1.

Introduction

Sedimentation engineering embraces the

identification, planning, analysis, and
613 remediation, principally in the context of civil
River Engineering and Sediment and hydraulic engineering practice, of projects or
Transport technical investigations to avoid and/or mitigate
problems caused by sedimentation processes.

These processes include erosion, entrainment,

transport, deposition, and compaction of
sediment.
1 2

Introduction Introduction
Sediment is any particulate matter that can be This course is concerned with sedimentation
transported by fluid flow and which eventually is problems involved in the development, use,
deposited as a layer of solid particles on the bed control and conservation of water and land
or bottom of a body of water. resources.

The overall goal is the economic management of Attainment of such goals would minimize the
our land resource so that the rate of normal engineering and aesthetic problems incurred by
geologic erosion, plus an allowable rate of man- sediment movement downstream.
made erosion will not exceed our ability to
sustain the soil.
3 4

Course objectives Introduction

Understand nature and scope of sedimentation
Sedimentation processes are natural processes
problems
that have been active throughout geological times
Understand fundamental aspects of fluid and have shaped the present landscape
dynamics that relate to sediment transport
Human activities also affect sedimentation
Understand the mechanics of sediment processes
entrainment, transport, and deposition External agents and forces driving these processes
Gain practice solving sediment-transport may include water, wind, gravity, and ice
equations Although each may be important locally, only the
Understand practical approaches to their hydrospheric forces of rainfall, runoff, and
solution streamflow are considered herein
5 6

1
 Introduction Introduction

When the transporting agent is water, it is called Transport depends on the shape, size, and weight
fluvial or marine sediment transport of the particle and the forces exerted on the
Aeolian sediment is that moved or deposited by particle by the flow
wind When these forces are diminished to the extent
Erosion will be defined as the detachment and that the transport rate is reduced or transport is
removal of rock particles by the action of water no longer possible, deposition occurs
Detachment of sediment particles in the erosion Sedimentation is of vital concern in the
process occurs through the kinetic energy of conservation, development, and utilization of our
raindrop impact, or by the force generated by soil and water resources
flowing water 7 8

Global aspects and changing roles Global aspects and changing roles

Awareness of sedimentation processes and the This type of comprehensive systems approach is
consequences of poor sediment-management also demanded by more stringent legal and
practices has increased among water resources regulatory requirements regarding sediment and
professionals and civil engineers. hydraulic processes in water bodies.

It has increasingly been realized that a Factors that have resulted in increased public
multidisciplinary approach to problem awareness and greater potential impacts to water
identification, quantification, and management is resources and the environment include the
often required to deal with the interrelated following:
effects of geomorphologic, environmental, and Growing global populations place increasing
engineering issues. pressures on land and water resources.
9 10

Global aspects and changing roles Global aspects and changing roles

Human settlements have increasingly occupied Considerable knowledge relative to the nature
areas more vulnerable to erosion and and influences of the forces involved in
sedimentation. sedimentation has been accumulated during the
past few decades, out of which has come:
One of the most significant factors affecting global
management and delivery of sediment has been realization that man, through his activities, can
the construction of dams on rivers. locally influence the so called “natural
sedimentation processes”
Scientific experts and governments worldwide
acknowledge strong scientific evidence better understanding of the fundamentals
demonstrating that human activities are changing involved in the creation or alleviation of
the Earth’s climate sedimentation problems
11 12

2
 Global aspects and changing roles Nature of sedimentation problems

This information is basic to the proper planning, Problems of erosion

design, installation, and maintenance of water
works of improvement for the development, use Geologic Erosion
and conservation of our soil and water resources
to ensure that Defined as erosion of the surface of the earth
under natural or undisturbed conditions
they will function as designed Results from tectonic uplift, earthquakes,
weathering, and chemical decomposition and the
will not create new problems, or accentuate long-term action of water, wind, gravity, and ice
existing ones
13 14

Problems of erosion Problems of erosion

Vast extent and great thickness of sedimentary Control of geologic erosion, by and large, is
rocks found on the earth’s surface provide often difficult to achieve because:
evidence of geologic erosion and deposition that
has occurred throughout geologic times natural conditions that have prevailed over
centuries of time cannot be changed
Varies in different places on the earth’s surface significantly to effect any great reductions in
because of differences in character of rocks and erosion
climatic and vegetative conditions
economic feasibility – the land on which
serious geologic erosion often occurs is
usually of lower agricultural productivity
15 16

Problems of erosion Problems of erosion

Justification for structural measures, therefore Accelerated Erosion

depends mainly on the reduction of severe Defined as the increased rate of erosion over the
sediment damages immediately downstream normal or geologic erosion brought by man’s
from the source activity
Erodibility of natural materials can be altered by:
disturbing the soil structure through plowing
or other tillage activities
changing the protective vegetative canopy by
grubbing, cutting, or burning existing
vegetation and introducing new species
17 18

3
 Problems of erosion Problems of erosion

altering the natural characteristics of stream Opening of new lands for agricultural purposes
channels, e.g., channel cross sections, necessarily disturbs the natural conditions
alignments, or gradients
indigenous forests are removed by ax and fire
Agricultural Activities
native grasslands are burned, overgrazed or
Widespread use of land and associated practices the sod turned and broken by plow
for agricultural purposes leads all other activities
resulting in accelerated erosion and the The rate of loss of productivity of farm fields
production of sediment so far as the world as a depends on the depth of erosion of plant-
whole is concerned producing topsoil
19 20

Problems of erosion Problems of erosion

Urbanization
Holding accelerated erosion of farmland to an
acceptable maximum needs When fully developed, are actually low sediment-
producing areas, because a large percentage of
continued efforts in education in respect to the land is protected against erosion by roofs,
the desirability of proper land use and streets, parking lots, etc
conservation
Erosion rates are high during actual construction
provisions for technical assistance in the
as a result of the removal of trees and other
planning and application of measures
vegetations, and excavation and grading activities

21 22

Problems of erosion Problems of erosion

Road and Highway Construction Altering Runoff Conditions

Serious erosion can occur when protective Erosion energy is a function of the volume of
vegetation is removed and steeply sloping cuts water and its velocity
and fills are left unprotected
When the volume of runoff water is increased,
Such erosion can create local problems of serious or is concentrated in natural or artificial
downstream sediment damages channels, its erosive energy is increased
Attention must be given to stabilize exposed
Many human activities as applied to the land
surfaces so that they do not become serious long-
generally either increase the amount of runoff
term sediment producing areas
water or concentrate the flow
23 24

4
 Problems of erosion Problems of erosion

clearing of land increases the percentage of Channel straightening, which increases the
runoff channel gradient and flow velocity, may initiate
channel erosion
water-collecting and disposal systems
concentrate the volume of flow Constricting the cross section of channels
Stream and River Control Works usually increases velocity and depth of flow and
the eroding and transport power of streams
Any structural work of improvement that changes
the direction of flow or increases the depth, Lowering of the bed of a main stream results in
duration, and velocity of flow may result in lowering the bed of tributary streams
erosion
25 26

Problems of erosion Problems of erosion

Degradation of channels may seriously Both the sediment load and flow conditions
depreciate adjacent land areas by lowering the were responsible for establishing the natural
ground-water tables needed to maintain regime of the channel prior to construction of
productivity capacity the dam
Clear water released from a reservoir
Construction of a dam influences downstream immediately picks up a new load downstream if
channel stability in two ways: the discharge is sufficient to erode the bed and
it traps the sediment load transport the sediment
Seriousness of the problem of degradation of
it changes the downstream natural flow channels below dams depends on the
characteristics erodibility of downstream channel materials
27 28

Problems of erosion Problems of erosion

Water Quality chemicals and wastes are assimilated onto and
into sediment particles
Sediment is not only the major water pollutant
by weight and volume but it also serves as a ion exchange occurs between solutes and
catalyst, carrier, and storage agent of other forms sediments
of pollution
Sediment has become a source of increased
sediment alone degrades water specially for concern as a carrier and storage agent of
municipal supply, recreation, industrial pesticide residue, adsorbed phosphorus,
consumption and cooling, hydroelectric nitrogen and other organic compounds, and
facilities, and aquatic life pathogenic bacteria and viruses
29 30

5
 Problems of sediment transport Problems of sediment transport

Impingement of Sediment Particles Paved flood channels that carry coarse sand
and gravel are subject to wear and need
Damage caused by the striking of objects by periodic repairs
sediment particles in transport is relatively rare,
but it does occur in isolated cases Torrential flood flows from mountain streams
may set huge boulders in rolling and bouncing
Bridge abutments and trestle piers on torrential motion down alluvial fans causing severe
streams transporting cobbles may be severely damage
damaged by boulders chipping away at wood or
concrete unless protection is afforded
31 32

Problems of sediment transport Problems of sediment transport

Main damage by transport of smaller particles Sediment in Suspension

is to turbines and pumps where sediment-laden
water causing excessive wear on runners, Effect of suspended sediment on the size,
vanes, propellers, and other accessory parts population, and species of fish in a stream

affects the light penetration in water and thus

Frequent replacement of worn parts may be reduces the growth of microscopic organisms
necessary to maintain efficient operation of on which insects and fish feed
hydroelectric generation equipment
injures the gills and breathing structure of
certain types of fish
33 34

Problems of sediment deposition Problems of sediment deposition

Deposition is the counterpart of erosion Deposits at Base of Eroding Slopes

Products of erosion may be deposited Usually heterogeneous deposits of rock and soil
immediately below their sources, or may be that have not been transported through any well-
transported considerable distances to be defined channel system
deposited in channels, on flood plains, or in lakes,
reservoirs, estuaries, and oceans Found mostly at the breaks of eroding slopes

All sediment deposition is not injurious. Some When occurred in farm fields, may completely
deposits may form fertile flood-plain or delta bury crops and fence lines
soils
35 36

6
 Problems of sediment deposition Problems of sediment deposition

May seriously damage high-cost property when, Flood-Plain Deposits

e.g., they cover highways or in urban areas,
streets, lawns, and gardens In the process of flood-plain deposition, natural
levees of coarser materials form adjacent to the
Sometimes cars, buildings, and other property are channels while the fine-grained materials
partially buried transported and deposited in areas behind the
levees

levees impair the surface drainage of flood

waters back to the channel

37 38

Problems of sediment deposition Problems of sediment deposition

fine-grained deposits reduce the permeability net result is the swamping of flood-plains
of flood-plain lands
Deposition of sediment on flood plains in urban
Deposits that occur on flood plains create areas can create serious local deposition on
numerous types of damages to crops and streets, highways, railroads, and in building and
developments other developments

may completely bury crops high maintenance and cleanup costs from
overbank deposition, year after year, often
infertile materials reduce the fertility and result in high damages
long-term productivity of flood plain soils
39 40

Problems of sediment deposition Problems of sediment deposition

Channel Deposits sediment deposited in navigable channels,

Deposition of sediment in drainage ditches, waterways, and harbors must be removed
irrigation canals, and in navigation and natural periodically to maintain required depths
stream channels create serious problem in loss of deposition of sediment in natural stream
services and cleanout costs channels greatly aggravates floodwater
sediment deposits in ditches result in the damages
raising of water tables and damage to crops
decreases the channel capacity and flood-
sediment in irrigation ditches reduces the rate carrying capacity
and volume of water delivered to irrigated
areas
41 42

7
 Problems of sediment deposition Problems of sediment deposition

clogging of diversions and turnouts with Deposits in Lakes and Reservoirs

sediment also reduce the amount of water
available in time of need When streamflow enters a natural lake or
reservoir, its velocity and transport capacity is
flushing of coarse-grained sediment onto reduced and its sediment load is deposited
irrigated areas reduces soil fertility, while
in artificial lakes with outlets, e.g., reservoirs,
colloidal sediment reduces permeability, both
the amount deposited depends on the
of which result in reduced productivity of
detention storage time, the shape of the
irrigated lands
reservoir, operating procedure, and other
factors
43 44

Problems of sediment deposition Problems of sediment deposition

in natural lakes that have no outlets the total spawning areas for certain species of fish are
incoming sediment load is deposited destroyed
In most natural lakes, the total rate of filling with operation of boats in such areas becomes
sediment is generally of less concern than the difficult
location of deposits
In artificial reservoirs, both the location of
Incoming fine-grained sediment, derived deposits and the loss of storage capacity are of
particularly from prolonged sheet erosion, is concern
inherently fertile so that accumulations not
only shoal specific areas but also provide a shoaling areas, particularly by deltaic deposits
fertile bottom encouraging prolific plant growth may seriously affect adjacent property values
45 46

Problems of sediment deposition Problems of sediment deposition

convert attractive beach areas into loss of storage capacity is usually a more
undesirable mud flats important factor than the location of deposits,
because the loss of carryover storage effects
damage fishing and boating activities the proper functioning of these reservoirs

Primary purpose of most reservoirs is to store reservoir evaporation losses may be increased
water for flood prevention, water supply, by changes in the area-capacity relation in a
irrigation, power reservoir caused by sediment accumulation

47 48

8
 Problems of sediment deposition Solutions of sedimentation problem

such turbid currents can also affect the Many problems arising from sediment transport
temperature and quality of the water released and deposition would be solved by stopping
to the extent that fish life may be damaged and upstream erosion completely
recreational use endangered Stopping erosion completely is not always
density or turbidity currents that carry fine physically possible, or economically feasible in the
sediments to the deepest parts of the reservoir solution of specific downstream sediment
can be generated problems

depending on the size, purpose and operation Solution of one problem in this manner can often
of the reservoir, can create excessively turbid lead to the creation of others, and in such
water for irrigation, municipal, or industrial use instances, alternatives must be considered
49 50

Solutions of sedimentation problem Solutions of sedimentation problem

Often various methods of reducing specific dredging

sediment damages must be considered in project
formulation. These include sediment bypassing,
etc.
erosion control of primary upstream sediment-
source areas Determination of the measure, or measures, to
use in a specific case and the evaluation of the
trapping sediment in sedimentation basins effectiveness of the measure, or combinations of
measures thereof, requires a through
allowance for storage of sediment in structure
understanding of the fundamental sedimentation
design
processes and principles involved
51 52

Basic considerations in solutions of Basic considerations in solutions of

sedimentation problem sedimentation problem
channel stability problem might be solely
Fully recognize and understand the nature of a caused by gravel-size particles
sediment problem and the part that fundamental
problem of drainage impairment might be
sedimentation processes play in creating the
caused by only the fine-size particles
problem
total sediment load may be of concern in the
determine whether the problem is caused by loss of storage capacity in the reservoir
the total amount of sediment delivered to the Once the nature of damage is determined and the
problem area or only some part of it, e,g., that characteristics (usually the particle size) of critical
contained within certain limitations of grain sediment are established, the source of this
size sediment must be determined
53 54

9
Basic considerations in solutions of Basic considerations in solutions of
sedimentation problem sedimentation problem
It is in this realm of sedimentology that the  hydraulic characteristics of flow transporting
specialist must draw heavily on his knowledge the sediment and influencing deposition
of the processes and principles of
sedimentation Determination of the type, or types, of measures
needed for the alleviation of damages follows the
Amount and character of sediment delivered to delineation of the source of damaging sediment
any point in a stream system is dependent on
It is usually necessary to consider and compare
 nature of materials being eroded alternative methods to arrive at measures that
are both economically possible and physically
 type of erosion
feasible
55 56

Approach to sedimentation problem 1.2. Properties of Sediments

Sedimentation processes (erosion, entrainment,
1. Identification of the problem (wider context) transport, and subsequent deposition of sediment)
2. Formulation of general objectives and desired depend on
state of knowledge properties of the sediment
3. Determination of problem dimensions and characteristics of the flow involved
analysis of physical system Those properties of most important in the
4. Formulation of hypotheses related to the problem sedimentation processes can be divided into
properties of
5. Generation of alternative solutions and cost
particles
estimates
sediment as a whole
6. Selection of optimum solution
57 58

Size and Shape of Sediment Particles Size and Shape of Sediment Particles
Size and shape of grains making up a sediment vary Sediment particles are classified, based on their size,
over a wide range into six general categories: Clay, Silt, Sand, Gravel,
Cobbles, and Boulders
meaningless to consider in detail the properties of
an individual particle Any single length or diameter that is to characterize
necessary to determine average or statistical values the size of a group of grains must be chosen either
arbitrarily or according to some convenient method of
Sediments are grouped into different size classes or measurement
grades
Natural sediment particles are of irregular shape

59 60

10
 Sediment Grade Scale Size and Shape of Sediment Particles
Sieve diameter is the length of the side of a square
sieve opening through which the given particle will
just pass
sieving is convenient to determine the size of sands
Sedimentation diameter is the diameter of a sphere of
the same specific weight and the same terminal
settling velocity as the given particle in the same
sedimentation fluid
size of silts and clay is generally expressed as
sedimentation diameter
61 62

Size and Shape of Sediment Particles Size and Shape of Sediment Particles
Nominal diameter is the diameter of a sphere of the Sphericity has been defined as the ratio of the
same volume as the given particle surface area of the sphere with the same volume as
the grain to the surface area of the particle
Those characteristics that seem most important to
engineers concerned with sediment transport are Roundness
shape and roundness depends on the sharpness or radius of curvature of
the edges
Shape
defined as the ratio of the average radius of curvature
describes the form of the particle without reference
of individual edges to the radius of the largest circle
to the sharpness of its edges
that can be inscribed within either the projected area
has been expressed in terms of true sphericity or a cross section of the grain
63 64

Shape of Sediment Particles Specific Weight of Sediment Particles

All sediments have their origin in rock material, and all
In studying the fall velocity, the shapes of the particles
constituents of the parent material can usually be
have been expressed by shape factor, , given by
found in the sediment

= As the materials become finer due to weathering and

abrasion, the less stable minerals tend to weather
faster and be carried away as fine particles or in
in which solution, leaving behind more stable components
= length of the longest perpendicular axis
Although quartz, because of its great stability, is by far
= length of the intermediate perpendicular axis the commonest mineral found in sediments moved by
= length of the shortest perpendicular axis water and wind, numerous other minerals also are
present
65 66

11
 Specific Weight of Sediment Particles Fall (Settling) Velocity of Particles
Directly characterize its reaction to flow
Although other materials besides quartz may be
present in appreciable quantities, the average specific Important for processes such as sedimentation and
gravity of sand is very close to that of quartz, i.e., 2.65, suspension
and this value is used often in calculations and analysis A constant velocity of a vertically falling particle in still
water
Reflects the integrated result of size, shape, surface
roughness, specific gravity and viscosity of the fluid
Its magnitude reflects a balance between the
downward acting force due to the submerged particle
weight and opposing forces due to viscous fluid
resistance and inertia effects (drag forces)
67 68

Fall Velocity of Particles Fall Velocity of Spheres

When Reynolds number = / is less than 0.1 − −1
for small particles in the silt-clay range viscous = =
resistance dominates and inertia is negligible 18 18

For particles coarser than 2 mm encounter resistance in which

from the inertia of the water as they fall, and viscosity = kinematic viscosity of the fluid
is unimportant
= specific weight of the fluid
For a sphere of diameter , the fall velocity, , for = specific weight of the sphere
values of Reynolds number = / less than = acceleration of gravity
approx 0.1 is given by Stokes law = specific gravity

69 70

Fall Velocity of Spheres Fall velocity of quartz spheres in air and water
Fall velocity over the entire range of Reynolds numbers,
in terms of the drag coefficient, , is given by
4 −
=
3

Drag coefficient in the Stokes range ( < 0.1) is given

by
24
=

For larger Reynolds numbers is still a function of

but it has been determined experimentally
71 72

12
 Fall Velocity of Non-Spherical Particles Fall Velocity of Non-Spherical Particles
Shape effect is largest for relatively large particles 10 0.01( − 1)
(> 300 ) which deviate more from a sphere than a = 1+ −1
small particle

Fall velocity of non-spherical sediment particles can be for 100 < ≤ 1000
determined from the following formulae:
= 1.1 ( − 1) for ≥ 1000
( − 1) in which
=
18 = sieve diameter
for 1 < ≤ 100 = specific gravity (= 2.65)
= kinematic viscosity coefficient
73 74

Fall velocities for d < 100 µm according to Stokes Fall velocities for particle sizes larger than 100 µm
according to US Inter-Agency Committee (1957)

75 76

Effect of Sediment Concentration on Fall Velocity Effect of Sediment Concentration on Fall Velocity
Fall velocity of a single particle is modified by the hindered settling is largely caused by the fluid
presence of other particles due to the mutual return flow induced by the settling velocities
interference of the particles
According to Richardson and Zaki, the fall velocity in a
If only a few closely spaced particles are in a fluid, fluid-sediment suspension can be determined as
they will fall in a group with a velocity that is higher
, = (1 − )
than that of a particle falling alone
in which
If particles are dispersed throughout the fluid, the
interference between neighbouring particles will ws,m = particle fall velocity in a suspension
tend to reduce their fall velocity – referred to as ws = particle fall velocity in a clear fluid
hindered settling c = volumetric sediment concentration
77 78

13
Effect of Sediment Concentration on Fall Velocity Influence of Sediment Concentration on Fall
Velocity (R small)
< 0.2 0.2 − 1 1 − 200 > 200
4.65 . . 2.39
4.35 4.45

= 4 for particles in the range of 50 to 500 )

Oliver formula read as

= (1 − 1.25 ) 1 − 0.75 .
,

which yields good results over the full range of

concentrations
79 80

Effect of Turbulence on Fall Velocity Effect of Turbulence on Fall Velocity

Spherical particles would settle more slowly in a fluid Presence of bursting processes characterized by lift-
oscillating in the vertical direction than in one at rest up of low-momentum fluid (bursts) and a down-rush
of high-momentum fluid to the bed (sweeps)
Reduction in fall velocity resulted from the nonlinear
relation between drag on the particle and their Asymmetric fluid motion in vertical direction with
velocity relative to the fluid relatively high (short duration) downward velocities
may result in a slight increase of the fall velocity
Another mechanism may be intensive eddy
production close to the bed inducing vertically
upward motions which may reduce the fall velocity
until the eddies dissolve at higher levels

81 82

Angle of Repose Porosity

Grains pile up on each other have an equilibrium
slope which is called the angle of natural repose ( ) The porosity ( ) is a measure of the volume of void
(∀ ) to total volume (∀ = ∀ + ∀ )
appears to be a function of size, shape and
porosity The volume of solid particle is:
increases with decreasing roundness
∀ = ∀ = (1 − )∀
sand sizes from 0.001 to 0.01 m show values in
the range of 30o to 40o Thus,
Referred to as the angle of internal friction is related ∀
to the particle stability on a horizontal or slopping bed = =
∀ 1+
Usually determined from the initiation of motion
experiment
83 84

14
 Void Ratio Size-Frequency Distribution
Because natural sediments are made up of grains with
The void ratio is a measure of the volume of void wide ranges of size, shape, and other characteristics, it
(∀ ) per volume of solid (∀ ) is natural to resort to statistical methods to describe
these characteristics
∀ Process of obtaining size distribution by separation of
= =
∀ 1− a sample into a number of size classes is known as
mechanical analysis
Results of such analyses of sediment are usually
presented as cumulative size-frequency curves, where
the fraction or percentage by weight of a sediment
that is smaller or larger than a given size is plotted
against the size
85 86

Size-Frequency Distribution Size-Frequency Distribution

Frequency distribution is characterized by

median particle size 50 which is the size at

which 50% by weight is finer or coarser

mean particle size = ∑( )/100 with =

percentage by weight of each grain size fraction
Normal size-Frequency distribution
curve
standard deviation =∑ − /100 or
= 0.5( ⁄ + ⁄ ) which is a
Cumulative frequency of normal measure based on graphic values
distribution i.e % finer-than curve
87 88

Size-Frequency Distribution Size-Frequency Distribution

geometric mean = . . , in which . log

and . are the grain sizes for which 84.1% and Skewness =
log
15.9% by weight, respectively, of the sediment is
finer
log
geometric standard deviation = . ⁄ .
2nd skewness =
log

log
Kurtosis =
log
89 90

15
 1.3. Characteristic Parameters Particle Mobility Parameter,
Particle Diameter, D* Plane bed
Reflects the influence of gravity, density and viscosity is the ratio of the hydrodynamic fluid (drag and lift)
and read as: forces and the submerged particle weight
⁄ Fluid force is proportional to and the
−1 ∗
∗ = submerged particle weight is proportional to
− , yielding a ratio of:
in which:
∗ ℎ
= median particle diameter of bed material = = =
−1 − ( − 1)
= specific gravity (= ρs/ρ)
= kinematic viscosity coefficient in which:
= acceleration of gravity = overall time-averaged bed-shear stress
91 92

Particle Mobility Parameter, Excess Bed-Shear Stress Parameter, T

∗ = overall bed shear velocity (τb = ρ u*2) The excess bed-shear stress parameter, , is defined as:
= specific gravity (= / )
− ,
ℎ = flow depth =
,
= energy gradient
in which:
Bed forms
, = critical time-averaged bed-shear stress
When bed forms are present, the grain-related or according to Shields
effective bed-shear stress ( ) instead of the overall
bed-shear stress ( ) should be used to calculate the
particle mobility parameter
93 94

Suspension parameter, Z Transport Rate,

Z reflects the ratio of the downward gravity forces Dimensionless transport usually represented as:
and the upward fluid forces acting on a suspended
=
sediment particle in a current and read as:
−1
= Another dimensionless expression is:
∗

in which: =
= particle fall velocity in clear fluid in which:
∗ = overall bed-shear velocity qt = volumetric total transport rate (m2/s)
= von Karman constant = acceleration of gravity (m/s2)
= ratio of sediment and fluid mixing coefficient d50 = median particle size of bed material (m)
95 96

16
 Transport Rate, 2. Fluid Velocities and Bed Shear Stresses
= particle fall velocity of bed material (m/s) Sediment transport processes are dominant near the
= specific density ( /) near-bed region

Volumetric sediment transport rate can also be Boundary layer flows

made dimensionless with the specific flow discharge A boundary layer is defined as the flow region next to
( ), yielding the discharge-weighted concentration a solid boundary where the flow field is affected by
the presence of the boundary
=
At the boundary, the velocity is zero
in which:
In a boundary layer, momentum is gained from the
= volumetric total transport rate (m2/s) main stream (or free stream) and contributes to the
= specific flow discharge (m2/s) boundary layer growth
97 98

Fluid Velocities and Bed Shear Stresses Equation of motion

A boundary layer is characterized by its thickness
defined in terms of 99% of the free-stream velocity

= [ = 0.99 ]

where is measured perpendicular to the boundary

and is the free-stream velocity

Although the boundary layer thickness is (arbitrarily)

defined in terms of 99% of the free-stream velocity,
the real extent of the effects of boundary friction on
Fluid forces and bottom shear stress
the flow is probably about 1.5 – 2 times
99 100

Equation of motion Equation of motion

For steady uniform (turbulent) flow By definition
= ∗
Δ Δ = (ℎ − )Δ Δ sin
resulting
or = (ℎ − ) sin ∗ = ℎ
where
or = ℎ−
= bed-shear stress, N⁄m
where = sin = energy gradient ∗ = bed-shear velocity, m⁄s
varies linearly with = fluid density, kg⁄m
= gravity acceleration, m⁄s
For z = 0 this yields = ℎ = ℎ
ℎ = flow depth, m
101 102

17
 Turbulent flow Turbulent flow
Turbulence is a random fluctuating velocity field which
interacts with and derives its energy from the mean
flow field

In turbulent flow the water particles move in very

irregular paths, causing an exchange of momentum
form one portion of fluid to another, and hence, the
turbulent shear stress (Reynolds stress)

In turbulent flow both viscosity and turbulence

contribute to shear stress

103 104

Turbulent flow Turbulent flow

According to the Reynolds’ procedure, the shear stress
in turbulent flow at height in a steady uniform flow =− = turbulence shear stress
can be described as: = time-averaged fluid velocity at height
= turbulent fluid velocity fluctuation in horizontal
= − direction
= turbulent fluid velocity fluctuation in vertical
or direction
= + = fluid density
in which = kinematic viscosity coefficient
= = viscous shear stress
105 106

Turbulent flow Turbulent flow

Although the time-averaged vertical velocity is

equal to zero ( = 0), the vertical turbulent
fluctuations are not equal to zero ( ≠ 0).
Consequently, the turbulence shear stress

=− ≠0

Turbulent shear stress is dominant in the major part

of the flow depth
107 108

18
 Turbulent flow Turbulent flow

In case of a smooth bottom the viscous shear stress

becomes dominant close to the bottom because the
turbulent fluctuations and die out near the The layer where viscous shear stress is dominant is
bottom and are equal to zero at the bottom ( = called the viscous sublayer ( )
at = 0)
109 110

Turbulent flow Turbulent flow

Between the viscous sublayer and the logarithmic Above the viscous sublayer the flow is turbulent
sublayer there is a transition sublayer, sometimes
called buffer sublayer
111 112

Turbulent flow Hydraulic Regimes

The roughness elements mainly influence the velocity
distribution close to the bottom by generating eddies
(with a size of the order of the roughness elements)

Further away, the eddies will rapidly be absorbed in

the general existing turbulence pattern

The type of flow regime can be related to the ratio of

the Nikurdase roughness ( ) and a length scale of the
Above the logarithmic sublayer there is an outer viscous sublayer ( / ∗ ) in which = kinematic
sublayer viscosity coefficient (m /s) and ∗ = bed-shear
velocity (m/s)
113 114

19
 Hydraulic Regimes Hydraulic Regimes
Based on experimental results, it was found that: ∗
Hydraulically rough flow, for = ≥ 70
∗ ⁄ ∗
Hydraulically smooth flow, for = ≤5
⁄ ∗
viscous sublayer does not exist and the flow velocity
Roughness elements are much smaller than the distribution is not dependent on the viscosity ( ) of
thickness of the viscous sublayer and do not affect the the fluid
velocity distribution
∗ ∗
Hydraulically rough flow, for = ≥ 70 Hydraulically transitional flow, for 5 < < 70
⁄ ∗

Bed roughness is so large that it produces eddies close The velocity distribution is affected by viscosity as well
to the bottom as by the bottom roughness
115 116

Mixing Length Mixing Length

fluid parcel travels over a length before its fluid parcel, located in layer 1 and having the velocity
momentum is transferred, i.e., before losing their , moves to layer 2 due to eddy motion
identity by mixing

117 118

Mixing Length Mixing Length

There is no momentum transfer during movement, i.e., This action will speed up the fluid in layer 2 which can
the velocity of the fluid parcel is still when it just be seen as a turbulent shear stress acting on layer 2
arrives at layer 2, and decreases to sometime later trying to accelerate layer 2
by momentum exchange with other fluid in layer 2
119 120

20
 Mixing Length Mixing Length
Assuming the vertical instantaneous velocity
fluctuation having the same magnitude

=−

where negative sign is due to the downward

movement of the fluid parcel
Products of velocity fluctuations were then formulated
horizontal instantaneous velocity fluctuation of the in terms of the mixing length
fluid parcel in layer 2 is
= − = =−
121 122

Turbulent sublayer Turbulent sublayer

Turbulent shear stress now becomes After substituting the above equation into the
turbulent shear stress and total shear stress
=− = equations, the viscous and turbulent shear stress
components are
Turbulent shear stress depends on the magnitude of
the velocity gradient and the mixing length = +

Mixing length is related to the distance to the wall: viscous turbulent

= The shear stress ( ) in the region close to the wall is

in which is the von Kármán constant ( ≅ 0.4) assumed to remain constant and equal to the
boundary shear stress = ∗
123 124

Turbulent sublayer Turbulent sublayer

In the turbulent sublayer the total shear stress ⁄
assuming the mixing length = 1−
contains only the turbulent shear stress ℎ
Total shear stress increases linearly with depth where von Kármán constant = 0.4, we get

( )= 1− ⁄ ∗
ℎ = =

By Prandtl’s mixing length theory Integration of the equation gives the famous logarithmic
velocity profile
=
∗
= ln
125 126

21
 Viscous sublayer Viscous sublayer
In the case of hydraulically smooth flow there is a Linear velocity distribution intersect with the
viscous sublayer logarithmic velocity distribution at the elevation
= 11.6 ⁄ ∗ yielding a theoretical viscous sublayer
Viscous shear stress is constant in this layer and equal
thickness
to the bottom shear stress
= 11.6
∗
= = = ∗
With shear velocities of the order of 0.1 m/s, the
Integrating and applying = 0 gives laminar sub-layer thickness in open-channel flow is
typically of the order of 0.1 mm, which is the size of
⁄ ∗ sands
= =
Generally speaking, a plane bed surface is hydraulically
linear velocity distribution in the viscous sublayer smooth for silts and clays
127 128

Transition (buffer) zone Transition (buffer) zone

Approximation to the velocity profile between ∗ 7 ∗ ∗
= 7 tan + tan − 0.52 tan
turbulent flow and the laminar sub-layer has been ∗ 7 3 7 7
given by Spalding
This formulation with argument in radians is useful for
∗
= + 0.1108 −1− − − ∗
2 6 4≤ ≤ 70

where = ( )⁄ ∗ logarithmic velocity profile is valid when

Alternatively, an explicit formulation has been ∗
proposed by Guo and Julien (2007) 70 < < 1000

129 130

Hydraulically rough and smooth boundaries Hydraulically rough and smooth boundaries
Natural boundaries are hydraulically smooth when
the surface grain roughness
∗
< or ∗ = <4
3
A transition zone exists where

or ∗
< <6 4< < 70
3
Turbulent flows are hydraulically rough when the grain
diameter far exceeds the laminar sub-layer thickness
>6 or ∗ > 70
131 132

22
 Hydraulically rough and smooth boundaries Hydraulically rough and smooth boundaries
In early experiments, Nikuradse glued sand particles Compared with the particle size distribution of the
and measured velocity profiles for turbulent flow over bed material, the roughness height has been shown to
boundaries with grain roughness height be approximately
≅3 or ≅ 6.8
On rough boundaries, the corresponding value of
= ⁄30 is always less than the surface grain diameter
Flow velocity at an elevation = is equal to
1 30 2.3 30 30
= ln ≡ log ≅ 5.75 log = 5.75 ∗ , and the velocity against a particle is thus
∗ roughly ≈6 ∗
In practice, gravel- and cobble-bed streams are Reference velocity is the velocity at = or =
considered hydraulically rough 8.5 ∗
133 134

Hydraulically rough and smooth boundaries Resistance to flow

Integration constant of the logarithmic velocity profile, In open channels, resistance to flow describes the
, corresponding to the elevation of zero velocity is property of the channel to reduce the mean flow
given as velocity
Smooth flow regime : Three commonly used parameters that define
= 0.11 ∗ resistance to flow and respective flow velocity
for ≤5
∗ relationships are:
Rough flow regime :
∗ 1) Darcy-Weisbach friction factor,
= 0.033 for ≥ 70
Transition regime : 8 8
∗ = = ∗
= 0.11 + 0.033 for 5 < < 70
∗
135 136

Resistance to flow Resistance to flow

2) Chezy coefficient, The following identities can be defined from =
∗
=
3) Manning coefficient,
⁄
1 8
= ⁄ ⁄ = = =
∗

where = hydraulic radius, = energy slope

Both and describe resistance to flow, but
describes flow conveyance

137 138

23
 Mean velocity Mean velocity
Averaging logarithmic velocity profile over the depth Neglecting the ⁄ℎ parameter in , the depth-
yields: averaged flow velocity does occur at
1 1 ∗ = ℎ⁄ ≈ 0.37ℎ, in which is the base of natural
= ( ) = ln logarithm ( ≈ 2.72)
ℎ ℎ
∗
This is the basis for the “six-tenths-depth rule” for
= ln measuring local average “vertical” velocity
ℎ
∗ ℎ Applying the average velocity ( ) equation in the
= − 1 + ln logarithmic velocity profile, the velocity distribution
ℎ
can also be expressed as:
∗ ℎ ∗ ℎ
≈ −1 + ln = ln
= ln
⁄ℎ − 1 + ln ℎ⁄
139 140

Mean velocity Mean velocity

The mean flow velocity for hydraulically smooth and ≅ 5.2
rough boundaries are, respectively
≅ 6.8
8 ∗ℎ The following relationship appears frequently in the
≡ ≅ 5.75 log + 3.25
∗ literature
and ≅3
ℎ 12ℎ The resulting mean velocity for hydraulically rough
≅ 5.75 log + 6.25 ≅ 5.75 log
∗ channels with plane surface can thus be approximated
For gravel-bed streams, Bray (1982) recommended by
≅ 3.1 8 4ℎ 2ℎ
= ≅ 5.75 log ≅ 5.75 log
≅ 3.5 ∗
141 142

Bed Shear Stress and Bed Friction Bed Shear Stress and Bed Friction
The overall time-averaged bed-shear stress is defined as: The frictions factor for laminar flow in a wide open
1 channel is:
= ℎ = =
8
64
in which: =
ℎ = water depth (m)
= energy line gradient
in which:
= depth averaged velocity (m/s)
= Chézy-coefficient ( = 8 ⁄ ), (m ⁄ ⁄s) = ℎ⁄ = Reynolds number
= friction factor of Darcy-Weisbach
= effective bed roughness height (m)
= fluid density (kg⁄m )
= acceleration of gravity (m⁄s )
143 144

24
 Bed Shear Stress and Bed Friction Bed Shear Stress and Bed Friction

Using ∗ = ℎ in , yields or using the corresponding expressions for integration

constant, , yields:
ℎ
= −1 + ln 12ℎ
Hydraulic smooth flow : = 18 log
3.3 ⁄ ∗
2.3 0.37ℎ
= log
11.4ℎ
= 18 log
⁄
or using = 0.4 and = 9.81 m/s2:
12ℎ
0.37ℎ Hydraulic rough flow : = 18 log
= 18 log
145 146

Bed Shear Stress and Bed Friction Bed Shear Stress and Bed Friction

12ℎ Rough regime equation can be approximated by the

Transitional flow : = 18 log Strickler formula (in the range of = 40 to 70 m . ⁄s):
+ 3.3 ⁄ ∗

12ℎ ⁄
= 18 log ℎ
+ 1.05 ⁄ = 25

8
The friction factor follows from: = Another widely used resistance equation is the Manning
equation, which reads as:
yielding for hydraulic rough flow:
⁄ ⁄
ℎ
12ℎ =
= 0.24 log
147 148

Bed Shear Stress and Bed Friction 3. Incipient motion

Water flowing over a bed of sediment exerts
in which forces on the grains that tend to move or entrain
⁄
= 0.045 = Manning coefficient them.
ℎ = water depth (m)
= effective bed roughness (m) Incipient motion describes the flow and
boundary conditions for which the transport of
Note: the water depth (ℎ) should be replaced by the sediment starts to occur.
hydraulic radius ( ) for narrow channels
Accurate estimate of the onset of sediment
motion is required
to prevent erosion of channel bed
149 150

25
 Incipient motion Incipient motion
to predict the risks of scouring below Finer sediments that contain appreciable
foundations (e.g. bridge pier foundations) fractions of silt or clay, or both, tend to be
cohesive and resist entrainment mainly by
to select rock armour material cohesion rather than by the weight of the
individual grains
Forces that resist the entraining action of the
flowing water differ according to the grain size In fine sediments groups of grains are entrained
and grain size distribution of the sediment. as units.
Course noncohesive sediments are moved as
For course sediments, e.g., sands and gravels, individual grains.
the forces resisting motion are caused mainly
Driving forces are strongly related to the local
by the weight of the particles
151
near-bed velocities. 152

Incipient motion Incipient motion

The threshold conditions between erosion and A sediment particle is at a
sedimentation of a single particle describe state of incipient motion
incipient motion. when one of the following
conditions is satisfied:
Initiation of motion is not merely deterministic
phenomenon but a stochastic process as well =

in turbulent flow conditions the velocities are =

fluctuating in space and time =
particle size, shape and position varies where = overturning moment due to and
randomly
153
= resisting moment due to and 154

Incipient motion Incipient motion

Real fluid moving past a body will exert a drag
force on the body

The fluid forces acting on a grain resting on the

bed consist of a drag ( ) and a lift force ( )

Drag force is consisted of friction drag and form

drag
155 156

26
 Incipient motion Incipient motion

The total drag is written as where projected area of the body to the plane
perpendicular to the flow direction
1 1 , drag and lift coefficients, depend on the shape
= = ∗ and surface roughness of the body and the
2 2 4 Reynolds number
particle diameter
The lift force is written in the same way fluid density
, ∗ average velocity and friction velocity close to the
1 1 bed
= = ∗ a coefficient
2 2 4 157 158

Incipient motion Incipient motion

Let us consider the steady flow over the horizontal The driving force is the flow drag force on the grain
bed composed of cohesionless grains
1
= ∗
2 4
The stabilizing force can be modelled as the friction
force acting on the grain
If ∗, , critical friction velocity, denotes the situation
where the grain is about to move, then the drag
force is equal to the friction force, i.e.

∗, = − − ∗,
159 160

Incipient motion Incipient motion

which can be re-arranged into We say that sediment starts to move if

4 ∗ > ∗, critical friction velocity ∗, = , ⁄

∗,
=
−1 3 + or

Shields parameter is defined as > , critical bottom shear stress , = ∗,

or
∗,
= > critical Shields parameter ∗,
−1 =
−1

161 162

27
 Incipient motion Incipient motion
0.24
the friction coefficient of the bed is =
12ℎ
log
8
=
∗
hydraulically rough flow ≥ 70
0.24
= 0.24
12ℎ =
log 12ℎ
3.3 ⁄ ∗ log
3.3 ⁄ ∗ +
∗
hydraulically smooth flow ≤5 ∗
transition flow 5< < 70
163 164

Incipient motion Incipient motion

Initial motion occurs when

instantaneous fluid force
moment
≥
instantaneous stabilizing
moment
+ ≥
+ + ≥ sin
165 166

Incipient motion Incipient motion

or At high the pressure force will be much larger
than the skin friction force and resulting will act
+
≥ tan through the centre of the particle ( = 0 and thus
− ⁄ = 1)
Assuming the ratio of the lift force and the At low the viscous friction force at the top of the
submerged particle weight is relatively small particle will dominate yielding > 0 and thus
<1
≥ tan
Substituting
where = ⁄ + depends on the Reynolds
number, 1
= ∗ = −
167 2 4 6 168

28
 Incipient motion Incipient motion
and rearranging, we have 4 tan
= tan =
3
∗
≥ tan
−1 = critical Shields parameter
or
= = relative density
≥
in which The factor depends on:

∗
hydraulic condition near the bed – ( ∗ = ∗ ⁄ )
= =
−1 − = ( ∗)

= mobility Shield parameter 169

particle shape and relative position 170

Incipient motion Incipient motion

Initiation of motion for a current over a plane bed, =

( ∗ ) Sheilds (1936)
Modified Shields diagram for granular material
, − value can only be obtained by iteration 171 172

Incipient motion Incipient motion

Modified Shields diagram

Initiation of motion and suspension for a current over a plane bed, = ( ∗)
173 174

29
 Incipient motion Incipient motion

3 distinct zones corresponding to 3 flow situations

1. Hydraulically smooth flow for = ∗ ⁄ ≤2
d is much smaller than the thickness of viscous sublayer
Initiation of motion for a current over plane bed, , = ( ) Sheilds (1936)
Grains are embedded in the viscous sublayer and hence, is
Influence of temperature is only significant for a independent of the grain diameter
particle diameter smaller than 600 µm 175 By experiments it is found = 0.1⁄ 176

Incipient motion Incipient motion

3 distinct zones corresponding to 3 flow situations 3 distinct zones corresponding to 3 flow situations
2. Hydraulically rough flow for ≥ 500 3. Hydraulically transitional flow for 2 ≤ ≤ 500
Viscous sublayer does not exist Grain size is the same order as the thickness of the
viscous sublayer
is independent of the fluid viscosity
There is a minimum value of = of 0.032
has a constant value of 0.06 177 corresponding to = 10 178

Incipient motion Comparisons

between Yang
Yang (1973) criteria for incipient motion has a equations and
form: laboratory data
when the particle Reynolds number < 70, i.e.
in hydraulically smooth and transition regime:
2.5
= + 0.66 ∗
0< < 70
log ∗ − 0.06

in the completely rough regime:

= 2.05 ∗
≥ 70
179 180

30
 Influence of criterion Influence of criterion
Data available on critical shear stress are based on
Complexity of defining a critical bed-shear stress
more or less arbitrary definitions of critical
for initiation of motion is mainly caused by
conditions
stochastic character of the driving fluid forces Intensities of motion near the critical or threshold
and the stabilizing resisting forces condition are defined as weak, medium and
general
lack of an unambiguous definition of
initiation of motion Weak movement indicates that a few or several of
the smallest sand particles are in motion in
isolated spots in small enough quantities so that
those moving on 1 cm2 of the bed can be counted
181 182

Influence of criterion Influence of criterion

Medium movement indicates that the condition

in which grains of mean diameter are in motion in
numbers too large to be countable

movement is no longer local in character

not yet strong enough to affect bed

configuration and does not result in
appreciable sediment discharge

Initiation of motion for a current over plane bed

183 184

Influence of criterion Influence of criterion

General movement indicates the condition in

which sand grains up to and including the largest
are in motion

movement is occurring in all parts of the

bed at all times

permanent grain movement at all location of

the bed surface

185 Initiation of motion and suspension for a current over plane bed186

31
 Influence of shape, gradation and size Influence of shape, gradation and size

− parameter is not much affected by the , , = ,

shape of the particles when the nominal diameter where
is used = critical bed-shear stress of fraction with size
, ,

, = critical bed-shear stress of average diameter

Gradation has no effect when the size range is
rather wide ( 90/ 50 > 3) because the larger log(19)
particles will be more exposed, while the smaller = = exposure coefficient
log 19 ⁄
particles are shielded by the larger particles
Particles smaller than have a much larger
Critical bed-shear stress of each fraction of the , because they are shielded between the large
bed material can be determined using particles
187 188

Influence of shape, gradation and size Influence of bed slope

Sediment particle
Critical bed shear stress of fine cohesionless flaky Longitudinal slope β
resting on a
sediments with particle sizes in the range of 10 to longitudinal bed will be
100 is proposed to be set in motion when the
sum of critical fluid
.
, ∗, force and gravity force
= 0.1 component is just
−
equal to the stabilizing
force
∗,
for 0.03 < <1
, + sin = cos tan
or
.
= 0.11 ∗ for ∗ < 10 = cos tan − sin
,
189 190

Influence of bed slope Influence of bed slope

Longitudinal slope β Longitudinal slope β

Critical fluid force for a horizontal bed ( , , ) is in which
given as
sin −
, , = tan = For downsloping flow <1
sin
Combining the above two equations, we have sin +
= For upsloping flow >1
, cos tan − sin sin − sin
= =
, , tan sin In terms of bed shear stress,

or , = , , , = , ,
191 192

32
 Influence of bed slope Influence of bed slope
Sediment particle
Transverse slope γ Transverse slope γ
resting on a
transverse sloping bed or , = cos tan − sin
will be set in motion
when the driving fluid Combining the above equation with the square of
force is equal to the , , = tan
stabilizing force
we obtain
=
, cos tan − sin
or =
, , tan
.
, + sin
tan
= cos 1−
= cos tan tan
193 194

Influence of bed slope Influence of bed form

Bed-shear stress ( ) over a bed consisting of bed forms
Transverse slope γ
is composed of a part
or , = , , - related to skin friction over the bed surface

or , = , , - related to non-uniform pressure distribution

over the bed form crest and eddy region
tan sin
in which = cos 1− = 1− . = +
tan sin
Sediment particle resting on the surface of bed form will
For a combination of a longitudinal and transverse bed slope be set in motion by the friction force or by the
turbulent fluctuations in the eddy region downstream of
, = , , the crest
195 196

Influence of cohesive material Influence of cohesive material

When the bed consists of silty and muddy Measured critical bed stress obtained from flume
materials, cohesive forces between the sediment tests was related to the critical bed-shear stress
particles cause a distinct increase of the strength according to Shields, yielding:
of soil against erosion
.
, = , ,
Important factor governing the erodibility of
cohesive soils is the rate of consolidation
in which
with the compaction of the soil the erosion = percentage of fines (mud, silt) smaller than
resistance rapidly increases 50 µm (in %, minimum value = 1%)

197 198

33
 Critical depth-averaged velocity Critical depth-averaged velocity
The earliest studies were related to critical Critical depth-averaged velocity can be derived
velocities of stones from the critical bed-shear stress using the Chézy
equation
Critical near bed velocity was related to the
particle diameter as follows Assuming hydraulic rough flow condition
( ∗ / > 70), the critical depth-averaged flow
.
, = velocity for the plane bed can be expressed
as
The near bed velocity is, however, not very well
defined and therefore it is preferred to use the 12ℎ
critical depth-averaged velocity as the = 5.75 ∗, log
characteristic parameter
199 200

Critical depth-averaged velocity Critical depth-averaged velocity

in which The above equation can be expressed as
= depth averaged critical velocity 12ℎ
= 5.75 −1 log
∗, = −1
= critical bed-shear velocity Using the above equation, = 3 and =
2 and Shields curve, the critical depth-
ℎ = water depth
averaged velocity for sand particles in the range
= = effective bed roughness of flat bed of 0.0001 to 0.002 m can be expressed as
= coefficient ( = 1 for stones ≥ 0.1 .
12ℎ
and = 3 for sand and gravel material) = 0.19 log
3
201 202

Critical depth-averaged velocity Critical depth-averaged velocity

for 0.0001 ≤ 50 ≤ 0.0005 m Critical depth-averaged velocity of coarse gravel
12ℎ and stone material suggested by Neill and
.
= 8.50 log Maynord are
3
⁄
for 0.0005 < 50 ≤ 0.002 m ℎ
Neill : = 1.4 −1
in which
⁄
= median particle diameter (m) ℎ
Maynord : = 1.3 −1
= diameter in which 90% finer (m)
203 204

34
 Incipient motion Initiation of suspension
For increased values of the bed-shear velocity,
the particles will be moving along the bed by
more or less regular jumps (saltation).

Sediment particles may go into suspension when

the value of the bed-shear velocity becomes
comparable to that of the particle fall velocity.

Particle only remains in suspension when the

turbulent eddies have dominant vertical velocity
components which exceed the particle fall
velocity.
Critical depth-averaged velocities for a plane bed 205 206

Initiation of suspension Initiation of suspension

Assuming that the vertical velocity component Criteria for initiation of suspension
( ′) of the eddies are of the same order of
magnitude as the vertical turbulence intensity ∗,
=1
, the critical value for initiation of suspension
can be expressed as: ∗,
or = =
. ( − 1) ( − 1)
= ≥
where
The vertical turbulence intensity has a u*,cs = critical bed-shear velocity for initiation of suspension
maximum value of the same order as the bed- ws = particle fall velocity in clear and still water
shear velocity ( ∗ ), both for hydraulic smooth d50 = particle diameter
and rough flow conditions ρs = density of sediment
= ∗ 207
ρ = density of fluid 208

Initiation of suspension Initiation of suspension

Another criteria for initiation of suspension given
or
by Engelund, based on a rather crude stability 16
analysis is: =
( ∗ ) ( − 1)
∗,
= 0.25 ∗,
∗ > 10: = 0.4,
From results of experimental investigation
carried out in Delft Hydraulics suggested that: or
∗, 4
1< ∗ ≤ 10: = , = 0.16
∗
( − 1)

209 210

35
 3.2 Design of Stable Channel Design of Stable Channel

The objective of an irrigation canal is to meet the The design of irrigation canals for sediment-
varying irrigation requirements during the laden water should consider aspects related to
irrigation season at the individual farms. the conveyance of irrigation water as well as the
transport of sediments.
Irrigation canals should be designed based on
the criterion that no sedimentation and no The design of a canal that has to convey a certain
erosion occur during a certain period. sediment load requires a set of equations related
to the water-sediment flow to provide the
The design of a stable cross section will be the unknown variables of bottom slope and cross
end result of this criterion. section (bottom width and water depth).

211 212

Design of Stable Channel Design of Stable Channel

Experience shows that there is no set method of In view of the design, irrigation canals can be
designing canals. divided into three categories:
 Design of the slope and cross section for open Canals with a rigid boundary:
channel flow in canals and laterals must to a
large extent be based on experience and The design is based on the determination of the
judgment. velocity at which any sediment entering into the
canal will not settle on the canal perimeter.
Stable channels can be developed where the
discharge, sediment load, and earth materials High velocities are allowed, but they should not
through which the canal is to be constructed are damage the lining or create large disturbances
uniform. in the water surface.
213 214

Design of Stable Channel Design of Stable Channel

Canals with an erodible boundary and carrying Canals with an erodible boundary and carrying
clean water: water with sediment:

The canal design is based on the determination The design principle assumes that the canal
of the maximum velocity for which the bed should transport the water as well as the
material in the cross section does not move. sediment.

The minimum cross section with a maximum The cross section must ensure flow velocities as
velocity that does not result in scouring of the large as possible to convey the sediment and at
bed should be the end result of the design. the same time not too large to prevent scouring
of the bed.
215 216

36
 Design Methods Regime Method
Design methods can be divided into four Transport of sediments is allowed as long as
categories: there is no net annual scour or deposition in
each channel.
Regime (empirical) method
A channel is assumed to be in regime
Tractive force method
(equilibrium) if it has not changed its
Permissible velocity method characteristics over a period of one or several
typical water years.
Rational method
For the regime theory a set of simple, but
empirical equations are available.
217 218

Regime Method Regime Method

Regime analysis originated with analyses on
Based on a set of empirical equations using
irrigation canals in India and Pakistan.
water depth, width, slope, flow velocity and
Engineers sought to determine what was unique discharge to develop relationships for stable
about those channels that said to be “in regime”. width, wetted perimeter, slope, and depth.
Obtain data from “regime” channels
The equations of the regime theory present a
Plotted the variables judged pertinent in long-term average profile rather than a
various combinations instantaneously variable state.
Tried thereby to discover what relations hold
among hydraulic and sediment characteristics
obtained in these channels 219 220

Regime Method Lacey Regime Method

Relations for the depth, width, and slope Some of the equations given by Lacey are
generally involve one or more loosely defined
quantities. =8
Suitable for large-scale irrigation systems with a = 2.67
wide range of discharges of fine fraction.

Equations may only be used within their validity = 1.17

range.
⁄
= ⁄
1750
221 222

37
 Lacey Regime Method Lacey Regime Method
where
The following steps are recommended when
= Lacey’s silt factor for a sediment size using the Lacey equations for the design of
earthen canals for a given diameter and
= sediment size (inches)
discharge :
= wetted perimeter (ft)
Determine the factor for the given
= discharge (ft3/s)
diameter
= hydraulic radius (ft)
Find the bottom slope and the perimeter
= mean velocity (ft/s)
= bottom slope
223 224

Lacey Regime Method Simons and Albertson’s Regime Method

Determine the velocity and area = ⁄ Channels are divided into five types:
and the hydraulic radius = ⁄ (1) Sand beds and banks

Determine the full cross section both by (2) Sand beds and cohesive banks with no
using the perimeter and the hydraulic sediment load
radius and by assuming that the cross (3) Sand beds and cohesive banks with
section is trapezoidal with a side slope of 1 sediment loads, 2000−8000 ppm
V: 2 H.
(4) Cohesive bed and banks
(5) Coarse noncohesive material
225 226

Simons and Albertson’s Regime Method Simons and Albertson’s Regime Method
Developed three groups of graphical stable depth, ft = 1.21 < 7 ft
relationships and equations to define stable
= 2 + 0.93 ≥ 7 ft
width and wetted perimeter, stable depth and
hydraulic radius, and stable velocity where
⁄
, = constant dependent on type of channel
stable wetted perimeter, ft =
= discharge, cfs
average width, ft = 0.9
Equations for stable flow rates
top width, ft = 1.087 + 2.17
= ( )
stable hydraulic radius, ft .
=
227 228

38
 Simons and Albertson’s Regime Method Simons and Albertson’s Regime Method

and
.
= =
ℎ

where
= average velocity
= kinematic viscosity
, , = constants dependent on type of
channel
229 230

Simons and Albertson’s Regime Method Blench’s Regime Method

From continuity
Utilizes a bed factor, a side factor, and a flow
= ℎ resistance equation to develop relationships for
For classification purposes, sediment sizes, stable width, depth and slope
channel banks, and beds are assumed to be sand
bed channel if the bed and bank materials are
medium to fine sands average width, ft =
Cohesive bed materials are those finer than fine
sand
Coarse bed materials are those with medium stable depth, ft ℎ=
particle size between 20 and 82 mm
231 232

Blench’s Regime Method Blench’s Regime Method

⁄ ⁄ side factor
stable slope = = 0.1 slightly cohesive banks (very fiable)
⁄ 1+
2330 = 0.2 moderately cohesive banks (silty clay loam)
= 0.3 highly cohesive banks (tough clay)
3.63
= ⁄ where
= discharge, cfs
bed factor = 1.9 1 + 0.012 = kinematic viscosity, sfs
= gravitational constant, ft/s2
= sediment concentration, ppm by weight
233
= median diameter of sediment, mm 234

39
 Regime Methods Regime Methods
Due to the great variation in soils usually found 1. The terrain crossed by the canal alinement may
along the alinement of the canals, the regime extend for long distances through great variation
formulas were not fully adopted. in soils.
However, they do serve as a starting point for  The major types of soil can be considered
design, and are useful for that purpose. separately in reaches, but within each soil
class, e.g., clay, silts, sands, and gravel, there is
Even so, factors other than the transport
a wide variation in size of particle and
characteristics of the canal enter into its design.
cohesion; and thus, in the erosion resistance
Some of these factors are:
of the soil.

235 236

Regime Methods Regime Methods

2. Some soils in place can resist erosion but may 4. Land use and its state of development also affect
erode easily when reworked in excavation and the selection of the canal cross section.
embankment.
 A new project area may require several years
3. After construction, canal sections must hold to develop a high water demand that
water as well as be stable. approaches design capacity.
 To prevent excess leakage, compacted earth  Others may require the full design rate of flow
lining, hard surface lining, and membrane at once.
lining are often used.

237 238

Regime Methods The Tractive Force Method

5. Methods of irrigation operation and maintenance Based on the principle of allowing no or
also affect the canal design. negligible motion of bed material
 Some canals remain in operation all the year, Most suitable for small-scale irrigation systems
while others are operated only during an with small discharges in soils of coarse material
irrigation season and are empty a period of ( ≥ 500 µm)
months each winter. − values of the Shields curve, which
represents conditions with frequent particle
 Maintenance and repairs can be made much
movements at all location of the bed, can not be
easier and at less expense in the canal that
recommended for the design of stable channel
operates only part of the year.
Application of experimental results is preferred
239 240

40
 The Tractive Force Method The Tractive Force Method

The tractive force depends on the shear stress at depends on the B⁄h ratio:
the bottom, which can be expressed as (Dahmen, for 1 < B⁄h < 4: = 0.77 . ( ⁄ )
1994):
for B⁄h ≥ 4: =1
= ℎ
where = density of water (1000 kg/m3)
where = bottom slope (m/m)
= tractive force (m2/s) B = channel bottom width (m)

= correction factor h = water depth (m)

241
= acceleration due to gravity (m/s2) 242

The Tractive Force Method The Tractive Force Method

The tractive force method was developed for the For the canal design a boundary shear stress of
threshold condition of sediment transport, which 80% of the critical shear stress is recommended:
occurs for the critical boundary shear stress along
the canal’s perimeter. = 0.75 ( in mm)

Field studies on very coarse material showed that The allowable shear stress is a function of the
the ‘critical shear stress’, above which motion median diameter and the sediment
would start, is approximately concentration of the water.
= 0.94 (N/m2)

243 244

The Tractive Force Method The Tractive Force Method

245 246

41
 The Tractive Force Method The Tractive Force Method
Experimental results of Paintal who measured
bed-load transport rates close to initiation of
motion (particle size = 0.0025 m, 0.008 m
and 0.022 m) found the following dimensionless
transport rates
= 10 for = 0.01
= 10 for = 0.02
= 10 for = 0.025
= 10 for = 0.03
= 10 for = 0.04
247 248

The Tractive Force Method The Tractive Force Method

in which Applying
= ⁄[ −1 ] . = dimensionless 12ℎ
transport rate = 5.75 −1 log
= bed-load transport rate, kg/sm
= ⁄ = specific density (−) and assuming = 1, = 2 , the critical
depth-averaged velocity for stone in uniform river
= median diameter of sediment, m
flow is
= ⁄ − = dimensionless
mobility parameter 6ℎ
= 0.02: = 0.8 −1 log

249 250

The Tractive Force Method The Tractive Force Method

6ℎ in which
= 0.03: = 1.0 −1 log
= 1.45/(1 + 3 ) = reduction factor related
to additional turbulence
For regions with additional turbulence = ⁄ = relative turbulence intensity ( =
production (downstream of structures: weirs, 0.15 for uniform flow; =
bridge piers) a reduction factor must be taken 0.35 for region downstream of
into account, yielding hydraulic jump)
6ℎ = standard deviation of time-averaged
= 0.8 −1 log velocity
= time-averaged velocity
251 252

42
 Schematization of Cross-section Schematization of Cross-section
The most practical cross-section that can be ̂ = 0.85 ℎ = 0.85 ̂
, ,
constructed is a trapezoidal cross-section
Analysis of computed and measured bed-shear For side slope 1:3
stresses along the cross-section of trapezoidal
channel (with a ratio of the surface width and ̂ , = 0.90 ℎ = 0.90 ̂ ,
depth larger than 5) has shown the following For side slope 1:4
results:
̂ , = ℎ ̂ , = 0.95 ℎ = 0.95 ̂ ,

̂ , = 0.75 ℎ = 0.75 ̂ , For side slope 1:6

For side slope 1:2253 254

Schematization of Cross-section Schematization of Cross-section

in which Side part will be stable if : ̂ , ≤ , ,
̂ , = maximum bed-shear stress in the in which
middle part of the channel
, , = critical bed-shear stress for the horizontal
̂ , = maximum bed-shear stress in the side middle part (N/m )
part (with side vertical to horizontal)
= cos 1 − tan ⁄tan
Bed-material will be stable when the prevailing = reduction factor (−)
maximum bed-shear stress is smaller than the = angle of side slope (°)
critical bed-shear stress = angle of repose (°)
Middle part will be stable if : ̂ ≤ = surface slope (−)
, , ,
255 ℎ = water depth (m) 256

Permissible velocity method Permissible velocity method

The minimum permissible velocity is defined as

that velocity that will not result in sedimentation
or induce the growth of aquatic weeds. This
velocity depends on the sediment transport
capacity of the canal.

The maximum permissible velocity is that velocity

that will not cause erosion. This velocity is
difficult to ascertain and is very variable; it can
only be estimated with experience and sound
judgment.
257 258

43
 Rational method Rational method

The design of irrigation canals typically tries to They might include an alluvial friction predictor
find the four unknown dimensions, which include (de Chézy, Manning, and Strickler), a sediment
the bottom slope , bottom width , water transport predictor and a minimum stream
depth ℎ and side slope . power or maximum sediment transport
efficiency.
The side slope should be based on the soil
properties and the (estimated) canal depth ℎ. Sometimes a regime relationship is used to
Hence, three equations are required to provide a relationship for and ℎ.
determine the remaining variables.
The rational method is useful for the design of
stable canals with very specific flow conditions.
259 260

Bed form is channel bed irregularity that is related Bed forms are symptomatic of local variations in
to the flow conditions. the sediment transport rate, which can give rise to
quasi-regular sequences of erosion and depositions
A free surface flow over an erodible sand bed
generates a variety of different bed forms and bed The type of bed configuration and the dimensions
configurations (arrays of bed forms). of the bed forms are dependent on the properties
of the flow, fluid, and bed material
Bed forms are relief features initiated by the fluid
oscillations generated downstream of small local Bed forms represent an important means of
obstacles over a bottom consisting of movable transverse and horizontal adjustments that natural
(alluvial) sediment materials. streams can make
261 262

Bed form Dimensions Shape Behavior and

Bed forms play significant role in the makeup of occurrence
Small-scale forms 10-2 – 102 m
hydraulic roughness of an alluvial channel (1) Sand bed streams
Ripples Wavelength, λ < 0.6 m Triangular profile Generally restricted
Effective bed roughness ( ) is also called the to finer than 0.6 mm
Height, δ < 0.04 m Gentle upstream slope
equivalent Nikuradse grain roughness, because it Discontinuous
Sharp crest
was originally introduced by Nikuradse in his pipe movement at
Steep downstream face velocities << the flow
flow experiments
Dunes λ = 3 to 15 h Similar to ripples but Upstream slope may
more variable be rippled
δ ≤ 1/3 h
consists of grain roughness generated by skin Discontinuous
>> ripples
friction forces and of form roughness generated by movement

pressure forces acting on the bed forms Out of phase with

Lower regime of roughness surface water
Ripples + Dunes
Form roughness dominant
263 264

44
Bed form Dimensions Shape Behavior and Bed form Dimensions Shape Behavior and
occurrence occurrence
Small-scale forms 10-2 – 102 m (2) Gravel-bed streams

(1) Sand bed streams Pebble 10-1 – 100 m Linear in flow Streamlined
clusters direction
Plane bed Bed surface devoid of Profile view Plan view Consist of a coarse
bed form obstacle clast ( ) with
collection of stoss- and
May not occur for lee-side particles
some ranges of depth
and bed material Ribs 100 – 101 m Transverse to flow Repeated ridges of
Antidunes Relatively low Sinusoidal profile Less common than direction coarse clasts, whose
height dunes spacing is roughly
Symmetrical than proportional to the size
Dependent on Occur in steep streams
dunes of the largest particle in
flow depth and the ridge crest
In phase with surface
velocity
water waves Steps 100 – 101 m Transverse to flow Stair-like sequence
May move upstream, direction formed by arrangement
downstream or remain of boulders and cobbles
Upper regime of roughness
Plane bed + Antidunes stationery across steep-gradient
Grain roughness dominant 265 channels 266

Bed form types in rivers according to Simons and Richardson

Bed form Dimensions Shape Behavior and

occurrence
Large-scale forms 101 – 103 m
Bars Length Variable Five main types:
comparable to (1) Point bars: form
channel width particularly on the inner
bank of meanders
(2) Alternate bars: distributed
periodically along one and
then the other bank of a
channel
(3) Channel junction bars:
develop where tributaries
enter a main channel
(4) Transverse bars (include
cliffs) may be diagonal to
the flow
(5) Mid-channel bars typical
of braided reaches 267 268

Ripples

Aeolian ripples, NE Arizona

269
http://www.geo.uu.nl/Research/Sedimentology/Staff/mvhattum/flume.html 270

45
 Crest line patterns

Straight-crested current ripples, Cambrian sandstone, Moraine Lake, Alberta

271 272
http://www.ualberta.ca/~jwaldron/gallerypages/bedforms.html

Current-generated linguoid ripples, Kennetcook River, Nova Scotia, Canada Lunate dunes in sand and gravel, Kennetcook River.
273
http://www.ualberta.ca/~jwaldron/gallerypages/bedforms.html 274
http://www.ualberta.ca/~jwaldron/gallerypages/bedforms.html

Bed form migration in lower and upper regime

275 276

46
Liu described in terms of suspension parameter ( ∗ ⁄ ) and Simons and Richardson described in terms of stream
a particle-related Reynolds number ( ∗ ⁄ ) power ( ) and median fall diameter ( )

277 278

Van Rijn used dimensionless

Van den Berg and Van bed-shear stress parameter
Gelder used
dimensionless particle
mobility parameter
and dimensionless particle
diameter parameter
and dimensionless
particle diameter
parameter

279 280

Can be predicted by empirical equations or diagrams

Small (mini) ripples
Wavelength, 500 to 1000
Height, 50 to 200
Mega ripples
= 0.5ℎ
.
= 0.02ℎ 1 − 10 −

281 for 1≤ ∗ ≤ 10 and 3 ≤ ≤ 10282

47
Dunes where
Van Rijn = 7.3ℎ = power coefficient of velocity in sediment
. . transport formula
= 0.11ℎ ⁄ℎ 1− (25 − )
= dune shape coefficient ( = 0.5 to 0.7)
Gill =ℎ 1− 1− , ⁄ /2
Sand waves
Yalin =ℎ 1− , ⁄ /6
= 10ℎ for ≥ 15
= 2ℎ
. ( )
. = 0.15ℎ 1 − 1−
Allen = 0.086ℎ
=ℎ . for ≥ 15
283 284

In the presence of bed-forms, the resistance to the flow

consists of two parts, one originating from the skin friction,
another due to the form pressure of the bed-forms, i.e.
= +

Fluid pressure and

stress distribution
along a dune

Skin-friction force
and fluid pressure
force on a dune
285 286

Total bed shear stress : Grain related bed shear stress :

, + − Δ 1
= ≈ + = + = = =
8
287 288

48
 = +
1 1
= +
8 8
1
= +
8
Form related bed shear stress : 1
=
8
1 1 where
= = =
2 8
= +
289 290

Effective roughness height, ( ) can be divided into

grain related part ( ) and form related part For the Chezy coefficient =
( ) as ℎ

= + 1 1 1
= +
The following concept of separation have also been ′ ′′
used
For shear velocity ∗ = ℎ
8 ℎ
For the Darcy-Weisbach friction factor =
∗ = ∗ + ∗
= +
291 292

Values for sand and gravel beds in the lower regime

∗ were reported by:
For the dimensionless shear stress =
−1
Kumphuis for 540 ≤ d90 ≤ 4200 µm = 2.5 d90
= + ′′ Gladki for 1000 ≤ d90 ≤ 20000 µm = 2.3 d84

For a slope = + ′′ Hey for 1000 ≤ d90 ≤ 20000 µm = 3.5 d84

Mahmood for 100 ≤ d90 ≤ 600 µm = 5.1 d84

For a hydraulic radius = Van Rijn = 3 d90 for θ < 1 (lower regime)

= 3 θ d90 for θ ≥ 1 (upper regime)

= + ′′
293 294

49
where The effective form roughness is related to the bed-
= grain related effective roughness height of a plane bed form height (∆), the bed-form steepness (∆/ ) and
= ∗⁄ − 1 = mobility parameter ( = for the bed-form shape ( ).
plane bed)
∗ = ⁄ = bed shear velocity ( ∗ = ∗ for plane bed) ∆
= ∆, ,
12ℎ 12ℎ
= 18 log = 18 log
+ 3.3 + /
∗
The overall form roughness is the summation of the
individual values
= Chézy coefficient
= ⁄ = kinematic viscosity coefficient of fluid- = , + , + ,
sediment mixture ( ≈ 10 m⁄s in near-bed region) 295 296

Ripples – bed forms with a length smaller than

the water depth
where
= overall current-related form roughness
∆
, = form-roughness related to ripples , = 20 ∆
, = form-roughness related to asymmetrical
dunes where
, = form roughness related to symmetrical ∆ , = ripple height and length
sand waves
= ripple presence factor ( = 1 for ripples
alone, = 0.70 for ripples superimposed
on dunes or sand waves)
297 298

Dunes – asymmetrical bed forms with a length Sand waves – symmetrical bed forms with a
of about 7 times the water depth length much larger than the water depth

= 1.1 ∆ 1− ∆ / The leeside slopes of symmetrical sand waves are

,
relatively mild. Hence, flow separation will not
occur. Therefore the form roughness of symmetrical
where sand waves is assumed to be zero.
∆ , = dune height and length
= form factor ( = 0.7 for field conditions) , =0

299 300

50
 ∆ ∆ Engelund and Hansen
Engelund : = 10 for dunes
ℎ = +
1 ℎ
Vanoni-Hwang : = 3.5 log − 2.3 for ripples
∆ = ℎ ∗ = ℎ
and dunes
= ℎ ∗ = ℎ
where
= form friction factor of Darcy-Weisbach ∗ ℎ ℎ
∆ = bed form height
= =
∗ ℎ ℎ
= bed form length
ℎ = water depth 12ℎ
= 2.5 ∗ ln
= ratio of lee-slope area and total bed-form area 301 2.5
302

Engelund and Hansen Engelund and Hansen

Using = ℎ and ∗ = ℎ Upper regime :

.
ℎ 12ℎ = for 0.7 < <1
= 2.5 ln
ℎ 2.5
. . for ≥1
= 0.3 + 0.7
.
12ℎ
or = 2.5 ln
2.5
Two regimes can be identified as follows
Lower regime : = 0.06 + 0.4 for ≤ 0.7
303 304

Engelund and Hansen Engelund and Hansen

Two cases are possible: Case II: ℎ and are known; find and
Case I: ℎ and are known; find and 1. guess ℎ (start ℎ = 0.33ℎ)
2. compute ∗ and
1. compute ∗, and
3. determine
2. determine
4. compute ℎ = ⁄ ℎ
3. compute ℎ = ⁄ ℎ
5. repeat until ℎ = constant
4. compute
6. compute
5. compute = ℎ .
7. compute from = ℎ
305 306

51
White et al., 1979 White et al., 1979
1. Compute particle parameter ∗, , , 1. Compute particle parameter ∗, , ,
/
−1 =0 for ∗ ≥ 60
∗ =
= 0.17 for ∗ ≥ 60
= 1 − 0.56 log( ∗) for 1 ≤ ∗ < 60 = log .
∗

0.23 2. Compute bed-shear velocity, ∗

= .
+ 0.14 for 1 ≤ ∗ < 60
∗
∗ = ℎ
307 308

White et al., 1979 White et al., 1979

3. Compute mobility parameter, 5. Compute average flow velocity, from

∗ ∗
= . =
−1 −1 . 5.66 log(10 ℎ/ )

4. Compute mobility parameter, 6. Compute Chézy coefficient,

− =
= 1 − 0.76(1 − )
− ∗
309 310

Brownlie Brownlie
Presented a method to predict flow depth (ℎ) as where:
a function of the main flow variables
∗ = . .
Lower regime :
. . . =
ℎ = 0.372 ∗ .
−1
Upper regime :
= 1/2( ⁄ + ⁄ )
. . .
ℎ = 0.284 ∗

for ≥ 0.006 or ≥ 1.74 ⁄ = . .

311 ℎ 312

52
 5. Sediment Transport in Steady and Uniform
Currents
Smith and McLean
Engineers engaged in river regulation and design
Based on the analysis of river data, they proposed:
and operation of canal systems have great need
= 1 + 700 − for methods of computing sediment transport.
where: Unfortunately, available methods or relations for
= effective current related bed roughness computing sediment discharge are far from
completely satisfactory with the results.
= ⁄ − = mobility parameter
= critical mobility parameter Plans for works involving sediment movement
= bed shear stress by water cannot be based strongly on such
relations.
= median particle diameter of bed material 313 314

Sediment Transport in Steady and Uniform Selection of Sediment Transport Formula

Currents The problem of the engineer is to select one or
more of these for solving particular problem.
At best these relations serve as guides to
planning and usually the engineer is forced to This selection is not straight forward since the
rely on experience and judgment in such works. results of different formulas often differ
drastically.
The guiding provided turns out to be important
especially when the conditions in the problem Selections are usually based on evaluating:
area differ from those in the experience of the data on which each formula is based,
planning engineers.
formulas by comparing observed sediment
Many formulas have been suggested over years. discharges in rivers with values calculated by
the formulas.
315 316

Modes of sediment transport Modes of sediment transport

Bed-load Suspended load

part of the total load which has more or less part of the total load which is moving without
continuous contact with the bed continuous contact with the bed as the result of
the agitation of the fluid turbulence
must be determined in relation to the effective
shear stress which acts directly on the grain related to the total bed shear stress
surface
may also contain some wash load (usually,
particles smaller than 50 µm)

317 318

53
 Modes of particle motion Basic approaches
rolling and sliding motion or both Basic approaches used in the derivation of
sediment transport functions or formulas are:

 Regime Approach
saltation motion
 Regression Approach

 Probabilistic Approach
suspended particle motion  Deterministic Approach

319 320

Regime approach Regression approach

Regime channel is an alluvial channel in dynamic Some researchers believe that sediment
equilibrium without noticeable long-term transport is such a complex phenomenon that no
aggradations, degradation, or change of channel single hydraulic parameter or combination of
geometry and profile. parameters can be found to describe sediment
Some site-specific quantitative relationships exist transport rate under all conditions.
among sediment transport rates or
concentration, hydraulic parameters, and channel Instead of trying to find a dominant variable that
geometry parameters. can determine the rate of sediment transport,
they recommend the use of regressions based on
Regime equations obtained from the regime
laboratory and field data.
concepts are mainly obtained from regression
analysis of regime canal data.
321 322

Regression approach Regression approach

Parameters used in these regression equations where

may or may not have any physical meaning = . ⁄ . .
relating to the mechanics of sediment transport.
= total sediment concentration in ppm by
Shen and Hung (1972) proposed the following weight
regression based on 587 sets of laboratory data = average fall velocity of sediment particles
in the sand size range:
Karim and Kennedy used nonlinear, multiple-
log = −107,404.45938164 + 324,214.74734085 regression analysis to derive relations between
flow velocity, sediment discharge, bed form
−326,309.58908739 + 109,503.87232539
geometry and friction factor for alluvial rivers
323 324

54
 Regression approach Regression approach
and for flows well above incipient motion
The uncoupled relations recommended by Karim
and Kennedy are: .
.
= 2.822
1.65
log = −2.279 +2.972 log 1.65
1.65 1.65
where
= volumetric total sediment discharge
∗ − ∗
+1.060 log log = water discharge per unit width
1.65 1.65
= gravitational acceleration
= median bed-material particle diameter
ℎ ∗ − ∗ = energy slope
+0.299 log log
1.65 = average flow velocity
325 326

Regression approach Regression approach

. .
∗ = bed shear velocity = ℎ = 9.82
1.65 ⁄ 1.65 ⁄ ′
∗, = Shields’ value of critical shear velocity at
incipient motion
where = Darcy-Weisbach friction factor
ℎ = water depth
The grain roughness factor ′ can be expressed as
If it is necessary to take into account the bed
configuration changes in the development of a 8
friction or velocity predictor, the above equation ′=
6.25 + 2.5 ℎ⁄2.5
should be replaced by:
The friction factor ratio ⁄ ′ can be computed as:
327 328

Regression approach Regression approach

May give fairly accurate results for engineering
= . + . . + . − . + . − .
′ purposes if the equation is applied to conditions
similar to those from where the equation was
for  ≤ 1.5
derived
= 1.20 for  > 1.5 Application of the equation outside the range of
′
data used for deriving the regression equation
where should be carried out with caution.
ℎ In general, regression equations without
= =
1.65 1.65 theoretical basis and without using
dimensionless parameters should not be used for
and = specific weight of water
329
predicting sediment transport in natural rivers. 330

55
 Probabilistic approach Einestein Approach
Physically, more realistic than deterministic Einstein pioneered sediment transport studies
approach because the stochastic nature
(turbulence) of the fluid forces on the grains is Assumptions
taken into account Beginning and ceasing of sediment motion can
be expressed in terms of probability
May also give more accurate results for
computation of transport rates close to initiation Movement of sediment bed-load is a series of
of motion steps followed by rest periods
Proposed equations are, however, not easily There is equilibrium between the number of
applicable because of the statistics involved particles eroded and deposited per unit area and
time ( = )
331 332

Einestein Approach Einestein Approach

Assumptions The deposition rate is defined by

Average step length is 100 times the particle =

diameter
where = average travel distance of a particle
The erosion or pick-up is given by
Using the above two equations and = , it
= −1
follows that
where = probability of erosion
= coefficient = −1
333 334

Einestein Approach Einestein Approach

The average travel distance of a particle was

found to be
pick-up probability
=
1−
where
1− = probability of deposition

= coefficient (= 100)
= particle diameter
335 336

56
 Einestein Approach Einestein Approach
The probability of erosion was expressed as a
Substituting in bed-load transport equation
function of the particle mobility parameter
1
= −1 =
1−
1 in which
or = = probability of erosion
∗ 1−
= ⁄ −1 = mobility parameter
in which τ = ℎ = effective bed-shear stress
= dimensionless bed-load transport ℎ = effective water depth
∗ = empirical coefficient (= 43.5) = energy gradient
= probability of erosion = β⁄β∗ = coefficient
337 338

Einestein Approach Van Rijn Approach

For graded bed material the transport is obtained It is assumed that the instantaneous bed-load
by summation of the transport rate per fraction transport rate is related to the instantaneous -
parameter as follows
=
. .
= −1 ∗
The -coefficient represents various effects related
to the behaviour of particles in a graded mixture in which
such as the hiding of smaller particles between the
− ,
larger particle ( ), variation of lift coefficient ( ) = = instantaneous shear stress
and variation of particle position β⁄β∗ ,
parameter
339 340

Van Rijn Approach Van Rijn Approach

= = instantaneous effective bed-shear stress The probability that > , (positive in flow
,
direction) or − > , , (negative against flow
, = instantaneous critical bed-shear stress direction) is:

= efficiency factor
> , − <− ,

The effective bed-shear stress is assumed to

have a normal distribution (mean value ̃ , 1
=
standard deviation ), as follows 2 .
, ,

1
= . +
2
, ,
341 342

57
 Van Rijn Approach Van Rijn Approach

The time averaged shear stress parameter

can be derives as
. .
1
. = .
−
2 , , , ,

in which
Gaussian shear stress distribution
. ⁄
=+
> , − <− , = shaded region
343 344

Van Rijn Approach I-integrals

. ⁄
=−

+ ̅ − , ,
=

− ̅ − , ,
=

= ̅ +
345 346

Van Rijn Approach Van Rijn Approach

The bed-load transport against the flow direction
has to be subtracted from the value in the flow Bed-load transport equation was recalibrated for
direction uniform flow, yielding

Based on the work of De Ruiter, the instantaneous . . . . .

= 0.1 −1 ∗
critical bed-shear stress was found to be
To apply the above equation the values of , , ,
, = 1.5 ̅ ,
and , , must be known
The standard deviation of the effective bed-shear
stress in uniform flow was found to be
= 0.4 ̅
347 348

58
 Deterministic approach Deterministic approach
Basic assumption Deterministic sediment transport formulas can be
expressed by one of the following forms
Existence of one-to-one relationship between
independent and dependent variables = −
Conventional, dominant, independent variables = −
 water discharge = −
 average flow velocity
 shear stress = −
 energy or water surface slope
= −
 sediment particle diameter
 water temperature, or kinematic viscosity = −
349 350

Deterministic approach Deterministic approach

where
Yang used laboratory data from a laboratory
= sediment discharge per unit width of channel flume with 0.93-mm sand as an example to
= water discharge
examine the validity of these assumptions
= average flow velocity
= energy or water surface slope
= shear stress
= stream power
= unit stream power
1, 2 … 6, 1, 2, … , 6 = parameters related to flow
and sediment conditions
= subscript denoting the critical condition at
incipient motion
351 352

Deterministic approach Deterministic approach

More than one value of total sediment discharge With stream power as independent variable, the
can be obtained for the same value of water correlation improves. Farther improvement made
discharge, velocity, slope, or shear stress with unit stream power as dominant variable
353 354

59
 Deterministic approach Bed load transport definition
Transport of particles by rolling, sliding and
saltating
Relationship between
dimensionless unit Can be defined as the product of particle
stream power, stream concentration ( ), particle velocity ( , m/s)
power, shear stress and layer thickness ( , m) as follows:
and 5.12-mm gravel
concentration =
Can also be defined as the product of the
number of moving particles per unit area
( , m ), the particle volume ( , m ) and the
particle velocity ( , m/s), as follows
355 356

Bed load transport definition Bed load transport characteristics

Equation of motion
=

Defining the particle velocity as the ratio of the

saltation length and the saltation period ( =
/ ) it follows that

= ⁄ = =

where = = / =eroded or deposited

volume of particles per unit area
and time (m/s)
357 358

Equation of motion Equation of motion

where Total mass of the sphere can be represented by

m = particle mass and added fluid mass
= particle relative velocity to the
flow
u = local flow velocity where = added mass coefficient
= longitudinal and vertical particle velocities The submerged particle weight is given by
= longitudinal and vertical particle accelerations

359 360

60
Equation of motion Equation of motion

The lift force is caused by the velocity gradient

The drag force caused by pressure and viscous skin
(shear effect) and by the spinning motion of the
friction forces can be expressed as
particle

where where
= drag coefficient = lift coefficient ( = 1.6 for viscous flow)
= 1⁄4 = cross-sectional area of the sphere ⁄ = velocity gradient
361 362

Equation of motion Equation of motion

The lift force is caused by the velocity gradient boundary conditions to solve equation of motion are
(shear effect) and by the spinning motion of the
particle bed level position
initial velocities

where flow velocity distribution

= lift coefficient ( = 0.4 for viscous flow)
= angular velocity of the particle 363 = . ⁄ ∗ + . 364

Saltation height Saltation length

Approximated by (inaccuracy = 10%) Approximated by (inaccuracy = 50%)

. .
where = 0.3 ∗
= saltation height
= median particle diameter
= dimensionless particle
parameter
= effective shear stress
( in case of bed form)
, = critical bed shear stress
= kinematic viscosity
= ρs/ρ = relative density

365 366

61
Particle velocity Particle velocity
On a downward sloping plane bottom can be
= 1⁄6
derived from a force balance consisting of the drag
= angle of repose
force , the gravity force and the friction
= angle of local bottom slope
force exerted by the bottom on the particle
, = fluid and particle velocities
during contact
+ − =0 The force balance yields
where 4 −1 tan sin( − )
.
= −
= 1/2⁄ − 3 sin
= − sin
= − tan cos Assuming = ∗ and = , , it follows
= 1⁄4 367 that 368

Particle velocity Particle velocity

.
, sin( − )
= − Van Rijn (inaccuracy 10%)
∗ sin
.
= + . −
where ∗
∗

∗ = bed shear velocity ( ∗ in case of bed Inaccuracy 50%

forms) .
= .
, = critical mobility parameter of Shields − .

on a horizontal bottom Engelund and Fredsøe

= mobility parameter
.
, = coefficients
= −
∗
369 370

Deterministic bed load transport formulas Meyer-Peter and Mueller formula

Meyer-Peter and Mueller formula .
=8 − 0.047
Based on extensive experiments performed is a
where
laboratory flume with a cross-section 2 x 2 m and
length of 50 m = = dimensionless bed load
Uniform bed material as well as particle mixtures
transport rate
were used in the experiment
= = dimensionless particle mobility
= 0.4 to 29 mm
parameter
Slope = 0.0004 to 0.02 = volumetric bed load transport rate m ⁄s
= ℎ = current related bed load bed shear
Depth ℎ = 0.1 to 1.2 m
371
stress m ⁄s 372

62
 Meyer-Peter and Mueller formula Frijlink formula
= mean particle diameter (m)
. Essentially is an approximation of the formula of
= ⁄ = bed-form factor or efficiency
factor Meyer-Peter-Muller
= 18 log 12ℎ⁄ = overall Chézy coefficient
. . ⁄
m ⁄ ⁄ =5 ∗
= 18 log 12ℎ⁄ = grain related Chézy
⁄ ⁄ The parameters are identical to those of Meyer-
coefficient m s
ℎ = water depth (m) Peter-Muller.
= energy gradient
= effective bed roughness (m)
= ⁄ = relative density
373 374

Bagnold’s Approach Bagnold formula

Bagnold (1966) introduced the stream power applied shear stress: = ℎ
concept for sediment transport based on general
physics. −
gravity component: = sin
From general physics, the rate of energy used in
transporting materials should be related to the where
rate of materials being transported.
= = mass of bed-load particles per unit
area (kg/m )
= slope angle
= solid volume of bed-load particles per unit
area (m)
375 376

Bagnold formula Bagnold formula

= sediment density (kg⁄m ) Tangential stress at the bed resisting the moving
= fluid density (kg⁄m ) bed-load grains
ℎ = flow depth (m)
= acceleration due to gravity (m⁄s ) = cos tan

Normal stress on the bed by moving grains where

tan = dynamic friction coefficient

= − cos = cos

377 378

63
 Bagnold formula Bagnold formula
The applied bed shear stress (at the base of moving The work ( ) required to be done by the grain-
bed-load layer) consists of a dispersive grain-shear shear stress in moving the bed-load particles is
stress ( , ) and an intergrannular fluid bed-shear
stress ( , ) = , = − cos tan − tan
= , + , Let = (volumetric transport rate)
From equilibrium requirement at the bed
= , = − cos tan − tan
, + =
The available fluid energy per unit area and time is:
, + − sin = − cos tan
, = − cos tan − tan = = ℎ
379 380

Bagnold formula Bagnold formula

Bagnold assumed that: = = efficiency factor (0.1 - 0.2)
tan = 0.6 = dynamic friction coefficient
= tan = = bed slope
− ( − )
ℎ = water depth (m)
where = gravity acceleration m⁄s
= volume of bed-load per unit
= volumetric bed-load transport area m ⁄m
m ⁄s
= ℎ = overall bed-shear stress (N/m2)
= depth-averaged velocity (m/s)
= energy gradient
381 382

Bagnold formula Van Rijn formula

Assumption
Note
Motion of bed load particles is dominated by
Bed load transport is independent of the particle particle saltations
diameter ( )
Definition
Transport rate is related to the overall bed-shear Bed load transport rate ( ) is defined as the
stress ( ) and not to the effective bed stress ( ) product of the particle velocity , the
saltation height ( ) and bed load concentration
Effect of bed form is not taken into account
( ) resulting in
=
383 384

64
 Van Rijn formula Van Rijn formula
Bed load concentration represented by = 0.1 −1 . . . . . for ≥ 3
∗

= 0.18 where
∗

where = volumetric bed-load transport rate m ⁄s

= − , , = dimensionless bed-shear
= volumetric bed-load concentration parameter
= maximum volumetric concentration = 0.65 = 18 log 12ℎ⁄3 = grain related Chézy-
= 0.053 −1 . . . . . coefficient m ⁄ ⁄s
∗
ℎ = water depth (m)
for 200 ≤ ≤ 2000 µm , = particle diameter (m)
385 386

Van Rijn formula Van Rijn formula

= depth-averaged velocity (m⁄s) = ⁄ =

= effective bed-shear
, = critical bed-shear stress according to stress (N⁄m )
Shields (N⁄m ) = ⁄ = bed form factor
⁄
∗ = −1 ⁄ = dimensionless = overall Chézy-coefficient (m ⁄ ⁄s)
particle parameter = ⁄ = overall bed-shear stress
= ⁄ = relative density (N⁄m )
= sediment density (kg⁄m )
= fluid density (kg⁄m )
= kinematic viscosity coefficient (m ⁄s)
= acceleration of gravity (m⁄s )
387 388

Bed load transport at steep slopes Bed load transport at steep slopes
Slopes influence the local near-bed velocity where
, = bed load transport on a sloping bed
Change threshold conditions
(m ⁄s)
Change the transport rate once the sediment is
= bed load transport on a horizontal bed m ⁄s
in motion
Transport rate relationship between on a sloping = = Bagnold shape factor
±
and horizontal bed
(+ for upsloping flow, − for downsloping flow)
, =
= angle of repose (tan = 0.6)
= local bed slope
389 390

65
 Bed load transport at steep slopes Bed load transport at steep slopes
Smart formula = longitudinal slope
.
ℎ = water depth
=4 −
= Chézy-coefficient m ⁄ ⁄s
= acceleration of gravity m⁄s
for ≥ 400 µm, = 0.0004 to 0.2 = ⁄ .
= sediment gradation
where coefficient
=∑ = mean particle diameter
= ⁄ −1 = dimensionless bed-
, = characteristic bed material sizes
load transport rate
= critical mobility parameter of Shields
= ℎ ⁄ −1 = dimensionless
corrected for the slope effect ( =
mobility parameter
, )
= volumetric bed load transport rate m ⁄s391 392

Bed load transport of non uniform material Bed load transport of non uniform material
Non-uniformity will result in selective transport where
processes = bed load transport rate integrated over
size classes (fractions)
General approach is to divide the bed material in
= percentage of size class
a number of size fractions and apply an existing
= specific density
formula for each size fraction with a correction
= acceleration of gravity
factor to account for the non uniformity effect
= bed form factor
= mobility parameter of size class
. . . .
= 8( − 1) − = correction for factor of size class
= critical mobility parameter based on the
for ⁄ ≥5 average diameter
393 394

Bed load transport of non uniform material Vertical sorting

=∑ = mean particle diameter Sorting of bed material refers to the range of sizes
= number of size fractions present
(19)
= Perfectly sorted bed material would contain all of
19 ⁄
the same diameter, where as poorly sorted one
= 1 for = contains a wide range of sizes
Particle diameter of the transported bed-load
Vertical sorting occurs in the upper layer of the
particles can be expressed as
bed covered with bed forms

= , /
395 396

66
 Vertical sorting Horizontal sorting

Occasional transport processes from largest bed In non-uniform bed material an armour layer
form height will be formed when
 courser particles of bed material have critical
Finer particles dominate in the upper layer stresses which are larger than the mean bed-
shear stress
Armouring of the bed occurs when an immobile
 there is no supply of sediment upstream or
fraction is present in the top layer of the bed
when the supply from upstream consists of
finer material than that of the local bed
The fraction , of particles of size class of
the armour layer is given by:
397 398

Horizontal sorting Horizontal sorting

,
, =
∑ ,

The fraction , of particles of size class of , = critical

the eroded material is given by: bed shear stress
of particles with
(1 − ) , diameter from
, =
∑(1 − ) , size class
where
= ℎ
, = fractions of size class of original bed material
= probability that a particle from size class will
remain stable 399 400

5.2 Suspended load transport Suspended load transport

When the values of the bed-shear velocity exceeds
Less massive particles may be carried upwards into
the particle fall velocity, the particles can be lifted
the main body of the flow to be transported in
to a level at which the upward turbulent forces will
suspension as suspended load, possibly once the
be comparable to or higher than the submerged
second threshold has been reached.
particle weight.
Onset of suspended load transport, involving Movement in suspension is maintained against
principally the sand fraction, has been linked to gravity by turbulent eddies of variable strength
the turbulent properties of flow close to the bed of and direction so that particles do not follow
a stream, which are themselves related to the predictable paths.
topography of the bed. Particle in suspension mode is occasional and
401
random. 402

67
 Suspended load transport Suspended load transport
Once suspended, the fate of a particle depends on Depth-integrated suspended-load transport rate
the balance between its fall velocity and the ( ) is defined as the integration of the product of
vertical component of flow associated with velocity ( ) and concentration ( ) from the edge
turbulent eddying. of the bed-load layer ( = ) to the water surface
( = ℎ)
This interaction usually results in a vertical
distribution of suspended sediment in which both
the concentration and average grain size decrease =
with distance up from the bed.
Rate decrease depends on the ratio of the fall 1
velocity and the bed-shear velocity ( / ∗ ). or = ℎ = ℎ
ℎ
403 404

Suspended load transport Suspended load transport

with

= ⁄ℎ

where
= volumetric suspended load transport (m2/s)
= fluid velocity at height above bed
= sediment concentration (volume) at height
above bed
= depth-averaged fluid velocity
= reference concentration at height = above bed
ℎ = water depth Definition sketch for suspended sediment transport
= dimensionless shape factor 405 406

Suspended load transport Suspended load transport

Total suspended sediment discharge is obtained Sometimes, the suspended load transport is
from integration of the unit suspended sediment given as a mean volumetric concentration
discharge over the entire width of the channel, or defined as the ratio of the volumetric suspended
load transport (= sediment discharge) and the
= flow discharge

The suspended load defines the amount of =

sediment passing a cross-section in suspension
over a certain period of time, thus
Mean concentration ( ) is approximately
equal to depth-averaged concentration ( ̅) for
=
fine sediments
407 408

68
 Sediment concentration unit Sediment concentration Unit

Units used in the measurement of sediment Other units include kg⁄m (1mg/l = 1g/m ),
concentration vary with the range of the volumetric sediment concentration , the
concentrations and the standard measurement concentration in parts per million , and the
techniques utilized in different countries concentration by weight

Most commonly used unit for sediment Concentration can be expressed as

concentration is mg/l which describes the ratio
of the mass of sediment particles to the volume  volume concentration ( ) in m ⁄m
of the water–sediment mixture
sediment volume ∀
= =
409
total volume ∀ 410

Sediment concentration Unit Sediment concentration

 percent volume concentration after multiplying For low concentration < 10 kg⁄m it follows that
with 100% = s

 weight concentration ( ) in kg/kg  concentration in parts per million ( ) by

weight in ppm
sediment wight = 10
= = =
total weight 1+ −1
 corresponding concentration in mg/l and
in which = / = specific density = 2.65 1 mg⁄
⁄ = s = 10 1 mg⁄l s =
+ 1 − 10
411 412

Mass balance equation for suspended sediment Diffusion model

(diffusion model) where
= time averaged concentration
= concentration fluctuation
by definition = 0 in steady uniform flow
Instantaneous upward sediment velocity at level
1= −
Instantaneous sediment concentration at a Instantaneous downward sediment velocity at
certain level 1 = + ′ level 2 = +
Instantaneous sediment concentration at a where
certain level 2 = − ′ = vertical velocity fluctuation
413 414

69
 Diffusion model Diffusion model
by definition = 0 in steady uniform flow For instantaneous steady uniform flow the time-
Time averaged upward transport is : averaged upward and downward transport rats
are equal giving
= − +
=
= + wc − −
or
= cw −
c − c =0
Time averaged downward transport is :
represents the downward transport of
= + − sediment by gravity
= c − wc + −w c represents the upward transport of
= − + 415
sediment by turbulent velocity fluctuations 416

Diffusion model Diffusion model

Vertical transport of sediment by turbulence where
( ) is proportional to the vertical = time-averaged sediment concentration at
concentration gradient ( / ) height above bed
dc = particle fall velocity (constant)
cw = −
dz = mixing coefficient at height above bed
where Above equation is valid for low concentration
= diffusion or mixing coefficient because fall velocity is assumed to be constant
Fall velocity affected by
dc
c + = 0 advection–diffusion equation return flow of the displaced fluid
dz
particle collisions
417 418

Diffusion model Fluid and sediment mixing coefficient

particle induced turbulence 1
constant : ϵ = κ u∗ h
modified drag coefficients α
1 z
group effects ϵ = κ u∗ h
linear : α h
overall effect can be represented by
z z
, = (1 − ) parabolic : ϵ = κ u∗ h 1−
h h
where parabolic-constant :
= particle fall velocity in a mixture z z z
, ϵ = κ u∗ h 1− for < 0.5
= particle fall velocity in clear still fluid h h h
= coefficient (4 to 5 for particles of 50 to 500 µm) z
ϵ = 0.25 κ u∗ h for ≥ 0.5
419 h 420

70
 Fluid and sediment mixing coefficient factor
where
Describes the difference in the diffusion of a
∗ = current-related bed-shear velocity fluid “particle” (or small coherent fluid structure)
ℎ = water depth and a discrete sediment particle
= vertical coordinate
= Von Karman constant (0.4)
Assumed to be constant over the flow depth
, = coefficient
Usually, the diffusion of the sediment particles < 1 because sediment particles cannot fully
(ϵ ) is related to the fluid mixing coefficient of a respond to the turbulent velocity fluctuations,
clear fluid (ϵ ), as follows: which were assumed to be one-dimensional

ϵ = ϵ
421 422

factor factor
> 1 for two-dimensional eddy motion
because of the presence of centrifugal forces Expresses the influence of the sediment particles
acting on the particles (higher density) causing on the turbulence structure of the fluid
the particles to be thrown to the outside of the
eddies with a consequent increase of the Extremely important in the upper regime with
effective mixing length high concentration ( ̅ > 10 kg/m ) because
this results in stratification and hence damping
Can be expressed by the following expression of turbulence

w w
β = 1+2 for 0.1 < <1
u∗ u∗
423 424

Concentration profiles Concentration profiles

Expressions for the sediment concentration integration of convection-diffusion equation yields:
profiles in the lower regime (low concentrations
< 10 kg⁄m ) can be obtained by integration constant : ϵ ∶ = ( ⁄ ∗) ( )/
of the convection-diffusion equation
⁄ ∗
cw + ϵ =0 linear : ϵ ∶ =

Hindering settling effects and turbulence ⁄ ∗

ℎ−
damping effects can be neglected in the lower parabolic : ϵ ∶ =
regime (w , = w and = 1) ℎ−
425 426

71
Concentration profiles Concentration profiles
where
parabolic-constant :
= concentration at height above the mean
bed
⁄ ∗
ℎ− = reference concentration at height =
∶ = for < 0.5
ℎ− above bed
ℎ = water depth
= fall velocity in clear water
⁄ ∗
∶ = ⁄ ∗ ⁄ ⁄ . ∗ = bed-shear velocity
ℎ− = ⁄ ∗ = suspension (Rouse) number

for ≥ 0.5 parabolic concentration profile

= Rose concentration profile
427 428

Concentration profiles Concentration profiles

Definition sketch for reference concentration

429 430

Reference concentration and reference level Van Rijn

Flat bed Flat bed
Van Rijn Thickness of the bed-load layer can be estimated
The most logical assumption for the reference by the saltation height of the bed-load particles
level ( = ) of the sediment concentration . .
profile is the upper edge of the bed-load layer = 0.3 ∗
( = )
Layer-averaged sediment concentration in the
Reference concentration ( ) is defined to be bed-load layer is
equal to the bed-load concentration
. .
= 0.18 at = = 0.3 ∗
= at = = ∗
431 432

72
Van Rijn Einstein
Flat bed Flat bed
assumed that the thickness of the bed-layer was
where equal to 2
= volumetric bed-load concentration average particle velocity, = 11.6 ∗
= maximum volumetric concentration (= 0.65) using the definition = δ
= dimensionless bed-shear stress parameter
= = .
=2
∗ = dimensionless particle size parameter 23.2 ∗ 23.2
= thickness of bed-load layer where
= volumetric bed-load transport rate
∗ = bed-shear velocity
= dimensionless bed-load transport rate
433 434

Engelund and Fredsøe Engelund and Fredsøe

Flat bed Flat bed
assumed that the thickness of the bed-layer was where
equal to 2
= particle mobility parameter
= 0.65 (1 + ) =2 = ∗⁄ − 1

− − 0.26
. = particle mobility parameter at the
with = 4.3 initiation of motion (Shields)
∗ = ⁄ 6 + 2.5 ln ℎ⁄2.5 = bed-shear
.
0.26 velocity
= 1+
−
435 436

Zyserman and Fredsøe Smith and McLean

Flat bed Flat bed

assumed that the thickness of the bed-layer was

proposed an empirical function defined at
equal to the zero velocity layer
= =2 (flat bed = )
They found =3 = grain roughness
.
0.331 −
= .
1 + 0.72 − = + 26.3( − )

437 438

73
Smith and McLean Reference concentration and reference level
Flat bed
Flat bed
average sediment concentration (volume) in the
bed-load layer was proposed to be

= 0.004 =
1 + 0.004
where
= maximum (volume) concentration (= 0.6)
= − ⁄ = transport stage parameter
= thickness of bed-load layer (= ) 439 440

Reference concentration and reference level Van Rijn

Bed forms Bed forms
Suspended load particles are entrained from effective reference concentration is defined as:
dunes by: .
= 0.015 .
 maximum velocities developed at the ∗
upsloping part
where
 turbulence mixing in the trough = reference concentration (volume)
= reference level above the mean bed ( = or ½Δ)
∗ = dimensionless particle parameter
= dimensionless bed-shear parameter
441 442

Suspended sediment size in case of Suspended sediment size in case of

non-uniform bed material non-uniform bed material
In natural conditions the bed material is non- = [1 + 0.011( − 1)( – 25)] for 0 < < 25
uniform
= for ≥ 25
 usually, the suspended sediment particles are
considerably smaller than the bed material where

Van Rijn = representative particle diameter of suspended

sediment
has applied the size fraction method to
= median particle diameter of bed material
determine a representative suspended sediment = geometric standard deviation of bed material
diameter ( ) which accounts for non-uniformity = ½( 84/ 50 + 50/ 16)
effects = dimensionless bed-shear stress parameter
443 444

74
Suspended load transport rate Einstein formula
Einstein formula where
based on a parabolic distribution of the fluid = volumetric suspended load transport (m2/s)
mixing coefficient and a logarithmic distribution ∗ = bed shear velocity due to the grain (m/s)
of the velocity = reference concentration (volume)
30.2 ℎ = /(11.6 ∗ )
= 11.6 ∗ c + ln
= reference level (= 2d) (m)
ℎ = water depth (m)
1−
= = particle diameter (m)
1− = /ℎ = dimensionless reference level
= /ℎ = dimensionless vertical coordinate
1− = /( ∗ ) = suspension number
= 0.216 ln
1− = correction factor
445 446

Bagnold formula Bagnold formula

Based on energy balance concept relating
Per unit time the suspended load sinks over a
suspended load transport to the work done by
vertical distance equal to fall velocity
the fluid
Normal fluid stress ( ) supporting suspended To keep the load in suspension (all particles
load per unit area is: remain statistically at the same level above the
bed), the normal stress must do work (per unit
= −
time and area)
where
= −
= solid volume of suspended load per unit
area (m ⁄m )
447 448

Bagnold formula Bagnold formula

Volumetric suspended load transport is defined where
as = , giving
= depth-averaged velocity
= − cos = bed-shear stress
= efficiency factor related to suspended load
Work done per unit time and width by the (= 0.01 to 0.02)
longitudinal gravity force component to keep the = efficiency factor related to bed load (= 0.1
fluid in motion is . Part of this energy to 0.2)
available to transport the suspended load
yielding, Per unit time and width the suspended load is
moved in longitudinal direction over a distance
, = 1− equal to
449 450

75
Bagnold formula Bagnold formula

Work done per unit time by the longitudinal where

gravity component is qs = volumetric suspended load transport (m2/s)
ws = fall velocity of sediment (m/s)
, = − = − = depth-averaged velocity
τb = overall current-related bed-shear stress (N/m2)
Suspended load transport can be derived from tanβ = bottom slope
ρs = sediment density (kg/m3)
the energy balance = , + , , yielding ρ = fluid density (kg/m3)
es = efficiency factor related to suspended load (= 0.01
1− to 0.02)
= eb = efficiency factor related to bed load (= 0.1 to 0.2)
− − β = bottom angle with horizontal datum
451 452

Bijker formula Bijker formula

Based on the concept of Einstein, Bijker proposed: current-related bed-load transport rate (m2/s) is
expressed as:
= 1.83 [ + ln(33 ℎ⁄ )]
. ⁄
= ∗
where
where
= suspended sediment transport rate (m2/s) ⁄ .
= = bed form factor
= bed-load transport rate (m2/s)
= grain related Chézy-coefficient
, = Integrals according to Einstein formula
= 18 log 12ℎ⁄
= = reference level
= coefficient (1 to 5)
453 454

Van Rijn formula Van Rijn formula

suspended load transport can be computed from: .
/ℎ − ( /ℎ)
=
= ℎ 1 − /ℎ 1.2 −
where where
= + = modified suspension parameter
= ( /ℎ)
. .
= 2.5 for 0.01 ≤ ≤1
Transport rate of suspension with an inaccuracy ∗ ∗
of about 25% can be expressed as (0.3 ≤ ≤ 3
and 0.01 ≤ /ℎ ≤ 0.1) and 0.01 ≤ ≤ 0.1
ℎ
455 456

76
Van Rijn formula Van Rijn formula
Using regression analysis, the computational ℎ = water depth (m)
results for depth range of 1 to 20 m, a velocity = depth-averaged velocity (m/s)
range of 0.5 to 2.5 m/s and a particle range of When the bed load transport and the suspended
100 to 2000 µm were represented by load transport are known, the total load
. . transport of bed material can be determined by
− 1 summation ( = + )
= 0.012 .
ℎ −1 ℎ ∗
ratio of the suspended load and total load
where transport
= volumetric suspended load transport (m2/s) 1
= critical depth-averaged velocity according to shields
= =
+ ⁄ +1
457 458

Van Rijn formula Ratio of suspended to total load

for reasons of simplicity the bed load transport is The comparison of suspended load to bed load
defined as delineates which mode of sediment transport is
dominant
=
It is found that sediment transport can be
where
subdivided into three zones describing which
= reference concentration mode of transport is dominant:
= effective transport velocity of bed-load particles
= reference level (= bed-load layer thickness) (1) Bed load

1 (2) mixed load

=
1 (3) suspended load
+1
ℎ
459 460

Ratio of suspended to total load Ratio of suspended to total load

For turbulent flow over rough boundaries, Bed load is dominant at values of ∗⁄ less than
incipient motion corresponds to ∗ ⁄ ≈ 0.2 about 0.5
461 462

77
Ratio of suspended to total load Ratio of suspended to total load

Mixed load is found where 0.5 < ∗ ⁄ < 2 in When ∗ ⁄ < 2, the ratio of the suspended load
which both the bed load and the suspended load to the bed load is approximately equal to
contribute to the transport 463 ∗⁄ 464

Ratio of suspended to total load Ratio of suspended to total load

The bed load and suspended load are In the case of mixed load, ∗ ⁄ < 2, the total
approximately equal when ∗ = load will be less than 5 times the bed load
465 466

Ratio of suspended to total load Ratio of suspended to total load

Suspended load is dominant when ∗ ⁄ > 2, and

gravitational effects on the particles are negligible
compared to turbulent mixing as ∗ ⁄ becomes
very large 467 468

78
 5.3 Hyperconcentrations Velocity profile and flow resistance

Refer to heavily sediment-laden flows in which Based on the Boussinesq hypothesis, the shear
the presence of fine sediments materially affects stress at height in a fluid-sediment mixture
fluid properties and bed material transport flow is defined as:

In general, the volumetric sediment concentration = , +

of hyperconcentrations ranges from 5–60%
where
The mass density of hyperconcentrations is
calculated from = fluid shear stress at height above bed
, = mixing coefficient in fluid-sediment
= + − mixture
469 470

Velocity profile and flow resistance Velocity profile and flow resistance
= kinematic viscosity coefficient in fluid- = , ℎ−
sediment mixture
= ⁄ = ℎ
= local velocity at height above bed
,
= [1 + − 1 ] = density of mixture at = 1− = , 1− ∗
ℎ ℎ
height
= sediment concentration at height above where
bed
= energy gradient
= dynamic viscosity coefficient ℎ = flow depth
̅ , = 1 + ( − 1) ̅ = mean density of layer above
Shear stress at height and bed-shear stress can
height to surface
be expressed as: ̅ = 1 + ( − 1) ̅ = mean density over depth
471 472

Velocity profile and flow resistance Velocity profile and flow resistance
̅ = mean concentration of layer above height
to surface In a fluid-sediment mixture the turbulence
̅ = mean concentration over depth structure and hence velocity profile and the flow
= specific density resistance are modified
= sediment density
= fluid density Analysis of measured velocity profiles in
sediment suspensions shows a decrease of the
Velocity gradient can be obtained from above near-bed velocities and an increase of the near-
equations: surface velocities
(1 + − 1 ̅ )(1 − /ℎ) ∗
=
1+ −1 ( , + )
473 474

79
Measured velocity profile in high-concentration Velocity profile and flow resistance
flows Defining the flow resistance as:
1
=
8
where
= bed-shear stress
= friction factor of fluid-sediment flow
= mixture density = (1 + – 1 ̅)
̅ = depth-averaged concentration
= fluid density

475 476

Velocity profile and flow resistance Velocity profile and flow resistance
both the mixture density and friction factor are No change of friction factor
in fully rough regime
important
Small decrease entering the
Increase concentration transition from the rough
regime (decreasing ) at
• Increase in mixture density the same relative roughness
( ⁄ )
• Increase of flow resistance
Increase for decrease of in
smooth regime

Small decrease by transition from smooth to laminar regime

Increase for decreasing in laminar regime

477 478

Velocity profile and flow resistance Velocity profile and flow resistance
Stratification phenomena affecting the velocity Ippen considered the effect of turbulence
profile and the flow resistance can be described as damping by viscous dissipation and mixing
against gravitation forces
 damping of turbulence because of turbulence energy
is consumed in keeping the particles in suspension
(1 + − 1 ̅)
(momentum transfer from fluid to particles) =
1 + 2.5
 damping of turbulence by increased viscous where
dissipation
= bed-load concentration (volume)
 reduction in mean flow energy by transfer of
momentum from the fluid to the particles (in the Based on energy considerations and analysis of
longitudinal direction) to keep the solid mass in velocity profiles measured in the Yellow river,
motion Wang found:
479 480

80
 Velocity profile and flow resistance Velocity profile and flow resistance
1.14 −1 ( , − ) ̅ = 0.33 + 0.11 ⁄ ∗ = zero velocity level
= + =3 = effective grain roughness
∗
= ∗⁄ − 1 = mobility parameter
where
= kinematic viscosity coefficient of fluid-
, = fall velocity of suspended sediment in sediment mixture (≈ 10 m ⁄s)
a fluid-sediment mixture (based on ̅)
Velocity distribution given by Van Rijn is: Itakura-kishi and McLean assumed constant -
value equal to the clear fluid value
= ln( ⁄ )
−1 + ln( ⁄ ) Effect of suspended sediment on the velocity
where profile was modeled by introduction of Monin-
= depth-averaged velocity Obukhov length scale ( ):
481 482

Velocity profile and flow resistance Velocity profile and flow resistance

=
∗ Velocity profiles are expressed as:
−1 ̅
1 ∗ ∗
where = 5.5 + 7 + <5
̅ = depth-averaged concentration ∗
= fall velocity in clear water 1 ∗ ∗
= Von Karman coefficient in clear water (= 0.4) = 8.5 + 7 + > 70
∗
Velocity gradient is given by
where
∗
= 1+
= viscosity coefficient of mixture
-coefficient = 7, based on analysis of measured = (1 + 2.5 )
velocity profiles in suspension ( ̅ < 0.65) = bed-load concentration
483 484

Velocity profile and flow resistance Concentration profiles and transport rates
Depth-averaged velocity can be obtained by In the upper regime with high or
integration over the depth: hyperconcentrations the effects of the
suspended sediments on the density, viscosity,
7ℎ 1 fall velocity and turbulence structure can not be
∗ℎ ∗
=3+ + ln for < 5 neglected
∗ 2
Einstein and Chien used the following set of
equations to compute concentration profile
7ℎ 1 ℎ ∗
=6+ + ln for > 70
∗ 2
, + =0

, = (1 − )
485 486

81
Concentration profiles and transport rates Concentration profiles and transport rates
Itakura and Kishi introduced the Nonin-Obukhov
= length scale and proposed:
1 + ( − 1) ̅ ( ⁄ )
⁄( ∗)
Ippen used a modified velocity gradient to ℎ−
compute the mixing coefficient distribution. The =
ℎ−
velocity gradient is modelled as:
where
∗ ⁄ℎ − = reference concentration at height = above bed
= = correction factor = 7 ℎ⁄
⁄ℎ − ln( ⁄ℎ)
= suspension number = ⁄( ∗ )
-parameter is an empirical parameter which = von Karman constant in clear water (= 0.40)
was found to be dependent on the -value and = fall velocity in clear water
parameter ℎ = water depth
487 488

Concentration profiles and transport rates Concentration profiles and transport rates
Van Rijn assumed that: where
= = = ⁄( ∗ )
where = stratification correction parameter
= fall velocity suspended sediment transport in
= damping factor dependent on concentration
clear water
= ratio of sediment and fluid mixing
= Von Karman constant (= 0.4)
. .
= + − > . > . / = ratio of sediment and fluid mixing

Van Rijn simplified method using modified Van Rijn modified suspension number must be
suspension number : used together with parabolic-constant fluid
mixing coefficient
= +
489 490

Concentration profiles and transport rates 5.4 Total Load Transport in Steady and Uniform
-parameter represents the turbulence damping Currents
and the hindered settling effect
Prediction Methods
Based on analysis of available -values, the Make distinction between bed and suspended-
following relationship was obtained (for > load transport
0.001 or > 2.5 kg/m3)
. . methods of Einstein, Bagnold, Bijker, Van Rijn
= . . ≤ ≤ . ≤ ⁄ ≤ .
∗ ∗
Directly give the total load transport rate
where
methods of Engelund-Hansen, Ackers-White,
= reference volume concentration Yang
= maximum volume concentration (= 0.65)
491 492

82
Einstein, Bagnold, Bijker, Van Rijn Engelund-Hansen
Total load transport of bed material particles can Based on energy balance concept
be obtained by summation of the bed load and Work (per unit time and width) required to
suspended load transport elevate a sediment load over a height equal to
the bed form height Δ is:
= +
= − ∆
Work (per unit time and width) done by the fluid
on moving the particles over a length equal to
the bed form length λ is:

= − , ∗
493 494

Engelund-Hansen Engelund-Hansen
Energy balance = yields = overall bed-shear velocity
∗

− 1 = friction coefficient = 2 ⁄
,
= ∗ = Chézy-coefficient
− ∆
,Δ = bed-form length and height (m)
where
= volumetric total load transport ⁄ Based on data analysis, ⁄Δ found to be
approximately constant, giving:
= effective bed-shear stress
, = critical bed-shear stress
= − ∗
= particle diameter ( )
495 496

Engelund-Hansen Engelund-Hansen
.
Using dimensionless transport parameter =

= -coefficient was determined by data fitting

−1 (approximately 100 flume data) giving = 0.1
.
. 0.1
= − =

Using = 0.06 + 0.4 , and assuming where

= 0.06, it follows that
=
− = 0.4 for lower regime −1
497 498

83
Engelund-Hansen Yang
∗ ℎ Assumed that the sediment transport is related
= = =
−1 − −1 to the unit stream power, defined as:

2 Total sediment concentration ( ), defined as

=
the ratio of the sediment and fluid discharge
per unit width, ( = ⁄ ), was expressed as:
rearranging:

0.05 −
=
∗ log = + log
−1 .

499 500

Yang Yang
Analysis of flume and field data resulted in: = energy gradient
= median particle diameter of bed material (m)
= 5.435 − 0.409 log ⁄ − 0.457 log ∗⁄
= fall velocity (based on of bed material)
= 1.799 − 0.409 log ⁄ − 0.314 log ∗⁄ (m/s)
∗ = bed-shear velocity (m/s)
where
= total load concentration in parts per million 2.5
= + 0.66
by weight (ppm) log ⁄ − 0.06
= depth-averaged velocity (m/s)
= depth-averaged velocity at initiation of ∗
for 1.2 < < 70
motion (m/s)
501 502

Yang Ackers-White
= 2.05 for ∗ Applied dimensional analysis to express mobility
≥ 70
and sediment transport rate in terms of some
Total transport rate (in kg/sm) is given by: dimensionless parameters
Based on analysis of 925 sets of flume and field
= 10 ℎ data, the following empirical formula was
where proposed
= total load transport rate (kg/sm) −
ℎ = water depth (m) =
∗

503 504

84
Ackers-White Ackers-White
where Their mobility parameter for sediment transport is
= total load transport (m3/s)
= depth-averaged velocity (m/s) ∗
= .
−1 5.66 log(10ℎ⁄ )
∗ = bed-shear velocity (m/s)
= particle mobility parameter
/
, , = coefficients −1
∗ =
= kinematic viscosity coefficient (m2/s)
= specific density
. .
= representative diameter of bed material (m) = 10 ∗ ∗

for 1 < D∗ < 60

505 506

Ackers-White Ackers-White
Revised and coefficients (HR Wallingford):
= 1 − 0.56 ( ∗) for 1 < D∗ < 60 . . .
= 10 ∗ ∗

9.66 for 1 < D∗ < 60

= + 1.34 for 1 < D∗ < 60 = 0.025 for D∗ ≥ 60
∗
6.83
0.23 = + 1.67 for 1 < D∗ < 60
Y = .
+ 0.14 for 1 < D∗ < 60 ∗
∗
= 1.78 for D∗ ≥ 60
= 0.025, = 0, = 1.5, = 0.17 Revision of the original coefficients is necessary
because it predicted transport rates which were
for D∗ ≥ 60 considerably too large for relatively fine
507
sediments ( < 0.2 mm) 508

6. Sediment transport in non-steady and non- Sediment transport in non-steady and non-
uniform currents uniform currents
Sediment transport capacity of a stream is
defined as the quantity of sediment that can be
carried by the flow without net erosion or
deposition
In non-steady and non-uniform flow the sediment
transport may be smaller (underload) or larger
(overload) resulting in net erosion or deposition
assuming sufficient availability of bed material Modeling approach depends on the relative rate
(no armour layers) of adjustment of the transport of sediment
509
particles to the new hydraulic conditions 510

85
 Sediment transport in non-steady flow Sediment transport in non-steady flow
River flood wave is a relatively slow process with Bed-load transport
a time scale of a few days Adjustment proceeds almost instantaneously for
Consequently, the sediment transport process in the transport taking place close to the bed
river flow can be represented as a quasi-steady Suspended-load transport
process
Adjustment proceeds relatively slowly because it
Therefore, the bed-load transport formulae and takes time and hence distance for particles to
the suspended sediment transport formulae for settle out from the suspension or to be mixed
steady and uniform flow case can be applied for into the flow, depending on ∗ ⁄
transport rate predictions
Van Rijn studied the adjustment of suspended
Sediment transport will adjust to the new load transport using two-dimensional vertical
hydraulic conditions 511 mathematical model 512

Sediment transport in non-steady flow Sediment transport in non-steady flow

Suspended-load transport Suspended-load transport
Adjustment length was defined as the length where:
after which the suspended load transport differs
= adjustment length
less than 5% from the new equilibrium suspended
ℎ = water depth
transport
= actual suspended load transport
Adjustment length in dimensionless form is given , = equilibrium suspended load transport
as: = fall velocity
∗ = bed-shear velocity
= , , = effective bed-roughness height
ℎ , ∗ ℎ
513 514

Sediment transport in non-steady flow Adjustment length of suspended sand transport

Suspended-load transport
⁄ℎ – parameter was held constant at a value
of 0.01

computer runs for the overloading and

underloading conditions were made

results were presented in graphical form

515 516

86
 Non-uniform suspended-load transport Non-uniform suspended-load transport
Generation of concentration profiles in a clear
flow (without initial sediment load) entering a
channel with an erodible channel bed

517 518

Non-uniform suspended-load transport Non-uniform suspended-load transport

Dimensionless height above bed, z/h

Dimensionless height above bed, z/h

Sediment concentration (ppm) 519 Sediment concentration (ppm) 520

Non-uniform suspended-load transport Transport of fine sediments across a trench or

channel perpendicular to the flow

flow velocity (m/s)

Dimensionless height above bed, z/h

height above bed, z

Sediment concentration (ppm) 521 concentration (ppm) 522

87
Transport of fine sediments across a trench or Transport of fine sediments across a trench or
channel perpendicular to the flow channel perpendicular to the flow

flow velocity (m/s) flow velocity (m/s)

height above bed, z

height above bed, z

concentration (ppm) 523 concentration (ppm) 524

Aggradation and Degradation of Alluvial- Aggradation and Degradation

Channel Beds
For various reasons, the sediment supply of a
Aggradation is general and progressive (long- stream may be increased or decreased beyond
term) buildup of the longitudinal profile of a the local transport capacity with the consequent
channel bed due to sediment deposition. change in bed elevation.

Degradation is general and progressive lowering In many cases, dams or other structures may
of the channel bed due to erosion, over relatively decrease the rate of supply to zero permanently
long channel length. or for considerable period of time.

Aggradation or degradation occurs over large Downstream of the structure, the bed will be
lengths and both are slow processes. degraded because of the clear stream will pick up
a new equilibrium load.
525 526

Aggradation and Degradation Degradation of Alluvial-Channel Beds

The rate of supply at the structure is zero but the If the rate at which sediment entering a given
transport rate is appreciable; therefore, the bed reach of the stream is less than that at which it is
scours. going out, the excess sediment will be picked up
from the bed and banks, and there will be
Beyond this point, the local load that the stream lowering of bed level unless the bed is non-
has picked up becomes the supply of the erodible.
succeeding section.
Thus for degradation to occur
The rate of change of the bed elevation then
becomes proportional to the rate of change of the
sediment discharge in the downstream direction. >0
527 528

88
 Degradation of Alluvial-Channel Beds Degradation of Alluvial-Channel Beds

Degradation occurring in a stream can proceed If an increase in slope is imposed at the d/s end,
either in the downstream or in the upstream u/s , and remaining the same, upstream
direction depending on the basic cause of progressing degradation will result.
degradation.
The composition of bed material of the degrading
If the reduction in the slope is caused by stream plays an important role in the process of
reduction in yearly bed material discharge , degradation.
reduction of bed material size or increase in
yearly water discharge at the upstream end,
downstream-progressing degradation will occur.

529 530

Degradation of Alluvial-Channel Beds Degradation of Alluvial-Channel Beds

Rotational degradation
If the bed material is uniform and the stream
slope is relatively large, the flow deficient of
sediment load picks up more sediment from
upstream reaches and relatively less from the
downstream reach.
As a result, stream slope reduces by rotation of
stream bed about some downstream control
where the water level is held constant.
Upstream and downstream progressing degradation531 532

Degradation of Alluvial-Channel Beds Degradation of Alluvial-Channel Beds

Rotational degradation Parallel degradation
Takes place by the removal of finer particles at
essentially a constant slope
Occurs if the bed material of the degrading
stream is non-uniform
Downstream-progressing degradation
Degradation downstream of high dams and
barrages
Increase in water discharge
Definition sketch of rotational degradation 533
Gravel mining 534

89
 Degradation of Alluvial-Channel Beds Degradation of Alluvial-Channel Beds
Downstream-progressing degradation Upstream-progressing degradation
Storage of bed material If a tributary is joining such a degrading stream,
Degradation at channel bifurcations the tributary also experiences upstream-
progressing degradation.
Upstream-progressing degradation
If the water level of the lake or the sea into which Execution of cut-off in a meandering river causes
the river discharges falls, an increased water increase in the bed slope in the cut-off leading to
surface slope is imposed on the river. degradation upstream of the cut-off and
aggradation downstream.
Hence, the river picks up material from the bed to
fulfill its increased transport capacity and
degradation occurs. 535 536

Degradation of Alluvial-Channel Beds Degradation of Alluvial-Channel Beds

Upstream-progressing degradation Control of degradation
In recent years, three methods have been tried to
control degradation
Artificial sediment feeding
When the stream is degrading due to deficiency
in sediment load, the degradation can be reduced
or arrested if properly estimated sediment load of
known sizes is fed every year on regular basis

537 538

Degradation of Alluvial-Channel Beds Degradation of Alluvial-Channel Beds

Control of degradation Control of degradation
Artificial armouring Construction of weirs or dams
Artificial armouring means formation of a If a weir or dam is constructed in a reach which is
complete cover on the bed with a layer of coarse degrading because of an upstream dam, it creates
material which is capable of resisting the shear backwater and reduces velocity, thereby reducing
exerted by the flow degradation
To stop washing away of fine material It can also cause aggradation which can offset the
underneath, a filter of graded material or degradation due to the dam upstream
geotextile can be used Construction of a series of dams also moderates
The armour thickness should be 0.8 to 1.0 m the flood, thereby reducing the transport capacity
539 540

90
 Local score near structures Scour around bridge pears
Scour is the local lowering of the stream bed Majority of bridge failures were due to scour of
around a hydraulic structure. foundation material.
Scour takes place around bridge piers, abutments, Such failure is primarily due to three causes:
spur dikes and breakwaters due to modification of
i. Inadequate knowledge about scour
flow pattern causing increase in local shear stress
phenomenon when the bridge was
which, in turn, leads to removal of material and
constructed.
hence scour.
ii. Inadequate data and knowledge about
Clear water scour design flood.
no upstream sediment transport <
iii. Increase in the loading on bridges due to
Live bed scour increase in the size of trucks and wagons and
upstream sediment transport > 541
frequency of loading. 542

Scour around bridge pears Scour around bridge pears

The total lowering of stream bed at any site can 4. Additional lowering of bed level can take
take place due to four reasons place due to concentration or non-uniform
flow distribution across the river width at the
1. Degradation taking place at bridge site due to
bridge
dam upstream
Earlier studies have indicated that depending on
2. Reduction in width in the vicinity of the
the type of pier and free stream conditions, an
bridge by providing embankments and guide
eddy structure comprising all or anyone or none
bunds can lead to lowering of bed level
of the vortex systems can form
3. Lowering of bed level that takes place due to
modification of flow pattern These include horseshoe vortex system, the wake
vortex system, and/or the trailing-vortex system
543 544

Scour around bridge pears Scour around bridge pears

Based on analysis of field and flume data,
Breusers et al. have found for a single pier in
uniform bed material:
ℎ , =
where
ℎ , = maximum scour depth below original river bed
= width of pier in plane of river cross-section
= coefficient related to ⁄
= coefficient related to ℎ ⁄
= coefficient related to shape of pier
= coefficient related to angle of attacking flow
545 546

91
 Scour around bridge pears Scour around bridge pears
ℎ , = ℎ , =

= depth –averaged flow velocity upstream of pier = 1 for flow normal to bridge piers
= critical depth-averaged flow velocity (upstream) = 1.3 for flow under angle of 15o and length-
ℎ = flow depth (upstream) width ratio of 4
=0 for ⁄ < 0.5 (no upstream transport) = 2 for flow under angle of 15o and length-
= 2( ⁄ - 0.5) for ⁄ = 0.5 to 1.0 (no upstream width ratio of 8
transport)
= 1 for ⁄ ≥ 1 length of the score hole ≈ 1 (upstream of the pier)
= 2 tanh ℎ⁄ yielding α2 = 2 for h/b ≥ 3
≈ 5 (downstream of the pier)
α2 = 1.5 for h/b < 1
= 1 for circular and round-nosed piers width of the score hole ≈ 2 on each side of the pier
= 0.75 for streamlined piers
= 1.3 for rectangular piers
547
where = pier width 548

Scour near groynes, spur-dikes, bridge Scour near groynes, spur-dikes, bridge
abutments abutments
Based on analysis of field and flume data,
Breusers proposed:
⁄
ℎ , = −ℎ
where
ℎ, = maximum scour depth near groyne
ℎ = mean water depth of contracted section before
scour
= discharge per unit width in contracted section
= coefficient depending on groyne and river
geometry ( ≈ 2 for straight channel and groyne
549
normal to bank) 550

Scour near groynes, spur-dikes, bridge Scour near groynes, spur-dikes, bridge
abutments abutments
Another approach is to assume that the cross- Scour near a short groyne will be considerably
sectional area of the contracted section smaller than according to the above equation
ultimately will be equal to that without the
groyne ℎ , = 0.5 to 1.5ℎ for = 1 to 3ℎ

= ℎ Scour downstream of weirs and barrages

Assuming that ≈ 1⁄3 ℎ , for a long
groyne ( > 10ℎ ) and ≈
ℎ , ≈ 3ℎ for > 10ℎ
551 552

92
Scour downstream of weirs and barrages Scour downstream of weirs and barrages
Based on experimental research in flumes, the Time scale was found to be:
time-dependent development of the scour depth .
in clear water flows was found to be: 330 −1 ℎ
= .
−
ℎ ( ) .
= in which:
ℎ
in which: = depth-averaged velocity just upstream
( = 0) of score hole
ℎ = maximum depth at time = critical depth-averaged velocity
ℎ = upstream water depth (initiation of motion)
= time at which ℎ = ℎ = specific density ( / )
553 554

Scour downstream of weirs and barrages Aggradation of Alluvial-Channel Beds

= coefficient depending on flow and turbulence In general, aggradation in a stream takes place
structure at the upstream end of scour hole when the stream is carrying more sediment than
(= 1.5 for two-dimensional flow without its transport capacity
structure,= 3 for very violent three- If the sediment load coming to a reach in a given
dimensional flow) time is greater than the sediment load going out
Generally accepted formulae for the maximum in the same time, i,.e. when
scour depth in the equilibrium situation are not
available <0
Rough estimate can be obtained from: the excess sediment gets deposited and the bed
level rises i.e.
−
ℎ , = ℎ >0
555 556

Occurrence of aggradation Occurrence of aggradation

Increase in sediment load Aggradation due to reduction in water slope (or
increase in water level)
Withdrawal of clear water
Aggradation at channel bifurcation

557 558

93
 Sediment deposition Deposition in channels
Deposition begins once the flow or shear velocity deposition in channels may occur due to:
falls below the settling velocity of particles, which
for a given particle size is less than that required geological process (time scale of centuries)
for entrainment.
shifting of shoals and banks (time scale of
decades)
reduction of local transport capacity (short
term effect)
Settling process is dominant in the downsloping
(deceleration) section and the middle section of
the channel
Erosion and deposition criteria 559 560

Deposition in channels Deposition in channels

Sediment pick-up from the bed into the
accelerating flow is dominant in the upsloping
(acceleration) section
Most relevant processes in the deposition/
erosion regions are:
Convection of sediment particles by the
horizontal and vertical fluid velocities
Mixing by turbulence and orbital motions
Settling of the particles due to gravity
Pick-up of the particles from the bed by
current and wave-induced bed-shear stresses561 562

Deposition in channels Deposition in channels

Assuming an exponential behaviour of the
decreasing suspended sediment transport, the
expression for the local suspended sediment
transport rate ( , ), in the channel is:

, − ,
= 1−
, − ,

in which:

, = incoming suspended sediment transport

per unit width
563 564

94
 Deposition in channels Deposition in channels
, = equilibrium suspended sediment Mayor-Mortensen-Fredsøe Formula
transport per unit width in the channel
Suspended sediment transport in the channel is
= width of streamtube of approaching flow
expressed as:
= width of streamtube in channel
= longitudinal coordinate ℎ
, = , exp − + , 1−
= coefficient ℎ

in which:
ℎ = water depth upstream of channel
ℎ = water depth in the channel
= ⁄ = coefficient
565
= particle fall velocity 566

Deposition in channels Deposition in channels

Mayor-Mortensen-Fredsøe Formula Fredsøe (1978) proposed a method to compute
the deposition due to bed-load transport in a
= depth-averaged mixing coefficient in channel for the special case of a current parallel
channel (= 0.085 ∗, ℎ ) to the channel axis
= longitudinal coordinate
The deposition rate (Δ ) per unit channel length
due to the suspended load transport is Drag and gravity forces
ℎ considered
, 1 − exp −
ℎ sin Resultant force inclined
∆ = sin
towards the bottom causes
− , 1 − exp − deposition
sin
567 568

Deposition in channels Deposition in channels

Assuming a small variation of the bed-load = dynamic friction angle (generally taken 25°)
transport over the side slope, the gravity
deposition is: = = time-scale parameter ( )
. = angle of initial side slope of the channel
, . .
∆ =2 + − = channel depth
tan
in which: Lean Formula
Δ = solid volume of sand transported from the suspended sediment transport in the channel is:
two banks into the channel during a period
of seconds ( ⁄ )
, = , − , − , 1 − exp 1 −
, = longitudinal bed-load transport at middle of
slope ( ⁄ ) 569 570

95
 Deposition in channels Deposition in channels
Lean Formula Lean Formula
in which: The deposition rate (Δ ) per unit channel length
due to the suspended load transport is:
=
∆ = , − , 1 − exp − sin
ℎ sin
= 1−
−1 + exp ℎ ⁄
Eysink-Vermaas Formula
= depth-averaged mixing coefficient upstream
of channel Suspended sediment transport in the channel is:
= depth-averaged velocity upstream of
channel , = , − , − , 1 − exp −
571
ℎ 572

Deposition in channels 7. Watershed Erosion and Yield Modeling

Eysink-Vermaas Formula Concepts
Prediction models play an important role in
in which:
prediction – meeting practical needs of soil
.
2 conservation goals: assessment, prediction,
= 0.015 1+ 1 + 4.1
∗, ∗, ℎ regulating programs
explanation – advancing the scientific
The backfill rate (Δ ) per unit channel length is: understanding of processes
Models are used wherever the costs or time
∆ = , − , 1 − exp − sin involved in making measurements are prohibitive
ℎ sin
Used for on-site concerns, off-site concerns, or both
573 574

7.1 Modeling Watershed Erosion and Yield Modeling Watershed Erosion and Yield
Modeling water and wind erosion is important Modeling permits the:

to understanding the processes governing soil 1) Understanding of the driving processes

erosion 2) evaluation of on-site and off-site impacts on
predicting runoff and soil erosion rates soil productivity and water and air pollution
on large scale
identifying or choosing appropriate measures
3) identification of strategies for erosion control
of erosion control.
4) assessment of the performance of soil
conservation practices for reducing water and
wind erosion.
575 576

96
 Watershed Watershed
An area of land that captures water in any form, Consist:
such as rain, snow, or dew, and drain into a channel network and the contributing or
common water body, i.e., stream, river, or lake interchannel areas, or
Surface drainage area above a specified point on stream channel and the upland or upstream
a stream enclosed by a topographic boundary or areas and adjacent or lateral overland flow areas
perimeter
Contributing area is called the watershed, the
Boundary is defined by the higher elevations or drainage basin or area, or the catchment
ridges that define which direction the rainwater All land is part of the watershed of some creek,
will flow stream, river or lake
577 578

Watershed Delineation Watershed Delineation Example

1. Identifying point of interest, i.e., the
downstream-most location
2. Drawing a line perpendicular to the counters
3. Picking the high points on a topographic map
4. Continuing until returning back to the point of
interest
Watershed algorithm: an image processing
segmentation algorithm that splits an image into
areas, based on the topology of the image
579 580

Point of Interest Erosion

Depending upon the scale of investigation and Processes by which soil particles are detached,
definition of the problem, point of interest can be: transported, and deposited by raindrop impacts,
by runoff on the soil surface, and by runoff in rills,
a position on a hillslope
concentrated flow areas, and stream channels
a property boundary at a construction site
Two-phase process consisting of the detachment of
the edge of a farm field individual particles from the mass and their
transport by erosive agents such as running water
delivery point to a stream channel
and wind
etc.
In different situations, the major processes leading
581 to sediment detachment will differ 582

97
 Erosion Sheet Erosion
There are four main types of erosion processes: Refers to the uniform detachment and removal of
soil, or sediment particles from the soil surface by
Sheet erosion
overland flow or raindrop impact evenly
Rill erosion distributed across a slop
Gulley erosion Together with rill erosion, sheet erosion is often
classified as ‘overland flow’ erosion, detaching
In-stream erosion
sediment from the soil surface profile only

For purpose of simplification, the two processes

are often considered together in erosion modeling
583 584

Rill Erosion Gully Erosion

Occurs when water moving over the soil surface Describes channels of concentrated flow that are
flows along preferential pathways forming an too deep to be obliterated by cultivation
easily recognizable channel
Gully flows differ from sheet and rill flows in that
These rill are generally small erosion features, and raindrop impact is not an important factor in terms
have been defined as being ‘flow channels that can of flow resistance or in sediment particle
be obliterated by tillage’ detachment
Rill initiation is controlled by the cohesive strength
of the soil and the shear forces exerted on the soil Identified as being major source of sediment
entering the waterways as they usually have high
Flow in rills acts as transporting agent for the delivery ratios when well connected to streams
removal of sediment downslope 585 586

Gully Erosion In-Stream Erosion

Two main states in gully development: Involves the direct removal of sediment from
1. Unstable gully initiation period stream banks (lateral erosion) or the stream bed
Sediment also enters the stream due to slumping
hydraulic erosion is predominant at the gully
of the stream bank resulting from bank erosion
bottom and rapid mass movement is
undercutting the stream bank
occurring
During high flow periods, a large proportion of the
2. period of stable sediment transport and
sediment that is transported through the stream
sedimentation at the gully bottom
network can originate from the stream channel
gully width is increasing due to lateral
These erosion types do not necessarily occur in
erosion and slow mass movement
587
isolation 588

98
Detachment, Transportation, and Deposition Detachment, Transportation, and Deposition
Soil particles are detached when the impact of Sediment particles are transported by raindrop
raindrops or the erosive force of flowing water is splash and by overland flow
in excess of the ability of the soil to resist erosion
Deposition of soil particles occurs when the weight
Particles detached in the interrill areas move to of the particle exceeds the forces tending to move
the rills by splash mechanisms and as a result of it
suspension and saltation in overland flow
sediment load > sediment transport capacity
The amount and rate of water and sediment
delivered to the rills determine rill erosion rates, Particles detachment and movement is
sediment transport capacity in rills, and rate of independent of processes in rill and stream
sediment deposition channels
589 590

Detachment, Transportation, and Deposition

591 592

Sediment Yield Sediment Yield

Final and net result of detachment, transport, and The bulk of the sediment is deposited at
deposition processes occurring from the watershed intermediate locations wherever the entraining
divide down to the point of interest where runoff waters are insufficient to sustain transport
sediment yield information is needed It is scattered to adjacent downslope positions
Not all of the eroded material is effectively sluiced It is deposited at the base of eroding slopes and
through the river systems and delivered to the sea at the bluff line bordering major river valleys

It overlays flood plains

It clogs stream channels and reservoirs

593 594

99
 Sediment Yield Sediment Yield
The percentage of sediment delivered from the The change (per unit area) of this downstream
erosion source to any specified downslope sediment movement, from the source to any given
location is affected by such factors as measuring point, is expressed by the delivery
ratio, , defined by
size and texture of erodible material
= 100
climate
where
land use = sediment yield at the measuring point
general physiographic position = total material eroded (gross erosion) from
the watershed and drainage system
595 upstream from the measuring point 596

Sediment Yield Watershed Processes Description

Values of and are given in units of mass per In terms of processes occurring
unit area per unit time (e.g. ton/ha/y)
In common usage, the sediment delivery ratio is on upland areas,
considered a percentage, of 100( / )
in small stream channels, and
The headwater regions yield almost all of the
sediment transported downstream, and channels over entire watersheds
and slopes in downstream areas contribute very
little sediment
Yield is controlled more by sediment supply than
transport capacity 597 598

Watershed Processes Description Upland Areas

as consisting of Upland processes
Zone 1 – drainage basin as a sediment and
runoff source Runoff,

Zone 2 – main river channels as a transfer sediment detachment, transportation and

component deposition, and
Zone 3 – alluvial fans, deltas, etc., as zones of sediment yield
deposition.
Considered together, these three elements form
the watershed
599 600

100
 Upland Areas Surface Runoff

Hydrologic and hydraulic processes deriving Result of precipitation and is the amount of water
erosion and sedimentation processes which appears in the stream channel network
during and after precipitation
rainfall amount and intensity, Direct flow of water over the soil surface and in
small, definable channels
runoff amount and rate, and
Overland flow consists of:
flow depth and velocity
sheet flow on the interrill areas and,
flow to, into, and within many small
concentrated flow channels or rills
601 602

Model Components Flow Routing

Dynamic models of channel change require a Flow routing from simple lag models that consider
number of components: only time delays to complex three-dimensional
representations of the routing processes
stream flow routing
Depending on the systems being modeled, flow
sediment load calculations routing can be classified as
changes in channel width and depth catchment routing – refers to the
transformation of precipitation or other basin
accounting for changes to curvature effect
inputs to the outflow from the basin
An alternative to such models are the more simple channel routing or reservoir routing – calculates
in-stream erosion models that do not consider the outflow from a stream for which the
changes to channel form 603 channel characteristics are known or assumed 604

Flow Routing Hydraulic Routing Methods

Based on the manner in which flow processes are Based on the conservation of mass and
represented by the routing algorithms, the routing momentum
of water through rivers and reservoirs can be
further categorized as Include the variations of the St Venants equation
that are commonly incorporated into many water
hydrologic quality models
hydraulic
Tend to require a more detailed description of the
semi-hydraulic physical properties of the system than hydrologic
Hydrologic routing models often employ spatially techniques
lumped forms of the continuity equation
605 606

101
 Hydraulic Routing Methods Process-Based Models
Their higher demands for computing power, as Also called physics-based erosion models attempt to
well as the quantity and quality of the data inputs address soil erosion on a relatively fundamental level
required to drive them, potentially limit the using mass balance differential equation
practical applicability of hydraulic models ( )
+ = + (1)
Semi-hydraulic models are a compromise between
hydrologic and hydraulic methods where
The transport of sediment includes further = sediment load
considerations, namely relating to the hydraulic = distance downslope
component of sediment transport relationships
= density of sediment particles
607 608

Process-Based Models Process-Based Models

= sediment concentration The storage term may be neglected for shallow and
= flow depth gradually varied flow, resulting steady-state
= time continuity equation

= rill erosion
= + (2)
= interill erosion
/ = change in sediment flow rate along
the slope
( )/ = change in sediment storage over time
609 610

Interaction Between , , , Interaction Between , , ,

The basic relationship between sediment load =− for deposition (5)

( ), transport capacity ( ), erosion rate ( ), and
deposition rate ( ) is: where
= rill erosion detachment capacity rate
or = ( − ) (3) = maximum erosion rate when sediment load is zero
= maximum deposition rate when transport capacity
where is a coefficient is zero

The rill erosion rate equation can be rewritten as:

= for erosion (4)
+ =1 (6)
611 612

102
 Interaction Between , , , Interaction Between , , ,
where is erosion rate. Rewriting Eq. 6 in terms of : with as the maximum deposition rate when
transport capacity is zero. This equation can be
= 1− (7) rewritten as:

with the maximum erosion rate given by rearranging = 1− (10)

Eq. 4
= (8) with the maximum deposition rate given as:
In a similar manner, the equation for rill deposition =− (11)
rate ( ) can be written as:
The coefficient is given by the ratio of the particle
+ =1 (9) fall velocity, , to the water discharge per unit
613 width, , as follows: 614

Interaction Between , , , Interaction Between , , ,

= (12) =1− (14)
where for deposition, which, in terms of relative sediment
= 0.5 for overland flow
load ( / ), can be written as:
= 1.0 for open channel flow
1
To summarize, the previous ten equations (Eq. 3-12) =1− (15)
⁄
show how sediment load may be different from
transport capacity; Eqs. 6 and 9 can be rewritten as: Note that Eq. 13 shows the potential relative erosion
rate, / , is a linear function of relative sediment
=1− (13) load. Eq. 15 shows relative deposition rate is
proportional to the reciprocal of relative sediment
for erosion, and 615 load 616

Interaction Between , , , Interaction Between , , ,

Curves in the schematic illustration of relationships between
potential erosion rate, transport capacity and deposition rate
suggest the following

1. Potential erosion rate is

=1− at its maximum when
sediment load is zero,
such as when clear
1
=1− water is directly
⁄ introduced into the
upstream end of a rill
or channel.

617 618

103
 Interaction Between , , , Interaction Between , , ,
Curves in the schematic illustration of relationships between Curves in the schematic illustration of relationships between
potential erosion rate, transport capacity and deposition rate potential erosion rate, transport capacity and deposition rate
suggest the following suggest the following

2. Relative erosion rate 3. Deposition rate is at

decreases linearly with its maximum when
increasing sediment transport capacity is
load until net erosion zero, such as when
ceases when sediment flow velocity is zero in
load exactly equals still water.
sediment transport
capacity.

619 620

Interaction Between , , , Transport Capacity

Curves in the schematic illustration of relationships between If the aim of a modeling application is to predict
potential erosion rate, transport capacity and deposition rate actual loads or specify differences between regions,
suggest the following
then results will be sensitive to the choice of the
4. Relative deposition rate sediment transport capacity relationship
decreases nonlinearly
from its maximum with Many factors influence transport capacity, including:
decreasing sediment
flow and flow velocity
load until net deposition
ceases when sediment size or density of particles
load exactly equals
sediment transport available sediment load
capacity.
resistance of the soil to detachment by flow
621 available sediment supply from interrill erosion622

Stream Channels Individual Channel Segments

The relative importance increases as interest in Flow in most channels in fields is spatially varied,
erosion and sediment yield extend to progressively with discharge increasing along the channel
larger land areas
Discharge along a single channel segment during a
Small channel (or stream) is permanent feature of runoff event, and in the absence of significant
the landscape that conveys water and sediment infiltration losses to the channel bed and banks,
from the upland areas to the major channels and can be assumed to vary directly with upstream
acts as a sediment source or sink contributing areas
If normal tillage can obliterate the concentrated
flow areas, they are termed rills. If not, they are
termed gullies or channels
623 624

104
 Individual Channel Segments Individual Channel Segments
If an initial discharge is allowed at the upper end of assuming uniform, but unsteady, flow
a segment to approximate flow from headwater
contributing areas, then the channel segment has Upland processes affecting water and sediment
an upstream inflow and increasing discharge in the supply to the stream channels also affect processes
downstream direction due to lateral inflow in the channels

Models use various options in approximating Localized changes in hydraulic conditions affect
spatially variable channel flow, including erosion, transport, and deposition of sediment in
rills and have similar effects if they occur in
selecting a characteristic discharge (the peak channels
discharge) and then assuming spatially varied,
but steady, flow 625 626

In-stream Sediment Transport Deposition

The concept of sediment transport capacity is Deposition processes require explicit consideration
largely used to describe sediment transport in in a catchment model
channel networks, as well as in overland flow
Deposition in land surface models
Bed-load transport equations commonly employed Many of the simpler empirical or conceptual
in the models are, amongst others, models do not explicitly model deposition
Meyer-Peter and Muller (1948) processes occurring in overland flow
Einstein (1950) Instead, a prescribed (or calibrated) sediment
delivery ratio is used to define the amount of
Bagnold (1977) eroded soil or sediment that moves into the
channel network
627 628

In-stream Deposition Small Watersheds

Upland processes and processes in individual
In-stream deposition is routinely incorporated in
channel segments are combined through the
models of channel evolution and development
channel network and interchannel areas to
The representation of deposition processes is influence runoff and sediment yield from
similar in many ways to the treatment of watersheds
deposition processes in overland flows Processes affecting watershed runoff and
sediment yield include
Sediment transport and deposition processes are interactions (e.g., channel junctions and backwater)
often simulated simultaneously using the sediment
land use
transport capacity concept
soil and cover characteristics, and
629 other factors varying over the drainage area 630

105
 7.2 Model Types Model Types
A wide range of models exists for use in Each model type serves a purpose, and a
simulating sediment transport process from particular model type may not categorically be
catchments considered more appropriate than others in all
These models differ greatly in terms of
situations
Their complexity
Processes considered Choice of a suitable model structure relies heavily
on the function that the model needs to serve
Data required for model calibration and model
use The ultimate factor determining a model’s value
is its simplicity relative to its explanatory power
In general there is no ‘best’ model for all
applications
631 632

Model Types Model Types

It is generally recognized that a good model Screening models are simple in concept and
should satisfy the requirements of designed to identify problem areas
– reliability
Assessment models need to predict with greater
– universal applicability accuracy because they are used for evaluating the
– easy use with a minimum data severity of erosion under different management
systems and may be intended as design tools for
– comprehensiveness in terms of the factors
the selection of conservation practices
and erosion processes included
– ability to take account of changes in climate
land use and conservation practice
633 634

Model Types Model Types

On cite consequences of erosion, such as the loss The most appropriate model will depend on the
of soil depth and declines in productivity, intended use and the characteristics of the
predictions of erosion are required for periods of catchment being considered
20–30 years either on an annual basis or as an
annual average over the time span Other factors affecting the choice of a model for
an application include:
More detailed, event models are needed for
Data requirement of the model including the
assessing the off-site effects
spatial and temporal variation of model inputs
and outputs

635 636

106
 Model Types Model Types
The accuracy and validity of the model In general, models fall into three main categories,
including its underlying assumptions depending on the physical process being
simulated by the model, model algorithms
The components of the model, reflecting the describing these processes and the data
model capabilities dependence of the model:
The objectives of the model user, including Empirical or statistical/metric
ease of use of the model, the scales at which Conceptual
model outputs are required and their form
(such as concentration vs. load) Physics based
The distinction between models is not sharp and
Hardware requirement of the model
637 therefore can be somewhat subjective 638

Empirical Models Empirical Models

Generally the simplest of all three model types Empirical models are often criticized

Based primarily on the analysis of observations ignoring the heterogeneity of catchment inputs
and seek to characterize response to these data and characteristics, such as rainfall and soil
type
The computational and data requirements for
such models are usually less, often being capable ignoring the non-linearity in the catchment
of being supported by coarse measurements system

Parameter values may be obtained by calibration, can not be extrapolated beyond their data
but are more often transferred from calibration at range, either to more extreme events or to
experimental sites other geographical areas
639 640

Empirical Models Empirical Models

generally based on the assumption of Nonetheless, empirical models are frequently

stationarity; that is, it is assumed that used in preference to more complex models
underlying conditions remain unchanged for
as they can be implemented in situations with
the duration of the study period
limited data and parameter inputs
also tend not to be event-responsive, ignoring particularly useful as a first step in identifying
the processes of rainfall-runoff in the sources of sediment
catchment being modeled
prediction of sediment delivery at regional
scale is commonly based on empirical methods
641 642

107
 Conceptual Models Conceptual Models
The goal of conceptual modeling is to describe the Play intermediary role between physics-based and
set of concepts used in a domain in order to empirical models
define a conceptual schema for the memory that
Tend to include a general description of
a software system should have of that domain
catchment processes, without including the
Typically based on the representation of a specific details of process interactions, which
catchment as a series of internal storages would require detailed catchment information

Usually incorporate the underlying transfer This allows them to provide an indication of the
mechanism of sediment and runoff generation in qualitative and quantitative effects of land use
their structure, representing flow paths in the and change, without requiring large amounts of
catchment as a series of storages spatially and temporally distributed input data
643 644

Conceptual Models Conceptual Models

Traditionally, conceptual models lump Tend to suffer from problems associated with the
representative processes over the scale at which identifiability of their parameter values
outputs are simulated
Most calibration techniques used for conceptual
Alternatively, lumped conceptual models may be models of medium complexity (say more than six
applied in a semi-distributed manner by parameters) are capable of finding only local
disaggregating a catchment into linked optima at best
subcatchments to which the model is applied
Parameter values for conceptual models have In general, simpler conceptual models have fewer
typically been obtained through calibration problems with model identification than more
against observed data complex models
645 646

Physics-Based Models Physics-Based Models

Physics-based models are based on the solution of In theory, the parameters used in physics-based
fundamental physical equations describing models are measurable and so are ‘known’
streamflow and sediment generation in a
catchment In practice, the large number of parameters
involved and the heterogeneity of important
Standard equations used in such models are characteristics means that these parameters must
often be calibrated against observed data
the equations of conservation of mass and
momentum for flow Where parameters cannot be measured in the
catchment they must be determined through
the equation of conservation of mass for
calibration against observed data
sediment
647 648

108
 Physics-Based Models Physics-Based Models
Uncertainties to model outcomes could be due to derivation of mathematical expressions
describing individual processes is subject to
lack of identifiability of large number of model numerous assumptions that may not be
parameters and non-uniqueness of ‘best fit’ relevant in many real world situations
solutions
discretization of the equations derived for use
errors in the measurement of important
with continuous spatial and temporal data
characteristics
differences between the scale at which model Error accumulation will need to be controlled in
algorithms are applied and the scale at which models which transfer output fluxes from one
measurements are made spatial element to the next as input
649 650

Distributed vs. Lumped Modeling Distributed vs. Lumped Modeling

Another way to view the range of models is the Typically, distributed models have involved
way in which they represent the area to which the 1. dividing the area of interest into cells (often
model is applied rectangular grids) at which basic computations
Parameters or variables can also be lumped in are undertaken
time as well as space, or both 2. outputs for each cell are then routed through
With increasing computing power distributed the system to produce catchment scale outputs
approaches have become more feasible Distributed models raise issues of
A distributed approach seems particularly increased data requirement
applicable to sediment transport modeling
651
effects of cell resolutions on model outputs 652

Distributed vs. Lumped Modeling Distributed vs. Lumped Modeling

Distributed models have the potential to assist
A compromise between fully distributed
managers requiring knowledge of the provenance
methodologies and lumped models are the semi-
of the major sources of pollutants or sediments
distributed models that break a catchment down
into a group of subcatchments Spatially distributed models should include only
those variables for which there exists sufficient
Ultimately the choice between lumped or information regarding their spatial distribution
distributed models depends on the desired output
Lumped models may suffice when
of the model and the nature of possible
management interactions estimates at the catchment outlet are sufficient
intervention can be applied spatially uniformly
653 654

109
 Temporal Resolution Universal Soil Loss Equation (USLE)
Event based models were developed to look at The most widely used and successful model to
the response of the modeled area to single storm predict soil loss from upland areas
events
Developed in the 1970s by the USDA, the model
For each event, the model time-step is of the
has undergone much research and a number of
order of minutes to hours for application to
modifications (e.g. MUSLE, USLE-M)
small plots or grid cells in a catchment
Alternatively, a larger temporal resolution was The model has also been upgraded to take into
used and models were applied to explore broad account additional information that has become
trends over time to changes available since the development of the USLE

Third approach is to use a continuous time step

655 656

Model output Input data

Input data requirements are low compared with
The typical output from the USLE is an annual
most other models
estimate of soil erosion from hillslopes
annual rainfall

an estimate of soil erodibility

land cover information

topographic information is required

657 658

Model Structure Model Structure

The procedure is founded on an empirical soil loss , , , , , = factors predicting the impacts of
equation that is believed to be applicable wherever rainfall energy and intensity ( ),
numerical values of its factors are available soil erodibility ( ), slope length
( ), slope steepness ( ), soil cover
and management ( ), and support
= practice ( )
The equation and its factors are based on
where observations of erosion and erosion processes
= computed average annual soil loss per unit rather than theoretically derived relationships
area (t/ha/y) Factors are estimated using tables and graphs in
659 the absence of observed data 660

110
 Model Structure Rainfall and Runoff Factor
Designed to predict the long-time average soil Computed as the product of rainfall storm energy
losses in runoff from specific field areas in ( ) and the maximum 30-min rainfall intensity
specified cropping and management systems ( ) divided by 100 for numerical convenience,
known as index
result should be seen as a long-term average
annual value
= 0.119 + 0.0873 log <= 76 mm/h
Although developed for application to small
hillslopes, the USLE and its derivatives have been
incorporated into many catchment scale erosion = 0.283 > 76 mm/h
and sediment transport modeling applications
661 662

Rainfall and Runoff Factor Rainfall and Runoff Factor

where is kinetic energy in megajoule per An approximate equation to estimate is:
hectare per millimeter of rainfall (MJ/ha • mm),
and is rainfall intensity in mm/h for a given .
= 0.417
period of constant rainfall intensity
Eq. (18) is applied over each interval in a storm, where is in MJ • mm/ha • h • y, and is the 2-y,
and the sum is rainfall energy 6-h rainfall amount in millimeters

Using index

= ⁄100
663 664

Soil Erodibility Factor Soil Erodibility Factor

, in terms of t • ha • h/ha • MJ • mm, is the soil If is plotted on the horizontal axis and is
loss rate per erosion index unit for a specified soil plotted on the vertical axis, then is the slope of
as measured on a unit plot the line through the origin expressing as a
function of
A unit plot is defined as a 72.6 ft, or 22.1 m, length The values are now typically obtained from a
of uniform 9% slope continuously clean-tilled nomograph (Foster et al., 1981) or the following
fallow condition equation:
Under these unit plot conditions, = 1, = 1,
.
and = 1 so that =1 =
. − + . − + . ( − )

=
665 666

111
 Soil Erodibility Factor Soil Erodibility Factor
The size of soil particles for very fine sand fraction
= (% +% ) × (100 − % )
ranges between 0.05 and 0.10 mm, for silt
content between 0.002 and 0.05, and clay < 0.002
where is particle-size parameter, is % of soil mm.
organic matter content, is soil structure code (1 = The soil organic matter content is computed as the
very fine granular; 2 = fine granular; 3 = medium or product of percent organic C and 1.72.
coarse granular; 4 = blocky, platy, or massive), and
profile permeability (saturated hydraulic conductivity)
Computed values of ranged from about 0.0026
class [1 = rapid (150 mm/h); 2 = moderate to rapid
to 0.092 t • ha • h/ha • MJ • mm, with most
(50–150 mm/h); 3 = moderate (12–50 mm/h); 4 =
slow to moderate (5–15 mm/h); 5 = slow (1–5 mm/h); agricultural soils having values in the range of
6 = very slow (<1 mm/h)]. 0.013 to 0.053
667 668

Slope Length and Steepness Factor Slope Length and Steepness Factor

is dimensionless and is the expected ratio of = 0.6[1 − exp(−35.835 )]

soil loss per unit area of a field slope to that from
a unit plot
= tan
100
The appropriate value can be obtained from
value can be obtained from the equation
where S is field slope (%) and is field slope
steepness in degrees
= 65.41 sin + 4.56 sin + 0.065
22.13

669 670

Cover and Management Factor Cover and Management Factor

is dimensionless and is the ratio of soil loss from is determined over time (cover and management
an area with specified cover and management to practices take time to implement, and their
that from an identical area in tilled and combined and interactive influences may take
continuous fallow months or years to stabilize) and on a mostly
empirical basis
is a measure of the combined effects of all cover
and management variables affecting soil loss and is Because vegetative cover develops over time and
the most difficult factor to estimate (under most with the seasons, as controlled by plant
conditions except the unit plot) in the USLE physiology, climate and weather, management,
soil characteristics, etc., it is highly dynamic and
highly variable
671 672

112
 Cover and Management Factor Cover and Management Factor
lumps an enormous amount of information on 3. Residue mulch as a measure of “on-ground”
biological, chemical, physical, and land use or protection from raindrop impact
management-induced variability into a single
coefficient 4. Incorporated residues affecting the top few
inches of soil
Various items affecting estimated factors include:
5. Tillage as it affects the soil, residues, etc
1. Cropstage periods to represent the seasonal
changes in effectiveness of plant cover 6. Land use residuals such as the influence of
plant roots, organic matter, and other factors of
2. Crop canopy as a measure of the degree of
interseasonal importance
protection provided by the canopy
673 674

Cover and Management Factor Cover and Management Factor

The computation of C is evaluated for six Period 2 (development)―End of period 1 until
cropstage periods defined as follows: canopy cover reaches 75 percent
Period F (rough fallow)―Inversion plowing to Period 3 (maturing crop)―End of period 2
secondary tillage until crop harvest. This period was
evaluated for three levels of final crop
Period SB (seedbed)―Secondary tillage for canopy
seedbed preparation until the crop has
developed 10 percent canopy cover Period 4 (residue or stubble)―Harvest to
plowing or new seeding
Period 1 (establishment)―End of SB until crop
has developed 50 percent canopy cover
675 676

Support Practice Factor Support Practice Factor

is dimensionless and is a factor used to The combined use of various practices is more
represent the ratio of soil loss with a specific effective than a single practice for controlling
support practice to the soil loss on a unit plot erosion in highly erodible soils.
The most important support practices for cropland In systems with various support practices,
are contour tillage, strip-cropping on the contour, values are calculated as follows:
and terrace systems
= + +
The values vary from 0 to 1 where the highest
values correspond to a bare without any support where is contouring factor for a given field
practices. slope, is strip cropping factor, and is terrace
677
sedimentation factor 678

113
 General Comments Erosion/Sediment Transport Models
The USLE, as an empirically derived and data
based model, shares the strengths and
weaknesses of such procedures
In terms of its main factors ( ), it is a linear
equation, but in terms of how physical features
and management practices affect the factors, it is
nonlinear
The USLE is intended to estimate long-term
average annual soil loss from upland areas
The emphasis in development of the equation was
on agricultural areas of the humid United States 679 680

Erosion/Sediment Transport Models 8. Pollutant Transport

Some of the factors which influence water quality
in rivers are
amount and type of pollutant which is
discharged into the river
type and location of the discharge (both of
which can affect the initial hydraulic mixing)
instream biological, physical, and chemical
processes (which depend on the
characteristics of both the receiving stream
and the pollutant)
681 682

Pollutant Transport Pollutant Transport

geometric and hydraulic characteristics of the Scope: natural transport processes of neutrally
river (which affect the transport processes) buoyant conservative substances by the river flow
Transport of dissolved or suspended substances is time dependent evolution of concentration in
essentially linked to the unsteady flow modelling river flow
in open channels conservative: does not absorb, does not adsorb,
Transport refers to the hydrodynamic process of does not decay, does not undergo chemical,
dispersion biological or nuclear transformation

interaction between differential convection Time scale of the mixing phenomena which
and turbulent diffusion interests us is generally much shorter than the
time over which significant changes in water
depends upon the flow velocity field discharge occur
683 684

114
 Pollutant Transport Transport Mechanisms
Steady-flow approach, which simplifies the Advection
transport equations, is usually justified Caused by the large-scale movement of water

Unsteady flow is represented as a series of Responsible for moving a pollutant along the
increasing (or decreasing) quasi-steady flow steps channel
Difference in advection with respect to either
Transport results at the end of each steady
time or space are responsible for most of the
flow step are used as initial conditions for the
spreading of a pollutant
next step
Absence of advection may cause a pollutant to
be delayed in its downstream movement
685 686

Transport Mechanisms 8.1 Advection diffusion process

Diffusion and dispersion or mixing The division of transport between advection and diffusion
Represent the net advection associated with the The transport of fluid with the fluid velocity defined as
turbulent (temporal) fluctuations of velocity and an average of particle velocities will not correspond to
the transport of molecules under their individual
concentration
molecular motions
Represent the net advective effects which are
not included explicitly in the advection terms
Temporary storage or exchange with dead
zones
advection
Represent the absence of advection in some process

regions 687 Center of mass 688

Advection diffusion process Advection diffusion process

Advection brings the concept of velocity, Assumptions
The fluid is not only advected with the fluid This molecule can occupy any of a number of
velocity, but it also diffused during the process of cells which have addresses numbered from 1,
advection due to molecules having velocities through , to
differing from the fluid velocity If the molecule is in any cell, let us say of address
, at time , however, we suppose that it can only
Description of advection-diffusion process in
move as far as cells with addresses – 1 and +
terms of random walk of a single representative
1 during a subsequent time Δ
molecule

689 690

115
 Advection diffusion process Advection diffusion process
Let = −1→
probability that it moves, on average, from – 1
to = ( – 1 → ) + 1− −1→ + +1→
+ ( +1→ )
probability that it moves, on average, from + 1
to = ( + 1 → )
probability that, on average, it stays at cell address
is → = 1– –1 → + +1→

= the probability that the molecule is in cell

address at time level Δ
691 692

Advection diffusion process Advection diffusion process

In words, the probability that the molecule is in Now if the fluid velocity corresponding to
cell address at time level ( + 1)Δ is equal to innumerable numbers of such motions is
the probability that it was in cell address ( – 1) supposed to be a continuous function of and ,
at time level Δ and moved one step to the it can be represented by a polynomial in and ,
right, plus it was in cell address at time level which is to say that its variations can be described
Δ and stayed where it was, plus the probability by a Taylor's series
that it was in cell address ( + 1) at time level ∆
Δ and moved one step to the left = + ∆ + + H. O. T
2!

∆
= − ∆ + − H. O. T
2!
693 694

Advection diffusion process Advection diffusion process

∆ If the flow is from left to right

= + ∆ + + H. O. T
2! ( –1 → ) = 1 ( → )=0 ( +1→ )=0

Substituting gives at Δ and Δ Pure advection =

∆ ∆ which expands through term-by-term

+ + ( −1→ − ( + 1 → ))
2 ∆ cancellation of the H.O.T., with Δ , Δ → 0, to

∆ ∆
− −1→ + +1→ = (H. O. T) + =0
2∆ ∆

695 696

116
 Advection diffusion process Advection diffusion process
If now Δ /Δ is taken as constant as Δ and Δ
In nature, Δ , Δ ≠ 0, but we can imagine of a
tend to zero, so that Δ /Δ represents the locally
sequence of ‘equivalent random walks’ with
constant velocity of the molecules, then we have
Δ 2/Δ kept constant with Δ , Δ → 0 to obtain
+ =0 advection equation
− =0 diffusion equation
In the event that ( – 1 → ) = ( + 1 → ) =
Δ , there is clearly no net motion and so no where
advection and in this case the Taylor expansion
of probability equation reduces to = Δ Δ 2/Δ = diffusion coefficient
∆ ∆
+ −∆ = (H. O. T)
2 ∆ 697 698

Advection diffusion process Advection diffusion process

In the case that Δ and Δ are taken down to The equations derived above have been
small numbers with Δ /Δ a constant identified expressed in terms of probability of finding a
with the mean molecular velocity, so that molecule in a certain interval
Δ 2/Δ is, by comparison, a very small number;
At the macroscopic level of our observations,
but to balance this,
however, the molecules can be seen as an agent
− → − + → ≪( − → + + → ) for advecting and diffusing any number of
macroscopic properties
probabilities expansion in Taylor’s series can be
approximated by advection-diffusion equation for density,

+ − = 0 advection-diffusion equation + − =0
699 700

Advection diffusion process Molecular diffusion

advection-diffusion equation for velocity , Fick’s Law
regarding as the momentum per unit mass of The ‘marked’ molecules of one fluid will diffuse in
fluid the other ‘neutral’ fluid with their flux being
proportional to their concentration gradient
+ − =0
= + +

where
= concentration of the tracer
= coefficient of molecular diffusion (m ⁄s)
701 702

117
 Molecular diffusion 8.2 Turbulent mixing process
Fick’s Law Turbulent diffusion occurs when water is flowing
in turbulent conditions, i.e. Reynolds number >
Molecular diffusion is very weak, being of the 2000
order of the dynamic viscosity coefficient
In steady but turbulent flow:
(10 m ⁄s)
average velocity constant at
e.g., a point source of dye in still water takes
instantaneous velocity varies in more or less
about 12 hr to attain a diameter of 1m
random manner in magnitude and direction
at

703 704

Turbulent mixing process Turbulent mixing process

A number of small particles of neutrally buoyant Mathematical description of this dispersion
substance released at will be dispersed one process is based on a semi-empirical theory which
from another at some distance downstream begins with a partial differential equation for the
conservation of tracer in a differential volume
Some quantity of pollutant released at some
element
point will grow into an ever expanding clouds as it
moves downstream
+ + +

= + +

705 706

Turbulent mixing process Turbulent mixing process

where instantaneous velocities , ,
, and = instantaneous velocity components = + = ̅+ = +
in the , and directions Instantaneous velocities have zero time-
Turbulent flow can be considered as resulting averages
from the superposition of turbulent fluctuations ′, ′, ′ = 0
on a steady time averaged flow
Motion of ‘marked’ neutrally buoyant particle can
Velocity field at a given point is composed of two then be considered as subject to three influences
parts:
molecular diffusion
time-averaged velocities , ̅,
time-averaged velocity
707 708

118
 Turbulent mixing process Turbulent mixing process
turbulent velocity fluctuations ( ̅+ ) ( ̅+ ) ( ̅+ )
+ + =
In order to relate the turbulent transport to the
mean flow field, we apply Reynolds averaging in
Averaging the above expression over a time
expressing the concentration at a point as the
period which is long compared to the turbulent
sum of a time averaged value and a fluctuating
fluctuations but short compared to the time
component,
scale of the dispersion event being modelled, we
= ̅+ have
Substituting this definition and the comparable
( ̅+ ) ( ̅+ ) ( ̅+ )
ones for , and into differential equation, we + + =
have in the direction
709 710

Turbulent mixing process Turbulent mixing process

Since time-averaged values of turbulent The semi-empirical approach consists in
fluctuations are zero , ̅ , , ̅ = 0 , we supposing that turbulent transport can be
obtain described in a manner analogous to molecular
diffusion,
̅ ̅ ̅
+ = − ̅
=1− ̃

where the subscript denotes the , , or axis where

The term represents turbulent transport ̃ = coefficient of turbulent diffusion
due to fluctuating local velocities
711 712

Turbulent mixing process Turbulent mixing process

Basic concepts of Prandtl used for evaluation of Basic concepts of Prandtl used for evaluation of
turbulent diffusion coefficient turbulent diffusion coefficient
The turbulent fluctuations are related to a A ‘mixing length ’ plays the same role as the
characteristic velocity known as the ‘friction’ mean free path in molecular gas diffusion
velocity ∗ theory, and is proportional to the distance to
the particle from the wall,
∗ = =
where where
= tangential stress at the bed = von Karman’s constant
= fluid density = distance from the bed
713 714

119
 Turbulent mixing process Turbulent mixing process
Turbulent diffusivities are much greater than Dropping the overbars we can now write the
molecular ones, which thus may be safely transport (or flux) equation for a neutrally
neglected in turbulent flow. buoyant substances in a three-dimensional
Distribution of the turbulent diffusivities in the velocity field
flow field is not uniform, but depends upon the
+ + +
orientation of fluctuating velocities and the
distance from the walls.
= ̃ + ̃ + ̃
In general the turbulent diffusion coefficients are
not the same in all directions and, moreover they The substance will be convected by time
vary as a function of the independent space averaged velocity , , and diffused by
variables , , . turbulent diffusivities ̃ , ̃ , ̃
715 716

Turbulent mixing process Turbulent mixing process

We refer to the overall process as dispersion Consequently the concentration distribution in
In principle, if the full three-dimensional velocity the vertical dimension becomes uniform very
field and the three diffusivities ̃ , ̃ , ̃ were quickly, long before uniformity in the horizontal
known, the above equation could be solved by direction is attained
analytical or numerical methods to yield the Therefore it is considered reasonable to average
function ( , , , ) for any appropriate initial the three-dimensional dispersion equation
and boundary conditions vertically over the flow depth
Because the river depth is usually small as
compared to the width, a neutrally buoyant
substance becomes fully mixed over the depth of
flow in a river in a relatively short time 717 718

Turbulent mixing process Turbulent mixing process

The depth-averaged dispersion equation is ℎ + (ℎ )+ (ℎ )
obtained by integrating three-dimensional
equation over depth of flow using the hypotheses = ℎ + ℎ
that differential convective transport and
turbulent diffusion processes may be combined in where
gradient diffusion terms , , = depth-averaged values
The result is the two-dimensional, depth- , = empirical mixing coefficients
averaged dispersion equation In general and are much larger than ̃
and ̃ , since they incorporate dispersion due to
differential convection as well as turbulent
719
diffusion 720

120
 Turbulent mixing process Turbulent mixing process
For uniform flow in a wide open channel,
= 5.93 ∗ ℎ
= 0.23 ∗ ℎ
Here is much larger than because it
incorporates differential convection due to the
logarithmic vertical velocity profile in
established flow
Attempts to model the dispersion process in
natural rivers have been concentrated in two
areas:
721 722

Turbulent mixing process

one-dimensional simulation of the dispersion Theoretical concept
downstream from an instantaneous plane The longitudinal dispersion of a tracer which is
source injection fully mixed over the cross section should behave
as a Fickian diffusion process, at least at large
two-dimensional simulation of the dispersion distances from the point at which the tracer in
downstream from an arbitrary source injected instantaneously
The dispersion equation is assumed to be of the
form

+ =

723 724

Theoretical concept −
( , )= exp −
4
where 2

= cross-sectional average concentration where

= cross-sectional average velocity and = concentration and volume of the
= flow area tracer injected instantaneously over
= longitudinal mixing coefficient (much greater the entire cross-section
than )
This solution is generally of little practical use for
If , , and are known and assumed engineering purposed, for three reasons
constant, the analytical solution to the one- 1. , and are never constant over any
dimensional dispersion equation is appreciable reach of a natural river
725 726

121
 2. experimental data in natural rivers do not where
support the assumption that the one-
dimensional mixing process can be considered = characteristic mixing length
to be Fickian = hydraulic radius
∗ = shear velocity
3. one-dimensional model implied by the
analytical solution can not be assumed until This distance is often well beyond the range of
the tracer has progressed a distance from the interest in water quality studies
source greater than

1.8
=
∗
727 728

Numerical solution of one-dimensional Finite difference approximation

convection equation
+ =
Numerical approach needs great care because of
artificial, or numerical, diffusion which is where
sometimes stronger than the physical diffusion = an artificial diffusion coefficient introduced
Convection equation for constant flow area is by the approximate nature of the finite
expressed as difference scheme
an essential quality criterion of a numerical
+ =0 scheme for the calculation of convection is the
value of
729 730

As long as is much smaller than , artificial since in pure convection the concentration of
diffusion does not compromise the simulation
marked fluid particles do not change, we can write
results
If is of the same order of magnitude as or =
greater than , the simulation results may well In the special case in which coincides with
appear plausible but be unrelated to the natural
phenomena being modelled ∆ = ∆
t
n+1
this procedure yields =
t

ξ n+1
n Δx - α α x
i-1 Δx i i+1 Δt

The particle of water which arrives at point + at future time

n x
+ departed from the point at time and followed the i Δx i+1
trajectory shown by dashed line 731 732

122
 estimating by interpolating linearly between ∆
≈ + ∆ +
2
and yields
∆
≈ − ∆ +
= 2

∆ − + ≈ − 2∆ + 2∆
=
∆

If we expand , and in a Taylor Using the identity Δ + = Δ and the relation

series around the point + 1, , dropping terms
which are higher than second order, we obtain =
733 734

we can express as
(∆ − )
+ =
2∆ ∆
= 1− ( − 2)
2∆
Thus for this scheme
which shows that we can express = 0 for =
(∆ − ) 1or = 2, but it reaches a maximum of
=
2∆ ∆
for = 1.5
8∆
In terms of Courant number
The same analysis for the case of ≤ 1, that is to
∆ say for the trajectory which cuts the -axis
=
∆ between points + 1 and , yields
735 736

∆ In a given time step, the solution of advection

= 1− equation is followed immediately by the solution
2∆
of
= 0 for = 0 and = 1, and
=
∆
= for = 0.5
8∆
The commonly used finite difference method for
Numerical solution of one-dimensional solution of the above equation is an explicit
diffusion equation centred scheme
t
n+1
In most one-dimensional dispersion models the
convection and diffusion are calculated in
separate steps 737
n
i-1 i–½ i i+½ i+1
x
738

123
 rate of accumulation in the volume is
We want to express conservation of mass in the
volume limited by points − ½ and + ½ −
≈
∆
The rate of mass diffusion into volume at − ½ is
By equating the rate of accumulation to the net
− inflow, we obtain the working relationship
≈ ⁄ ⁄
⁄
− −
∆ −
rate of diffusion out at + ½ is = ⁄ ⁄
( ⁄ − ⁄ ) −
− −
≈ ⁄ ⁄ − ⁄
⁄
− ⁄
−
739 740

9. RIVER ENGINEERING
The aims of river engineering works are
The scheme is stable as long as maximizing the benefits that can be obtained
from rivers.
∆
≤ 0.5 River engineering is concerned with planning,
( − )( − )
design, construction and maintenance of works in
order to attain a better serving of human needs.
River engineering works can be divided into four
main types:
1. Channel regulation for low and high discharges
2. Discharge regulation
741 742

RIVER ENGINEERING RIVER ENGINEERING

3. Water level regulation Changes in hydraulic conditions are usually
4. Bank and/or flood protection taking place quite rapidly in comparison with
geological time
All of these works change the natural river
environment and morphology to a greater or The resulting changes in bed topography (bed
lesser degree levels and river width) normally take much
longer
River engineering structures have to be built
according to the aesthetic standards which insist In river engineering works the delayed response
on preserving the natural look of the river to changes in hydraulics and morphology must
Environmental considerations influence be taken into account
decisively the design
743 744

124
RIVER ENGINEERING RIVER ENGINEERING
It should be emphasized that in a river system 2. Adaptations towards new equilibrium
often many transient processes are active at the conditions
same time, all of them having been initiated in Changes in the river boundary conditions
different periods and human activities may result in long-
Changes in rivers can be divided into two classes lasting even permanent changes in the
river environment
1. Fluctuations around the mean values
River engineering works may vary greatly in size
Changes in bed-form dimensions and in their effect on a river’s behaviour
Steeper bend profiles  They may provide a strictly local
Changes in non-uniform reaches caused improvement which hardly changes the
by flow variations 745
upstream or downstream river reaches 746

RIVER ENGINEERING RIVER ENGINEERING

 They may be designed to alter completely a Two broad types of human-induced change can
river regime, thus affecting the river over be identified
almost its entire length  Direct or channel-phase changes : brought
about by direct modification to the channel
Elements which may need changing to promote
itself
human use of a river are the river bed, the river
discharge and the water level, which may affect River regulation
each other  Water storage by reservoir
If dangerous or expensive outcomes are to be  Diversion of water
avoided, the consequences of interference need Channel modifications
to be understood
747
 Bank stabilization 748

RIVER ENGINEERING RIVER ENGINEERING

Channel modifications Land-use changes
 Channel straightening  Changes in agricultural practices
 Stream gravel extraction  Building construction
 Indirect or land-phase changes : result from  Urbanization
activity in extra-channel areas
 Mining activity
Land-use changes
Land drainage
 Removal of vegetation, especially
deforestation  Agricultural drains

 Afforestation  Storm water sewerage systems

749 750

125
River regulation River regulation
Upstream impact Downstream impact
 Local base-level is raised to a position at  Reduction in the magnitude of flood peaks by
which the water surface intersects the as much as 90 per cent
original bed
 Marked decrease in the sediment load,
 Maximum rise in height determined by the
especially in those reaches immediately
crest of the dam spillway
below the dam
 With the reduction in transporting ability as a
stream enters a reservoir, a depositional  Degradation of the channel bed, typically at
wedge is constructed and the channel rates much higher than in natural rivers, by
gradient locally lowered sediment-free water
751 752

River regulation River regulation

Downstream impact Straightening
 Since degradation is usually at a maximum River is shortened by artificial cut-offs, thereby
close to the dam and progressively declines steepening the gradient and increasing flow
downstream, channel slope tends to become velocity and transport capacity
flatter, although the amount of degradation
can vary considerably along a river The modified stream attempts to establish a new
equilibrium gradient through a combination of
Extend often over long stretches of river
upstream progressing degradation and
Rivers respond to channelization both within and downstream aggradation
beyond the modified reach
753 754

River regulation River regulation

Resectioning Levee construction
Widening and/or deepening of the river channel Channel banks are artificially raised to confine
to increase its conveying capacity and thereby floodwaters
reduce the incidence of overbank flooding Bank protection
Widening has the effect of reducing unit stream Use of structures such as gabions and steel piles
power and therefore sediment discharge, so that to control bank erosion
deposition may occur Clearing and snagging
Deepening may increase the susceptibility of the Removal of obstructions from the watercourse,
river banks to erosion and may trigger upstream thereby decreasing resistance and increasing
progressing degradation within tributaries flow velocity
755 756

126
River regulation River regulation
In general, training structures are primarily In the regularized channel, however, the
designed to: resistances are considerably lower and the river
 contract the channel by means of training does not require such a slope as it had before to
structure overcome the resistances
 preserve the cut-off areas to be utilized for In order to reinstitute the balance between the
agriculture following their sedimentation transporting power and the resisting force the
slope must decrease
 produce a new, stable bank at the training
structures Together with the decrease of the slope due to
In the unimproved river a balance exists channel deepening, a further purpose of
between the energy of water and the resistance regularization, namely lowering of ground water
level, is attained
757 758

River regulation Longitudinal profile and determination of

equalized profile
Channel deepening can be:
 desired, taking into account diminution of River meanders, bed-forms such as the riffle-
overflows pool sequence and large-scale dunes cause
local, within-reach slope variations
 harmful, taking into account agriculture, if
due to the lowering of ground water levels it However, mean reach slopes are averages over
causes excessive drying of soils this local variability and are mutually adjusted to
cross-section and plan-form properties in the
medium-term steady-state time scale

759 760

Longitudinal profile and determination of Longitudinal profile and determination of

equalized profile equalized profile
Consecutive reach slopes combine to create the Long profile adjustments are emphasized as
complete long profile, which reflects long-term being necessary to maintain sediment transport
geological development influenced by tectonic with the available discharge and given channel
and morphogenetic histories as well as recent characteristics
channel pattern adjustments However, mutual adjustment of slope, plan-form
Rivers tend to develop a concave – upward form and cross-section properties characterize the
of the long profile true response of alluvial streams to the
multivariate environmental controls of run-off,
flood magnitude and frequency, sediment yield
and size
761 762

127
Longitudinal profile and determination of Longitudinal profile and determination of
equalized profile equalized profile
The interaction of these controls determines the Bed elevation changes result from aggradation
slope at the reach scale, and their downstream and degradation, which reflect alterations in the
variation determines the spatial adjustments of river’s transport capacity relative to sediment
slopes which create the complete long profile supply
Factors affecting changes in stream bed elevation
The course component of bed material controls
channel slope  Aggradation: increasing bed elevation
Upstream control; e.g. glacio-fluvial
Long-profile adjustments include changes of
sediment entering at headwater
channel bed elevation, channel gradient and
overall profile shape Downstream control; e.g. rising base level
763 764

Longitudinal profile and determination of Longitudinal profile and determination of

equalized profile equalized profile
 Aggradation: increasing bed elevation  Degradation: decreasing bed elevation
Basin-wide control; e.g. increasing sediment Basin-wide control; e.g. decreasing
yield: stream flow ratio caused by climatic sediment yield: stream flow ratio caused by
change, vegetation clearance, etc. climatic change, conservation measures, etc.
 Degradation: decreasing bed elevation
To compute the design cross-section we take
Upstream control; e.g. bed-load equalized slope, possibly on the long distance
entrapment
Downstream control; e.g. falling base level The designed alignment usually involves some
and Knick point recession shortening of a natural river course due to
765
rectification by means of cut-offs 766

Longitudinal profile and determination of Longitudinal profile and determination of

equalized profile equalized profile
Prior to the adoptation of the equalized slope it  straight line, if the slope differences at the
is necessary to determine ‘fixed’ points: beginning and the end of the reach are
 natural, e.g. rocky bars, major tributaries, negligible
rapid changes of valley width,  curve, in case the differences are large
 artificial, e.g. weirs, sills, bridges, etc.
A tributary acts in various ways depending upon
Most often carried out regularization is only to the ratio of discharge and quantity of brought in
make the existing slope even sediment
In the first case the design of slope consists in
making it even between the ‘fixed’ points
767 768

128
Longitudinal profile and determination of Longitudinal profile and determination of
equalized profile equalized profile
 If the tractive force of the river is not sufficient  A mountain tributary of the lowland river
to get rid of the sediment brought in from the increases quantity of sediment which the river
tributaries it can be easily increased by is unable to transport and hence the ‘fixed’
steepening the slope which is obtained by point will rise, steepening continuously the
shortening the river course slope below, and decreasing above the
tributary
 If conversely, bottom and bank erosion is too
great the slope should be decreased by means
of sills, dams, etc.

769 770

Determination of normal stage and normal Determination of normal stage and normal
discharge discharge
Variation of discharge over the year is a general The stage which is taken as the basis of the
characteristic of practically every river attempted regularization is the normal stage and
 Main cause of this is undoubtedly the uneven the corresponding discharge is the normal
distribution of rain discharge
The regularization design should be adjusted to all To this normal stage the training structures
discharge variations that happen in a given river should be first of all adopted
River regulation is associated with a certain water For diverse structures, we assume different
stage, defined with respect to the purpose for normal stages, which result from the following
which we attempt the works, e.g. flood control, statement
water intake, navigation, bank protection, etc. 771 772

Determination of normal stage and normal Determination of normal stage and normal
discharge discharge

 Mean high water is a normal stage for  The absolute lowest stage is a normal stage
mountain stream engineering for design intake especially for large towns

 The absolute highest stage or probable stage The choice of a normal stage is decided by
with low frequency, e.g. 1% or 0.1% are the different factors:
basis for computations of levee spacing and a) character of a river and its size
capacity of storage reservoirs
b) variation of stages and discharges
 Mean annual flow is a normal stage for
lowland (sand-bed) river engineering c) quantity and character of transported
sediment
773 774

129
Determination of normal stage and normal 10. Sediment Control Methods
discharge
Considered in two general areas
d) aim of regulation: water supply, irrigation
the land surface
and/or drainage, navigation, bank
protection, etc. the fluvial channels and associated water
bodies
The normal stage for the same river may be
different in its different reaches and is usually The concept and the objective of watershed
lower downstream protection, management, and development are
based on the following premises:
1. Using each hectare of watershed land within its
capability for sustained use
775 776

Sediment Control Methods Sediment Control Methods

2. Applying cultural, vegetative, and supporting 4. Developing the water and related resources to
mechanical practices the extent necessary to meet present and
reasonably foreseeable future needs
3. Protection from most floods and control of
sediment production to the extent of: 5. Improving and managing vegetative cover of
trees and grass for sustained timber and forage
a) Maximizing net economic benefits from yields as well as watershed protection
floodwater and sediment damage reduction
6. Developing the most effective and economical
b) Reducing reservoir deposition and other forms combination of:
of sediment damages
a) Vegetative and supporting mechanical
c) Improving water quality and related purposes practices on the surface of the land
777 778

Sediment Control Methods Role of Planning Team

b) Structural measures Watershed project planning is a coordinated
7. Achieving a balance in water management analysis by a team of technicians representing
between the needs of each hectare, the various disciplines. The principal disciplines are:
watershed as a whole, and the river basin economics engineering
The primary objectives of watershed protection hydrology general soil science
programs, in broad terms, are:
geology plant technology
a) Preservation and beneficial use of soil and
Each is dependent on and interrelated with the
water resources
others in developing and selecting an
b) Reduction of erosion, floodwater, and economically feasible system of improvements to
sediment damages 779
meet the needs of a watershed. 780

130
 Role of Planning Team Watershed Survey
The information gathered for a specific watershed Objective
is presented in a final watershed work plan that: The objective of a watershed survey, in the context of
defines and describes the problems watershed protection, is to obtain the required
encountered; information:
presents the recommended preventive or 1) To quantitatively define the problems
corrective measures for control of the encountered;
problems; 2) To develop a program of works of improvement
to reduce, alleviate, or eliminate the problems;
evaluates the control measures for both
physical and economic effectiveness as a basis 3) To evaluate the effects of the selected works of
for justification on a project. improvement.
781 782

Watershed Survey Watershed Survey

Each watershed is a case unto itself and must be 3) Determining causal factors
so considered
4) Determining alternative preventive or
Six sequential steps that apply in dealing with corrective measures
sedimentation problems by means of watershed
projects are: 5) Selecting and evaluating the most feasible
preventive or corrective measures
1) Identifying and determining sedimentation
problems, existing and potential 6) Installing and maintaining preventive or
corrective measures
2) Determining the rates at which the problems
occur or may be expected to occur
783 784

Sedimentation Problems Sedimentation Problems

The unwanted deposition of eroded material is the extent of sediment deposition in channels
the most common expression of a sedimentation and in irrigation and drainage ditches and its
problem influence on watershed problems
Sedimentation problems begin with the
deposition of sediment in existing reservoirs
detachment of soil or rock materials by water or
wind the effect and extent of sediment in transport
Information must be obtained concerning the on water supplies
severity of the problems encountered
the location of the source of sediment in terms
the extent, both as to area and depth, of of erosion
damaging deposition on floodplain lands
785 786

131
 Survey Procedures Survey Procedures
Deposition on flood-plain land is often obvious to Diminished productivity
the observer, especially if overbank flow has can be recognized by comparisons of areal
recently occurred photographs taken in different years
If textural changes either to a composition coarser determined by interview or other historical
or finer than the original floodplain soils are not information
obvious on the surface, they must be determined
by boring Scouring of flood-plain land can be recognized by
the existence of channels eroded in flood-plains
The swamping of flood-plain land requires a
Severity of deposition in existing reservoir must
knowledge of past use of the land
be determined by reservoir sedimentation survey
787 788

Survey Procedures Survey Procedures

Stream bank erosion, although obvious in some Active gulling is obvious to an observer
instances, may require comparisons of aerial
photos rates of gullying to determine annual
contributions to sediment yield must be
On sloping, cultivated land sheet erosion can be developed either by comparison of aerial
recognized by photos taken in different years or by formulas

rills in fields The volume of material produced by stream bank

erosion must be determined by
deposition at the base of cultivated slopes
comparison of aerial photos taken at different
general appearance to trained eyes times
789 790

Survey Procedures Survey Procedures

interviews with local landowners often provide Where channel beds are composed of
excellent information on the rates of such noncohesive materials, the application of bed
erosion load transport equations can give estimates of
Degrading stream bed can be recognized by the the volume of materials moved each year
lack of deposition in the channel or by obvious
erosion of the bed Quantitative estimates of the production of
eroded materials are required in order to
Comparisons of available differing-age cross develop, recommend, and evaluate watershed
sections afford a means of developing rates of control measures that reduce generation of this
erosion and volumes of material produced from downstream damaging material
this source
791 792

132
 Reporting Erosion and Sediment Information Reporting Erosion and Sediment Information

The data must be presented in a manner that Tabulations in work plans are summations of
allows an evaluation of the information work sheets concerning these features developed
during field and office work
Locate and indicate existing sedimentation
problems on a map that is included in the final Sediment yields are shown in the tabulations if
work plan they can be determined with reasonable accuracy

Delineate by their area limitations critical Both current sediment yields and expected yields
sediment source areas after project completion are indicated in order
that the effects of the program in a given
Show swamping by conventional symbol watershed may be evaluated
793 794

Control Measures Control Measures

Reduction of erosion incident to cultivation, Cultural developments, including roads,
grazing, and timber production on watershed urbanization, mining, and industrial projects
land is a first step to correcting most sediment require increasing consideration in sediment
problems control programs
Principal sources of sediment Where upland sheet erosion problems prevail,
Sheet erosion and channel type erosion, control is usually by use of land treatment
including gullying where agriculture is measures
predominant land use
Structural measures are needed for control of
Gullying and stream channel erosion in forest channel type erosion
and range lands
795 796

Land Treatment Methods for Watershed Protection Land Treatment Methods for Watershed Protection
Land treatment measures are basic elements of 1. Vegetative Treatment – more effective in reducing
watershed projects and are the first increment in erosion and resultant sediment yields than in
project evaluation reducing in peak flow
They have significant effect in reducing a. Conservation cropping systems that encompass
sedimentation damages, particularly in the the growing of crops in combination with needed
highland agricultural areas and in areas where cultural and management measures
damaging sediment is derived primarily from b. Cover cropping with close-growing grasses,
sheet erosion legumes, or small grain in a cropping system
c. Critical area planting that is achieved by
establishing vegetative cover on excessive
sediment-producing areas
797 798

133
Land Treatment Methods for Watershed Protection Land Treatment Methods for Watershed Protection
d. Crop residue use – leaving plant remains Preservation and improvement of vegetative
e. Hayland planting cover provides a fourfold means of reducing
erosion and sediment yield in watersheds
f. Mulching – not produced on the site applied
i. The plant materials intercept the rainfall and
g. Pasture planting on new pastureland minimize the effect of raindrop impact
converted from other uses
ii. Increase infiltration and thereby reduce the
h. Tree planting in open areas to establish a stand
rate of surface runoff
of forest trees
i. Woodland interplanting in sparsely or iii. The roots and plant stems help to bind the
inadequately stocked stands soil into an erosion resistant mass
799 800

Land Treatment Methods for Watershed Protection Land Treatment Methods for Watershed Protection
3. Supporting Mechanical Field Practices
iv. Vegetation increases the roughness of the
ground surface reducing the velocity of a. Contour farming
overland flow and thereby its capacity to b. Contour furrowing on rangeland
erode and transport sediment
c. Contour strip-cropping
2. Protecting Existing Vegetative Cover on Forest and d. Terraces
Grazing Land
e. Diversions
a. Fire protection f. Grassed waterways
b. Grazing management g. Irrigation ditch and canal lining
h. Grade stabilization structures
801 802

Structural Measures Structural Measures

Watershed structural measures, the second They may include
element of a watershed project, are installed for:
1) Any form of earth work, either excavation or
land stabilization – primarily are used to fill
prevent land destruction or to reduce 2) Works of concrete, masonry, metal, or other
production of damaging sediment materials
water flow control – control damaging water 3) Vegetative planting associated with such
flows and water-borne sediment structural work

storage to provide user benefits

803 804

134
 Structural Measures Reservoirs
The major watershed structural measures Reservoirs are the basic structural measure in
commonly used include: watershed projects in which flood prevention is a
significant purpose
1) Reservoirs, both detention and multi-purpose
Flood retarding structures are detention dams
2) Stream channel improvement and with a fixed-capacity principal spillway and an
stabilization works emergency spillway
3) Debris and sedimentation basins The volume of storage between the inlet to the
principal spillway and the emergency spillway is
4) Levees, dikes, floodways and flood water for detention of floodwaters that are released at
diversions predetermined rate
805 806

Reservoirs Stream Channel Improvement and Stabilization

Multiple-purpose spillway reservoirs, in addition May consist of
to floodwater detention, may embody additional
capacity for such uses as municipal water supply, Excavation, which includes enlargement or
irrigation water, and recreation straightening of existing channels
Both types of reservoirs are effective in two ways: Removing brush and snags from existing
1) By the control of floodwater channels without excavation
2) By the impoundment of water-borne sediment These kinds of stream channel improvement,
The trap-efficiency may vary from 50% to nearly although used to control water flow, may affect
100% depending on the grain-size of incoming flood peaks by increasing the velocity of flow for
sediment and other factors the same volume of water
807 808

Stream Channel Improvement and Stabilization Stream Channel Improvement and Stabilization
Since they usually are used in combination with The method or combination of methods selected
floodwater retarding structures, peak flows with depends on the nature of the problem
system of structural measures are significantly
less than without the system In broad terms protective or control methods can
be classed be permeable or impermeable, flexible
Channel stabilization is required if channels or rigid, and permanent or temporary
themselves are deteriorating by down cutting,
bank erosion, or head cutting The control may be for bank stability, stream
gradient control, stream gradient reduction, or a
In general, channel improvement and stabilization combination of these purposes
methods fall into various categories
809 810

135
Stream Channel Improvement and Stabilization Stream Channel Improvement and Stabilization
In general, the purposes of channel stabilization 4) To hold the stream thalweg gradient and
structures are: prevent lowering with accompanying bank
undercutting
1) To produce a protective blanket that resists
5) To convey the water and sediment load with
forces of flowing water
maximum efficiency and minimum long-time
2) To create bank roughness and thereby reduce maintenance requirements
the velocity and erosion forces acting on the 6) To give water conveyance capacity and
streambank adjacent farmland drainage benefits
3) To divert flow away from erodible banks In all cases, continuity in design and treatment as
opposed to spotty control is of prime importance
811 812

Stream Channel Improvement and Stabilization Stream Channel Improvement and Stabilization
In any channel rectification plans, full
Installing stabilization works in anticipation that a
consideration is given to the
problem will develop can result in needless costs
stream hydraulics and erroneous location and type of control
probable flood frequencies, peaks and For this reason, delay of channel improvement or
durations (with and without flood detention certain phases of improvement is sometimes
provisions) desirable
stream periphery materials
suspended and bed load characteristics

813 814

Stream Channel Improvement and Stabilization Debris Basin

The problem may be A debris basin is a reservoir designed specifically
to trap sediment and debris
meander control
The capacity to be provided may be equal to
reduction in sediment load
the volume of debris expected to be trapped at
increase in conveyance capacity
the site during the planned useful life of the
rapid flood passage structure

or a combination of rectification needs the volume of debris anticipated during only

one or more major storms
815 816

136
 Debris Basin Sediment Control at River Intakes
Sediment control starts already with influencing
Debris basins are designed to reduce the amount the approaching flow
of generally coarse-grained sediment and debris
deposed in downstream channels and reservoirs, Experience shows that it is much easier to
or on highways, railroads, or urban and prevent the intrusion of sediment moving close
agricultural areas to the bed than of sediment in suspension

Any reduction in peak flow that may take place as The first step towards sediment control at an
a result of temporary water storage in debris intake, therefore, should be encouraging the
basins is incidental to their primary purpose concentration of sediment in the fluid layers
close to the bed

817 818

Sediment Control at River Intakes Control of Sediment Moving Close to the Bed

This can be achieved by applying appropriate Sediment rejection by means of sills and bars
river training measures capable of decreasing using spur dykes, training walls, etc as auxiliary
the flow velocity and of suppressing turbulence devices

Control of Sediment Moving Close to the Bed

Various methods and techniques use the
principle of the application of a horizontal
diversion separating the upper layers containing
mostly pure water from the sediment-laden
lower layers
819 820

Control of Sediment Moving Close to the Bed Control of Sediment Moving Close to the Bed
Sediment extraction by means of sediment Sediment ejection by means of vortex tube
tunnels and flashing devices sediment ejectors

821 822

137
 Control of Suspended Sediment Control of Suspended Sediment

Prevention of suspended sediment from entering Creating a settling basin before withdrawing the
the intake is very difficult, particularly when it is water by means of diversion dam
rather uniformly distributed over the water
depth

The settling process can be speeded by adding

chemicals

823 824

Control of Suspended Sediment Control of Suspended Sediment

Constructing a settling tank after diverting the Constructing a settling tank after diverting the
water water
The settling tank can be the longitudinal type The settling tank can be the circular type
(straight flow) (vertical vortex)

825 826

Other Structural Measures Other Structural Measures

They may be designed to control erosion, provide The ideal balance between initial cost and
proper drainage, distribute irrigation water, operation and maintenance costs for a structure
improve efficiency of water use, or for other are attained when the annual costs are a
purposes that provide desirable soil-water minimum for the planned life of the structure
relations for agriculture The ideal balance can be attained in the design of
the structure only to the degree that accurate
Among other supplemental structural measures
data are available on:
that may be considered in watershed protection
projects are 1. Durability of the materials involved
2. Cost of construction
levees and dikes, tide gates, floodways,
pumping plants, and floodwater diversions 3. Operation and maintenance costs
827 828

138
829

139