Fast coastal wave climate prediction using autoencoders-based models
• Surrogate autoencoder is proposed for spatially distributed nearshore wave prediction. • MDA and K-means organize study area into homogeneous macro-areas for enhanced analysis. • Accurate reconstruction of H s , T p , and Dir wave , with RMSE of about ∼ 10 %–15%. • The model enables near real-time predictions for efficient coastal forecasting.
Authors
- Elisa Castro (ORCID: https://orcid.org/0000-0003-2884-9295)
- Claudio Iuppa (ORCID: https://orcid.org/0000-0001-9257-2230)
- Luca Cavallaro (ORCID: https://orcid.org/0000-0002-9356-7289)
- Enrico Foti (ORCID: https://orcid.org/0000-0001-8407-1890)
- Luca Patané (ORCID: https://orcid.org/0000-0002-5488-9365)
- Rosaria Ester Musumeci (ORCID: https://orcid.org/0000-0002-1022-0571)
- Maria Gabriella Xibilia (ORCID: https://orcid.org/0000-0001-7723-2051)
Institutions
- University of Messina (IT)
- University of Catania (IT)
Publication Details
- Journal
- Ocean Engineering
- Published
- 2026-05-28
- DOI
- https://doi.org/10.1016/j.oceaneng.2026.126118
- Primary Topic
- Ocean Waves and Remote Sensing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- European Commission
- Ministero dell'Istruzione e del Merito