Applications of SETRAC


The model has been applied successfully to well documented case studies in Austria , Switzerland and France. The SETRAC model has also been applied successfully to replicate Laboratory experiments modelling the deposition behaviour in a bedload retention basin as well as the development of an armour layer.

 Model application at Sessladbach

After the extreme events from August 2005 several torrents were documented. One oft these torrents is the Sessladbach in Tyrol, Austria. The catchment area is about 9.9 km² and the main channel is 3 km long. The mean slope is 0.25, the steepest parts are in the middle reaches with channel gradients up to 0.4. The slope of the fan is 0.18. About 18,000 m³ of bedload were mobilized. The sediment erosion and deposition was mapped in the field in order to be compared with SETRAC calculations. The duration of the whole event was 24 hours with a peak discharge of about 25 m³/s at the fan apex.  For the simulation of a supply limited case, sediment stock was limited by possible erosion depths in the reaches.  Figure 1 shows a comparison of the reconstructed bedload transport and results obtained with SETRAC simulations. Neglecting energy losses due to form roughness yields in an overestimation of the observed bedload transport. All the sediment stock is depleted during the rising limb of the flood hydrograph. Considering form roughness loses, the transported bedload volumes are close to the observations. The temporal evolution of the sediment discharge is also in agreement with the event documentation. Also other simulations of extreme events show the importance of form roughness losses at steep slopes.

 bedload transport comparison

Figure 1: Comparison of modelled and reconstructed bedload transport for the Sessladbach

 Recalculation of laboratory experiments

For the optimisation of a bedload retention basin a physical scale model has been built at the Institute of Mountain Risk Engineering. The sediment deposition was scanned after each run with a 3d Laserscan device. The slope upstream the basin is 4% and in the basin 1%.  It is a Froude scaled model with a physical sale of 1:30. The peak discharge was 27 l/s and the duration was 35 minutes. During the experimental run 695 kg of sediment were deposited in the basin. In order to compare the deposition pattern with SETRAC simulations two longitudinal profiles were taken as reference. Figure 2 shows these profiles and the deposition in the basin after an experimental run. The input hydro- and sedigraphs were discretised by several steps and are also shown.

For the SETRAC simulation cross-sections were taken every 25cm. The sediment transport module in SETRAC was calibrated with the critical Shields parameter. A comparison between the modelled deposition heights and the longitudinal sections is made in Figure 3. Apart from the scour caused by the hydraulic jump at the begin of the basin, which cannot be modelled with the simplified hydraulics of SETRAC, the deposition heights obtained by the numerical model are very close to the observations. Also the deposition volume is in accordance with the measured deposition.

 deposition Experiment

Figure 2: Deposition pattern and input hydrographs for he HQ150 laboratory run

 modelled deposition

Figure 3: Comparison between measured and modelled deposition heights