Paper Title
Evaluation of Flood Discharges in Tsitsa River: South Africa

Abstract
The design of large hydraulic structures must consider both magnitude and frequency of extreme floods expected in a given catchment so as to minimise risk of structural failure. While probable maximum flood forms the basis of flood evaluation, the choice for the exact magnitude of flood needs to weigh economic factors as well. Flood evaluation for dam design in South Africa entails evaluation by five main methods for return periods ranging from 2 years to 200 years. Flood magnitudes from the five methods are compared against each other based on professional judgment and prior knowledge of the catchment. The design flood is taken as the mean value of the best results. Since most dams have catchments far in excess of 15 km2, the rational method is used mostly to validate results from the other methods. This study investigated the performance of the rational method, standard design flood, and regional maximum flood. The standard design flood and the regional maximum flood were consistently in agreement for all the return periods considered. The rational method performed satisfactorily for long return periods, but tended to overestimate the flood for shorter return periods. Keywords - Rational Method, Regional Maximum Flood, Standard Design Flood