Premium
A region‐based hierarchical model for extreme rainfall over the UK, incorporating spatial dependence and temporal trend
Author(s) -
Atyeo Jonathan,
Walshaw David
Publication year - 2012
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/env.2155
Subject(s) - covariate , spatial dependence , bayesian probability , bayesian hierarchical modeling , extreme value theory , hierarchical database model , tail dependence , environmental science , bayesian inference , econometrics , climatology , geography , statistics , computer science , mathematics , multivariate statistics , geology , data mining
In this paper, we develop a Bayesian hierarchical model for extreme rainfall over the entire U.K. The model is based around a strong regional structure, so that for each of nine distinct physiographic regions, we impose a distinct model for the marginal behaviour that employs site‐specific covariate information, and a model for the dependence structure. Using data collected at 204 locations over a 40‐year period, the model is designed to meet the challenges posed by the problem, specifically: the distinct extremal behaviour in the different regions; the need to model extremal dependence for up to 25 sites within a region; and the possibility of temporal trends within the extreme rainfall climate. Copyright © 2012 John Wiley & Sons, Ltd.