With the morphological models that are in use today, such as Delft2D-MOR, the scour that is calculated near the tip of hard coastal structures like breakwaters and groynes is often underestimated. This is most likely due to the fact that large horizontal turbulent eddies, which are formed when flow goes round these structures, locally enhance the stirring and horizontal dispersion of sediment. In the present models these turbulent eddies are not resolved, because the time and length scales used are usually too large. Their integrated effect is also not accounted for via 2D-turbulence models. In this study therefore a new version of the morphological model Delft2D-MOR is used in which a recently developed sub-grid scale eddy viscosity model (SGS model) is applied. With this sub-grid scale turbulence closure model and with appropriate small grid cell sizes and time steps large horizontal eddies can be calculated. The modelling of turbulence in this way is called Horizontal Large Eddy Simulation (HLES). The aim of this study is to investigate whether the large turbulent eddies that form when flow passes a coastal structure, are indeed resolved in this new version of the morphological model Delft2D-MOR and what their influence is on the shape, depth and extent of a scour hole. The influence of the large horizontal eddies on the morphology is studied for two test cases: 1. A hypothetical test-case of current induced scour near a groyne. 2. A realistic test-case of tidal induced scour near the moles of IJmuiden harbour. The results from these test-cases show that with the new version of the model large horizontal eddies are indeed resolved, both for the hypothetical test-case and the realistic test-case. From the morphological results from the hypothetical groyne test-case it can be seen that the predicted scour is greater in the case where the large horizontal eddies are resolved compared to the cases where they are not resolved. For the realistic IJmuiden test-case the location and extent of the scour hole are
predicted much more according to reality with the new eddy resolving model compared to the "standard" non eddy resolving model. The results of this study shows that is feasible to apply HLES using Delft2D-MOR, even for a realistic case, including tidal currents, and that the accuracy of predicting long-term morphological changes with this new version of the morphological model Delft2DMOR has increased compared to that of the standard version.