How variations of the duration and time to peak of the Chicago Design Storm affect the hydraulic response, as well as the areas contributing to peak runoff, of a synthetic urban catchment area

Sammanfattning: With an expanding urbanization in the world, and thus the expansion of impermeable surfaces, the risk of pluvial floods is an increasing factor that needs to be considered. This, in combination with increasing rain intensities and frequency of rain events indicates a problem both today and for the future. With this in mind, it is an advantage to increase the knowledge of how different variations of extreme rainfall affects the hydraulic response of urban catchments, as well as which areas in urban environments contribute to the flood peak. The aims of this study are, with a particle tracking approach, to investigate how the peak runoff contributing areas differ geographically depending on the duration and time to peak of the rainfall event. This also includes the evaluation of what sizes of urban catchment areas are relevant to include when modelling the hydraulic response of Swedish urban catchment in relation to the characteristics of the hyetograph. The catchment area used in this study is made synthetically to represent a generic Swedish urban catchment with regards to the proportions of hardened surfaces, buildings and low points, as well as the slope of the catchment. Various variants of the Chicago Design Storm were implemented in the model. This included three different durations of 2-, 4- and 6 hours of which each, separately, constituted of three different time to peak that is decided by an r-value when creating the design storms. The r-values used in this study is 0.1, 0.4 and 0.8 where the values correlates to an early-, centred- and late peak of the hyetograph. To be able to investigate the peak contributing area, a particle tracking approach was initially used as an equivalent to tracers where the particles are first evenly distributed over the catchment area to then be concentrated to the locations that shows a variation in in the peak contributing area. This was done by using the modelling program MIKE 21 Flow Model FM powered by DHI, which also was used to run the hydrodynamic simulations of the inundation. The results of the hydrodynamic simulations showed that the rain events generated more runoff as the duration was extended. In addition, the timing of the peak of the rainfall intensity also had an impact on the result as the runoff increased with increasing r-value. Thus, as the peak of the hyetograph is delayed, it imposes an increasing risk of severe flooding. Furthermore, with the use of particle tracking, it could be concluded that the different design storm had an influence on the peak contributing distance where the distance grew larger when the duration of the rainfall event was extended and when the peak of the storm was delayed.