Influence of Baseline Air Masses and Wildland Fires on Air Quality in the Western United States

Abstract

Air quality in the western United States (U.S.) is influenced by local and long-distance anthropogenic sources, as well as natural sources such as wildfires and stratospheric intrusions. The U.S. Environmental Protection Agency regulates several major pollutants in the U.S. through the National Ambient Air Quality Standards (NAAQS). These regulated pollutants include particulate matter (PM) and ozone (O₃). PM and O₃ can cause or exacerbate a number of lung and heart conditions, and the primary NAAQS are set at levels determined to be safe for human health. However, many of the sources of PM and O₃ in the western U.S.--including wildfires, stratospheric intrusions, and long-range transport of Asian urban and biomass burning pollution plumes--cannot be controlled through U.S. regulations. As the O₃ and PM NAAQS are reviewed and potentially tightened in the coming years, it will be important to develop methods to determine the contribution of each pollution source to observed concentrations of PM and O₃. This dissertation focuses on several key uncertainties related to PM and O₃ concentrations in the western U.S. Each analysis conducted for this dissertation centers on data collected at the Mount Bachelor Observatory (MBO, 2.8 km a.s.l., 43.98° N, 121.69° W), a mountaintop research site in central Oregon, U.S. The first component of this dissertation is an analysis of the contribution of baseline O₃ to observed O₃ concentrations in two western U.S. urban areas, Enumclaw, Washington (WA) and Boise, Idaho, during 2004 - 2010. Baseline O₃ is the concentration of O₃ in an air mass resulting from natural sources and long-range transport of anthropogenic sources. Therefore, baseline O₃ in the western U.S. does not include impacts from regional pollution sources; however, it does include influences from Asian pollution and stratospheric intrusions. Sites measuring baseline concentrations in the western U.S. are located away from urban areas. For this analysis, I compared O₃ data from two baseline sites (MBO and Cheeka Peak, WA) to O₃ concentrations in the two urban areas on days when backward air mass trajectories showed transport between the baseline and urban sites. I found that the urban areas studied had relatively low O₃ on the days with a strong influence from baseline air masses (28.3 - 48.3 ppbv). These data suggested that there was low production of O₃ from urban emissions on these days, which allowed me to quantify the impact of baseline O₃ on urban O₃ concentrations. A regression of the Boise and MBO O₃ observations showed that free tropospheric air masses were diluted by 50% as they were entrained into the boundary layer at Boise. These air masses can contain high O₃ concentrations (>70 ppbv) from Asian pollution sources or stratospheric intrusions, indicating that these sources can greatly contribute to urban surface O₃ concentrations. In addition, I found that the elevation and surface temperature of the urban areas studied impacted baseline O₃ concentrations in these areas, with higher elevation and greater surface temperatures leading to greater O₃ concentrations. The second and third components of this dissertation are analyses of the impact of wildland fires on PM and O₃ concentrations in the western U.S. For both of these analyses, I calculated pollutant enhancement ratios for PM, O₃, and other species in wildland fire plumes observed at MBO during 2004 - 2013. Enhancement ratios are the enhancement of a species in a plume normalized by the enhancement of a long-lived species such as carbon monoxide (CO) or carbon dioxide (CO₂). I identified the source and transport time of each of the 55 plumes observed at MBO using backward air mass trajectories and satellite data. During the first two days of transport, PM enhancement ratios (relative to CO) were greater than documented emission factors (on average 0.29 µg m<super>-3</super> ppbv<super>-1</super> compared to an average emission factor of 0.16 µg m<super>-3</super> ppbv<super>-1</super> for temperate regions), which I attributed to net production of Secondary Organic Aerosol (SOA). At transport times of more than two days, PM enhancement ratios were relatively low, which I attributed to loss of PM (through deposition and/or cloud processing) exceeding net SOA production in the fire plumes. My research also showed that SOA production from non-methane organic carbon (NMOC) would be necessary to explain the high aerosol scattering/CO₂ enhancement ratios (20.46 - 187.18 Mm<super>-1</super> ppbv<super>-1</super>) observed in 2012 - 2013. My analysis of three plumes transported greater than two days showed that O₃ production occurred in the plumes transported in the boundary layer but not in the plume transported in the free troposphere. This suggests that there may be a connection between plume transport altitude and O₃ production, which could be attributed to a higher proportion of peroxyacetyl nitrate (PAN) in free tropospheric plumes. A further analysis showed that PAN, a temperature dependent reservoir for nitrogen oxides (NO_x), was on average 48% of the observed reactive nitrogen (NO_y) in the six plumes observed in 2013. Because O₃ production in wildland fire plumes is NO_x-limited, decomposition of PAN to NO_x during transport could lead to downwind O₃ production. The fourth component of this dissertation is a comparison of daily satellite observations of CO with daily in situ observations of CO at MBO during 2004 - 2013. For this analysis, I used CO observations from the Atmospheric Infrared Sounder (AIRS) instrument on board the Aqua satellite. CO observations from MBO were compared with the satellite observations during the daily morning and afternoon satellite overpasses. I found that the daily variations in CO observed at MBO were also evident in the daytime AIRS CO observations, particularly during spring (r = 0.52 in spring). North American wildland fires were a major factor impacting the lower correlation between MBO and AIRS CO observations in the summer and autumn (r = 0.26 and 0.16, respectively), reflecting the inability of AIRS to quantify high CO in the boundary layer. The MBO and AIRS CO observations also showed the same inter-annual variability in spring (r = 0.77). During this season, CO and O₃ concentrations at MBO are often influenced by long-range transport of Asian pollution (Asian LRT) and Upper Troposphere/ Lower Stratosphere (UT/LS) air masses. Through a case study, I showed that the correlation between the MBO and AIRS CO observations can help us understand high O₃ enhancements resulting from Asian LRT.