The influence of storm movement and temporal variability of rainfall on urban pluvial flooding : 1D-2D modelling with empirical hyetographs and CDS-rain

Abstract: Pluvial floods are formed directly from surface runoff after extreme rain events. Urban areas are prone to suffer from these floods due to large portions of hardened surfaces and limited capacity in the stormwater infrastructure. Previous research has shown that catchment response is influenced by the spatio-temporal behaviour of the rainstorm. A rainstorm moving in the same direction as the surface flow can amplify the runoff peak and temporal variability of rainfall intensity generally results in greater peak discharge compared to constant rainfall. This research attempted to relate the effect of storm movement on flood propagation in urban pluvial flooding to the effect from different distributions of rainfall intensity. An additional objective was to investigate the flood response from recent findings on the temporal variability in Swedish rain events and compare it to the flood depths produced by a CDS-rain (Chicago Design Storm), where the latter is the design practice in flood modelling today. A 2D surface model of an urban catchment was coupled with a 1D model of the drainage network and forced by six different hyetographs. Among them were five empirical hyetographs developed by Olsson et al. (2017) and one a CDS-rain. The rainstorms were simulated to move in different directions: along and against the surface flow direction, perpendicular to it and with no movement. Maximum flood depth was evaluated at ten locations and the model results show that storm movement had negligible effect on the flood depths. The impact from the movement was likely limited by the big difference in speed between the rainstorm and the surface flow. All evaluated locations showed a considerable sensitivity to changes in the hyetograph. The maximum flood depth increased at most with a factor of 1.9 depending on the hyetograph that was used as model input. The CDS-rain produced higher flood depths compared to the empirical hyetographs, although one of the empirical hyetographs produced a similar result. Based on the results from this case study, it was concluded that storm movement was not as critical as the temporal variability of rainfall when evaluating maximum flood depth.