Timing of aeolian sand transport events on a narrow beach

Abstract

Modelling aeolian transport in coastal zones is important to understand aeolian transport dynamics and it allows us to secure future stability of dunes as natural coastal defences. Nonetheless, modelled aeolian transport rates based solely on wind-data largely overpredict the actual transport and hence deposition on the foredune. The Coastal Group of Utrecht University has developed a new aeolian transport model, including the most important features suggested by previous studies that limit aeolian transport. These features include wave-runup, storm surges, surface moisture content and more realistic flow patterns of the wind on a beach (local wind conditions). The aim of this study is to evaluate and validate the performance of the model for its compatibility with actual transport events, with a focus on the timing of the model. The intensity of actual transport events was visually quantified with the use of high quality photos taken from the Coast3D tower, located at the beach of Egmond aan Zee. The intensity was quantified based on the size of moving dry patches of sand, the formation of streamers and the formation and movement of sand strips. In total there are five classes defined whereby class 0 indicates no aeolian transport, class 1 indicates a low intensity and the transport intensity gradually increases up to class 4, indicating an extremely high transport intensity. Based on a combination of wind speed and the classified transport intensity, three types of transport are distinguished. The first type of transport event is an unlimited transport event, meaning that the aeolian transport is equal to the wind potential. The second type of transport event is a limited transport event, whereby the transport rate is lower than the wind potential. The last type of transport event is an inhibited transport event, defined as events with no aeolian transport. For each of these transport events, the outcome of the model is evaluated. The model performed correctly for 60% of the time for the unlimited aeolian transport events and limited aeolian transport events. For the other 40% of the events the model generated a zero-transport rate, because the wind speed or wind direction did not exceed the implemented threshold. For the inhibited transport events, the transport rates were strongly reduced which is an important improvement, however even with this strong reduction, the transport rates remain too high to represent actual transport rates. For limited and inhibited events, the moisture content was the most important limiting factor. Nonetheless, the largest improvement for all transport types (unlimited, limited and inhibited) in reducing the overprediction in transport rate, was the result of using local wind conditions. Overall, the timing of the model for generating unlimited, limited and inhibited transport rates has improved greatly compared to the models based only on wind data. Further improvements can be accomplished by re-evaluating the calculations of the minimum width required for maximum transport (critical fetch) and re-evaluating the values for the wind speed and wind speed direction threshold.