%0 Journal Article %1 JGRC:JGRC20669 %A Weiss, Jérôme %A Bernardara, Pietro %A Benoit, Michel %D 2014 %J Journal of Geophysical Research: Oceans %K Homogenity RFA wa %P n/a--n/a %R 10.1002/2013JC009668 %T Formation of homogeneous regions for regional frequency analysis of extreme significant wave heights %U http://dx.doi.org/10.1002/2013JC009668 %X Regional frequency analysis (RFA) can reduce uncertainties in the estimations of return levels, provided that homogeneous regions can be delineated. In the framework of extreme marine events, a physically-based method to form homogeneous regions by identifying typical storms footprints is proposed. First, a spatio-temporal declustering procedure is employed to detect storms generating marine extremes. Second, the identification of the most typical storms footprints relies on a clustering algorithm based on a criterion of storm propagation. The resulting regions are readily explicable: sites from a given region are likely to be impacted by the same storms, and any storm impacting a region is likely to remain enclosed in this region. This procedure is fairly simple to implement, as the only information required is the time of occurrence of the observed extremes. An application to the estimation of extreme significant wave heights from the numerical sea-state database ANEMOC-2 is given. Six regions, both physically and statistically homogeneous, are delineated in the North-East part of the Atlantic Ocean. It is shown that the identification of storms footprints allows the increase of the overall statistical homogeneity. Combined with RFA, the proposed method highlights regional differences in the spatial extent and intensity of storms.

@article{JGRC:JGRC20669, abstract = {Regional frequency analysis (RFA) can reduce uncertainties in the estimations of return levels, provided that homogeneous regions can be delineated. In the framework of extreme marine events, a physically-based method to form homogeneous regions by identifying typical storms footprints is proposed. First, a spatio-temporal declustering procedure is employed to detect storms generating marine extremes. Second, the identification of the most typical storms footprints relies on a clustering algorithm based on a criterion of storm propagation. The resulting regions are readily explicable: sites from a given region are likely to be impacted by the same storms, and any storm impacting a region is likely to remain enclosed in this region. This procedure is fairly simple to implement, as the only information required is the time of occurrence of the observed extremes. An application to the estimation of extreme significant wave heights from the numerical sea-state database ANEMOC-2 is given. Six regions, both physically and statistically homogeneous, are delineated in the North-East part of the Atlantic Ocean. It is shown that the identification of storms footprints allows the increase of the overall statistical homogeneity. Combined with RFA, the proposed method highlights regional differences in the spatial extent and intensity of storms.}, added-at = {2014-04-28T11:17:52.000+0200}, author = {Weiss, Jérôme and Bernardara, Pietro and Benoit, Michel}, biburl = {https://www.bibsonomy.org/bibtex/2c03f0e90cbf3a535efcfafdebf39bc91/marsianus}, description = {Formation of homogeneous regions for regional frequency analysis of extreme significant wave heights - Weiss - Journal of Geophysical Research: Oceans - Wiley Online Library}, doi = {10.1002/2013JC009668}, interhash = {c223a474cc08d6c52e898f728ecb9af7}, intrahash = {c03f0e90cbf3a535efcfafdebf39bc91}, issn = {2169-9291}, journal = {Journal of Geophysical Research: Oceans}, keywords = {Homogenity RFA wa}, pages = {n/a--n/a}, timestamp = {2014-04-28T11:17:52.000+0200}, title = {Formation of homogeneous regions for regional frequency analysis of extreme significant wave heights}, url = {http://dx.doi.org/10.1002/2013JC009668}, year = 2014 }