Abstract
This study presents a comprehensive classification of wave energy resources in the Mediterranean and Black Seas based on 63 years of ERA5 reanalysis data. To evaluate the spatial and temporal variability of wave energy potential, five key wave parameters were analyzed. Correlation and feature selection methods were applied to prevent multicollinearity among the parameters and improve model efficiency, resulting in the creation of four distinct input datasets. Using the k-means clustering algorithm, the study area was divided into five wave energy classes, each representing different wave-climate characteristics. Unlike conventional wave energy studies, the inclusion of additional parameters—mean wave direction (dir), wave spectral peakedness (Qp), and wave directional width (wdw)—enabled a more precise regional definition of energy classes. Capacity factor analyses conducted for five different WEC types demonstrate how the classification can guide device selection and deployment strategies. This study highlights the spatial variability of regional energy potential, challenging the prevailing perception that the Mediterranean and similar semi-enclosed seas are homogeneous low-energy areas. According to the results the ISWEC facility off Pantelleria Island is located in Class 5, representing the highest energy levels. In addition, facilities established by Eco Wave Power off the coasts of Jaffa and Gibraltar are clustered within the high-energy Class 4. This methodology can also be applied to other semi-enclosed seas, offering an in-depth framework for classifying regions with similar energy characteristics and providing a solid foundation for marine renewable energy research and for understanding the spatial distribution of wave energy in southeastern Europe.