Sandbar behaviour along a man-made curved coast

Sandbars, submerged ridges of sand roughly parallel to the shoreline, change continuously under time-varying wave conditions. Sandbars migrate onshore and offshore, and meanwhile crescentic patterns may develop or get destructed. So far, most studies have been based on sandbars at straight coasts. Curved coasts impose an alongshore variability in coastline orientation and consequently in the local wave conditions, which in turn may affect bar behaviour. Further insight in the effects of coastline curvature on bar behaviour is highly relevant with the increasing trend of constructing concentrated, km-scale nourishments that locally alter the orientation of the coastline. The aim of this thesis is to improve the understanding of sandbar behaviour along the curved coast of the dynamic, km-scale Sand Engine nourishment (The Netherlands) at daily to seasonal timescales. Accuracies of morphologic data were assessed first. Hereto, bed elevations zb were inverted from X-band radar and optical video with two algorithms differing in wavenumber-frequency retrieval, i.e. a FFT method and a cross-spectral method, respectively. Validation with in situmeasurements shows a systematic bias in bed elevation, i.e. too deep at small depths and too shallow at larger depths. A 2.3 m bias by the radar-FFT approachfor shallow depths (zb⩾−6 m) may relate to an inhomogeneous wave field in the 960x960 m analysis tiles. Next to bed elevation accuracies, accuracies were determined of breaker lines in time-averaged, low-tide video images that approximate the position of the sandbar. Consistent with earlier studies, an O(10 m) accuracy was found. To quantify sandbar behaviour, breaker lines were extracted from a 2.4-year data set of daily, low-tide video images, and their cross-shore position and rate of patterning were determined for a box north and west of the Sand Engine’s tip. Both the northern and western side show a seasonal signal of onshore and offshore migration. The northern side differs from the western side in the shape of the bar, the depth of the bar crest and the alongshore wavelength of the patterns. Besides, timing differences in the formation and destruction of patterns, as well as coupling to patterns in the shoreline, are found between both sides. Those geometric and timing differences seem to relate to alongshore differences in wave characteristics imposed by the curved coast, i.e. breaker wave height and local wave angle, respectively. How a curved coast, through local differences in the wave angle and the resulting flow field, contributes to alongshore variability in crescentic pattern formation was systematically explored with a non-linear morphodynamic model. The alongshore positioning and growth rate of patterns are shown to vary with the local breaker angle. Increasingly oblique angles lower the growth rates as the alongshore currents intensify and wave heights reduce with refraction. Growth rates are higher at strongly curved coasts where the wave climate is low oblique. This implies that rip currents, associated with pronounced patterns and potentially threatening swimmer safety, may increase in number and strength with the implementation of such km-scale nourishments that change the local coastline orientation