To reduce computational efforts, surrogate models have been developed for dune erosion prediction. Current surrogate models can describe the relationship between the XBeach input and output (Athanasiou, 2022) and provides a prediction of a morphological indicator based on a parameterized input (profile shape parameters and hydrodynamics). In this research, first steps are taken to set-up a surrogate model that is able to deal with spatial input and the prediction of actual profile shapes along the Holland Coast. Using a convolutional neural network structure (U-Net), several model set-ups are explored and scaled-up to a realistic storm scenario along the Holland Coast.

BaCla is a new stochastic parametric precipitation model to estimate rainfall associated with Tropical cyclones (TCs) in a computationally efficient way. It is validated for a number of calibration cases along with the current benchmark IPET deterministic method. The results of these models are compared with the observed StageIV-based rainfall. Two predictive parameters, the pressure deficit [hPa] (Δ𝑃) and maximum sustained wind speed [m/s] (vmax) are tested for the BaCla model. Frank copula is used by the BaCla model to predict the peak amount of rainfall. It employs the adapted Holland wind profile to create a 1D rain profile. Asymmetry may be added to make a 2D rain profile. The calibration study challenges the assumption that the radius of maximum wind speed (Rvmax) is equal to the radius of maximum precipitation (Rpmax) in the BaCla model. Additionally, it is observed that the radial fit overestimates rainfall in scenarios where the peak amount of rainfall is less than 2.8mm/hr. These observations lead to modifications in the relationship between Rvmax and Rpmax, the radial fit threshold, and the fitting coefficients of the adapted Holland wind profile for rainfall distribution in the improved BaClHa model. The new BaClHa model is verified with new and calibrated sets of TCs. The BaClHa model shows potential in achieving good accuracy along with global applicability at a limited computational expense.

All over the world coastal communities are at risk due to sea-level rise and intensifying weather conditions. Many sandy beaches are eroding as a result of human-induced factors. Currently, the preferred coastal protection measure in the United States are beach nourishments. In Europe, there also is a general shift from hard to soft coastal protection measures. However, beach nourishments are not a long-term solution. Recently, in the Netherlands, a new concept called a large-scale (mega-feeder) nourishment has been introduced (the Sand Engine). Numerous studies on this new concept have been conducted. However, not for a mega-feeder nourishment nearby a tidal inlet system. About 10\% of the world's beaches consist of barrier islands. Emphasizing the importance of investigating the development of a mega-feeder nourishment nearby a tidal inlet system, under various hydrodynamic conditions. Therefore the research question is as follows: “How does a nearby tidal inlet system influence the development of a mega-feeder nourishment?” The research question is answered by investigating the effects various hydrodynamic conditions have on the development of a mega-feeder nourishment nearby a tidal inlet system. This is done for fixed morphodynamic features, such as the dimensions of the tidal basin and the dimensions and orientation of the tidal inlet. The only variable morphodynamic feature is the alongshore position of the mega-feeder nourishment. Four distinct hydrodynamic scenarios are modelled to investigate their effects on a mega-feeder nourishment. The tidal range (η), significant wave height (Hs), peak wave period(Tp) and peak wave direction (Dp) are varied. This resulted in the following hydrodynamic scenarios: 
•Mild wave conditions: (η = 1.5m; Hs=1.0m and Dp=0°);
•Oblique wave conditions: (η = 1.5m; Hs=1.0m and Dp=-45°);
•Storm wave conditions: (η = 1.5m; Hs=variable and Dp=0°);
•High tidal range: (η = 3.0m; Hs=1.0m and Dp=0°).
These hydrodynamic conditions and their effect on a mega-feeder nourishment are modelled by utilizing a process-based numerical model called Delft3D. In Delft3D, two locations of the mega-feeder nourishment per hydrodynamic scenario are evaluated. A mega-feeder nourishment is placed at an alongshore distance of 2 kilometers and 5 kilometers from the tidal inlet. This to get insight in the tidal flow nearby a tidal inlet and up to what alongshore distance this tidal flow affects the development of a mega-feeder nourishment. The hydrodynamic conditions were simplified, meaning steady wave characteristics and a single M2 tidal constituent. Using real time-varying hydrodynamic conditions yields similar results compared to the simplified hydrodynamic conditions. Therefore, simplifying the hydrodynamic conditions is justified. The results show that there will be additional erosion near a tidal inlet if the mega-feeder nourishment is located inside the influence of the tidal inlet. The influence is the alongshore distance where the currents owing to the tidal inlet (residual currents) still affects the total alongshore sediment transport (larger than 50 m³/6y/m). The alongshore distance of the influence increases with an increasing tidal range (tidal prism). However, there is no shoreline retreat owing to the tidal inlet at the location (2km from the tidal inlet) of the mega-feeder nourishment over a time period of 6 years. Only the adjacent coast on the inlet-side of the mega-feeder nourishment erodes significantly more than without a tidal inlet, with an increasing magnitude in the shoreline retreat towards the tidal inlet. Hence, it is expected that if the mega-feeder nourishment is placed close to the tidal inlet (i.e. several hundreds of meters), then the influence of the currents owing to the tidal inlet will enhance the shoreline retreat at the location of a mega-feeder nourishment. To conclude, the tidal inlet does influence the development of a mega-feeder nourishment nearby a tidal inlet (order several hundreds of meters). However, this is not necessarily seen in the retreat of the shoreline but instead in deeper depth contours. The governing process in the sediment transport at a mega-feeder nourishment is the incident wave angle for small tidal ranges (η < 1.5m) and mild wave conditions (Hs > 1m). However, for a large tidal range (η > 3.0m) the residual currents owing to the tidal inlet will become the governing process.

Wave runup is generated by energy which remains after wave breaking and travels farther to the coast in the form of a bore. It can be seen as a thin wedge of water running up the beach face (Brocchini and Baldock, 2008). Under storm conditions runup is responsible for beach and dune erosion and accurate runup predictions are therefore required (Ruggiero et al., 2001; Stockdon et al., 2005). For runup and its components, the time-mean setup component and the time-varying swash component, empirical parameterizations have been developed in the past, but they cannot be validated for storm conditions due to a lack of data (Stockdon et al., 2005). The data gap can be filled by numerically simulated runup, for example with the process-based XBeach model. XBeach is a depth-averaged model which predicts nearshore hydrodynamics and can be used in a phase-averaged or a phase-resolving mode. However, both the significant incident and infragravity swash is underpredicted by the phase-averaged XBeach Surfbeat model (Palmsten and Splinter, 2016; Stockdon et al., 2014), which does not resolve incident wave motions. In order to predict runup under storm conditions with confidence the performance of XBeach under mild conditions should be assessed. Here runup simulated with XBeach Surfbeat and the phase-resolving XBeach Non-hydrostatic for the intermediate reflective beach of Duck was compared to measurements of the SandyDuck'97 experiment, where mild offshore conditions were present. A 2DH model was set up using measured bathymetry and forced with measured frequency-directional spectra. The hydrodynamics responsible for a difference in runup prediction were investigated and their origin in the cross shore was identified. It was shown that the prediction of significant incident and infragravity swash can be improved by using the phase-resolving XBeach Non-hydrostatic model instead of the XBeach Surfbeat model, while performance for setup remains similar. Incident swash predictions are improved by resolving the incident wave motions. The major part of the improvement in infragravity swash predictions is driven by differences in infragravity wave transformation between the two XBeach models. A small part also originates within the swash zone, for which incident bore merging can be a possible explanation. The difference in infragravity wave height predictions between the two XBeach models mainly develops in the surf zone where a different response to directional spreading and different degrees of shoaling most likely can explain the difference in infragravity wave height. Against expectations no correlation with the groupiness of the incident waves or with the phase difference between wave group and infragravity wave was found. A small part of the difference in infragravity wave height predictions is already present near the offshore boundary and probably results from interaction processes between high and low frequency wave boundary conditions. It can thus be said that on intermediate reflective beaches, where both incident and infragravity waves play a role, the resolving of incident wave motions is a necessity to predict runup accurately. In these situations the phase-resolving XBeach Non-hydrostatic model should therefore be used, instead of the phase-averaged XBeach Surfbeat model. Here only an intermediate reflective beach and a small range of energetic conditions were included. More types of beaches and storm condition forcing should be investigated to be able to validate the empirical parameterizations but also to further indicate applicability ranges of the two XBeach models. Also, more attention should be paid to hydrodynamic differences at the boundary and in the surf zone in order to find more conclusive reasons for differences between the two XBeach models.