Section 3 Complete

Section – 3
Urban Road Drainage Runoff

GUIDELINE 6
AACRA DRAINAGE DESIGN MANUAL
TABLE OF CONTENTS
3. URBAN ROAD DRAINAGE RUNOFF 3-1
3.1 Introduction 3-1
3.2 Travel Time 3-1

3.2.1 Time of Concentration Error! Bookmark not defined.
3.3 Design of Urban Road Drainage Runoff 3-4

3.3.1 The Rational Formula 3-4
3.3.2 SCS Graphical Peak Discharge Method 3-4
3.3.3 SCS Unit Hydrograph Method 3-8
3.4 Flood Frequency Analysis 3-14

3.4.1 Selection and Preparation of Data 3-14
3.4.2 Calculation of Plotting Positions for Observed Floods 3-17
3.4.3 Types of Frequency Distributions 3-18
3.4.4 Plotting Flood Frequency Curves 3-18
3.4.5 Flood Frequency Distributions 3-19
3.4.6 Modifications to Data 3-22
3.5 Data Requirements for the Urban Road Drainage System 3-28
3.5.1 Watershed Mapping 3-28
3.5.2 Soil Maps 3-28
3.5.3 Land Use Mapping 3-28
3.5.4 Flood Histories 3-29
3.5.5 Existing Drainage Facilities 3-29
3.5.6 Utility Plans 3-29
3.6 Worked Examples 3-29
3.6 References 3-34
TABLES
Table 3.1: Manning's Roughness Coefficient (n) for Overland and Sheet Flow 3-2
Table 3.2 Intercept Coefficients for Velocity vs. Slope Relationship of Eq. 3.5 3-3
Table 3.3: Runoff Coefficients for Rational Formula * 3-5
Table 3.4: Runoff Curve Numbers 3-8
Table 3.5 : Runoff Curve Numbers with Hydrologic Conditions 3-9
Table 3.6: Ia/P for Selected Rainfall Depths and Curve Numbers 3-10
Table 3.7: Coefficients for SCS Peak Discharge Method 3-11
Table 3.8 (Fp) for Pond and Swamp Areas 3-11
Table 3.9: Ratios of Dimensionless Unit Hydrograph and Mass Curve 3-15
Table 3.10: Frequency Factors KY for Use with Log Pearson Type III Distribution, Positive Skew
Coefficients 3-21
Table 3.11: Frequency Factors KY for Use with Log Pearson Type III Distribution, Negative Skew
Coefficients 3-22
Table 3.12: Values of KN for Outlier Test – 5% Significance Level Values 3-24
Table 3.13: Values of ß for Known Values of Skew g and Sample Size N 3-24
Table 3.14: Values of θ for Known Values of Skew g and Sample Size N 3-25
Table 3.15: Guide for Treatment of Outliers 3-26
Table 3.16: Values of δ for Known Values of Exceedance probability and Skewness Index 3-27
Table 3.17: Computations of Curvilinear UH – Example 3.1 3-30
Table 3.18: Annual Flood Series - Akaki at Akaki Basin (Awash) 3-32
Table 3.19: Computation of Frequency Curve Coordinates for Example 3.3 3-33
FINAL - November 2004 Section – 3 Urban Road Drainage Runoff

Page i
GUIDELINE 6
FIGURES
Figure 3.1: Example of Hydrograph Synthesis 3-12
Figure 3.2: Dimensionless Unit Hydrograph and Mass Curve 3-14
Figure 3.3: Dimensionless Curvilinear Unit Hydrograph and Equivalent Triangular Hydrograph 3-16
Figure 3.4: The Computed Curvilinear UH - Example 3.1 3-30
Figure 3.5: Example 3.5: Given IDF Curve for Example 3.2 3-31
Section – 3 Urban Road Drainage Runoff FINAL - November 2004

Page ii
GUIDELINE 6
SYMBOLS used in SECTION 3
A = drainage area (km2, hectares)Runoff Coe

AEP = annual exceedence probability
Ai = area for cover type i
ARI = average recurrence interval (yrs)
C = runoff coefficient
C0, C1, C2 = coefficients
Ci = runoff coefficient for the cover type ith
CL = confidence limits
CN = runoff curve number
Cw = weighted runoff coefficient
D = unit duration of the rainfall excess
F =frequency factor for the Normal distribution for the selected probability level
Fp = pond and swamp adjustment factor
g = coefficient of Skewness
i = rainfall intensity (mm/h)
Ia = the initial abstraction (mm)
IDF = Intensity-Duration-Frequency
k = number of segments
k1= intercept coefficient
KN =value for sample size of N annual floods
KY = frequency factor
L = flow length (m)
Li = flow length of the ith segment (m)
M = mean of the logarithms of the peak annual discharges
M = mean of the logs of the annual floods
M = mean of the logs of the annual floods
m = rank of the flood in the series
M = Mean
n = Manning’s roughness coefficient
N = number of years of record
P = the depth of precipitation (mm)
P2 = 2-year, 24-hour rainfall depth (mm)
PP = plotting position
Q = the depth of runoff in (mm)
Q = peak flow (m³/s)
Page iii
GUIDELINE 6
qa = the adjusted peak discharge (m3/s)

qp = the peak discharge (m3/s)
qu = the unit peak discharge (m3/s/km2/mm of runoff)
QY = the discharge having an AEP of 1 in Y years (m3/s)
R = the hydraulic radius (m)
S = the maximum potential retention (mm)
S = slope of the surface (m/m, %)
S = standard deviation of the logarithms of the peak annual discharges
S= Standard Deviation
SCS = Soil Conservation Services, USA
Tt = the travel time (min)
Tti = the travel time in the ith segment (min)
UH = unit hydrographs
V = velocity (m/s)
Vi = velocity in the ith segment (m/s)
XH = high outlier threshold (in log units)
Xi = the logarithm to base 10 of an annual peak discharge
XL = low outlier threshold (in log units)
Y = recurrence interval (yr)
ß = an adjustment factor
tc = time of concentration (hours)
tc = time of concentration (min.)
tp= the time to peak (hours)
tr = recession time (hours)
α = a constant.
δ = parameter for determining the standard error of a Pearson III distribution
θ = an adjustment factor

Page iv
GUIDELINE 6
3. URBAN ROAD DRAINAGE RUNOFF
3.1. Introduction
Construction of urban highways and arterial roads substantially alter the perviousness of the areas. Large
amounts of permeable areas are replaced with hard surfaces. In the older portions of large cities, the
pavements may represent over 50% of the total urban impervious area. The diminution of permeable
surfaces lessens the depression storage and infiltration. The paved surfaces speed up the conveyance of
runoff. Thus, urban highways result in greater quantities of runoff at higher rates than would occur under pre-
highway conditions. Stormwater management aims at minimizing, or preferably eliminating entirely, these
development-caused increases in runoff.
An important step in the design of road drainage is the assessment of peak flows because it is the discharge
that a given structure must be sized to handle. Of course, the peak flow varies for each different storm, and it
becomes the designer's responsibility to size a given structure for the magnitude of storm that is determined
to present an acceptable risk in a given situation. Peak flow rates can be affected by many factors in a
watershed, including rainfall, basin size, and the physiographic features.
This Section describes methods which can be used for determination of design discharges in urban
environment. The methods fall in two general categories; un-gauged sites, and gauged sites. The methods
will be discussed in sufficient details as applicable to Addis Ababa.
3.2. Travel Time
The travel time of runoff is an important design parameter . Travel times of surface runoff must be estimated
for design of inlets and pipe drainage systems. Some peak discharge methods use the time of concentration
as input to obtain rainfall intensities from the intensity-duration-frequency curves. Hydrograph times-to-peak,
which are in some cases computed from times of concentration, are used with the hydrograph methods.
Channel routing methods use computed travel times in routing hydrographs through channel reaches. Thus,
estimating travel times are central to a variety of hydrologic design problems.
3.2.1 Time of Concentration
The time of concentration, tc, is defined as the time required for a particle of water to flow from the
hydraulically most distant point in the watershed to the outlet or design point. Factors that affect the time of
concentration are the length of flow, the slope of the flow path, and the roughness of the flow path.
(a) Sheet-Flow Travel Time
Sheet flow is a shallow mass of runoff on a plane surface with the depth uniform across the sloping surface.
Typically flow depths will not exceed 50 mm. Such flow occurs over relatively short distances, rarely more
than about 100 m, but most likely less than 25 m.
Sheet flow rates are commonly estimated using a version of the kinematic wave equation. The original form
of the kinematic wave time of concentration is:
0.6
6.92 ⎡ nL ⎤
t c = 0.4 ⎢ (3.1)
i ⎣ S ⎥⎦
where, tc = time of concentration (minutes)

Page 3-1
GUIDELINE 6
n = Manning’s roughness coefficient

i = rainfall intensity (mm/h)
S = slope of the surface (m/m)
L = flow length
Values of n can be obtained from Table 3.1.
Table 3.1: .Manning's Roughness Coefficient (n) for Overland and Sheet Flow
(Source: HDS 2, 1996)
N Surface n Surface
0.011 Smooth asphalt Cultivated
Soils
0.012 Smooth concrete 0.06 Residue cover 20%
0.013 Concrete lining 0.17 Residue cover > 20%
0.014 Good wood 0.13 Range (natural)
0.014 Brick with cement mortar Grass
0.015 Vitrified clay 0.15 Short grass prairie
0.015 Cast iron 0.24 Dense grasses
0.024 Corrugated metal pipe 0.41 Bermudagrass
0.024 Cement rubble surface Woods*
0.05 Fallow (no residue) 0.40 Light underbrush
0.80 Dense underbrush
*When selecting n for woody underbrush, consider cover to a height of about 30 mm. This is the only part of
the plant cover that will obstruct sheet flow.
Some hydrologic design methods, such as the rational equation, assume that the storm duration equals the
time of concentration. Thus, the time of concentration is entered into the IDF curve to find the design
intensity. However, for Eq. 3.1 i and tc are mutually dependent. Therefore, the computation of tc is an
iterative process. An initial estimate of tc is assumed and used to obtain i from the intensity-duration-
frequency curve for the locality. The tc is computed from and used to check the initial value of i. If they are
not the same, then the process is repeated until two successive tc estimates are the same.
To avoid the necessity to solve for tc iteratively, HDS 2 (1996) uses the
following variation of the kinematic wave equation:
0.8
5.476 ⎡ nL ⎤
t c = 0.5 ⎢ ⎥ (3.2)
P2 ⎣ S ⎦
Where, P2 = 2-year, 24-hour rainfall depth (mm)

An upper limit of L=91 m is recommended for the use of Eq. 3.2.
(b) Velocity Method
The velocity method is applicable to sheet flow, shallow concentrated flow, pipe flow, or channel flow. It is
based on the concept that the travel time, Tt in minutes, for a flow segment is a function of the length of flow
(L, in meters) and the velocity (V, in m/s):
L
Tt = (3.3)
60V
The travel time is computed for the principal flow path. When the principal flow path
consists of segments that have different slopes or land covers, the principal flow path
should be divided into segments and Eq. 3.3 used for each flow segment. The

Page 3-2
GUIDELINE 6
time of concentration is then the sum of travel times:
k k
⎛ L ⎞
t c = ∑ Tti = ∑ ⎜⎜ i ⎟⎟ (3.4)
i −1 i =1 ⎝ 60Vi ⎠
Where, k = number of segments and the subscript i refers to the flow segment
The velocity of Eq. 3.3 is a function of the type of flow (overland, sheet, rill and gully flow, channel flow, pipe
flow), the roughness of the flow path, and the slope of the flow path.
A number of methods have been developed for estimating the velocity.After short distances, sheet flow tends
to concentrate in rills and then gullies of increasing proportions. Such flow is usually referred to as shallow
concentrated flow. The velocity of such flow can be estimated using an empirical relationship between the
velocity and the slope:
V = k1S0.5 (3.5)
Where, V = velocity (m/s)

S = Slope (percent)
k1= intercept coefficient
The value of k1 is a function of the land cover, with values for selected land covers given in Table 3.2 .
Table 3.2 Intercept Coefficients for Velocity vs. Slope Relationship of Eq. 3.5
Land Cover/ Flow Regime k1

Forest with heavy ground litter; hay meadow (overland flow) 0.076
Trash fallow or minimum tillage cultivation; contour or strip cropped; 0.152
woodland (overland flow)
Short grass pasture (overland flow) 0.213
Cultivated straight row (overland flow) 0.274
Nearly bare and untilled (overland flow); alluvial fans in western mountain 0.305
regions
Grassed waterway (shallow concentrated flow) 0.457
Unpaved (shallow concentrated flow) 0.491
Paved area (shallow concentrated flow); small upland gullies 0.619
Flow in gullies empties into channels or pipes. Manning's equation can be used to estimate average flow
velocities in pipes and open channels:
R 2 / 3 S 1/ 2
V = (3.6)
n
Where, R = the hydraulic radius (m)
It may be pointed out that S in Eq. 3.5 is in percentage while that in Eq. 3.6 is expressed as a fraction.

Page 3-3
GUIDELINE 6
3.3. Design of Urban Road Drainage Runoff

3.3.1 The Rational Formula
One of the most commonly used equations for the calculation of peak discharges from small areas is the
rational formula. The rational formula is given as:
CiA
Q= (3.7)
360
Where, Q = peak flow in m³/s

i = rainfall intensity for the design storm (mm/h)
A = drainage area (hectares)
C = dimensionless runoff coefficient assumed to be a function of the cover of the watershed
and often the frequency of the flood being estimated
The quotient, 1/360, is a unit conversion that can be approximated as 0.00278.
The assumptions in the rational formula are as follows:
The drainage area should be smaller than 80 hectares.

The peak discharge occurs when all of the watershed is contributing.
A storm that has a duration equal to tc produces the highest peak
discharge for this frequency.
The rainfall intensity is uniform over a time duration equal to the time of
concentration.
The frequency of the computed peak flow is equal to the frequency of the
rainfall intensity. In other words, the 10-year rainfall intensity, i, is
assumed to produce the 10-year peak discharge.
Ground cover determines the runoff coefficient, C. Some tables of C provide for variation due to slope, soil,
and the return period of the design discharge. The runoff coefficient is a volumetric coefficient that relates the
peak discharge to the "theoretical peak" or 100 percent runoff, occurring when runoff matches the net rain
rate. Hence C is also a function of infiltration and other hydrologic abstractions. Some typical values of C for
the rational formula are given in Table 3.3.
In cases where basin contain varying amounts of different covers, a weighted

runoff coefficient for the entire basin can be determined as:
Cw =
∑C A i i
(3.8)
A
where, Cw = weighted runoff coefficient

Ci , Ai = runoff coefficient and area respectively for cover type i
A = total drainage area. Coe
3.3.2 SCS Graphical Peak Discharge Method
Soil Conservation Services, USA (SCS) has developed a fairly versatile procedure for determination of peak
discharge. The procedure also applies to urban environment. The input for many peak discharge estimation
methods, includes variables to reflect the size of the contributing area, the amount of rainfall, the potential
watershed storage, and the time-area distribution of the watershed. The SCS Graphical Peak Discharge
Method is based on similar input.

Page 3-4
GUIDELINE 6
The method uses a soil classification system that consists of four groups, which are identified by the letters
A, B, C, and D. Soil characteristics that are associated with each group are as follows:
Group A: deep sand, deep loess; aggregated silts
Group B: shallow loess; sandy loam
Group C: clay loams; shallow sandy loam; soils low in organic

content; soils usually high in clay
Group D: soils that swell significantly when wet; heavy plastic clays;
certain saline soils.
Table 3.3: Runoff Coefficients for Rational Formula *

Type of Drainage Area C

Business:
Commercial area 0.70-0.95
Neighbourhood areas 0.50-0.70
Residential:
Single-family areas 0.30-0.50
Multi-units, detached 0.40-0.60
Multi-units, attached 0.60-0.75
Suburban 0.25-0.40
Apartment dwelling areas 0.50-0.70
Industrial:
Light areas 0.50-0.80
Heavy areas 0.60-0.90
Parks, cemeteries 0.10-0.25
Playgrounds 0.20-0.40
Railroad yard areas 0.20-0.40
Unimproved areas 0.10-0.30
Lawns:
Sandy soil, flat, < 2% 0.05-0.10
Sandy soil, average, 2 to 7% 0.10-0.15
Sandy soil, steep, > 7% 0.15-0.20
Heavy soil, flat, < 2% 0.13-0.17
Heavy soil, average 2 to 7% 0.18-0.22
Heavy soil, steep, > 7% 0.25-0.35
Streets:
Asphaltic 0.70-0.95
Concrete 0.80-0.95
Brick 0.70-0.85
Drives and walks 0.75-0.85
Roofs 0.75-0.95
The method takes into account various types of urban developments and above soil types for predicting peak
discharge. Important steps involved in the process are described below.
Step 1: Determine Depth of Direct Runoff, Q (mm)
For the purpose of SCS rainfall-runoff relationship, the total rainfall was separated into three components:
direct runoff (Q), actual retention (F), and the initial abstraction (Ia). The retention F was assumed to be a

Page 3-5
GUIDELINE 6
function of the depths of rainfall and runoff and the initial abstraction. The development of the equation
yielded:
( P − I a )2
Q= (3.9)
(P − I a ) + S
Where, P = the depth of precipitation (mm)
Ia = the initial abstraction (mm)
S = the maximum potential retention (mm)
Q = the depth of direct runoff (mm).
Equation 3.9 requires two unknowns to be estimated, S and Ia. The retention S should be a function of the
following five factors: land use, interception, infiltration, depression storage, and antecedent moisture.
Empirical evidence resulted in the following equation for estimating the initial abstraction:
I a = 0. 2 S (3.10)
Substituting Eq. 3.10 into Eq. 3.9 yields the following equation, which contains the single unknown S:
Q=
( P − 0.2 S )
2
(3.11)
P + 0.8S
Additional empirical analyses were made to estimate the value of S. The studies found that S was related to
soil type, land cover, and the hydrologic condition of the watershed. These are represented by the runoff
curve number (CN), which is used to estimate S by:
25400
S= − 254 (3.12)
CN
A curve number is an index that represents the combination of a hydrologic soil group and a land use and
treatment class. Empirical analyses suggested that the CN was a function of three factors: soil group, the
cover complex, and antecedent moisture conditions. Table 3.4 and Table 3.5 present the CN values for
urban land use.
Step 2: Determine Ia/P Ratio
The parameter Ia/P is required to estimate peak discharge rates. Ia denotes the initial abstraction, and P is
the 24-hour rainfall depth for a selected return period. The Ia/P value can be obtained from Table 3.6 for a
given CN and P. For a given 24-hour rainfall distribution, Ia/P represents the fraction of rainfall that must
occur before runoff begins.
Step 3: Determine Unit Peak Discharge qu
Unit peak discharge is obtained from the following relation:
qu = 0.000431(10) 0 + C1 log(t c ) + C 2 [log(t c )]

C 2
(3.13)
Where, qu = unit peak discharge m3/s/km2

tc = time of concentration in hours
C0, C1, C2 are the coefficients from Table 3.7.

Page 3-6
GUIDELINE 6
Step 4: Determine Peak Discharge qp
The following equation can be used to compute a peak discharge with the SCS
method:
qp =qu A Q (3.14)
Where, qp = the peak discharge in m3/s

qu = the unit peak discharge m3/s/km2/mm of runoff
A = the drainage area in square kilometers
Q = the depth of runoff in mm.
Step5: Adjustment for Ponding or Swampy Areas (if required)
Eq. 3.14 is based on the assumption that the topography of the area permits unimpeded flow into ditches,
drains, and streams is relatively unimpeded. Where ponding or swampy areas occur in the watershed, a
considerable amount of the surface runoff may be retained in temporary storage. The peak discharge rate
should be reduced to reflect this condition of increased storage. Values of the pond and swamp adjustment
factor (Fp) are provided in Table 3.8. The adjustment factor values are a function of the percent of the total
watershed area in ponds and swamps (PPS). If the watershed includes significant portions of pond and
swamp storage, then the peak discharge of Eq. 3.14 can be adjusted using the following:
qa=qpFp (3.15)
Where, qa = the adjusted peak discharge in m3/s.

Page 3-7
GUIDELINE 6
Table 3.4: Runoff Curve Numbers

3.3.3 SCS Unit Hydrograph Method
For basins without measured data, synthetic methods can be used to develop unit hydrographs. These
methods tend to be somewhat inflexible in that they use standard shapes for the unit hydrographs. A number
of different synthetic unit hydrograph procedures have evolved. One of the most widely used is the SCS
method. This method is presented in this Section.
As the name suggests, a design hydrograph is a hydrograph that has characteristics that are believed to
represent the flood conditions at the limit considered acceptable. The design hydrograph is usually
generated using a design-storm hyetograph and a unit hydrograph. However, the design hydrograph could
also be an actual storm hydrograph that was experienced at the design location. In either case, the design
hydrograph may have an exceedence frequency associated with it. In the case of design-storm modelling, it
is common to assume that the frequency of the runoff hydrograph is the same as the frequency of the design
hyetograph. An example of hydrograph synthesis is presented in Figure 3.1.

Page 3-8
GUIDELINE 6
Table 3.5 : Runoff Curve Numbers with Hydrologic Conditions

The Soil Conservation Service has developed a synthetic unit hydrograph procedure that has been widely
used in their conservation and flood control work. The unit hydrograph used by the SCS is based upon an
analysis of a large number of natural unit hydrographs from a broad cross section of geographic locations
and hydrologic regions. This method is easy to apply. The input parameters are the peak discharge, the area
of the watershed, and the time to peak. With these parameters, a standard unit hydrograph is constructed.
Dimensionless unit hydrographs, based on an extensive analysis of measured data, are used in this method.
Unit hydrographs were evaluated for a large number of actual watersheds and then made dimensionless by
dividing all discharge ordinates by the peak discharge and the time ordinates by the time to peak. An
average of these dimensionless unit hydrographs (UH) was computed. The time base of the dimensionless
UH was approximately 5 times the time to peak, and approximately 3/8 of the total volume occurred before the
time to peak; the inflection point on the recession limb occurs at approximately 1.7 times the time to peak, and the

Page 3-9
GUIDELINE 6
UH has a curvilinear shape. The average dimensionless UH is shown in Fig. 3.2 and the discharge ratios for
selected values of the time ratios are given in Table 3.1.
Table 3.6: Ia/P for Selected Rainfall Depths and Curve Numbers
Source: HDS 2, 1996

Page 3-10
GUIDELINE 6
Table 3.7: Coefficients for SCS Peak Discharge Method

Source: HDS 2, 1996
Table 3.8 (Fp) for Pond and Swamp Areas

Source: HDS 2, 1996

Page 3-11
GUIDELINE 6
Figure 3.1: Example of Hydrograph Synthesis

Source: HDS 2, 1996
The curvilinear unit hydrograph can be approximated by a triangular UH that has similar characteristics;
Figure 3.3 shows a comparison of the two dimensionless unit hydrographs. It is important to recognize that
the triangular UH is not a substitute for the curvilinear UH. The curvilinear UH is always used in hydrologic
computations. The triangular unit hydrograph is only used to develop an expression for computing the peak
discharge of the curvilinear unit hydrograph. While the time base of the triangular UH is only 8/3 of the time
to peak (compared to 5 for the curvilinear UH), the areas under the rising limbs of the two UHs are the same
(i.e., 37.5 percent).
The area under a hydrograph equals the depth of direct runoff Q, which is 1 mm for a unit hydrograph; based
on geometry the runoff volume is related to the characteristics of the triangular unit hydrograph by:
q p (t p + t r )
1
Q= (3.16)
2
Where, tp and tr are the time to peak and the recession time, respectively; and qp is the peak discharge.
Solving Eq. 3.16 for qp and rearranging yields:

⎛ ⎞
Q⎜ 2 ⎟
qp = ⎜ ⎟ (3.17)
t p ⎜ 1 + tr ⎟
⎜ tp ⎟
⎝ ⎠
Letting K replace the contents within the brackets yields:
KQ
qp = (3.18)
tp

Page 3-12
GUIDELINE 6
Because tr = 1.67tp, qp in m3/s, tp in hours, and Q in mm, it is necessary to divide qp by the area A in km2.
Adjusting for the units of the variables, Eq. 3.18 becomes:
2.5 AQ 0.2083Q
qp = = (3.19)
12t p tp
The constant 0.2083 of Eq. 3.19 reflects a unit hydrograph that has 3/8 of its area under the rising limb. For
mountainous watersheds the fraction could be expected to be greater than 3/8, and therefore the constant of
may be near 0.26. For flat, swampy areas the constant may be on the order of 0.13.
The time to peak of can be expressed in terms of the unit duration of the rainfall excess and the time of
concentration. Figure 3.3 provides the following two relationships:
tc + D = 1.7 tp (3.20)
If the lag equals 0.6 tc, then
D/2 + 0.6tc = tp (3.21)
Solving Eq. 3.20 and Eq. 3.21 for D yields:
D = 1.33 tc (3.22)
Therefore, tp can be expressed in terms of tc:
tp = D/2 + 0.6tc = 2/3tc (3.23)
Expressing Eq. 3.19 in terms of tc rather than tp yields:
0.3125 AQ
qp = (3.24)
tc
For a unit hydrograph, the depth of runoff Q would equal 1 mm.

Page 3-13
GUIDELINE 6
Figure 3.2: Dimensionless Unit Hydrograph and Mass Curve

Source: HDS 2, 1996
3.4. Flood Frequency Analysis

A flood frequency analysis is utilised to determine the magnitude of a flood with a particular probability of
exceedance from a statistical analysis of recorded floods. This section deals with the frequency analysis of
floodwaters, which occur, at or near the same location as that of the recorded events. Frequency methods
can also be used to transfer information from other gauged catchments.
3.4.1 Selection and Preparation of Data

To undertake a flood frequency analysis the data must constitute a random sample from a homogenous
population. Most catchments change over time in terms of the land use. It is particularly true for a city like
Addis Ababa. The changes in land use have impact on the homogeneity of the catchment over time and
therefore on the recorded data.
Designers should use caution when using flood frequency methods to predict flows and note what changes
have occurred in the catchment over time. Fortunately, the available evidence indicates that unless changes
to the catchment involve large proportions of the total area or large changes in the storage on the catchment,
the effects on flood magnitude are likely to be small. Also, the effects are likely to be larger for small floods
than for the large floods that are of the primary interest in design (Institute of Engineers, 2001).

Page 3-14
GUIDELINE 6
Table 3.9: Ratios of Dimensionless Unit Hydrograph and Mass Curve

Source: HDS 2, 1996
There are two methods available for selecting the data series to be used for the flood frequency analysis:
(a) Annual Flood Series
This is the simplest and most commonly used method. The series comprises the data set of the maximum
instantaneous discharge for each year of records. The maximum discharge is selected for each year
regardless of whether it is a major flood event or not.

Page 3-15
GUIDELINE 6
Figure 3.3: Dimensionless Curvilinear Unit Hydrograph and Equivalent Triangular Hydrograph
Source: HDS 2, 1996
(b) Partial Flood Series
The partial flood series consists of all events with peak discharges larger than a selected base value,
regardless of the number of such events in a year.
The partial series has an important advantage over the annual series in that depending on the base value
selected, small events that are not really floods are excluded from the data set. In the annual series however
these small events may remain in the data set due to a very dry year for instance and thereby skew the
analysis.
It is critical in selecting peak flows for the partial flood series that the peaks are from independent events. It is
common that within the flow hydrograph for a single storm a number of peaks are noted. Only one peak
should be selected from the single event for inclusion into the partial flood series.
(c) Comparison and Choice of Series
Due to the different data sets for the two methods the frequency analysis of the two series are also different.
Theoretical analysis of the relationship between the probabilities of the annual and partial series has shown
that the difference is relatively small when the recurrence interval (Y) is 10 year or greater.
The recommended series therefore is dependent upon Y:
i. Y greater than or equal to 10 years

Page 3-16
GUIDELINE 6
Annual series is generally preferred as it is generally simpler to calculate yet provides almost
identical results to the partial series. Most major stormwater infrastructure is designed for rare floods
such as the 1 in 50 or 1 in 100 year events and as such the annual series is most commonly used.
This is especially applicable to bridges and culverts in a city environment like Addis Ababa.
Ii Y is less than 10 years
The partial series is generally preferred when considering storms with a frequency less than 1 in 10
years. For these minor storms all floods are of interest wherever they are the highest for a particular
year or not. The annual series however may omit many of these storms. The annual series is
appropriate for estimating floods of low ARI for the design of temporary works etc.
3.4.2 Calculation of Plotting Positions for Observed Floods
Observed peak discharges can be plotted on a frequency diagram. The plotted values may be used to draw
a straight line or smooth curve as an empirical probability distribution, or to check the adequacy of an
analytically fitted distribution.
The value of the AEP or ARI used for each observed flood is referred to as the ‘plotting position’. It should be
noted that the value of the AEP or ARI should not be used as an estimate of the actual flood discharge. The
estimate should be obtained from straight or curved fitted distribution.
(a) Annual Flood Series
When analyzing an annual flood series, a general formula (Blom, 1958) for calculating plotting position,
PP(m), of an observed flood in terms of exceedance probability as a fraction is:
m− α
PP(m ) = (3.25)
N + 1 − 2α
Where
m= rank of the flood in the series (the largest flood having the rank m = 1)
N= number of years of record
α= a constant.
The value of α depends on the type of distribution. A log Pearson III distribution is recommended for fitting
the annual series (Institute of Engineers, 2001). The suitable value of α depends on the parameters of the
distribution but on average a value of 0.4 has been found suitable. The plotting position for the annual series
data can therefore be simplified to:
m − 0.4
PP(m ) = (3.26)
N + 0.2
If the plotting position exceedance probability is expressed in the form 1 in YP(m), then:
N + 0.2
PP(m ) = (3.27)
m − 0.4
It should be noted that the event related to the recurrence interval is the exceedance of a given discharge
within any one year and the ARI is the average interval between years in which a given discharge is

Page 3-17
GUIDELINE 6
exceeded, whether once or more than once. The ARI for the partial series however is different and relates to
the average interval between exceedances of the given discharge.
As noted previously for ARI’s greater than 10 years the differences are small and the ARI from the annual
series can be a useful approximation of the normal ARI from the partial series.
(b) Partial Flood Series
For the partial series, only a plotting position in the form of an ARI of YP years should be calculated. The
plotting position determined from Equation 3.27 should be used.
3.4.3 Types of Frequency Distributions
The general form of the data distributions dictate the type of plotting and analytical processes used in
analyzing the data. As such, it is necessary to have a basic understanding of the form of the distributions.
(a) Annual Series
When analyzing the data set of an annual flood series, most commonly the points will cluster around a
medium sized flood magnitude. The number of points representing very low magnitude floods or very high
magnitude floods will be relatively few. The resulting probability distribution is of the form of a bell shaped
curve with almost always positive skew. If logarithms of the flood magnitudes are assessed the distribution
may become approximately symmetrical or often have a negative skew. It may be possible to approximate
the log of the magnitudes as a Normal distribution. Whether actual or logarithmic values of the flood
magnitude are used, the bell shaped distribution is similar to many standard types of probability distributions,
and allows the use of standard probability graph papers and standard statistical techniques in the analytical
analysis of the distribution.
(b) Partial Series
When assessing the data set for a partial flood series there is commonly a large number of points near the
base magnitude selected as the cut-off for inclusion of points in the set. The number of points reduces as the
magnitude increases. This type of distribution can be approximated by a negative exponential distribution.
3.4.4 Plotting Flood Frequency Curves
(a) General Considerations
Plotting flood frequency curves permits the frequency analysis to be presented graphically and also provides
a useful visual check on the adequacy of analytically fitted probability distributions.
Flood data are plotted in the form of a cumulative frequency distribution. Discharge or its logarithm are
plotted on the ordinate and AEP or ARI on the abscissa. The simplest way of plotting flood data is on special
graph paper with the abscissa graduated to conform to a particular type of probability distribution. Plotting
points using this kind of graph paper permits points to occur more evenly spaced and to fall approximately in
a straight or curved line.
(b) Plotting of Annual Series
Logarithmic Normal probability graph paper should be used to plot the annual series. Each data point is
plotted with magnitude on the logarithmic scale and annual exceedance calculated by the plotting position
formulae on the probability scale.

Page 3-18
GUIDELINE 6
A straight or curved line can be drawn through the plotted points or alternatively the plotted points can be
used to provide a visual check on an analytically fitted distribution. A straight line on log Normal paper
involves the assumption that the skew of the logarithmic values is zero.
(c) Plotting of Partial Series
Plotting of the partial flood series should be undertaken on semi-log (or log-linear) paper. For each flood,
magnitude is plotted on the linear scale and ARI calculated by the plotting position formula is plotted on the
logarithmic scale. A straight line or curve can be drawn through the points to approximate the distribution
(d) Extrapolation of Frequency Curves
Most often records do not extend to the required AEP or ARI for design purposes and as such the trends of
the known values must be extrapolated. To undertake this extrapolation the fitting curve needs to have a
small curvature or be straight. Due to extorted scales small variations along the line mean large numerical
values are involved. Large errors can occur and care and caution are needed in extrapolating data.
3.4.5 Flood Frequency Distributions
This section concentrates on flood frequency distributions for the annual flood series. As mentioned
previously the data points in the annual series have a bell-shaped distribution similar to many standard types
of probability distributions, and are thus readily analysed by standard statsical techniques.
The log Pearson Type III distribution is recommended for general use. That is, the logarithms of the annual
floods are assumed to be distributed in accordance with the Pearson Type III probability distribution.
(Institute of Engineers, 2001)
(a) Steps in Fitting a Log Pearson III Distribution to Annual Series Data
Step 1: Select data points for the annual series as previously discussed.
Step 2: Plot data on log Normal probability paper.
Step 3: Calculate estimates of the moments of the populations of logarithms of floods. If Xi is the
logarithm to base 10 of an annual peak discharge Qi and N years is the length of record, then for the
logarithmic values:
Mean (M)
1
∑
N
M = 1
Xi (3.28)
N
Standard Deviation (S)
S = [ Σ(Xi2) – {1/N}( ΣXi)2 }0.5 for i = 1…N (3.29)

N-1
Co-efficient of Skewness (g)
g = N2 Σ(Xi3) – 3N ΣXi Σ(Xi2) + 2(ΣXi)3 for i = 1…N (3.30)

3
N(N-1) (N-2)S
Step 4: The observed data should be tested for outliers, which are values that depart significantly from
the trend of the remaining data.
Page 3-19
GUIDELINE 6
Step 5: Flows for a range of AEP’s should then be calculated from the formula:
Log (QY) = M + KYS
Where,
QY = the discharge having an AEP of 1 in Y
M= mean of the logarithms of the peak annual discharges
KY = frequency factor found from Table 3.10 and Table 3.11 for the required AEP of 1 in
Y
S= standard deviation of the logarithms of the peak annual discharges
Step 6: Confidence limits should then be calculated for each of the AEP’s for which discharge was
calculated
Step 7: The discharges and confidence limits calculated should be plotted on the graph from Step 2.
Smooth curves should be drawn through the computed values. The flood frequency curve should then
be carefully compared with the plotted observed data. The confidence limits, which indicate the probable
range of the flood population, provide assistance in assessing whether the calculated distribution gives a
satisfactory fit to the observed data.

Page 3-20
GUIDELINE 6
Table 3.10: Frequency Factors KY for Use with Log Pearson Type III Distribution, Positive Skew
Coefficients
Source: Institute of Engineers Australia, 2001

Page 3-21
GUIDELINE 6
Table 3.11: Frequency Factors KY for Use with Log Pearson Type III Distribution, Negative Skew
Coefficients
3.4.6 Modifications to Data
Outliers are events that depart significantly from the trend of the remaining data. The retention, modification
or deletion of these values can appreciably affect the fitted distribution and estimates made from it.
(a) High Outliers
High outliers create a particular problem for flood frequency analysis. The outliers can generally be classified
into:
Errors in the data. In this event data should be corrected where possible;

Page 3-22
GUIDELINE 6
Changes in the catchment conditions under which an extreme event occurred. In this event an
attempt should be made to adjust the magnitude of the extreme event to be homogenous with the
remainder of the data;
A rare event with AEP much lower than those of the other high ranking observed events; or
An extreme event resulting from an unusual type of phenomenon that is different to those causing
the other events.
The assessment of the importance of these events and effects must necessarily contain a large subjective
element and the suggested procedure involves a broad classification of strong, moderate and weak prior
belief. The second stage of the assessment of outliers involves applying a statistical test. The equation used
to indicate high outliers is:
XH = M + ß KN S (3.31)
Where,
XH = high outlier threshold in log units
M= mean of the logs of the annual floods, excluding zero and very low events, and previously discarded
outliers
S= standard deviation of logs of floods
KN = value for sample size of N annual floods (Table 3.12)
ß= an adjustment factor depending on N and g, the skew of the logs of floods

Page 3-23
GUIDELINE 6
Table 3.12: Values of KN for Outlier Test – 5% Significance Level Values

Table 3.13: Values of ß for Known Values of Skew g and Sample Size N

Page 3-24
GUIDELINE 6
(b) Low Outliers
Values at the low end of the observed range that are lower than the trend of the remainder of the data, can
distort the fitted probability distribution. The following equation can be used to detect low outliers:
XL = M – θ KN S (3.32)
Where
XL = low outlier threshold in log units
M= mean of the logs of the annual floods, excluding zero and very low events
θ= an adjustment factor depending on N and g, the skew of the logs of floods
Table 3.14: Values of θ for Known Values of Skew g and Sample Size N
(c) Treatment of Outliers
When outliers have been detected a suitable treatment of these points is necessary. Table 3.15 provides
guidance for the treatment of outliers.

Page 3-25
GUIDELINE 6
Table 3.15: Guide for Treatment of Outliers

(d) Confidence Limits
Different random samples of flood magnitude data will result in different estimates of the population
distribution and as such an estimate of confidence limits is useful. The following method can be used for log
Pearson III distributions.
At a selected AEP of 1 in Y, the confidence limits CL(QY) about the discharge QY are given by:
Log(CL(QY)) = log QY +/- F x δ x S (3.33)

N½
Where logarithms are to base 10 and
F= frequency factor for the Normal distribution for the selected probability level
δ= parameter for determining the standard error of a Pearson III distribution
S= standard deviation of logarithms of flows.
The 5% and 95% confidence limits are usually used and for these F = 1.645.

Page 3-26
GUIDELINE 6
Table 3.16: Values of δ for Known Values of Exceedance probability and Skewness Index

Page 3-27
GUIDELINE 6
Table 3.16 continued from previous page
3.5. Data Requirements for the Urban Road Drainage System

The discussion so far has focussed on determination of design discharges for the drainage system and its
components. The following description provides a summary of data required for the storm drainage system.
3.5.1 Watershed Mapping
Mapping of the watershed includes identification of the watershed boundaries, its topographic features,
details of ground cover and the existing drainage patterns. In Addis Ababa, the sources of this information
include: Ethiopian Mapping Authority (EMA), Bureau of Urban Development (Addis Ababa City
Administration), National Urban Planning Institute (NUPI), Addis Ababa City Roads Authority (AACRA), and
Addis Ababa Water Supply and Sewerage Authority (AAWSSA).
3.5.2 Soil Maps
Soil maps include records of the soil types and hydrologic soil groups. The data are useful for determination
of the runoff and design of structural components of the drainage system. This information can be obtained,
among others, from: Ethiopian Mapping Authority (EMA), Bureau of Urban Development (Addis Ababa City
Administration) and National Urban Planning Institute (NUPI).
3.5.3 Land Use Mapping
Nature of the land use is directly related to the runoff it generates from a given storm. While designing a
drainage system, the designer should not only know the current land use of the area to be drained but also
the expected changes in the future. The information is available from: Bureau of Urban Development (Addis
Ababa City Administration),

Page 3-28
GUIDELINE 6
3.5.4 Flood Histories
Flood histories and high water marks provide useful information for design of the drainage system and its
individual components. The gauge records for Addis Ababa rivers are available at Ministry of Water
Resources. This information in relation to flood damages are compiled, up to their study periods, in the two
Flood Protection Stormwater Drainage Studies made for the City by BCEOM (1987) and JICA (1998).
3.5.5 Existing Drainage Facilities
A description of the existing drainage facilities is essential for effective new designs as well as any
improvements. The information consists of size, age, condition, shape and material etc. As-built drawings
can provide most of the required information. There is no readily available data base for the existing drainage
facilities. The major ones, up to its period of study are, however, included in the Flood Protection and
Stormwater Drainage Study made by BCEOM (1987) for the City.
Information is also required on design and performance of the existing drainage systems. It will be necessary
for the designer to determine as to how the existing system will function under the new loading. If
unavailable, field surveys should be performed to obtain this information.
3.5.6 Utility Plans
The move towards integrated City level management of utility lines is commencing. As such integrated utility
plans are not readily available at one place. The required Utility plans can, therefore, be obtained from the
individual utility lines providing institutions the major ones of which include: Addis Ababa Water Supply and
Sewerage Authority (AAWSSA), Ethiopian Telecommunication Corporation (ETC) and Ethiopian Electric
Power Corporation (EEPCo).
3.6. Worked Examples

Example 3.1 The objective is to determine the curvilinear UH for a 1.2 km2 watershed that has been
commercially developed. The flow length is 1982 m, the slope is 1.3 percent, and the soil is of group B.
Assume that a time of concentration of 1.34 hours was computed.
For commercial land use and soil group B, the watershed CN is 92 (see Table 3.5). For 1 mm of rainfall
excess, Eq. 3.24 provides a peak discharge of:
The time to peak is:
tp = 2/3(tc) = 0.893h
and the time base of the UH is:
tb = 5 tp = 4.5h
The ordinates of the SCS curvilinear UH can be determined using the values of Table 3.9. The curvilinear
UH will be approximated for selected values of t/tp; the SCS For selected values of t/tp, the curvilinear UH is
computed in the Table 3.17 and shown in Figure 3.4. The curvilinear UH can be considered a D-hour UH,
with D computed by Eq. 3.22 as:

Page 3-29
GUIDELINE 6
Thus the UH should be reported on an interval of 0.178 hour, or about 0.2 hour, and all computations
performed at that interval.
Table 3.17: Computations of Curvilinear UH – Example 3.1
Figure 3.4: The Computed Curvilinear UH - Example 3.1
Example 3.2 This example demonstrates iterative use of Eq. 3.1 to determine time of concentration using
Intensity-Duration-Frequency (IDF) curves. Flow Length is 50 m on short grass (n = 0.15) and at a slope of
0.005 m/m. Use a 2-year return period.
When Eq. 3.1 is used, it must be solved iteratively. From Eq. 3.1:
113.6
tc = (3.34)
i 0 .4
The value of i is obtained from an IDF curve for the locality of the project. For this example, the IDF curve is
provided in Fig.3.5. The problem assumes that a 2-year return period is specified. An initial tc of 12 minutes
will be used to obtain the intensity from the given IDF curve. The initial intensity is 116 mm/h.

Page 3-30
GUIDELINE 6
Using this Eq. 3.1a gives a tc of 17.0 minutes. Since this differs from the assumed tc of 12 minutes, a second
iteration is necessary.
Using a duration of 17.0 minutes with the IDF curve gives a rainfall intensity of 78 mm/h, which when
substituted into Eq. 3.1a yields an estimated tc of 19.9 minutes. Again, this differs from the assumed value of
17.0 minutes, so another iteration is required.
For this iteration, the rainfall intensity is found from IDF curve using a duration of 19.9 minutes. This gives an
intensity of 72 mm/h. With Eq. 3.34 the estimated tc is 20.5 minutes. While the change in tc for this iteration is
small, one more iteration will be made.
For a duration of 20.5 minutes, the intensity is 71 mm/h Eq 3.34 gives a tc of 20.6 minutes. Since this would
not change the intensity, a time of concentration of 20.6 minutes is used for this flow path.
Figure 3.5: Example 3.5: Given IDF Curve for Example 3.2
Example 3.3
The following flood frequency analysis is for the maximum mean daily peak data of Akaki at Akaki Basin
(Awash). The catchment area to the recording station is approximately 884sq. kilometres. The Ministry of
Water Resources has supplied the flood information data in the form of an annual series.

Page 3-31
GUIDELINE 6
Table 3.18: Annual Flood Series - Akaki at Akaki Basin (Awash)
Rank Year Peak Flow X X2 X3 AEP

M m3/s log10 %
1 1996 374 2.57 6.62 17.02 3.3
2 1990 278 2.44 5.97 14.59 8.8
3 1993 276 2.44 5.96 14.55 14.3
4 1989 224 2.35 5.52 12.98 19.8
5 1991 215 2.33 5.44 12.70 25.3
6 1998 203 2.31 5.32 12.27 30.8
7 1981 201 2.30 5.31 12.23 36.3
8 1997 196 2.29 5.25 12.04 41.8
9 1984 189 2.28 5.19 11.81 47.3
10 1982 172 2.24 5.00 11.17 52.7
11 1985 158 2.20 4.83 10.61 58.2
12 1992 153 2.18 4.77 10.43 63.7
13 1988 148 2.17 4.71 10.24 69.2
14 1995 148 2.17 4.71 10.22 74.7
15 1983 139 2.14 4.59 9.83 80.2
16 1994 96 1.98 3.92 7.77 85.7
17 1986 69 1.84 3.38 6.20 91.2
18 1987 37 1.56 2.44 3.82 96.7
Totals 39.80 88.94 200.49
i. The data set is arranged in order of magnitude as presented in Table 3.18 above.
ii. The plotting position is calculated using Equation 3.25.

PP(m) = m – 0.4 x 100 = m – 0.4 x 100%
N + 0.2 17.8
iii. Statistics of the Annual Series are calculated.

Summations of the logarithms of the discharges listed in Table 3.18 and of their squares and
cubes. For the data set the summations are as follows:
ΣX = 39.80
ΣX2 = 88.94
ΣX3 = 200.49
These values are substituted into the Equations 3.28 – 3.30.
M = 2.21
S = 0.23487
g = -1.60

Page 3-32
GUIDELINE 6
Table 3.19: Computation of Frequency Curve Coordinates for Example 3.3
AEP KY KYS LOG QY QY

% = M + KYS m3/s
99 -3.388 -0.796 1.414 25.9
95 -1.962 -0.461 1.749 56.1
90 -1.329 -0.312 1.898 79.1
80 -0.675 -0.159 2.051 112.5
50 0.254 0.060 2.270 186.2
20 0.817 0.192 2.402 252.3
10 0.994 0.233 2.443 277.3
5 1.093 0.257 2.467 293.1
2 1.166 0.274 2.484 304.8
1 1.197 0.281 2.491 309.7
0.5 1.216 0.287 2.496 313.0
0.2 1.231 0.289 2.500 315.6

Page 3-33
GUIDELINE 6
3.5 References
Austroads-Australia (1994),“Waterway Design- A Guide to the Hydraulic Design of Bridge, Culverts and
Floodways”, AUSTROADS National Office, Sydney.
Ethiopian Road Authority (ERA - 2001), “Drainage Design Manual”, Federal Democratic Republic of Ethiopia.
Institute of Engineers, Australia (2001), “Australian Rainfall and Runoff – A Guide to Flood Estimation
Volume 1”.
HDS 2 (1996)- Federal Highway Administration, “Highway Hydrology”, US Department of Transportation.
HDS 4 (2001)- Federal Highway Administration, “Introduction of Highway Hydraulics”, US Department of

Transportation.
HEC 22 (1996) - Federal Highway Administration, ”Urban Drainage Design Manual”, US Department of
Transportation.
TS -79-225 (1979), Federal Highway Administration “Design of Urban Highway Drainage”, US Department of
Transportation.

Page 3-34

Section 3 Complete

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Section 3 Complete

Uploaded by

Copyright:

Available Formats

Section – 3

Urban Road Drainage Runoff

3. URBAN ROAD DRAINAGE RUNOFF 3-1

3.1 Introduction 3-1

3.2 Travel Time 3-1

3.3 Design of Urban Road Drainage Runoff 3-4

3.4 Flood Frequency Analysis 3-14

3.6 Worked Examples 3-29

3.6 References 3-34

FINAL - November 2004 Section – 3 Urban Road Drainage Runoff

Section – 3 Urban Road Drainage Runoff FINAL - November 2004

SYMBOLS used in SECTION 3

A = drainage area (km2, hectares)Runoff Coe

qa = the adjusted peak discharge (m3/s)

Section – 3 Urban Road Drainage Runoff FINAL - November 2004

3. URBAN ROAD DRAINAGE RUNOFF

3.2. Travel Time

3.2.1 Time of Concentration

(a) Sheet-Flow Travel Time

where, tc = time of concentration (minutes)

FINAL - November 2004 Section – 3 Urban Road Drainage Runoff

n = Manning’s roughness coefficient

Values of n can be obtained from Table 3.1.

Where, P2 = 2-year, 24-hour rainfall depth (mm)

(b) Velocity Method

Section – 3 Urban Road Drainage Runoff FINAL - November 2004

time of concentration is then the sum of travel times:

Where, V = velocity (m/s)

Land Cover/ Flow Regime k1

Where, R = the hydraulic radius (m)

FINAL - November 2004 Section – 3 Urban Road Drainage Runoff

3.3. Design of Urban Road Drainage Runoff

Where, Q = peak flow in m³/s

The assumptions in the rational formula are as follows:

 The drainage area should be smaller than 80 hectares.

In cases where basin contain varying amounts of different covers, a weighted

where, Cw = weighted runoff coefficient

3.3.2 SCS Graphical Peak Discharge Method

Section – 3 Urban Road Drainage Runoff FINAL - November 2004

Group A: deep sand, deep loess; aggregated silts

Group B: shallow loess; sandy loam

Group C: clay loams; shallow sandy loam; soils low in organic

Table 3.3: Runoff Coefficients for Rational Formula *

Type of Drainage Area C

FINAL - November 2004 Section – 3 Urban Road Drainage Runoff

Step 2: Determine Ia/P Ratio

Step 3: Determine Unit Peak Discharge qu

Unit peak discharge is obtained from the following relation:

qu = 0.000431(10) 0 + C1 log(t c ) + C 2 [log(t c )]

Where, qu = unit peak discharge m3/s/km2

Section – 3 Urban Road Drainage Runoff FINAL - November 2004

Step 4: Determine Peak Discharge qp

Where, qp = the peak discharge in m3/s

Step5: Adjustment for Ponding or Swampy Areas (if required)

Where, qa = the adjusted peak discharge in m3/s.

FINAL - November 2004 Section – 3 Urban Road Drainage Runoff

Table 3.4: Runoff Curve Numbers

3.3.3 SCS Unit Hydrograph Method

Section – 3 Urban Road Drainage Runoff FINAL - November 2004

Table 3.5 : Runoff Curve Numbers with Hydrologic Conditions

FINAL - November 2004 Section – 3 Urban Road Drainage Runoff

Section – 3 Urban Road Drainage Runoff FINAL - November 2004

Table 3.7: Coefficients for SCS Peak Discharge Method

Table 3.8 (Fp) for Pond and Swamp Areas

The drainage area should be smaller than 80 hectares.