Physical complexity to model morphological changes at a natural channel bend

This study developed a two-dimensional (2-D) depth-averaged model for morphological changes at natural bends by including a secondary flow correction. The model was tested in two laboratory-scale events. A field study was further adopted to demonstrate the capability of the model in predicting bed deformation at natural bends. Further, a series of scenarios with different setups of sediment-related parameters were tested to explore the possibility of a 2-D model to simulate morphological changes at a natural bend, and to investigate how much physical complexity is needed for reliable modeling. The results suggest that a 2-D depth-averaged model can reconstruct the hydrodynamic and morphological features at a bend reasonably provided that the model addresses a secondary flow correction, and reasonably parameterize grain-sizes within a channel in a pragmatic way. The factors, such as sediment transport formula and roughness height, have relatively less significance on the bed change pattern at a bend. The study reveals that the secondary flow effect and grain-size parameterization should be given a first priority among other parameters when modeling bed deformation at a natural bend using a 2-D model.