The Sand Engine was constructed in 2011 with the main goal to enhance the natural dune growth. It is the first mega nourishment in the world comprising of 21.5 Mm³ of sediment. There is a large difference between the predictions of the dune growth (25 ha in 20 years) and the actual measurements (1 ha in 4 years). Therefore a more advanced method is needed to predict the dune growth due to the Sand Engine. At the Sand Engine the aeolian sediment transport is determined by the availability limiting processes, e.g. sediment sorting and hydraulic mixing, instead of the wind transporting capacity. AeoLiS is a model that calculates the aeolian sediment transport based on the availability limiting processes. From a sediment budget analysis at the Sand Engine it can be concluded that 58% of all sediments deposited in the dunes originate from the low lying beaches that are regularly reworked by waves. For this reason a model is needed that takes the interaction between both the aeolian and hydrodynamic processes into account.

An Integrated Hydro- Aeolian- Morphodynamic (IHAM) model is used to combine the two processes. The main objective of this research is to verify and applicate the IHAM model applied at the Sand Engine (approx. 8 km of coastline). The IHAM model uses AeoLiS for the aeolian sediment transport, Delft3D Flexible Mesh (FM) for the subaqueous sediment transport and SWAN for the wave transformation calculation. The variables that are exchanged are the wave height calculated by SWAN, the water level calculated by FM and the bed level changes calculated by AeoLiS and FM. When an aeolian and hydrodynamic model are integrated there is a mixed area in which both models are relevant. An additional calibration is required on this mixed area.

If the IHAM model applied at the Sand Engine is compared to the measurements available between 2011 - 2016 a total Brier Skill Score of 0.8 is obtained, which is judged as ‘excellent' according to the classification of Sutherland2004. Especially the tip of the peninsula and the area just north of the tip show large morphological activity that are accurately modelled by the IHAM model. However, the modelled areas south of the Sand Engine and the adjacent coastal areas show larger differences. Overall, the dune growth prediction of the IHAM model compares very well with the measured dune growth between 2012 - 2015. The measured dune growth varies between 14 - 19 m³/m/yr, while the IHAM model computes an average dune growth of 14 m³/m/yr.

If the dune growth prediction of the AeoLiS standalone model is compared to the IHAM model a 3 times smaller dune growth prediction is obtained. For the IHAM model an additional transport from the mixed area towards the dunes is taken into account, due to the inclusion of waves and surge levels, which results in a much better dune growth prediction. Emphasising the need to integrate aeolian and hydrodynamic models.

If the IHAM model, applied at the Sand Engine, is extended to a period of 10 years by repeating the boundary conditions of the first 5 years the IHAM model shows reliable dune growth predictions. After approximately 10 years the IHAM model shows excessive erosion and can no longer be applied. If the IHAM model is applied at the entire Delfland coast, the dune growth prediction of 11 m³/m/yr is close to the measured dune growth of 7.5 ±2 m³/m/yr. However, excessive erosion is observed at the coastal areas away from the Sand Engine.

During the final stage of this thesis it became apparent that the excessive erosion is most likely due to a bug in the bed load transport calculation of FM. The bed load and suspended load transport are in balance at the adjacent coastal areas, resulting in a more or less stable coastline location. At the Sand Engine this balance results in erosion of the peninsula. Therefore a possible bug in FM that undermines the balance and results in erosion, will have a larger influence on the adjacent coastal areas than on the Sand Engine. Despite the bug in FM the IHAM model results of the Sand Engine are very well aligned with the measurements of the first 5 years and show reliable predictions for approximately 10 years. When the bug is fixed the IHAM model will most likely be able to show accurate model results for the entire Delfland coast and for simulation periods longer than 10 years.