CHAPTER FOUR

GRADUALLY VARIED FLOW (GVF)

A steady non-uniform flow in a prismatic
channel with gradual changes in its water-surface
elevation is termed as gradually-varied flow
(GVF).
 The back water produced by a dam or weir
across a river and
 the drawdown produced at a sudden drop in a
channel are few typical examples of GVF.
 In a GVF, the velocity varies along the channel and
 consequently the bed slope, water surface slope, and
energy slope will all differ from each other.
 Regions of high curvature are excluded in the analysis
of this flow.
1
…CON’T
The two basic assumptions involved in
the analysis of GVF are: -
1. The pressure distribution at any section
is assumed to be hydrostatic
Slope of the channel is small.
In GVF bed slope, water surface slope
and energy slope are all different.
Hydrostatic normal acceleration =0

2
…CON’T
2. The resistance to flow at any depth is
assumed to be given by the
corresponding uniform flow equation,
such as Manning’s formula, with the
condition that the slope term to be used
in the equation is the energy slope and
not the bed slope.
Thus, if in a GVF the depth of flow at a
section is y, the energy slope Sf is given
by:
3
Differential equation of GVF

4
…CON’T
 Consider the total and specific energy H and
E respectively of a gradually-varied flow .

 Differentiation the above equation with respect to x

shows that:-

 In this equation the meaning of each term is as

follows:- 5
…CON’T
1. Represents the energy slope
 Denoting it by Sf:-

2. Denotes the bottom slope

3. Represents the water surface slope

relative to the bottom of the channel.
6
…CON’T
4.

If a value of the kinetic energy correction factor α greater than

unity is to be used,

7
Other forms of GVF equation
1. If K= conveyance at any depth y and
Ko= conveyance corresponding to the normal depth
yo, then:-

8
9
…CON’T
2. if Qn represents the normal discharge at a depth y
and Qc denotes the critical discharge at the same
depth yc and Q is the given discharge of the GVF at a
depth y.

10
Classification of flow profiles
In a given channel, y0 and yc are two fixed depths if
Q, n and So are fixed.
Also there are three possible relations between y0
and yc as:-

and
further there are two cases where doesn’t exist ,
i.e. when
the channel bed is horizontal (So=0),
when the channel has an adverse slope, (So is –ve).
based on the above, the channels are classified into
five categories 11
…CON’T
classification of channels

12
… CON’T

13
…CON’T
Depending upon the channel category and region of
flow, the water-surface profiles will have either of
the following characteristic shapes.
1. Back water curves: if the depth of flow
increases in the direction of flow.
2. Drawdown curves: if the depth of flow
decreases in the direction flow.
 The dynamic equation of GVF expresses the
longitudinal surface slope of flow with respect to the
channel bottom is given by:-

14
15
16
…con’t
the behavior of dy/dx at certain key depths is noted by
studying the differential equation of GVF.

The water surface meets the normal depth line

asymptotically
 As y →∞ dy /dx →So The water surface meets a very
large depth as a horizontal asymptote /tends to be
horizontal/
 As y →yc dy /dx → ∞ The water surface meets the
CDL vertically /normally/
In reality, high curvatures at the critical depth zone violet
the assumption of GVF.
17
…con’t
At critical depth the curves are indicated by
dashed lines to remind that the GVF
equation is strictly not applicable in that
Neighborhood.
As y →0 dy /dx → ∞ The water surface
meets the channel bottom normally.
Based on the above information, the various
possible gradually-varied flow profiles are
grouped into twelve types.

18
19
...CON’T
In reality the GVF profiles, especially M1, M2and
H2 profiles, are very flat.

20
All curves in region 1 &3 have positive slopes, known as back water
curves and all curves in region 2 have negative slopes and are
referred as draw down curves. 21
some features of flow profiles
A). Type M flow profiles
Obstruction to the flow, such as weirs, dams,
control structures and natural features,
such as bends, produce M1 curves.

22
…CON’T
The M2 profiles occur at a sudden drop
in the bed of the channel, at constriction
type of transitions and at the canal outlet into
pools.

23
…con’t
Where a supercritical stream enters a mild-slope channel,
the M3 type of profile occurs.
The flow leading from a spillway or a sluice gate to a mild
slope forms a typical example.
The beginning of the M3 curve is usually followed by a small
stretch of rapidly varied flow and the down-stream is
generally terminated by a hydraulic jump.
Compared to M1 and M2 profiles, M3 curves are of
relatively short length

24
B). Type S profiles
The S1 profile is produced when the flow from a
steep channel is terminated by a deep pool created by
an obstruction, such as weir or dam.
At the beginning of the curve, the flow changes from
the normal depth (super critical flow) to subcritical
flow through a hydraulic jump

25
…CON’T
Profile of S2 type occurs at the entrance
region of a steep channel leading from a
reservoir and at the break of grade from mild
slopes to steep slope. Generally S2 profiles
are of short distance.

26
… CON’T
Free from a sluice gate with a steep slope on
its downstream is of the S3 type.
The S3 curve also results when a flow exists
from a steeper slope to a less steep slope.

27
c)Type C profiles
C1 and C3 curves are very rare and are highly
unstable
D)Type H profiles
It is obvious that there is no region 1 for a horizontal channel
ay yo=∞.
The H2 and H3 profiles are similar to M2 and M3 profiles
respectively. However, the H2 curve has a horizontal
asymptote.

28
e)type A profiles
Adverse slopes are rather rare and A2 and A3 curves
are similar to H2 and H3 curves respectively.
The profiles are of very short length.

29
 Control section
A control section is defined as a section in which
a fixed relationship exists between the
discharge and depth of flow.
Weirs, spillways, sluice gates are some typical
examples of structures which give rise to control
sections.
The critical depth is also a control point.
Any GVF profile will have at least one control
point.
Subcritical flows have controls in the
downstream end while supercritical flows are
governed by control section existing at the
upstream end of the channel section.
30
…CON’T
A few typical control sections are indicated in
the figure below.

31
32
 Analysis of flow profiles
To determine the resulting water-surface profile
in a given case, one should be in a position to
analyses the effect of various channel sections
and controls connected in series.
Break in grade
Simple situation of a serious combination of
two channel sections with differing bed
slopes are connected in the figure below.
The grade changes acts as a control section and
this can be classified as a natural control.
Various combinations of slopes and the resulting
GVF profiles are presented in the figure below.
It may be noted that in some situation there can
be more than one possible profile.
33
… CON’T
 Procedure to draw the profile of GVF in grade
transition
Draw the longitudinal section of the system.
Calculate the critical depth and normal depths
of various reaches
draw the CDL and NDL in all reaches. Since yc
doesn’t depend upon the slope CDL will be
constant above the channel bed in all slopes.
Mark all the controls, both the imposed as well
as natural controls.
Identify the possible profiles.
The normal depth for the mild slope is lower
than that of the milder slope in this case the
second depth acts as a control.
34
1. The following figures show GVF profiles at breaking grade

35
36
37
38
GVF Computations
Almost all major hydraulic engineering
activities in the free surface flow involve the
computation of GVF profiles.
The flow computation is needed to analyze
problems such as
Determination of effect of a hydraulic
structure on the channel
Inundation due to a dam or weir
construction
Estimation of flood zone
39
Cont…
Methods of GVF computations
1. Numerical method
2. Graphical method
Out of these the graphical method is
practically obsolete and is seldom used.
Further the numerical method is the most
extensively used technique. In the form of a
host of available compressive Software’s it is
the only method available to solve practical
problem in natural channels.
The direct integration technique is essentially
of academic interest. 40
1. Numerical method
The numerical solution procedure to solve GVF problems
can be broadly classified into two categories as:
1. Simple numerical methods
These were developed primarily for hand computation.
they usually attempt to solve the energy equation either in
the form of the differential energy equation of GVF or in
the form of the Bernoulli equation.
Two commonly used simple numerical methods to solve
GVF problems are :
A. Direct-step method
B. Standard-step method
2. Advanced Numerical Methods
These are normally suitable for use in digital computers
as they involve a large number of repeated calculations.
They attempt to solve the differential equation of GVF.
41
 A. The Direct Step Method (Distance from Depth)
The direct step method is a simple method applicable to
prismatic channels.
Depths of flow are specified and the distances between
successive depths are calculated (x.
The equation may be used to determine directly (with means
explicit) the distance between given differences of depth y.
Consider the differential energy equation of GVF.
𝒅𝑬
= 𝑺𝒐 − 𝑺𝒇
𝒅𝒙
∆𝑬
Writing this in the finite difference form = 𝑺𝑶 − 𝑺𝒇
∆𝒙
Where 𝑺𝒇 =average friction slope in the reach Δx.
∆𝑬
∆𝒙 = ………..(4) and Between two sections 1 and 2
𝑺𝒐 −𝑺𝒇
(𝑬𝟐 −𝑬𝟏 )
(𝑿𝟐 −𝑿𝟏 ) = ∆𝑿 = 𝟏 …………(5)
𝑺𝑶 − (𝑺𝒇𝟏 +𝑺𝒇𝟐 )
𝟐
This equation is used to calculate the GVF profile.
42
B. Standard step method
While the direct step method is suitable for use in
prismatic channels and there are some basic
difficulties in applying it to natural channels.
As already indicated in natural channels the
cross sectional shapes are likely to vary from
section to section and also the cross- section
information is known only at a few locations
along the channel.
Thus the problem of computation of the GVF
profile for a natural channel can be stated as:
Given stage at one section it is required to
determine the stage at the other section.
The sequential determination of the stage as a
solution of the above problem will lead to the
GVF profile.
43
…cont
The solution of the above problem is obtained by
trial and error solution of the basic energy
equation.
Consider figure 4.6 which shows two sections 1
and 2 at a distance Δx.
Calculations are assumed to proceed up stream.
equating the total energy at section 1 and 2.
𝑽𝟐𝟐 𝑽𝟐𝟏
𝒁𝟐 + 𝒚𝟐 + 𝜶𝟐 = 𝒁𝟏 + 𝒚𝟏 + 𝜶𝟏 + 𝒉𝒇 + 𝒉𝒆
𝟐𝒈 𝟐𝒈
Where hf = friction loss and he = eddy loss.
The frictional loss hf can be estimated as
𝟏
𝒉𝒇 = 𝒔𝒇 ∗ 𝚫𝐱 = (𝑺𝒇𝟏 +𝑺𝒇𝟏 )𝚫𝐱
𝟐 𝟐
𝟐
𝑸 𝒏
Where Sf = 𝟐 𝟒Τ𝟑
𝑨 𝑹 44
…Cont
There is no rational method for estimating the eddy
loss but it is usually expressed as

𝜶𝟏 𝑽𝟐𝟏 − 𝜶𝟐 𝑽𝟐𝟐
𝒉 𝒆 = 𝒄𝒆
𝟐𝒈

Where 𝒄𝒆 is a coefficient having the values as below?.

Nature of transition Values of coefficient Ce
Expansion Contractio
n
1. No transition (prismatic channel) 0.0 0.0
2. Gradual transition 0.3 0.1
3. Abrupt 0.8 0.6
45
…CONT

46
…cont
 Denoting the stage = 𝒁 + 𝒚 = 𝒉and the total
energy by H, and using the suffixes 1 and 2 to
refer the parameters to appropriate sections,
𝑽𝟐
𝑯=𝒉+ 𝜶 and the total energy at section 2 becomes
𝟐𝒈
𝑯𝟐 = 𝑯𝟏 + 𝒉𝒇 + 𝒉𝒆 ……(4.6)
 The problem can now be stated as: knowing
H1and the geometry of the channel at sections
1 and it is required to find h2.
 This is achieved in the standard-step method
by the trial and error procedure outlined
below. 47
…cont
Procedure
Select a trial value of h2 and calculating H2, hf
and he and check whether Eq.(4.6) is satisfied.
If there is a difference ,improve the assumed
value of h2 and repeat calculations till the two
sides of Eq(4.6) match to an acceptable degree
of tolerance.
On the basis of the ith trial, the (i+1)th trial
value of h2 can be found by the following
procedure suggested by Henderson.
Let HE be the difference between the left-hand
side and right-hand side of Eq(4.6) in the ith
trial,
48
 …Cont
i.e𝑯𝑬 = 𝑯𝟐 − 𝑯𝟏 + 𝒉𝒇 + 𝒉𝒆
The object is to make HE vanish by changing the depth
𝑽𝟐𝟐 𝑽𝟐𝟏
y2.Hence 𝒅𝒚𝟐 = 𝒅𝒚𝟐 ൤𝒚𝟐 + 𝒁𝟐 + 𝜶𝟐 𝟐𝒈 − 𝒚𝟏 + 𝒁𝟏 − 𝜶𝟏 𝟐𝒈 − 𝟏𝟐 ∆𝒙 𝑺𝒇𝟏 + 𝑺𝒇𝟐
𝒅𝑯𝑬 𝒅
−

49
 …cont
𝒅𝑺𝒇 𝒅 𝒏𝟐 𝒒𝟐 𝑺𝒇
For a wide rectangular channel = = 𝟑. 𝟑𝟑
𝒅𝒚 𝒅𝒚 𝒚𝟏𝟎Τ𝟑 𝒚
𝒅𝑺𝒇𝟐 𝟑.𝟑𝟑𝑺𝒇𝟐 𝟑.𝟑𝟑𝑺𝒇𝟐
Hence =− =
𝒅𝒚𝟐 𝒚𝟐 𝑹𝟐

𝐝𝐇𝐄 𝟐 𝟏.𝟔𝟕𝐒𝐟𝟐. 𝚫𝐱
leading to = 𝟏 − 𝟏 + 𝐂𝐞 𝐅𝟐 +
𝐝𝐲𝟐 𝐑𝟐
𝒅𝑯𝑬 𝜟𝑯𝑬
 If = and Δy2 is chosen such that ΔHE =HE
𝒅𝒚𝟐 ∆𝒚𝟐
−𝑯𝑬
∆𝒚𝟐 =
𝟏. 𝟔𝟕𝑺𝒇𝟐. 𝜟𝒙
𝟏− 𝟏 + 𝑪𝒆 𝑭𝟐𝟐 +
𝑹𝟐
 The negative sign denotes that ∆𝒚𝟐 is of opposite sign to that of
H E.
 it may be noted that if the calculations are performed in the
downward direction, as in super critical flow the third term in the
denominator will be negative. 50
 …cont
For the computation the following data are needed:
 Discharge Q
 Length of the reach Δx
 Area A as function of y
 Hydraulic radius R as function of y
 Roughness coefficient ( n or C)
 Carioles coefficient 
 The computation of the flow profile by the standard
step method is arranged in tabular form
Stat trial Elevatio y h or Z A H R Sf 𝑺𝒇 Length hf he Total HE ∆𝒚𝟐
-ion n of bed depth stage (m2) 𝑽𝟐 Total (m) Units Units of (m) (m) head (m) (m)
z(m) (m) (z+y) 𝟐𝒈 head of reach (m)
of
(m) (m) (m) 10-4 10-4 (m)
1(a) 1(b) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

51
…cont
 Each column of the table is explained as
follows:
1. Column 1(a) → station & column 1(b) trials
2. Column 2→elevation of bed (m)
3. Column 3→depth of flow y (m)
4. Column 4 → stage h (m) = column 2+column 3
5. Column 5→ area A(m2)
𝑽𝟐
6. Column 6→ velocity head ( ) (m)
𝟐𝒈
7. Column 7→total head H (m) = column 4+column 6
𝑨
8. Column 8→hydraulics radius R= ( )
𝑷
𝑸𝟐 𝒏𝟐
9. Column 9→water surface slop Sf =
𝑨𝟐 𝑹𝟒Τ𝟑
𝟏
ഥ
10. Column 10→average 𝑺𝒇 = (𝑺𝒇𝟏 +𝑺𝒇𝟏 )
𝟐

52
…cont
11. Column 11→ length of reach L (m)
12. Column 12 → frictional loss (m) 𝒉𝒇 = 𝒔𝒇 ∗ 𝑳
or column 10 x column 11
𝒗𝟐
13. Column 13→transiton loss he (m) 𝒉𝒆 = 𝒄𝒆 ∆
𝟐𝒈
14. Column 14 → total head (m) = column 7 of pervious section +
(column 12+column 13) of the present section
15. Column 15 → change of energy ∆𝑯𝑬 = column 7 – column 14
16.Column16 →change of water depth (m)
−∆𝑯𝑬
∆𝒚𝟐 =
𝟏. 𝟔𝟕𝑺 𝒇𝟐 ∆𝑿
𝟏 − 𝟏 + 𝒄𝒆 𝑭𝟐𝟐 +
𝑹𝟐
Note: -1. The trial continues until column 7 = column 14 (i.e. ∆𝑯𝑬 = 𝟎)
2. The next trial depth 𝒚𝟐𝟐 = the previous trial depth (𝒚𝟐𝟏 +∆𝒚𝟐 )
53
End of chapter two

54