Section 4 Complete
Section 4 Complete
TABLE OF CONTENTS
FIGURES
Figure 4.1. Terms in the Energy Equation 4-7
Figure 4.2: Nomograph for Calculation of Normal Depth 4-14
Figure 4.3: Permissible Shear Stress for Cohesive Soils 4-16
TABLES
Table 4.1: Principal Features of the River System in Addis Ababa 4-1
Table 4.2: Probable Flood Peak Discharges of Addis Ababa Rivers 4-1
Table 4.3: Channel Characteristics of Addis Ababa Rivers 4-2
Table 4.4: Values of Roughness Coefficient ‘n’ for Uniform Flow 4-8
Table 4.5: Impact of Changes in Sediment Supply 4-11
Table 4.6: Qualitative Response of Alluvial Channels 4-12
Table 4.7: Manning’s Roughness Coefficients for Roadside Drains 4-13
Table 4.8: Grass Cover Classification for Use of Table 4.9 4-13
Table 4.9: Permissible Shear Stress for Various Protective Material 4-15
The city of Addis Ababa is situated at altitudes varying from 2,350 to 2,700 meters. Undulating and flat
topography of city spreads to the southern slopes of Intoto Ridge. The urban land includes valleys formed by
five major river systems flowing towards south. Flat topography gradually slopes to the south easterly
direction.
River System Basin Area (km2) River Length (km) River Slope
West Akaki 172.2 35.6 1/50 – 1/100
Little Akaki 30.8 20.5 1/25 – 1/100
Kebena 59.8 23.9 1/20 – 1/100
(Upper Kebena) (54.8)
(Lower Kebena) (5.0)
Bantyiketu 29.3
(Bantyiketu) (5.4) 1/100
(Kechene) (13.6) 11.2 1/20-1/50
(Kurtume) (10.3) 9.3 1/20-1/50
Hanku 11.1 8.6 1/50 – 1/70
Flood runoff analysis was conducted by Nippon Koei and Associates (JICA, 1998) using rational method in
conjunction with the triangular unit hydrograph. River discharges for return periods ranging from 2- to 30-year
were computed. Table 4.2 presents the probable flood peak discharges for the river systems.
Based on the river surveys and the hydraulic analysis Nippon Koei and Associates (JICA, 1998) computed
river channel characteristics. Channel widths and the river carrying capacities were among the important
information compiled. Table 4.3 presents the river channel characteristics.
Table 4.2 indicates that Addis Ababa rivers have steep bed gradient. The beds are generally composed of
volcanic rock (JICA, 1998). This would indicate quite stable beds.
In simpler cases, the procedures can be used readily and without the help of computer programs. This
section covers such procedures. More involved cases of complex cross-sections of natural streams or the
unsteady nature of the flow, enormous amount of computations are required which require the use of
computers. This Chapter presents an overview of computer modeling for such complex cases.
Specific energy head is the energy head relative to the channel bottom. The total energy head is the specific
energy head plus the elevation of the channel bottom with respect to some datum.
The locus of the energy head from one cross section to the next defines the energy grade line.
Specific Energy (E) is defined as the energy head relative to the channel bottom. If the channel is not too
steep (slope less than 10 percent) and the streamlines are nearly straight and parallel (so that the hydrostatic
assumption holds), the specific energy E becomes the sum of the depth and velocity head:
E = y + α (V2/2g) (4.1)
Where,
y = flow depth (m)
α = velocity distribution coefficient
V = mean velocity (m/s)
g = gravitational acceleration (9.807 m/s2)
Due to the presence of a free surface and also to friction along the channel boundary, the velocities are not
uniformly distributed in the channel section. As a result, the velocity head of an open channel is usually
greater than the average velocity head computed as (Q/At)2/2g. A weighted average value of the velocity
head is obtained by multiplying the average velocity head, above, by a velocity distribution coefficient, α,
defined as:
n
α = ∑(Ki3/Ai2)/(Kt3/At2) (4.2)
i=1.
Where,
Ki = conveyance in subsection (Section 4.3.3b)
Kt = total conveyance in section
Ai = cross-sectional area of subsection (m2)
At = total cross-sectional area of section (m2)
n = number of subsections
The Froude number (Fr) is an important dimensionless parameter in open channel flow. It represents the
ratio of inertia forces to gravity forces and is defined by:
Fr = V/[(gd/α).5 (4.3)
Where,
V = mean velocity = Q/A (m/s)
d = hydraulic depth = A/T (m)
A = cross-sectional area of flow (m2)
T = channel top width at the water surface (m)
Q = total discharge (m3/s)
This expression for Froude number applies to any single section channel of nonrectangular shape.
The flow in open channels at which the energy content of the fluid is at a minimum.
Critical depth is the depth of maximum discharge when the specific energy is held constant. Froude number
at critical depth has a value of one. Following relation holds at critical depth:
Depths greater than critical occur in subcritical flow and the Froude number is less than one. In this state of
flow, small water surface disturbances can travel both upstream and downstream, and the control is always
located downstream.
In supercritical flow the depth is less than the critical depth and the Froude number is more than one. Small
water-surface disturbances are always swept downstream in supercritical flow, and the location of the flow
control is always upstream.
A hydraulic jump occurs when the flow changes from supercritical to subcritical. This takes place as an
abrupt transition associated with dissipation of energy. There are significant changes in depth and velocity
before and after the jump. The hydraulic jump is often employed to dissipate energy and control erosion at
highway drainage structures.
A steady flow is one in which the discharge passing a given cross-section is constant with respect to time.
The maintenance of steady flow in any reach requires that the rates of inflow and outflow be constant and
equal. When the discharge varies with time, the flow is unsteady.
A uniform flow is one in which the velocity and depth remain constant along channel length, while they vary
in non-uniform flow. Uniform flow can only occur in a prismatic channel, which is a channel of constant cross
section, roughness, and slope in the flow direction. However, non-uniform flow can occur either in a prismatic
channel or in a natural channel with variable properties.
A non-uniform flow in which the depth and velocity change gradually in the flow direction, such that vertical
accelerations can be neglected, is referred to as a gradually-varied flow; otherwise, it is considered to be
rapidly-varied.
Steady Flow
Uniform Flow
Non-uniform Flow
− Gradually Varied Flow
− Rapidly Varied Flow
Unsteady Flow
The steady uniform flow case and the steady non-uniform flow case are the most fundamental types of flow
treated in highway engineering hydraulics.
The following equations are most commonly used to analyze open channel flow. For simpler cases a hand-
held calculator solution of these equations can serve the purpose while for complex cases they form part of
computer programs.
Manning’s Equation is the most widely used depth-velocity relation for the open-channel flow. For a given
depth of flow in a channel with a steady uniform flow, the mean velocity, V, can be computed with Manning’s
equation:
The selection of Manning’s ‘n’ is generally based on observation; however, considerable experience is
essential in selecting appropriate ‘n’ values. The range of ‘n’ values for various types of channels and
floodplains is given in Table 4.4.
For a given channel geometry, slope, roughness, and a specified value of discharge Q, a unique value of
depth occurs in steady uniform flow. It is called the normal depth. The normal depth is used to design
artificial channels in steady, uniform flow and is computed from Manning’s Equation:
If the normal depth computed from Manning’s Equation is greater than critical depth, the slope is classified
as a mild slope. Conversely, if the normal depth is less than critical depth, the slope is a steep slope. Thus,
uniform flow is subcritical on a mild slope and supercritical on a steep slope.
In channel analysis, it is often convenient to group the channel properties in a single term called the channel
conveyance K:
The conveyance represents the carrying capacity of a stream cross-section based upon its geometry and
roughness characteristics alone and is independent of the streambed slope.
The concept of channel conveyance is useful when computing the distribution of overbank flood flows in the
stream cross section and the flow distribution through the opening in a proposed stream crossing.
The shear stress at the boundary (το) for steady uniform flow, determined by applying the conservation of
mass and momentum principles to the control volume is given by the following equation:
το = γRSf (4.9)
The continuity equation is the statement of conservation of mass in fluid mechanics. For the special case of
one dimensional, steady flow of an incompressible fluid, it assumes the form:
Where, Subscripts 1 and 2 refer to successive cross sections along the flow path.
The energy equation expresses conservation of energy in open channel flow as energy per unit weight of
fluid with a dimension of head, it is therefore called the energy head. The energy head is composed of
potential energy head (elevation head), pressure head, and kinetic energy head (velocity head). These
energy heads are scalar quantities that give the total energy head at any cross section when added. In
comparing an upstream open channel cross section designated 1 and a downstream cross section
designated 2, the energy equation is:
The stage ‘h’ is the sum of the elevation head ‘z’ at the channel bottom and the pressure head, or depth of
flow ‘y,’ i.e. h = z + y. The terms in the energy equation are illustrated graphically in Figure 4.1. The energy
equation states that the total energy head at an upstream cross section is equal to the energy head at a
downstream section plus the intervening energy head loss. The energy equation can only be applied
between two cross sections at which the streamlines are nearly straight and parallel so that vertical
accelerations can be neglected.
Stage-discharge relation for open channels can be determined accurately from stream gauging records if
these are available. The alternative is a theoretical approach such as Slope-Area Method.
The Slope-Area Method is a single-section analysis method and essentially involves solution of Manning’s
Equation for normal depth of flow. The method presumes the existence of steady uniform flow. The
procedure employed by this method has been explained in the worked example in Section 6.8
Manning’s ‘n’ is affected by many factors and its selection in natural channels depends heavily on
engineering experience. Once the Manning’s ‘n’ values have been selected, it is highly recommended that
they be verified with historical high water marks and/or gauged stream flow data. In computer modelling, this
is done as main part of the calibrating process.
Manning’s ‘n’ values for artificial channels are more easily defined than for natural stream channels. Table
4.4 presents typical ‘n’ values for both artificial and natural stream channels.
Rivers are seldom in a state of perfect equilibrium. Some rivers achieve a state of practical equilibrium
throughout long reaches. The stable streams are also called "poised" by engineers and "graded" by
geologists. However, significant changes do occur over a short period of time or over a period of years even
is the stable streams.
River reaches react to interventions like bridge, culverts or training works. Engineering design must consider
the possible impact of the proposed interventions to enhance the natural tendency of the stream toward
stable conditions. To do so, an understanding of the direction and magnitude of change in channel
characteristics caused by the actions of man and nature is required. Following factors contribute to this
understanding:
A large numbers of variables are interrelated and can respond to changes in a river system. Precise
prediction of the river behaviour is a complex task. Some useful methods have been developed to predict the
response of channel systems to changes both qualitatively and quantitatively. Quantitative methods are
computer based and depend on fairly advanced techniques of hydrodynamics and sediment transport which
makes them irrelevant to this Manual. However, the following description covers the qualitative approach
which is quite useful for predicting in general terms the response of a river to external changes particularly
where limited data precludes quantitative analysis.
Usually the data are not sufficient for quantitative estimates, and only qualitative estimates are possible.
Qualitative assessment aims to predict general river behaviour from the observed changes in the watershed,
evolving morphology or construction of hydraulic structures. Research supports the following general
statements (HIRE, 1990):
1. Depth of flow is directly proportional to water discharge and inversely proportional to sediment
discharge;
5. Slope of stream channel is inversely proportional to water discharge and directly proportional to
sediment discharge and grain size; and
6. Sinuosity of stream channel is proportional to valley slope and inversely proportional to sediment
discharge.
The above statements apply to natural rivers and not necessarily to artificial channels with bank materials
that are not representative of sediment load. The relations will, however, help to determine the response of
water conveying channels to change. One proportionality generally supported by the research data is QS ~
Qs D50, where Q and Qs are water and sediment discharge respectively; S is the channel bed slope.
Research has lead to the conclusions (HIRE, 1990) the response of a river system to changes in variables
can be stated in terms of variables given in Table 4.5. Cf is volumetric concentration of the fine sediment, W
is the channel width, yo is the channel depth and V is the average velocity while other variables used in Table
4.5 have been defined above. A plus (+) sign signifies an increase in the value of the variable and a minus (-)
sign signifies a decrease in the value of the variable. The letter B indicates an increase in the product SQ¼
and a shift toward a braided condition and the letter M indicates a reduction in SQ¼ and a shift toward the
meandering condition. No attempt is made here to determine whether or not the channel braids or meanders.
hor Meander
As discussed in Section 4.3.1, the proportionality QS ~ Qs D50 can be used to predict changes in channel
profiles caused by changes in water and sediment discharge. From what is known on the subject, it is
possible to talk qualitatively about changes in channel profile, changes in river form and changes in river
cross section both at a station and along the river channel using the other relations presented above. Table
4.6 presents impact of independent variables of discharge; sediment load and vegetation etc can have on
river form, flow regime and channel stability.
An important aspect of the design of the roadside drains is the assessment of their stability, for which it is
essential to know the shear stress to which a given bank or bed material can be safely subjected to. Table
4.9 presents the allowable shear stresses for various protection measures.
Note: Values listed are representative values for the respective coefficients, n, varying with the flow depth.
Category Condition
A Taller than 750 mm (Excellent Stand)
B 300 – 600 mm (Good Stand)
C Uncut 150 – 300 mm (Good Stand)
D Uncut 50 – 150 mm (Good Stand)
E Cut to 40 mm (Good Stand)
τp (Pa)
4.1.5.1.1.1.1 Protective Cover 4.1.5.1.2 Category
Vegetation Class A 177
Class B 101
Class C 48
Class D 29
Class E 17
Gravel: D50 = 25 mm 19
D50 = 50 mm 4.1.5.1.2.1.1 38
Due to the centrifugal force induced by the flow, the water surface at the outer bend experiences a super
elevation. The super elevation can be computed by the following equation:
V 2T
SE = (4.12)
gR c
Where,
SE = super elevation (m)
V = mean velocity (m/s)
g = acceleration due to gravity (9.807 m/s2)
Rc = mean radius of the bend
4.4.2 Freeboard
An important factor in design of road side channels is the vertical distance from the water surface to the top
of the channel known as the freeboard. It should be sufficient to prevent waves or fluctuations in the water
surface from overflowing the sides. For road side drains a minimum free board of 0.15 m should be used
(HEC15 1988). Steep gradient channels should have a higher freeboard with a maximum equal to the depth
of design flow. This is to prevent the overflow caused by splashing and surging. The lining material should
extend to the freeboard.
Table 4.9 presents permissible shear stresses for a range of protective material. The following simple
relation provides guidance to select minimum rock size for the riprap (HEC 15):
τp = 628.3d50 (4.13)
For cohesive soils, Figure 4.3 can be used to determine the permissible stress.
The current version (version 3) of HEC-RAS supports one dimensional steady and unsteady flow water
surface profile calculations. It is designed for interactive use in a multi-tasking, multi-user network
environment. The current version can be used for steady flow water surface profiles and unsteady flow
computations. The future versions will include mobile boundary sediment transport calculations.
The steady flow computations yield water surface profiles for gradually varied flow. It can handle single river
reach, dendritic system or a full network of channels. Hydraulic design of bridges and culverts, so often
treated in city drainage systems like that of Addis Ababa are also built in the model. The model out put
options includes a range of variables used in design of lining of the road side drains and also for scour and
erosion protection measures around bridges and culverts.
The unsteady flow capability of the model can be used to track variations of the flow field in time and space.
The input includes flow hydrograph at an upstream point and the out put consists of a water levels, flow
velocities and other parameters along river length in time.
Because of its universal acceptance and applicability of its hydraulic features to the road drainage, HEC-RAS
is recommended for computation of open-channel flow and other related aspects.
Given: A roadside drainage channel is trapezoidal with a bottom width of 1.22 m and 1V:4H side
slopes. The bed slope is 0.01 m/m and the design flow rate is 0.28 m3/s.
Find: Calculate the design depth of flow and the required diameter (D50) of a gravel riprap that is to
be used as a permanent channel lining.
Solution:
(1) Solve for normal depth of flow, y, using Fig. 4.2 Nomograph
n=0.03 (Given in Fig. 4.2 Example). Alternatively use Table 4.7 (n=0.033 for 25
mm D50 and depth 0.15-0.6 m)
Q(n) = 0.0084
y/B = 0.14
y = 0.18 m
(2) Determine permissible shear stress for D50 = 25 mm
Now because τd < τp, accept D50 of approximately 25 mm. Otherwise repeat with
another riprap diameter.
Given: A drain lined with a good stand of native grasses (approximately 0.2 m in height). The drain is
trapezoidal with a bottom width of 1.2 m and side slopes of 1V:4H. The ditch slope is 0.010 m/m.
Find: The maximum discharge for which this lining will be stable and the corresponding flow depth.
Solution:
(1) From Table 4.9 , the native grass has is Category C and from the permissible shear stress is:
τp = 48 Pa
(4) Finally determine the Manning’s n value from Table 4.4 and solve for Q from Manning’s equation:
n = 0.030 and
4.7 References
Ethiopian Road Authority (ERA - 2001), “Drainage Design Manual”, Federal Democratic Republic of Ethiopia.
HEC 15SI (1988) - Federal Highway Administration, ”Design of Roadside Channels with Flexible Lining”, US
Department of Transportation.
HEC-RAS (2001), “Hydrologic Engineering Center River Analysis System – version 3”, US Army Corps of
Engineers.
Japan International Cooperation Agency (JICA – 1998), “The Study on Addis Ababa Flood Control Project in
the Federal Democratic Republic of Ethiopia – Volume II Main Report”, Nippon Koei Co., Ltd and Nikken
Consultants, Inc.