Developing and Testing a Simple Approach for Projecting Temperature Trends to Facilitate Public Health Preparedness Open Access

Abstract

Extreme heat events (EHEs) are expected to intensify in North America in the 21st
century, posing challenges outside current public health capacity. Adaptation will likely
be required to minimize health impacts. Most adaptation to date has been in response to
extreme weather events, e.g., the response to the European heat wave of 2003. The US
has not experienced a similarly dramatic EHE and likely has a significant preparedness
deficit. Scenario-driven table-top exercises are one way to facilitate preparedness, but
these require credible projections of climatic shifts. Downscaled projections are
computationally expensive, unavailable for most localities, and most public health
agencies cannot make their own. Given the advent of warming trends in many locales,
our objective is to determine whether it is appropriate to use historical weather data to
project future temperature trends. Such an approach is relatively straightforward,
intuitive, inexpensive, scalable, and generalizable compared with downscaled projections.
To test this proposition, we used a dataset of historical weather for selected major US
cities to identify appropriate indicators for projection, test multiple methods for
extrapolating historical trends, project findings forward by climate region and city, and
compare findings with available downscaled projections. Average daily temperature,
average daily maximum temperature, and maximum of the daily maximum temperature
in June-July-August (JJA) from 1950-2010 were selected as indicators. Stepwise
autoregressive methods were used to generate polynomial projections of the indicators.
The projections most consistent with historical data occurred with the indicators of
central tendency and showed mostly positive trends (e.g., average daily temperatures in
JJA in the Southeast are projected to increase by 5.6°F by 2035, and 10.0°F by 2055).
Negative trends for extremes (maximum of the daily maximum temperature in JJA)
resulted for three regions. The trends were determined to be a function of the
extrapolation method and greater variability in historical data for these regions. Available
downscaled regional projections were in concordance with our findings that much of the
US will experience warming in the future. We conclude that this approach has promise
but may systematically underestimate the magnitude of likely future temperature changes
in regions with variable historical temperature trends.

Table of Contents

Introduction.
Background.
Significance.
Goals and Objectives.
Methods
Study Design.
Study Setting.
Site Selection.
Data Sources and Management
Data Analysis and Outcomes
Indicators.
Projections.
Comparisons.
Results
Study Sample.
Indicators.
Variability of Historical Data
Projections.
Comparisons with Downscaled GCM Outputs.
Discussion
Indicators.
Variability of Historical Data
Projections.
Comparisons.
Study Strengths and Limitations
Conclusion
References
Tables and Figures.

About this Master's Thesis

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Rollins School of Public Health
Department	Environmental Health
Degree	MPH
Submission	Master's Thesis
Language	English
Research Field	Health Sciences, Public Health
Keyword	Public Health Preparedness Heat Waves Temperature Trends Extreme Heat Events Climate Change
Committee Chair / Thesis Advisor	Hess, Jeremy, Emory University
Partnering Agencies	CDC

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Developing and Testing a Simple Approach for Projecting Temperature Trends to Facilitate Public Health Preparedness ()	2018-08-28 12:28:19 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions