A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Probabilistic modelling of extreme storms along the Dutch coast
Abstract Due to the unprecedented growth in population and economic development along the coastal zone all over the world, knowledge about future extreme oceanographic events will assist in ensuring human and property safety. This will be a task with increasing significance in the light of projected climate change impacts. A joint estimation of extreme storm events' variates of deep water wave conditions was performed. It can be used for multivariate descriptions of wave climate variates, such as wave height, period, steepness, and storm duration. The storm sequences can be simulated and extrapolated from limited observational data for optimal structure protection strategies and various disaster risk analysis, like erosion or overtopping. The analysis not only shows the effectiveness of the proposed statistical approaches for improving multivariate modelling of the storm parameters but also highlights the most compatible approach for the Dutch wave climate data from 1979 to 2009. We used the Monte-Carlo method and four methods to construct the dependency structures, based on copula functions, physical relationship and extreme value theory. The marginal probabilistic distribution functions of wave climate variables and the joint probability were then obtained. The simulated data group performs a reasonable similarity to the field measurements according to the goodness-of-fit test, and the Gaussian copula model was found to be the best wave climate simulation method for the Dutch coast.
Highlights Sea storms can threaten the coastal zone by means of erosion or flooding. The multivariate statistical technique is used to simulate storm events. Four methods are proposed and compared to construct the dependency structures. The Gaussian copula method is found to be the best option for the Dutch dataset.
Probabilistic modelling of extreme storms along the Dutch coast
Abstract Due to the unprecedented growth in population and economic development along the coastal zone all over the world, knowledge about future extreme oceanographic events will assist in ensuring human and property safety. This will be a task with increasing significance in the light of projected climate change impacts. A joint estimation of extreme storm events' variates of deep water wave conditions was performed. It can be used for multivariate descriptions of wave climate variates, such as wave height, period, steepness, and storm duration. The storm sequences can be simulated and extrapolated from limited observational data for optimal structure protection strategies and various disaster risk analysis, like erosion or overtopping. The analysis not only shows the effectiveness of the proposed statistical approaches for improving multivariate modelling of the storm parameters but also highlights the most compatible approach for the Dutch wave climate data from 1979 to 2009. We used the Monte-Carlo method and four methods to construct the dependency structures, based on copula functions, physical relationship and extreme value theory. The marginal probabilistic distribution functions of wave climate variables and the joint probability were then obtained. The simulated data group performs a reasonable similarity to the field measurements according to the goodness-of-fit test, and the Gaussian copula model was found to be the best wave climate simulation method for the Dutch coast.
Highlights Sea storms can threaten the coastal zone by means of erosion or flooding. The multivariate statistical technique is used to simulate storm events. Four methods are proposed and compared to construct the dependency structures. The Gaussian copula method is found to be the best option for the Dutch dataset.
Probabilistic modelling of extreme storms along the Dutch coast
Li, F. (author) / van Gelder, P.H.A.J.M. (author) / Ranasinghe, R. (author) / Callaghan, D.P. (author) / Jongejan, R.B. (author)
Coastal Engineering ; 86 ; 1-13
2013-12-17
13 pages
Article (Journal)
Electronic Resource
English
Probabilistic modelling of extreme storms along the Dutch coast
| Elsevier | 2014
Probabilistic modelling of extreme storms along the Dutch coast
| Online Contents | 2014
Probabilistic modelling of extreme storms along the Dutch coast
| British Library Online Contents | 2014
Statistics of extreme stormsurge levels at tide-gauges along the Dutch coast
| British Library Conference Proceedings | 1999