A Hierarchical Model for Serially Dependent Extremes: A Study of Heat Waves in the Western U.S.

• Activity Number: 396
• Type: Invited
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: JABES-Journal of Agricultural, Biological, and Environmental Statistics
• Author(s): Brian J. Reich* and Benjamin Shaby and Dan Cooley
• Companies: North Carolina State University and Penn State and Colorado State University
• Keywords: Bayesian ; Climate change ; Extreme value analysis ; Pareto ; Time series data

Abstract

Heat waves take a major toll on human populations, with negative impacts on the economy, agriculture, and human health. As a result, there is great interest in studying the changes over time in the probability and magnitude of heat waves. In this paper we propose a hierarchical Bayesian model for serially-dependent extreme temperatures. We assume the marginal temperature distribution follows the generalized Pareto distribution (GPD) above a location-specific threshold, and capture dependence between subsequent days using a transformed max-stable process. Our model allows both the parameters in the marginal GPD and the temporal dependence function to change over time. This allows Bayesian inference on the change in likelihood of a heat wave. We apply this methodology to daily high temperatures in nine cities in the western US for 1979-2010. Our analysis reveals increases in the probability of a heat wave in several US cities.