Chapter 5

Hydraulic Processes:
Open-Channel Flow
Open-channel flow refers to that flow whose top surface is exposed to atmospheric pressure. The
topic of open-channel flow is covered in detail in textbooks such as Chow (1959), Henderson
(1966), French (1985), Townson (1991), Chaudhry (1993), Jain (2001), and Sturm (2001).
5.1 STEADY UNIFORM FLOW
Thissection describes thecontinuity, energy, andmomentum equations for steady uniform flow in
open channels. Consider the control volume shown in Figure 5.1.1 in which the channel cross-
section slope and boundary roughness are constant along the length of the control volume. For
uniformfiowthevelocity is uniform throughout thecontrol volume, sothat VI = Vl for thecontrol
volume inFigure 5.1.1. Hencefor a uniform flow, QI = Ql,A
I
=Al, VI = Vl, andYI =Yl. The
depth of flow in uniform open-channel flow is alsoreferred to as the normal depth. Figure 5.1.2
shows an open-channel flow, in an aqueduct of the Central Arizona Project.
5.1.1 Energy
Theenergy equationforopen-channel flow canbederived inasimilar manner astheenergy equation
forpipeflow (equation 4.2.13) using thecontrol volume approach. Insection 3.4,thegeneral energy
equationforsteady fluid flowwasderived asequation (3.4.20). Consideringopen-channel flow inthe
control volume in Figure 5.1.1, the energy equation can be expressed as
(5.1.1)
Assume the energy correction factor (section 3.7)is a = 1.0. Referto equation (3.6.4) for the
definition of a. The shaft work termis dWs/dt = 0 because no pump or turbine exists. Because
hydrostatic conditions prevail, the terms (p/ p+e
u
+gz) can be taken outside the integral in
equation (5.1.1):
(5.1.2)
113
114 Chapter 5 Hydraulic Processes: Open-Channel Flow
CD
--
-- ................
.....................
.....................
....................
Uniform flow
CD
--
--
--
................
--
--................
--
--
---
I-
Figure 5.1.1 Open-channel flow: uniform and nonuniform flow.

The terms JPVd4 =m are the mass rate of flow at sections 1 and2 andthe terms
J(pV
3/2)d4
= (pV
3/2)A
=m~ 2 , so that equation (5.1.2) becomes
A
dH (P2 ).. Vi (P1 ).. Vf
di= p+eUz+
gz
2 m+mT- p+eU1+
gz
1 m-mT
Dividing through by mg andrearranging yields
P1 Vf P2 Vi [euz - eUj 1 dH]
-+Zl+-=-+Z2+-+ --.---
Y 2g Y 2g g mg dt
(5.1.3)
(5.1.4)
(5.1.5)
Similar to equation (4.2.10), theterms in square brackets represent theheadloss h
L
dueto viscous
stress (friction). This energy loss due to friction effects per unit weight of fluid is denoted as h
L
.
The energy equation for one-dimensional flow in an open-channel is
P1 Vf P2 Vi
- +Zl +a1 - =- +Z2 +a2- +hL
Y 2g Y 2g
5.1 Steady Uniform Flow 115
Figure 5.1.2 Hayden-Rhodes Aqueduct, Central Arizona Project. (Courtesy of the U.S. Bureauof
Reclamation (1985), photograph by Joe Madrigal Jr.)
where we have put back in the energy correction factor (see section 3.7). Pressure is
hydrostatically distributed, and thus p/y +z is constant at each section in the control volume,
so that pdY = Yl and pl/y = Y2 The energy equation for nonuniform open-channel flow is
expressed as
(5.1.6)
For uniform flow, VI = V2 and Yl = Y2, so
(5.1.7)
By dividing both sides by L, the length ofthe control volume (channel), the following headloss per
unit length of channel, Sf' is obtained as
(5.1.8)
so that the friction slope equals the channel bottom slope. The channel bottom slope
So = tan e, where e is the angle of inclination. If e is small 100), then tan e~ sin e=
(ZI - Z2)/L.
116 Chapter 5 Hydraulic Processes: Open-Channel Flow
5.1.2 Momentum
Theforces acting uponthefluid control volume inFigure5.1.1 arefriction, gravity, andhydrostatic
pressure. Thefriction force, F
f
, is the product of thewall shearstress 'toandthe area overwhich it
acts, PL, where P is the wetted perimeter of the cross-section, thus
Ff = -'toPL
(5.1.9)
where the negative signindicates that the friction force acts opposite to the direction of flow. The
gravity force F
g
relates to the weight of the fluid yAL, where y is the specific weight of the fluid
(weight per unitvolume). Thegravity force onthefluid is thecomponent ofthe weightacting inthe
direction of flow, that is,
r, =yALsine
(5.1.10)
The hydrostatic forces are denoted as F, and F
2
, and are identical for uniform flow so that
F j -F2 = o.
For a steady uniform flow, the general formof the integral momentum equation (3.5.6) in the x
direction is
(5.1.11)
cs
or
where I>x(pV. A) = O. Because F, = F2, then by equation (5.1.12) F
g
+F
f
= 0, or
cs
yALsine - 'toPL = 0
For esmall, So ~ sin eso
yALSo ='toPL
(5.1.12)
(5.1.13)
(5.1.14)
(5.1.15)
which statesthat for steady uniform flow thefriction andgravity forces are inbalance andSo =Sf.
Solving equation (5.1.14) for the wall shear stress (for steady uniform flow) yields
yALSo
'to=--
PL
or
'to = yRSo = yRS
j
(5.1.16)
where R = AlP is the hydraulic radius. Equation (5.1.16) expresses theeffects offriction through
thewall shearstress 'toas represented from a momentum viewpoint andthrough therateof energy
dissipation Sfrepresented froman energy viewpoint. Consequently, equation (5.1.16) expresses a
linkage between the momentum and energy principles.
Theshearstress 'to forfullyturbulent flow canbeexpressed as afunction of density, velocity, and
resistance coefficient C
f
as
(5.1.17)
5.1 SteadyUniformFlow 117
Equating (5.1.16) and (5.1.17) yields
(5.1.18)
and solving for the velocity gives
v= (5.1.19)
yc;
Defining C = V2g/Cf, then equation (5.1.19) can be simplified to the well-known Chezy
equation
(5.1.20)
where C is referred to as the Chezy coefficient.
Robert Manning (1891, 1895) derived the following empirical relation for C based upon
experiments:
(5.1.21)
(5.1.22)
wheren is the Manning roughness coefficient. Values of n are listedin Table5.1.1. Values of n for
natural channels have been also published by the U.S. Geological Survey (Barnes, 1962). Sub-
stituting C fromequation (5.1.21) into equation (5.1.20) results in the Manning equation
V = R
2
/
3

n
whichis valid for SI units and So = Sf.
Table 5.1.1 Values of theRoughness Coefficient n
(Boldface figures arevalues generally recommended indesign)
Type of channel anddescription
A. Closed conduits flowing partly full
A-I. Metal
a. Brass, smooth
b. Steel
1. Lockbar andwelded
2. Riveted andspiral
c. Castiron
1. Coated
2. Uncoated
d. Wrought iron
1. Black
2. Galvanized
e. Corrugated metal
1. Subdrain
2. Storm drain
A-2. Nonmetal
a. Lucite
b. Glass
c. Cement
1. Neat, surface
2. Mortar
Minimum Normal Maximum
0.009 0.010 0.013
0.010 0.012 0.014
0.013 0.016 0.017
0.010 0.013 0.014
0.011 0.014 0.016
0.012 0.014 0.015
0.013 0.016 0.017
0.017 0.019 0.021
0.021 0.024 0.030
0.008 0.009 0.010
0.009 0.010 0.013
0.010 0.011 0.013
0.011 0.013 0.015
(Continued)
Table 5.1.1 (Continued)
Type of channel and description Minimum Normal Maximum
d. Concrete
1. Culvert, straight and free of debris 0.010 0.011 0.013
2. Culvert withbends, connections, and some debris 0.011 0.013 0.014
3. Finished 0.011 0.012 0.014
4. Sewerwithmanholes, inlet, etc., straight 0.013 0.015 0.017
5. Unfinished, steel form 0.012 0.013 0.014
6. Unfinished, smooth woodform 0.012 0.014 0.016
7. Unfinished, roughwoodform 0.015 0.017 0.020
e. Wood
1. Stave
0.010 0.012 0.014
2. Laminated, treated
0.015 0.017 0.020
f Clay
1. Common drainage title
0.011 0.013 0.017
2. Vitrified sewer
0.011 0.014 0.017
3. Vitrified sewer with manholes, inlet, etc.
0.013 0.Ql5 0.017
4. Vitrified subdrainwith openjoint
0.014 0.016 0.018
g. Brickwork
1. Glazed
0.011 0.013 0.Ql5
2. Linedwith cement mortar
0.012 0.015 0.017
h. Sanitary sewers coated with sewage slimes,
0.012 0.013 0.016
with bendsand connections
i. Pavedinvert, sewer, smoothbottom
0.016 0.019 0.020
j. Rubblemasonry, cemented
0.018 0.025 0.030
B. Linedor built-up channels
B-1. Metal
a. Smooth steel surface
1. Unpainted
0.011 0.012 0.014
2. Painted
0.012 0.013 0.017
b. Corrugated
0.021 0.025 0.030
B-2. Nonmetal
a. Cement
1. Neat, surface
0.010 0.011 0.013
2. Mortar
0.011 0.013 0.Ql5
b. Wood
1. Planed, untreated
0.010 0.012 0.014
2. Planed, creosoted
0.011 0.012 0.Ql5
3. Unplaned
0.011 0.013 0.015
4. Plank withbattens
0.012 0.015 0.018
5. Linedwithroofing paper
0.010 0.014 0.017
c. Concrete
1. Trowel finish
0.011 0.013 0.015
2. Float finish
0.013 0.015 0.016
3. Finished, with gravelon bottom
0.015 0.017 0.020
4. Unfinished
0.014 0.017 0.020
5. Gunite, goodsection
0.016 0.019 0.023
6. Gunite, wavy section
0.018 0.022 0.025
7. On goodexcavated rock
0.017 0.020
8. On irregular excavated rock
0.022 0.027
d. Concrete bottomfloat finished with sides of
1. Dressed stone in mortar
0.015 0.017 0.020
2. Random stonein mortar
0.Q17
0.020 0.024
118
Table 5.1.1 (Continued)
Type of channel anddescription Minimum Normal Maximum
3. Cement rubble masonry, plastered 0.016 0.020 0.024
4. Cement rubble masonry 0.020 0.Q25 0.030
5. Dryrubble or riprap 0.020 0.030 0.035
e. Gravel bottom with sides of
I. Formed concrete 0.017 0.020 0.Q25
2. Random stone in mortar 0.020 0.023 0.026
3. Dryrubble or riprap 0.023 0.033 0.036
f Brick
I. Glazed 0.011 0.013 0.015
2. In cement mortar 0.012 0.015 0.018
g. Masonry
I. Cemented rubble 0.017 0.025 0.030
2. Dryrubble 0.023 0.032 0.035
h. Dressed ashlar 0.013 0.015 0.017
i. Asphalt
I. Smooth 0.013 0.013
2. Rough 0.016 0.016
j. Vegetal lining 0.030 0.500
C. Excavated or dredged
a. Earth, straight and uniform
I. Clean, recently completed 0.016 0.018 0.020
2. Clean, after weathering 0.018 0.022 0.Q25
3. Gravel, uniform section, clean 0.022 0.025 0.030
4. Withshort grass, few weeds 0.022 0.027 0.033
b. Earth, winding andsluggish
I. No vegetation 0.023 0.Q25 0.030
2. Grass, someweeds
0.Q25 0.030 0.033
3. Dense weeds or aquatic plants in deepchannels 0.030 0.035 0.040
4. Earthbottom andrubble sides
0.Q28 0.030 0.035
5. Stony bottom andweedy banks 0.025 0.035 0.040
6. Cobble bottom andcleansides 0.030 0.040 0.050
c. Dragline-excavated or dredged
I. No vegetation
0.Q25 0.028 0.033
2. Light brush on banks 0.035 0.050 0.060
d. Rockcuts
I. Smooth anduniform
0.Q25 0.035 0.040
2. Jagged andirregular 0.035 0.040 0.050
c. Channels not maintained, weeds andbrush uncut
I. Dense weeds, highas flow depth 0.050 0.080 0.120
2. Cleanbottom, brush on sides 0.040 0.050 0.080
3. Same, highest stage of flow 0.045 0.070 0.110
4. Dense brush, highstage 0.080 0.100 0.140
D. Natural streams
D-l. Minor streams (topwidth at flood stage <100 ft)
a. Streams on plain
1. Clean, straight, full stage, no rifts or deeppools 0.025 0.030 0.033
2. Sameas above, but more stones and weeds 0.030 0.035 0.040
3. Clean, winding, some pools andshoals 0.033 0.040 0.045
4. Same as above, but some weeds and stones 0.035 0.045 0.050
5. Sameas above, lowerstages, more ineffective
0.040 0.048 0.055
slopes andsections
(Continued)
119
120 Chapter 5 HydraulicProcesses: Open-Channel Flow
Table 5.1.1 (Continued)
Type of channel anddescription
Minimum Normal Maximum
0.030 0.040 0.050
0.040 0.050 0.070
0.025 0.030 0.035
0.030 0.035 0.050
0.020 0.030 0.040
0.025 0.035 0.045
0.030 0.040 0.050
0.035 0.050 0.070
0.035 0.050 0.060
0.040 0.060 0.080
0.045 0.070 0.110
0.070 0.100 0.160
0.110 0.150 0.200
0.030 0.040 0.050
0.050 0.060 0.080
0.080 0.100 0.120
0.100 0.120 0.100
6. Same as 4, but more stones
7. Sluggish reaches, weedy, deep pools
8. Very weedy reaches, deeppools, or floodways
with heavy stand of timber andunderbrush
b. Mountain streams, no vegetation in channel,
banks usually steep, trees andbrush along banks
submerged at highstages
1. Bottom: gravels, cobbles, andfew boulders
2. Bottom: cobbles with large boulders
D-2. Flood plains
a. Pasture, no brush
1. Short grass
2. High grass
b. Cultivated areas
1. Nocrop
2. Mature row crops
3. Mature field crops
c. Brush
1. Scattered brush, heavy weeds
2. Light brush andtrees, in winter
3. Light brush andtrees, in summer
4. Medium to dense brush, in winter
5. Medium to dense brush, in summer
d. Trees
1. Dense willows, summer, straight
2. Cleared landwith tree stumps, no sprouts
3. Same as above, but with heavy growth of sprouts
4. Heavy stand of timber, a few down trees, little
undergrowth, flood stage below branches
5. Same as above, but with flood stage reaching branches
D-3. Major streams (topwidth at flood stage> 100ft).
Then value is less thanthat for minor streams of similar
description, because banks offerless effective resistance.
a. Regular section with no boulders or brush
b. Irregular andrough section
Source: Chow (1959).
Manning's equation in SI units can also be expressedas
Q= ! AR
2
/
3
S6/
2
n
0.045
0.050
0.075
0.Q25
0.035
0.050
0.070
0.100
0.060
0.080
0.150
0.060
0.100
(5.1.23)
For V in ft/sec and R in feet (U.S. customaryunits), equation (5.1.22) can be rewritten as
V = 1.49R
2/3S
6
/
2
(5.1.24)
n
and equation (5.1.23) can be writtenas
_ 1.49 2/3 1/2
Q--AR SO
n
(5.1.25)
whereA is in ft
2
and So = Sf. Table 5.1.2 lists the geometric function for channel elements.
5.1 Steady UniformFlow 121
Table 5.1.2
Geometric Functions for Channel Elements
Section: Rectangle
Trapezoid
Triangle
r-
B
-;1 r-B--j
I II
U
1
z L.l
f-Bw-J
f-Bw-l
AreaA
BwY (Bw+zy)y
zl
1 .
g(8-sm
Wetted B
w+2y 2Y\h +z2
1
1
8d
o
perimeter P
Hydraulic
BwY (Bw+zy)y zy
(1- sin 8)d
radius R
B
w+2y B
w+2YVl
+z2 2V1 +z2
4 8 0
Top width B
]d
o
B
w
B
w
+2zy 2zy
or
(Bw+2zy)(SB
w
+6yVl +z2)
2Jy(d
o-Y)
4(2 sin 8+38 - secas 6)
2dR 1M SB
w
+6y
\-4z/V1 +Z2
8
3d
o8(8
- sin e)sin (8/2)
3Rdy + Ady 3y(B
w+2y)
3y(B
w
+zy)(B
w
+2yVl +z2)
3y
where 8 = 2 cas - 1 ( 1 -
Source: Chow (1959) (with additions).
Todetermine thenormaldepth(forunifonn flow), equation(5.1.23)or (5.1.25)canbe solvedwith
a specified discharge. Becausethe original shear stress"toin equation(5.1.17)is for fully turbulent
flow, Manning's equation is valid only for fully turbulent flow. Henderson (1966) presented the
following criterion for fully turbulent flow in an open channel:
n
6
JRSf ? 1.9 x 10 -13 (R in feet) (S.1.26a)
n
6
JRSf ? 1.1 x 10 -13 (R in meters) (S.1.26b)
EXAMPLE 5.1.1
An8-ft widerectangular channel witha bedslopeof 0.0004 ftlft has a depthof flow of 2 ft. Assuming
steadyuniform flow, determine the discharge in the channel. The Manning roughness coefficient is
n=0.015.
SOLUTION Fromequation (5.1.25), the discharge is
Q = 1.49
n
= 1.49 (8)(2)[ (8)(2) ]2/3(00004)1/2
0.015 8 +2(2) .
= 38.5ft3/s
122 Chapter 5 Hydraulic Processes: Open-Channel Flow
EXAMPLE 5.1.2
SOLUTION
EXAMPLE 5.1.3
SOLUTION
Solve example 5.1.1 using SI units.
Thechannel width is 2.438 m, with a depthof flow of 0.610 m. Using equation (5.1.23), thedischarge is
? /3
_ 1 [ (2.438)(0.610) ]- (0.0004)1/2
- 0.015 (2.438)(0.610) 2.438 +2(0.610)
Determine thenormal depth (foruniformflow) if thechannel described inexample 5.1.1 hasaflow rateof
100 cfs.
Thisproblem is solved using Newton's method wish Q; defined by equation (5.1.25):
( )
5/3
1.49 1/2 Bwy;
Q--S
J - n 0 (B
w
+2y;) 2/3
5/3 (8 )5/3
Q. = 1.49 (0.0004)1/2 (8y;) =1.987 Y;
J 0.Q15 (8+2y;) 2/3 (8+2yi/
3
Using a numerical method such as Newton's method (see Appendix A), the normal depthis 3.98 ft.
5.1.3 Best Hydraulic Sections for Uniform Flow in Nonerodible Channels
The conveyance of a channel section increases with an increase in the hydraulic radius or with a
decrease in the wetted perimeter. Consequently, the channel section with the smallest wetted
perimeterfor a givenchannel sectionarea will havemaximumconveyance, referredtoas the best
hydraulic section or the cross-section of greatest hydraulic efficiency. Table 5.1.3 presents the
geometricelementsof the best hydraulicsectionsfor six cross-section shapes. These sectionsmay
not always be practical becauseof difficulties in construction and use of material. The concept of
Table 5.1.3 Best Hydraulic Sections
Wetted Hydraulic Hydraulic
Cross-section Area perimeter radius Top width depth
A
p
R T D
Trapezoid, half of a hexagon
v'3i
2v'3y
%Y 1V3y %Y
Rectangle, half of a square
2i
4y
Y2Y
2y
Y
Triangle, half of a square
i
2V2y
!V2y
2y
Y2Y
Semicircle
Ii
1ty
%Y
2y
~ y
Parabola, T =2V2y
1
3
/21 'l3/2
y
Y2Y
2V2y
%Y
Hydrostatic catenary 1.39586i 2.9836y 0.46784y 1.917532y 0.72795y
Source: Chow (1959).
EXAMPLE 5.1.4
SOLUTION
5.1 Steady Uniform Flow 123
besthydraulic sectionis onlyfornonerodible channels. Eventhough thebesthydraulic section gives
the minimum area for a given discharge, it maynot necessarily have the minimum excavation.
Determine the cross-section of greatest hydraulic efficiency for a trapezoidal channel if the design
discharge is 10.0 m
3/sec,
thechannel slope is 0.00052, and Manning's n = 0.025.
From Table 5.1.3, thehydraulic radius should beR =y/2, sothat thewidth B and area A are
2V3y 1
B =-3- = 1.155y (becauseB =Jpfor halfof a hexagon)
A = v'3l = 1.732l
Manning's equation (5.1.23) is used to determine thedepth:
1 1 ( ) 2/3
Q= = 0.025(1.732y2) (0.00052)1/2 = 10
so
10 x 0.025 X 2
2/3
_ 8/3
1.732(0.00052)1/2 - y
Thus, y = 2.38 m, sothat B = 2.75 mandA = 9.81 m
2
.
5.1.4 Slope-Area Method
The slope-area methodcan be used to estimate the flood discharge through a channel or river
reach of length L1x with known cross-sectional areas of flow at the upstream, A
w
and down-
stream, Ad, endsof the reach. Theuse of high-water marksfroma flood anda survey of the cross
sections allow computation of the cross-sectional areasof flow. Manning's equation (5.1.25) can
be expressed as
Q=KVSo
(5.1.27)
1.49 2/3
whereK is the conveyance factor expressed as K =--AR . Conveyance is a measure of the
n
carryingcapacityof a channel since it is directlyproportional to the discharge Q. The average
conveyance factor is the geometric meanof the conveyance factors at the upstream, K
w
and the
downstream, K
d
, ends of the channel reach, i.e.,
The discharge is then expressed as
K= VKuKd (5.1.28)
(5.1.29)
where Sis the waterslope given as S = (zu - Zd), Zu andZd are the watersurface elevations at the

upstream and downstream ends of the reach, respectively.
Alternatively the friction slope, Sf could be used in equation (5.1.29), Q=K/Sf, where
(Chow, 1959)
Sf=
(5.1.30)
(5.2.1)
124 Chapter 5 Hydraulic Processes: Open-Channel Flow
Thedifference inwater surface elevations is referred toasthefall. Thek is a factor toaccount for a
contraction and expansion of a reach. For a contracting reach V
u
< V
d
so k = 1.0 and for an
expanding reach (Vu> V
d
) sok = 0.5. Thefirst approximation would compute thedischarge using
Q=K/Sf with the friction slope computed ignoring the velocity heads. Using the first approx-
imation of Q, theupstreamanddownstreamvelocity heads arecomputedforthenextapproximation
of the friction slope, which is used to compute the second approximation of the discharge. The
procedure continues computing the new friction slope using the last discharge approximation to
compute the new discharge. This process continues until the discharges approximations do not
change significantly.
5.2 SPECIFIC ENERGY, MOMENTUM, AND SPECIFIC FORCE
5.2.1 Specific Energy
The total heador energy head, H, at anylocation in an open-channel flow can be expressed as
V
2
H=y+z+ 2g (5.1.6)
which assumes thatthevelocity distribution is uniform (i.e., a = 1) andthepressure distribution is
hydrostatic (i.e., p = yy). Using thechannel bottom as thedatum (i.e., z = 0) thendefine thetotal
headabove the channel bottom as the specific energy
V
2
E=y+-
2g
Using continuity (V = Q/A), the specific energy can be expressed in terms of the discharge as
Q2
E=Y+2gA2 (5.2.2)
Specific energy curves, such as areshown in Figure 5.2.1 and5.2.2, canbe derived using equation
(5.2.2).
y
ConstantQ
Q2
E=y+ --2
2gA
Figure5.2.1 Specific energy.
(5.2.3)
5.2 Specific Energy, Momentum, and Specific Force 125
y
Ql
E=y+-
2gA
2
Figure 5.2.2 Specific energy showing subcritical andsupercritical flow ranges.
Criticalflow occurswhenthe specific energy is minimumfor a givendischarge (i.e.,dEIdy = 0),
so that

dy gA3 dy
Referring toFigure5.2.1,thetop-width is defined asT = dAIdysoequation (5.2.3)canbeexpressed
as
(5.2.4)
or
(5.2.5)
To denote critical conditions use T
e,
An V
e,
andYe' so
(5.2.6)
(5.2.7)
or
A
e
g t:
Equation (5.2.6) or (5.2.7) can be used to determine the critical depth and/or the critical velocity.
Rearranging equation (5.2.7) yields
The hydraulic depth is defined as D = AIT so equation (5.2.7) becomes
= 1
gD
c
(5.2.8)
(5.2.9)
126 Chapter 5 Hydraulic Processes: Open-Channel Flow
or
This is basically the Froude number, F
r
, which is I at critical flow:
{
<1subcritical flow
F, = ~ = 1critical flow
gD >1supercritical flow
(5.2.10)
(5.2.11)
Figure 5.2.2 illustrates the range of subcritical flow and the range of supercritical flow
along with the locationof the critical states. Note the relationship of the specificenergy curves
and the fact that Q3 >Q2 >Q1. Figure 5.2.1 illustrates the alternate depths YI and Y2 for which
1 =2 or
V
2
V2
1 2
Y1 +-=Y2+-
2g 2g
(5.2.12)
For a rectangular channel De = Ac/T
e
= Ye, so equation (5.2.10) for critical flow becomes
(5.2.13)
If welet qbe theflow rate per unit widthof channelfor a rectangularchannel, i.e., q =Q/Bwhere
T = B, the width of the channel (or q = Q/n then equation (5.2.6) can be rearranged,
T
eQ2
/ ( g T ~ y n =q2 / (gYe) = 1, and solved for Yc to yield
EXAMPLE 5.2.1
SOLUTION
_ (q2) 1/3
Ye -
g
Compute thecritical depth for thechannel in example 5.1.1 using a discharge of 100 cfs.
Using equation (5.2.13), V
e
= ,;gy; = Q/A = 100/8Ye, so
100 (100)2/3
y ~ / 2 =8/8 or Yc = 8/8 = 1.69ft
Alternatively, using equation (5.2.14) yields
(
(100/ 8)2) 1/3
Ye = = 1.69ft
g
(5.2.14)
EXAMPLE 5.2.2
Fora rectangular channel of 20ft width, construct afamily of specific energy curves forQ= 0,50, 100,
and300cfs. Draw thelocus of thecritical depth points onthese curves. Foreach flow rate, what is the
minimum specific energy found from these curves?
SOLUTION
5.2 Specific Energy, Momentum, and SpecificForce 127
The specific energy is computed using equation (5.2.1):
Computing critical depths for the flow rates using equation (5.2.14) withq = Q/B yields
Q=O:
Q= SOds:
Q= 100ds:
Q= 300ds:
Computed specific energies are listedin Table 5.2.1.
The specific energy curves are shown in Figure 5.2.3. The minimum specific energies are:
Q=SOds:
Q= 100ds:
Q=300ds:
Emin = 0.868
Emin = 1.379
Emin = 2.868
Table 5.2.1 Computed Specific Energy Values for Example 5.2.2
Specific energy, E (ft-1b/lb)
Depth, y (ft) Q=O Q = 50 Q = 100 Q = 300
0.5 0.50 0.89 2.05 14.86
0.6 0.60 0.87 1.68 10.57
0.8 0.80 0.95 1.41 6.41
1.0 1.00 1.10 1.39 4.59
1.2 1.20 1.27 1.47 3.69
1.4 1.40 1.45 1.60 3.23
1.6 1.60 1.64 1.75 3.00
1.8 1.80 1.83 1.92 2.91
2.0 2.00 2.02 2.10 2.90
2.2 2.20 2.22 2.28 2.94
2.4 2.40 2.42 2.47 3.02
2.6 2.60 2.61 2.66 3.13
2.8 2.80 2.81 2.85 3.26
3.0 3.00 3.01 3.04 3.40
3.5 3.50 3.51 3.53 3.79
4.0 4.00 4.01 4.02 4.22
4.5 4.50 4.50 4.52 4.68
5.0 5.00 5.00 5.02 5.14
128 Chapter 5 Hydraulic Processes: Open-Channel Flow
y(ft)
7
7 E(ft)
EXAMPLE5.2.3
SOLUTION
Figure 5.2.3 Specific energy curves for example 5.2.2.
Arectangular channel 2mwide hasaflowof2.4m
3/s
at adepth of 1.0m.Determine whether critical depth
occurs at (a)a section where ahump ofilz = 20 emhighis installedacrossthe channelbed, (b) a side
wall constriction (withno humps)reducingthe channel widthto 1.7 m, and (c) both the hump and
side wall constrictions combined. Neglect headlosses of the hump and constriction caused by
friction, expansion, and contraction.
(a) The computation is focused on determining the critical elevation change in the channel bottom
(hump) L1z
cr
it that causes a critical depth at the hump. The energy equation is E = Emin +L1z
ct
it or
L1z
ct
it = E- Emin, where E is the specific energy of thechannel flow andE
rnin
is the minimum specific
energy, which isat critical depth bydefinition. If<1z
ct
it :::: L1z, thencritical depthwilloccur. Using equation
(5.2.2) yields
Q2 q2
E = y+ 2gA2 = y+ 2gy2
which canbe solved for q:
Differentiating this equation with respect to y because maximum q and minimum E are equivalent
(see Figure 5.2.4) yields
dq =!!.- [ /2g(y2E _ y3)] = 0
dy dy V
2 3
Yc = "3 Emin or Emin = :??c
Tocompute specific energy, use
q2 (2.4/2.0)2
E =Y+ -22 =1.0+ 2 =1.073m
gy 2(9.81)(1)
5.2 Specific Energy, Momentum, and SpecificForce 129
Y
2
Ye=gEmin
qconstant
E
min E
Y
Subcritical
Supercritical
Econstant
q
5.2.2 Momentum
Figure 5.2.4 Specific energy curve andY versus q for constant E.
Next compute E
min
using Emin = 3/2Ye, where Ye = (l /g) 1/3 (equation (5.2.14)):
[
2] 1/3
= (2.4/2.0) =0 28
Ye 9.81 .5 m
So Emin = 3/2(0.528 m) = 0.792 m. Then ~ Z e r i t =E- Emin = 1.073 - 0.792 = 0.281 m. In this
caseM =20 em=20/100 m =0.2 m<Merit = 0.281 m. Therefore, Ye does not occurat the hump.
(b) The critical depth at the side wall constrictionis
= [(2.4/1.7)2] 1/: 0 588
Ye 9.81 . m
Thus E
min
= (3/2)Ye = (3/2)(0.588) = 0.882 m. E is computed above as E = 1.073 m. Because
Emin = 0.882 m<E = 1.073 m, critical depthdoes notoccurat theconstriction. Remember thatenergy
losses arenegligible sothatthespecific energy intheconstriction andupstreamoftheconstriction mustbe
equal. For critical flow to occur, the constriction width can be computed as follows: Emin =
E = 1.073 m = (3/2)Ye, so that Ye = 0.715 m. Then using equation (5.2.14), 0.715 = [(2.4/B
e
)2/
]
1/3
9.81 andBe = 1.267 m.
(c) With both the hump and the side wall constriction, Yc is 0.588 m, so Emin = 0.882 m. Then
Merit =E- Emin = 1.073 - 0.882 = 0.191 m.
Because ~ z = 20/100 m = 0.20 m>M
erit
=0.191 m, critical depth will occur at the hump with a
constriction.
Applyingthemomentumprinciple(equation3.5.6) toa shorthorizontalreachof channelwithsteady
flow (Figure 5.2.5), we get
(2.5.6)
where
(5.2.15)
130 Chapter 5 Hydraulic Processes: Open-Channel Flow
Control volume
w
- - - - - - - - ~ F j -----
-----
9 sin9-S
o
Z1 rc--------dx Z2
--------------------------- --Dawm
Figure5.2.5 Application of momentum principle.
The momentum entering from the upstream is - P ~ I VIQandthe momentum leaving the control
volume is P ~ 2 V2Q, where ~ is called the momentum correction factor that accounts for the
nonuniformity of velocity (equation 3.7.8), so that
(5.2.16)
Theforces are
(5.2.17)
The hydrostatic forces are
(5.2.18)
and
(5.2.19)
where YI andY2 arethe distances to thecentroid. Thegravity force F
g
due to theweight Wof the
water is W sin e= pgAdx sin e, where W = pgAdx. Because the channel slope is small,
So ~ sin e, andthe force due to gravity is
r, = pgAdxSo (5.2.20)
where A = (AI +A2)/2 is the average cross-sectional area of flow. The external force due to
friction created byshearbetween thechannel bottom andsidesof thecontrol volume is - 'toPdx
where 'to is the bed shear stress and P is the wetted perimeter. From equation (5.1.6),
'to = yRSf = pg(A/P)Sf. So the friction force is then
For our purposes here we will continue to use F
g
and F;.
Substituting equations (5.2.15) through (5.2.21) into (3.5.6) gives
yAIYl - y
A2Y2
+Wsin e- F; = - P ~ I VIQ+P ~ 2 V2Q
(5.2.21)
(5.2.22)
which is the momentum equation for steadystate open-channel flow.
It should beemphasized thatintheenergy equation theFf(lossduetofriction) is a measure of the
internal energy dissipated in the entire mass of water in the control volume, whereas F; in the
5.2.3 Specific Force
5.2 Specific Energy, Momentum, and Specific Force 131
momentum equation measures the losses dueto external forces exerted on thewaterby the wetted
perimeter of thecontrolvolume. Ignoring thesmalldifference between theenergy coefficient a and
the momentum coefficient ~ in gradually varied flow, the internal energy losses are practically
identical withthe losses due to external forces (Chow, 1959). For uniform flow, F
g
= Fj.
Application of the energyand momentum principles in open-channel flow can be confusing at
first. It is important to understand the basic differences, even though the two principles may
produce identical or very similar results. Keep in mind that energy is a scalar quantity and
momentum is a vector quantity and that energy considers internal losses in the energyequation
and momentum considers external resistance in the momentum equation. The energy
principle is simpler and clearer than the momentum principle; however, the momentum
principle has certain advantages in application to problems involving high internal-energy
changes, suchas the hydraulic jump (Chow, 1959), which is discussedin section5.5 on rapidly
varied flow.
Forashorthorizontal reach(control volume) withe= 0andthegravity forceF
g
= Wsine= 0, the
external force of friction Fj, canbe neglected so F; = 0 andF
g
= O. Alsoassuming ~ I = ~ 2 ' the
momentum equation (5.2.22) reduces to
(5.2.23)
Substituting VI = Q/A
I
and V2 = Q/A2, dividing through by 'Y and substituting l/g = ph and
then rearranging yields
The specific force F (Figure 5.2.6) is defined as
F=Q2 +Ay
gA
(5.2.24)
(5.2.25)
Y
----T----+1
1-- -----
YlC Y
j
2
__1__ _ ~ ---.
F
Figure 5.2.6 Specific force curves.
132 Chapter 5 Hydraulic Processes: Open-Channel Flow
which has units of ft3 or m
3
. The minimum value of the specific forcewithrespectto the depthis
determined using
which results in
dF = _ TQ2 +A = 0
dy gA2
(5.2.26)
(5.2.27)
Referto Chow (1959) or Chaudhry (1993) for theproofandfurther explanation of d(Ay) / dy = A.
Equation (5.2.27) reduces to - V
2
/ g +A/T = 0 where the hydraulic depthD = A/T, so
V
2
V
2
-=D or -=1 (5.2.28)
g gD
which we have already shown is the criterion for critical flow (equation (5.2.9) or (5.2.10.
Therefore, at critical flow the specific force is a minimum for a given discharge.
Summarizing, critical flow is characterized by the following conditions:
Specific energy is minimum for a given discharge.
Specific force is minimum for a givendischarge.
Velocity headis equal to half the hydraulic depth.
Froude number is equal to unity.
Two additional conditions that are not proven here are (Chow, 1959):
The discharge is maximum for a givenspecific energy.
Thevelocity offlow inachannel of small slopewithuniformvelocity distribution isequaltothe
celerity of small gravity waves in shallow watercaused by local disturbances.
Whenflow is at or nearthecritical state, minorchanges inspecific energy nearcritical flow cause
majorchanges in depth(seeFigures 5.2.1 or 5.2.2), causing the flow to be unstable. Figure5.2.7
illustrates examples of locations of critical flow.
(a)
--
(c)
l nnft Hydrauic jump
------- Yc
I
(d)
Figure 5.2.7 Example locations ofcritical flow. (a) Critical depth near free overfal!; (b) Change ingrade
of channel bottom; (c) Flow over a broad-crested weir; (d) Flow through hydraulic jump.
EXAMPLE 5.2.4
SOLUTION
5.2 Specific Energy, Momentum, and SpecificForce 133
Compute the specific force curves for the channel andflow rates usedin example 5.2.2.
Thespecific force values arecomputed using equation (5.2.25) withthevalues presented in Table 5.2.2.
The curves are plotted in Figure 5.2.8.
Table 5.2.2 Computed Specific Force Curve Values for Example 5.2.4
Specific force, F (fr')
Depth, y (ft) Q=O Q = 50 Q = 100 Q = 300
0.1 0.10 38.92 155.38 1397.62
0.2 0.40 19.81 78.04 699.16
0.4 1.60 11.30 40.42 350.98
0.6 3.60 10.07 29.48 236.52
0.8 6.40 11.25 25.81 181.09
1.0 10.00 13.88 25.53 149.75
1.2 14.40 17.63 27.34 130.86
1.4 19.60 22.37 30.69 119.42
1.6 25.60 28.03 35.30 112.94
1.8 32.40 34.56 41.03 110.04
2.0 40.00 41.94 47.76 109.88
2.2 48.40 50.16 55.46 111.92
2.4 57.60 59.22 64.07 115.83
2.6 67.60 69.09 73.57 121.35
2.8 78.40 79.79 83.95 128.31
3.0 90.00 91.29 95.18 136.58
3.5 122.50 123.61 126.94 162.43
4.0 160.00 160.97 163.88 194.94
4.5 202.50 203.36 205.95 233.56
5.0 250.00 250.78 253.11 277.95
y(ft)
4.5
4
3.5
3
2.5
2
1.5
20 40 60 80 100 120 140 160 180 F(tt3)
Figure 5.2.8 Specific force curves.
134 Chapter 5 Hydraulic Processes: Open-Channel Flow
5.3 STEADY, GRADUALLY VARIED FLOW
5.3.1 Gradually Varied Flow Equations
Several types of open-channel flow problems can be solved in hydraulic engineering practice
using the concepts of nonuniform flow. The first to be discussed are gradually varied flow
problems in which the change in the water surface profile is small enough that it is possible to
integrate the relevant differential equation from one section to an adjacent section for the
change in depth or change in water surface elevation. Consider the energy equation (5.1.6)
previously derived for nonuniform flow (Figure 5.1.1) using the control volume approach (with
al = a2 = 1):
(5.1.6)
Because hi. = SfL= Sftu lettingAy = Y2 - YI andliz = Zl - Z2 = SoAx, thenequation (5.1.6)
can be expressed as
(5.3.1)
Rearranging yields
and then dividing through by tu results in
Ay =So _ Sf _ (Vi _Vl) 2-
tu 2g 2g tu
Taking the limit as tu ~ 0, we get
lim (AY) = dy
Llx-->O tu dx
and
lim (Vi _Vl) (2-) _ (V
2
)
Llx-->O 2g 2g tu - dx 2g
Substituting these into equation (5.3.3) andrearranging yields
(5.3.2)
(5.3.3)
(5.3.4)
(5.3.5)
(5.3.6)
dy d (V
2
)
dx +dx 2g = So - Sf
[
(
V2)]
d V
2
d 2 d
The secondtermdx (2) can be expressed as --g- 2, so that equation(5.3.6) can be
simplified to g dy dx
dy [1 +d ( ~ ; ) ] =S-S
dx dy 0 'f
(5.3.7)
5.3 Steady, Gradually Varied Flow 135
'-dA= Tdy
.J.------T-----+l,01
dy
T 1,,-----"_'I
(a) (b)
Figure5.3.1 Definition of top width (T= dAldy). (a) Natural channel; (b) Rectangular channel.
or
dy
dx
So -Sf
(5.3.8)
Equations (5.3.7) and(5.3.8) aretwoexpressions of thedifferential equationfor gradually varied
flow. Equation (5.3.8) can alsobe expressed in terms of the Froude number. First observe that
(5.3.9)
By definition, the incremental increase in cross-sectional area of flow dA, due to an incre-
mental increase in thedepthdy, is dA = Tdy, where Tis thetopwidthof flow (seeFigure5.3.1).
AlsoAIT = D, which is the hydraulic depth. Equation (5.3.9) can nowbe expressed as
(5.3.l0a)
= _F
2
r
(5.3.l0b)
where F; = ~ = ~ . Substituting equation (5.3.lOb) into(5.3.8) andsimplifying, wefind
ygD AygD
that the gradually varied flow equation in terms of the Froude number is
dy
dx
So -Sf
l-F2
r
(5.3.11)
EXAMPLE 5.3.1
Consider a vertical sluice gate in a wide rectangular channel (R =AlP = ByI(B+2Y) ~ y because
B 2y). The flow downstream of a sluice gateis basically a jet that possesses a venacontracta
(seeFigure 5.3.2). Thedistance from thesluice gatetothevenacontracta asaruleisapproximated as
thesameasthesluice gateopening (Chow, 1959). Thecoefficients ofcontraction for vertical sluice
gates areapproximately 0.6, ranging from 0.598 to0.611 (Henderson, 1966). Theobjective of this
problem is to determine the distance from the venacontracta to a point b downstream where the
depthofflow isknown tobe0.5mdeep. Thedepth offlow at thevenacontracta is0.457 mfora flow
rate of 4.646m
3/s
per meter of width. Thechannel bed slope is 0.0003 andManning's roughness
factor is n = 0.020.
O.6h
136 Chapter 5 Hydraulic Processes: Open-Channel Flow
T - ; ~ __~
h
h
L----+-t
SOLUTION
Figure 5.3.2 Flow downstream of a sluice gate in a wide rectangular channel.
To compute the distance, Axfrom Ya toYb, thegradually varied flow equation (5.3.1) can be used,
where Lly =Y2 - Yl. Solving for Ax, weget
Thefriction slope is computed using Manning's equation (5.1.22) with average values of thehydraulic
radius
_ ~ 2/3 1/2
Vave - Rave Sf
n
so
S = [nVave] 2
'f R
2
/
3
ave
Let us use the following values for thisexample:
Now we get
Location
a
b
Y(m)
0.457
0.500
R = Y (m)
0.457
0.500
V(mlS)
10.17
9.292
V
2/2g
(m)
5.27
4.40
_ 10.17+9.292 _ 9 73 rnf
Vave - 2 -. s
_ 0.457+0.500 _ 0 479
Rave - 2 -. m
S = [0.020(9.73)]2 = 0.101rnfm
'f 0.479
2
/
3
5.3 Steady, Gradually Varied Flow 137
The distance from a to b, Ax, is
_ (0.500 - 0.457) +(4.40 - 5.27) _ - 0.827 _ 821
Ax - 0.0003 _ 0.101 - _ 0.101 -. m
The distance from the sluice gateto b is ( 0 ~ ~ ~ 7 ) +8.21 == 8.97 m.
5.3.2 Water Surface Profile Classification
Channelbedslopesmaybeclassifiedasmild(M),steep(S),critical(C), horizontal (H)(So = 0), and
adverse(A) (So <0). Todefinethevarious typesof slopesfor the mild, steep, andcritical slopes,the
normal depthYn and critical depthYc are used:
Mild:
Yn>Yc
Yn 1
(S.3.12a) or ->
Yc
Steep:
Yn < or
Yn 1
(5.3.12b) -<
Yc
Critical:
Yn = Yc or
Yn = 1
(S.3.12c)
Yc
Thehorizontal andadverseslopesarespecialcasesbecausethenormaldepthdoesnot existfor them.
Table5.3.1lists the typesandcharacteristics of the varioustypesof profiles andFigure5.3.3shows
the classification of graduallyvariedflow profiles.
Table 5.3.1 Types of Flow Profiles in Prismatic Channels
Channel
Designation Relation of Y to Yn and Yc
slope
Zone I Zone 2 Zone 3 Zone I Zone 2 Zone3 General type of curve Type of flow
Horizontal None
Y > Yn >
Yc
None None
So == 0 H2
Yn >Y >Yc
Drawdown Subcritical
H3
Yn
> Yc > Y
Backwater Supercritical
Mild MI
Y > Yn >
Yc
Backwater Subcritical
O<So s, M2
Yn >Y >Yc
Drawdown Subcritical
M3
Yn > Yc >Y
Backwater Supercritical
Critical CI
Y > Yc Yn
Backwater Subcritical
So == Sc>O C2
Yn
==y
== Yc
Parallel to channel bottom Uniform-critical
C3
Yc Yn >Y
Backwater Supercritical
Steep SI
Y > Yc >
Yn
Backwater Subcritical
So s. >0 S2
Yc >Y >Yn
Drawdown Supercritical
S3
Yc > Yn > Y
Backwater Supercritical
Adverse None Y> (Yn)* >
Yc
None None
So <0 A2 (Yn)*
>Y >Yc
Drawdown Subcritical
A3 (Yn)* > Yc > Y
Backwater Supercritical
*Yn in parentheses isassumed a positive value.
Source: Chow (1959).
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
1
j
\
-- \
{In)
FIll"'" 5.3.3 1'\OW "",Jiles (fiO
m
ChOW (\
9
59)).
dyldx "'+
--_... - - - ~
- "':'
--
5.3 Steady, Gradually VariedFlow 139
The three zones for mild slopes are definedas
Zone 1:
Zone 2:
Zone 3:
Y>Yn>Ye
Yn>Y>Yc
Yn>Ye>Y
The energygradeline, water surface, and channel bottomare all parallelfor uniformflow, i.e., Sf =
So = slopeofwatersurfacewheny = Yn.FromManning'sequationforagivendischarge,Sf<SoifY>Yn'
Nowconsider the qualitative characteristics using the three zones.
Zone 1 (Ml profile):
Zone2 (M2profile):
Zone3 (M3 profile):
Y> Yn; then Sf <So or So - Sf = +
F, <1 since Y >Ye, so 1 - F; = +
dy So -Sf +
by equation (5.3.11), -d =-1--
2
=- = +
x -F, +
thenYincreases withx so that Y--+ Yn
Y <Yn; then Sf >So or So - Sf = -
F, <1 since Y >Ye, so 1 - F; = +
dy So -Sf -
by equation (5.3.11), -d = -1--
2
= - = -
x -F, +
thenY decreases withx so that Y--+ Yc
Y <Yn; then Sf >So or So - Sf = -
F, >1 since Y <Ye so 1 - F; = -
dy So -Sf -
by equation (5.3.11), -d =-1--
2
=- = +
x -F,
thenY increases withx so that Y--+ Yc
.EXAMPLE 5.3.2
SOLUTION
This analysis can be made of the other profiles. The results are summarizedin Table 5.3.1.
For the rectangular channel described in examples 5.1.1, 5.1.3, and 5.2.1, classify the type of slope.
Determine thetypesof profiles thatexistfordepths of5.0 ft, 2.0ft, and1.0ft withadischarge of 100fels.
Inexample 5.1.3, thenormal depth is computed asYn = 3.97ft, andinexample 5.2.1, thecriticaldepthis
computed as Ye = 1.69ft. Because Yn > Ye' this is a mildchannel bed slope.
For a flow depthof 5.0 ft, 5.0 >Yn >Ye, so that an Ml profile witha backwater curveexists (refer to
Table 5.3.1 and Figure 5.3.3). The flow is subcritical,
For a flow depth of 2.0ft, Yn >2.0>Ye' sothat an M2profile witha drawdown curveexists. Theflow is
subcritical.
For a flow depthof 1.0ft, Yn >Ye > 1.0, sothat anM3profile witha backwater curveexists. Theflow is
supercritical.
Prismatic Channels withChanges in Slope
Consider a channel that changes slope from a mild slope to a steep slope. The critical depth is the
same for each slope; however, the normal depth changes, for the upstreammild slope, Ynl >Ye, and
for the downstreamsteep slope,Yn2 <Ye' The only control is the critical depth at the breakin slopes
where flow transitions from a subcritical flowto a supercritical flow. The flow profiles are an M2
140 Chapter 5 Hydraulic Processes: Open-Channel Flow
Figure 5.3.4 Water surface profile for prismatic channel with slope change from mild to steep.
profile for theupstream reach andanS2profile for thedownstream reachas shown in Figure 5.3.4.
Thespecific energy forthereaches, E
nl
andE
n2
, andthespecific energy forcriticalconditions canbe
computed. Achannel thatchanges slope from asteep slope toamildslope ismorecomplicatedinthat
a hydraulic jumpforms. Thisjumpcould form ontheupper steep slope or onthe lowermildslope.
5.3.3 Direct Step Method
In example 5.3.1 we computed the location where a specified depth occurred, using the energy
equation. Thisprocedure canbeextended tocompute reachlengths for specified depths at eachend
of a reach. Computations areperformed stepbystepfrom oneendof thechannel reachby reach to
the otherend. Thisprocedure is calledthe direct stepmethod and is applicable only to prismatic
channels.
The gradually varied flow equation can be expressed in termsof the specific energy. Equation
(5.3.1) can be rearranged to
(5.2.12)
where Sf is the average friction slope for the channel reach. Solving for Llx,
(5.3.13)
(5.3.14)
EXAMPLE
Thedirectstepmethod is basedonthisequation. Manning's equation is usedtocompute thefriction
slope at the upstream and downstream ends of eachreachfor the specified depths using
n
2
y2
Sf =-2.-22-R-
2
-j 3
The average friction slope, Sf =! (Sfl +Sf
2
) , is used in equation (5.3.13). The computation
procedure must be performed f r o ~ downstream to upstream if the flow is subcritical and from
upstream to downstream if the flow is supercritical.
Atrapezoidal channel has theflowing characteristics: slope = 0.0016; bottom width = 20ft; and side
slopes = 1vertical to2horizontal (z = 2).Water atadownstream location isanembankment where the
water depth is5.0ftjustupstreamoftheembankment and the discharge is400cfs.Determine thedistance
upstream to where theflow depth is 4.60 ft. (Adapted from Chow, 1959.) (Assume a = 1.10.)
SOLUTION
5.4 Gradually Varied Flowfor Natural Channels 141
Thecritical depth isYc = 2.22 ft andthenormal depth isYn = 3.36 ft, sotheflow issubcritical. Because the
depth of5.0ft isgreater than thenormal depth of3.36 ft, thisisanMl profile. Then thedepth will decrease
proceeding intheupstream direction because dy/ dx = + in thedownstream direction. Inother words,
going upstream thedepth will approach normal depth. Sowewill consider thefirst reach, going upstream
to a depth of 5.0to 4.8ft andthe second reach upstream is from a depth of 4.8to 4.6ft depth.
First reach; AtY = 5.0ft then A = 150 fr', R = 3.54 ft, V = 2.67 ftIsec, Sf = 0.000037, andE = 5 +
(I.I) (2.67)z/[2(32.2)] = 5.123 ft, andatY = 4.80ft then A = 142.1 ftz, R = 3.43 ft, V = 2.82 ft/sec,
Sf = 0.00043, andE = 4.936 ft. Using equation 5.3.14,
- 1
where Sf =2: (0.00037+0.00043) =0.00040,
Ez-E1 (5.123-4.938)
then~ x = So _ Sf 0.0016_ 0.00040 = 156ft.
Second reach: Aty =4.60ft,A = 134.3 ftZ,R = 3.3lft, V = 2.98 ftIsec,Sf= 0.00051,andE = 4.752ft.
- 1
Sf =2: (0.00043 +0.00051) = 0.00047, and~ x = 163 ft.
Sothedistance upstream to a depth of 4.60 ft is 156 + 163 = 319ft. Theprocedure canbe continued
upstream to where thenormal depth occurs.
5.4 GRADUALLY VARIED FLOW FOR NATURAL CHANNELS
5.4.1 Development of Equations
As an alternate to the procedure presented above, the gradually varied flow equation can be
expressed in termsof thewatersurfaceelevation for application to naturalchannelsbyconsidering
w = z +y wherewis the water surfaceelevation abovea datumsuchas meansea level. The total
energy H at a sectionis
V
2
V
2
H= z+y+a- = w+a- (5.4.1)
2g 2g
including the energycorrection factor a. The changein total energyhead withrespect to location
along a channel is
The total energyloss is due to friction losses (Sf) and contraction-expansion losses (Se):
dH
-= -Sf-Se
dx
(5.4.2)
(5.4.3)
S, is the slope termfor the contraction-expansion loss. Substituting (5.4.3) into (5.4.2) results in
(5.4.4)
The frictionslope Sfcan be expressed using Manning's equation (5.1.23) or (5.1.25):
whereK is defined as the conveyance in SI units
K = !AR
2
/
3
n
(5.4.5)
(5.4.6a)
142 Chapter 5 Hydraulic Processes: Open-Channel Flow
or in U.S. customary units as
(5.4.6b)
(5.4.8)
for equations (5.1.23) or (5.1.25), respectively. The friction slope (from equation 5.4.5) is then
Q2 Q2[1 1]
Sf =K2 =T Kr +Ki (5.4.7)
withtheconveyance effect ~ =~ [ ~ +~ ] ,where K, andK
2
aretheconveyances, respectively,
K 2 K
1
K
2
at the upstream and the downstream ends of the reach. Alternatively, the friction slope can be
determined using an average conveyance, i.e., X = (Kl +K2) /2andQ=XS}/2; then
Q2
Sf=-
X
2
The contraction-expansion loss term S; can be expressed for a contraction loss as
(
V2) (V
2
V
2)
for d a- = a 2 ~ -a1.-l >0
2g 2g 2g
(5.4.9)
and for an expansion loss as
(5.4.10)
Thegradually variedflow equationfor anatural channel isdefined bysubstituting equation (5.4.7)
intoequation (5.4.4):
or
_ Q2 [ ~ +~ ] _s, = dw+!!..- (aV
2)
2 Kr Ki dx dx 2g
(5.4.11a)
(5.4.11b)
Rearranging yields
(5.4.12)
EXAMPLE 5.4.1
SOLUTION
Derive anexpression forthechange inwater depth asa function of distance along a prismatic channel
(i.e., constant alignment and slope) for a gradually varied flow.
We start with equation (5.3.11):
5.4 Gradually Varied Flowfor Natural Channels 143
with
Then
dy
dx
So -Sf
(
1- Q2T)
gA3
EXAMPLE 5.4.2
SOLUTION
Todetermine a gradually varied flow profile, this equation is integrated.
Forariversection witha subcritical discharge of6500ft3/S, thewatersurface elevation at thedownstream
section is5710.5 ft withavelocity headon.72 ft.Thenextsection is500ft upstream withavelocity head
of 1.95 ft. The conveyances for the downstream and upstream sections are 76,140 and 104,300,
respectively. Using expansion and contraction coefficients of 0.3 and 0.1, respectively, determine the
water surface elevation at the upstream section.
Using equation (5.4.12), the objective is to solve for the upstream water surface elevation WI:
Thefriction slope termis
1 + 1 ]500=2.79ft
2 Ki +KI 2 76,140
2
104,300
2
Because /2g)] >0, a contraction exists for flow fromcross-section 1 to cross-section 2. Then
The water surface elevation at the upstream cross-section is then
WI = 5710.5+1.77+2.79+0.177 = 5715.2ft
5.4.2 Energy Correction Factor
In section 3.7, the formula for the kinetic energy correction factor is derived as
(3.7.4)
144 Chapter 5 Hydraulic Processes: Open-Channel Flow
C
(channel)
Figure 5.4.1 Compound channel section.
which can be approximated as
(5.4.13)
where V is the mean velocity.
Consider a compound channel section as shown inFigure 5.4.1 that hasthree flow sections. The
objective is to derive an expression for the energy coefficient in terms of the conveyance for a
compound channel, sothatthevelocity head fortheentire channel isu(V
2
/2g) where Vis themean
velocity in the compound channel. Equation (5.4.13) can be expressed as
N
EV/A;
;=1
U;:::j--N-
V
3
E A;
;=1
(5.4.14)
where Nis thenumber of sections (subareas) of thechannel (e.g., inFigure 5.4.1, N = 3), Vis the
mean velocity in eachsection (subarea), andAi is the cross-sectional area of flow in each section
(subarea).
The mean velocity can be expressed as
N
EViAi
V = ~
N
EAi
i=l
(5.4.15)
Substituting Vi = Q;/Ai andequation (5.4.15) for V into(5.4.14) and simplifying yields
(5.4.16)
N(Q )3
E -.!. Ai
i=l Ai
[
N 1
3
'" Q;
Z:: iii A; N
; ~ 1 EA-
t.', C,)
U=----...,,-----
5.4 Gradually Varied Flowfor Natural Channels 145
Nowusingequation (5.4.5) for each section, we get
1/2
Qi =s;
and solving for S)./2 yields
(5.4.17)
(5.4.18)
Assuming that thefriction slopeis thesamefor all sections, Sf; =SfU= 1, ... , N), thenaccording
to equation (5.4.18)
(5.4.19)
This leads to
(5.4.20)
and the total discharge is
(5.4.21)
Substituting the aboveexpression for LQi and
into equation (5.4.16) and simplifying, we get
N(K3)(N )2
~ At ~ A i
a = (tK;)3
1=1
or
whereAt and K, are the totals.
The friction slopefor the reach is
(5.4.22)
(5.4.23)
(5.4.24)
S = (2:Qi)2
'f, 2:K
i
(5.4.25)
EXAMPLE 5.4.3
by eliminating QN/ K
N
fromequations (5.4.19) through (5.4.21).
Forthecompound cross-section at river mile 1.0shown inFigure 5.4.2, determine theenergy correction
factor a. Thedischarge is Q = 11,000 cfs andthewater surface elevation is 125 ft.
146 Chapter 5 Hydraulic Processes: Open-Channel Flow
! " " " - - - - - - - ~ - - - - - - -
50' I 115' 60'
.1 150'
n=.0451 n =.025 In=.045
70'
n=.04
Cross-section at river mile 1.0
Q=11,000cfs
WS=125ft
C,=.3
C
c
=.1
- - - - - - - ~ - - - - - - - 125'
......- ... 110'
100'
Elev.---l
70'
n=.04
AM 1.0
L= 2640ft
AM 1.5
aV
2
Vhf
---
--
2C
u
-f-a
2
-
..
L
r----1115'
105'
Elev.---l
I
Cross-section at river mile 1.5
Q=10,500 cfs
WS=? ft
C,=.3
C
c
=.1
Figure 5.4.2 Cross-section andreach length datafor example 5.4.3 (from Hoggan (1997.
SOLUTION Step1 Compute thecross-sectional areas offlow fortheleft overbank (L), channel (C), andrightover-
bank (R):
A
L
= 1050ft
2,A
c
= 3000fr,A
R
= 1050fr
Step2 Compute the hydraulic radius for L, C, andR:
R
L
= 1050/85 = 12.35 ft; R
c
= 3000/140 = 21.40 ft; RR = 1050/85 = 12.35 ft
Step3 Compute the conveyance factor for L, C, andR:
2/3 ()( )2/3
K = 1.486A
LR
L = 1.486 1050 12.35 = 208 700
L n 0.04 '
K = 1.486(3000)(21.4)2/3 = 1 716300
c 0.02 "
K
R
= 208,700(K
R
= Kd
Step4 Compute totals At andK;
K, = K
L
+K
c
+K
R
= 2,133,700
At = A
L
+A
c
+A
R
= 5100 ft
2
Step5 Compute K
3/A2
and LK3/A2:
KiiAI = 8.25 x 10
9
K ~ / A ~ =561.8 x 10
9
K U A ~ = 8.25 x 10
9
LK
3
/A
2
= 578.3 X 10
9
Step6 Useequation (5.4.24) to compute a:
a = E(%) (At? = (578.3 x 10
9)(5100)2
= 1.55
(K
t)3
(2,133,700)3
ExAMPLE5.4.4 ForthedatainFigure 5.4.2, startwith theknown water surface elevation atrivermile1.0anddetermine the
water surface at rivermile 1.5 (adapted from Hoggan, 1997).
(5.4.2)
5.4 Gradually Varied Flow for Natural Channels 147
SOLUTION Computations are presented in Table 5.4.1.
Table 5.4.1 Standard Step Backwater Computation
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)
Water Average Energy
River surface Hydraulic Manning Average friction Friction correction
mile elevation Area radius roughness Conveyance conveyance slope loss factor Velocity v
2
Ll(*)
W
k
A R n K
K
Sf
h
L
K
3/A2
a V
a"2g
h ** Wk*** 0
(ft
2
) (ft) (1) (10
9
) (ftlsec) (ftlsec) (ftlsec) (ft) (ft)
1.0 125.0 1050 12.35 0.040 208,700 8.25
3000 21.43 0.020 1,716,300 561.80
1050 12.35 0.040 208.700 8.25
5100 2,133,700 578.30 1.55 2.16 0.11
1.5 126.1 666.0 9.37 0.045 97,650 2.10
2426.5 17.97 0.025 989,400 164.50
666.0 9.37 0.045 97,650 2.10
3758.5
1,184,700 1,659,200 0.000042 0.111 168.70 1.43 2.79 0.17 -0.060.02 125.07
1.5 125.0 600 8.57 0.045 83,000 1.59
2300 17.04 0.025 905,300 140.22
600 8.57 0.045 83,000 1.59
3500
1,071,300 1,602,350 0.000043 0.113 143.30 1.43 3.00 0.20 -0.09 0.03 125.05
(A )z" K
3/A2
*a= t w J I
(K
t
)3
"h
o
= /2g) I for Ll(aV
2
j2g) <0 (loss due to channel expansion); h
o
= /2g) I for /2g) >0 (loss due to channel
contraction).
'''Wz = WI +hdh
o
= 125.0+( -0.06)+0.111+0.02 = 125.066 125.07.
Source: Hoggan (1997).
5.4.3 Application for Water Surface Profile
The change in head withrespect to distance x along the channel has beenexpressed in equation
(5.4.2) as
dH dw d ( V
2
)
dx = dx +dx a 2g
Thetotalenergy losstermis dHj dx = - Sf - Se, where Sfis thefriction slope defined byequation
(5.4.7) or equation (5.4.8) andS, is theslope of thecontraction or expansion loss. Thedifferentials
dwandd[a(V
2
j2g)] aredefined overthechannel reachas dw= Wk - Wk+ 1 andd[a(V
2
j2g)] =
V
2
V
2
ak+1 1 - ak 2;' where wedefine k + 1at thedownstreamandk at theupstream. River cross-
sections are normally defined from downstream to upstream for gradually varied flow. Equation
(5.4.2) is now expressed as
Vf Vf+l
Wk +ak 2g = Wk+l +ak+l +Sfdx +Se
dx
(5.4.26)
The standard step procedure for water surface computations is described in the following
steps:
a. Startat apointinthechannel where thewatersurface isknown orcanbeapproximated. Thisis
thedownstream boundary condition forsubcritical flow andtheupstreamboundary condition
148 Chapter 5 Hydraulic Processes: Open-Channel Flow
forsupercritical flow. Computation proceeds upstreamforsubcritical flow anddownstreamfor
supercritical flow. Why?
b. Choose a water surface elevation Wk at theupstream endof thereach for subcritical flow or
Wk +1 at thedownstream endof thereach for supercritical flow. Thiswatersurface elevation
will be slightly lower or higher depending upon the type of profile (see Chow (1959);
Henderson (1966); French (1985); or Chaudhry (1993)).
c. Next compute the conveyance, corresponding friction slope, andexpansion and contraction
loss terms in equation (5.4.26) using the assumed water surface elevation.
d. Solve equation (5.4.26) for wk+ 1 (supercritical flow) or Wk (subcritical flow).
e. Compare thecalculated water surface elevation Wwiththeassumed water surface elevation w'.
If thecalculated andassumed elevations donotagree within anacceptable tolerance (e.g.,0.01
ft), thensetwk+l =Wk+l (forsupercritical flow) andWk =Wk (forsubcriticalflow)andretum
to step(c).
Computer models for determining water surface profiles using the standard step procedure
include the HEC-2 model and the newer HEC-RAS model. HEC-RAS River Analysis System
(developed by the U.S. Army Corps of Engineers (USACE) Hydrologic Engineering Center)
computes water surface profiles for one-dimensional steady, gradually varied flow in rivers of
any cross-section (HEC, 1997a-c). HEC-RAS can simulate flow through a single channel, a
dendritic systemof channels, or a full network of openchannels (sometimes calleda fullylooped
system).
100 0 100 200
I"II! I I
Scale (ft)
Figure5.4.3 Map oftheRed Fox River indicating cross-sections forwater surface profile analysis (from
U.S. Bureau of Reclamation (1957)).
5.4 Gradually Varied Flowfor Natural Channels 149
HEC-RAS can model sub- or supercritical flow, or a mixture of each within the same analysis.
A graphical user interface providesinput data entry,data modifications, and plots of streamcross-
sections, profiles, and other data. Program options include inserting trapezoidal excavations on
cross-sections, and analyzing the potential for bridge scour. The water surface profile through
structuressuchasbridges,culverts,weirs,andgatescanbecomputed.TheWorldWideWebaddress
to obtain the HEC-RAS model is www.hec.usace.army.mil.
, EXAMPLE 5.4.5 Aplanview of theRedFoxRiver in California is shown in Figure 5.4.3, along with thelocation of four
cross-sections. Perform the standard stepcalculations to determine thewater surface elevation at cross-
section 3 for a discharge of 6500ft
3/s.
Figures 5.4.4a, b, andc are plotsof cross-sections at 1, 2, and3,
respectively. Figures 5.4.5a, b, andcaretheareaandhydraulic radius curves forcross-sections 1,2, and3,
respectively. Useexpansion andcontraction coefficients of 0.3and0.1, respectively. Manning's rough-
ness factors are presented in Figure 5.4.4. The downstream starting water surface elevation at cross-
section 1is 5710.5 ft above mean sealevel. Thisexample wasoriginally adapted bytheU.S. Army Corps
ofEngineers frommaterial developed bytheU.S. BureauofReclamation (1957). Distance between cross-
sections 1and2 is 500ft, between cross-sections 2 and3is 400ft, andbetween cross-sections 3 and4 is
400ft.
SOLUTION The computations for this example are illustrated in Table 5.4.2.
1700 1500
600(690, 0) 800 1000
Station (It)
Cross-section No. 1
(a)
400 200
(20,25)
f I I I
(1635,25)
1:\
A. A
3
A A,
AI
,
'"
n 0.10
n f 0.05
n 0.05 n 0.10
-f+
1--1=0.03
(110,18)
(415,17)
(650,14)
(1020,14)
(1590,14)
(710,13)
(675,1)
\
(7111)
5710
5700
o
5705
6 5715

m
5725
5720
5730
5725
g 5720
c
.2 5715

ijj 5710
5705
I
A, A, A A
3
v,L"J,-
I
n =0.10
n =0.05 I n =0.10

20)
I-
J
n =0.03
(110,20)

(1195,18) (640,18)
r\.
(575,9.5)
I
(58,,4) (61
1,4)
o 200 400 600 800 1000
Station (It)
Cross-section No.2
(b)
1200 1400
Figure 5.4.4 Cross-sections of the Red FoxRiver (fromU.S. Bureau of Reclamation (1957)).
ISO Chapter 5 Hydraulic Processes: Open-Channel Flow
5710
1000 400 600 800
Station (tt)
Cross-section No.3
(c)
200
I
A
3
A,
1 A
2
1~ = 0 . 1 1
" =0.03
,,=o.OS
(40.25)
I
(875,25)-
"""
1--,,=0.10
WI I
_I...-
(850,22
(260,22) .....
-
l370,18.7)\
f< (600.20)
I 1\
(560,17:3)
(420.15)1 \
J \
I) (s+.7.5)
(500,7.1)
5715
5720
5705
0
~ 5725
~
~
~
w
5730
1000 400 600 800
Station (tt)
Cross-section No.4
(Ii)
200
I AJ
A
3
1 A, A
2
TJ
" .. 0.10
I,,-
"i
o
.
os ,,=
"i
o
.
os
I
-..0.
10 0.036
(700.26)
..,..,
1 ~ 0 , 2 2 )
30,26) (130,24)
7(460.22
330,23) I--
""r
i
1/
(400'1
10
)
5715
5730
5705
0
5710
~ 5725
~
,g 5720
~
W
Figure 5.4.4 (Continued)
572118 16
5720
~ 5716
.9
1ii
~
m
5710
o
~ ~ ~ 400 ~ ~ 1 ~ 1 ~ 1400 1 ~ 1 ~
Area(tt2)
(a)
Figure 5.4.5 Areaelevation andhydraulic radius-elevation curves forcross-sections I to4. (a) Cross-
section 1;(b) Cross-section 2; (c) Cross-section 3; (d) Cross-section 4 (from U.S. Bureau of Reclamation
(1957)).
o 16
57211r-8
5720
5710
5708
0 200 400 600 800 1000 1200 1400 1600 1800
Area (tt2)
(b)
A
2
14 12 10 8 4
572118 16
5720
A
1
5718
g
5716
c
0
il
>
5714
CD
iii
5712
5710
5708
0 200 400 600 800 1000 1200 1400 1600 1800
Area (fI2)
(c)
5721
18 16 14 4 2 0
5720
5718
g
5716
c
0
..

5714
iii
400 1400
Area (tt2)
(d)
Figure 5.4.5 (Continued)
151
152 Chapter 5 Hydraulic Processes: Open-Channel Flow
Table 5.4.2
Standard Step Backwater Computation for Red Fox River
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)
Water Average Energy
Cross- surface Hydraulic Manning Average friction Friction correction
section elevation Area radius roughness Conveyance conveyance slope loss
K
J factor
Velocity E ( E)
Wk + !
A R n K K S/1O-
3
) h
L
A2
a
V u2g U 2g
h
o
W
k
(ftl)
(ft)
(ft/ft) (ft) (10
6
) (ft/sec) (ft) (ft) (ft) (ft)
1 5710.5* 420 7.0 0.03 76,100
1.0 15.5 3.72
2 5714.7 470 7.6 0.03 90,100 3311.1
260 2.5 0.05 14,200 42.0
730 104,300 90,200 5.19 2.60 3353.5 1.58 8.90 1.95 + 1.77 0.18 5715.07**
5715.0
500 7.85 0.03 97,800 3741.8
300 2.7 0.05 17,300 57.5
800 115,100 95,600 4.62 2.31 3799.3 1.59 8.13 1.63 +2.09 0.21 5715.1
3 5718.0 1145 5.85 0.03 184,100 149,600 1.89 0.76 1.0 5.68 0.50 + 1.13 0.115717.1
5717.1 970 5.6 0.03 151,500 133,300 2.38 0.95 1.0 6.70 0.70 +.93 0.095717.1
*Known starting watersurface elevation.
**Wk+1 = 5710.7+1.77+2.60 +0.18= 5715.07 = 5715.1; a. = (A,/ ,Kl!Af /(K,)3; h, = eel (a.V
2/2g)
Ifor <0 (loss due to
channel expansion); h
o
= /2g) I for /2g) >0 (lossdue to channel contraction); Wk+ 1 =Wk + /2g) + h
L
+ h.,
Source: Hoggan (1997).
5.5 RAPIDLY VARIED FLOW
Rapidly variedfiowoccurs when awater flow depth changes abruptly overaveryshortdistance. The
following are characteristic features of rapidly varied flow (Chow, 1959):
Curvature of the flow is pronounced, so that pressure distribution cannot be assumed to be
hydrostatic.
The rapid variation occurs over a relatively short distance so that boundary friction is
comparatively small and usually insignificant.
Rapid changes of water area occur in rapidly varied flow, causing the velocity distribution
coefficients a andpto be much greater than 1.0.
Examples of rapidly varied flow are hydraulic jumps, transitions in channels, flow over
spillways, flow in channels of nonlinear alignment, and flow through nonprismatic channel
sections such as flow in channel junctions, flow through trash racks, and flow between bridge
piers.
The discussion presented in this chapter is limited to the hydraulic jump. The hydraulic
jump occurs when a rapid change in flow depth occurs from a small depth to a large depth
such that there is an abrupt rise in water surface. A hydraulic jump occurs wherever super-
critical flow changes to subcritical flow. Hydraulic jumps can occur in canals downstream
of regulating sluices, at the foot of spillways, or where a steepchannel slopesuddenly becomes
flat.
Figure 5.5.1 illustrates ahydraulic jumpalong with thespecific energy andspecific force curves.
Thedepths of flow upstream anddownstream of thejumparecalled sequent depths or conjugate
depths. Because hydraulic jumps are typically short in length, the losses due to shear along the
wetted perimeter aresmall compared tothepressure forces. Neglecting these forces andassuming a
horizontal channel (F
g
= 0), themomentum principle canbe applied as in section 5.2.3 to derive
y
Figure 5.5.1 Hydraulic jump.
equation (5.2.24):
Hydraulic jump
y
5.5 Rapidly Varied Flow 153
Specific force curve
Q2 _ Q2 _
- +AlYl = - +A2Y2 (5.2.24)
gAl gA2
Consider a rectangular channel of width B > 0, so Q =Al VI =A2V2,A
l
= BYl,A2 = BY2,
)11 = yI!2, and )12 = y2/2:
(5.5.1)
Simplifying yields
Q2 ( ~ _ ~ ) = ! B 2 ( y ~ -yi)
g Yl Y2 2
Q2 ( ) 1 2 (2 2)
g Y2 -Yl ="i
B
Y1Y2 Y2 -Yl
B
2y2V2
1
~ (Y2 - Yl) ="iB2YIY2(Y2 +Yl)(Y2 - vr)
YlVf 1
-g- = "iY2(Y2 +Yl)
Dividing byYi, we get
2Vf =Y2 (Y2 +1)
gYl Yl Yl
(5.5.2)
The Froude number for a rectangular channel is F
r 1
= VI! JgDl = VI! Viii; therefore
equation (5.5.2) reducesto
or
(5.5.3)
(5.5.4)
using the quadratic formula and discarding the negative roots.
154 Chapter 5 Hydraulic Processes: Open-Channel Flow
Alternatively, Q= BY2 V2 and F
rz
= V2//iYi. could have been used to derive
-1+V
1+8F
;z) (5.5.5)
Equations (5.5.4)and (5.5.5) can be usedto findthe sequent depths of a hydraulicjump. The use of
hydraulic jumps as energy dissipaters is further discussed in Chapters 15 and 17.
EXAMPLE5.5.1
SOLUTION
EXAMPLE5.5.2
SOLUTION
EXAMPLE5.5.3
SOLUTION
Consider the8-ft widerectangular channel usedinexamples 5.1.1,5.1.3,and5.3.1 withadischarge of 100
cfs.If a weirwereplacedin thechannel andthedepth upstream of theweirwere5 ft, would a hydraulic
jump formupstream of the weir?
Forthedischarge of 100cfs,thenormal depthisYn = 3.97ft from example 5.1.3,andthecriticaldepthis
Ye = 1.69. Because Ye <Yn <5 ft, a hydraulic jump would not form. As a resultof Yn >Ye, a mildslope
exists. For a jump to form, Yn <Ye, whichis a steepslope.
For example 5.3.1, determine whether a hydraulic jump will occur.
Thenormal depthandcriticaldepthmustbecomputed andcompared. Usingequation (5.2.14) withq =
4.646 m
3/s
per meter of width, we get
_ (q2) 1/3 _ (4.646
2
) 1/3 _
Ye - g - 9.81 -1.30m.
The depths of flow at Ya = 0.457m and at Yb = 0.5 m, so this flow is supercritical flow.
Next, the normal depthis computed using Manning's equation (5.1.23):
Q = ! AR
2f3
56
/ 2
n
or
4.646 =
ThusYn = (5.365)3
/
5 = 2.74 m. Under theseconditions Yn > Ye > 0.5 m, an M3 water surfaceprofile
existsand a hydraulic jump occurs. If normal depthoccurs downstream of thejump, what is the depth
before thejump? Usingequation (5.5.5) withY2 = 2.74m, we find that Yl is 0.5 m.
A rectangular channel is 10.0 ft wide and carries a flow of 400 cfs at a normal depth of 3.00.
Manning's n = 0.017. An obstruction causesthe depthjust upstream of the obstruction to be 8.00ft
deep. Will a jump formupstream fromthe obstruction? If so, howfar upstream? What type of curve
will be present?
Firstadetermination mustbemadewhether ajumpwillformbycomparing thenormal depthandcritical
depth. Usingequation (5.2.14), we find
(
q2) 1/3 ( 40
2
) 1/3
Ye = g = 32.2 = 3.68 ft
5.5 Rapidly VariedFlow 155
andYc>Yn = 3.0ft, therefore thechannel issteep. BecauseYn <Yc<8ft, asubcritical flowexistsonasteep
channel, a hydraulic jumpforms upstreamof theobstruction. If thedepthbefore thejumpis considered to
benormal depth.y, = Yl = 3, thentheconjugate depthy, canbecomputed using equation (5.5.4), where
the Froude number is
so
Yl (
Y2 = 2 -1+
~ ~ ( - 1 +
= 4.45 ft
1+
8(40)2)
g(3)3
Nextthe distance Ax fromthe depth of 4.45ft to the depth of 8 ft is determined using equation (5.3.1):
y2 y2
SoAx + 2 ~ =i1y + 2 ~ + Sfi1x
which can be rearranged to yield
y2 y2
(So -Sf)i1x =Y3 +-1. -Y2 _---.l
2g 2g
= E3-E2
so
E3-E2 M
Ax=--=--
So -Sf So -Sf
Tosolvefor i1x, first compute E
2
andE
3
:
Depth
(ft)
8
4.45
80
44.5
R
(ft)
3.08
2.35
Y
(ft/s)
5.00
8.99
0.388
1.25
E
(ft)
8.388
5.70
Rave
(ft)
2.72
Yave
(ft/s)
7.00
Now compute SffromManning's equation using equation (5.1.24) with Yave andRave, andrearrange to
yield
S = n
2
Y;ve = 0.017
2
x 7
2
= 0.00168
'f 2 . 2 2 R ~ ~ ~ 2.22 x 2.72
4
/
3
Compute So using Manning's equation withthe normal depth:
1.49 (30)2/3
400 = 0.017 x 3 x 10 x 16 jS;
156 Chapter 5 Hydraulic Processes: Open-Channel Flow
M 8.388 - 5.71 322 f h di f h
Thus So = 00100 Now usingfix =--= = t, t e istance rom t e
, .., So - Sf 0.0100 - 0.00168
conjugatedepthof thejumpYz = 4.45ft downstream to thedepthY3 of 8 ft (locationof the obstruction) is
322ft. In otherwords,the hydraulic jump occurs approximately 322ft upstreamof the obstruction. The
type of water surface profile after the jump is an S1 profile.
EXAMPLE 5.5.4
SOLUTION
Ahydraulic jump occursin a rectangular channel 3.0 mwide. The waterdepthbeforethejump is 0.6 m,
andafterthejumpis 1.6 m.Compute(a)theflow rateinthechannel,(b) thecritical depth, and(c) thehead
loss in the jump.
(a) To compute the flowrate knowing Yl = 0.6 m and yz = 1.6m, equation (5.5.4) can be used:
Yl [
Y2 =2 -1+
8
q
2 ]
1+-
gyj
EXAMPLE 5.5.5
in which F
r 1
= q/ ;;;i has been substituted:
0.6[ 81 ]
1.6 = T - 1 + 1 + 3
9.81(0.6)
6.33 = VI +3.775q2
40.07 = 3.7751
q = 3.26 m
3/s
per meter widthof channel
and
Q = 3q = 9.78 m
3/s.
(b) Critical depth is computedusing equation(5.2.14):
_ (q2) 1/3 _ (3.26
2
) 1/3 _
Yc - g - 9.81 - 1.03 m
(c) The headloss in the jump is the change in specific energy before and after the jump
hi. = M = 1-2
so
V
2
V
2
h: = YI + ---l - Y2 - ~
2g 2g
so
VI = Q/A
I
= 9.78/(3 x 0.6) = 5.43 mls
V2 = Q/A
2
= 9.78/(3 x 1.6) = 2.04rn1s
h = 0 6 5.43
2
_ _ 2.04
2
L . +2(9.81) 1.6 2(9.81)
= 0.6 +1.5- 1.6 - 0.21
h: = 0.29m.
Derive an equation to approximate the headloss (energy loss) of a hydraulic jump in a horizontal
rectangular channel in terms of the depths before and after the jump, Y, and Yz, respectively.
SOLUTION
5.5 Rapidly Varied Flow 157
Theenergy loss canbe approximated by
ht: = 1 -2
= (Yl +~ : ) - (Y2+~ ! )
Thevelocities canbe expressed as VI =Q/A
1
=q/Yl andV2 = q/Y2, so
1 q2 1 q2
b: =Yl +-- - Y2 - --
2gyi 2 g y ~
Thebalance between hydrostatic forces andthe momentum flux per unit width of thechannel canbe
expressed using equation (5.2.23):
where A = (1)(yl) andy, = Yl /2, so
Ii- Ii = pq(!L _!i)
2 1 2 2 Y2 Yl
Solving, weget
l (Yl +Y2)
-=Y1Y2 --
g 2
Substituting this equation intotheabove equation for theheadloss andsimplifying gives
Prismatic Channels with Change in Slope
Consider the channel that changes slopefroma steepslopeto a mildslopeshown in Figure5.5.2.
InthiscaseYnz >Yn! . Theconjugate depthofYn! isy' computed usingequation (5.5.4). Ify' <Ynz. the
hydraulic jumpoccurs upstreamof theslopebreakwithcontrolat A, andif y' >Ynz. thejumpoccurs
downstream of the breakin slopewith control at B. If (En! - till
i
!) >E
nz
then a hydraulic jump
occursonthe downstream mildslope. If E
nz
> (En! - till
iI
) . thenthehydraulic jumpoccursonthe
upstream steepslope. till
iI
is theenergy lossin thejumpinreach1. Another waytolookat thisis if
Ey' >E
nz
controlis at Bbecause the energy loss till
iI
in the hydraulic jump in reach 1is not large
enoughto decrease the energy fromEn! to E
nz.
Mildslope
Yn,
Figure 5.5.2 Rapidly varied flow caused by slope change from steep to mild.
158 Chapter 5 Hydraulic Processes: Open-Channel Flow
5.6 DISCHARGE MEASUREMENT
5.6.1 Weir
A weir is a device (or overflow structure) that is placed normal to the direction of flow. Theweir
essentially backs up water sothat in flowing overthe weir, the water goes through critical depth.
Weirs have been usedforthemeasurement ofwater flow inopenchannels for many years. Weirs can
generally beclassified assharp-crestedweirs andbroad-crestedweirs. Weirs arediscussed indetail
in Bas et al. (1984), Brater et al. (1996), andReplogle et al. (1999).
Asharp-crestedweir isbasically athinplatemounted perpendicular totheflow withthetopofthe
plate having a beveled, sharp edge, which makes the nappe spring clear from the plate (see
Figure 5.6.1). Therateof flow is determined bymeasuring thehead, typically in a stilling well(see
Figure 5.6.2) at a distance upstream from the crest. The headH is measured using a gauge.
Suppressed Rectangular Weir
These sharp-crested weirs areaswide asthechannel, andthewidth ofthenappe isthesamelengthas
thecrest. Referring toFigure 5.6.1, consider anelemental areadA = Bdhandassume thevelocity is
J2ih; then the elemental flow is
(5.6.1)
Thedischarge isexpressed byintegrating equation (5.6.1) overtheareaabove thetopof theweir
crest:
H H
Q= JdQ = ViiBJh
1
/
2dh
=~ ViiBH
3
/
2
o 0
(5.6.2)
v
2
a
2g
Drawdown
Figure5.6.1 Flow over sharp-crested weir.
tr
1
1
1
--------B-------+I
I
u u
~

Stilling
-
well
tdh
C
~
Crest
Figure5.6.2 Rectangular sharp-crested weir without endcontraction.
5.6 Discharge Measurement 159
Friction effects havebeenneglected inderivation ofequation (5.6.2). Thedrawdown effectshown in
Figure 5.6.1 andthecrest contraction indicate that the streamlines arenot parallel or normal to the
area in the plane. To account for theseeffects a coefficient of discharge Cdis used, so that
Q
= (563)
3 v- ..
where Cdis approximately 0.62. Thisisthebasicequation for a suppressed rectangular weir, which
can be expressed moregenerally as
Q = C
w
BH
3
/
2
(5.6.4)
2
c, istheweircoefficient, c, = c,"3 /li. ForU.S. customary units,c, ::::; 3.33, andfor SI
units C
w
::::; 1.84.
If thevelocity of approach Va where His measured is appreciable, thentheintegration limitsare
(5.6.5a)
(5.6.5b)
v
2
H +J* [( 2) 3/2 (2)3/2]
Q= /liB h
1
/
2dh
= CwB H+ -
v
2
...!L
2g
(
V
2
) 3/2
When 2a ::::; 0, equation (5.6.5a) can be simplified to
g V
2
3/2
Q= CwB ( H+2;)
Contracted Rectangular Weirs
Acontractedrectangular weirisanother sharp-crestedweirwithacrestthatisshorter thanthewidth
of the channel and one or two beveled end sections so that water contracts bothhorizontally and
vertically. Thisforces the nappe width to be less thanB. The effective crest lengthis
B' = B- 0.1 nH (5.6.6)
where n= 1if theweiris placed against onesidewallof thechannel so that thecontraction onone
side is suppressed andn=2 if the weir is positioned so that it is not placed against a sidewall.
Triangular Weir
Triangular or V-notch weirs aresharp-crested weirs thatareusedforrelatively smallflows, buthave
theadvantage thattheycanalsofunction forreasonably largeflows aswell. Referring toFigure5.6.3,
Figure5.6.3 Triangular sharp-crested weir.
160 Chapter 5 Hydraulic Processes: Open-Channel Flow
the rate of discharge through an elemental area, dA, is
dQ = Cd/fih dA
where dA = 2xdh, and, x = (H- so dA = 2(H- Then
dQ = Cd/fih[2(H -
and
H
Q=
o
= CdC85) fig
(5.6.7)
(5.6.8)
(5.6.9)
= C",H
S
/
2
Thevalue of C
w
for a valueof e= 90 (themostcommon) is C
w
= 2.50for U.S. customary units
and C
w
= 1.38 for SI units.
Broad-Crested Weir
Broad-crested weirs (refer to Figure5.6.4) are essentially critical-depth weirs in that if the weirs
are high enough, critical depthoccurs on the crest of the weir. For critical flow conditions
2 1/3 3
Ye = (q jg) and E = 2
Ye
for rectangular channels:

(2)3/2
Q=B.q=Bj;i=Byg\3
E)
=B 3 y'gE
3
/
2
or, assuming the approach velocity is negligible:
(
2)3/2
Q=B 3 y'gH
3
/
2
Q= C
wBH
3
/
2
Figure5.6.5illustrates a broad-crested weir installation in a concrete-lined canal.
(5.6.10)
__1__ _
i Y1
P
y2
c
28
Figure 5.6.4 Broad-crested weir.
5.6 Discharge Measurement 161
/
/
/ Survey point forestablishing gauge zero
/ / reference. L/4toLl3from end. onweir
/ / center line
/
/
/
Cross-section
rr
Y2 d
Top of canal
l_J

Section along center line
Recorder
1----'
L..., I
-* 1 I
F : l
Figure 5.6.5 Broad-crested weirin concrete-lined canal (from Bos et al. (1984)).
EXAMPLE5.6.1
Arectangular, sharp-crested suppressed weir3 mlongis 1.0 mhigh. Determine thedischarge when the
headis 150 mm.
SOLUTION Using equation (5.6.4), Q = 1.84BH1.5, the discharge is
(
150 ) 1.5
Q= 1.84(3) 1000 = 0.321 m
3/s
5.6.2 Flumes
Boset al. (1984) provideanexcellentdiscussionof flumes. Aweir is a control sectionthat is formed
by raising the channel bottom, whereasaflume is formedby narrowing a channel. When a control
sectionisformedbyraisingboththechannelbottomandnarrowingit, thestructureis usuallycalleda
flume. Figure 5.6.6 shows a distinctionbetween weirs and flumes.
Weirscanresult inrelativelylargeheadlossesand,if thewater has suspendedsediment,cancause
deposition upstream of the weir, resulting in a gradual change of the weir coefficient. These
disadvantages can be overcome in many situations by the use of flumes.
Figure5.6.7illustratesthe general layout of a flow measuringstructure.Most flow measurement
and flow regulating structures consist of (a) a converging transition, where subcritical flow is
accelerated and guided into the throat without flow separation; (b) a throat where the water
162 Chapter 5 Hydraulic Processes: Open-Channel Flow
Broad-crested weir
Long-throated flume
Usually flume
Cross-sections are through control
atweir crest orflume throat
Figure 5.6.6 Distinction between a weir and a flume (from Bas et al. (1984)).
Tailwater
. Converging.. channel
Distance to transition Diverging
gauging station'\ C transition
Throat I
:
Approach channel I I I
I
Upstream channel
entrance
Figure 5.6.7 General layout of a flow-measuring structure (from Baset al. (1984)).
accelerates to supercritical flow so that the discharge is controlled; and (c) a diverging transition
where flow velocity is gradually reduced to subcritical flow andthe potential energy is recovered.
Oneof the morewidely usedflumes is the Parshall flume (U.S. Bureau of Reclamation, 1981),
which is a Venturi-type flume illustrated in Figure5.6.8. The discharge equation for these flumes
withwidths of 1 ft (0.31 m) to 8 ft (2.4 m) is
(5.6.11)
Extend wingwall
intocanal bank~
as required A.;'
<.. ... I I
'" I
P
-------
-----
Flow
Throat
r
Diverging
Section Section
H
b
PLAN
PROFILE
Water surface
----
---
W A
gA
B C 0 E F G M N P R
FREEFLOWCAPACITY
3
MINIMUM MAXIMUM
FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. FT. IN. CFS CFS
0 6 2
7
1
4 ~
2 0 1 3
1
1
3.
2 0 1 0 2 0 1 0 0 4
1
2 11
1
1 4 .05 3.9
iii 2 8 2 2
9 2
10
1 11
1
2 10 1 3 1
10.
2 6 1 0 1 6 1 0 4
1
3 6
1
1 4 .09 8.9
8 8 8 2 2
1 0 4 6 3 0 4 4
1
2 0 2 9
1
3 0 2 0 3 0 1 3 9 4 1 0 ~ 1 8 .11 16.1
8 4 4
1 6 4 9 3 2 4 7
1
2 6 3
4 ~
3 0 2 0 3 0 1 3 9 5 6 1 8 .15 24.6
8 8
2 0 5 0 3 4 4 10
1
3 0 3 11
1
3 0 2 0 3 0 1 3 9 6 1 1 8 .42 33.1
8 2
3 0 5 6 3 8 5 4
1
4 0 5 1
1
3 0 2 0 3 0 1 3 9 7 3
1
1 8 .61 50.4
4 8 2
4 0 6 0 4 0 5
10.
5 0 6 4
1
3 0 2 0 3 0 1 6 9 8 1 0 ~
2 0 1.3 67.9
8 4 4
5 0 6 6 4 4 6 4
1
6 0 7
6.
3 0 2 0 3 0 1 6 9 10 1
1
2 0 1.6 85.6
2 8 4
6 0 7 0 4 8 6 1 0 ~ 7 0 8 9 3 0 2 0 3 0 1 6 9 11 3
1
2 0 2.6 103.5
8 2
7 0 7 6 5 0 7 4
1
8 0 9
111.
3 0 2 0 3 0 1 6 9 12 6 2 0 3.0 121.4
4 8
8 0 8 0 5 4 7 10
1
9 0 11
1 ~
3 0 2 0 3 0 1 6 9 13 8
1
2 0 3.5 139.5
8 4 4
Figure 5.6.8 Standard Parshall flume dimensions. 103-D-1225 (fromU.S. Bureau of Reclamation (1978)).
163
164 Chapter 5 Hydraulic Processes: Open-Channel Flow
whereQis thedischarge infe Is, Wis thewidthof theflume throat, andHa is the upstreamheadinft.
For smaller flumes, e.g., 6-inchflumes,
(5.6.12)
and for 9-inchflumes
(5.6.13)
EXAMPLE 5.6.2
SOLUTION
Water flows through aParshall flume with athroat width of4.0ft atadepth of2.0ft. Whatis theflow rate?
Using equation (5.6.11).
Q=4.0(4.0)(2.0)(1.522)(4.0
r 26
=47.8ft
3
/s
Figure 5.6.9 illustrates a flow-measuring structure for unlined (earthen) channels that are longer, and
consequently more expensive, than structures forconcrete-lined channels. Forconcrete-lined channels,
the approach channel andsides of thecontrol section are already available.
5.6.3 Stream Flow Measurement: Velocity-Area-Integration Method
As for weirsandflumes, streamflow is not directly measured. Instead, water level is measured and
streamflow is determined from a rating curve, which is the relationship between water surface
elevation and discharge.
Upstream
channel
Figure 5.6.9 Flow-measuring structure for earthen channel with rectangular control section (from Bos
et al. (1984)).
Battery
5.6 Discharge Measurement 165
Gas cylinder
Figure5.6.10 Water level measurement using a bubble gauge recorder. Thewater level is measured as
thebackpressure onthebubbling streamofgas byusing amercury manometer (fromRantz et al.(1982)).
Water surface level (elevation) canbe measured manually or automatically. Crest stagegauges
areusedtomeasure flood crests. Theyconsist ofawooden staffgaugeplacedinsideapipewithsmall
holesfor water entry. Corkin thepipefloats as thewaterrises andadheres to the staff(scale) at the
highest water level. Bubble gauges (shown in Figure 5.6.10) sense the water surface level by
bubbling acontinuous streamofgas(usually carbondioxide) intothewater. Thepressure required to
continuously forcethegas streamout beneath thewatersurface is a measure of thedepthof water
over the nozzle of the bubble stream. The pressure is measured with a manometer assembly to
provide a continuous record of water level in the stream (gauge height).
Rating curves are developed usinga set of measurements of discharge and gaugeheightin the
stream. Thedischarge is Q= AVwhere Vis themeanvelocity normal tothecross-sectional areaof
flowA,which is afunction of thegaugeheight. Soinordertomeasure discharge thevelocity andthe
gaugeheightmustbe determined. Ina streamor river, thevelocity varieswithdepth, as discussed in
Chapter3 (Figure 3.7.1). Therefore, the velocity must be recorded at various locations anddepths
across the stream.
Referring toFigure5.6.11, thetotaldischarge iscomputed bysumming theincremental discharge
calculated from each measurement i, i = 1, 2, ... , n, of velocity Vi, and depth Yi- These
Vi = mean of velocities
at 0.2and0.8depth
Figure 5.6.11 Computation of discharge from stream gauging data.
166 Chapter 5 Hydraulic Processes: Open-Channel Flow
.....-- Wading rod
(a)
Revolution counter - ~ - l r : ! ; . J
Anemometer
Support cable
Rudder with adjustable
balancing weight
(h)
Figure 5.6.12 (a) Propeller- and (b) Price-type current meters (from James (1988)).
measurements represent average values over the width L1wi of the stream. The total discharge is
computed using
n
Q= l:ViYi
L1wi
i=l
(5.6.14)
Boththeory andexperimental evidence indicate thatthemeanvelocityinavertical section canbe
closely approximated by the average of the velocities at 0.2 depth and 0.8 depth below the water
surface, as shown in Figure 5.6.11. If the streamis shallow, it may be possible to take a single
measurement of velocity at a 0.6 depth.
Tomeasure thevelocity inastream, acurrent meter, which isanimpellordevice, canbeused.The
speed at which the impellor rotates is proportional to the flow velocity. Figure 5.6.12a shows a
propeller typecurrent meterona wading rodandFigure5.6.l2b shows aPricecurrentmeter, which
is the most commonly used velocity meter in the United States. Refer to Wahl et al. (1995) for
detailed descriptions on the stream-gauging program of the U.S. Geological Survey.
PROBLEMS
5.1.1 Compute the hydraulic radius and hydraulic depth for a
trapezoidal flood control channel with abottomwidth of20ft,side
slopes 2:1 (h:v), and a topwidth of 40 ft.
5.1.2 Compute the hydraulic radius and hydraulic depth for a
trapezoidal flood control channel with a bottomwidth of4m,side
slopes 2:1 (h:v), and a topwidth of 8 ft.
5.1.3 Compute the hydraulic radius and hydraulic depth for a
36-inch diameter culvert witha depthof flow of 24 in.
5.1.4 Compute the hydraulic radius and hydraulic depth for a
1.5-m diameter culvert witha depth of flow of 1.24m.
5.1.5 A 2-m wide rectangular channel with a bed slope of
0.0005 has a depth of flow of 1.5 m. Manning's roughness
coefficient is 0.015. Determine the steadyuniform discharge in
the channel.
5.1.6 Determine theuniform flow depth in a rectangular channel
2.5 m wide with a discharge of 3 m
3/s.
The slope is 0.0004 and
Manning's roughness factor is 0.015.
5.1.7 Determine the uniform flow depth in a trapezoidal channel
with a bottom width of 8 ft and side slopes of 1 vertical to
2 horizontal. The discharge is 100 ft
3
/s. Manning's roughness
factor is 0.015and the channel bottom slope is 0.0004.
5.1.8 Determine the uniform flow depth in a trapezoidal channel
with a bottom width of 2.5 m and side slopes of 1 vertical to
2 horizontal witha discharge of 3 m
3/s.
The slope is 0.0004 and
Manning's roughness factor is 0.015.
5.1.9 Determine the cross-section of the greatest hydraulic effi-
ciency for a trapezoidal channel withside slope of 1 vertical to
2horizontal if thedesigndischarge is 10m
3
Is. Thechannel slopeis
0.001 andManning's roughness factor is 0.020.
5.1.10 Fora trapezoidal-shaped channel (n = 0.014andslopeSo
of 0.0002 witha 20-ftbottomwidthandsideslopes of 1vertical to
1.5 horizontal), determine the normal depth for a discharge of
1000 cfs.
5.1.11 Showthatthebesthydraulic trapezoidal section isone-half
of a hexagon.
5.1.12 Atrapezoidal channel hasa bottom width of 10ft andside
slopes of 2:1 (h:v). The channel has a slope of 0.0001 and a
Manning's roughness of 0.018. If the uniform flow depth is 4 ft,
what is the discharge in the channel?
5.1.13 Compute the normal depth of flow in a 36-in diameter
culvert with a slope of 0.0016 and Manning's n of 0.015 for a
discharge of 20 cfs.
5.1.14 A 6-ft diameter concrete-lined sewer has a bottom
slope of 1.5 ft/mi Find the depth of flow for a discharge of
20 cfs.
5.1.15 Compute the uniform flow depthin a trapezoidal channel
with a bottom width of 20 ft, slope of 0.0016, Manning's n of
0.025, andsideslopes of 2:1(h:v) fora discharge of 500cfs.What
is the velocity of flow?
5.1.16 Design atrapezoidal concrete-lined channel (n = 0.015) to
convey 100 cfs on a slope of 0.001. Assume the use of a best
hydraulic section for the design.
5.1.17 Design atrapezoidal concrete-lined channel (n = 0.015) to
convey 20 m
3/s
on a slopeof 0.0001. Assume the use of a best
hydraulic section for the design.
5.1.18 Phillips and Ingersoll (1998) presented equations for
determining the Manning's roughness factor for gravel-bed
Problems 167
streams using the relative roughness defined as (R/ d
so)
where
R is the hydraulic radius and d
so
is the median grain size. The
equation was verified for Arizona (for the range in d
so
0.28 to
0.36 ft) is n = (0.0926RI/
6)/[1.46
+ 2.2310g(R/d
so)].
Use
this equation to develop a graph illustrating the relation of n, R,
andd
so.
5.1.19 Using the slope-area method, compute the flood dis-
charge through a river reach of 800 ft havingknownvalues of
water areas, conveyances, and energy coefficients of the up-
stream and downstream end sections. The fall of the water
surface is 1.0 ft.
Au = 11,070 ft
l
Ad = 10,990 ft
l
t; = 3.034 X 10
6
K
d
= 3.103 X 10
6
au = 1.134
ad = 1.177
5.2.1 Solveexample 5.2.2for discharges of 0, 25, 75, 125, and
200fe/s.
5.2.2 Rework example 5.2.3 for a 30-cmhigh humpand a side
wall constriction that reduces the channel widthto 1.6m.
5.2.3 Compute thecriticaldepthforthechannel inproblem 5.1.5.
5.2.4 Compute thecritical depthforthechannel inproblem5.1.6.
5.2.5 Rework example 5.2.4withdischarges of0, 25,75,125,and
200 cfs.
5.2.6 Compute thecriticaldepthina36-indiameter culvertwitha
slopeof 0.0016 for a discharge of 20 cfs.
5.2.7 Compute the critical flow depth in a trapezoidal channel
with a bottom width of 20 ft a slope of 0.0016, Manning's
n of 0.025, and side slopes of 2:1 (h:v) for a discharge of 500
cfs.
5.2.8 Atrapezoidal channel has a bottomwidthof 10ft andside
slopes of 2:1 (h:v). Manning's roughness factor is 0.018. For a
uniformflow depth of 2.9ft, whatisthenormal slope(correspond-
ingtouniformflowdepth) andthecriticalslopeofthechannel fora
discharge of 200 cfs?
5.3.1 Resolve example 5.3.1 for a channel bed slopeof 0.003.
5.3.2 A 2.45-m wide rectangular channel has a bed slope of
0.0004 andManning's roughness factorof 0.015. For a discharge
of 2.83 m
3/sec,
determine the type of water surface profile for
depths of 1.52m, 0.61 m, and 0.30 m.
5.3.3 Rework problem 5.3.2witha bed slopeof 0.004.
5.3.4 Ifthechannel ofproblem5.1.10ispreceded byasteepslope
and followed by a mild slope and a sluice gate as shown in
Figure P5.3.4, sketch a possible water surface profile with the
elevations to a scaleof 1in to 10ft. Consider a discharge of 1500
cfs. For this discharge, the normal depthfor a slopeof 0.0003 is
8.18ft and for a slopeof 0.0002 is 9.13ft.
168 Chapter 5 Hydraulic Processes: Open-Channel Flow
So= 0.0002
3'
So= 0.0003
Figure P5.3.4
M
M
(a)
M
----1------
H
(b)
-- ~
--
-..........--------- ----. .--
. ~ : t - - - - - ~ i - - - - - -
H
------- Y
e
critical depth
_._.- Y
n
normal depth
(e)
Figure P5.3.5
. ~ . - -
---
---
5.3.5 Sketch possible water surface profiles for the channel in
Figure P5.3.5. Firstlocate andmarkthecontrol points, thensketch
theprofiles, marking eachprofile withtheappropriate designation.
Show anyhydraulic jumps that occur.
5.3.6 Show thatfor depths lessthanthenormal depth (y <Yn) that
Sf> So and that for Y >Yn thenSf<So.
5.3.7 Show that dyfdx = + for the Sl profile; dyldx = - for
the S2 profile; and that dy/ dx = + for the S3 profile.
5.3.8 Using the gradually varied flow equation dyldx =
(So - Sf)/ (1 - F;) define dy/ dx forcriticalflow andforuniform
flow.
5.3.9 A trapezoidal channel with a bottom width of 20 ft, a
slope of 0.0016, Manning's n of 0.025, and side slopes of 2:1
(h:v)has a discharge of 500 cfs. An obstruction in this channel
causes a backwater profile with a depth of 6.5 ft just upstream
of the obstruction. What would be the depth of flow 200 ft
and 400 ft upstream of the obstruction? How far upstream
does the normal depth occur? Assume an energy coefficient
of 1.1.
5.3.10 Consider a concrete (n=0.013) wide rectangular channel
thatdischarges 2.0m
3
Isperunitwidth offlow. Thechannel bottom
slope is 0.0001. There is a stepriseof 0.2m. Determine the flow
depth downstream of the step assuming notransition loses. Does
the water rise or fall at the step?
5.3.11 Consider a 5-mwide rectangular channel with a discharge
of 12.5 m
3/s.
Thedepthofflow upstreamofthestepis2.5m.There
is a stepriseof0.25 minthebottomofthechannel. Determine the
flow depthdownstream of the stepassuming notransition losses.
Does the water riseor fall at the step?
5.3.12 Consider the trapezoidal channel with a bottom width of
20ft, a slope of 0.0016, Manning's n of 0.025, andsideslopes of
2:1(h:v) having a discharge of 500cfs. Now a stepriseof 1.0ft is
placed in the channel bottom. Determine the flow depthdown-
stream of thebottom stepassuming notransition losses. Doesthe
water surface rise or fall at the step?
5.3.13 A 5-mwide rectangular channel with tworeaches, each
withadifferent slope, conveys 50m
3
Isof water. Thechannel slope
for thefirst reach is 0.001 andthena sudden change to a slope of
0.010. The Manning's n for the channel is 0.Q15. Perform the
necessary computations to sketch the water surface profile and
define the typeof profiles.
5.3.14 Consider a concrete (n=0.013) wide rectangular channel
(R= y) thatdischarges 2.0m
3/sec
perunitwidth offlow. Theslope
of the channel is 0.001. A low damcauses a backwater depth of
2.0 m immediately behind (upstream of) the dam. Compute the
distance upstream of the damto where the normal depthoccurs.
5.3.15 Consider a concrete (n =0.013) wide rectangular chan-
nel (R= y) that discharges 2.0m
3/sec
perunit widthof flow. The
slopeof the channel is 0.0001. A lowdamcauses a backwater
depth of 2.0 m immediately behind (upstream of) the dam.
Compute the distance upstream of the damto where the normal
depth occurs.
5.3.16 Consider aconcrete trapezoidal channel with a4-mbottom
width, side slopes of 2:1 (h:v), and a bottom slope of 0.005.
Determine the depth 150m upstream froma section that has a
measured depthof 2.0m.
5.3.17 Awide rectangular channel changes inslopefrom0.002 to
0.025. Sketch thewater surface profile fora discharge of 1.7m
3/s1
m andManning's n=0.025.
5.3.18 A 5-mwide rectangular channel with two reaches, each
with adifferent slope, conveys 40m
3
Is of water. Thechannel slope
forthefirst reach is 0.0005 andthena sudden change toa slope of
0.015 so that critical flow occurs at thetransition. Determine the
depths of flow at locations 10m, 20mand30mupstream of the
critical depth. The Manning's n for the channel is 0.015.
5.3.19 A 500-ft, 6-ft diameter reinforced concrete pipe culvert
(n= 0.012) is used to convey stormwater from a detention re-
servoir to a downstream flood control channel. The slope of the
Problems 169
culvert is0.02. Theoutletof theculvert isplaced at anelevation so
thatit willnot besubmerged. Fora discharge of 230cfs, compute
the water surface profile. Develop a spreadsheet to perform the
computations.
5.3.20 A 500-ft, 6-ft diameter reinforced concrete pipe culvert
(n= 0.012) isused toconvey stormwater from adetention reservoir
toa downstream flood control channel. Theslope of theculvert is
0.001.Theoutlet oftheculvert isplaced at anelevation sothatit will
not be submerged, andthe flow falls freely intothe flood control
channel. Foradischarge of?5cfs,compute thewater surfaceprofile.
Develop a spreadsheet toperform thecomputations.
5.4.1 Rework example 5.4.2using equation (5.4.8), Sf = (jl /7(2.
5.4.2 Resolve example 5.4.3witha discharge of 10,000 cfsanda
downstream water surface elevation of 123.5 ft.
5.4.3 Rework example 5.4.4using a discharge at rivermile1.0of
8000cfs anda discharge of7500 cfs at rivermile 1.5. Thewater
surface elevation at rivermile1.0is 123.5 ft. Allotherdataarethe
same.
5.4.4 Consider a starting (assumed) water surface elevation of
5719.5 ft at cross-section 1 for example 5.4.5 anddetermine the
water surface elevation at cross-section number 4.
5.4.5 Consider a starting (assumed) water surface elevation of
5717.6 ft at cross-section 1 for example 5.4.5 anddetermine the
computed water surface elevation at cross-section 4.
5.4.6 Perform the backwater computations at mile 2.0 for the
situation in example 5.4.4. TheManning's n values at mile2 are
0.02 for the main channel and 0.04 for the overbanks. Use an
assumed trialwatersurface elevation of 125.5 ft. Cross-sections at
miles 2.0and 1.5are the same.
5.4.7 Formostnatural channels andmany designed channels, the
roughness varies alongthewettedperimeterofthechannel. Inorder
toperformnormal flowcomputations forthesecompositechannels
it is necessary tocompute thecomposite (equivalent or effective)
roughness factor. Forthecomposite channel inFigure 5.4.2, com-
putetheeffective roughness factor (n
e
) at rivermile1.0forawater
surface elevation of 125 ft using thefollowing equations:
and
where Pi and n, are, respectively, the wetted perimeter and
N
Manning's n for eachsubsection of thechannel; P= L, Pi is the
i=l
wetted perimeter of the complete channel section, and N is the
number of subsections of the channel.
5.4.8 UsetheU.S. Army Corps ofEngineers HEC-RAS computer
codeto solve Example 5.4.4.
5.5.11 A hydraulic jump is formed in a lO-ft widechannel just
downstreamofasluicegateforadischarge of450cfs.Ifthedepths
of flow are 30 ft and 2 ft just upstream and downstream of the
gate, respectively, determine thedepth of flow downstream of the
jump. What is the energy loss in the jump? What is the thrust
[F
gate
= y(M)] on the gate? Illustrate the thrust on the specific
force and specific energy diagrams for this problem.
5.5.12 Consider a 40-ft widehorizontal rectangular channel with
adischarge of 400cfs.Determine theinitialandsequent depths of
a hydraulic jump, if the energy loss is 5 ft.
5.6.1 A rectangular, sharp-crested weir with end contraction is
1.6mlong. How highshould it beplaced ina channel tomaintain
an upstream depth of 2.5 mfor 0.5 m
3/s
flow rate?
5.6.2 Fora sharp-crested suppressed weir(C
w
= 3.33) of length
B = 8.0ft, P = 2.0ft, andH = 1.0ft, determine the discharge
over the weir. Neglect the velocity of approach head.
5.6.3 Rework problem 5.6.2 incorporating the velocity of ap-
proach head(equation (5.6.5a)).
5.6.4 Rework example 5.6.2using equation (5.6.5b).
5.6.5 A rectangular sharp-crested weir with end contractions is
1.5mlong. How highshould theweircrestbeplacedina channel
to maintain an upstream depth of 2.5 m for 0.5 m
3/s
flow rate?
5.6.6 Determine theheadona60V-notch weirforadischarge of
150 lis. Take Cd = 0.58.
5.6.7 The head on a 90 V-notch weir is 1.5 ft. Determine the
discharge.
5.6.8 Determine the weircoefficient of a 90 V-notch weirfor a
headof 180mmfor a flow rate of 20 lis.
5.6.9 Determine the required headfor a flow of 3.0 m
3/s
over a
broad-crested weir 1.5 mhighand3 mlongwitha well-rounded
upstream comer (C
w
=1.67).
5.6.10 Water flows through aParshall flume withathroatwidth of
4.0 ft at a depth of 7.5 ft. Determine the flow rate.
5.6.11 Water flows through aParshall flume withathroatwidth of
5.0 ft at a depth of 3.4 ft. Determine the flow rate.
5.6.12 Thefollowing information wasobtained froma discharge
measurement on a stream. Determine the discharge.
170 Chapter 5 Hydraulic Processes: Open-Channel Flow
5.4.9 UsetheU.S. ArmyCorps ofEngineers HEC-RAS computer
code to solveexample 5.4.5.
5.5.1 Consider a 2.45-m wide rectangular channel with a bed
slopeof0.0004 andaManning's roughness factor of0.015. Aweir
is placed in the channel and the depth upstream of the weir is
1.52mforadischarge of5.66m
3
Is. Determine whether ahydraulic
jump forms upstream of the weir.
5.5.2 A hydraulic jump occurs in a rectangular channel 4.0 m
wide. Thewaterdepthbefore thejumpis 0.4mandafterthejump
is 1.7m. Compute theflow ratein thechannel, thecriticaldepth,
andthe headloss in thejump.
5.5.3 Rework example 5.5.3 withaflow rateof450cfsat anormal
depthof 3.2 ft. All other data remain the same.
5.5.4 Rework example 5.5.4if thedepth before thejumpis 0.8m
and all other data remain the same.
5.5.5 A rectangular channel is 3.0 m wide (n = 0.018) with a
discharge of 14m
3/s
at a normal depth of 1.0m. An obstruction
causes thedepthjust upstream ofthe obstruction tobe2.7mdeep.
Will a jump formupstream of the obstruction? If thejump does
form, howfar upstream is it located?
5.5.6 Rework example 5.5.4if the depth after thejumpis 1.8m
and all other data remain the same.
5.5.7 A lO-ft wide rectangular channel (n=0.015) has a dis-
charge of 251.5 cfs at a uniform flow (normal) depth of 2.5ft. A
sluicegateat thedownstream endof thechannel controls theflow
depth just upstream of thegatetoadepthz. Determine thedepth z
sothat a hydraulic jumpis formed just upstreamof thegate. What
is thechannel bottom slope? Whatis theheadloss (energy loss) in
the hydraulic jump?
5.5.8 A3-mwiderectangular channel (n = 0.02)hasa discharge
of 10m
3
Isat a uniformflow (normal) depthof0.8m.Asluice gate
at thedownstream endof thechannel controls theflow depth just
upstream of thegateto a depth z. Determine thedepth z sothat a
hydraulic depthis formed just upstream of the gate. Whatis the
channel bottom slope? What is the headloss (energy loss)in the
hydraulic jump?
5.5.9 A 5-m wide rectangular channel with two reaches, each
with a different slope, conveys 40 m
3
Is of water. The channel
slope for the first reach is 0.015 and then a sudden changeto a
slope of 0.0005. The Manning's n for the channel is 0.Ql5.
Does a hydraulic jump occur in the channel? If there is a
hydraulic jump, where does it occur: on the first reach or the
second reach?
5.5.10 A 5-mwide rectangular channel with tworeaches, each
with a different slope, conveys 80 m
3/s
of water. The channel
slope for the first reach is 0.01 and then a sudden change to a
slope of 0.001. The Manning's n for the channel is 0.Ql5.
Does a hydraulic jump occur in the channel? If there is a
hydraulic jump, where does it occur: on the first reach or the
secondreach?
Distance frombank(ft)
o
12
32
52
72
92
100
Depth (ft)
0.0
0.1
4.4
4.6
5.7
4.3
0.0
Meanvelocity (ft)
0.00
0.37
0.87
1.09
1.34
0.71
0.00
References 171
REFERENCES
Barnes, H.H.Jr.. Roughness Characteristics ()fNatural Channels, U.S.
Geological Survey, Water Supply Paper H!49, U.S. Government Printing
Office, Washington, DC, 1962.
Bos, M. G., 1. A. Replogle, and A. J. Clemmens, Flow Measuring
Flumes for Open Channel System, JohnWiley & Sons, NewYork, 1984.
Brater, E. E, H. W. King, 1. E. Lindell, andC. Y. Wei, Handbook of
Hydraulics, 7thedition, McGraw-Hili, NewYork, 1996.
Chaudhry, M. H., Open-Channel Flow, Prentice Hall, Englewood
Cliffs, NJ, 1993.
Chow, V. T.,Open-Channel Hydraulics, McGraw-Hili, NewYork, 1959.
French, R. H., Open-Channel Hydraulics, McGraw-Hili, NewYork,
1985.
Henderson, EM., Open-Channel Flow, Macmillan, NewYork, 1966.
Hoggan, D. H., Computer-Assisted Floodplain Hydrology and Hy-
draulics, 2nd edition, McGraw-Hili, NewYork, 1997.
Hydrologic Engineering Center(HEC), HEC-RAS RiverSystemAna-
lysis System, User's Manual, Version 2.2, U. S. ArmyCorpsof Engineers
Water Resources Support Center, Davis, CA, 1997a.
Hydrologic Engineering Center (HEC), HEC-RAS River Analysis
System, Hydraulic Reference Manual, Version 2.0, U.S. Army Corps of
Engineers WaterResources Support Center, Davis, CA, 1997b.
Hydrologic Engineering Center (HEC), HEC-RAS River Analysis
System, Applications Guide, Version 2.0, U. S. Army Corpsof Engineers
Water Resources Support Center, Davis, CA, 1997c.
Jain, S. C., Open-Channel Flow, JohnWiley & Sons, Inc., NewYork,
2001.
James, L. G., Principles of Farm Irrigation System Design, JohnWiley
andSons, Inc., NewYork, 1988.
Manning, R., "On the Flowof Water in OpenChannels and Pipes,"
Transactions Instituteof CivilEngineers of Irelund, vol. 20, pp. 161-209,
Dublin, 1891; Supplement, vol. 24, pp. 179-207, 1895.
Phillips, J. v., and T. L. Ingersoll, "Verification of Roughness Co-
efficients for Selected Natural and Constructed Stream Channels in
Arizona," U.S. Geological Survey Prof. Paper 1584, 1998.
Rantz, S. E. et aI., Measurement and Computation of Stream Flow,
vol. 1, Measurement of StageandDischarge, WaterSupply Paper2175,
U.S. Geological Survey, 1982.
Replogle, J. A, A 1. Clemmens, andC. A Pugh,"Hydraulic Design of
FlowMeasuring Structures," in Hydraulic Design Handbook, editedbyL.
W. Mays, McGraw-Hili, NewYork, 1999.
Sturm, T., Open Channel Hydraulics, McGraw-Hili, New York,
2001.
Townson, J. M.,Free-Surface Hydraulics, Unwin Hyman, London, 1991.
U.S. Bureau of Reclamation, Guide for Computing Water Surface
Profiles, 1957.
U.S. Bureau of Reclamation, Design of Small Canal Structures, U.S.
Government Printing Office, Denver, CO, 1978.
U.S. Bureau of Reclamation, Water Measurement Manual, U.S. Gov-
ernment Printing Office, Denver, CO, 1981.
Wahl, K. L., W. O. Thomas, Jr., and R. M. Hirsch, "Stream-Gauging
Program of theU.S.Geological Survey," Circular 11123, U.S.Geological
Survey, Reston, VA, 1995.