HySwash: a hybrid model for nearshore wave processes

Climate change-induced sea level rise and increasing storm severity are significant stressors that threaten the livability of coastal areas worldwide. This research presents a site-specific metamodel based on SWASH (Simulating WAves till SHore) numerical model simulations that aim at simplifying in a fast and efficient manner the prediction of hydrodynamic variables along cross-shore profiles. To accomplish this, a large synthetic database of offshore wave and sea level conditions is created and downscaled using numerical modeling together with sampling, selection, and interpolation techniques. All these mathematical methods permit to replace the computationally intensive cost of classical dynamical downscaling. In addition, the metamodel uses dimensionality reduction techniques that allow to account for a comprehensive analysis of the primary patterns governing the coastal hydrodynamic behavior. The proposed tool has been numerically validated in three different idealized coral reef profiles, showing good skill at reproducing the spatial evolution of wave setup, wave heights associated with different frequency bands, and wave runup. The flexibility and robustness of the method make it very convenient for being used in coastal risk assessments, early warning systems, or climate change projections.This research would not have been possible without funding from the Spanish Ministry of Science and Innovation, project Beach4cast PID2019-107053RB-I00. AR is funded by a Concepción Arenal predoctoral scholarship from the Universidad de Cantabria. LC acknowledges the funding from the Juan de la Cierva Formación FJC2021-046933-I/MCIN/AEI/10.13039/501100011033 and the European Union “NextGenerationEU”/PRTR