morphological
characteristics, which determine the bank erosion.
4.2.5 Need for setting up of data base
The Mission observed that data relevant for bank erosion and its cause are scattered and not
stored in one data base, which would allow easy reference and could form the basis for good
studies into the causes of bank erosion.
Data for studying river processes can be divided in (1) hydrological and sediment transport data
and (2) morphological data. In Vietnam hydrological data are collected and stored by the
Ministry of Water Resources. These include:
• stages
• discharge
• sediment concentrations
and these can be used to determine rating curves, slopes and sediment rating curves, and
subsequently be utilized to determine statistical parameters of stages and discharges plus
summarized information on sediment transport characteristics of the rivers, like sediment loads.
In addition to these data the Mission proposes to set up a special data base in which
morphological data of relevance for bank erosion, bank protection works and the response of the
rivers to bank protection works are collected. This data base should include the following
information:
• historical and recent maps
• satellite images
• cross-sections and special sounding maps
• borings and other information on bank composition
• bank erosion rates
• scour depths
• information on bank protection works: when constructed, typical cross-section,
construction method, as-made drawings, subsequent soundings, etc. figures on
maintenance over the years, etc.
• drawings with impact of river training works on upstream and downstream reaches
All this information should be included for as many years as possible. The proposed data base
can be set up at the Ministry, but probably it is advantageous to make the Water Resources
Institutes in Hanoi (for the Red River) and in Ho Chi Minh City (for the Mekong River)
responsible for the filling of the data base with data and with the maintenance of it. A special
study should be carried out how this data base should be set up, though it seems logic to use a
GIS like ArcInfo or Arc as basis for this type of information (which is mostly spatial).
4.2.6 Has bank erosion increased over the years and if so, what are the possible causes?
During the visits of the Mission some provinces in the South were claiming that bank erosion is
increasing (and more specifically since 1994). Some possible causes for an increase in bank
erosion are listed below:
• increase of frequency of large floods, e.g. as an effect of climate change
• increase flow in main channels due to reduction in overland flow
• change in flow distribution over channels in Mekong delta (acknowledged for
upstream reaches in Cambodia)

32
• (substantial) degradation of river bed downstream of dam in Da River resulting in bank
erosion
• change in flow distribution at confluence of Da and Red rivers
• psychological: more damage due to development (houses, roads etc.)
Some observations on this matter are given hereafter, partly on the basis of some data on yearly
floods in the Mekong River in the period 1978-1998, provided by the Southern Institute for
Water Resources:
• the flood discharges of the Mekong River do not seem to have increased in the period
1978-1998 (see Figure 11(a));
• the distribution of the flow over the two Mekong channels does not seem to have
changed (See Figure 11(b)).
(a) Combined flow of Tien and Hau channels
(b) Relative flood discharge distribution
TC = Tan Chau, VN = Vam Nao, MT = My Thuan, CD = Chau Doc, CT = Can Tho
Figure 11. Some information on flood peaks of the branches Lower
Mekong River in the period 1978-1998
The Mission proposes to do more detailed studies on these and other aspects of bank erosion
along the Mekong and Red Rivers, once the data base as proposed in Section 2.6 is available and
can be used.
4.2.7 Response of river to bank protection works and consequences for future
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
3
5
7
9
11
13
15
17
19
21
Year
Q station / Q Tan Chau
VN/TC
MT/TC
CD/TC
CT/TC
0
5000
10000
15000
20000
25000
30000
35000
40000
123456789101112131415161718192021
Year (start ing from 1978)
Flood discharge (m3)
TC+CD
MT+CT

33
As far as the Mission understood bank protection works in Vietnam are carried out on the basis
of where the need arises. When an important reach is eroding bank protection works are
implemented over a certain length. Once bank protection works have been implemented,
economic development of the protected areas is accelerated. This creates a commitment to
properly maintain the bank protection and to prevent that the protected area is eroded in
subsequent years. Bank protection works however cause changes in the river system, which are
noticeable both downstream and sometimes in later stages also upstream and which may
necessitate additional bank protection works. The continued morphological development of the
reach with bank protection works may thus induce the need for additional river engineering
works both upstream and downstream. It is important to realize this in advance in view of
decision making on protection, for proper siting of the bank protection works and to limit the
funds required later for maintenance and extension of the bank protection works
An illustrative example is the case of Chandpur
bank protection in Bangladesh. Figure 12 shows a
processed satellite image of the Lower Meghna
along which Chandpur town is located. The
number of inhabitants of Chandpur is about 1
million people.
The Lower Meghna is a very large river which
carries the combined flow of Ganges,
Brahmaputra and Upper Meghna Rivers. The
flood discharge of the river exceeds every year
100,000 m3/s. Probably due to neotectonics the
Lower Meghna River is moving Eastward. Due to
this shift of the river, the town of Chandpur
becomes more and more exposed. Chandpur
Town is being defended since mid 20-ies, but the
upstream reach continues to erode. Since the 20-
ies the scour depths are increasing and have now
reached 70 m below floodlevel. At some places
the slopes of the present bank protection are in
the order of 1:1 and very unstable. The yearly
extensive maintenance works are much in excess
of initial estimate of the required maintenance.
By now it is realized that also upstream bank
protection and river training measures are needed, but these will be very expensive and in a
recent study ithe option to abandon the town was included and not immediately rejected!
Mirroring the situation around the diagonal of Figure 12 shows a striking resemblance with the
conditions at Tan Chau and may indicate the future conditions at that location. At the same time
though it should be realized that the Mekong River is morphologically much slower than the
Lower Meghna. Nevertheless the Mission holds the opinion that upstream and downstream
effects of the bank protection works and morphological developments upstream and downstream
of planned works should be taken into account in the design of the bank protection works. This
also stresses the need for the development of a long-term strategy for bank protection works
along both the Mekong River in Vietnam and the Red River.
Figure 12 Chandpur along Lower Meghna
River

34
4.3 Prediction of bank erosion
4.3.1 Introduction
During its visit to Vietnam the Mission has assessed how in Vietnam predictions are made for
future bank erosion. Such predictions are required for decisions regarding the implementation of
bank protection works or evacuation of the threatened population. In the following sections
some observations made in Vietnam are presented and some suggestions for the improvement of
the prediction of bank erosion rates are given.
4.3.2 Present practice in Vietnam and ongoing studies
Not much information is available on how prediction of future bank erosion rates are made in
Vietnam. As far as understood by the Mission, until now predictions of future bank erosion rates
are in particular an extrapolation of the river behaviour in the past.
Recently however, some interesting studies into bank erosion along the Mekong River were
carried out by the Southern Institute for Water Resources in Vietnam. This study was funded by
the Ministry for Science and Technology and the most important findings were published
recently in Le Manh Hung & Dinh Cong San (2002).
Although the report is in Vietnamese, still the Mission observed that promising results have
been obtained which pave the way for improved and more scientifically based prediction
methods for bank erosion rates. The Mission would like to point at two highlights from this
study. Figure IV.a of Le Manh Hung & Dinh Cong San (2002) the yearly bank erosion rates at
Sadec are plotted versus the parameter R/B (radius of curvature over width of the river). This
figure is copied here as Figure 13. This result is in line with the work of Hickin & Nanson
(1984), but the interesting aspect of this study is that the data included in this figure relate to the
bank erosion at any point along the bend, whereas the Hickin & Nanson (1984) method formally
only only holds for the one location where the bank erosion is maximum.
Figure 13. Local bank erosion rates of the Mekong River at Sa Dec versus
relative curvature (source Le Manh Hung & Dinh Cong San (2002))

35
Results like included in Figure 13 suggest that this method can be elaborated to a prediction
method for a whole bend.
Another interesting idea in the study of Le Manh Hung & Dinh Cong San (2002) is the use of a
method initially proposed by Popov (no reference available) is used which reads:
max
0
i
xi HH
F
BLT H H
α
⎡⎤
−
=⎢⎥
−
⎣⎦
(4.1)
where Bxi = yearly bank erosion at point i along the bend (m/yr), F = total eroded area (m2), L =
length of eroding area (m), T = time elapsed between the two subsequent (…), Hmax1 = bend
scour depth at point i, H = average water depth, and H0 = maximum bend scour depth. Although
for any application still F and Hmaxi have to be predicted, this method when found appropriate
opens the possibility to predict the variation of the bank erosion along a bend.
The Mission understands that the study of Le Manh Hung & Dinh Cong San (2002) will be
continued with another study with funding from the same source, and they are convinced that
these studies, in combination with results from studies abroad (as summarized in Section 4.3.3)
and the introduction of numerical models (see Section 4.3.4), will result in improved prediction
methods for bank erosion rates.
4.3.3 Methods used elsewhere
There is not much theoretical or even empirical work related to quantitative prediction of bank
erosion rates. Brice (1984) found that bank erosion rate increased linearly with drainage area for a
number of USA rivers. Here three different approaches are discussed. One approach deals with
estimates of yearly erosion and was developed by Hickin and Nanson (1984) for rivers in Canada.
Klaassen and Masselink (1992) used the same approach for the Brahmaputra River in Bangladesh.
The second approach was developed in Bangladesh by Klaassen et al (1993) and improved and
updated by Sarker and Khayer (2002). The third approach, included in numerical models in which
the bank erosion is simulated, links the momentary bank erosion to the local conditions (either
flow or bank height).
The first method was developed by Hickin and Nanson (1984), who did an extensive study of
bank erosion rates along sand-bed rivers in Western Canada using aerial photographs. They found
that erosion rate was a function of the radius of curvature to width ratio R/W, with a maximum at
R/W = 2.5. This can be written as:
(
)
2.5
M R/B = M . f(R/B) (4.2)
For f(R/W) an empirical relation was derived:
for 1 <R/B < 2.5 f(R/B) = 2/3 (R/B-1) (4.3a)
for R/B > 2.5 f(R/B) = 2.5 B/R (4.3b)
The maximum erosion rate occurs for R/B = 2.5 and is defined as M2.5 (in m/year) and this
maximum rate is proportional to total streampower Ù, which is defined as:

36
55
= Q = g Q i
h
τ
ρ
Ω
(4.4)
where Ω = total stream power (in Watt/m'), Q5 = discharge exceeded once in 5 years (m3/s).
M2.5 is inversely proportional to a bank-strength parameter YB (dimension N/m2) which is a
function of bed material size (given as a figure in Hickin & Nanson (1984)):
2.5
B
=
Mh.Y
Ω
(4.5)
The use of this method in other parts for other river systems elsewhere must be done with care but
can give some idea of possible bank erosion rates. Application in Vietnam can be considered once
the method has been re-calibrated on the basis of bank erosion data of both the Mekong and the
Red River. This may not necessarily result to the same calibration coefficients, in view of the
different characteristics of the two rivers (see Chapter 4.2.2).
The second method was developed for the Jamuna (=Brahmaputra) River, which is one of the
largest braided rivers in the world. The river is very dynamic, complex and chaotic in nature and
its changing bank lines cause major problems for the people living along its course and on its
islands. Comprehensive studies on the Jamuna River started only in the 1960-ies. Studies by
Coleman (1969) and Bristow (1987) are considered as milestones in the growing understanding
of the morphological processes of the complex river system. Later, in the late 1980-ies and early
1990-ies, studies intensified in the context of the construction of the Jamuna Bridge and the
Flood Action Plan (FAP). These studies gave particular attention to the interaction between the
river and the different types of interventions and a number of empirical prediction tools were
developed on the basis of low flow satellite imagery (see e.g. Klaassen & Masselink (1992) and
Klaassen et al (1993).
Most of the empirical prediction tools developed were derived using dry-season satellite images
from the mid 1970-ies to the late 1980-ies, having a coarse resolution of 80 x 80 m. Although
the Jamuna River is very dynamic, these coarse resolution images were not sufficient to estimate
annual changes with reasonable accuracy. Therefore, intervals of 2 to 3 years were used. Due to
the very chaotic and complex nature of the river, however, it was recognized that the longer the
time interval, the more uncertain the predictions.
The study by Sarker and Khayer (2002) aimed to develop and update those prediction tools
using a series of annual dry-season satellite images with a relatively fine resolution of 30 x 30
m, collected during the period 1992-2000. Based on these images, more accurate predictions
could be made of such morphological processes as: channel abandonment, migration of
bifurcation, and bank erosion along outflanking curved channels. The study introduced so-called
“sedimentary features” as a new prediction tools, helping morphologists to better interpret the
behaviour of the river. These sedimentary features are contraction bars, sharpened bars, sand
wings, sand tongues and “bank side” bars. Identified from satellite images, they can be
considered indicators of morphological behaviour. Data extracted from the satellite images are
divided into two groups based on the presence and absence of sedimentary features, and
correspondingly, two sets of prediction tools were developed for each of the above mentioned

37
morphological processes. A summary of prediction techniques for the different morphological
processes with and without sedimentary features is presented in the Table 2.
Bifurcation
migration
(m/y)
Bank erosion
along concave
bend (m/y)
Bank erosion
along
straight reach
(m/y)
Bank erosion
along
convex bend
(m/y)
Channel
migration
(m/y)
Presence
of
sedment
ary
feature
Probability (p) of
channel
abandonment as a
function of
deviation angle
(ϕ) Avg.
value
Range Avg.
value
Range Avg.
value
Range Avg.
value
Range Avg.
value
Range
Yes p=0 for ϕ ≤20°
p=1 for ϕ ≥65°
1,200 200
to
2,500
230 60
to
400
215 90
to
330
175 50
to
410
950 0
to
3,200
No p=0 for ϕ ≤20°
p=1 for ϕ ≥110°
670 -1,400
to
2,600
130 0
to
330
--- --- --- --- --- ---
Table 2 Advantages of including the effect of sedimentary features in the prediction methods for
the Jamuna River (Sarker & Khayer, 2002)
Table 2, copied from Sarker & Khayer (2002),shows that the presence of sedimentary features
facilitates predictions with higher accuracy. Moreover, in addition to the predictions for channel
abandonment, bifurcation migration and bank erosion along concave bank, they make it possible
to predict for bank erosion along the straight and convexly curved reaches and channel
migration within the braid belt.
An example of the prediction of bank erosion along a curved channel, which is the most
relevant for the conditions in Vietnam, is given in the below Figure 14.
* ‘m’ stands for migration rate
Figure 14 Prediction method for bank erosion along outer bends in the Jamuna River as an
example of the methods developed by Sarker & Khayer (2002)
Type of
bifurcation
Mi
grat
i
on rate
*m = 670 m/y
m (90%) = -200 to 2000
m/y
l
ong
(L>20km)
I
s any sedimentary
f
eature available
within 3 km
downstream of the
bifurcation?
m
(50%)
=
670
m
/
y
m (90%) = -1400 to 2600 m/y for L = 0 to 10
km
m (90%) = -500 to 1800 m/y for L = 10 to 20
km
no
yes
m
(50%)
=
1200
m
/
y
m (90%) = 200 to 2500
m/y
s
h
ort
(L
<
20km)

38
The use of the empirical predictions based on low flow satellite imagery either without or with
sedimentary features relating to morphology of rivers and developed for the Jamuna River can
be effectively used to make predictions of bank erosion rates and the celerity of other
morphological processes. In principle it can also be applied to the Mekong and Red River. It
should be studied whether it is feasible and required to include sedimentary features into the
prediction method.
The third approach uses local conditions like the local shear stress, the local velocity the local
sediment transport and/or the local bank height for estimating the bank erosion rates. Typically
also here calibration coefficients have to be introduced which represent a.o. the bank properties.
Althought the approach may be more generally applicable, these calibration coeffieicnets have to
be determined for each river separately.
Prediction methods which take into account the local conditions have been proposed by e.g.
Ariathurai and Arulanandan (1978), Crosato (1990), Mosselman (1992), DHI (1996) and
Shishikura (1996). A typical example of such a predictor is the predictor of Mosselman (1992):
B
c
ac
-
n = E
t
δ
τ
τ
δ
τ
(4.6)
where nB = bank line position and in the right side the local shear stress and the criritical shear
stress are present. The coefficient Ea has to be calibrated to the local conditions.
DHI & SWMC (1996) have proposed an alternative prediction method which reads:
b
b
b
zs
= · + ·
tH
B
nt
αβ
∂
∂
∂∂
(4.7)
where zb = bed level near the bank, sb = bed load transport near the toe of the bank and Hb =
height of the bank. The first term agrees with a proposition of Mosselman (1992), but the
generally accepted term with excess shear strength (see Equation (4.1)) is missing and the
second term with transport capacity is questionable because the sediment transport capacity (sb)
influences only indirectly the bank erosion mechanism and this indirect influence is already
accounted for by the sediment balance. The DHI bank erosion relation implies that higher banks
(larger Hb) are less prone to erosion. This is in accordance with Hickin & Nanson (1984).
4.3.4 Use of mathematical models
Recently mathematical models have been introduced which also predict bank erosion and the
subsequent planform changes. These models are all 2D models, in which the following features
are included:
• orthogonal or curvi-linear grid (where in bank erosion studies the curvi-linear grid is
preferred)
• 2D flow computation
• helical flow computed later (and impact of helical flow on main flow not accounted
for; acceptable for not too sharp bends)
• bed load
• suspended load via the advection-diffusion approach
• bank erosion.

39
Different models are available, the best known commercial ones being Delft-3D Rivers from
WL/Delft Hydraulics and MIKE21 from DHI (Danish Hydraulic Institute). The Mission
understood that MIKE21 will be implemented in both Institutes for Water Resources Research
in Vietnam under a DANIDA funded project. In MIKE 21 C (the curvi-linear version of
MIKE21), the bank erosion can be simulated in parallel with the sediment transport and
hydrodynamic simulations. In each time step, the eroded bank material is included in the
solution of the sediment continuity equation. The bank erosion model incorporated is the one
already described in Section 4.3, notably:
b
b
b
zs
= · + ·
tH
B
nt
αβ
∂
∂
∂∂ (4.8)
The extra sediment which is discharged into the river from this source is included in the
sediment continuity equation. The accumulated bank erosion causes a retreat of the bank line
position and thereby also in the extent of the modeling area. The MIKE 21 C incorporates these
plan form changes by re-generating the curvilinear grid during the simulation when the bank line
changes exceed a certain pre-defined threshold. In this manner, the morphological model
becomes a plan form model.
The Mission welcomes the introduction of this numerical model, but warns that such models are
not a panacee. Such models are as good as the ingredients on which they are based. In particular
the bank erosion modelling part needs to be thoroughly tested and calibrated versus
observations. In this respect it would be useful to use MIKE21C in the ongoing research project
on bank erosion.
4.3.5 Need for additional studies
For the improvement of methods for prediction of bank erosion rates, a good understanding of
the underlying processes is a must. The Mission suggests that in addition to the further
continuation of the study of Le Manh Hung & Dinh Cong San (2002), a number of other studies
should be carried out. These should include, but are not limited to only:
• study of scour phenomena in the Mekong and the Red River, linking scour depths to
overall planform features, to come up with prediction methods for scour depths as the
trigger for bank erosion
• more studies into the time dependency and the probabilistic nature of scour processes
• studies on the basis of low flow satellite imagery to develop prediction methods,
without or with the inclusion of sedimentary features
• comparative study of bank erosion features in the Mekong and the Red River
• modelling and prediction of geotechnical stability for sites where bank protection
measures are considered.
(NB. some additional remarks are given in Chapters 4.4 and 4.5)

40
4.4 Geotechnical aspects
The basic causes for bank erosion are the tractive forces due to currents and waves. Primarily
these forces result in surface erosion (grain by grain erosion). However, because the impact of
the hydraulic loading often is higher at deeper levels at the riverbed and because of the
difference in erosion resistance of the different soillayers, in many cases pure surface erosion
will be accompanied by steepening of the river bank. Geotechnical instabilities may therefore be
evoked as a more or less interrupting part of the total (combined hydraulic-geotechnical) bank
erosion process.
Many aspects, generic but especially local, may influence the response of the riverbanks among
which the features and size of bank sliding and slumping and the rate of bank retrogression.
Specific local conditions are also important for the suitability and applicability of the mitigating
bank protection measures. In the foregoing chapters an overview of the important aspects in
relation to “causes” and “bank protection” is given. Because the geotechnical features were
described in a more qualitative way up till here this section 4.4 treats the (entries to) a
quantitative assessment of bank stability in some more detail. In addition information on bank
erosion processes is described in supplements 1, 2, 3 and 4.
After a brief description of the geological structures of the Mekong delta and the soil types in
4.4.1, the main geotechnical aspects and some possibilities for modelling and stability prediction
are described in 4.4.2 and 4.4.3 respectively. In section 4.4.4 special attention is paid to the flow
slide mechanism which may initiate bank failure involving large bank retrogression within a
short period of time. Therefore, in locations where this mechanism may occur, the consequences
of bank erosion could be very serious. In section 4.4.5 some observations and experiences
acquired during the meetings and site visits in the Mekong and Red river deltas are described
and illustrated.
Bank protection measures are descibed and discussed in section 4.5.
4.4.1 Geological structures and effect on soil characteristics and erosion mechanism
The present surface level of the flat Mekong Delta plain varies between 1 to 5 m above mean sea
level. The depositional sequence in the delta consists of thick silty to sandy layers originated
during the Eocene to Pleistocene period, overlain by Holocene sediments of variable character.
The depth of the Pleistocene – Holocene boundary ranges from 15 m near the Cambodian border
and the Southeast region to about 110 m in the coastal areas. This geological sequence together
with the relatively dynamic character of the Mekong rivers and their tributaries during the last
millennia mean that the erosion processes in and along the rivers are predominantly governed by
the young Holocene sediments.
The Holocene sediments consist of clay, silt and sand, ranging from very clayey soil to clean
sand. As in most deltas and estuaries the stratification is often very complicated and
heterogeneous. For instance along the Tien river fine to middle sand is present below the top
clayey or silty layers almost everywhere between the Cambodian border to Vinh Long. However
the top boundary of the sand deviates to a high extent.

41
The dynamic meandering character of the main and secondary rivers in the Mekong Delta
suggests that the present upper 10 to 20 m are deposited very recently from the geological point
of view. It means that the consistency or in-situ density of the different types of sediments
generally will be relatively low. As a result the soils will be rather vulnerable to erosion attack
and bank instability and in case of sand, under special conditions, susceptible for liquefaction.
From the geotechnical point of view the distinction between clay and sand is very important
with respect to bank erosion and bank stability. Grain to grain surface erosion is very likely in
granular soil as sand and gravel because of the absence of cohesion whereas the phenomenon is
unlikely in clay because of cohesive resistance. Therefore, because surface erosion progresses
slowly in clayey banks, in many cases erosion will progress by means of mass failures due to
undermining in the sand layer underneath the clay. These mass failures may have different
dimensions from one large failure to successive smaller ones.
Another difference between clay and sand of great relevancy concerns the permeability to
groundwater flow. Within the time scales involved in the erosion and bank stability phenomena
this difference in permeability coefficient (factor between 10-2 to 10-5) leads to a complete
different response with respect to seepage and in relation to origin and dissipation of excess pore
water pressures.
The variety in alluvial stratification is illustrated in Figure 15 with the soil profiles in a number
of locations along the Tien river.
Figure 15. Alluvial stratification along the Tien river

42
4.4.2 Main geotechnical aspects
The geotechnical features and mechanisms will be explained by means of Figure 16 showing the
subsequent stages of the erosion process (left side) and the procedure for determination of
riverbank stability (right side).
During flood conditions in the river, the initial riverbed and bank geometry (upper left in Figure
16) will be subjected to large currents and erosion. Bank recession with time can be estimated
by using a procedure as shown at the right hand side. According to the flow pattern in the river
the current will be highest at deeper levels. Together with the sensitivity for surface erosion of
the different soil layers the erosion rate can be assessed. In the upper diagram at the right side,
soillayer 1 is supposed to be weakest which leads to the eroded cross section at a certain point of
time as shown in the middle left figure. However in case the erosion resistance of soillayer 2 or
3 had been the lowest underneath a cohesive toplayer, the eroded section would have been more
of an undercutting type. For the determination of the erosion characteristics of the different soil
types erosion tests in the laboratory have to be carried out, especially for cohesive soils. For
granular soils data can be taken from literature. However, because this method for cross section
determination as function of time requires detailed information on river and sediment
characteristics, soilprofile and erosion properties, frequent measuring of bed profiles and
interpretation from the river engineering point of view often is a better alternative.
flow slide
Figure 16. Procedure for evaluating riverbank stability (US Army, 1981)
The middle left figure shows a bank geometry with such steep slope inclination that bank
instability and slumping is obvious as is indicated in the lower left figure. The analysis of
riverbank changes caused by collapse and slumping is analogous or similar to conventional

43
slope stability analysis of an excavated slope. Slope stability analysis for (circular or non-
circular) slip surfaces can be used to judge the stability of the bankprofile caused by erosion (as
well as that of a possible eroded profile in the future) and the safety of the slope of the slope
protection design. It requires data about the shear strength as function of depth in the various
layers, as is indicated in the lower right figure by the dependency of shear stress as function of
normal stress. Bank failure results when the induced shear stresses exceed the shear strength of
the bank soils. Increase in shear stress result from increase in slope height or steepness, increase
in external loads, and rapid drawdown of the river. Decreases in shear strength of the soil result
from an increase in pore-water pressure, soil expansion, or shear movements. In case a loose
sandlayer is present also a flow slide may take place which often results in much larger bank
regression in a short period of time. This is indicated at the right side in the lower left figure. For
a prediction of the final bank recession for say after one year, the total erosion is equal to the
cumulative bank recession caused by surface erosion and slope failures.
In case of banks of natural (non-regulated) rivers, the river is in a continuous process of
morphological changes and a large area can be affected by seasonal changes. Erosion processes
of banks and shorelines are very complicated. In general, there is no one universal explanation
and solution; each case needs its own analysis and treatment. As an example, in Figure 17 the
complexity of the bank erosion problem is illustrated by a variety of processes that act together.
Figure 17. Example of physical components of bank erosion (US Army, 1981)
As illustrated in Figure 17, a typical bank consists of different soils deposited in distinct layers,
such as clay, sand, silt, etc. These soils do not permanently stand at a vertical face, but form an
angled slope that varies with the soil and groundwater conditions. In many cases this slope is
formed after a series of failures whose nature depends on whether the soil is cohesive (clay) or
granular (sand, silt, gravel, etc.). Cohesive soils generally slide along a straight or curved plane
and rotate or move downward. Granular soils will erode gradually as a result of surface erosion
or will drop or suddenly flow downwards. Height is a factor because high steep slopes impose
greater stresses and are likely to suffer more severe stability problems than low ones.

44
The internal strength of soils can be decreased by groundwater and seepage flows within the
bank. For instance, rainwater or river water at high stages (also high tides) is naturally absorbed
and seeps down to lower levels. Especially after a drop in the river level, the weight of saturated
soil is increasing causing potential instability. Soils, such as coarse sand, are permeable and
allow rapid and free passage of water. Impermeable soils, such as clay, do not allow the free
flow of water except through cracks or other openings, and the over-pressured conditions may
last for a long time leading at certain moment to the movement of a soil block (especially, in a
case of rapid and large drop in the outside water levels).
In addition the stability and the recession rate of the bank depends on:
- the added weight of buildings and other structures near the top edge of the bank (extra soil
stress contribute to slope failure);
- wave action, both due to wind (in estuaries, along coastal shorelines and river stretches with
a large fetch) and navigation will erode or undercut the bank. The effect of wave action will
be more extreme at steep slopes;
- roots of trees penetrate the clay layer and provide a path for seepage to the sand layer
beneath. On the other hand tree roots will also strengthen the soil and increase the slope
stability;
- as the clay starts to fail, cracks are formed at the surface, providing a path for seepage to
penetrate the soil, further weaken the clay, and accelerate the failure process;
- water-level fluctuations, especially rapid drawdown, will reduce the slope stability in
impermeable soils. In these soils the internal porepressure cannot follow the external
conditions through which the passive resistance reduces. In addition seepage out of the slope
arises by which surface instability is introduced and increases the risk of slope failure. In a
layered stratum seepage water penetrates the clay, reaches the permeable sand layer and
exits by flow along the lower clay surface;
- in addition, surface flow can erode the bank face, causing gullies and deposits of eroded
material on the beach below. The seepage leaving the bank at the upper clay layer can also
cause surface erosion;
- also holes made by animals (rabbits, rats, worms, termites, etc.) may intensify the erosion,
for instance because of widening of these holes by drainage water;
- intensive sand mining near the toe of the bank will create scour holes and subsequent
(additional) erosion.
4.4.3 Modelling and stability prediction
Modelling and geotechnical stability prediction is of great importance for sites where bank
protection measures are considered. In this respect the stability calculation forms an essential
part in the slope protection design process. It requires quantitative information of the local
conditions both with respect to river and erosion characteristics (cross-sectional geometry) as to
the stratification of the subsoil and the soil properties. The degree to which predictions are
relevant and applicable strongly depends on the reliability of the quantitative input data.
On the contrary quantitative modelling and stability prediction is less relevant within an overall
analysis of the erosion process and bank erosion rates along the river (as explained before in
4.4). Although insight in the geotechnical processes and failure modes (in relation to soil
conditions based on borings) is required, these issues can be conceived as a part of the total
erosion process. Pure geotechnical modelling and stability calculations are less relevant and an

45
empirical and more qualitative approach is much more appropriate. Nevertheless, indications
can be obtained on basis of quantitative analysis by specific bank stability and toe erosion
model. As an example reference is made here to the bank toe erosion developed by dr. A. Simon
and dr. E. Langendoen at the National Sedimentation Laboratory (USA), available on the world
wide web (http://msa.ars.usda.gov/ms/oxford/nsl/cwp_unit/bank.html).
A number of calculation methods to determine the slope stability have been developed in
geotechnical institutes throughout the world. These are powerful tools for the everyday
engineering practice for solving two-dimensional stability problems. The methods are based on
well-established design methods and have been widely (and still are) used for a number of
decades.
For a stability calculation usually an assumption is made with respect to the plane along which
the soil mass may slide, supposing that at each point of that plane the maximum shear stress or
shear strength is available. The reciprocal of the degree to which this available strength has to be
mobilized (as an average along the potential slip plane) gives the stability factor SF of that
plane. By considering many of these planes the most critical sliding plane with the lowest
stability factor SF can be found. With a powerful computer stability model the critical slip plan
can be found very rapidly and accurately. Because of the simple rotational equilibrium the
circular plane is taken as the basic (standard) assumption for most slope and subsoil conditions.
However for more complicated slopes and complex soillayer and/or loading conditions also non-
circular slip planes have to be considered. In special cases with structures of high importance
including combination of stiff parts (concrete foundations of rock soil) and soft soils, finite
elements methods with a realistic model of soil behaviour give better and more realisctic results
(for instance the computer programma PLAXIS).
For slope stability the Bishop method is the most usual. In this method the potentially sliding
mass is divided in vertical slices and for each slice vertical equilibrium is found. The safety
factor is found by comparison of the driving moment by the sum of slice weights (and external
loading if present) with the resisting moment given by the shear stresses alomng the potental
sliding plane.
Besides accurate stability analysis several special options are available such as:
- user-defined zones that the plane will not cross (for example sheet piling);
- input of geotextiles;
- introduction of special loading as water pressures, surcharge (distributed or pointload),
earthquake loading (by additional horizontal and vertical acceleration);
- internal pore pressures and degree of consolidation;
- (in addition to standard circular slip planes) module for slope stability with user-defined
slip plane;
- import a settled geometry calculated by a separate settlement model or a pore pressure
distribution as a result of a groundwaterflow calculation;
- graphical output of safety contours and of stress components along slip plane.

46
MStab 9.8 : 295114NW.ST I.sti
Stieltjesweg 2
2628 CK
Phone +31-15-2693500
Fax +31-15-2610821
date drw.
ctr.
form.
-
A4
13-2-2004
CO-303100 DIJKVERSTERKING ZEDERIK
Dp295+114m Alt2 Be b=10m+2.5. Kr8.0m -
Annex
-
Critical Circle Bishop
Xm : 22,50 [m]
Ym : 26,00 [m] Radius : 35,20 [m]
Safety : 1,41
Materials
13. vb
14. 8 hv onder
12. 13 nd
11. 12 d
10. 11 kt
9. 10 hv
8. 9 gz
7. 8 hv
6. 7 gl
5. 6 gz
4. 5 hv
3. 4 gz
2. 3 bv
1. 2 k
0. 1 z
25,000 55,000
00
1
2
3
4
5
67
8
910
11
12
01
2
34
5
6
14
8
9
10
11
12 13
Figure 18. Example of slope stability calculation.Cross section of a dike with berm and slip
circle
4.4.4 Flow slides in sand
A flow slide can be described as an instability that occurs in a fairly gentle underwater slope
consisting of loose to medium dense sand, causing liguefied sand to flow out into an even more
gentle slope. Compared to a common shear failure it generally involves a much larger mass of
sand. It is therefore often accompanied by a sudden and large decline of the coastline or
riverbank. A flow slide may occur after a change in slope geometry, for example steepening
and/or deepening of the channel as a result of scour erosion. A rather small but quick change
may trigger the liquefaction of a vast amount of sand. Also vibration or seepage out of the slope
or a riverbank due to tidal water level variation may introduce the phenomenon. As a
consequence the local subsoil conditions required for liquefaction are often of greater
importance than the exact reason for triggering the liquefaction.
Flow slides occur in many parts of the world, for instance along the banks of the Mississippi, in
several regions in Bangla Desh and in the south western part of the Netherlands. In this respect
some remarking events are reported in literature from which it can be concluded that the chance
of occurrence of flow slides is largest where turbulent hydraulic conditions and young geologic
sub soil conditions are combined. It is therefore not surprising that a number of well-known flow
slides areas are situated in river deltas. For detailled information, see Stoutjesdijk, T.P., De

47
Groot, M.B. and Lindenberg, J (1994) and Stoutjesdijk, T.P., De Groot, M.B. and Lindenberg, J
(1998).
The phenomenon also regularly occurs because of human activity during dredging works in
loose sand and along the slopes of artificial sand islands (e.g in Beafort Sea near Canada). It
means that in case critical subsoil conditions are present or expected, special preventive
measures have to be taken.
Dimensions of a flow slide
The possible size and dimensions of a flow slide seem to be unlimited. In literature extreme
amounts of displaced sand of hundreds of millions cubic meters are reported. However the
dimensions depend on the degree of liquefaction susceptibility of the sand (related to the density
of the sand) and on geometrical conditions like the channel depth, the steepness of the channel
and thickness of the sensitive sandlayer. In case the sandlayer is covered by a cohesive layer, the
size of the slide is influenced by the thickness ratio of toplayer and loose sandlayer. A relative
thick cohesive toplayer will discourage the continuation of the flow slide process (by sinking in
the liquefied sand) and reduce the distance of retrogression of the river bank.
An important geometrical feature for the damage is the average slope that remains after the
slide. Depending on the conditions described above the average endslope may vary between 1 to
5 and 1 to 20 (occasionally 1 to 30). The final cross section can in most cases be characterised
by a very gentle slope which passes into a rather steep upper end (see following figure). As an
example: suppose a depth of 30 m of the river close to the bank consisting of low density sand.
Because of erosion the slope has been steepened to 1 to 3 and a flow slide is initiated at a depth
of say 20 m. Because the loose sand is present up to 25 m, the liquefaction may occur to this
depth. If the top of the loose sand is at the surface, the end slope will be between 1 to 7 and 1 to
20. Application of this endslope means retrogression of the upper bank boundary beween
approximately 50 m and more than 200 m (see Figure 19). Because this may happen within
several hours to maximum one day, a flow slide often represents a drastic, far-reaching and
sometimes dangerous mechanism.
The same example but now with a cohesive clay layer above the sand (lower part Figure 19).
This claylayer reduces the thickness of the sand susceptible for liquefaction. The final slope in
the toplayer after the instability will be between 1 to 1 and 1 to 2. Suppose the thickness of the
claylayer is 10 m covering 15 m of loose sand. Based on the most extreme endslope in the sand
of 1 to 20, the maximum possible retrogression of the bank is reduced to less than 120 m. It will
be even more reduced because of the restraining effect due to the sinking of the clay in the
liquefied sand. The chosen thickness ratio of 10/15 means that the maximum possible endslope
in the sand decreases to 1 to 10 resulting in a maximum bank retrogression of about 40 m.

48
Figure 19. Schematization of flow slides for sand and clay
In plan view the result of a flow slide can be characterized by a shell form which is illustrated in
the Figure 20.
after flow slide in sand
Shear failure
initial erosion
Average slope
after flow slide with
cla
y
on to
p
initial erosion
Average slope
Maximum 200 m
maximum 40 m

49
Figure 20. Plan view of a flow slide
Physical explanation of a liquefaction flowslide
Liquefaction may occur in loose sand because this material has the tendency to decrease in
volume under influence of a change in shear stress τ’. This is shown in Figure 21.
Figure 21. Physical principle of liquefaction phenomenon
However, in case the shear stress change takes place quickly, the decrease in volume ∆e (or
reduction in pore volume) is not possible when the pores are fully saturated with water. It means
that the tendency to decrease in volume leads to an increase in the pore water pressure u. From
the basic soil mechanic rule σ = σ, + u (total stress = effective stress + pore pressure) an increase
in pore pressure involves a decrease in effective stress. And because the effective stress is
directly related to the shearing resistance, according to the Coulomb law τ’ = σ, tan ϕ, the effect
of the change shear stress in saturated loose sand is a decrease in shearing resistance. If this sand
is present in or just below a slope it will reduce the stability of the (underwater) slope and the
slope may fail in a relative short period of time.
CONTRACTION
' '
-
e
'
τ
’

50
The increase in pore pressure may be so large that the initial effective stress is fully cancelled. In
this case the sand is liquefied and it reacts like a thick liquid. The consequence is that an
apparently stable slope may flow into the river. Because it is the change in shear stress that
causes the excess pore pressure, the initiation of liquefaction can be triggered in a slope that is
much more gentle than the angle of repose. Therefore flow slides may happen in a slope with
inclination 1 to 3 or 1 to 4. Because liquefaction means that the sand reacts like a liquid, the
final slope after the event can be very flat: between 1 to 7 and 1 to 25.
Flow slide prevention
For the prevention of the flow slide mechanism two basic entries are available:
- reduction of liquefaction potential of the sand: in general this means densification of the
loosely packed sand. In general this method (for instance vibro flotation) is rather expensive
and not applicable in slopes because the densification itself may trigger liquefaction and the
initiation of a flow slide;
- prevention of the triggering mechanism: For scour induced liquefaction of slopes it can be
realized by full protection of the slope. In the Netherlands slope covering protection has
been very sussessful because the occurrence of flow slides has been reduced almost
completely.
4.4.5 Observations and remarks from meetings and site visits
During the meetings in Long Xuyen, Cao Lanh, Vinh Long, detailled information regarding
procedures for selection of river bank locations to be protected, protection design and protection
methods have been provided by the provincial authorities (SARD) and the representatives of the
Ministry for Agriculture and Rural Development (MARD). This information and the
observations during the subsequent site visits give rise to the following remarks from the
geotechnical point of view:

during the meetings with provincial authorities of SARD it became quite clear that bank
erosion takes place in a large number of locations along the rivers in the Mekong delta as
well as in the Red River delta. Bank erosion should be conceived as a feature of the natural
behaviour in the Mekong rivers and to a high extent also for the Red River within the man-
made flood protection system. Bank erosion cannot be fully prevented without complete
regulation of the river;
 because of the limited financial means bank protection works should be considered very
carefully and based on decent cost benefit analysis. In principal the mission agrees on the
basic entries for selection and priorization developed by the MARD ministry as explained
during the meetings. However the result of the selection procedure being the priority list for
locations to be protected in short term is not completely clear;
 in the Netherlands much experience has been obtained with slope protection by gabions.
However gabions are only seldom used in heavily attacked conditions and not to protect
slopes over a depth of more than ten meters. It has never been used to a depth of about 45 m
as is presently in construction in Tan Chau in the Mekong delta. For conditions like those in
Tan Chau (province An Giang) the gabion slope protection makes high demands on the
quality of gabion wire (avoidance of rust), degree of filling, exact placement and prevention
of downward moving of the elements (e.g. due to wave action), etc. Frequent monitoring and
inspection of the performance of the gabions protection and adequate maintenance and repair
of damage is therefore of vital importance;

51
 in a number of cases revetments and gabion protection are stopped just beyond the toe of the
slope (maximum about 20 meters). Sometimes even before the toe. It is recommended to
continue these protections to the deepest point in the river or to make a decent toe structure.
On the whole, in the design stage one should better anticipate on possible future changes of
the riverbed, e.g. the development of a scour hole;
 in the design stage the stability of the original (unprotected) and future (protected) slope is
assessed. Some information concerning these geotechnical analyses has been supplied to the
mission members. This information contains remarkable low values for the shear strength of
the present soillayers. Regarding the types of soil and the unit weights we should expect at
least 30% higher values. Additional information on types of sheartests and design procedure
will be appreciated by the Dutch mission;
 based on the reported surprising low values for the shear strength, the stability factor for the
Sa Dec gabion protection is only about 0.85, which suggest that the slope is not stable.
Therefore it appears very important to re-analyse the calculation procedure and the basis of
the low shear strength values ;
Figure 22. Sa Dec slope protection
 During the mission information on subsoil stratification and soil properties has been
provided for some locations where protection works are carried out or recently completed
(e.g. Sa Dec: see above). For other locations including those with serious erosion,
information on subsoil stratification and soil properties is scarce, not available or badly
accessible. It means that prediction of future erosion or explanation of former erosion is
rather difficult. The set-up of an generic geotechnical/geological data base (on provincail or
national level) is recommended;
 In front of the Vinh Long boulevard strong erosion takes place (Vinh Long province). From
comparison of results of recent soundings with respect to those of 2000 it is known that
scour has increased and reached a depth of about 30 m with bank steepness approaching 1 to
1 (Figure 23). To the mission members the situation appears to be very serious. However up
till now only the upper meters of the slope are protected by gabions and real effective
measures are not planned;
concrete
gabions on geotextile
slope 1 to 3
minus 12 m
minus 25 m
Critical slip circle SF = 0.85
sandbags
concrete
gabions on geotextile
slope 1 to 3
minus 12 m
minus 25 m
Critical slip circle SF = 0.85
sandbags

52
Figure 23. Serious erosion near Vinh Long (average slope of riverbed approaches 1 to 1 but no
protection is planned)
 in many locations sandmining is carried out. These activities are primarily meant for
economic reasons supplying construction materials. In a number of locations however these
activities seem to be very uncontrolled from the river management point of view.
Concentrated and extensive sandmining can threaten the stability of the riverbanks and
intensify bank erosion. It is therefore recommended to promote the sandmining activities in
locations where it may help to shift the deep channels away from the bank, for instance the
removal of sandbars in the river;
 navigation, especially fast sailing ships close to the bank (of river and canals) may intensify
the bank erosion and should be regulated;
 concrete piles to prevent instability of slopes have been applied in the bank protection works
in Tan Chau and Sa Dec. The mission is still not convinced that this is the most optimum
solution. Moreover, in case piles are used, they should be driven to such a depth that the
benefit is applicable for the entire length of the critical slope. In this respect the depth of the
piles in the Sa Dec slope protection is far insufficient (see Figure 22);
 during the mission there were some indications that the flow slide mechanism could play a
role in the bank erosion process and in the rate of bank regression in Vietnam. It is very
important to recognize this special bankfailure mechanism because it may cause very large
bank retrogression in a very short period of time. During the meeting with SIWRR on
November 5 is has been explained that a bank sliding may involve 50 m retrogression as a
maximum. At the eroding site near Phong Van village along the Red River many cracks
have been observed up till a distance of 35 m from the bank edge. It seems very unlikely that
this is caused by (the beginning of) a conventional shear failure;
some gabions
minus 30 m
bed profile 2003
unprotected
bed profile 2000
some gabions
minus 30 m
bed profile 2003
unprotected
bed profile 2000

53
 because the flow slide mechanism may involve large bank retrogression in a short period of
time and because the phenomenon often occurs unexpectedly, it is important to know the
potential locations with subsoil conditions (loosely packed fine sand) highly susceptible for
liquefaction. Although no strong clues for such locations
 Between K82 and K84 at the left bank of the Red River (district Chay Giang in Hung Yen
province) a number of cracks in the dike are reported by the provincial authorities. The
cracks arose after dike repair and the recent construction of an extensive berm at the
riverside (see Figure 24). The local soil profiel contains a 6 m thick soft clayey (mud) layer
at a depth of about 8 m below the dike crest. It was suggested by the authorities that the
crack pattern was related to the initiation of a large shear failure through the soft mudlayer.
The geotechnical specialist of the mission however believes this in not very likely because of
the very gentle average inclination (including the new constructed berm) and the fact that the
mud layer has been completely consolidated due to the loading by the dike for so many
centuries. A more credible explanation for the cracking should be found in the compression
of the mudlayer owing to the extra loading of the large berm and, as a consequence in
settlement at the surface. Because this settlement also occurs at the toe of the dike and only
to a minor extent at the dike crest, it means an increase of the distance in between these two
points and longitudinal cracking in of the surface of the dike slope. The cracks at the surface
of the berm most probably are caused by drying and shrinking of the toplayer, perhaps partly
due to unsufficient densification. Although more unlikely, due to the fairly steep slope of the
dike the cracks in the upper part of the dike might also be an indication of a shallow shear
failure in the dike itself. Because the origin of the phenomenon has to be found in
deformation (above the mudlayer) and compression (of the mudlayer) the execution of finite
element calculations (e.g by using PLAXIS) can be very helpful to understand this problem;
Figure 24. Cracks in dike left bank Red river between K82 and K84 in Hung Yen province.
Compression of mud layer during centuries – extra compression of the mud layer because of
(recent) berm accompanied by surface settlement and cracks which may go on for many years
 it became very clear to the mission members that MARD has to operate with an enormous
lack of budget both with respect to the Mekong delta and the Red River delta. It comes
forward in the procedure for priorization and selection of protection projects and in the
absence of sufficient budget for maintenance of protection structures. That’s why the
ministry is unable to allocate the majority of protection projects proposed by the involved
provinces. An illustrative example is the yearly budget of in total 150.109 VN dongs (10
million US$) that is available for the whole flood defence system in the Red River delta.
mud layer
Compression of the
mud layer mud layer
Compression of the
mud layer

54
This budget, which comprises all new projects and maintenance works including dikes, bank
protection and sluices, represents far less than one percent of the total invested value in the
delta
 due to the lack of budget the MARD ministry is forced to act in a fairly ad hoc way with
primary attention for short term solutions. In some respect the policy seems to be based on
disaster management in stead of taking measures in anticipation to future development and
based on decent cost-benefit analyses.

55
4.5 Bank protection
4.5.1 General
There is a large number of non-structural and structural measures for coping the floods and bank
erosion. Among them one may distinguish the following two types of measures measures: Non-
structural measures and Structural measures. Some examples are given below.
Structural measures
- bank protection in case of big problems (revetments, groins, etc)
- local measures (even by individual person) for small/local erosion problems
- zoning along riverbanks (free space for natural developments)
- removing of sand bands which reduce the cross-section and shifting the current to one of
the banks; it should be considered to make an economic study on the quality of sand and
and selling possibility as buiding material or for land reclamation purposes.
Non-structural measures
- prediction and warning systems
- evacuation plans
- relocation people to other areas
- tree planting
- floating tree
Mainly, the structural bank protection measures will be discussed here based on onservations
during the mission, however, some remarks on non-structural measures will also be given.
It is believed that after 1994/1998 the floods in the Mekong River have increased their intensity:
higher water levels and higher discharges then before. The high floods illustrate this in 1961,
1966, 1978, 1991, 1996, 1999, 2000, 2001 and 2002. These higher floods cause more bank
erosion. But other factors contributing to an increase in bank erosion are:
• Construction of dykes,
• The increase in population results in more development (construction of houses,
factories, etc.) close too the banks,
• The channels connecting the Tien River and the Hau River tend to erode and to increase
their capacity. The maximum discharge measured in this connecting channel Vam Nao
Canal is 15.000 m3/s. Originally it was a small man made canal.
Before 1994 bank erosion along the branches of the Mekong River did not cause serious
problems. The increased floods have resulted in an urgent need to construct new bank
protections.
First the observations made by the Mission regarding the existing bank protections will be
summarized in Section 4.5.2. Next the Mission proposes to develop an Integrated Plan for river
bank protections as part of a wider Master Plan (4.5.3) an to start with a structured learning
process to improve knowledge necessary for an optimal design of bank protections (4.5.4).

56
4.5.2 Observations on bank protections
Old protections
The old bank protections along the branches of the Mekong River were constructed by the local
people to protect the upper part of the bank against erosion by river current, ship- and wind
waves mainly. These constructions have not been evaluated on their economic effectiveness.
Indigeous methods using vegetation to protect the bank against erosion have been applied
successfully to some degree. Probably the experiences with these methods have been
documented in the standard 14 TCN – 84 – 91. No other design manual seems to exist. Often the
bank is still in its natural state without a dyke or embankment with a flood free road on top.
Recent bank protections: the revetments near Tan Chau and Sa Dec
Recently bank protections have been designed to stabilize the bank, for example a revetment
near Tan Chau and a series of groins or a revetment near Sa Dec. Their construction will be
completed soon. Irrigation projects do not seem to extend towards a river bank. The overland
flow is not considered as a disaster; therefore only bank protections near towns are designed to
stop overland flow and to prevent inundation during floods.
Design tools used for the Tan Chau revetment are experience and a mathematical model,
while a mathematical model, a physical model and experience are applied for the design of the
Sa Dec series of groins. Mission has learnt that representatives of the People Committee and the
local DDMFC offices usually decide the choice of design tools, and to which extend
For both protections the same design report has been prepared. It is a technical report with a lot
of information, without figures. Separate books with design figures indicate that the protection
has a length of a few hundred meters. The slope of the upper 10 m of the protection has a steep
slope 1 in 2, the deeper part has a slope 1 in 3. The design depth of the channel in front of the
protection is about 25 meter.
For Tan Chau only a combination of a revetment and a stone gabion layer with bamboo trees
had been elaborated. This stone gabion layer had been designed to accelerate sedimentation
upstream of the protection. For Sa Dec two alternatives, a series of groins and a revetment had
been compared. The revetment costs half the price of the series of groins. Therefore a revetment
had been selected.
Other protections
Nowadays the riparian population use sand bags made of a fabric to construct emergency
protections. If these bags are exposed to sunlight, the strength of the fabric will deteriorate in a
few years. And the protection will get damaged and finally it will be lost. Simple wooden
protections have some effect to protect against wave attack.
The connection to the rigid concrete structure founded on concrete piles above the cages of a
protection by gabions allows the formation of a gap. As part of maintenance of the revetment
this gap should be filled and repaired. The Mission saw an example of this damage in the
waterfront in Vinh Long close to Cuu Long Hotel.
At a ferry terminal along the Mekong River the Mission saw some older gabions badly
damaged. These gabions were made of unprotected metal wire.

57
Rock or rip rap has been applied at several locations at a small scale
Concrete piles and concrete sheet walls can be found at places where ships are moored. At the
locations visited by the Mission the surface of the concrete did not show damage or
irregularities.
Besides the traditional materials also to use vegetation, as a bank protection should be
investigated more in detail. Examples of type of vegetation, which can be used to improve the
strength of the bank to resist the eroding forces by the river, are for example vertiver grass,
catkin grass, palm trees or acacia trees. A disadvantage of vertiver grass is that it will die after
more than 14 days of continued inundation during a flood.
4.5.3 Proposal for an Integrated Plan for River Bank Protections
Introduction
Many locations along the Mekong River in Vietnam experience increased bank erosion since
1994. The need for priority ranking of these locations and the potentially complicated
morphological interactions require an Integrated Plan or a Strategic Plan to realize the most
economic and efficient control of the bank erosion problem partly by planning new bank
protections.
The local agencies wish a flood and bank erosion management plan for each province (MARD)
and wish guarantees for continuation of financing of projects (already planned and future
projects)
The potential benefits of an Integrated Plan for bank protections are not jet recognized at all
levels, because only a few bank protections are under development. It is expected that in due
time the response of the river to bank protections will be observed, but at that time costly
corrections are the only option (Figure 25). Therefore the Mission recommends to prepare an
Integrated Plan for the construction of bank protections along the rivers in the Mekong delta. For
example the SARD expects that the protection in Tan Chau in future has to be extended in
upstream direction. In a final stage this can be large protection work guiding the river bend. The
same holds for the revetment protection now under construction in Sa Dec. Already now a new
budget is requested for extension works in 2004. The Integrated Plan can be a tool to prevent
unpleasant financial surprises for the decision makers.
An Integrated Plan for bank protections should take into account the long term, short term,
socio-economic and environmental aspects, while not neglecting the importance of the rivers
and canal system for inland navigation should be recommended. And an Integrated Plan should
be a part of a Master Plan for the whole delta (i.e., for Tien River, Hau River in Mekong delta
and for Red River). This Master Plan should include: water management, drinking water supply,
floods, dikes, bank erosion, zoning along riverbanks (free space for natural developments),
navigation, ecology, relocation of riparian people to other areas, legislation for living along
riverbanks, etc. It was mentioned in discussions that some regulations on that aspect already
exists but the Mission was not able to see these documents.
An Integrated Plan for bank protections should deal with the following elements, varying from a
high conceptual level to a detailed level of the bank protection structure:

58
1 Objectives
2 Priority for implementation
3 Illegal land reclamation
4 Selection type of bank protection
Moreover, in the learning process of improving the actual approaches more attantion should be
paid to such aspects as:
- design methods and design parameters,
- pilot testing,
- emergency actions and flood fighting, and
- (update) Design Guidelines Standards
Description of components of an Integrated Plan for Bank Protections
Objectives
In general the objective is to control the bank erosion along the branches of the Mekong River
for the lowest costs.
What is the more specific objective or what are the objectives? The answer on this question can
be for example:
- A river alignment with the shortest total length of bank protections or
- A complete conservation of the present alignment or
- Mainly a conservation of the present alignment, but with small corrections in case of islands;
one branch will be close gradually and the procentual discharge in remaining branch will
increase to 100 %.
A consequence is that finally in due time almost all banks have a bank protection.
Priority for implementation
Which bank protection works can be implemented independent of each other? Distinguish a
protection of a straight bank, an outer bend, an inner bend and a bifurcation point. A sequence
will be selected for the implementation of the works to avoid unwanted morphological
consequences. The presently applied criteria for priority of river bank protection works:
I dike safety
II urban areas/cities
III (densely) populated villages
IV agricultural land
Often bank protection works are composed of combinations of different structural measures,
such as:
- bank protection in case of big problems (revetments, series of groins, etc)
- local measures (even by individual person) for small/local erosion problems
- removing of sand bands which reduce the cross-section and shifting the current to one of
the banks; it should be considered to make an economic study on the quality of sand and
and selling possibility as buiding material or for land reclamation purposes,
- light protections: vegetation, such as trees, planting along threatened banks,
- temporary protections, for example by floating trees.
In addition to the structural measures the Plan should also contain non-structural measures:
- prediction and warning systems for extreme floods and of erosion including better
transfer of information from studies carried out by Water Resources Institutes and others

59
- flood fighting and evacuation plans during emergengy situations.
An example of a small correction in the river planform resulting in some land reclamation:
Sketch 1 Present situation
Sketch 2 Example of potential river training measures:
1 Stabilization of the bifurcation with bank protections
2 Create sedimentation in the small branch, for example by a series of groins
Sketch 3 Final situation where 2 small islands are unified into 1 island
Figure 25. Possible effect of planform correction measures
2
1

60
Maintenance is a general problem in Vietnam; no or little money for maintenance is available.
Because of scarce of funds the maintenance is considered as spending money without direct
profit (thus wasting money). A convincing strategy must be formulated in the Plan where the
role of proper maintenance can be expressed in terms of saving money.
Illegal land reclamation
At some places private owners extend their plot into the river by a small reclamation. Often this
an illegal activity but it not corrected systematically by the authorities. To support the legal
framework a smooth curve should be determined to indicate the maximum land reclamation.
This curve should be communicated and discussed with all stakeholders. If land reclamation
crosses this curve, a correction to remove this illegal part has a strong backing. This will became
an important aspect as in future space in a river becomes more rare and the unit price of land
will increase.
Selection type of bank protection
Which the type of hydraulic structure should be applied and where? Examples of hydraulic
structures are revetments, series of permeable or impermeable groins, bottom screens,
combinations with vegetation and other measures. Other measures might include an
investigation in the effect of the floating fish farms on the hydraulic load on the bank.
The main advantages and disadvantages of types of bank protections are summarized in Table 3.
Type advantages disadvantages
Revetment Narrow width of the
structure.
Fixation of the bank line
Solid groins Flexible to future changes
in bank line.
Strong current along the
bank in the groin field.
Wide zone of the bank
required.
Impermeable groins Flexible to future changes
in bank line.
Good sedimentation
downstream of groin.
Wide zone of the bank
required.
Bed protection Flexible for future
extension of the
protection.
Only partial solution for
bank erosion problem.
Vegetation Economic Effect varies in time.
A spacious area is required
Table 3 Comparison of different protection types.
Design method and design parameters
For a deterministic design of a bank protection the decisive flow velocity is an important design
parameter. This parameter has been determined as the maximum measured flow velocity (in
1997). A statistical analysis and an extrapolation to the selected frequency of re-occurrence is a
better method, which allows a differentiation of safety levels. However, this method can only be
applied if sufficient field data are available. Therefore a regular monitoring should be set up of
the erosion process and the hydrodynamic load on the bank at all important places.

61
Above a revetment the flow velocity varies significantly. Therefore it is important to mention
where the maximum flow velocity was measured and to determine a maximum velocity
(including an estimate of extreme turbulence intensity) as a function of the location above the
revetment or bank.
The maximum hydraulic load on a bank is a combination of a maximum depth averaged flow
velocity and turbulence intensities, maximum ship waves and flow velocities and maximum
wind waves. The design method should provide a standard calculation method for the maximum
hydraulic load.
The slope of the revetment had been tested for its geo-technical static stability. However, its
morphological effects on scour in front of the revetment could not be found in the design report
The shape of a bank protection has often many geometric parameters to be selected. Not for all
those parameters exist design rules, therefore some are selected by expert judgement only. It is
recommend to develop methods to find the optimal shape regarding the technical aspects and the
total costs of the investment and maintenance. This can be a sensitivity analysis for a limited
number of parameters (main parameters). Such methods do not exist for bank protections.
The Mission recommends the comparison of more elaborated alternative designs for a bank
protection to be sure that the optimum will be selected between costs and effectiveness.
In general a revetment has an upstream termination and a downstream termination. The Mission
recommends to add a downstream termination to the design of the revetment near Tan Chau and
to strengthen the terminations of the revetment near Sa Dec.
Pilot testing
The present state of the selection of a certain type bank protection is guided by a limited
knowledge. The selection of pilot structures with several sections, which differ in top layers,
slopes, falling aprons and materials. In a pilot revetment these structures will be tested in
different sections. Their performance will be monitored during a few years. Finally the results
will be evaluated in a cost- benefit analysis.
Emergency actions and flood fighting
Flood fighting is concentrated in the North of Viet Nam. It ‘s development in the Mekong delta
had just started. Local people take emergency actions by themselves. A manual guiding these
actions and the preparation of these actions will improve the effectiveness to save the banks.
Standard Guidelines
For the design of bank protections national guidelines are available, Standard 14 TCN – 84 – 91
and 14 TCN 130 – 2002. However, these standards are old and not representing the actual state
of developments.
Finally, an update should be prepared of the Standard guidelines for bank protections. Design
engineers will use these to design new river training structures.
National data-base:
Systematic approach: monitoring/surveying/instrumentatin, prediction, maping of problems,
selection/selection criteria, cost-benefit also for urgent cases where no money is available,
priority versus money/budget, etc. Design standards and guidelines are needed. The old design
documents are usually outdated.

62
Better-coordinated support by Research Institutes is needed. It includes the better
contacts/support planning agencies as well as design offices.
4.5.4 Bank protection materials and structures
Observations
Gabions are sensitive for damage by settlement of the stones and consequently rocking of the
stones by wave attack. This rocking results in due time nearly always in damage of the wire
basket. Therefore the lifetime is rather short and the construction will probably require
maintenance measures.
In general gabions are sensitive for damage by settlement of the stones inside the cage. This
results in a space in the cage and consequently rocking of the stones by wave attack. This
rocking results in due time nearly always in damage of the gauze. Therefore the lifetime is rather
short and the construction will probably require maintenance measures.
The cages are flexible and after some they will slide slightly along the slope of the revetment..
Conclusions on conceptual design of bank protections
In the past bank protections along the branches of the Mekong River were designed to protect
against wind- and ship waves mainly. Indigeous methods using vegetation to protect the bank
against erosion have been applied successfully. Probably the experiences with these methods
have been documented in the standard 14 TCN – 84 – 91. No other design manual seems to
exist. Often the bank is still in its natural state without a dyke or embankment with a flood free
road on top.
Recently bank protections have been designed to stabilize the bank, for example a revetment
near Tan Chau and a series of groins near Sa Dec. Their construction will be completed soon.
Irrigation projects do not seem to extend towards a riverbank. The overland flow is not
considered as a disaster; therefore only bank protections near towns are designed to stop
overland flow and to prevent inundation during floods.
Several basic concepts have been developed for
bank protections along riverbanks, such as
revetments, permeable and impermeable groins,
falling aprons and bottom protections. A revetment
has standard an upstream and a downstream
termination to guide the flow and to anticipate the
formation of a local scour hole by the flow
separation at the downstream termination.
Guidelines are missing for the selection of a
concept or a combination of different concepts for a
bank protection in a certain location.
The Mission recommends the comparison of more elaborated alternative designs for a bank
protection to be sure that the optimum will be selected between costs and effectiveness.
The slope of a revetment is tested for its geo-technical static stability. However, its
morphological effects are sometimes neglected, for example on the scour in front of the
revetment. A gentle slope will reduce the depth of the scour hole.

63
4.6 Summary of personal observations by Pilarczyk (just as a remainder)
Bank erosion is a serious problem in Vietnam. There are many different rivers and channels with
different geological composition and different soil conditions, and different pattern of siltation
and erosion process. Moreover, the flood and bank erosion damage is not always easy to
separate from each other.
In Mekong area people do not consider floods as a disaster problem as long as it is not an
extreme big flood; often, it is a disaster when there is no usual/moderate annual flood.
Advantages: water and soil improvement for crops, better fishery conditions, navigation
Disadvantage: big floods (when nearly the whole area/province(s) is flooded
Flood 2000 is one of the biggest historical flood for Mekong Delta (most area was inundated
about 1.5m or more for a number of weeks)
Government Decree nr. 86 (2003) gives the mandate to MARD/DDMFC to manage the erosion
problems of riverbanks and seashores nationwide (all over the country).
Legal aspects/questions:
- how to anticipate to the erosion problems (instead of the present emergency management)
- how to manage the flood plain, especially in the areas without dikes (Mekong Delta and partly
Central Vietnam).
In Government Act from 1998 the new Water Law was formulated including forming of
National Water Resources Council and Regional River Basin Organizations. However, these
River Basin Organizations are not active yet, still in a primary stage of organization and they do
not fulfill a practical rol in water management. On the other side, a number of overall actions
and future planning items mentioned in this report can be a part of activities of these new
agencies.
At this moment these River basin Organizations are still a learning process and they need strong
support from the involved Ministries and Provinces to become real powerful organizations.
Reasons of erosion
In Mekong Delta as well as in Red River area it was suggested by people that the eropsion
problems have increased from early 90-ies. Increase of population and associated increase of
occupation of riverbank areas in combination with destroying or removing of the natural
vegetation have probably resulted in accelerated erosion in recent years. In some areas increase
in number of small dikes and embankments for roads has also reduced the space for water
resulting in increase of water levels and flow velocities.
In Mekong Delta there is observed an accumulation of extreme floods in last few years; extreme
floods in nearly each year in the period of 1977 to 2002. This has, of course, resulted in
accumulation of erosion problem is a very short time-period.
In HCMC area (Saigon River), erosion problems are evident especially at sites where natural
banks have been adapted for local urban and industrial developments.
Besides erosion of main rivers an increased erosion of (inter-) connecting canals is observed.
During the dry seasons the erosion of lower parts of banks continue endangering the stability of
the bank. On the upper part of slopes the drying process results in cracking and slope erosion
due to land saturation/overflow after heavy rains.

64
Reasons of increase of erosion (especially, in Mekong Delta):
- increase of flood size
- increase of flood frequency
- increase discharges/velocities
- monsoon winds (surface waves)
- navigation (ship waves)
- sand mining (at some location it can be a main reason of local erosion problems)
- increase of fishery area
- increase of population and use of riverbanks, also for economic activities
-
In Red River area the hydrograph and the induced flow regime and morphological changes is
partly influenced by operation of Hoa Binh reservoir. Some people refer increase of erosion in
recent years to the operation of Hoa Binh reservoir. However, this effect is not studied in detail
yet. In Red River more erosion is observed in the downstream areas influenced by tidal
movement (in contradiction to Mekong Delta). The main erosion problems/damages are
observed at the end of the flood season.
Bank erosion problems in Central Vietnam are somewhat different than in North and South due
to specific characteristics of rivers, specific soil conditions and nearly no dikes: short rivers with
a high gradient, rapid increase of water levels and discharges after the monsoon rains. There is a
strong relationship between river hydrograph and erosion probability. Some local studies are
carried out with support of HWRU Hanoi.
Physical understanding of mechanisms of erosion is not studied in depth yet. Especially, the
geotechnical aspects of erosion and stability (steep foreshore, saturation, sliding, etc.) need more
attention.
Much statistical data is available at various institutions (local agencies, research institutes,
universities, ministries) but systematic overview/data bank is still missing and not available.
Main/most common erosion mechanism: scour of underwater section induces the sliding of
(saturated) upper part of slope, usually after drop-down the high water. Removing of natural
vegetation accelerates erosion process.
The following classes of erosion are often applied for classification of erosion problems:
I 1-5m/year
II 5-10m/y
III 10-20m/y
IV > 20m/y
In Mekong Delta (i.e., Thuong Phuoc area above Tan Chau, but also in Sa Dec and Vinh Long
area) there are a large number of kilometers of riverbanks with erosion rate of 40m/year or even
more; a few thousands hectares loss of land, and tens of casualties, and a few thousands of
households relocated..
Social/socio-economic problems: loss of houses/properties, loss of land, loss of work, possible
casualties, etc.

65
The cost of lost of land, properties and of necessary resettlements should be taken into account
in the cost-benefit analyse, including social/socio-economic repercussions for the society.
However, the unit costs should be better/more uniform defined, eventually taking into account
the potential value of certain area/land in respect to the planned/prognosed/expected
developments.
Methodological items
Prediction of erosion is a difficult task. Systematic survey will help to understand the erosion
processes and will help to make a better prediction.
Prediction of bank erosion:
- measurements and experience/data bank
- data collection and processing; monitoring by surveying, satellite image, GIS/GPS, radar
etc.; the use of new monitoring techniques should be stimulated
- analytical methods/formulae
- modelling
- prediction; extrapolation to coming future
-
NB.
- Physical modeling can be considered as support for projects (design purposes) but not as a tool
for prediction of bank erosion.
- Available (international) analytical formulas on erosion/scour prediction can be
improved/adapted to local conditions by calibration using survey data’s; purely experimental
formulas have limited (local) value and only applicable as indicator or help for extrapolation of
collected data’s.
- There is already much experience on the use of the modern monitoring techniques in Vietnam,
however, this knowledge is spread over a number of Institutes and Universities belonging to
different Ministries. The lack of cooperation is the main weak point and the main reason of not
fully use these techniques for practical application. Dissemination/exchange of knowledge and
experience/courses/ publications/reports, etc. should be further stimulated. These organizations
will need to be monitored and assisted to increase their effectiveness.
- More attention should be paid to monitoring and prediction of scour and geotechnical state of
bank/dike structures. Monitoring/surveying scour (especially in bends) can help early definition
of critical sections.
Surveying capacity is not adequate to the problems; the actual expenditure on monitoring is very
low. Surveying on Red River (1 x per year, about 170 standard river cross-sections) is executed
by a survey vessel from Ministry of Transport. Systematic data processing and storage in central
data bank should be improved. The diagrams on erosional trends of monitored srcoss-sections
are not always available (or difficult to find).
The information on critical sections is not available. Usually, the critical sections are recognized
as such after collapsing of a certain section. The monitoring of such a sections takes usually
places in the scope of preparation of rehabilitation/mitigation measures/projects.
With good survey strategy, definition of critical sections, data collection and proper/on time
data processing one may earn money/investment back due to better selection capability. Good
communication between various (national and local) agencies is crucial for optimal use of
available resources and creating a data base. Definition of critical sections can be done by

66
analyse of surveying data supplemented with results from mathematical models (i.e. MIKE 21-
C or DELFT 1D or 2D).
Zonation along the riverbanks (free belts) should be stimulated/improved.
Prediction of erosion: monitoring/survey- data base- mathematical modeling- calibration-
validation- prediction-quality consideration
Hydraulic components of prediction of erosion: critical velocity- dominant discharge tidal
movement- survey sediment- design criteria
Combat erosion:
Because of lack of funds the actual policy is focused mainly on bank protection in urban areas
(and very limited in agricultural areas). However, within these limited funds it is even not
possible to solve the all urgent cases.
The groins, usually applied on limited scale/on too short stretches or too small in size, have little
effect in solving problems. In some cases it should be recommended to use large groins working
as flow guiding structures to be able to deflect the flow/currents.
Design and Construction methods
The hydraulic conditions are often very severe; the velocities during floods can be up to 5m/s.
Also the execution conditions can be very severe, say up to 3m/s. These conditions require
rather heavy designs and proper placement equipment. Usually, the execution takes place when
velocities are reduced to <2m/s. The actual design codes must be upgraded to be able to tackle
properly these heavy conditions in designs.
Standard for big projects: revetments, gabion mattresses, sometimes also (closed/open)groins.
Gabion stone mattresses are usually of low quality (without corrosion/abrasion protection) and
thus, of short life-time.
Local/traditional materials and systems: bamboo, sandbags, bamboo’s anchored in gravel/stone
baskets, Dragon stone/bamboo rolls (cylinders filled with stone).
In general, the construction methods seems to be properly chosen, however , often based on a
limited number of alternatives (if any). However, the projects are not executed/implemented to
the whole planned extend and the flanking erosion can still be a potential danger for the already
realized structures/protections.
Preservation of the existing/natural vegetation should be stimulated and wherever possible, new
appropriate vegetation should be planted (for example, vetiver, bamboo, etc).
Some alternative solutions can be considered: combination of existing floating fish-houses with
concept of underwater vanes (with variable depth/submergence); it can create a
flexible/replaceable solution/measure. More standarization and large-scale equipment can reduce
the cost of construction.
Quality control of submitted designs is not always optimal. Approval is often provided after start
of the work, and the realization of recommended improvements is not always controlled.

67
Local and National data base is needed
The morphological studies of riverbanks are usually conducted by (Water Resources) Research
Institutes (including Universities). However, statistical and economic data on flood and bank
erosion damage is not yet incorporated in the activities of the planning institutes.
Integrated Water Management Planning should include flood and riverbank damage, protection
and maintenance, including cost-benefit considerations. These information’s can be used as
input for selection and planning of mitigation measures, and the optimization of the use of
available funds.
Little (or no) attention is paid to the optimization of the cost of the particular project; more
alternatives should be considered before final choice is made. More attention should be paid to
preservation of natural vegetation and the use of vegetation as a protective measure against
erosion.
Capacity building and cooperation:
There is a need for increase of capacity for copying this problem (capacity building).
Limited technical capacity of local offices: it needs improvement
Support from abroad for Research Institutes; it needs improvement
Support from research Institutes to design offices and local managing agencies should be
improved.
Education of young generation should be improved: upgrading programs at Universities and
more education/training abroad (IHE, AIT, etc.). In respect to the latest, it was recognized that
the poor working knowledge of English often forms an obstacle in practical realization of this
goal. The further improvement of English must be stimulated.
Dissemination/exchange of knowledge and experience/courses/ publications/reports, etc. should
be further stimulated in all fields related to bank erosion and protection.
Also, local and National database is needed.
Need for Master Plan and Spatial Planning
There is a need for general strategy (general concept of approach to erosion problems). Man has
to learn to live with erosion; only local measures/protection related to stability of river system
and its economic values will be implemented based on cost-benefit/cost optimization. It is
practically impossible to protect everybody/everything. The strategy should focus on:
- preserving natural systems as far as possible, and
- stepwise regulation of rivers and canals (applying, groins, revetments, vegetation, etc)
The Master Plan should consider such items as:
- overview of potential danger areas (mapping of problem areas)
- estimate of economic values
- cost-benefit analyse
- selection criteria and priorities
- suggestions for short- and long-term planning/actions
- functional and cost effective designs (methodology and tools)
Need for a long-term strategy for bank protection (in relation with Master Plan). However, bank
erosion and protection have often/usually direct interrelationship with safety aspects (stability of
dikes, boulevards, roads, urban and/or industrial properties, and other infrastructural objects)

68
and should be treated within one total system. The interrelationship with dike safety is very
strong in case of Red River.
The proposals for follow-up are required/should be formulated (in cooperation of IHE-RWS-
DDMFC-HWRU-NRA) on Capacity building on bank erosion forecasting and planning
mitigation measures. This should include Strategic Planning of bank erosion and measures and
creating local and National database.
The National Scientific Research Project: Study on forecasting and prevention bank erosion for
the Cuu Long River (Mekong Delta), 2001-2004 (MOSTE-MARD).
The mission was very satisfied with this national project. It demonstrates the serious intention to
solve the erosion problems in Vietnam and is a good example of cooperation between two
ministries and various agencies/institutes. The scope of the project is very ambitious and there
are high expectations concerning final results. However, the international support for this project
seems to be not adequately organized resulting may be in not optimal use of available research
potential.
This project should be evaluated at the end of 2004 and the conclusions and recommendations
should be formulated for future.
On one side, it would probably possible that some extension/supplementary studies for Mekong
Delta still will be needed, and on the other side, based on this experience from this project, the
similar projects can be formulated for other river basins, i.e., Red River, Central Vietnam. In
case of Red River the potential impact of bank erosion on safety of dikes must be studied in
detail.
In the follow-up more attention should be paid to international support concerning international
expertise and tools.
This follow-up project can be a part of the intended program on capacity building on riverbank
erosion and mitigation. It should also include a subject of new monitoring techniques (satellite
image, remote sensing, GIS-techniques, radar, etc.) where all institutions working in this field
should be invited to work together.
This project should also focus on establishment of operational mathematical models for main
river basins.

69
5. Analyse, final conclusions and recommendations of bank erosion mission
5.1 General considerations on Mekong and Red River delta’s
The Mekong Delta and Red River have experienced an unprecedented pace of infrastructure
development in the last few decades. Although such development has contributed significantly
to economic growth within the delta’s and nationally, it has also resulted in the genesis of a
number of adverse impacts on the society and natural environment. Many impacts have
originated from the failure to recognise the deltas as a dynamic biophysical system, and an
emphasis on rapid economic development based often on export commodity production.
This chapter provides an overview of major issues relating to the general aspects of flooding and
bank erosion and protection, and sets out various approaches for its protection, including
institutional, managerial, technical, economic and scientific approaches. The roles played by
public awareness and national and international co-operation are also discussed. In addition, it
offers advice for aid agencies for priority activities for technical and financial assistance for a
better integration of bank protection concerns in environment protection and infrastructural
development activities. Annexed is a checklist for environmental impact assessments of projects
affecting the river environment and other pertinent reference materials.
The main objectives are:
• to strengthen national and regional capabilities in erosion prevention and management;
• to develop a regional bankerosion monitoring and information management network;
• to strengthen the ability of country to implement and enforce international environmental
conventions and codes;
• to develop and initiate sustainable financing mechanisms which will support ongoing activities
beyond the life of the project.
First and foremost conclusion and recommendation is that future infrastructure development
should incorporate a greater appreciation of the system characteristics of the Mekong Delta.
An activity in one part of the delta will generate impacts in other areas. The analysis of
environmental and socio-economic costs of proposed projects will need to be carried out, not
only for the project areas, but also for the entire delta (AMRC, 2001). An improved
coordination of activities between the discrete project areas, promoted in the first instance by
an increased level of inter-provincial cooperation, would minimise the generation of cross-
project and cumulative impacts.
The prediction and gauging of physical, environmental and social impacts arising from
infrastructure development in the Mekong Delta and Red River are often hampered by the lack
of adequate baseline and pre- and post-implementation monitoring data. Monitoring of
existing situation and projects will assume increasing importance in providing input for future
projects, as the increase in the number of projects within the delta’s will bring about a
corresponding increase in cumulative impacts. It is not to say that data do not exist; various
government agencies and institutions both within and outside the Mekong Delta region have
carried out a number of studies, but their temporal coverage is often too short to enable trends to
be identified. Data collection may not be of benefit initially, but their utility grows with time.
There is also an urgent need to improve the coverage of environmental data on the Mekong
River catchment. This is especially important in light of uncertainty over the effects of current
and future dam construction on the delta. It needs to be a fundamental change in the planning
and design of projects, namely a move away from the “defensive” approach that pervades many

70
recent infrastructure projects. Instead of total control, prevention and elimination, emphasis
should be placed on partial control, amelioration and, in general, adaptation to the natural
environmental conditions. For example, there is more long-term benefit in replacing the current
approach of flood-control, involving much investment in hard infrastructure for defending the
delta plain from overbank flooding, with strategies which involve re-routing overbank flow and
partial protection.
Although it is fundamentally the conceptual basis behind infrastructure development which
requires change if it is to strike a balance with the natural environment, much can be achieved
through the improved design and management of existing infrastructure. For example, simple
morphological modification of the river/canal network, such as the elimination of sharp bends,
excessive branches or dead-ends, may substantially decrease sedimentation and associated
hydrological problems caused by rivers, while the enlargement of culverts and water
intakes/outlets along dikes may be all that is required to improve the throughflow of flood
waters on the delta plain, thus ameliorating problems associated with flow stagnation and
obstruction of overbank sediment transport. The solution to bank erosion problems in the
rivers/canals may lie in appropriate dredging practices, bank design and community education
on the use of river banks and vegetation.
5.2 Measures to cope with or counter bank erosion
Various bank protection methods as structural measures and a number of non-structural
measures can be considered for solving a particular problem of bank erosion.
River bank protection works have the positive impact of saving valuable land, which would
of been otherwise lost to the river. However, there is also a general concern that river bank
protection works in one reach may cause increased erosion at another unprotected reach,
and even on the opposite river bank, by changing river flow patterns. It can be considered
that such effects are usually localized. However, as the river is naturally meandering and
bank erosion occurs naturally, it is difficult, if not impossible, to measure the effects of
bank protection works on other reaches. However, this can be true for smaller rivers and
canals.
There is a large number of non-structural and structural measures for copying the floods and
bank erosion. Among them one may distinguish the following measures:
Non-structural measures
- floodplain/riverbank zoning
- code of regulation
- changing people’s attitude on floods and bank erosion
- flood forecasting
- (early) flood warning/erosion warning
- evacuation
- flood proofing
- control of human activities
- organizational (emergency) measures
- insurance
- etc.,

71
Structural measures
- summer dikes/embankments (structural or non-structural?)
- dikes (levees)
- floodwalls
- discharge sluices/pumping stations
- detention (retention) basins/areas
- storage dams/reservoirs
- chanalisation
- channel re-sectioning (deepening/widening)
- channel re-alignment
- diversion channels
- floodplain platforms/mounds
- (temporary) geo-membrane barriers
- bank protection
- vegetation
- etc.,
-
Individual measures or combination of measures can be considered in particular situation.
Some of the measures listed above will be discussed in following sections by placing them in
more broad context of problem mitigation.
Response of river to bank protection works and consequences for future
As far as the Mission understood bank protection works in Vietnam are carried out on the basis
of where the need arises. When an important reach is eroding bank protection works are
implemented over a certain length. Once bank protection works have been implemented,
economic development of the protected areas is accelerated. This creates a commitment to
properly maintain the bank protection and to prevent that the protected area is eroded in
subsequent years. Bank protection works however cause changes in the river system, which are
noticeable both downstream and sometimes in later stages also upstream and which may
necessitate additional bank protection works. The continued morphological development of the
reach with bank protection works may thus induce the need for additional river engineering
works both upstream and downstream. It is important to realize this in advance in view of
decision making on protection, for proper siting of the bank protection works and to limit the
funds required later for maintenance and extension of the bank protection works the Mission
holds the opinion that upstream and downstream effects of the bank protection works and
morphological developments upstream and downstream of planned works should be taken into
account in the design of the bank protection works. This also stresses the need for the
development of a long-term strategy for bank protection works along both the Mekong River in
Vietnam and the Red River.
5.3 Legal and socio-economic aspects
5.3.1 Institutional aspects and approaches
The effectiveness of national institutions in charge of environmental protection and erosion
mitigation in many countries remains limited. In particular, their resources are often small with
marginal influence on preparation of national development plans and development-related
decision-making. The failure to create effective national infrastructures equipped with

72
interdisciplinary expertise and adequate resources is a major constraint in river basin and coastal
environmental management.
Sustainable management of the river and coastal areas requires a variety of expertise and, above
all, a good understanding of the cross-sectoral nature of environmental issues. While narrow,
sectoral technical expertise exists in most countries, greater efforts should be devoted to training
experts in interdisciplinary skills. This requirement for more training of interdisciplinary experts
is certainly highlighted whenever developing countries have to respond to significant flood and
erosion disasters and water pollution incidents.
The national environmental policies and practices of most countries are embodied in legislation.
Since environmental concern is relatively recent, environmental provisions are often scattered in
a number of sectoral laws. Only a few countries have enacted comprehensive environmental
laws, with clear definitions of the functions and powers of the authority(ies) responsible for their
implementation and enforcement. Where environmental legislation has been enacted,
fragmentary application and weak enforcement have been major problems.
Often rivers and coastal environmental problems are shared by neighbouring countries.
Therefore, their solutions require international (bilateral, regional or global) co-operation.
Legally binding international agreements (conventions, protocols, etc.) can play a decisive role
in organising and maintaining such co-operation.
5.3.2 Legal aspects in Vietnam
A national water resources policy should state the principles, procedures and direction which
will be taken with respect to broad issues in the sector. In many countries, policy is developed
through a process of investigation and consultation and is used as a basis for legislation. In
Vietnam, policy development should be in harmony with, and guided by, the provisions of the
Law on Water Resources (LWR) and other legislation and with national goals and objectives.
The LWR is a framework document, which requires secondary legislation to bring it into effect.
While the LWR gives a great deal of valuable guidance for management and development of
water resources, it does not answer all of the important policy questions. Further work will be
required to develop both policy and legislation onimportant topics coming under the LWR.
Policy development priorities should be established to ensure that the most important topics are
addressed first. Although many of the responsibilities for water resource management have been
centralized (through the LWR) in MARD, it will be important to develop policy
recommendations through an open process in which all ministries, agencies and provinces with
an interest in the issues are able to participate.
Government Decree No. 86/2003/ND-CP (18 July 2003) regulates the functions, tasks and
organizational structure of the Ministry of Agriculture and Rural Development (MARD). One
of the functions is the state management over the irrigation and water services nationwide. The
tasks on irrigation and water services include such items as:
a) Manage construction, exploitation, usage and protection irrigation works, drainage works
for rural area in a unified manner.
b) Manage river basin, exploitation, usage and river integrated development per approved
plans in unified manner.
c) Manage construction, dike protection, prevention flood and storm works, and tasks
related to prevent and combat against flood, storm, drought, and landslide along the river
and coastline in a unified manner.

73
There are also some tasks formulated on science and technology related to the development of
necessary programs and new technology application in the domain of irrigation/water services
and others, including management of standards and information of science and technology in the
domain of the ministry in accordance with legal documents.
Additionally, the international cooperation in the domain of irrigation/water services is
mentioned as an important task of the ministry.
The Department of Dyke Management and Flood Control (DDMFC) is the responsible
implementing ministerial agency for flood mitigation and bank erosion problems in Vietnam, in
collaboration with provincial units.
Large-scale problems and resulting mitigation activities/projects are usually organized and
financed by the MARD/DDMFC on behalf of the Central Government, in collaboration with
relevant provinces. Some local projects are often organized by provincial, district or local
bodies, or even by individual persons. However, the ministerial funds for tackling of flood
damage and large-scale bank erosion are rather limited in comparison with the scale of the
problem. The selection of the sites for national financial support/projects is based on the
following (socio-economic) criteria:
Priority I: Populated urban areas and/or safety of dikes
Priority II: Densely populated rural areas (larger villages) and/or high economic values
Priority III: Rural areas
Priority IV: others
In general, bank erosion is strongly related to the flooding and it is often difficult to separate the
bank erosion damage from flood damage. The figures on damage amount are not very consistent
but one has to think in terms of some millions US dollars per year, in tens of human casualties ,
some tens or even (locally)hundred meters of land losses, and hundreds of demolish houses,
often in combination with relocation to other areas.
These indicative damage figures must be confronted with the annual budget of DDMFC of
only/about 10 million US dollars per year.
The above stresses the hard need for national Master Plan on Spatial Planning and River Basins
Management, which should also include the national long-term strategy for bank protection and
risk-based economic selection criteria. The Water Resources Planning Institutes in cooperation
with the Water Resources Research Institutes should prepare the basic ingredients to these new
policy documents.
Other relevant legal documents
The following important legal documents for water resources management have been issued:
- Law on Water Resources No 08/1998/QH10 dated 20/05/1998.
- Ordinance on Exploitation and Protection of Hydraulic Works No 36L/CTN-dated 10/09/1994.
- Ordinance on Dykes and Dams No 26/2000/PL-UBTVQH10 dated 7/9/2000.
- Ordinance on Flood and Storms Controlling (amended and supplemented
24/8/2000).
Decrees of Guidelines on the implementation of the Law on Water Resources and the above-
mentioned Ordinances have also been enacted. Based on the Law on Water Resources, the
Government has issued the Decree No 179/1999/ND-CP stipulating the implementation of the
Law on Water Resources.

74
The National Council of Water Resources was also established in accordance with the Decision
No 67/2000/QD-TTg dated 15/6/2000.
In conformity with the current situation and to the Law on Water Resources, the Standing
Committee of the National Assembly has compiled three Ordinances (revised):
- Ordinance on Exploitation and Protection of Hydraulic Works (currently being amended and to
be approved).
- Ordinance on Dykes and Dams (already amended and issued No 26/2000/PL-UBTVQH10
dated 7/9/2000).
- Ordinance on Flood and Storms Controlling (amended and supplemented 24/8/2000).
- There is also a legal system of water fees for use of water for irrigation. Water fees and their
calculation are based on a national decree that the government cabinet promulgated in August 1984
(112 HDBT, 1984). This decree specifies that all organizations and individuals benefiting from
irrigation, drainage and other hydraulic public services, have to pay a water fee to hydraulic
companies.
5.3.3 Socio-economic and environmental aspects of flooding and bank erosion
Flooding and resulting bank erosion are a natural and recurrent phenomenon in the Mekong and
Red River Delta. Especially in the Mekong Delta, it is a process, which drives the evolution of
the delta plain and provides constraint to the human and economic development of the delta
(AMRC, 2001). Due to the low elevation and relief of the delta plain, floods in the Mekong
Delta are typically prolonged and aggravate the problem of poor drainage.
Another socio-economic effect of flooding and poor drainage is an increased cost of
infrastructure development and maintenance. For example, major roads need to be constructed
on an embankment, and buildings on high foundations, mounds or stilts. Roads, which are
submerged during the flood season, require frequent maintenance and the prolonged period
during which they remain impassable hinders communication, trade and transportation.
Damages due to flooding and erosion amount to tens of billions of Vietnamese dong (VND) per
annum.
However, not all socio-economic effects of flooding are adverse. Sediment deposition effected
by floods plays an important role in rejuvenating soil over geological time scales. Although it is
debatable whether the annual contribution of soil nutrients through flood-related sedimentation
is sufficiently significant to improve crop growth, it is without doubt that overbank flooding and
the associated sedimentation contribute to improved soil properties in the long-term, through the
creation of higher, better-drained land (e.g. along levees), by flushing out accumulated toxins in
the soil, and by counteracting unfavourable changes to the physical and chemical properties of
the soil, e.g. in the absence of replenishment with new material, the soil may become compacted
and partially reduced with age, hindering root growth and nutrient uptake and increasing the
possibility of toxicity. Furthermore, the annual flooding brings increased opportunities for
fisheries activities.
Sedimentation and erosion also present a challenge to the human utilisation of the Mekong Delta
waterways. The upper delta experiences very rapid rates of channel migration (banks erosion
rates are commonly up to 20 m / year), caused by the lateral accretion of point-bars and mid-
channel bars / islands, and the downstream migration of mid-channel bars. Mid- and lower delta
channels are more stable (bank erosion rates are commonly 5-10 m / year), and channel change
here is mainly caused by the slow accretion of elongated point-bars and mid-channel bars.

75
Bank erosion is considered a serious socio-economic problem in the upper delta provinces of An
Giang and Dong Thap provinces. Problems are especially severe at Tan Chau on the Mekong
branch in An Giang, where erosion rates attain 30 m/year, and a large number of households
have had to be relocated due to destruction of their dwellings through bank collapse. Bank
erosion has resulted in major disruptions to local livelihoods, and financial burden on the
provincial government (cost up to the present amounts to hundreds of billions of VND) by
necessitating the relocation of inhabitants and localised bank protection works. Losses due to
bank erosion appear to have increased in the last decade, probably due to the growing urban
population and the resultant concentration of activity and capital along the waterfront. The
severity of erosion at Tan Chau is largely attributable to the sharp meander-bend morphology,
which focuses the river flow energy onto the concave bank (where the town is situated). The
gradual downstream rotation of the point-bar on the opposite bank has resulted in a progressive
downstream shift in the zone of erosion; stretches of river bank upstream of Tan Chau, which
formerly experienced severe erosion are now experiencing bank accretion.
Other erosion hotspots further downstream within An Giang (e.g. at Long Xuyen) are mostly
associated with the downstream migration of mid-channel bars, which creates a shifting zone of
erosion downstream and to the sides of the bar, and a zone of accretion to its upstream. High
flow velocities in main rivers and canals contribute to bank erosion and enhanced transport of
sediment in the main canals. Upon entering the smaller tertiary canals, much of this sediment is
deposited due to an abrupt drop in flow velocities, adding to the cost of canal maintenance.
Large-scale bank stabilization through hard engineering produces often undesirable side-effects
such as rapid channel aggradation, and accelerated downstream erosion/sedimentation.
Sedimentation, which accompanies bank erosion, also represents an economic cost in places,
through the shoaling of navigation channels, the stranding of wharves, docks and other water
transport infrastructure, and the blocking of entrances to canals. However, sedimentation in the
main distributary channels is often regarded by many as an economic benefit, given the
predominantly sandy nature of channel sediments, and the increasing demand for construction
sand driven by urban expansion. Numerous sand dredging operations exist along most of the
length of both the Mekong and the Bassac branches and Red River; an individual operation may
extract volumes in the order of 104 m3 /year from the bed of the channels.
Many of the environmental problems resulting from large-scale infrastructure development
interventions in the Mekong Delta may be viewed as a consequence of failure to recognise the
delta as an environmental system. At a larger scale, the entire delta may be viewed as a
component of the river catchment system. In this context, the delta is a sink and a transfer zone
for matter derived from the more upstream parts of the catchment and transported downstream.
However, this sediment formerly distributed over a large area of the delta plain is now
accommodated in canals, which manifests itself in rapid siltation rates and high cost of
maintenance of canals.
Given the inherent role of deltas as sediment sinks, and the rapid rates of geomorphic processes
driven by a large river discharge and sediment load, the deltas are the highly dynamic
biophysical environmental systems (AMRC, 2001). As such, the deltas are in a constant state of
evolution. Such environmental change is apparent at many different spatial scales, for example,
a mid-channel bar undergoes accretion and downstream migration within a channel system,
which evolves through channel shifts within the meander belt and occasional avulsions, and
which itself is part of the expanding delta system. Trends in geomorphic evolution may be

76
progressive, cyclical or episodic, and there is commonly a link between the spatial and temporal
scales of evolution; namely, that small-scale geomorphic features, such as mid-channel bars,
evolve over short time scales while the evolution of larger-scale features, such as the channel,
and the encompassing delta system takes place over longer time scales. Analogous relationships
may be observed in the biological environment; for instance, the time required for the
establishment of a viable forest ecosystem far exceeds that required for the establishment of
individual trees, which compose it.
Reference:
AMRC, 2001: Environmental Issues and Recent Infrastructure Development in the Mekong
Delta: review, analysis and recommendations with particular reference to largescale water
control projects and the development of coastal areas; Takehiko ‘Riko’ Hashimoto, Working
Paper No. 4, Australian Mekong Resource Centre, University of Sydney, June 2001.
5.3.4 Some observations during the mission on institutional, legislation and economic
matters
Institutional matters:
The actual institutional arrangements should be reviewed as far as they concern bank erosion
and protections. At present the responsibility for priority listing, design and tools as
mathematical modeling and physical modeling, research, management and maintenance of bank
protections has been distributed over several institutes and offices, while other organizations
have to be consulted regularly (on navigational, financial, sociological (compensation for
damage), ecological and agricultural aspects. A small redistribution of responsibilities might be
considered.
The (intended) Master Plan should include criteria for the selection of proposals for bank
protection projects. The Provinces send these proposals to the Ministry for approval and
funding. The decision making process might be reviewed including the funding of river training
projects. At present no budget is available for maintenance of existing bank protections. The
structural lack of funding for maintenance activities needs urgently attention.
Legislation:
The Province arranges an area for resettlement of the people who have lost their land by bank
erosion. In addition the victims of bank erosion receive only small compensation for their loss (a
few hundred thousand Dong per family).
In the past no legislation had been made to prevent construction of buildings too close to the
bank. In a new law it is decided that in urban areas a margin of 50 m from the riverbank should
be free from (permanent) buildings. In practice it is difficult to apply this law strictly.
Administrative and financial problems
• A permanent shortage of finance even on several very serious/danger occasions
(i.e., evidence of very steep underwater profile in front of road and/or buildings) .
This results in a slow start to project activities and later on often , on partly
realization (completion) of projects;
• The initial request from the local agencies for the river bank protection works is
often refused because of lack of budget and it is waiting for collapse/failure to may
apply for the additional finance from the national disaster funds;

77
• A number of small stretches were protected (or are under construction) with the
grants provided by international donors.
• The main technical problems encountered were those with the existing designs in
the inception period and the problems with the meandering of the Mekong River
and red river.
• This mission concerns bank erosion and protection and not flood management and
mitigation. However riverbank erosion should be a major concern when planning
flood protection works and is worthy of further comment. This is because it is usual
practice to locate the dikes as near to the river bank as possible in order to
maximize the areas protected. The top-soils on the river banks are generally very
fertile and intensively farmed. However if the dikes are located close to an eroding
riverbank then they will eventually be undermined and collapse. For example it may
be considered appropriate to locate a dike 50 m from a river bank. However
Mekong River bankerosion (and some locations of Red River) can be up to 20 m a
year (or even more) and if this is the case, then the dike will have to be moved in a
few years. The same is the case when the properties are buit close to the bank.
• As stated previously, bank protection works are expensive and cannot be
automatically considered in a flood protection project, particularly in rural areas,
where in economic terms the cost can be higher than the value of the assets, which
are to be protected. The final choice/decision, however, should be taken based on a
proper cost-benefit analyse/criteria; the further development of the use of this
technique should be stimulated.
• Evacuation and resettlement are often applied as emergency actions in Vietnam. An
important aspect of resettlement is whether the resettled population is satisfied with
their new socio-economic and environmental conditions, and whether sufficient
assistance has been provided by the government/provinces (?), both financially and
in logistic terms for the re-settlers to be able to regain acceptable living conditions.
The Mission is not aware of any studies on the conditions of the re-settled
population and hence they advocate that such a study is carried out. Only when the
results of such studies are available, it can be assessed whether relocation of
population threatened by bank erosion is an acceptable alternative.
• To reduce the problems associated with evacuation and resettlement the Floodplain
zonation should be applied on a larger scale.
The mission identified also some issues which are not so explicitly stated in the Law:
 Development of national policy and policy-making mechanisms including floods and bank
erosion;
 Development of a strategy or action plan on bank erosion problems and mitigation, including
even the necessity of drastic measures (as illustrated with the Chandpur case below);
 Future management and maintenance, including financing;
 Development of a strategy or action plan to guide capacity building in water resources
specifically related to floods, bank erosion, river training, ecology, and environmental and socio-
economic impact.
Some of these issues will be discussed below (in the following sections).
5.4 Integrated River Basin Planning and Management
Planning and management of water resources needs to be carried out in an integrated manner,
taking into account all water issues, needs and possible solutions in a balanced way. Effective

78
planning can help to achieve coordination between sectoral water use, resolution of conflicts at
an early stage, and coordination between water related aspects such as land use, wastewater
discharge, etc. Planning is also an important part of the state management of water, as specified
in the Law on Water Resources (LWR). Improved models and other decision support tools and
planning procedures are needed, as are improved stakeholder consultation and involvement
practices.
An Integrated Water Management Plan should be developed and implemented in relation to all
activities associated with the flood mitigation and associated problems of bank erosion. This
plan should include:
- details of specific regional items and problems, and general national philosophy/strategy
- details of monitoring of impacts on physical resources, human and economic development and
quality of life values;
- details of measures to mitigate any anticipated negative impacts.
This Integrated Water Management Plan should be supplemented by other documents, such as:
Master Plan on Flood and Bank Erosion Mitigation, Environmental Management Plans,
Resettlement Monitoring Programs and Action Plans, etc..
The development of appropriate Management Plans should ensure monitoring of problems and
possible consequences, and the implementation of appropriate mitigation measures as required.
It will enhance the country’s sustainable economic and social development and contribute to the
growth of the regions.
5.4.1 Need for a long-term strategy for bank protection
The Mission noticed that no long-term strategy has been developed for either the Mekong River
in Vietnam or the alluvial part of the Red River. As a consequence bank protection works are
implemented when and where the need is highest and at places where at that moment the river is
eroding. The effect of this is clearest for the Red River. Although the Mission did only visit one
side of the river (the other side is ”out of view” for the province the Mission was visiting, it is
clear that the many implemented river bank protection works are creating a fixed river with a
very curved pattern. This planform may not necessarily be the best planform of the river inview
of the continuously rising flood levels. Possibly a straighter course might have advantages.
The Mission holds the opinion that for both rivers a long-term strategy for bank protection and
river fixation has to be elaborated at short notice. Bank protection works implemented in the
coming years should fit into such a long-term strategy.
The long-term strategy for the Mekong and the Red River will probably be different. Although
the Mission was not in a position to do an extensive study of the best strategy for each of the
rivers, it suggests that the best strategy is determined by the river characteristics on the one hand
and socio-economic aspects on the other hand. For the Mekong River the long-term strategy
could consist of:
• stabilization of the flow distribution around the stable islands (in view of vested interest)
• stabilization of islands
• protection of important areas with bank protection works and upstream river training works
For the Red River the preferred strategy could be:
• a continuous bank protection along all outer bends
• in due time narrowing of the river to induce bed degradation to counter-act the present
aggradational trend.

79
A proper and appropriate long-term strategy for bank protection works and river training can
only be developed when the river characteristics are properly understood and this calls for a in-
depth study of the morphology of both rivers and other relevant aspects. Unless such a study is
carried out and such a long-term strategy has been developed and is accepted by the population,
the risk is present that investments in bank protection works in due time may turn out to be a
less effective than initially anticipated.
Also Floodplain zonation can be an important instrument to regulate and direct the use of the
floodplain and riverbanks. This measure should be included in the Master Plan of the Ricver
Basisns and in the Strategy for bank protection.
5.4.2 Proposed Integrated Strategy Plan for River Banks or Master Plan
Many locations along the Mekong Delta and Red River in Vietnam experience increased bank
erosion in recent years. The need for priority ranking of these locations and the potentially
complicated morphological interactions require a Master Plan or an Integrated Strategic Plan to
realize the most economic and efficient control of the bank erosion problem.
The potential benefits of a Master Plan for bank protections are not jet recognized, because only
a few bank protections are under development. It is expected that in due time the response of the
river to bank protections will be observed, but at that time costly corrections are the only option.
Therefore the Mission recommends to prepare a Master Plan for the construction of bank
protections along the rivers in the Mekong Delta and Red River Delta.
The Master Plan can be a tool to prevent unpleasant financial surprises for the decision makers.
A Master Plan for river training should deal with the following elements, varying from a high
conceptual level to a detailed level of the bank protection structure:
1 Objectives
2 Priority for implementation
3 Illegal land reclamation
4 Selection type of bank protection
5 Design method and design parameters
6 Pilot testing
7 Update Design Guidelines Standard
The description of components of the Master Plan is given Section 4.5.3.
Note: the Mission is not expecting that the items mentioned above cover fully the content of the
Strategy or Master Plan and therefore, they should be interpreted just as indication and/or
examples.
5.4.3 Floodplain zonation
Floodplain zonation can be an important instrument to regulate and direct the use of the
floodplain. This is important for at least two reasons. Firstly, the floodplain is important for the
conveyance and storage of floodwater. This property should be maintained and hence it is
important to indicate what is acceptable in the floodplains of the Mekong River and the Red
River. Secondly, it should be prevented that people settle in areas which might be eroded within
the foreseeable future and for which there are no plans to protect the area by bank protection
works. This calls for legislation to regulate settling and other activities and it calls for
enforcement of this legislation.

80
In The Netherlands already in 1908 the so-called River Law was introduced, with the specific
purpose of floodplain zonation of the floodplains of the main rivers in the Netherlands (Rhine
and Meuse). An essential part of the River Law are maps on which it is specifically indicated
which areas of the floodplain are important for conveyance and which for storage. Building of
houses, raising of the ground level and building of roads in areas reserved for conveyance is in
general not allowed unless a permit of Public Works Department is obtained. Such permits
usually come with requirements which should be fulfilled before the permit can be used. Such
requirements usually consist of compensatory measures like the lowering of other parts of the
floodplain. In areas reserved for storage more activities are allowed, but still a permit is needed.
In the River Law no provision is made for areas threatened by bank erosion because th emain
rivers in The Netherlands had been regulated already in the 2nd half of the 19th century, and bank
erosion is prevented by long series of groynes and revetments.
The Mission has understood that in Vietnam no legislation exists that can be used for floodplain
zonation. Nevertheless legislation similar to the Dutch River Law can be made and
implemented. Different from the Dutch “maps” in Vietnam it is important to identify also areas,
which are reserved for future bank erosion. In these areas settlements should not be allowed. A
crucial aspect of such type of legislation is enforcement. The Mission holds the opinion that
mechanisms should be developed which allow for better enforcement. Experience from The
Netherlands (and other countries) can be of use in this respect.
In conclusion, coordination is needed with regard to policy and legislation, information
management, water resource planning, operational programs, and emergency response. At the
central level this will involve such things as development of a more coordinated strategy for the
water resources sector, approval of budgets, approval of investment projects and river basin
management plans, and improved communication and dispute resolution between sectors and
major water users.
Coordination and clear definition of roles is needed between water management agencies and
organizations from the central down to the local level. Water management agencies at all levels
need training, facilities, financial support and operational guidelines for effective water
management.
It will be necessary to build coordination in both a “top-down approach” and a “bottom-up
approach”. The National Water Resources Council (NWRC) can play a roll in this process and
help to establish policies and processes for coordination at the ministry and river basin level.
The Council can also initiate developing of methodology/principles and guidelines for inter-
ministry and basin level coordination and for the necessary policy and technical documents
(Master Plans, Strategic Bank Erosion Plans, Resettlement Plans, Monitoring Plans, etc.).
In respect to the financing of bank protection projects the reference can be made to the Decree 112
HDBT, 1984. This decree specifies that all organizations and individuals benefiting from irrigation,
drainage and other hydraulic public services, have to pay a water fee to hydraulic
companies/agencies. It is worthy to investigate the possibility of applying this regulation also to
bank protection projects where local interest of people is directly present.

81
5.5 Capacity building
5.5.1 Technical and scientific approaches
Long-term policies and management decisions need to be based, as far as possible, on facts
collected, analysed and interpreted according to scientific criteria. Knowledge of the main
causes and effects of floods, erosion, pollution and physical degradation is, in most cases,
sufficient to provide reliable advice as to practical control measures. Nevertheless, further
research is needed because existing databases and our understanding of the processes shaping
the natural state of the water environment are generally inadequate for reliable predictions about
changes and trends. The importance of monitoring “key” morphological and environmental
indicators cannot be over-stressed. By appropriate monitoring, in combination with modeling
techniques, it is possible to track progress in achieving technical,environmental and economic
objectives and to learn from this experience the relative effectiveness of different
approaches/measures.
Education and training is an area of identified priority for the Vietnam/MARD. As part of the
process of reform, modernization and integration with the international community, Vietnam/
MARD is keen to access formal training from developed countries such as Australia, Denmark,
and Netherlands. In the scope of strengthening of higher education and training in Vietnam a
number of specific programs already exists (i.e., Coastal Engineering at HWRU Hanoi). In this
perception it is also desirable to upgrade the existing curriculums at Universities (i.e., HWRU)
and to establish scholarship programs abroad for flood mitigation, river training, bank erosion
and protection, environmental aspects, new technologies and other related items.
Implementation of these items will contribute to the capacity building on the bank erosion
mitigation and tackling of future problems.
5.5.2 Staffing of Dike Department
Where the future River Basin Organizations (RBO’s) are the institutions, which will deal with
overall river basin development, the Dike Department (DDMFC) is the institution which should
have and take the responsibility for the long-term strategy for river training and bank protection
of both the Mekong Delta and the Red River. This should be done in close cooperation with the
RBO’s. An important task for the Dike Department in this respect is the preparation of a
yearbook on the development of the two rivers, on the basis of information provided by the
respective provinces and local districts. In this yearbook the progress of the construction of the
bank protection works along the two rivers should be reported upon together with the
morphological response of the rivers. Water Resources Research Institutes should support the
Dike Department in these activities.
Another important task is the maintenance of the data base as proposed in Chapter 4.2, Section
. Finally the Dike Department could serve as central point for the collection of experiences with
the implementation of the bank protection works by the provinces. Possibly a regular meeting
(e.g. twice a year) could be arranged where the officials of the different design offices of the
provinces along the Mekong and the Red River exchange their experiences with planning and
implementation of bank protection works and the morphological response of the rivers. This
could be the basis of the yearbooks.
For these central activities some additional specialized staff may be needed. If so, the Mission
very much favours such an extension of the staff of the Dike Department, as it feels that the
Dike Department must be allowed to play this central role.

82
5.5.3 Need for training of local staff
The Mission has visited a number of provinces and has had discussions with a number of
officials and staff of provincial engineering bureaus. Although in general the Mission is
impressed by the dedication of the staff involved, at the same time the Mission feels that the
design work is done on the basis of a number of standard (somewhat outdated) guidelines. There
is limited scope for original solutions based on special local conditions. Moreover there is
limited feeling for the special aspects of implementing bank erosion works in large and dynamic
morphological systems like the Mekong and the Red River. For this reason the Mission proposes
to prepare and implement a special training course for the different engineers involved in bank
protection works along the Mekong and the Red River, where the experience gained by the
individual members of the Mission in The Netherlands and abroad is transferred to the
Vietnamese staff of the provincial engineering firms. This course should be designed in such a
way that the experience presently gained by the provincial engineering firms is mobilized as
well and a proper exchange of experiences is achieved.
5.5.4 Modelling support for designs
During their field visits and during the visits to the Water Resources Institutes and the
University both in Ho Chi Minh City and Hanoi several times the use of models as support for
bank protection studies was mentioned. In principle two types of models can be used in bank
protection studies, notably:
• physical models, both fixed bed and mobile-bed;
• mathematical models allowing to study flow, sediment transport, morphological changes,
bank erosion.
The Mission favours the use of models, but at the same time wants to make a number of critical
remarks on the use of models (see also Sections 4.3.4 and 4.3.5):
• Physical models subject to scale effects (e.g. scour depth not on scale in distorted mobile
bed models which simulates the overall morphological behaviour correct)
• It is an illusion to think that bank erosion can be studied in a mobile bed model: modelling
of bank erosion is at present (and probably also in the far future) not possible in a
quantitative way.
• Physical models can be used to study local scour near bank protection works. Until now this
cannot be studied with mathematical models.
• Mathematical models require schematisations and proper (bank erosion) equations. Bank
erosion predictors are not developed up to the level that with mathematical models bank
erosion can be predicted accurately.
• Nowadays mathematical models can be used advantageously to simulate morphological
changes, but they need proper calibration and verification.
• In large projects often a combination of models is used (e.g. a mathematical model for a
long river reach and a physical model for details like local scour). This is called a hybrid
modelling approach.
• Models need prototype data for calibration and verification. Models are as good as the field
data available. This implies that much more field data before and after the construction of
bank protection works are needed.
The present situation in Vietnam regarding modelling is the following. Within Vietnam some
knowledge on physical modelling is available, but in view of the Mission this knowledge should

83
be updated on the basis e.g. of the extensive experience with physical modelling for bank
protection works in Bangladesh and the improved insight into scale effects obtained over the last
two decades (Struiksma, 1983; Struiksma & Klaassen, 1987; Haque, 1999). A special course on
possibilities and limitations of physical models seems appropriate.
Within a support project for the Water Sector in Vietnam, DANIDA is funding the
implementation of MIKE 21C in both Water Resources Research Institutes. This
implementation includes the following modules of MIKE 21C: flow, sediment transport,
morphological changes and bank erosion. A critical attitude towards these models will be
required. These models are as good as the understanding of the relevant processes is, and as far
as bank erosion is concerned this understanding is quite limited. The maximum these models
can provide is an indication of where bank erosion might occur. Predictions based on field data
(see Chapter 4) might for the time being be the better alternative. The Mission assumes that as
part of the implementation sufficient training in the use of MIKE 21C will be given. This
training should be sufficiently critical.
Research capabilities at HWRI and involvement of Universities in bank erosion studies and
monitoring
Research capabilities at the two Water Research Institutes are limited. The same holds for the
funding of research into the characteristics of the two main rivers and the morphological
response of the rivers to bank protection and river training measures, although the funding of
Ministry of Science of a study into bank erosion in Vietnam is a very promising sign. The
Mission favours the involvement of the Hanoi Water Resources Universities into these studies.
BSc and MSc student can carry out research into the field of bank protection works and river
training at low costs, and the WRI’s could provide some funding for the students as an incentive.
These studies should also include field studies to study the response of the river system to bank
protection works. Simple measuring methods should be developed to monitor morphological
changes. For the analysis of the data use can be made of GIS.
5.6 Data and Information Management
Water resources data and information management need to be improved in order to support the
policy, planning and operational needs of improved integrated water resources management.
This will include improved inventories and assessments of surface water, groundwater and water
quality, river hydrograph and morphology, flood and bankerosion damage, improved accuracy
and electronic management of data, and in particular, better sharing and dissemination of data
and information. Coordinated data systems, planning and decision tools and public information
procedures are needed. The need for and the components of data base for bank erosion are
discussed in Section 4.2.5.
The National Water Resources Council and its Office (ONWRC) can assist MARD to
coordinate actions toward the development of a water sector information system at the national
and basin levels. Implementation of such a system would likely be the responsibility of MARD
and other key agencies.

84
5.6.1 Monitoring, data base and evaluation (see also Sections 4.2.4 and 4.2.5)
The Mekong River and Red River are actively meandering and the main technical problem
is from riverbank erosion undermining the flood protection dikes and other protection
structures. Therefore estimates of past and future erosion rates should be an important part
of the investigations for all flood protection and bank protection projects on the Mekong
River and Red River Monitoring and Evaluation of the works completed are important for
long term management of the flood defence system, for setting priorities for new works and
adjusting Operation and Maintenance (O&M) requirements.
The monitoring of important physical processes affecting the flood protection facilities
must include:
• rates of river bank erosion along the dike and in particular immediately upstream
and downstream of river bank protection works:
• dike crest levels, width and slopes;
As stated several times, river bank erosion is the process that most threatens the flood
protection facilities. Without such erosion the dikes (also boulevards, roads) would only
require routine maintenance, except when overtopped by rare flood events. If river bank
erosion undercuts the dike during high river levels then flooding may occur through
breaches in the protection. Erosion rates must be measured at locations, especially all along
the dikes, so that the examination of the records will indicate priority areas which require
action soonest and those which appear stable or at least require no action for several years.
Beyond the simple use of the rate of erosion, it is recommended that specialists be
periodically requested/hired for an interpretation and evaluation of the records. A specialist
may perhaps separate out various influences such as unusual bank materials, perhaps
changes to the natural river regime, for example those caused by local mining of sand and
gravels from the river, unusually high drainage discharges through the river bank caused by
irrigation facilities and most importantly, the large, powerful effects of river meanders.
Inspection and monitoring of areas immediately upstream and downstream of river bank
protection works are important as often the length of works is affected by available finance
and it may be that erosion is continuing upstream and downstream and that the works may
become isolated and fail from the ends.
Dike crest levels should be surveyed at least every two years or more frequently in areas
subjected to settlement and/or excessive vehicle use, for example when mined sands and gravels
are transported. The crest levels should not be allowed to fall more than 0.2 m below design
levels as this is the amount of freeboard allowed for dike wear and tear. Excessive wear and tear
in particular locations should be investigated. In sheltered reaches or protected by natural
vegetation, the freeboard allowance for waves can be reduced and if circumstances change, such
as buildings demolished or trees cleared, and such areas become exposed to waves, then the dike
should be raised accordingly.
5.6.2 Requirement for operation and maintenance
The flood and bank protection works will not be sustainable if the dikes, bank protection/
revetments and toe protection/ and drainage/pumping structures are poorly operated and
maintained through lack of operation and maintenance. Maintenance programmes are
necessary to ensure that the flood defence facilities are fully serviceable before flood

85
periods. Operation procedures are necessary for the closing and opening of the water
control structure gates or stop logs.
It should be recommended that an Operation and Maintenance Sections (O&M) be formed
at provincial agencies. The O&M must have finance, staff and resources to undertake
many activities including the continual inspection of all the dike/bank facilities, to carry out
routine and periodic maintenance, be able to perform emergency activities if necessary
during the flood season, undertake system operation, monitor rates of river bank erosion
and prepare survey, design, drawing and cost estimates for special works which may
require additional inputs from consultants.
It is also recommended that donor assistance is required for technical and financial assistance
with O&M activities. Technical assistance is necessary to assist the proposed Agency with on-
the-job and formal training and the rehabilitation of a number of existing water control
structures located downstream of the Urgent Phase area that are in an extremely poor condition
and in an urgent need of repair. Donor finance is required to fully equip and operate the Agency
for several years and for the rehabilitation of the structures.
5.6.3 Recommendations for further implementation work
A critical location can be defined as one where the dike was immediately adjacent to an
eroding riverbank, the dike crest had partially collapsed, causing a traffic hazard, and the
dike could not be realigned because of adjacent properties.
It is again emphasized that the first priority deserves the selection/definition of the
urgent/critical phase works, which may have improved the flood protection facilities
protecting urban and valued economic areas and may have reduced the risk of flooding or
high economic damage. Therefore the recommendations for further work include:
• investigations for increasing the knowledge on the critical sites/locations, especially
when bank erosion endangeres the level of flood protection or valuable
properties.Therefore, so that it is more accurately known what are the erosion rates
and the gradient of underwater slopes, it is recommended that additional river level
gauging stations should be installed and the amount and frequency of monitoring
should be increased as soon as possible;
• the construction of further lengths of river bank protection works in critical areas to
secure the flood protection facilities and prevent the loss of urban and agricultural
land;
• assistance with O&M of the dikes, structures and bank protection works to maintain
the level of protection.
Investigations for the riverbank protection works should include the possible (future) large-
scale morphological changes of the rivers and the future development/destination of the
areas under consideration.
Probably the main design consideration with riverbank protection works is the cost as they are
expensive to construct on rivers such as the Mekong or Red River. The use of rock and gabions
is far cheaper than methods using concrete. However the geotextile and good quality mattresses
have to be imported (i.e., rock filled Reno mattresses, 6 × 2 × 0.3 m thick galvanized wire
baskets). For local projects, wherever possible, the use of local rock and local materials and
alternative structures should be stimulated. However, in case of local solutions and materials the
proper design guidelines should be established. Also, the placing of rock for transitions at the

86
upstream and downstream ends of the mattresses and also at the toe of the works should be
considered on a larger scale.
It is concluded that still considerable further inputs are required to maintain the present/ required
level of flood protection, or increase the level of protection to that required by a local
circumstances (i.e., safety of cities) and also to secure the protection by the construction of
further lengths of river bank protection works.
5.6.4 Environmental Impact assessment (EIA)
Environmental risk and impact assessments should be mandatory for all projects in these areas,
which are likely to affect human health and well being, environmental quality or biodiversity
(see also Annex). The “precautionary principle” should be one of the main guiding principles in
preparing the assessments, and the assessments should be carried out before any such project is
implemented. Economic incentives and cost-recovering fees should also be considered as
environmental management tools.
5.7 Conclusions and recommendations in short (as a reminder)
• Need for better monitoring of riverbanks and yearly reporting (including local and/or
provincial Data-base)
• Better prediction of bank erosion rates and scour
• Better insite in erosion processes including geotechnical aspects
• Numerical modeling of river morphology needs further development and can become an
important tool for prediction of erosion
• More clearnes in Non-structural measures and procedures (early warning, evacuation plans,
compensation, etc.)
• Design rules for Structural measures (traditional vs. advanced, including vegatation)
• Need for Design and Maintenance Manuals (including alternatives) and Guides for
Environmental Impact Assessment
• Maintenance aspects should be included in the design
• Legal framework, funding and prioritizing criteria needs more clearness/transparency
• Need for Master Plan for whole rivers
• Need for a long-term strategy for bank protection
• Offices and Technical staff
- (often) limited number of personnel
- (sometimes) necessity of training
- better support by Institutes
- some improvement of capabilities of Institutes and co-operation (a.o., shearing
tools/models) is still needed
- better support for design and cost-benefit studies
• More international (IHE) education supported by Vietnamese experience/data base/monitoring
• Learn from international experience (negative/positive)
(i.e., permeable groins, combined floating alternatives, environmental friendly protection)
• Better comparison of alternatives (functioning and costs)
• Effect of partly completed projects ? (response of the river)
• Improvement of surveying and data processing
• More integrated approach and visual aspects (urban area)

87
• Continue to investigate and evaluate the human impacts on the riverbank erosion on river
systems.
• Need for long term studies including future developments
• Involve people in (open) discussion of the problems (the boss is not always right)
• Continue to develop and perfect the technology for forecasting, warning, and mitigation of the
erosion on riverbanks on the Vietnamese river systems.
• Establish a co-operation platform (periodic consultations of all agencies working in the
same/common field)
In conclusion: there is little money and many problems in Vietnam; the actual management
likes more as the emergency (disaster) management rather than the planned management. On the
other side, it should be said that DDMFC under this very difficult/limited possibilities is
doing/providing a good and responsible work and products. However, as listed above, still some
improvements in approach and working methods is possible to optimize the ffectiveness of these
limited funds.
5.8 Proposed Action Plan 2004-2006
1. Preparation of project proposals on Capacity Building for flood and riverbank erosion
mitigation (half 2004);
2. IHE Delft education for engineers from Water Resources Research Institutes on river
training, erosion prediction and protection measures; (start in September 2004);
3. Organizing of a Workshop/Short course on Erosion Prediction and Bank Protection, in
combination with presentation of the final results from the National Program: Program KC. 08:
Enviroment and Natural Disaster Prevention; project kc.08.15; Project title: Research on the
causes and the solutions to prevent riverbank erosion and deposition for the Lower Mekong
Delta River System (LMDRS); (proposed November 2004);
4. Organizing a follow-up for National bankerosion Program for Mekong and Red River.
5. Organizing a Vietnamese Working Group (supplemented with Dutch experts) on Upgrading
Design Standards for bank protection;
6. Proposals for extending of Curriculum at HWRUCE with riverbank erosion and mitigation
7. Inventory of literature on bank erosion related issues (2005)
8. Working Group on Strategy of Bank Erosion Mitigation (2005-2006)
9. Short course on bank erosion mitigation: system approach/methodology, prediction
techniques, planning, design and monitoring (2006);
10. Implementation and evaluation plan (end 2006/2007), including proposals for upgrading
legal documents and regulations.

88
References
AMRC, 2001: Environmental Issues and Recent Infrastructure Development in the Mekong
Delta: review, analysis and recommendations with particular reference to largescale water
control projects and the development of coastal areas; Takehiko ‘Riko’ Hashimoto, Working
Paper No. 4, Australian Mekong Resource Centre, University of Sydney, June 2001.
Ariathurai, R. & K. Arulanandan (1978), Erosion rates of cohesive soils, Journ. Hydr. Div.,
ASCE, Vol.104, No.HY2, pp.279-283
Bristow, C.S. (1987), Brahmaputra River: channel migration and deposition, In: Recent
developments in fluvial sedimentology, Eds. F.G. Ethridge, R.M. Flores & M.D. Harvey, Soc.
Economic Paleontol. and Mineral., Spec. Publ. No.39, pp. 63-74
Coleman, J.M. (1969), Brahmaputra River: channel processes and sedimentation, Sedimentary
Geol., Vol.3, Nos.2-3, pp.129-239
Crosato, A. (1990), Simulation of meandering river processes, Communications on Hydr. and
Geotech. Engrg., No.90-3, Delft Univ. of Technology, ISSN 0169-6548
DHI & SWMC (1996), Mathematical morphological model of Jamuna River; Jamuna Bridge
Site. First forecast report for Government of Bangladesh, World Bank & Jamuna Multipurpose
Bridge Authority, Danish Hydraulics Institute & Surface Water Modelling Centre
Experco (1994), Preliminary study Hanoi dyke rehabilitation sub-project, Irrigation and Flood
Protection Rehabilitation Project, Volume 1
Hickin, E.J. & Nanson, G.C. (1984), Lateral migration rates of river bends, Journ. Hyd. Eng.
ASCE, Vol. 110, no. 11, pp. 1557-1567
Hoffmans, G.J.C.M. and Verheij, H.J. (1997), Scour manual, Balkema, Rotterdam, 205 pp
Jagers, H.R.A. (2003), Modelling planform changes of braided rivers, Twente University, Ph.D.
Thesis, 318 p., fig., tab., ref. + cd-rom , ISBN 90-9016879-6
Jansen, P.Ph., van der Berg, J., de Vries, M. & Zanen, (…) (Ed.), Principles of river engineering.
The non-tidal alluvial river, London, Pitman Publ. Cy. (also 1994 reprint, DUT)
Klaassen, G.J. & Masselink, G. (1992), Planform changes of a braided river with fine sand as
bed and bank material, Proc. 5th Intern. Symp. on River Sedimentation, Karlsruhe, Germany, pp.
…- … (13 pages)
Klaassen, G.J., E. Mosselman & H. Bruehl (1993), On the prediction of planform changes in
braided sand-bed rivers, In: Wang, S.S.Y. (Ed.), Adv. in Hydro-Sci. and -Engrg., Ed.), Publ.
Univ. Mississippi, pp.134-146.

89
Le Manh Hung & Dinh Cong San (2002), Xoi lo bo song cuu long (Bank erosion Mekong
River), Southern Institute for Water Resources
Mosselman, E. (1992), Mathematical modelling of morphological processes in rivers with
erodible cohesive banks, Communications on Hydr. and Geotech. Engrg., No.92-3, Delft Univ.
of Technology, ISSN 0169-6548.
Ngaonh, M.T. and Akira, Y. (2003?), River bank erosion along in the Mekong Delta, Origin
unknown
Nguyen Tuan Anh & Tran Xuan Thai (2000), Some problems on river bed and flood passage of
the Red River, International European-Asian Workshop Ecosystem & Flood 2000, Hanoi,
Vietnam, 7 pages
Olesen, K.W. & Tjerry, S. (2003), Morphological modelling of the Chaktomuk junction,
Pilarczyk, K.W. (1998), Dikes and revetments, A.A. Balkema, Rotterdam
Przedwojski, B., Blazejewski, R and Pilarczyk, K.W. (1995), River training techniques:
fundamentals, design and applications, A.A. Balkema, Rotterdam/Brookfield
Sarker, M.H. & Khayer, Y. (2002), Developing and updating empirical methods for predicting
morphological changes in the Jamuna River, EGIS, Dhaka, EGIS Technical Note Series 29
Shishikura, T. (1996), Morphological changes due to river bank protection, IHE Delft/Delft
Hydraulics, M.Sc. thesis no. 285
Stoutjesdijk, T.P., De Groot, M.B. and Lindenberg, J (1994), Engineering Approach to Coastal
Flow Slides, Proceedings Int. Conf. Coastal Eng., 1994 (ICCE'94),Kobe, Japan. New York,
ASCE, pp 3350-3359
Stoutjesdijk, T.P., De Groot, M.B. and Lindenberg, J.(1998), Flow slide prediction method:
influence of slope geometry, Canadian Geotechnical Journal 35, pp. 43-54
US Army (1981), Streambank Erosion Control Evaluation and Demonstration (Main Report),
US Army Corps of Engineers, Final Report to Congress, 1981

90
Annex
ENVIRONMENTAL IMPACT ASSESSMENT REPORTS FOR PROJECTS
AFFECTING THE RIVER AND COASTAL ENVIRONMENT
In situations where an EIA is deemed necessary, the following checklist may be used as
guidance for the EIA work and the form/content of the EIA Report. This checklist is based on a
format devised by FINNIDA in their draft “Guidelines for Environmental Impact Assessment in
Development Assistance” (1989).
I. Existing environment and site selection considerations
• Assess whether there are possible alternative sites or locations which could be considered in
project siting.
• For each alternative site, whether there are, within or nearby, assess natural conditions and
man-made activities.
• Whether the project location might cause conflicts with the abovementioned land or resource
uses, interests, values or communities.
• Whether the project locations are affected by major natural hazards (e.g. floods, hurricane,
volcanic activity).
• The extent of existing development in the area and whether there are already significant
environmental problems (e.g. water quality, bank/coastal erosion, habitat damage, over-fishing)
in the project vicinity.
• Any relevant human health/disease concerns in the project vicinity.
• Whether the surrounding area can provide adequate supporting facilities.
• Whether there are relevant environmental policy (including EIA) guidelines.
• Whether there are any relevant planning/land use policy considerations (including coastal zone
management plan, economic development zones).
• Whether there is relevant international legislation (international waterways, etc.).
II Site preparation and construction
• Identify relevant site preparation and construction activities and components.
• Identify and predict impacts on natural and socio-economic conditions.
III. Project operation
• Identify alternative design, manufacturing processes, raw materials, fuels, etc., which could be
considered.
• Identify for each alternative relevant activities/components.
• Identify the extent to which discharges (particularly to marine/estuarine environments) may
cross regional or national boundaries.
• Identify and predict impacts on natural and socio-economic environment.
IV. Mitigation and monitoring measures
• Plan adequate mitigation of harmful impacts.

91
Appendices
Appendix I: Mission participants
Mission participants from the Netherlands
1. Krystian W. Pilarczyk: Bank and dike protection expert, DWW
2. Maarten van der Wal: River engineering expert, DWW
3. Ruud (Marinus Cornelis Johannes) Bosters: Hydraulic engineering and hydrology expert,
DWW
4. Jaap (Jacob) Lindenberg: Geotechnical expert, GeoDelft
5. Gerrit J. Klaassen: River engineering and river morphology expert, UNESCO-IHE
Mission participants from Vietnam
6. Nguyen Huu Phuc: Chief of Master planning division of DDMFC
7. Nguyen Huy Dzung: Hydraulic engineering expert, DDMFC
8. Nguyen Si Nuoi: Vice Director of DDMFC (only Red River mission)
9. Do Ngoc Thien: Vice Director of DDMFC (only Mekong mission)
10. Dang Quang Thanh: Hydraulic engineer, DDMFC (only Mekong mission)
Adresses
1. DWW: Ministry of Transport, Public Works and Water Management; Directorate-General of
Road Public Works and Water Management; Road and Hydraulic Engineering Institute: Van
der Burghweg 1; P.O. Box 5044; NL-2600 GA Delft. Tel. ++.31.15.2518437, Fax
++.31.15.2518555
2. GeoDelft: Stieltjesweg 2; P.O. Box 69; NL-2600 AB Delft
3. UNESCO-IHE: Westvest 7; P.O. Box 3015; NL-2601 DA Delft. Tel. ++.31.15.2151715,
Fax ++.31.15.2122921
4. DDMFC: Ministry of Agriculture and Rural Development (MARD); Department of Dike
Management and Flood Control: A4 Building; 2 Ngoc Ha Street; Ba Dinh; Hanoi; Vietnam.
Tel. (84-4)7335690, Fax (84-4)7335701
Abbreviations
DWW: Road and Hydraulic Engineering Institute
UNESCO-IHE: Institute of Water Education
MARD: Ministry of Agriculture and Rural Development
DDMFC: Department of Dike Management and Flood Control

92
Appendix II: Contacts in Vietnam
Hanoi Water Resources University, Second Base, Ho Chi Minh City
1. Dr. Duong Van Vien: Vice Director
2. Nguyen Van Dien: Expert
3. Nguyen Van Thiet: Expert
4. Miss Vu Hoang Anh: Teacher
5. Miss Mai: Teacher
Sub Institute of Geography, Ho Chi Minh City
6. Prof. Nguyen Sinh Huy: Expert on hydraulics
7. Miss Nguyen Thi Hanh: Secretary of Prof. Nguyen Sinh Huy
Sub-Institute for Water Resources Planning, Ho Chi Minh City
8. To Van Truong: Director
9. Nguyen Ngoc Anh: Deputy Director
10. Nguyen Xuan Hien: Deputy Director of SIWRP
11. Tran Minh Khoi: Vice Head of Center for Water Quality
12. John B. Cantor: Consultant for AusAID
13. Brian Cummings: Consultant for AusAID
Southern Institute of Water Resources Research (SIWRR), Ho Chi Minh City
14. Dr. Le Manh Hung: Director of Center for River Training and natural Disaster Prevention
15. Ass.prof. Dr. Hoang Van Huan: Deputy Director of SIWRR
16. Dr. Tang Duc Thang: Deputy Director of SIWRR
17. Nguyen The Bien: Deputy Chief of Department
18. Dinh Cong San: Deputy Director of Center for River Training & Natural Disaster Prevention
Service of Agriculture and Rural Development of An Giang Province (SARD), Long
Xuyen
19. Do Van Hung: Vice Deputy
20. Pham Van Le: Director of AN Giang Provincial Department of Water Resources
21. Vuong Huu Tien
Engineering Company for SARD of An Giang Province, Long Xuyen
22. Mai Anh Vu: Expert of Consultant Company for Rural Development
23. Tran Huu Ha: Expert of Consultant Company for Rural Development
People's Committee of Tan Chau
24. Do Thanh Trung: Vice chairman of People Committee of Chau District
Service of Agriculture and Rural Development of Dong Thap, Cao Lanh
25. Huynh The Phien: Vice Director
26. Nguyen Chap Kinh; Deputy Director of Dong Thap Provincial Department of Water
Resources

93
Service of Agriculture and Rural Development of Vinh Long, Vinh Long Town
27. Phan Nhat Ai: Director of SARD
28. Ha Thanh Thang: Head of technical department
Sub-Department of Dike Management and Flood Control of Vinh Long Province
29. Nguyen Van Thanh: Director
Engineering Company for SARD of Vinh Long Province
30. Nguyen Trong Minh: Deputy Director of Design Consultant Company Of Vinh Long
Province
31. Tran Quoc Hoai: Designer
Service of Agriculture and Rural Development of Ha Tay
32. Luu Van Hai: Vice Director
Sub-Department of Dike Management and Flood Control of Ha Tay Province
33. Vu Xuan Phieu: Deputy Director of Sub-Department
34. Nguyen Tuan Khai
Service of Agriculture and Rural Development of Hung Yen
35. Bui Xuan Bai: Vice Director
Sub-Department of Dike Management and Flood Control of Hung Yen Province
36. Doan Quang Viet: Director
37. Ho Trong Khai: Deputy Director
38. Hoang Xuan Vang: Deputy Director of DDMFC and Head of Hydraulic Construction
Project Management
39. Tran Van Bui: Hydraulic Contruction Project Management
40. Vu Van Hanh: Director of Consultant Company
Vietnam Institute for Water Resources Research
41. Ass.Prof.Dr. Tran Dinh Hoi: Deputy Director
42. Ass.Prof.Dr. Tran Xuan Thai: Director of River Engineering Research Centre
43. Nguyen Ngoc Quynh: Vice Director of River Engineering Research Centre
44. Miss Dao Thu Trang: International Cooperation Section
45. Ho Viet Cuong: River Engineering Research Centre
46. Nguyen Dang Giap
Hanoi Water Resources University, Hanoi
47. Prof. Dr. Le Kim Truyen: Rector of HRWU
48. Ass.Prof. Dr. Vu Minh Cat: Head of Division of Science and International Co-operation
49. Tran Thanh Tung: Lecturer on river and coastal engineering
50. Ass.Prof. Dr. Le Dinh Thanh: Head of Coastal Engineering Faculty
51. Dr. Le Xuan Roanh: Secrectary of Coastal Engineering Faculty
52. Ass. Prof. Dr. Vu Thanh Te: Hydraulic Engineering Faculty
53. Ass. Prof. Dr. Do Tat Tuc: Head of Hydrology and Environmental Faculty

94
Adresses
1. Hanoi Water Resources University (HWRU), Second Base; 2 Truong Sa; Binh Thanh
District; Ho Chi Minh City. Tel. (08)8404743, Fax (08)8400542
2. Sub Institute of Geography: 1 Mac Dinh Chi Street; District 1; Ho Chi Minh City. Tel.
(08)8234347
3. Sub-Institute for Water Resources Planning (SIWRP): 253A, An Duong Vuong; District 5;
Ho Chi Minh City. Tel. (84-8)8350850, Fax (84-8)8351721
4. Southern Institute of Water Resources Research (SIWRR): 2A Nguyen Bieu St.; District 5;
Ho Chi Minh City. Tel. (08)8380990 (Dinh Cong San), (08)8362821 (Nguyen The Bien),
Fax (08)9235028
5. Hanoi Water Resources University (HWRU): 175 Tay Son Street; Dong Da District; Hanoi;
Vietnam
6. Service of Agriculture and Rural Development (SARD) of An Giang Province: 4 Nguyen
Du Street; My Binh Ward - Long Xuyen City. Tel. (076)853257, Fax (076)856705
7. Engineering Company for SARD of An Giang Province: 2-3 Le Hong Phong; Long Xuyen
city; An Giang Province.
8. People's Committee of Tan Chau: Van Phong HDND - UBND Huyen Tan Chau, An Giang:
Duong 1/5 Thi Tran Tan Chau. Tel. (076)822201
9. Service of Agriculture and Rural Development of Dong Thap Province: 154 Quoc Lo 30; Xa
My Tan - TX. Cao Lanh. Tel. (067)852532, Fax (067)853514
10. Department of Water Resources Dong Thap Province: 154 Quoc Lo 30; Xa My Tan - TX.
Cao Lanh. Tel. (067)851092, Fax (067)853028
11. SARD of Vinh Long: 107/2. Pham Hung; P. 9; TX. Vinh Long. Tel. (070)830981, Fax
(070)827635
12. Engineering Company for SARD of Vinh Long Province: 107/2. Pham Hung; P. 9; TX.
Vinh Long. Tel. (070)830981, Fax (070)827635
13. Sub-Department of Dike Management and Flood Control of Vinh Long Province: 107/2.
Pham Hung; P. 9; TX. Vinh Long. Tel. (070)830981, Fax (070)827635
14. SARD of Ha Tay:
15. Sub-DDMFC of Ha Tay Province:
16. SARD of Hung Yen:
17. Sub-DDMFC of Hung Yen Province:
18. Vietnam Institute for Water Resources Research: 171 Tay Son; Dong Da; Hanoi. Tel. (84-
4)8523766, Fax (84-4)5634809
19. River Engineering Research Centre: 299 Tay Son; Dong Da; Hanoi. Tel. (84-4)8536524,
Fax (84-4)5634478
20. Hanoi Water Resources University: 175, Tay Son Street; Dong Da; Hanoi. Tel.
(84-4)8533083, Fax (84-4)8534198

95
Appendix III: Mission program and schedule
Schedule for mission visit to Mekong Delta
Date Meeting/location Program
24 HWRU, 8:30-9:00 Introduction
Nov. Second Base, 9:00-10:00 Presentation on Mekong River bank erosion
2003 Ho Chi Minh City 10:00-11:30 Questions and discussion
12:00-13:30 Lunch
SIWRP, 13:30-14:00 Introduction
Ho Chi Minh City 14:00-15:30 Presentation on Mekong River bank erosion
15:30-16:30 Questions and discussion
Overnight in Ho Chi Minh City
25 SWRRI, 8:30-9:00 Introduction
Nov. Ho Chi Minh City 9:00-10:00 Presentation on Mekong River bank erosion
2003 10:00-11:30 Questions and discussion
12:00-13:30 Lunch
Thanh Da Island 13:30-16:30 Bank erosion field visit to Sai Gon River
Overnight in Ho Chi Minh City
26 (Travelling) 7:00-11:30 By car to An Giang province (260 km)
Nov. 11:30-12:00 Check-in at hotel in Long Xuyen
2003 12:00-13:30 Lunch
SARD of An Giang, 13:30-14:30 Introduction and short briefing
Long Xuyen 14:30-16:30 Questions and discussion
Overnight in Long Xuyen
27 Tan Chau 8:30-11:30 Bank erosion field visits by boat
Nov. 12:00-13:00 Lunch
2003 Long Xuyen 13:00-18:30 Visits to revetment and bank erosion sites
Overnight in Long Xuyen
28 (Travelling) 8:00-11:00 By car to Cao Lanh in Dong Thap province (50 km)
Nov. 11:00-11:30 Check-in at hotel
2003 12:00-13:30 Lunch
SARD of Dong 14:00-15:30 Introduction and short briefing
Thap, Cao Lanh 15:30-16:30 Questions and discussion
Overnight in Cao Lanh
29 Sa Dec 8:30-10:30 Visit to Sa Dec revetment system (50 km)
Nov. (Travelling) 10:30-11:30 By car to Vinh Long province (30 km)
2003 11:30-12:00 Check-in at Cuu Long Hotel in Vinh Long town
12:00-13:30 Lunch
SARD of Vinh Long, 13:30-14:00 Introduction and short briefing
Vinh Long town 14:00-15:30 Questions and discussion
Vinh Long town 15:30-17:00 Visit to revetment system and bank erosion sites
Overnight in Vinh Long town
30 (Travelling) 8:30-11:30 By car to Ho Chi Minh city (140 km)
Nov. 11:30-12:00 Hotel registration
2003 12:00-13:30 Lunch
Afternoon Free time
Overnight in Ho Chi Minh City
1 SIWRP, 9:00-12:00 Evaluation of meetings and field visits
Dec. Ho Chi Minh City
2003 12:00-13:30 Lunch
(Travelling) 17:00-19:00 By plane to Hanoi
Overnight in Hanoi

96
Schedule for mission visit to Red River Delta
Date Meeting/location Program
2 (Travelling) 7:30-8:30 By car from Hanoi to Ha Tay province
Dec. SARD of Ha Tay 8:30-9:30: Meeting with SARD
2003 Ha Tay Province 9:30-14:00 Visits to revetments and bank erosion sites
14:00-15:00 Lunch
Overnight in Hanoi
3 (Travelling) 7:30-9:30 By car to Hung Yen province (60 km)
Dec. SARD of Hung Yen 9:30-10:30 Meeting with SARD
2003 Hung Yen town 10:30-11:30 Field visit to revetment
11:30-13:00 Lunch time
Hung Yen Province 13:30-16:30 Visits to revetments and bank erosion sites
Overnight in Hanoi
4 VIWRR, Hanoi 8:00-11:30 Visit Vietnam VIWRR
Dec. 12:00-13:30 Lunch
2003 HWRU, Hanoi 13:30-16:30 Visit HWRU
Overnight in Hanoi
5 DDMFC, Hanoi 8:30-11:30 Presentation of mission results for DDMFC
Dec. 12.00:13:30 Lunch
2003 Hanoi Afternoon Free time
19.00 Farewell party (DDMFC, ICD, PI, WRRI and HWRU
of MARD; representative of RNE)
Overnight in Hanoi
6 Hanoi Day Free time
Dec. 20:00 Departure to Amsterdam
End of mission
Abbreviations
HWRU: Hanoi Water Resources University
SIWRP: Sub-Institute for Water Resources Planning
SIWRR: Southern Institute of Water Resources Research
SARD: Service of Agriculture and Rural Development
DDMFC: Department of Dike Management and Flood Control
VIWRR: Vietnam Institute for Water Resources Research

97
Appendix IV: Field visits Mekong River
Field visit 1: Thanh Da Island, Sai Gon River, Ho Chi Minh City, 25
November, 14:00
Participant from Southern Institute of Water Resources Research
1. Nguyen The Bien
Report
The visit consists of a boat trip around Thanh Da Island. Thanh Da is situated in a meander of
the Sai Gon river, northeast of the center of Ho Chi Minh City, and became an island after the
digging of the Kenh Thanh Da channel (in the 1960s), which short-cuts the meander. Presently,
the meander (depth 20 m) takes 84% of the discharge of the Saigon, while the channel takes
16% (depth 10 m). The tidal difference is about 2,5 m.
The right bank and the channel banks are mostly occupied by slums. On the left bank quite a
few expensive villas are under construction or have been recently constructed. Further upstream
on the left bank of the meander, there are a number of shipping docks. Approximately 200 small
ships per day come through the meander, about 100 small ships go through the much narrower
channel. Cargo vessels have a speed limit of 20 km/h. Tourist vessels have a limit of 40 km/h.
The natural vegetation seems to be a trunkless palm species (cay dua), with leaves up to 5 m.
Where the palms grow there seems to be no bank erosion. Where the palms have been cleared,
bank erosion occurs. All bank protection has been constructed by the inhabitants of the banks
and depending on their means, may vary from sandbags to stone walls, vertical or inclined.
Locally, also private land reclamation has been practised.
Field visit 2: Tan Chau, An Giang Province, 27 November, 10:00
Participant from People's Committee of Tan Chau
1. Do Thanh Trung
Report
After a short meeting with the People's Committee of Tan Chau, a boat trip was made up the
river, to the North until the Cambodian border. Bank erosion is observed at Tan Chau, where the
ruins of a French colonial building are now being swallowed by the river. At Vinh Hoa the West
bank shows erosion and circular slides over a length of several kilometres.
On the way back from the border, a side channel west of the main channel is entered. It is
commented that this channel is getting wider all the time.
Driving back to Long Xuyen, a stop is made at the Song Vam Nao, a connecting river which
takes more and more of the discharge from the Tien river to the Hau river. High flow speeds
occur and are accompanied by erosion on the southeast bank.
Field visit 3: Long Xuyen, An Giang Province, 27 November, 16:30
Report
Returning from Tan Chau, the revetment system of Long Xuyen is visited. It consists of
interlocked concrete blocks on the waterline and gabions below the waterline. It seems in a good
condition.

98
After this, a boat trip is made on the Hau river, where the island of Cu Lao Ong Ho is sailed
around. Bank erosion is observed on both the western and northern side of this island. The
smaller and very narrow island which lies in between Cu Lao Ong Ho island and Long Xuyen
has suffered from erosion at the north side, where it has shortened a few 100 m.
Field visit 4: Sa Dec, Dong Thap Province, 28 November, 9:30
Report
A construction site in Sa Dec is visited. On the site a dam was built to close a side channel.
Presently on the river bank a new protection structure is built, consisting of sand bags, gabions
and a concrete revetment. The sand bags are used to grade the underwater slope of the river
bank. After the desired slope has been established, gabion matresses are let down from a barge.
The matresses have a dimension of 2x10x0,5 m and are put in place making use of a GPS. Free-
divers check the correct placement and make underwater pictures. The divers can go as deep as
30 m. At Sa Dec the construction depth is about 25 m.
The concrete revetment consists of interlocked slabs of approximately 0,15x0,8x0,8 m, which
are made on site.
Field visit 5: Vinh Long Town, 29 November, 15:30
Report
With a speedboat the Tien river is explored upstream. Some bank erosion is visible and
furthermore some concrete permeable groins are being observed. After a recreational island is
visited. Also here there is some bank erosion.

99
Appendix V: Meetings Mekong River
Meeting 1: HWRU, Second Base, Ho Chi Minh City, 24 November 8:30
Participants from HWRU, Second Base
1. Dr. Duong Van Vien: Vice Director
2. Nguyen Van Dien: Expert
3. Nguyen Van Thiet: Expert
4. Miss Vu Hoang Anh: Teacher
5. Miss Mai: Teacher
Participants from from Sub Institute of Geography
6. Prof. Nguyen Sinh Huy: Expert on hydraulics
7. Miss Nguyen Thi Hanh: Secretary of Prof. Nguyen Sinh Huy
Meeting report
Mr. Duong Van Vien gives a powerpoint presentation about the Mekong Delta, concerning flow
regimes, bank erosion and geology. In the presentation, a number of specific technical data are
given. Miss Nguyen Thi Hanh gives some additional geologic explanations.
HWRU will provide a copy of the presentation.
Meeting 2: Sub-Institute for Water Resources Planning, Ho Chi Minh City,
24 November 13:30
Participants from Sub-Institute for Water Resources Planning
1. To van Truong: Director
2. Nguyen Ngoc Anh: Deputy Director
3. Tran Minh Khoi: Vice Head of Center for Water Quality
4. Nguyen Xuan Hien: Deputy Director of SIWRP
5. John B. Cantor: Consultant for Vietnam - Australia Water Resources Management
Assistance Project
6. Nguyen Dinh Thanh
7. Pham Anh Tuan
8. Dang Thanh Lam
9. Ho Trong Tien
10. Do Duc Dung
11. Luu Van Thuan
Meeting report
Mr. To van Truong opens the meeting with an introduction explaining about the planning
institute. Mr. Nguyen Xuan Hien holds a presentation about flood control planning in the
Mekong Delta. A print of the presentation is handed out.
Mr. Tran Minh Khoi holds a presentation about the water quality monitoring project.
Mr. Pilarczyk mentions that the Dutch embassy is interested in the Mekong water managament
programs. The mission will concentrate on bank erosion and is wondering to what extent this is
an issue for the Planning Institute. Mr. To van Truong indicates that it is hard to gather sufficient
data and that planning is concentrated on sites which are economically important or where ther
live a lot of people.

100
Mr. Cantor indicates that bank erosion is an important issue also for the Mekong River
Comission. A meeting is proposed on Monday morning 1 December at 9:00 at SIWRP.
Meeting 3: Southern Institute of Water Resources Research, Ho Chi Minh
City, 25 November, 8:30
Participants from Southern Institute of Water Resources Research
1. Dr. Le Manh Hung: Director
2. Dr. Hoang Van Huan: Deputy Director
3. Dr. Tang Duc Thang: Deputy Director
4. Nguyen The Bien: Deputy Chief of Department
5. Dinh Cong San: Deputy Director of Center for River Training & Natural Disaster Prevention
Meeting report
Mr. Hoang Van Huan opens the meeting and introduces SIWRR. Of the people present, Mr.
Dinh Cong San can speak English. Mr. Nguyen The Bien will come on the afternoon excursion.
Mr. Le Manh Hung holds a presentation about the research about bank erosion in the Mekong
Delta. Mr. Dinh Cong San gives some additional explanation on the whiteboard. The following
questions are formulated to the mission:
1. Prediction of bank erosion?
2. Critical velocity?
3. Dominant discharge/tidal movement?
4. Survey sediment transport?
5. Construction measures for bank protection?
After the break, Mr. Pilarczyk proposes that one or two people from SIWRR join the final
meeting on December 1 at the Sub-Institute for Water Resources Planning. SIWRR thinks this is
a good idea.
Mr. Hoang Van Huan holds a presentation on structures to counteract bank erosion.
Mr. Lindenberg asks whether there is generally a deeper, non-cohesive, fine sand layer which is
sensible to liquefaction (and which would induce collapse of the upper layer). What is the
process of erosion? Is the erosion fast or slow?
Mr. Le Manh Hung says that the erosion in the upper reaches, where there is no tidal influence is
much more gradual and less sudden than in the lower reaches. In a few occasions, the bank has
suddenly collapsed over a width of 50 m.
Mr. van der Wal asks whether there are some design standards or guidelines. Furthermore, to
what extent is there undermining of the existent protection?
Mr. Hoang Van Huan answers that generally people drop sandbags right in front of the bank.
There are some standards but not really guidelines. The standards take undermining into account
extending the construction 5 to 10 m beyond the toe.
Mr. Pilarczyk asks whether most erosion takes place where the river is deepest? What is the
influence of sand-mining?
Mr. Hoang Van Huan answers that this is correct. The problem is aggravated because sand
mining takes place where the river is deepest because the sand is coarser.
Mr. Pilarczyk asks why Vietnam resists to the application of Vetiver grass. Mr. Le Manh Hung
answers that Vietnam also applies natural solution, e.g. coconut trees are apllied in the less
flooded zones where the groundwater level doesn't vary that much. However, the Vetiver doesn't
survive in the places which are flooded for continued periods, nor in the tidal area with saline

101
groundwater. Still, he would like to receive more information, for maybe the possibilities are
underestimated.
Mr. Klaassen asks whether there is any legislation as to how far away from the river you should
live. The answer is that in recent years there is some regulation, but it varies a lot and is very
flexible. As to accreted areas, they may be used temporarily, but should not be used finally.
When was the Vam Nao connection canal dug? Has the Tonle Sap recently developed and may
its coming about be at the origin of bank erosion? Nobody knows when the channel was dug.
The water always flows from Tien to Hau.
Meeting 4: SARD of An Giang, Long Xuyen, 26 November 14:00
Participants from SARD of An Giang
2. Do Van Hung: Vice deputy
3. Pham Van Le: Vice Head of Department of Water Resources
4. Vuong Huu Tien
Participants from Engineering Company for SARD of An Giang Province
5. Tran Huu Ha: Expert of Consultant Company for Rural Development
6. Mai Anh Vu: Expert of Consultant Company for Rural Development
Meeting report
Mr. Do Van Hung presides the meeting and does a presentation about An Giang province,
flooding and bank erosion. A paper in Vietnamese is handed out, which contains the data about
flooding and bank erosion. It mentions that since 1994 bank erosion has increased, 3.000 houses
were relocated, etc. Dzung will later on make a written translation of the highlights of the data in
the paper.
Mr. Pilarczyk asks whether some event occured around 1994 which might be linked to the
increased bank erosion.
Mr. Do Van Hung says that since 1997 big floods seem to be an annual event, where before they
only occured every 2 or 3 years. Due to (low) dike construction the floods are higher than
before. Besides, the population increased a lot, which makes that the problem attracts more
attention,esp. where people started living on the banks. The water level difference between dry
and wet season is about 5 m.
In 2002 some 600 m of revetment were constructed in the Tan Chau area. Cost: 102·109 VND.
Heigth of revetment: 15 m, width 50 m, therefore the slope is about 1:3. In Long Xuyen City
some 700 m of revetment were constructed. For the construction gabions and bamboo is used.
Mr. Mai Anh Vu tells that they use a model (MIKE) for the calculation of flow velocities. This
is used for the design of bank potections, according to standard 14TCN-84-91.
Meeting 5: SARD of Dong Thap, Cao Lanh, 28 November 14:00
Participant from SARD of Dong Thap
1. Huynh The Phien: Vice Director
Participant of Department of Water Resources Dong Thap Province (= sub-Department of Dike
Management and Flood Control)
2. Nguyen Chap Kinh: Deputy Chief
Meeting report

102
Mr. Huynh The Phien opens the meeting. Kinh comments about bank erosion in Dong Thap
Province, as explained in a Vietnamese paper, which is handed out at the beginning of the
meeting. A brief bank erosion history is given about the problem locations Hong Ngu, Can Tho
and Sa Dec. Sa Dec is the most affected place. Presently 900 m of bank protection exists and
there is the desire to extend this. According to Mr. Huynh The Phien sand mining is the main
cause for bank erosion. Mr. Pilarczyk observes that sand mining doesn't need to be a problem as
long as it is done at the right place. Mr. Huynh The Phien says that sand mining is regulated, but
hardly controlled.
Meeting 6: SARD of Vinh Long, Vinh Long Town, 29 November 13:30
Participants from SARD of Dong Thap
1. Phan Nhat Ai: Director of SARD
2. Ha Thanh Thang: Head of technical department
Participant of sub-Department of Dike Management and Flood Control of Vinh Long Province
3. Nguyen Van Thanh: Director
Participants from Engineering Company for SARD of Vinh Long Province
4. Minh: Consultant
5. Hoai: Designer
Meeting report
Mr. Phan Nhat Ai opens the meeting. Mr. Nguyen Van Thanh gives a presentation about Vinh
Long Province and the occuring bank erosion. A Vietnamese paper with bank erosion is handed
out during the meeting.
Meeting 7: Sub-Institute for Water Resources Planning, Ho Chi Minh City, 1
December 2003, 9:00
Participants from SIWRP:
1. John Cantor: Vietnam - Australia Water Resources Management Assistance Project
2. Brian Cummings: Vietnam - Australia Water Resources Management Assistance Project
3. Dr. Le Manh Hung
Participant fromSIWRR:
4. Dinh Cong San
Meeting report
Mr. Cantor tells about the work of AUSaid for the River Basins Organization, dealing with
water management in the Mekong Delta. A list of important water resources issues has been
drawn up by the steering committee. Mr. Cantor mentions that there are a number of
institutional reasons for which the River Basins Organization has not yet so much influence on
the water resources management, meaning the policies descending from Hanoi.

103
Appendix VI: Field visits Red River
Field visit 1: Ha Tay Province, 2 December, 9:30
Report
Driving East, the dividing dikes of the Day River were passed. The Day river plain serves as an
outlet for floods on the Red River, which are then diverted through the Ha Binh reservoir, which
is located at the entrance of the Day River.
A first stop at the Red River is made near Chu Minh. Here the bank is high and near vertical,
being completely dried out, despite the recent end of the wet season. Bank erosion has occured
'as far as the eye can see'. In front is the Minh Chau sand bar, which forces the flow to the
eroded (South) bank.
After, the bank protection at Phu Cuong is visited. It consists of loose stones in a concrete
framework at the top of the bank, and of gabions with smaller stones on the toe of the bank. At
one point the toe is being eroded and needs to be refixed.
Next, a stop is made at Phong Van, where there is a severe bank erosion and collapsing, and
where numerous cracks testify a large amount of slides.
Near Tong Lenh some short groins from loose stone are observed, which have been constructed
by the locals a few years ago. Presently, this bank is no longer threatened and the toe of the
groins is not even submerged (the scour hole around it is clearly visible), the main river channel
being shifted to the other side of the river. It seems that the river channel shifts to the other side
every ten years or so.
Field visit 2: Hung Yen Province, 3 December, 10:30
Report
In the morning, a revetment is visited at the Red River near Hung Yen town. The revetment is
located just North of a large bridge under construction. The revetment consists of pitched riprap
in a concrete framework. The toe is protected with gabions which form a staircase. As a
transition, riprap has been deposited on the 'steps'. The gabions are in bad shape, being corroded
and damaged, although the structure is only a few years old.
South of the bridge, the bank is eroding, showing cracks and slides.
In the afternoon, the site of Thanh Cong is visited. Here, in 1982 groin construction was started
to counteract bank erosion at the toe of the dike. The groins are approx. 100 m length and are
covered in pitched riprap. The last 20 m are covered in gabions, which are in bad state. When
after a few years the groins didn't seem to be effective, a retired ambankment was constructed in
1985. Eventually, the dike didn't collapse and groin construction was continued. Presently there
are 6 groins, the last being constructed in 2002.
Another 5 groins are visited at Tu Dan. They are of the same construction. Construction was
started because even at a 1:3,5-slope the bank was not stable and continued to slide. After groin
construction the sliding stopped.
Finally, a visit is paid to the site of Phi Liet, where a revetment will be constructed. On the
opposite site of the river, illegal sand mining is taking place.

104
Appendix VII: Meetings Red River
Meeting 1: SARD of Ha Tay, 2 December, 8:00
Participants from SARD of Ha Tay
8. Luu Van Hai: Vice Director
Participants from Sub-Department of Dike Management and Flood Control of Ha Tay Province
9. Vu Xuan Phieu
10. Nguyen Tuan Khai
Meeting report
Mr. Nguyen Si Nuoi opens the meeting, introducing the participants. Mr. Luu Van Hai explains
about Ha Tay Province and its geographic features. Along the Da River (a tributary of the Red
River) bank erosion occurs on several locations over lenghts of several hundreds of m.
The land loss was 40 ha since 1994. A number of other locations and rivers with bank erosion
are summed up.
Among the causes of bank erosion is the operating of Ha Binh reservoir, due to regulating the
flow in the flood season and trappinf of sediment in the reservoir. Important aspects for bank
erosion are the formation of sand bars in the middle of the channel and the fact that the banks are
sandy. In opposition to the Mekong, most bank erosion in the Red River occurs in the dry
season.
2.000 People were moved, another 2.000 still have to be moved.
Groin construction is expensive, dry season water depth ranging between 10 and 20 m.
Floods may occur in the Da, the Lo and the Red River on different moments.
Monitoring of discharge, water level and flow velocity occurs hourly in one station. Over the
whole Red River, 168 cross-sections have been surveyed since 1992.
There are a lot of data, but it is not clear whether they are systematically organised nor whether
anything is being done with them.
Bank protection is not carried out with anticipation but only after damage has occured.
Anually, approx. 55 109 VND is spent on bank protection in the whole Red River Delta (19
provinces), comprising both maintenance and new structures. Whether the money goes to
maintenance or new structures varies greatly and depends on the severity of the floods.
Meeting 2: SARD of Hung Yen, 3 December, 9:30
Participants from SARD of Hung Yen
1. Mr. Bui Xuan Bai: Vice Director
Participants from Sub-Department of Dike Management and Flood Control of Hung Yen
Province
2. Mr. Doan Quang Viet: Director
3. Mr. Ho Trong Khai: Deputy Director
4. Mr. Hoang Xuan Vang: Deputy Director of DDMFC and Head of Hydraulic Construction
Project Management
5. Mr. Tran Van Bui: Hydraulic Contruction Project Management
6. Mr. Vu Van Hanh: Director of Consultant Company
Meeting report

105
Mr. Bui Xuan Bai explains about Hung Yen Province and its geographic features. In the flood
season, there are a lot of places where piping occurs. Furthermore, in many places bank erosion
occurs and is eroding the toe of the dikes.
The Province is surrounded by a ring dike. Breaches occured frequently in the Northwest of the
province in the 19th century. The last dike breach was in 1945.
Meeting 3: Vietnam Institute for Water Resources Research, 4 December,
8:00
Participants from VIWRR
1. Mr. Tran Dinh Hoi: Deputy Director
2. Mr. Tran Xuan Thai: Director of River Engineering Research Centre
3. Mr. Nguyen Ngoc Quynh: Vice Director of River Engineering Research Centre
4. Miss Dao Thu Trang: International Cooperation Section
5. Mr. Ho Viet Cuong: River Engineering Research Centre
6. Mr. Nguyen Dang Giap
Meeting report
Mr. Tran Dinh Hoi formally opens the meeting and tells about the history and activities of the
institute. Mr. Tran Xuan Thai holds a presentation about bank erosion in the Red River. Within
VIWRR, the River Engineering Research Centre is in charge of bank erosion research. A video
shows the consequences of the flood season in 2002.
Modelling attempts are being made, for example with MIKE-21, but a lot of problems were
encountered. In the Red River delta, bank erosion does not only affect those who live on the
banks, but also the dike system, and therefore much more people. For this reason, bank erosion
in the rest of Vietnam is not so important.
Apart from the natural causes (high flow velocity), an important cause of bank erosion is sand
mining. There is more bank erosion in areas with tidal influence. Bank erosion has a tendency to
continue in downstream direction rather than in upstream direction. In some cases bank erosion
is continuous, in others it only occurs during some years. Bank erosion is strongest at the end of
the flood season (end of september), when the water level starts going down. If nothing is done,
the river bed will shift and bank erosion will continue downstream.
The most serious bank erosion sites are enumerated and show how quickly the Red River may
change its course.
For bank erosion prediction the Russian formulas of Ibadzade and Popov were used. Bank
erosion is combatted with revetments or groins. The revetments have improved a lot. The groins
were constructed esp. in the 1960s. Mostly they are impermeable, recently also permeable groins
are built to divert the river flow in service of navigation. The groin consists of a soil core
covered by stone. At the top a drainage system is provided for. The concrete revetment blocks
have a thickness of 0,12 m. At the toe of the groins stones with a diameter of approx. 0,8 m
('dragon stones') are deposited to prevent scourholes.
In Vietnam presently only short-term solutions are applied, but there is a great necessity for
long-term solutions.
A digital copy of the presentation will be provided.
Discussion
Mr. van der Wal asks how VIWRR managed to improve the revetments.

106
Answer: Before everything was done by the local people in a purely empirical way. Besides
there is more money now. Before, 3 109 VND per km was invested, now 25 109 VND.
A groin of 50 m may cost 2 109 VND.
The 'dragon stone' is essentially a large, cylindric gabion. Design is purely empirical. The
dragon stone has a length of 10 m. Mr. van der Wal wonders whether this is not too short, as it
can cover only a scour hole of about 4 m depth.
It is stated that there is little money for satellite images to improve prediction. Mr. Klaassen
indicates that also protection structures are very expensive, so it is not a matter of money but a
matter of approach.
Mike 21C is used for flow velocity, morphology and bank erosion. The data for the model and
the calibration come from a measuring station. Presently the government is investing in more
accurate data collection.
Meeting 4: Hanoi Water Resources University, 4 December, 13:30
Participants from HWRU
1. Prof. Dr. Le Kim Truyen: Rector of HRWU
2. Dr. Vu Minh Cat: Head of Division of Science and International Co-operation
3. Tran Thanh Tung: Lecturer on river and coastal engineering
4. Prof. Dr. Le Dinh Thanh: Head of Coastal Engineering Faculty
5. Prof. Dr. Le Xuan Roanh: Secrectary of Coastal Engineering Faculty
6. Ass. Prof. Dr. Vu Thanh Te: Hydraulic Engineering Faculty
7. Ass. Prof. Dr. Do Tat Tuc: Head of Hydrology and Environmental Faculty
Meeting report
Dr. Le Kim Truyen opens the meeting and tells about the structure and activities of the
University. Dr. Do Tat Tuc holds a presentation about bank erosion studies in central Vietnam,
concerning the Thu Bon river. 13 Sections with severe erosion were defined. The Thu Bon river
is a distributary of the Vu Gia river. Originally, ik took about 20% of the discharge of the Thu
Bon river, but due to the shortcut of a meander, almost all the water started flowing into the Thu
Bon river, changing the dynamics and causing bank erosion. It is the desire to stabilize the river
mouth. In the flood season the flow velocity may reach 2 to 3 m/s, leading to surface erosion.
Erosion prediction, or rather meander shifting is predicted in an empirical way and is not based
on theory or models. The University (doing research in central Vietnam) and the two research
institutes (research in respectively Mekong and Red River) apply different ways of prediction.
The HWRU mainly uses remote sensing and survey data. Western formulae and theories badly
apply to the Vietnamese rivers, due to differences in morphology etc. Mr. Pilarczyk pleads for
setting up a good databank with hydraulic and morphologic data.
There is a strong correlation between the hydrograph and bank erosion. Mr. Klaassen argues that
this would be a good base to predict bank erosion.
The University doesn't use erosion formulae (e.g. the Russian ones), because the erosion is also
strongly influenced by factors like sand mining, which the formulae do not take into account.
The available data consist of hydrographs and cross-sections. The maximum flow velocity in the
central rivers is about 3 m/s, the maximum discharge about 20.000 m3/s.
In the old city of Hoi An, vertical stone walls in combination with rock matresses (for bed
protection) are used to stabilize the banks.
Mr. Pilarczyk recommends to do some kind of systematic data research, e.g. by means of M.Sc.
students. Furthermore he stresses that good cooperation with the Research Institutes (esp. the

107
Second Base in Ho Chi Minh City) is of crucial importance. Mr. Cat says the funds for this are
lacking.
Final meeting: Presentation of mission results, DDMFC, 5 December 2003,
8:30
Participants
1. Mr. Nguyen Sy Nuoi: Deputy Director of DDMFC
2. Mr. Do Ngoc Thien: Deputy Director of DDMFC
3. Mr. Nguyen Huu Phuc: Head of Master Planning and Bank, Coastal Erosion Section
4. Ms. Cao Thi Lua: Director of Dyke Engineerin Consultant Center
5. Ms. Nguyen Thi Hien: Vice Director of Dyke Engineerin Consultant Center
6. Mr. Nguyen Tien Toan: Vice-chieft of Dyke Management Section
7. Some staffs of DDMFC
8. Mr. Tran Thanh Tung (HWRU)
9. Mr. Roanh (HWRU)
10. Nico Bakker (Dutch Embassy)
Meeting report
Mr. Sy Nuoi opens the meeting, introducing the participants and indicating that bank erosion is
becoming a serious problem. DDMFC hopes that after the mssion a program may be started to
upgrade the capacity to cope with bank erosion in Vietnam. DDMFC hopes that the Embassy
can take an active part in this.
Mr. Bakker indicates that the Embassy aims to support Vietnam in water management, esp. in
flood control. For the Mekong river, the embassy is cooperating with the Mekong River
Comission. Priority is given on the development of standards and guidelines, as well as
capacity-building. The Embassy prefers to give financial aid to projects for which there is
already basic funding by the ADB. Future financial aid is partly depending upon this principle.
Mr. Phuc holds a presentation about bank erosion, treating backgrounds, contexts, causes,
reasons and need for measures to be taken on various administrative levels. His presentation
concentrates on planning and management.
Mr. Bakker comments that it is important to link DDMFC to the planning institute in the south.
Furthermore, China plans to build another 7 to 9 dams in the Mekong, so the hydrograph of the
river will change. Data collection should be cooperated with the other Mekong countries.
Mr. Phuc agrees that an umbrella structure should be set up and adds that this is also the case for
the Red River, which also originates in China.
The dilemma of Mr. Pilarczyk is that the master planning is very important but in practice even
the money for the most urgent protection is lacking.
Mr. Phuc says there is no fixed budget, the money which is available is disaster orientated. Mr.
Sy Nuoi indicates that on average 2% to 3% of the GDP s spent on disaster mitigation.
Mr. Pilarczyk holds a presentation on the findings of the mission.
After lunch, Mr. Lindenberg gives a presentation on the geotechnical aspects of bank erosion.
He is followed by Mr. Klaassen, who gives a presentation on river morphology. Mr. van der Wal
gives a presentation on bank protection structures. Mr. Kerssens gives a presentation on the
Second Red River Basin Sector Project.

108
Mr. Sy Nuoi indicates that the work of the mission has been very useful to the DDMFC and
thanks the mission for their effort. Mr. Pilarczyk says thanks on behalf of the mission for the
warm reception of the mission and stresses that the most important thing is that DDMFC uses
their own experience. He will do his best to to something iwth the suggestion to start capacity
building on erosion prediction and monitoring, but of course it depends on money from the
embassy as well. Furthermore that a nationwide approach should be followed comprising
Mekong and Red River and other areas which suffer from this problem.

109
Appendix VIII: Damage overview (collected data/not fully representative)
An Giang Province, Mekong Delta
Consequences of bank erosion between 1996 and 2003
1. Land loss: 123 ha
2. Flooded houses: 170
3. Damaged houses: 41
4. Number of families already relocated: 2.932
5. Number of families still to be relocated: 5.896
6. Approximate cost of damage: 18,4·109 VND
Damage per location between 1997 and 2003
Location Land loss Nr. of families moved
Tan Chau 68 ha 835
Cho Moi 23 ha 454
Long Xuyen 21 ha 1.312
Phu Tan 5 ha 180
Other locations 6 ha 151
Total 123 ha 2.932
Damage per year (in VND)
1996 1997 1998 1999 2000 2001 2002 2003 Total
2,4·109 2,5·109 0 0,3·109 7,3·109 1,4·109 4,0·109 0,5·109 18,4·109
Cost of bank protection
1. Tan Chau: 69·109 VND for a bank protection structure with a length of 0,61 km (113·109
VND per km) and a depth of approx. 45 m (slope ≈ 1:3);
2. Long Xuyen: 45·109 VND for a total protected length of 1,27 km (35,4·109 VND per km) on
3 locations with a depth of approx. 25 m (slope ≈ 1:3).
Dong Thap Province, Mekong Delta
Big floods occured in 1961, 1966, 1978, 1991, 1996, 2000, 2001 and 2002. During this up to
90% of the area of the province got inundated. Bank erosion occurs in 8 districts, over a total
length of 106 km.
Consequences of bank erosion
1. Total land loss: 50 ha
2. 1992: Big erosion in Hong Ngu with 28 persons dead or missing and 20·109 VND of damage
3. Sa Dec: Continuing erosion over a length of 10 km, destroying 2 bridges, 10 km of road, a
hospital, a school and other public buildings, with a total damage of 100·109 VND
4. Number of families already moved: 280
5. Number of families to be moved until 2005: Approx. 5.080
6. Budget for family relocation: 4,5·109 VND (≈0,9·106 VND/family)
Cost of bank protection
1. Sa Dec: 50·109 VND for bank protection over a total length of 0,92 km (54,3·109 VND per
km), with an approximate depth of 25 m and a slope of 1:3;

110
2. Requested for new works: 100·109 VND.
Vinh Long Province, Mekong Delta
Consequences of bank erosion
1. Number of families moved between 1995 and 2002: 951
2. Land loss: 22 to 25 ha/year
3. Approximate cost of damage: 142·109 VND/year
Cost of bank protection
1. 355·109 VND for a total protected length of 9,01 km (39,4·109 VND per km) on 15
locations.
Ha Tay Province, Red River
Consequences of bank erosion
1. Number of people already relocated: 2.000
2. Number of people still to be relocated: 2.000
Cost of bank protection
1. Annual DDMFC budget for bank protection (maintenance and new structures, distribution
depends greatly on severity of flood year): Approx. 55 109 VND for the whole Red River
Delta (19 provinces);
2. Total annual DDMFC budget for Red River Delta: 160 109 VND
Hung Yen Province, Red River
Consequences of bank erosion since 1994
1. Land loss: 55 ha

111
Appendix IX: Supplementary informations
Vetiver Network Viet Nam
Home
Photos
News
The Vetiver
Network
Contact us
For further information about VNVN, please send us email at Vetiver
Network Viet Nam. We have available The green book in both
Vietnamese and English. We also have pictorial brochures about
vetiver, as well as CDs prepared by Paul Truong.
Please email, fax, call, or write to let us know the quantity of each you
would like – we'll see what we can do and let you know!
Please let us know a little about yourself, your background and
interests as related to vetiver, your organization (if any), and anything
else you think might contribute to furthering the goal of VNVN. Or, if
you wish, just write to say hello!
In addition, for the names and contact information of other people
working on vetiver, please go to other vetiver contacts.
Vetiver Network Viet Nam
Pham Hong Duc Phuoc (Mr)
Acting Coordinator
University of Agriculture and Forestry
International Relations Office
Ho Chi Minh City
Viet Nam
Tel: (84) (8) 8966946
Mobile: (84) 0913920173
Fax: (84) (8) 8960713
Email: Vetiver Network Viet Nam
http://www.vetiver.org/
http://www.vetiver.org/TVN_FRONTPAGE_ENGLISH.htm
http://www.vetiver.org/VetiverNetworkVietNam/Contact.htm
Vetiver Network Viet Nam: phdphuoc@hcm.vnn.vn

112
http://www.gisdevelopment.net/magazine/gisdev/2002/oct/dcrmrv.shtml
http://www.gisdevelopment.net/application/natural_hazards/floods/nhcy0009pf.htm
Using remotely sensed data to detect changes of riverbank in Mekong River,
Vietnam
Pham Bach Viet, Lam Dao Nguyen and Ho Dinh Duan
Information and Remote Sensing Division -
Institute of Physics, Hochiminh City
Email: vientham@hcm.vnn.vn
Introduction
Traditional methodologies in study of riverbank change require conventional surveys, repeated
measurements to identify and to evaluate changes. Hydrology, geomorphology and geology make use of
data obtained from their surveys as input in their mathematical modeling. Recent studies on Mekong
River have been focused on erosion processes of shorelines at hot spots1). The common feature to all
these studies is that they are localized in extent. Remote sensing techniques offer another approach to
this issue - the use of satellite imagery combined with other digital data to extract information and derive
certain measurements, as in an assessment of channel migration of Thu Bon River using scanned data-
aerial photos and satellite imagery2). A typical study of channel migration in Yellow river (China) made
use both analog and digital data with a time sequential imageries of 19 dates from 1976 to 19943).
This paper presents an application of time-series satellite digital data of different sources composed of
optical and radar imageries in shoreline change detection and to demonstrate a capability of remotely
sensed data with digital processing and GIS analysis for river studies in a large area.

113
http://www.ihe.nl/we/dicea/default.htm?/we/dicea/cress.htm
Coastal and River Engineering Support
System
A cooperation project of the Netherlands Ministry of Public Works
(Rijkswaterstaat), IHE-Delft and TU-Delft
Introduction
This program is intended as a support for design and planning of coastal and
river projects and is not intended to replace the judgement of a qualified
engineer on a particular project. The contents of CRESS are not to be used for
advertising, publication or promotional purposes. Citation of trade names does
not constitute an official endorsement or approval of the use of such
commercial products.The issuing partners do not accept liability for
interpretations or implementations made by users of this program.
Nederlandse versie; voorlopig alleen RWS-Cress, met uitgebreide help-files.
English version; via this link also the French, Indonesian and Chinese version of Cress are downloadable.
Notice:
A
t this moment there are two versions of Cress available. One version is
originally developed by IHE-Delft, the other by Rijkswaterstaat. At this moment
both programs are being merged. The final product wil look like the RWS-
Cress version.
For Information, you may contact:
Rijkswaterstaat, Bouwdienst ir. C. Dorst Postbus 20000, 3502 LA Utrecht
UNESCO-IHE-Delft ir. M. van der Wegen P.O. Box 3015, 2601 DA Delft
TU-Delft ir. H.J. Verhagen P.O. Box 5048, 2600 GA Delft
For help use the helpfiles of the site of the Hydraulic Engineering Department of the
Faculty of Civil Engineering of TU Delft
Coastal and River Engineering Support System
A cooperation project of the Netherlands Ministry of Public Works
(Rijkswaterstaat), IHE-Delft and TU-Delft
Introduction
Cress is available for use under DOS and under Windows (3.1, '95 and '98). The DOS
version is also (partly) available in French and in Bahassa Indonesia. A windows

114
version is available in Chinese.
In case of problems, you may consult the faq-list
All versions are downloadable:
download RWS-Cress (English version)
download RWS-Cress (Dutch version)
download IHE-Cress (DOS version)
download IHE-Cress (Windows version)
download IHE-Cress (DOS version, French)
download IHE-Cress (DOS version, Bahasa Indonesia)
download IHE-Cress (Windows version, Chinese version)
After downloading, you have to decompress the files, using an unzip program (for
example PKunzip).
For both versions a manual is available in HTML format:
go to IHE-Cress DOS manual
go to IHE-Cress Windows manual
In a separate file you will find the Table of contents
Reprint or republication of this program should give appropriate credit to the
International Institute for Infrastructural, Hydraulic and Environmental Engineering,
IHE-Delft, P.O. Box 3015, 2601 DA Delft, The Netherlands, http://www.ihe.nl
Philosophy of the package
In mathematical modelling of coastal processes there is in general a tendency to
make programs more sophisticated and more advanced. The consequence of this
modelling is that models become usually more specialized, and also more difficult to
handle.
Although much effort is paid to the user friendliness of systems, general systems
require much input, which has to be defined in some way. Most programs nowadays
can be handled relatively easy only if one is familiar with the program.
On the other hand, 90 % of the problems in engineering are rather standard
problems. These problems require only the application of very few formulae.
Continuous research is going on to improve the quality of such formulae, although
also here is a tendency to concentrate on the more exotic cases. This is very
understandable, because for a researcher the challenge of such problems is much
more attractive. For the design engineer, this development is not so attractive,
because for his daily work he is therefore often condemned to use outdate reference
material. Especially engineers working in smaller companies or agencies have
difficulties is accessing the latest developments. The Shore Protection Manual is still
their major source of reference information.
Because application of a dedicated program requires familiarity with the input
structure, many designers having a minor problem, will not use such dedicated
programs. The time they have to invest in learning how to handle the program is too

115
much in comparison with the importance of the problem. So in such cases designers
often go back to graphs and design manuals. To overcome this problem, IHE has
developed a very simple package, called CRESS (Coastal and River Engineering
Support System). In fact, CRESS is a collection of small routines, each containing a
formula, or group of formulae, important in coastal and river engineering. The input
and output is highly standardized, and is both available in numerical and graphical
form. Working with CRESS is fast and simple, the package is designed in such a way
that is works on all types of machines, even on the slow ones, and that it does not
require a lot of memory. Uncompressed it still fits on a 720 Kb diskette.
For a design several steps have to be taken. CRESS does not automatically transfer
data from one step to the next one. In this case the user is forced to think about the
input and all the intermediate results. CRESS does not prevent the user to apply a
formula outside its range of application. Often in engineering it is useful to do so,
however it has to be done with great care. Tendencies and sensitivities are very
important. The graphical routines of CRESS are made in such a way, that the
sensitivity of a given parameter can be shown easily in the diagram. The available
Help routines allow the user to find some background information on applied formula,
but also lists of constants, which can be entered (for example the Manning
coefficients).
For more detailed background information sometimes is referred to literature, but
mostly to the IHE lecture notes. We try to make more references to Dicea-files. These
files are available via the web-site of IHE (www.ihe.nl) and can be downloaded if
needed.
Being an educational institute, our first aim in developing CRESS was not to provide a
handy tool for designers, but to develop an instrument for the training of our students.
In our view a design engineer has to be able to understand the physical background
of the formulae used, and has to know the sensitivity of the various input parameters.
In most cases it is not necessary that a designer can derive all formulae used, neither
it is necessary to know the formulae by heart. However, for training in real design, one
has to apply formulae quite often, especially in order to develop a feeling for the
ranges of validity and for the sensitivity. Doing this with a (programmable) pocket
calculator takes a lot of time (with the risk of many data entry errors) which is in our
opinion not well spent time.
In CRESS all these formulae have been placed. Applying CRESS goes very fast, and
therefore all the available time can be used for evaluation of the given output.
Because of the fast and flexible graphical output, students get develop a feeling for
ranges in a relative short time.
It is our intention to keep CRESS on the level of the state-of-the-art in coastal and
river engineering. When research results in new approaches to design problems, and
in practical application this leads to acceptable results in the design process, such
new developments will be implemented as soon as possible in CRESS. Examples of
new developments in CRESS are the Breakwater armour unit formula of Van der
Meer (as alternative of Hudson), the longshore transport formula of Queens (as
alternative of Cerc), the new Delft run-up formula (as alternative of the 8Htan()
formula).
It is not our intention to build out CRESS as a sophisticated tool able to solve all major
problems in coastal and river engineering. For example, a very simple routine is

116
available to compute a backwater curve. This routine is fast, and also the input can be
generated very fast. However, this routine is not able to compute a backwater curve in
a network, with a time dependent flow, etc. For such problems, one has to use a
package, specially designed to solve that kind of problems.
Licensed users
Licensed users are IHE participants graduated in branches Hydraulic Engineering
branch a and b in 1992 or later years. IHE participants graduated in previous years
can become licensed users by writing to the program administrator. Licensed users
will be informed on updates of the program. They are entitled to receive updated free
of charge. In order to get an update by mail the licensed users should send a
formatted diskette to the program administrator (for the Windows-version 3 diskettes).
IHE will copy the update on that floppy and return it to the sender. However, a faster
way is to download the program directly from the net. It is allowed to copy the
program for third parties. It is appreciated when names and addresses of these third
parties are send to the program administrator.
Program administrators:
ir. H. J. Verhagen, TU Delft
phone: +31.15.2785067
e-mail: h.j.verhagen@ct.tudelft.nl
ir M. van der Wegen, IHE-Delft
phone: +31.15.2151811
fax: +31.15.2122921
e-mail: mvw@unesco-ihe.org

117
Supplement: Short review on bankerosion and protection
1. Causes of erosion and failure
The major causes of bank erosion are:
1. Channel bed degradation;
2. Channel meander;
3. Sand-wave sediment transport;
4. Divided flow conditions;
5. Channel restrictions;
6. Varied streamflow (current-velocity-related tractive forces);
7. Water-level fluctuations (including rapid drawdown triggering slumpages);
8. Long-duration water levels and discharges;
9. Wave action (including ship-induced waves, currents and drawndowns);
10. Poor control of overbank drainage; removal of bank soil by seepage of water through zones
of low erosion resistance (piping) with slabbing and caving of overlying soils or, weather-
induced spalling/cracking of upper bank surface soils, eventually in combination with
surface flow;
11. Devastation of (natural) vegetation;
12. Dredging/sand mining near the bank toe and resulting (scour) holes.
Figure 1: General river flow velocity profiles

118
Figure 2: Detailed river flow velocity profiles
Figure 3: Wave generation by ships

119
Figure 4: Ship-induced waves
The analysis of streambank changes caused by soil erosion is analogous or similar to
conventional stability analysis of an excavated slope. Bank recession with time can be estimated
by using a procedure as shown below. This conceptual procedure combines erosion
characteristics and conventional soil parameters used in limit equilibrium slope stability
analyses. Erosional changes in geometry, such as toe recession and/or bed degradation, can
precipitate slope failure with resulting top retreat of the streambank. The bank recession with
time is equal to the cumulative bank recession caused by erosion and slope failures.
Figure 5: Procedure for evaluating streambank stability (Us Army, 1981)

120
To evaluate streambank stability, it is necessary to estimate changes in geometry due to erosion
and slope movements. Bank recession or bed degradation estimated from the laboratory
relationships developed for tractive (current) erosion is an approximation because it does not
take into account such things as accretion along the bank, secondary currents, and bed
degradation as eroded soil from upstream is deposited at the reach of the river under
consideration. A sediment transport analysis which includes hydraulic sorting and armouring
would be necessary to include the effects of deposition. In addition to changes in geometry due
to current erosion, bank failure causes changes in geometry. Bank failures results when the
induced shear stresses exceed the shear strength of the bank soils. Increases in shear stress can
result from increase in slope height or steepness, increase in external loads, and raoid drawdown
of the river. Decreases in shear strength of the soil can result from an increase in pore-water
pressure, soil expansion, or shear movements.
A general overview of typical failure modes related to bank recession is given in the figures 6 to
8.
Figure 6: Bank collapsing

121
Figure 7: Series of bank slidings (opposite site of the river)

122
Figure 8a: Failure modes of river banks

123
Figure 8b: Failure modes of river banks

124
2. Cliff erosion
Shorelines are areas of (practically) unending conflict among the natural forces in wind, water
and land. Due to these forces shoreline materials are washed-out. Usually, when the eroded
material is replaced by an equal quantity from other areas, the shore remains (dynamically)
stable. However, the local, often momentary, changes can still have catastrophic effects to the
shoreline condition and adjacent properties. In case of riverbanks of natural (non-regulated)
rivers, the river is in continuous process of morphological changes and a large area can be
affected by seasonal changes.
Erosion processes of banks and shorelines are rather complicated and are treated extensively in
a number of textbooks and publications. In general, there is no one universal explanation and
solution; each case needs its own analysis and treatment.
Erosion problems of banks and coastal shorelines can be illustrated by a bluff shoreline where a
variety of forces and processes act together (see figure 9). However, it should be remembered
that it represents only one type of (combined) erosion, and that each case must be treated
individually.
Figure 9: Physical components of bank erosion (US Army, 1981)
The most prevalent causes of bank erosion and especially bluff erosion are scour at the toe
(base) by currents and waves and instability of the bluff materials themselves . The erosion rate
and (in-)stability depend strongly on the type and composition of the soil. Therefore the slope
stability problems are difficult to analyze correctly without (local) expertise in geotechnical
engineering.
As figure 13 illustrates, a typical bluff often consists of different soils deposited in distinct
layers, such as clay, sand, silt, etc. These soils do not permanently stand at a vertical face, but
form an angled slope that varies with the soil and groundwater conditions. This slope forms
following a series of failures whose nature depends on whether the soil is cohesive (clay) or

125
granular (sand, silt, gravel, etc.) Cohesive soils generally slide along a circular or curved arc,
and the soil moves downward and rotates along the failure surface. With granular soils, on the
other hand, vertical sides blocks of soil will drop or the soil will suddenly flow down an inclined
plane. Height is a factor because high bluffs impose greater stresses and are likely to suffer more
severe stability problems than low bluffs.
The internal strength of soils can be decreased by groundwater and seepage flows within the
bank (bluff). For instance, rainwater or river water at high stages (also high tides) is naturally
absorbed and seeps down to lower levels (or after a drop in water levels). In case of low banks
the area can be frequently flooded providing additional source of soil saturation. The weight of
saturated soil is increasing causing potential instability.
Permeable soils, such as coarse sand, allow rapid and free passage of water. Impermeable soils,
such as clay, do not allow the free flow of water except through cracks or other openings, and
the over-pressured conditions may last for a long time leading at certain moment to the
movement of a soil block (especially, in a case of rapid and large drop in the outside water
levels).
In figure 9, the large tree’s roots penetrate the clay layer and provide a path for seepage to the
sand layer beneath. Likewise, as the clay fails, cracks form at the surface, providing a path for
seepage to penetrate the soil, further weaken it, and accelerate the failure process. water can also
enter the clay along the existing circular failure surface, leading to further movement.
Once seepage penetrates the clay and reaches the permeable sand layer, it passes freely to the
lower clay layer, where it flows along the clay’s surface and exits the bluff face. This seepage
can increase the risk of a slope failure. In addition, surface flow can erode the bluff face, causing
gullies and deposits of eroded material on the beach below. The seepage exiting the bluff at the
clay layer can also cause surface erosion.
The added weight of buildings and other structures near the top edge of the bank/bluff can
increase soil stresses and contribute to slope failure.
The other major cause of shoreline problems is current and wave action at the toe (including
ship waves and ship currents and depressions). The eroded material is deposited at the toe and
sorted by currents and /or waves. However, during severe wave activity (estuary and coastal
shorelines, or river stretches with a relatively large fetch), waves can reach the bank/bluff itself
and erode or undercut the toe. Also, the slope of the bottom is important to wave action on the
bank. If the bottom slopes are steep, deep water is closer to shore, more severe wave activity is
possible, and maintenance of a protection is more difficult. Flat bottom slopes, on the other
hand, result in shallower water near the bank, which inhibits heavy wave action at the bank/toe
and provides for potentially better protection conditions.

126
3. Bank erosion in stable river systems [from US streambank Manual]
For this discussion local instability refers to bank erosion that is not symptomatic of a dis-
equilibrium condition in the watershed (i.e., system instability) but results from site-specific
factors and processes. Perhaps the most common form of local instability is bank erosion along
the concave bank in a meander bend which is occurring as part of the natural meander process.
Local instability does not imply that bank erosion in a channel system is occurring at only one
location or that the consequences of this erosion are minimal. As discussed earlier, erosion can
occur along the banks of a river in dynamic equilibrium. In these instances the local erosion
problems are amenable to local protection works such as bank stabilization measures. However,
local instability can also exist in channels where severe system instability exists. In these
situations the local erosion problems will probably be accelerated due to the system instability,
and a more comprehensive treatment plan will be necessary.
Overview of Meander Bend Erosion
Depending upon the academic training of the individual, streambank erosion may be considered
as either a hydraulic or a geotechnical process. However, in most instances the bank retreat is
the result of the combination of both hydraulic and geotechnical processes.
The material may be removed grain by grain if the banks are non-cohesive (sands and gravels),
or in aggregates (large clumps) if the banks are composed of more cohesive material (silts and
clays). This erosion of the bed and bank material increases the height and angle of the
streambank which increases the susceptibility of the banks to mass failure under gravity. Once
mass failure occurs, the bank material will come to rest along the bank toe. The failed bank
material may be in the form of a completely disaggregated slough deposit or as an almost intact
block, depending upon the type of bank material, the degree of root binding, and the type of
failure (Thorne, 1982). If the failed material is not removed by subsequent flows, then it may
increase the stability of the bank by forming a buttress at the bank toe. This may be thought of as
a natural form of toe protection, particularly if vegetation becomes established. However, if this
material is removed by the flow, then the stability of the banks will be again reduced and the
failure process may be repeated.
As noted above, erosion in meander bends is probably the most common process responsible for
local bank retreat and, consequently, is the most frequent reason for initiating a bank
stabilization program. A key element in stabilization of an eroding meander bend is an
understanding of the location and severity of erosion in the bend, both of which will vary with
stage and plan form geometry.
As streamflow moves through a bend, the velocity (and tractive force) along the outer bank
increases. In some cases, the tractive force may be twice that in a straight reach just upstream or
downstream of the bend. Consequently, erosion in bends is generally much greater than in
straighter reaches. The tractive force is also greater in tight bends than in longer radius bends.
This was confirmed by Nanson and Hickin (1986) who studied the migration rates in a variety of
streams, and found that the erosion rate of meanders increases as the radius of curvature to width
ratio (r/w) decreased below a value of about 6, and reached a maximum in the r/w range of 2 to
3. Biedenharn et al. (1989) studied the effects of r/w and bank material on the erosion rates of
160 bends along the Red River in Louisiana and also found that the maximum erosion rates were
observed in the r/w range of 2 to 3.
However, the considerable scatter in their data indicate that other factors, particularly bank
material composition, were also modifying the meander process. The severity and location of
bank erosion also changes with stage. At low flows, the main thread of current tends to follow
the concave bank alignment. However, as flow increases, the flow tends to cut across the convex

127
bar to be concentrated against the concave bank below the apex of the bend. Friedkin (1945)
documented this process in a series of laboratory tests on meandering in alluvial rivers. Because
of this process, meanders tend to move in the downvalley direction, and the zone of maximum
erosion is usually in the downstream portion of the bend due to the flow impingement at the
higher flows. This explains why the protection of the downstream portion of the bend is so
important in any bank stabilization scheme. The material eroded from the outer bank is
transported downstream and is generally deposited in the next crossing or point bar. This
process also results in the deposition of sediment along the upper portion of the concave bank.
This depositional feature is often a good indicator of the upstream location to start a bank
protection measure.
Streambank Erosion and Failure Processes
The terms streambank erosion and streambank failure are often used to describe the removal of
bank material.
Erosion generally refers to the hydraulic process where individual soil particles at the bank’s
surface are carried away by the tractive force of the flowing water.
The tractive force increases as the water velocity and depth of flow increase. Therefore, the
erosive forces are generally greater at higher flows.
Streambank failure differs from erosion in that a relatively large section of bank fails and slides
into the channel. Streambank failure is often considered to be a geotechnical process. A detailed
discussion of the erosion and failure processes discussed below is provided by Thorne (1993).
Identifying the processes responsible for bank erosion is not an easy task and often requires
some training. The primary erosion processes are parallel flow, impinging flow, piping,
freeze/thaw, sheet erosion, rilling/gullying, wind waves, and vessel forces.
Parallel flow erosion is the detachment and removal of intact grains or aggregates of grains from
the bank face by flow along the bank. Evidence includes: observation of high flow velocities
close to the bank; near-bank scouring of the bed; under-cutting of the toe/lower bank relative to
the bank top; a fresh, ragged appearance to the bank face; absence of surficial bank vegetation.
Impinging flow erosion is detachment and removal of grains or aggregates of grains by flow
attacking the bank at a steep angle to the long-stream direction. Impinging flow occurs in
braided channels where braid-bars direct the flow strongly against the bank, in tight meander
bends where the radius of curvature of the outer bank is less than that of the channel centerline,
and at other locations where an in-stream obstruction deflects and disrupts the orderly flow of
water. Evidence includes: observation of high flow velocities approaching the bank at an acute
angle to the bank; braid or other bars directing the flow towards the bank; tight meander bends;
strong eddying adjacent to the bank; near-bank scouring of the bed; under-cutting of the
toe/lower bank relative to the bank top; a fresh, ragged appearance to the bank face; absence of
surficial bank vegetation.
Piping is caused by groundwater seeping out of the bank face. Grains are detached and entrained
by the seepage flow (also termed sapping) and may be transported away from the bank face by
surface run-off generated by the seepage, if there is sufficient volume of flow. Piping is
especially likely in high banks or banks backed by the valley side, a terrace, or some other high
ground. In these locations the high head of water can cause large seepage pressures to occur.
Evidence includes: pronounced seep lines, especially along sand layers or lenses in the bank;
pipe shaped cavities in the bank; notches in the bank associated with seepage zones and layers;
run-out deposits of eroded material on the lower bank. Note that the effects of piping erosion
can easily be mistaken for those of wave and vessel force erosion (Hagerty, 1991a,b).
Freeze/thaw is caused by sub-zero temperatures which promote freezing of the bank material.
Ice wedging cleaves apart blocks of soil. Needle-ice formation loosens and detaches grains and

128
crumbs at the bank face. Freeze/thaw activity seriously weakens the bank and increases its
erodibility. Evidence includes: periods of below freezing temperatures in the river valley; a
loose, crumbling surface layer of soil on the bank; loosened crumbs accumulated at the foot of
the bank after a frost event; jumbled blocks of loosened bank material.
Sheet erosion is the removal of a surface layer of soil by non-channelized surface run-off. It
results from surface water draining over the bank edge, especially where the riparian and bank
vegetation has been destroyed by encroachment of human activities. Evidence includes: surface
water drainage down the bank; lack of vegetation cover, fresh appearance to the soil surface;
eroded debris accumulated on the lower bank/toe area.
Rilling and gullying occurs when there is sufficient uncontrolled surface run-off over the bank to
initialize channelized erosion. This is especially likely where flood plain drainage has been
concentrated (often unintentionally) by human activity. Typical locations might be near
buildings and parking lots, stock access points and along stream-side paths. Evidence includes: a
corrugated appearance to the bank surface due to closely spaced rills; larger gullied channels
incised into the bank face; headward erosion of small tributary gullies into the flood plain
surface; and eroded material accumulated on the lower bank/toe in the form of alluvial cones
and fans.
Wind waves cause velocity and shear stresses to increase and generate rapid water level
fluctuations at the bank. They cause measurable erosion only on large rivers with long fetches
which allow the build up of significant waves. Evidence includes: a large channel width or a
long, straight channel with an acute angle between eroding bank and longstream direction; a
wave-cut notch just above normal low water plane; a wave-cut platform or run-up beach around
normal low-water plane. Note that it is easy to mistake the notch and platform produced by
piping and sapping for one cut by wave action (Hagerty, 1991a,b).
Vessel Forces can generate bank erosion in a number of ways. The most obvious way is through
the generation of surface waves at the bow and stern which run up against the bank in a similar
fashion to wind waves. In the case of large vessels and/or high speeds these waves may be very
damaging. If the size of the vessel is large compared to the dimensions of the channel
hydrodynamic effects produce surges and drawdown in the flow. These rapid changes in water
level can loosen and erode material on the banks through generating rapid pore water pressure
fluctuations. If the vessels are relatively close to the bank, propeller wash can erode material and
re-suspend sediments on the bank below the water surface. Finally, mooring vessels along the
bank may involve mechanical damage by the hull. Evidence includes: use of river for
navigation; large vessels moving close to the bank; high speeds and observation of significant
vessel-induced waves and surges; a wave-cut notch just above the normal low-water plane; a
wave-cut platform or "spending" beach around normal low-water plane. Note that it is easy to
mistake the notch and platform produced by piping and sapping for one cut by vessel forces
(Hagerty, 1991a,b).
Ice rafting erodes the banks through mechanical damage to the banks due to the impact of ice-
masses floating in the river and due to surcharging by ice cantilevers during spring thaw.
Evidence includes: severe winters with river prone to icing over; gouges and disruption to the
bank line; toppling and cantilever failures of bank-attached ice masses during spring break-up.
Other erosion processes (trampling by stock, damage by fishermen, etc.) could be significant but
it is impossible to list them all.
Serious bank retreat often involves geotechnical bank failures as well as direct erosion by the
flow. Such failures are often referred to as "bank sloughing" or "caving," but these terms are
poorly defined and their use is to be discouraged. Examples of different modes of geotechnical
stream bank failure include soil fall, rotational slip, slab failure, cantilever failure, pop-out

129
failure, piping, dry granular flow, wet earth flow, and other failure modes such as cattle
trampling. Each of these is discussed below.
Soil/rock fall occurs only on a steep bank where grains, grain assemblages or blocks fall into the
channel. Such failures are found on steep, eroding banks of low operational cohesion. Soil and
rock falls often occur when a stream undercuts the toe of a sand, gravel or deeply weathered
rock bank. Evidence includes: very steep banks; debris falling into the channel; failure masses
broken into small blocks; no rotation or sliding failures.
Shallow slide is a shallow seated failure along a plane somewhat parallel to the ground surface.
Such failures are common on banks of low cohesion. Shallow slides often occur as secondary
failures following rotational slips and/or slab failures. Evidence includes: weakly cohesive bank
materials; thin slide layers relative to their area; planar failure surface; no rotation or toppling of
failure mass.
Rotational slip is the most widely recognized type of mass failure mode. A deep seated failure
along a curved surface results in back-tilting of the failed mass toward the bank. Such failures
are common in high, strongly cohesive banks with slope angles below about 60o .Evidence
includes: banks formed in cohesive soils; high, but not especially steep, banks; deep seated,
curved failure scars; back-tilting of the top of failure blocks towards intact bank; arcuate shape
to intact bank line behind failure mass.
Slab-type block failure is sliding and forward toppling of a deep seated mass into the channel.
Often there are deep tension cracks in the bank behind the failure block. Slab failures occur in
cohesive banks with steep bank angles greater than about 60o. Such banks are often the result of
toe scour and under-cutting of the bank by parallel and impinging flow erosion. Evidence
includes: cohesive bank materials; steep bank angles; deep seated failure surface with a planar
lower slope and nearly vertical upper slope; deep tension cracks behind the bank-line; forward
tilting of failure mass into channel; planar shape to intact bank-line behind failure mass.
Cantilever failure is the collapse of an overhanging block into the channel. Such failures occur
in composite and layered banks where a strongly cohesive layer is underlain by a less resistant
one. Under-mining by flow erosion, piping, wave action and/or pop-out failure leaves an
overhang which collapses by a beam, shear or tensile failure. Often the upper layer is held
together by plant roots. Evidence includes: composite or layered bank stratigraphy; cohesive
layer underlain by less resistant layer; under-mining; overhanging bank blocks; failed blocks on
the lower bank and at the bank toe.
Pop-out failure results from saturation and strong seepage in the lower half of a steep, cohesive
bank. A slab of material in the lower half of the steep bank face falls out, leaving an alcove-
shaped cavity. The over-hanging roof of the alcove subsequently collapses as a cantilever
failure. Evidence includes: cohesive bank materials; steep bank face with seepage area low in
the bank; alcove shaped cavities in bank face.
Piping failure is the collapse of part of the bank due to high groundwater seepage pressures and
rates of flow. Such failures are an extension of the piping erosion process described previously,
to the point that there is complete loss of strength in the seepage layer. Sections of bank
disintegrate and are entrained by the seepage flow (sapping). They may be transported away
from the bank face by surface run-off generated by the seepage, if there is sufficient volume of
flow. Evidence includes: pronounced seep lines, especially along sand layers or lenses in the
bank; pipe shaped cavities in the bank; notches in the bank associated with seepage zones; run-
out deposits of eroded material on the lower bank or beach. Note that the effects of piping
failure can easily be mistaken for those of wave and vessel force erosion.
Dry granular flow describes the flow-type failure of a dry, granular bank material. Other terms
for the same mode of failure are ravelling and soil avalanche. Such failures occur when a

130
noncohesive bank at close to the angle of repose is undercut, increasing the local bank angle
above the friction angle. A carpet of grains rolls, slides and bounces down the bank in a layer up
to a few grains thick. Evidence includes: noncohesive bank materials; bank angle close to the
angle of repose; undercutting; toe accumulation of loose grains in cones and fans.
Wet earth flow failure is the loss of strength of a section of bank due to saturation. Such failures
occur when water-logging of the bank increases its weight and decreases its strength to the point
that the soil flows as a highly viscous liquid. This may occur following heavy and prolonged
precipitation, snow-melt or rapid drawdown in the channel. Evidence includes: sections of bank
which have failed at very low angles; areas of formerly flowing soil that have been preserved
when the soil dried out; basal accumulations of soil showing delta-like patterns and structures.
Other failure modes could be significant, but it is impossible to list them all. Cattle trampling is
just one example of a common failure mode.

131
4. Bank erosion and planform changes
4.1 Overview
In this chapter planform changes are discussed. The emphasis on natural changes. Broadly
speaking a distinction is made to gradual and sudden changes in channel planform. Gradual
changes are usually linked to bank erosion. Sudden changes occur when cutoffs develop. Both
changes are discussed in this chapter.
In this chapter causes and rates of bank erosion and some prediction methods are discussed.
Section 2 is mainly dealing with bank erosion along meandering rivers. Section 3 discusses cutoffs
of meandering rivers, but the analysis can be applied to braided rivers as well. Next planform
changes of braided rivers are reviewed. The main difference between the two is that bank erosion
along a meandering river can be predicted fairly well, while bank erosion along braided rivers is
much more subject to stochastic behaviour. Also other planform changes in braided rivers are
reviewed. Section 5 discusses the predictability of planform changes.
4.2 Bank erosion of meandering rivers
General
Bank erosion is a common feature of all alluvial rivers. Rates of lateral erosion for various rivers
are greatly different due to variations in geological structure and sedimentological composition of
the valley material. Rates of lateral migration vary from a few meters per year (Brice 1984, USA
rivers) via some tens of meters per year (Mahakam River, Kalimantan, Indonesia) and 50-75
m/year (Missouri, Mississippi) to values of 30 to 750 m/year for the Brahmaputra River, occuring
mainly during the flood season (Coleman, 1969; Klaassen & Masselink, 1992).
Bank erosion is most prominent in river bends due to the increase of velocities in the outer bend
and the spiral flow which tends to deepen the outer bend. The rate of bank erosion is determined
by the strength of the bank on the one hand and the fluid forces on the other hand. Often a
distinction is made between noncohesive and cohesive banks.
Bank erosion usually tends to increase the length of the river, especially if the erosion is directed
in lateral direction and not so much in downstream direction. Cutoffs are the typical phenomenon
balancing this gradual lengthening.
Bank erosion processes
Bank erosion can be due to fluvial entrainment of individual particles (mainly by large velocities
and shear stress on the banks), undermining of the toe of the bank and subsequent soil-mechanical
failure or liquefaction by overpressure in fine sand during falling water levels. Coleman (1969)
observed that for the Branhmaputra River the majority of the failures was due to current
undermining and subsequent failure of the levee deposits. For the toe erosion it holds that the
material which enters into the river has to be eroded. Hence the bank erosion is controlled on the
one hand by the instability of the banks and on the other hand by the sediment transport capacity
of the flow near the outer bend. For more details on bank erosion processes see the article by
Osman and Thorne (1988). The article of Hagerty (1991) draws attention to the possible existance
of horizontal layers layers of different permeability in the banks and their effects.

132
Bank erosion rates
There is not much theoretical or even empirical work related to quantitative prediction of bank
erosion rates. Brice (1984) found that bank erosion rate increased linearly with drainage area for a
number of USA rivers. Here two different approaches are discussed. One approach deals with
estimates of yearly erosion and was developed by Hickin and Nanson (1984) for rivers in Canada.
Klaassen and Masselink used the same approach for the Brahmaputra River in Bangladesh. The
other approach, included in models in which the bank erosion is simulated, links the momentary
bank erosion to the local conditions (either flow or bank height).
Hickin and Nanson (1984) did an extensive study using photographs of sand and of the radius in
Western Canada. They found that erosion rate was a function of the radius of curvature to width
ratio R/W, with a maximum at R/W = 2.5. This can be written as:
M(R/B) = M2.5 . f(R/B) (1)
For f(R/W) an empirical relation was derived:
for 1 <R/B < 2.5 f(R/B) = 2/3 (R/B-1) (2a)
for R/B > 2.5 f(R/B) = 2.5 B/R (2b)
The maximum erosion rate occurs for R/B = 2.5 and is defined as M2.5 (in m/year) and this
maximum rate is proportional to total streampower Ù, which is defined as:
Ω = Q5 τ h-1 = ρ g Q5 i (3)
where Ω = total stream power (in Watt/m'), Q5 = discharge exceeded once in 5 years (m3/s).
M2.5 is inversely proportional to a bank-strength parameter YB (dimension N/m2) which is a
function of bed material size:
2.5
B
=
Mh.Y
Ω (4)
Example: Given:
Q
5 = 800 m3/s i = 10-3 h = 5 m
D
bank = 1 mm R/B = 5
Solution: Dbank = 1 mm -> YB = 80 N/m2
R/W = 5 -> f(R/W) = 0.5
Ω = ρg i Q5 = 8000 Watt/m'
M
2.5 = Ω. h-1YB-1 = 20 m/year
M = M2.5 . f(R/W) = 10 m/year.
The use of this relation in other parts of the world must be done with care but can give some idea
of possible bank erosion rates.
The other approach uses local conditions like the local shear stress, the local velocity the local
sediment transport and/or the local bank height for estimating the bank erosion rates. Typically

133
also here calibration coefficients have to be introduced which represent a.o. the bank properties.
Althought the approach may be more generally applicable, these calibration coefficients have to be
determined for each river separately.
Prediction methods which take into account the local conditions have been proposed by e.g.
Ariathurai and Arulanandan (1978), Crosato (1990), Mosselman (1992), DHI (1996) and
Shishikura (1996). A typical example of such a predictor is:
B
c
ac
-
n = E
t
δ
τ
τ
δ
τ
(5)
where nB = bank line position and in the right side the local shear stress and the criritical shear
stress are present. The coefficient Ea has to be calibrated to the local conditions.
4.3 Cutoffs in meandering rivers
Natural cutoffs occur when channels start to develop that cut short a river bend. Cutoffs occur both
in meandering channels and in braided channels. For the development of such a cutoff channel
there are two criteria should be applied:
(1) The shear stress should exceed the critical shear stress.
(2) The sediment entering the potential cutoff channel should be less than the sediment
transport capacity of the cutoff channel.
In meandering rivers cutoff channels develop usually in the floodplains. These floodplains are
often characterized by cohesive soils and vegetation. Both aspects have a pronounced influence on
the resistance to erodibility. Hence often those cutoffs do not develop easily. If the dimensions of a
cutoff are characterized by the cutoff ratio ë (see Klaassen & van Zanten, 1989), then typical
ratio's of ë for meandering rivers are 5 to 30. For a particular river, often use can be made of
"scars" in the terrain to study old natural cutoffs and hence to decide on a typical cutoff ratio for a
particular river. Artificial cutoffs can be made by making use of the analysis of these natural
cutoffs. Such artificial cutoffs are succesfull even for smaller ë if the depth during flood is
sufficiently large. The shear stress should be larger than the critical shear stress. Often does the
critical shear stress become smaller in vertical direction because of the decreasing influence of
cohesion (the lower strata contain usually less fine sediments). The pilot channel should have
sufficient width. Initially the scouring process will promote vertical erosion, only later will the
pilot channel widen.
In braided sand-bed rivers cutoffs occur frequently. Cutoff ratio's between 1.0 and 1.5 have been
observed (see Klaassen & Masselink,1992). This is due to the fact that during flood the critical
shear stress is exceeded everywhere. Also on the bars that are present in areas where potential
cutoffs may occur. Hence whether a cutoff occurs is dependent on the ratio of the transport
capacity in the potential cutoff and the quantitites of sediment entering at the upstream bifurcation.
As shown by Mosselman et al (1993) in particular the angle of the upstream angle is an important
parameter for the occurrence of a cutoff, implying that the sediment distribution at the bifurcation
(which is very sensitive to the geometry of the bifurcation) plays a key role. For more details on
the occurrence of cutoffs in braided sand bed rivers see Klaassen & van Zanten (1989).

134
4.4 Planform changes of braiding rivers
See the articles of Klaassen and Masselink (1992) and Klaassen et al (1993).
4.5 On predictability and modelling of bank erosion and planform changes
Traditionally river planforms have been classified as alternate bars, meandering and braiding (see
Leopold et al, 1964). As far as bank erosion is concerned, a major difference exists between
braided rivers on the one hand and straight rivers with alternate bars and meandering rivers on the
other hand. It seems that for straight rivers with alternate bars and for meandering rivers the future
planform changes can be predicted with some accuracy. Bank erosion occurs mainly along the
outer bend, where the velocities are largest. Recently this has also been applied in the development
of mathematical models for the prediction of the planform changes along these meandering
channels (Crosato, 1990; Mosselman, 1992). It is of course obvious that the duration of the flood
is important input parameter. This is not fully taken into account in the prediction method of
Hickin & Nanson (1984).
Two types of mathematical models are presently available to model bank erosion. MEANDER of
WL/Delft Hydraulics is specifically useful for the simulation and prediction of meandering for
decades ahead. The prediction is essentially deterministic (given a certain discharge hydrograph),
but it has been applied in a stochastic way by varying the input discharge hydrographs and
assuming the occurrence of cutoffs depending on the extremity of the flood. Calibration of the
model on field data (changes of meandering patterns over the past decades, e.g. from satellite
imagery) is required. The model has been applied in several WL/Delft Hydraulics studies. In
MIKE21 of DHI a bank erosion module is coupled to a 2D flow, sediment transport and
morphology module. In practice this model, which has been used for predicting planform changes
along the Brahmaputra River, can only be used a few years ahead, because the required computer
capacity is still quite high.
For braided river systems the predictability is substantially less. Clearly some chaotic behaviour is
apparent, not because of the discharge hydrograph but mainly because of the highly non-linear
character of the system. It seems that small differences in location of bars in the system can induce
the rapid formation of cutoffs and the closing of existing channels. This holds especially for
braided sand bed rivers like the Lower Brahmaputra River in Bangladesh. See Klaassen &
Masselink (1992).Although some planform elements remain to be present over a long period, in
particular severe bank erosion shifts rapidly form one place to another in a few years. For the time
being mathematical models for planform changes for this type of rivers are not advocated. See
Klaassen et al (1994) and Jagers (2003).

135
5. Survey and Data Collection
Preliminary Analysis for Bank Erosion Reconnaissance
Country: Place: Date:
Problem definition
Qualitative description of occured damage and expected future damage due to bank erosion on
specific locations. Damage may be material or non-material.
Documentary Survey
Documents and general information about the river system to be gathered and consulted before a
field visit:
1. Both recent and old topographical maps at various scales
2. Morphological maps
3. Geological maps
4. Aerial photographs and satellite data
5. Small and large scale nautical river charts (bathymetry)
6. Longitudinal and tranversal river profiles
7. Tidal tables for the lower river reaches
8. Meteorological and hydrological data
9. Statistic data about flooding
10. Old newspapers and other historical documents about flooding and bank erosion
11. Technical reports of existing or planned protection structures along the river
Detailed information
1. Land features (sources: maps, photographs)
a. Land use near the river banks: Recreation/nature conservation & national parks/agriculture/
fish farming/housing/infrastructure/industries
b. Elevation of the land behind the banks
c. Widths of land threatened by inundation or erosion
d. Presence of flood protection structures: Levees and dikes
2. Bank features (sources: maps, photographs, transversal cross-sections)
a. Bank profiles: Steep/smooth
b. Bank protection structures: Revetments/groins
3. River features and bathymetry (sources: maps, photographs, charts, technical reports)
a. Type of river: Meandering/braiding
b. River mouth configuration: Single or multiple delta outlets
c. Stability of the river bed: Stable/degrading
d. Location of important tributaries
e. Large scale human interference affecting hydraulic and morphologic conditions:
channelization work/changes in upstream reservoir operation/changes in land use
f. Specific river data: Flow velocities, flood discharges, bed gradients, depth contours, currents,
sediment loads of the river and its main tributaries (m3/year), flooding frequencies

136
4. Tides in the lower river reaches (source: tidal tables)
a. Mean high water level: MSL + ... m
b. Mean low water level: MSL - ... m
c. Mean tidal range: ... m
d. Mean high water springs: MSL + ... m
e. Mean low water springs: MSL - ... m
f. Mean spring range: ... m
g. Mean high water neaps: MSL + ... m
h. Mean low water neaps: MSL - ... m
i. Mean neap range: ... m
j. Mean tidal current: ... m/s
5. Waves (source: technical reports)
a. Amount of ship traffic, type and speed of ships
b. Fetch lengths, depth and wind characteristics
c. Spectrum of wave heights
d. Spectrum of wave periods
e. Spectrum of wave directions
6. Bank erosion data (source: technical reports)
a. Eroded volumes (m3/year)
b. Eroded widths (m/year)
c. Eroded lengths (m)
7. Soil data
a. Profiles from CPT's and borings
b. Soil characteristics: sieve curves, d10, d50, d90, angle of internal friction, cohesion, etc.

137
Field inspection
Province: Community: Location (km): Date:
Land features
a. Land use near the river bank: Recreation/nature conservation & national parks/agriculture/
fish farming/housing/infrastructure/industries
b. Elevation of the land behind the bank: Low/high, flat/rising
c. Evidence of frequent overbank flows: Sand splays/overbank erosion/crop damage
d. Width of land threatened by inundation or erosion
e. Presence of flood protection structures: Levees and dikes
Bank features
a. Bank outline: Straight/wide curve/narrow curve
b. Bank heights and angles
c. Bank failures mechanisms: Slab/rotational failures
d. Bank stratigraphy: Material and thickness of the observed layers
e. Vegetation: Distribution, size, type, and approximate age on the bank and in the channel
f. Signs of erosion and bank instability: Tension cracks, collapsed banks, gullies
g. State of erosion: Little/significant/severe/very severe
h. Highwater and flood marks (debris trapped in vegetation)
i. Presence and type of (nearby) bank protection structures: Groins/revetments
j. Condition of bank protection structures: Good/reasonable/eroded in one or more places/being
undermined/destroyed
River features
a. Characterization of river reach: Lower reach/delta/middle reach/upper reach/meandering/
braiding
b. Presence of terraces (inactive floodplains) and berms (active floodplains)
c. Width and depth of river channel at the level of the floodplain, the top bank, berms and
terraces (if present) at about every 15 to 20 channel widths along the channel
d. Nearby presence and significance of tributaries
e. Flow velocity: High/medium/low
f. Presence of strong currents, turbulent flow, eddies
g. Approximate Manning roughness for the active channel, berms and floodplain
h. Sediment load: High/medium/low
i. Amount and size of sediment deposited at and just downstream of tributary confluences
j. Bed material: Gravel/coarse sand/fine sand/silt/clay
k. Bed stability of the river reach: Stable/degrading, presence of knickpoints & knickzones
l. Condition (scour) and influence of nearby river structures (bank protection, bridges, drop
inlet structures, culverts, grade control structures, water intakes, pipelines) on the river flow
regime and morphology
Sediment
a. Sediment sources: banks/land behind the banks/watershed (upland) erosion
b. Sediment carriers: tributaries/gullies/drainage ditches/river bed
c. Sediment samples from low-lying hinterland
d. Sediment samples from each stratigraphic bank layer
e. Sediment samples from thalweg and some other points in a cross section
f. Sediment samples from tributary mouth bars

138
Questionnaire for local authorities and people (i.e. fishermen)
Province: Community: Location (km):
Date:
Flooding
a. How often does flooding occur?
b. Are there very big differences between the heights and duration of the floods?
c. What is the worst flooding in memory?
Bank erosion
a. How much bank erosion has occured in the last year?
b. Since when does the bank erosion occur?
c. Does the amount of erosion greatly vary from one year to another?
d. How much damage has been caused by bank erosion in the last 10 years?
River profile
a. Has the river depth in front of the bank greatly changed in the last few years?
b. Are there scour holes in front of the bank?
c. Is the river depth in front of the bank subject to seasonal changes?
d. Has the river bed been degrading over the last 10 years or is the river bed relatively stable?
Currents
a. How strong is the current near the bank?
b. How often do turbulent flows and eddies occur near the bank?
Waves
a. What are the maximum wave heights which occur on the river?
b. Are these waves caused by ships or do they occur during storms?
c. How often do these waves occur?
d. What are the wave directions?
Winds
a. When do the strongest winds occur?
b. What is the direction and speed of these winds?
c. What is the prevailing wind direction throughout the year?
Other remarks
Collect
a. Maps
b. Documents and interviews
c. Dated pictures of erosion
d. Drawings: cross-sections, protection structures
e. Historical information (changes in time)
f. Soil samples

139
6. Types of bank protection
The choice of viable alternatives from a seemingly endless variety of methods to protect a
streambank against any of several causes of damage and failure is a difficult task. Therefore, as
much knowledge, experience, and guidance as possible should be utilized from the efforts of
others (past and present), the results of previous investigative work, and design directives (as
described in the number of textbooks and design Manuals and as available from other sources).
The following categories of streambank protection techniques, from which more specific items
can be derived, are not mutually exclusive and often could be used in combination:
1. Direct bank protection with materials more resistant to erosion than the underlying soil due
to greater density, mat-type construction, or reinforcement of the bank soils (as with
vegetation).
2. Spur Dikes (Groynes) or Hard Points or other devices to slow the flow velocities along the
bank.
3. Groynes or other techniques to shape the channel alignment locally or extensively so as to
direct the flow away from the bank or reduce sharp curvature of the channel.
4. Protection of the toe of the bank to prevent undercutting.
5. Grade control of the channel bottom.
6. Improving the structural stability of the soil mass of the streambank.
7. Controlling flow of water entering channel over its bank.
Any substantial changes affecting the whole channel should be given very careful consideration
so as not to initiate a "domino effect" of new problems resulting from an improper action to
solve the original problem. If whole-channel improvements are necessary, the best scheme
usually interferes the least with the natural stream condition. As much of the natural channel as
possible should be used, and the water and sediment flow characteristics through the reach
should be changed as little as possible.
A wide variety of both natural and man-made materials are currently available to control bank
erosion. These include rock riprap, concrete blocks in various configurations, concrete mats,
vegetation schemes, etc.. All of these materials gave unique advantages and disadvantages
depending upon the size of the area to be protected, the cause of the bank instability, the
magnitude of hydraulic loading, the availability of the material, and the cost.
Bank protection materials in high-energy environments (turbulent flow, high velocity, waves)
must be placed on appropriate granular or fabric filters (geotextile) to prevent the loss of bank
material to the penetrating currents and waves. Usually, in low-energy environment, quarry-run
riprap (wide-graded rock from the quarry) of sufficient size and thickness might perform well
without filter layer underneath.

140
Figure 10: Failure mechanisms related to slope protection
Rock is the most used material for protection against bank erosion, although the methods of
application and design vary widely. It will likely continue to be the first choice of bank
protection materials where material of sufficient size is available and affordable, because of
durability, flexibility, easy repair, and other advantages.
• A riprap blanket is flexible and is neither impaired nor weakened by slight movement of the
bank resulting from settlement or other minor adjustments.
• Local damage or loss is easily repaired by the placement of more rock.
• Construction is (usually) not complicated and so special equipment or construction practice
is not necessary (besides the sites with large depth and heavy currents).
• Riprap is usually durable and recoverable and may be stockpiled for future use.
• The cost-effectiveness of locally available riprap provides a viable alternative to many other
types of bank protection.
• Riprap stability increases with increasing thickness as more material is available to move to
damaged areas and more energy is dissipated before it reaches the filter and streambank.
Figure 11: Design components of typical revetment

141
Costs of protective schemes vary widely, depending upon the extent of the problem to be solved,
the availability of locally available materials, and the size of the problem (project).
The cost-effectiveness of riprap from a local source remains strongly competitive with other
long-term protection types, and it is usually a very effective erosion protection device. In
addition to riprap, the rock-dominated methods also afford some promise toward effectively
controlling streambank erosion. For example, rock toes with suitable upslope vegetation
function well in some situations. Similarly, the techniques of using rock hard points, jetties, and
windrow provide adequate protection when properly designed, but some initial erosion should
be anticipated before the units become effective. Gabions offer an effective bank protection
technique where suitable riprap is not available.

142
7. Techniques of bank protection
Note: Usually, before placing a revetment the bank should be previously graded to a stable
slope.
Quarry-run rock protection consist of wide-graded-stone material with maximum size limited to
about 0.5 m and relatively large percentage of fines; the recommended thickness is usually much
larger than the traditional riprap to allow washing out of the fines and creating a natural (poor)
filter.
Riprap consist of durable, relatively small-graded stone material (D85/D15<2) with average size
(diameter) defined by the local hydraulic loading, and usually placed in two layers on granular
filter of geotextile (eventually with a cushion layer inbetween).
Figure 12: Application of riprap for bank protection
Gabion protection consists of wire cages filled with small stone or waste brick material. Usually,
filter material (or geotextile) is placed between the gabions and the original bank.

143
Stone mattresses/Reno mattresses/Crushed stone and wire mat protection consist of natural or
crushed stone (about 0.10m max. diameter) placed in a flat wire-basket., or protected by a weir
mat (on the top) anchored to the underlying soil.
Riprap-filled cells or grates consist of a cellular-type containment with bottom and top opening
that can be square, hexagonal, etc, filled with fine riprap.
Waste materials (minestone, slags, silex, building waste, etc.) are often locally available and
(mostely) inexpensive waste products from the industry. They are placed in a number of layers
to obtain the prescribed thickness. Possible environmental impact must be taken into
consideration.
Figure 13: Transformation from riprap to stone pitching
Concrete blocks and slabs

144
Concrete mats
Concrete-filled geo-mattresses (Fabriform) consist of a fabric envelope filled with pumpable
sand and cement grout. A geotextile or a bedding layer are usually provided under the mat.
Figure 14: Block mats
Soil-cement
Asphalt (bituminous revetments)
Windrow revetment consists of a mound of stone placed on the ground, or partially or totally
buried, immediately adjacent and parallel to the general alignment of the eroding bank. As bank-
line caving reaches the windrow, the stone is undercut, thereby falling down the bank and
protecting the bank the bank against further erosion (see figure 15).

145
Figure 15: Windrow revetment, definition sketch (US Corps, 1981)
Figure 16a: Reinforced revetment

146
Figure 16b: Reinforced revetment (US Corps 1981)
Reinforced revetment has a bank-line toe of erosion-resistant material placed riverward of the
high bank, reinforced intermittently by stone-filled tiebacks extending landward from the toe
into the riverbank (see figure 16).
Earth core dikes are mounds of sand fill (or clay) extending riverward of the bank line and
protected on the upstream face by a stone toe and covered by a (relatively) thin layer of stone.
Composite revetment has a bank-line toe of erosion-resistant material, an upper bank treatment
covering the zone of normal seasonal fluctuations, and a freeboard zone that is generally
vegetated (see figure 17).

147
Figure 17: Composite revetment
Grout-filled paper bags consist of nylon reinforced bags filled with sand and cement grout and
placed individually on a prepared slope.
Transverse spur dikes (river groynes) is a standard protective technique in a concave bend of a
meandering stream with (usually) noncohesive banks and insignificant suspended load. Several
design attention points can be mentioned:
• Spur dikes can reduce near-bank velocities to one-half of those that occur without a dike
field;
• Spacing-to-length ratios as high as three may be effective in protecting concave banks with
spur dikes; however, the ratio was found to vary with discharge. Spacing-to-length ratio for
specific projects can be determined by previous experiments in similar circumstances or site-
specific model studies (or prototype pilot project).
• Speu dike root (section extending landward into bank) should be protected from scour
caused by vortices set up along the upstream and downstream faces.
• The spur dike should be (usually) aligned perpendicular to the bank or current.
• Aprons (bottom armour layer, usually fine rock) are effective in limiting the depth of scour
at the spur dike’s toe ( the point of maximum scour will move more downstream from the toe
of the spur dike, improving the structural integrity of the spur dike).
• Existing equations for prediction of scour at spur dikes in concave bends are questionable
and should be interpreted as indicative only.

148
Figure 18: River groins
Permeable spur dikes (or fences, vanes)
The advantages and dis-advantages of permeable groins compared with solid groins or spur
dikes can be summarized as below:
1. The vulnerability for floating debris and ice (United Nations, 1953, Central Board of
Irrigation and Power, 1989 en Federal Highway Administration, 1995);
2. Danger for the inland navigation when the groins are submerged. This holds also for
impermeable groins. (United Nations,1953 en Central Board of Irrigation and Power, 1989);
3. Less effectiveness to guide the flow then impermeable groins (Federal Highway
Administration, 1995).
The authors have different opinions on the degree of bank protection provided by permeable
groins compared to the effectiveness of impermeable groins. The effectiveness of permeable
groins depends probably significantly on the sediment transport in the river. The most important
advantages of permeable groins are:
1. Less deep scour holes, according to the Federal Highway Administration even no scouring at
all if the permeability exceeds 70 % (United Nations, 1953, Central Board of Irrigation and
Power, 1989 en Federal Highway Administration, 1995);

149
2. The flow pattern has less variations, and this benefits the navigation (United Nations, 1953,
Central Board of Irrigation and Power, 1989 en Federal Highway Administration, 1995);
3. Less erosion at the connection of a groin and the bank. (Federal Highway Administration,
1995);
4. Permeable groins are cheaper to construct then impermeable groins (Central Board of
Irrigation and Power, 1989);
5. Permeable groins have a wider construction window (Central Board of Irrigation and Power,
1989);
6. A short construction period (Central Board of Irrigation and Power, 1989).
Hard points consist of two components: a short spur of erosion-resistant material extending
from the bank riverward, and a root of stone placed in a trench excavated landward from the
bank line to preclude flanking (see figure 19). The structure protrude only a short distances into
the river channel. The structures are especially adaptable in long, straight (or slightly concave)
reaches not subjected to high direct attack.
Figure 19: Hard point with section detail
Refusals consist of erosion-resistant material placed in a trench excavated landward at the
upstream end of each revetment segment to prevent flanking.
Breakwaters are structures whose primary purpose is to protect the banks from any erosion that
may be caused by wave action.
Parallel spurdikes are longitudinally placed structures for guiding the stream; however, they
may also fulfil the same function as breakwaters (reduction of wave action).

150
Used automobile tires can be installed to provide protection in the form of either a wall
(bulkhead) filled with sand and capped with concrete, or of a mattress placed on the bank, which
was previously graded to a stable slope.
Geobags filled with sand or clay
Figure 20: Geobags before and after collapse

151
Figure 21: Concrete geomattresses before and after collapse
Geosystems (geotubes, geocontainers, etc.) - geotube protection is a fabric tube placed parallel
to the direction of the flow and filled with either sand or gravel (or sand-cement). Various sizes
are available. When exposed, UV protection is needed. Note: geotubes can also be applied for
construction of spur dikes and breakwaters.
Tree retard systems generally consist of groups of trees cabled together, placed perpendicular to
the bank line, and anchored in place using cables with fabricated weights. A small stone root is
constructed into the bank line to anchor the landward end of the tree and protect the landward
end of each retard from flanking by overtopping flows.
Vane dikes are low-elevation, within-the-channel fills of stone or lower grade material that hold
the high-velocity erosive flows away from the banks and encourage the accumulation of
sediment on the landward side. The flow is allowed to course both ends and overtop the
structure to create and preserve environmentally desirable shallow, braided channels.

152
Figure 22a: Stream guiding by bandals
Figure 22b: Stream guiding by fences and/or vanes

153
Fencing (or vane screens) of various configuration
Sheet-piles and/or bulkheads
Vegetative protection, consisting of either grass or shrubbery, is often provided in conjunction
with some of the methods described previously. Protection consisting only of vegetation can be
applied at low hydraulic loadings.
In general, most bank protection techniques are not economically justified from the cost-benefit
point of view. The choice of protection is usually dictated by the importance of problem and the
financial ability. Considering optimisation, one option may be to perform only minimal
protection first, then repair as necessary, e,g., windrow revetment, low-elevation structures,
intermittent bank-line revetment, or hard points. Also, low-grade or waste materials may be
satisfactory at certain conditions, e.g., minestone, slags or poor quality rock. However, nearly in
all situations, the most important conclusion is to provide effective protection at the toe of the
bank.
Prediction of when, where, and extent of bank erosion and/or bank instability remains a difficult
matter. The forces contributing to bank instability are generally (theoretically) known and
(partly) understood; however, application of the theoretical principles to the real world (practice)
are complicated by the many processes acting simultaneously throughout a given river reach.
Streams displaying very active tendencies to erode their banks often seem to reverse themselves
and display periods of relative stability. These processes will continue to make the prediction of
erosion indeterminate, and most efforts to control the erosion will be based on after-the-fact
information (systematic monitoring and analyse).

154
Figure 23: Alternative toe protections

155
8. References
1. Andrews, E.D. (1982), Bank stability and channel width adjustment, East Fork River, Wyoming, Water
Resources Res., AGU, Vol.18, No.4, pp.1184-1192
2. Ariathurai, R. & K. Arulanandan (1978), Erosion rates of cohesive soils, Journ. Hydr. Div., ASCE, Vol.104,
No.HY2, pp.279-283
3. Biglari, B. (1989), Cut-offs in curved alluvial rivers, Delft Hydraulics, Report Q553/IHE Delft, M.Sc. thesis
no. (...)
4. Burger, J.W., G.J. Klaassen & A. Prins (1988), Bank erosion and channel processes in the Jamuna River,
Bangladesh, Prepared for River bank erosion Symposium, Dhaka (Bangladesh)
5. Christensen, B.A. (1989), Riverbank stability analysis I: Theory (discussion), J. Hydr. Engrg., ASCE,
Vol.115, No.7, pp.1010-1013.
6. Coleman, J.M. (1969), Brahmaputra River: Channel processes and sedimentation, Engineering Geology,
Vol. 3, no. 2/3, pp. 129-239.
7. Crosato, A. (1990), Simulation of meandering river processes, Communications on Hydr. and Geotech.
Engrg., No.90-3, Delft Univ. of Technol., ISSN 0169-6548
8. Darby, S.E. (1994), A physically-based numerical model of river channel widening, Dissertation, Univ.
Nottingham.
9. Darby, S.E. & C.R. Thorne (1994), Prediction of tension crack location and riverbank erosion hazards along
destabilized channels, Earth Surface Processes and Landforms, BGRG, Vol.19, No.3, pp.233-245.
10. Darby, S.E. & C.R. Thorne (1996), Numerical simulation of widening and bed deformation of straight sand-
bed rivers. I: Model development, Journ. Hydr. Engrg., ASCE, Vol.122, No.4, pp.184-193
11. Darby, S.E., C.R. Thorne & A. Simon (1996), Numerical simulation of widening and bed deformation of
straight sand-bed rivers. II: Model evaluation, Journ. Hydr. Engrg., ASCE, Vol.122, No.4, pp.194-202
12. Escarameia, M. (1998), River and channel revetments, Thomas Telford, London
13. Fenwick, G.B. (1969), State of knowledge of channel stabilization in major alluvial rivers, Technical
Report 7 of the Committee on Channel Stabilization of the US Army Corps of Engineers
14. Furbish, D.J. (1991), Spatial autoregressive structure in meander evolution, Geol. Soc. Am. Bull., Vol.103,
pp.1576-1589
15. García, M. & Y. Niño (1993), Dynamics of sediment bars in straight and meandering channels: experiments
on the resonance phenomenon, Journ. Hydr. Res., IAHR, Vol.31, No.6, pp.739-761
16. Guidelines and Design manual for Standarized bank protection Structures, Ministry of Water Resources,
Bangladesh, 2001
17. Gurnell, A.M., S.R. Downward & R. Jones (1994), Channel planform change on the River Dee meanders,
1876-1992, Regulated Rivers: Res. & Management, Vol.9, No.4, pp.187-204
18. Hagerty, D.J. (1989), Riverbank stability analysis I: Theory (discussion), Journ. Hydr. Engrg., ASCE,
Vol.115, No.7, pp.1013-1014
19. Hagerty, D.J. (1991), Piping/sapping erosion. I: Basic considerations, Journ. Hydr. Engrg., ASCE, Vol.117,
No.8, pp.991-1008
20. Hagerty, D.J. (1991), Piping/sapping erosion. II: Identification - diagnosis, Journ. Hydr. Engrg., ASCE,
Vol.117, No.8, pp.1009-1025
21. Hasegawa, K. (1989), Universal bank erosion coefficient for meandering rivers, Journ. Hydr. Engrg., ASCE,
Vol.115, No.6, pp.744-765
22. Hickin, E.J. & Nanson, G.C. (1984), Lateral migration rates of river bends, Journ. Hyd. Eng. ASCE, Vol.
110, no. 11, pp. 1557-1567
23. Hoffmans, G.J.C.M. & H.J. Verheij (1997), Scour manual, A.A.Balkema, Rotterdam
24. Hooke, J.M. & C.E. Redmond (1992), Causes and nature of river planform change, In: Billi, P., Hey, R.D.,
Thorne, C.R. & Tacconi, P. (Ed.), Dynamics of gravel-bed rivers, , Wiley, pp.557-571
25. Hooke, J.M. (1995), River channel adjustment to meander cutoffs on the River Bollin and River Dane,
northwest England, Geomorphol., Vol.14, No.3, pp.235-253.
26. Howard, A.D. (1983), Simulation model of meandering, In: River Meandering, Proc. Conf. Rivers 1983,
New Orleans, Ed. C.M. Elliott, ASCE, 1984, pp.952-963
27. Howard, A.D. & T.R. Knutson (1984), Sufficient conditions for river meandering: a simulation approach,
Water Resources Res., AGU, Vol.20, No.11, pp.1659-1667.

156
28. Howard, A.D. (1992), Modeling channel migration and floodplain sedimentation in meandering streams, In:
Carling, P.A. & Petts, G.E. (Ed.), Lowland floodplain rivers, Geomorphological perspectives, Wiley, pp.1-
41
29. Jagers, H.R.A. (2003), Modelling planform changes of braided rivers, Twente University, Ph.D. Thesis, 318
p.: fig., tab., ref. + cd-rom , ISBN 90-9016879-6
30. Klaassen,G.J. & Masselink, G. (1992), Planform changes of a braided river with fine sand as bed material,
Proc. 5th Intern. Symp. on River Sedimentation, Karlsruhe (FR Germany), pp. 459-471
31. Klaassen, G.J., E. Mosselman & H. Brühl (1993), On the prediction of planform changes in braided sand-
bed rivers, In: Wang, S.S.Y. (Ed.), Adv. in Hydro-Sci. and -Engrg., Publ. Univ. Mississippi, pp.134-146.
32. Klaassen, G.J. & B.H.J. van Zanten (1989), On cutoff ratios of curved channels, Proc. 23rd Congress IAHR,
Ottawa (also Delft Hydr. Publ. No.444)
33. Lagasse, P.F., J.D. Schall, F. Johnson, E.V. Richardson & F. Chang, Stream Stability at highway
structures, Hydraulic engineering circular no 20, Federal highway administration, U.S. Department of
Transportation
34. Lawler, D.M. (1993), The measurement of river bank erosion and lateral channel change: a review, Earth
Surface Processes and Landforms, BGRG, Vol.18, No.9, pp.799-821.
35. Manual on Rock in Hydraulic Engineering, report 169, CUR-RWS/DWW, Gouda, 2000
36. Mosselman, E. Huisink, M., Koomen, E. abd Seymonsbergen, A.C. (1993), Morphological changes in large
braided sand-bed river, Proc. 3rd Intern. Geomorphology Conf., Hamilton, Ontario (also Delft Hydraulics
Publ. 480)
37. Nghiên cứu dự báo phòng chống xói lở bờ sông Cửu Long, Trung tâm nghiên cứu chỉnh trị sông & phòng
chống thiên tai, Tp. Hồ Chí Minh, 2001
38. Osman, A.M. & Thorne, C.R. (1988), Riverbank stability analysis, I: Theory, Journ. Hydr. Engng., Vol. 114,
No. 2, pp. 134-150
39. Pilarczyk, K.W. (ed.) (1998), Dikes and Revetments, A.A. Balkema, Rotteram
40. Przedwojski, B., R. Blazejewski & K.W. Pilarczyk (1995), River Training Techniques, A.A.Balkema,
Rotterdam
41. Rivertraining and bank protection, Flood control series no. 4, United Nations, 1953
42. Sarker, M.H. (1996), The morphological processes in the Jamuna River, M.Sc. Thesis H.H.290, IHE, Delft.
43. Shishikura, T. (1996), Morphological changes due to river bank protection, IHE Delft/Delft Hydraulics,
M.Sc. thesis no. 285
44. The WES Stream Investigation ans Streambank Stabilization Handbook, U.S. Army Corps of Engineers,
Vicksburg, MS, 1997
45. Thorne, C.R. (1982), Processes and mechanisms of river bank erosion, In: Gravel-bed rivers: fluvial
processes, engineering and management, Eds. R.D. Hey, J.C. Bathurst & C.R. Thorne, Wiley, pp.227-271
46. Thorne, C.R. & S.R. Abt (1993), Analysis of riverbank instability due to toe scour and lateral erosion, Earth
Surface Processes and Landforms, BGRG, Vol.18, No.9, pp.835-843.
47. Thorne, C.R., S.R. Abt & S.T. Maynord (1993), Prediction of near-bank velocity and scour depth in
meander bends for design of riprap revetments, In: River, coastal and shoreline protection; Erosion control
using riprap and armourstone, Eds. C.R. Thorne, S.R. Abt, F.B.J. Barends, S.T. Maynord & K.W. Pilarczyk,
Wiley, pp.115-133
48. Thorne, C.R. & A.M. Osman (1988), Riverbank stability analysis. II: Applications, Journ. Hydr. Engrg.,
ASCE, Vol.114, No.2, pp.151-172
49. US Army (1981), Streambank Erosion Control Evaluation and Demonstration (Main Report), US Army
Corps of Engineers, Final Report to Congress, 1981.
50. Varma, C.V.J., K.R. Saxena & M.H.Rao (ed.) (1989), River behaviour management and training,
Publication no. 204, Vol I, Central board of irrigation and power, Malcha Marg. Chanakya Puri, New-
Delhi
51. Wang, Z.B., R.J. Fokkink, M. de Vries & A. Langerak (1995), Stability of river bifurcations in 1D
morphodynamic models, Journ. Hydr. Res., IAHR, Vol.33, No.6, pp.739-750