Open Channel Hydraulics

Faculty Name (Civil and Water Resource Engineering)

Chapters
1. Open channel flows and its classification

2. Energy and Momentum principles in open channel

flows

3. Uniform Flow and its Computations

4. Gradually Varied Flow and its Computation

5. Rapidly Varied Flow

6. Introduction to unsteady flow

By: Getenet N 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Chapter 3
Uniform Flow and its Computations

INTRODUCTION
 A flow is said to be uniform if its properties remain
constant with respect to distance.

 As mentioned earlier, the term uniform flow in open

channels is understood to mean steady uniform flow.

By: Getenet N 3/18/2025

 Faculty Name (Civil and Water Resource Engineering)

The depth of flow remains constant at all sections in a

uniform flow (Fig. 3.1).

Considering two sections 1 and 2, the depths

y1  y2  y0
and hence A1  A2  A0
 Faculty Name (Civil and Water Resource Engineering)

 Since Q =AV = constant , it follows that in uniform flow

V1 = V2 = V.

 Thus in a uniform flow, the depth of flow, area of cross-section and

velocity of flow remain constant along the channel.

 The trace of the water surface, channel bottom slope and energy
slope are parallel in uniform flow.

 Thus in uniform flow water surface slope, channel bed slope and
energy slope are the same.

By: Getenet N

5 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Expressing the velocity of a uniform flow

 For hydraulic computation the mean velocity of a

turbulent uniform flow in open channels is usually
expressed approximately by a so-called uniform flow
formula.

 Most practical uniform flow formula can be expressed in

the following general form

 Where V is the mean velocity, R is the hydraulic radius,

S is the energy slope, x and y are exponents and C is a
factor of flow resistance
By: Getenet N

6 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

CHEZY EQUATION
 By definition there is no acceleration in uniform flow. By applying the
momentum equation to a control volume encompassing sections 1
and 2, distance L apart, as shown in Fig. 3.1,

P1  W sin  Ff  P2  M 2  M1 ( 3.1)

• where P1 and P2 = the pressure forces and

• M1 and M2 = the momentum fluxes at section 1 and 2
respectively
• W = weight of fluid in the control volume and
• Ff = shear force at the boundary.
Since the flow is uniform,
P1  P2 and M1  M 2

By: Getenet N 13/18/2025

 Faculty Name (Civil and Water Resource Engineering)

Also W   AL and Ff   0 PL

 where τ0 = average shear stress on the wetted perimeter of length

P.
 and γ = unit weight of water.
 Replacing sin by S 0 (= bottom slope), Eq. (3.1) can be written as
 ALS 0   0 PL
A
or  0   S 0   RS 0 (3.2)
P

This equation plays a very important role in developing flow equations

which are common to all shapes of channels.

By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 Expressing the average shear stress  as  0  kV 2

0

 where k =a coefficient which depends on the nature of the surface

and flow parameters, Eq. (3.2) is written as

kV 2   RS0

Leads to V  C RS 0 (3.3)

 1
C
Where  k = a coefficient which depends on the nature of
the surface and the flow.

This is know as chezy velocity coefficient.

 Faculty Name (Civil and Water Resource Engineering)

Chezy resistance factor

Several forms of expressions for the Chezy coefficient have been
proposed by different investigators in the past. A few selected ones are
listed below:
1. Pavlovski Formula 1 X ( 3.4)
C R
n

in which 
x  2.5 n  0.13  0.75 R n  0.10 
and n = Manning’s coefficient.
 Faculty Name (Civil and Water Resource Engineering)

2. Ganguillet and Kutter Formula

1 0.00155
23  
n S0
C
 0.00155 n
1  23   (3.5)
 S0  R

in which n = Manning’s coefficient

3. Bazin’s Formula
87.0 (3.6)
C
1 M R

in which M = a coefficient dependent on the surface roughness.

By: Getenet N

11 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

MANNING’S FORMULA
 A resistance formula proposed by Robert Manning, an Irish
engineer, for uniform flow in open channels, is
1 23 12
V  R S0 ( 3.7)
n
where n = a roughness coefficient known as Manning’s n.

This coefficient is essentially a function of the nature of boundary

surface.

It may be noted that the dimensions of n are L T 

1 3

By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Equation (3.7) is popularly known as the Manning's

formula.

Owing to its simplicity and acceptable degree of

accuracy in a variety of practical applications, the
Manning’s formula is probably the most widely used
uniform flow formula in the world.

Comparing Eq. (3.7) with the Chezy formula, Eq. (3.3),

we have
 Faculty Name (Civil and Water Resource Engineering)

MANNING’S ROUGHNESS COEFFICIENT

 In the Manning's formula, all the terms except n are capable

of direct measurement.
 The roughness coefficient, being a parameter representing the
integrated effects of the channel cross-sectional resistance, is
to be estimated.
 The selection of a value for n is subjective, based on one's
own experience and engineering judgement
 However, a few aids are available which reduce to a certain
extent the subjectiveness in the selection of an appropriate
value of n for a given channel.
 Faculty Name (Civil and Water Resource Engineering)

 A comprehensive list of various types of channels their

descriptions with the associated range of values of n are listed.
 Some typical values of n for various normally encountered
channel surfaces prepared from information gathered from
various sources are presented in Table 3.2.

By: Getenet N
15 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

By: Getenet N
16 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

By: Getenet N
17 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Empirical Formulae for n

 Many empirical formulae have been presented for estimating

Manning's coefficient n in natural streams.

 These relate n to the bed-particle size.

 The most popular form under this type is the Strickler formula:
16
d 50
n
21.1
 Where d50 is in meters and represents the particle size in which 50 per
cent of the bed material is finer. For mixtures of bed materials with
considerable coarse-grained sizes.
16
d 90
 The above equation has been modified by Meyer n
26
By: Getenet N
18 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 where d90 = size in meters and in which 90 per cent of the particles are
finer than d90 .

 This equation is reported to be useful in predicting n in mountain

streams paved with coarse gravel and cobbles.

Factors Affecting n
 The Manning’s n is essentially a coefficient representing the integrated
effect of a large number of factors contributing to the energy loss in a
reach.

 Some important factors are:

(a) surface roughness, (b) vegetation,
(c) cross-section irregularity and (d) channel alignment of channel.

The chief among these are the characteristics of the surface.

By: Getenet N
19 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

EQUIVALENT ROUGHNESS
 In some channels different parts of the channel perimeter may have
different roughnesses.

 Canals in which only the sides are lined, laboratory flumes with glass
walls and rough beds, rivers with a sand bed in deep water portion and
flood plains covered with vegetation, are some typical examples.

 This equivalent roughness, also called the composite roughness,

represents a weighted average value for the roughness coefficient.

 Several formulae exist for calculating the equivalent roughness.

 All are based on certain assumptions and are approximately effective

to the same degree. One such method of calculation of equivalent
roughness is given below.

By: Getenet N
20 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 Consider a channel having its perimeter composed of N

types of roughness. P1 , P2 ,..., Pi ,..., PN are the lengths
N parts andn1 , n2 ,..., ni ,...nN
of these
are the respective roughness coefficients (Fig. 3.8).

Let each part Pi be associated with a Ai area such

partial
N
that

A  A  A  ...  A  A  total area
i 1
i 1 2 N

By: Getenet N
21 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

It is assumed that the mean velocity in each partial area is

the mean velocity
V for the entire area of flow, i.e.
V1  V2  ...  Vi  ...VN  V
By the Mannning’s formula
V1n1 V2 n2 Vi ni VN nN
S 12
0  2 3  2 3  ...  2 3  ...  2 3
R1 R2 Ri RN
Vn
 23
R
where n = equivalent roughness 23
 Ai  ni Pi 2 3
  
 A nP2 3

22
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

ni3 2 Pi
Ai  A 3 2
n P
 n 32
P
 A  A A
i
i
32
n P
i

n
 n 32
i Pi
2/3

P 23
( 3.8)
 This equation affords a means of estimating the equivalent roughness
of a channel having multiple roughness types in its perimeter.

23
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

• EXAMPLE 1 An earthen trapezoidal channel (n = 0.025) has a bottom

width of 5.0 m, side slopes of 1.5 horizontal:1 vertical and a uniform flow
depth of 1.1m. In an economic study to remedy excessive seepage from
the canal two proposals, viz. (a) to line the sides only and (b) to line the
bed only are considered. If the lining is of smooth concrete (n=0.012),
determine the equivalent roughness in the above two cases.
• Solution
• Case (a) : Lining on the side only
• Here for the bed n1  0.025 and P1  5.0 m

n2  0.012 and P2  2 1.10  1  1.5

2
• For the sides:

 3.966 m

24
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

P  P1  P2  5.0  3.966  8.966 m

Equivalent roughness,

n
5  0.025
1.5
 3.966 0.012 
1.5 2 3

8.966 23

0.085447
  0.020
4.31585
Case b: Lining on the bottom only
P1  5.0 m n1  0.012
P2  3.966 m n2  0.025
P  8.966
By: Getenet N
25 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Equivalent roughness

n
5  0.012
1.5
 3.966 0.025 
1.5 2 3

8.96623

0.079107
  0.018
4.31585
 Faculty Name (Civil and Water Resource Engineering)

UNIFORM FLOW COMPUTATIONS

 The Manning's formula and the continuity equation, Q =AV form the
basic equations for uniform-flow computations. The discharge Q is
then given by
1
Q  AR 2 3 S 01 2 (3.9)
n
 K S0
1
K AR 2 3
 where, n is called the conveyance of the channel and
expresses the discharge capacity of the channel per unit longitudinal
slope.

 The term nK  AR 2 3 is sometimes called the section factor for

uniform-flow computations.
example
 Faculty Name (Civil and Water Resource Engineering)

23
For a given channel, AR is a function of the depth of flow.
For example, consider a trapezoidal section of bottom
width =B and side slope m horizontal: 1 vertical. Then,
A  B  my y 
P  B  2 y m2  1 
B  my y  B  my 
53
y5 3
R
B  2 y m 1
2

 AR 2 3 
 B  2 y m2  1 
23
 f  B, m, y 

(eq 3.10)

example

By: Getenet N
28 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

For a given channel, B and m are fixed and AR  f ( y )

23

Figure 3.9 shows the relationship of eq.10 in a non

dimensional manner by plotting   AR8 3 vs y B for
2 3

different values of m B
 Faculty Name (Civil and Water Resource Engineering)

 It may be seen that for m  0 , there is only one value y B

for each value of  , indicating that for , m  0 is a single-
valued function of y.

 This is also true for any other shape of channel provided

that the top width is either constant or increases with
depth.
 we shall denote these channels as channels of the first
kind. key takeaway

Qn
Since AR 2 3  and if n and So are fixed for a
S0
channel, the channels of the first kind have a unique depth
in uniform flow associated with each discharge.
By: Getenet N
30 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 This depth is called the normal depth.

 Thus the normal depth is defined as the depth of flow at which a given
discharge flows as uniform flow in a given channel.

 The normal depth is designated as yo , the suffix ‘0’, being usually

used to indicate uniform-flow conditions.

 The channels of the first kind thus have one normal depth only. While
a majority of the channels belong to the first kind, sometimes one
encounters channels with closing top width.

 Circular and ovoid sewers are typical examples of this category.

Channels with a closing top-width can be designated as channels of
the second kind.
 Faculty Name (Civil and Water Resource Engineering)

By: Getenet N
32 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 The variation of AR with depth of flow for few channels of this

23

second kind is shown in Fig. 3.10.

23
 It may be seen that in some ranges of depth, AR is not a single-
valued function of depth.
 For example, the following regions of depth have two values for a
given value of : AR 2 3
(i) y/D>0.82 in circular channels,
(ii) y/B>0.71 in trapezoidal channels with m = -0.5,
(iii) y/B>1.30 in trapezoidal channels with m = -0.25.
 Thus in these regions for any particular discharge, two normal depths
are possible.

 As can be seen from Fig. 3.10, the channels

23
of the second kind will
have a finite depth of flow at which AR ,and hence the discharge for
a given channel, is maximum.
 Faculty Name (Civil and Water Resource Engineering)

Types of Problems
 Uniform flow computation problems are relatively simple. The available
relations are:
1.Manning's formula
2.Continuity equation
3.Geometry of the cross-section.
 The basic variables in uniform flow situations can be the discharge Q ,
velocity of flow V, normal depth Yo, roughness coefficient n , channel
slope So and the geometric elements (e.g. B and m for a trapezoidal
channel).
 There can be many other derived variables accompanied by
corresponding relationships. From among the above, the following five
types of basic problems are recognised.
 Faculty Name (Civil and Water Resource Engineering)

 Problems of the types 1, 2 and 3 normally have explicit solutions

and hence do not present any difficulty in their calculations.

 Problems of the types 4 and 5 usually do not have explicit solutions

and as such may involve trial-and-error solution procedures. A
typical example for each type of problem is given below.
 Faculty Name (Civil and Water Resource Engineering)

EXAMPLE 2 A trapezoidal channel 5.0 m wide and having a side slope

of 1.5 horizontal: 1 vertical is laid on a slope of 0.00035.The roughness
coefficient n=0.015. Find the normal depth for a discharge of 20 m3/s
through this channel.
Solution
Let y0  normal depth
Area A  5.0  1.5 y0  y0

Wetted perimeter P  5.0  2 3.25 y0

 5.0  3.606 y0

RA P
5.0  1.5 y0  y0
5.0  3.606y0 
36
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

The section factor Qn

AR 23

S0

Algebraically, yo can be found from the above equation by the trial-

and-error method. The normal depth is found to be 1.820 m.

5.0  1.5 y0  y05 3  20  0.015

53
 16.036
5.0  3.606y0 23
0.00035 12

37
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

EXAMPLE 3 A concrete-lined trapezoidal channel (n=0.0155) is to have

a side slope of 1.0 horizontal: 1 vertical. the bottom slope is to be 0.0004.
Find the bottom width of the channel necessary to carry 100 m3/s of
discharge at a normal depth of 2.50 m.
Solution
Let B = bottom width. Here yo = normal depth= 2.20 m
Area A  B  2.5 2.5

Wetted perimeter  
P  B  2 2  2.5  B  7.071
Qn 100 0.015
  75  AR 2 3 B  2.5 2.55 3  75.0
S0 0.0004 B  7.0712 3
By trial-and-error B = 16.33 m.

38
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Computation of Normal Depth

 the calculation of normal depth for a trapezoidal channel involves a
trial-and-error solution.
 This is true for many other channel shapes also. Since practically all
open channel problems involve normal depth, special attention
towards providing aids for quicker calculations of normal depth is
warranted.
 A few aids for computing normal depth in some common channel
sections are given below.
Rectangular Channel
(a) Wide Rectangular Channel

39
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

For a rectangular channel, (Fig. 3.13)

Area A  By0
Wetted perimeter P  B  2 y0

By0 y0
Hydraulic radius R 
B  2 y0 1  2 y0 B
As yo/B , the aspect ratio of the channel decreases, R yo . Such
channels with large bed-widths as compared to their respective depths are
known as wide rectangular channels.

In these channels, the hydraulic radius approximates to the depth of flow.

Considering a unit width of a wide rectangular channel,

40
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

A  y0 , R  y0 and B  1.0
Q 1
 q  discharge per unit width  y05 3 S 01 2
B n
35
 qn 
y0    (3.11)
 S0 

This approximation of a wide rectangular channel is found applicable for

rectangular channels with y0 B < 0.02.

(b) Rectangular Channels with y0 B  0.02

Qn
For these channels  AR 2 3
S0

41
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

AR 2 3 
 By 0 53

 y 0 B 53
B 83

B  2 y0 2 3 1  2 y0 B 2 3
0 
23 53
Qn AR
 83    0  ( eq 3.12)
1  20 
83 23
S0 B B
y0
where 0 
B
Equation (eq 12) when plotted as   0  vs  0 will provide a
non-dimensional graphical solution aid for general
application.

Qn

Since S0 B8 3 , one can easily find y0 B from this plot for
any combination of Q, n, So and B in a rectangular channel.
42
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Trapezoidal Channel

Following a procedure similar to the above, for a trapezoidal section of

side slope 1:m, (Fig. 3.14)
Area A  B  my0  y0

Wetted perimeter 
P  B  2 m2  1y0 
B  my0 y0
Hydraulic radius RA P
B  2 m2  1y0 
43
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Non-dimensionalising the variables,

Qn
 AR 
 B  my0  y05 3
53

 
23
23
S0 B  2 m  1 y0
2

AR 2 3

Qn

1  m0  0     , m
53 53

B8 3 
S0 B8 3 1  2 m 2  1
0 
23 0
(3.13)
y0
where 0 
B
A curve of  vs  0 with m as the third parameter will provide
a general normal depth solution aid. It may be noted that m
=0 is the case of a rectangular channel.
There is a table in the appendix which will be very useful in
quick solution of a variety of uniform-flow problems.

44
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

EXAMPLE 4 Solve the problem of Example 2 by using Table 3A.1.

Solution
For example
Qn 16.036 AR 2 3
   0.21936  8 3
S0 B 83
583
B

Looking at Table 3A.1 under m = 1.5

  0.21485 for y0 B  0.360
  0.22619 for y0 B  0.370
By interpolation, for   0.21936 , y0 B  0.364
Hence y0  0.364 5.0  1.820m

45
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Circular Channel
Let D be the diameter of a circular channel (Fig. 3.15) and 2 be the,
angle in radians subtended by the water surface at the center.
A = area of the flow section
=area of the sector-area of the triangular portion

46
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

1 2 1
 r0 2   2r0 sin r0 cos
2 2 2

D
2  sin 2 
8
1 2

 r0  2  r02 sin 2
2

P = wetted perimeter  2r0  D

r0  y0  2 y0 
Also cos   1  
r0  D 
Hence   f  y0 D 
1
Q  AR 2 3 S 01 2
n
47
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Assuming n = constant for all depths

Qn A5 3 D10 3 2  sin 2 

53
 23  53
S0 P 8 D 2 3

Non-dimensionalising both sides

AR 2 3 1   sin 2 
53
Qn
 83 
S0 D 83
D 32 23

   y0 D  eq 3.14
The functional relationship of Eq. 3.14 has been evaluated for various
values of yo/D and is given in Tables in Appendix. Besides AR 2 3 / D 8 3, other
geometric elements of a circular channel are also given in the table which is
very handy in solving problems related to circular channels.

48
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 Using this table, with linear interpolations wherever necessary, the

normal depth for a given D, Q, n and So in a circular channel can be
determined easily.

 As noted earlier, for depths of flow greater than 0.82D , there will be
two normal depths in a circular channel.

 In practice, it is usual to restrict the depth of flow to a value of 0.8D

to avoid the region of two normal depths.

 In the region y/D>0.82, a small disturbance in the water surface

may lead the water surface to seek alternate normal depths, thus
contributing to the instability of the water surface.

49
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

EXAMPLE A trunk sewer pipe of 2.0 m diameter is laid on

a slope of 0.0004. Find the depth of flow when the discharge
is 2.0m3/s .(Assume n=0.014.)
Solution AR 2 3 Qn 2.0  0.014
 
0.0004  2.0
83
D8 3 S0 D8 3

 0.22049

From Table 2A.2 AR 2 3 y0

83
 0.22004 at  0.62
D D
 0.22532 at y0 D  0.63

AR 2 3 D8 3  0.22049, y0 D  0.621
By interpolation, for
y0  1.242 m
The normal depth of flow

50
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

HYDRAILICALLY-EFFICIENT CHANNEL SECTION

 The conveyance of a channel section of a given area increases with a

decrease in its perimeter.

 Hence a channel section having the minimum perimeter for a given area
of flow provides the maximum value of the conveyance.

 With the slope, roughness coefficient and area of flow fixed, a

minimum perimeter section will represent the hydraulically-efficient
section as it conveys the maximum discharge.

 This channel section is also called the best section.

 Of all the various possible open channel sections, the semicircular shape
has the least amount of perimeter for a given area.
By: Getenet N 51 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

(a) Rectangular Section

Bottom width = B and depth of flow = y
Area of flow A  By  constant
Wetted perimeter P  B  2y
A
  2y
y
dP A
If P is to be minimum with A = constant  2 20
dy y
Which gives A  2 ye2
ye
ye  Be 2 , Be  2 ye and Re 
• 2
By: Getenet N
52 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 the suffix ‘e‘ denotes the geometric elements of a hydraulically-efficient

section.
 Thus it is seen that for a rectangular channel when the depth of flow is
equal to half the bottom width i.e. when the channel section is a half-
square of bottom width, a hydraulically-efficient section is obtained.

53
By: Getenet N
3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

(b) Trapezoidal Section

Bottom width = B , side slope = m horizontal: 1 vertical
Area A  B  my y  constant
A
B   my (3.15)
y
Wetted perimeter P  B  2 y m2  1
A
  my  2 y m 2  1
y
Keeping A and m as fixed, for a hydraulically-efficient section,

By: Getenet N
54 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

•

A  2 1 m  m y
2
 2
e


Be  2 ye 1  m2  m 

2 1  m  my
2 2

22 1  m  m y
R 
e
e
 ye 2
2
e

By: Getenet N
55 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

A hydraulically-efficient trapezoidal section having the proportions given

by Eq.s (3.18) through (3.19) is indicated in Fig. 3.19.
Let O be the center of the water surface. OS And OT arc
perpendiculars drawn to the bed and sides respectively.

OS  ye OR
OT  OR sin 
m2  1

By: Getenet N
56 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Substituting for Be form Eq. (3.17), 1

OR  Be  mye

Be  2 ye 1  m2  m  2

OR  ye 1  m2
OT  OS  ye
Thus the proportions of a hydraulically-efficient trapezoidal
section will be such that a semicircle can be inscribed in it

In the above analysis, the side slope m was held constant. However, if m
is allowed to vary, the optimum value of m to make Pe most efficient
is obtained by putting dPe = 0. Form Eqs (3.18) and (3.16)
dm

By: Getenet N
57 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)


Pe  2 A 2 1  m 2  m  (3.20)

dPe
Setting = 0 in Eq. (3.20) gives
dm
1
mem   cot  em  60
3
where the suffix ’em’ denotes the most efficient section. Further,

 1 
Pem  2 yem  2 1  1 3    2 3 yem (3.21a)
 3
 1  2
Bem  2 yem  2 1  1 3   yem (3.21b)
 3 3
By: Getenet N
58 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 1  2 (3.21c)
A   2 11 3   yem  3 yem
2

 3
If L = length of the inclined side of the canal, it is easily seen that
2
Lem  yem  Bem from Eq.(3.21b)
3

Thus the hydraulically most efficient trapezoidal section is one-half of a

regular hexagon.

Using the above approach, the relationship between the various

geometrical elements to make different channel shapes hydraulically
efficient can be determined. Table 3.3 contains the geometrical relation
of some most efficient sections.

By: Getenet N
59 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

By: Getenet N
60 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

EXAMPLE 3.14 A slightly rough brick-lined trapezoidal channel with

n=0.017 carrying a discharge of 25.0 m3/s is to have a longitudinal slope of
0.0004. Analyze the proportions of
(a) an efficient trapezoidal channel section having a side of 1.5 horizontal:
1 vertical,
(b) the most efficient-channel section of trapezoidal shape.
Solution
Case (a): m = 1.5
For an efficient trapezoidal channel section,


Ae  2 1  m2  m ye2 
  2 1  1.5  1.5  ye2  2.1056ye2
2

 

By: Getenet N
61 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Re  ye 2, Q  25.0 m3 s
23
1 2  y 
25   (2.1056ye ) e  (0.0004)1 2
(0.017)  2 

 ye  2.830 m
Be  2  2.830 1  1.5  1.5 
2

 
 1.714 m

By: Getenet N
62 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Case (b): For the most-efficient trapezoidal channel section

1
mem   0.57735
3
yem
Aem  1.732 y , 2
em Rem 
2
23
1 2  yem 
25   (1.732 yem )  (0.0004)1 2
(0.017)  2 

83
yem  19.4753  yem  3.045 m
2
Bem   3.045  3.516 m
3

By: Getenet N
63 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

COMPOUND SECTIONS
 Some channel sections may be formed as a combination of
elementary sections.
 Typically natural channels, such as rivers, have flood plains which
are wide and shallow compared to the deep main channel.
 Figure 3.21 represents a simplified section of a stream with flood
banks. Channels of this kind are known as compound sections.

By: Getenet N
64 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 Consider the compound section to be divided into subsections by

arbitrary lines.
 These can be either extensions of the deep channel boundaries as in
Fig. 3.21 or vertical lines drawn at the edge of the deep channels.

 Assuming the longitudinal slope to be the same for all subsections, it is

easy to see that the subsections will have different mean velocities
depending upon the depth and roughness of the boundaries.

 Generally, overbanks have larger size roughness than the deeper main
channel.
 If the depth of flow is confined to the deep channel only ( i.e. y<h),
calculation of discharge by using the Manning's formula is very simple.

By: Getenet N
65 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 However, when the flow spills over into the flood plain ( i.e. y>h), the
problem of the discharge calculation is complicated as the calculation may
give a smaller hydraulic radius for the whole stream section and hence the
discharge may be underestimated.

 This underestimation of the discharge happens in a small range of y , say

h<y<ym, where ym = maximum value of y beyond which the
underestimation of the discharge as above does not occur.

 For a value of y >ym , the calculation of the discharge by considering the

whole section as one unit would be adequate.

 For values of y in the range h<y<ym , the channel has to be considered to

be made up of sub-areas and the discharge in each sub-area determined
separately.
By: Getenet N
66 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 The total discharge is obtained as a sum of discharges through all such

sub- areas.
 The value of ym would depend upon the channel geometry. However,
for practical purposes the following method of discharge estimation
can be adopted.
(i) The discharge is calculated as the sum of the partial discharges in
the sub-areas; for e.g. units 1,2 and 3 in Fig.3.21.

Q p   Qi   Vi Ai

(ii) The discharge is also calculated by considering the whole section as

one unit, (portion ABCDEFGH in Fig. 3.21), say Qw.

By: Getenet N
67 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

(iii) The larger of the above two discharges, Qp and Qw, is adopted as
the discharge at the depth y.

For determining the partial discharges Qi and hence Qp in step one

above, two methods are available.

1. Posey’s method

 In this method, while calculating the wetted perimeter for the sub
areas, the imaginary divisions (FJ and CK in Fig. 3.21) are considered
as boundaries for the deeper portion only and neglected completely in
the calculation relating to the shallower portion. This way the shear
stress that occurs at the interface of the deeper and shallower parts is
empirically accounted for.
By: Getenet N
68 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

2. Zero shear method

 Some investigators mostly in computational work, treat

the interface as purely a hypothetical interface with zero
shear stress. As such, the interfaces are not counted as
perimeter either for the deep portion or for the shallow
portion.

By: Getenet N
69 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

EXAMPLE 3.16 For the compound channel shown in Fig.

3.22 determine the discharge for a depth of flow of (a) 1.20

m and (b) 1.60 m. Use Posey's method for computing
partial discharges.

By: Getenet N
70 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Solution
Case (a):
(i) Partial area Discharge by Posey’s Method
Sub-area 1: A1  7.0  0.3  2.1 m 2 P1  0.3  7.0  7.3 m
R  2.1 7.3  0.288 m

QP1 
1
    
 2.1  0.288 0.0002
23
12

0.02
 0.647 m3 s
Similarly QP3  0.647 m3 s

Sub-area 2: A2  3.0 1.2  3.6 m 2

By: Getenet N
71 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

P2  3.0  1.2  1.2  5.4 m R2  3.6 5.4  0.667 m

 3.6 0.667 0.0002  1.943 m3 s

1
QP 2 
23 12

0.02
QP  total discharge by partial areas
 0.647  0.647  1.943  3.237 m3 s

(ii) By the Total-Section Method

A  2.1  2.1  3.6  7.8 m 2

P  0.3  7.0  0.9  3.0  0.9  7.0  0.3  19.4 m

R  7.8 19.4  0.402 m

By: Getenet N
72 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 7.80.402 0.0002
1
QW 
23 12

0.02
 3.005 m3 s
Since QW  QP , the discharge in the channel is taken as
Q  QP  3.237 m3 s

Case (b): y0  1.6 m

(i) Partial Area Discharge by Posey’s Method
Sub-area 1: A1  7  0.7  4.9 m 2
P1  0.7  7.0  7.7 m
R1  4.9 7.7  0.636 m

 4.9 0.636 0.0002

1
QP1 
23 12
= Qp3 =2.563m3/s
0.02
By: Getenet N
73 3/18/2025
  2Faculty
.563Name 3
m (Civil
s and Water Resource Engineering)

Sub-area 2: A2  3 1.6  4.8 m 2

P2 = 1.6x2 +3 = 6.2m
R2 = 4.8/6.2 = 0.774m

 4.8 0.774 0.0002

1

23 12
QP 2
0.02
 2.862 m3 s
QP  2.563  2.563  2.862  7.988 m 2 s
(ii) By the Total –Section Method
A  4.9  4.9  4.8  14.6 m 2
P =0.7+7+0.9+3+0.9+7+0.7 = 20.2m

By: Getenet N
74 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

R  14.6 20.2  0.723 m

 14.6  0.723 0.0002

1
QW 
23 12

0.02

 8.315 m3 s

Since QW  QP , the discharge in the channel is taken as .

Q  QW  8.315 m3 s

By: Getenet N
75 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

A1  7.0  0.3  2.10 m 2

P1  0.3  7.0  7.3 m
R1  2.1 7.3  0.288 m

 0.647 m3 s
QP3  0.647 m3 s
By: Getenet N
76 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Sub-area 2: Area A2  3.0 1.2  3.60 m 2

Perimeter P2  3.0  0.9  0.9  4.8 m
R2 = 3.6/4.8 = 0.75m

 3.6  0.750  0.0002

Partial discharge 1

23 12
QP 2
0.02
 2.10 m3 s
Total discharge by partial areas QP  QP1  QP 2  QP 3

QP  0.647  2.100  0.647  3.395 m3 s

(ii) By the Total-Section Method:
Area A  2.1  2.1  3.6  7.8 m 2
Perimeter P  0.3  0.7  0.9  3.0  0.9  0.7  0.3  19.4 m

By: Getenet N
77 3/18/2025
 R  7.8 19 .4 
Faculty 0.402
Name m Water Resource Engineering)
(Civil and

 7.8  0.402  0.0002

1
Qw 
23 12
Discharge
0.02
 3.005 m3 s

Since Qp  Qw, the discharge in the channel is taken as

Q  Q p  3.395 m3 s

By: Getenet N
78 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

GENERALISED-FLOW RELATION (Reading Assignment)

Since the Froude number of the flow in a channel is
V Q 2 F 2 A3 (3.57)
F 
gA T g T

If the discharge Q occurs as a uniform flow, the slope S0 required to

sustain this discharge is, by the Manning’s formula,
Q2n2 (3.58)
S0  2 4 3
AR

Substituting Eq. (3.57) in Eq. (3.58) and simplifying

By: Getenet N
79 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

or F 2 gn2 P 4 3
S0 
TA1 3
(3.59)
P4 3
 1 3  f y
S0
2 2
F gn TA
For a trapezoidal channel of side slope m,

S0


B  2 m  1y 2
 43

F 2 gn2 B  2my0 B  my  y0 1 3 (3.60a)

Non-dimensionalising both sides, through multiplication by B1 3,

 S0 B 
13
 2 2  

1 2 m 1  2

43

(3.60)
 F gn  1  2 m 1  m 13

By: Getenet N
80 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

 S 0 B1 3 
in which  y0 B. Designating  2 2   S*
 F gn 

generalized slope S*  f m,  (3.61)

Equation (3.60) represents the relationship between the various

elements of uniform flow in a trapezoidal channel in a generalized
manner.
The functional relationship of Eq. (3.60) is plotted in Fig. 3.23. This
figure can be used to find, for a given trapezoidal channel, (a) the bed
slope required to carry a uniform flow at a known depth and Froude
number and (b) the depth of flow necessary for generating a uniform flow
of a given Froude number in a channel of known bed slope.

By: Getenet N
81 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

For a rectangular channel, m =0 and hence Eq. (3.60) becomes

S* 
1  2 
43
(3.62a)
 13

For a triangular channel, B = 0 and hence Eq. (3.60) cannot be used.

However, by redefining the generalized slope for triangular channels,
by Eq. (3.60a).

1 3 1 m 
 S*t  2 
S 0 y1 3 2

2 2 2 
 (3.63)
F gn  m 

By: Getenet N
82 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Roots and Limit Values of S. for Trapezoidal Channels

Equation (3.60) can be written as

S*3 
1 2 1 m 2

4

(3.64)
1  2m 3 1  m 

This is a fifth-degree equation in  , except for m = 0 when it reduces to a

fourth, degree equation. Out of its five roots it can be shown that (a) at
least one root shall be real and positive and (b) two roots are always
imaginary. Thus depending upon the value Of m and S* , there may be one,
two or three roots.
dS*
The limiting values of S* are obtained by putting, 0
d
which results in

8 1  m 2 1  m 1  2m   1  2 1  m 2 
1  10m  10m    02 2
(3.65)
By: Getenet N
83 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

By: Getenet N
84 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)


8 1  m 2 1  m 1  2m   1  2 1  m 2 
1  10m  10m    0
2 2
(3.65)
Solving Eq. (3.65) the following significant results are obtained
1. Rectangular channels (m=0), a single limiting
value with S = 8/3 and  = l/6 is obtained.
*
2. Between m = 0 and m = 0.46635 there are two
limiting values.
3. At m = 0.46635, the two limit values merge into
one at S*= 2.l545 and =0.7849.
4. For m > 0.46635, there are no limiting points.
These features are easily discernible from Fig. 3.23.

By: Getenet N
85 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

CRITICAL SLOPE AND LIMIT SLOPE

The slope of a channel which carries a given discharge as a uniform

flow at the critical depth is called the critical slope, S c . The condition
governing the critical slope in any channel can be easily obtained
from Eq. (3.60) by putting F=1.0. For trapezoidal channels, by
13
S B
denoting the generalized critical slope, c 2  S*c and yc   c , the
gn B
behaviour of S *c can be studied using Fig 3.23.

All the conclusions derived in the previous section for S* will also
apply to S*c  c relationship.

By: Getenet N
86 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)
Sc
For a channel of given shape and roughness S*c will have a least value
under conditions corresponding to a limit value of S c . The least value of
is called the limit slope, S Lc Keeping the critical slope and limit slope in
mind, Fig. 3.23 can be studied to yield the following points:

1. For a trapezoidal channel of given geometry and roughness, a given

depth of critical flow can be maintained by one and only one critical slope.

However, for a given critical slope there can be more than one critical
depth.

2. For channels of the second kind ( m is negative) and for rectangular

channels (m=0), only one limit slope exists.

By: Getenet N
87 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

Slopes flatter than this cannot be critical and the slopes steeper than
this can be critical at two different depths. For a rectangular
channel, the limit value of S *cis 8/3 at  c = 1/6.
3. when m  0.46635 , any slope can be critical and for each slope
there will be only one critical depth. There are no limit slopes in
this range.
For m = 0.46635, the limit value of S*c is 2.15446 at c = 0.7849.
4. For 0  m  0.46635 , there are two values of S Lc1 limit slopes,
and S Lc 2 with S Lc1  S Lc2 .
(a) For S Lc2  S c  S Lc1 , there are three critical depth for each
value of ;
Sc

By: Getenet N
88 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

the largest of these, however, may be impracticably large.

(b) For S c  S Lc1 or Sc  S Lc 2 there are two critical depths.
(c) For S c  S Lc1 or S c  S Lc2 , there is only one critical depth for
each value of the slope.

By: Getenet N
89 3/18/2025
 Faculty Name (Civil and Water Resource Engineering)

End of Chapter 3

Thank you

By: Getenet N
3/18/2025