Estimation des niveaux marins extrêmes avec et sans l’action des vagues le long du littoral métropolitain

Abstract : Accurate knowledge of the statistical distribution of extreme sea levels is of the utmost importance for the characterization of flood risks in coastal areas, with a particular interest devoted to extreme water levels because they may induce the most dramatic consequences. Research was funded by the French Ministry of Ecology, Sustainable Development and Energy to identify the risk of flooding from the sea in France. The aim is to provide values on design levels along the French coasts by a statistical method of extreme value analysis. These levels must include the effect of the three following components: tide, meteorological surge and wave set-up. The principle is as follows: an analysis is carried out at the harbours, where seal level observations are available, then the result is interpolated between the harbours. Different approaches are tested. In particularly, the following specific items are studied:- the tide surge dependence, with two different types of dependence: a temporal dependence and an amplitude dependence;- the interpolation method: with the comparison of a site-by-site analysis (SSA) with a Regional Frequency Analysis (RFA), and a 1-D with a 2-D interpolation;- the estimation of the wave set-up, based on the state of art of parametric formula;- the surge wave dependence, with the bivariate laws of extreme values. The final result is two profiles of the 100-year water level: one for the still water level (tide and meteorological surge), and the other for the sea level with the wave set-up. The highest sea levels are located, for the English Channel and Atlantic coasts at the Saint-Michel-bay (because of the tide), and for the Mediterranean coast around Marseille. The analysis shows that the temporal tide surge dependence has no effect on the estimation of the sea level extreme values. In contrast, the model of the amplitude tide surge dependence shows some interesting results for few harbours. In comparison with the SSA, the RFA tends to smooth the result. RFA estimates are higher along the Mediterranean coast, and similarly along the English Channel and Atlantic coasts. RFA would be recommended for estimating return levels out of the SSA validity domain. Because of the small number of observation sites, a 1-D interpolation, along a smoothed coastline, is preferred. The wave set-up is calculated with the formula of Dean and Walton [2009].The surge wave dependence is medium along the Mediterranean coast. Variations of the dependence factor are more important along the English Channel and the Atlantic coasts, with a maximum at the bay of the Seine and some minima at the bay of Saint-Michel and Calais. Some ideas are provided to perform the methodology for further work
Complete list of metadatas

Cited literature [72 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/tel-01277151
Contributor : Abes Star <>
Submitted on : Monday, February 22, 2016 - 11:02:09 AM
Last modification on : Tuesday, March 27, 2018 - 6:02:05 PM
Long-term archiving on : Monday, May 23, 2016 - 12:00:27 PM

File

TH2015PEST1102_complete.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01277151, version 1

Collections

Citation

Xavier Kergadallan. Estimation des niveaux marins extrêmes avec et sans l’action des vagues le long du littoral métropolitain. Mécanique des fluides [physics.class-ph]. Université Paris-Est, 2015. Français. ⟨NNT : 2015PESC1102⟩. ⟨tel-01277151⟩

Share

Metrics

Record views

539

Files downloads

1084