Graduation of Boudewijn van Heijningen

13 July 2021 15:00 till 17:00 - Location: Online - By: Webredactie | Add to my calendar

The probabilistic assessment of trench sedimentation in sand wave fields: Based on a case study at the Borssele Wind Farm Zone

  • Professor of graduation: Dr. Ir. M.A. De Schipper (TU Delft)

  • Supervisors of graduation: Dr. Ir. O. Morales Napoles (TU Delft), Ir. W.P. de Boer (TU Delft, Deltares), Dr. Ir. B.W. Borsje (Twente University), Ir. A.J.H. Hendriks (Boskalis)

Offshore wind energy plays an essential role in the energy transition from fossil to renewable resources in the Netherlands (Rijkswaterstaat, 2018). Export cables transfer the energy produced at the sites to the onshore grid. The installation of the cables can often not be simultaneously executed with dredging the trenches. A typical period between these operations is one month (Boskalis, 2017). During this period the trench should remain sufficiently open to install the cable, fulfilling the depth requirements. Since large parts of the North Sea are covered by sand wave fields (Krabbendam, 2018), the question arises what the effect of this bedform on the trench sedimentation is. 

 

To assess trench sedimentation, two models were used: the one-dimensional model SedPit and an 2DV-sand wave model. A more elaborate probabilistic assessment, by means of a Standard Monte Carlo (SMC) – analysis, was possible for SedPit, while Latin Hypercube Sampling (LHS) had to be used in the 2DV-model. Based on a case study at the Borssele Wind Farm Zone, the trade-off between model complexity and uncertainty assessment was investigated. As base case, without the sand wave perturbations, it was found that the uncertainties in trench sedimentation are reduced in the 2DV-model compared to the results obtained with SedPit. On the other hand, the range in sedimentation rates obtained by the 2DV-model is fully covered by SedPit. More precise estimations can be achieved by the 2DV-model, but results in an increase of computation time.


Results show that although the migration of sand waves is neglectable on the time scale considered, their effect on trench sedimentation itself is not. Due to an increase of in particular bed load transport, the sedimentation of trenches located near sand wave crests increases significantly. However, located near sand wave troughs, sedimentation rates are decreased. The predicted sedimentation by SedPit is furthermore underestimated near sand wave crests and overestimated close to sand wave troughs. Adjusting the calibration parameter for sand transport within the ranges set by Van Rijn (2013), does not lead to more precise results. For the probabilistic assessment of sedimentation in trenches located close to sand wave crests and troughs, therefore at least the 2DV-model in combination with LHS is required; for other cases SedPit with a SMC-analysis can serve as first estimate.