Accurate and computationally efficient prediction of wave run-up is required to mitigate the impacts of inundation and erosion caused by tides, storm surges, and even tsunami waves. The conventional methods for calculating wave run-up involve physical experiments or numerical modeling. Machine learning methods have recently become a part of wave run-up model development due to their robustness in dealing with large and complex data. In this paper, an extreme gradient boosting (XGBoost)-based machine learning method is introduced for predicting wave run-up on a sloping beach. More than 400 laboratory observations of wave run-up were utilized as training datasets to construct the XGBoost model. The hyperparameter tuning through the grid search approach was performed to obtain an optimized XGBoost model. The performance of the XGBoost method is compared to that of three different machine learning approaches: multiple linear regression (MLR), support vector regression (SVR), and random forest (RF). The validation evaluation results demonstrate that the proposed algorithm outperforms other machine learning approaches in predicting the wave run-up with a correlation coefficient (R² ) of 0.98675, a mean absolute percentage error (MAPE) of 6.635%, and a root mean squared error (RMSE) of 0.03902. Compared to empirical formulas, which are often limited to a fixed range of slopes, the XGBoost model is applicable over a broader range of beach slopes and incident wave amplitudes.•The optimized XGBoost method is a feasible alternative to existing empirical formulas and classical numerical models for predicting wave run-up.•Hyperparameter tuning is performed using the grid search method, resulting in a highly accurate machine-learning model.•Our findings indicate that the XGBoost method is more applicable than empirical formulas and more efficient than numerical models.

Momentum Conservative Schemes for Shallow Water Flows - Volume 4 Issue 2 - S. R. Pudjaprasetya, I. Magdalena

For safety and survival at sea and on the shore, wave predictions are essential for marine-related activities, such as harbor operations, naval navigation, and other coastal and offshore activities. In general, wave height predictions rely heavily on numerical simulations. The computational cost of such a simulation can be very high (and it can be time-consuming), especially when considering a complex coastal area, since these simulations require high-resolution grids. This study utilized a deep learning technique called bidirectional long short-term memory (BiLSTM) for wave forecasting to save computing time and to produce accurate predictions. The deep learning method was trained using wave data obtained by a continuous numerical wave simulation using the SWAN wave model over a 20-year period with ECMWF ERA-5 wind data. We utilized highly spatially correlated wind as input for the deep learning method to select the best feature for wave forecasting. We chose an area with a complex geometry as the study case, an area in Indonesia’s Java Sea. We also compared the results of wave prediction using BiLSTM with those of other methods, i.e., LSTM, support vector regression (SVR), and a generalized regression neural network (GRNN). The forecasting results using the BiLSTM were the best, with a correlation coefficient of 0.96 and an RMSE value of 0.06.