Single Site Extreme Wave Analysis in the Pacific Ocean Comparing Stationary and Non-stationary GEV Models
F. Calderón-Vega *
Department of Civil Engineering, Universidad de Guanajuato, Juárez 77, Zona Centro, P.C. 36000 Guanajuato, Gto, Mexico and Laboratori d’Enginyeria Marítima, Universitat Politècnica de Catalunya, Jordi Girona 1-3, Mòdul D1, Campus Nord, 08034, Barcelona, Spain.
C. Mösso
Laboratori d’Enginyeria Marítima, Universitat Politècnica de Catalunya, Jordi Girona 1-3, Mòdul D1, Campus Nord, 08034, Barcelona, Spain and Centre Internacional d'Investigació dels Recursos Costaners, Jordi Girona 1-3, Mòdul D1, Campus Nord, 08034, Barcelona, Spain.
A. D. García-Soto
Department of Civil Engineering, Universidad de Guanajuato, Juárez 77, Zona Centro, P.C. 36000 Guanajuato, Gto, Mexico.
E. Delgadillo-Ruiz
Department of Civil Engineering, Universidad de Guanajuato, Juárez 77, Zona Centro, P.C. 36000 Guanajuato, Gto, Mexico.
*Author to whom correspondence should be addressed.
Abstract
The adequate knowledge of the weather behavior is very important for the design and management of socioeconomical, environmental and sustainability human interests in the coasts and oceans. In the present study an extreme value analysis of maximum significant waves recorded at a buoy located in the Pacific Ocean was carried out. The analysis was carried out from two perspectives, by considering a Generalized Extreme Value (GEV) model with stationary distribution (i.e., the time variations are not accounted for), and by considering a non-stationary GEV model, which incorporates the monthly seasonality of maximum observed values in time increments; the maximum significant wave behavior was parameterized using harmonic functions for the distribution measures. Both approaches were compared for a single buoy. In the study a seasonality effect was found, which was also present at the Gulf of Mexico in previous studies, and which cannot be captured by a stationary model.
Keywords: GEV model, stationary, non-stationary, seasonality, projections.