Rational Monte Carlo method for flood frequency analysis in urban catchments

SummaryThe January 2011 flash flood at Toowoomba, Australia was substantially larger than other high ranking observed floods. Flood frequency analysis (FFA) is often performed with limited hydrological data so the occurrence of an exceptional flood provides valuable data. Peak 1 h rainfall intensities during the storm varied from <10 to >1000 year ARI across the catchment. Average recurrence interval (ARI) estimates of the resulting streamflow peak discharge were 220 year ARI for a General Extreme Value distribution fitted to the Annual Series and 450 year ARI based on a log–Pearson 3 distribution. An independent method referred to as the Rational Monte Carlo method (RMC) was developed in order to provide an independent check of the ARI estimate. The RMC method is a simple derived distribution approach where the Rational equation links observed rainfall intensity at a reference pluviograph to the peak flood discharge. The RMC 2011 flood ARI estimate was 475–515 years, comparable with the log–Pearson 3 method. Although the proposed method has limitations, the RMC approach shows promise as an alternative and independent FFA method for urban catchments. For all methods, the ARI estimates were sensitive to whether the January 2011 peak discharge was included in the analysis. Overall, the study highlights the inherent difficulty in extrapolating ARI estimates beyond the range of the available historical record.

► Rational Monte Carlo method proposed for flood frequency analysis in urban catchments.
► Alternative method for Annual Series flood frequency analysis demonstrated for 2011 Toowoomba flash flood.
► Difficulty in extrapolation of flood probability estimation beyond historical record is demonstrated.