Sensitivity of dune erosion to wave obliquity : Comparison of field observations to a numerical model

Large sections of the western Netherlands are protected from flooding by dunes on the Dutch North Sea coast. As part of the national safety assessment of primary flood defenses, the failure probability of dunes due to erosion and secondary failure mechanisms is regularly assessed using numerical models. In the current 1-dimensional assessment method, which is valid for dunes along an alongshore uniform coast, shore-normal waves are assumed. However, for more complex dune coasts, e.g., curved coasts at the heads of barrier islands, the angle of wave incidence will likely vary significantly spatially. Two opposing conceptual models exist regarding the potential effect of oblique waves on dune erosion: 1) refraction of oblique waves leads to lesser wave energy at the dune foot, resulting in lesser dune erosion; 2) oblique waves drive an alongshore current, increasing bed shear and sediment concentrations in the nearshore, leading to greater offshore transport by the undertow and greater dune erosion. The model currently used in the dune safety assessment (XBeach) utilizes numerical relations for (interacting) physical processes that are dependent upon the angle of wave incidence to compute dune erosion. This study aims to assess the ability of XBeach 2DH to support assessment of dune erosion under oblique wave attack, and considers likely consequences of, and key developments for, a future safety assessment methodology for complex dune coasts.

The RealDune/REFLEX field experiment (Van Wiechen et al., 2024a) was used to assess the agreement between simulated and observed dune erosion processes under varying angles of wave incidence. At the scale of individual processes, such as the cross-shore mean current velocity variation and sediment concentration, differences are present between observations and the numerical model, but these differences are not found to correlate strongly with the angle of wave incidence. The sensitivity of simulated dune erosion to the angle of wave incidence in the XBeach 2DH model was found to be similar to that observed during the experiment, with no discernible difference between simulated and observed dune erosion relating to the angle of wave incidence found. Although these findings are the result of the analysis of only one field experiment, they point towards the potential to use the XBeach 2DH model to assess dune erosion under oblique wave angles for the purpose of the safety assessments. Further field validation of dune erosion under varying angles of wave attack is recommended to confirm this finding.

Given previously reported sensitivity of dune erosion to obliquity in XBeach (Den Heijer, 2013; De Winter and Ruessink, 2017; Abdulrahim, 2022), shore-oblique wave incidence on alongshore uniform coasts may increase simulated dune erosion by ~10–30% relative to shore normal wave incidence, rising to 15-45% for curved coasts. Although failure probabilities are site-specific and not quantified explicitly in this study, increased dune erosion will most likely increase the failure probability. Based on estimates from previous related research (Arcadis/Deltares, 2024), the differences in dune erosion described above may likely alter failure probabilities by up to a factor of 2. Inclusion of the effect of the angle of wave incidence in the Dutch safety assessment framework requires three key steps: 1) A reliable and well validated numerical model that is capable of simulating dune erosion under conditions of varying wave obliquity. 2) Quantification of the dune erosion signal due to wave obliquity in the numerical model, and quantification of the uncertainty of this signal. 3) Quantification of the effect of the modelled dune erosion signal due to wave obliquity on the failure probability of dunes and its subsequent effect on the safety assessment.