01-27820. National Ambient Air Quality Standards for Ozone: Proposed Response To Remand

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Proposed response to remand.

SUMMARY:

On July 18, 1997, in accordance with sections 108 and 109 of the Clean Air Act (Act), EPA completed its review of the national ambient air quality standards (NAAQS) for ozone (O₃) by promulgating revised primary and secondary standards (62 FR 38856; henceforth, “1997 final rule”). On May 14, 1999, the United States Court of Appeals for the District of Columbia Circuit (D.C. Circuit) remanded the O₃ NAAQS to EPA to consider, among other things, the alleged beneficial health effects of O₃ pollution in shielding the public from the “harmful effects of the sun's ultraviolet rays.” 175 F. 3d 1027 (D.C. Cir. 1999). Today's action provides EPA's proposed response to that aspect of the court's remand. As explained more fully below, based on its review of the air quality criteria and NAAQS for O₃ completed in 1997, and its additional assessment of the potential beneficial effects of tropospheric O₃, EPA has provisionally determined that the information linking changes in patterns of ground-level O₃ concentrations likely to occur as a result of programs implemented to attain the 1997 O₃ NAAQS to changes in relevant exposures to UV-B radiation of concern to public health is too uncertain at this time to warrant any relaxation in the level of public health protection previously determined to be requisite to protect against the demonstrated direct adverse respiratory effects of exposure to O₃ in the ambient air. Further, the Administrator notes that it is the Agency's view that associated changes in UV-B radiation exposures of concern, using plausible but highly uncertain assumptions about likely changes in patterns of ground-level ozone concentrations, would likely be very small from a public health perspective. As a result, the revised O₃ NAAQS will remain set at a level of 0.08 parts per million (ppm), with a form based on the 3-year average of the annual fourth-highest daily maximum 8-hour average O₃ concentrations measured at each monitor within an area. The primary standard provides increased protection to the public, especially children and other at-risk populations, against a wide range of health effects directly induced by breathing O₃ in the ambient air, including decreased lung function (primarily in children active outdoors), increased respiratory symptoms (particularly in highly sensitive individuals), hospital admissions and emergency room visits for respiratory causes (among children and adults with pre-existing respiratory disease such as asthma), inflammation of the lung, and possible long-term damage to the lungs. The secondary standard provides increased protection to the public welfare against effects on vegetation, such as agricultural crop loss, damage to forests and ecosystems, and visible foliar injury to sensitive species associated with direct exposure to O₃ in the ambient air. Today's action constitutes EPA's proposed response to the part of the remand of the 1997 O₃ NAAQS by the D.C. Circuit related to whether tropospheric O₃ has a beneficial effect with regard to attenuation of naturally occurring solar radiation. Other issues related to the 1997 O₃ NAAQS are now before the D.C. Circuit for proceedings consistent with the February 27, 2001 opinion of the United States Supreme Court in this case, Whitman v. American Trucking Associations, 531 U.S. 457 (2001), and are not addressed by today's action.

DATES:

Comments on this proposed response must be received by January 14, 2002.

ADDRESSES:

Submit written comments (in duplicate if possible) on this proposed response to: Air and Radiation Docket and Information Center (6102), Attn: Docket No. A-95-58, U.S. Environmental Protection Agency, 1200 Pennsylvania Ave., NW., Washington, DC 20460. Electronic comments are encouraged and can be sent directly to EPA at: [e-mail]. Comments will also be accepted on disks in WordPerfect in 8.0/9.0 file format. All comments in electronic form must be identified by the docket number, Docket No. A-95-58.

FOR FURTHER INFORMATION CONTACT:

Susan Lyon Stone, Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency (C539-01), Research Triangle Park, NC 27711; e-mail [e-mail]; telephone [phone].

SUPPLEMENTARY INFORMATION:

Docket

A docket containing information relating to EPA's review of the O₃ primary and secondary standards (Docket No. A-95-58) is available for public inspection at the EPA's Air and Radiation Docket and Information Center, 401 M Street, SW., Washington, DC 20460 in room M-1500, Waterside Mall (ground floor). This docket incorporates the docket from the previous review of the O₃ standards (Docket No. A-92-17) and the docket established for the air quality criteria document (Docket No. ECAO-CD-92-0786). The docket may be inspected between 8 a.m. and 5:30 p.m. on weekdays, excluding legal holidays. A reasonable fee may be charged for copying.

Availability of Related Information

Certain documents are available from the U.S. Department of Commerce, National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161. Available documents include:

(1) The Review of the National Ambient Air Quality Standards for Ozone: Assessment of Scientific and Technical Information (“Staff Paper”) (EPA-452/R-96-007, June 1996, NTIS #PB-96-203435; $67.00 paper copy and $21.50 microfiche). (Add a $3.00 handling charge per order.)

(2) Air Quality Criteria for Ozone and Other Photochemical Oxidants (“Criteria Document”) (three volumes, EPA/600/P-93-004aF through EPA/600/P-93-004cF, July 1996, NTIS #PB-96-185574; $169.50 paper copy and $58.00 microfiche).

A limited number of copies of other documents generated in connection with the review of the standard, such as documents pertaining to human exposure and health risk assessments and the relationships between ground-level O₃, ultraviolet-B (UV-B) radiation, and health effects, can be obtained from: U.S. Environmental Protection Agency Library (MD-35), Research Triangle Park, NC 27711; telephone [phone]. These and other related documents are also available for inspection and copying in the EPA docket.

Electronic Availability

The Staff Paper and documents pertaining to human health risk and exposure assessments are available on the Office of Air and Radiation, Policy and Guidance Web site at: http://www.epa.gov/​ttn/​oarpg/​t1sp.html. The O₃ NAAQS 1996 proposal and 1997 final rule are available at the same Web site, at: http://www.epa.gov/​ttn/​oarpg/​t1pfpr.html.

Children's Environmental Health

This proposed response to the court's remand, reaffirming the 1997 8-hour O₃NAAQS, specifically takes into account children as the group most at risk to the direct inhalation-related effects of O₃ exposure, and was based on studies of effects on children's health (U.S. EPA, 1996a; U.S. EPA, 1996b) and assessments of children's exposure and risk (Johnson et al., 1994; Johnson et al., 1996a,b; Whitfield et al., 1996; Richmond, 1997). The 8-hour O₃ primary standard protects children's health with an adequate margin of safety from the direct adverse effects associated with inhalation exposures to ground-level O₃, after considering potential indirect beneficial effects of ground-level O₃ related to its attenuation of UV-B radiation and resultant adverse health effects. The public is invited to submit or identify peer-reviewed studies and data, of which EPA may not be aware, that assess results of early life exposure to the direct effects of breathing ground-level O₃ or to changes in UV-B radiation, and associated health effects, that may result from changes in ground-level O₃.

Implementation Activities

When the 8-hour primary and secondary O₃ standards are implemented by the States, utility, automobile, petroleum, and chemical industries are likely to be affected, as well as other manufacturing concerns that emit volatile organic compounds (VOC) or nitrogen oxides (NO_X). The extent of such effects will depend on implementation policies and control strategies adopted by States to assure attainment and maintenance of the standards.

The EPA will develop appropriate policies and control strategies to assist States in the implementation of the 8-hour primary and secondary O₃ NAAQS. The resulting implementation strategies will then be published for public comment in the future.

Table of Contents

The following topics are discussed in today's preamble:

I. Background

A. 1997 Revision of the O₃ NAAQS

1. Legislative Requirements

2. Review of Air Quality Criteria and Standards for O₃

B. Ozone NAAQS Litigation and Remand

1. Litigation Summary

2. Remand on Health Benefits Issue

C. Atmospheric Distribution of O₃ and UV-B Radiation

D. Related Stratospheric O₃ Program

II. Rationale for Proposed Response to Remand on the Primary O₃ Standard

A. Direct Adverse Health Effects from Breathing O₃ in the Ambient Air

1. Health Effects Associated with O₃ Inhalation Exposures

2. Human Exposure and Risk Assessments

B. Potential Indirect Beneficial Health Effects Associated with Ground-level O₃

1. Health Effects Associated with UV-B Radiation Exposure

2. Relationship Between Ground-level O₃ and UV-B Radiation Exposure

3. Evaluation of UV-B Radiation-related Risk Estimates for Ground-level O₃ Changes

C. Consideration of Net Adverse Health Effects of Ground-level O₃

D. Proposed Response to Remand on the Primary O₃ NAAQS

III. Rationale for Proposed Response to Remand on the Secondary O₃ Standard

A. Direct Adverse Welfare Effects

B. Potential Indirect Beneficial Welfare Effects

C. Proposed Response to Remand on the Secondary O₃ NAAQS

IV. Administrative Requirements

A. Executive Order 12866: OMB Review of “Significant Actions'

B. Executive Order 13045: Children's Health

C. Executive Order 13132: Federalism

D. Executive Order 13175: Consultation and Coordination with Indian Tribal Governments

E. Unfunded Mandates Reform Act

F. Regulatory Flexibility Analysis/Small Business Regulatory Enforcement Fairness Act

G. Paperwork Reduction Act

H. National Technology Transfer and Advancement Act

I. Executive Order 13211:Energy Effects

V. References

I. Background

A. 1997 Revision of the O₃ NAAQS

On July 18, 1997, in accordance with sections 108 and 109 of the Act, EPA completed its review of the NAAQS for O₃ by promulgating revised primary and secondary standards (“1997 final rule”). These standards were based on EPA's review of the available scientific evidence linking direct exposures to ambient O₃ to adverse health and welfare effects at levels allowed by the then current O₃ standards. The revised primary and secondary standards were each set at a level of 0.08 ppm, with an 8-hour averaging time and a form based on the 3-year average of the annual fourth-highest daily maximum 8-hour average O₃ concentrations measured at each monitor within an area. ^[1] The new primary standard was established to provide increased protection to the public, especially children and other at-risk populations, against a wide range of O₃-induced respiratory health effects due to inhalation exposures, including decreased lung function, primarily in children active outdoors; increased respiratory symptoms, particularly in highly sensitive individuals; hospital admissions and emergency room visits for respiratory causes, among children and adults with pre-existing respiratory disease such as asthma; inflammation of the lung; and possible long-term damage to the lungs. The new secondary standard was established to provide increased protection to the public welfare against direct O₃-induced effects on vegetation, such as agricultural crop loss, damage to forests and ecosystems, and visible foliar injury to sensitive species.

1. Legislative Requirements

Two sections of the Act govern the establishment, review, and revision of NAAQS. Section 108 (42 U.S.C. 7408) directs the Administrator to identify certain pollutants which “may reasonably be anticipated to endanger public health or welfare” and to issue air quality criteria for them. These air quality criteria are to “accurately reflect the latest scientific knowledge useful in indicating the kind and extent of all identifiable effects on public health or welfare which may be expected from the presence of [a] pollutant in the ambient air * * *.”

Section 109 (42 U.S.C. 7409) directs the Administrator to propose and promulgate “primary” and “secondary” NAAQS for pollutants identified under section 108. Section 109(b)(1) defines a primary standard as one “the attainment and maintenance of which, in the judgment of the Administrator, based on [the] criteria and allowing an adequate margin of safety, are requisite to protect the public health.” A secondary standard, as defined in section 109(b)(2), must “specify a level of air quality the attainment and maintenance of which in the judgment of the Administrator, based on [the] criteria, [are] requisite to protect the public welfare from any known or anticipated adverse effects associated with the presence of [the] pollutant in the ambient air.” ^[2]

Section 109(d)(1) of the Act requires periodic review and, if appropriate, revision of existing air quality criteria and NAAQS. Section 109(d)(2) requires appointment of an independent scientific review committee to review criteria and standards and recommend new standards or revisions of existing criteria and standards, as appropriate. The committee established under section 109(d)(2) is known as the Clean Air Scientific Advisory Committee (CASAC), a standing committee of EPA's Science Advisory Board.

2. Review of Air Quality Criteria and Standards for O₃

An overview of the last review of the O₃ air quality criteria and standards is presented in section I.C of the preamble to the 1997 final rule. In summary, the 1997 review was initiated in August 1992 with the development of a revised Air Quality Criteria Document for Ozone and Other Photochemical Oxidants (henceforth, the “Criteria Document”). Multiple drafts of the Criteria Document were reviewed by CASAC and the public, resulting in a final Criteria Document (U.S. EPA, 1996a) that reflected CASAC and public comments.^[3] The EPA also prepared a staff paper, Review of National Ambient Air Quality Standards for Ozone: Assessment of Scientific and Technical Information (henceforth, the “Staff Paper”).^[4] Multiple drafts of the Staff Paper were also reviewed by CASAC and the public, resulting in a final Staff Paper (U.S. EPA, 1996b) that reflected CASAC and public comments.^[5]

On November 27, 1996 EPA announced its proposed decision to revise the NAAQS for O₃ (61 FR 65716, December 13, 1996; henceforth, “1996 proposal”), as well as its proposed decision to revise the NAAQS for particulate matter (PM). To ensure the broadest possible public input on these proposals, EPA took extensive and unprecedented steps to facilitate the public comment process, including the establishment of a national toll-free telephone hotline and provisions for electronic submission of comments. The EPA also held several public hearings, participated in numerous meetings across the country, and held two national satellite telecasts to provide direct opportunities for public comment and to disseminate information to the public about the proposed standard revisions. As a result of this intensive effort to solicit public input, over 50,000 comments were received on the proposed revisions to the O₃ NAAQS by the close of the public comment period on March 12, 1997.

The final rule, published on July 18, 1997, presented EPA's rationale for its final decision, and addressed the major issues raised in comments on the 1996 proposal. A comprehensive summary of all significant comments, along with EPA's response to such comments (U.S. EPA, 1997; henceforth, “Response to Comments”), can be found in the docket for the 1997 rulemaking (Docket No. A-95-58 ^[6] ). The 1997 final rule presented EPA's decision to replace the existing 1-hour primary and secondary standards ^[7] (each set at a level of 0.12 ppm, with a 1-expected-exceedance form, averaged over 3 years ^[8] ) with 8-hour standards, each set at a level of 0.08 ppm, with a form based on the 3-year average of the annual fourth-highest daily maximum 8-hour average O₃ concentrations measured at each monitor within an area (as determined by 40 CFR part 50, appendix I).

B. Ozone NAAQS Litigation and Remand

1. Litigation Summary

Following promulgation of the revised 8-hour O₃ NAAQS, numerous petitions for review of the standards were filed in the D.C. Circuit. American Trucking Associations v. EPA, No. 97-1441 (ATA). Oral argument was held on December 17, 1998 and the Court of Appeals rendered its opinion on May 14, 1999. American Trucking Associations v. EPA, 175 F. 3d 1027 (D.C. Cir. 1999). A divided panel found that section 109 of the Act, 42 U.S.C. 7409, as interpreted by EPA in setting the revised O₃ (and PM) NAAQS, effected an unconstitutional delegation of legislative authority. Id. at 1033-1040. The court remanded the O₃ standards with instructions that EPA should articulate an “intelligible principle” for determining the degree of residual risk to public health permissible in setting revised NAAQS. Id. In addition, the court also directed that, in responding to the remand, EPA should consider the alleged beneficial health effects of O₃ pollution in shielding the public from the “harmful effects of the sun's ultraviolet rays.” Id. at 1051-1053.

In 1999, EPA petitioned the Court of Appeals for rehearing en banc on a number of aspects of the court's decision in the ATA case. Although the petition for rehearing was granted in part and denied in part, the court declined to review its ruling with regard to the potential beneficial effects of O₃ pollution. American Trucking Associations v. EPA, 195 F. 3d 4, 10 (D.C. Cir. 1999). The court did note, however, that it “expressed[ed] no opinion, of course, upon the effect, if any, that studies showing the beneficial effects of tropospheric ozone * * * might have upon any ozone standards * * *” Id. On January 27, 2000, EPA petitioned the Supreme Court for certiorari on the constitutional issue and two other issues, but did not request review of the Court of Appeals ruling regarding the alleged beneficial health effects of O₃. The EPA's petition for certiorari was granted on May 22, 2000; oral argument was subsequently held on November 7, 2000; and an opinion was issued on February 27, 2001. Whitman v. American Trucking Associations, 531 U.S.457 (2001). The U.S. Supreme Court reversed the judgment of the D.C. Circuit on the constitutional issue, holding that section 109 of the Act does not delegate legislative power to the EPA in contravention of the Constitution, and remanded the case to the D.C. Circuit for proceedings consistent with its opinion. Since EPA did not seek Supreme Court review of the Court of Appeals' decision relating to potential beneficial health effects of O₃, EPA is moving forward to address that aspect of the lower court's remand independently.

2. Remand on Health Benefits Issue

The Court of Appeals' ruling concludes that “EPA cannot ignore the possible health benefits of ozone.” ^[9] American Trucking Associations v. EPA, 175 F. 3d 1027, 1033 (D.C. Cir. 1999). According to the court “[p]etitioners presented evidence that, according to them, shows the health benefits of tropospheric ozone as a shield from the harmful effects of the sun's ultraviolet rays—including cataracts and both melanoma and non-melanoma skin cancer.” Id. at 1051. In rejecting EPA's interpretation of the Act that it need not consider alleged indirect beneficial effects of tropospheric O₃ in shielding the public from potentially harmful, but naturally occurring, UV-B radiation from the sun, the court concluded that “legally * * * EPA must consider the positive identifiable effects of a pollutant's presence in the ambient air in formulating air quality criteria under section 108 and NAAQS under section 109.” Id. at 1052. As a result, the court directed EPA to “determine whether * * * tropospheric ozone has a beneficent effect and, if so, then to assess ozone's net adverse health effect.” Id. at 1053. Today's action sets forth EPA's proposed response in that regard.

C. Atmospheric Distribution of O₃ and UV-B Radiation

The focus of the 1997 review of the air quality criteria and standards for O₃ and related photochemical oxidants was on public health and welfare effects associated with direct exposure to ambient levels of O₃ in the lower troposphere, essentially at ground level. People are directly exposed to ground-level O₃ simply by breathing ambient air; similarly, plants are directly exposed through their respiratory processes. Ground-level O₃ is not emitted directly from mobile or stationary sources but, like other photochemical oxidants, commonly exists in the ambient air as an atmospheric transformation product. Ground-level O₃ formation is the result of chemical reactions of VOC, NO_X, and oxygen in the presence of sunlight and generally at elevated temperatures. As a principal ingredient in photochemical smog, elevated episodic concentrations of ground-level O₃ typically occur in the summertime. High concentrations may be found in and downwind of major urban centers as well as across broad regions of elevated precursor emissions. A detailed discussion of atmospheric formation, ambient concentrations, and health and welfare effects associated with direct exposure to O₃ can be found in the Criteria Document and Staff Paper.

Naturally occurring O₃ is found in two sections of the earth's atmosphere, the stratosphere and the troposphere. The demarcation between these two layers varies between about 8 and 18 kilometers (km) above the earth's surface. As illustrated in Figure 1, depicting the vertical profile of O₃, most naturally occurring O₃ (> 90 percent) resides in the stratosphere, with the remaining O₃ (< 10 percent) in the troposphere. The band of O₃ between about 15 and 30 km is commonly known as the “ozone layer.”

Man-made air pollution has significantly perturbed the natural distribution of O₃ in both layers. It is now widely accepted that emissions of long-lived chlorofluorocarbons (CFCs) and other compounds can deplete the natural O₃ layer in the

stratosphere. And, as summarized above, much shorter lived emissions of VOC and NO_X can markedly increase “smog” O₃ in the lowest portion of the troposphere, which is termed the planetary boundary layer. This fluctuating planetary boundary or “mixing” layer of the troposphere can extend as high as 1 to 3 km above the ground. Assuming a fairly high summertime O₃ pollution reservoir of 65 parts per billion (ppb) in a typical 1 km mixing layer, Cupitt (1994) estimated that pollution would add less than 1 percent to the expected total vertical profile of tropospheric and stratospheric O₃ (i.e., “total column” O₃) that would occur in the natural environment.

Ozone at ground level and throughout the troposphere is chemically identical to stratospheric O₃. Stratospheric O₃ occurs far too high to present any threat of direct respiratory-related adverse effects to people or plants from ambient ground-level exposures, but is known to provide a natural protective shield from excess radiation from the sun by absorbing UV-B radiation ^[10] before it penetrates to ground level. Recognizing that exposure to UV-B radiation has been associated with adverse health and welfare effects, EPA and international scientific, regulatory, and legislative organizations have for some time focused on understanding the effects of UV-B radiation and on controlling the man-made pollution that is causing the depletion of the O₃ layer in the stratosphere, as discussed in section I.D below.^[11]

During the 1997 review, EPA recognized that tropospheric O₃ also absorbs UV-B radiation (U.S. EPA, 1996a, p. 5-79), such that ground-level O₃ formed by man-made pollution has the potential to provide some degree of additional shielding beyond the natural levels that would otherwise occur in the absence of man-made pollution. The relationship between ground-level O₃ and UV-B radiation, as well as the health effects associated with exposure to UV-B radiation and consideration of the UV-B radiation-related health risks associated with changes in ground-level O₃ are discussed in section II.B below. In response to the remand on the health benefits issue, EPA's assessment of the net adverse health effects of ground-level O₃ is discussed in section II.C below, as a basis for today's proposed decision on the primary O₃ NAAQS, summarized in section II.D below.

D. Related Stratospheric O₃ Program

In the 1970s, scientists first grew concerned that certain chemicals could damage the earth's protective stratospheric O₃ layer, and these concerns were validated by the discovery of thinning of the O₃ layer over Antarctica in the southern hemisphere. Because of the risks posed by stratospheric O₃ depletion and the global nature of the problem, leaders from many countries decided to work together to craft a workable solution. Since 1987, over 175 nations have signed a landmark environmental treaty, the Montreal Protocol on Substances that Deplete the Ozone Layer. The Protocol's chief aim is to reduce and eventually eliminate the production and use of man-made O₃ depleting substances, such as CFCs. By agreeing to the terms of the Montreal Protocol, signatory nations ratifying the Protocol—including the United States—commit to take actions to protect the stratospheric O₃ layer and to reverse the damage due to the use of O₃ depleting substances.

In 1990, Congress amended the Act by adding title VI (sections 601-618) to address the issue of stratospheric O₃ depletion.^[12] Most importantly, the amended Act required the gradual end to the production of certain chemicals that deplete the O₃ layer.^[13] In addition, the Act requires EPA to develop and implement regulations for the responsible management of O₃ depleting substances in the United States. The EPA has developed several regulatory programs under these authorities that include: ending the production and import of O₃ depleting substances (57 FR 33754, July 30, 1992) and identifying safe and effective alternatives (59 FR 13044, March 18, 1994), ensuring that refrigerants and halon fire extinguishing agents are recycled properly (58 FR 28660, May 14, 1993), banning the release of O₃ depleting refrigerants during the service, maintenance, and disposal of air conditioners and other refrigeration equipment (60 FR 40420, August 8, 1995), and requiring that manufacturers label products either containing or made with the most harmful O₃ depleting substances (58 FR 8136, February 11, 1993). Because of their relatively high O₃ depletion potential, several man-made compounds, including CFCs, carbon tetrachloride, methyl chloroform, and halons were targeted first for phaseout. The EPA continues to develop additional regulations for the protection of public health and the environment from effects associated with the depletion of the stratospheric O₃ layer.

Besides implementing and enforcing stratospheric O³ protection regulations in the U.S., EPA continues to work with other U.S. government agencies and international governments to pursue ongoing changes to the Montreal Protocol and other treaties. These refinements to the Protocol and other treaties are based on ongoing scientific assessments of O₃ depletion that are coordinated by the United Nations Environment Programme (UNEP) and the World Meteorological Organization (WMO), with cooperation from EPA and other agencies around the globe (UNEP, 1998; and WMO, 1998).

In addition to these regulatory and scientific activities, EPA maintains several education and outreach projects to help protect the American public from the health effects of overexposure to ultraviolet (UV) radiation. Chief among these projects is the UV Index, a tool that provides a daily forecast of the next day’s likely UV levels across the United States. ^[14] The UV Index, which EPA launched in partnership with the National Weather Service, serves as the cornerstone of EPA's SunWise School Program, the goal of which is to educate young children and their caregivers about the health effects of overexposure to the sun, as well as simple steps that people can take to avoid overexposure. ^[15]

II. Rationale for Proposed Response To Remand on the Primary O₃ Standard

Today's action presents the Administrator's proposed response to the remand, reaffirming the 8-hour O₃ primary standard promulgated in 1997, based on: (1) Information from the 1997 criteria and standards review that served as the basis for the 1997 primary O₃ standard, including the scientific information on health effects associated with direct inhalation exposures to O₃ in the ambient air, consideration of the adversity of such effects for individuals, and human exposure and risk assessments (section II.A below); (2) a review of the scientific information in the record of the 1997 review (but not considered as part of the basis for the 1997 standard) on the health effects associated with changes in UV-B radiation, the association between changes in ground-level O₃ and changes in UV-B radiation, and predictions of changes in ground-level O₃ levels likely to result from attainment of alternative O₃ standards ^[16] (section II.B below); and (3) consideration of the net adverse effects of ground-level O₃, taking into account both direct adverse inhalation-related health effects and the potential for indirect beneficial health effects associated with the shielding of UV-B radiation by ground-level O₃ (section II.C below).

A. Direct Adverse Health Effects From Breathing O₃ in the Ambient Air

This section briefly summarizes information on the direct adverse health effects from breathing O₃ in the ambient air, information as to when those effects become adverse to individuals, and insights gained from human exposure and risk assessments intended to provide a broader perspective for judgments about protecting public health from the risks associated with direct O₃ inhalation exposures.^[17]

1. Health Effects Associated With O₃ Inhalation Exposures

Based on information from human clinical, epidemiological, and animal toxicological studies, an array of health effects has been attributed to short-term (1 to 3 hours), prolonged (6 to 8 hours), and long-term (months to years) exposures to O₃. Long-established acute health effects ^[18] induced by short-term exposures to O₃, generally while individuals were engaged in heavy exertion, include transient pulmonary function responses, transient respiratory symptoms, and effects on exercise performance.^[19] The 1997 review included substantial new information on similar effects associated with prolonged exposures at concentrations as low as 0.08 ppm and at moderate levels of exertion. Other health effects associated with short-term or prolonged O₃ exposures include increased airway responsiveness, susceptibility to respiratory infection, increased hospital admissions and emergency room visits, and transient pulmonary inflammation. The 1997 review also included new information on chronic health effects ^[20] associated with long-term exposures. This array of effects is briefly summarized below, followed by considerations as to when these physiological effects could become medically significant such that they should be regarded as adverse to the health of individuals experiencing them.

a. Effects of Short-Term and Prolonged O₃ Exposures

(i) Pulmonary function responses. Transient reductions in pulmonary function have been observed in healthy individuals and those with impaired respiratory systems (e.g., asthmatic individuals) as a result of both short-term and prolonged exposures to O₃. The strongest and most quantifiable exposure-response information on such responses has come from controlled human exposure studies, which clearly show that reductions in lung function are enhanced by increased levels of activity involving exertion and by increased O₃ concentrations. Numerous such studies of exercising adults have demonstrated decrements in lung function both for exposures of 1-3 hours at ≥0.12 ppm O₃ and for exposures of 6.6 hours at ≥0.08 ppm O₃, providing conclusive evidence that O₃ levels commonly monitored in the ambient air induce lung function decrements in exercising adults. Further, numerous summer camp studies provide an extensive and reliable data base on comparable lung function responses to ambient O₃ and other pollutants in children and adolescents. The extent of pulmonary function decrements varies considerably among individuals, pulmonary function generally tends to return to baseline levels shortly after short-term exposure, and effects are typically attenuated upon repeated short-term exposures over several days.

(ii) Respiratory symptoms and effects on exercise performance. Various transient respiratory symptoms, including cough, throat irritation, chest pain on deep inspiration, and shortness of breath, have been induced by O₃ exposures of both healthy individuals and those with impaired respiratory systems. Increasing O₃ exposure durations and levels have been shown to elicit increasingly more severe symptoms that persist for longer periods in increasingly larger numbers of individuals. Symptomatic and pulmonary function responses follow a similar time course during an acute exposure and the subsequent recovery, as well as over the course of several days during repeated exposures. As with pulmonary function responses, the severity of symptomatic responses varies considerably among subjects. For some outdoor workers or active people who are highly responsive to ambient O₃, respiratory symptoms may cause reduced productivity, may curb the ability or desire to engage in normal activities, and may interfere with maximal exercise performance.

(iii) Increased airway responsiveness. Increased airway responsiveness is an indication that the airways are predisposed to bronchoconstriction, with a high level of bronchial responsiveness being characteristic of asthma. As a result of increased airway responsiveness induced by O₃ exposure, human airways may be more susceptible to a variety of stimuli, including antigens, chemicals, and particles. For example, healthy subjects after being exposed to O₃ concentrations as low as 0.20 ppm for 1 hour and 0.08 ppm for 6.6 hours have experienced small increases in nonspecific bronchial responsiveness, which usually resolve within 24 hours. Because enhanced response to antigens in asthmatics could lead to increased morbidity (i.e., medical treatment, emergency room visits, hospital admissions) or to more persistent alterations in airway responsiveness, these health endpoints raise concern for public health, particularly for individuals with impaired respiratory systems.

(iv) Increased susceptibility to respiratory infection. When functioning normally, the human respiratory tract, like that of other mammals, has numerous closely integrated defense mechanisms that provide protection from the adverse effects of a wide variety of inhaled particles and microbes. Evidence that inhalation of O₃ may break down or impair these defense mechanisms comes primarily from a very large number of laboratory animal studies with generally consistent results. One of the few studies of moderately exercising human subjects exposed to 0.08 ppm O₃ for 6.6 hours reported decrements in alveolar macrophage function, the first line of defense against inhaled microorganisms and particles in the lower airways and air sacs. While no single experimental human study or group of animal studies conclusively demonstrates that human susceptibility to respiratory infection is increased by exposure to O₃, taken as a whole, the data suggest that acute O₃ exposures can impair the host defense capability of both humans and animals, potentially resulting in a predisposition to bacterial infections in the lower respiratory tract.

(v) Hospital admissions and emergency room visits. Increased summertime hospital admissions and emergency room visits for respiratory causes have been associated with ambient exposures to O₃ and other environmental factors. Numerous studies consistently have shown such a relationship, even after controlling for modifying factors, as well as when considering only O₃ concentrations <0.12 ppm. Individuals with preexisting respiratory disease (e.g., asthma, chronic obstructive pulmonary disease) may generally be at increased risk of such effects, and some individuals with respiratory disease may have an inherently greater sensitivity to O₃. On the other hand, individuals with more severe respiratory disease are less likely to engage in the level of exertion associated with provoking responses to O₃ exposures in healthy humans. On balance, it is reasonable to conclude that evidence of O₃-induced increased airway resistance, nonspecific bronchial responsiveness, susceptibility to respiratory infection, increased airway permeability, airway inflammation, and incidence of asthma attacks suggests that ambient O₃ exposure could be a cause of increased hospital admissions, particularly for asthmatics.

(vi) Pulmonary inflammation. Respiratory inflammation can be considered to be a host response to injury and indicators of inflammation as evidence that respiratory cell damage has occurred. Inflammation induced by exposure of humans to O₃ may have several potential outcomes: (1) Inflammation induced by a single exposure (or even several exposures over the course of a season) could resolve entirely; (2) repeated acute inflammation could develop into a chronic inflammatory state; (3) continued inflammation could alter the structure and function of other pulmonary tissue, leading to disease processes such as fibrosis; (4) inflammation could interfere with the body's host defense response to particles and inhaled microorganisms, particularly in potentially vulnerable populations such as children and older individuals; and (5) inflammation could amplify the lung's response to other agents such as allergens or toxins. Exposures of laboratory animals to O₃for periods ≤8 hours have been shown to result in cell damage, inflammation, and increased leakage of proteins from blood into the air spaces of the respiratory tract. In humans, the extent and course of inflammation and its constitutive elements have been evaluated by using bronchoalveolar lavage (BAL) to sample cells and fluid from the lung and lower airways. Several such studies have shown that exercising humans exposed (1 to 4 hours) to 0.2 to 0.6 ppm O₃ had O₃-induced markers of inflammation and cell damage, with the lowest concentration of prolonged O₃ exposure tested in humans, 0.08 ppm for 6.6 hours with moderate exercise, inducing small but statistically significant increases in these endpoints. Thus, it is reasonable to conclude that repeated acute inflammatory response and cellular damage is potentially a matter of public health concern; however, it is also recognized that most, if not all, of these effects begin to resolve in most individuals within 24 hours if the exposure to O₃ is not repeated. Of possibly greater public health concern is the potential for chronic respiratory damage that could be the result of repeated O₃ exposures occurring over a season or a lifetime.

b. Potential Effects of Long-Term O₃ Exposures

Epidemiologic studies that have investigated potential associations between long-term O₃ exposures and chronic respiratory effects in humans thus far have provided only suggestive evidence of such a relationship. Most studies investigating this association have been cross-sectional in design and have been compromised by incomplete control of confounding variables and inadequate exposure information. Other studies have attempted to follow variably exposed groups prospectively. The findings from such studies conducted in southern California and Canada suggest small, but consistent, decrements in lung function among inhabitants of the more highly polluted communities; however, associations between O₃ and other copollutants and problems with study population loss have reduced the level of confidence in these conclusions. Other epidemiologic studies have attempted to find associations between daily mortality and O₃ concentrations in various cities around the United States. Although an association between ambient O₃ exposure in areas with very high O₃ levels and daily mortality has been suggested by these studies, the data are limited.

In a large number of animal toxicology studies, “lesions” ^[21] in the centriacinar regions of the lung (i.e., the portion of the lung where the region that conducts air and the region that exchanges gas are joined) are well established as one of the hallmarks of O₃ toxicity. Under certain conditions, some of the structural changes seen in these studies may become irreversible. It is unclear, however, whether ambient exposure scenarios encountered by humans result in similar “lesions” or whether there are resultant functional or impaired health outcomes in humans chronically exposed to O₃.

The epidemiologic lung function studies generally parallel those of the animal studies, but lack good information on individual O₃ exposure history and are frequently confounded by personal or copollutant variables. Thus, the Administrator recognizes that there is a lack of a clear understanding of the significance of repeated, long-term inflammatory responses, and that there is a need for continued research in this important area. In summary, the collective data on long-term exposure to O₃ garnered in studies of laboratory animals and human populations have many ambiguities. Nevertheless, the currently available information provides at least a biologically plausible basis for considering that repeated inflammation associated with exposure to O₃ over a lifetime may result in sufficient damage to respiratory tissue such that individuals later in life may experience a reduced quality of life, although such relationships remain highly uncertain.

c. Adversity of Effects for Individuals

Some population groups have been identified as being sensitive to effects associated with exposures to ambient O₃ levels, such that individuals within these groups are at increased risk of experiencing such effects. Population groups at increased risk include: (1) Active children and outdoor workers who regularly engage in outdoor activities; ^[22] (2) individuals with preexisting respiratory disease (e.g., asthma or chronic obstructive lung disease); ^[23] and (3) some individuals, referred to as “hyperresponders,” who are unusually responsive to O₃ relative to other individuals with similar levels of activity or with a similar health status and may experience much greater functional and symptomatic effects from exposure to O₃ than the average individual response.

In making judgments as to when the effects discussed above become significant enough that they should be regarded as adverse to the health of individuals in these sensitive populations, the Administrator has looked to guidelines published by the American Thoracic Society (ATS) and the advice of CASAC. Based on these guidelines, with CASAC concurrence, gradations of individual functional responses (e.g., decrements in forced expiratory volume (FEV₁), increased airway responsiveness) and symptomatic responses (e.g., cough, chest pain, wheeze) were defined, together with judgments as to the potential impact on individuals experiencing varying degrees of severity of these responses.^[24]

In judging the extent to which such impacts represent effects that should be regarded as adverse to the health status of individuals, an additional factor considered is whether such effects are experienced repeatedly by an individual during the course of a year or only on a single occasion. While some experts would judge single occurrences of moderate responses to be a “nuisance,” especially for healthy individuals, a more general consensus view of the adversity of such moderate responses emerges as the frequency of occurrence increases. Thus, EPA has concluded that repeated occurrences of moderate responses, even in otherwise healthy individuals, may be considered to be adverse since they could well set the stage for more serious illness.

2. Human Exposure and Risk Assessments

To put judgments about health effects that are adverse for individuals into a broader public health context, the Administrator has taken into account the results of human exposure and risk assessments. ^[25] This broader context includes consideration, to the extent possible, of the particular population groups at risk for various health effects, the number of people in at-risk groups likely to be exposed to O₃ concentrations shown to cause health effects, the number of people likely to experience certain adverse health effects under varying air quality scenarios, and the kind and degree of uncertainties inherent in these assessments. These quantitative assessments add to our understanding of the overall body of evidence linking O₃ inhalation exposures to adverse health effects. The EPA believes, and CASAC concurred, that the models used in these assessments were appropriate and that the methods used represent the state of the art.

a. Exposure Analyses

The EPA conducted exposure analyses to estimate O₃ exposures for the general population and two at-risk populations, active children who regularly engage in outdoor activity (i.e., “outdoor children”) and “outdoor workers,” living in nine representative U.S. urban areas.^[26] Exposure estimates were developed for a baseline year (e.g., 1993, 1994), using monitored O₃ air quality data (i.e., the “as is” scenario), as well as for simulated air quality conditions reflecting attainment of the 1-hour NAAQS and various alternative standards. The exposure analyses provide: (1) Estimates of the number of people exposed in each of these population groups to various O₃ concentrations, and the number of occurrences of such exposures, under different regulatory scenarios,^[27] which are an important input to the risk assessment conducted for certain adverse health effects (summarized in the next section); and (2) estimates of the frequency of occurrences of O₃ “exposures of concern,” ^[28] which help to put into broader perspective other O₃-related health effects that could not be included in the risk assessment (summarized below).

The computer model used in these analyses, the probabilistic NAAQS exposure model for O₃ (pNEM/O₃), combines information on O₃ air quality with information on patterns of human activity to produce estimates of O₃ inhalation exposures. This model has been developed to take into account the most significant factors contributing to total O₃ inhalation exposure including: the temporal and spatial patterns of ground-level O₃ concentrations throughout an urban area; the variations of O₃ levels within a comprehensive set of “microenvironments”; ^[29] the temporal and spatial patterns of the movement of people throughout an urban area; and the effects of variable exertion levels (represented by ventilation rates), associated with a range of activities that people regularly engage in, on O₃ uptake in exposed individuals. The analysis of these key factors incorporated extensive data bases, including, for example, data from ground-level O₃ monitoring networks in these areas, data from numerous research studies that characterized the activity patterns of the general population and at-risk groups as they go about their daily activities (e.g., from indoors to outdoors, moving from place to place, and engaging in activities at different exertion levels),^[30] and census data on relevant factors such as age, work status, home location and type of air conditioning system present, and work place location.

The regulatory scenarios examined in the exposure analyses include both 1-hour O₃ standards, at levels of 0.12 ppm (the 1979 NAAQS) and 0.10 ppm, and 8-hour standards, at levels of 0.07, 0.08, and 0.09 ppm, with 1- and 5-expected exceedance forms, i.e., the range of alternative 8-hour standards recommended in the Staff Paper and supported by CASAC as the appropriate range for consideration in this review. These estimates were also used to roughly bound exposure estimates for concentration-based forms of the standards under consideration (e.g., the second- and fifth-highest daily maximum 8-hour average O₃ concentration, averaged over a 3-year period).^[31] The estimated exposures are based on a single year of air quality data and reflect what would be expected in a typical or average year in an area just attaining a given standard over a 3-year compliance period; additional analyses were done to estimate exposures that would be expected in the worst year of a 3-year compliance period.

Based on the results of the exposure analyses, children who are active outdoors (representing approximately 7 percent of the population in the study areas) appear to be the at-risk population group examined with the highest percentage and number of individuals likely to experience exposures of concern. Estimated exposures of concern varied significantly across the urban areas examined in this analysis, with far greater variability associated with the 1-hour NAAQS in contrast to the more consistent results associated with alternative 8-hour standards.^[32] Despite this variability across areas, general patterns can be seen in comparing alternative standards. For example, for aggregate estimates of the mean percent of outdoor children likely to experience exposures of concern within the seven nonattainment areas: the range of estimates associated with the 1-hour NAAQS is approximately 0.3-24 percent, whereas for alternative 8-hour standards (of the same 1-expected-exceedance form as the 1-hour NAAQS), the ranges are approximately 3-7 percent for a 0.09 ppm standard, 0-1 percent for a 0.08 ppm standard, and essentially zero for a 0.07 ppm standard. Within any given urban area, these differences in estimated exposures of concern between alternative standards are statistically significant.

In looking more specifically at a comparison between 8-hour standards at the 0.09 ppm and 0.08 ppm levels, aggregate estimates of the mean percentage of outdoor children likely to experience exposures of concern are estimated to be approximately 3 percent at the 0.08 ppm level (ranging from 2-10 percent in the nine areas), increasing to approximately 11 percent at the 0.09 ppm level (ranging from 7-29 percent in the nine areas).^[33] Thus, based on these analyses, a standard set at 0.09 ppm would allow more than three times as many children to experience exposures of concern as would a 0.08 ppm standard, with the number of children likely to experience such exposures increasing from approximately 100,000 to more than 300,000 in these nine areas alone. These exposures of concern are judged by EPA to be an important indicator of the public health impacts of those O₃-related effects for which information is too limited to develop quantitative estimates of risk, but which have been observed in humans at a level of 0.08 ppm for 6- to 8-hour exposures. Such effects include increased nonspecific bronchial responsiveness (related, for example, to aggravation of asthma), decreased pulmonary defense mechanisms (suggestive of increased susceptibility to respiratory infection), and indicators of pulmonary inflammation (related to potential aggravation of chronic bronchitis or long-term damage to the lungs).

In taking these observations into account, the Administrator and CASAC recognize the uncertainties and limitations associated with such analyses, including the considerable, but unquantifiable, degree of uncertainty associated with a number of important inputs to the exposure model. A key uncertainty in model inputs results from limitations in the human activity data base that may not adequately account for day-to-day repetition of activities common to children, such that the number of people who experience multiple occurrences of high exposure levels may be underestimated. Small sample size also limits the extent to which ventilation rates associated with various activities may be representative of the population group to which they are applied in the model. In addition, the air quality adjustment procedure used to simulate air quality distributions associated with attaining alternative standards, while based on generalized models intended to reflect patterns of air quality changes that have historically been observed, contains significant uncertainty, especially when applied to areas requiring very large reductions in air quality to attain alternative standards or to areas that are now in attainment with the 1-hour NAAQS.^[34]

b. Risk Assessments

The EPA conducted an assessment of health risks for several categories of respiratory effects considering the same population groups, alternative air quality scenarios, and urban areas that were examined in the human exposure analyses described above. The objective of the risk assessment was to estimate to the extent possible the magnitude of risks to population groups believed by EPA and CASAC to be at greatest risk either due to increased exposures (i.e., outdoor children and outdoor workers) or increased susceptibility (e.g., asthmatics) while characterizing, as explicitly as possible, the uncertainties inherent in the assessment. While different risk measures are provided by the assessment, EPA has focused on “headcount risk” estimates which include: (1) Estimates of the number of people likely to experience a given health effect and (2) estimates of the number of incidences of a given health effect likely to be experienced by the population group of interest (n.b., some individuals likely experience that given health effect more than once in a year). While the estimates of numbers of people and incidences of effects are subject to uncertainties and should not be viewed as demonstrated health impacts, EPA believes they do represent reasonable estimates of the likely extent of these effects on public health given the available information.

This risk assessment builds upon earlier O₃ risk assessment approaches developed during the previous O₃ NAAQS review. The risk models produce estimates of risk by taking into account: (1) Exposure-response or concentration-response relationships used to characterize various respiratory effects of O₃ exposure; (2) distributions of population exposures upon attainment of alternative standards resulting from the exposure analyses described above; and (3) distributions of 1-hour and 8-hour daily maximum O₃ concentrations upon attainment of alternative standards, developed as part of the exposure analyses. The assessment addresses a number of adverse lung function and respiratory symptom effects as well as increased hospital admissions, as discussed below.

(i) Adverse lung function and respiratory symptom effects. Risk estimates have been developed for several of the respiratory effects observed in controlled human exposure studies to be associated with O₃ exposure for which sufficient quantitative dose-response information was available. These effects include lung function decrements (measured as changes in FEV₁) and pain on deep inspiration (PDI).^[35] More specifically, these effects, or health endpoints, are defined not only in terms of physiological responses, but also the amount of change in that response judged to be of medical significance (as discussed in section II.A.3 above). For decrements in FEV₁ responses, risk estimates are provided for the lower end, midpoint, and upper end of the range of response considered to be an adverse health effect (i.e., ≥ 10, 15, or 20 percent FEV₁ decrements), while for PDI responses, risk estimates are provided for moderate and severe responses. Although some individuals may experience a combination of responses, risk estimates could only be provided for each individual health endpoint rather than various combinations of functional and symptomatic responses.

The exposure-response relationships used to characterize these functional and symptomatic effects were based on the controlled human exposure studies, and were applied to “outdoor children,” “outdoor workers,” and the general population.^[36] These exposure-response relationships were combined with the results of the exposure analyses, which provided distributions of population exposures estimated to occur upon attainment of alternative standards, in terms of both the number of individuals in the general population, outdoor workers, and outdoor children exposed and the number of occurrences of exposure.

Following from the results of the exposure analyses showing outdoor children to be the population group experiencing the greatest exposures, this population group also has the highest estimated risk in terms of the percent of the population, and the numbers of children, likely to experience the health effects included in the assessment. As expected, the risk estimates exhibit the same general patterns in comparing alternative standards as was observed in the results of the exposure analyses. Estimated risk varied significantly across the urban areas examined, with greater variability associated with the 1-hour NAAQS than with alternative 8-hour standards, and, within any given urban area, the differences in risk estimated for the various 1-hour and 8-hour standards analyzed were statistically significant.

In looking more specifically at a comparison between 8-hour standards at the 0.09 ppm and 0.08 ppm levels, aggregate estimates of the number of outdoor children in the nine areas likely to experience moderate (≥ 15 percent) and large (≥ 20 percent) FEV1 decreases and moderate or severe PDI are summarized in the 1997 final rule.^[37] For example, for large FEV ₁ decreases (≥ 20 percent), approximately 2 percent of outdoor children (58,000 children) would likely experience this effect one or more times per year (100,000 occurrences) at the 0.08 ppm standard level, increasing to approximately 3 percent of outdoor children (97,000 children and 220,000 occurrences) at the 0.09 ppm standard level. Based on this assessment, a standard set at 0.09 ppm would allow approximately 40-65 percent more outdoor children to experience these functional and symptomatic effects than would a 0.08 ppm standard, and approximately 70-120 percent more occurrences of such effects in outdoor children per year.

In considering these observations, the Administrator and CASAC have recognized that there are many uncertainties inherent in such assessments, not all of which can be quantified. Some of the most important caveats and limitations in this assessment include: (1) The uncertainties and limitations associated with the exposure analyses discussed above; (2) the extrapolation of exposure-response functions, consistent with CASAC's recommendation, that projects some biological responses below the lowest-observed-effects levels to an estimated background level of 0.04 ppm; and (3) the inability to account for some factors which are known to affect the exposure-response relationships (e.g., assigning children the same symptomatic response rates as observed for adults and not adjusting response rates to reflect the increase and attenuation of responses that have been observed in studies of lung function and symptoms upon repeated exposures).^[38]

(ii) Excess respiratory-related hospital admissions. A separate risk assessment was done for increased respiratory-related hospital admissions as reported in several epidemiologic studies.^[39] The assessment looked only at one urban area, New York City, for which adequate air quality information was available to assess population risk. Increased respiratory-related hospital admissions for individuals with asthma were modeled using a probabilistic concentration-response function based on the results of an epidemiologic study in New York City (Thurston et al., 1992) and estimated distributions of daily maximum 1-hour average O₃ concentrations upon attainment of alternative standards at various monitors in New York City (developed as part of the exposure analysis discussed above).^[40] The resulting risk estimates are for excess respiratory-related hospital admissions (i.e., those attributable to O₃ concentrations above an estimated background O₃ level of 0.04 ppm) for asthmatic individuals over an O₃ season.

Similar to the risk assessment discussed above for lung function and respiratory symptom effects, reductions in hospital admissions for respiratory causes for asthmatic individuals and the general population are estimated to occur with each change in the level of alternative 8-hour standards from 0.09 ppm to 0.07 ppm. In looking more specifically at a comparison between 8-hour standards at 0.09 ppm and 0.08 ppm levels, a standard set at 0.09 ppm is estimated to allow approximately 40 more excess hospital admissions of asthmatics within an O₃ season in New York City for respiratory causes as compared to a 0.08 ppm standard, which represents approximately a 40 percent increase in excess O₃-related admissions, but only approximately a 0.3 percent increase in total admissions of asthmatics. The EPA believes that while these numbers of hospital admissions are relatively small from a public health perspective, they are indicative of a pyramid of much larger numbers of related O₃-induced effects, including respiratory-related hospital admissions among the general population, emergency and outpatient department visits, doctors visits, and asthma attacks and related increased use of medication that are important public health considerations.

In taking these observations into account, the Administrator recognizes the uncertainties and limitations associated with this assessment. These include: (1) The inability at this time to quantitatively extrapolate the risk estimates for New York City to other urban areas; (2) uncertainty associated with the underlying epidemiologic study from which the concentration-response relationship used in the analysis was drawn; and (3) uncertainties associated with the air quality adjustment procedure used to simulate attainment of alternative standards for the New York City area.^[41]

B. Potential Indirect Beneficial Health Effects Associated With Ground-level O₃

This section is drawn from information in the record of the 1997 review with regard to the effect of ground-level O₃ on the attenuation of UV-B radiation and potential associated health benefits. All relevant record information was reviewed, including EPA documents, published articles, oral testimony at public meetings, and written comments submitted during the rulemaking. This section summarizes information on the health effects associated with UV-B radiation exposure and the relationship between ground-level O₃ and UV-B radiation, and evaluates estimates of UV-B radiation risks that have been attributed to reductions in ground-level O₃ projected to result from attainment of the 1997 O₃ NAAQS.

1. Health Effects Associated With UV-B Radiation Exposure

It has long been recognized that exposure to sunlight has a positive effect on health. Sunlight is essential to the human body because of its biosynthetic action. More specifically, UV radiation induces the conversion of ergosterol and other vitamin precursors present in normal skin to vitamin D, an essential factor for normal calcium deposition in growing bones.^[42] Sunlight is also an important controlling agent of recurrent daily physiological alterations known as circadian rhythms. Lighting cycles have been shown to be important in regulating several types of endocrine function. However, it is also recognized that excessive exposure to solar radiation can result in adverse health effects, which are particularly associated with UV-B radiation.

The following summary of information on the adverse human health effects associated with exposure to UV-B radiation focuses on the three major organ systems whose tissues are commonly exposed to solar radiation: the skin, eyes, and immune system.^[43] It is these three systems that are potentially subject to damage from increased UV-B radiation as a result of the absorption of solar energy by molecules present in the cells and tissues of these organs. The biologically effective dose of radiation that actually reaches target molecules generally depends on the duration of exposure at particular locations, time of day, time of year, behavior (i.e., “sun avoidance,” which is an intentional decrease in exposure, for example, by using clothing, sunscreens, and sunglasses to shield from solar radiation; and “sun seeking,” which is an intentional increase in exposure to solar radiation, for example, by sunbathing), and, for the skin, characteristics that include pigmentation and temporal variations (e.g., changes in the pigmentation due to tanning).

a. Effects on the Skin

The most common form of solar damage to the skin is sunburn. Susceptibility to sunburn and the ability to tan are the basis for a classification system of six skin phenotypes. The most sensitive individuals (skin type I) are very light-skinned, with red or blonde hair and blue or green eyes (U.S. EPA, 1987, ES-33). The most resistant individuals (skin type VI) are darkly pigmented even without exposure to solar radiation. Susceptibility to sunburn may be a risk factor for skin cancer.

Among light-skinned populations, skin cancer is among the most common kinds of cancer. The three types of skin cancer that have been associated with exposure to solar radiation include two common types of nonmelanoma skin cancers, squamous cell carcinoma (SCC) and basal cell carcinoma (BCC), and melanoma, a far less common form of cancer. Various types of evidence support the conclusion that increases in solar radiation in general, and UV-B radiation in particular, increase skin cancer morbidity and mortality. Epidemiological studies are the primary source of information providing evidence of associations between UV-B radiation and the occurrence of skin cancer in humans. In addition, experimental studies on animals, and animal and bacterial cells, have helped define the action spectra for particular biological endpoints, which describe how effective radiation of specific wavelengths is in causing a biological effect, and also the possible mechanisms by which damage can occur.

(i) Nonmelanoma skin cancer (NMSC). Based on surveys, particularly in the U.S. and Australia, prolonged exposure to the sun is considered to be the dominant risk factor for NMSC (U.S. EPA, 1987, ES-33). It has been observed that NMSC tends to develop on sites that are most frequently exposed to the sun (e.g., head, face, and neck). Outdoor workers, who are subject to greater exposure to solar radiation, tend to have higher incidence rates of NMSC. A latitudinal gradient exists for the flux of UV-B radiation (i.e., the amount of radiation transmitted through the atmosphere), with fluxes generally higher in lower latitudes. A similar latitudinal gradient is generally seen in incidence rates of NMSC. Skin pigmentation provides a protective barrier that reduces the risk of developing NMSC, such that light-skinned individuals, who are more susceptible to sunburn and have blue or green eyes, are more likely to develop NMSC. The risk of NMSC is highest among individuals with a genetic predisposition to abnormal skin pigmentation (e.g., people with xeroderma pigmentosum).

Both types of NMSC result from the malignant transformation of keratinocytes, the major structural cells of the skin. Cumulative long-term exposure to UV radiation is the exposure of concern for both types of NMSC. More specifically, the incremental increase in cumulative lifetime exposure to UV-B radiation is the metric used to estimate the risk of increased incidence of NMSC (U.S. EPA, 1987, ES-3). Epidemiological evidence, however, also indicates that exposure to solar radiation may play different roles in the etiology of SCC and BCC. In particular, SCC is more likely to develop on sites receiving the highest cumulative UV radiation doses (e.g., nose), and the development of SCC is more strongly associated with cumulative exposure to UV radiation. Relative to SCC, BCC is more likely to develop on sites that are not normally exposed to the sun, such as the trunk. For a given cumulative level of exposure to solar radiation, the risk of developing SCC may be greater than the risk of developing BCC.

Results from experimental studies suggest that UV-B radiation may be the most important component of solar radiation that causes variations in the incidence of NMSC. UV radiation has been demonstrated to produce nonmelanoma skin tumors in animals, and UV-B wavelengths have been shown to be the most effective part of the UV spectrum in producing these tumors. Mechanisms by which this damage can occur have been demonstrated in laboratory animals. UV-B radiation has been shown to cause a variety of DNA lesions, to induce neoplastic transformation in cells, and to be a mutagen in both animal and bacterial cells.

Dose-response relationships for NMSC are generally estimated in terms of a biological amplification factor (BAF), which is defined as the percent change in tumor incidence that results from a 1 percent change in UV-B radiation. While there is considerable uncertainty in such estimates, results from several studies have produced an overall BAF range that is 1.8 to 2.85 for all nonmelanoma skin tumors (U.S. EPA, 1987, ES-34). The BAF estimates are generally higher for males than females and for SCC than BCC, and generally increase with decreasing latitude. Key uncertainties in these estimates include, for example, uncertainties in the actual doses of UV-B radiation received and in the underlying baseline incidence rates in populations. Additional uncertainty is introduced in estimating the change in mortality from NMSC associated with changes in UV-B radiation, reflecting in part discrepancies of reporting between death certificates and hospital diagnoses. Based on published estimates, rates of metastasis among SCCs and BCCs varied by one to two orders of magnitude, with rates estimated to be approximately 2 to 20 percent for SCC and 0.0028 to 0.55 percent for BCC. The overall fatality rate for NMSC has been estimated to be approximately 1 to 2 percent, with three-fourths to four-fifths of the deaths attributable to SCC (U.S. EPA, 1987, ES-34).^[44]

(ii) Melanoma. Melanoma is a serious, life-threatening skin cancer that is far rarer and generally much more aggressive than NMSC. Melanoma is a malignant cancer of the melanocytes, the pigment producing cells in the skin. While the development of melanoma is associated with cumulative lifetime exposure to UV radiation, there are several histological forms of melanoma that vary in their relationships to exposure to solar and UV-B radiation, sites on the body, skin pigmentation, and possibly in precursor lesions. Assessment of incidence by type is not consistent among registries, thus complicating attempts to evaluate the relationship between melanoma and solar radiation (U.S. EPA, 1987, ES-36).

The relationship between exposure to UV-B radiation and melanoma is not as clear as the relationship between exposure to UV-B radiation and NMSC. The EPA (1987) noted limitations in the evidence linking solar radiation to melanoma. For example, no animal models were identified in which exposure to UV-B radiation experimentally induces melanoma, and no in vitro models for malignant transformation of melanocytes. Despite these limitations, EPA (1987) recognizes that a large array of evidence does support the conclusion that solar radiation is one of the causes of melanoma. Melanin, the principal pigment in the skin, effectively absorbs UV radiation, such that darker skin provides more protection from UV radiation. Light-skinned races, whose skin contains less protective melanin, have higher incidence and mortality rates from melanoma than do dark-skinned races. Lighter members of light-skinned races, including those who are unable to tan or who tan poorly, have a higher incidence of melanoma than do darker members of light-skinned races. In addition, as was the case in NMSC, the risk of melanoma is highest among individuals with a genetic predisposition to abnormal skin pigmentation (e.g., people with xeroderma pigmentosum).

Sun exposure seems to induce freckling, which is an important risk factor for melanoma, and sun exposure leading to sunburn apparently induces melanocytic moles, which are also a risk factor for melanoma. Additional evidence suggests that melanoma risk may be associated with childhood sunburn. However, other evidence suggests that childhood sunburn may be a surrogate for an individual's pigmentation characteristics or be related to mole development, rather than being a separate risk factor (U.S. EPA, 1987, ES-37).

Most studies that have used latitude as a surrogate for sunlight or UV-B exposure have found an increase in melanoma incidence or mortality correlated with proximity to the equator. Other evidence, however, creates uncertainty about the relationship between solar radiation and melanoma. Some ecologic epidemiology studies, conducted primarily in Europe or in countries close to the equator, have failed to find a latitudinal gradient for melanoma. In addition, outdoor workers generally have lower incidence and mortality rates from melanoma than indoor workers, which appears to be incompatible with the hypothesis that the cumulative dose from exposure to solar radiation causes melanoma. Unlike SCC and BCC, most melanoma occurs on sites of the body that are not habitually exposed to sunlight. This evidence suggests that exposure to solar radiation, or UV-B, is not solely responsible for variations in the incidence and mortality from melanoma (U.S. EPA 1987, ES-37).

Considering the available evidence, EPA (1987) concluded that UV-B radiation is a likely component of solar radiation that causes melanoma, either through the initiation of tumors or through suppression of the immune system. The EPA (1987) also recognized that significant uncertainties exist in characterizing associations between solar radiation and melanoma, including the appropriate action spectrum to be used in estimating doses, the best functional form for a dose-response relationship, and the best way to characterize dose (e.g., peak value, cumulative summer exposure).

b. Effects on the Eyes

Evidence suggests that adverse effects on the eye are associated with exposure to UV-B radiation. Effects likely include increases in cataract incidence or severity and increased incidence of retinal disorders and retinal degeneration. Cataracts are characterized by the gradual loss of transparency of the lens due to the accumulation of oxidized lens proteins. Many possible mechanisms exist for the formation of cataracts, and UV-B radiation may play an important role in some mechanisms. Epidemiological and laboratory evidence indicates that the exposure of concern in the development of cataracts is the cumulative lifetime exposure to UV-B radiation.

Although the cornea and aqueous humor of the human eye screen out significant amounts of ultraviolet-A (UV-A) and UV-B radiation, nearly 50 percent of radiation at 320 nm is transmitted to the lens. Transmittance declines substantially below 320 nm, so that less than 1 percent is transmitted below approximately 290 to 300 nm. However, results of laboratory experiments on animals indicate that short-wavelength UV-B (i.e., below 290 nm) is perhaps 250 times more effective than long-wavelength UV-B (i.e., 320 nm) in inducing cataracts. Thus, while epidemiological studies indicate that the prevalence of human cataracts varies with latitude and UV radiation in general (U.S. EPA, 1987, ES-40), significant uncertainty exists about the action spectrum to be used in any estimation of dose associated with variations in solar radiation.

c. Effects on the Immune System

Information on the effects of UV-B radiation on the immune system comes primarily from laboratory animal studies. High doses of UV radiation cause a depression in systemic hypersensitivity reactions, resulting in an inability of the animal to respond to an antigen presented to the animal through unirradiated skin, whereas relatively lower doses cause a depression in local contact hypersensitivity, resulting in an inability to respond to an antigen presented through UV-irradiated skin. Both of these immunosuppressive effects of UV radiation have been found to reside almost entirely in the UV-B portion of the solar spectrum (U.S. EPA, 1987, ES-39).

Information about the effects of UV radiation on the human immune system, however, is much more limited. Preliminary studies indicate the UV radiation may prevent an effective immune response to micro-organisms that infect via the skin. Because UV-B can produce systemic immunologic change, the possibility exists that changes in UV-B radiation exposure could result in effects on diseases whose control requires systemic rather than local immunity. Without more complete information from laboratory or epidemiological studies, the nature of an exposure of concern cannot be estimated. Immunologic studies have not assessed the effects of long-term, low-dose UV-B irradiation, such that the magnitude of risk from this type of exposure cannot be assessed (U.S. EPA, 1987, ES-40).

2. Relationship Between Ground-level O₃ and UV-B Radiation Exposure

a. Relevant Atmospheric Factors

The relationships between ground-level O₃ and UV radiation occur in the context of a much larger dynamic of the earth's atmospheric systems. The sun is, of course, overwhelmingly the main source of a wide band of electromagnetic radiation, including the ultraviolet. The total atmosphere blocks a significant portion of the range of this incoming solar radiation before it reaches ground level, including much of the more energetic wavelengths that are shorter than visible light. The UV spectrum (100-400 nm) is comprised of UV-C (100-280 nm), UV-B (280-320 nm), and UV-A (320-400 nm). The most energetic component, UV-C, is completely blocked or absorbed by oxygen (O₂) and O₃ in the atmosphere. The middle range, UV-B, is efficiently but not completely absorbed by total column O₃. Ultraviolet-A radiation (320-400 nm) in wavelengths above 350 nm is not absorbed by O₂ or O₃, nor is visible light (4000-900 nm) ^[45] (U.S. EPA, 1987, ES 35). The absorption of UV-B by O₃ varies across the spectrum, being much stronger for wavelengths of 300 nm and below than for the upper region near 320 nm (Cupitt, 1994). Because the amount of atmospheric O₃ traversed by sunlight varies with the sun angle, atmospheric absorption is more complete in winter months and both early and late in the day, as compared to the absorption around mid-day near the summertime solar zenith. Therefore, a decrease in total column O₃ from naturally occurring conditions is of greater concern during times of higher sun angles, and for the more energetic portion of the UV-B range.

The underlying annual and diurnal patterns of UV-B penetration to the ground layer are driven primarily by three factors: (1) The change in apparent sun angle with the surface that occurs as the earth travels around the sun; (2) the diurnal change in apparent sun angle caused by the earth's rotation; and (3) the solar/meteorologically driven annual change in the amount of O³ in the stratosphere. Stratospheric O₃ over U.S. latitudes shows a characteristic peak in the spring months, falling steadily thereafter through summer and fall (Fishman et al., 1990; Frederic et al., 1993). The combination of the annual sun cycle and the stratospheric O₃ cycle means that peak UV-B radiation reaching the troposphere tends to occur in late June to early July, and falls steadily thereafter (Frederick et al., 1993). The annual peak in ground-level O₃ concentrations, which extends in most areas from May through September, generally overlaps the UV-B radiation peak (e.g., U.S. EPA, 1996a, Figure 4-23).

As noted in the EPA's SunWise Program communications, UV-B radiation exposure is of most concern between the hours of 10 am and 4 pm, peaking around mid-day. Ground-level O₃ patterns vary, but in urban areas, summertime peaks tend to occur between noon and 4 pm (U.S. EPA, 1996a, Section 4.4). This obviously overlaps with peak incoming UV-B radiation. The pattern of vertical mixing in the atmosphere is such that morning ground-level measurements probably do not accurately reflect “mixing-layer” concentrations (U.S. EPA, 1996a, p. 3-44).^[46]

The relationship between ground-level O₃ and solar radiation, including UV-B radiation, is complex and mediated by a number of atmospheric factors. It is not limited to the simple absorption of energy. At a fundamental level, the variation in apparent solar radiation is a primary cause of meteorological fluctuations that strongly influence the build-up and transport of anthropogenic air pollution. Further, as discussed in Chapter 3 of the Criteria Document, UV-B radiation that penetrates the stratosphere to the mixing layer plays a key role in the processes leading to the formation of photochemical smog, including the formation of ground-level O₃. In fact, increased penetration of UV-B radiation to the troposphere due to stratospheric O₃ depletion would likely increase ground-level concentrations of O₃ in most urban and many rural areas of the U.S. (U.S. EPA, 1996a, p. 3-5). The chain of indirect events triggered by increased penetration of UV-B radiation can result in both increases and decreases in aerosol and acid rain formation (U.S. EPA, 1996a; pp. 3-38 to 39), with attendant further feedbacks through heterogeneous chemistry and aerosol scattering of UV-B radiation. All of these complex processes could, under varying conditions, increase or decrease the amount of UV-B radiation that actually reaches ground level relative to an unperturbed case. The reactions can further affect the concentrations of radiatively important substances such as methane, ozone, and particles, and could affect local, regional, and global climate.

Setting aside the direction and magnitude of these complex indirect effects of UV-B radiation penetration on ground-level air pollution, and assuming appropriate sun angles and cloud density, the marginal effect of ground-level O₃ on the absorption of UV-B radiation by the earth's atmosphere can be considered separately. Because of increased scattering of incident UV-B radiation by the denser layer air molecules, droplets, and particles nearer the surface, tropospheric O₃ can absorb somewhat more UV-B radiation than an equal amount of O₃ in the stratosphere (Brühl and Creutzen, 1989). The extent to which this increase in unit effect occurs depends on the relative concentrations and character of aerosols in the troposphere as compared to the stratosphere.

A further consideration is the relative effectiveness of ground-level O₃ in absorbing those spectra of UV-B radiation wavelengths most likely to cause health effects. The “effective dose” of UV-B radiation can be expressed as a function of two factors, the intensity of radiation (by wavelength) reaching the earth's surface and the action spectrum. The wavelength-dependent effect of O₃ on reducing the intensity of radiation in the UV-B range is summarized above. The action spectrum describes how effective radiation at particular wavelengths is at causing a particular biological effect or a response in an instrument. Action spectra allow the estimation of the potential effects of simultaneously changing radiation at different wavelengths by different amounts, as happens with changing O₃ levels. Laboratory and field studies have been used to estimate and adopt action spectra conventions for various biological endpoints (e.g., Madronich, 1992). As noted above, uncertainty exists about the action spectra as well as how to specify appropriate dose metrics for particular health endpoints. Even estimates of the range of wavelengths considered to be generally biologically active vary within the UV-B radiation spectrum. These different action spectra have different sensitivities to changes in total column O₃, which are formalized as numerical radiation amplification factors (RAF).^[47] In general, a 1 percent change in total column O₃ will produce greater than a 1 percent change (e.g., 1.1 to 1.8 percent) in effective radiation dose for particular effects.

Nevertheless, as noted above, typical summertime ground-level O₃ pollution in the eastern U.S. is less than 1 percent of total column O₃. Even considering the relative effectiveness of ground-level O₃ in reducing UV-B radiation and the amplification of effective dose, such pollution could add a few percent at most to naturally occurring biologically effective UV-B radiation shielding.^[48] Viewed from one perspective and holding all other factors constant, the assumed typical O₃ pollution level is providing some “improvement” or incremental UV-B radiation shielding above the natural conditions that would otherwise exist in the mixing layer. It should also be noted that, if typical summertime O₃ levels were assumed to approximate the estimated continental background of about 40 ppb for daylight hours (U.S. EPA, 1996b, p. 20-21), this too would represent an “improvement” over the natural conditions that would exist in the mixing layer without the influence of international transport of O₃.^[49]

The extent to which changes in ground-level O₃ concentrations would translate into changes in UV-B radiation-related health effects in various locations cannot, however, be adequately viewed by reference to uniform assumptions applicable for specific sun angle, latitude, time of day, cloud cover, and the presence of other pollutants. ^[50] In the real world, all of these factors vary with location, season, meteorology, and time of day. Moreover, the complex causal relationships noted above among all of these factors mean that neither static calculations holding other factors constant (e.g., Cupitt, 1994) nor simple empirical associations between measured ground-level O₃ and UV-B radiation (e.g., Frederick et al., 1993) provide an adequate basis for assessing the “net” shielding associated with control strategy driven changes in ground-level pollution in various locations over an extended time period. Moreover, as for the direct effects of O₃, the extent of resultant UV-B radiation-related health effects is also heavily dependent on the variation of these physical changes superimposed on the activity patterns and other factors that determine population exposures and sensitivities to UV-B radiation, and on the extent to which significant biological responses can be attributed in part to episodic peak exposures as well as to long-term cumulative exposures.

Assessing the effective O₃ layer shielding is considerably more difficult for ground-level O₃ than for stratospheric O₃ because of its far greater spatial and temporal variability and the much smaller contribution made by ground-level O₃. Some insights into the relative variability of these two layers are provided in Fishman et al. (1990), which compares satellite measurements of stratospheric O₃ with “residual” tropospheric O₃, a measure that actually excludes the lowest portion of the ground-layer O₃ in the mixing layer. For the summer months, the long-term spatial variability in the amount of ozone in the stratosphere across the lower 48 U.S. States is about 7 percent (Figure 8c), while the variability in the tropospheric “residual” is nearly 4 times greater, at about 25 percent (Figure 9c). By comparison, the spatial variability in ground-level O₃ measurements across regions and cities in the U.S. is far greater (U.S. EPA, 1996a, Chapter 4) reaching 200 percent and higher for comparable long-term measurements. Within an area as small as the Los Angeles basin alone, for example, the median ground-level 8-hour O₃ values in different locations varied by more than a factor of 2 (Table 28; Johnson et al., 1996c). The satellite information also shows a marked contrast in the seasonal variations in O₃ for these two layers. The variation in the summer/winter stratospheric O₃ column over the U.S. is only about 2 to 4 percent, while the variation in seasonal “residual” tropospheric O₃ is about 50 to 80 percent (Figures 8a,c;9a,c; Fishman et al., 1990). Again, the variability is even greater for ground-level measurements (e.g., U.S. EPA, 1996a, Figure 4-23; Frederick et al., 1993)

Although Fishman et al. (1990) do not compare daily variations in stratospheric O₃ above the U.S., it is reasonable to conclude that the spatial and annual/seasonal temporal stability evidenced by this large stratospheric reservoir would result in far more stable day-to-day and diurnal patterns as compared to ground-level O₃. The high variability of daytime O₃ concentrations for these temporal scales is amply documented in the Criteria Document (U.S. EPA, 1996a, Figure 4-23).

The spatial and temporal stability of the expansive and deep stratospheric O₃ reservoir means that assessments of the effects of long-term declines or restoration can reasonably assume that short-term and local-scale variations in important factors such as cloud cover, other pollutants, temperature, and activity patterns beneath this layer will tend to “even out” over time, permitting more confidence in the magnitude and direction of such assessments. In contrast to the stability of the stratospheric O₃ layer, the large spatial and day-to-day variability outlined above for ground-level O₃ means that geographical or temporal variations in other factors such as weather, other pollutants, and human activity patterns may not “even out” in particular areas under assessment. Moreover, it is reasonable to assume that the variations in ground-level O₃ are not independent of the variations in many of these other factors. Such variability may have a substantial impact on the outcome of any assessment of the relative effects of a change in ground-level O₃ strategies or standards. This, combined with the many local- and regional-scale interactions among all of these factors, would complicate any such ground-level O₃ assessment.

b. Factors Related to Area-Specific Assessment

An enumeration of factors that would be important in assessing the potential UV-B radiation-related consequences of a more stringent O₃ NAAQS in any geographical area serves to illustrate the complexities discussed above. Analogous to the factors that were important in the respiratory effects exposure and risk assessments discussed above section II.A.2, these UV-B radiation-related factors include: the temporal and spatial patterns of ground-level O₃ concentrations throughout a geographic area where reductions are likely to occur, and the variations in O₃ concentrations within a comprehensive set of “microenvironments” relevant to UV-B radiation exposures; the associated temporal and spatial patterns of UV-B radiation flux in such microenvironments; the temporal and spatial patterns of movement of people throughout the microenvironments within the geographic area; and the effects of variable behaviors (e.g., the use of sunscreen, hats, sunglasses) within the range of activities that people regularly engage in, on the effective dose of UV-B radiation that reaches target organs such as the skin.

While analogous to the respiratory-related factors, there are a number of important differences between these sets of factors that arise, for example: (1) Due to the indirect nature of the relationship between changes in ground-level O₃ and UV-B radiation-related health effects (in contrast to the direct relationship between ground-level O₃ and inhalation-related health effects); (2) the long-term nature of the relevant exposures that are associated with UV-B radiation's chronic health effects (in contrast to the short-term exposures associated with acute inhalation effects); (3) the different types of parameters that are relevant to assessing dermal exposures (in contrast to those that are important in assessing inhalation exposures); and (4) the importance of skin type in characterizing the sensitive populations (in contrast to characterizing sensitive populations in terms of activity levels and respiratory health status). Further, as was done in EPA's assessment of respiratory effects, it is important to characterize the exposure-related factors specifically to address the relevant at-risk sensitive population groups. As noted in section II.B.1, the sensitivity to UV-B radiation effects varies among U.S. demographic groups, such that it could be important to incorporate census data on relevant characteristics (e.g., age at time of exposure, skin pigmentation) that affect an individual's susceptibility.

Aspects of each of these factors are discussed briefly below, and areas where current information or modeling tools are insufficient to address these factors at this time are noted.

(i) Estimation of area-specific and microenvironment changes in ground-level O₃. Implementation of a more stringent O₃ standard would, over time, further reduce O₃ concentrations across the U.S., but would affect various areas in different ways. Depending on the strategies adopted, in some locations peak concentrations would be reduced significantly during the O₃ season, while the lower concentrations that occur on far more numerous days could increase. In such areas, the long-term cumulative effect could be little net change, or even a small increase in cumulative shielding. In other areas, the entire distribution of O₃ could be reduced. The assessment of the acute respiratory health effects of O₃ appropriately focused on the higher portion of this distribution, using a simple roll-back approach discussed above (section II.A.2.a) to simulate changes in air quality patterns during the O₃ season based on available air quality monitoring data. For assessment of chronic effects such as those associated with UV-B radiation, however, where long-term cumulative exposures are of central importance, the mid to lower portion of the distribution would also be important. Also the distribution across the entire year, for which O₃ monitoring data is not generally available in many parts of the country, could potentially be important. The mid to lower portion of the distribution is much more strongly influenced by complex atmospheric chemistry, such that more sophisticated, area-specific modeling may be needed.

In addition, although not relevant to assessing direct respiratory effects, the vertical distribution of O₃ concentrations up through the mixing layer becomes important in assessing the effect of O₃ in shielding UV-B radiation. The current lack of routine vertical profile measurements means that little is known about the relative effect of ground-level control strategies on O₃ in the mixing layer.

With regard to characterizing changes in O₃ concentrations within microenvironments relevant to UV-B radiation exposure, it is clear that this set of microenvironments would differ in some respects from the set of microenvironments that were relevant for respiratory effects. For example, while indoor microenvironments can reduce exposure to both ambient O₃ and UV-B radiation, outdoor microenvironments that are relevant for inhalation exposure do not reflect the characteristics that are important for UV-B radiation exposure. For example, while not relevant to inhalation exposure, microenvironments shaded by the presence of trees, buildings, and other structures in many heavily occupied areas could be important to characterize because they would tend to have greatly reduced UV-B radiation exposures even when at the same ground-level O₃ concentration as a sunny microenvironment.

(ii) Estimation of temporal and spatial patterns of UV-B radiation flux. Relative to the assessment of respiratory effects, the assessment of the effect of O₃ shielding on UV-B radiation-related health effects requires the additional step of estimating how changes in the temporal and spatial patterns of O₃ concentrations result in changes in the patterns of UV-B radiation. Given a three-dimensional pattern of O₃ levels, a first-order approximation of UV-B penetration to the Earth's surface can be readily made. The factors that influence radiation flux through the stratosphere are fairly well characterized, and most directly related to the modest changes in stratospheric O₃ and large variations in sun angle that depend on latitude, time of year, and time of day (U.S. EPA, 1987). Nevertheless, beyond these factors, and in addition to changes in ground-level O₃, a number of other (second-order) factors in the boundary layer and the rest of the troposphere can affect the amount of UV-B radiation reaching potentially affected populations. One such factor is cloud cover, which can reduce UV-B radiation reaching the earth's surface by 50 percent or more (Cupitt, 1994). Another such factor is the presence of UV-B radiation scattering and absorbing aerosols. Depending on local circumstances and the strategy chosen, aerosol-related UV-B radiation exposure might increase or decrease as a result of ground-level O₃ reductions (U.S. EPA, 1996a, Chapter 3). Both O₃ and aerosols can affect local climate as well as UV-B radiation, and this could affect cloud cover as a further indirect consequence of a reduction strategy. While any such indirect effects might be expected to be small for modest O₃ changes, it is not currently possible to predict the magnitude or the sign of their net effect on UV-B radiation penetration.

(iii) Estimation of temporal and spatial patterns of movement of people throughout microenvironments. While population densities are high in areas with the highest ground-level O₃ concentrations, people may not receive their highest exposure to UV-B radiation in such locations. Reductions in O₃ shielding would presumably be most significant in outdoor recreational areas such as the beach or rural open areas where many people likely receive a disproportionate share of their cumulative sun exposure. Local or regional meteorological factors can, however, cause ground-level O₃ concentrations to be lower in many such areas, particularly in the western United States. For example, O₃ concentrations in the heavily populated Los Angeles area tend to be lowest at the coast and increase inland; in this case, smog-related O₃ would be providing the least shielding where the potential for exposure to UV-B radiation is the highest. The extensive database on human activity patterns, which was used in the assessment of respiratory effects, does not include parameters that relate to people's movement through the types of outdoor microenvironments that are relevant to the assessment of UV-B radiation exposure. For example, additional data would be needed to conduct an exposure analysis that could account for the fraction of UV-B radiation exposure that is incurred during outdoor recreational activities in non-shaded microenvironments. EPA believes that reliable estimation of the change in UV-B radiation exposure associated with reducing ground-level O₃ would be hindered by not taking such factors into account.

(iv) Effects of variable behaviors on effective dose of UV-B radiation. Another important factor to be considered in assessing the potential UV-B radiation-related effects of a change in ground-level O₃ is that human behavior affects UV-B radiation exposures. When people choose to shield themselves from UV-B radiation exposure with clothing and sunscreens, and by timing their outdoor activities to avoid peak sun conditions, they are affecting a parameter that is important in assessing UV-B radiation-related effects. The generally well-known risks associated with too much sun exposure are such that many people limit their own as well as their children's exposure through such measures, regardless of the status of the protective stratospheric O₃ layer or variable amounts of ground-level O₃ pollution. While some sun exposure is generally beneficial to health, limiting excessive sun exposure would remain important for a person's health even if the stratospheric O₃ layer were fully restored to its natural state.^[51]

Since sun-seeking or sun-avoidance behaviors can tend to maximize or minimize exposure to UV-B radiation, not factoring such behavioral data into an area-specific exposure assessment would hinder reliable estimation of the increased exposure associated with reducing ground-level O₃. Changes in behavior in the past, specifically increases in sun-seeking behaviors, are believed to be the primary reason for the increases in skin cancer incidence and mortality observed in the U.S. by the 1980's (U.S. EPA, 1987). Conversely, future rates of skin cancer could be reduced to the extent that people choose to change their behavior by increasing sun-avoidance behaviors.

Public awareness of the risks associated with overexposure to UV radiation seems to be having an effect on behavior. In 1987, EPA noted that behaviors causing increased UV-B radiation exposure were apparently reaching an upper limit (U.S. EPA, 1987, ES-35). The effect of increased awareness of the health consequences of UV-B radiation exposure on decreasing the number of harmful exposures is not likely to show up, in terms of reducing the incidence and mortality rates of skin cancers, for many years. Nevertheless, ignoring its effects would tend to bias exposure estimates in an area-specific assessment of the UV-B radiation-related effects of smog reduction strategies.

Based on the discussion of factors above, the Administrator believes that more information is needed to address these factors before reliable area-specific quantitative assessment of potential UV-B radiation-related consequences of a more stringent O₃ NAAQS would be possible. EPA intends to seek additional information relevant to such quantitative assessment. EPA is now requesting comment on the factors discussed above.

3. Evaluation of UV-B Radiation-Related Risk Estimates for Ground-level O₃ Changes

As should be clear from the discussion above, a full risk assessment of UV-B radiation-related effects resulting from a moderate change in ground-level O₃ would be an extremely challenging enterprise that appears to be beyond current data and modeling capabilities. Nevertheless, three analyses (Cupitt, 1994; U.S. DOE, 1995; Lutter and Wolz, 1997) have developed estimates that attempt to bound the potential indirect UV-B radiation related effects associated with replacing the former 1-hour O₃ NAAQS with an 8-hour O₃ standard. All three analyses essentially reflect a static comparison of two separate O₃ concentrations on a national basis, and include, either explicitly or implicitly, numerous assumptions needed while excluding the important area-specific issues and factors outlined above.

The most thoroughly documented calculations are those provided in Cupitt (1994), an EPA white paper developed as an initial scoping analysis of the issues, in preparation for potential consideration in the Regulatory Impact Analysis (RIA) that would accompany the O₃ NAAQS regulatory package. The paper discusses many of the important factors and uncertainties outlined above, summarizes key background information to provide perspective, and includes a discussion and table summarizing the many simplifying assumptions that were needed to permit the development of quantitative estimates. Cupitt's analysis evaluates changes resulting from cumulative exposures under two scenarios, including one that compares estimates of NMSC incidence associated with an assumed reduction of daytime summer O₃ of 10 ppb in O₃ that would occur uniformly throughout 30 eastern States and the District of Columbia and within an assumed atmospheric mixing layer that ranged up to 2 km in altitude. Assuming no other relevant factors changed over the several decade exposure period that would be required, the resulting increase in NMSC incidence for this extreme scenario was estimated eventually to reach “between 0.6% and 1%.” While these percentages are small—indeed too small to be measurable (Cupitt, 1994)—if taken at face value, they would not necessarily be judged as trivial because of the large baseline of NMSC. For reasons outlined below, however, even these small percentage estimates appear to be substantially overstated and cannot be considered reliable.

The Cupitt paper was never formally published, but it was subjected to internal agency peer review and commentary by experts at EPA's Office of Research and Development (ORD) (Childs, 1994; Altshuller, 1994). While finding the exposition, including recognition of the difficulties in such an approach, to be “very acceptable,” the reviewers noted substantial uncertainties in basic data and concerns about the numerous simplifying assumptions that called the numerical results into significant question. Examples of data uncertainties noted by the reviewers include: (1) The accuracy of column O₃ (in Dobson units) and UV measurements used; (2) the fact, recognized in Cupitt (1994), that the predicted UV-B radiation flux changes are at the “noise” level and could not be reliably detected statistically or attributed to the change in ground-level O₃ concentration; (3) data on effects of aerosols are limited, yet ignoring such effects in estimating the O₃—UV-B radiation relationship was “erroneous;” and (4) data to permit dynamic assessment of the feedback between increased UV radiation and increased O₃ is limited to uncertain models, and this potential feedback mechanism was ignored in the analysis (Childs, 1994).

Reviewers also questioned a number of the simplifying assumptions that could have “substantial impact” on the resulting risk estimates. Among these were: (1) The assumed mixing height of 2 km, which reviewers considered too high on average, especially for the eastern United States (By overstating the thickness of the pollution-related layer of the atmosphere that is the focus of the control strategies designed to attain the NAAQS, this factor would bias the estimates upwards by as much as a factor of 2.); (2) the assumption that the ozone mixing ratio is the same at the earth's surface as it is at 2 km, when the vertical profile varies through the diurnal cycle (Because vertical mixing increases through the day, this assumption would be most important in the earlier portion of daylight hours.); (3) the assumption that neither aerosols nor O₃ production cycles themselves exert either positive or negative feedback on UV-B penetration (As noted in the previous section, a dynamic consideration of these factors could change the direction of the result in particular areas.); (4) the assumption that NMSC might result from episodic exposures, when, in fact, NMSC results from cumulative doses (This assumption affects only separate and far smaller estimates Cupitt made for episodic changes, essentially invalidating those results.); (5) the assumption that all people would be susceptible based on assumed exposure factors; and (6) the assumption that behavioral patterns, demographic patterns, and meteorological factors and other factors related to actual exposures remain constant over time (Childs, 1994; Altschuller, 1994).

These reviewers capsulized their conclusions regarding the quantitative results of this analysis as follows:

In summary, (1) the numbers resulting from these calculations are quite small, and (2) the limitations of the accuracy and reliability of the input to the calculations produces numbers that cannot be defended, whether large or small. (Childs, 1994).

As noted in the discussion above, this is not simply a matter of uncertain and small risk estimates. On balance, several of the problems noted above served to inflate the overall estimates, and, depending upon local conditions and the control strategy assumed, could even call the direction of the results into question for some locations. Further, a significant bias, not highlighted in the cited reviews, is how well the assumed 10 ppb change in daytime O₃ levels averaged over an entire summer season (and over half the U.S.) reflects what might occur in response to the revised O₃ NAAQS.^[52] In fact, this assumed change, as well as the assumptions regarding its spatial and vertical extent, are significantly larger than could reasonably be expected based on the revisions to the O₃ standard promulgated in 1997.

To provide a fair comparison, it is necessary to convert the 1-hour standard into its nearest 8-hour equivalent. As documented in the Staff Paper (U.S. EPA, 1996b), the nearest equivalent 8-hour standard would have a level of about 0.09 ppm. Superficially, this might appear to support a 10 ppb difference compared to the 0.08 ppm 8-hour standard set in 1997, until considering that these standards are stated in reference to extreme high values in the distribution (e.g., the average of the 4th-highest daily maximum concentrations). Cupitt's analysis assumed that a “mixing layer” up to 2 km deep over a very large geographical region would experience a change of 10 ppb in daylight average O₃ for an entire O₃ season. This scenario would require a challenging regional strategy that would, on average, reduce each day for the over 150 day O₃ season by 10 ppb. Yet, the 0.08 ppm 8-hour O₃ standard would require that only the fourth-highest day of the ozone season be reduced by about 10 ppb, as compared to the previous standard. Based on available O₃ trends information, strategies that reduce peak O₃ days would have far less effect on the far more numerous days toward the middle and lower-parts of the O₃ season distribution (e.g., U.S. EPA, 1996a, Figures 4-2, 4-3). In fact, as reported in the Response to Comments document, based on earlier RIA projections of long-term O₃ reductions that might occur with the 0.08 ppm 8-hour O₃ standard, the magnitude of the assumed average change appears to be overstated by more than a factor of 3 (U.S. EPA, 1997). When considered with the excessively high assumed mixing layer, the overly large geographical area requiring reductions (over 30 States), and the assumption that the entire population would be at the same risk as the more sensitive subpopulations, it is EPA's judgment, based on the record, that these readily identified biases could well be on the order of a factor of 10. EPA solicits comment on the assumptions discussed above.

More subtle are the uncertainties and potential bias inherent in an essentially static comparison of two different O₃ values that are assumed to be uniform over a very large area. Dynamic, real-world strategies would involve a number of alternative local and regional scale approaches that vary significantly in time and space, with a variety of possible outcomes with respect to the middle and lower portions of the distribution that is most relevant to estimating long-term summer averages over a period of decades into the future. An example of such local strategy-dependent outcomes would be control of NO_X emissions across a metropolitan area, which could reduce O₃ concentrations at downwind peak monitors, but also result in localized increases in lower concentrations in the center city area (National Academy of Sciences, 1991, Figure 11-2). As noted in section II.B.2 above and in Altshuller (1994), the interrelated indirect results from reduced O₃ and UV-B radiation could trigger feedbacks through increased O₃, aerosol, or cloud cover that could partially or fully offset the initial O₃ effects on UV-B radiation. Available data and assessment tools do not permit a reasonable quantitative assessment of these second-and third-order indirect effects (Altshuller, 1994; Childs, 1994).

Other potential problems associated with ignoring area-specific considerations in an O₃/UV-B risk analysis summarized in the previous section include the assessment of local physical factors (e.g., buildings) that reduce UV-B radiation exposure in outdoor microenvironments, meteorological conditions (e.g., sea breeze) or local emissions patterns that reduce pollution in high UV-B radiation exposure microenvironments, behavioral adjustments to information concerning UV-B radiation risk over time, and local differences in the proportion of sensitive populations. Even Cupitt's assumption that 90 percent of exposure occurs during the summer O₃ season embeds an assumption about long-term personal behavior for which little empirical evidence exists.

In summary, the Cupitt (1994) white paper was useful for its intended purpose as a scoping analysis to identify the potential issues arising in any attempt to assess the potential shielding provided by changes in ground-level O₃. It established that any effects of even fairly large long-term O₃ reductions in ground-level O₃ would be quite small, but as evidenced in the comments of the peer review and the discussion above, available data and modeling tools fall far short of permitting reliable quantitative risk estimates for consideration in standard setting or benefits assessments.

The analysis of this issue by U.S. Department of Energy (DOE) staff (1995) is summarized in a statement submitted as a part of public comments at a CASAC meeting. The exposition is far less complete than that of Cupitt, and it is quite difficult to reconcile the range of estimates for NMSC, the lower bound of which are less than Cupitt, while the upper bound estimates are more than double his. The analysis apparently starts with the same assumptions regarding a constant change in summertime O₃ of 10 ppb through a 2 km mixing layer, but important information about the other assumptions is lacking. In any event, the paper does not appear to improve upon the methodology in the Cupitt analysis.^[53] Given that the U.S. DOE statement must share the limitations outlined above for Cupitt and the fact that the analytical approach is not well documented nor peer reviewed, no reliance is placed on the quantitative results presented in the U.S. DOE submission.

The work of economic analysts Lutter and Wolz (1997) provides a “preliminary analysis” of UV-B radiation screening by tropospheric O³. Here, the exposition permits a more direct comparison with that of Cupitt, and it appears that many of the same simplifying assumptions were used—either explicitly or implicitly. This paper relied upon Cupitt's assumption that the NAAQS revision might bring about a summertime average of 10 ppb reduction in O₃ in areas not attaining the standard. As discussed above, based on the record, EPA believes this substantially overstates the likely effect of the NAAQS revision. Their assumption of a constant mixing ratio for the 10 ppb change that would extend well above the planetary boundary layer, up to 10 km, also introduces upward bias into their upper-bound risk estimates. The resultant apparent dose appears to be a factor of 4 larger than the upper bound used by Cupitt and U.S. DOE staff. The other quantitative inputs to the analysis differed to a more modest degree from those used by Cupitt. In the end, the upper bound estimate of NMSC is more than double that of Cupitt, due largely to the unwarranted assumption of a 10 km mixing height.

Again, because the quantitative assessment shares most of the limitations cited above for Cupitt, and actually adds substantial bias in a key assumption, EPA has placed no reliance on the quantitative risk estimates for NMSC from Lutter and Wolz (1997) or to the secondary estimates derived in the U.S. DOE analyses. EPA solicits comment on the assessments discussed above.

At the end of the 1997 O₃ NAAQS review, EPA published the final RIA, containing, among other requirements, an analysis addressing all of the quantifiable benefits of the O₃ NAAQS. This analysis, which was reviewed by other Federal agencies and approved for release by the Office of Management and Budget (OMB), concluded that the available scientific and technical information would not permit reliable quantitative estimates of any effect of changing the O₃ NAAQS on UV-B radiation-related effects. Based on the present examination of all of the available information in the record, the Administrator believes that this remains a sound conclusion.

C. Consideration of Net Adverse Health Effects of Ground-Level O₃

In considering the net adverse health effects of ground-level O₃, EPA has focused on characterizing and weighing the comparative importance of the potential indirect beneficial health effects associated with the attenuation of UV-B radiation by ground-level O₃ (section II.B above) and the direct adverse health effects associated with breathing O₃ in the ambient air (section II.A above). The same key factors considered by EPA in its 1997 review of the O₃ standard are again considered here in characterizing the additional information on potential beneficial effects and in comparatively weighing this information relative to the direct adverse effects. Beyond quantitative assessments of exposure and risk that were central to EPA's 1997 review, these factors include the nature and severity of the effects, the types of available evidence, the size and nature of the sensitive populations at risk, and the kind and degree of uncertainties in the evidence and assessments. In recognition of the complexity and multidimensional nature of such a comparison, no attempt is made to characterize all the relevant effects or associated risks to public health with a common metric.

The available record information on the potential indirect beneficial health effects associated with ground-level O₃ includes information from studies of health effects caused by exposure to UV-B radiation and studies that focus on the consequences of unnaturally high exposures to UV-B radiation due to depletion of the stratospheric O₃ layer, as well as analyses that attempt to focus specifically on the consequences of assumed changes in tropospheric O₃ levels. The nature and severity of the effects of UV-B radiation exposure on the skin, eye, and immune system are discussed above (section II.B.1), as is the nature of sensitive populations at risk for these effects. These effects, especially on the skin and eye, are generally understood to be associated with long-term cumulative exposure to UV-B radiation and to have long latency periods from cumulative exposures, especially those early in life. People with light skin pigmentation make up the primary at-risk population for effects on the skin, especially for NMSC, while at-risk populations for other effects are not as well understood. For NMSC, uncertainties in the evidence generally relate to uncertainties in the relevant action spectra and BAFs, as well as in factors related to characterizing the severity of the different types of NMSC. Based on the record information, for the other effects, the role of UV-B radiation is less well understood (e.g., as to relevant action spectra, BAFs, the nature of exposures of concern), although cumulative exposure to UV-B radiation is thought to play a causal role. These characterizations are derived from the large body of epidemiologic and toxicologic evidence that served as the basis for the reference document by EPA (1987).

The record includes a quantitative assessment conducted by EPA (1987, App. E) of the health risks associated with changes in exposure to UV-B radiation attributable to changes in the stratospheric O₃ layer. This assessment models the relationship between wide-scale changes in global/regional levels of stratospheric O₃, resulting from emissions of O₃ depleting substances with long-atmospheric lifetimes, and changes in UV-B radiation flux as a function of latitude for three broad regions across the United States.^[54] As discussed above (section II.B.2), because changes in the stratospheric O₃ layer are relatively uniform across broad regions, varying across the U.S. primarily with latitude, information on localized spatial and temporal patterns of exposure-related variables (e.g., changes in ground-level O₃, meteorological conditions, human activity patterns) are not relevant in producing credible estimates of risk associated with changes in stratospheric O₃. This is in sharp contrast to the nature of the information necessary to produce credible estimates of risk associated with changes in exposures to UV-B radiation projected to result from changes in ground-level O₃ that would be associated with attainment of alternative 8-hour standards for O₃.

An evaluation of the available analyses that have produced estimates of health risks associated with changes in ground-level O₃ (section II.B.3 above) identifies major limitations in available information that resulted in the need for the analyses to incorporate broad and unsupportable assumptions. These limitations are particularly important with regard to information on spatial and temporal patterns of changes in ground-level O₃ likely to result from various future emission control strategies, relevant meteorological conditions and atmospheric chemistry leading to a cascade of broader indirect effects, and human demographic and activity patterns likely to result in exposures of concern. For the reasons discussed above, these limitations are judged to be of central importance in any such analysis. Thus, in light of such limitations, the Administrator agrees with internal and external reviewers in proposing to conclude that the available scientific and technical information would not permit credible quantitative estimates of these potential beneficial effects.^[55] Thus, available analyses based on such limited information cannot serve as credible estimates of potential beneficial effects associated with the presence of ground-level O₃ due to man-made emissions of O₃ forming substances.

Further, in setting aside the available quantitative analyses, EPA notes that our above evaluation of a number of critical factors in the analyses provides reasons for believing that the public health impacts of any potential beneficial effects associated with ground-level O₃ are likely very small, albeit unquantifiable at this time (section II.B.2). In giving qualitative consideration to the available evidence on potential indirect beneficial effects of ground-level O₃, EPA believes it is appropriate to weigh this information in the context of the body of evidence on adverse effects caused by direct inhalation exposures to ground-level O₃ that formed the basis for the 1997 O₃ primary standard.

As an initial matter, as discussed in the 1997 final rule, the Administrator focused primarily on quantitative comparisons of risk, exposure, and air quality in selecting both the level (62 FR 38867-8) and form (62 FR 38869-72) of the 1997 O₃ primary standard. More specifically, she looked at comparisons of both those risks to public health that can be explicitly quantified in terms of estimated incidences and the size of the at-risk population (e.g., children) likely to experience adverse effects, as well as those for which quantitative risk information is more limited, but for which quantitative estimates of the number of children likely to experience exposures of concern could be developed (as discussed in section II.A.2 above). In considering these comparisons, she recognized that although there were inherent uncertainties in these estimates, the underlying assessments took into account extensive data bases on the spatial and temporal patterns of air quality and directly relevant human activity patterns likely to result in inhalation exposures of concern. Further, the Administrator took into account CASAC's advice that the assessment methods were appropriate and state-of-the-art, and that the results should play a central role in her decision.

Beyond the quantitative information on direct adverse effects, with regard to the qualitative evidence suggestive of potential serious, chronic adverse effects on public health associated with long-term inhalation exposures, the Administrator judged that such information was too uncertain and not well enough understood at the time to serve as the basis for establishing a more restrictive 8-hour standard in terms of either level (62 FR 38868) or form (62 FR 38871). This conclusion was consistent with CASAC's advice that further research into potential chronic adverse effects in humans should be continued, and the results considered in the next review (62 FR 38871).

In weighing the available information on potential indirect beneficial effects of ground-level O₃, the Administrator considers this information in the same light as the information on potential direct chronic adverse effects associated with long-term inhalation exposures to ground-level O₃. In both instances, the potential health effects are serious and likely to develop over many years, with important periods of exposure likely occurring in childhood. Different population groups are likely affected, however, by these potential adverse and beneficial effects. Urban populations and people with impaired respiratory systems (e.g., people with asthma), who are disproportionately from certain minority groups, are most at-risk for the direct inhalation-related effects, whereas fair-skinned populations are most generally, but not exclusively, at-risk for the indirect beneficial effects related to exposure to UV-B radiation. Although different types of uncertainties are inherent in the record information on these effects, in both cases, the uncertainties related to ground-level O₃ are so great as to preclude the development of credible estimates of the size of the affected population or the probability of the occurrence of such effects. In the case of indirect effects related to ground-level O₃, EPA believes that the use of plausible but unsubstantiated assumptions would likely lead to the conclusion that the potential impacts on public health are likely very small; no such conclusions have yet been drawn with regard to the public health impacts of potential direct chronic adverse effects related to inhalation exposures. After considering these factors, the Administrator now provisionally concludes that, much like the qualitative evidence on direct adverse effects potentially associated with long-term inhalation exposures, the newly considered available evidence on potential indirect beneficial effects is not well enough understood at this time to serve as the basis for establishing a less restrictive 8-hour standard than was promulgated in 1997. Rather, the Administrator believes that the most recent evidence and analyses of potential long-term, indirect beneficial effects should be considered in the next review in conjunction with the most recent information on long-term, direct adverse effects.

D. Proposed Response to Remand on the Primary O ₃ NAAQS

After carefully considering the scientific information available in the record on adverse effects on public health associated with direct inhalation exposures to O₃ in the ambient air and on the potential for indirect benefits to public health associated with the presence of ground-level O₃ and the resultant attenuation of naturally occurring UV-B radiation from the sun, taking into account the weight of that evidence in assessing the net adverse health effects of ground-level O₃, and for the reasons discussed above, the Administrator proposes to respond to the remand by reaffirming the 8-hour primary O₃ standard promulgated in 1997. In proposing to leave unchanged the 1997 O₃ standard at this time, the Administrator has fully considered the available information in the record of the 1997 O₃ NAAQS review on potential beneficial health effects of ground-level O₃. Based on such consideration, she has provisionally determined that the information linking changes in patterns of ground-level O₃ concentrations likely to occur as a result of programs implemented to attain the 1997 O₃ NAAQS to changes in relevant exposures to UV-B radiation of concern to public health is too uncertain at this time to warrant any relaxation in the level of public health protection previously determined to be requisite to protect against the demonstrated direct adverse respiratory effects of exposure to O₃ in the ambient air. Further, the Administrator notes that it is the Agency's view that associated changes in UV-B radiation exposures of concern, using plausible but highly uncertain assumptions about likely changes in patterns of ground-level ozone concentrations, would likely be very small from a public health perspective.

In the past, the Administrator has been confronted with situations where there has been both quantifiable and unquantifiable evidence, and has moved forward with a NAAQS decision. The inability to quantify all related effects does not preclude the Agency from making a NAAQS decision, particularly in situations where there is strong quantifiable evidence of significant adverse health effects. Moreover, in this case, as noted above, EPA believes the potential beneficial effects are not quantifiable at this time and likely very small from a public health perspective. Accordingly, the Administrator believes it is inappropriate to wait for additional information on such effects prior to responding to this remand.

The 0.08 ppm, 8-hour primary standard is met at an ambient air quality monitoring site when the 3-year average of the annual fourth-highest daily maximum 8-hour average O₃ concentration is less than or equal to 0.08 ppm. Data handling conventions are specified in a new appendix I to 40 CFR part 50, as discussed in the 1996 proposal and 1997 final rule.^[56]

In proposing to respond to the remand by reaffirming the 1997 primary O₃ standard at this time, the Administrator recognizes, however, that relevant information on indirect potentially beneficial health effects of ground-level O₃ is now available that was not part of this rulemaking record. In addition, she notes that the next periodic review of the O₃ NAAQS has now been initiated by EPA's ORD with a call for information (65 FR 57810; September 26, 2000). Thus, to ensure that the next review of the O₃ criteria and standards can be based on a comprehensive and current body of relevant scientific information, EPA encourages the submission of new scientific information on the relationships between ground-level O₃, associated attenuation of UV-B radiation and other indirect effects of the presence of O₃ in the ambient air, and effects on public health such as those associated with changes in relevant exposures to UV-B radiation.

In looking ahead to the next review, EPA anticipates that the available information may warrant a fuller examination of relevant public health policy factors in weighing the net adverse health effects associated with ground-level O₃. Such factors could include, for example, the extent to which the proximate cause of the effects is natural or man-made; the extent to which the effects are in excess of naturally occurring background levels; the extent to which the exposures of concern are affected by human behavior patterns; the time course of exposure-response relationships; and environmental justice issues that arise in any analysis of risk trade-offs involving different sensitive populations. To help inform this aspect of the next review, EPA also solicits comments on whether these and other factors should be considered to be relevant in weighing the net adverse health effects of ground-level O₃.

III. Rationale for Proposed Response To Remand on the Secondary O₃ Standard

This notice also presents the Administrator's proposed response to the remand, reaffirming the 8-hour O₃ secondary standard promulgated in 1997, based on: (1) Information from the 1997 criteria and standards review that served as the basis for the 1997 secondary O₃ standard, including the scientific information on welfare effects associated with direct exposures to O₃ in the ambient air, with a focus on vegetation effects, and assessments of vegetation exposure, risk, and economic values and (2) a review of the scientific information in the record of the 1997 review (but not considered as part of the basis for the 1997 standard) on the welfare effects associated with changes in UV-B radiation, the association between changes in ground-level O₃ and changes in UV-B radiation, and predictions of changes in ground-level O₃ levels likely to result from attainment of alternative O₃ standards.

A. Direct Adverse Welfare Effects

As discussed in the 1997 final rule, direct exposures to O₃ have been associated quantitatively and qualitatively with a wide range of vegetation effects such as visible foliar injury, growth reductions and yield loss in annual crops, growth reductions in tree seedlings and mature trees, and effects that can have impacts at the forest stand and ecosystem level. Visible foliar injury can represent a direct loss of the intended use of the plant, ranging from reduced yield and/or marketability for some agricultural species to impairment of the aesthetic value of urban ornamental species. On a larger scale, foliar injury is occurring on native vegetation in national parks, forests, and wilderness areas, and may be degrading the aesthetic quality of the natural landscape, a resource important to public welfare. Growth and yield effects of O₃ have been well documented for numerous species, including commodity crops, fruits and vegetables, and seedlings of both coniferous and deciduous tree species. Although data from tree seedling studies could not be extrapolated to quantify responses to O₃ in mature trees, long-term observational studies of mature trees have shown growth reductions in the presence of elevated O₃ concentrations. Even where these growth reductions are not attributed to O₃ alone, it has been reported that O₃ is a significant contributor that potentially exacerbates the effects of other environmental stresses (e.g., pests). In addition, growth reductions can indicate that plant vigor is being compromised such that the plant can no longer compete effectively for essential nutrients, water, light, and space. When many O₃-sensitive individuals make up a population, the whole population may be affected. Changes occurring within sensitive populations, or stands, if they are severe enough, ultimately can change community and ecosystem structure. Structural changes that alter the ecosystem functions of energy flow and nutrient cycling can alter ecosystem succession.

Based on key studies and other biological effects information reported in the Criteria Document and Staff Paper, it was recognized that peak O₃ concentrations equal to or greater than 0.10 ppm can be phytotoxic to a large number of plant species, and can produce acute foliar injury and reduced crop yield and biomass production. In addition, O₃ concentrations within the range of 0.05 to 0.10 ppm have the potential over a longer duration of creating chronic stress on vegetation that can result in reduced plant growth and yield, shifts in competitive advantages in mixed populations, decreased vigor leading to diminished resistance to pest and pathogens, and injury from other environmental stresses. Some sensitive species can experience foliar injury and growth and yield effects even when O₃ concentrations never exceed 0.08 ppm. Further, the available scientific information supports the conclusion that a cumulative seasonal exposure index is more biologically relevant than a single event or mean index.

To put judgments about these vegetation effects into a broader national perspective, the Administrator has taken into account the extent of exposure of O₃-sensitive species, potential risks of adverse effects to such species, and monetized and non-monetized categories of increased vegetation protection associated with reductions in O₃ exposures. In so doing, the Administrator recognized that markedly improved air quality, and thus significant reductions in O₃ exposures would result from attainment of the 0.08 ppm, 8-hour primary standard. In looking further at the incremental protection associated with attainment of a seasonal secondary standard, she recognized that areas that would likely be of most concern for effects on vegetation, as measured by the seasonal exposure index, would also be addressed by the 0.08 ppm, 8-hour primary standard.

B. Potential Indirect Beneficial Welfare Effects

This section is drawn from the limited information in the record of the 1997 review with regard to the effect of ground-level O₃ on the attenuation of UV-B radiation and potential associated welfare benefits.^[57] While this information suggests the potential for effects on plants and aquatic organisms, EPA (1987, ES-40—ES-43) recognizes that relevant studies are limited and the uncertainties are great due in part to problems in study designs, such that quantitative conclusions cannot be drawn.

With regard to effects on vegetation, while some plant cultivars tested in the laboratory were determined to be sensitive to UV-B radiation exposure, these experiments have been shown to inadequately replicate effects in the field, such that they do not reflect the complex interactions between plants and their environment. The only long-term field studies of crops involved soybeans, producing suggestive evidence of reduced yields under conditions simulating changes in total column O₃ over an order of magnitude greater than those projected to occur as a result of changes in ground-level O₃ associated with attainment of the 1997 O₃ NAAQS. Beyond the limited studies of crops, EPA (1987, ES-41) notes that little or no data exist on UV-B radiation effects on trees and other types of natural vegetation, or on possible interactions with pathogens. While it is noted that changes in UV-B radiation levels could alter the results of competition in natural ecosystems, no evidence is available to evaluate this effect. Further, it is recognized that UV-B radiation may both inhibit and stimulate plant flowering, depending on the species and growth conditions. Recognizing that interactions between UV-B radiation and other environmental factors are important in determining potential UV-B radiation effects on plants, EPA (1987, ES-42) notes that extensive, long-term studies would be required to address these interactions.

With regard to effects on aquatic organisms, EPA (1987, ES-42) notes that while initial experiments show that increased UV-B radiation has the potential to harm aquatic life, difficulties in experimental designs and the limited scope of the studies prevent the quantification of potential risks. Some study results suggest that most zooplankton show no effect due to increased exposure to UV-B radiation up to some threshold exposure level, with exposures above such threshold levels eliciting notable effects. For species under UV-B stress, such effects could include reduced time spent at the surface of the water, which is critical for breeding in some species, possibly leading to changes in species diversity. It is also noted that, as do all other living organisms, aquatic biota cope with exposure to UV-B radiation by avoidance, shielding, and repair mechanisms, although uncertainty exists as to the extent to which such mitigation mechanisms would occur (U.S. EPA, 1987, ES-43). It is recognized that determination of UV-B radiation exposure in aquatic systems is complex because of the variable attenuation of UV-B radiation in the water column, and that further research is needed to improve our understanding of how UV-B radiation exposure affects marine species, particularly given their world-wide importance as a source of protein.

C. Proposed Response To Remand on the Secondary O₃ NAAQS

After considering the scientific information available in the record on adverse welfare effects associated with direct exposure to O₃ in the ambient air and on the potential indirect benefits to public welfare related to attenuation of naturally occurring UV-B radiation, the Administrator provisionally concludes that there is insufficient information available on UV-B radiation-related effects to warrant any relaxation in the level of public welfare protection previously determined to be requisite to protect against the demonstrated direct adverse effects of exposure to O₃ in the ambient air. Thus, the Administrator proposes to respond to the remand by reaffirming the 8-hour secondary O₃ standard promulgated in 1997, which is identical to the 8-hour primary O₃ standard.

As recognized above in section II.B.4 with regard to consideration of health effects, the Administrator also recognizes that relevant information on indirect potentially beneficial welfare effects of ground-level O₃ is now available that was not part of this rulemaking record. In addition, as previously noted, the next periodic review of the O₃ NAAQS is now being initiated by EPA's ORD with a call for information. Thus, to ensure that the next review of the O₃ criteria and standards can be based on a comprehensive and current body of relevant scientific information, EPA encourages the submission of new scientific information on the relationships between ground-level O₃, associated attenuation of UV-B radiation and other indirect effects of the presence of O₃ in the ambient air, and effects on public welfare such as those associated with changes in relevant exposures to UV-B radiation.

IV. Administrative Requirements

A. Executive Order 12866: OMB Review of “Significant Actions”

Under Executive Order 12866, the Agency must determine whether a regulatory action is “significant” and, therefore, subject to OMB review and the requirements of the Executive Order. The order defines “significant regulatory action” as one that may:

(1) Have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities;

(2) Create a serious inconsistency or otherwise interfere with an action taken or planned by another Agency;

(3) Materially alter the budgetary impact of entitlements, grants, user fees, or loan programs or the rights and obligations or recipients thereof; or

(4) Raise novel legal or policy issues arising out of legal mandates, the President's priorities, or the principles set forth in the Executive Order.

In view of its important policy implications, this proposed action has been judged to be a “significant regulatory action” within the meaning of the Executive Order. The EPA has submitted this proposed action to OMB for review. Changes made in response to OMB suggestions or recommendations will be documented in the public record and made available for public inspection at EPA's Air and Radiation Docket and Information Center (Docket No. A-95-58).

Since today's proposed response to the remand is a reaffirmation of the revisions to the O₃ NAAQS previously promulgated in 1997, no new RIA has been prepared. The RIA (1997) prepared in conjunction with the 1997 revision to the O₃ NAAQS is available in the docket, from EPA at the address under “Availability of Related Information,” and in electronic form as discussed above in “Electronic Availability.”

The Clean Air Act and judicial decisions make clear that the economic and technological feasibility of attaining ambient standards are not to be considered in setting NAAQS, although such factors may be considered in the development of State plans to implement the standards. Accordingly, although a RIA was prepared for the 1997 decision to revise the O₃ NAAQS, neither that RIA nor the associated contractor reports have been considered in issuing this proposal.

B. Executive Order 13045: Children's Health

Executive Order 13045, entitled “Protection of Children from Environmental Health Risks and Safety Risks” (62 FR 19885, April 23, 1997), requires Federal agencies to ensure that their policies, programs, activities, and standards identify and assess environmental health and safety risks that may disproportionately affect children. To respond to this order, agencies must explain why the regulation is preferable to other potentially effective and reasonably feasible alternatives considered by the agency.

Today's proposed response to the remand, reaffirming the 1997 primary O₃ NAAQS, specifically takes into account children as the group most at risk to the direct inhalation-related effects of O₃ exposure, and was based on studies of effects on children's health (U.S. EPA, 1996a; U.S. EPA, 1996b) and assessments of children's exposure and risk (Johnson et al., 1994; Johnson et al., 1996a,b; Whitfield et al., 1996; Richmond, 1997). The 1997 revision to the primary O₃ NAAQS was promulgated to provide adequate protection to the public, especially children, against a wide range of direct O₃-induced health effects, including decreased lung function, primarily in children who are active outdoors; increased respiratory symptoms, primarily in highly sensitive individuals; hospital admissions and emergency room visits for respiratory causes, among children and adults with respiratory disease; inflammation of the lung and possible long-term damage to the lungs. This proposed response to the remand affirming the 1997 primary O₃ NAAQS maintains the level of protection of children's health established by the standard set in 1997. Therefore, today's proposed action does comply with the requirements of E.O. 13045.

C. Executive Order 13132: Federalism

Executive Order 13132, entitled “Federalism” (64 FR 43255, August 10, 1999), requires EPA to develop an accountable process to ensure “meaningful and timely input by State and local officials in the development of regulatory policies that have federalism implications.” “Policies that have federalism implications” is defined in the Executive Order to include regulations that have “substantial direct effects on the States, on the relationship between the national government and the States, or on the distribution of power and responsibilities among the various levels of government.”

Today's proposed response to the remand does not have federalism implications. It will not have substantial direct effects on the States, on the relationship between the national government and the States, or on the distribution of power and responsibilities among the various levels of government, as specified in Executive Order 13132. The proposed response to the remand only reaffirms the previously promulgated ozone standard and would not alter the relationship that has existed under the Clean Air Act for 30 years, in which EPA sets NAAQS and the states implement them through submission of SIPs, in accordance with the requirements of the Clean Air Act. Thus, Executive Order 13132 does not apply to this action. In the spirit of Executive Order 13132, and consistent with EPA policy to promote communications between EPA and State and local governments, EPA specifically solicits comment on this proposed action from State and local officials.

D. Executive Order 13175: Consultation and Coordination With Indian Tribal Governments

Executive Order 13175, entitled “Consultation and Coordination with Indian Tribal Governments” (65 FR 67249, November 6, 2000), requires EPA to develop an accountable process to ensure “meaningful and timely input by tribal officials in the development of regulatory policies that have tribal implications.” “Policies that have tribal implications” is defined in the Executive Order to include regulations that have “substantial direct effects on one or more Indian tribes, on the relationship between the Federal government and the Indian tribes, or on the distribution of power and responsibilities between the Federal government and Indian tribes.”

This proposed response to the remand does not have tribal implications. It will not have substantial direct effects on tribal governments, on the relationship between the Federal government and Indian tribes, or on the distribution of power and responsibilities between the Federal government and Indian tribes, as specified in Executive Order 13175. This is because this proposed response to the remand leaves unchanged the 1997 final rule. Thus, Executive Order 13175 does not apply to this rule.

E. Unfunded Mandates Reform Act

Title II of the Unfunded Mandates Reform Act of 1995 (UMRA), Public Law 104-4, establishes requirements for Federal agencies to assess the effects of their regulatory actions on State, local, and tribal governments and the private sector. Under section 202 of the UMRA, EPA generally must prepare a written statement, including a cost-benefit analysis, for proposed and final rules with “Federal mandates” that may result in expenditures to State, local, and tribal governments, in the aggregate, or to the private sector, of $100 million or more in any one year. Before promulgating an EPA rule for which a written statement is needed, section 205 of the UMRA generally requires EPA to identify and consider a reasonable number of regulatory alternatives and adopt the least costly, most cost-effective or least burdensome alternative that achieves the objectives of the rule. The provisions of section 205 do not apply when they are inconsistent with applicable law. Moreover, section 205 allows EPA to adopt an alternative other than the least costly, most cost-effective or least burdensome alternative if the Administrator publishes with the final rule an explanation why that alternative was not adopted. Before EPA establishes any regulatory requirements that may significantly or uniquely affect small governments, including tribal governments, it must have developed under section 203 of the UMRA a small government agency plan. The plan must provide for notifying potentially affected small governments, enabling officials of affected small governments to have meaningful and timely input in the development of EPA regulatory proposals with significant Federal intergovernmental mandates, and informing, educating, and advising small governments on compliance with the regulatory requirements.

As noted above, EPA cannot consider in setting a NAAQS the economic or technological feasibility of attaining ambient air quality standards, although such factors may be considered to a degree in the development of State plans to implement the standards. Accordingly, and for the reasons discussed in the 1996 proposal and 1997 final rule, EPA has determined that the provisions of sections 202, 203, and 205 of the UMRA do not apply to this proposed action. The EPA acknowledges, however, that any corresponding revisions to associated State implementation plan requirements and air quality surveillance requirements, 40 CFR part 51 and 40 CFR part 58, respectively, might result in such effects. Accordingly, EPA will address unfunded mandates as appropriate when it proposes any revisions to 40 CFR parts 51 and 58.

F. Regulatory Flexibility Analysis/Small Business Regulatory Enforcement Fairness Act

Under the Regulatory Flexibility Act (RFA), 5 U.S.C. 601 et seq., EPA must prepare a regulatory flexibility analysis assessing the impact of any proposed or final rule on small entities. Under 6 U.S.C. 605(b), this requirement may be waived if EPA certifies that the rule will not have a significant economic impact on a substantial number of small entities. Small entities include small businesses, small not-for-profit enterprises, and governmental entities with jurisdiction over populations less than 50,000 people.

Today's proposed response to the remand, reaffirming the 1997 primary O₃ NAAQS, does not establish any new regulatory requirements affecting small entities. On the basis of the above considerations and for the reasons discussed in the 1996 proposal and 1997 final rule, EPA certifies that today's proposed action will not have a significant economic impact on a substantial number of small entities within the meaning of the RFA, as affirmed by the D.C. Circuit in American Trucking Associations v. EPA, 175 F. 3d 1027 (D.C. Cir. 1999). Based on the same considerations, EPA also certifies that the new small-entity provisions in section 244 of the Small Business Regulatory Enforcement Fairness Act (SBREFA) do not apply.

G. Paperwork Reduction Act

Today's proposed response to the remand does not establish any new information collection requirements beyond those which are currently required under the Ambient Air Quality Surveillance Regulations in 40 CFR part 58 (OMB #2060-0084, EPA ICR No. 0940.15). Therefore, the requirements of the Paperwork Reduction Act do not apply to today's proposed action.

H. National Technology Transfer and Advancement Act

Section 12(d) of the National Technology Transfer and Advancement Act of 1995 (NTTAA), Public Law No. 104-113, Section 12(d) (15 U.S.C. 272 note) directs EPA to use voluntary consensus standards in its regulatory activities unless to do so would be inconsistent with applicable law or otherwise impractical. Voluntary consensus standards are technical standards (e.g., materials specifications, test methods, sampling procedures, and business practices) that are developed or adopted by voluntary consensus standards bodies. The NTTAA directs EPA to provide Congress, through OMB, explanations when the Agency decides not to use available and applicable voluntary consensus standards. Today's proposed response to the remand does not involve technical standards. Therefore, EPA did not consider the use of any voluntary consensus standards.

I. Executive Order 13211: Energy Effects

This proposed response to remand is not a “significant energy action” as defined in Executive Order 13211, “Actions Concerning Regulations That Significantly Affect Energy Supply, Distribution, or Use” (66 FR 28355 (May 22, 2001)) because it is not likely to have a significant adverse effect on the supply, distribution, or use of energy. This is because this proposed response to the remand leaves unchanged the 1997 final rule. Thus, Executive Order 13211 does not apply to this rule.

V. References

Altschuller, A.P. (1994) Memorandum to L.T. Cupitt re: Addendum to My Review of Your Manuscript “Calculations of the Impact of Tropospheric Ozone Changes On UV-B Flux and Potential Skin Cancers” EPA Docket A-95-54, IV-D-2694, Appendix B 17.

American Thoracic Society. (1985) Guidelines as to what constitutes an adverse respiratory health effect, with special reference to epidemiologic studies of air pollution. American Review of Respiratory Disease. 131: 666-668.

Brii­hl, C. and Creutzen, P.J. (1989) On the Disproportionate Role of Tropospheric Ozone as a Filter Against Solar UV-B Radiation. Geophys. Res. Letters, 16:703-706. Docket A-9-54, IV-D-2694, Appendix B 10.

Childs, N. (1994) Memorandum to L. Grant re: Relationships of Reductions in Tropospheric Ozone to UV-B Penetration to Earth's Surface, EPA Docket A-95-54, IV-D-2694, Appendix B 16.

Creutzen, P.J. (1992) Ultraviolet on the Increase Nature 356:104-105. Docket A-95-54, IV-D-2694, Appendix B 11.

Cupitt, L.T. (1994) Draft memorandum, Calculations of the Impact of Tropospheric Ozone Changes on UV-B Flux and Potential Skin Cancers, EPA ORD/AREAL Docket A-95-54, IV-D-2694, Appendix B 2.

Fishman, J.; Watson, C.E.; Larsen, J.C.; and Logan, J.A. (1990) Distribution of Tropospheric Ozone Determined From Satellite Data J. Geophys. Res. 95:3599-3617. Docket A-95-54, IV-D-2694, Appendix B 1.

Frederick, J.E.; Koob, A.E.; Weatherhead, E.C. (1993) Empirical Studies of Tropospheric Transmission in the Ultraviolet: Broadband Measurements. J. Applied Meteorology 32:1883—1892. Docket A-95-54, IV-D-2694, Appendix B 13.

Galindo, I.; Frenk, S.; Bravo, H. (1995) Ultraviolet Irradiance Over Mexico City J. Air and Waste Manage. Assoc. 45:886-892. Docket A-95-54, IV-D-2694, Appendix B 14.

Johnson, T. (1994) Letter report: Enhancements to the pNEM summer camp methodology. Prepared by IT/Air Quality Services for U.S. EPA, OAQPS; Research Triangle Park, NC, March.

Johnson, T.; Capel, J.; Mozier, J.; McCoy, M. (1996a) Estimation of ozone exposures experienced by outdoor children in nine urban areas using a probabilistic version of NEM. Prepared by IT/Air Quality Services for U.S. EPA, OAQPS; Research Triangle Park, NC, August.

Johnson, T.; Capel, J.; McCoy, M.; Mozier, J. (1996b) Estimation of ozone exposures experienced by outdoor workers in nine urban areas using a probabilistic version of NEM. Prepared by IT/Air Quality Services for U.S. EPA, OAQPS; Research Triangle Park, NC, August.

Johnson, T.; Capel, J.; McCoy, M. (1996c) Estimation of ozone exposures experienced by urban residents using a probabilistic version of NEM. Prepared by IT/Air Quality Services for U.S. EPA, OAQPS; Research Triangle Park, NC, April.

Liu, S.C.; Mckeen, S.A.; and Madronich, S. (1991) Effect of Anthropogenic Aerosols on Biologically Active Ultraviolet Radiation Geophysical research Letters 18:2265-2268. Docket A-95-54, IV-D-2694, Appendix B 3.

Lutter, R. and Wolz, C. (1997) UV-B Screening by Tropospheric Ozone: Implications for the National Ambient Air Quality Standard. Envir. Sci. Technol. 31:142A-146A. Docket A-9-54, IV-D-2694, Appendix B 7.

Madronich, S. (1992) Implications of Recent Total Atmospheric Ozone Measurements for Biologically Active Ultraviolet Radiation Reaching Earth's Surface, Geophys. Res. Letters 19:37-40. Docket A-95-54, IV-D-2694, Appendix B 8.

National Academy of Sciences (1991) Rethinking the Ozone Problem in Urban and Regional Air Pollution, National Academy Press, Washington, District of Columbia.

Richmond (1997) Supplemental ozone exposure and health risk analyses. Internal memorandum from Harvey M. Richmond to Karen M. Martin, U.S. EPA, AQSSD/OAQPS/OAR, RTP, NC, dated February 11, 1997. Docket No. A-95-58 Item IV-A-1.

Seckmeyer, G. and McKenzie, R.L. (1992) Letters to Nature Increased Ultraviolet Radiation in New Zealand (45° S) Relative to Germany (48° N) Nature 359:135-137, Docket A-95-54, IV-D-2694, Appendix B 12.

Thiele, J.J.; Traber, M.G.; Tsang, K.; Cross, C.E.; Packer, L. (1997) In vivo exposure to ozone depletes vitamins C and E and induces lipid peroxidation in epidermal layers of murine skin. Free Radical Biol. Med. 23:385-391.

Thurston, G.D.; Ito, K.; Kinney, P.L.; Lippmann, M. (1992) A multi-year study of air pollution and respiratory hospital admissions in three New York State metropolitan areas: results for 1988 and 1989 summers. Journal of Exposure Analysis and Environmental Epidemiology. 2:429-450.

UNEP (1998) Environmental Effects of Ozone Depletion; Elsevier Science S.A., The Netherlands.

U.S. DOE (1995) Statement of Marvin Frazier at Clean Air Scientific Advisory Committee (CASAC) Ozone Review Panel, Public Meeting, March 21, 1995, at 203-218 transcript. Docket A-95-54, IV-D-2694, Appendix B 9.

U.S. EPA (1987) Assessing the Risk of Trace Gases That Can Modify the Stratosphere. Volume 1 Executive Summary, Docket A-95-54, IV-D-2694, Appendix B 4.

U.S. EPA (1995a) Fact sheet: Health Effects of Overexposure to the Sun, EPA 430-F-95-003, Office of Air and Radiation Docket A-95-54, IV-D-2694, Appendix B 5.

U.S. EPA (1995b) Fact sheet: UV Radiation, EPA 430-F-95-003, Office of Air and Radiation, Docket A-95-54, IV-D-2694, Appendix B 18.

U.S. EPA (1996a) Air quality criteria for ozone and related photochemical oxidants. Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report nos. EPA/600/AP-93/004a-c.

U.S. EPA (1996b) Review of the national ambient air quality standards for ozone: assessment of scientific and technical information. OAQPS staff paper. Research Triangle Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-452/R-96-007. Available from: NTIS, Springfield, VA; PB96-203435.

U.S. EPA (1997) Responses to Significant Comments on the 1996 Proposed Rule on the National Ambient Air Quality Standards for Ozone, Office of Air and Radiation, Docket A-95-54, July 1997.

U.S. EPA (1999a) Science Advisory Board Committee: Advisory Council on Clean Air Act Compliance Analysis (Council) Health and Ecological Effects Subcommittee (HEES), Summary Minutes of Public Meeting, Date: June 28-29, 1999.

U.S. EPA (1999b) The Benefits and Costs of the Clean Air Act 1990 to 2010; Office of Policy Analysis and Review; EPA report no. EPA-410-R-99-001.

Whitfield, R.G.; Biller, W.F.; Jusko, M.J.; Keisler, J.M. (1996) A probabilistic assessment of health risks associated with short-term exposure to tropospheric ozone. Report prepared for U.S. EPA, OAQPS. Argonne National Laboratory; Argonne, IL.

Wolff, G.T. (1995a) Letter from Chairman of the Clean Air Scientific Advisory Committee to the EPA Administrator, dated November 28, 1995. EPA-SAB-CASAC-LTR-96-002.

Wolff, G.T., (1995b) Letter from Chairman of Clean Air Scientific Advisory Committee to the EPA Administrator, dated November 30, 1995. EPA-SAB-CASAC-LTR-96-002.

Wolff, G.T., (1996) Letter from Chairman of Clean Air Scientific Advisory Committee to the EPA Administrator, dated April 4, 1996. EPA-SAB-CASAC-LTR-96-006.

WMO (1994) Scientific Assessment of Ozone Depletion: 1994, Global Ozone Research and Monitoring Project—Report No. 37, published in 1995; Office of Air and Radiation Docket A-95-54, IV-D-2694, Appendix B25.

WMO (1998) Scientific Assessment of Ozone Depletion: 1998, Global Ozone Research and Monitoring Project—Report No. 44, published in 1999.

List of Subjects in 40 CFR Part 50

• Environmental protection
• Air pollution control
• Carbon monoxide
• Lead
• Nitrogen dioxide
• Ozone
• Particulate matter
• Sulfur oxides

Dated: October 31, 2001.

Christine Todd Whitman,

Administrator.

Footnotes

1.  The form of a standard refers to the air quality statistic that is used to determine whether an area attains the standard.

2.  Welfare effects as defined in section 302(h) (42 U.S.C. 7602(h)) include, but are not limited to, “effects on soils, water, crops, vegetation, man-made materials, animals, wildlife, weather, visibility, and climate, damage to and deterioration of property, and hazards to transportation, as well as effects on economic values and on personal comfort and well-being.”

3.  In a November 28, 1995 letter from the CASAC chair to the Administrator, CASAC advised that the final draft Criteria Document “provides an adequate review of the available scientific data and relevant studies of ozone and related photochemical oxidants” (Wolff, 1995a).

4.  The Staff Paper evaluates policy implications of the key studies and scientific information in the Criteria Document, identifies critical elements that EPA staff believes should be considered, and presents staff conclusions and recommendations of suggested options for the Administrator's consideration.

5.  In separate letters from the CASAC chair to the Administrator, CASAC advised that the primary standard and secondary standard sections of the final draft Staff Paper provide “an adequate scientific basis for making regulatory decisions” concerning the O₃ standards (Wolff, 1995b, 1996).

6.  This docket incorporates by reference the docket from the previous O₃ NAAQS review (Docket No. A-92-17) and the docket established for the Criteria Document (Docket No. ECAO-CD-92-0876).

7.  These 1-hour O₃ standards were originally set in 1979 (44 FR 8202, February 8, 1979) and reaffirmed in 1993 (58 FR 13008, March 9, 1993).

8.  The 1-hour standards are attained when the expected number of days per calendar year with maximum hourly average concentrations above 0.12 ppm is equal to or less than one, averaged over 3 years (as determined by 40 CFR part 50, appendix H).

9.  For the reasons discussed in the Response to Comments (U.S. EPA, 1997, pp. 128-135), EPA did not consider in the 1997 review adverse health effects caused by the potential increase in UV-B radiation that could result from reductions in ground-level O₃ brought about by control programs implemented to attain a revised O₃ NAAQS.

10.  UV-B radiation refers to the region of the solar spectrum within the range of wavelengths generally from 280-290 nanometers (nm) at the lower end, to 315-320 nm at the upper end.

11.  For example, in 1977 and again in 1990, Congress added provisions to the Act to address stratospheric O₃ depletion and the resultant increase in exposure to UV-B radiation.

12.  Title VI replaced the provisions regarding stratospheric O₃ depletion enacted in 1977. 42 U.S.C. 7671.

13.  Both the Act and the Montreal Protocol, however, provide for limited “essential use exemptions” for the continued production and import of very small quantities of CFCs and other O₃ depleting substances needed for certain essential uses, for example, for metered dose inhalers used by people with asthma and other respiratory diseases.

14.  Information about the UV Index is available from the EPA Stratospheric Ozone Hotline at [phone] or at http://www.epa.gov/​sunwise/​uvindex.html.

15.  Information about EPA's SunWise School Program is available at http://www.epa.gov/​sunwise/​.

16.  In complying with the direction of the Court of Appeals in its remand on the health benefit issue, we have considered the large amount of relevant information in the record of the 1997 review, and in doing so, have based this proposed response on all the information available to the court in reaching its decision.

17.  See the 1996 proposal and 1997 final rule for more complete summaries and the Criteria Document and Staff Paper for more detailed discussion.

18.  “Acute” health effects of O₃ are defined as those effects induced by short-term and prolonged exposures to O₃. Examples of these effects are functional, symptomatic, biochemical, and physiologic changes.

19.  The 1-hour O₃ primary NAAQS set in 1979 was generally based on these acute effects associated with heavy exercise and short-term exposures.

20.  “Chronic” health effects of O₃ are defined as those effects induced by long-term exposures to O₃. Examples of these effects are structural damage to lung tissue and accelerated decline in baseline lung function.

21.  Differing views have been expressed by CASAC panel members regarding the use of the term “lesion” to describe the O₃-induced morphological (i.e., structural) abnormalities observed in toxicological studies. Section V.C.8 of the Staff Paper describes and discusses these degenerative changes in more detail.

22.  Exertion increases the amount of O₃ entering the airways and can cause O₃ to penetrate to peripheral regions of the lung where lung tissue is more likely to be damaged.

23.  While not necessarily more responsive than healthy individuals in terms of the magnitude of pulmonary function decrements or symptomatic responses, these individuals may be at increased risk since the impact of O₃-induced responses on already-compromised respiratory systems may more noticeably impair an individual's ability to engage in normal activity or may be more likely to result in increased self-medication or medical treatment.

24.  These gradations and impacts are summarized in the 1996 proposal and discussed in the Criteria Document (Chapter 9) and Staff Paper (section V.F, Tables V-4 and V-5).

25.  See the 1996 proposal (61 FR 65723-6) and 1997 final rule (62 FR 38860-1) for a more complete summary of these assessments. A detailed description of the exposure and risk models and their application at the time of the 1996 proposal are presented in the Staff Paper and associated technical support documents (Johnson et al., 1994; Johnson et al., 1996 a,b; McCurdy, 1994a; Whitfield et al., 1996). Following proposal, supplemental exposure and risk analyses were done to analyze the specific standard proposed and alternative standards on which comment was solicited, as well as to refine the procedures used to simulate O₃ concentrations upon attainment of alternative standards (Richmond, 1997).

26.  The areas include a significant fraction of the U.S. urban population, 41.7 million people, the largest urban areas with major O₃ nonattainment problems, and two large urban areas that are in attainment with the 1-hour NAAQS.

27.  Estimates of “people exposed” reflect the number of people who experience exposures to a given concentration of O₃, or higher, at least one time during the period of analysis, and estimates of “occurrences of exposure” reflect the number of times a given O₃ concentration is experienced by the population of interest.

28.  “Exposures of concern” refer throughout to O₃ exposures at and above 0.08 ppm, 8-hour average, at moderate exertion. Such exposures are particularly relevant to a consideration of a number of health effects, discussed in section I.A.1 above, that have been observed in controlled human studies under these exposure conditions, but for which data were too limited to allow for quantitative risk assessment. Exposures at and above 0.12 ppm, 1-hour average, at heavy exertion, are also of concern; however, the focus here is on 8-hour average exposures since exposure estimates are higher for the 8-hour average effects level of 0.08 ppm at moderate exertion than for the 1-hour average effects level of 0.12 ppm at heavy exertion.

29.  The five indoor and two outdoor microenvironments included in this exposure model account for the highly localized variations in O₃ concentrations to which people are exposed that are not directly reflected in the concentrations measured at ambient ground-level O₃ monitoring sites.

30.  See, for example, Tables V-8 and V-9 in the Staff Paper, pp. 83-84.

31.  As discussed in section IV and appendix A of the Staff Paper.

32.  The observed area-to-area variability reflects differences in the shape of air quality distributions and differences in the relationships between 1-hour and 8-hour peak concentrations across urban areas, as well as differences in the percentage of homes with air conditioning (which impacts exposure estimates when individuals are indoors) and the frequency of warm versus cool days (which impacts exposure estimates because different sets of human activity patterns are used for warm versus cool days in the exposure model) across the nine urban areas (Richmond, 1997).

33.  Based on the supplemental analyses that used the third-highest concentration-based form of the standards (Richmond, 1997).

34.  A more complete discussion of uncertainties and limitations is presented in the Staff Paper and technical support documents (Johnson et al., 1996a,b; Richmond, 1997).

35.  Each of the effects is associated with a particular averaging time and, for most of the acute (1- to 8-hour) responses, effects also are estimated separately for specific ventilation ranges [measured as equivalent ventilation rate (EVR)] that correspond to the EVR ranges observed in the studies used to derive exposure-response relationships.

36.  While these studies only included adults aged 18-35, findings from other clinical studies and summer camp field studies in several locations across the U.S. and Canada indicate changes in lung function in healthy children similar to those observed in healthy adults exposed to O₃ under controlled laboratory conditions.

37.  Based on the supplemental analyses that used the third-highest concentration-based form of the standards (Richmond, 1997).

38.  A more complete discussion of assumptions and uncertainties is presented in the Staff Paper and the technical support documents (Whitfield et al., 1996; Richmond, 1997).

39.  Several studies, mainly conducted in the northeastern U.S. and southeastern Canada have reported excess daily respiratory-related hospital admissions associated with elevated O₃ levels within the general population and, more specifically, for individuals with asthma.

40.  The model is described in more detail in Whitfield et al. (1996) and results from the supplemental analysis are presented in Richmond (1997).

41.  A more complete discussion of these uncertainties and limitations is presented in the Staff Paper and technical support documents (Whitfield et al., 1996; Richmond, 1997).

42.  Evidence of this effect is found in Galindo et al., (1995), who reported on the increased risk of rickets associated with decreased incident UV-B radiation due to air pollution.

43.  The reference document available in the record for the information in this section is the EPA document “Assessing the Risk of Trace Gasses that Can Modify the Stratosphere” (U.S. EPA, 1987.)

44.  More recent estimates or mortality rates from NMSC may be found on the American Cancer Society's Web site http://www.cancer.org,, under cancer type “Skin, Nonmelanoma,” then under “Nonmelanoma Skin Cancer—Overview.”

45.  The shorter (blue) wavelengths of visible light are, however, scattered by atmospheric gases, which is responsible for the “blue” sky characteristic of days with low pollution and less than full cloud cover.

46.  The mixing layer (relevant to the vertical “thickness” of ground-level O₃) develops and grows in height through the day.

47.  The RAF is defined as the percent increase in effective dose divided by the percent decrease in total column zone (Madronich, 1992).

48.  For reasons discussed below, any such shielding would vary widely from day to day, even in the summer O₃ season.

49.  This estimated continental background is due in part to natural sources of emissions in North America and in part to the long-range transport of emissions from both anthropogenic and natural sources outside of North America.

50.  Adding to the complexity of understanding this relationship are the results of high-dose animal toxicology studies that suggest more research is needed into the direct effects of ground-level O₃ on the skin. Tests by Thiele et al. (1997) suggest that long-term exposure to O₃ can deplete vitamin E in the skin, and this could make the skin more susceptible to the effects of UV-B radiation (U.S. EPA, 1997). Therefore, reducing long-term ground-level O₃ exposure might serve to reduce skin problems. Even a relatively small O₃ effect here could partially or completely offset any small UV-B radiation mediated effect estimated based on O₃—UV-B interactions alone.

51.  Because of the high baseline risk of effects under natural conditions, as well as the increased risk posed by stratospheric O₃ depletion, medical authorities and governmental bodies have developed campaigns to effect such changes in behavior. The EPA and the National Weather Service (NWS) developed the UV Index. The Index provides a forecast of the expected risk of overexposure to the sun and indicates the degree of caution that should be taken when working, playing, or exercising outdoors. The EPA also developed the SunWise School Program to be used in conjunction with the UV Index. This program is designed to educate the public, especially children and their care givers, about the health risks associated with overexposure to UV radiation and encourage simple and sensible behaviors that can reduce the risk of sun-related health problems later in life (U.S. EPA, 1995a, b).

52.  Cupitt provides no rationale for the selection for this value where it first appears in a Table, which is characterized as addressing “questions from OMB.”

53.  In addition to estimates for NMSC, the U.S. DOE statements also provided estimates for melanoma skin cancers and cataracts. As discussed above, the quantitative relationship between cumulative UV-B exposure and the latter effects are not as well established as for NMSC. Given the lack of documentation and the additional uncertainties over those for NMSC, neither the U.S. DOE estimates of such effects nor the uncritical reliance on them by Lutter and Wolz (1997) should not given quantitative credence.

54.   Since the EPA's 1987 risk assessment on stratospheric ozone depletion, numerous changes have been made to the model to reflect the commitments made since 1987 by the United States, under amendments to the Montreal Protocol, for reductions in production of various ozone depleting chemicals and to incorporate more accurately the latest scientific information.

55.  This conclusion was also reached by the Health and Ecological Effects Subcommittee of the Advisory Council on Clean Air Compliance Analysis, a part of EPA's Science Advisory Board, in conjunction with their review of the “The Benefits and Costs of the Clean Air Act 1990 to 2010” (EPA, 1999b)

56.  Subsequent to the 1997 final rule, EPA has promulgated further revisions to 40 CFR part 50 with regard to the applicability of the 1-hour O₃ standards (65 FR 45182; July 20, 2000). In addition, EPA notes that recent legislation addresses the timing of future actions on nonattainment designations with regard to the 8-hour O₃ standards (Pub. L. No. 106-377, 114 Stat. 1441 (2000)).

57.  The information in this section is drawn primarily from the EPA document “Assessing the Risk of Trace Gasses that Can Modify the Stratosphere” (U.S. EPA, 1987).