Multiple Approaches to Understanding the Intersection of Climate Change, Air Quality & Public Health Open Access

Stowell, Jennifer (Fall 2019)

Permanent URL:


An overwhelming majority of climate scientists have declared the validity of climate change and its potential threat to the environment and to living organisms on the planet. Though it is still taken lightly by some, temperatures across the globe are rising—especially near the poles where sheets of ice help to balance the conditions we experience in the troposphere. In fact, temperature anomalies have been tracked for many years by different entities, including both governmental and scientific organizations. Overwhelmingly, the temperature anomalies they have tracked deviate little from one another and follow nearly the exact same trend of warming in our atmosphere. Many skeptics consider the major cause of changes in climate to be natural increases of energy from the sun. While it is true that the sun has a natural, oscillating pattern of high and low energy, it has been operating on the same 11-year cycle for hundreds of years and with only slight increases in trend. However, global temperatures began to deviate from normal patterns in the 1950s and have continued to rise ever since. Hence, there must be something else causing temperatures to accelerate to such high levels. Aside from human development and some natural depletion of vegetation and forests, little else has changed in the total environment, save human activity. So, while there could be something at play of which we are unaware, it seems very likely that human activity is contributing to the problem in a significant way. This is evident when looking at levels of carbon dioxide (CO2). Since the industrial revolution, CO2 has continued to surpass natural levels that were seen prior to large-scale fossil fuel use. These CO2 concentrations, similar to temperature, are also trending higher with little to no decrease.

An important issue that we must consider when approaching the climate change issue, is the effect that it can have on living organisms. One of the most readily apparent ways in which climate change affects living things is through large decreases in air quality. Climate change can have both direct and indirect paths of affecting the air that we breathe. For instance, a direct path involves the direct formation of ozone (O3) in the troposphere where it can harm human health. With respect to indirect pathways, the effects of climate change on wildfire activity have been evident in the past few decades. This is considered an indirect effect of climate on air quality because rising temperatures can increase the incidence of wildfires—which in turn can dump toxic chemicals and particles into the atmosphere through the dissipation of smoke plumes.

In this dissertation, the effect of climate change on air quality is approached in three different ways and on three different spatial scales. The first objective looks at the future changes in harmful tropospheric O3 on a national scale. This approach separates differences in concentration according to source: climate change or emissions policies. Using modeling methods to separate and predict future O3 by source adds to our understanding of the potential dangers to human health that we could experience in future years. With future O3 concentrations predicted by source, we can then project the impact on multiple morbidities and premature mortality. Results from this analysis showed that while the effect on morbidity varies between locations, it was evident that climate change could impact future mortality via O3 exposure. However, the real culprit of excess mortality due to O3 was emissions policy. Looking at two different predicted emissions scenarios, there was a significant difference in effects on mortality for scenarios in which allowed emissions are not restricted by emissions policies.

Another objective estimates the association of present-day wildfire activity on cardiorespiratory events on a statewide scale. Using health records form the state of Colorado during the fire seasons (May-August) of 2011-2014, we can estimate the association between smoke PM2.5 exposure and adverse health outcomes. Using a two-stage modeling approach, we calculated the contribution of smoke PM2.5 to total PM2.5. Separating the contributions allowed us to examine the effects due solely to smoke PM2.5. It was evident from our results that smoke PM2.5 was associated with many respiratory morbidities, but not associated with cardiovascular disease. We also conducted stratified analyses on both age and sex. While no significant difference was observed for sex, several differences were apparent for age. One of the most striking results was the odds ratio or expected increase in risk of asthma exacerbation due to smoke PM2.5 exposure. The results suggest that risk increases by over 8% (95% CI: 1.06, 1.11) for every 1 µg/m3 increase in smoke PM2.5 exposure. This result is higher than risk reported in previous publications. One conclusion that we might draw is that smoke PM2.5 may be more toxic than background, ambient PM2.5.

A third objective builds upon the results from the Colorado wildfire study and attempts to estimate future wildfire health impacts on a regional scale in the western US. Using complicated chemical transport models with and without included fire sources, we were again able to separate out smoke PM2.5, but in this instance, we are investigating potential future changes and health burden due to additional smoke PM2.5 exposure in the 2050s. This involved taking the difference between future smoke PM2.5 and present smoke PM2.5 in order to estimate potential smoke PM2.5 increases we could expect in addition to our present exposure. Through this process and adopting the risk measurement and the incidence of emergency department visits from respiratory outcomes in the Colorado study, we were able to project the future health burden from smoke PM2.5.We observed a few hotspots that seem to be highly affected by future smoke PM2.5 concentrations. These areas included northern Idaho, Nevada, and the coast of Oregon. However, it was also important to keep in mind the population distribution in comparison with the increased effects on human health. When looking at the results compared to changes in population, Montana stood as another area for concern. This was due to its relatively high increase in wildfire PM2.5 events and an overall decline in population that is expected by the 2050s.

Taken together, these three aims help us understand more about the relationship between air quality and climate change. And, in turn, this allows for us to draw out potential risk to human health that could be seen in the future. Looking at 3 different approaches with each on a different spatial scale allowed us to explore some of the assumptions that we might draw from future exposures to O3 and wildfire smoke PM2.5. Moving forward, it will be important to expand on these future impacts and find ways to attach monetary and other important values to these expected changes. This type of analysis could be beneficial in that it can be a tool for both informing policy and emergency response plans as we look to the changes that may be expected in the future.

Table of Contents

Front Matter: List of Figures, Equations & Tables. vi

1. Introduction. 1

1.1 Impact of climate change on air quality and human health through multiple pathways. 1

1.2 Direct effects of climate change: tropospheric ozone. 2

1.2.1 Toxicology of tropospheric O3 3

1.3 Direct effects of climate change on wildfire. 3

1.3.1 Indirect effects of climate change: PM2.5 from smoke. 4

1.3.2 Toxicology of wildfire PM2.5 5

1.4 Intersection between health, climate, air quality and wildfire events. 6

2. Significance. 7

2.1 Study Rationale. 7

2.2 Description of aims. 8

2.3 Synopsis of purpose and intent 8

1.5 References. 10

3. Separating the Effects of Climate Change and Emissions. 13

3.1 Abstract 13

3.2 Background. 14

3.3 Data & Methods. 17

3.3.1 Dynamical downscaling for O3 due to changes in climate and air pollution emissions. 18

3.3.2 Statistical downscaling for O3 changes due to climate change. 20

3.3.3 Future O3 changes due to changes of air pollution emissions. 22

3.3.4 Population health impact of future O3 changes. 22

3.4      Results. 24

3.4.1 Future O3 changes due to climate change. 24

3.4.2 Future O3 levels due to climate change and changes in emissions. 27

3.4.3 Future O3 change due to changes of air pollution emissions. 27

3.4.4 Population health impact of future O3 30

3.5      Discussion. 32

3.5.1 Summary and impact of results. 32

3.5.2 Comparison with current literature. 33

3.5.3 Strengths and limitations 35

3.6 Study conclusions and future directions. 35

3.7 References. 37

3.8 Supplemental 43

4. Health Effects of Smoke Exposure. 48

4.1 Abstract 48

4.2 Introduction. 50

4.2.1 Increasing threat of wildfires. 50

4.2.2 Particulate matter from smoke differs from ambient concentrations. 50

4.2.3 Epidemiological approaches to studying the effects of wildfire smoke exposure. 51

4.2.4 Importance of isolating smoke-related particulate matter 52

4.3 Data & Methods. 52

4.3.1 Health data. 52

4.3.2 PM2.5 and meteorological data. 53

4.3.3 Epidemiological modeling methods. 54

4.4 Results. 56

4.4.1 Exposure modeling and smoke contribution to PM2.5 levels. 56

4.4.2 Epidemiological modeling. 62

4.4.3 Stratified analysis. 67

4.5 Discussion. 68

4.5.1 Summary and impact of results. 68

4.5.2 Comparison with current literature. 69

4.5.3 Strengths and limitations 72

4.6 Study conclusions and future direction. 73

4.9 References. 75

4.10 Supplemental 81

5. Future health impacts due to smoke-related PM2.5 exposure. 111

5.1 Abstract 111

5.2 Introduction. 112

5.3 Data & Methods. 112

5.3.1 Study Domain. 112

5.3.2 Exposure Modeling Framework. 112

5.3.3 Isolation of Smoke PM2.5 and Calculation of Increase for 2050-2059. 112

5.3.4 Epidemiological Metrics & Health Impact Assessment 112

5.4 Results. 113

5.4.1 Isolation of Smoke PM2.5 and Calculation of Increase for 2050-2059. 113

5.4.2 Epidemiological Metrics & Health Impact Assessment 113

5.4.3 Statewide Potential Increased Burden. 113

5.4.3 Potential Increases in Health Care Costs. 113

5.5 Discussion. 113

5.5.1 Summary and impact of results. 113

5.5.2 Comparison with current literature. 113

5.5.3 Strengths & limitations. 113

5.6 Study Conclusions and future directions. 113

5.7 References. 114

6. Conclusions. 115

6.1 Contribution of Aim 1. 115

6.2 Contribution of Aim 2. 116

6.3 Contribution of Aim 3. 118

6.4 Summary & future directions. 118


About this Dissertation

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.
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files