Next Article in Journal
Aerosol Retrieval over Land from the Directional Polarimetric Camera Aboard on GF-5
Next Article in Special Issue
Assessing Green Solutions for Indoor and Outdoor Environmental Quality: Sustainable Development Needs Renewable Energy Technology
Previous Article in Journal
Assessment of Two Drought Indices to Quantify and Characterize Drought Incidents: A Case Study of the Northern Part of Burundi
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Atmosphere
Volume 13
Issue 11
10.3390/atmos13111883
atmosphere-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Investigation of the Parameters Influencing Baseline Ozone in the Western United States: A Statistical Modeling Approach

by
Matthew Ninneman
1,*,
Irina Petropavlovskikh
2,3,
Peter Effertz
2,3,
Duli Chand
4 and
Daniel Jaffe
1,5
1
School of Science, Technology, Engineering and Mathematics, University of Washington Bothell, 18115 Campus Way NE, Bothell, WA 98011, USA
2
Cooperative Institute for Research in Environmental Sciences, 216 UCB, University of Colorado, Boulder, CO 80309, USA
3
National Oceanic and Atmospheric Administration Global Monitoring Laboratory, 325 Broadway, Boulder, CO 80305, USA
4
Atmospheric Sciences and Global Change Division, Pacific Northwest National Laboratory, P.O. Box 999, Richland, WA 99354, USA
5
Department of Atmospheric Sciences, University of Washington, 3920 Okanogan Lane, Seattle, WA 98195, USA
*
Author to whom correspondence should be addressed.
Atmosphere 2022, 13(11), 1883; https://doi.org/10.3390/atmos13111883
Submission received: 20 October 2022 / Revised: 7 November 2022 / Accepted: 10 November 2022 / Published: 11 November 2022
(This article belongs to the Special Issue Science and Technology of Indoor and Outdoor Environment)
Browse Figures
Versions Notes

Abstract

Ground-level ozone (O3) is a key atmospheric gas that controls the oxidizing capacity of the atmosphere and has significant health and environmental implications. Due to ongoing reductions in the concentrations of O3 precursors, it is important to assess the variables influencing baseline O3 to inform pollution control strategies. This study uses a statistical model to characterize daily peak 8 h O3 concentrations at the Mount Bachelor Observatory (MBO), a rural mountaintop research station in central Oregon, from 2006–2020. The model was constrained by seven predictive variables: year, day-of-year, relative humidity (RH), aerosol scattering, carbon monoxide (CO), water vapor (WV) mixing ratio, and tropopause pressure. RH, aerosol scattering, CO, and WV mixing ratio were measured at MBO, and tropopause pressure was measured via satellite. For the full 15-year period, the model represents 61% of the variance in daily peak 8 h O3, and all predictive variables have a statistically significant (p < 0.05) impact on daily peak 8 h O3 concentrations. Our results show that daily peak 8 h O3 concentrations at MBO are well-predicted by the model, thereby providing insight into what affects baseline O3 levels at a rural site on the west coast of North America.

1. Introduction

Ground-level ozone (O3) production occurs due to photochemical reactions between oxides of nitrogen (NOx = nitric oxide (NO) + nitrogen dioxide (NO2)) and volatile organic compounds (VOCs). High concentrations of O3 are harmful to human and ecosystem health [1,2,3]. Consequently, O3 is subject to regulatory action by the U.S. Environmental Protection Agency (EPA). Compliance with the O3 National Ambient Air Quality Standard (NAAQS) set by the U.S. EPA is achieved when the annual fourth-highest maximum daily 8 h average (MDA8) O3 concentration is no more than 70 parts per billion (ppb), averaged over a three-year period [4]. Numerous studies have highlighted the successes and challenges of meeting this standard across the U.S. [5,6,7,8]. Due to the health and regulatory implications of surface O3, it is important to have a comprehensive understanding of the factors controlling O3 levels.
One factor that affects ambient O3 concentrations at surface U.S. sites is the amount of U.S. background O3 (USBO), defined as the O3 that would be present in the absence of U.S. anthropogenic emissions. However, since USBO cannot be observed directly, we instead refer to baseline O3, which is the O3 concentration observed at a rural or remote site that has not been influenced by recent, local emissions [9]. Previous work has investigated baseline O3 concentrations throughout the U.S. [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26]. For example, Ambrose et al. [25] found that baseline O3 at the Mount Bachelor Observatory (MBO)—a rural mountaintop research station in central Oregon—was strongly influenced by long-range transport and upper tropospheric/lower stratospheric intrusions. In addition, Zhang and Jaffe [26] identified smoke and precursor emissions from wildfires as a source of baseline O3 at MBO.
Baseline O3 levels can have major implications for air quality management, especially in the western U.S. (WUS). In many parts of the WUS, mean seasonal baseline O3 concentrations comprise up to 70% of the national O3 standard [24,27,28], making it challenging for those areas to attain the O3 NAAQS. These high baseline O3 values are due in part to stratospheric intrusions [6,16], intercontinental transport of O3 [17], and increased wildfire activity over the past two decades [9,29].
Although many previous studies have assessed baseline O3 concentrations in the U.S., little has been done to examine the effect of individual meteorological and chemical parameters on baseline O3 at rural high-elevation sites in the WUS. This is noteworthy for three reasons. First, such locations are far from major anthropogenic pollution sources. Second, such sites are frequently impacted by free tropospheric air. These characteristics make rural high-elevation sites well-suited for investigating what influences baseline O3 concentrations. Third, as discussed above, baseline O3 is a large fraction of the ambient O3 for much of the WUS. Therefore, knowing which meteorological and chemical variables exert the greatest influence on baseline O3 in the WUS is essential for creating pollution control strategies aimed at lowering O3 levels in the region.
In the present study, we investigated the impact of several meteorological and chemical parameters on O3 concentrations at MBO from 2006–2020 by using a machine learning/statistical model that is described in the next section. The goal of this work is to gain a better understanding of what affects baseline O3 in the WUS.

2. Materials and Methods

The Mount Bachelor Observatory (MBO; 43.98° N, 121.69° W, 2764 m a.s.l.) is a rural mountaintop research station located in central Oregon that was established in 2004 [30]. Continuous measurements of O3, carbon monoxide (CO), air temperature (Tair), barometric pressure (BP), relative humidity (RH), and other chemical and meteorological parameters have been made at MBO since its inception [25,31]. The MBO data have been used previously in a number of trend and model assessments [9,26,32,33,34,35]. Since MBO is a high-elevation site located far from major urban areas, it is an ideal site for examining the variables that affect baseline O3 in the WUS.
In this study, hourly averaged data for several meteorological and chemical variables were used to investigate what influenced O3 concentrations at MBO from 2006–2020. O3 was measured using a Dasibi 1008-RS analyzer (Dasibi Environmental Corporation, Glendale, CA, USA) from 2006–2014 and an EcoTech Serinus 10 analyzer (EcoTech, Warren, RI, USA) from 2014–2020 [25,32,36]. CO was measured using a Thermo 48C-Trace Level Enhanced analyzer (Thermo Fisher Scientific, Waltham, MA, USA) from 2006–2012 and a Picarro G2502 Cavity Ring-Down Spectrometer (Picarro, Santa Clara, CA, USA) from 2012–2020 [25,32,37]. Aerosol scattering was measured using a TSI nephelometer (TSI Incorporated, Shoreview, MN, USA) [38,39,40]. Water vapor (WV) mixing ratios were calculated following the methodology of Bolton [41]. RH measurements were also included.
Figure 1 shows the diurnal variability in median O3 concentrations for each season. Consistent with previous studies conducted at MBO [25,36,42], we generally see a daytime minimum and nighttime maximum in O3 due to upslope flow during the day and downslope flow at night. This influenced how we chose to average the data, as discussed below.
For this work, all hourly data were converted to running 8 h averages. Only 8 h averages calculated using at least six valid data points were included in the analysis. The 8 h averaged O3 data were used to calculate the daily peak 8 h O3 (hereafter referred to as peak 8 h O3). To determine peak 8 h O3, we considered a 24 h period starting at 12:00 LST. This was because O3 concentrations at MBO typically exhibit a nighttime maximum and a daytime minimum (see Figure 1), consistent with other high-elevation sites [43,44,45,46,47,48]. As a result, we selected the 24 h window to start at 12:00 LST to allow for more variability in the timing of peak 8 h O3. This 24 h window was also used to calculate the 8 h averages of all other hourly data. The 8 h averages of the other hourly data for the middle hour of the 8 h period when peak 8 h O3 concentrations occurred were used in our analysis.
The 8 h averaged CO, aerosol scattering, RH, and WV mixing ratio data were used to help constrain a Generalized Additive Model (GAM), which was run using the “mgcv” package in R [49]. CO and aerosol scattering were used as model constraints because higher CO and aerosol scattering values correspond to more polluted airmasses, which likely contain higher concentrations of O3 and its precursors. RH and WV mixing ratio were used to constrain the GAM because both are anticorrelated with O3 in rural, low-NOx environments [50,51]. Daily, 1° × 1° tropopause pressure data from the Atmospheric Infrared Sounder (AIRS) were used to further constrain the GAM [52]. Data were obtained for the ascending orbits, which move from south-to-north across the Equator at 13:30 local time [52]. Due to the 1° spatial resolution, the data are regionally representative. Daily regional tropopause pressure was included as a model constraint because peak 8 h O3 at MBO is likely more influenced by lower-O3 air originating from the boundary layer when the daily regional tropopause pressure is higher. In contrast, when the daily regional tropopause pressure is lower, peak 8 h O3 at MBO is likely more influenced by higher-O3 air originating from the free troposphere.
We also tested other variables as model inputs, but they were not part of the final model configuration. Specifically, we ran the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model for each day using 1° × 1° Global Data Assimilation System (GDAS) meteorological data to calculate 24 h back trajectories for MBO. These were used to compute the direct transport distance and transport quadrant of airmasses. Even though these two variables help characterize the airmasses affecting MBO, they were not retained as model inputs because they were poorly correlated with peak 8 h O3 (result not shown). Additionally, we tested observed, 8 h averaged Tair and BP as model constraints because high temperatures and stagnant conditions are often conducive to O3 formation. However, neither Tair nor BP were retained as model inputs because we found that peak 8 h O3 concentrations at MBO are weakly dependent on both parameters (results not shown).
GAMs are statistical models that use a sum of smooth functions of predictive variables to model a response variable [29,49,53,54]. Previous studies have used GAMs and other statistical models to meteorologically adjust trends in urban O3 [55], examine the effect of wildfire smoke on urban O3 concentrations [29,54], predict high-O3 events in the Houston metropolitan area [56], and predict the impact of O3 on net ecosystem production at a forested site in the Czech Republic [57]. However, to our knowledge, this is the first study to use a GAM to investigate what influences baseline O3 levels in the WUS.
In this analysis, peak 8 h O3 was the response variable, and the predictors are listed in Table 1. In addition to the model constraints discussed previously, “Year” and “day-of-year” were included as predictive variables due to the interannual and seasonal variability in the meteorological conditions impacting O3 concentrations. One GAM simulation was done for the full 2006–2020 time period to assess the impact of each predictor on peak 8 h O3 concentrations at MBO. Our approach for configuring the GAM was similar to the one used by Gong et al. [29]. For the smoothing function associated with each predictor, we used penalized cubic regression splines (CRSs) with 10 degrees of freedom to account for the complex, nonlinear relationship between peak 8 h O3 and the predictive variables. Then, the seven selected predictors were added into the model one at a time to determine whether they decreased the Akaike information criterion (AIC) and increased the adjusted coefficient of determination (R2) [49,58,59]. Figure 2 shows that the AIC and adjusted R2 decreased and increased, respectively, when the predictors were included in the model. Since the AIC continuously declined as each predictor was added, it is unlikely that our model is overfit.
We further investigated baseline O3 concentrations in the WUS from 2006–2020 by examining the seasonal variability in hourly and peak 8 h O3 at MBO and O3 measured via ozonesonde at Trinidad Head (THD), CA. THD is a rural coastal site in northern California; therefore, its O3 profile measurements are characteristic of baseline O3 values [21]. Since the average ambient pressure at MBO is approximately 730 hPa, THD O3 profile data collected at the 680–780 hPa level during 798 balloon flights were included in our analysis so that the seasonality in O3 at the two sites could be compared.

3. Results and Discussion

The seasonal variability in O3 concentrations at MBO and THD (680–780 hPa) is shown in Figure 3. Median O3 levels at MBO using all data ranged from 43 ppb in fall to 49 ppb in spring, median peak 8 h O3 levels at MBO ranged from 46 ppb in winter to 53 ppb in spring, and median O3 levels at THD ranged from 48 ppb in winter to 55 ppb in summer. These values are (1) in line with mean baseline O3 concentrations of approximately 50 ppb reported for 15 other high-elevation sites in the WUS [16] and (2) about 61–79% of the 70 ppb O3 NAAQS. The slightly lower seasonal concentrations of hourly O3 at MBO were likely attributable to daytime upslope flow of boundary-layer air containing more moisture and less O3 compared to free tropospheric air. Meanwhile, the comparable seasonal values of peak 8 h O3 at MBO and O3 at THD suggest that the two sites are affected by similar airmasses.
Figure 4 compares the observed versus GAM-predicted peak 8 h O3 at MBO from 2006–2020. Our model effectively predicted peak 8 h O3, with an adjusted R2 of 0.61. All seven predictive variables had a statistically significant impact on peak 8 h O3 levels (p < 0.05). The effects of RH, WV mixing ratio, aerosol scattering, CO, and daily regional tropopause pressure on peak 8 h O3 are shown in Figure 5, Figure 6, Figure 7, Figure 8 and Figure 9. Peak 8 h O3 generally decreased with increasing RH and WV mixing ratio (Figure 5 and Figure 6). Since MBO is in a rural, NOx-sensitive environment, these relationships are likely due to increased removal of O3 by hydrogen oxide radicals (HOx = hydroxyl radical (OH) + hydroperoxyl radical (HO2)) at higher RH and WV mixing ratios [51]. Aerosol scattering and CO values up to approximately 30 Mm−1 and 300 ppb, respectively, had a positive relationship with peak 8 h O3 (Figure 7 and Figure 8). This is consistent with higher levels of aerosol scattering, CO, O3, and O3 precursors in more polluted airmasses. The response of peak 8 h O3 to aerosol scattering and CO values greater than about 30 Mm−1 and 300 ppb, respectively, is less clear because such high values are infrequently observed at MBO (see x-axes for Figure 7b and Figure 8b). This led to the large model uncertainty at very high aerosol scattering and CO values. As shown in Figure 9, peak 8 h O3 concentrations at MBO slightly decreased with increasing daily regional tropopause pressure. This is likely due to the lesser influence of free tropospheric air containing less moisture and more O3 when daily regional tropopause pressure is higher.
Figure 10 shows the residuals (observed peak 8 h O3—GAM-predicted peak 8 h O3) for the full 15-year period, binned by GAM-predicted peak 8 h O3 concentrations. Median residuals for all bins were close to 0 ppb, indicating that the seven-parameter model was unbiased across the O3 distribution. This further demonstrates that our model was successful in predicting peak 8 h O3 concentrations at MBO.
It needs to be noted that ongoing climate change will impact the parameters affecting baseline O3 in the WUS. For example, U.S. wildland fires have burned more than 3.2 million ha y−1 in 10 of the past 18 years, and this increase in wildfire activity has primarily taken place in the WUS [60]. Climate-related factors such as higher summertime temperatures and drought have contributed to the increasing wildfire activity [61,62]. At MBO, more wildfires have led to higher aerosol optical thickness (AOT) values over the past decade, especially during summer and fall (Figure 11). Specifically, monthly AOT values were approximately 0.4 in August and September 2017, August 2018, and September 2020. Since wildfires are expected to increase in the future [63,64,65], such high monthly AOT values at MBO may become more common. This may lead to increased suppression of O3 at MBO because high aerosol concentrations reduce solar radiation, which is not conducive to high O3 levels [66]. However, increased O3 suppression due to wildfires will only occur if MBO is increasingly impacted by fresh smoke plumes with very high aerosol loading. If increasing wildfires instead lead to an increase in the number of aged smoke plumes affecting MBO, then an increase in the number of high-O3 days at MBO will likely occur. This is because aged smoke plumes have lower aerosol loading, and O3 and aerosol scattering have a positive relationship at MBO, particularly at lower aerosol scattering values (Figure 7). Furthermore, higher temperatures and drier conditions during non-smoky periods in the WUS will likely lead to more high-O3 days at MBO. This is because WV mixing ratios and O3 are anticorrelated, and O3 production and temperature are positively correlated [26,67]. Overall, due to (1) the effects of climate change on O3 and (2) the impact of O3 levels at MBO on downwind O3 concentrations [68,69], future studies should consider reinvestigating the variables influencing O3 concentrations at this rural site.
Two other items should also be the focus of future studies. First, future work should use a GAM to predict hourly O3 concentrations at MBO. This will likely lead to an understanding of which variables have the greatest impact on the diurnal cycle of O3 at MBO. Second, future studies should use a GAM constrained with surface observations of daily maximum O3 and satellite observations of free tropospheric O3 to predict daily maximum O3 concentrations for MBO. For this analysis, the smoothing functions associated with the two predictive variables listed above will need to be penalized CRSs with two degrees of freedom. If this work is undertaken by future studies, the results may show how much daily maximum O3 concentrations at MBO are affected by transport of free tropospheric O3. Such findings would further improve our understanding of what influences baseline O3 concentrations in the WUS.

4. Conclusions

This study examined the effects of several meteorological and chemical variables on peak 8 h O3 at the Mount Bachelor Observatory (MBO) from 2006–2020. The analysis was completed using a Generalized Additive Model (GAM) constrained by seven parameters (Table 1). Over the 15-year period, our model successfully predicted the observed peak 8 h O3, with an adjusted R2 of 0.61. All predictive variables—year, day-of-year, daily regional tropopause pressure, and 8 h averaged relative humidity (RH), aerosol scattering, carbon monoxide (CO), and water vapor (WV) mixing ratio—significantly affected peak 8 h O3 concentrations (p < 0.05). Our results show that peak 8 h O3 levels at MBO were well-captured by the seven-parameter model. Therefore, since meeting the national O3 standard continues to be challenging for much of the western U.S. (WUS), future work should consider using this study’s methodology to assess what influences baseline O3 concentrations at other rural or remote sites in the region. This will help inform pollution control strategies aimed at reducing O3 levels in the WUS.

Author Contributions

Conceptualization, M.N. and D.J.; methodology, M.N. and D.J.; software, M.N.; formal analysis, M.N.; investigation, M.N.; resources, I.P., P.E., D.C. and D.J.; data curation, D.C. and D.J.; writing—original draft preparation, M.N.; writing—review and editing, M.N., I.P., P.E., D.C. and D.J.; visualization, M.N.; supervision, D.J.; funding acquisition, I.P., P.E. and D.J. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Science Foundation, grant number AGS-1447832, and the National Oceanic and Atmospheric Administration, grant numbers RA-133R-16-SE-0758 and NA17OAR4320101. The APC was funded by the National Science Foundation, grant number AGS-1447832.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data for the Mount Bachelor Observatory are publicly available via the University of Washington’s Research Works Archive (https://digital.lib.washington.edu/researchworks/discover?scope=%2F&query=%22mt.+bachelor+observatory%22&submit=&filtertype_0=title&filter_relational_operator_0=contains&filter_0=data, accessed on 18 April 2022). Trinidad Head O3 profile data for individual balloon flights can be found on the NOAA GML website (https://gml.noaa.gov/dv/data/index.php?category=Ozone&type=Balloon&site=THD, accessed on 14 July 2022). The tropopause pressure data used in this study are publicly available via the NASA GES DISC database (https://doi.org/10.5067/Aqua/AIRS/DATA303, accessed on 20 September 2022). The aerosol optical thickness data from the MODIS Aqua satellite used in this study can be found on the MODIS Adaptive Processing System Services website (http://dx.doi.org/10.5067/MODIS/MYD08_M3.006, accessed on 20 June 2022). These data sources are cited in the References [52,70,71,72].

Acknowledgments

The authors thank Kai-Lan Chang of the Cooperative Institute for Research in Environmental Sciences at the University of Colorado for providing a single file that contained the Trinidad Head O3 profile data.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Laurence, J.A. Ecological effects of ozone: Integrating exposure and response with ecosystem dynamics and function. Environ. Sci. Policy 1998, 1, 179–184. [Google Scholar] [CrossRef]
  2. Lippmann, M. Health effects of tropospheric ozone. Environ. Sci. Technol. 1991, 25, 1954–1962. [Google Scholar] [CrossRef]
  3. Zhang, J.; Wei, Y.; Fang, Z. Ozone Pollution: A Major Health Hazard Worldwide. Front. Immunol. 2019, 10, 2518. [Google Scholar] [CrossRef] [PubMed]
  4. U.S. EPA. NAAQS Table. Criteria Air Pollutants. 2016. Available online: https://www.epa.gov/criteria-air-pollutants/naaqs-table (accessed on 30 June 2022).
  5. Jaffe, D.A.; Ninneman, M.; Chan, H.C. NOx and O3 Trends at U.S. Non-Attainment Areas for 1995–2020: Influence of COVID-19 Reductions and Wildland Fires on Policy-Relevant Concentrations. J. Geophys. Res. Atmos. 2022, 127, e2021JD036385. [Google Scholar] [CrossRef] [PubMed]
  6. Langford, A.O.; Alvarez, R.J., II; Brioude, J.; Fine, R.; Gustin, M.S.; Lin, M.Y.; Marchbanks, R.D.; Pierce, R.B.; Sandberg, S.P.; Senff, C.J.; et al. Entrainment of stratospheric air and Asian pollution by the convective boundary layer in the southwestern U.S. J. Geophys. Res. Atmos. 2017, 122, 1312–1337. [Google Scholar] [CrossRef]
  7. Nussbaumer, C.M.; Cohen, R.C. The Role of Temperature and NOx in Ozone Trends in the Los Angeles Basin. Environ. Sci. Technol. 2020, 54, 15652–15659. [Google Scholar] [CrossRef]
  8. Simon, H.; Reff, A.; Wells, B.; Xing, J.; Frank, N. Ozone Trends Across the United States over a Period of Decreasing NOx and VOC Emissions. Environ. Sci. Technol. 2015, 49, 186–195. [Google Scholar] [CrossRef]
  9. Jaffe, D.A.; Cooper, O.R.; Fiore, A.M.; Henderson, B.H.; Tonnesen, G.S.; Russell, A.G.; Henze, D.K.; Langford, A.O.; Lin, M.; Moore, T. Scientific assessment of background ozone over the U.S.: Implications for air quality management. Elem. Sci. Anthr. 2018, 6, 56. [Google Scholar] [CrossRef]
  10. Dolwick, P.; Akhtar, F.; Baker, K.R.; Possiel, N.; Simon, H.; Tonnesen, G. Comparison of background ozone estimates over the western United States based on two separate model methodologies. Atmos. Environ. 2015, 109, 282–296. [Google Scholar] [CrossRef]
  11. Emery, C.; Jung, J.; Downey, N.; Johnson, J.; Jimenez, M.; Yarwood, G.; Morris, R. Regional and global modeling estimates of policy relevant background ozone over the United States. Atmos. Environ. 2012, 47, 206–217. [Google Scholar] [CrossRef]
  12. Fiore, A.M.; Jacob, D.J.; Bey, I.; Yantosca, R.M.; Field, B.D.; Fusco, A.C.; Wilkinson, J.G. Background ozone over the United States in summer: Origin, trend, and contribution to pollution episodes. J. Geophys. Res. Earth Surf. 2002, 107, ACH-11-1–ACH 11-25. [Google Scholar] [CrossRef]
  13. Fiore, A.; Jacob, D.J.; Liu, H.; Yantosca, R.M.; Fairlie, T.D.; Li, Q. Variability in surface ozone background over the United States: Implications for air quality policy. J. Geophys. Res. Earth Surf. 2003, 108, ACH-19-1–ACH-19-12. [Google Scholar] [CrossRef]
  14. Fiore, A.M.; Oberman, J.T.; Lin, M.Y.; Zhang, L.; Clifton, O.E.; Jacob, D.J.; Naik, V.; Horowitz, L.W.; Pinto, J.P.; Milly, G.P. Estimating North American background ozone in U.S. surface air with two independent global models: Variability, uncertainties, and recommendations. Atmos. Environ. 2014, 96, 284–300. [Google Scholar] [CrossRef]
  15. Lefohn, A.S.; Emery, C.; Shadwick, D.; Wernli, H.; Jung, J.; Oltmans, S.J. Estimates of background surface ozone concentrations in the United States based on model-derived source apportionment. Atmos. Environ. 2014, 84, 275–288. [Google Scholar] [CrossRef]
  16. Lin, M.; Fiore, A.M.; Cooper, O.R.; Horowitz, L.W.; Langford, A.O.; Levy, H., II; Johnson, B.J.; Naik, V.; Oltmans, S.J.; Senff, C.J.; et al. Springtime high surface ozone events over the western United States: Quantifying the role of stratospheric intrusions. J. Geophys. Res. Earth Surf. 2012, 117, D00V22. [Google Scholar] [CrossRef]
  17. Miyazaki, K.; Neu, J.L.; Osterman, G.; Bowman, K. Changes in US background ozone associated with the 2011 turnaround in Chinese NOx emissions. Environ. Res. Commun. 2022, 4, 045003. [Google Scholar] [CrossRef]
  18. Parrish, D.D.; Ennis, C.A. Estimating background contributions and US anthropogenic enhancements to maximum ozone concentrations in the northern US. Atmos. Chem. Phys. 2019, 19, 12587–12605. [Google Scholar] [CrossRef]
  19. Parrish, D.D.; Young, L.M.; Newman, M.H.; Aikin, K.C.; Ryerson, T.B. Ozone Design Values in Southern California’s Air Basins: Temporal Evolution and U.S. Background Contribution. J. Geophys. Res. Atmos. 2017, 122, 11–166, 182. [Google Scholar] [CrossRef]
  20. Parrish, D.D.; Faloona, I.C.; Derwent, R.G. Observational-based assessment of contributions to maximum ozone concentrations in the western United States. J. Air Waste Manag. Assoc. 2022, 72, 434–454. [Google Scholar] [CrossRef]
  21. Stauffer, R.M.; Thompson, A.M.; Oltmans, S.J.; Johnson, B.J. Tropospheric ozonesonde profiles at long-term U.S. monitoring sites: 2. Links between Trinidad Head, CA, profile clusters and inland surface ozone measurements. J. Geophys. Res. Atmos. 2017, 122, 1261–1280. [Google Scholar] [CrossRef]
  22. Wang, H.; Jacob, D.J.; Le Sager, P.; Streets, D.G.; Park, R.J.; Gilliland, A.B.; van Donkelaar, A. Surface ozone background in the United States: Canadian and Mexican pollution influences. Atmos. Environ. 2009, 43, 1310–1319. [Google Scholar] [CrossRef]
  23. Yan, Q.; Wang, Y.; Cheng, Y.; Li, J. Summertime Clean-Background Ozone Concentrations Derived from Ozone Precursor Relationships are Lower than Previous Estimates in the Southeast United States. Environ. Sci. Technol. 2021, 55, 12852–12861. [Google Scholar] [CrossRef] [PubMed]
  24. Zhang, L.; Jacob, D.J.; Downey, N.V.; Wood, D.A.; Blewitt, D.; Carouge, C.C.; van Donkelaar, A.; Jones, D.B.; Murray, L.; Wang, Y. Improved estimate of the policy-relevant background ozone in the United States using the GEOS-Chem global model with 1/2° × 2/3° horizontal resolution over North America. Atmos. Environ. 2011, 45, 6769–6776. [Google Scholar] [CrossRef]
  25. Ambrose, J.L.; Reidmiller, D.R.; Jaffe, D.A. Causes of high O3 in the lower free troposphere over the Pacific Northwest as observed at the Mt. Bachelor Observatory. Atmos. Environ. 2011, 45, 5302–5315. [Google Scholar] [CrossRef]
  26. Zhang, L.; Jaffe, D.A. Trends and sources of ozone and sub-micron aerosols at the Mt. Bachelor Observatory (MBO) during 2004–2015. Atmos. Environ. 2017, 165, 143–154. [Google Scholar] [CrossRef]
  27. Jaffe, D.A.; Zhang, L. Meteorological anomalies lead to elevated O 3 in the western U.S. in June 2015. Geophys. Res. Lett. 2017, 44, 1990–1997. [Google Scholar] [CrossRef]
  28. Jaffe, D.A.; Fiore, A.M.; Keating, T.J. Importance of background ozone for air quality management. The Magazine for Environmental Managers, 1 November 2020; 1–5. Available online: https://pubs.awma.org/flip/EM-Nov-2020/jaffe.pdf (accessed on 19 July 2022).
  29. Gong, X.; Kaulfus, A.; Nair, U.; Jaffe, D.A. Quantifying O3 Impacts in Urban Areas Due to Wildfires Using a Generalized Additive Model. Environ. Sci. Technol. 2017, 51, 13216–13223. [Google Scholar] [CrossRef]
  30. Jaffe, D.; Prestbo, E.; Swartzendruber, P.; Weisspenzias, P.; Kato, S.; Takami, A.; Hatakeyama, S.; Kajii, Y. Export of atmospheric mercury from Asia. Atmos. Environ. 2005, 39, 3029–3038. [Google Scholar] [CrossRef]
  31. Baylon, P.; Jaffe, D.A.; Hall, S.R.; Ullmann, K.; Alvarado, M.J.; Lefer, B.L. Impact of Biomass Burning Plumes on Photolysis Rates and Ozone Formation at the Mount Bachelor Observatory. J. Geophys. Res. Atmos. 2018, 123, 2272–2284. [Google Scholar] [CrossRef]
  32. Gratz, L.E.; Jaffe, D.A.; Hee, J.R. Causes of increasing ozone and decreasing carbon monoxide in springtime at the Mt. Bachelor Observatory from 2004 to 2013. Atmos. Environ. 2015, 109, 323–330. [Google Scholar] [CrossRef]
  33. Zhang, L.; Jacob, D.J.; Boersma, K.F.; Jaffe, D.A.; Olson, J.R.; Bowman, K.W.; Worden, J.R.; Thompson, A.M.; Avery, M.A.; Cohen, R.C.; et al. Transpacific transport of ozone pollution and the effect of recent Asian emission increases on air quality in North America: An integrated analysis using satellite, aircraft, ozonesonde, and surface observations. Atmos. Chem. Phys. 2008, 8, 6117–6136. [Google Scholar] [CrossRef]
  34. Gaudel, A.; Cooper, O.R.; Ancellet, G.; Barret, B.; Boynard, A.; Burrows, J.P.; Clerbaux, C.; Coheur, P.-F.; Cuesta, J.; Cuevas, E.; et al. Tropospheric Ozone Assessment Report: Present-day distribution and trends of tropospheric ozone relevant to climate and global atmospheric chemistry model evaluation. Elem. Sci. Anthr. 2018, 6, 39. [Google Scholar] [CrossRef]
  35. Qu, Z.; Henze, D.K.; Cooper, O.R.; Neu, J.L. Impacts of global NOx inversions on NO2 and ozone simulations. Atmos. Chem. Phys. 2020, 20, 13109–13130. [Google Scholar] [CrossRef]
  36. Weiss-Penzias, P.; Jaffe, D.A.; Swartzendruber, P.; Dennison, J.B.; Chand, D.; Hafner, W.; Prestbo, E. Observations of Asian air pollution in the free troposphere at Mount Bachelor Observatory during the spring of 2004. J. Geophys. Res. Earth Surf. 2006, 111, D10304. [Google Scholar] [CrossRef]
  37. Chen, H.; Karion, A.; Rella, C.W.; Winderlich, J.; Gerbig, C.; Filges, A.; Newberger, T.; Sweeney, C.; Tans, P.P. Accurate measurements of carbon monoxide in humid air using the cavity ring-down spectroscopy (CRDS) technique. Atmos. Meas. Tech. 2013, 6, 1031–1040. [Google Scholar] [CrossRef]
  38. Briggs, N.L.; Jaffe, D.A.; Gao, H.; Hee, J.R.; Baylon, P.M.; Zhang, Q.; Zhou, S.; Collier, S.C.; Sampson, P.D.; Cary, R.A. Particulate Matter, Ozone, and Nitrogen Species in Aged Wildfire Plumes Observed at the Mount Bachelor Observatory. Aerosol Air Qual. Res. 2016, 16, 3075–3087. [Google Scholar] [CrossRef]
  39. Zhou, S.; Collier, S.; Jaffe, D.A.; Briggs, N.L.; Hee, J.; Sedlacek, A.J., III; Kleinman, L.; Onasch, T.B.; Zhang, Q. Regional influence of wildfires on aerosol chemistry in the western US and insights into atmospheric aging of biomass burning organic aerosol. Atmos. Chem. Phys. 2017, 17, 2477–2493. [Google Scholar] [CrossRef]
  40. Zhou, S.; Collier, S.; Jaffe, D.A.; Zhang, Q. Free tropospheric aerosols at the Mt. Bachelor Observatory: More oxidized and higher sulfate content compared to boundary layer aerosols. Atmos. Chem. Phys. 2019, 19, 1571–1585. [Google Scholar] [CrossRef]
  41. Bolton, D. The Computation of Equivalent Potential Temperature. Mon. Weather Rev. 1980, 108, 1046–1053. [Google Scholar] [CrossRef]
  42. Reidmiller, D.R.; Jaffe, D.A.; Fischer, E.V.; Finley, B. Nitrogen oxides in the boundary layer and free troposphere at the Mt. Bachelor Observatory. Atmos. Chem. Phys. 2010, 10, 6043–6062. [Google Scholar] [CrossRef]
  43. Aneja, V.P.; Businger, S.; Li, Z.; Claiborn, C.S.; Murthy, A. Ozone climatology at high elevations in the southern Appalachians. J. Geophys. Res. Earth Surf. 1991, 96, 1007. [Google Scholar] [CrossRef]
  44. Aneja, V.P.; Li, Z. Characterization of ozone at high elevation in the eastern United States: Trends, seasonal variations, and exposure. J. Geophys. Res. Earth Surf. 1992, 97, 9873–9888. [Google Scholar] [CrossRef]
  45. Lefohn, A.S.; Shadwick, D.S.; Mohnen, V.A. The characterization of ozone concentrations at a select set of high-elevation sites in the eastern United States. Environ. Pollut. 1990, 67, 147–178. [Google Scholar] [CrossRef]
  46. Lefohn, A.S.; Mohnen, V.A. The Characterization of Ozone, Sulfur Dioxide, and Nitrogen Dioxide for Selected Monitoring Sites in the Federal Republic of Germany. J. Air Pollut. Control Assoc. 1986, 36, 1329–1337. [Google Scholar] [CrossRef]
  47. Mohnen, V.A.; Hogan, A.; Coffey, P. Ozone measurements in rural areas. J. Geophys. Res. Earth Surf. 1977, 82, 5889–5895. [Google Scholar] [CrossRef]
  48. Naja, M.; Lal, S.; Chand, D. Diurnal and seasonal variabilities in surface ozone at a high altitude site Mt Abu (24.6°N, 72.7°E, 1680 m asl) in India. Atmos. Environ. 2003, 37, 4205–4215. [Google Scholar] [CrossRef]
  49. Wood, S.N. Generalized Additive Models: An Introduction with R; Chapman & Hall/CRC: Boca Raton, FL, USA, 2006. [Google Scholar]
  50. Kavassalis, S.C.; Murphy, J.G. Understanding ozone-meteorology correlations: A role for dry deposition. Geophys. Res. Lett. 2017, 44, 2922–2931. [Google Scholar] [CrossRef]
  51. Lu, X.; Zhang, L.; Shen, L. Meteorology and Climate Influences on Tropospheric Ozone: A Review of Natural Sources, Chemistry, and Transport Patterns. Curr. Pollut. Rep. 2019, 5, 238–260. [Google Scholar] [CrossRef]
  52. Texeira, J. AIRS/Aqua L3 Daily Standard Physical Retrieval (AIRS-only) 1 degree × 1 degree V006. Available online: https://disc.gsfc.nasa.gov/datasets/AIRS3STD_006/summary (accessed on 20 September 2022).
  53. Hastie, T.J.; Tibshirani, R.J. Generalized Additive Models; Chapman & Hall/CRC: Boca Raton, FL, USA, 1990. [Google Scholar]
  54. McClure, C.D.; Jaffe, D.A. Investigation of high ozone events due to wildfire smoke in an urban area. Atmos. Environ. 2018, 194, 146–157. [Google Scholar] [CrossRef]
  55. Camalier, L.; Cox, W.; Dolwick, P. The effects of meteorology on ozone in urban areas and their use in assessing ozone trends. Atmos. Environ. 2007, 41, 7127–7137. [Google Scholar] [CrossRef]
  56. Sun, W.; Palazoglu, A.; Singh, A.; Zhang, H.; Wang, Q.; Zhao, Z.; Cao, D. Prediction of surface ozone episodes using clusters based generalized linear mixed effects models in Houston–Galveston–Brazoria area, Texas. Atmos. Pollut. Res. 2015, 6, 245–253. [Google Scholar] [CrossRef]
  57. Jurán, S.; Edwards-Jonášová, M.; Cudlín, P.; Zapletal, M.; Šigut, L.; Grace, J.; Urban, O. Prediction of ozone effects on net ecosystem production of Norway spruce forest. iForest-Biogeos. For. 2018, 11, 743–750. [Google Scholar] [CrossRef]
  58. Cavanaugh, J.E.; Neath, A.A. The Akaike information criterion: Background, derivation, properties, application, interpretation, and refinements. WIREs Comput. Stat. 2019, 11, e1460. [Google Scholar] [CrossRef]
  59. CFI Team. Adjusted R-Squared. Available online: https://corporatefinanceinstitute.com/resources/knowledge/other/adjusted-r-squared/ (accessed on 1 October 2022).
  60. National Interagency Fire Center. Fire information: Statistics. Available online: https://www.nifc.gov/fire-information/statistics (accessed on 5 July 2022).
  61. Aldersley, A.; Murray, S.J.; Cornell, S.E. Global and regional analysis of climate and human drivers of wildfire. Sci. Total Environ. 2011, 409, 3472–3481. [Google Scholar] [CrossRef] [PubMed]
  62. Decker, Z.C.J.; Zarzana, K.J.; Coggon, M.; Min, K.-E.; Pollack, I.; Ryerson, T.B.; Peischl, J.; Edwards, P.; Dubé, W.P.; Markovic, M.Z.; et al. Nighttime Chemical Transformation in Biomass Burning Plumes: A Box Model Analysis Initialized with Aircraft Observations. Environ. Sci. Technol. 2019, 53, 2529–2538. [Google Scholar] [CrossRef]
  63. Moritz, M.A.; Parisien, M.-A.; Batllori, E.; Krawchuk, M.A.; Van Dorn, J.; Ganz, D.J.; Hayhoe, K. Climate change and disruptions to global fire activity. Ecosphere 2012, 3, 1–22. [Google Scholar] [CrossRef]
  64. Pechony, O.; Shindell, D.T. Driving forces of global wildfires over the past millennium and the forthcoming century. Proc. Natl. Acad. Sci. USA 2010, 107, 19167–19170. [Google Scholar] [CrossRef]
  65. Val Martin, M.; Heald, C.L.; Lamarque, J.-F.; Tilmes, S.; Emmons, L.K.; Schichtel, B.A. How emissions, climate, and land use change will impact mid-century air quality over the United States: A focus on effects at national parks. Atmos. Chem. Phys. 2015, 15, 2805–2823. [Google Scholar] [CrossRef]
  66. Buysse, C.E.; Kaulfus, A.; Nair, U.; Jaffe, D.A. Relationships between Particulate Matter, Ozone, and Nitrogen Oxides during Urban Smoke Events in the Western US. Environ. Sci. Technol. 2019, 53, 12519–12528. [Google Scholar] [CrossRef]
  67. Sillman, S.; Samson, P.J. Impact of temperature on oxidant photochemistry in urban, polluted rural and remote environments. J. Geophys. Res. Earth Surf. 1995, 100, 11497–11508. [Google Scholar] [CrossRef]
  68. Baylon, P.M.; Jaffe, D.A.; Pierce, R.B.; Gustin, M.S. Interannual Variability in Baseline Ozone and Its Relationship to Surface Ozone in the Western U.S. Environ. Sci. Technol. 2016, 50, 2994–3001. [Google Scholar] [CrossRef] [PubMed]
  69. Wigder, N.L.; Jaffe, D.A.; Herron-Thorpe, F.L.; Vaughan, J.K. Influence of daily variations in baseline ozone on urban air quality in the United States Pacific Northwest. J. Geophys. Res. Atmos. 2013, 118, 3343–3354. [Google Scholar] [CrossRef]
  70. Research Works Archive. Search: Mt. Bachelor Observatory. Available online: https://digital.lib.washington.edu/researchworks/discover?scope=%2F&query=%22mt.+bachelor+observatory%22&submit=&filtertype_0=title&filter_relational_operator_0=contains&filter_0=data (accessed on 18 April 2022).
  71. NOAA GML. GML Data Finder. Available online: https://gml.noaa.gov/dv/data/index.php?category=Ozone&type=Balloon&site=THD (accessed on 14 July 2022).
  72. Platnick, S.; Hubanks, P.; Meyer, K.; King, M.D. MODIS Atmosphere L3 Monthly Product (08_M3). Available online: http://dx.doi.org/10.5067/MODIS/MYD08_M3.006 (accessed on 20 June 2022).
Atmosphere 13 01883 g001
Figure 1. Diurnal variability in the median O3 concentration at MBO for spring (MAM), summer (JJA), fall (SON), and winter (DJF) 2006–2020.
Figure 1. Diurnal variability in the median O3 concentration at MBO for spring (MAM), summer (JJA), fall (SON), and winter (DJF) 2006–2020.
Atmosphere 13 01883 g001
Atmosphere 13 01883 g002
Figure 2. (a) Magnitude of adjusted R2 and AIC with additional variables for MBO from 2006–2020. (b) Changes in adjusted R2 and AIC with additional variables for MBO from 2006–2020. See Table 1 for the parameter names corresponding to the parameter numbers.
Figure 2. (a) Magnitude of adjusted R2 and AIC with additional variables for MBO from 2006–2020. (b) Changes in adjusted R2 and AIC with additional variables for MBO from 2006–2020. See Table 1 for the parameter names corresponding to the parameter numbers.
Atmosphere 13 01883 g002
Atmosphere 13 01883 g003
Figure 3. Seasonal O3 concentrations at MBO and THD from 2006–2020. Hourly O3 data were used to generate the MBO boxplots, and 680–780 hPa O3 data collected via ozonesonde were used to generate the THD boxplots. The bottom and top whiskers denote the 10th and 90th percentile values, respectively, the central rectangles span the 25th percentile to the 75th percentile, and the horizontal lines within the central rectangles represent the median values.
Figure 3. Seasonal O3 concentrations at MBO and THD from 2006–2020. Hourly O3 data were used to generate the MBO boxplots, and 680–780 hPa O3 data collected via ozonesonde were used to generate the THD boxplots. The bottom and top whiskers denote the 10th and 90th percentile values, respectively, the central rectangles span the 25th percentile to the 75th percentile, and the horizontal lines within the central rectangles represent the median values.
Atmosphere 13 01883 g003
Atmosphere 13 01883 g004
Figure 4. Observed versus GAM-predicted peak 8 h O3 at MBO from 2006–2020. The solid red line and the dashed blue line are the trendline and 1:1 line, respectively.
Figure 4. Observed versus GAM-predicted peak 8 h O3 at MBO from 2006–2020. The solid red line and the dashed blue line are the trendline and 1:1 line, respectively.
Atmosphere 13 01883 g004
Atmosphere 13 01883 g005
Figure 5. (a) Observed peak 8 h O3 (ppb) versus observed 8 h average RH (%) at MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by 8 h average RH. The red squares are centered on the median 8 h average RH and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of 8 h average RH (%) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed 8 h average RH values. The spline smoothing function for 8 h average RH is on the y-axis, with the label including its degrees of freedom (4.66). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds.
Figure 5. (a) Observed peak 8 h O3 (ppb) versus observed 8 h average RH (%) at MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by 8 h average RH. The red squares are centered on the median 8 h average RH and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of 8 h average RH (%) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed 8 h average RH values. The spline smoothing function for 8 h average RH is on the y-axis, with the label including its degrees of freedom (4.66). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds.
Atmosphere 13 01883 g005
Atmosphere 13 01883 g006
Figure 6. (a) Observed peak 8 h O3 (ppb) versus observed 8 h average WV mixing ratio (g kg−1) at MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by 8 h average WV mixing ratio. The red squares are centered on the median 8 h average WV mixing ratio and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of 8 h average WV mixing ratio (g kg−1) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed 8 h average WV mixing ratios. The spline smoothing function for 8 h average WV mixing ratio is on the y-axis, with the label including its degrees of freedom (4.86). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds.
Figure 6. (a) Observed peak 8 h O3 (ppb) versus observed 8 h average WV mixing ratio (g kg−1) at MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by 8 h average WV mixing ratio. The red squares are centered on the median 8 h average WV mixing ratio and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of 8 h average WV mixing ratio (g kg−1) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed 8 h average WV mixing ratios. The spline smoothing function for 8 h average WV mixing ratio is on the y-axis, with the label including its degrees of freedom (4.86). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds.
Atmosphere 13 01883 g006
Atmosphere 13 01883 g007
Figure 7. (a) Observed peak 8 h O3 (ppb) versus observed 8 h average aerosol scattering (Mm−1) at MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by 8 h average aerosol scattering. The red squares are centered on the median 8 h average aerosol scattering value and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of 8 h average aerosol scattering (Mm−1) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed 8 h average aerosol scattering values. The spline smoothing function for 8 h average aerosol scattering is on the y-axis, with the label including its degrees of freedom (7.58). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds. Note that the x-axes for panels (a) and (b) are plotted on a logarithmic scale.
Figure 7. (a) Observed peak 8 h O3 (ppb) versus observed 8 h average aerosol scattering (Mm−1) at MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by 8 h average aerosol scattering. The red squares are centered on the median 8 h average aerosol scattering value and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of 8 h average aerosol scattering (Mm−1) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed 8 h average aerosol scattering values. The spline smoothing function for 8 h average aerosol scattering is on the y-axis, with the label including its degrees of freedom (7.58). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds. Note that the x-axes for panels (a) and (b) are plotted on a logarithmic scale.
Atmosphere 13 01883 g007
Atmosphere 13 01883 g008
Figure 8. (a) Observed peak 8 h O3 (ppb) versus observed 8 h average CO (ppb) at MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by 8 h average CO. The red squares are centered on the median 8 h average CO concentration and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of 8 h average CO (ppb) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed 8 h average CO concentrations. The spline smoothing function for 8 h average CO is on the y-axis, with the label including its degrees of freedom (6.74). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds. Note that the x-axes for panels (a) and (b) are plotted on a logarithmic scale.
Figure 8. (a) Observed peak 8 h O3 (ppb) versus observed 8 h average CO (ppb) at MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by 8 h average CO. The red squares are centered on the median 8 h average CO concentration and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of 8 h average CO (ppb) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed 8 h average CO concentrations. The spline smoothing function for 8 h average CO is on the y-axis, with the label including its degrees of freedom (6.74). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds. Note that the x-axes for panels (a) and (b) are plotted on a logarithmic scale.
Atmosphere 13 01883 g008
Atmosphere 13 01883 g009
Figure 9. (a) Observed peak 8 h O3 (ppb) versus observed daily regional tropopause pressure (hPa) for MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by daily regional tropopause pressure. The red squares are centered on the median daily regional tropopause pressure and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of daily regional tropopause pressure (hPa) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed daily regional tropopause pressure values. The spline smoothing function for daily regional tropopause pressure is on the y-axis, with the label including its degrees of freedom (3.81). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds.
Figure 9. (a) Observed peak 8 h O3 (ppb) versus observed daily regional tropopause pressure (hPa) for MBO from 2006–2020. The black dots show the individual data points, and the connected red squares show the median peak 8 h O3 concentration, binned by daily regional tropopause pressure. The red squares are centered on the median daily regional tropopause pressure and the median peak 8 h O3 concentration for each bin. (b) Partial response plot showing the effect of daily regional tropopause pressure (hPa) on model-predicted peak 8 h O3 at MBO from 2006–2020. The tick marks on the x-axis denote the density of observed daily regional tropopause pressure values. The spline smoothing function for daily regional tropopause pressure is on the y-axis, with the label including its degrees of freedom (3.81). The solid line shows the smooth curve, and the dashed lines indicate 2 standard error bounds.
Atmosphere 13 01883 g009
Atmosphere 13 01883 g010
Figure 10. Residuals (observed peak 8 h O3 − GAM-predicted peak 8 h O3) at MBO from 2006–2020, binned by GAM-predicted peak 8 h O3 concentrations. The dashed red line denotes 100% agreement between observed and GAM-predicted peak 8 h O3 (i.e., residuals = 0 ppb). The components of each boxplot have the same meanings as in Figure 3.
Figure 10. Residuals (observed peak 8 h O3 − GAM-predicted peak 8 h O3) at MBO from 2006–2020, binned by GAM-predicted peak 8 h O3 concentrations. The dashed red line denotes 100% agreement between observed and GAM-predicted peak 8 h O3 (i.e., residuals = 0 ppb). The components of each boxplot have the same meanings as in Figure 3.
Atmosphere 13 01883 g010
Atmosphere 13 01883 g011
Figure 11. Monthly aerosol optical thickness (AOT) at MBO from 2006–2020. AOT data with 1° resolution were taken from the MODIS Aqua satellite.
Figure 11. Monthly aerosol optical thickness (AOT) at MBO from 2006–2020. AOT data with 1° resolution were taken from the MODIS Aqua satellite.
Atmosphere 13 01883 g011
Table 1. List of parameters used to constrain the Generalized Additive Model (GAM) for this study.
Table 1. List of parameters used to constrain the Generalized Additive Model (GAM) for this study.
Data Source *Parameter NumberParameter Name
(Unit)		Description
11Year (unitless)Year
12DOY (unitless)Day-of-year
23RH_8h (%)8 h average relative humidity
24Scattering_8h (Mm−1)8 h average aerosol scattering
25CO_8h (ppb)8 h average carbon monoxide
26WV_8h (g kg−1)8 h average water vapor mixing ratio
37Tropopause_Pres (hPa)Daily, satellite-derived regional tropopause pressure
* (1) Calculation, (2) MBO data archive, and (3) ascending orbit of the Atmospheric Infrared Sounder (AIRS).
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Ninneman, M.; Petropavlovskikh, I.; Effertz, P.; Chand, D.; Jaffe, D. Investigation of the Parameters Influencing Baseline Ozone in the Western United States: A Statistical Modeling Approach. Atmosphere 2022, 13, 1883. https://doi.org/10.3390/atmos13111883

AMA Style

Ninneman M, Petropavlovskikh I, Effertz P, Chand D, Jaffe D. Investigation of the Parameters Influencing Baseline Ozone in the Western United States: A Statistical Modeling Approach. Atmosphere. 2022; 13(11):1883. https://doi.org/10.3390/atmos13111883

Chicago/Turabian Style

Ninneman, Matthew, Irina Petropavlovskikh, Peter Effertz, Duli Chand, and Daniel Jaffe. 2022. "Investigation of the Parameters Influencing Baseline Ozone in the Western United States: A Statistical Modeling Approach" Atmosphere 13, no. 11: 1883. https://doi.org/10.3390/atmos13111883

APA Style

Ninneman, M., Petropavlovskikh, I., Effertz, P., Chand, D., & Jaffe, D. (2022). Investigation of the Parameters Influencing Baseline Ozone in the Western United States: A Statistical Modeling Approach. Atmosphere, 13(11), 1883. https://doi.org/10.3390/atmos13111883

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop