Seminar Announcement

A Bayesian Climate Ensemble, Uncertainty, and Variability in Annual Temperature Profiles

Tamara Greasby, National Center for Atmospheric Research

Monday, October 8, 2012

4:00pm, room 223 Weber Building

 

Abstract

One goal of the North American Regional Climate Change Assessment Program (NARCCAP) is
to better understand regional climate model output and the impact of the boundary conditions
provided by the global climate model, the regional climate model itself, and interaction between
the two. This talk will outline a statistical framework for quantifying this output, and the associated
uncertainty, spatially and temporally, with the aim of understanding the contributions of each model
and how those contributions vary over space. At the most basic level, a Bayesian method is used
to create an overall estimate of climate and climate change along with gaining an understanding
of the uncertainty contributed by each model. Regression is performed in an effort to understand
that uncertainty and how much is due to RCM, GCM, and any potential interaction. These results
will be compared for different climactic regions and metropolitan regions. In the next step, annual
profiles of average daily temperature are computed and hierarchical methods are employed to
simultaneously represent how these temperature profiles change over time, how they vary spatially,
and how they vary between different climate models. The discussion of the results of this analysis
will focus on changes in seasonality and other measures of interest to impacts researchers.