This study reveals characteristics of the Mediterranean Seas wave climate by illustrating the results of a spatial assessment of extreme significant wave heights. The assessment was based on a 37-year wind and wave hindcast database covering the entire area at 10-km resolution. A point-wise GEV (Generalized Extreme Values) model was employed to generate the assessments results. Overall, the spatial model proved capable of providing an accurate description of extreme return levels of significant wave heights and their spatial variability, especially on a basin-wide scale and with greatest precision on the mesoscale. However, return level estimates are found less reliable in certain coastal areas because the traditional point-wise approach is not refined enough to address the entire wave spectrum of an area as complex as the Mediterranean Sea. Therefore, MSLP (Mean Sea Level Pressure) fields were incorporated as covariates to improve localized assessment. These covariates represent meteorological forcing and allow analysis of the role of different cyclonic regimes in defining wave features and their spatial variability. Finally, the broad temporal span and high spatial resolution of the hindcast database allow for EOF (Empirical Orthogonal Function) and CCA (Canonical Correlation Analysis) analysis of wind and waves fields. This analysis validates spatial wave distribution assessment, revealing that the main processes governing the Mediterranean Seas wave climate can be attributed to four main modes.
Spatio-temporal modelling of extreme wave heights in the Mediterranean Sea
Sartini, L.;Besio, G.;Cassola, F.
2017-01-01
Abstract
This study reveals characteristics of the Mediterranean Seas wave climate by illustrating the results of a spatial assessment of extreme significant wave heights. The assessment was based on a 37-year wind and wave hindcast database covering the entire area at 10-km resolution. A point-wise GEV (Generalized Extreme Values) model was employed to generate the assessments results. Overall, the spatial model proved capable of providing an accurate description of extreme return levels of significant wave heights and their spatial variability, especially on a basin-wide scale and with greatest precision on the mesoscale. However, return level estimates are found less reliable in certain coastal areas because the traditional point-wise approach is not refined enough to address the entire wave spectrum of an area as complex as the Mediterranean Sea. Therefore, MSLP (Mean Sea Level Pressure) fields were incorporated as covariates to improve localized assessment. These covariates represent meteorological forcing and allow analysis of the role of different cyclonic regimes in defining wave features and their spatial variability. Finally, the broad temporal span and high spatial resolution of the hindcast database allow for EOF (Empirical Orthogonal Function) and CCA (Canonical Correlation Analysis) analysis of wind and waves fields. This analysis validates spatial wave distribution assessment, revealing that the main processes governing the Mediterranean Seas wave climate can be attributed to four main modes.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.