Hybrid dunes are a coastal flood defense structure that combines the sandy and wave- dissipating capacity of natural dunes with the robustness of hard structures. Their increasing application in coastal environments highlights the need for a better understanding of how these systems respond to storm forcing. This study investigates the morphodynamic behavior of hybrid dunes under storm conditions through a large-scale field experiment at the Sand Engine (NL), aiming to improve predictive insight into erosion mechanisms and structural performance. High-resolution LiDAR, pressure sensors, and acoustic instruments were deployed across four engineered dune configurations, ranging from a fully sandy dune to hybrid setups with embedded sea dike and seawall cores, to monitor morphological changes and hydrodynamic forcing during five consecutive storms.
The main research question is: What are the dominant erosion mechanisms and structural influences of embedded hard elements in hybrid dunes during consecutive storm events?
The findings show that hybrid dunes transition through three consistent erosion phases: (1) initial sand-dominated retreat characterized by notching, slumping, and offshore sediment transport, (2) reduced erosion once hard elements become partially exposed, and (3) structure-dominated stability after full exposure, during which landward retreat halts. Core geometry strongly influenced behavior: The Dike-in-dune (S1) supported smoother energy dissipation and profile adjustment, while the wall-in-dune (S4) halted retreat more abruptly but caused localized vertical beach erosion. In contrast, the Sandy dune (S2) continued to retreat by over 7 m, with cumulative erosion volumes exceeding 11 m3 /m. Erosion volumes correlated significantly with 20 min time-averaged waterlevel, measured 80 m seaward of the dune foot. Field observation identified the total water level as the dominant driver of erosion. Once exposed, hard structures reduced wave run-up and changed energy distribution, influencing sediment mobility. Oblique wave conditions likely disrupted longshore sediment supply, especially toward unarmored sections, amplifying erosion near structural transitions. These interactions underscore the need to jointly assess cross- and longshore processes when evaluating hybrid dune behavior. Key design factors include sand cover thickness, core geometry, and especially transition zone reinforcement. Structural flanking and undermining were critical failure mechanisms, as seen in the collapse of the Dike-in-Dune and Dike sections during the fourth storm. These results highlight the importance of integrated design strategies addressing not only structural shape, but also lateral sediment continuity and toe stability.
Hybrid dunes shift erosion from retreat-driven to structure-constrained behavior, offering localized resistance under storm conditions. However, their long-term performance depends on robust, systemscale design, balancing sand volume, structural exposure, and sediment pathways to deliver adaptable and resilient coastal protection.

This study focuses on the infragravity (IG) waves at 5 measurement sites (Brouwershavense Gat 2 (BG2), Oosterschelde 4 (OS4), Cadzand, Hansweert and Bath) in the Scheldt region, examining their characteristics in these waters as well as to what extent they can be modelled using SWAN (Simulating WAves Nearshore). To facilitate this, data collected during the occurrence of four storms (Corrie, Dudley, Eunice and Franklin) that traversed the North Sea in January and February 2022 has been used.

A spectral analysis has been used to provide insight into the IG and sea-swell (SS) wave field. To this extent, the significant wave height for the IG (0.005 – 0.04 Hz) and SS (0.04 – 0.33 Hz) frequency bands have been derived, in addition to the storm averaged wave period. Measurement device BG2 documented the largest storm averaged significant wave height for both the IG and SS frequency bands (0.256 m and 3.46 m during storm Corrie, respectively). The measurement device with the smallest storm averaged significant waveheight values was Hansweert, which measured 0.024 m and 0.217 m for the IG and SS wave frequencies, also during storm Corrie. Most of the wave energy for the measurement devices in the Western Scheldt, Hansweert and Bath, can be attributed to waves with frequency values larger than the upper boundary for SS waves (T < 3 s). The fraction of the total wave energy contained within the IG and SS frequency bands ranged from 0.1 to 0.56 for these two measurement devices. Conversely, the IG and SS frequency bands contain 0.81 to 0.96 of the total wave energy for measurement device Cadzand, BG2 and OS4. The correlation between the IG and SS significant wave heights is mostly strong for the BG2, OS4 and Cadzand measurement devices (0.657 to 0.956) and generally moderate for Hansweert and Bath (0.474 to 0.87), with one outlier equivalent to 0.128). A bispectral analysis was used to offer more insight regarding the components contributing to the total IG wave field. The results at Bath proved to be erroneous for all storms except Corrie, as the derived free IG (FIG) contribution was negative. The origin remains unclear, although it appears to be related to unexplained spurious bursts in the wave elevation signal. The bispectral analysis for the remainder of the locations lead to maximum TIG wave heights ranging from 0.05 m to 0.12 m for Hansweert, 0.15 m to 0.39 m for BG2, 0.11 m to 0.35 m for OS4 and 0.21 m to 0.34 m for Cadzand for the 4 storms. The contribution of the bound IG (BIG) and FIG energy at Bath and Hansweert appears to fluctuate heavily, undergoing rapid changes on an hourly basis. For BG2, OS4 and Cadzand, the FIG energy contribution is strongly dominant as it generally amounted to > 0.75 of the TIG energy.

The capabilities of the SWAN model were evaluated by simulating FIG waves in the North Sea basin. The model relates incident SS wave energy to reflected FIG wave energy with FIG source lines based on the reflection parametrization of Ardhuin et al. (2014). The hourly SS wave inputs were provided by Copernicus Marine Service. The predictive skill was used to qualitatively assess to what extent the model is able to reproduce FIG waves in the Scheldt region. The model was unable to correctly predict the FIG waves Hansweert and Bath, where the predictive skill values for all storms ranged from 0.0003 to 0.0009 for Hansweert, and were 0 for Bath. Better predictive skill values were obtained for BG2 (0.5044 to 0.6565), OS4 (0.3327 to 0.4293) and Cadzand (0.6466 to 0.7510), but there is still room for improvement. The lacking skill values can be attributed to not enough FIG wave energy being able to penetrate the estuarine waters. Reflective lines were implemented along the coastline of the Western Scheldt in an attempt to force more FIG energy into the estuary. This only led to a minor increase for the predictive skill values (O(0.002) for Hansweert, 0 for Bath and O(0.0001) for BG2, OS4 and Cadzand) but from a spatial standpoint, it appears that there are locations in the estuary where the absolute increase was O(0.02), which is significant as there are areas in the estuary where the significant FIG wave height has doubled as a consequence of the implementation of the reflective lines.

Sea dikes have long been one of the main coastal defense infrastructures in the Netherlands, protecting against flooding and storm surges. As climate change intensifies severe phenomena and sea level rises, dikes must be heightened to protect the coast efficiently against future threats. Meanwhile, nature-based solutions are emerging as viable alternatives. One of these solutions is the use of salt marshes as coastal protection in front of dikes instead of heightening the dike, which could potentially lead to a more cost-effective solution. A salt marsh vegetated foreshore not only dampens the incoming waves but also adapts to the advancing waters of the sea level rise by trapping sediment.
This study’s main objective is to assess the efficacy of salt marshes in mitigating wave run-up and overtopping on a real-scale sea dike with a vegetated foreshore.
To carry out these large-scale experiments were carried out at the Deltares Delta Flume. Experiments were carried out for low (0.75 m), medium (1.5m) and high (2.5m) water depths above the foreshore, designed to simulate the extreme storm scenarios experienced by the Friesland dikes in the northern Netherlands. In addition to these conditions, the tests were performed for three different qualities of vegetation, good damaged, and mowed vegetation. First, the experiments ran for a fully vegetated foreshore, then after several runs the vegetation was assumed damaged as an important part of the biomass had eroded. Finally, the vegetation is mowed to leave a bare foreshore which is the baseline scenario.
The acquired run-up results are compared with empirical equations from literature. In addition, the results were compared with those of similar research on salt marshes. Finally, the other important objective of this study is to quantify the reduction in wave loads according to vegetation conditions by introducing certain damping ratio parameters.
To measure the important parameters for the completion of this study, three wave gauges offshore and three wave gauges close to the toe of the dike were used for the measurement of the wave characteristics. For the wave run-up a camera was placed above the dike slope to capture the run-up events over time, and at the same time a LIDAR laser scanner recorded the same slope to collect data on run-up and overtopping.
From the wave analysis, the spectral parameters offshore and at the toe of the dike were calculated. From this, it was possible to have a first estimate of the wave attenuation by comparing the incident significant wave height at these two locations. The results reveal a 14 to 30 % decrease in wave height for low conditions. For medium and high storm conditions the attenuation was decreased to a range between 5 to 15% and 4 to 19% respectively.
The video process run-up measurements were in good agreement with the literature with RMSE = 0.18 - 0.19 m. An additional method was used to validate the results of the run-up. The detection of the run-up was completed visually, by tracking the waves that exceed the markers on the slope, for part of three experiments to compare their signal with the signal derived from the video process. The bias between the signals ranged from 0.017 to 0.022 m, indicating a strong agreement between the methods. One of the most important findings from this study is that the reduction in run-up is not directly attributed to the presence of vegetation itself, but rather to the role played by the significant wave height at the toe of the dike. The run-up was reduced between 2 and 16% for low storm conditions and a reduction between 4 and 24% for medium storm conditions, with an average value of 10.3%. At the same time, the reduction of run-up from damaged vegetation to fully vegetated foreshore is up to 9% with an average of 7.7%. The laser scanner was also used to measured the wave run-up. The signal obtained from the laser is compared with the signal of the visual detection method and reveals a bias between 0.032 and 0.037 m. The run-up 2% results were also compared with the same results from the video camera, revealing a 0.14 m (7%) deviation which may be due to the accuracy of the laser.
The laser scanner also obtained the overtopping results using the virtual overtopping method. From this method, the virtual volume and the virtual overtopping discharge can be calculated and then compared with the equations from the literature. The comparison reveals a strong agreement between the calculated and predicted values, which proves that the equations predict accurately the overtopping discharges for the case of a living dike. For low storm conditions, a decrease of 2 to 50 % of the maximum virtual overtopping volumes is possible, while for medium conditions a maximum volume decrease of 5 to 54% according to the crest height. For the highest storm conditions, this volume decrease is calculated to be between 20 to 28%.

In 1990 the Dutch government passed legislation that dictates the South West Texel coast must be maintained with regard to the 1990 coastline. The coast is currently maintained by applying shoreface and beach nourishments with an interval of approximately 3 years, because it experiences erosion. The problems with this approach are that the mobilization of the dredging vessels is expensive, dredging vessels cause nuisance in the ecological system and moreover, the dredging vessels emit carbon dioxide and therefore contribute to climate change. The first goal of this study was to to investigate if an alternative coastal protection strategy is able to mitigate the drawbacks of the current scheme. Consequently, the application of a mega nourishment was chosen over a cross-shore dam or an outer delta nourishment. To optimally design a mega nourishment, a morphological analysis was executed that focused on elements in the sediment balance of the system. Using Vaklodingen and JARKUS data, volume changes of each transect were identified for 4 parts of the transect: the dune, the beach, the shallow part of the foreshore and the Molengat channel. It was found that dunes started growing from the moment that the nourishments were structurally applied. When nourishments are aborted, it is expected that the shrinking of the dunes will resume, just like the pre-1990 situation. The Molengat in the transects was found to be moving towards shore before the nourishment program started, and its landward movement was not influenced by the implementation of the nourishment program. The Molengat filling rate was projected into the future, leading to a fill date in transects 880-930 by 2035, and in transects 930-1013 by 2055. In transects 1013-1108 and 1108-1210 the Molengat has already been filled. The late projection filling date of transects 930-1013 is a reason to put the mega nourishment at that location, combined with the largest natural erosion occurring at the same location. The application of a mega nourishment is done by putting 20 years worth of sediment at transect range 945-1053 in one construction project. Simultaneously, the mega nourishment aims to diffuse sediment alongshore in the shallow part of the transects, providing sediment to adjacent transects. The sediment supply to other transects could be able to limit the shrinking of the adjacent dunes, or even let them grow. However, this requires further investigation. Delft3D Model results showed that sediment transport in this area is mainly caused by waves on the NUN-shoal, and by the current in the Molengat ebb-tidal channel. The model resolution was not fine enough to accurately describe the accretion and erosion patterns alongshore. Yet, by the use of results of morphological updates it could be determined that the mega nourishment will likely cause accretion south of the nourishment area. This research can be used as a stepping stone for future design steps of a mega nourishment, and also for acquiring knowledge about future behaviour of different parts of the bottom profiles in the South West Texel area.

Living Breakwaters have evolved from traditional breakwaters to create multi-purpose structures that provide environmental and social benefits. These breakwaters promote a healthy habitat for flora and fauna, while achieving the same structural capabilities as conventional structures. Reefy has developed a modular living breakwater that can serve both as an artificial reef and a stable breakwater. This study aims to model the impact of a Reefy breakwater on the transmitted wave height, using a process-based numerical model, namely XBeach.

This study builds upon a previous one conducted by van den Brekel [2021], where scaled experiments were performed in a physical wave flume, to investigate the hydrodynamic and ecological functionalities of the Reefy breakwater. In the wave flume experiments a total of 15 structure configurations together with 35 wave conditions, both regular and irregular, were tested. The first 7 of the configurations were 2D structures with a relatively simple shape and a porosity of 20%, and the rest were complex 3D structures with a porosity above 20%. The wave flume was recreated within the numerical model. The influence of the breakwaters on the wave heights are examined through the comparison between the calculated transmission coefficient of the wave flume and the numerical model. However, XBeach, and more specifically XBeach non hydrostatic+ mode which was used, does not resolve the complex interaction between waves, friction and porosity of the structures. As a result, a simple method had to be found to compensate for the loss of these phenomena, while successfully calculating the transmission coefficient.

The findings of this analysis demonstrate that treating the breakwater as an impermeable change in bathymetry, coupled with a 15% reduction in structure height, produces the desired outcomes. From the 35 wave conditions only the 8 irregular wave conditions were chosen to be of interest due to time constraints. Among these 8 wave conditions that were used as an input in XBeach, only 5 of them were satisfying the criterion that describes the range for which the numerical model should produce accurate results (kd<2). The transmission coefficient was defined as the ratio of the transmitted wave height behind the breakwater, and incident wave height in front of the structure. The wave decomposition method used was a modified Guza split method. In pursuit of finding a valid solution, the breakwaters were modeled firstly as impermeable structures with a decreased height, secondly as impermeable structures with a decreased width, thirdly as vegetation with the help of the vegetation module, and lastly the maximum wave steepness criterion as applied in the model was increased (maxbrsteep up to 1.4), without modifying the structure height.

Among the suggested solutions and after being validated for 13 experiments with both simple and complex structures, decreasing the width did not achieve the desired results while the vegetation module failed to produce consistent outcomes. As far as transmission coefficient is of interest, a 15% decrease in the structure height had the smallest mean absolute percentage error of almost 10% and a root mean square error of 8.2%. On the other hand, choosing to increase the maximum wave steepness (to a value of 1.4) is not a valid option for the complex 3D structures, but illustrates even better behavior than a change in the structure height when simple structures with 20% porosity are concerned. To summarize, this thesis provides alternatives to replicate in a simplified way the impact of porous structures, such as Reefy breakwaters, without the need of an extensive and time expensive numerical model.

Coastal dunes serve as the primary defence mechanism against coastal storms for many coastal communities around the world. Vegetation plays a role in increasing dune resiliency as it enables dune growth, however not enough is known about its effects during storms. Research has shown that the current climate crisis will increase the intensity of coastal storms as the sea level rises and global temperatures increase. It is therefore of utmost importance to understand the effects of vegetation on coastal dunes. Particularly with current trends of nature-based solutions to plant vegetation in dune restoration projects.

In this master thesis, the impact of Hurricane Ian on two barrier islands is analyzed and modelled. The primary focus of the study was to investigate how dune vegetation influenced the erosional effects of the storm, with a particular emphasis on enhancing the existing methodologies for incorporating vegetation into morphological models such as XBeach. The research findings derived from the data analysis revealed that the most resilient dunes are high, broad and have a dense vegetation coverage. Moreover, the model outcomes highlight the substantial improvement in predictive accuracy achieved by integrating vegetation as a bed roughness coefficient within the XBeach model. Adding vegetation to the model directly influences current velocities, but does not affect water levels, wave heights, or infragravity waves. The primary influence of vegetation becomes pronounced when an island is inundated or breached, significantly reducing the currents and sediment transport caused by water level gradients. An additional effect of vegetation was observed in a comparative analysis between vegetated and unvegetated models. This analysis highlighted a delay in dune crest lowering when vegetation was present, showcasing the importance of vegetation in shaping the response of barrier islands during storm events.

Numerical modelling can help understand the complex processes that shape barrier islands during storms. This research emphasized the necessity of including vegetation in XBeach models to enhance their predictive capabilities. The best predictions occur when high-resolution bathymetric data is combined with land use land cover (LULC) data to include vegetation as a constant bed roughness parameter. Furthermore, reducing the land classes to four different ones based on dune vegetation zones improves the results of the model and facilitates the calibration of bed friction coefficients.

The most effective models applied in this study demonstrated impressive skill, ranging from good to excellent, accurately predicting breaches in the precise areas they occurred for both islands under investigation. This research contributes to the continued improvements of modelling with XBeach and provides a detailed method of analyzing the effects of dune vegetation on dune erosion to determine the impact of Hurricanes in coastal dunes.

The coastal safety of the dunes are of high importance for low laying countries (Chen et al., 2017). This is especially true for the Netherlands as it is used officially as a safety measure against flooding from extreme weather since 1984 (Rijkswaterstaat, 1989). As of current 1D modelling of the waves is an acceptable method for the Dutch coast to assess the safety of the dunes (Den Heijer, 2013). For assessing the dune coastal safety the 98t h percentile of the runup R2% (Stockdon et al., 2006) is used for defining the wave response during a storm on the dunes. The current dune coastal safety studies consider bound waves (infragravity and sea- swell waves) propagating from offshore to nearshore. However, due to the Dutch coast being located at a very large semi-enclosed basin additional free infragravity waves can be present during a storm next to the bound waves that are reflected from other coasts (Daouk, 2020) as leaky waves (Herbers et al., 1995), and potentially increasing the R2% magnitude. Here we show that the R2% with the additional free infragravity waves in the wave domain can have a significant response on the dune coastal safety. The additional free infragravity waves can have tens of centimetres of R2% magnitude increase that is dependent on the slope of the domain. For the additional free infragravity waves a higher R2% is observed for steeper slopes, while a larger R2% growth is observed at the milder slopes when the free infragravity waves are initially introduced to the slope. The dune coastal safety assessments conducted thus far do not consider the effects from leaky waves from other coasts, therefore underestimating the safety provided from the dunes against flooding. With the research conducted by considering additional free infragravity waves with the traditionally bound waves originating in a storm, a better insight can be provided for dune coastal safety. As a result, companies and governmental bodies such as Rijkswaterstaat in the Netherlands that assess the coastal safety of the dunes can benefit by having a more complete representation of the coastal hydrodynamics that can as a result prevent flooding or provide dune nourishment after the episodic event (Sallenger, 2000).

San Andrés is a Colombian-Caribbean Island located 800 km from the Colombian coast. On its Eastern side there is a barrier reef formation protecting the island from offshore incident waves. Due to the protected environment created by the coral reef, sandy beaches can be formed on the East side of the island.
During normal conditions, waves come predominantly form the East (90% of the wave climate), have significant wave height of 2 m and period of 8 s. Besides, San Andrés is situated on the Caribbean hurricane route, which can cause an enormous damage to the island. The storm season at San Andrés is between October and December, which is also when major erosion events take place.
The economy of San Andrés is mostly built upon tourism, specially related to its biodiverse ecosystems and Caribbean beaches. The island’s ecological environment is composed by mangroves, seagrasses, and coral reefs, attracting a wide spectrum of fauna and flora to its ecosystems.
During the Masterplan for Coastal Erosion (PMEC), San Andrés was pointed out as a location in which coastal erosion is problematic. In a follow-up of this Masterplan, the island was elected to be part of a program in which solutions against coastal erosion would be presented. This research is part of this project, as a parallel trajectory to get a more profound understanding of the system and the possible mitigation measures that could be applied on the island.
With increasing urbanization and frequency of extreme weather events, erosion is becoming a problem with which San Andrés and its residents are repeatedly having to deal. Erosion is specially problematic for the Northern part of the island, called Spratt Bight. This region is not only the most densely populated area of the island, but also economically and touristically very important. Its beach presents periodically eroding patterns during storm seasons, when wave action drives the sediment towards the East, decreasing its beach width almost to none. A decreasing beach width has a direct negative impact on tourism, making coastal erosion in Spratt Bight not only a coastal safety problem, but also an economic issue.
This study aims to look into the main hydro- and morphological processes driving coastal erosion in Spratt Bight and, using the Building with Nature philosophy, propose a set of solutions to mitigate this problem. To reach this objective data analysis and literature research has been carried out, after which different environmental conditions were modelled using the numerical model Delft3D.
During these activities it was found that independently of its direction, waves approaching San Andrés break upon the coral reef and induce a water level set up inside the coral lagoon. The difference in water level in- and outside the lagoon generates a current and sediment transport, which is directed towards the western opening in the coral reef.
When the Northern waves approach the island (1.5% of the wave climate), the same water level set-up phenomenon is observed. However, as waves are approaching form the North, they not only break upon the reef, but are also able to enter the sheltered lagoon through the western opening in the coral reef. These waves are able to bend around the reef reaching the shore and the headland on the Northern part of the island, inducing a longshore current and a sediment transport that is southeastward directed. The result is that Northern waves are mostly responsible for a strong westward and erosive sediment transport pattern. These waves are mostly observed between October and March, which coincides with the storm season in San Andrés.
Besides, it was found that the Eastern waves are responsible for restoring the (dynamic) equilibrium profile of Spratt Bight Beach. However, this restoring force has a less strong intensity, taking more time to restore the beach than to disrupt its equilibrium.
The solutions proposed include seagrass restoration to enhance ecology, restrain sediment transport and attenuate wave heights; the beneficial reuse of dredged material, to nourish Spratt Bight Beach; and finally, the implementation of artificial coral reefs as breakwaters to prevent the newly nourished sediment to be lost from the system. Besides, artificial coral reefs enhance the ecosystem by attracting fauna and flora increasing biodiversity. All proposed solutions have a positive impact on the beaches and therefore on tourism and the economy of the island. This makes them multifunctional solutions, serving the main goal of protecting the beach while at the same time creating benefits for other functions and values in the area. Following in this way the prescriptions of the Building with Nature design approach by van Eekelen and Bouw (2020).

Estimating wave runup is of crucial importance during a coastal risk assessment. Currently, runup levels are usually calculated using empirical formula derived from experiments conducted at beaches all over the world, and during different beach states. During this research, conditions are analyzed where runup reaches the dune toe and above. The main objective of this research is to examine the changes in the runup characteristics, which are the wave setup, the infragravity swash motions (low frequency wave motion) and the incident swash motions (high frequency wave motion). In order to examine these conditions, a fieldwork experiment has be conducted at the Sand engine near Kijkduin (South Holland). a dune of 150 (m) in width and 5.5 (m) above NAP (Nieuw Amsterdams Peil) have be constructed near the waterline. Instruments have been installed which have monitored the incoming wave conditions and the behavior of the hydrodynamic conditions at the waterline. One of these instruments was an instrument new to measurements conducted at the coast. This was a LLC (Line scanning Low Cost) LiDAR, which was used to track the runup during extreme wave conditions.
The LLC LiDAR was evaluated in terms of measuring the nearshore hydrodynamics by comparing the water level measured by the LLC LiDAR to the water level measured by a pressure sensor. The water levels from both instruments compared well with an 𝑅2 ranging from 𝑅2 = −0.86. The match in the peaks of the water levels were lower, showing 𝑅2 in the range of 0.34 − 0.78.
Data from the Fieldwork Experiment showed an increase in incident swash wave height during collision conditions when compared to swash conditions. Although the nearshore wave field showed similar conditions between the swash and collision conditions, the swash height showed an increase. Several parameters have been investigated to find the dependence of runup height to the offshore wave or beach characteristics. A best fit is found between runup and the beach slope (𝑅2 = 0.86), also indicating that swash motions are significant. Runup heights during these conditions are also compared to an empirical runup formula. This comparison indicated that runup elevations could still be predicted during conditions where runup reaches above the dune toe elevation.

Each year several fatalities due to rip currents are reported in the Netherlands. One of these accidents was during May 11th in 2020 near the Scheveningen harbour, were several surfers lost their lives. During the hydrodynamic and meteorological accident were quite severe, with a significant wave height of 3.0m and wind speeds of 11 m/s, which is likely to have resulted in a boundary rip at the northern mole. On top of that extraordinary amounts of sea foam amassed at the northern harbour mole, disorienting anyone who got trapped in the area. It is likely that due to the wind and wave conditions and the resulting rip current it was near impossible for the surfers to exit the vicinity of the mole, especially with thick layer of foam. The Scheveningen harbour area was modelled in Delft3D and subjected to tide wind and waves. Assessment of the influence of wind, waves and tide on rip current intensity was performed by comparing the model results from a combined run with the rip velocities obtained by superimposing rip velocities per individual forcing condition. Lastly, the model was run with conditions resembling the aforementioned accident as to check whether the model was able hindcast this situation. The main wind and wave directions found to lead to offshore directed rip transport come from 30-70 degrees relative to shore normal. Retentive rips also occur, mainly when the wind-wave forcing comes from -70 to 0 degrees relative to shore normal, although rips occur less frequently in this range.

An increasing amount of the worlds coastal regions suffer from structural erosion due to anthropogenic influences. A prime example for this phenomenon is the Oosterschelde, Netherlands, where the partial closure of an estuary lead to a decline of the local intertidal flats, threatening the valuable functions they provide as an ecosystem and for shore protection. A sustainable measure to slow down the decline, is the usage of wave disrupting oyster reefs, which reduce the wave energy on the flat, thus reducing erosion. As of now, the conditionality and extend of the wave attenuation effect of such reefs is still widely unknown. This is thwarting a more extensive usage of oyster reefs for erosion management, since the actual effects of such a reef are highly uncertain prior to construction. This thesis was designed to investigate and quantify wave attenuation capabilities of an artificial oyster reef in a macrotidal environment, improving the understanding of the impact of such structures on their physical environment. A wave analysis was carried out on a reef using field data, allowing for the quantification of attenuation, the identification of processes, a parameterization of the results and an insight on the predictability of attenuation. Extensive field data was acquired on seven locations on a transect over the investigated reef, including pressure and velocity data. The reef was built using shell filled wire gabions and is orientated perpendicular to the main wind direction. It has dimensions of approximately 120 m length, 8 m width and 0.8 m height. The reef crest is located 0.5 m above mean low water level and 2.3 m below mean high water level, being emerged approximately 30 % of the time. Incoming waves are characterized by maximum significant wave heights of 0.8 m and maximum mean periods of 3 seconds. Wave attenuation was quantified by comparing the energy flux in front of (seaward) and behind (shoreward) the reef under different hydrodynamic conditions. The conditionality of attenuation was found to be best represented by the relative submergence ds /Hm0,i , which represents the ratio of the freeboard on top of the reef and the incoming wave height. Attenuation was found until a ds /Hm0,i ratio of approximately 4. Compared to the bare tidal flat in front of the reef, the reef attenuated energy up to 40 times more effectively per distance. The spatial footprint of the reef was found to vary strongly with different ds /Hm0,i ratios, but being roughly of the same order of magnitude as the reef length. Reflection, overtopping, wave breaking and friction were identified as the main mechanisms behind energy attenuation. An attempt to parameterize these was done using dimensionless numbers, which can be used to predict attenuation by using incoming wave, and reef characteristics. The determined processes were best represented by the parameters ds /Hm0,i and L01 /ds. Three empirical relations describing wave attenuation over breakwaters were used to investigate whether existing formulas can predict wave attenuation over the observed oyster reef. The measured and calculated results show a strong correlation of R ~ 0.9 but need to be tuned to reach reasonable root mean square errors. This implies that such existing formulas can be adjusted to represent oyster reefs, making them a possible tool to predict oyster reef performance in the future. The findings show that the present oyster reef is effective in attenuating waves and contribute to the overall understanding of how oyster reefs can reduce erosion. A better insight on the quantity of attenuation and the spatial footprint of the reef's effects was achieved. The footprint of the reef's effect is hereby limited to an area close to it, meaning that the reefs impact on erosion should not be overestimated. Such a reef should thus rather be seen as a complementary measure against erosion, stabilising the local area. As a next step, setting up a numerical model is recommended, where the wave attenuation can be linked to actual erosion quantities. Quantitative research on additional oyster gabions can further be used to refine existing empirical relations, creating some predictability of transmission values over oyster reefs.

The Netherlands has 254 km of coastline that is covered with sandy dunes. As these dunes function as primary flood defences, intensive research has been conducted into dune erosion in recent years. This research mainly took place in laboratories, because all boundary conditions could be controlled and measurements could take place very accurately. However, it remains difficult to simulate a natural wave climate in a flume. In addition, there is also the risk of forgetting certain (as yet unknown) natural processes. Partly because of this, the demand for accurate field data and field measurement techniques has increased over the years. This thesis proposes a new experimental method to investigate dune erosion in the field, in which a 40 feet open-top shipping container is used as an in-field wave flume. Two containers are placed in between the low- and high waterline of the Sand Engine, near Kijkduin. The experimental method is validated by looking at the hydrodynamics within the container and by comparing dune erosion within this container with dune erosion from existing field and laboratory studies. This validation consists of four experiments in which several research hypotheses are tested to see if the wave propagation and erosion processes inside the containers work as expected. In general, it can be concluded that the experiments carried out inside the containers were in good agreement with the tested theories. These theories included that: the avalanching process is present, higher dunes have a bigger eroded volume, the dune toe follows the rising water level and the slope of the foredune remains similar during the erosion process. As this was the first time this method has been used, some recommendations are made. One of those findings is that the use of the measurement systems should be the same in both containers. A good reason to use this contained erosion method again is for the investigation of the grain size influence in erosion processes.

Worldwide, rubble mound breakwaters are designed and built to shelter and protect coastal areas from overtopping and flooding, especially harbours and shorelines. Rubble mound breakwaters are essential to preserve desired hydraulic conditions within the hinterland, avoiding damage to inhabited or industrial areas. This research focuses on the wave overtopping of rubble mound breakwaters as failure mechanism. To assess the wave overtopping engineers can adopt multiple methods. These methods can be ordered in increasing degree of accuracy and costs: empirical methods, neural networks, numerical models and physical laboratory experiments. The preliminary design phase is a highly iterative process. Using physical laboratory experiments within this phase is an expensive choice, therefore empirical methods are often preferred. Nevertheless, this research revealed that empirical methods, e.g. the so called EurOtop 2018, show significant shortcomings in assessing average overtopping quantities over rubble mound breakwaters, even more when the geometrical complexity of the structure increases (presence of protruding wave wall). This thesis re-calibrated the roughness coefficient proposed by the original EurOtop 2018 approach, referred to as the updated EurOtop 2018 method. Numerical models are proposed as a possible solution between empirical methods (which can be carried out quickly given their low complexity) and physical laboratory experiments (which need more time but are characterised by high accuracy). They have been increasingly used and accepted in the past decades. Following this trend, the Joint Industry Project (JIP) CoastalFOAM was launched with the objective to develop and validate a numerical model (OpenFOAM, waves2Foam, OceanWaves3D and JIP additions; referred to as CoastalFOAM) capable of accurately modelling the wave-structure interaction of rubble mound breakwaters. This research aims to calibrate and evaluate the CoastalFOAM model to assess small to large wave overtopping of rubble mound breakwaters with protruding or non-protruding wave wall, considering 500 waves. The evaluation of CoastalFOAM shows that this numerical model can be used, instead of empirical methods (e.g. updated EurOtop 2018), to assess the average overtopping discharges. CoastalFOAM showed excellent agreement with measurements for large and medium overtopping cases, while resulting less accurate for small overtopping cases. The analysis revealed, however, that the average overtopping discharge as a quantity is not capable of identifying the magnitude of large overtopping events (without modelling all 500 waves). This can be considered critical in determining whether the structure is safe enough in terms of Serviceability Limit State (SLS) or Ultimate Limit State (ULS). Consequently, this thesis proposes a new methodology to assess the maximum overtopping volume within a storm, applying the concept of wave focusing and using the NewWave theory. The input variables for the NewWave profile are extracted from the spectral properties at the toe of the breakwater. A first order wave maker is used to generate the NewWave profile, making the methodology sensitive for the degree of non-linearity of the considered wave conditions. This NewWave methodology offers improved accuracy to assess the maximum overtopping volumes when compared to the EurOtop 2018 approach. Furthermore, this research proposes to apply the inverse EurOtop 2018 technique to assess the average overtopping discharge using the NewWave maximum overtopping volume. However, the accuracy of this methodology is lower than that obtained with the updated EurOtop 2018 guidelines. This study shows that CoastalFOAM can be used, instead of the current empirical methods, to assess the average overtopping discharge within the design cycle of rubble mound breakwaters. However, according to what has emerged, CoastalFOAM shows to be less accurate in calculating small overtopping discharges as opposed to medium and large. On the other hand, when considering the maximum overtopping volume as design criterion the proposed NewWave methodology showed to be the most efficient and accurate.

This report presents the results of a research concerning long waves in the North Sea. ‘Long waves’ in this research is a collective name for various types of waves that are longer than the well-known sea-swell waves, here referred to as ‘short waves’. Wave types that are referred to as long waves are infragravity waves, meteo-tsunami’s, seiches and tides. Besides their extensive length, they distinguish themselves from regular sea-swell waves by the mechanisms responsible for their generation and the scale of their impact. Whereas short waves are generated by wind, long waves are generated by short wave-group forcing, the breaking of short waves nearshore and large atmospheric pressure variations. Long waves play a crucial role in the processes of coastline erosion and the breaching of flood defenses, as well as in the formation of rip currents and seiches. These events can lead to severe damage to hydraulic structures and possibly even casualties. A better understanding of them is thus vital for coastal safety.
Rijkswaterstaat has been collecting data regarding the occurrence of long waves on the North Sea for over a decade. Their measurement campaign started in a time when digital broadcasting signals and digital storage space were not well developed yet, making sending and storing the water surface elevation time-series not feasible. The data is therefore filtered by a FIR filter, reducing the size of the data, but also reducing the information that is contained in the data. All frequencies of approximately 0,0125 퐻푧 and higher are removed by the FIR filter. As a consequence a significant part of the infragravity signal is missing and it is not possible to perform a bispectral analysis (used to determine the bound long wave contribution). Nowadays, digital broadcasting signals and digital storage space have plenty of capacity to send and store the data, making the filter unnecessary.
Analysis of the spatial- and temporal distribution of long waves in the North Sea, shows that the majority of long waves have a yearly averaged significant wave height of 2,4 - 3,3 cm and a yearly averaged mean wave period of 125 to 140 cm. Based on data from the full decade, a clear trend in the spatial distribution of long waves cannot be distinguished, offshore and nearshore locations show approximately the same long wave properties. The seasonal analysis shows that the mean significant long wave height averaged out over all locations per season varies from 2,8 cm during summer, 2,9 cm during spring, 3,3 cm during winter to 3,5 cm during fall. For the mean wave period, the mean value averaged out over all locations per season varies from 127,5 s during summer, 128,1 s during spring, 135,6 s during winter to 136,3 s during fall. Predictions of the extreme wave conditions show that the significant wave height with a return period of 10.000 years (Dutch design condition for coastal flood defenses) are around 40-60 cm, although there is quite some uncertainty in these predictions. This uncertainty is due to a relatively short measurement period, usually 30 years of data is used for extreme wave predictions, and remarkable observations in the data.
Time-series analysis of the hourly significant wave height of long waves and predicted bound long waves (Hasselmann, 1962) (Hasselmann, 1963) shows that during mild weather conditions significant wave heights of long waves are low, with minimal contributions by the bound long waves. At the onset of a storm, the significant wave height of both long waves and bound long waves rapidly increase. At the end of the storm, they decrease to the pre-storm values. This shows that the generation of long waves occurs mainly during heavy weather conditions. Hourly significant wave heights of long waves, and especially bound long waves, show a strong correlation with significant short wave heights. The relative contribution of bound long waves (퐸푏푙푤/퐸푙푤) during a storm increases up to values of 60%, which is considerably higher than contributions of bound long waves found in other studies (Herbers et al., 1994). The high ratio of the relative bound long wave contribution, in combination with very low bound long wave activity during mild conditions, shows that the generation mechanism involving wave-group forcing is a major generation mechanism of long waves in the North Sea.

This report presents the methodology and findings of a study on the impacts of free long waves on dune erosion predictions in XBeach. In this paper a distinction is made between two types of infragravity waves: bound infragravity waves and free infragravity waves. Bound infragravity waves are generated by short-wave group forcings and are bound to the propagation of the wave groups. Conversely, free waves are infragravity waves that are released from the wave-group in the surfzone, reflect at the shoreline, and propagate back out of the surf zone. In a semi-enclosed basin with mild bathymetric changes such as the North Sea, free long waves that reflect off one coast can propagate large distances and impact other coastlines.

Modelling the response of a coastline during a storm event requires an offshore boundary condition that represents the incident short and long waves. The boundary condition for the incident infragravity waves is often derived by assuming a local equilibrium between the directionally spread sea-swell waves and bound infragravity waves following the K. Hasselmann (1962) method. This approach has proven to be problematic for two main reasons. First, the method tends to overestimate the incident infragravity wave height by assuming a local equilibrium is achieved at the model boundary; however, observations have shown that the transfer of energy from the short-wave groups to the underlying bound wave is gradual on a sloping bed. Second, by applying the equilibrium K. Hasselmann (1962) method it is implied that only bound infragravity waves are present at the boundary underestimating the total incident infragravity wave energy.

An analysis of XBeach in surfbeat mode revealed that the model can confidently predict infragravity wave behaviour on a natural beach slope with two well-developed bars during a storm event by calibrating the roller breaker slope coefficient and wave breaker coefficient. Infragravity wave heights were reasonably accurately predicted in the surfzone, but the model generally overpredicted the most energetic infragravity waves.

Further, the study investigated the impact of free infragravity waves on dune erosion predictions by simulating the behaviour of two 1D planar beach slopes to the inclusion of free long waves at the boundary. The study revealed that on a beach with a 1:35 slope the relative increase in dune erosion volume was 44.1% when the bound infragravity energy is equal to the free infragravity energy relative to the case excluding free waves. For the same conditions, a 20.6% increase was observed in the maximum runup. For a milder 1:70 slope a 36.4% and 6.1% increase in dune erosion and maximum runup was observed.

The results demonstrate that the dune response of a coastline is sensitive to the inclusion of incident free infragravity waves at the boundary. Moreover, neglecting the presence of free infragravity waves at the boundary may underpredict the dune response during a storm event.

Lighthouses are structures subjected to extreme weather conditions which have to resist strong surge. Therefore to ensure their survival over time, it is necessary to determine the loads that these structures will face during its lifetime. However, load prediction on lighthouses is troublesome due to the multiple factors that affect the loading. First, the geometries of lighthouses vary considerably from one to another, not usually representing the classic cylindrical shape by which they are modelled in laboratories. Second, climate change transforms the statistical properties of the sea states and will bring more extreme events in the near future. As so as sea level rise. Lastly, the bathymetry around a lighthouse is considerably different to the ones used to model lighthouses in labs, which can affect wave propagation and breaking. Therefore loading is affected. As the traditional physical methods are both expensive and time consuming, researchers have developed computational tools to accurately represent reality. In this report, waves2Foam (a toolbox within OpenFOAM) will be used to reproduce some tests performed on a physical wave flume. The goal is to represent the conditions of the lab in the computational model so that this model can then be used to represent different geometries and wave loads over lighthouses in an easy and cheap way. Additionally, the most of the current tests were performed with waves directly hitting the lighthouse structure. However in this study, the waves break over a shoal and then the mass of water reaches the structure. This causes a smaller maximum load but the load is applied during a longer time that when a wave breaks directly on the lighthouse. In this study the physical model performed by Piermodesto Caputo (year 2017/2018) and supervised by Professor Renata Archetti and Dr. Alessandro Antonini (co-supervisor) will be modeled using waves2Foam (OpenFOAM). The physical model was performed in Plymouth University laboratory "COAST". The goal of the model was to determine the loads that the Dubh Artach lighthouse will face in the future taking into account sea level rise and extreme weather conditions due to climate change.

Velocities derived from X-band radar were compared to depth averaged ADCP measurements in a complex tidal inlet system at Ameland, the Netherlands. Inclusion of depth assimilation and ensemble averaging in radar calculations led to smaller differences between ADCP and radar. The observed differences were clustered and related to water level elevations, wind velocities, wave periods, wave heights, spatial coherence in radar output and error metrics of the radar fitting procedure. Larger waves and higher wind velocities were observed to benefit radar agreement with ADCP results. Rising water levels benefitted agreement in east west direction. Falling water levels benefitted agreement in north south direction. Confidence intervals of the fitting procedure were observed to coincide with differences between ADCP and radar and potential for filtering based on them was shown. Nevertheless, an unclarified tendency towards northwestern bias, which may be specific to the comparison locations, remains. The radar at Ameland monitors the whole inlet system and provides current velocities everywhere in its range. This study shows that its currents are in good agreement with ADCP depth averaged currents throughout most of the tidal cycle. Furthermore, it stresses radar’s potential for better monitoring of the coast and for cost effective coastal field measurements to obtain large datasets, even in hydrodynamically very complex regions.

Many coastlines fronted by a coral reef are threatened by wave driven flooding. For this reason, waves in coral reef environments has been topic of many studies. However, most are focused on a 1D simplification of the coral reef environment. These studies neglect the impact of 2D processes, for example wave directional spread. In this study an attempt is made to understand and investigate the influence of wave directional spread on wave runup. XBeach Non-Hydrostatic with an additional vertical layer is used to perform a great amount of digital experiments. A synthetic database is created consisting of 540 different model runs. The analysis showed that wave directional spread influences the wave runup by affecting the wave setup and low frequency runup. So, for the most accurate runup predictions it is advised to include directional spreading in the analysis. Furthermore, the complete exclusion of directional spreading in the runup analysis will lead to an over prediction of the runup, so current 1D models give conservative estimations of the runup. Lastly, it was found that the exact wave directional spread is not that important but it is advised to obtain information about the wave field that is dealt with and take that into account in the runup analysis.

Wave impact on vertical structures with overhangs is studied. 3 research questions are addressed: 1. How does the entrapped air change during the process of wave impacts ? 2. How is the wave loading influenced by the presence of horizontal overhang ? 3. How is the wave loading influenced by the presence of air pocket?

The Sand Engine is a new innovation in coastal protection, a mega feeder nourishment. This pilot project was constructed in 2011 along the Delfland coast, which is historically prone to erosion. Since its construction, the Sand Engine is intensively being monitored to track the morphological development.

The objective of this thesis is therefore to assess how the morphology of the Sand Engine is evolving over time and how this evolution contributes to the sediment budgets of the Delfland coast. In this thesis the first five years of morphological evolution of Sand Engine and its surroundings are investigated.

A data analysis is performed on bi-monthly bathymetry measurements, covering the first five years of morphodynamic evolution of the Sand Engine and the surrounding coastal cell. The high temporal and spatial resolution in both alongshore and cross-shore direction, provided the opportunity to perform research in great detail and precision.

Volume changes and sediment transports are successfully derived. Sediments are redistributed in both alongshore and cross-shore direction over the coastal cell. The accretive areas slowly extend alongshore while the erosive part of the Sand Engine remains in the same position. The Sand Engine has a positive net contribution to the sediment budget of 5.8km of coast after 5 years, which is an extension of 3.2km. 4.2Mm^3 of the initially nourished volume of 17.5Mm^3, has been redistributed after 5 years.

A Gaussian curve is fitted to several iso-lines of constant altitude to describe the plan-form adjustment of the Sand Engine at different altitudes. The Gaussian parameters are useful in accurately describing the spatio-temporal behaviour. Yet large differences in adjustment rates are found over the altitude. The decrease in cross-shore extent is fastest around 0m+NAP, and decreases to near zero at -8m+NAP.

Cross-shore profile behaviour is investigated by inspection of the temporal evolution of characteristic cross-sections in accretive and erosive locations along the Sand Engine. Erosive and accretive profiles behave considerably different. The profiles are adjusting rapidly and bed-level activity varies considerably over the altitude, with a maximum morphological activity in the intertidal zone, rapidly decreasing with increasing depth.

The results show that the Sand Engine spreads alongshore and feeds the adjacent coastal sections in the five years after construction. A mega feeder nourishment is therefore capable of supplying sediments to the adjacent coast.
The first year morphodynamic response was much stronger than in any subsequent year. The current results therefore give an improved view on the long-term development of the Sand Engine.