CAPE PENINSULA UNIVERSITY OF TECHNOLOGY
FACULTY OF ENGINEERING & THE BUILT ENVIRONMENT
DEPARTMENT OF CIVIL ENGINEERING AND GEOMATICS
HMG581S HYDROLOGICAL MODELLING TUTORIAL 3
The following maximum annual flood peaks have been extracted from the flow record for a certain
river.
Year Discharge Q (m3/s)
1959/60 191
1960/1 259
1961/2 179
1962/3 283
1963/4 425
1964/5 84
1965/6 245
1966/7 314
1967/8 180
1968/9 391
1969/70 1133
1970/1 289
1971/2 379
1972/3 253
1973/4 337
1974/5 165
1975/6 586
1976/7 348
1977/8 189
1978/9 185
1979/80 157
1980/1 218
1981/2 244
1982/3 411
1983/4 671
1984/5 428
Question 1
Use frequency factors to estimate the 100-year flood peak, assuming each of the following
probability distributions:
a) Normal [5 marks]
b) Log-normal [5 marks]
c) Log Pearson Type III [5 marks]
d) Gumbel [5 marks]
Page 1 of 2
Question 2
Plot the data together with each fitted distribution. Which distribution fits the data best? Estimate
the return period of the highest recorded flood peak. [30 marks]
Question 3
A dam is to be constructed at the site where the record was obtained. The contractor has built a
coffer dam to protect the site against river discharges up to 750 m3/s. What is the probability (risk)
that the construction site will be flooded at least once during the 5-year construction period? Give
some indication of the reliability of this estimate. [10 marks]
Question 4
Discuss the following:
a) What are the assumptions underlying flood frequency analysis? [0 mark]
b) Discuss the relationship between return period (T), risk (R), and design life (N) [0 mark]
c) Discuss the two general methods (and procedure to follow) of flood frequency analysis and
indicate their limitation [0 mark]
You may exclude question 4 in your submission.
Due date: 15 August 2024 @09h50
Page 2 of 2
