Final Report: Air Quality Impacts of Extreme Weather Events: Historical Analysis and Future Projection

EPA Grant Number: R835204
Title: Air Quality Impacts of Extreme Weather Events: Historical Analysis and Future Projection
Investigators: Wang, Yuhang , Deng, Yi , Loadholt, Jay , Park, Taewon , Song, Yongjia , Zeng, Tao , Zhang, Henian , Zhang, Yuzhong , Zou, Yufei
Institution: Georgia Institute of Technology , Georgia Environmental Protection Division
EPA Project Officer: Hunt, Sherri
Project Period: June 1, 2012 through May 31, 2015 (Extended to May 31, 2016)
Project Amount: $749,859
RFA: Extreme Event Impacts on Air Quality and Water Quality with a Changing Global Climate (2011) RFA Text |  Recipients Lists
Research Category: Air Quality and Air Toxics , Water Quality , Climate Change , Air , Water


Atmospheric systems are strongly affected by extreme weather events. Previous studies have demonstrated large sensitivities of air quality to the variability in atmospheric systems. The primary objective of this project was to characterize and understand the links between air quality extremes and regional weather/climate systems such that air quality and its extreme projections with explicit uncertainty estimates can be developed.

Summary/Accomplishments (Outputs/Outcomes):

In this project, we made use of extensive observations of ozone and fine particulate matter (PM2.5) over the United States, meteorological reanalysis, and regional and global model simulation results to understand the trends of surface ozone and PM2.5 and their extremes over the United States. We decomposed the original research objectives into several tasks in order to gain a holistic understanding of the science issues. Rather than solely relying on climate model projections, we took the approach of understanding the past observations, deriving the key mechanisms, and understanding how these mechanisms affect air quality and its extremes in climate projections. While we believe that this is a valid and fruitful approach to address the key science questions in this project, we found in the course of this study that many aspects of our findings differ from the generally accepted conceptual framework of ozone and PM2.5. We took a cautious approach to verify the analysis results by either applying multiple statistical analysis methods or developing new methodologies. Since the results do not always conform to conventional views, the writing of the research results also took longer because of our desire to reconcile with the conventional view if possible or present the new findings in journals that may draw more attention from the science community and the public.

The recent publication of the paper “Climate-driven ground-level ozone extreme in the fall over the Southeast United States” in the Proceedings of National Academy of Sciences of the United States of America already has been reported by 84 news outlets. In this study, we developed a new statistical approach to solve the problem of collinearity between temperature and relative humidity. High ozone concentrations usually occur in the summer over the United States. However in extreme cases, such as in October 2010 over the Southeast United States, ozone during the fall reached the summer level. We find a large contribution by enhanced emissions of biogenic isoprene to ozone extremes from water-stressed plants under a dry and warm condition. It also explains the puzzling fact that the two extremes of high October ozone over the region all occurred in the 2000s with lower anthropogenic emissions than during the 1980–1990s. The occurrences of drying and warming in the fall, projected by climate models, will likely lead to more active photochemistry, enhanced biogenic isoprene and fire emissions, an extension of the ozone season from summer to fall, and an increase of secondary organic aerosols in the Southeast, posing challenges to regional air quality management.

It would have been interesting to find similar extreme PM2.5 cases that can clearly demonstrate the effects of climate variation and show previously unknown mechanisms as in the ozone study. However, over the United States we find generally a larger impact due to the reduction of emissions than climate variation. On the other hand, there was a well known extreme PM2.5 event during January 2013 in China. The initial report by the New York Times using the observations at the U.S. Embassy set an avalanche of news reports followed by changes of environmental policies by the government. With funding support from both this project and the National Science Foundation, we show that the unprecedented haze event is due to the extremely poor ventilation condition, which has not been seen in the past 3 decades. Analysis of observation data suggests that it was triggered by significant boreal cryosphere changes (i.e., the Arctic sea ice loss in the preceding autumn and extensive boreal snowfall in earlier winter). Climate model sensitivity experiments corroborated the substantial impacts of these boreal cryosphere changes on regional ventilation and dispersion over eastern China. Extreme haze events in winter will likely occur at a higher frequency in China as a result of the changing boreal cryosphere, requiring more stringent pollution controls to mitigate the climate change penalty. Current climate models underestimate the cryospheric impacts on wintertime air quality over eastern China due to poor winter snow simulations. As the air quality improvement due to emission reduction slows down in the United States, we expect to find PM2.5 extremes driven by similar regional climate systems not necessarily around North America.

To gain a better general understanding of extreme weather events and air quality extremes in different regions and seasons, we examined the relationship between extreme ozone and PM2.5 events and the representative meteorological parameters, such as daily maximum temperature (Tmax), minimum relative humidity (RHmin) and minimum wind speed (Vmin), using the location-specific 95th or 5th percentile threshold derived from historical data (30 years for ozone and 10 years for PM2.5). We found that ozone and PM extremes were decreasing over the years, reflecting EPA’s tightened standards and effort on reducing the corresponding precursor’s emissions. Annual ozone extreme days were highly correlated with Tmax and RHmin, especially in the Eastern United States. They were positively correlated with Vmin in urban stations and negatively correlated with Vmin in rural and suburban stations. The overlapping ratios of ozone extreme days with Tmax were fairly constant, about 32 percent, and tended to be high in fall and low in winter. Ozone extreme days were most sensitive to Tmax, then RHmin and least sensitive to Vmin. The majority of ozone extremes occurred when Tmax was between 300 K and 320 K, RHmin was less than 40 percent, and Vmin was less than 3 m/s. The number of annual extreme PM2.5 days was highly positively correlated with the extreme RHmin/Tmax days, with correlation coefficient between PM2.5/RHmin highest in urban and suburban regions and the correlation coefficient between PM2.5/Tmax highest in rural area. Tmax has more impact on PM2.5 extremes over the Eastern United States. Extreme PM2.5 days were more likely to occur at low RH conditions in the Central and Southeastern United States, especially during spring time, and at high RH conditions in the Northern United States and the Great Plains. Most extreme PM2.5 events occurred when Tmax was between 300 K and 320 K and RHmin was between 10 percent and 50 percent. Extreme PM2.5 days usually occurred when Vmin was under 2 m/s. However, during spring season in the Southeast and fall season in Northwest, high winds were found to accompany extreme PM2.5 days, likely reflecting the impact of fire emissions.

The next obvious question is how the progression of air quality extremes over the last 30 years compares to the changes of the medians. The observation record of PM2.5 is too short, so we focused on ozone observations in different regions of the United States. The analysis suggests that the statistical distributions of summertime ozone is moving towards to a more “natural” state and the effects of further emission reduction will not produce the large decreases that we have seen in the past. Considering that it is unrealistic to reach a zero-emission state, this study has direct relevance to the realism of any further tightening of the ambient ozone standard. The results imply that the room for using emission control strategies to significantly improve summertime ozone is quite limited over the Eastern United States. Much of the future improvements will not come from the reduction of the extremes as in the past but from that of median changes in the summer. In a separate study, we show that a large portion of the improvement of average ozone concentrations over the Eastern United States is due to regional climate in the summer and that this trend will likely to continue. These two studies suggest that regional climate change has had a beneficial effect on summertime ozone over the Eastern United States, and its effect will likely become even more significant in the future. This is in contrast to the Western United States, where regional climate change tends to increase ozone concentrations in the summer. The same analysis is applied in spring. We find that the increase of ozone is again affected by regional climate change. The effect of long-range transport from Asia has been suggested to be a major reason for the increase of springtime ground-level ozone. However, our study suggests that the mechanism through which ozone produced from Asian emissions potentially affects the ground level concentrations over the United States is more complex, and that springtime ozone would have increased even without the increase of emissions in Asia. Lastly, we applied statistical analyses to understand the regional characteristics of organic matter relative to black carbon. The analysis results appear to challenge the prevailing view that secondary organic aerosol (SOA) formation due to biogenic emissions over the Southeast is regional in nature. We do find that anthropogenic emissions play an important role as in previous studies, and our results suggest that their role is essential for SOA formation over different regions of the United States. The role of regional climate variation also is likely to be important and will require further studies.

Journal Articles on this Report : 3 Displayed | Download in RIS Format

Other project views: All 9 publications 3 publications in selected types All 3 journal articles
Type Citation Project Document Sources
Journal Article Zhang H, Wang Y, Park T-W, Deng Y. Quantifying the relationship between extreme air pollution events and extreme weather events. Atmospheric Research 2017;188:64-79. R835204 (Final)
  • Full-text: ScienceDirect-Full Text HTML
  • Abstract: ScienceDirect-Abstract
  • Other: ScienceDirect-Full Text PDF
  • Journal Article Zhang Y, Wang Y. Climate driven ground-level ozone extreme in the fall over the Southeast United States. Proceedings of the National Academy of Sciences of the United States of America 2016;113(36):10025-10030. R835204 (Final)
  • Full-text: PNAS-Full Text PDF
  • Abstract: PNAS-Abstract & Full Text HTML
  • Journal Article Zou Y, Wang Y, Zhang Y, Koo J-H. Arctic sea ice, Eurasia snow, and extreme winter haze in China. Science Advances 2017;3(3):e1602751 (9 pp.). R835204 (Final)
  • Full-text: Sciences Advances-Full Text HTML
  • Abstract: Sciences Advances-Abstract
  • Other: Sciences Advances-Full Text PDF
  • Supplemental Keywords:

    extreme events, air quality projection, climate change, uncertainty

    Relevant Websites:

    Yuhang Wang Group | Georgia Tech Exit

    Progress and Final Reports:

    Original Abstract
    2012 Progress Report
    2013 Progress Report
    2014 Progress Report