The volume of material required for the construction of new and expansion of existing beach sites is an important parameter for coastal management. This information may play a crucial role when deciding which beach sites to develop. This work examines whether artificial neural networks (ANNs) can predict the spatial variability of nourishment requirements on the Croatian coast. We use survey data of the nourishment volume requirements and gravel diameter from 2016 to 2020, fetch length, beach area and orientation derived from national maps which vary from location to location due to a complex coastal configuration on the East Adriatic coast, and wind, tide, and rainfall data from nearby meteorological/oceanographic stations to train and test ANNs. The results reported here confirm that an ANN can adequately predict the spatial variability of observed nourishment volumes (R and MSE for the test set equal 0.87 and 2.24 × 104, respectively). The contributions of different parameters to the ANN’s predictive ability were examined. Apart from the most obvious parameters like the beach length and the beach areas, the fetch length proved to be the most important input contribution to ANN’s predictive ability, followed by the beach orientation. Fetch length and beach orientation are parameters governing the wind wave height and direction and hence are proxies for forcing.

Croatia’s coast located on the eastern Adriatic is rich with small gravel beaches with limited fetch. This leads to a specific low-energetic wave climate compared to most other beaches, while their gravel composition makes them unique. Most management of these beaches is performed without understanding the sediment transport occurring on the beaches. XBeach-Gravel is a numerical model capable of simulating bed-level change on gravel beaches, but lacks validation in the case of low significant wave height (under 2.5 m) and low peak periods (under 6 s), conditions that are present on the eastern Adriatic. Based on measurements performed in both laboratory conditions in a water canal in Hannover and actual storm wave conditions on Ploče beach, calibration of the model is performed. Model results are compared between laboratory conditions and field conditions for comparable wave conditions. XBeach-Gravel can simulate low-energetic events resulting in berm formation and berm buildup with a high Brier skill score if calibrated. Simulation of laboratory conditions requires high transport coefficient values and shows more sediment transport than similar wave conditions in the field. Calibration for field conditions is dependent on geodetic survey data capable of isolating wave events with dominant cross-shore transport, but once calibrated, XBeach-Gravel can achieve good to excellent Brier skill score values in simulating sediment change in low-energetic wave conditions on the eastern Adriatic.

Long-term time series of wave parameters play a critical role in coastal structure design and maritime activities. At sites with limited buoy measurements, methods are used to extend the available time series data. To date, wave hindcasting research using machine learning methods has mainly focused on filling in missing buoy measurements or finding a mapping function between two nearshore buoy locations. This work aims to implement machine learning methods for hindcasting wave parameters using only publicly available Copernicus data. Ensemble regression and artificial neural networks were used as machine learning methods and the optimal hyperparameters were determined by the Bayesian optimization algorithm. As inputs, data from the MEDSEA reanalysis wave model were used for the wave parameters and data from the ERA5 atmospheric reanalysis model were used for the wind parameters. The results of this study show that the normalized RMSE of the test data improved by 29% for Rijeka and 12% for Split compared to the original MEDSEA wave hindcast at buoy locations. The proposed method was extremely efficient in removing bias in the original MEDSEA hindcasts (e.g., NBIAS = -0.35 for Rijeka) to negligible values for both Split and Rijeka (NBIAS < 0.03).

Wave data play a critical role in offshore structure design and coastal vulnerability studies. For various reasons, such as equipment malfunctions, wave data are often incomplete. Despite the interest in completing the data, few studies have considered constructing a machine learning model with publicly available wind measurements as input, while wind data from reanalysis models are commonly used. In this work, ANNs are constructed and tested to fill in missing wave data and extend the original wave measurements in a basin with limited fetch where wind waves dominate. Input features for the ANN are obtained from the publicly available Integrated Surface Database (ISD) maintained by NOAA. The accuracy of the ANNs is also compared to a state-of-the-art reanalysis wave model, MEDSEA, maintained at Copernicus Marine Service. The results of this study show that ANNs can accurately fill in missing wave data and also extend beyond the measurement period, using the wind velocity magnitude and wind direction from nearby weather stations. The MEDSEA reanalysis data showed greater scatter compared to the reconstructed significant wave heights from ANN. Specifically, MEDSEA showed a 22% higher HH index for expanding wave data and a 33% higher HH index for filling in missing wave data points.