Process-based cross-shore modelling of barred beaches

Abstract

Coastal profile models are commonly used for hind- and forecasting studies of nearshore bathymetry, often in response to human interference in the nearshore, for instance related to implementation of shoreface nourishments. They are developed to predict wave heights and currents in a line perpendicular to the coast. Based on these wave heights and currents, sediment transport rates are computed and from these the bed level changes are determined. At the present stage of research, the predictive capability of these models generally is rather low in quantitative sense. Actually, these models are still in their infancy. Much of the behaviour of waves, currents and sediment transport near the shore is still unknown. This thesis describes a ‘state-of-the-art’ cross-shore profile model. The model computations are compared with measurements in a laboratory and in nature. From this comparison it appears that waves and currents in te nearshore can be described at least with the same accuracy using a single representative wave as when computing all waves independently. Modelling the undertow velocity under breaking waves remains difficult however. In particular because of the use of the relatively simple linear wave theory to compute part of the undertow. Dally and Brown (1995) have shown that the use of stream function wave theory leads to more accurate predictions of the undertow velocities. Another aspect is the insufficient landward shift of the maximum undertow velocity. The measured maximum undertow velocities are located more shoreward than those computed with the model. Model predictions might improve by making the amount of water that is transported shoreward in a breaking wave travel with the wave somewhat longer than is predicted in the present models. The model reasonably predicts sediment transport rates near the shore. However, the presence of ripples is of crucial importance. In this thesis, computed ripple dimensions from three ripple formulae are compared with the dimensions measured in the laboratory and in nature. The best predictors is adapted to improve agreement with the measurements. Nevertheless, uncertainties remain large. Much is still unknown on the generation of ripples under waves and currents. Predicting nearshore sandbar migration by a coastal profile model remains a challenge. The migration of a sandbar in the laboratory could rather well be predicted, provided that the undertow under breaking waves is accurately predicted and the effect of ripples on the sandbed is taken into account. Predicting nearshore sandbar migration in nature is difficult because of inaccurate predictions of the undertow under breaking waves and uncertainties on the dimensions of sand ripples.