Urban Flood Monitoring Using an Integrated River Basin-Urban Flood Modeling Approach

Abstract:

Abstract: Urban flooding has been increasing steadfastly and caused steeply rising losses/damages during the recent decades. Reliable urban flood modeling, with capability in forecasting of scalable inundation information, is in great demand in contemporary urban planning and hazard management. Water level of river channels running through urban area defines critical boundary condition for urban flooding development and evolution, as the rivers can be source or sink of urban inundation water. However, the dynamic interactions between natural river network and urban drainage network (particularly pipe system) are usually not explicitly presented at the river basin scale in current urban flood models. In this study, the Storm Water Management model (SWMM) is adopted for rainfall-runoff simulation for urban area, forming a module integrated into the Dominant river tracing-Routing Integrated with VIC Environment (DRIVE) model framework which provides extensive modeling of hydrological processes for the river basin upstream/downstream to the urban area. The coupled model (referred to as DRIVE-Urban) deals with the challenges in delineating the complex responses of the combined drainage networks from natural and urban environment to extreme rainfall. The DRIVE-Urban model is tested in two severe flooding cases caused by short-time heavy rainfall and Typhoon individually at Haikou, the largest urbanized city in Hainai Province, China. The results show the model successfully captures 61.5% and 69.0% of the monitored road inundations respectively for the two cases. The model derives a mean relative error of -13.8% for the mean depth of the inundated area and 4.2% for inundation extent, which is promising, while potentials to improve are also indicated.

Keywords: Hydrological model; urban flood; SWMM; inundation algorithm;