Modeling of Future Extreme Storm Surges at the NW Mediterranean Coast (Spain)

Storm surges are one of the main drivers for extreme flooding at the coastal areas. Such events can be characterized with the maximum level in an extreme storm surge event (surge peak), as well as the duration of the event. Surge projections come from a barotropic model for the 1950–2100 period, und...

Full description

Bibliographic Details
Published in:Water
Main Authors: Jue Lin-Ye, Manuel García-León, Vicente Gràcia, María Isabel Ortego, Piero Lionello, Dario Conte, Begoña Pérez-Gómez, Agustín Sánchez-Arcilla
Format: Article in Journal/Newspaper
Language:English
Published: MDPI AG 2020
Subjects:
Online Access:https://doi.org/10.3390/w12020472
https://doaj.org/article/b2c6f61cfb144e88acdfe9fdf883a017
Description
Summary:Storm surges are one of the main drivers for extreme flooding at the coastal areas. Such events can be characterized with the maximum level in an extreme storm surge event (surge peak), as well as the duration of the event. Surge projections come from a barotropic model for the 1950–2100 period, under a severe climate change scenario (RCP 8.5) at the northeastern Spanish coast. The relationship of extreme storm surges to three large-scale climate patterns was assessed: North Atlantic Oscillation ( <math display="inline"> <semantics> <mrow> <mi>N</mi> <mi>A</mi> <mi>O</mi> </mrow> </semantics> </math> ), East Atlantic Pattern ( <math display="inline"> <semantics> <mrow> <mi>E</mi> <mi>A</mi> <mi>W</mi> <mi>R</mi> </mrow> </semantics> </math> ), and Scandinavian Pattern ( <math display="inline"> <semantics> <mrow> <mi>S</mi> <mi>C</mi> </mrow> </semantics> </math> ). The statistical model was built using two different strategies. In Strategy #1, the joint probability density was characterized by a moving-average series of stationary Archimedean copula, whereas in Strategy #2, the joint probability density was characterized by a non-stationary probit copula. The parameters of the marginal distribution and the copula were defined with generalized additive models. The analysis showed that the mean values of surge peak and event duration were constant and were independent of the proposed climate patterns. However, the values of <math display="inline"> <semantics> <mrow> <mi>N</mi> <mi>A</mi> <mi>O</mi> </mrow> </semantics> </math> and <math display="inline"> <semantics> <mrow> <mi>S</mi> <mi>C</mi> </mrow> </semantics> </math> influenced the threshold and the storminess of ...