Wave modeling with unstructured mesh for hindcast, forecast and wave hazard applications in the Mediterranean Sea

• Published in: Applied ocean research Vol. 122; p. 103118
• Main Authors: Lira-Loarca, Andrea, Cáceres-Euse, Alejandro, De-Leo, Francesco, Besio, Giovanni
• Format: Journal Article
• Language: English
• Published: Barking Elsevier Ltd 01.05.2022; Elsevier BV
• Subjects: Buoys; Directional spectra; Flooding; Forecast; Hindcast; Mediterranean Sea; Methods; Risk assessment; Shallow water; Storms; Unstructured mesh; Vulnerability; Wave model; Wave parameters; Mediterranean Sea; Wave model; Unstructured mesh; Hindcast; Mediterranean Sea; Risk assessment; Forecast
• ISSN: 0141-1187; 1879-1549
• DOI: 10.1016/j.apor.2022.103118

A wave model based on an unstructured grid in the Mediterranean Sea is presented, which provides high-resolution in near-shore regions using the wave model WAVEWATCH III v6.07. The hindcast comprises hourly time series of integrated wave parameters and 2D directional spectra in selected locations, from January 1st 1979 until December 31st 2020, whereas a daily forecast simulation provides predictions for the following 5 days. The hindcast is validated against buoys and several satellite missions. Results show that the model provides a good performance for storm events and mean conditions in the Mediterranean Sea with normalized mean absolute error (NMAE) lower than 15% in 70% of the basin, spatial index of agreement (d1) above 0.6, and the model under/overestimations are below 35% compared with in-situ data. Next, a methodology for coastal risk assessment is developed, leveraging the detailed information provided by the model in shallow waters and relying on a Storm Power Index (SPI), Coastal Vulnerability Index (CVI) and Risk Index (RI). The methodology is applied to the coast of Liguria, Italy, as a region with a high social, economical and touristic value which has experienced catastrophic coastal flooding episodes in recent years. Results of risk assessment were presented for the storm event of October 2018, known to have caused severe damages in the Ligurian coastline. The results provide a SPI of 3 and 5 at the beginning and peak of the storm, respectively, leading to RI of 3–5 depending on the characteristics and socioeconomic importance of the coastal stretches. Therefore, it is concluded that the methodology estimates the risk in an efficient and adequate way for its implementation in an operational risk forecasting system. •An unstructured grid wave model for the Mediterranean Sea is presented providing an efficient model with high-resolution in near-coast regions.•A hindcast of hourly data from January 1st 1979 until December 31st 2020 is calibrated and validated against buoy and satellite data.•A hazard and risk assessment is presented and applied to the coast of Liguria, Italy.