Joint probability analysis for estimation of extremes

Conditions required to cause flooding often involve more than one source variable such as large waves combined with a high sea level causing coastal flooding, or high river flow combined with a high sea level causing river flooding. In order to estimate the probability of flooding, one needs to know not only the high and extreme values of each variable, but also their likelihood of occurring simultaneously. This work summarises the terminology and the types of method available for joint probability analysis, and discusses the issues associated with data selection and event definition. It then describes the development and testing of methods for incorporation of temporal dependence into an approach involving Monte Carlo simulation. An illustration study shows the effect of introducing firstly short-term clustering, then seasonality and long-term trend. Three further case studies are used to illustrate some key points in the appropriate use and interpretation of joint probability methods.

[1]  B. Gouldby,et al.  The joint probability of waves and water levels in coastal defence design. , 2002 .

[2]  Edward F. Thompson,et al.  Recommended practice for extreme wave analysis , 1994 .

[3]  P. Prinos,et al.  Statistical models for bivariate extremal analysis of a spatial process , 2008 .

[4]  P. J. Hawkes,et al.  Joint Probability: Dependence Mapping and Best Practice , 2006 .

[5]  T. A. Buishand,et al.  Bivariate extreme-value data and the station-year method , 1984 .

[6]  N. Doorn,et al.  Extreme values for coastal, estuarine, and riverine environments , 2008 .

[7]  P. Young,et al.  Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.

[8]  Cecilia Svensson,et al.  Dependence between extreme sea surge, river flow and precipitation in eastern Britain , 2002 .

[9]  Jonathan A. Tawn,et al.  The multivariate Gaussian tail model: an application to oceanographic data , 2000 .

[10]  Barry L. Nelson,et al.  Modeling and generating multivariate time-series input processes using a vector autoregressive technique , 2003, TOMC.

[11]  S. Coles,et al.  Modelling Extreme Multivariate Events , 1991 .

[12]  B. Gouldby,et al.  A simulation method for flood risk variables , 2007 .

[13]  B. Gouldby,et al.  Statistical simulation of flood variables: incorporating short‐term sequencing , 2008 .

[14]  J. Vrijling,et al.  JOINT PROBABILITY DISTRIBUTIONS FOR WAVE HEIGHT, WIND SETUP AND WIND SPEED , 2005 .

[15]  Eric P. Smith,et al.  An Introduction to Statistical Modeling of Extreme Values , 2002, Technometrics.

[16]  Barry L. Nelson,et al.  Modeling and generating multivariate time series with arbitrary marginals and autocorrelation structures , 2001, Proceeding of the 2001 Winter Simulation Conference (Cat. No.01CH37304).

[17]  Jonathan A. Tawn,et al.  Statistical Methods for Multivariate Extremes: An Application to Structural Design , 1994 .

[18]  Nicholas J. Garrity,et al.  Evaluation of event and response approaches to estimate the 100-year coastal flood for pacific coast sheltered waters , 2007 .

[19]  Cecilia Svensson,et al.  Dependence between sea surge, river flow and precipitation in south and west Britain , 2004 .

[20]  Janet E. Heffernan,et al.  Dependence Measures for Extreme Value Analyses , 1999 .

[21]  Ben Gouldby,et al.  The joint probability of waves and water levels in coastal engineering design , 2002 .

[22]  P. J. Hawkes,et al.  Use of Joint Probability Methods in Flood Management: A guide to best practice , 2006 .