Tag Archives: 2007 Staff Paper

EPA’s New Ozone Rule: Part 18

In our last post, we saw how EPA’s CASAC reacted strongly to its decision to make the secondary standard of ground-level ozone identical to the primary standard. That influenced EPA to reconsider its decision as reported in the document National Ambient Air Quality Standards for Ozone, Final Preamble, 2011 (p. 215):

In reconsidering the 2008 final rule in the 2010 proposal, the Administrator agreed with the conclusions drawn in the 2006 Criteria Document, 2007 Staff Paper and by CASAC that the scientific evidence available in the 2008 rulemaking continues to demonstrate the cumulative nature of O3 – induced plant effects and the need to give greater weight to higher concentrations. Thus, the Administrator concluded that a cumulative exposure index that differentially weights O3 concentrations represents a reasonable policy choice for a secondary standard to protect against the effects of O3 on vegetation during the growing season. The Administrator further agreed with both the 2007 Staff Paper and CASAC that the most appropriate cumulative, concentration-weighted form to consider is the sigmoidally weighted W126 form.

As EPA noted before, the amount of protection the primary standard would give to vegetation is uncertain, but the hint is that EPA is now prepared to err on the side of regulation. In this excerpt (p. 216), EPA argues that we can’t be sure that the primary standard can protect vegetation as well as the W126 standard. A comparison is hard to make because the results of such a comparison would likely differ from year to year, and because we don’t have enough data in the areas where the secondary standard might do the most good, in rural areas. (The paragraph sign [¶] indicates a paragraph break that I introduced that wasn’t there in the original text. The “8-hour average standard” is the primary standard, which averages ozone readings taken during an eight-hour period.):

The Administrator noted that… EPA proposed a second option of revising the then-current 8-hour average secondary standard by making it identical to the proposed 8-hour primary standard. The 2007 Staff Paper analyzed the degree of overlap expected between alternative 8-hour and cumulative seasonal secondary standards using recent air quality monitoring data. Based on the results, the 2007 Staff Paper concluded that the degree to which the current 8-hour standard form and level would overlap with areas of concern for vegetation expressed in terms of the 12-hour W126 standard is inconsistent from year to year and would depend greatly on the level of the 12-hour W126 and 8-hour standards selected and the distribution of hourly O3 concentrations within the annual and/or 3-year average period.

¶ The 2007 Staff Paper also recognized that meeting the then current or alternative levels of the 8-hour average standard could result in air quality improvements that would potentially benefit vegetation in some areas, but urged caution be used in evaluating the likely vegetation impacts associated with a given level of air quality expressed in terms of the 8-hour average form in the absence of parallel W126 information. This caution was due to the concern that the analysis in the 2007 Staff Paper may not be an accurate reflection of the true situation in non-monitored, rural counties due to the lack of more complete monitor coverage in many rural areas. Further, of the counties that did not show overlap between the two standard forms, most were located in rural/remote high elevation areas which have O3 air quality patterns that are typically different from those associated with urban and near urban sites at lower elevations. Because the majority of such areas are currently not monitored, there are likely to be additional areas that have similar air quality distributions that would lead to the same disconnect between forms. Thus, the 2007 Staff Paper concluded that it remains problematic to determine the appropriate level of protection for vegetation using an 8-hour average form. [emphasis mine — MHK]

Now here is the real rationale behind the secondary rule: cumulative exposure hurts plants more than it hurts humans. But why that should be? That question I can’t answer. The document continues (p. 217):

The Administrator also noted in the 2010 proposal that CASAC recognized that an important difference between the effects of acute exposures to O3 on human health and the effects of O3 exposures on welfare [of vegetation — MHK] is that vegetation effects are more dependent on the cumulative exposure to, and uptake of, O3 over the course of the entire growing season (Henderson, 2006c). The CASAC O3 Panel members were unanimous in concluding the protection of natural terrestrial ecosystems and managed agricultural crops requires a secondary O3 standard that is substantially different from the primary O3 standard in form, averaging time, and level (Henderson, 2007).

That concludes the EPA’s rationale in the document. Again, it seems to me that the decision was based on a judgement call. You may agree with me that there is less of a moral imperative to safeguard property and crops than there is safeguarding human life, so when evaluating the secondary standard, it makes even more sense to compare gains and losses. True, a secondary standard might improve agricultural crops, but is it worth the additional cost to industry to maintain that standard? That question is especially hard to answer when we don’t know exactly how much benefit the secondary standard would bring us above and beyond the primary standard. It’s a very tricky question. More about this in my final comments on the subject. In the meantime, let’s discuss how EPA standards are implemented.

EPA’ s New Ozone Rule: Part 15

A major innovation of EPA’s 2010 revision of the ozone standard was the introduction of what is called a secondary standard that is different from the primary standard. The secondary standard has existed before, but it was always set identical to the primary standard. To summarize the two standards:

  • The primary standard is intended to protect the public health. It is currently based on the fourth-highest 8-hour average ozone concentration reading in a year.
  • The secondary standard is meant to protect property, economic interests, and other concerns. It is based on a cumulative ozone concentration over time. Ozone readings are taken hourly between 8 a.m. and 7 p.m., adjusted by what is called the W126 rule, and then summed during a three-month period. Units are in ppm-hours. See what I wrote in this blog about the secondary standard in the post “EPA’s New Ozone Rule: Part 6.” To view, click here.

Now if one standard was consistently stricter than the other, the EPA could simply adopt the stricter standard. That it felt necessary to formulate two standards can only mean that in some places one standard will be harder to meet, and in other places the other standard will be the stricter. The EPA wants to meet both standards everywhere, a condition we Orthodox Jews call being machmir for both shitos.

What I don’t understand yet is why the primary standard, which is meant to safeguard public health, is based on a highest one-time average, whereas the secondary standard, meant to protect property, is based on a cumulative measure. A cumulative standard makes sense, because research shows that the extent of damage to plants caused by ozone depends on cumulative exposure. But perhaps damage to human health also depends on cumulative exposure, just as the damage caused by radiation to human health depends on cumulative exposure. Why not make the primary standard cumulative as well? Be that as it may, currently the primary standard remains based on a highest one-time average, while the secondary standard remains identical to the primary standard.

What I want to do in this post is quote EPA in its own words why it felt a new secondary standard was necessary, discussed in the document National Ambient Air Quality Standards for Ozone, Final Preamble, 2011.

From the outset, the EPA is clear that the secondary standard was formulated because of ozone’s effects on plants (p. 196):

…The 2006 Criteria Document concluded that O3 exposure indices that cumulate differentially weighted hourly concentrations are the best candidates for relating exposure to plant growth responses…

It is interesting that the EPA recognized the value of a secondary standard long before 2010 (p. 197):

At the conclusion of the 1997 review, the biological basis for a cumulative, seasonal form was not in dispute. There was general agreement between the EPA staff, CASAC, and the Administrator, based on their review of the air quality criteria, that a cumulative, seasonal form was more biologically relevant than the previous 1-hour and new 8-hour average forms (61 FR 65716).

The EPA also explained why, rather than summing up straight ozone concentrations, it chose to sum up modified values, referred to as the W126 form. Using W126 values gives more weight to higher concentrations and much less weight to lower concentrations that would exist either naturally without human activity, or from foreign sources beyond the control of the U.S. government (p. 198):

Regarding the first consideration, the 2007 Staff Paper noted that the W126 form, by its incorporation of a continuous sigmoidal weighting scheme, does not create an artificially imposed concentration threshold, yet also gives proportionally more weight to the higher and typically more biologically potent concentrations, as supported by the scientific evidence. Second, the index value is not significantly influenced by O3 concentrations within the range of estimated PRB [policy-relevant background, the level of ozone not caused by human activity in the U.S. — MHK], as the weights assigned to concentrations in this range are very small.

Nevertheless, the EPA retained a secondary standard identical to the primary standard until 2010. Initially, the EPA felt that if the primary standard was made more strict, it would be sufficient for the secondary standard were made identical to it. A separate secondary standard that was cumulative would provide no additional protection unless it was made very strict, which can’t be justified because our knowledge of the effects of low-level ozone on vegetation is so uncertain (the paragraph sign [¶] indicates a paragraph break that I inserted. P. 209):

In considering the appropriateness of establishing a new standard defined in terms of a cumulative, seasonal form, or revising the 1997 secondary standard by making it identical to the revised primary standard, … EPA first considered the 2007 Staff Paper analysis of the projected degree of overlap between counties with air quality expected to meet the revised 8-hour primary standard, set at a level of 0.075 ppm, and alternative levels of a W126 standard based on currently monitored air quality data. This analysis showed significant overlap between the revised 8-hour primary standard and selected levels of the W126 standard form being considered, with the degree of overlap between these alternative standards depending greatly on the W126 level selected and the distribution of hourly O3 concentrations within the annual and/or 3-year average period. On this basis, as an initial matter, EPA concluded that a secondary standard set identical to the proposed primary standard would provide a significant degree of additional protection for vegetation as compared to that provided by the then-current 0.084 ppm secondary standard.

¶ In further considering the significant uncertainties that remain in the available body of evidence of O3-related vegetation effects and in the exposure and risk analyses conducted for the 2008 rulemaking, and the difficulty in determining at what point various types of vegetation effects become adverse for sensitive vegetation and ecosystems, EPA focused its consideration on a level for an alternative W126 standard at the upper end of the proposed range (i.e., 21 ppm-hours). The 2007 Staff Paper analysis showed that at that W126 standard level, there would be essentially no counties with air quality that would be expected both to exceed such an alternative W126 standard and to meet the revised 8-hour primary standard – that is, based on this analysis of currently monitored counties, a W126 standard would be unlikely to provide additional protection in any monitored areas beyond that likely to be provided by the revised primary standard.

The EPA states again that with the lack of extensive monitoring in rural areas, it is unsure how much additional protection a separate secondary standard would provide. At this point, it decided to err on the side of less regulation. Note that the term “8 hour standard” refers to the primary standard, which averages readings over eight-hour periods (p. 210):

The EPA also recognized that the general lack of rural monitoring data made uncertain the degree to which the revised 8-hour standard or an alternative W126 standard would be protective in those areas, and that there would be the potential for not providing the appropriate degree of protection for vegetation in areas with air quality distributions that result in a high cumulative, seasonal exposure but do not result in high 8-hour average exposures. While this potential for under-protection using an 8- hour standard was clear, the number and size of areas at issue and the degree of risk was hard to determine. However, EPA concluded at that time that an 8-hour standard would also tend to avoid the potential for providing more protection than is necessary, a risk that EPA concluded would arise from moving to a new form for the secondary standard despite significant uncertainty in determining the degree of risk for any exposure level and the appropriate level of protection, as well as uncertainty in predicting exposure and risk patterns.

…EPA concluded at that time that establishing a new secondary standard with a cumulative, seasonal form would result in uncertain benefits beyond those afforded by the revised primary standard and therefore may be more than necessary to provide the requisite degree of protection.

Eventually, though, the EPA changed its mind. Why will be discussed in the next post.

EPA’s New Ozone Rule: Part 9

Exactly what was the EPA’s reasoning behind lowering the maximum ground-level ozone concentration from 75 ppb to 70 ppb? This is the opening paragraph of the discussion in EPA’s National Ambient Air Quality Standards for Ozone, 2011, page 35.

This section presents the rationale for the Administrator’s final decision that the O3 primary standard, which was set at a level of 0.075 ppm in the 2008 final rule, should instead be set at 0.070 ppm. In developing this rationale, the Administrator recognizes that the CAA [Clean Air Act — MHK] requires her to reach a public health policy judgment as to what standard would be requisite to protect public health with an adequate margin of safety, based on scientific evidence and technical assessments that have inherent uncertainties and limitations. This judgment requires making reasoned decisions as to what weight to place on various types of evidence and assessments, and on the related uncertainties and limitations. Thus, in selecting a final level, the Administrator is seeking not only to prevent O3 levels that have been demonstrated to be harmful but also to prevent lower O3 levels that may pose an unacceptable risk of harm, even if the risk is not precisely identified as to nature or degree.

What the EPA is saying is that it isn’t enough for the maximum concentration of ground-level ozone allowable to be set just below the minimum known to cause harm. Rather, the limit must be low enough so that even if the harm is not certain but only possible, the risk of harm is low enough to be acceptable. Question is, how low must the risk be to considered acceptable? The document itself states that risk must be taken into consideration even when it can’t be precisely identified. But does that mean that any level of risk, no matter how low, is unacceptable? That would be setting a very high standard indeed. And if that is not so, what is the maximum level of risk that is acceptable? What is the cutoff point?

Unspoken is the realization that it is politically unwise to try to impose tougher rules on the public than is necessary to achieve the objective. To do so is to impose unnecessary economic hardship that could provoke a backlash. And indeed, we saw that backlash in September 2011. The EPA can’t admit that fact, but it is nevertheless true.

There have been a number of controlled studies examining human exposure to ozone, but most have been at the 80 ppb level1. However, studies by William C. Adams, researcher (now retired) at the University of California at Davis did expose humans to ozone at average concentrations as low as 40 ppb2. Besides exposing his subjects to steady concentrations, Adams attempted to mimic the natural environment by slowly increasing and then decreasing the ozone concentration, much as the ambient ozone concentration grows in the morning, peaks in midday, and then declines toward evening. Adams found no statistically significant difference in lung function compared to breathing filtered (ozone-free) air at the 40 ppb and 60 ppb levels. However, a later analysis of Adam’s data by the EPA did find a small statistical difference at the 60 ppb level3. EPA finds this of concern, because a small statistical drop of lung function among healthy adults could manifest itself much more forcefully among those with lung disease4.

Still, most controlled studies on ozone exposure do not test beneath the 80 ppb level. Yet the EPA notes that there is no evidence that the harmful effects of ozone stop at the 80 ppb level (start with a very high concentration of ozone and slowly lower it. The concentration level where harmful effects would stop is known as the threshold). In fact, it can be inferred that such effects extend well below that level, because of the variability of responses of the test subjects5. I believe this means that if 80 ppb was the threshold level, then if you exposed test subjects to that concentration, you would see a number of small responses, but they would all be roughly equal to each other. If, on the other hand, some test subjects experience effects much more than others, even though the effects are still small, that indicates that the effects occur well below the 80 ppb level. And small effects for healthy people can mean big effects for those with respiratory disease.

The above concerned controlled studies of subjects of laboratory experiments. EPA also looked at epidemiological studies, studies of what is happening to populations in their day-to-day lives6. Some found thresholds between 25 and 50 ppb. Other studies never found a threshold because the damage that ozone inflicted seemed linear with the concentration. As I understand this, this means that if the concentration was reduced by a specific percentage (for example, a 20% reduction), measureable effects are reduced by the same percentage multiplied by fixed factor (say a 2% reduction in concentration results in a 3% decrease in effects, a 4% reduction results in a 6% decrease in effects, and so on). On the other hand, you might expect that at a concentration near the threshold level, a further reduction would result in a greater decrease of effects (say a 2% reduction results in a 3% in effects, but a 3% reduction results in a 25% decrease in effects, and a 4% reduction results in a 95% decrease in    effects)6. These studies never saw this sort of effect, so they could not conclude there was any threshold for ozone.

The EPA also looked at studies that did subset analysis looking only at days whose ozone concentration did not exceed certain ozone concentrations (such as 80 ppb and 61 ppb), and still found associations between those concentrations and lung function decrements)6.

Regarding the existence of a threshold for the effects of ozone, the EPA concluded:

Based on the above considerations, the 2007 Staff Paper recognized that the available evidence neither supports nor refutes the existence of effect thresholds at the population level for morbidity and mortality effects, and that if a population threshold level does exist, it would likely be well below the level of the then current standard and possibly within the range of background levels. Taken together, these considerations also support the conclusion that if a population threshold level does exist, it would likely be well below the level of the 0.075 ppm, 8-hour average, standard set in 2008.7

But if the EPA needed to pick the lowest allowable concentration, should it have chosen the lowest threshold found by the studies, 25 ppb? That would not be possible, because the background level of ground-level ozone (the concentration of ozone in the U.S. that is either naturally occurring or coming from outside the U.S. and over which the U.S. government has no control. The background level varies with location and season8.) is often above that level of 25 ppb9. This being the case, setting the standard at 25 ppb would have been an impossible demand. (In fact, the 2007 Staff Paper found that below 35 ppb, it was difficult to tell effects from ozone from effects from other air pollutants9.) Even 50 ppb would be an extremely difficult and expensive goal to meet.


  1. U.S. Environmental Protection Agency, National Ambient Air Quality Standards for Ozone, Final Preamble, 2011, p.38
  2. Adams,W.C., Comparison of chamber 6.6-h exposures to 0.04-0.08 ppm ozone via square-wave and triangular profiles on pulmonary responses Inhalation Toxicology vol. 18: pp. 127-136. For the abstract, click here.
  3. U.S. Environmental Protection Agency, National Ambient Air Quality Standards for Ozone, Final Preamble, 2011, p.38
  4. ibid.p.39
  5. ibid.p.40
  6. ibid.p.42
  7. ibid.p.43
  8. For a detailed discussion of background ozone levels, see U.S. Environmental Protection Agency, Integrated Science Assessment for Ozone and Related Photochemical Oxidants, Third External Review Draft, June 2012, Section 3.4, “Background Ozone Concentrations”, p.3-32ff.
  9. U.S. Environmental Protection Agency, National Ambient Air Quality Standards for Ozone, Final Preamble, 2011, p.42, p.107