Quantification of maintenance dredging uncertainty on the Waal

Large-scale river projects are executed on the Dutch rivers to meet increased flood conveyance standards. Because the projects bring variation to the standardised river profile, they are expected to induce increased river dynamics. The consequences are higher dredging costs and more disruption to inland navigation. All river intervention plans have been optimised to meet the dredging guidelines of Rijkswaterstaat, but it is important not to look at just the cumulative effect of individual projects but also at the combined effect. Furthermore, the dredging volumes are calculated in a deterministic way: the river discharge is simulated by means of a single representative yearly hydrograph that is repeated for every year in the simulation. The uncertainties in dredging volumes in time and in space are not known this way. Understanding the stochastic response of the required dredging can help the river manager better attend the hindrance for inland navigation. It will help the river manager to decide where and how to interfere in the river system. This study is conducted to determine the increase and uncertainty of the dredging demand on the Waal River, before and after the execution of the Room for the River programme. Numerical calculations with a 2D depth averaged Delft3D model are executed to determine the dredging effort and navigability. To include uncertainty in the model response a Monte Carlo Simulation is performed with discharge variation on the upstream boundary. The simulations conducted in this study show that the increase in dredging is less than can be expected based on the impact assessment studies of the individual river intervention works in the Room for the River programme which stresses the need for integrated impact assessment. The stochastic approach shows that the increase is relatively small compared to the uncertainty in dredging, but nevertheless significant, because considerable increases are predicted at locations that are already heavily dredged. The simulations shows that the dredging capacity on the Waal is insufficient to handle sudden large dredging peaks. If a flood event on the Waal is followed by a rapid fall of the discharge this leads to disruption to navigation for long periods of time. These peaks would require an unrealistically large dredging capacity so river management will have to consider structural measures to reduce the dredging effort, or to accept a lower quality of navigation. Structural measures are expensive and it is therefore illogical to try to mitigate all dredging work by structural measures. The dredging demand is not distributed evenly along the river: It is almost permanent at some locations and more sporadic at others. Locations that show structural dredging can be considered for locations to execute structural measures. Such location are predominantly found on the large river bends in the Boven-Waal. The information acquired from in this study can be used to develop mitigating measures to limit the maintenance dredging effort and the hindrance for navigability. In designing measures to mitigate the dredging effort, an emphasis should be put on more sustainable ways to maintain the river; The lay-out of the Dutch river system should be designed in such way that it keeps itself navigable. The results of the simulations put the use of a single representative hydrograph up for discussion. The hydrograph in the deterministic calculations is yearly repeated, so the natural yearly variance is not taken into account and the results of the deterministic method are also not representative for the Monte Carlo Simulation.. Information acquired from the Monte Carlo Simulations can be used to develop better representative hydrographs; The hydrograph needs to be as detailed as possible to better simulate reality.