River Training and Flood Control: Chapter Five
River Training and Flood Control: Chapter Five
1
concerned with the most suitable alignment and height of marginal embankments for
disposal of floods and may also include other measures of channel improvement for the
same purpose. Thus high water training can also be called Training for Discharge.
Flood protection can be either passive or active.
Passive protection: Such as building on high grounds or on stilts above high water marks;
may be even today resorted to as a temporary or isolated protection measure.
Active protection: Allows the continuation of normal human activities in the flood plain
during the great part of the flood events.
Complete elimination of flood hazards and ensuing damage for any given region flood plain is
practically not feasible, due to the stochastic nature of flood events. It is very seldom, if ever,
envisaged from the engineering and economic point of view, since it would lead to
unacceptable financial outlays, far outweighing expected damages to property or crops. It
follows, therefore, that what characterizes any proposed flood-control project is the extent to
which flood damages are expected to be reduced, and not by any means their complete
disappearance.
It seems inevitable, therefore, that the problem of damages caused by exceptional flood events
will stay with us as long as flood plains remain an important area of human activity.
Engineering methods for flood control and protection are the following:
i. Stream training and regulation: Works concerning cross-section, alignment,
longitudinal slope & roughness of stream, with the scope of increasing its conveyance
capacity.
ii. Reduction of peak discharges: By means of flood routing through retention reservoirs
for temporary storage of floodwaters.
iii. Flood protection by dykes and levees.
iv. Attenuation of flood waves: Through diversion to other channels, or to less critical areas.
As flood protection works are actually designed to only reduce the frequency and extent of
expected inundation damages, the first question which must be answered when considering any
river training project is the determination of the design discharge.
Economic benefits from a proposed protection project over the expected useful life span of the
works should be equal to or greater than the compound cost of the project. In evaluation of the
former, it is generally an accepted practice that only tangible and direct benefits should be
2
drawn into consideration, i.e. those measurable in money equivalent on the one hand, and those
accrued directly as the result of the project, on the other hand. Overall cost of the proposed
project should include all expenditures required for its completion, operation and maintenance,
interest and depreciation.
A simple optimization procedure for a flood control project is schematically shown below.
3
adopted to alter the river cross section and alignment must obviously be designed in
accordance with that stage of the river at which the maximum movement of sediment
takes place during any period under consideration. Although there is maximum activity
of the bed of the river at high flow stages but such stages are maintained only for a
short duration. On the other hand, there is little movement of sediment at low stages,
which persist for a very long duration. In between the two there is a stage at which the
combined effect of forces causing sediment movement and the time for which these
forces are maintained is maximum. This is therefore the most important stage as it has a
considerable influence on the configuration of the river. This stage is somewhere near
the mean water.
4
deviates towards the channel to be eroded, while the bottom flow together with the bed load
passes under the mat.
5
Due to the scour under the bandalls, the screen has to be lowered at certain intervals. Due to
the concentration of the water in the channel, the velocities increase and the channel scours.
Bandalls can only be placed in limited depths, and are only effective within a small range of
water level variations. This is, therefore, not suitable for rivers with rapid water level variation.
During rising levels, the bandalls will have to be removed again.
Bottom panels: May be considered as an improvement on bandalls. The main principle of
bottom panels is to induce helical current, thus directing the bed flow and the bed load in a
direction differing from the main current. This principle can be used to direct the bed load from
the main channel to the sides and to close the secondary channels. They can be placed only
during low levels. They will be most effective at stages about twice the height of the panels.
The angle with the main flow should be about 45°.
Recurrent Dredging: For bend cutting, closing of secondary branches, dredging
crossings, etc. Many types of dredgers exist which include:
Bucket dredgers
Grab dredgers
Cutter dredgers
Hopper dredgers
Trailer (or suction) dredgers
Dustpan dredgers
Hopper dredgers are in general not suitable to work on rivers. Grab and bucket dredgers are
used only for special circumstances, e.g. rocky soil, etc. For river dredging normally the cutter
dredger, the dustpan dredger, and the trailer dredger are used.
The use of a cutter dredger is only effective in removal of layers with a thickness larger than
1.00 m. For shallower cuts the dustpan dredger and the trailer dredger are more suitable.
6
5.3.2 Permanent River Training Works: The types of channel regulation by
permanent river training works are:
Closure of secondary branches
Short-cut of bends
Groynes
Flow guiding structures
Embankments
Bank protection
Bed protection
Elimination of obstacles
The river training structures are either in the flow direction or parallel to it, or they are
perpendicular (or at some angle) to the flow direction.
i) Closure of secondary branches: In case of complete closure of a secondary
branch, after a new equilibrium has been reached, all water and sediment will follow the
remaining branch. When the final equilibrium in the new channel has been reached, a new
slope, width and depth will be present.
ii) Short-cut of bends (or artificial cut-off)
When a river contains sharp bends, bend cutting may be required for several reasons.
1. to improve the situation for navigation – e.g. reduce the length
2. to improve flood discharge capacity – It may also be used to direct the river from
the curved flow which may be endangering valuable land and property
3. to stop severe bank erosion
- to straighten out the approach of the river for some structures to be constructed
just below the bend
Bend cutting is executed by dredging a new channel along a much shorter but stable alignment.
To get a stable channel, a gentle bend should be made.
7
Morphological consequences of short-cutting are:
(a) Energy gradient of the stream along the cut-off channel and upstream of it is increased;
hence also its sediment transport capacity.
i. From the mild slope upstream of A to the steeper slope of the cut-off;
ii. From the steeper slope of the cut-off to the mild slope downstream of B.
Generally, erosion starts first not far from point A and deposition will take place in the vicinity
of point B. With time, the erosion moves upstream from point A (back erosion), and
sedimentation advances downstream from point B. The back erosion and progressing
deposition are carried on until eventually a new longitudinal slope is established, more or less
similar to the original slope.
(b) Because of reduced stream storage, peak discharge downstream of the cut-off is likely to
be higher than before. The streambed upstream of point A is lowered with time, while
downstream from point B it will be raised above the original streambed before the
construction of the cut-off. This reduction in channel capacity in the downstream part may
cause flooding at higher discharges, since the water can no longer be contained within the
stream channel. Therefore, in most cases, meander short-cutting alone is not sufficient to
prevent the stream from overflowing its banks during the flood protection design
discharge; hence additional means are necessary, such as channel improvement or dykes.
8
iii) Marginal Embankments (Dykes and Levees)
A levee or dyke is a structure mainly for flood protection by controlling the river and not by
training it. The alignment should follow the normal pattern of meandering of the river. They
are constructed of earth materials and may be provided at one or both sides of the river. The
design of dykes or levees is just like embankment dams.
Like embankment dams they are likely to fail due to overtopping, piping, seepage, etc. They
are designed to hold water up to the maximum anticipated H.F.L. without the possibility of
overtopping and withstanding all external pressures. Therefore, the necessary conditions are
met by providing sufficient freeboard, bed width, top width and stone protection on slopes. As
the height increases it becomes necessary to provide key trenches, zoned sections, etc. to make
the embankment stable. Freeboard may be between 0.3 m and 1.5 m above H.F.L.
9
i. Economics: concerning cost of dykes – low dykes are cheaper to build, not only
because of minor volume of earthwork but due to other construction considerations –
e.g. protection against piping; clearing foundation area from vegetation, roots, boulders,
or organic matter; compaction in layers; etc. Dykes are usually built along extended
stretches of the stream; hence cost of dyking scheme is sensitive to additional height.
ii. Safety: Failure with low dykes will result in only minor damages, while with
high dykes, both the material damage and human suffering are likely to be more severe.
10
Design Criteria for Guide Banks
i. Length of Clear Waterway: to be provided between the guide banks or the
abutments of the work is given by:
The length, L, of the overall waterway (or length of the work or the structure) between the
guide banks or the abutments of the work is obtained by adding the thickness of the piers (in
case of bridges) to P. Generally, L≈ 1.1 to 1.25 P.
11
Downstream curved heads – sweep angle of 45° to 60° and half radius of upstream curved
head; R1 = ½ R
(See also the sketch)
iv. Cross section of Guide Banks:
- Top width not less than 3 m
- Constructed of locally available material, usually sand (earthen, soil)
- Side slopes not steeper than 2:1 (H:V)
- Free board of 1.25 to 1.5 m above anticipated flood level
12
Depth of scour below high flood level (H.F.L.) = K Rs
Where
In launching, since the apron will not form a uniform stone carpet as by hand packing, the
thickness of the apron in launched position is assumed to be 1.25T. Thus, the quantity of stone
required per meter length of the launching apron will be
(√5) D x 1.25T = 2.80 TD
Width of launching apron = 1.5D (usually).
Figure 5.8 Launching apron
13
Qmax = 6000 m³/s
H.F.L. = 104 m
River bed level = 100 m
Average diameter of bed material = 0.10 m
Design and sketch a guide bund including the launching apron to train the river.
14
15
v) Groynes or Spurs: are small jetties (jetties = structure built out into a water body
as a breakwater or landing place for boats, etc), solid or permeable, constructed of timber,
sheet piling, vegetation, and stone rubble, etc. They usually project into the stream
perpendicularly to the bank, but sometimes are inclined in the upstream or downstream
direction. The main purpose of groynes is to reduce channel width and to remove the damage
of scour from the banks.
Functions of Groynes:
Training the river along a desired course by attracting, deflecting, or repelling the flow in
the river;
Creating a slack flow with the objective of silting up the area in the vicinity;
Protecting the river bank by keeping the flow away from it;
Contracting a wide river channel, usually for the improvement of depth for navigation.
Groynes can be classified according to methods and materials of construction into:
16
i. Impermeable groynes
ii. Permeable groynes
i) Impermeable groynes (solid groynes): do not permeate appreciable flow of
water through them. They consist of rockfill, sand and gravel, or soil as available in the river
bed, protected on the top and sides by strong stone facing (pitching) or concrete blocks. Slopes
vary between 2:1 to 3:1 depending on the material used. Since the head of the groyne is
subjected to severe attack by the stream, thicker stone pitching and launching apron is
provided. Since the head of the groyne is subjected to severe attack by the stream, it needs
protection. Hence at the groyne head a launching apron is provided and also the thickness of
the pitching is increased.
17
Figure 5.10 Normal groyne (or deflecting groyne)
Groynes pointing downstream have the property of attracting the flow towards them, and are
called attracting groynes.
(a) (b)
Figure 5.12 Plan and cross section of a groyne (typical groyne structure)
The groynes are, therefore, generally aligned either perpendicular to the bank or pointing
upstream. The perpendicular alignment is generally used on convex banks, and the upstream
pointing alignment is generally used on concave banks.
When the length of an upstream pointing groyne is small, such that it changes only the
direction of flow, without repelling it, it is called deflecting groyne.
More extensive studies of local scour around groynes have been carried out by A.M. Gill. The
results of his laboratory investigations appear to indicate the following:
18
(a) Scour depth depends on the depth of flow, and it grows with the increase in depth,
(b) Scour depth depends on the bed material size. For the same value of the ratio τ c/τ0: τ0
= γRI, coarse sand will be scoured deeper than fine sand; on the other hand, for the same
absolute value of the shear stress, fine sand will be eroded deeper than the coarse sand
(because τc for coarse sand is higher than for fine sand).
(c) Bed load movement does not appreciably affect the scour depth. Once the movement of
alluvial bed is started, maximum depth of scour tends to remain constant for a given
depth of flow.
For design purposes, Gill has proposed an empirical formula (the coefficient has been rounded
up):
For 0< L/d < 25, where L = length of groynes, and d = undisturbed water depth; formula
proposed by Liu is used:
where S = scour hole depth measured below the mean bed level
Fr = Froude number of undisturbed flow
For L/d > 25, empirical formula obtained from field observations on rock dykes is applied:
Groynes may be constructed either singly or in series, depending upon the need. When
constructed in series, they are more effective as they create a pool of almost still water between
them, which resist the current and gradually accumulate silt between them, thus forming
19
almost a permanent bank after a certain time. The choice of using them in series arises, if the
reach to be protected is long, or if a single groyne is neither strong enough to deflect the
current nor quite effective for silt deposition upstream and downstream of itself.
Design considerations
1. Length of groynes: Length depends on the position of the original bank line and the
designed normal line of the trained river channel. Too long groynes on easily erodible rivers
are susceptible to damage and failure. In such cases, groynes of shorter length may be provided
and then they may be extended gradually as silting between them proceeds.
2. Spacing of groynes: Since the length of the bank to be protected by each groyne
depends on the length of the groyne, the spacing depends on their length. It is, therefore, taken
as a certain proportion of their lengths. Other factors affecting spacing are:
i. Width of the river: For rivers of equal flood discharges, a larger ratio of spacing
to length of groynes may be used for Wide River than for a narrow one.
ii. Location of groynes: Large spacing may be used for convex banks than for
concave banks (e.g. on concave banks, spacing = length of groynes; on convex banks,
spacing = 2 to 2.5 times length of groynes).
iii. Type of construction (or type of groyne): Permeable groynes may be spaced
farther apart than solid or impermeable ones. Generally, empirical rule of thumb
specifies spacing as
- One to two times channel width, or,
- 1 to 5 times the groyne length
According to laboratory tests carried out in the Delft Hydraulic Laboratory, there appears to
exist a semi-empirical dependency between the spacing L between the groynes and a
theoretically derived parameter. Results of this test seem to indicate that the best flow guiding
by groynes is obtained when only one strong eddy is formed between each pair.
20
Figure 5.13 Groyne spacing and flow field between groynes
The reasoning follows that the energy required to sustain the backward flow between the
groynes can only be available if the energy loss in the stream, IL, is smaller than average
velocity head (this is possible only if the level of the water surface in the stagnation point of
the downstream groyne is higher than the water level at point A of upstream groyne), v2/2g.
Writing energy equation between the head points of two adjacent groynes and substituting
Manning’s/ Chezy´s equation to express the mean velocity, it can be easily shown that
; Manning’s equation
; Chezy’s equation
Example:
(a) For d = 4m, n = 0.035
The spacing between groynes should be about
21
(c) For d = 5 m and C = 40 m1/2/s
The spacing between the groynes should be about:
In practice, however, the distance L (or spacing) would be taken somewhat less in order to be
on the safe side.
22
Figure 5.14 Slope reduction by drops
23
b) Bottom Sills
Before the erosion of the original bed between the sills takes place, there is no influence of the
sills on the flow. Eventually such erosion will go on until a new equilibrium slope is
established, milder than the original one. At that stage, the system will form a cascade of small
drops. Bottom sills should always be built strong enough to act as low retaining walls after the
erosion has taken place.
b) Riprap Revetment
Riprap is a layer or facing of rock, dumped or hand-placed to prevent erosion, scour, or
sloughing of a structure or embankment. Materials other than rock are also referred to as
24
riprap; for example, rubble, broken concrete slabs, and preformed concrete shapes (slabs,
blocks, rectangular prisms, etc.).
In this context, riprap is defined as:
"A flexible channel or bank lining or facing consisting of a well graded mixture of rock,
broken concrete, or other material, usually dumped or hand-placed, which provides protection
from erosion."
As described above, riprap is a flexible revetment. Flexibility of the riprap mass is due to
individual particles acting independently within the mass.
Rock riprap is the most widely used and most desirable type of revetment. It is compatible with
most environmental settings. The term "riprap" is most often used to refer to rock riprap. Rock
riprap is further subdivided by placement method into dumped riprap, hand-placed riprap, and
plated or keyed riprap.
Riprap is composed of three sections: the armor or stone layer, the filter layer, and the toe
protection. Typical armor is composed of rough, angular rock. The second component, the
underlying filter layer, supports the stone against settlement, allows groundwater to drain
through the structure, and prevents the soil beneath from being washed through the armor layer
by waves or groundwater seepage.
Dumped riprap is graded stone dumped on a prepared slope in such a manner that
segregation will not take place. Dumped riprap forms a layer of loose stone; individual stones
can independently adjust to shifts in or movement of the base material.
Advantages associated with the use of dumped rock riprap include:
The riprap blanket is flexible and is not impaired or weakened by minor movement of the
bank caused by settlement or other minor adjustments.
Local damage or loss can be repaired by placement of more rock.
Construction is not complicated.
When exposed to fresh water, vegetation will often grow through the rocks, adding
esthetic and structural value to the bank material and restoring natural roughness.
Riprap is recoverable and may be stockpiled for future use.
Hand-placed riprap is stone laid carefully by hand or by derrick following a definite
pattern, with the voids between the larger stones filled with smaller stones and the surface kept
25
relatively even. The need for interlocking stone in a hand-placed revetment requires that the
stone be relatively uniform in size and shape (square or rectangular).
Advantages associated with the use of hand-placed riprap include:
The even interlocking surface produces a neat appearance and reduces flow turbulence at
the water revetment interface.
The support provided by the interlocking of individual stones permits the use of hand-
placed riprap revetments on steeper bank slopes than is possible with the same size loose
stone riprap.
With hand-placed riprap, the blanket thickness can usually be reduced to 150 to 300 mm
less than a loose riprap blanket, resulting in the use of less stone.
Disadvantages associated with hand-placed riprap include:
Installation is very labor-intensive, resulting in high costs.
The interlocking of individual rocks in hand-placed revetments results in a less flexible
revetment; as mentioned above, a small shift in the base material of the bank can cause
failure of large segments of the revetment.
By their nature, hand-placed rock riprap revetments are more expensive to repair than are
loose rock revetments.
26