|
Activity Number:
|
32
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #305491 |
|
Title:
|
Environmental Risk Evaluation: A Bayesian Hierarchical Approach for Extreme Temperature Over Space and Time
|
|
Author(s):
|
Hongfei Li*+ and Jonathan Hosking and Huijing Jiang
|
|
Companies:
|
IBM T. J. Watson Research Center and IBM Research Division and Georgia Institute of Technology
|
|
Address:
|
1011 Kitchawan Rd, Yorktown Heights, NY, 10598,
|
|
Keywords:
|
Environmental risk ; Extreme value theory ; Point process ; Spatio-temporal ; Extreme temperature
|
|
Abstract:
|
Extreme high temperature is an aspect of environmental risk that involves various threats of adverse effects on living organisms and environment. We propose a Bayesian hierarchical approach to model spatially and temporally varying environmental extreme values and thus to evaluate environmental risk. Data are assumed to follow a generalized extreme value distribution with three parameters that vary over space and time. The spatial component is modeled by a spatial kernel function and the temporal component is captured through a dynamic linear model. We use a Markov chain Monte Carlo algorithm for Bayesian inference. The approach is applied to 30 years of temperature data for the eastern United States. We study whether there is any trend in the magnitude and frequency of extreme temperatures, and thus whether environmental risk in terms of extreme high temperature is increasing over time.
|
