Seite - 483 - in Book of Full Papers - Symposium Hydro Engineering

The numerical model was validated by using the experimental result provided by Cui et al. (2003) who conducted experiments to investigate how the sediment supplied to a channel as discontinuous pulses were moved to the downstream of the channel. The channel was 45 m long, 0.5 m wide, and 0.7 m deep. The slope was 0.0108, the discharge was 0.009 m3/s, the water depth was 0.0325 m, and the velocity was 0.55 m/s. The sediment pulse, 7.5 cm long and 3.5 cm high, was placed at 8 m away from the inlet of the channel where the sediment was supplied toward the downstream of the channel. The median size of the sediment (D50) was 2.0 mm with the geometric mean of 1.83 mm (Run-2). Fig. 1 shows the change of the sediment discharge over time. After four minutes of the simulation, a large amount of the sediment was discharged from the sediment pulse. As time passed, the sediment discharge from the pulse decreased. The sediment spread to the downstream of the channel. Alternate bars were developed and migrated to the downstream of the channel. The bars were dissipated at one hour and six minutes. This complies with the experimental results. Fig. 2 shows the comparison of longitudinal bed changes between numerical and experimental results. At four minutes, the bed elevation at the sediment pulse decreased dramatically, and the sediment was transported to the downstream of the channel (Fig. 2(b)). At 18 minutes, the bed elevation at the sediment pulse decreased and the sediment was gradually diffused to the downstream (Fig. 2(c)). After more time passed, the bed reached an equilibrium state (Figs. 2(d)). The transport and diffusion of the sediment pulse found in the experiment was well simulated numerically. Fig. 1 Numerical results of sediment discharge changes with time for Run-2: The Tmes are in (hour:min) 483