2004 has had the lowest ozone smog levels since states began measuring the stuff back in the 1970s. Based on preliminary data from around the country, days exceeding EPA's tough new 8-hour ozone standard declined an average of about 50 percent below 2003, which was itself a record year.
A combination of continuing emission reductions and favorable weather explains the improvements. Weather is the single largest factor affecting year-to-year variations in smog levels. All else equal, cool, wet, and windy years will have less ozone than warm, dry, and calm ones. But weather is only part of the story. During the last 30 years most of the country has had several years that were cooler and/or wetter than 2004, but never have smog levels been anywhere near this low.
The charts below will give you an idea of how extraordinary 2004 was. Figure 1 shows the average number of days per year exceeding EPA's 1-hour and 8-hour ozone standards at the nation's ozone monitoring sites from 1975 to 2003. For each standard, the chart includes the exceedance rate for all sites operating in any given year (generally about 700 to 1,200), and also for the 261 sites that operated continuously from 1983-2003. The percentages at right give the decline in the number of 8-hour and 1-hour exceedances since 1975.
Note that 2003 was the best year on record, barely edging out 2000, and that the average number of 8-hour ozone exceedances varies greatly from year to year. Annual variations in weather create the large short-term variability in smog levels, but superimposed on this is a long-term decline in ozone exceedances due to emission reductions.
Figure 1. National Trend in Days per Year Exceeding EPA's Ozone Standards, 1970-2003
Figure 1 can't be extended through 2004 until all states have reported their 2004 ozone data. However, preliminary data for 2004 are available on the web for several metropolitan areas and states. Figure 2 compares 8-hour ozone exceedances in 2003 and 2004 for several of these areas. For each area, the chart gives the average number of 8-hour ozone exceedances for all monitoring locations that had data for both years. A similar figure at the end of this column provides data for the worst location in each of the areas in Figure 2.
Note the large declines almost across the board. I wasn't able to locate data for the Midwest in time for this column, but EPA's web site reports that there wasn't a single "ozone action day" in 2004 in all of Illinois, Michigan, Ohio, Wisconsin, and Minnesota, while southern Indiana (the portion in the Louisville, Kentucky metro area) had just one.
Figure 2. Average Number of Days per Year Exceeding the 8-hour Ozone Standard during 2003 and 2004
Overall, 8-hour ozone exceedance days declined an average of about 50 percent between 2003 and 2004, meaning that 2004 is not only the best year on record, but the best by a large margin.
You wouldn't know this from reading activists' reports on air quality, which continue to tell a deceitfully gloomy story. Dangerous Days of Summer from Environmental Defense (ED) and Danger in the Air from the Public Interest Research Group (PIRG) are the two latest entries. Neither report mentions that 2003 and 2004 were the best years in history for ozone. PIRG does mention that 2003 and 2004 were better than 2002, but attributes it all to weather.
Nevertheless, as you might expect, activists are always ready with a press release in years when air pollution rises. When ozone levels spiked upward during the hot, dry summer of 2002, a Clean Air Trust press release proclaimed "New Survey Finds Massive Smog Problem in 2002." But no activist press releases highlighted the spectacular decline in ozone levels the next year, or the record-low ozone levels of the last two years.
Other potential, but unmentioned contributors to the recent ozone improvements are a 60 percent reduction in coal-fired power plant NOx emissions during the May-September "ozone season," implemented in May 2004 under EPA's NOx SIP Call regulation, and an ongoing reduction of about eight percent per year in total automobile emissions due to fleet turnover to cleaner vehicles. Activists avoid mentioning these reductions, because they undermine their claims that urban "sprawl" increases air pollution and that power-plant emissions are increasing.
ED's Dangerous Days commits the full range of deceptions pioneered by the American Lung Association (ALA) in its annual State of the Air series, such as: inflating pollution levels, exaggerating the harm from current air pollution levels and the number of people living in areas that exceed EPA standards, downplaying positive trends, and creating the impression that there will be little or no future improvement without stringent new regulations.
For example, Dangerous Days claims the New York metro area exceeded the 8-hour ozone standard on 22 percent of summer days during 2001-2003. But the average site in the New York area exceeded the 8-hour standard on 10 percent of summer days -- less than half of ED's claim. ED's number is higher than even the worst site in the New York area (Jackson Township, NJ), which exceeded the 8-hour standard 20 percent of the summer. ED likewise inflated ozone levels in all of the country's metro areas.
But Environmental Defense's ozone inflation is even worse than this, because most people in the New York area live in places with the lowest ozone levels. Monitoring sites in the five boroughs of New York City averaged 3.7 percent of summer days exceeding the 8-hour standard, or 1/6th of ED's claim. Likewise, ED claimed Los Angeles exceeded the 8-hour standard on 50 percent of summer days. But about half of Angelenos live in areas that never exceed the 8-hour standard. ED also fails to distinguish between moderate and high ozone. Most ozone exceedances involved relatively low ozone levels. The average site in the New York metro area exceeded the higher 1-hour ozone standard on only 2 percent of summer days, compared with 10 percent for the 8-hour standard.
Dangerous Days also exaggerates the number of people who live in areas that violate EPA's air standards. According to the report, "Nearly 160 million Americans live in areas where ozone smog levels exceed national standards...Some 99 million Americans live in areas that exceed annual fine particle standards."
Both of these numbers are based on the populations of counties designated as "non-attainment" areas by EPA. But this has little to do with actual pollution levels, because EPA designates whole regions as non-attainment areas even if only a single monitoring location violates a federal standard. This makes sense for air quality planning, but not for determining air pollution exposure. Thus, 94 to 99 percent of people in San Diego, Chicago, Las Vegas, and Phoenix live in areas that meet all EPA ozone standards, but EPA counts everyone in those areas as breathing dirty air.
The claim for fine particulate matter (PM2.5) is misleadingly high for an additional reason: EPA designated some counties as PM2.5 non-attainment areas not because they exceed the PM2.5 standard, but because they are believed to contribute to violations elsewhere. All told, ED overestimates by more than a factor of two the number of people living in areas that violate EPA standards.
Dangerous Days also implies air pollution is responsible for rising asthma rates: "Asthma has increasingly gained attention as a nationwide epidemic and a symbol of the manifold health impacts of air pollution. It is the nation's fastest growing chronic disease..." Yet air pollution can't be a cause of rising asthma, because air pollution of all kinds has been falling nationwide at the same time that asthma has been rising.
Air pollution can aggravate pre-existing respiratory disease, but its impact is nothing close to what groups like ED claim. For example, when the Clinton-era EPA developed the 8-hour ozone standard, it predicted that going from full national attainment of the 1-hour standard to full national attainment of the 8-hour standard would reduce hospital admissions for asthma by 0.6 percent, despite the 8-hour standard's much greater stringency. Data from around the U.S. show that asthma hospitalizations are lowest in July and August-when ozone and, in many areas PM, are highest. Air pollution has gained the "national attention" referred to by ED not because of its overall importance as a cause of disease and disability, but because of its rhetorical power to generate eye-catching headlines, donations, and research funding.
PIRG's Danger in the Air makes ED's Dangerous Days look like a model of reliable analysis. To arrive at its claims about ozone exceedances, PIRG simply adds up the ozone exceedances at each monitoring location in a city or state and calls that the number of exceedances for the area. Thus, PIRG claims Colorado exceeded the 8-hour ozone standard 60 times in 2003, even though the worst location in the state had 15 exceedances, and the average location had less than four. Despite a national average of about four 8-hour ozone exceedances per year in 2003 (see Figure 1), PIRG managed to cook up 4,583 exceedances-a particularly masterful feat when you consider that there are only 365 days in a year. PIRG's method is meaningless for determining health risks or anything else about actual air pollution, but it's great for generating big, scary numbers.
I've often criticized the media for their mostly gloomy and misleading accounts of air quality issues. For example, despite the substantial decline in ozone exceedances since the 1970s (see Figure 1), in a story on ALA's State of the Air 2004 the Washington Post asserted "Ozone pollution has declined slightly over the past 30 years" (emphasis added). But many reporters around the country have noticed this years' unusually low pollution levels and have let the public know about it. Even here, however, most stories gave the impression that mild weather was the sole cause, and failed to discuss the long-term decline in smog-forming emissions or to compare smog levels in 2004 with much higher smog levels in previous years that had favorable weather.
Will air pollution remain just as low next year? That depends largely on the weather. Either way, emissions will continue to decline and the long-term trend will continue downward. Regardless, environmental activists are sure to tell us the sky is falling.
Figure 3. Number of Days per Year Exceeding the 8-hour Ozone Standard at the Worst Site in Each State or Metropolitan Area during 2003 and 2004
Notes: Los Angeles-San Bernardino (LA-SB) and the San Joaquin Valley (SJV) are excluded in order to keep the vertical scale from being too compressed for the other areas. For the record, the 2003 and 2004 values at the worst sites in these areas are as follows: LA-SB, 72 and 62; SJV, 116 and 94. Arvin, the worst area in the SJV, might end up with a few additional ozone exceedances over the next couple of weeks.
Joel Schwartz is an adjunct fellow at Reason Foundation and visiting scholar at American Enterprise Institute.
 Note that is a statement about nationwide average ozone levels. It doesn't mean that ozone declined everywhere in 2004 or that 2004 was the best year everywhere. Some areas experienced ozone increases in 2004, and 2004 was not necessarily a record year in all areas of the country.
 Results are based on analysis of hourly ozone data for 1970 through 2003 downloaded from EPA at http://www.epa.gov/ttn/airs/airsaqs/detaildata/downloadaqsdata.htm.
For a list of ozone action days, in EPA's Region 5, see http://www.epa.gov/region5/air/ozoneday/pastoad.html. Days highlighted in red are ozone action days (OAD) and clicking on the date gives a map showing where the alert occurred. The only OAD occurred in on August 3, 2004. According to EPA Region 5 staff, the OAD listed for May 1, 2004 is a system test and not an actual OAD.
 Data sources:
M. J. Pitzl, "Summer's Ozone Bad Only Once," Arizona Republic, October 2, 2004, p. 3B. Data for 2004 go through September 30th, except for California, which goes through October 5th. Houston sometimes has a few 8-hour ozone exceedances in October. To eliminate the chance for bias, I used data only for April through September in both 2003 and 2004 when comparing ozone levels in Texas.
 Even this may be an overestimate. ED counted only ozone exceedances that occurred between Memorial Day and Labor Day and then divided by 99 days to get their percentage. Because of time constraints, I've counted ozone exceedances occurring at any time of the year, but I still divide by 99 days. Thus, my numbers represent an upper limit on the percentage of summer days with ozone violations. The actual percentage will be lower in some cases.
Environmental Protection Agency, "National Ambient Air Quality Standards for Ozone: Proposed Decision," Federal Register, December 13, 1996, pp. 65715-65750
Elevated ozone occurs mainly during the summer, because ozone formation increases with increasing sunlight and temperature. Elevated PM occurs mainly in the summer in the east and in the rural west, and in the winter in urban areas of the west. For seasonal PM data see, for example, http://www.netl.doe.gov/publications/proceedings/ 02/PM25/1.1.2Tanner_p.pdf,
http://capita.wustl.edu/CAPITA/CapitaReports/PMFineAn/ PM_vs_Tran/finalreport/BltPhx_PMvsWnd_FinalReport.html, R. A. Eldred et al., "Composition of PM2.5 and PM10 Aerosols in the Improve Network," Journal of the Air & Waste Management Association, vol. 47 (1997), pp. 194-203.
For seasonal trend in asthma symptoms and hospitalization see, for example, Spokane Regional Health District, Asthma in Spokane County (Spokane, Washington: April 2002), http://www.srhd.org/information/pubs/pdf/factsheets/ AsthmaInSpokaneCounty.pdf, K. Tippy and N. Sonnenfeld, Asthma Status Report, Maine 2002 (Augusta, Maine: Maine Bureau of Health, November 25, 2002), J. F. Gent et al., "Association of Low-Level Ozone and Fine Particles with Respiratory Symptoms in Children with Asthma," Journal of the American Medical Association, vol. 290, no. 14 (2003), pp. 1859-1867, Texas Department of Health, Asthma Prevalence, Hospitalizations and Mortality - Texas, 1999-2001 (Austin: November 21, 2003), http://www.tdh.state.tx.us/cphpr/asthma/asthma.pdf, J. K. Stockman et al., California County Asthma Hospitalization Chart Book, Data from 1998-2000 (Sacramento: California Department of Health Services, September 2003), http://www.ehib.org/cma/papers/ Hosp_Cht_Book_2003.pdf, K. R. Wilcox and J. Hogan, An Analysis of Childhood Asthma Hospitalizations and Deaths in Michigan, 1989-1993 (Lansing, Michigan: Michigan Department of Community Health, undated), http://www.michigan.gov/documents/ Childhood_Asthma_6549_7.pdf, R. L. Wahl et al., "Trends in Asthma Mortality and Hospitalization in Detroit, Michigan, 1990-1996," A Public Health Response to Asthma, Atlanta, Centers for Disease Control, February 9-11 1999, Michigan Deparment of Community Health, Epidemiology of Asthma Fact Sheet (Lansing, Michigan: January 2000), http://www.michigan.gov/documents/Asthma_factsheet_6385_7.pdf, Canadian Institute for Health Information, Asthma Hospitalizations Declining, Still Number One Cause of Hospitalizations for Children (September 26, 2001), http://secure.cihi.ca/cihiweb/dispPage.jsp? cw_page=media_26sep2001_e.
 D. V. Cohn, "Particles as Well as Ozone Foul Region's Air; Lung Association Report Ranks Areas among Worst in U.S.," Washington Post, April 29, 2004, p. B1