Chapter Eight

Frequency
Analysis
Introduction to frequency Analysis
 Flow frequency
 Flood probability
 Risk, Reliability and Safety factor
Introduction
Water resource: systems must be planned for future
events for which no exact time of occurrence can
be forecasted.
Hence, the hydrologist must give a statement of the
probability of the stream flows
These probabilities are important to the economic
and social evaluation of a project.
For major projects, the failure of which seriously
threatens human life, a more extreme event, the
probable maximum flood, has become the standard for
designing the Hydraulic structures.
 :
Frequency analysis is the hydrologic term used
to describe the probability of occurrence of a
particular hydrologic event (e.g. rainfall, flood,
drought, etc.)

For planning and designing of water resources

development projects, the important parameters
are river discharges and related questions on the
frequency & duration of normal flows
Frequency analyses try to answer the following
problems:
(1) Given n years of daily stream flow record for stream X,
what is the maximum (or minimum) flow Q that is likely to recur
with a frequency of once in T years on average?

(2) What is the return period associated with a maximum (or

minimum) flow Q ?
Given n years of stream flow data for stream X and L years design life
of a certain structure, what is the probability P of a discharge Q being
exceeded at least once during the design life L?.
Hydrologic extremes
Extreme events
 Storm
 Floods
 Droughts
 One of the major problems in hydrologic design is the
estimation of maximum floods.
These estimations are used to design hydrological and
hydraulic dimensions of
 spillways
 bridges and sewers,
 dams and protection embankments,
 diversion canals, etc.
Accurate estimation of flood frequency discharge increases
safety of the structures
..
 Hydrologic design scale

Hydrologic design scale – range in magnitude of the

design variable within which a value must be selected
Design considerations
 Safety
 Cost
Do not design small structures for large peak values (not
cost effective)
Do not design large structures for small peak values
(unsafe)
Balance between safety and cost.
 Hydrologic design level – magnitude of the hydrologic event
to be considered for the design of a structure or projects.
 Three approaches for determining design level
 Empirical/probabilistic
 Risk analysis
 Hydro economic analysis
Probability
The probability P(A) of an event A is the relative number of occurrences
of event A after a very large number of trials.
It will be convenient to define the sum of the probabilities of all the
possibilities as one.

S (P(X ) = 1
i i

This means that a probability can range from zero to one

0≤P≤1
The T-year Flood
We talked about an event, the T-year flood.
For example, we talked about a 100-year flood in
Arba Minch
The probability of the event T is 1/T
For example, in any year, the chance of getting the
100-year flood is 1/100
The probability that the event T does not occur is
1- 1/T
If p is probability of success, then (1-p) is the
probability of failure.
Hydro-economic Analysis

 Probability distribution of hydrologic event and damage

associated with its occurrence are known
 As the design period increases, capital cost increases, but the
cost associated with expected damages decreases.
 In hydro economic analysis, find return period that has
minimum total (capital + damage) cost.
 Selection of Data
In order to have meaningful estimates from flood frequency analysis,
The data to be analyzed describe random events.
The data is homogeneous.
The population parameters can be estimated from the
sample data.
It is of good quality.
 If the data available for analysis do not satisfy any of the listed
assumptions, then the estimates are not considered reliable.
 Moreover, the data should be
(i) relevant, (ii) adequate, and (iii) accurate. For flood frequency analysis,
either annual flood series or partial duration series may be used.

13
 Relevance; implies that the data must deal with the
problem
 Adequacy; refers primarily to length of record,. The
observed record is merely a sample of the total
population of floods that have occurred and may occur
again. If the sample is too small, the probabilities derived
cannot be expected to be reliable
 Accuracy; refers primarily to the problem of
homogeneity
FLOW DURATION CURVES. (FDC

FDC is graphical representation of a given discharge
against % of time that the discharge is exceeded

The FDC only applies for the period for which it was
derived.

If this is a long period, say more than 10 to 20 years,
the FDC may be regarded as a probability curve or
flow frequency curve,

which may be used to estimate the percentage of
time that a specified discharge will be equaled or
exceeded in the future.
 The over all slope of the FDC is an indication of the flow
variability in the stream.
 Specific uses that have been made of duration curves
are
 Assessing the hydropower potential of run-of-river
plants
 Determining minimum flow release
 Water quality studies
 Sediment yield studies
 Comparing yield potential of basins
:
 :

The shape of the flow-duration curve gives a good

indication of a catchment’s characteristics response to its
average rainfall
steeply historycurve results from a very variable
sloped
discharge,
very flat slope indicate little variation in flow regime,
Probability analysis seeks to define the flood flow
: of p being equaled or exceed in
with probability
any year.
The probability of each events can be calculated
as follow P 
m
N 1

The recurrence interval T, (also called return

period) of frequency is calculated as follow
1
T 
P
Consider, for example, a list of flood magnitudes of a
river arranged in descending order as shown in Table
The length of record is 50 years.

20
 Theoretical Distributions of Floods
:
Chow has shown that most frequency-distribution functions
applicable in hydrologic studies can be expressed by the following
equation known as the general equation of hydrologic frequency
analysis:

Where xT = value of the variate X of a random hydrologic series with a

return period T,
x- = mean of the variate,
σ = standard deviation of the variate,
K = frequency factor which depends upon the return period, T and
the assumed frequency distribution.
 Parameter estimation is important in flood
frequency analysis. Any set of flood occurrences
can be described using some statistical
distribution.
 Statistical distribution may be mathematically
represented through functional relationships
which may contain a set of parameters.
 The well-known parameter estimation techniques
 Analytical/ Frequency factor method

Chow proposed using:

where

xT x  KT s
xT Estimated event magnitude
K T Frequency factor
T Return period
x Sample mean
s Sample standard deviation
Methods of Analytical Frequency
Analysis

 when larger extrapolations of T are involved,

 Gumbel extreme-value,
 Log-Pearson Type III, and
 log normal distributions have to be used.
Extreme-Value Type I Distribution (Gumbel’s Method)

This extreme value theory of Gumbel is only

applicable to annual extremes.
In the Gumbel method the data are ranked in
ascending order and it makes use of the
probability of non-exceedence
q=1-P
The return period T is therefore given by
T = 1 / P = 1 / (1-q).
The :
original Gumbel equation describes the probability of non-
exceedence (q)
q=exp[-exp{-y}]

applicable to an infinite sample size (i.e. N

→ ∞).
 Gumbel’s Equation for Practical use
The above equation giving the value of the variate x with recurrence
interval T is used as
xT  x  K n  1
Where = standard deviation of the sample

 n 1 
( x  x) 2
N1
K= frequency Factor expressed as

yT  y n
K 
sn
In which yT = reduced variate, a function of T and given as
:
yT   [ln . ln T
T1 ]
Reduced mean, a function of sample size N and is
yn  given in table
(For N  , yn  0.577 )
Standard deviation, a function sample size N
and is given in table
(For N  , y  1.2825 )
 Table : Reduced mean yn in Gumbel's extreme value
:
distribution, N = sample size

N 0 1 2 3 4 5 6 7 8 9

10 0.4952 0.4996 0.5035 0.507 0.51 0.5128 0.5157 0.5181 0.5202 0.522

20 0.5236 0.5252 0.5268 0.5283 0.5296 0.5309 0.532 0.5332 0.5343 0.5353

30 0.5362 0.5371 0.538 0.5388 0.5396 0.5402 0.541 0.5418 0.5424 0.543

40 0.5436 0.5442 0.5448 0.5453 0.5458 0.5463 0.5468 0.5473 0.5477 0.5481

50 0.5485 0.5489 0.5493 0.5497 0.5501 0.5504 0.5508 0.5511 0.5515 0.5518

60 0.5521 0.5524 0.5527 0.553 0.5533 0.5535 0.5538 0.554 0.5543 0.5545

70 0.5548 0.555 0.5552 0.5555 0.5557 0.5559 0.5561 0.5563 0.5565 0.5567

80 0.5569 0.557 0.5572 0.5574 0.5576 0.5578 0.558 0.5581 0.5583 0.5585

90 0.5586 0.5587 0.5589 0.5591 0.5592 0.5593 0.5595 0.5596 0.5598 0.5599

100 0.56
 Table : Reduced standard deviation Sn in Gumbel's extreme
value distribution, N = sample size
0 1 2 3
N 4 5 6 7 8 9

0.9496 0.9676 0.9833 0.9971

10 1.0095 1.0206 1.0316 1.0411 1.0493 1.0565

1.0628 1.0696 1.0754 1.0811

20 1.0864 1.0915 1.0961 1.1004 1.1047 1.1086

1.1124 1.1159 1.1193 1.1226

30 1.1255 1.1285 1.1313 1.1339 1.1363 1.1388

1.1413 1.1436 1.1458 1.148

40 1.1499 1.1519 1.1538 1.1557 1.1574 1.159

1.1607 1.1623 1.1638 1.1658

50 1.1667 1.1681 1.1696 1.1708 1.1721 1.1734

1.1747 1.1759 1.177 1.1782

60 1.1793 1.1803 1.1814 1.1824 1.1834 1.1844

1.1854 1.1863 1.1873 1.1881

70 1.189 1.1898 1.1906 1.1915 1.1923 1.193

1.1938 1.1945 1.1953 1.1959

80 1.1967 1.1973 1.198 1.1987 1.1994 1.2001

1.2007 1.2013 1.202 1.2026

90 1.2032 1.2038 1.2044 1.2049 1.2055 1.206

1.2065 07/02/2009 Bedlu.G

100
 Procedur :
e
These equations are used under the following 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  n  1 and
x x. Find
for the given data.
2 Using a table determine yn sand
n appropriate to given
N. yT
3 Find for a given T.
4 Find K.
x
5 Determine the required T
 Example 1
:
Annual maximum recorded floods in a certain river, for the period
1951 to 1977 is given below.
Year Max.Flood(m3/s) Year Max.Flood(m3/s) Year Max.Flood(m3/s)
1951 2947 1960 4798 1969 6599
1952 3521 1961 4290 1970 3700
1953 2399 1962 4652 1971 4175
1954 4124 1963 5050 1972 2988
1955 3496 1964 6900 1973 2709
1956 2947 1965 4366 1974 3873
1957 5060 1966 3380 1975 4593
1958 4903 1967 7826 1976 6761
1959 3757 1968 3320 1977 1971
Verify whether the Gumbel extreme-value I distribution fit
the recorded values.
Estimate the flood discharge with return period of (i) 100
years and (ii) 150 years by graphical extrapolation.
 Solution
:
The flood discharge values are arranged in
descending order and the plotting position return
period TP for each discharge is obtained as
Tp=(N+1)/M=(27+1)/M=28/M
Where
m = order number. The discharge magnitude Q
can be plotted against the corresponding TP on a
Gumbel extreme probability paper.
The statistics x(mean) and σn-1(Stan.Dev) for
the series are next calculated and are shown in
table below.
Flood Discharge(m3/s) Rank(m)

7826 1
: P=M/(N+1)

0.035714286
Tp=(N+1)/M

28
Flood Discharge(m3/s)

3873
Rank(m) P=M/(N+1) Tp=(N+1)/M

15 0.535714286 1.86666667

6900 2 0.071428571 14 3757 16 0.571428571 1.75

6761 3 0.107142857 9.333333333 3700 17 0.607142857 1.64705882

6599 4 0.142857143 7 3521 18 0.642857143 1.55555556

5060 5 0.178571429 5.6 3496 19 0.678571429 1.47368421

5050 6 0.214285714 4.666666667 3380 20 0.714285714 1.4

4903 7 0.25 4 3320 21 0.75 1.33333333

4798 8 0.285714286 3.5 2988 22 0.785714286 1.27272727

4652 9 0.321428571 3.111111111 2947 23 0.821428571 1.2173913

4593 10 0.357142857 2.8 2947 24 0.857142857 1.16666667

4366 11 0.392857143 2.545454545 2709 25 0.892857143 1.12

4290 12 0.428571429 2.333333333 2399 26 0.928571429 1.07692308

4175 13 0.464285714 2.153846154 1971 27 0.964285714 1.03703704

4124 14 0.5 2
N 27
mean 4263
Stan. Dev. 1433
 Using these the discharge XT for some chosen return
:
interval is calculated by using Gumbel's formulae .
From Table for N = 27, yn = 0.5332 and Sn = 1.1004.
Choosing T = 10 years,

yT  [ln . ln TT 1 ] yT = -[ln(ln(10/9))] = 2.25037

yT  y n K = (2.25307-0.5332)/1.1004 = 1.56
K
sn

xT  x  K  n  1 xT = 4263 + (1.56*1432.6) =
6499m3/s.

x
Similarly, values of T are calculated for
two more T values as shown below.
 T (Years)
XT [obtained by above Eq.
(m3/s)
5 : 5522
10 6499
20 7436

When these values are plotted on Gumbel probability paper, it is

seen that these points lie on a straight line according to the property
of the Gumbel's extreme probability paper.
Then by extrapolation of the theoretical xT Vs T relationship, from
this plot, at
T = 100 years, xT = 9600m3/s and at
T = 150 years, XT = 10700m3/s.
[By using above Eq., x100 = 9558m3/s and
x150 = 10088m3/s.]
07/02/2009 Bedlu.G
 Confidence Limits for the fitted
data
The value of the variate for a given return
period XT determined by Gumbel's method can
have errors due to the limited sample data
used .an estimate of the confidence limits
of the estimate is desirable.
The confidence interval indicates the limits
about the calculated value between which the
true value can be said to lie with a specific
probability based on sampling errors only
For a confidence probability c, the
: interval of the variate xT is
confidence
bound both value x1 and x2 given by
x1/2 = xT± f (c) Se
Where
f(c) = function of the confidence probability c
determined by using the table of normal variate as
 Example 2:
:
Data covering a period of 92 years for a certain river yielded the
mean and standard deviation of the annual flood series as 6437
and 2951 m3/s respectively. Using Gumbel's method, estimate the
flood discharge with a return period of 500 years.
What are the
(a) 95% and
(b) 80% confidence limits for this estimate?
 Solution:
: for N = 92 years,
From Table
yn = 0.5589, and Sn =1.2020. Then
y500 = -[ln((ln(500/499))] = 6.21361
K500 = (6.21361 - 0.5589)/1.2020 =
4.7044,
X500 = 6437 + 4.7044*2951
=20320m3/s. s  probable..error b  N
e
n 1

From above Eq. b  (1  1.3K  1.1K )

2

b=√(1+1.3*4.7044+1.1*4.7044^2)=5.61
Se=5.61*(2951/ √92)=1726
 (a) For the 95% confidence probability f(c) = 1.96 and
:
x1/2 =20320xT ± (1.96*1726)
which results in x1 = 23703m3/s and
x2 = 16937m3/s.
Thus the estimated discharge of 20320m3/s has a 95%
probability of lying between 23700 and 16940m3/s.
(b) For 80% confidence probability, f(c) = 1.282 and
x1/2 = 20320xT ± (1.282*1726), which results in
x1 = 22533m3/s and
x2 =18107m3/s.
Thus the estimated discharge of 20320 m3/s has an 80%
probability of lying between 22533 and 18107m3/s.
 Exercise(Quiz)
Flood frequency computations for the river X by using
Gumbles method, yielded the following results:
 :
II Log-Pearson Type III Distribution
In this distribution the variate X is first transformed into logarithmic
form and the transformed data is then analyzed.
Z = log x
For this z series, for any recurrence interval T,

Where
Kz = a frequency factor which is a function of
recurrence interval T and the coefficient of skew
C s,
the corresponding value of xT is obtained by
: xT = antilog(zT)
Generally a minimum of 30 years of data is considered
as essential in order to use flood frequency analysis
:
 Example :
For the annual flood series data given in above 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.
 : Solution:
The variate z = log x is first calculated for all the discharges in table below.
Then the statistics z , σz and Cs are calculated from above table to obtain

z = 3.6071
σz = 0.1427
Cs = (27*0.0030)/(26)(25)(0.1427)3 = 0.043

The flood discharge for a given T is calculated as below. Here, values of Kz

for given T and
Cs = 0.043 are read from table .
Year Flood x (m3 /s) z = log x Year Flood x (m3/s) z = log x
1951 2947 : 3.4694 1965 4366 3.6401
1952 3521 3.5467 1966 3380 3.5289
1953 2399 3.38 1967 7826 3.8935
1954 4124 3.6153 1968 3320 3.5211
1955 3496 3.5436 1969 6599 3.8195
1956 2947 3.4694 1970 3700 3.5682
1957 5060 3.7042 1971 4175 3.6207
1958 4903 3.6905 1972 2988 3.4754
1959 3751 3.5748 1973 2709 3.4328
1960 4798 3.6811 1974 3873 3.588
1961 4290 3.6325 1975 4593 3.6621
1962 4652 3.6676 1976 6761 3.83
1963 5050 3.7033 1977 1971 3.2947
1964 6900 3.8388
 T Kz (from Kzσz ZT = z + XT =
(years)
:
table for
Cs =
Kzσz antilog ZT
(m3/s)
0.043)

100 2.358 0.3365 3.9436 8782

200 2.616 0.3733 3.9804 9559

1000 3.152 0.4498 4.0569 11400

Drought analysis
• Type of draught
 Meteorological
 Agricultural
 Hydrological
The objective of drought analysis is to characterize the
magnitude, duration, and severity of meteorological,
agricultural, or hydrological drought in a region of interest.
Drought can be characterized by the following terms

• Duration, D = length of period for which X <X0;

• Severity, S = cumulative deviation from X0;
• Intensity (or magnitude), I = S/D.
Risk, Reliability and Safety factor
Risk and Reliability: The designer of a hydraulic structure
always faces a nagging doubt about the risk of failure of his
structure.
 The probability of occurrence of an event (x≥xT) at least
once over a period of n successive years is called the risk,
R.
R= 1 - (Probability of non-occurrence of the n event in n years).
 1
Re 1  R  1  
The reliability Re, is defined as  T
 Example
A bridge has an expected life of 25
years and is designed for a flood
magnitude of return period 100 years.
(a) What is the risk of this hydrological
design? (b) If 10% risk is acceptable,
what return period will have to be
adopted?
 Solution:
(a) The risk R for n = 25 years and T = 100
years is:

(c) say 240 years. Hence to get 10%

acceptable risk, the bridge will have to be
designed for a flood of return period T = 240
years.
 EXAMPLE
A cofferdam has been built to protect homes in a floodplain until a
major channel project can be completed. The cofferdam was built
for the 20 –yr flood event. The channel project will require 3-yr to
complete. What are the probabilities that:
(a) The cofferdam will not be overtopped during the 3 yr (the
reliability)?
(b) The cofferdam will be overtopped in any one year?
(c) The cofferdam will be overtopped at least once in 3 yr (the
risk)
(d) The cofferdam will be overtopped only in the third year?
Solution:
(a) Reliability = (1 – 1/T)n = (1 – 1/20)3 =
0.86
(b) Prob = 1 /T = 0.05

(d) Risk = 1 – Reliability = 0.14

(e) Prob =(1-p)(1-p)p =0.95 2. (0.05) = 0.045