Comparison of flood simulation results using rainfall data sourced at different temporal patterns and spatial scales

Thivanka Dharmasena, Lloyd HC Chua, Hongkui Zhang


Recent trends in the earth’s weather have seen an increase in the occurrence and severity of floods. According to the Intergovernmental Panel on Climate Change, “A changing climate leads to changes in the frequency, intensity, spatial extent, duration, and timing of extreme weather and climate events, and can result in unprecedented extreme weather and climate events”. Climate change can therefore results in sea level rising, as well as increased intensity of storms. This puts a heavy stress on storm drainage systems which may not be designed to cope with such extreme events. The impacts of increased flood peaks and flow volume due to urbanization and population growth are expected to exacerbate the risk of floods in major cities in the future.

Flood prediction in dense urban environments is a difficult task for engineers due to the complex interactions between flow with land form and structures. This study was carried out with the aim of assessing prediction accuracy using rainfall data that spanned across a range of spatial and temporal scales. The professional version of PCSWMM (Stormwater Management Model), developed by Computational Hydraulics International, which incorporates a combination 1-D modelling for storm sewers with 2-D surface flooding was used to investigate historical floods in Geelong, Australia. The study compared rainfall collected from point measurements to spatial patterns of rainfall distribution to differential spatial patterns and point measured rainfall against an ensemble approach to study dependence on temporal behaviour. Three historical floods, in excess of 1 in 50 year events were therefore assessed and simulation results using a single rain gauge measured, 1987 Australian Rainfall and Runoff which is a single design storm approach, 2016 Australian Rainfall and Runoff which is an ensemble design storm approach and weather radar predicted rainfall, were compared against observations. The results indicate that the use of spatial rain data (weather radar acquired rainfall data) did not result in improvements to the single rain gauge approach. Comparisons of the design storm using an ensemble of temporal patterns did not significantly improve predictions using a single rain gauge.

 Click here to download a static PDF version of the presentation.

 Click here to watch recorded presentation on YouTube.

Permanent link: