How historical information can improve estimation and prediction of extreme coastal water levels: application to the Xynthia event at La Rochelle (France)

Abstract

Abstract. The knowledge of extreme coastal water levels is useful for coastal flooding studies or the design of coastal defences. While deriving such extremes with standard analyses using tide-gauge measurements, one often needs to deal with limited effective duration of observation which can result in large statistical uncertainties. This is even truer when one faces the issue of outliers, those particularly extreme values distant from the others which increase the uncertainty on the results. In this study, we investigate how historical information, even partial, of past events reported in archives can reduce statistical uncertainties and relativise such outlying observations. A Bayesian Markov chain Monte Carlo method is developed to tackle this issue. We apply this method to the site of La Rochelle (France), where the storm Xynthia in 2010 generated a water level considered so far as an outlier. Based on 30 years of tide-gauge measurements and 8 historical events, the analysis shows that (1) integrating historical information in the analysis greatly reduces statistical uncertainties on return levels (2) Xynthia's water level no longer appears as an outlier, (3) we could have reasonably predicted the annual exceedance probability of that level beforehand (predictive probability for 2010 based on data until the end of 2009 of the same order of magnitude as the standard estimative probability using data until the end of 2010). Such results illustrate the usefulness of historical information in extreme value analyses of coastal water levels, as well as the relevance of the proposed method to integrate heterogeneous data in such analyses.

High-resolution marine flood modelling coupling overflow and overtopping processes: framing the hazard based on historical and statistical approaches

“…lack of precision in spa- tiotemporal resolution, bathymetry or forcing data) are transferred to the statistical analysis and generate uncertainties as to the results (Caires and Sterl, 2005;Mínguez et al, 2012). Bulteau et al (2015) developed a method (called HIBEVA for Historical Information in Bayesian Extreme Value Analysis) for using historical data from archives to analyse extreme water level values. The flexibility and overall Bayesian framework of HIBEVA justify its use in this study to estimate the marginal probability distribution of significant wave heights via a combination of observation data and model output.…”

Intérêt du "Data-Rescue" en marégraphie : des archives papiers à l’étude de l’évolution du niveau marin. Exemple de la reconstruction de la série marégraphique de Saint-Nazaire depuis le 19ème siècle

“…Entre 1863 et 2014, une augmentation globale du niveau moyen de la mer de l'ordre de +1.03 ± 0.05 mm/an est observée à Saint-Nazaire. Ce résultat tend à être plus faible en comparaison de ce qui est rapportée dans la littérature pour des périodes similaires : entre 1860 et 2010, la tendance estimée à Brest est de +1.32 ± 0.07 mm/an, celle de la série composite du Pertuis d'Antioche est de +1,38 ± 0.07 mm/an et celle de Liverpool estimée à + 1.44 ± 0.13 mm/an (GOURIOU, 2012 (BULTEAU et al, 2015). L'ensemble des évènements de tempêtes identifiés seront intégrés dans la base de données TEMPETES dont l'objectif est de capitaliser tout type d'information sur les événements de tempêtes et de submersions historiques (GILOY et al, 2018).…”

Evaluación probabilística del peligro por lahares en el flanco NE del Volcán Popocatépetl

“…Sin embargo el hecho de tener dos fuentes del flujo con localizaciones distintas antes de confluir en la población, representa una partición del espacio de muestreo en dos subconjuntos independientes, generando las condiciones para que su análisis probabilístico apropiado sea el de tipo bayesiano (Walpole et al, 2007). La probabilidad Bayesiana consiste en utilizar datos históricos como información previa para estimar una distribución a priori para los paráme-tros de distribución de probabilidad (Bulteau et al, 2015). Esta técnica proporciona un marco adecuado debido a su capacidad natural para el manejo de las incertidumbres, conduciendo a la representación probabilística de las estimaciones dadas por los modelos; además, permite la construcción modular en modelos complejos facilitando la inferencia en estos sistemas (Coles y Tawn, 2005;Bulteau et al, 2015).…”

“…Posteriormente se realiza un muestreo representativo de todos los casos posibles, es decir se generan escenarios hipotéticos que están dentro de las magnitudes posibles registradas en eventos históri-cos del volcán Popocatépetl, usando la técnica de Muestreo Hipercúbico Latino (LHS por sus siglas en inglés) (Núñez et al, 1999). Finalmente se llega al análisis probabilístico bayesiano (Coles y Tawn, 2005;Bulteau et al, 2015) de esta amenaza en la población de Santiago Xalitzintla. (Capra et al, 2004).…”