Prepared by:

Engr. Marlowe L. Labusnog

By the end of this lecture, the student must be
able to:
 estimate flood discharge on a catchment
using rational method;
 apply the unit hydrograph to estimate peak
flood flows; and
 apply statistics for flood frequency analysis.
 A flood is any high flow that overtops either natural or
artificial embankments along a stream.
 Estimating the magnitude of flood peak discharge:
1. Rational method
2. Empirical method
3. Unit – hydrograph technique
4. Flood – frequency studies
• Assumptions:
1. The computed peak rate of runoff at the outlet
point is a function of the average rainfall rate
during the time of concentration, that is, the peak
discharge does not result from a more intense
storm of shorter duration, during which only a
portion of the watershed is contributing to runoff
at the outlet.
2. The time of concentration employed is the time
for the runoff to become established and flow
from the most remote part of the drainage area to
the inflow point of the sewer being designed.
3. Rainfall intensity is constant throughout the storm
duration.
Source: Subramanya (2008)
 Rational Formula:

where: Q p = peak discharge

C = average runoff coefficient of the catchment
𝒏
𝒊 𝟏 𝒊 𝒊
𝒏
𝒊 𝟏 𝒊
i = rainfall intensity
A = watershed area

where: Q p is in m3/s, i in mm/hr, and A is in hectares.

Source: Chow, et al. (1988)
Source: Chow, et al. (1988)
 Kirpich (1940)

where: t c = time of concentration, min

L = length of channel/ditch from headwater
to outlet, ft
S = average watershed slope, ft/ft

 California Culverts Practice (1942)

where: L = length of longest watercourse, mi

H = elevation difference between divide and
outlet, ft
 Izzard (1946)

where: i = rainfall intensity, in/h

c = retardance coefficient
c = 0.0070 for very smooth pavement; 0.012
for concrete pavement; 0.06 for dense turf
L = length of flow path, ft
S = slope of flow path, ft/ft
Note: The product of i times L should be ≤ 500.
 Federal Aviation Administration (1970)

where: C = rational method runoff coefficient

L = length of overland flow, ft
S = surface slope, %
 Kinematic wave formulas Morgali & Linsley
(1965); Aron & Erborge (1973)

where: L = length of overland flow, ft

n = Manning roughness coefficient
i = rainfall intensity in/h
S= average overland slope, ft/ft
Source: Chow, et al. (1988)
 SCS lag equation (1973)
𝟎.𝟕
𝟎.𝟖

𝒄 𝟎.𝟓
where: L = hydraulic length of watershed (longest flow path), ft
CN = SCS runoff curve number
S = average watershed slope, %
Note: Adapted to small urban basins under 2000 acres
Equation assumes that t c = 1.67 x basin lag
 SCS average velocity charts (1975, 1986)

where: L = length of flow path, ft

V = average velocity in feet per second from Fig.
3-1 of TR 55 of various surfaces
Source: Chow, et al. (1988)
 A 500ha watershed has the land use/cover and corresponding runoff
coefficient as given below:

 The maximum length of travel of water in the watershed is about 3000m and
the elevation difference between the highest and outlet points of the
watershed is 25m. The maximum intensity duration frequency relationship of
the watershed is given by
𝟎.𝟏𝟓𝟐𝟑

𝟎.𝟗𝟒𝟓
where i = intensity in cm/h, T = return period in years and D = duration of the
rainfall in hours. Estimate the (i) 25 year peak runoff from the watershed and (ii)
the 25 year peak runoff if the forest cover has decreased to 50ha and the
cultivated land has encroached upon the pasture and forest lands to have a total
coverage of 450ha.
1. Gumbel (Extreme Value Type I) Method
2. Log – Pearson Type III Distribution
3. Normal Distribution
4. Log – Normal Distribution
5. Gamma Distribution, etc.
 Probability is empirically defined as the number of times a
specific event occurs from the total number of events
measures.
 Flood Probability – probability analysis seeks to define the
flood with a probability P of being equaled or exceeded in
any year.
 Formula:
1. P = 1/T
2. P = m/(N+1) or T = (N+1)/m
where:
P = the probability that a certain event (flood flow) is
equaled or exceeded in any year
T = return period or recurrence interval
= the average interval in years within which a given
event will be equaled or exceeded
N = period of record
m = the number of times the mth largest flood in the
data series has been equaled or exceeded.
 The probability of occurrence of an event equal to or
larger than a value x0 is
𝒚

where

where
= mean
= standard deviation of the variate X
 In practice it is the value of X for a given P that is
required and as such the equation is transposed as

 Noting that the return period T = 1/P and designating yT =

the value of y, commonly called the reduced variate, for a
given T

or
 The value of the variate X with a return period T is

Note: The equation above is of the same from as the general equation
of hydrologic frequency analysis.
where

 The equations above constitute the basic Gumbel’s

equations and are applicable to an infinite size (i.e.
N→∞).
 For a sample size, N, the Gumbel’s equation is given by

where
σn-1 = standard deviation of the sample of size N

K = frequency factor

yT = reduced variate which is a function of T

= reduced mean (function of sample size N); N→∞, →0.577
Sn = reduced standard deviation (function of sample size N); N→∞,
Sn→1.2825
 Procedure to estimate the flood magnitude corresponding to a given
return period based on an annual flood series.
1. Assemble the discharge data and note the sample size N. Here the
annual flood value is the variate X. Find and for the given
data.
2. Using Tables 7.3 & 7.4 determine and appropriate to given N.
3. Find for a given T.
4. Find the frequency factor K.
5. Determine the required flood magnitude .
Source: Subramanya (2008)
 If flood flow on a large watershed has an average value of 1200
cubic meters per second with a variance of 62,500 (cubic meters per
second)2, what is the probability that a flood will be equaled to or
exceed 2000 m3/s?
Solution:
𝒆 𝒚
where:

𝒆 𝟒.𝟔𝟖𝟏 𝟑
and

𝟑
 Annual maximum recorded floods in the river Bhima at Deorgaon, a
tributary of the river Krishna, for the period 1951 to 1977 is given
below. Verify whether the Gumbel extreme-value distribution fit the
recorded values. Estimate the flood discharge with recurrence
interval of (i) 100 years and (ii) 150 years by graphical extrapolation.

Source: Subramanya (2008)

Source: Subramanya (2008)
1. A hydraulic structure on a stream has been designed for a
discharge of 350 m3/s. If the available flood data on the stream is
for 20years and the mean and standard deviation for annual flood
series are 121 and 60m3/s respectively, calculate the return period
for the design flood by using Gumbel’s method.
2. In a frequency analysis of rainfall based on 15years of data of
10minutes storm, the following values were obtained:
Arithmetic mean data = 1.65cm
Standard deviation = 0.45cm
Using Gumbel’s extremal distribution, find the recurrence interval of a
storm of 10 minutes and depth equal to 3.0cm. Assume the sample size
to be very large
3. For a data of maximum-recorded annual floods of a river the mean
and the standard deviation are 4200m3/s and 1705m3/s
respectively. Using Gumbel’s extreme value distribution, estimate
the return period of a design flood of 9500m3/s. Assume infinite
sample size.
 For this method, the variate is first transformed into logarithmic form
(base 10) and the transformed data is then analyzed.

𝑻 𝒛 𝒛
where:
Kz = frequency factor w/c is a function of recurrence interval T and the
coefficient of skew Cs
σz = standard deviation of the Z variate sample
𝟐
𝒛

Cs = coefficient of skew of variate Z

𝟑
𝒔 𝟑
𝒛
= mean of the z values
N = sample size = number of years of record
Source: Subramanya (2008)
Source: Subramanya (2008)
 Sometimes, the coefficient of skew Cs, is adjusted to
account for the size of the sample by using the following
relation proposed by Hazen (1930).

 However, the standard procedure for use of log-Pearson

Type III distribution adopted by U.S. Water Resources
Council does not include this adjustment for skew.
 When the skew is zero, i.e. Cs = 0, the log-Pearson Type
III distribution reduces to log normal distribution.
 For the annual flood series data given in the Gumbel’s
distribution example, estimate the flood discharge for a return
period of (a) 100 years (b) 200 years and (c) 1000 years by
using log-Pearson Type III distribution.
 The probability of occurrence of an event (X ≥ XT) at least once
over a period of n successive years is called the risk, . Hence,

where:

 The reliability Re, is defined as

𝒏

𝒆
 The safety factor of any water resource development project for
a given hydrological parameter M is
𝒂𝒎
𝒎
𝒉𝒎
where:
(SF)m = Safety factor (for the parameter M; i.e. flood discharge
magnitude, max. river stage, reservoir capacity & free
board)
Cam = Actual value of the parameter M adopted in the design of the
project
Chm = Value of the parameter M obtained from the hydrological
considerations only
Cam – Chm = safety margin
 A bridge has an expected life of 25years and is designed for a
flood magnitude of return period 100years. (a) What is the risk
of this hydrologic design? (b) If a 10% risk is acceptable, what
return period will have to be adopted?
 Analysis of annual flood series of a river yielded a sample
mean of 1000m3/s and standard deviation of 500m3/s.
Estimate the design flood of a structure on this river to provide
90% assurance that the structure will not fail in the next
50years. Use Gumbel’s method and assume the sample size
to be very large.
Bedient, et al. (2013). Hydrology and Floodplain Analysis.
Pearson Education Limited. England.
Chow, et al.(1988). Applied Hydrology. McGraw-Hill Book
Co. Singapore.
Mays, L.W. (2011). Water Resources Engineering. John
Wiley & Sons, Inc. USA.
Subramanya, K. (2008). Engineering Hydrology. Tata
McGraw-Hill Publishing Company Limited. India.
HEC-HMS Technical Reference Manual