Improving the modelling of flow over groynes : modelling of laboratory cases

Groynes are ubiquitous structures present in the Rhine for more than a century. When groynes are submerged during high flow periods, they cause additional friction (or energy loses). Modelling of the energy losses caused by groynes is essential for a proper management of the river system.Several models exist for considering the effect of groynes in the large-scale models employed by Rijkswaterstaat for operational purposes. Nevertheless, there is uncertainty about it and the impact when, for instance, the crest level is lowered or groynes are streamlined. To gain insight into this matter, laboratory experiments are being conducted in the Delft University of Technology where, for varying discharge, velocity and water level are measured in a 5m wide, 30m long flume. In this project we model the laboratory experiments considering groynes as subgrid features, as it is done in the operational models by Rijkswaterstaat. We find that, regardless of the sub-model for computing the energy losses caused by groynes, the energy losses are overestimated, as the predicted discharge along the groyne field is smaller than measured. This result contradicts previous results that showed that energy loses were underestimated. It is also in conflict with the idea that, when using a particular sub-model (i.e., Tabellenboek), the energy losses are always underestimated, as that sub-model was derived from measurements under hydraulically-smooth conditions. Also, the differences between sub-models are significantly smaller than the difference with the measured data. This, together with the fact that the measured total discharge as derived from the cross-sectional measurements varies up to 15 %, points to the need of accurate measurements for a proper model comparison.