DEBRE TABOR UNIVERSITY

GAFAT INSTITUTE OF TECHNOLOGY

DEPARTMENT OF HYDRAULIC AND WATER

RESOURCES ENGINEERING

SEDIE DIVERSION SMALL SCALE IRRIGATION PROJECT

1
Introduction
1.1. General
Water is a critical resource for agricultural productivity, particularly in regions where rainfall is
erratic or seasonal. In many developing countries, small-scale irrigation projects play a vital role
in enhancing food security, improving livelihoods, and fostering sustainable rural development
(FAO, 2017). The Sedie Diversion Small Scale Irrigation Project is one such initiative,
designed to harness local water resources to support agricultural activities in the Sedie catchment
area.

By diverting water from nearby surface sources through engineered channels and structures, the
project aims to ensure a consistent and regulated water supply to farmlands. This facilitates year-
round crop production, reduces dependence on unpredictable rainfall, and helps improve the
resilience of local farming communities to climate variability (IWMI, 2018). In addition, small-
scale irrigation has been shown to significantly boost agricultural productivity and income in
similar agro-ecological contexts (World Bank, 2016).

This thesis provides a comprehensive study of the Sedie Diversion Small Scale Irrigation Project,
covering aspects such as planning, hydrological assessment, engineering design, implementation
strategies, and socio-economic impacts. It also examines the environmental implications and
operational challenges associated with the project. The study aims to contribute valuable insights
into the design and management of community-based irrigation systems, thereby informing
future efforts in sustainable water resource development.

1.1.1. Location and Accessibility

This irrigation project is located mainly at in East Gojjam zone of Amhara National Regional
State. It is located in Hulet Eju Enese wereda, Addiss Alem Kebele. The specific location of the
project site is called Afkira got. The proposed irrigation project is to be undertaken on Sedie
River, the headwork structures are specifically located at an altitude of about 2491.95masl, and
geographical coordinates of 372962.68 E, 1203919.51N (UTM).

The project Headwork site is accessible 33 km from Motta.

2
1.2. Objective of the Study

1.2.1. General Objective

The main objective of the Sedie Diversion Small Scale Irrigation Project is to design
and evaluate an efficient, sustainable, and community-focused irrigation system that
utilizes diverted surface water to improve agricultural productivity and enhance the
livelihoods of farmers in the Sedie area.

1.2.2. Specific Objectives

 To assess the availability and reliability of water sources for diversion and irrigation
purposes.
 To design appropriate hydraulic structures and conveyance systems suitable for small-
scale irrigation.
 To estimate the potential increase in crop production and income generation as a result of
the irrigation intervention.
 To provide recommendations for the sustainable management and operation of the
irrigation system.

1.3. Scope of the Study

This study focuses on the planning, design, and evaluation of the Sedie Diversion Small Scale
Irrigation Project located in the Sedie area. The scope is limited to technical, hydrological,
environmental, and socio-economic aspects directly related to the development and
implementation of a small-scale surface water diversion irrigation system.

The study covers:

 Site selection and assessment, including topographic and soil suitability analysis.
 Hydrological analysis, including estimation of streamflow and irrigation water demand.
 Design of irrigation structures, such as diversion weir, canals, and control structures.

3
  Command area analysis, focusing on the land to be irrigated and its suitability for
selected crops.
 Environmental and socio-economic impact assessment, including potential benefits and
challenges.
 Operation and maintenance planning for long-term sustainability of the system.

The study does not include large-scale irrigation development, groundwater extraction, or
detailed agronomic studies beyond crop water requirements. It is primarily intended to support
smallholder farmers and local authorities with practical solutions for improved water use and
agricultural productivity.

1.4. Methodology
The methodology adopted for this study involves a combination of field investigation (which
was given from advisor), data collection, technical analysis, and design procedures to assess and
develop the Sedie Diversion Small Scale Irrigation Project. The main steps include:

1.4.1. Data Collection

Primary Data

 Field surveys to gather topographic data using GPS and leveling instruments.
 Soil sampling and testing for texture, infiltration rate, and fertility.
 Streamflow measurement to estimate water availability for irrigation

Secondary Data
 Meteorological data (rainfall, temperature, evapotranspiration) from nearby weather
stations.
 Demographic and socio-economic data from local agricultural offices and community
interviews.
 Existing reports and maps from relevant government agencies.
1. Hydrological Analysis
 Estimation of available discharge using historical flow records and field measurements.
 Calculation of irrigation water demand based on crop type, area coverage, and climatic
conditions using the FAO CROPWAT model.
2. Irrigation System Design
4
  Design of diversion weir, intake structure, and canal systems using standard hydraulic
equations.
 Determination of canal alignment, slope, and dimensions based on topographic survey
results.
 Selection of construction materials and structural dimensions in accordance with design
standards and local availability.
3. Command Area Analysis
 Mapping and delineation of the irrigable area using GIS tools and field observations.
 Assessment of land suitability based on slope, soil, and land use.
4. Environmental and Socio-Economic Assessment
 Identification of potential environmental impacts (e.g., waterlogging, erosion) and
mitigation measures.
 Evaluation of the project's potential to improve agricultural productivity, income, and
food security through stakeholder consultations.
5. Feasibility and Sustainability Assessment
 Technical feasibility analysis based on design outputs and resource availability.
 Cost estimation and economic analysis including benefit-cost ratio and return on
investment.
 Recommendations for operation and maintenance strategies to ensure long-term
sustainability.

2. HYDROLOGICAL ANALYSIS
2.1 General
Hydrological data is used in the design of hydraulic and irrigation structures .To deduce from its
analysis a few significant figures such as minimum and maximum discharge of the river at the
proposed location of the required. This is done by determining the design storm from the available

5
rainfall data and then synthesizing the flood structure. To obtain these figures a flow record of
many years as much as possible is hydrograph.

The same is true for Sedie River, i.e. the ungauged flood hydrograph is developed for 24 years
maximum daily rainfall data obtained from Motta and Felege birehan meteorological station,
which has similar climatologically, feature to the project area.

2.1 Data Availability

There is peak daily rainfall from the nearby station, because of non-existence of gauged flow
data, no discharge data. Since our attention is determining runoff discharge, it is possible to
convert peak daily rainfall data to runoff through different methods.

Basic hydrological data required are;

 Hydro meteorological data like temperature, wind velocity, humidity, etc.
 Flood records
 Evaporation data
 Cropping pattern, crops and their consumptive use
These data should be stationary, consistent, and homogeneous when they are used for frequency
analyses or to simulate a hydrological system.

2.2. Peak Rainfall Determination

Maximum design discharge is the peak river discharge that corresponds to a certain return
period. The maximum design discharge Qmax. Is used in the design to determine the back water
curve results from constructing the weir ,which enables to predict the highest water level that
occurs average once every T years ,where T is the selected return period of the discharge . The
Qmax. Determines the water afflux on the weir and hence the height of the weir, wing walls and
the cross bridge is estimated .It also gives information to determine the height of dykes if
required.

To estimate the magnitude of flood peak the following alternative methods are available.
1 Rational Method
2 Empirical formula Method
3 Synthetic Unit Hydrograph Technique (Snyder’s method)
4 Flood Frequency Analysis Method

6
 5 USSCS (United States Soil Conservation Service) Method
1. Rational Method
The rational formula is found to be suitable for peak flow prediction in small catchments up to
50km2 in area. It finds considerable application in urban drainage designs and in designs of small
Culverts and Bridges. The basic equation of rational method is given by

Qp=1/3.6(C*Itc,p*A)

Where: Qp is peak discharge (m^3/s)

C -runoff coefficient

Itc, p-The mean intensity of precipitation (mm/Hr) for a duration equal to tc

And an exceedence probability p.

A- Drainage area in km2

The use of this method to compute Qp requires parameters; Tc, (Itc, p) and C ..
Limitation:
a. Calculation of weighted run off coefficient is by far difficult as the catchments covered
by different land features with varying area coverage (which is not known for sedie
River catchments.
b. This method is applicable for small areas up to 50km2.
c. Estimation of Itc, p requires some other regional constants based on catchments
behavior.
Because of the above limitations, rational method is not convenient for the determination of peak
flood for Sedie River.

2. Empirical formula Method

The empirical used for estimation of flood peak are essentially regional formula based on
statistical correlation of the observed peak and observed catchments parameters.

Generally, this method is given as a function of catchments area.

Qp = f (A)

7
For example, Admassu developed an empirical formula through regression analysis of 42
catchments in Ethiopia with area ranging from 200-9980 km2.

Qp = Q (1+kt*Cv) ---------------------------general formula

Q =0.87*A^.7---------------------------------Dr. Admassu’s relation

Where: A-Catchments area (km2)

Kt-frequency factor

Kt =
−√ 6
x [ ( [ ])]
0.57721+ ln ln
T
T −1

T = return period

Cv = The average Coefficient of variation (= 38 for most cases)

The formula is safely adopted for most Ethiopia basins under the given area range, however; the
basin area under our consideration is not in the domain and hence we can’t use this method to
estimate the peak discharge.

To developed unit hydrographs for catchments, detailed information about the rain fall is needed.
Then the resulting flood hydrograph are obtained.

3. Synthetic Unit Hydrograph Technique (SNYDER’S METHOD)

Such information would-be available only at few locations and in majority of catchments the
data would normally be scanty .In order to construct unit hydrograph for such areas, empirical
equations of regional validity that relate the salient hydrograph characteristics to the basin
catchments are available. Unit hydrographs derived from such relationships are known as
Synthetic Unit Hydrographs.

Snyder’s Method

Snyder (1938) developed a set of empirical equation for synthetic unit hydrographs in USA. This
equation used with some modifications in many other countries and so called Snyder’s Synthetic
Unit Hydrograph.

8
The first of the Snyder’s equation relates the basin lag tp, defined as the interval from the mid
point of the unit rain fall excess to the peak of unit hydrograph, to the basin characteristics as ,

Tp = Ct (L*Lca) hr

L – Basin length measured along the watercourse from the basin divide to the gauging station in
km.

Lca – distance along the watercourse from the gauging station to appoint opposite the watershed
centered in km.

Ct – regional constant, representing watershed slope &storage.

Better correlation of basin lag tp with catchments parameter, (L*Lca)/√ S is obtained by et al .as

[ ]
n
L−Lca
Tp = Ctl
√S
Where Ctl and n are basin constants & s is basin slope Snyder as gives standards duration tr hrs
of effective rainfall

tp
Tr = hrs
5.5

The peak discharge Qps [m^3/s] of a hydrograph of standard duration tr hrs is given by Snyder
as

2.78∗Cp∗A
Qps =
tp

Where: A-Catchments area km2

Cp – a regional constant

If anon standard rain fall duration tr is adopted, instead of the standard value tr derive a unit
hydrograph, the value of the basin lag’s affected. The modified basin lag is given by:

Trh
tp’= tp, where tp – basin lag in hrs for an effective of r=¿
21 tr
t + ¿
21 p 4

The peak discharge for anon standard effective rainfall of duration Tr in m^/s is

9
 2.78Cp∗A
QP =
tp

When Tr = tr, QP=Qps

Snyder as gives the time base (tb) of unit hydrograph

tp
Tb=3+ days=(72+3 tp)
8

Finally, to assist in the sketching of unit hydrographs at 50 percentage &75% of the peak have
been found US catchment’s by the US army corps of engineers. These widths are given by:

5.18 W 50
W50 = 1.08
∧W 75 =
q 1.75

Where W50 – width of unit hydrograph in hr at 50% peak discharge

W75 - width of unit hydrograph in hr at 75% peak discharge

Q = QP/A, peak discharge per unit catchments area in m3/s/km

Since the coefficients Ct and Cp vary from region to region , in practical application. It is
advisable that the value of these coefficients are determined from known unit hydrograph of a
meteorologically homogeneous catchments and other used in the basin under study.

4. Flood Frequency Analysis Method

When the stream flow peaks are arranged in the descending order of magnitude, they constitute
statically array whose distribution can be expected in terms of frequency of occurrence. The
probability ‘p’of each event being equal to or exceeded (plotting position) formula.

m
P=
N +1

Where m = order number of the events and N=Total number of events in the data.

The recurrence interval (T) return interval, is calculated as

1
T=
p

10
But in our case there is no measured flow data instead it is possible to determine the probability
of occurrence of daily maximum rainfall (rain fall frequency analysis).

The general equation for flood frequency analysis is:

XT = µ+k*sd-------------------------------------- (Chow 1951)

Where XT = Value of variant(X) of random hydrologic series with return period (T)

µ = Mean value of variant

sd = standard deviation of variant

k = frequency factor which depends up on the return period (t) and assumed frequency
distribution.

Nevertheless, according to the data available it is rainfall frequency analysis.

Table 2.1 Guidelines for selecting Design floods [source R. Baban, page30]

№ Structure Return
period
1 Spillway for projects with storage of more than 60*106m3/sec 100 (a)
2 Barrage and minor dams with storage less than 60*106m3/sec 100(a)
3 Spill way of small reservoirs dams in the country sides ,not endangering 10-20(b)
urban resident
4 As above (3) but located so as endanger other structure Or urban residences 50-100(b)
in case of failure
5 Diversion weir 50-100(a)
6 Small bridges on main highways 50-100(b)
a- Subramanian S1989
b- Nemec 1972

Table 2.2 Present practice for estimating peak discharges from catchments areas may be grouped
as follows.

11
 Area in square Methods Commonly Used
mile
<1 Inflation approach, Rational Method
<100 Over flow Hydrograph, Rational Method, Flood frequency method
100- 2000- Unit Hydrograph, Flood frequency Method.
>2000 Flood routing and flood frequency
(From NOVAK)

2.2.1. Rainfall Frequency Analysis

Due to the lack of flow (discharge) data we are forced to analysis the peak daily rain fall for
computation of peak discharge. In the analysis of rainfall frequency, the probability of
occurrence of a particular extreme rainfall (24 hr maximum rainfall) is important. Such
information is obtained by the frequency analysis of point rainfall depth. Then the probability of
occurrence of point rainfall (24 hr maximum rainfall) is estimated for a recurrence interval of 50
years, for diversion weirs.

The prediction of peak flows from rainfall over a catchments involves estimation of daily
maximum rainfall for a given return period and conversion of the daily maximum rainfall to run
of hydrograph at the desired location.
Table 2.3 Maximum daily rainfall Values of 24 years for Motta and Felege birehan
meteorological station
S.No. Year Motta Max. Felege Birhan Max. RF
RF
1 1984 34 61.6
2 1985 46.2 75.9
3 1986 46.9 81.3
4 1988 38 43.5
5 1989 32.3 43.3
6 1990 56.8 68.8
7 1991 42.9 65.5
8 1992 53 53.2
9 1993 44.7 44
10 1994 47.1 45
11 1995 46.4 65.5
12 1996 67.9 98.4
13 1997 34.3 51
14 1998 48 83
15 1999 36.5 50
12
 16 2000 47.1 144
17 2001 68.4 39.8
18 2002 43.7 50.8
19 2003 51.3 40.8
20 2004 51.3 59.7
21 2005 46.7 71.4
22 2006 40.4 44
23 2007 57.4 42
24 42

2.3 Data Processing and Quality Checking

2.3.1 Data Consistency Test

The daily heaviest rainfall data of Motta methodological station from 1984 to 2007 is used for
the design. Hence 24years heaviest rain fall data is available. These data should be checked for
its consistency and adequacy Consistency If the conditions relevant to the recording of a rain
gauge station have undergone a significant change during the period of record, inconsistency
would arise in the rainfall data of that station. Inconsistency of the records can be checked by
using a double mass curve (DMC) technique and this method is mostly applicable when there are
number of neighboring data stations. But there is only one gauging station and it is to check the
consistency of the data. Some of the common causes for inconsistency of the records are:

 Shifting of rain gauge station to a new location

 The neighboring hood of the station undergoing a marked change
 Change in the ecosystem due to calamities, such as forest fires, land slide and
 Occurrence of observational error from a certain date

Adequacy refers to the length of record, but scarcity of data collecting stations is often a
problem. The observed recorded is merely a sample of the total population of floods that have
occurred and may occur again. If the sample is too small the probability derived cannot be
expected to be reliable and mostly it is good to have records more than thirty years. Before
proceeding to the other analysis, the adequacy of rainfall data series should be checked and it
should be realized. The data series could be considered and adequate if relative standard

Table 2.3 Maximum daily rainfall Values of 24 years for Motta Station

13
 No. Year Max. Rank Descending Cumulativ R - Ravg (R - (R -
RF e Ravg)2 Ravg)3
1 1984 34 1 68.4 68.4 -11.60 134.463 -1559.215
2 1985 46.2 2 67.9 136.3 0.60 0.365 0.221
3 1986 46.9 3 57.4 193.7 1.30 1.701 2.218
4 1987 13 4 56.8 250.5 -32.60 1062.488 -34632.693
5 1988 38 5 53 303.5 -7.60 57.697 -438.254
6 1989 32.3 6 51.3 354.8 -13.30 176.779 -2350.427
7 1990 56.8 7 51.3 406.1 11.20 125.533 1406.497
8 1991 42.9 8 48 454.1 -2.70 7.268 -19.592
9 1992 53 9 47.1 501.2 7.40 54.822 405.909
10 1993 44.7 10 47.1 548.3 -0.90 0.803 -0.719
11 1994 47.1 11 46.9 595.2 1.50 2.263 3.403
12 1995 46.4 12 46.7 641.9 0.80 0.647 0.520
13 1996 67.9 13 46.4 688.3 22.30 497.476 11095.784
14 1997 34.3 14 46.2 734.5 -11.30 127.596 -1441.301
15 1998 48 15 44.7 779.2 2.40 5.780 13.896
16 1999 36.5 16 43.7 822.9 -9.10 82.734 -752.536
17 2000 47.1 17 42.9 865.8 1.50 2.263 3.403
18 2001 68.4 18 40.4 906.2 22.80 520.030 11858.851
19 2002 43.7 19 38 944.2 -1.90 3.594 -6.814
20 2003 51.3 20 36.5 980.7 5.70 32.538 185.599
21 2004 51.3 21 34.3 1015 5.70 32.538 185.599
22 2005 46.7 22 34 1049 1.10 1.219 1.346
23 2006 40.4 23 32.3 1081.3 -5.20 26.997 -140.270
24 2007 57.4 24 13 1094.3 11.80 139.338 1644.773
Sum 1094.3 1094.3 3096.930 -14533.801
Mean 45.596 45.596
Std 11.604 11.604
Skewness -0.388

Note: the above table was arranged by excel calculation.

R
Ravg = = 45.596 N =24
N
Std =
√
(R−Ravg)2
N −1
= 11.604

Data reliability checking

The data series should be consistent and adequate if relative standard error (σ<10%), where σ is
relative standard error.
Sy
σ = se/xm, se=
√N

14
 11.604
se= =¿ 2.369
√ 24
2.369
Then, =¿ 5.195% <10%
45.569
Hence, the recorded data for 24 years is reliable and adequate.

Test For Out Lairs

This test helps to avoid those data lie out of the range in between the lowest datum and the
highest datum .The lowest datum and the highest datum are calculated as follows.
Let Y=Log X;
Lowest datum=RL=10^ (YL)
Highest datum =RH=10^ (YH)
Where YL=Yavg-KN*Syn-1
YH = Yavg + KN*Syn-1
Where; Yavg – mean of the data
Syn-1 - standard deviation of the data
KN - factor from Corresponding to number of year data.
Yavg = 1.64
Sn-1 =0.14
KN=2.408
N=24
YL= 1.64 - 2.408*0.14
YL =1.30288
Therefore x = 101.326 = 20.085
Low outlier threshold, RL= 20.085mm
The actual recorded lower value 13mm<20.085mm, it is not ok!
Therefore the smallest datum is 13 mm. Since there is no any data lower than this, all the
available data can be if they satisfy condition in next step.
YH=1.64+2.408*0.14
=1.977
High out lier threshold, RH=10^1.954=94.842 mm
The actual recorded lower value 68.4mm<94.842 mm, it is ok!
15
Therefore; the highest datum is 94.842mm. Since there is no any data higher than this, all the
available data can satisfy our condition.

Table 2.3. Revised data of daily heaviest rainfall (after outlier tests) for Motta station

Max Descendin (R- (R-

Year Rank cumulate R-Ravg Y=LogX
RF g Ravg)^2 Ravg)^3
1 34 1 68.4 68.4 -13.01 169.339 -2203.62 1.531
2 46.2 2 67.9 136.3 -0.813 0.661 -0.537 1.665
3 46.9 3 57.4 193.7 -0.113 0.013 -0.001 1.671
4 38 4 56.8 250.5 -9.013 81.235 -732.174 1.580
5 32.3 5 53 303.5 -14.713 216.474 -3184.99 1.509
6 56.8 6 51.3 354.8 9.787 95.785 937.439 1.754
7 42.9 7 51.3 406.1 -4.113 16.917 -69.581 1.632
8 53 8 48 454.1 5.987 35.844 214.594 1.724
9 44.7 9 47.1 501.2 -2.313 5.350 -12.375 1.650
10 47.1 10 47.1 548.3 0.087 0.008 0.001 1.673
11 46.4 11 46.9 595.2 -0.613 0.376 -0.230 1.667
12 67.9 12 46.7 641.9 20.887 436.265 9112.247 1.832
13 34.3 13 46.4 688.3 -12.713 161.621 -2054.70 1.535
14 48 14 46.2 734.5 0.987 0.974 0.961 1.681

15 36.5 15 44.7 779.2 -10.513 110.524 -1161.94 1.562

16 47.1 16 43.7 822.9 0.087 0.008 0.001 1.673

16
 17 68.4 17 42.9 865.8 21.387 457.402 9782.435 1.835
18 43.7 18 40.4 906.2 -3.313 10.976 -36.365 1.640
19 51.3 19 38 944.2 4.287 18.378 78.786 1.710
20 51.3 20 36.5 980.7 4.287 18.378 78.786 1.710
21 46.7 21 34.3 1015 -0.313 0.098 -0.031 1.669
22 40.4 22 34 1049 -6.613 43.732 -289.204 1.606
23 57.4 23 32.3 1081.3 10.387 107.889 1120.637 1.759
Sum 1081.3 1081.3 38.271
Mean 47.013 47.013 1.6640
Stnd 9.298 9.298 0.0843
Skewness 0.671 0.671 0.1326

R
Ravg = = 47.013 N =23
N
Std =
√
(R−Ravg)2
N −1
= 9.298

Data reliability checking

The data series should be consistent and adequate if relative standard error (σ<10%), where σ is
relative standard error.
Sy
σ = se/xm, se=
√N
9.298
se= =¿1.939
√23
1.939
Then, =¿ 4.135% <10%
47.013
Hence, the recorded data for 23 years is reliable and adequate.

Test For Out Lairs

This test helps to avoid those data lie out of the range in between the lowest datum and the
highest datum .The lowest datum and the highest datum are calculated as follows.
Let Y=Log X;
Lowest datum=RL=10^ (YL)

17
 Highest datum =RH=10^ (YH)
Where YL=Yavg-KN*Syn-1
YH = Yavg + KN*Syn-1
Where; Yavg – mean of the data
Syn-1 - standard deviation of the data
KN - factor from Corresponding to number of year data.
Yavg =1.664
Sn-1 =0.0843
KN=2.448
N=23
YL= 1.664 - 2.448*0.0843
YL =1.458
Therefore x = 101.459 = 28.7
Low outlier threshold, RL=28.7mm
The actual recorded lower value 28.7mm<32.3mm, it is ok!
YH=1.664+2.448*0.0843
=1.87
High out lier threshold, RH=10^1.87=74.1mm
The actual recorded lower value 68.4mm<74.1mm, it is ok!

Therefore; the highest datum is 74.1mm. Since there is no any data higher than this, all the
available data can satisfy our condition.

To analyze the maximum discharge expected in T years we can use the frequency distribution
function listed below, but the data in hand may fit to only one of them. Therefore, before
employing the methods it have to be checked for the fittest one. Among the distribution systems,
the following are the commonly known:

1. Normal distribution method

2. Gumbel’s Method

3. Log Pearson type Three distribution Method

4. Log Normal distribution Method

18
The suitability of the methodology for different countries is different. Example, Germany log
Pearson type three, UK GEV and USA Log Pearson type three. However, in case of Ethiopia no
institute proposed a certain methodology. The sample statistics of data distribution should be
tasted for goodness of fit criteria as satisfactory basis for selection.

Taking return period of 50 years the design rainfall is determined by the following methods was
tested for Motta station, because it fits higher value of outliers but not the lower.

1. Normal Distribution Method

RT = Ravg + KT*SD

Where: RT = Annual Maximum rainfall T years return Period.

Ravg = Mean rain fall data

KT = Frequency factors

SD = Standard deviations

( )
2
2.51557+ 0.80285 w+0.01033 w
KT =W − 2 3
1+ 1.143279 w+0.1992 w + 0.00135 w

W =¿, p =1/p =1/50 =0.02

W = 2.797

KT = 2.348

RT= Ravg + KT*SD

RT = 47.013+2.348* 9.298

RT = 68.844

2. Gumbel Distribution’s method

The distribution is applicable to extreme hydro logic events such as maximum daily rainfall, rain
intensity and peak flood flows and expressed by an equation;

RT = Ravg + KT*SD ------------------------------------------ (-***)

Where Ravg =mean of the annual maximum daily rainfall

19
 RT= Annual maximum rainfall of T years return period (design storm)

KT= Frequency factor expressed as;

Yt −Yn
KT = ----------------------------------------------------- (**)
Sn

Yt be a reduced variant, a function of T and is given by

[
Y t=−ln −ln 1− ( 1
T )] -------------------------------------------------------------- (*)
Where: Yn= reduced mean, it is a function of sample size.

Sn = reduced standard deviation which is also a function of sample size.

SD = Standard deviation of sample size.

SD =
√ (R−Ravg)2
N −1

To determine the distribution parameters when it is applied to asset of data distribution the
following steps are followed.

1. Assemble the maximum daily rainfalls data and note the sample size N. Here the daily
rainfall data is the variety R. Find Ravg and SD.

2. Using table and determine Yn and Sn appropriate to given N.

3. Find Yt for a given T by equation (*)

4. Find KT by equation(**)

5. Determine the required RT by equation (***)

R
Ravg = = 47.013
N

SD =
√ (R−Ravg)2
N −1
=9.298

Y t=−ln −ln 1−
[ ( )] 1
50
= 3.901

For N = 23
20
Yn = 0.5283

Sn = 1.0811

3.9−0.5283
KT = = 3.121
1.0811

RT = Ravg + KT*SD

= 47.013+3.121* 9.298

= 76.027

3. Log Pearson Type Three Distribution Method

This distribution is extensively used in USA for project sponsored by U.S government. In this
the variant is first transformed into logarithmic form (base 10) and the transformed data is then
analyzed. If R is the variant of random hydrologic series of Z variants

Where Z = logR ---------------------------------------------- (*)

Are first obtained. For this Z series, for any recurrence interval T

ZT=Zavg + kz*SD----------------------------------------------- (**)

Where kz =frequency factor which is a function of recurrence interval T and the coefficient of
skew Cs.

SD = Standard deviation of the Variant sample.

SD =
√ (Z −Zavg)2
N −1

And, Cs = coefficient of skew of variant Z.

3
N (Z−Zavg)
Cs =
( N −1)(N −2)SD 3

N = sample size = number of year of recorded

21
The variation of kz =F (Cs, T) is given in table. After finding for ZT by equation (**), the
corresponding value of RT is obtained by equation (*) as:

RT = R50 = antilog Z50

SD =
√ (Z −Zavg)2
N −1
= 0.0843

3
N (Z−Zavg)
Cs =
( N −1)(N −2)SD 3

Cs = 0.1326

Kz = 2.125

ZT=Zavg + kz*sd

=1.664+ (2.125)*0.0843

Z50 = 1.8432

R50 = antilog Z50

R50 = 69.69mm

4. Log normal Distribution Method

Log normal distribution method is especial type of log Pearson type three distribution method
with Cs = 0

From table for Cs = 0, T50, kz = 2.054

Z50=Zavg + kz*SD

=1.664+2.054*0.0843

Z50 =1.837

R50 = Antilog Z50 = 68.74 mm

Table 2.6 Summary of result for 50 years return period

Methods Maximum Storm (P,mm) Calculated

22
 (For Return period of 50 years)

Log Pearson Type Three Method 69.69

Normal Distribution Method 68.844

Log normal Distribution Method 68.74

Gumbel Method 76.027

2.3.2. D-Index Calculation for Motta Station

D-index calculation used to select the best distribution method for the better goodness of the
given data. Hence, in this study it was used to determine the best statistical distribution to
estimate the peak rainfall by using some highest value. Therefore, the value having minimum d-
index will be the best fit to the distribution.

D-index = XI-XT or (ABS XI-XT)

Table 6 statistical parameter

Statistical parameter Rainfall data x y= log x

mean/µ/ 47.013 1.664
Std v/αn-1/ 9,298 0.0843
Skew /g/ 0.671 0.1326
Number of data 23
Weibull ranked Probability plotting position P=m/(n+1)

P=1/50 = 0.02

m= 0.02(24+1) =0.5

From the rainfall data, top values

2001: 68.4mm (Rank 1)

1996: 67.9mm (Rank 2)

2007: 57.4mm (Rank3)

23
 1990: 56.8mm (Rank 4)

1992: 53mm (Rank 5)

2004: 51.3mm (Rank 6)

Table 7 D-index and design rainfall for each method for Motta

Log person type

Normal log normal Gumbel
III
RANK XI XI-XT XI-XT XI-XT XI-XT
1 68.4 5.283 3.280 3.826 1.208
2 67.9 8.026 7.555 7.555 4.442
3 57.4 0.309 0.277 0.277 2.374
4 56.8 0.793 1.137 1.137 0.296
5 53 1.563 1.008 1.008 1.964
6 51.3 1.981 1.282 1.282 1.873
Sum 17.956 14.540 15.086 12.157
sum/mean 0.382 0.309 0.321 0.259
Design
68.844 69.692 68.744 76.027
Rain fall
Here, even though Gumbel method gives the best-fit D-Index value,

Table 2.4 Maximum daily rainfall Values of 24 years for Felege Birhan Station

Max. R- (R -
(R - Ravg)3
No. Year RF Rank Descending Cumulative Ravg Ravg)2
1 1984 61.6 1 144 144 0.58 0.335 0.194
2 1985 75.9 2 98.4 242.4 0.58 0.335 0.194
3 1986 81.3 3 83 325.4 14.88 221.390 3294.093
4 1987 43.5 4 81.3 406.7 20.28 411.245 8339.698
5 1988 43.3 5 75.9 482.6 -17.52 306.980 -5378.538
6 1989 68.8 6 71.4 554 -17.72 314.028 -5564.837
7 1990 65.5 7 68.8 622.8 7.78 60.515 470.760
8 1991 53.2 8 65.5 688.3 4.48 20.063 89.865
9 1992 44 9 65.5 753.8 -7.82 61.165 -478.365
10 1993 45 10 61.6 815.4 -17.02 289.709 -4931.085
11 1994 65.5 11 59.7 875.1 -16.02 256.667 -4112.021
12 1995 98.4 12 53.2 928.3 4.48 20.063 89.865
13 1996 51 13 51 979.3 37.38 1397.202 52226.250
14 1997 83 14 50.8 1030.1 -10.02 100.417 -1006.263
15 1998 50 15 50 1080.1 21.98 483.084 10617.779
16 1999 144 16 45 1125.1 -11.02 121.459 -1338.577
17 2000 39.8 17 44 1169.1 82.98 6885.542 571356.546
24
 18 2001 50.8 18 44 1213.1 -21.22 450.324 -9556.246
19 2002 40.8 19 43.5 1256.6 -10.22 104.465 -1067.724
20 2003 59.7 20 43.3 1299.9 -20.22 408.882 -8267.937
21 2004 71.4 21 42 1341.9 -1.32 1.745 -2.304
22 2005 44 22 42 1383.9 10.38 107.727 1118.118
23 2006 42 23 40.8 1424.7 -17.02 289.709 -4931.085
24 2007 42 24 39.8 1464.5 -19.02 361.792 -6881.587
Sum 1464.5 1464.5 12674.843 594086.794
Mean 61.021 61.021
Std 23.475 23.475
Skewness 1.913

Note: the above table was arranged by excel calculation.

R
Ravg = = 61.021 N =24
N
Std =
√
(R−Ravg)2
N −1
= 23.475

Data reliability checking

The data series should be consistent and adequate if relative standard error (σ<10%), where σ is
relative standard error.
Sy
σ = se/xm, se=
√N
23.475
se= =¿4.792
√24
4.762
Then, =¿7.853% <10%
61.021
Hence, the recorded data for 24 years is reliable and adequate.

Test For Out Lairs

This test helps to avoid those data lie out of the range in between the lowest datum and the
highest datum .The lowest datum and the highest datum are calculated as follows.
Let Y=Log X;
Lowest datum=RL=10^ (YL)
Highest datum =RH=10^ (YH)
25
 Where YL=Yavg-KN*Syn-1
YH = Yavg + KN*Syn-1
Where; Yavg – mean of the data
Syn-1 - standard deviation of the data
KN - factor from Corresponding to number of year data.
Yavg = 1.761
Sn-1 = 0.142
KN=2.408
N=24
YL= 1.761 - 2.408*0.142
YL =1.42388
Therefore x = 101.447 = 26.538
Low outlier threshold, RL= 26.538mm
The actual recorded lower value 39.8mm>26.538mm, it is ok!
Therefore the smallest datum is 39.8mm. Since there is no any data lower than this, all the
available data can be if they satisfy condition in next step.
YH=1.761+2.408*0.142
=2.098
High out lier threshold, RH=10^2.098=125.314 mm
The actual recorded highest value 144mm< 125.314 mm, it is ok!

Therefore; the highest datum is 144mm. Since there is no any data higher than this, all the
available data can satisfy our condition.

To analyze the maximum discharge expected in T years we can use the frequency distribution
function listed below, but the data in hand may fit to only one of them. Therefore, before
employing the methods it have to be checked for the fittest one. Among the distribution systems,
the following are the commonly known:

1. Normal distribution method

2. Gumbel’s Method
3. Log Pearson type Three distribution Method

26
 4. Log Normal distribution Method
The suitability of the methodology for different countries is different. Example, Germany log
Pearson type three, UK GEV and USA Log Pearson type three. However, in case of Ethiopia no
institute proposed a certain methodology. The sample statistics of data distribution should be
tasted for goodness of fit criteria as satisfactory basis for selection.

Taking return period of 50 years the design rainfall is determined by the following methods was
tested for Felege berhan, because it fits the lower and higher value of outliers.
1. Normal Distribution Method
RT = Ravg + KT*SD
Where: RT = Annual Maximum rainfall T years return Period.
Ravg = Mean rain fall data
KT = Frequency factors
SD = Standard deviations

( )
2
2.51557+ 0.80285 w+0.01033 w
KT =W − 2 3
1+ 1.143279 w+0.1992 w + 0.00135 w
W =¿, p =1/p =1/50 =0.02
W = 2.80
KT = 2.05
X50=Xmean + KT*σx = 57.41+2.05*16.31
RT = 90.91mm
2. Gumbel Distribution’s method
The distribution is applicable to extreme hydrologic events such as maximum daily rainfall, rain
intensity and peak flood flows and expressed by an equation;
RT = Ravg + KT*SD ------------------------------------------ (-***)
Where Ravg =mean of the annual maximum daily rainfall
RT= Annual maximum rainfall of T years return period (design storm)
KT= Frequency factor expressed as;
Yt −Yn
KT = ----------------------------------------------------- (**)
Sn
Yt be a reduced variant, a function of T and is given by
27
 [ ( )]
Y t=−ln −ln 1−
1
T
-------------------------------------------------------------- (*)

Where: Yn= reduced mean, it is a function of sample size.

Sn = reduced standard deviation which is also a function of sample size.
SD = Standard deviation of sample size.

SD =
√ (R−Ravg)2
N −1
To determine the distribution parameters when it is applied to asset of data distribution the
following steps are followed.
6. Assemble the maximum daily rainfalls data and note the sample size N. Here the daily
rainfall data is the variety R. Find Ravg and SD.
7. Using table and determine Yn and Sn appropriate to given N.
8. Find Yt for a given T by equation (*)
9. Find KT by equation(**)
10. Determine the required RT by equation (***)
R
Ravg = = 61.021
N

SD =
√ (R−Ravg)2
N −1
= 23.475

[
Y t=−ln −ln 1− ( 1
50 )]
= 3.901

For N = 24 [by interpolation]

Yn = 0.53
Sn = 1.08
Frequency factor, KT= (3.9-0.53)/1.08 = 3.12
Y=Xmean + Z*XStandared deviation =57.41+3.12*16.31 = 108.31mm
3. Log Pearson Type Three Distribution Method
This distribution is extensively used in USA for project sponsored by U.S government. In this
the variant is first transformed into logarithmic form (base 10) and the transformed data is then
analyzed. If R is the variant of random hydrologic series of Z variants
28
Where Z = logR ---------------------------------------------- (*)
Are first obtained. For this Z series, for any recurrence interval T
ZT=Zavg + kz*SD----------------------------------------------- (**)
Where kz =frequency factor which is a function of recurrence interval T and the coefficient of
skew Cs.
SD = Standard deviation of the Variant sample.

SD =
√ (Z −Zavg)2
N −1
And, Cs = coefficient of skew of variant Z.
3
N (Z−Zavg)
Cs =
( N −1)(N −2)SD 3
N = sample size = number of year of recorded
The variation of kz =F (Cs, T) is given in table. After finding for ZT by equation (**), the
corresponding value of RT is obtained by equation (*) as:
RT = R50 = antilog Z50

Table 2.5 Calculation of design storm by log Pearson type III method.
Year Max. RF Z=logR (Z-Zavg)2 (Z-Zavg)3
1984 61.6 1.790 0.001 0.000
1985 75.9 1.880 0.014 0.002
1986 81.3 1.910 0.022 0.003
1987 43.5 1.638 0.015 -0.002
1988 43.3 1.636 0.015 -0.002
1989 68.8 1.838 0.006 0.000
1990 65.5 1.816 0.003 0.000
1991 53.2 1.726 0.001 0.000
1992 44 1.643 0.014 -0.002
1993 45 1.653 0.012 -0.001
1994 65.5 1.816 0.003 0.000
1995 98.4 1.993 0.054 0.013
1996 51 1.708 0.003 0.000
1997 83 1.919 0.025 0.004
1998 50 1.699 0.004 0.000
1999 144 2.158 0.158 0.063
29
 2000 39.8 1.600 0.026 -0.004
2001 50.8 1.706 0.003 0.000
2002 40.8 1.611 0.023 -0.003
2003 59.7 1.776 0.000 0.000
2004 71.4 1.854 0.009 0.001
2005 44 1.643 0.014 -0.002
2006 42 1.623 0.019 -0.003
2007 42 1.623 0.019 -0.003
Sum 1464.5 42.261 0.462 0.064
Mean 61.021 1.761 0.019 0.003
Std 0.142
Cs 1.072

Design Period, T=50

Probability, P=1/T=1/50=0.02
K= (Cs,y/6)=0.54/6=0.09
W= (Ln (1/P2)) 0.5 = (ln (1/0.022))0.5=2.80
Frequency Factor,
KT= (w ((2.515517+0.802853*w+0.010328*w2)/(1+1.432788*w+0.189269*w2+0.001308*w3)))
= 2.05
Standard Normal Variance,
Z=KT+(KT2-1)*K+1/3*(KT3-6*KT)*K2-(KT2-1)*K3+KT*K4+1/3*K5=2.33
Y=Ymean + Z*Std = 1.74+2.33*0.12 =2.01
Design Rainfall, X50 = 10Y = 102.01
RT= 103.41mm
4. Log normal Distribution Method
Log normal distribution method is especial type of log Pearson type three distribution method
with Design Period, T=50.00
Probability, P=1/50=0.02
K=(Cs/6)=0/6=0.000
W= (Ln (1/P2)) 0.5 = (ln (1/0.022)) 0.5 =2.80
Frequency Factor,
KT= (w ((2.515517+0.802853*w+0.010328*w2)/(1+1.432788*w+0.189269*w2+0.001308*w3)))
=2.05

30
Standared Normal Variance,
Z=KT+ (KT2-1)*K+1/3*(KT3-6*KT)*K2-(KT2-1)*K3+KT*K4+1/3*K5=2.05
Y=Ymean + Z*σy =1.74+2.05*0.12=1.982
Design Rainfall, X50 = 10Y = 101.982 = 96.01mm
Table 2.6 Summary of result for 50 years return period
Methods Maximum Storm P(mm) Calculated
(For Return period of 50 years)

Normal 90.91
Log Pearson Type III 103.41
Log Normal 96.01
Gumbel 108.31

2.3.3. D-Index Calculation Felege Birhan station

D-index calculation used to select the best distribution method for the better goodness of the
given data. Hence, in this study it was used to determine the best statistical distribution to
estimate the peak rainfall by using some highest value. Therefore, the value having minimum d-
index will be the best fit to the distribution.

D-index = XI-XT or( ABS XI-XT)

1. Normal
PR= (Rank)/ (N+1)

W= (LN (1/PR^2)) ^0.5

KT=(W-((2.515517+0.802853*W+0.010328*W^2)/
(1+1.432788*W+0.189269*W^2+0.001308*W^3)))

XI’=Xmean +x*KT

XI-XI’ =ABS (XI-XI’)

2. Log normal
PR= (Rank)/ (N+1)

W= (LN (1/PR^2)) ^0.5

31
KT=(W-((2.515517+0.802853*W+0.010328*W^2)/
(1+1.432788*W+0.189269*W^2+0.001308*W^3)))

Y=Ymean +y*KT

XI’=10^Y

XI-XI’ =ABS (XI-XI’)

3. Log person Type III

PR= (Rank)/ (N+1)

W= (LN (1/PR^2)) ^0.5

K=CS, Y/6

Z=(W-((2.515517+0.802853*W+0.010328*W^2)/
(1+1.432788*W+0.189269*W^2+0.001308*W^3)))

KT=Z+ (Z^2-1)*K+1/3*(Z^3-6*Z)*K^2-(Z^2-1)*K^3+Z*K^4+1/3*K^5

Y=Ymean +y*KT

XI’=10^Y

XI-XI’ =ABS (XI-XI’)

4. Gumbel
PR= (Rank)/ (N+1)

YT=-LN (LN ((1/PR)/ ((1/PR)-1)))

KT= (YT-Yn)/Sn

XI’=Xmean +x*KT

XI-XI’ =ABS (XI-XI’)

Rank XI Normal Log Pearson Log Gumbel

Type III Normal
XI -'XI' XI -'XI' XI -'XI' XI -'XI'
1 98.40 12.74 6.36 10.32 1.33
2 83.00 3.03 1.33 2.77 3.28
3 81.30 5.12 5.63 5.92 1.48
4 75.90 2.71 4.48 4.12 0.78

32
 5 71.40 0.74 3.31 2.54 0.02
6 68.80 0.39 3.47 2.44 0.56
Sum 24.74 24.59 28.12 7.46
Sum/Mean 0.43 0.43 0.49 0.122
Point Rainfall 90.91 103.41 96.01 108.31
Design Point 108.31
Rainfall =

Peak Design Rainfall

We adopted the design storm (maximum daily rainfall) obtained by Gumbel Method (108.31mm,
because it gives the best-fit D-Index value. Taking the conservative value of flood for design of
structure will make the structure safe and stable.
Design P=108.31mm.

Area Estimation

Rain gauges represent only point measurements. In practice however, hydrological analysis
requires knowledge of the precipitation over an area.

Several approaches have been devised for estimating areal precipitation from point
measurements.

From Catchment area of Motta=10.582km2 and for Felege berhan =12.533km2 this means it is
not contribution for weir site or outlet point.

The Thiessen polygon

The Thiessen polygon method gives more accurate estimation than the simple arithmetic mean
Estimation as the method introduces a weighting factor on rational basis.
Furthermore, rain gauge stations outside the catchment area can be considered effectively
By this method.
P 1∗A 1+ P 2∗A 2
Areal rainfall=
A 1+ A 2
Where A1=10.582km2 and A2=12.533km2
P1=76.027mm and P2=108.31mm
Areal Rainfall = 93.53mm

33
Therefore Design P=93.53mm.

2.4 Design Discharge Determination

2.4.1 USSCS (United States Soil Conservation Service)

This method of hydrograph synthesis used by united states; under a department of agriculture
and soil conservation service originated from conservation that a hydrograph could be represent
in a simple geometric form as a triangle.
The design flood, which is expected ton, occur during period of the diversion scheme, is
therefore determined by USSCS method to this end the storm that is estimated by Gumbel is
adopted.
Table 2.7 USSCS Table for Calculation of Design Discharge.
Step designation/formula symbol unit value

1 Area of catchments (this can A km2 23.12

be determined from 1:50000
scale topographic maps or
Arial photo graphs )

2 3 4 5
Elevation(m) Difference (m) Distance (km) Difference Tc (hr)
Length(m)
H = 13300.00
H1 =0.00 0 0 0

H2 =350.00 350 1400 1.4 0.15

H3 =560.00 560 8300 8.3 0.96

H4 =122.00 122 3600 3.6 0.65

6 Time of concentration Tc hr 1.76

Tc=0.948*(L3/H)0.385
7 Rain fall excess duration D hr 0.30
D=Tc/6 ; if Tc<3hrs
D=1hr.if Tc>3hrs
8 Time to peak Tp hr 1.20
34
 Tp = 0.5D+0.6Tc
9 Time base of hydrograph Tb hr 3.22
Tb=2.67Tp
10 Lag time TL=0.6Tc TL hr 1.05
11 Peak rate of discharge m3/smm 4.03
created by 1mm run off qp
excess of whole of the
catchments qp =
(0.21A)/Tp

12 13 14 15 16 17 18 19
Duration Daily Point Rainfall Rainfall Areal to Areal Incremental decreasi
Rainfall Profile Profile Point rainfall Rainfa rainfall ng order
(1) (2) 1*2 ratio (3) ll
1*2*3
Hr mm % mm % mm mm Number
0-0.30 93.53 25 23.38 55.13 12.89 44.46 1
0.30-0.60 33 30.86 70.75 21.84 40.03 2
0.60-0.90 43 40.22 76.75 30.87 35.71 3
0.9-1.20 48 44.89 79.55 35.71 30.87 4
1.20-1.50 53 49.57 80.75 40.03 21.84 5
1.50-1.80 58 54.25 81.95 44.46 12.89 6

12 Fill 0-Dhr, D-2Dhr, …5D-6Dhr

13 Determine the magnitude of the daily rain fall with the given recurrence interval by applying
statistical method
13Determine the magnitude of the daily rainfall with the given recurrence interval by applying
statistical method.

35
14 Read from given Annex ----fig---, the rain fall profile
(%) occurring in D, 2D, 3D, 4D, 5D, 6Dhrs and put in 14.
15 Multiply col.13 and col.14 to find the rainfall profile (mm) enter in 15.
16 Read from table ----area to point rainfall ratio for different duration in particular catchments.
17 Multiply col.15 and col.16
18 Calculate incremental rainfall by deducting the current Arial rainfall from the preceding Arial
rainfall as written in 18.
19 Assign order to the rainfall depths in descending order 1-6
20 21 22 23 24 25
Rearr Rearranged the Cumulative Time of incremental hydrograph
anged incremental rainfall
order rainfall
No mm mm Time of Time to peak(hr) Time to end
beginning(hr) (hr)
6 12.89 12.89 0.30 1.50 3.52
5 21.84 34.73 0.60 1.80 3.82
4 30.87 65.59 0.90 2.10 4.12
3 35.71 101.31 1.20 2.40 4.42
2 40.03 141.34 1.50 2.70 4.72
1 44.46 185.79 1.80 3.00 5.02

20 From 19 mention the rearranged order as6,5,4,3,2,1 (arbitrary but

considering ascending and descending feature of hydrograph ordinate
where peak value is middle of the hydrograph).
21 Fill in the corresponding incremental rain fall value to the rearranged
order of 20 from 17.
22 Fill in the cumulative rainfall value of 21 by adding with the rainfall
value in preceding duration.
23 Fill in the time of beginning of the hydrograph 0,D,2D…,5Dhr

36
24 Fill in the time peak as Tp,D+Tp,2D+Tp,…,5D+Tp or add Tp in every
value of 23 and mention in 24.
25 Fill in the time of end as Te, D + Tb, 2D+Tb,…5D+Tb
26 CN = 90.38 (Given depending on the landuse characterictics)

No Description/Formula Symbol Unit Example

27 Find the maximum potential deference b/n rainfall (P) and S m CN=90.38
direct run off (Q), which is given by the following formula. S=27.04
25400
 254
CN
S=
CN= value of corresponding to AMC -III
28 Substituting the value of “S” in the following formula, giving the relation b/n direct run
off (Q) and rainfall (P).
( P  0.2 S )^2
Q= ( P  0.8S )

29
Substituting the value of P1 30 31
as mentioned in col. 20, in P(mm) Q(mm)
the above formula and fined 12.89 1.62
find the corresponding value 21.84 6.21
of Q (28) enter; Enter the
30.87 12.35
value of Q in col. 31.
35.71 16.01
40.03 19.44
44.46 23.07

37
32 33 34 35 36 37

Duration Value Incremental Peak Time of Time to Time to Composite

of Q run off runoff for beginnin peak end hydrograph
increment g
Col.(24) Col.(25)
Col.(23)

hr mm mm m3/s hr hr hr

0-0.30 1.62 1.62 6.54 0.30 1.50 3.52

0.30-0.60 6.21 4.59 18.50 0.60 1.80 3.82
0.60-0.90 12.35 6.14 24.74 0.90 2.10 4.12
0.9-1.20 16.01 3.66 14.78 1.20 2.40 4.42
1.20-1.50 19.44 3.43 13.82 1.50 2.70 4.72
1.50-1.80 23.07 3.64 14.66 1.80 3.00 5.02

32 Enter the same time as in col.12, 0-D,D-2D,2D-3D,…,5D-6D.

35 There are the value of Qas found out in col.31 corresponding to the value of P
36 F incremental runoff by reducing the value of col;33 by preceding value.
37 Multiply col. 34 and peak rate of run off corresponding to 1mm run off excess as
found incol.11
38 Plot triangular hydrograph with time of beginning, peak time and time to end as
mentioned in 23,24,25 and peak run off as mentioned in col.36
39 Plot composite hydrograph by adding all the triangular hydrographs .The resultant
hydrograph will be composite hydrograph of desired return period. The coordinate of
the peak of hydrograph will give the peak run off with desired return period.

38
Hydrographs of hydrology
Time Ordinate Of Hydrograph (m3/s)
Hr 1 2 3 4 HT
0
0.33 0 0.33
0.66 1.9 0 2.56
0.99 4 10 0 14.99
1.32 6.5 20.5 34.2 0 62.52
1.65 8.5 31 65 18 124.15
1.98 7.9 42.25 98.5 37.92 188.55
2.31 6.5 36.5 128.82 55.5 222.50
2.64 5 30 110 75.65 200.29
2.97 3.5 23.5 87.75 63.8 181.52
3.3 2.5 12.6 69 53.85 141.25
3.63 2 11.25 52.25 43.25 112.38
3.96 0 4.5 33 32.25 73.71
4.29 0 14.9 21 40.19
4.62 0 11.25 15.87
4.95 0 4.95
5.28

39
 Composite of Hydrograph
250

200
Peak Runoff m3/s

H1
150
H2
H3
100
H4
50 HT

0
0 1 2 3 4 5 6

Duration

Figure 2: Composite of Hydrograph

2.7. Flood mark Method

During field assessment and topographic survey, the flood characteristics have been requested
from the local aged farmers and it has been concluded the flow would not overflow above bank
of the river. This is just to check the design flood what we have determined using rough
simulation methods such as SCS for ungagged catchments. That was the main purpose of taking
flood marks information during field assessment and that will be checked using the stage
discharge analysis in the following section. Tail water depth of the river is equal to the flood
depth and amount at the proposed weir site before construction of the weir. It is used to
crosscheck peak flood estimated by the SCS unit hydrograph method with flood mark method
and to see the flood feature after the hydraulic jump. During field visit, the flood mark of the
river at the proposed diversion site was decided to be inside the river bank based on dwellers
information and physical indicative marks. The data required to do stage discharge analysis has
been discussed in the following section.

40
 Cross section of weir
2500
2498
f(x) = − 0.777400281385997 ln(x) + 2497.7302001996
2496
Elevation

2494
2492
2490
2488
0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00

Cumulative Distance

Figure 2-5: The River Cross-Section for Sedie Weir Axis

a) Average river bed

Slope Average riverbed slope of a River is estimated by two different techniques. One is by end
area method and the other is by using best fit line method. Designers have adopted the end area
method output for further analysis. Average water surface slope is calculated to be, S = 0.0295
The water level of the river is taken at different points along the river channel around the head
work site. Surveying work done for 395.00m length. And then, average water surface slope is
considered as the river bed slope.

b) Manning’s Roughness coefficient for the river bed and banks (average)
The Manning’s roughness coefficient is taken from standard table based on the river nature. At
this site the river course is well defined, with fixed width, length and excellent form of river
shape The River banks are defined and relatively smooth. So, Manning’s roughness coefficient
(n = 0.04) is adopted.

c) Discharge of the river Input data:

Velocity V (m/sec.) =1/n*s0.5*R2/3;
Where, n=river bed material manning coefficient = 0.04, s = river bed material manning
coefficient = 0.0295 and R = Hydraulic radius

41
Discharge Q (m3/sec.) = WET AREA (m) * Velocity V (m/sec.) =A*V

Table 2-8: River discharge computation at different stages of flow

Hydrulic
Wet Total Top Wet Velocity Discharge Q
Elevation Depth Radius R
Area Perim Length Perim V(m/s) (m3/s)
(m)
2491.95 0 0 0 0 0 0 0 0
2492.25 0.3 1.47 16.1 8.03 8.06 0.18 1.369 11.033
2492.55 0.6 4.34 21.69 10.79 10.9 0.4 2.331 25.409
2492.85 0.9 7.9 25.71 12.75 12.96 0.61 3.088 40.026
2493.15 1.2 11.94 25.71 14.07 11.65 1.02 4.351 50.689
2493.45 1.5 16.34 31.03 15.27 15.76 1.04 4.408 69.464
2493.52 1.57 17.47 31.64 15.56 16.08 1.09 4.548 73.129
2493.75 1.8 21.1 33.59 16.48 17.11 1.23 4.929 84.341
2494.05 2.1 26.22 36.11 17.67 18.44 1.42 5.425 100.032
2494.35 2.4 31.7 38.63 18.86 19.77 1.6 5.874 116.128
2494.55 2.6 35.55 40.36 19.67 20.68 1.72 6.164 127.474
2494.95 3 43.81 44.38 21.61 22.78 1.92 6.633 151.103
2495.25 3.3 50.57 48.86 23.8 25.06 2.02 6.861 171.949
2495.55 3.6 58.13 54.64 26.65 27.99 2.08 6.997 195.838
2495.85 3.9 66.61 61.41 30 31.4 2.12 7.086 222.504
From the above stage, discharge table the maximum flood estimated from the SCS design flood
depth interpolated is 3.9m from the riverbed.

Stage Discharge Curve

2497
2496
2495
Elevation

2494
2493
2492
2491
2490
0 50 100 150 200 250
Discharge, m3/s

Figure 2-7: Stage Discharge Curve

42
2.8. Selected Design flood
The computed flood amount using SCS method above and hence design flood is calculate as
222.504m3/s.

Based on the stage discharge analysis of the project at d/s of the weir section or design discharge
capacity of the river channel before the construction or intervention, the downstream high (HFL)
flood level can be fixed by the selected designed flood in the previous section. Therefore, the
D/S HFL=riverbed elevation + depth of Tail-water (as obtained from SCS flood at a depth of
3.9m). D/S HFL=2491.95+3.9 = 2495.85m.a.s.l

3. HEAD WORK DESIGN

3.1 General
In selecting or comparing the best site, it was practically important to put criteria’s such as
considering the economic considerations, the availability of enough command and other
selection or comparison flood the channel has to accommodate, the length of the weir crest,
topographic suitability, the relative criteria’s. The headwork site is situated at 372962.68E and
1203919.51m N and riverbed elevation of 2491.95m above sea level. At this site the river course
is well defined, with fixed width, length and excellent form of river shape. The stream bed is
covered by Shallow alluvial deposits with bolder which comes from the mountains.

3.2 Headwork Type Selection

After detail observation of the available water balance, the target command area and the cost of
future idle canal length, it is decided to design a diversion weir with commanding elevation.
Weir type selected for the given height and flood amount is broad crest weir type which could be

43
subjected to the expected hydrostatic and hydrodynamic forces that can be generated from peak
flood. The weir is to be constructed with masonry material with a 200mm thickness RCC
capping.

Having decided upon the location of weir, the actual site is selected with the following
considerations:

 A reasonably wide and well-defined channel with reliable banks is favored

 The associated canal alignment should enable adequate command without excessive
excavation or embankment
 With respect to the adjoining land surface, the elevation of water surface upstream of the
weir should not be so low as to require an excessively high weir to divert the water at the
intake
 Easy arrangement of flow diversion during construction
 Availability of construction material at the nearest place
o Sustainable for environment effects
o The flow velocity should be controlled to protect river bed from erosion and
structures scouring
o The desired amount of water should be diverted for most of the time -Water level
fluctuations in front of the intake should be decreased
o Under optimum conditions the site should be located
o Neither very close to the irrigable area
o Nor very far -River bank (support) & river bed stability & water tightness

3.3. Hydraulic Design of the Structure

3.3.1. Weir Height Determination

The following major factors have been reasonably assessed during the field study and the final
topographic map whether the target area can be commanded at less weir height or not. Finally
using necessary parameters, the weir crest elevation have been determined as follows:

 Maximum command area elevation

 Deriving head of the intake structure

44
  Main canal slope
 Turn out and head regulator loss
 Lowest Point of river center

Table 3-1: Weir Height Determination

Description Value Unit

Command area level 2492.76 m
Canal Length 395.00 m
Manning n 0.02 -
Slope 0.001 m
Design bed width 0.50 m
Depth of canal flow 0.40 m
Freeboard 0.20 m
Loss across the field 0.05 m
Loss on turnout 0.05 m
Head Regulator 0.10 m
Outlet Sill level 2493.35 m
River bed elevation 2491.95 m

3.4 Selection of Weir Type

The selection of the most suitable type of headwork (weir) for a particular site under various
governing factors requires a lot of judgment& experiences, after collection of the secondary
design data for sedie river medium scale project headwork site and considering the basic
feasibility governing factors of project such as

 Economy
 Performance (function),
 Availability of construction material such as embankment materials at the selected site
 Soil type
 Simplicity for construction

There are the following common types of weirs

 Masonry weirs with vertical or slightly slopping U/S and D/S face Rock fill weirs
 Concrete weir with sloping glacis
Source (Irrigation and Hydraulic Engineering Structures Santos Kumar Garg1976)
Under Sedie river medium irrigation project the masonry weirs with masonry is selected.
45
 In deciding the type of the weir, the following conditions should be considered.
 Economy of construction
 Size of the project
 Physical feasibility
 Foundation condition
3.5 Design of Masonry Weirs
The design of the weir is the design of both structural and hydraulic design. Hydraulic design is
the determination of crest level and weir length. However, structural design includes empirical
dimension & checking of stability.

3.5.1 Hydraulic design of the weir height determination

 Design discharge Q peak = 222.504m3/sec
 Command area Elevation =2492.76m
 River bed level=2491.95m
 Proposed length of the canal=395.00m
 Head loss due to canal slope = Canal slope*length of canal=0.001*395m = 0.395
 Water head loss at the turnout = 0.05m
 Driving head for head regulator= 0.10m
 Total water head loss = 0.1+0.395+0.05= 0.545m
 Crest level of the weir = max command area level+ Total losses = 2492.76m +0.545m =
2493.305m
Weir height = Crest Level of the Weir – River Bed Level
=2493.305m -2491.95m = 3m

3.5.2. Weir Dimensioning

3.5.2.1. Weir Crest Length Determination
Flow over the Weir crest
a. Crest Length
 Length of the weir depends on the Physical features of the given site. The effect of the
weir length on the upstream watershed,
It should be adequate to pass the design flood safely and etc.
 Actual river section width from bank to bank = 32m
46
 Since the formation of the river reach at the selected weir axis is stable, so the Actual
river section width of the over flow section of the river is adopted for the design.
b. Discharge over the weir section
 From the hydrological analysis, for 50 years return period the design discharge is
222.504m3/sec

3.5.2.2. Top and Bottom Width Determination

Bligh’s formula can be used to approximately estimate the top and bottom width of a broad crest
weir. However, the dimension values obtained by using this formula should be checked
structurally in the stability section. Top width T in m,

h
T=
√ ρ−1
h+ He
Bottom width B in m, B=
√ ρ−1
Q= CLHe3/2
He¿ ¿= ¿ = 2.67m
2.67
Hence, Top width T = = 2.34m
√2.3−1
2.67+3
Bottom width, B= = 4.97m  5m
√ 2.3−1
Regimes scour depth(R)
13
 q2 
 
f
R=1.35   where q=Q/L=222.504/30 = 6.95 m3 s m and f=1.7

R = 4.12
Where, h, is Height of weir (m), He is specific energy head (over flow depth + approaching
velocity head (m)), ρ is specific weight of weir body (2.35 for cyclopean concrete, 2.3 for
masonry and 2.4 for reinforced concrete). In selecting the specific weight of weir body, it is
advisable to take lower values, since during construction, the quality may be deteriorated and
hydraulic parameters may be under estimated. For our case, masonry weir body capped with
RCC is selected and the value of G =2.3 Therefore taking L=30m, H=3m and

Table 3-2: Weir Top and Bottom Width Determination

47
 Parameter Unit Values
Discharge coeff., C - 1.7
Design discharge,Q m3/s 222.504
Specific Energy Head,He m 2.13
Water Depth U/S After Construction m 0
Specific Gravity,G m 2.3
Top Width m 2.34
Bottom Width,B m 4.97

This is an initial weir dimension which may undersize or oversize the structure. Therefore, to
design resistant from all forces and to make the structures economical, the dimensions are
checked by stability analysis and finally the dimensions becomes.

Weir height 3m
Top Width 2m
Bottom Width 5m

3.5.3. Determination of Upstream and Downstream High Flood Level

Determination of high flood level for both u/s and d/s case of the weir is essential in determining
the wall heights. D/s high flood level is required because, the flood after passing the weir, will
join the natural channel section and therefore From the stage discharge curve prepared in
hydrology Section, the high flood level before construction (i.e. D/s HFL) corresponding to the
design flood is found to be 2495.85m a.m.s.l.
Therefore, top level of the downstream High flood is, D/s HFL = 2495.85m a.m.s.l.
However u/s High flood level is determined by assuming the designed height of weir would be
implemented and therefore
U/s HFL = U/s bed level + weir height + Design (Hd) implying that to find the u/s HFL, some
procedural calculations are required as stated in the following section. Hd is the depth of water
over the weir crest. This is calculated by the selected weir type (broad crested weir) formula. The
velocity head, ha is computed from the approach velocity as shown below

48
 2
Va
ha=  Where g: acceleration due to gravity = 9.81m/sec2, Va is Approach velocity
2g
Q
determined by Va=
¿¿
L is Weir crest length + under sluice length = 32m, hd is flow depth over the weir and also,
hd= He- ha
Q
hd= He- ha ( L(h+hd )¿)2
=¿ ¿ ¿
2g
By using Goal Seek of MS Excel, hd is found to be 1.95m
 Therefore, ha = He-hd = 2.67m-1.95m = 0.72m Therefore:
Therefore:
This facilitates the solution for u/s TEL= weir crest level + He=2493.305m +2.67=2,495.975m
Therefore, U/s HFL =U/s TEL- velocity head =2,495.975-0.72m = 2,495.255m
 Afflux Determination Afflux = U/s HFL- D/s HFL = 2495.85m- 2,495.255m = 0.60m.
Design of Weir Wall
The weir wall is proposed to be trapezoidal cross-section with u/s face vertical and d/s face
with slope 1:1.
Depth of Sheet Piles
R.L of bottom of upstream (U/S) pile= U/s HFL-1.5B, B= Bottom of Weir
=2,495.255-1.5*4.97
=2,488.58m

Therefore, depth of U/S pile (d 1 ) = Bottom of upstream (U/S) pile -River bed level

d 1 =2491.95m -2,488.58m
=2.92m  2.9m
R.L of bottom of downstream (D/S) pile
=D/S HFL– 2B, B= Bottom of Weir
=2495.85-2*4.97
=2485.91m

Depth of D/S pile (d 2 ) = River bed level - R.L of bottom of downstream (D/S) pile
=2,488.58m -2485.91m

49
 =2.67  2.7m
Impervious Floor
Seepage head, Hs= Pond level – Bed Level
=2491.95m -2,488.58m
=3.37m
'
By Bligh s theory, the total creep length (L) is given by:
'
L=CHs where, C=Bligh s
Creep coefficient taken as (5-9) for grave
l foundation
Let us take C=9
L=9*3.37
= 30.33m

Length of downstream impervious floor, l 2

Hs
L 2 =2.21*C 10

=2.21*9
√ 3.37
10
=11.54m11.55m

Length of upstream impervious floor, L 1

L 1 =L- (L 2 +B+2d 1 +2d 2 )

=30.33-(11.55 +4.97+2*2.9+2*2.7)
=+ve

For the u/s impervious floor let us take nominal value of L 1 = 3m

Therefore, total length of impervious floor b, will be

b=L 1 +B+L 2
=3+4.97+11.55
=19.52m

Total creep length changed into=b+2d 1 +2d 2

50
 =19.52+2*2.9+2*2.7
=26.72m
Protection Work
a) D/S protection work
The total length of d/s floor and d/s protection work is given by

=L 2 +L 3
Hsq
10q s
=18C , q = Q/L=222.504/30 = 6.95 m3 s m

3.15 * 20.83
√
3.37∗6.95
=18*9 10∗75 10 * 75

=32.4m

Length down stream protection=L1+L2-L3

=32.4-11.55=20.85m
Hence provide 1m thickness d/s loose talus of 20m in length .
Note: This length of 20m can be partly provided as blocks over inverted filter and partly as
launching apron.
Up stream Protection Work
The length of upstream talus (L4) may be kept equal to half the length of talus.
L3
L4= 2 =20.85/2=10.43m
Thickness of the impervious floor by Bligh’s theory.
 Seepage head=3.37m
 Creep length=26.72m
The maximum ordinate of the H.G. Line above the bottom of the floor for the downstream
portion at the function of weir wall:
3.37
H= ∗30.33=3.94 m
26.72
The thickness of D/S floor at this point is then obtained by;

51
 H 3.94
( ) ¿=4.03 m
t= 1.33 G  1 =1.33*( G−1
Thickness of D/s Floor after 5m from the function of the weir wall.
HS
(16.1  5)
H=HS- L
=1.62m
1.62
)
t=1.33*( 1.24 =1.05m
NOTE: The up stream impervious floor thickness is assumed to be nominal i.e. t=.8m and for
floor length beneath the weir 1m thickness is provided.

3.5.4 Stilling Basin Design

During the flood season, when high flood occurs over the weir crest water falls from the
maximum reservoir level of u/s to the d/s tail water and the difference b/n the u/s and d/s energy
grade line becomes very high. Therefore ,the energy must be dissipated before it reaches the
natural river source: other wise it causes damage to d/s of the apron. The energy tends to
dissipate through a hydraulic jump d/s of the weir .To control the location of the jump stilling
basin is designed.
Conventional Method
It is experimental formula to determine the length and depth of stilling basin.
hf
L=3*
1
hf
D= 2
Where: f/2 >=d>=f/3
L=Length of basin
D=depth of basin
h=over flow depth
f= u/s water level +velocity head –d/s water level.
f=2,495.255m +0.72-2495.85m =1.41m
h=4.91m

Length of basin, L=3* 4.91* 1.41 =7.89m

52
 1
4.91 * 1.41
Length of basin, d= 2
f/2> =d>=f/3, .71<d<.47
Correction for mutual interference
D d  D
'
* 
Correction=19* b  b 
Where, D=Depth of pile whose effect is required on the another pile
D=2.7-1.74 = 1.96
b ' =Distance b/n two piles
=19.52m
b=Total floor length, b=19.52m
d=the depth of the pile on which the influence occur (d=2.9-0.8=2.1m)

Correction=19*
√ 1.96
19.52 √
∗
2.1∗1.96
19.52
=3.99% (+ve)
b) Downstream pile:

b=19.52m, d2=2.7m ,   2.69m,  1.93

100    2
E  cos  1  
   
=51.15%
100    1
D  cos  1  
   
=34%

Thickness correction for  E

   D 
 E  E  * 1.74
 d2 
=3.73%(-ve)
Correction for mutual interference
D d D
'
* 
Chsorrection= -19* b  b 
53
 =-19*
√ 1.96
19.52
∗
√
2.1∗1.96
19.52
=-3.99%(-ve)

Corrected  E 51.15  3.73  3.99

=43.43%
Percentage pressure at A
 C1   ECorrected
 ECorrected  * lengthofd / sfloor
= b
4.46−44.43
=43.43+ ∗11.55
19.52
=59.21%
Residual head, h=0.59*3.37
=1.87m
h 1.87
 1.5
Thickness of the floor = G  1 1.24  2.2m………….Ok
Percentage pressure at B
4.46−44.43
=43.43+ ∗5
19.52
Residual head, h=0.5*3.15
=1.59m
1.59
1.28
Thickness of the floor = 1.24 1.74m…………….Ok

'
Hence the floor is safe by Khosla s theory

Checking of the thickness of the floor by Kohsla’s theory

1. Exit gradient
Total length of the impervious floor, b=19.52m
Depth of down stream pile, d2=2.7m

54
 19.52  1  1   1.93
2

 = 2.7 , 2

H S *1 1
0.11
GE= d 2 *  *  =0.09 < 9 ………..ok
2. Up lift pressure
a. Up stream pile
29.5
 7.41
b=19.52m, d1=2.9m, 2.9 ,  4.24

t=0.8(assumed)

100    2
E  COS  1  
     C1 100  E
,
=32.28% =67.72%
100    1
D  cos  1  
     D1 100   D

=22.32% = 77.68%
 c1
Thickness correction for
  d   C1 
 c1  1  *t

 d1  =2.75%

55
3.5.5 Stability analysis of weir
Dynamic case
 Uplift pressure is considered for the weir wall.
 Water wedge weight is considered for weir crest only
3
 Unit weight of water and masonry is taken to be 9.81 and 22.4 KN/m respectively.
 Moment is taken about the toe per meter width

No Item Forces Lever Moments

(KN) arm (KN-m)
(m)
vertical Horizontal Overturning Restor
ing
56
 1 PH1 = 4.91*ɤw*3.37 151.73 1.575 238.975
2 PH2 = 0.5*ɤw*3.372 48.67 1.05 51.1035
3 PU = 0.5*ɤw(4.91+3.37)*5 -288.6 4.87 1405.482
4 PH3 = 0.5*ɤw*1.852 -16.79 0.617 10.36
5 WW  w * AW 354.84 4.784 1697.5
4
6 W1 m * A1 120.96 6.55 792.28
8
7 W2 m * A2 233.86 3.87 905.02
3
8 Foundation reaction 3.65P1  P2  8.882P1  2 P2 

 V 197.67  H 48.67
M O 600 M R 1484.45

Safety factors

So 
M R

1484.45
2.47 1.5
Overturning stability, M O 600
Safe

Ss 
H 
48.67
0.25 0.75
Sliding stability, V 197.67
OK!
❑
∑
❑
M
648.36
Check for tension, x= ❑
= =3.28
197.67
∑
❑
V

e=¿
B
2
5
| |
B 5
−x∨¿= −3.28 =0.78 , = =0.83
2 6 6
B
e = 0.78¿ =0.83, OK! No tension.
6

3.5.6 Design of Under Sluice

This structure has crest at low level to develop a deep channel pocket, which will help to bring
low dry weather discharge towards this pocket, thereby ensuring easy division of water in to the

57
canal through the head regulator. This opening will also help in scouring and removing the
deposited silt from the under-sluice pocket.
Designed with the discharge of;
1) 20% of the max. Flood, Q=0.2*222.504m3/sec =44.5m3/sec
2) Therefore, Q sluice will be max. of the above. Q sluice=44.36m3/sec

3.5.7 Silt excluder

It is a structure which excludes the silt from irrigation water as the name implies .it separates silt
laden proton of the water from the upper silt free portion.it consists of a series of parallel tunnels
of low height. the tunnels are constructed in the pocket parallel to the flow of water in the river.
The height of the tunnels depends up on the silt distribution in the flow of the water.

3.5.8 Design of silt excluder

It is design that the bottom layer of water, which is highly charged with silt and sediment, will
pass down the tunnels and escape over the floor of the under-sluice ways. Since the gate of the
under -sluice ways shall be kept open up to the top of the tunnels. Then the clear water enters in
to the canal through the head regulator. Head regulator is kept 1.2 to 1.5 higher than the crest
level of the under sluice. (1.25)

3.5.9 Design procedure

Crest level of under sluice=2493.305m
Crest level of the head regulator=2493.305+1.25=2494.555m
The excluder covers usually two or three bays of under sluice of the weir
By width under sluice =2m (Garge, 2003)
Design discharge 15% to 20% of cannel discharge =0.2*222.504m3/sec =44.5m3/sec
A minimum velocity of 2m/s is usually adopted through the tunnel in order to keep the sediment
free from deposit v=2m/s
Area of cross section
(A)=Q/V=44.5m3/sec/2= 22.25m2
Height of tunnels generally varies from 0.8To1.2m for boulder stage (Garge, 2006)
Height of under sluice 1 (h) =crest level of head regulator - crest level of under sluice
2494.555m - 2493.305m =1.25m

58
Total clear width=A/h=22.25m2/1.25=17.8m
For clear span of 0.04m, (assume)
Number of tunnel (Assume=1)
Assume thickness of divide wall =0.3
Overall width 0.04*1+0.3=0.345=0.35m
By considering this, the opening size of the gate is 0.35*1.25m
The gate for under sluice is to be vertical sheet metal of size 1.25 by 1m on the side of the weir,
and for the closure of the opining space. providing some extra dimensions for the groove
indentation gross area of sheet metal for the gate will be (allowing 5cm insertion for grooves),
the grooves is to be provided on the walls using iron frames at the two sides of the gate opening.

4. WATER DEMAND, AGRONOMY AND WATER DELIVERY

ASPECTS
4.1 General
There is no vital requirement other than water for crops. It has a number of functions in the
process of growth.
The function of water with respect of growth of plant and its yield are:
1. Solvent for gaseous, minerals and other soluble food.
2. Conduct and translocation of solutions in cell and tissues.
3. As an active reagent in photosynthesis and hydrolysis.
The factors that affect the water requirement of plants are: type of soil, type of plant,
metrological variables like sunshine, temperature, humidity, rainfall and wind etc.
4.2 Crop water requirement
Crop water requirement is defined as” the depth of water needed to meet the water loss through
evapotranspiration (ETcrop) of a disease free, growing in large fields, under non-restricting
conditions including soil water and fertility and achieving full production potential under the
given growing environment.” (FAO-24, 1994)
The water requirement of crops may be contributed from different sources such as irrigation
requirement, effective rainfall, soil moisture storage and ground water contributions.
CWR=IR+ER+S+GW
59
Where, CWR=crop water requirement
IR=irrigation requirement
ER=effective rainfall
S=carry over soil moisture in the root zone
GW=ground water contribution
a. Irrigation requirement of crops (IR)
It is defined as the part of water requirement of crops that should be fulfilled by irrigation.
IR=CWR-(ER+S+GW)
b. Effective rainfall
It is defined as the rainfall that is stored in the root zone and can be utilized by crops. All the
rainfall that falls is not useful or effective. The different methods used to calculate ER from
monthly total rainfall data are as follows;
1. Fixed percentage effective rainfall
The effective rainfall is taken as affixed percentage of the monthly rainfall
ER=% of total rainfall
2. Dependable rainfall
An empirical formula developed by FAO/AGLW based on analysis for different arid and sub-
humid climates. This formula is as follows
ER=0.6*total rainfall -10 ---------------- for total rainfall<70mm
ER=0.8*total rainfall-24 ---------------- for total rainfall>70mm
3. Empirical formula for effective rainfall
This formula is similar to FAO/AGLW formula (see dependable rainfall method above) with
some parameters left to the user to define. The formula is as follows;
ER=a*total rainfall-b ---------------- total rainfall<Z mm
ER=c*total rainfall-d ---------------- total rainfall>Z mm
Where a, b, c, d and Z are variables to be defined by the user.
4. Method of USDA soil conservation service
The effective rainfall is calculated according to the formula developed by USDA soil
conservation service, which is as follows.
ER=total rainfall*(125-0.2*total rain fall)/125 ------- total rainfall<250mm
ER=125+0.1*total rainfall ------- ------- ------- -------total rainfall>250mm
60
From the above different methods, the USDA is preferable to find the ER

61
 FAO (2017). The Future of Food and Agriculture – Trends and Challenges. Food and
Agriculture Organization of the United Nations.

 IWMI (2018). Revitalizing small-scale irrigation for rural development. International Water
Management Institute.

 World Bank (2016). Confronting Drought in Africa’s Drylands: Opportunities for Enhancing
Resilience. Washington, DC.

62