3D numerical modelling of asynchronous propagation characteristics of flood and sediment peaks in three gorges reservoir

Abstract

The asynchronous propagation characteristics of flood and sediment peaks downstream (i.e., sediment peak lagging behind flood peak) are crucial for the operation of sediment peak release to reduce the sedimentation of the Three Gorges Reservoir (TGR). In the flood season of 2013, the sediment peak regulation was conducted in the TGR, based on the asynchronous propagation characteristics of flood and sediment peaks, and promising results were achieved towards reducing the reservoir sedimentation. In this study, the processes of flood propagation and sediment transport over a stretch of 280 km, upstream of the TGR on the Yangtze River were studied through a three-dimensional (3D) semi-implicit cross-scale hydroscience integrated system model (SCHISM), focussing on the asynchronous propagation characteristics of flood and sediment peaks downstream. It was demonstrated that firstly, the model efficiently and accurately reproduced the processes of flood propagation and sediment transport during the 2013 flood season compared with the measured results. The sensitivity analysis showed that the horizontal grid used in the numerical modelling had a greater influence on the model results than the vertical grid. The Manning coefficient was not sensitive to flood peaks propagation with a large flow discharge. Next, the validated model was adopted to investigate the asynchronous propagation characteristics of the flood and sediment peaks under different TGR operational scenarios. The results indicated that the impounded level ahead of the Three Gorges Dam (TGD) had an insignificant influence on the occurrence time and magnitude of the flood peak, but had a significant influence on the occurrence time and magnitude of the sediment peak. The time of sediment peak lagging behind flood peak increases as the impounded level in the TGR increased. The peak value of sediment concentration gradually decreased owing to the sediment deposition. The research results could provide a scientific basis for further optimising the sediment peak regulation, and the 3D numerical model could provide a reliable real-time tool to predict the arrival time of the flood and sediment peaks in the TGR.