Waves and water level setup during storms can create overwashing flows across barrier islands. Overwashing flows can cause erosion, barrier breaching, and inlet formation, but their sediments can also be deposited and form washover fans. These widely different outcomes remain difficult to predict. Here we suggest that a breach develops when the sediment volume transported by overwashing flows exceeds the barrier subaerial volume. We form a simple analytical theory that estimates overwashing flows from storm characteristics, barrier morphology, and dune vegetation, and which can be used to assess washover deposition and breaching likelihood. Our theory suggests that barrier width and storm surge height are two important controls on barrier breaching. We test our theory with the hydrodynamic and morphodynamic model Delft3D as well as with field observations of 21 washover fans and 6 breaches that formed during hurricane Sandy. There is reasonable correspondence for natural but not for developed barrier coasts, where traditional sediment transport equations do not readily apply. Our analytical formulations for breach formation and overwash deposition can be used to improve long-term barrier island models.

Along sandy coasts, the seaward expansion of dunes starts with the development of embryonic dunes. Vegetation is crucial for their initiation and the subsequent increase in dune height and volume when sediment supply is sufficient. During severe storms, the plants’ tolerance to and recovery from hydrodynamic disturbances, such as exposure to saline water during high water levels and their (partial) removal during storms, is vital to the long-term (months to years) resilience of the dune building process. Accordingly, areas with high embryo dune abundance have been correlated to wider beaches, attributed both to increased wind-driven sediment supply and increased wave attenuation during storms. Recent observations have shown that alongshore variations in subtidal sandbar morphology may also lead to variations in wave attenuation and foredune erosion, following a series of extreme storms. With our research we aim to determine whether subtidal bar characteristics play a role in long-term (months to years) embryo dune development. We first analysed a data set of 112 annual bathymetric profiles (spaced 250 m alongshore) and topographic (airborne Lidar) measurements in addition to observations of embryo dune presence derived from aerial photographs, spanning 50 km along the Dutch coast from 2010 to 2016. Embryo dune area extraction was done by supervised classification of vegetation pixels on the beach. Using a linear regression model, we found that profiles with a more seaward vegetation extent significantly correlated to shallower subtidal bars, in particular during stormy years. Second, to study the timing and alongshore variability of individual erosion events in more detail, we analysed 10 years (2005-2015) of half-hourly images of a 4-km stretch of the same coast, near Egmond aan Zee, in addition to the annual data. These images provide unique observations of the entire bar-beach-dune system, allowing for the concurrent analysis of bar morphology, embryo dune areas and, crucially, embryo dune exposure to saline water and wave action during storm events. At the conference, we will further explain the observed spatial and temporal (storm-driven) variability in embryo dune development.

Foredune growth results predominantly from sand that is blown from the beach and backshore. Predictions of multi-year potential sand supply that are based on time series of wind speed and direction measured at a regional (offshore or coastal) meteorological station, however, often grossly overpredict measured deposition volumes on the foredune. This is commonly ascribed to supply limiting factors, such as beach surface moisture or shell deposits, or to fetch limitations. Here we show that differences between regional and local (i.e., on the beach) wind characteristics can also contribute substantially to this overprediction. Using wind data collected during a five-week field experiment on a Dutch beach backed by a 20-m high, steep (1:2) foredune we found that the wind speed on the beach is lower and that the wind direction on the backshore is more alongshore than expected from the regional wind data. Both the difference in speed and direction were a function of the regional wind direction, with the largest speed reduction (to about 70% of the regional value) for shore-normal winds and the largest alongshore deflection (about 15 degrees) for shore-oblique winds. When these functional dependencies are applied to a 10-year series of regional wind data, we found that the potential annual onshore sand transport at our site, predicted with the aeolian sand transport equation of Hsu (1971), reduces from 86 to 24 m3/m. The latter is now comparable, although still somewhat higher than the measured annual deposition volume of 10 to 15 m3/m. Further analysis of the computations shows that most of this reduction is due to the difference between regional and local wind speed. In future work we will explore how much of the remaining overprediction is due to surface moisture and fetch limitations.

Natural coastal foredune systems often contain blowouts, through which beach sand is blown into the more landward dunes. Along many developed coasts, blowouts have long been considered a safety hazard, endangering the strength of the foredune as the primary sea defense. Natural blowouts have thus been actively vegetated to promote sand accumulation in the foredune. Recent studies have, however, illustrated that the cessation of sand input in the landward dunes has contributed to a reduction in biodiversity. Nowadays, the safety of coastal foredunes can be assessed with sufficient accuracy allowing for the reintroduction of blowouts. As there is little knowledge on how to optimize blowout layout, the present approach is largely ‘learning by doing’. As a result, many different layouts, varying in blowout width, plan view and orientation, have been adopted in various dune restoration projects. The aim of this study is to model the airflow through an existing man-made blowout and to validate the model results using field observations. We expect that a better understanding of airflow patterns will help in optimizing the design of future blowouts as part of dune restoration projects. The open source Computational Fluid Dynamics (CFD) package OpenFOAM was used to model wind flow through a man-made trough blowout in Dutch National Park Zuid-Kennemerland. The length of the blowout extends roughly 100 m through the foredune; its width narrows from 100 m at the seaward entrance to 20 m at its narrowest part after which it widens again. The deepest part is around 7 m above mean sea level (MSL) while the crest of the surrounding foredune is at 21 m above MSL. The blowout orientation is nearly parallel to the dominant SW wind direction and oblique with respect to the approximately N-S coast line. The field data comprises long-term (many months) observations of wind speed and direction at four locations on the blowout basin and depositional lobe. The model is able to reproduce the observed topographical steering of the wind towards the blowout normal under oblique wind approach as well as the wind-speed acceleration toward the narrowest part of the blowout. Consistent with the observations, the degree of steering and acceleration depend strongly on the wind approach angle, not on the wind speed. As a next step we envision the modeling of different blowout topographies to determine the blowout shape that potentially maximizes the sand transport toward the landward dunes.

Blowouts are characteristic features of many natural coastal foredunes. These dynamic bowl- or trough-shaped depressions act as conduits for aeolian transport of beach sand into the more landward dunes. Along many inhabited coasts foredunes and their blowouts have been planted with vegetation to retain the sand in the foredune, facilitate blowout closure and hence function as sea defense. The resulting vegetated and uniform foredune has, subsequently, contributed to a widespread reduction in the biodiversity of the backdunes. Present-day dune management therefore increasingly involves artificially creating blowouts to maintain and improve backdune biodiversity. The design criteria are high, aiming to postpone or prevent blowout closure as long as possible. Such dune restoration projects often follow a learning-by-doing approach, as information on the underlying aeolian processes, including airflow patterns that steer blowout development, is scarce. Here, we focus on airflow patterns measured in a man-made trough blowout in Dutch National Park Zuid-Kennemerland excavated in winter 2012. The blowout is approximately 100 m long and up to 11 m deep, and has a trapezoidal plan view that narrows from 100 to 20 m in the landward direction. It is approximately aligned with the dominant southwesterly wind direction and hence obliquely with the roughly N-S coastline. Four ultrasonic 3D anemometers, sampling at 10 Hz, were installed in winter/spring 2017 from the mouth of the blowout, across its basin, on to the depositional lobe and have been operational since. The wind recordings at a nearby weather station operated by the Royal Netherlands Meteorological Institute serve as the offshore reference. Wind speed-up through the blowout varied with offshore wind approach angle, and was generally strongest (140%) when the wind was aligned with the blowout axis up to approximately 30° to the south of this axis. Intriguingly, winds approaching with the same angle from the north did not accelerate. We suspect that this asymmetry in speed-up is invoked by the asymmetric blowout shape, with a substantially steeper northern than southern sidewall. Wind deceleration on the lobe was also a function of offshore wind approach angle, with the largest deceleration (40%) for winds approaching from the north of the blowout axis. Winds with approach angles up to 70° were all steered into the blowout, to become approximately aligned with the blowout axis at the landward blowout end. On the lobe, however, the wind closely followed the offshore wind direction. Future work will focus on modelling air flow patterns with computational fluid dynamics, and exploring the relationship between the airflow patterns, blowout morphology and sand transport pathways using additional field observations.