Investigating the evolution of gravel bar at river confluence during flood events using a 2D many-fraction river morphodynamic model

The knowledge of river morphology is fundamental and useful information for engineering and habitat restoration purposes. Many interesting phenomena such as armoring and downstream fining significantly affect the quality of riverine habitats. These phenomena could be even more devastating when they occur during the extreme flood events. Therefore, the development of meso-scale bedforms during flood events and the change in their sediment composition have become important topics of study. However, the complex interactions between flood flow, nonuniform particles and sediment transport make these problems difficult to tackle. In this study, we develop a 2D (two-dimensional) many-fraction FE (finite element) morphodynamic model to investigate the evolution of gravel bar during flood events. The proposed model adopts the characteristic dissipative Galerkin (CDG) scheme such that the convection-dominated bar evolution can be computed without numerical instabilities. A two-year record of DEM (digital elevation model) is obtained by airborne Lidar at the confluence of the Xin-Dian River (Taiwan), which is used to verify the model results. The proposed model is further applied to determine the best operation scheme of the Feitsui reservoir for mitigating blockage of river confluence by the gravel bar and sustaining the bed composition favorable to the river biota.