Flow and Sediment Transport in a Laboratory Model of a stretch of the Elbe River

Application Challenge 4-02 © copyright ERCOFTAC 2004

Overview of Tests

The BAW performed measurements in a hydraulic model of the river Elbe between km 506.0 and km 513.0 for various discharge conditions with fixed and movable beds. The horizontal length scale of the model was 1:110 and the vertical scale 1:40. It should be noted that scaling issues are not relevant as CFD is to reproduce the laboratory model situation. The relation of lengths between river and laboratory model can be seen from Fig. 1. In the following, all quantities are at laboratory scale. In the model, the river is 54 m long and 3 m wide. The AC concerns a case with low-water discharge of 12.54 l/s which corresponds to Q = 350m3/s in the natural river, starting from a flat movable bed. The basic curved channel cross-section was trapezoidal with vertical walls; then the groynes were inserted and finally the bottom was covered with a 20 cm thick layer of plastic model sediment material with an initially flat surface. This material has a density of 1053 kg/m3 and a median diameter d50 = 2.1 mm. After switching on the flow, at the inlet 1070 cm3/min of the model sediment was fed in permanently and this was also the mean bed-load transport rate in the channel. The experiment was run for 6 hours and the resulting bed level in the model was measured at about 2000 points by stereo photogrammetry after that time. However, a state close to equilibrium was established already after 2 to 3 hours and the surface velocity was then measured at a number of cross-sections by particle tracking velocimetry (PTV). The cross-sections at which these measurements were taken are indicated in Fig. 1.

Test Case EXP-1

Description of Experiment

Only one series of measurements is available which has been described already in overview section above.

Boundary Data

The topography of the bed at the beginning of the measurements is provided. This consists of the flat model sediment bed and the topography of the groynes and side walls. The bed topography changes and the new topography is given for the equilibrium situation after 6 hours. The roughness of the bed is not known directly so that roughness coefficients (Manning-Strickler coefficients) must be calibrated in CFD calculations (see CFD simulations section). The roughness height of the side walls (also groynes) was about 4 mm. The characteristics of the sediment material have been given in the overview section and also the flow rate and the rate of sediment inflow. The water level at outflow has also been measured. The discharge is given but some assumptions about the velocity distribution at inflow must be made with the aid of available measurements; hence there is some uncertainty about the inflow conditions but on the whole the uncertainty about the boundary data is not large, certainly much lower than for the real river.

Measurement Errors

The possible error in the measurement of surface elevation is of the order of 2 mm. The velocity could be measured by PTV with an accuracy of 5% and the bed profiles could be determined to an accuracy of 1 to 2 mm. Here some uncertainties were introduced by the existence of short-term changes in bed form like dunes moving through the model.

Measured Data

The data that have been measured were mentioned already in the overview section. The topography at the start of the experiment and the grid system for the calculation are given in file grid_bott0.dat. Velocity measurements and bed topography at the end of the experiment are given in files mvel.dat and mbot.dat respectively.