Activity Number:
|
322
- ENVR Student Paper Award Winners
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics and the Environment
|
Abstract #322759
|
|
Title:
|
A Geostatistical Approach to Modeling Climate Teleconnections
|
Author(s):
|
Joshua Hewitt* and Jennifer A Hoeting and James Done and Erin Towler
|
Companies:
|
Colorado State University and Colorado State University and National Center for Atmospheric Research and National Center for Atmospheric Research
|
Keywords:
|
spatial ;
climate ;
teleconnection ;
hierarchical ;
bayesian ;
eof
|
Abstract:
|
We propose a geostatistical model to address spatial problems that violate a central tenant of spatial statistics: that distant processes are weakly correlated or independent. The model draws on ideas from spatially varying coefficient models and additionally uses predictive process ideas to overcome modeling challenges in applications. We adopt a hierarchical Bayesian framework to conduct inference and make predictions and use this model to improve predictions of average Winter precipitation in Florida by accounting for teleconnection effects with Pacific Ocean sea surface temperatures. We also discuss physical motivations and interpretations for our model.
|
Authors who are presenting talks have a * after their name.