Modélisation des surcotes sur les côtes françaises, Manche et Atlantique
Résumé
The risks of coastal flooding are directly related to water level variations and, in particular, to storm surge and wave set-up. The majority of existing models underestimate the maximum water level occurring during storms. The current study focuses on improving the storm surge modeling of for the Atlantic and English Channel French coast, in the framework of the Previmer-Surcotes project. Three principal subjects are explored to improve the predictive quality of the model: the sensitivity of the model to the parameterization of the drag coefficient, the influence of the interactions between the tide and water level, and the quantification of the relative contributions of wave set-up and storm surge to the total water level. This study focuses on four sites (Dunkerque, Saint-Malo, Le Conquet, and La Rochelle), by studying the storm surge on time scales of storm events (storms in November 2007, February 2008, March 2009, and February 2010) and of multi-annual variability (2003-2010). First of all, the study of the model sensitivity to different parameterizations of the drag coefficient, in comparison to the constant drag coefficient, shows that the formulation of Moon et al. [2007] improved the modeled storm surges (differences of up to 10 cm in certain cases). The interactions between the tide and the storm surge have also a significant contribution in the storm surge prediction, which is particularly important for the sites of Dunkerque, Saint-Malo, and La Rochelle. Finally, in-situ observations and preliminary modeling studies of wave set-up show that, for storm events like Johanna (February 2008), the wave set-up represents half of the total elevated water level at a site like Truc Vert beach.