Abstract
Various empirical methods have evolved over the years in South Africa to estimate either design floods, design rainfall, catchment response time, and/or Areal Reduction Factors (ARFs). The verification of any empirical method requires the use of observed data not used during the calibration process, while observed data is also required for validation purposes. In the case of ARFs, which are used to convert average design point rainfall depths to an areal (catchment) design rainfall depth, all the calibration/verification data sets remain only estimated sample values of design rainfall. Subsequently, this paper presents an independent application and validation of the regional geographically-centred ARF method (Pietersen 2023) against the currently recommended geographically-centred ARF method (Alexander 2001) by incorporating the different ARF estimates into the Rational Method (RM) to highlight the impact thereof on the resulting flood estimates. In applying a ranking-based goodness-of-fit selection procedure, the RM in combination with the newly derived regional geographically-centred ARF method (Pietersen 2023) resulted in the best deterministic flood estimates when compared to the at-site statistical flood peaks. Apart from the ARFs, catchment response time, design rainfall, and weighted runoff coefficients are all regarded as key input parameters for design flood estimation in ungauged catchments.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call