Climate change is accelerating coastal erosion, posing growing risks to infrastructure, ecosystems, and livelihoods across the world’s coastal zones. While global and regional studies increasingly provide projections of physical shoreline change, they often fail to integrate these with the spatial distribution of exposed assets or the potential for adaptation measures. As a result, their utility for informing where and how interventions, such as beach nourishment, could serve as viable adaptation options remains limited. Current research typically treats coastal erosion, exposure, and adaptation in isolation, leaving a critical gap in support for strategic coastal planning, where cross-country and cross-boundary insights could enable more coordinated and effective responses.

Here we present a continental-scale framework that integrates shoreline-change modelling, infrastructure exposure, and nourishment suitability to inform coastal adaptation across Europe. We first compile a novel database of 1060 historical nourishment interventions spanning 70 years, capturing the spatial distribution, implementation motives, and coastal contexts, including borrow area, nourishment type, and coastal type, under which nourishments have been applied. These insights inform a Nourishment Suitability Assessment Model (NSAM), which evaluates future nourishment potential based on four key factors: coastal type, policy context, sediment availability, and prior experience. We then project shoreline retreat by combining ensemble-based sea-level rise scenarios with ambient erosion trends derived from multi-decadal satellite imagery. The analysis applies the Bruun Rule with spatially variable coastal slopes and probabilistically combines sea-level rise-induced retreat with ambient trends. Applied to over 27000km of sandy coastline, this produces spatially explicit retreat projections that are intersected with exposure data to identify high-risk beaches.

Our results show that over 98% of transects are projected to erode under sea-level rise by 2100, with median shoreline retreat reaching -80m under SSP5-8.5. In contrast, ambient change alone shows strong regional variation, with 41% of historically unstable coasts eroding but a median accretion of 17m at the European scale. Combined projections indicate that while 85% of European sandy coasts are projected to erode by mid-century under SSP2-4.5, only 11.6% of these coincide with direct infrastructure exposure. By focusing on high-risk beaches, defined as stretches with over 1km of exposed assets, this research identifies up to 841 exposure hotspots (covering 1497km) under SSP5-8.5 in 2100, helping to prioritise sites in need of adaptation. Among these exposed sites, nourishment suitability varies considerably, shaped by both physical conditions and institutional capacity. The NSAM’s traffic light classification offers a first-order indication of where nourishment could form part of context-specific adaptation strategies, with suitability highest where conditions align with historical precedent and institutional barriers are low.

Key uncertainties are identified, particularly in slope representation and the assumed independence between ambient and sea-level rise-induced retreat. These highlight priorities for future model refinement. The framework demonstrates how integrated hazard, exposure, and adaptation assessment can enable more targeted and context-specific coastal management. By synthesizing forward-looking erosion risk with backward-looking adaptation practice, this work helps bridge the gap between impact assessment and action, showing that while many coasts are at risk, not all are equally suitable for nourishment, and a context-specific approach is necessary.

Coastal systems provide numerous services to communities worldwide, including flood protection, recreational spaces, and biodiversity support. However, these valuable environments face increasing pressure from climate change, sea-level rise, and growing coastal populations. Coastal management aiming to preserve these services is evolving from traditional "grey" towards sandy Nature-based Solutions (NBS). The success of NBS inherently relies on natural processes driving their evolution. NBS typically aim to fulfil multiple objectives, while having lasting impact crossing multiple domains (such as the nearshore, beach and dunes). This increases the need quantitative tools that describe sediment transport and morphodynamic development across the nearshore-dune system.

This dissertation addresses the challenge of modelling coastal evolution across multiple domains through four complementary aims: (A) enabling process-based description of dune development, (B) demonstrating the technical feasibility of coupling numerical models, (C) mapping sediment pathways across the nearshore-dune system, and (D) demonstrating the practical utilisation of these tools for informing NBS design.

To enable the process-based description of coastal dune development in engineering contexts, the existing aeolian transport model AeoLiS is enhanced with landform-shaping processes, including vegetation growth, topographic steering of wind flow, and avalanching. The goal was enabling the simulation of realistic coastal dune evolution at scales relevant to engineering applications. The enhanced model successfully reproduces four distinct dune landforms under real-world conditions. The simulation of barchan dunes closely matched dimensions and migration rates observed in Morocco. Simulated parabolic dunes reproduced migration rates and seasonal dynamics observed in Brazil. The simulation of an embryo dune field captured the seasonality and spatial sheltering effect as observed at De Hors, Netherlands. Finally, the spatial patterns and volumetric changes of excavated foredune notches along the Dutch coast were successfully replicated. Simulating this blowout development on an engineering scale demonstrated the model's practical applicability.

The nearshore and dune domains function as an integrated system, yet they are typically studied separately using different models. To address this limitation, a coupling framework was developed that enables continuous exchange of bed levels, wave heights, and water levels between three process-based models: Delft3D Flexible Mesh, SWAN, and the enhanced AeoLiS model. The framework was applied to the Sand Engine mega-nourishment in the Netherlands, accurately reproducing observed marine-driven longshore erosion (4.1 Mm3) and aeolian deposition patterns. The simulation results enable the quantification of interactions between domains. Aeolian extraction of sediment from the shared sediment budget reduces marine-driven longshore dispersion by 2%. Meanwhile, marine-driven morphodynamics cause variations in foredune growth; up to 24% lower dune growth in sheltered areas and 7% higher growth along accretive beaches in the fifth year. As these morphological deviations accumulate over time, the effect of these cross-domain interactions intensifies over time. This study demonstrates the feasibility of coupling models at scales relevant to coastal management—spanning multiple kilometres over five years.

Understanding coastal evolution requires more than quantifying morphological changes; it demands insight into how sediment moves through the system. To address this need, a Lagrangian particle-tracking approach was developed that accounts for morphodynamic-driven burial. This extends Lagrangian analysis beyond traditional timescales to multi-year periods. Applied to the Sand Engine simulation, the simulated pathways reveal movement patterns impossible to detect through conventional Eulerian approaches. We analysed both the dispersion of nourished sediment and the origin of accumulated sediment at the Sand Engine's flanks. During initial stages, nourished sediment dispersal is severely restricted by rapid burial. The displacement of particles is reduced by an order of magnitude compared to displacement expected along undisturbed coastlines. Sediment accumulation south of the perturbation includes both direct supply from the nourishment (41%) and indirect contributions from updrift sources (59%). As the coastline perturbation diffuses over time, transport patterns gradually transition toward those of an undisturbed coast.
These result enable improved understanding of sediment movement between sources to sinks and thereby the causal relationships that drive coastal change. Rather than indirectly inferring connectivity from Eulerian morphological comparisons, sediment pathways explicitly visualize sediment movement between domains.

The practical utility of the developed tools was demonstrated by examining how specific design choices might influence dune evolution in both large-scale mega-nourishments and smaller beach nourishments. For the Sand Engine, the impact of multiple alternative design considerations on dune growth was explored. The simulations indicated that removing the artificial dune lake would increase dune growth by 42,000 m3 over 10 years, with effects extending 1,200 m alongshore. Reducing coarse sediment content enhanced dune growth by 65%, and lowering the crest elevation increased sediment availability through more frequent mixing. In a smaller-scale application, we explored using certain decisions in nourishment design to steer sediment supply as an abiotic condition for vegetation growth. Incorporating a lagoon in beach nourishment design restricted sediment supply toward the foredune, reducing total dune growth by 62%. Under the assumption that vegetation depends on sediment burial, vegetation growth was limited, maintaining wider blowout entrances. As a result, the simulated amount of backdune deposition was 23% higher for the scenario with lagoon. Particle tracking analysis revealed that the lagoon design more than doubled the contribution from nearshore sediment sources to backdune accumulation. These applications demonstrate how the developed tools can support the design of NBS aiming to achieve objectives across multiple domains.

This dissertation advances our ability to model coastal processes across domain boundaries, connecting the marine and aeolian environments that have traditionally been studied separately. The development from enhanced dune modelling to coupled nearshore-dune frameworks, advanced sediment pathway analysis, and practical design applications creates new possibilities for coastal management. While challenges remain in computational efficiency and process representation, these tools enable coastal engineers and managers to examine how interventions impact evolution across the nearshore-dune system. As coastal regions continue to face increasing pressures, the ability to connect domains in morphodynamic modelling provides a valuable tool for integrated coastal management.

Addressing the impact of sea-level rise in coastal areas requires sustainable management strategies based on informed decision-making, and thus a shoreline projection is important. This research explores methods to improve shoreline change projections by obtaining a long-term shoreline change trend, empirically from \acrfull{sds} observations, and implementing a coastal impact model to project the future shoreline. We defined a heuristic approach to obtain a clean-\enquote{primary}- shoreline-change signal from all raw \acrshort{sds} observations that are available at transect. Long-term trend analysis from the primary shoreline-change signals results in a more accurate long-term shoreline-change trend. The improvement is also observed in Truc Vert, a meso-tidal case previously well known for low \acrshort{sds} accuracy, where the primary series results in accurate long-term shoreline trend regimes. Our projections in open sandy transects show that long-term trends predominantly influence shoreline change patterns and regimes compared to the retreat due to sea level rise. Using input comparable to the input of an earlier study, projections in seven out of eight regions indicate less retreat, except in North Carolina where the regional sea-level rise projections are relatively high. Compared to a hybrid, physics- and process-based, model, our projections showed higher progradation and erosion in \enquote{hot spots}, revealing limitations in linear trend extrapolation for capturing non-linear shoreline behaviour. We can conclude that future shoreline-change projections can be improved by using more accurate long-term shoreline-change trends. The framework's applicability is best suited to open sandy beaches with minimal human intervention. This highlights the need for better coastal typology classification that formalizes the concept of coastal type and coastal squeeze in the backshore of the beach. The outlook for the development of the shoreline projection framework is to explore the non-linear behaviour of shoreline change. This can be captured by decomposing the primary shoreline-change series into different signals that better represent the shoreline-change behaviour.

Sand waves are rhythmic seabed patterns which develop as a result of the interaction between tidal flow and perturbations in the seabed. Due to their size and dynamic behaviour sand waves can pose a threat to offshore infrastructure such as cable trenching. To ensure the safety of these cables, it is crucial to understand the interactions between sand waves and trenches. Residual currents, currents that are not induced by the tide, can exert a significant influence on sand wave dynamics due to the non-linear relationship between flow velocity and sediment transport. Hence, this thesis aims to investigate the effects of residual currents on the morphological interactions between sand waves and trenches. A Delft3D FM model is used to simulate a case study of a trench within a group of sand waves in the Belgium Continental Shelf. The influence of residual currents, mostly induced by wind and atmospheric pressure, is studied during a period of a storm and a period with representative residual currents. It is found that during the tidal cycle at the peak of the storm, the residual currents cause an increase of 66% in sediment infilling into the trench compared to tide-only conditions. Since this is in the same order of magnitude as the background infilling due to tidal dynamics, this is less than expected. The residual currents are however strong enough to reverse the sediment transport patterns of the sand wave and trench system. During the representative residual currents, the contribution of residual currents is found to be insignificant, especially compared to the observed influence of the spring-neap tide. This study concludes that the overall effects of the residual currents, induced by wind and atmospheric pressure, can alter sediment transport patterns but the magnitudes are insignificant in the case study under investigation. It is recommended, for further research and for effective design and maintenance of offshore infrastructure procedures, to keep in mind that morphodynamics of sand wave and trench systems are highly subject to the variations in tidal dynamics induced by the spring-neap tide.

Sand waves can have a significant impact on offshore activities, and with the global push for sustainability and more renewable energy sources, these negative impacts will become more prevalent. Sand waves in the North Sea have a height up to a couple of meters and a wavelength between 100 and 1000 meters. The migration or change in asymmetry of sand waves can cause free spans in export cables leading to buckling or vibration causing failure. The migration can also cause cables to be exposed on the sea bed and in danger from offshore activities like anchors and fishing boats.

This research focuses on utilizing satellite data to determine sand wave characteristics. Satellites like Sentinel-1 and Sentinel-2 which are used in this research have global availability of data over multiple years at a 10 meter resolution. Sentinel-1 has a Synthetic Aperture Radar (SAR) which creates images by sending out microwave signals and recording the strength and the time delay of the returning signals. Sentinel-2 has an optical instrument which creates images by recording the reflected light from the sun on the earth's surface. Sand waves were detected through the change in Sea Surface Roughness due to the current interaction with the sea bed as described in the Alpers-Hennings Model. The difference in resulting sand wave characteristics between SAR and optical images were determined.

Three initial areas of interest off the coast of the Netherlands were chosen based on their different characteristics. Hoek van Holland contains short irregular sand waves. Holland Kust Zuid contains long regular sand waves. The third location is Alkmaar, which contains no sand waves and was used to determine the results of the methodology when a location has no sand waves present. The first step was to determine the environmental conditions necessary for sand waves to be visible in satellite images. Image collections for an area of interest over a set period of time were filtered based on the environmental parameters. These parameters include mean glint angle and cloud cover for optical images, and wind and current speed for both optical and SAR images. Threshold values were determined through literature. The wind speed should be between 3 and 12 m/s, the current speed should be greater than 0.4 m/s, and the mean glint angle for the latitude of the North Sea is less than 56 degrees. The image collection for the year 2021 for the three different areas of interest was filtered based on these parameters. The final collection after filtering was then manually checked for the visibility of sand waves. This number was lower than the initial filtered collection which means that there are additional factors that need to be accounted for.

A methodology was created to determine the sand wave characteristics over an area by applying a Fast Fourier Transformation (FFT) and calculating the sand wavelength and wave angle from the resulting signal. This was first applied to the three different areas of interest within the North Sea, Hoek van Holland, Holland Kust Zuid, and Alkmaar. Using the resulting image collection at these locations the sand wave characteristics were calculated. Although there are no sand waves at Alkmaar, due to the methodology, sand wave characteristics are still calculated. Also, the methodology works better at Holland Kust Zuid with long regular sand waves compared to at Hoek van Holland where sand waves are shorter and irregular.

To determine at which scale the satellite images should be viewed at to obtain information on the sand wave characteristics, the data was split into different area sizes and a FFT performed. The resulting average values and spread of the sand wavelength and wave angle were compared at different area sizes for optical and SAR images, as well as Multi-Beam Sonar (MBES). This resulted in an area of 5 by 5 kilometers which allows for the correct signals to be read across different possible sand wavelengths. Also, the effect of different sources of noise on the FFT and calculation of the sand wave characteristics for optical and SAR images was determined. From this it was seen that ships within a satellite image containing sand waves introduce a very strong signal within the FFT. This prevents the sand wave characteristics from being calculated correctly. Wind farms within a satellite image affect SAR images, and not optical images. Suspended sediment transport and clouds, which only occur within optical images, have different effects. Optical images are not affected by sediment transport because the current direction will always be perpendicular to sand wave crests and therefore does not influence the calculation of sand wave characteristics. Clouds, similarly to ships, introduce a strong signal within the FFT which blocks the signal of the sand waves. Dark patches (like rain cells), which effect SAR images, block the visibility of sand waves within satellite images and therefore there no signal corresponding to sand waves within the FFT can be found.

Then, the area of interest was increased to the entire southern North Sea. Satellite images over the North Sea were downloaded in 5 by 5 km tiles for a single satellite pass over per year where the environmental parameters met the conditions necessary to view sand waves. As for the local scale cases, a FFT was applied to each tile and the sand wave characteristics were calculated including sand wavelength, wave angle, density, and spatial frequency. The differences in characteristics were compared from year to year. Typically, the difference is less than 200 meters in sand wavelength, however there are points where the difference is much larger. This is due to noise affecting the calculated value. Additionally, the data for wind speed and current speed for the date and time of the satellite image was determined. From this it is seen that optical images can contain sand waves at lower wind and current speeds, at the lower end of the boundary conditions. Sand waves are only visible in SAR images when the current speed is very high. Also, the results were compared to the water depth of the North Sea. At depths less than 40 meters which occurs on the Dutch Continental Shelf sand waves are almost always visible with the correct environmental parameters. At depths greater than 40 meters, which occurs on the West side of the sand wave field, sand waves are not visible consistently. The calculated sand wavelength over the North Sea was also compared to wavelength obtained from MBES. This resulted in low correlation values for both optical and SAR. As a result, noise present in satellite images should be taken into account.

By both applying the methodology to smaller scale cases and to the entire North Sea, it is possible to detect sand waves in different environments. Using the change in sea surface roughness due to sand waves it is possible to see sand waves in both optical and SAR satellite images. By applying the Fourier transformation the average sand wave characteristics over an area can be calculated. The methodology is limited by environmental parameters and noise that can be present in the satellite images. SAR and optical images require specific wind and current conditions, while optical images also requires specific mean glint angles and a low cloud cover. Different sources of noise also have a negative impact, adding signals to the FFT that do not correspond to the sand waves. Although the methodology proposed in this thesis is successful in determining the average sand wave characteristics over an area, more research for this topic would increase the possibilities. This includes using additional conditions for filtering image collections for sand waves, removing sources of noise, utilizing higher resolution satellite data, and testing the methodology on different sand wave fields that have different characteristics, such as symmetry. There are still many possibilities that can be explored in using satellite images to determine sand wave characteristics.

Sand wave is one of the large-scale seabed morphological features with a typical wave height of metres and a wavelength of hundreds of metres. Its mobile nature causes a nuisance to offshore infrastructure and marine transport. Pre-sweeping is a construction procedure commonly involved in submarine utilities installation to dredge existing sand waves but recovery of sand waves is usually observed after dredging. The underlying processes of sand wave recovery remain relatively underexplored. Given the potential implications for improved dredging planning, this thesis aims to investigate the drivers of sand wave recovery and examine the modelling performance of Delft3D FM on the recovery process through a case study. The numerical model is the first 3D morphological model specifically for simulating sand wave recovery after pre-sweeping. With the incorporation of DelftBlue and MORFAC, computational efficiency has been significantly enhanced. The results show that the transverse slope effect and the oblique current flow supply sediment into the dredged area causing sand wave recovery. This study also elaborates on the two stages in the recovery: an adaption period referring to the morphological adaption of the dredged profile in response to the forcing and a subsequent regeneration period referring to the growth of a feature developed during the adaption period. Although the precise estimation of the recovery duration by Delft3D FM remains challenging, this pioneering research provides an explanation on how tidal flow drives the recovery and reveals the accomplishment of Delft3D FM in modelling the phenomenon. Further exploration of this subject, such as the impact due to storm and the influence of trench configuration, is suggested to further enrich our understanding of sand wave recovery.

The Wadden Sea serves multiple roles: it acts as a protective barrier against severe wave conditions, is a natural habitat for diverse flora and fauna and has recreational purposes. This research provides a comprehensive analysis of the intricate morphological dynamics of the Wadden Sea. By implementing various model schematizations, the study aims to enhance understanding of complex interactions between tidal inlets and basins, sediment transport dynamics, the influence of forcing actors and the importance of grid resolution. The ultimate goal is to improve the process-based models that are used to perform morphological forecasts, serving as a valuable tool for efficient coastal management and ensuring the preservation of this unique tidal ecosystem.

The study incorporates several schematizations, including the spatial extent of the modelling grid (domain schematization), grid resolution and various forcing components applied to the model boundaries and grid. Analysing water and sediment exchange between basins and through the Ameland tidal inlet has revealed distinct differences in hydrodynamics and morphodynamics across the model schematizations.

The single-inlet domain schematization encompassed only the Ameland tidal inlet region, while the more expansive multi-inlet schematization captured the entire Wadden Sea area. The single-inlet schematization does not account for hydrodynamic interactions between the Ameland basin and neighboring basins. This impacts sediment transport and morphological development in the Ameland inlet. A multi-inlet approach, encompassing the entire Wadden Sea, grants a more comprehensive model. The model with Wadden Sea domain schematization is essential for obtaining accurate results, as it showed significant differences in sediment transport compared to the single-inlet model. Modeling practices that exclude transport dynamics between tidal basins risk misrepresenting the natural processes in play.

Around the Ameland inlet, ebb dominance is observed in most regions, but there are exceptions in the marginal flood channel and at the outer side of the ebb-tidal delta. The Ameland model shows more ebb-dominance than the Wadden Sea model. Sediment transport during ebb periods exceed than during flood periods and differences in sediment transport between model schematizations are more pronounced during ebb than during flood.

Regarding the drivers of change, the spring-neap tidal cycle is particularly indicative of periods with meaningful residual sediment transport. Wind, especially from southwesterly directions, has a significant impact on water transport dynamics and a limited impact on sediment transport through the inlet.

Two grid schematizations are used, one with a base resolution of 60 by 60 meter and another with a high-resolution grid of 30 by 30 meter near the inlet. An enhanced grid resolution results in marginal changes in simulation outcomes, offering limited advantages for the conducted short-term morphological forecasts.

Overall, enhanced model schematizations can significantly improve short-term morphological forecasts of the Ameland tidal inlet, providing a more accurate and comprehensive representation of the complex hydrodynamic and sediment transport processes. Leveraging high-performance computing (HPC) for executing Delft3D-FM models has been central in this research, offering possibilities to enhance model resolution and reduce computational time.

The benefits of HPC come with the added challenge of mastering new software, handling larger models and interpreting more extensive data sets. A deep understanding of the physical processes and careful cost-benefit consideration is crucial, as increased computational power does not automatically resolve all modeling limitations. As such, the effective use of HPC in coastal modeling requires a balance between its sophisticated capabilities and the complexity it introduces.

Since early history, humans have been attracted to coastal areas. This can be related to the economic benefits of these areas due to access of ocean navigation, coastal fisheries, tourism and recreation (Seas and Plans, 2011). Around 40% of the world’s population lives within 100 km of the coast (Seas and Plans, 2011). People are drawn to sandy beaches in particular because of its aesthetics and value for specific economic amenities (Luijendijk et al., 2018). Nevertheless, as these beaches are dynamic both in time and space, proper coastal management is required to prevent loss of land and secure future coastal life.

Up till now, studies into coastal erosion have been conducted locally, resulting in site specific observations. However, the promising results of using satellite imagery in the field of coastal engineering allowed studies to be performed at larger spatial scales. This can lead to the identification of areas with similar characteristics, resulting in methodological standardization of approaching a specific problem. A first step toward this new approach of studying shoreline evolution was taken by Luijendijk et al. (2018) who presented a global dataset of annual shoreline positions for sandy beaches over the period 1984-2016 using satellite derived shorelines
(SDS). However, the drivers (causes) of shoreline evolution on a global scale were still unknown, making it only suitable for identifying areas of structural shoreline change, but less suitable for deriving coastal management solutions. Therefore, the research objective in this study is to identify and characterize drivers of shoreline evolution on a global scale using SDS.

This study focused on dynamic sandy beaches, or hotspots, extracted using a method developed by Kras (2019). In this method, using a 2.5-kilometer moving window, transects showing structural shoreline changes and similar characteristics both in space and time were grouped. The small size of the moving window led to locally created hotspots, 95% of which had a spatial extent of less than 10 kilometers, allowing to study shoreline drivers with small to moderate spatial scales (∼10 kilometer). Therefore, the main focus in this study lies on seasonality as a natural driver of shoreline evolution and three anthropogenic drivers: reclamations, nourishments and littoral drift barriers. As seasonality shows inter-annual variability, the temporal resolution of the SDS is increased from annual to monthly.
Using time series decomposition methods, different parameters are extracted that can be used to link the drivers to the SDS. Besides temporal parameters, also parameters related to spatial characteristics are considered. These parameters can be split into identification parameters, used for identifying a driver, and informative parameters, providing knowledge on the behavior of the driver. These parameters were developed and tested using local case studies. Results from these local case studies showed that the identification parameters showed similar behavior along the case studies. This implies that the identification parameters correctly reflect a driver’s behavior. Next, identification of the drivers was verified on a larger scale, all transects within hotspots on West-European coastlines. Verification was done on hundreds of samples using literature or manual inspection of satellite images. Using precision scores, the fraction of true positives to the total identified cases, optimal settings were derived for identification of the drivers. These settings resulted in a pattern of driver identification and characterization along the West-European coastline that is supported by literature.

With the optimal settings for identification determined, the methods were deployed on a global scale. The global dataset consisted out of 3033 prograding and 2121 retreating hotspots containing over 58 thousand transects in total. For these hotspots, SDS were generated over the period 1984-2021 with a monthly temporal resolution. This resulted in a global dataset of more than 26 million monthly shoreline positions. Two other processes, in addition to a seasonal change in wave height, were found to be able to generate seasonal variations in coastline positions from this global dataset. At the Red Sea, even though the wave climate is low in energy (Langodan et al., 2017), the coast is characterized by seasonal behavior. However, in this basin, seasonal variations in water levels rather than the wave climate best described this pattern. In addition to varying wave height and water level, seasonal beach morphology can also be caused by a shift in wave direction. This was observed in southern and western parts of Australia. Furthermore, non-seasonal beaches were primarily seen in low-energetic wave regions where neither of these other two processes occurred, as is the case in the Mediterranean. In regions where seasonal shoreline fluctuations are caused by differences in wave energy, minimum shoreline positions were found at the start of the summer. However, the period in which minimal shoreline positions are observed may be observed at a different time of the year in regions where seasonal shoreline behavior is driven by water level variations or a shift in wave direction. The identification of reclamations pointed out that this driver was especially linked to shoreline evolution in the Middle East and East-Asia. Furthermore, while the amount of constructions of reclamations remained
constant on a global level over the period 1987-2017, in these two areas an increase was observed. Opposite behavior was found by the identification of nourishments, as this driver was identified more often in the period 2007-2017 compared to the two decades before that. Moreover, nourishments were mostly observed in Western countries, for example the USA and the Netherlands. Nevertheless, also in Non-Western countries, an increment over time in the amount of nourishments could be detected. This indicates that throughout the entire world the use of nourishments as a measure to prevent coastal erosion is increasing. Shoreline evolution linked to littoral drift barriers was mostly observed in North-America, Europe and Africa. Downdrift (erosive) hotspots were mostly observed in Africa while in North-America and Europe mostly updrift (accreting) hotspots were linked to littoral drift barriers. On a global level, a combination of an updrift and downdrift hotspots (a pair) was observed in only 2% of all hotspots.

The outcomes above can support local-scale studies by identifying the drivers of shoreline evolution, describing their characteristics and even create standardization by analyzing areas with similar behavior. Hence, it can be concluded that spatiotemporal parameters describing the behavior of a driver can be used to identify and characterize drivers on a global scale using SDS. Nevertheless, not all drivers of shoreline evolution were included in this study. Therefore, to include drivers with larger spatial scales, hotspots should be extracted by using a larger moving spatial window. Furthermore, by increasing the spatiotemporal resolution on which this extraction is based, accuracy of the spatial extent of the hotspots is expected to increase. The small proportion of pairs identified for littoral drift barriers can be partly explained by the erroneous spatial extent of some hotspots. Finally, drivers are identified independently from each other neglecting their interactions. Even though interactions might be complex, drivers should not be identified independently as this will rather
require local studies than support them. Still, even though refinement and further development of the methods is required, this research has shown that identifying the drivers of shoreline development on a global scale using SDS has great potential for sustainable coastal management in the face of future challenges

Coastal areas around the world have always been densely populated areas. However, sea-level-rise and an increase in single extreme events due to climate change, threaten the coastal areas and their inhabitants. Governmental organizations, coastal managers and various private parties thrive for better insights into long-term shoreline behavior for sustainable decision-making. Currently, these insights are gathered by process-driven models that are often time expensive, require local input and calibration, and are often limited by equations. However, satellite imagery proved to be a promising data source to derive historic shoreline behavior on a global scale. Machine Learning (ML) algorithms are suggested to create an extra in-depth understanding of these satellite images. The increase in the availability of both satellite data and ML algorithms opens possibilities for shoreline forecasting on a global scale. This research aims to improve forecasting of shorelines by creating a global model in which cross-time series information can be used. To encourage a meaningful exchange of time series information, a clustering forecasting approach is proposed.This research builds upon the yearly SDS data behind the Shoreline Monitor of Deltares. The SDS dataset captures shoreline evolution and behavior on a global scale by quantifying it in time series. The SDS dataset consists of transects every 500 m along the global coast and contains 33 years (1984-2016) of shoreline evolution. First, in order to cluster transects, the features for clustering were defined. Here, clustering features were divided into time series features and hydraulic and geomorphic features. Whether transect clustering could improve shoreline forecasting, based on these time series and hydraulic and geomorphic features, was explored. For this thesis, it was decided to focus on time series features for further clustering usage. Eventually, a shape-based transect representation was chosen as feature for input of the clustering algorithm. Around 350,000 globally distributed sandy transects were assigned to nine different main clusters, with a semi-unsupervised clustering approach. These nine clusters captured global sandy shoreline behavior ranging from extreme erosion to extreme accretion and proved to be a practical tool to quantify the (distribution of) shoreline behavior on different spatial scales. Subsequently, a second supervised step was developed to create sub-clusters to gain more insights into the trends of the nine clusters. With this supervised sub-clustering step, three sub-clusters were generated for each of the five largest clusters, allowing for the distinction between accelerating and decelerating behavior in the last decade of time series per cluster. Global, country-level and local case studies showed the potential of these sub-cluster refinements. Hereafter, the nine main clusters were separately used as input for four different forecasting algorithms. This resulted in predictions per cluster with a more reliable and more accurate seven-year forecast for >95 % of the transects, sometimes improving accuracy up to factor 15, compared to recently published work. Furthermore, a multi-method approach was created to determine the most reliable forecast, on a local scale by combining overall accuracy and information gained during clustering. Time series clustering enhances the forecasting of transects based on overall accuracy and local reliability. The multi-method approach for selecting the appropriate prediction on a local level, could be a practical tool for governments to apply. Besides, a vulnerability assessment explores the possible application of the predictions and underlines the need for global databases regarding the coastal zone. However, further research should consider incorporating hydraulic and geomorphic features to strengthen the clustering of transects. Hence, shoreline forecasting based on historical data should be considered as a valuable tool for sustainable decision-making regarding coastal zones.

Offshore wind farms are being built at an unprecedented pace in the North Sea. The Dutch government is aiming for a CO2 neutral energy supply by 2050 and vast cost reductions have turned (offshore) wind energy into a worthy competitor for other (green) energy sources. Most of the planned wind farms in the Dutch North Sea are located in areas where the seabed is covered with sand waves. In the North Sea these sand waves have lengths of 100-1000 m, heights of 1-8 m and they migrate with rates of up to 10 m/year (Morelissen et al., 2003). Sand wave migration and changes in shape, may cause a significant rise or drop in local bed level. This bed level variation over time could decrease the stability of foundations or bed protections or cause exposure of cables and pipelines.
Various offshore infrastructural projects, like offshore wind farms, thus require long term (30-50 years) predictions of the seabed dynamics. Currently data-driven methods are used to determine the range of expected bed levels. However, the uncertainty in these predictions is significant, with sand waves being the largest source of uncertainty. Furthermore, no real understanding of the systems at hand forms the base of these predictions. Not many attempts have been made to accurately model sand wave dynamics in real-life situations using a process based model. Since sand waves often have steep slopes in migration direction, a need for small numerical grid sizes arises. On the other hand oceanic hydrodynamics are affected by large scale bed forms. To include (the influence of) these bed forms, large model domains are required. This makes numerical modelling of sand wave fields rather difficult due to the balance between grid sizes and computation time. The newly developed Delft3D Flexible Mesh (FM) model may be able to overcome some of these problems.
Through the use of unstructured grids, the desired level of detail can be reached in certain sand wave areas. In combination with the possibility to run simulations in parallel, on multiple cores, computation times can be reduced significantly. However, the Delft3D FM model has not yet been used for the prediction of sand wave dynamics. The aim of this research is to find the opportunities and challenges of the Delft3D FM model for quantitative modelling of sand wave dynamics in the North Sea.

Through two case studies various opportunities and challenges for predicting sand wave dynamics using Delft3D FM are discovered. The Delft3D FM model showed a significant reduction of computation times for a 2DV case using a single core compared to Delft3D-4. For parallel simulations, using multiple cores, an approximately linear further reduction of computation time is observed in a 2DV setting. Furthermore, the possibility of unstructured grids presents a solution for the small grid sizes needed in sand wave areas. Other computational
gains are realized through a morphological scale factor (which is also
ncorporated in Delft3D-4), optimized time-step management and a new type of boundary conditions. This new boundary condition imposes both water level and flow velocity over depth in both horizontal directions. In this way the local flow conditions are accurately represented and the influence of bed forms outside the model domain can easily be incorporated. This potentially eliminates the need for buffer zones.
Challenges for the application of Delft3D FM on sand wave cases are found in amongst others the availability of data. When less data on local hydrodynamics is available the accurate representation of processes driving sand wave dynamics becomes more difficult. Furthermore the inclusion of sub-grid processes, like the growth and migration of megaripples, could be problematic. In the model case studies no calibration was carried out. This calibration could potentially increase the effort needed to apply Delft3D FM to real life sand wave cases.
Furthermore, the Delft3D FM model is still in development and significant differences between the results of different versions of Delft3D FM were observed. Care should be taken when applying a new version of Delft3D
FM. The model is however being developed in collaboration with users which ensures quick feedback and thus stimulates improvement of the results between versions.
Recommended research includes extended exploration of the 3D effects influencing sand wave dynamics. Furthermore, improvement of morphological results and optimization of the model set-up will increase the applicability
of Delft3D FM in an engineering setting. A model run forced by two Riemann boundaries showed improved morphological results, although the hydrodynamics were not well represented. These results might indicate
where differences with reality originate. Examples of such differences are overestimation of peak velocities, exclusion of wind-driven currents and exclusion of processes like suspended sediment transport, free surface waves
and grain size sorting. Further exploration of these factors could enhance the predictive capacities of Delft3D FM. Applying the model at other locations in the North Sea will help determine the overall applicability of the model.
Through this extended research the full potential of the Delft3D FM model can be discovered and prepared for future engineering applications. Insights gained into the processes influencing sand waves dynamics, through the use of Delft3D FM, could pave the way for more nature based solutions, thus reducing the need for dredging. In this way Delft3D FM could contribute to reducing risks, costs and environmental impact of offshore construction projects in sand wave areas.

In recent years, the awareness of climate change and sea level rise has grown. Especially in low lying countries like the Netherlands, with a large river delta with a high economical and natural value, extreme variations in water levels will have great impact. Former used solutions proofed to be expensive and in some cases disturbed the dynamics of nature. Consequently, the need of new solutions of flood protections grows in which the negative impact on nature environments should be limited. The Delta21 plan proposes a future-proof design for the Dutch Southwest delta, that aims to tackle the growing need for flood protection, while contributing to nature and storing energy all at once. However, the coastal structure, located in the Haringvliet mouth, is likely to produce large disturbances in the environment. Since the Haringvliet mouth is located in the 'Voordelta', a Natura2000 area, it is essential to study the initial response of the proposed interventions to the hydraulics and morphodynamics that shape the system. Hence, the objective of this research is to determine the initial impact of the Delta21 plan on the large scale multi-year average hydro- and morphodynamics of the Northern Voordelta. To this end, a literature review was performed to analyse the evolvement of the Northern Voordelta since the closure of the Grevelingen and Haringvliet estuaries by the Delta Works in 1970. This study showed that the closure caused the tidal prisms to reduce strongly resulting in the tidal propagation to change from cross­shore to longshore. The remaining ebb tidal deltas started to follow the characteristics of a short basin, in which water levels offshore and nearshore are in phase. Flow velocities reduced strongly and channels started to fill up, at the foreshore waves started to dominate the morphodynamics. Waves began to 'bulldoze' the former ebb-tidal deltas onshore, creating shore parallel sandbars like the Bollen van de Ooster and the Hinderplaat. Besides, the construction of the Maasvlakte 2 triggered further sedimentation of the Haringvliet mouth, landward of the Hinderplaat, by sheltering the area from NW waves. The initial response of the system to Delta21 has been studied using a short-term morphostatic 2DH computational model simulation. The complexity of the processes inside the study area caused the need of the process based model Delft3D, in which flow, waves and sediment transport computations can be carried out. To reduce computational times, the applied modelling setup made use of reduction techniques and schematisations of the hydraulic forcing, while maintaining the natural character of the system. The computed flow and sediment transport patterns were used as indicators, in order to predict and compare the feedback of the system of the various scenarios. Model results showed that the tidal propagation inside the Haringvliet estuary, which is opened in the Delta21 scenario, changes to cross-shore again. Similar to the situation prior to the closing, the area will start to follow the characteristics of a long-basin. This implies that water levels in the estuary are out of phase with the offshore water levels. A cross-shore tide dominated flow prevails in the area, while the Grevelingen delta remains unchanged in the Delta21 scenario. The cross-shore action influences the offshore wave- and setup-driven residual currents. In the same offshore area, the Delta21 pump discharge develops a similar cross-shore acting influence area. Model results further showed that the Delta21 geometry causes a 'tidal squeeze', similar to the present-day processes around Maasvlakte 2. Altogether, the large disturbances to the hydrodynamic forcing, triggers morphodynamic activity. Concluding from literature and the model results, this leads to the erosion and formation of ebb and flood channels in the Haringvliet mouth and the formation of an ebb- and flood-tidal delta around the inlet. Processes in and around the ebb-tidal delta adjust towards a balance between the wave and tidal action. These findings agree with the known processes of a mixed-energy tidal inlet system that is adjusting towards a dynamic equilibrium. Conceptualised results are shown in Figure 1. Concluding, the analysis on the initial response of the system, as well as the expected initial dynamic behaviour, provides useful insight and a good basis for future optimisation studies on the Delta21 design. This study reveals the disturbed areas caused by the presence of Delta2 l and relates it to the contribution of the various hydrodynamic processes. Furthermore, the development of this type of study can be significantly relevant to the evaluation of other preliminary studies on solutions to sea level rise.

The coastal area provides important services such as valuable habitats for wildlife, resources for regional development, and buffer zones for the land against natural disasters such as storm surges. But these narrow coastal areas experience pressure from both land and ocean side. In order to regulate sustainable coastal development, protect coastal areas from natural forces, implement ecosystem protection strategies and mitigate the impacts of climate change on coastlines, it is crucial to observe and quantify the changes along coastlines that are vulnerable to these pressures. Among the various coastal environments, the focus of this master thesis is on muddy coasts. A muddy coast is defined as a coastal depositional environment that is usually formed along an unsheltered coastline exposed to low energy conditions and that consists mainly of fine sediments, which are smaller than 63 micrometres in diameter. So far, few methods have been developed to detect sandy coasts and the rate of change of sandy coastlines at global scale, but little is known about the detection and behaviour of muddy coasts. The main objective of this thesis is to develop an automated classification method to identify muddy coasts along all the coastlines of the world. The main objective can be translated into the following research question: How can a reliable global mud classification be obtained by analysing the characteristics of muddy coasts using publicly available satellite remote sensing techniques and globally available coastal datasets? To obtain a reliable global mud classification, a method based on the spectral properties of individual mud patches and on the physical geographical characteristics of muddy coasts is proposed.
The spectral properties of individual mud patches are studied at pixel level using multispectral satellite images and the physical geographical characteristics of a muddy coastal system are studied using seven globally available coastal datasets. The information from the multispectral images and the information from the coastal datasets are used as the input of the development of a hybrid coastal transect classification model, in order to obtain a more reliable and robust classification model that can detect muddy coasts at global scale. With supervised machine learning the hybrid coastal transect classification model is developed, which can classify global coastlines into five coastal types: plain beaches, muddy coasts, coastal cliffs, vegetated coasts and other. 85 percent of the muddy coasts are correctly recognised by the hybrid classification model. The accuracy of the hybrid model is still increasing significantly, meaning that the model could perform even better if more training data is added to train the model. The large amount of training and validation data at both pixel and transect level resulted in a reliable, robust global mud classification model. With the addition of the coastal datasets, the model performs better than when only the multispectral satellite images are used for classifying muddy coasts. The hybrid classification model is used to classify 100000 global coastal transects from which 12 percent are classified as muddy coasts; 60 percent of the classified muddy coasts are in the tropics.

The Sandwindmill system could decarbonize the Dutch coastal protection whilst harmonizing with the building with nature approach. The system consists of a pipeline connecting an offshore borrow location to a nearshore area, where the mined sediment is discharged. The equipment is powered by wind turbines. As long as the wind blows sufficiently, sediment is pumped through the system, leading to a near-continuous nourishment. After the sediment has left the pipes, natural forces should distribute the material. Despite the considerable environmental benefit, one substantial challenge is the financial competitiveness with the traditional hoppers. This research aims to provide insights that can enable the system to be also financially competitive to the hoppers. In the first part, an integral analysis of the Sandwindmill is carried out. The second part of this thesis focuses on the optimization of the dispersion of the nourished sediment.

In the first part, all sub-systems of the Sandwindmill concept are treated separately. Wind data, theoretical formulations, and an exploration of the mining options are used to identify optimization opportunities. From the assessment, it becomes apparent that the interdependency of these sub-systems complicates the cost-optimization. Hence, a competitive system design requires an accurate harmonization of these sub-systems. Three main conclusions are drawn. First of all, the costs per cubic meter decrease with an increasing nourishment volume. Finding the marginal costs is essential in determining the feasibility of the system for certain volumes. Secondly, it is concluded that - given an annual nourishment volume - the pump capacity and windmill size should be attuned. Their cost-optimum is found at a set-up that leads to a yearly operational time of approximately 70%. Lastly, the analysis shows that the application of batteries to support the system in case of lower wind velocities can contribute to a more economical system. This is mainly the case if the system has wave-induced limited operational times.

Due to the costs of the displacement of the pipe outlet and the financial benefits of nourishing large volumes, the second part of the research aims at generating guidelines to designing a dispersive nourishment. Both wave- and tide-dominated systems are a potential field of application for the Sandwindmill. North-Holland is selected as an appropriate case study, containing the Marsdiep tidal inlet and a wave-dominated closed coastal section. A process-based coastal area model is set up to determine the sensitivity of different nourishment strategies on the dispersion at a one-year timescale. A tidal channel wall nourishment at the Marsdiep and a shoreface nourishment at Callantsoog are assessed.

One of the key findings of the second part is that the strong tidal velocities at the Marsdiep inlet are more capable of transporting the nourishment than the conditions at the wave-dominated coast. After one year, the distance travelled by a significant part of the nourished sediment is a factor four higher for the Marsdiep nourishment. Secondly, the research shows that the grain size plays a vital role in the local and regional dispersion. Therefore, the borrowed grain size should be part of the system design. Under the evaluated scenarios, the nourishment application under forcing conditions with a higher transporting potential has a smaller effect in the longer term. This implies that the optimal operational time could be determined solely based on the cost-optimization and should hence be around 70%.

The world's coasts are at risk: an estimated 24% of the world's beaches are eroding. Extreme events such as storms continuously threaten the coastal region and pose a serious risk of coastal flooding. Under the influence of climate change, the seaward pressures are only expected to increase. Sea level rise is accelerating, storm intensities and frequencies will increase, and precipitation will become more extreme. At the same time human pressures such as coastal tourism and rapid urbanisation impose stresses to the coastal system. An estimated 23% of the world's population lives in the coastal zone where natural hazards cause a direct flood risk. Protection of the coastal region becomes more and more relevant. Coastal management strategies need to take the local circumstances into account. Therefore, a good understanding of the coastal zone is needed, but especially in data-poor regions there is a lack of geomorphological information such as slope and grain size. This thesis presents an alternative approach for acquiring this data with the ability to run on a global scale using remote sensing technologies. Remote sensing, satellite imagery in particular, has proven to be a promising new technology to monitor coastal regions at large temporal and spatial scales and is applied here to determine coastal slopes and sediment sizes. The approach has shown promising results and paves the way for an estimate of the grain size at beaches around the world. Some hurdles remain to be taken but the ability to do measurements of coasts around the globe using satellite imagery will change the way we study our coasts forever. This research is a major step forwards in the global monitoring of the world's coast, but only shows the beginning of the endless possibilities.

Research on the physical, ecological and environmental processes in the coastal zone often rely on the availability of accurate bathymetric information. The traditional method to acquire these data is by conducting depth soundings from a vessel, which is a very accurate way for measuring bottom elevations in all kinds of water, but has as a downside that the acquisition of data is slow. The rate data collection is limited by the vessel speed and swath, which for instance results in months of surveying to accurately map the entire Dutch coastal zone. Therefore, it is mostly applied in targeted surveys where bathymetric information is required for instance in construction project or detailed coastal analysis and modeling. It is not also feasible to conduct an entire survey for such projects and therefore remote sensing data has gained popularity in the past decades.
Multiple methods of bathymetry derivation from satellite data has been topic of investigation, One of those is the concept of light attenuation, which was already covered in publications from the 1970s but is still used in many (coastal) researches. A simplification of the complex radiative transfer theory was made by assuming a log-linear relationship between reflected radiance and water depth. This simplification implies that calibration with known elevation data is needed to obtain an accurate bathymetry. These empirical Satellite Derived Bathymetry (SDB) methods were successfully used in multiple researches the past years, but those mostly focused on single locations with typical good SDB conditions (clear water with a high reflective bottom). The main goal of this research was to gain insight in the possible applications and limitations of empirical SDB methods when optical conditions are not as ideal as in previous researches. A multi-case evaluation of these methods was conducted at several different locations on earth with each having particular conditions that affected the reflected radiance ranging from the "most ideal" case with clear water to a case where SDB is hardy possible because of high levels of turbidity.
Promising results were obtained in cases where one can be certain that the measured reflectance is coming from the bottom, so in clear waters with hardly any suspended sediment and waves that disturb the path of light to the bottom. The results, with vertical errors below 50 cm, were in accordance to earlier researches. Also in more complex cases, where water clarity was affected by some levels of turbidity, a reliable bottom was still derived, but vertical errors where increasing. The determining factor whether it was possible to use SDB was if measured reflectance was coming from the sea floor. The full potential of a single satellite image was further analyzed by spatial analyses which showed that a small amount of data already gives enough information of calibrate a model to an accurate SDB. However, these results all rely on the initial reflectance values measured by the satellite's sensor These initial values were analyzed more elaborately without the use of calibration data. The concept of the attenuation of light in water was applied to satellite images to assess the bottom derivation without performing any calibration of the algorithm. It was found that, even after correction for atmospheric effect, measured reflectances on images that were captured at different time varied per image, which hindered the temporal analysis. With the addition of ground-truth data, these fluctuations are eliminated by vertically shifting the bottom to the absolute correct level, but when no data is available, differences in measured reflectance does not necessarily be caused by vertical bottom changes. A contribution of these fluctuations was found in the solar zenith angle, which is differs over the year.
With SDB it is possible to derive accurate bathymetric maps under good optical condition. When these conditions get more challenging, the uncertainty in the bottom reflection increases, resulting in an increase of uncertainty in SDB as well.

Knowledge of seafloor topography (bathymetry) is increasingly important as coastal environments are unprecedentedly stressed by climate change and anthropogenic pressure. The bathymetry of shallow nearshore waters is yet marginally monitored due to costly and time-intensive survey techniques. Methods to obtain satellite derived bathymetry (SDB) have become increasingly valuable. Mapping temporal change is however challenging, because the majority of these methods remain heavily dependent on situ observations. This thesis introduces an SDB approach to estimate temporal bathymetric changes, which omits the need for synchronous in situ data. The approach is based on a reference image correction method that enables direct comparison of multitemporal imagery and temporal extrapolation of a conventionally-trained bathymetry estimation model (BEM). Research focused on pre-processing multispectral imagery, developing a bathymetry estimation model and estimating bathymetry for times of absent in situ data. The proposed method is demonstrated with a case study in the Dutch Wadden Sea; a site characterised by dynamic morphology, high turbidity and homogeneous bottom type. A log-linear estimation model is obtained by linear regression on in situ observations and the three visible bands of Sentinel-2 imagery. Scarcity of high-quality Sentinel-2 imagery is managed by combing multiple images into a six-month composite. The availability of two sets of vaklodingen in situ observations allowed for training and testing two bathymetry estimation models (BEM 2016 and BEM 2019) and for cross-validating the depth estimates after a three-year extrapolation of these models. Bathymetry is estimated for times of absent in situ data by temporal extrapolation of the two estimation models. The extrapolation showed estimation of shallow bathymetric structures in up to four metre water depth with an RMSE of approximately one metre. Additionally, the migration direction of these bathymetric structures is successfully estimated. Within the tested three-year time frame, predictive power did not decrease. These results imply that estimation performance is governed by composite quality and predictive power of the bathymetry estimation model. The limited influence of temporal extrapolation on estimation performance suggests that the availability of high-quality satellite imagery and one set of non-synchronous in situ observations is sufficient to estimate bathymetry for times of absent in situ data. The proposed method potentially provides a tool for mapping temporal bathymetric changes of nearshore zones across the globe.

Today's coastal zones are densely inhabited as the majority of the world's population lives in these attractive areas. The shorelines in coastal zones are shaped by complex spatial and temporal variable interactions between natural forcings like changes in mean sea-level, tides, wave and wind conditions, and storm surges. Besides, natural hazards such as coastal erosion, tropical cyclones, hurricanes, typhoons, floods, salt intrusion and tidal surges threaten a major part of the world's population. Furthermore, climate change is likely to increase the risk of natural hazards. As a response, humans changed the world's shorelines and the forcing-driven processes that work on them, to increase protection against hazards and keep supporting their activities. These human interventions are deployed discontinued in time, disperse spatially and might result in negative consequences leading to human-induced hazards. This research focuses on one particular hazard to coastal communities, coastal erosion, which is extended to a term referred to as shoreline evolution by incorporating coastal accretion as well. Up till now, detailed local-scale studies are able to expose human and natural drivers of shoreline evolution and provide a possibility to make a step towards intentional rather than accidental coastal engineering. Digital imaging and, more recently introduced, satellite imagery proved to be a promising new technology to measure and monitor shoreline evolution at bigger temporal and spatial scales. Nevertheless, the opportunity to develop a model that exposes the drivers of shoreline evolution on a planetary scale remains unexplored. This is due to the required computational effort as well as the large variability in coastal systems around the world. With an ever-increasing data availability, data-driven models incorporating Machine Learning (ML) proved to be an efficient alternative approach to heavy computing classical process-driven models in civil engineering practice. Next to this, a coastal classification can be used as a means to inventory the aforementioned variability. Therefore, the research objective in this study is to explore the possibility of exposing and classifying the drivers of shoreline evolution on a planetary scale, by employing ML on satellite imagery. Approximately 390.000 km of shoreline is analyzed for the past 33 years. This resulted in a statistically derived classification of natural and human-induced sandy shoreline evolution. By elaborating on this classification, it is found that natural and human-induced shoreline evolution accounts for approximately 16 and 25% of the total of globally exposed and classified shoreline evolution signals respectively. All outcomes in this research can support detailed local scale investigations and therefore provide an enhanced opportunity to make a step towards intentional rather than accidental coastal engineering. Hence, it is concluded that the developed and applied methods that employ ML on satellite imagery can be used to expose and classify (in)direct human and natural influences on sandy shoreline evolution using spatial and temporal characteristics on a planetary scale. 57% of the shoreline evolution signals is present in a regime with complex and combined (compound) influences, which still requires (rather than supports) local scale investigations to determine the correct influence or driver. More research is required to elaborate on the opportunities that can enhance insight in the compound regime, to improve the obtained results and advance the applicability of this study.

Accelerated sea level rise (SLR) is predicted to have multiple adverse impacts on the coastal zone, aggravating phenomena such as coastal erosion on sandy coasts. For climate change adaptation planning and informing policy, morphodynamic changes occurring at coastlines are becoming increasingly important. In this study, a calibrated Delft3D model forced by real-time wave conditions, was applied to simulate and assess the morphological behaviour of the Delfland coast in response to accelerated SLR over a 30-year time period. The calibrated Delft3D model uses a novel acceleration technique called brute-force merged (BFM) proposed by Luijendijk et al. (2019), which enables the modelling of multi-decadal predictions, with significant gain in computational effort. An assumption of the study was that no nourishments take place, i.e. no additional sediment supply. The Sand Engine (Zand Motor in Dutch), currently located along this coast was also excluded from the model, thereby assuming a straight unnourished coastline. A selection of six SLR scenarios was simulated, including a no SLR scenario used as the reference case. The chosen scenarios covered the full bandwidth of accelerated SLR projections translated for the Dutch coast up to 2100, assuming increased mass loss from the Antarctic ice sheet, a hypothesis proposed by DeConto and Pollard (2016). These projections therefore exceed those presented in the IPCC AR5. Based on the recent literature, the SLR rates selected, ranging from 3 mm/year to 120 mm/year, are assumed plausible and useful for the modelling study. Model outputs that are assessed include erosion and sedimentation plots, and volume changes, particularly erosion volumes. Analysis shows that no major change to the general coastal system behaviour occurs due to accelerated SLR; erosive sections in the south remain erosive and accretive regions in the northern part remain accretive. This is influenced mainly by gradients in alongshore sediment transport and presence of structures. Erosion volumes increase with higher SLR rates, indicating an increase in erosion rate due to accelerated SLR. Volume changes were calculated in different alongshore sections and in different depth/elevation zones in the cross-shore direction in order to identify regions more vulnerable to accelerated SLR. It is determined that the southern section is most impacted by SLR causing increased erosion, particularly in the subtidal zone. Processes driving the observations and trends are discussed in the study. With significant SLR, dune erosion also occurs due to water levels and waves being able to reach higher elevations. The dunes along the Delfland coast are the primary sea defence which protects the hinterland from flooding, therefore it is critical to consider potential dune erosion due to accelerated SLR. Implications of the model study’s findings are briefly discussed, in the context of coastal maintenance policy and implementation of nourishments. Evaluating the added value of Delft3D as a coastal impact model is another objective of the study. Delft3D shows a number of benefits, including detailed analysis at multiple spatio-temporal scales. Another is the inclusion of transport processes in both alongshore and cross-shore directions, which is not the case for the Bruun Rule or 1-D coastline models. A limitation of Delft3D is that it does not include aeolian transport processes, and so dune growth cannot be simulated.