Study of Sediment Impacts on Flood Risk by Numerical Modeling

Abstract

Sediments seriously increase flood risk. The lack of sediment data, as well as uncertainty in sediment yield and transport modeling, hampers the incorporation of sediment into flood risk assessment/early-warning systems. In this study, we targeted a 10km reach that experienced a flash flood with high sediment inflows in 2019 and simulated sediment impacts on flood propagation and flood risk using a 1D hydraulic model. Two sediment inflow types were considered: sudden inflow from a debris flow and constant inflow from upstream. For the former type, different inflow times before, during and after the peak discharge were tested; for the latter type, different sediment concentrations and medium grain sizes were tested in the scenario design. Our results showed that when sediments inflowed into the reach in a short time, they greatly increased the water level by up to 3.7m in the reach near the inflow point (i.e., 500–700m upstream and downstream of the point). When sediments constantly inflowed from upstream, they influenced the water level of the whole reach but raised the water level by up to 1.2m in the cross-sections, which showed strong bed aggradation. Sudden sediment inflow influenced the reach near the inflow point and might increase its flood risk to the highest level, whenever it occurred before, during or after the peak discharge. Constant sediment inflow affected little the flood risk of the peak discharge but increased the risk afterward when the sediment concentration was higher than 40 kg m -3 . Cross-sections showing increased flood risk were characterized by decreasing bed slopes or increasing widths. In view of the impacts of different sediment inflow types on flood risk, we developed a method to identify potential sediment risk. Together with current risk assessment by water flow, we proposed a framework to incorporate sediment into flood risk assessment.

Keywords: Flash flood risk, Sediment supply, Channel morphology, Q-W relation, Hydrodynamic modeling