Air qual. epa

fede

ral r

egiste

r

42529

WednesdayAugust 4, 1999

Part III

EnvironmentalProtection Agency40 CFR Part 58Air Quality Index Reporting; Final Rule

VerDate 18-JUN-99 16:03 Aug 03, 1999 Jkt 183247 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\04AUR3.XXX pfrm01 PsN: 04AUR3

42530 Federal Register / Vol. 64, No. 149 / Wednesday, August 4, 1999 / Rules and Regulations

ENVIRONMENTAL PROTECTIONAGENCY

40 CFR Part 58

[FRL–6409–7]

RIN 2060–AH92

Air Quality Index Reporting

AGENCY: Environmental ProtectionAgency (EPA).ACTION: Final rule.

SUMMARY: Today, EPA adopts revisionsto the uniform air quality index used byStates for daily air quality reporting tothe general public in accordance withsection 319 of the Clean Air Act (Act).These changes include the addition ofthe following elements: a new categorydescribed as ‘‘unhealthy for sensitivegroups;’’ two new requirements, first, toreport a pollutant-specific sensitivegroup statement when the index isabove 100, and second, to use specificcolors if the index is reported in a colorformat; new breakpoints for the ozone(03) sub-index in terms of 8-houraverage 03 concentrations; a new sub-index for fine particulate matter (PM2.5);and conforming changes to the sub-indices for coarse particulate matter(PM10), carbon monoxide (CO), andsulfur dioxide (SO2). In addition, EPA ischanging the name of the index from thePollutant Standards Index (PSI) to theAir Quality Index (AQI). This documentdiscusses the development of relatedinformational materials on pollutant-specific health effects and sensitivegroups and on precautionary actionsthat can be taken by individuals toreduce exposures of concern. Thisdocument also discusses theinterrelationship between the uniformair quality index and other programsthat provide air quality information andrelated health information to the generalpublic, including State and local real-time air quality data mapping andcommunity action programs.EFFECTIVE DATE: October 4, 1999.ADDRESSES: A docket containinginformation relating to EPA’s revisionsof the air quality index (Docket No. A–98–20) is available for public inspectionin the Air and Radiation Docket andInformation Center, U.S. EnvironmentalProtection Agency, South ConferenceCenter, Room M–1500, 401 M St., SW,Washington, DC 20460, telephone (202)260–7548. The docket may be inspectedbetween 8 a.m. and 5:30 p.m. onweekdays, and a reasonable fee may becharged for copying. For the availabilityof related information, seeSUPPLEMENTARY INFORMATION.

FOR FURTHER INFORMATION CONTACT:Terence Fitz-Simons, EPA (MD–14),Research Triangle Park, NC 27711,telephone (919) 541–0889, e-mail [email protected]. For healtheffects information, contact Susan LyonStone, EPA (MD–15), Research TrianglePark, NC 27711, telephone (919) 541–1146, e-mail [email protected] INFORMATION: Incompliance with President Clinton’sJune 1, 1998 Executive Memorandum onPlain Language in government writing,this package is written using plainlanguage. Thus, the use of ‘‘we’’ or ‘‘us’’in this package refers to EPA. The useof ‘‘you’’ refers to the reader and mayinclude industry, State and localagencies, environmental groups andother interested individuals.

Availability of Related InformationCertain documents are available from

the U.S. Department of Commerce,National Technical Information Service,5285 Port Royal Road, Springfield, VA22161. Available documents include:

(1) The Review of the NationalAmbient Air Quality Standards forOzone: Assessment of Scientific andTechnical Information (‘‘Staff Paper’’)(EPA–452/R–96–007, June 1996, NTIS# PB–96–203435, $67.00 paper copy and$21.50 microfiche). (Add a $3.00handling charge per order.)

(2) Review of the National AmbientAir Quality Standards for ParticulateMatter: Policy Assessment of Scientificand Technical Information (‘‘StaffPaper’’) (EPA–452/R–96–013, July 1996,NTIS # PB–97–115406, $47.00 papercopy and $19.50 microfiche). (Add a$3.00 handling charge per order.)

The guidance documents associatedwith this rulemaking are available fromEPA’s Office of Air Quality Planningand Standards in Research TrianglePark, NC. Requests for thesepublications can be mailed to: TerenceFitz-Simons, EPA (MD–14), ResearchTriangle Park, NC 27711. Your requestmay also be phoned in to Terence Fitz-Simons at 919–541–0889, or sent by e-mail to [email protected].

(1) Guideline for Public Reporting ofDaily Air Quality—Air Quality Index(AQI) (EPA–454/R–99–010).

(2) Guideline for Developing anOzone Forecasting Program (EPA–454/R–99–009).

The following document is availablefrom EPA’s Office of Mobile Sources(OMS) in Ann Arbor, MI. Requests forthis publication can be mailed to:Michael Ball, US EPA—NationalVehicle and Fuel Emissions Laboratory(NVFEL), 2000 Traverwood Dr., AnnArbor, MI 48103. Your request may alsobe phoned in to Michael Ball at 734–

214–4897, or sent by e-mail [email protected].

(1) Community Action Programs:Blueprint for Program Design (EPA 420–R–98–003).

Table of Contents

I. BackgroundA. What Are the Legislative Requirements?B. What Is the History of the Air Quality

Index?C. What Programs Are Related to the AQI?1. Ozone and Particulate Matter NAAQS

Revisions2. Real-Time Data Reporting Initiative

(Ozone Mapping Project)3. Community Action Programs

II. Rationale for Final RevisionsA. What Revisions Did We Propose?1. What Were the Proposed General

Changes?2. What Were the Proposed Changes to the

Sub-Indices?B. What Were the Significant Comments

and Our Responses?1. Comments and Responses on General

Changes2. Comments and Responses on Changes to

the Sub-IndicesC. What Are the Final Revisions?1. What Are the General Changes?2. What Are the Changes to the Sub-

Indices?D. What Are the Related Informational

Materials?III. Regulatory and Environmental Impact

AnalysesA. Executive Order 12866: OMB Review of

‘‘Significant Actions’’B. Regulatory Flexibility Analysis/Small

Business Regulatory EnforcementFairness Act

C. Unfunded Mandates Reform ActD. Paperwork Reduction ActE. Executive Order 13045: Children’s

HealthF. Executive Order 12848: Environmental

JusticeG. Executive Order 12875: Enhancing

Intergovernmental PartnershipsH. Executive Order 13084: Consultation

and Coordination with Indian TribalGovernments

I. National Technology Transfer andAdvancement Act

J. Congressional Review ActIV. References

I. Background

A. What Are the LegislativeRequirements?

Section 319 of the Act governs theestablishment of a uniform air qualityindex for reporting of air quality. Thissection directs the Administrator to‘‘promulgate regulations establishing anair quality monitoring systemthroughout the United States whichutilizes uniform air quality monitoringcriteria and methodology and measuressuch air quality according to a uniformair quality index’’ and ‘‘provides fordaily analysis and reporting of air


42531Federal Register / Vol. 64, No. 149 / Wednesday, August 4, 1999 / Rules and Regulations

1 Significant harm levels are those ambientconcentrations of air pollutants that present animminent and substantial endangerment to publichealth or welfare, or to the environment, asestablished in 40 CFR 51.151.

2 Intermediate index values of 200, 300, and 400were defined and are the basis for the Alert,Warning, and Emergency episode levels included in40 CFR part 51, appendix L, as part of thePrevention of Air Pollution Emergency Episodesprogram. This program requires specified areas tohave contingency plans in place and to implementthese plans during episodes when high levels of air

pollution, approaching the SHL, are in danger ofbeing reached. Changes to this emergency episodeprogram will be proposed in the near future.

Below an index value of 100, historically anintermediate value of 50 was defined either as thelevel of the annual standard if an annual standardhas been established (for PM10 and SO2), or as aconcentration equal to one-half the value of theshort-term standard used to define an index valueof 100 (for O3 and CO). Coarse or inhalableparticulate matter, PM10, refers to particles with anaerodynamic diameter less than or equal to anominal 10 micrometers.

3 PM2.5 refers to particles with an aerodynamicdiameter less than or equal to a nominal 2.5micrometers.

quality based upon such uniform airquality index * * *’’.

B. What Is the History of the Air QualityIndex?

In 1976, we established a nationallyuniform AQI, called the PollutantStandards Index (PSI), for use by Stateand local agencies on a voluntary basis(41 FR 37660). This uniform index wasestablished in light of a study conductedby EPA and the President’s Council onEnvironmental Quality (CEQ, 1976).This study found that the 55 urban areasin the U.S. and Canada reporting anindex of air quality used 14 differentindices, in conjunction with differentcautionary messages, such that inessence 55 different indices were beingused to report air quality. This diversityof indices sent a confusing messageabout air quality to the public. Based inpart on this study, we developed anindex to meet the needs of State andlocal agencies that has the followingadvantages: it sends a clear andconsistent message to the public byproviding nationally uniforminformation on air quality; it is keyed asappropriate to the national ambient airquality standards (NAAQS) and thesignificant harm level (SHL) 1 whichhave a scientific basis relating airquality and public health; it is simpleand easily understood by the public; itprovides a framework for reflectingchanges to the NAAQS; and it can beforecasted to provide advanceinformation on air quality.

The PSI, which is also commonlyreferred to by some State and localagencies as the AQI, includes sub-indices for O3, PM, CO, SO2, andnitrogen oxide (NO2), which relateambient pollutant concentrations toindex values on a scale from 0 through500. This represents a very broad rangeof air quality, from pristine air to airpollution levels that present imminentand substantial endangerment to thepublic. The index has historically beennormalized across pollutants bydefining an index value of 100 as thenumerical level of the short-term (i.e.,averaging time of 24-hours or less)primary NAAQS for each pollutant andan index value of 500 as the SHL.2 Such

index values serve to divide the indexinto categories, with each category beingidentified by a simple informativedescriptor. The descriptors are intendedto convey to the public informationabout how air quality within eachcategory relates to public health, withincreasing public health concerns beingconveyed as the categories range to theupper end of the scale. Additionalinformation about the general healtheffects associated with each category,and precautions that sensitive groupsand the general public can take to avoidexposures of concern, has been madeavailable through an informationalbooklet, updated as appropriate, thatalso presents and explains the PSI (EPA,1994).

In 1979, we made changes to the AQI,in part to reflect revisions to theNAAQS for O3, and to establishrequirements for AQI reporting (44 FR27598). The requirement for State andlocal agencies to report the AQI appearsin 40 CFR part 58.50, and the specificrequirements (e.g., what to report, howto report, reporting frequency,calculations) are in appendix G to 40CFR part 58.

C. What Programs Are Related to theAQI?

Historically, State and local agencieshave used primarily the AQI, or otherAQIs, to provide general information tothe public about air quality and itsrelationship to public health. In recentyears, many States and local agencies, aswell as EPA, have been developing newand innovative programs and initiativesto provide more information to thepublic, in a more timely way. Theseinitiatives, including real-time datareporting through the Ozone MappingProject and community action programs,can serve to provide useful, up-to-date,and timely information to the publicabout air pollution and its effects. Suchinformation will help individuals takeactions to avoid or reduce exposures ofconcern and can encourage the public totake actions that will reduce airpollution on days when levels areprojected to be in air quality categoriesof concern to local communities. Thus,these programs are significantly

broadening the ways in which State andlocal agencies can meet the nationallyuniform AQI reporting requirements,and are contributing to State and localefforts to provide community healthprotection and to attain or maintaincompliance with the NAAQS. We andState and local agencies recognize thatthese programs are interrelated withAQI reporting and with the informationon the effects of air pollution on publichealth that is generated through theperiodic review, and revision whenappropriate, of the NAAQS.

The most recent revisions to the O3

and PM NAAQS, the Ozone MappingProject, and community action programsare discussed briefly below. In light ofthe interrelationships among theseprograms, we have developed today’srevisions to the uniform AQI with thegoal of creating a revised AQI that caneffectively serve as a nationally uniformlink across these programs. In so doing,we intend to support and encourageState and local participation in real-timedata reporting initiatives and thedevelopment and implementation ofcommunity action programs that servepublic education and health protectiongoals.

1. Ozone and Particulate Matter NAAQSRevisions

On July 18, 1997, we revised theprimary NAAQS for O3 and PM basedon a thorough review of the scientificevidence linking exposures to ambientconcentrations of these pollutants toadverse health effects at levels allowedby the previous NAAQS. In particular,we replaced the 1-hour O3 NAAQS withan 8-hour O3 NAAQS andsupplemented the PM NAAQS with 24-hour and annual standards for fineparticulate matter (measured as PM2.5 3).These decisions were challenged in theU.S. Court of Appeals for the District ofColumbia Circuit, and on May 14, 1999,the Court remanded them to the Agencyfor further consideration, principally inlight of constitutional concernsregarding section 109 of the Act asinterpreted by EPA. American TruckingAssociations v. EPA, Nos. 97–1440, 97–1441 (D.C. Cir. May 14, 1999). On June28, 1999, the U.S. Department of Justiceon behalf of EPA filed a petition forrehearing seeking review of the Court’sdecision by the entire Court of Appeals.The EPA is continuing to assess whatfurther legal or administrativeproceedings may be appropriate inresponse to the Court’s decision, as well



4 Under section 319, the levels that areappropriate for this purpose do not necessarilydepend on the NAAQS levels that may beappropriate under section 109. Depending on howthe Agency chose to set an ambient standard, forexample, it might conclude that the standard doesnot need to preclude certain effects falling belowthe level of public health concern, and at the sametime set the AQI in such a way as to assure thatsensitive individuals who might experience thoseeffects receive notification and advice on actionsthey might take to avoid them. Similarly, AQIvalues might be set that are higher than thestandard would permit but that would require moreserious health warnings. This is not to say,however, that the levels of the 1997 NAAQS areirrelevant to decisions on the AQI breakpoints. Tothe contrary, the levels of the 1997 NAAQS areuseful surrogates for a series of scientificconclusions reached in the NAAQS rulemakings,based on the revised air quality criteria, regardingthe nature, extent, and severity of health effectsassociated with varying concentrations of PM andO3 in the air. Accordingly, later sections of thisnotice make reference as appropriate to relevantlevels of the 1997 NAAQS.

5 CASAC is a scientific advisory committeeestablished under the Act to review the scientificcriteria and standards and to advise theAdministrator on revision of the NAAQS, asappropriate.

as its relevance to other rulemakingssuch as this one.

With respect to the presentrulemaking, we have concluded that itis appropriate to proceed with finalaction on the proposed AQI revisions.As indicated previously, section 319 ofthe Act requires the Agency to establisha uniform air quality index, and thisrequirement is independent of thestatutory provisions governingestablishment and revision of theNAAQS. Moreover, there is no statutoryrequirement that the AQI be linked tothe NAAQS, although EPA has usedNAAQS levels in the past as referencepoints for the establishment of specificbreakpoints within sub-indices. Nothingin the Court’s opinion alters theconclusions EPA reached in revising theair quality criteria for PM and O3 undersection 108 of the Act, or in the NAAQSrulemakings, concerning the occurrenceof specific health effects at varyingconcentrations of PM and O3 in the air.Regardless of the outcome of the remandas to the NAAQS themselves, we believethe scientific record and conclusionsunderlying them are more thansufficient as a basis for decisions on thelevels at which the public should benotified about health risks associatedwith daily air quality.4

We do not regard this notificationfunction as involving the constitutionalconcerns raised in the Court’s opinion.The AQI has no bearing on pollutioncontrol requirements for specificsources; nor does it serve to implementthe NAAQS involved in the litigation.Rather, it provides information on airquality and health that will helpindividual citizens take prudent, self-protective actions to avoid or reduceexposures of concern and to avoidcontributing to air pollution on dayswhen unhealthy air quality is projected.

In this regard, the AQI is essentially away of conveying scientific/medicaladvice to the public in an easilyunderstood form.

As indicated below, there was broadsupport in public comments formodifying and expanding the use of theAQI to take into account the expandedunderstanding of air quality-healthrelationships that resulted from EPA’sreview of the latest scientificinformation on the effects of PM and O3.Other proposed revisions were designedto enhance the effectiveness of the AQIgenerally. The function the AQI servesof conveying to the public informationon daily air quality and associatedhealth risks is clearly important, and theseason of higher pollution levels isimminent. For all the above reasons, wesee no reason to delay final action onthe proposed revisions of the AQI. Theremainder of this section discussesaspects of the O3 and PM NAAQSrulemakings as they relate to today’saction.

As a result of the reviews of thescientific information upon which the1997 NAAQS for O3 and PM are based,an expanded understanding emerged asto the nature of the relationshipsbetween exposure to ambientconcentrations of these pollutants andthe health effects likely to beexperienced, especially near the level ofthe NAAQS. We and the Clean AirScientific Advisory Committee(CASAC) 5 recognized that for thesepollutants there may be no thresholdsbelow which health effects are not likelyto occur, but rather a continuum ofeffects potentially extending down tobackground levels. As ambientconcentrations increase, the proportionof individuals likely to experienceeffects and the seriousness of the healtheffects increase. Thus, the 1997standards were not considered risk free.While the standards were intended toprotect public health with an adequatemargin of safety, in accordance withsection 109(b) of the Act, including thehealth of sensitive groups, exposures toambient concentrations just below thenumerical level of the standards mayresult in exposures of concern for themost sensitive individuals. Conversely,exposures to ambient concentrationsjust above the numerical level of thestandards are not likely to result inexposures of concern for most healthypeople. This expanded understanding isreflected in the forms of the newstandards, which allow for multiple

days above the numerical level of thestandards.

These understandings were alsoreflected in CASAC’s advice to theAdministrator during the O3 NAAQSreview, urging expansion of the publichealth advisory system (i.e., a uniformAQI) and communication to the publicof the apparent nonthreshold nature ofthe health effects. More specifically, anumber of CASAC panel membersrecommended ‘‘that an expanded airpollution warning system be initiated sothat sensitive individuals can takeappropriate ‘exposure avoidance’behavior’’ (Wolff, 1995). Consistent withthis advice, in the preamble to theproposed revisions to the O3 NAAQS(61 FR 65733–65734), the Administratorrequested comment on the usefulness ofproviding specific health effectsinformation when ambientconcentrations are around the numericallevel of the standard, theappropriateness of using the AQI toconvey such information to the public,the possible addition of two new AQIcategories (one just above and one justbelow the numerical level of thestandard) and associated descriptorsand levels, as well as related healtheffects and cautionary statements.

Broad support for modifying the AQIwas received in public comments onthis aspect of the O3 NAAQS proposal,as discussed in the final ruleestablishing revisions to the O3 NAAQS(62 FR 38873–38874). Commentersoverwhelmingly endorsed expandingthe use of the AQI for various reasons,although many expressed concern withthe possible category descriptorssuggested in the proposal (i.e.,‘‘moderately good’’ and ‘‘moderatelyunhealthful’’). Many commenters feltthat an expanded AQI could helpparticularly sensitive people take actionto minimize their exposures, and thatthe AQI could be combined withcommunity action programs to reduceambient concentrations when thenumerical level of the standard wasforecasted to be exceeded. Somecommenters endorsed increasing thespecificity of health and cautionarystatements related to the AQI categories.Commenters from State and localagencies encouraged us to develop anyapproaches to revising the AQI inconsultation with them, specifically inthe areas of sharing real-time monitoringdata, risk communication with thepublic, and coordination of a nationalprogram.

2. Real-time Data Reporting Initiative(Ozone Mapping Project)

The Ozone Mapping Project is part ofEPA’s Environmental Monitoring for



Public Access and Community Tracking(EMPACT) initiative—a new approachto providing timely environmentalinformation to communities. It is acooperative effort of the EPA, State andlocal air pollution control agencies, andregional organizations including theMid-Atlantic Regional Air ManagementAssociation (MARAMA), the NortheastStates for Coordinated Air UseManagement (NESCAUM), the northeastOzone Transport Commission (OTC),the Lake Michigan Air DirectorsConsortium (LADCO), SouthEast StatesAir Resource Managers (SESARM), andCentral States Air Resource Agencies(CenSARA). During the summer of 1998,EPA’s Office of Air Quality Planningand Standards assumed coordination ofthe project.

The Ozone Map provides simple andtimely information about ground-levelO3. During the 1998 O3 season it wasavailable on EPA’s AIRNOW web site(http://www.epa.gov/airnow) and onsome local television and news reports.It is an animated contour map thatshows concentrations of O3, incategories ranging from good tomoderate to varying degrees ofunhealthy, based on AQI values, as theydevelop across the eastern UnitedStates. In 1998, the map was createdfrom real-time, hourly O3 data providedby a network of more than 400 airmonitoring stations from South Carolinato Wisconsin and Maine. Whenaccessed on a computer, cautionarystatements for each category could bedisplayed by running a cursor over thelegend. Also available on the AIRNOWweb site were still maps of maximumvalues and forecasted values, andarchived animated maps. In 1999, theozone mapping coverage is beingexpanded to include 31 States and over1500 monitors across the eastern andcentral U.S., and California. In addition,TV weather service providers areplanning to carry the Ozone Map andforecasts as part of their traditionalweather packages for local TV stations.

Along with the Ozone Map, theAIRNOW web site contains informationabout O3 health effects in the ‘‘HealthFacts’’ section, and emission reductionactivities in the ‘‘What You Can Do’’section. It also provides links to real-time data, and community actionprogram web sites, that are maintainedby State and local agencies around thecountry. The goals of the web site are to:(1) Provide real-time air pollution datain an understandable, visual format, (2)provide information about the publichealth and environmental effects of airpollution, and (3) provide the publicwith information about ways in which

they can protect their health and actionsthey can take to reduce pollution.

3. Community Action ProgramsThe implementation of community

action programs (also referred to asvoluntary action programs or episodicemission control programs) is becomingincreasingly popular across the countryas an innovative approach used toreduce emissions of O3 precursors, CO,and PM. Motivation for implementationof this type of program often stems fromlocal government and business concernsabout the NAAQS attainment status ofthe area and the restrictions, additionalcontrols, and costs associated withbeing classified as a nonattainment area.Many areas are also motivated by publichealth concerns and believe thatincreasing the amount of air qualityinformation available to sensitivepopulations raises awareness andresults in significant health benefits.Specific goals which are usuallyassociated with community actionprograms include: (1) Educate thepublic and enhance protection of publichealth; (2) attain or maintain NAAQSattainment status and the associatedeconomic benefits; (3) meet specificemission reduction targets; and (4)manage/reduce traffic congestion.

Community action programs areusually voluntary and generally providemultiple steps that the public, business,and industry can take to reduceemissions when higher levels of airpollution are forecast to occur,including in particular transportation-related measures such as trip reduction,postponement of certain activities suchas vehicle refueling, and maintenance ofcars. The programs emphasize educatingthe public about the impact ofindividual activities on local air qualityand the basics of air pollution. Theeducational component of theseprograms also helps to create a stronglink between environmental goals andassociated public health benefits.

Most of these programs are based onthe categories of the AQI and make useof the AQI descriptors and relatedhealth effects and cautionary statementson action days. By linking action daysto the AQI, local control programs hopeto alter individual behavior to reduceemissions and to reduce exposures tothe population. In addition to reducedpollutant exposure of the generalpopulation due to improved air quality,there are other health benefits directlyassociated with community actionprograms that can be enhanced bylinkage to the AQI. Different populationgroups are more sensitive to the harmfuleffects of the different air pollutantsincluded in the AQI, and the revisions

to the AQI being adopted today, togetherwith related informational materials,will significantly improve theeffectiveness of communications withthese groups. Public education, orprograms directly targeting thesegroups, may provide the mostsignificant benefits of a communityaction program. Forecasting days withelevated pollution levels, and thencommunicating effectively about airquality and associated health effects,may help these groups selectively limittheir outdoor activities and, therefore,limit their potential for exposures ofconcern.

We are committed to providing Statesand local agencies with support in theirefforts to meet air quality standards, toinform the public about air quality, andto educate the public about the impactsof air pollution. The revisions to theAQI being adopted today have as a goalthe creation of a revised AQI that caneffectively serve as a nationally uniformlink across the range of programs (e.g.,real-time data reporting initiatives,community action programs) that havethese functions.

In support of community actionprograms, we have developedinformational materials related to theAQI, including the health effects andcautionary statements associated witheach category and more detailed healtheffects information (see section II.D.),available on the AIRNOW web site, thatState and local agencies may use toenhance their community actionprograms. Focusing on transportationmeasures that are often a majorcomponent of community actionprograms, EPA’s OMS has developed areport entitled, ‘‘Community ActionPrograms: Blueprint for ProgramDesign.’’ This document describes themajor steps needed to put together asuccessful episodic control program andprovides criteria that State and localagencies can use to examine andevaluate their own programs. The reportis available from OMS (see Availabilityof Related Information).

II. Rationale for Final RevisionsIn developing the revisions to the AQI

that are being adopted today, we soughtextensive input from State and localagencies and from the public. Wesponsored a workshop with State andlocal agencies, participated in numerousmeetings, prepared and made availablea staff draft revision to the AQI sub-index for O3 for use during the 1998 O3

season, and conducted several focusgroups across the nation to obtainpublic input on the effectiveness of draftrevisions to the AQI and related O3

maps and informational materials. A



6 For NO2, the index ranges from 200 to 500, sincethere is no short-term NAAQS for this pollutant.

detailed history of the process leading tothe proposal and the rationale for theproposed revisions are described morefully in the December 9, 1998 proposalnotice (63 FR 67818–67834). The sub-sections below contain a description ofthe revisions we proposed, a discussionof the significant comments we receivedand our responses to them, and asummary of the AQI we are adoptingtoday.

A. What Revisions Did We Propose?

The primary consideration thatshaped the proposed revisions was theimportance of providing nationallyuniform health information associatedwith daily ambient levels of the airpollutants included in the index,consistent with the requirement ofsection 319 of the Act for an index toachieve national uniformity in daily airquality reporting. More specifically, theproposed changes to the AQI sub-indices for O3 and PM reflected the 1997revisions to the O3 and PM NAAQS. Theproposed general changes to thestructure of the AQI were based on theexpanded understanding that emergedduring the O3 and PM reviews as to thenature of the relationships betweenexposure to ambient concentrations ofthese pollutants and the health effectslikely to be experienced, considerationof the implications of changes for theother pollutants, and broad input fromState and local agencies and the public.The proposed general changes to theAQI, together with related informationalmaterials, were intended to expand theuse of the AQI to provide morepollutant-specific health information,especially when ambient concentrationsare close to the level of the primaryNAAQS.

1. What Were the Proposed GeneralChanges?

a. Categories and related descriptors,index values and colors. The AQIcurrently incorporates the pollutants O3,PM, CO, SO2, and NO2. Index valuesrange from 0 to 500 6, and the index issegmented into five categories named bydescriptor words that were chosen tocharacterize the relationship betweendaily air quality and public health. Toreflect better the current understandingof the health effects associated withexposure to these air pollutants, weproposed to revise the AQI categoriesand descriptors, and to associatespecific colors with the categories asshown below in Table 1.

TABLE 1.—PROPOSED CATEGORYINDEX VALUES, DESCRIPTORS, ANDCOLORS

Index values Descriptor Color

0–50 ............. Good ................... Green51–100 ......... Moderate ............. Yellow101–150 ....... Unhealthy for sen-

sitive groups.Orange

151–200 ....... Unhealthy ............ Red201–300 ....... Very unhealthy .... Purple301–500 ....... Hazardous ........... Maroon

These proposed changes reflected theaddition of a new category above an AQIof 100, created by dividing the current‘‘unhealthful’’ category into twocategories.

When air quality is in the ‘‘unhealthyfor sensitive groups’’ range, people thatare in the sensitive group, whether thesensitivity is due to medical conditions,exposure conditions, or inherentsensitivity, may experience exposures ofconcern. However, exposures to ambientconcentrations in this range are notlikely to result in exposures of concernfor most healthy people. The descriptor‘‘unhealthy for sensitive groups’’ waschosen to convey this message clearly.Participants in focus groups (SAIC 1998)clearly understood that ‘‘sensitivegroups’’ does not refer to the generalpublic, indicating that this descriptoreffectively communicates the intendedhealth message. This category wouldinclude a caution that while perhaps ofinterest to all citizens, would be ofparticular interest to individuals andfamilies of individuals who aremembers of sensitive groups.

As air quality moves into the‘‘unhealthy’’ range, exposures areassociated with an increase in thenumber of individuals who couldpotentially experience effects andincludes a greater proportion ofmembers of the general public. Based oninput received in the development ofthe proposal, the descriptor‘‘unhealthy’’ appropriately characterizesair quality in this range.

In addition to an increasing number ofexposures of concern, when air qualitymoves into the ‘‘unhealthy’’ range andabove, individuals who were affected atlower levels, typically members ofsensitive groups, are likely toexperience more serious health effectsthan members of the general public. Toreflect this understanding, it isappropriate to convey two messages inthe cautionary statements for both the‘‘unhealthy’’ and ‘‘very unhealthy’’categories. One message is directed tomembers of sensitive groups, and theother is directed to the general public.The use of a distinct cautionary message

for members of sensitive groups isentirely consistent with an original goalthat the index be based on therelationships between pollutantconcentrations and adverse healtheffects within various groups, e.g.,aggravation of disease in people withrespiratory disease and incidence ofrespiratory effects in healthy people.Guidance on pollutant-specificcautionary statements related to thecategories of the AQI is discussed belowin section II.D.

Consistent with the overarching goalof national uniformity in the reportingof air quality, we proposed that thespecific colors listed in Table 1 beassociated with each category. While theAQI can be reported without the use ofcolors (through text and numbers alone),when the index is reported using colors,we proposed to require that only thesespecified colors be used. Threeexamples of AQI reports that use colorare the color bars that appear in manynewspapers, the color scales on Stateand local agency web sites, and thecolor contours of the Ozone Map. Weparticipated in many discussions withState and local agencies andassociations regarding which specificcolors should be associated with theAQI categories, particularly above anindex value of 100. These discussionstypically were in the context of eitherthe Ozone Mapping Project orcommunity action programs. It wasclear that the color associated with acategory can be part of the health effectsand cautionary message being conveyed.Were various State and local agencies touse different colors to represent thesame category, and thus the same levelof air quality, it could well send aconfusing message about air quality andassociated health effects to the public.

As an alternative to requiring the useof specified colors, we solicitedcomment on the option ofrecommending, rather than requiring,the use of these colors when reportingagencies choose to report the AQI incolor format. In soliciting comment onthis alternative, we sought to allowcommunities maximum flexibility inAQI reporting, while still preserving anationally uniform AQI. We, therefore,requested that commenters addressingthis issue discuss how this more flexibleapproach would satisfy the statutorylanguage requiring a nationally uniformAQI if different colors may be usedacross the nation to represent the samerange of air quality.

b. Reporting requirements. Weproposed to change 40 CFR part 58.50to require reporting of the AQI in all



7 A complete list of MSAs and their boundariescan be found in the Statistical Abstract of theUnited States (1998).

Metropolitan Statistical Areas (MSAs) 7

with a population over 350,000, insteadof all urbanized areas with a populationover 200,000. This change was proposedfor consistency with the othermonitoring regulations in part 58, whichare or will be based on MSAs. Thisproposed change would not, however,have a significant impact on who isrequired to report, since virtually thesame number of cities would be coveredunder the proposed reportingrequirement as are covered under theexisting requirement.

Consistent with early input from Stateand local agencies, we proposed tochange the rounding conventions usedto calculate index values correspondingto pollutant concentrations at and abovethe numerical level of the NAAQS to beconsistent with the roundingconventions used in defining theNAAQS for each pollutant. This wouldavoid situations where a health advisorycould be issued that describes the air asunhealthy, when in fact the numericallevel of the standard has not beenexceeded.

The proposed rule retained therequirements to identify the area forwhich the AQI is being reported, thetime period covered by the report, the‘‘critical’’ pollutant for which thereported AQI value was derived, theAQI value, and the associated categorydescriptor. Recognizing that manyagencies use a color format to report theAQI, the proposed rule added therequirement to report the associatedcategory color if a color format is used.Because different sensitive groups are

at-risk from different pollutants, issuingadvisories for all sensitive groups whomay be affected at AQI values greaterthan 100 clearly improves public healthprotection. Therefore, the proposed ruleencouraged, but did not require, thatAQI reports include: appropriate healtheffects and cautionary statements, allAQI values greater than 100, the AQI forsub-divisions of the MSA (if there areimportant differences in air qualityacross sub-divisions of the MSA),possible causes for high index values,and the actual pollutant concentrations.These topics were also discussed in ourdraft ‘‘Guideline for Public Reporting ofDaily Air Quality—Pollutant StandardsIndex (PSI)’’ that was made available onthe AIRLINKS web site.

The proposed rule emphasized theimportance of forecasting the AQI byspecifying that forecasted values shouldbe reported, when possible, but did notrequire that forecasted values bereported. Given the importance of theO3 sub-index in a large number ofMSAs, and the use of an 8-houraveraging time for calculating the O3

sub-index value, forecasting the O3

index value is now more beneficial thanbefore. For a health advisory system tobe effective, people need to be notifiedas early as possible to be able to avoidexposures of concern. Because the O3

sub-index is based on 8-hour O3

averages, forecasting O3 concentrationsclearly would have increased value inproviding cautionary statements to thepublic. We recognized that many Stateand local air agencies are alreadyissuing health advisories based on

forecasted O3 concentrations. Since wehave determined that forecasting wouldadd much to the benefits of AQIreporting, we indicated that we wouldbe making available guidance onstarting a forecasting program (EPA1999b) in an area or MSA whereforecasting is not presently done.Included in the document is guidanceon using hourly O3 concentrations aspredictors for 8-hour averages.

c. Index name. Many State and localagencies encouraged us to change thename of the PSI to the Air QualityIndex, or AQI, since many agenciesalready use the name AQI whenreporting the AQI value to the public.Most participants in the focus groupspreferred the name AQI, commentingthat it more clearly identified the indexas relating to the quality of the air ratherthan to environmental pollution ingeneral. Based on these considerations,we solicited comment on changing theindex name from Pollutant StandardsIndex (PSI) to Air Quality Index (AQI).

2. What Were the Proposed Changes tothe Sub-Indices?

To conform to the proposed generalchanges to the AQI discussed above,and to reflect the recent revisions to theO3 and PM NAAQS, we proposedchanges to the sub-indices for O3, PM,CO, and SO2; no conforming changes arenecessary for the NO2 sub-index. Theproposed sub-indices are summarizedbelow in Table 2, in terms of pollutantconcentrations that correspond tobreakpoints in the index, and arediscussed in the following sections.

TABLE 2.—PROPOSED BREAKPOINTS FOR O3, PM2.5, PM10, CO, AND SO2 SUB-INDICES

AQI value

O3 PMCO, 8-hr

(ppm)SO2, 24-hr

(ppm)8-hr (ppm) 1-hr (ppm) PM2.5, 24-hr(µg/m3)

PM10, 24-hr(µg/m3)

50 ................ 0.07 ............. ................................. 15 50 4 0.03100 .............. 0.08 ............. 0.12 65 150 9 0.14150 .............. 0.10 ............. 0.16 * 100 250 12 0.22200 .............. 0.12 ............. 0.20 * 150 350 15 0.30300 .............. 0.40 (1-hr) ... 0.40 * 250 420 30 0.60400 .............. 0.50 (1-hr) ... 0.50 * 350 500 40 0.80500 .............. 0.60 (1-hr) ... 0.60 * 500 600 50 1.00

* If a different SHL for PM2.5 is promulgated, these numbers will be revised accordingly.

a. Proposed ozone sub-index. On July18, 1997, we revised the O3 primaryNAAQS to replace the 1-hour standardwith a new standard with an 8-houraverage at a level of 0.08 ppm and aform based on the 3-year average of theannual fourth-highest daily maximum 8-

hour average O3 concentrationsmeasured at each monitor within anarea (62 FR 38856–38896). Theseproposed revisions were based onfindings from the most recent review ofthe NAAQS indicating that the newprimary standard will provide increased

protection to the public, especiallychildren active outdoors and othersensitive groups, against a wide range ofO3-induced health effects, includingdecreased lung function; increasedrespiratory symptoms; hospitaladmissions and emergency room visits



for respiratory causes, among childrenand adults with pre-existing respiratorydisease such as asthma; inflammation ofthe lung; and possible long-term damageto the lungs. In setting this standard, werecognized that there is no apparentthreshold below which health effects donot occur, that the standard is not riskfree, and, thus, that exposures ofconcern are possible below thenumerical level of the standard for someextremely sensitive individuals.

We proposed to set an index value of100 equal to the level of the 8-hour O3

standard. Recognizing the continuum ofhealth effects, we considered the resultsof a quantitative risk assessment(Whitfield et al., 1996) in selecting 8-hour O3 concentrations to correspond toindex values of 50, 150 and 200. Sinceno human health effects informationwas available for 8-hour average O3

concentrations at significantly higherlevels, we proposed to retain thebreakpoints at the upper end of the AQIscale (between the ‘‘very unhealthy’’and ‘‘hazardous’’ categories and the SHLwhich corresponds to the top of the PSIscale of 500) in terms of the existing 1-hour average concentrations.

These proposed revisions reflect thenew 8-hour O3 NAAQS and will inalmost all areas result in a moreprecautionary index than the current 1-hour sub-index. However, werecognized that a very small number ofareas in the U.S. have atypical airquality patterns, with very high 1-hourdaily peak O3 concentrations relative tothe associated 8-hour averageconcentrations. In such areas, the use ofthe current 1-hour sub-index may bemore precautionary on a given day thanthe proposed 8-hour sub-index. Toallow for the reporting of the moreprecautionary sub-index value, weproposed to retain the 1-hour sub-indexat and above AQI values of 100 and toallow the reporting of the higher of thetwo O3 sub-index values. Thus, both thenew 8-hour and the current 1-hour sub-indices, as shown in Table 2, wereincluded in the proposed appendix G.Since for the large majority of areas the8-hour sub-index will be moreprecautionary, we did not propose torequire all areas to calculate both sub-index values. Rather, we proposed toallow areas the flexibility to calculateboth sub-index values and, when bothsub-index values are calculated, torequire that the higher value bereported. We specifically solicitedcomment on this proposed approach.

b. Proposed PM sub-index. On July18, 1997, we revised the PM NAAQS byadding a new set of standards for fineparticles, or PM2.5, set at levels of 15 µg/m3 (annual) and 65 µg/m3 (24-hour

average) (62 FR 38652–38760). Theserevisions were based on findings fromthe most recent review of the PMNAAQS that recently published studieshave indicated that serious healtheffects were more closely associatedwith the levels of the smaller particlesubset of PM10. These health effectsinclude premature mortality andincreased hospital admissions andemergency room visits, primarily in theelderly and individuals withcardiopulmonary disease; increasedrespiratory symptoms and disease inchildren and individuals withcardiopulmonary disease; decreasedlung function, particularly in childrenand individuals with asthma; andalterations in respiratory tract defensemechanisms. In addition, PM10

standards were retained at the samelevels of 50 µg/m3 (annual) and 150 µg/m3 (24-hour average) to continue toprovide protection against health effectsassociated with the coarse particlesubset of PM10, including aggravation ofasthma and respiratory infections. Toreflect these revisions to the PMNAAQS, we proposed to add a new sub-index for PM2.5, and to make conformingchanges to the sub-index for PM10,consistent with the proposed generalchanges to the AQI. The proposed sub-indices are summarized in Table 2 anddiscussed below.

Proposed new PM2.5 sub-index.Consistent with the historical method ofselecting breakpoints of the AQI, weproposed to set an index value of 100at the level of the 24-hour PM2.5

NAAQS, 65 µg/m3, and an index valueof 50 at the level of the annual NAAQS,15 µg/m3. Also consistent with the basicstructure of the AQI, the proposedupper bound index value of 500corresponds to the SHL, established insection 51.16 of the CFR under thePrevention of Air Pollution EmergencyEpisodes program. The SHL is set at alevel that represents an imminent andsubstantial endangerment to publichealth. When we propose revisions tothe Prevention of Air PollutionEmergency Episodes program, theproposal will include a SHL for PM2.5.In the interim, we proposed to establisha PM2.5 concentration of 500 µg/m3 to beassociated with a PM2.5 index value of500.

For intermediate breakpoints in theAQI between values of 100 and 500,PM2.5 concentrations were proposedthat generally reflect a linearrelationship between increasing indexvalues and increasing PM2.5 values. Theavailable scientific evidence of healtheffects related to population exposuresto PM2.5 concentrations between the 24-hour NAAQS level and the proposed

PM2.5 concentration to be associatedwith a PM2.5 index value of 500 suggesta continuum of effects in this range,with increasing PM2.5 concentrationsbeing associated with increasingly largernumbers of people likely experiencingserious health effects (62 FR 38675; StaffPaper, p. VII–27). The proposedgenerally linear relationship betweenAQI values and PM2.5 concentrations inthis range, rounded to increments of 50µg/m3 to reflect the approximate natureof such a relationship, is consistent withthis evidence.

Proposed conforming changes to thePM10 sub-index. Consistent with theretention of the levels of the PM10

NAAQS, we proposed to retain the PM10

sub-index generally and to add a newbreakpoint at an index value of 150 toconform to the proposed additional AQIcategory. We proposed that thisbreakpoint be set at a PM10 24-houraverage concentration of 250 µg/m3, themid-point between the breakpointsassociated with index values of 100 and200. We believe that the PM10 sub-index, with this conforming change,remains appropriate for the publichealth protection purposes of the AQI.

c. Proposed conforming changes tothe CO and SO2 sub-indices. Since thecurrent AQI sub-indices reflect thecurrent NAAQS for CO and SO2, theonly change we proposed for these sub-indices was to add a breakpoint to eachsub-index at an index value of 150 toconform to the proposed additional AQIcategory. We proposed that thesebreakpoints be set at concentrations atthe mid-points between the breakpointsassociated with index values of 100 and200, consistent with the approachdescribed above for conforming changesto both the 1-hour O3 sub-index and thePM10 sub-index. These proposedbreakpoints are summarized in Table 2and will be reviewed in conjunctionwith the future reviews of the CO andSO2 NAAQS.

B. What Were the Significant Commentsand Our Responses?

This section describes the significantcomments we received on proposedrevisions to the index and our generalresponses to them. More detailedcomment summaries and responses arecontained in a Response to CommentsDocument that is available in the docket(see ADDRESSES).

1. Comments and Responses on GeneralChanges

a. Categories and related descriptors,index values and colors. With regard tothe proposed changes to the generalstructure of the index, we receivedcomments that focused on two major



issues. The first major issue waswhether to add a category above orbelow the standard, or both. In addition,related to that issue were commentsabout the proposed descriptor for thecategory we proposed to add above thelevel of the standard. The second majorissue regarded the particular colors,listed in Table 1, we proposed toassociate with each category.

With regard to the general structure ofthe index, most commenters supportedour proposal to add a category above thelevel of the standard. However,commenters from environmental groupsand several States suggested adding acategory below the level of the standardto provide additional caution formembers of sensitive groups, instead of,or in addition to one above. Thesecommenters expressed the view that theproposed sub-indices, that added acategory above the standard, did notsufficiently caution members ofsensitive groups about health effectsoccurring below the level of thestandard. Specifically, their commentswere in reference only to potentialhealth effects occurring below the 8-hour O3 and 24-hour PM2.5 standards.Regarding health effects below the PM2.5

standard, one State commenter tookexception with the statement in theproposal that an additional categorybelow the standard, while perhapsmeaningful for O3, would not be anappropriate distinction for the otherpollutants in the index. This commenternoted that ‘‘such a distinction would bemore imperative for other pollutants,especially for PM where the level of the24-hour standard may be less protectiveof sensitive groups than the ozonestandard.’’ (Docket No. A–98–20, IV–D–19). Agreeing with the importance ofcautioning sensitive groups below thelevel of the 24-hour PM2.5 standard,another commenter noted ‘‘We believethat adding a category below the levelof the standard is of particularimportance with respect to fineparticles.’’ (Docket No. A–98–20, IV–D–11). Regarding the O3 sub-index, someof the States and the environmentalgroups that endorsed adding a categorybelow the level of the standardsupported that position by noting thatwe and CASAC stated that extremelysensitive individuals may be affecteddown to background levels of O3. Onecomment from an environmental groupnoted that:The CASAC recognized that for O3 and fineparticle pollution, ‘‘there are no discerniblethresholds below which health effects are notlikely to occur in the most sensitiveindividuals’ as it was advising EPA to setnew health standards. We agree with CASACand support the idea of setting ‘‘an expanded

air pollution warning system (to) be initiatedso that sensitive individuals can takeappropriate exposure avoidance behavior,’’however EPA has misrepresented the healththreat with the levels it has proposed.(Docket No. A–98–20, IV–D–17).

A State commenter that supportedadding a category below the level of thestandard observed that adding such acategory would be consistent withEPA’s conclusion ‘‘that exposures toambient concentrations just below thenumerical level of the standard mayresult in exposures of concern for themost sensitive individuals.’’ (Docket No.A–98–20, IV–D–19).

We understand and agree with theissues related to communication of riskbelow the levels of the 24-hour PM2.5

and 8-hour O3 standards. For the PM2.5

sub-index, we have addressed concernsabout health effects below the level ofthe 24-hour PM2.5 standard by revisingthe PM2.5 sub-index so sensitive groupsare cautioned below the 24-hour PM2.5

standard. Based on review of thesuggested revisions to the PM2.5 sub-index that we received in comments, webelieve this approach fully addressestheir concerns. The revision isdiscussed in section II.B.2 below.

For better communication of healthrisk below the 8-hour O3 standard, wehave addressed the issues raised bycommenters by revising the O3 sub-index. We have expanded the‘‘moderate’’ range of the 8-hour O3 sub-index to make it more precautionary.When air quality is in the ‘‘moderate’’range of the 8-hour O3 sub-index, wehave provided health effects andcautionary statements, available in ourAQI Reporting Guidance document(EPA, 1999a) (discussed in section II.D),that may be used by State and localagencies to caution unusually sensitiveindividuals below the level of the 8-hour O3 standard. This revision isdiscussed in section II.B.2 below.

We do not believe it is necessary orappropriate to change the generalstructure of the index by adding a newcategory below the level of the standardto caution extremely sensitiveindividuals. Based on the concerns ofState and local agencies that theaddition of two new categories wouldunduly complicate the index, we areadding just one new category tomaintain the degree of simplicitystrongly supported by State and localagencies, none of whom advocated theaddition of two new categories. Asdescribed in section II.A.1 above, webelieve that adding a category above thelevel of the standard makes a distinctionthat is useful for members of sensitivegroups without alarming the general

public. As noted by one Statecommenter:

We are satisfied and support the proposedcategory index values, descriptors and colors.[We] believe that the Air Quality Index * * *has been a very effective communication toolduring the ozone season. It has been ourexperience that a category above the standardprovides the proper communication to theaffected populations without alarming ordesensitizing others. (Docket No. A–98–20,IV–G–04).

Further, given the changes we havemade to the PM2.5 sub-index, and theexpanded ‘‘moderate’’ range and thecautionary statements we have madeavailable in guidance for use below thelevel of the 8-hour O3 standard, we donot believe a category below the level ofthe standard to caution members ofsensitive groups would be anappropriate distinction for any of thepollutants included in the index. Webelieve that the approach we haveadopted retains the simplicity of theindex while allowing for more detailedcautionary information to be madeavailable to the public whenappropriate.

With regard to the descriptor‘‘unhealthy for sensitive groups,’’ somecommenters expressed the view thatthis descriptor is misleading because itencompasses a large segment of thepopulation. In addition, they argued, thepublic will not know that for certainpollutants healthy people, especiallyhealthy children, are members ofsensitive groups. Noting that it isprudent policy to assume that most riskcommunication regarding air qualityimpacts will be limited to the generaldescriptors, some of these commentersrequested that if we continue todistinguish sensitive groups from thegeneral population, that the descriptorbe changed from ‘‘unhealthy forsensitive groups’’ to ‘‘unhealthy forchildren and other sensitive groups,’’ sothat the public would receive a clearmessage that children are members of asensitive group that may be at increasedrisk from exposure to ozone. (DocketNo. A–98–20, IV–D–2, IV–D–4 and IV–D–11). We agree with the view of thesecommenters, based on the responses ofparticipants in the focus groups, that thepublic will not know that healthypeople, including healthy children, maybe at risk when air quality is in the‘‘unhealthy for sensitive groups’’ range.The suggested descriptor, however, isonly appropriate for pollutants forwhich children are a sensitive group.Since the sensitive groups differ fromone pollutant to another, and childrenare only part of the sensitive group forO3, PM2.5 and NO2, this descriptor is notappropriate for the other pollutants. For



example, the descriptor ‘‘unhealthy forchildren and other sensitive groups’’would not be appropriate for use in theCO sub-index, where people with heartdisease are the group most at-risk. Useof this descriptor when CO levels areabove an index value of 100 could leadto confusion about the health effectsassociated with high levels of CO.Therefore, we do not believe it would beuseful or prudent to adopt thedescriptor ‘‘unhealthy for children andother sensitive groups.’’ To increasepublic awareness that healthy childrenare members of the sensitive group forO3, we are adding the requirement thatwhen the AQI value is above 100,reporting agencies include in theirpublished report a statement describingthe sensitive group for that particularpollutant. The reporting requirement forpollutant-specific statements describingsensitive groups is discussed below insection II.C.1.b on reportingrequirements, and listed in appendix G.We believe that the requirement foragencies to report the pollutant-specificstatements identifying the groups atrisk, when air quality is above an indexvalue of 100, will more effectivelycommunicate the risk associated withspecific air pollutants, and therebybetter help members of the publicreduce personal exposure. To the extentpossible with AQI reporting, thisrequirement will also ensure that thepublic is informed that children are partof the sensitive group for O3. Thisrequirement will not only improveprotection for healthy children, but alsohealthy adults, the elderly, and peoplewith heart and lung disease. We believethat another good way to address thislack of awareness is to educate thepublic, and the media and health careprofessionals that inform the public,about the health effects messageassociated with the category ‘‘unhealthyfor sensitive groups.’’ To helpaccomplish the goal of educating thepublic, we will be expanding thedevelopment of education and outreachmaterials and activities as described insection II.D below.

With regard to the colors listed inTable 1, we received commentsconcerning both the particular colorsassociated with the different categoriesand whether specific colors should berequired or recommended. The majorityof commenters, including most Stateand local agencies commenting,supported our proposed color scheme.Many of those (commenters that didsupport it), had used the same or asimilar color scheme associated witheither community action programs orozone maps. Commenters that had used

the same or a similar color schemenoted that it effectively andappropriately portrayed the full range oflocal air quality values. On the otherhand, some environmental groups andseveral States commented that the colorred should be used for the category justabove standard, instead of the colororange that we proposed. Primarily,these commenters expressed the viewthat the color orange would not send asufficiently strong message that thestandard has been exceeded. In theproposal we indicated that because thecolor red sends a strong cautionarymessage, it is most appropriately usedwhen effects are likely to occur in thegeneral population, and when moreserious effects are likely in members ofsensitive groups. Many of thesecommenters noted that since up to 30percent of the population could beconsidered to be in the sensitive groupfor O3, when the standard is exceededthe general public should be alerted.These commenters expressed the viewthat it is appropriate to use the color redjust above the level of the standard bothto alert the public of potential healthrisks and to encourage emissionreduction actions. An environmentalgroup commented:

While individuals that are sensitive to poorair quality may look at the daily listing in thenewspaper or call a message recorded by thestate or local air agency, we know fromexperience that air quality does not receivebroad public attention until it is predicted orreaches the level of ‘‘code red.’’ At that point,the television and radio media announcesthat people should restrict outdoor activityand take steps to not add more pollution tothe air by carpooling, using less electricity,or using mass transit. (Docket No. A–98–20,IV-D–17).

Another commenter from a Stateagency noted:

Considering that the definition of sensitiveindividuals for ozone includes healthy activechildren and outdoor workers, a clearunambiguous message needs to be sent to thepublic so that they can respond accordingly.For parents of active children, a messagewhich states that air quality is unhealthy,and displays it using the color red, sends aclear message—even though it may carrywith it the risk that individuals not in thesensitive population might also takeexposure avoidance measures. Issuing amessage that air quality is unhealthy forsensitive individuals and displays it with acode orange runs the risk of having sensitiveindividuals, or those guiding sensitiveindividuals (i.e., doctors and parents) notprescribe any avoidance action because of theambiguity of the message. (Docket No. A–98–20, IV-G–19).

Additionally, these commenterssuggested that the color orange be usedfor the category they wanted us to add

below the level of the standard, asdescribed above.

In considering these comments, werecognize that the NAAQS are set toprotect public health with an adequatemargin of safety, including the health ofsensitive groups. When the standardsare met, public health is protected.Exposures to ambient concentrationsjust above the numerical level of thestandards are not likely to result inexposures of concern for most healthypeople. This is especially true for the 8-hour O3 standard, which has aconcentration-based form designed tooffer more protection from higherconcentrations than from multiplesmaller exceedances of the standard.The form of the 8-hour O3 standardallows for multiple days above the levelof the standard, provided the 3-yearaverage of the fourth-highest maximumconcentrations does not exceed the levelof the standard. This means that publichealth is protected, even when there aremultiple days each year when ambientO3 concentrations are above the level ofthe standard, as long as the standard ismet. Therefore, it is inappropriate onany given day to express a high level ofconcern when air quality just exceedsthe level of the standard. Besidessending an inaccurate health effectsmessage by using the color red with thecategory ‘‘unhealthy for sensitivegroups,’’ another concern is thepotential loss of credibility that couldresult from repeatedly sending a signaldisproportionate to the expectedincidence of noticeable symptoms. Ifthis were to happen, the AQI could losethe power to influence people’sbehavior to protect their health. Onecommenter from a State agencyexpressed this concern:

One of our key concerns * * * is that thegeneral public will become ambivalent if weforecasted 20, 30, or more Code Red daysover the course of an ozone season. Underthis scenario, people may not take adequateprecautions to protect themselves when anactual unhealthy level is reached. (DocketNo. A–98–20, IV–G–05).

A commenter from another Stateagency expressed a similar view:

It is important to make sure that thisgeneral message is not jeopardized bytreating the new 85 ppb, 8-hour standard asthe bright line between healthy andunhealthy. The Code Red message will not beconsidered credible if it is issued between 40to 60 times a summer in our area. Last yearthere were 54 days * * * where the 8-hourstandard was exceeded. (Docket No. A–98–20, IV–G–13).

From the comments we have receivedand from our focus group research, webelieve that the color red sends toostrong a message for use in the



‘‘unhealthy for sensitive groupscategory.’’ Additionally, based on thecomments of State and local agenciesthat have used the same or a similarcolor scheme, we believe that the colororange sends an appropriate healthmessage and yet a strong message thatthe standard has been exceeded. OneState commenter noted that theirenvironmental agency:has been using a green/yellow/orange/redcommunication system since 1993. Themedia has used the red, orange and yellowair quality codes to convey a ‘‘the air is notclean’’ message. In general, the media hasused Code Red to convey a message that airpollution is or will be at a near emergencylevel. Code Orange has connoted ‘‘verydirty.’’ Code Yellow has, in general, beenused to characterize air pollution as not toobad—but still not clean. (Docket No. A–98–20, IV–G–13).

Another State commenter noted:We disagree, however, with * * * [the]

assertion that the ‘‘Code Orange’’ message inthe PSI does not adequately protect publichealth. Our experience * * * has been thatthe health message can be effectivelydelivered for Code Orange levels. We havereceived much feedback from the generalpublic about our ozone action day program,and the resounding message has been: Thankyou for this program, I can now plan my dayto avoid exposure to high levels of ozone.(Docket No. A–98–20, IV–G–05).

In addition, ozone mapping projectshave successfully represented airquality using the full AQI color scheme.In the Ozone Mapping Project,described in section I.C.2, the proposedAQI color scheme was used successfullyduring the 1998 O3 season. ParticipatingState and local agencies and regionalorganizations have selected the samecolor scheme for use in the 1999 O3

season. Having used the proposed colorscheme in their local O3 map, onemetropolitan air agency noted that‘‘EPA’s proposed color schemecommunicates clearly in a logicalprogression which in our experience isalready understood by the public andthe media.’’ (Docket No. A–98–20, IV–G–11).

Because we believe the proposedcolor scheme effectively andappropriately communicates the healtheffects message that was the basis forsetting the O3 and PM standards, wehave adopted the color scheme asproposed. However, we strongly agreewith the views expressed bycommenters that it is important for thehealth effects message associated withthe category ‘‘unhealthy for sensitivegroups’’ to be effectively communicatedto the public, health care providers andthe media. It is very important thatmembers of sensitive groups, which for

some pollutants includes healthychildren and adults, be alerted topotential health risks and that thegeneral public be motivated to takeemissions reductions measures when airquality is above the level of thestandard. In response to the concernsexpressed by these commenters, we areplanning to significantly step up thedevelopment of education and outreachmaterials and increase activities to getthis message out, as discussed in sectionII.D below.

Only two commenters recommendedagainst requiring specific colors. Thefirst commenter did so on the groundsthat requiring specific colors would beunenforceable, and may lead tofrustration and conflict. Whileapplauding our goal of establishing aconsistent message, and agreeing that itis good to have as much nationalconsistency as possible, this commenternoted that efforts to legislate aestheticsare uncomfortable, unwieldy andultimately unnecessary. (Docket No. A–98–20, IV–D–11). The secondcommenter noted that some States mayelect to use Code Red for ozone actionprograms at levels other than what isbeing proposed and the regulationshould not preclude them from doingthat. (Docket No. A–98–20, IV–D–19).On the other hand, there was verystrong support in the comments for usto require that agencies that use color,use specific colors in AQI reporting. Allof the other commenters that addressedthis issue, including a commenter froman environmental organization,supported requiring specific colors forall State/local agencies using a colorformat. The commenter from anenvironmental group noted:

EPA states that revisions to the PSI haveas a goal the creation of a nationally uniformlink across a range of programs. We urge thatthis uniformity be achieved through the useof a national public health warning systemthat is clear to the public. To this end, wedo support the EPA requiring that whencolors are used by a state in its PSI, that thesame color system incorporated in the PSI,and not variants, be utilized by such state.(Docket No. A–98–20, IV–D–21).

One of the many State commentersagreeing with us that such arequirement was necessary for nationaluniformity, noted that ‘‘Specific colors* * * associated with each categoryshould be required for nationaluniformity and ease of understanding.Anything less would defeat the purposeof a national index for comparing airquality in different locales.’’ (Docket No.A–98–20, IV–D–07). Another Statecommenter made the point that‘‘Consistency of message is important,especially if the regional nature of many

air pollution problems is to becommunicated effectively.’’ (Docket No.A–98–20, IV–D–01).

In response to the first commenter’sobjections, we do not believe thatrequiring specific colors presents anyparticular enforceability problems. Thisrequirement is one of many contained inthe 40 CFR part 58 Ambient Air QualitySurveillance requirements and would beenforceable in the same manner and tothe same extent as any otherrequirement of this section. As such, webelieve there is no difference inenforceability between this and arequirement for the use of particulardescriptors or air quality index values.We expect to work with EPA RegionalOffices to ensure that they monitor Stateimplementation of the revised AQI andwork with the States to encouragecompliance.

With regard to comments that ourrequirement would preclude States fromusing other color schemes and actionlevels in their voluntary programs, it isimportant to note that the AQI addressesthe reporting of measured air qualityand does not impose any requirementsor limitations on community actionprograms based on air quality forecasts.We recognize that a nationally uniformcolor scheme for AQI reporting will, asa practical matter, complicate a State’sefforts to use other color schemes inaction programs based on predicted airquality, but they remain free to do sounder our regulations.

Because it is the fundamental goal ofthe AQI to provide nationally uniforminformation about daily air quality andthe public health messages that areappropriately associated with variousdaily air quality levels, in a format thatis timely and easily understood, wecontinue to believe that requiringspecified colors when the AQIcategories are reported in color format isboth necessary and appropriate. Neitherof the commenters opposing thisrequirement addressed how a moreflexible approach of recommendingspecific colors, thereby allowing the useof different colors to represent the samerange of air quality, would satisfy thestatutory language requiring a nationallyuniform air quality index. Therefore, weare adopting the requirement, asspecified in appendix G below, thatwhen State and local agencies report theAQI in a color format, that the specificcolors listed in Table 1 be associatedwith each category.

b. Reporting requirements. Wereceived significant comments onseveral issues related to the reportingrequirements, including the populationthreshold and other aspects of thereporting requirements, the appropriate



method of monitoring and reporting thePM sub-indices, the effect of AQIchanges relative to the SHL program,and the effective date of the final rule.Since we received no significantcomments on our proposal to change therounding conventions for calculatingthe index to make them consistent withthe rounding conventions used indefining the NAAQS, we are adoptingthat revision as proposed. With regardto the population threshold, onecommenter expressed the view that thechange from requiring AQI reporting inurbanized areas with a populationgreater than 200,000, to requiringreporting in MSAs with populationsgreater than 350,000, would raise thethreshold for the requirement andappear to mean that large segments ofthe U.S. population would not haveaccess to AQI reporting. (Docket No. A–98–20, IV–D–03). We have adopted therequirement for AQI reporting in MSAswith populations greater than 350,000 tobe consistent with the State/Local AirMonitoring Stations (SLAMs)monitoring regulations in 40 CFR part58, since AQI reporting is based oninformation from SLAMS monitors thatare located and reported within thecontext of MSAs. The use of MSAs alsoprovides for more stable reporting areassince MSAs are usually defined bycounty boundaries that typically do notchange, whereas the boundaries forurbanized areas are very irregular, mayinclude parts of counties, and maychange with each census. In selectingthe MSA population threshold of350,000, we tried to make the newreporting requirement equivalent to theold one. Under the new requirement,virtually the same number of cities willbe required to report the AQI as werepreviously. Because urbanized areas andMSAs are not equivalent, we realize thatsome areas will be required to report theAQI that were not required to do sobefore this rulemaking, and vice versa.The regulation does not preclude anyarea from reporting the AQI, and weencourage State and local air agencies toreport the AQI whenever possible sothat people will be informed about localair quality.

Another commenter noted that someMSAs fall within the boundaries ofmore than one State, and requested thatwe identify which of the two or morereporting agencies would be responsiblefor reporting the AQI for the MSA.(Docket No. A–98–20, IV–G–15). Weexpect that decisions about AQIreporting in multi-State MSAs will bemade by participating agencies in thesame manner as decisions aboutactivities to implement the standards

through the State Implementation Plans(SIPs). Guidance for air quality planningand implementation in MSAs that fallwithin the boundaries of more than oneState generally calls for the participatingState and local agencies to identify, inthe SIPs for those States, who will beresponsible for the preparation andsubmission of the required elements,including AQI reports. Where a local orregional planning organization has beendesignated to carry out suchrequirements, such an organization isthe appropriate one to report the AQI.In any case, we encourage AQI reportingon the sub-MSA level, especially wherethe AQI differs within the MSA.

Another commenter urged us toexpand the requirement for AQIreporting to areas with populations lessthan 350,000, if these areas are likelynot to be in attainment for the 8-hour O3

standard. To support this position, thecommenter noted that O3 can betransported long distances downwindfrom where it is generated, resulting inserious air quality problems indownwind rural and smaller urbanareas. (Docket No. A–98–20, IV–G–27).We agree with this commenter thatdownwind areas may be significantlyaffected by transport of O3 andprecursors. In section 5 of appendix G,we encourage States to evaluate airquality in affected areas downwind ofMSAs to identify the potential forsignificant transport-related air qualityimpacts and to expand their AQIreporting to address these situations. Wehave also changed the language in thissection such that the affected area neednot be contiguous to the reporting MSA.

On a related topic, one commenternoted an example in which a MSA witha population greater than 350,000, hasnot registered AQI values in excess of 50(such that AQI reporting would bediscretionary), although values above100 are registered infrequently at anational monument within the larger airbasin. (Docket No. A–98–20, IV–G–17).This commenter requested that werevise the reporting requirements to addan air quality consideration to thepopulation threshold as a secondcomponent of AQI reporting. To addressone part of this comment, we encourageState and local air agencies to report theAQI and issue forecasts for nationalparks or monuments whenever possible,since these are places people go to foractivities that often involve prolongedor vigorous exertion, thereby increasingthe risk from air pollution. We haveworked with the National Park Serviceto develop appropriate guidance forvisitors and staff to use when indexvalues are expected to be above 100 forO3. To address the other part of this

comment, section 8 of appendix Gdescribes exceptions under which AQIreporting becomes discretionary, eitherfor one pollutant or the entire index, forareas with good air quality. Regardingthese exceptions, a State commentersuggested that we require a minimum of2 years at an AQI value lower than 50before allowing agencies to ‘‘opt out’’ ofreporting the AQI for a particularpollutant, so that for example, oneunusually good O3 season would notmake it possible for an agency to avoidreporting high index values insubsequent O3 seasons. (Docket No. A–98–20, IV–D–06). We believe thatrequiring 2 years of index values lowerthan 50 before allowing State and localagencies discretion in reporting, whileappropriate in some situations, may beunnecessary in others. We agree withthis commenter that it is appropriate torequire reporting of higher index values,even if air quality has been goodthroughout the previous year. Therefore,we have revised section 8 of appendixG, such that when the criteria for anexemption are no longer met, theresponsible agency is required to reportthe AQI. Another commenter expressedthe view that we should strengthen theminimum notification requirements, sothat when the AQI value exceeds 100,State and local agencies are required toreport the index to all three media(print, radio and television) to helpensure that the public is informed thatthe standard has been exceeded. (A–98–20, IV–E–3) We agree that it is importantto inform the public when the AQI isabove 100, and therefore havestrengthened the reporting provisions insection 6 of appendix G. In particular,when the AQI exceeds 100, reportingagencies should expand reporting to allmajor news media, and at a minimum,should include notification to the mediawith the largest market coverages for thearea in question.

Looking at these reporting provisionsmore broadly, we believe that it wouldbe very beneficial for reporting agenciesto educate the media about alternativesources for this information, such asweb sites and community actionprograms. Many State and local agencieshave web sites that provide quick accessto timely and accurate air quality andrelated information. For State and localagencies participating in the OzoneMapping Project, the media could bedirected to the AIRNOW web site as asource of information about O3 airquality and associated health effects foryesterday, today and tomorrow. Inaddition, this web site provides in-depth information about O3 healtheffects, sources of emissions and simple



measures people can take to improve airquality. Community action programsalso provide timely and accurateinformation, and are often used toinform the public when air quality ispredicted to be above an index value of100. Tools and programs such as thesecan significantly improve the timelinessof AQI reporting and provide additionaluseful information. We believe that, inthe near future, the AQI will be reportedby the regional and national media inways, such as the Ozone Map, that willnot be limited to specific MSAs. Thistype of approach will help provide AQIreporting for areas that would otherwisenot be covered, including, in somecases, rural and small urban areas andnational parks.

Regarding reporting the PM sub-indices, one commenter requested thatwe clarify whether PM2.5 and PM10

should be treated as one pollutant (e.g.,reported simply as PM) or two differentpollutants (e.g., reported separately).(Docket No. A–98–20, IV–D–19). Weexpect State and local air agencies toreport PM2.5 and PM10 separately, sincethere are two separate sub-indices withdifferent sensitive groups, and differenthealth effects and cautionarystatements. In response to thiscomment, we have added clarifyinglanguage to section 9 in appendix G. Inaddition, many commenters noted thatat the present time there is very littlemonitoring for PM (both PM2.5 andPM10) that is suitable for use in dailyAQI reports, and requested guidance forthe use of non-reference methods for thepurpose of AQI reporting. Since PM isoften measured at intervals longer thanevery 24-hours, State and local agenciesare encouraged to use monitoring datafrom continuous PM monitors for use inAQI reporting, whenever possible. Asnoted by commenters, due to the lack ofappropriate monitoring information, atthis time it may not be possible to reportthe AQI for PM in many locations. Toassist State and local agencies in the useof non-reference methods, we haveadded language to section 10 ofappendix G stating that non-referencemethods may be used for the purpose ofAQI reporting if it is possible todemonstrate a simple linear relationshipbetween the non-reference and thereference methods.

Regarding the effect of changes to theAQI on the SHL program, we receivedtwo significant comments. Onecommenter noted that our proposedchanges to the categories, to standardizethem such that the upper bound falls onan even number, rounded to 50 (e.g.,200), and lower bound falls on an oddnumber (e.g., 201), resulted in the AQIbreakpoint of 200 being the upper

bound of the ‘‘unhealthy’’ category,rather than the lower bound of the ‘‘veryunhealthy’’ category, as it has beenhistorically. Since the AQI breakpoint of200 is also commonly used as the ‘‘AlertLevel,’’ or the first stage of an airpollution emergency episode inexample guidance associated with theSHL program, this commenter requestedthat we leave the AQI value of 200 asthe lower breakpoint of the ‘‘veryunhealthy’’ category, so that emergencyepisodes would start when air quality isclassified as ‘‘very unhealthy’’ andinclude appropriate-sounding healtheffects and cautionary statements.(Docket No. A–98–20, IV–D–22). We areadopting the breakpoints as proposed,because we believe that it is importantto be consistent in the treatment of thecategory boundaries (e.g., 51 to 100, 101to 150, 151 to 200, etc.). When wepropose revisions to the requirements ofthe SHL program, we plan to change allreferences to the ‘‘Alert Level’’ so theywill refer to air quality that exceeds the‘‘Alert Level,’’ rather than to air qualitythat reaches the ‘‘Alert Level.’’ However,State and local agencies should notchange their emergency episode plans atthis point simply because we areadopting this consistent approach tosetting AQI breakpoints. Eventually,some agencies may have to reviseemergency episode plans because wehave revised the AQI value of 200 forthe 8-hour O3 sub-index. But we do notexpect States to make any revisions totheir emergency episode plans until wepromulgate the revised requirements.Finally, several commenters noted thatin the proposal, we did not specify aneffective date for the final revisions.Some of these commenters suggestedthat we extend the effective date, withsuggestions ranging from 60 days tomore than a year after publication. Weare adopting an effective date of 60 daysafter publication. We believe that thiswill allow adequate time for State andlocal agencies to revise daily AQIreports. We recognize that it may takelonger to revise related informationalmaterials, such as printed documents, orrelated programs that agencies maywant to revise. However, since thisrulemaking applies only to therequirements for daily reporting of airquality, we believe an effective date of60 days is adequate.

c. Index name. All commenters thatexpressed a view on the index namesupported changing the name of theindex from the Pollutant StandardsIndex (PSI) to the Air Quality Index(AQI), because this name clearlyidentifies the index as relating to thequality of the air. Accordingly, we are

changing the name of the index to theAir Quality Index, or AQI.

2. Comments and Responses on Changesto the Sub-Indices.

All of the comments we received onproposed changes to the sub-indicesfocused on the sub-indices that wereadded for O3 (8-hour) and PM2.5. Sincewe did not receive specific commentson the conforming changes we proposedto the CO, SO2 and PM10 sub-indices,we are adopting these sub-indices asproposed.

a. Ozone sub-index. We receivedsignificant comments on two issuesrelated to the O3 sub-index. The firstgroup of comments was in response toour request for comment on retainingthe 1-hour O3 sub-index in addition tothe 8-hour O3 sub-index. The secondgroup of comments focused on theappropriateness of providingprecautionary language below the levelof the 8-hour O3 standard. Regarding the1-hour sub-index, almost all of thecomments that addressed this issuesupported retaining the 1-hour O3 sub-index. However, one State commenterexpressed the view that the proposalwas unclear regarding how areas thathave not attained the 1-hour O3

standard are to use the new 8-hour O3

sub-index. This commenter also notedthat it might be confusing to report theAQI based on the 8-hour O3 sub-indexin an area where the 1-hour O3 standardhad not yet been attained. (Docket No.A–98–20, IV–D–07). We are requiringthat all State and local agencies thatreport the AQI for O3 calculate the 8-hour O3 sub-index, even if the reportingarea has not attained the 1-hourstandard. In addition to calculating the8-hour O3 sub-index, which is required,the reporting agency may also calculatethe 1-hour O3 sub-index, but this is notrequired. However, if the reportingagency calculates both O3 sub-indexvalues, it is required to report the higherindex value of the two. The AQI doesnot relate to attainment status; rather, itis a tool for reporting daily air qualityand associated health information. Weare retaining the 1-hour O3 sub-indexonly because we recognize that there area very small number of areas in the U.S.that have atypical air quality patterns,with very high 1-hour daily peak O3

concentrations relative to 8-houraverage concentrations. In such areas,an index value greater than 100 mightbe calculated using the 1-hour sub-index, even when the 8-hour sub-indexmight be below 100. For these areas, theuse of the 1-hour sub-index is clearlymore precautionary. Because our majorinterest is that appropriateprecautionary messages be issued, we



8 See 63 FR 67819, 67829 (Dec. 9, 1998).9 See 62 FR 38669–71, 38676–77 (July 18, 1997).

are not retaining a complete 1-hour O3

sub-index with ‘‘good’’ and ‘‘moderate’’categories. Likewise, when ambient 8-hour O3 concentrations are greater than0.374 ppm, reporting agencies mustcalculate the index value using the 1-hour O3 sub-index. This is because nohuman health effects information isavailable for higher 8-hour average O3

concentrations to use as a basis forselecting 8-hour breakpoints and fordeveloping appropriate health effectsand cautionary statements. We believethat since State and local agencies arerequired to report the name of thepollutant responsible for an index valuegreater than 100, but not the associatedaveraging period, using the 8-hour O3

sub-index should not be confusing inareas that have not yet attained the 1-hour O3 standard.

Regarding the issue of alertingsensitive individuals below the level ofthe 8-hour O3 standard, somecommenters not only suggested addinga category below the level of thestandard, but also suggested reducingthe lower bound of the ‘‘moderate’’category. (Docket No. A–98–20, IV–D–11, IV–D–17, IV–D–19, IV–G–21). Weare not adding a category below thelevel of the standard as discussed insection II.B.1. above. However, to besomewhat more precautionary, we haveexpanded the ‘‘moderate’’ range byreducing the lower bound of thiscategory from 0.070 ppm to 0.065 ppmO3, 8-hour average. We believe thatsetting the breakpoint between the‘‘good’’ and ‘‘moderate’’ categories atthis lower level, is appropriate, based inpart on risk estimates done inconjunction with the review of the O3

NAAQS which suggested that risk tohealthy people likely becomesnegligible at this level (Whitfield et al.,1996). This change is also responsive tocomments from State agencies that theproposed range of the ‘‘moderate’’category was so narrow (spanning only15 ppb O3, as compared to 20 ppb rangeused in the Ozone Map in 1998) that itwould be more difficult to forecastaccurately and also would provide tooquick a transition from good tounhealthy. (Docket No. A–98–20, IV–D–10, IV–G–04). Conversely, an industrygroup and a State commenter tookexception to issuing a ‘‘limited healthnotice’’ for O3 that we proposed as thepurpose of the ‘‘moderate’’ category.(Docket No. A–98–20, IV–D–12, IV–G–14). The State commenter objected tothe use of the term ‘‘health notice’’below the level of the standard becauseit implies that the standard is notprotective of public health. In additionto stating that the ‘‘limited health

notice’’ associated with moderate airquality is inconsistent with the 8-hourO3 standard because the standard isintended to protect public health, eventhe health of sensitive populations, withan adequate margin of safety, theindustry commenter expressed the viewthat we should omit from our materialsthe health effects and cautionarystatements suggesting that air qualitymeeting the level of the standard is athreat to health. We agree with theindustry and State commenters thatsince the 8-hour O3 standard is intendedto protect public health, including thehealth of sensitive groups, with anadequate margin of safety, that the term‘‘limited health notice’’ may bemisleading. However, we continue tobelieve that it is appropriate to provideguidance with cautionary language forextremely sensitive individuals, notpopulations or groups, below the levelof the standard. This approach isconsistent with the advice of CASAC,and the way we discussed expandingthe use of the AQI, specifically tocaution extremely sensitive individualsbelow the level of the O3 standard, inthe O3 proposal and final decisionnotices.

b. PM2.5 sub-index. We received anumber of comments regarding thePM2.5 sub-index, almost all of themfocusing on our proposal to set theindex value of 100 at the level of the 24-hour standard (65 µg/m3). Somecommenters recommended setting anindex value of 100, or otherwiseproviding for cautionary messages, atconcentrations lower than 65 µg/m3.One commenter, for example, stated thatunder the proposal ‘‘many areas of thecountry will likely violate the annualstandard of 15 µg/m3 without ever (orhardly ever) reaching a PSI of 100 or acategory indicating some degree ofunhealthfulness. This situation willresult in an inconsistent andinappropriate message to the public,especially given the severe health effectsassociated with fine particles.’’ (DocketNo. A–98–20, IV–D–11).

In light of these comments, we havereexamined the basis for selecting PM2.5

AQI breakpoints and agree that the sub-index as proposed would not adequatelycaution sensitive groups about potentialrisks associated with short-termexposures to PM2.5. This is essentiallybecause the proposed PM2.5 sub-indexwas developed using the Agency’shistorical approach to selecting indexbreakpoints, which on examination doesnot correspond well with the way thePM2.5 standards were intended tofunction. The historical practice hasbeen simply to set the AQI value of 100at the level of the short-term standard

for a pollutant (in this case, the 24-hourPM2.5 standard) and the AQI value of 50at the level of the annual standard, ifthere is one, or at one-half the level ofthe short-term standard.8 This methodof structuring the index is appropriatefor a ‘‘typical’’ suite of air-qualitystandards, which includes a short-termstandard designed to protect against thehealth effects associated with short-termexposures and an annual standarddesigned to protect against health effectsassociated with long-term exposures. Insuch cases, the short-term standard ineffect defines the level of healthprotection provided against short-termrisks and thus is a useful benchmarkagainst which to compare daily air-quality concentrations.

In the case of the PM2.5 standards,however, EPA took a different approachto protecting against health risksassociated with short-term exposures.For reasons discussed in the preambleto the final standards, the annual and24-hour PM2.5 standards were designedto work together for this purpose, andthe intended level of protection againstshort-term risk is not defined by the 24-hour standard but by the combination ofthe two standards working in concert.Indeed, the annual PM2.5 level of 15 µg/m3 was intended to serve as theprincipal vehicle for protection againstshort-term PM2.5 exposures (by reducingthe entire distribution of PM2.5

concentrations in an area), with theshort-term standard serving essentiallyto provide supplemental protection inspecial situations. 9 Given the respectiveroles of the two standards, setting theAQI value of 100 at the level of the 24-hour standard would not reflect theshort-term health risks associated withlower concentrations, which the annualstandard was designed to address.Accordingly, we agree that it isappropriate to caution members ofsensitive groups below the level of the24-hour standard and believe thisshould be done in a way that reflects theintended roles of both standards inprotecting against short-term risks.

It would also be inappropriate tocompare daily air-quality concentrationsdirectly with the level of the annualstandard (by setting the AQI value of100 at that level), because the annualstandard represents an average of manydaily concentrations rather than dailyvalues per se. In the circumstances, webelieve the guiding principle for PM2.5

should be to set the AQI value of 100in a way that, at least conceptually,reflects the general level of healthprotection against short-term risks



10 See 62 FR 38670, 38677 (July 18, 1997).11 See 63 FR 67824, 67832 (Dec. 9, 1998).

12 As discussed in the proposal, should the finalSHL for PM2.5, when promulgated, be different fromthis concentration, we will revise this PM2.5 sub-index accordingly.

provided by the annual and 24-hourstandards in combination. Thisapproach, although inexact, isconsistent with the historical approach,in that the underlying logic of thatapproach, as applied to a typical suiteof standards, is also to set the AQI valueof 100 in a way that reflects the level ofprotection provided against short-termrisks—that is, by setting it at the levelof the short-term standard that providesthe protection. In the case of PM2.5, asindicated above, the level of the 24-hourstandard (65 µg/m3) is too high to reflectthe intended level of protection, and thelevel of the annual standard (15 µg/m3)is too low. Between the two values, theavailable health studies indicate acontinuum of risks associated withincreasing PM concentrations, althoughwith significant uncertainties as to theextent of the risk associated with singlepeak exposures.10 Consistent with EPA’sgeneral practice of setting AQIbreakpoints in symmetrical fashionwhere health effects information doesnot suggest particular levels,11 weconcluded that it is appropriate to setthe AQI value of 100 at the mid-pointof the range between the annual and the24-hour PM2.5 standards (40 µg/m3).Given that decision, we also concludedthat it is appropriate to retain the levelof the annual standard for an AQI valueof 50, as proposed, and to set the AQIlevel of 150 at the level of the 24-hourstandard.

To reiterate, the purpose of setting theAQI value of 100 somewhat below thelevel of the 24-hour standard was toreflect the dual role of the annual and24-hour PM2.5 standards in protectingagainst short-term risks, and the aimwas to select a breakpoint that wouldserve as a rough surrogate for thegeneral level of protection provided bythe two standards in combination.Given the nature of the standards andthe available health information, a moreexact approach was not possible. In thisregard, setting the breakpoint at themid-point of the range between theannual and 24-hour standards, asopposed to a level somewhat higher orlower within that range, simplyreflected EPA’s general practice ofsetting symmetrical breakpoints asindicated above, and does not imply anysort of health-effects threshold. Inparticular, it does not reflect a judgmentabout the extent of the risk associatedwith single peak concentrations ofPM2.5, as to which the available healthinformation is inconclusive, or the levelat which EPA might set a 24-hourstandard if the annual standard did not

serve as the primary vehicle forprotection against such concentrations.As with other AQI breakpoints, it alsohas no effect on the degree of controlrequired of specific sources.

In short, EPA’s decision to treat theannual standard as the principal vehiclefor protecting against short-term PM2.5

concentrations, although judged to bethe best approach based on the availablehealth information, does present adifferent situation than that involved inprevious AQI rulemakings. As discussedin the preamble to the final standards,the annual standard was intended toreduce all PM2.5 concentrations,including short-term peaks, in an areasufficiently to protect public health withan adequate margin of safety, aside fromspecial situations which the 24-hourstandard was designed to address. Asone commenter suggested, however, itwould be possible for an area to violatethe annual standard without everexperiencing (or seldom experiencing)daily peaks that exceeded the level ofthe 24-hour standard. Moreover, itmight be difficult, if not impossible, topredict in advance whether the annualstandard will be attained in a given area.For these reasons, as well as theuncertainties in the available healthinformation, it is inherently difficult tojudge the significance of single peakconcentrations when they occur. Inview of the various uncertaintiesinvolved, particularly sensitiveindividuals may wish to avoid exposureto such concentrations, especiallyconcentrations that approach the levelof the 24-hour standard. To facilitatesuch choices, consistent with thepurposes of the AQI and the advice ofCASAC, we believe that cautioningmembers of sensitive groups in therange of 40 to 65 µg/m3 is appropriate.

We did not receive any comments onthe proposal to establish a concentrationof 500 µg/m3 to be associated with aPM2.5 index value of 500, or our methodof selecting the intermediatebreakpoints. Therefore, we are adopting500 µg/m3 as the upper bound of theindex.12 For intermediate breakpoints inthe AQI between values of 150 and 500,we have adopted PM2.5 concentrationsthat generally reflect a linearrelationship between increasing indexvalues and increasing PM2.5 values. Asdiscussed in the proposal, the generallylinear relationship between AQI valuesand PM2.5 concentrations in this range,rounded to increments of 50 µg/m3 toreflect the approximate nature of such a

relationship, is consistent with thehealth effects evidence that was thebasis for the PM standards.

C. What Are the Final Revisions?The sub-sections below only

summarize changes to the regulatorytext. They do not describe all aspects of40 CFR part 58.50 or appendix G.

1. What Are the General Changes?Based on the proposed structure of

the AQI, the comments we received andour responses to them, as discussedabove, we are adopting the followingchanges to the general structure andreporting requirements to the AQI.

a. Categories and related descriptors,index values and colors. We areadopting the index values, descriptorsand associated colors listed in Table 1above.

b. Reporting requirements. We arerevising 40 CFR 58.50 to requirereporting of the AQI in all MSAs witha population over 350,000. In appendixG, we are adopting roundingconventions to be used to calculateindex values that are consistent with therounding conventions used in definingthe NAAQS for each pollutant.

The final rule retains therequirements to identify the area forwhich the AQI is being reported, thetime period covered by the report, the‘‘critical’’ pollutant for which thereported AQI value was derived, theAQI value, and the associated categorydescriptor. The final rule adds tworequirements: (1) To report theassociated category color if a colorformat is used and, (2) to report thepollutant-specific sensitive group forany reported index value greater than100. The final rule encourages, but doesnot require, that AQI reports include:appropriate health effects andcautionary statements, all AQI valuesgreater than 100, the AQI for sub-divisions of the MSA (if there areimportant differences in air qualityacross sub-divisions of the MSA),possible causes for high index values,and the actual pollutant concentrations.

In the case of rural or small urbanareas that are significantly affected bypollutants transported from a MSAwhere the AQI is reported, the final rulerecommends that the MSA report theAQI for the affected areas as well. Inaddition, when the AQI is greater than100, reporting agencies should expandAQI reporting to include all major newsmedia. The final rule continues to allowagencies to discontinue reporting forany pollutant, if index values for thatpollutant have been below 50 for anentire season or a year. However, if insubsequent years pollutant levels rise



such that index values for that pollutantwould be above 50, then the final rulerequires that AQI reporting for thatpollutant resume. The final ruleemphasizes the importance offorecasting the AQI by specifying thatforecasted values should be reported,when possible, but does not require thatforecasted values be reported.

c. Index name. We are adopting thename the Air Quality Index or AQI.

2. What Are the Changes to the Sub-Indices?

Based on the proposed sub-indices,the comments we received and ourresponses to them, as discussed above,we are adopting new sub-indices

corresponding to the 8-hour O3 standardand the PM2.5 standards, as well asconforming changes to the CO, 1-hourO3, PM10, and SO2 sub-indices. Theadopted breakpoints for the O3 (8-hourand 1-hour) PM2.5, PM10, CO and SO2

sub-indices are listed in Table 3.

TABLE 3.—BREAKPOINTS FOR O3, PM2.5, PM10, CO, AND SO2 SUB-INDICES

AQI value

O3 PM

CO, 8-hr (ppm) SO2, 24-hr (ppm)8-hr (ppm) 1-hr (ppm) PM2.5, 24-hr

(µg/m3)PM10, 24-hr

(µg/m3)

50 ................ 0.06 ............. ................................. 15 50 4 0.03100 .............. 0.08 ............. 0.12 40 150 9 0.14150 .............. 0.10 ............. 0.16 65 250 12 0.22200 .............. 0.12 ............. 0.20 * 150 350 15 0.30300 .............. 0.40 (1-hr) ... 0.40 * 250 420 30 0.60400 .............. 0.50 (1-hr) ... 0.50 * 350 500 40 0.80500 .............. 0.60 (1-hr) ... 0.60 * 500 600 50 1.00

* If a different SHL for PM2.5 is promulgated, these numbers will be revised accordingly.

These sub-indices are presented inmore detail in appendix G to reflect thechanges to the numerical roundingconventions for calculating indexvalues.

D. What Are the Related InformationalMaterials?

The primary documents associatedwith the AQI and this rulemaking, areour guidance on AQI reporting,‘‘Guideline for Public Reporting of DailyAir Quality—Air Quality Index (AQI)’’(EPA 1999a), and our guidance on AQIforecasting, ‘‘Guideline for Developingan Ozone Forecasting Program’’ (EPA1999b). These documents are availableon AIRLINKS (http://www.epa.gov/airlinks). The AQI Reporting documentcontains information regarding the AQIrequirements and recommendations,example AQI reports, and a list of MSAsrequired to report the AQI. It alsoincludes pollutant-specific healtheffects and cautionary statements for usewith the index, for O3, PM2.5, PM10, CO,and SO2. The AQI Forecastingdocument explains the steps necessaryto start an air pollution forecastingprogram. Included in the document isguidance on using hourly O3

concentrations as predictors for 8-houraverages.

Other related informational materialsare also available. The brochure ‘‘ThePollutant Standards Index’’ (EPA 1994)contained general information about thehealth effects and air quality, andgeneral precautions that sensitivegroups and the general public can taketo avoid exposures of concern. It isbeing revised to be consistent with thenew name (i.e., the Air Quality Index

brochure), with final revisions to theAQI, and will identify sensitive groupsin the health effects statements for eachof the pollutants, and include thepollutant-specific health effects andcautionary statements discussed above.A colorful fact sheet, called the ‘‘AirQuality Guide,’’ provides informationabout the AQI, O3 health effects and thesources of ground-level O3 is availableon the AIRNOW web site. A revisedbooklet, ‘‘SMOG—Who Does It Hurt?,’’provides information for the generalpublic about O3 health effects and isbased on scientific information gainedin the recent review of the O3 standard.‘‘SMOG—Who Does It Hurt?’’ wasdesigned to provide, in simple language,enough detail for individuals tounderstand who is at most risk from O3

exposure and why, the nature of O3

health effects, and a detailedexplanation of how individuals canreduce the likelihood of exposure usingcommon everyday activities asexamples. We are also developing ashorter, summary pamphlet about O3

health effects to complement the‘‘SMOG—Who Does It Hurt?’’ booklet.We expect the AQI brochure, ‘‘SMOG—Who Does It Hurt?’’ and the shortersummary pamphlet about O3 healtheffects to be available in paper formatand on the AIRNOW web site early inthe 1999 ozone season. In addition, wewill translate the Air Quality Guide, theAQI brochure, ‘‘SMOG—Who Does ItHurt?’’, and the shorter summarypamphlet into Spanish. These materialswill be available on a Spanish page onthe AIRNOW web site.

There are other materials available onthe AIRNOW web site that provide

general information about O3.Information about ground-level ascontrasted to stratospheric O3 may befound in EPA’s publication ‘‘Ozone:Good Up High, Bad Nearby.’’ The EPA’svideo, ‘‘Ozone Double Trouble’’ alsoprovides information about ground-leveland stratospheric O3 and the healtheffects associated with exposure toground-level O3, or smog.

In addition to the products discussedabove, to address the concerns ofcommenters that when air quality is inthe ‘‘unhealthy for sensitive groups’’range the public will not understandthat the standard has been exceeded orwho is at risk, we are going tosignificantly increase education andoutreach related to the AQI. At thispoint, we are still in the process ofplanning specific new products oractivities, but have decided whatgeneral direction these efforts will take.First, we plan to increase our contactswith the news providers to better informthem about the importance of includingaccurate, timely and understandableinformation in their broadcasts andreporting, and to enlist them as fullpartners in the implementation of theAQI. Second, we plan to form newassociations with health care providersto keep them informed about airpollution health effects, since theseprofessionals are the most trustedsource of health effects information.Third, we plan to increase directoutreach to the public through a varietyof means, including materials tailored toschool-age children, the Spanish-speaking community, and others.Finally, we plan to work with publichealth interest organizations to support



their efforts to provide more immediateand interactive education and outreachto all of these groups.

III. Regulatory and EnvironmentalImpact Analyses

A. Executive Order 12866: OMB Reviewof ‘‘Significant Actions’’

Under Executive Order 12866, theAgency must determine whether aregulatory action is ‘‘significant’’ and,therefore, subject to Office ofManagement and Budget (OMB) reviewand the requirements of the ExecutiveOrder. The order defines ‘‘significantregulatory action’’ as one that may:

(1) Have an annual effect on theeconomy of $100 million or more oradversely affect in a material way theeconomy, a sector of the economy,productivity, competition, jobs, theenvironment, public health or safety, orState, local, or tribal governments orcommunities;

(2) Create a serious inconsistency orotherwise interfere with an action takenor planned by another Agency;

(3) Materially alter the budgetaryimpact of entitlements, grants, user fees,or loan programs or the rights andobligations or recipients thereof; or

(4) Raise novel legal or policy issuesarising out of legal mandates, thePresident’s priorities, or the principlesset forth in the Executive Order. TheEPA has determined that the revisionsto air quality index reporting in thisfinal rule would not have an annualeffect on the economy of $100 millionor more or adversely affect in a materialway the economy, a sector of theeconomy, productivity, competition,jobs, the environment, public health orsafety, or State, local, or tribalgovernments or communities, andtherefore did not prepare a regulatoryimpact assessment. The OMB hasadvised us this final decision should beconstrued as a ‘‘significant regulatoryaction’’ within the meaning of ExecutiveOrder 12866. Accordingly, this actionwas submitted to the OMB for review.Any changes made in response to OMBsuggestions or recommendations will bedocumented in the public record andmade available for public inspection atEPA’s Air and Radiation DocketInformation Center (Docket No. A–98–20).

B. Regulatory Flexibility Analysis/SmallBusiness Regulatory EnforcementFairness Act

Under the Regulatory Flexibility Act(RFA), 5 U.S.C. 601 et seq., EPA mustprepare a regulatory flexibility analysisassessing the impact of any proposed orfinal rule on small entities. Under 6

U.S.C. 605(b), this requirement may bewaived if EPA certifies that the rule willnot have a significant economic impacton a substantial number of smallentities. Small entities include smallbusinesses, small not-for-profitenterprises, and governmental entitieswith jurisdiction over populations lessthan 50,000 people.

Today’s final decision to revise theAQI program modifies existing airquality reporting requirements forMSA’s with populations over 350,000people. Today’s final decision will notestablish any new regulatoryrequirements affecting small entities. Onthe basis of the above considerations,EPA certifies that today’s final decisionwill not have a significant economicimpact on a substantial number of smallentities within the meaning of the RFA.Based on the same considerations, EPAalso certifies that the new small-entityprovisions in section 244 of the SmallBusiness Regulatory EnforcementFairness Act (SBREFA) do not apply.

C. Unfunded Mandates Reform ActTitle II of the Unfunded Mandates

Reform Act of 1995 (UMRA), PublicLaw 104–4, establishes requirements forFederal agencies to assess the effects oftheir regulatory actions on State, local,and tribal governments and the privatesector. Under section 202 of the UMRA,EPA generally must prepare a writtenstatement, including a cost-benefitanalysis, for proposed and final ruleswith ‘‘Federal mandates’’ that mayresult in expenditures to State, local andtribal governments, in the aggregate, orto the private sector, of $100 million ormore in any 1 year. In addition, beforeEPA establishes any regulatoryrequirements that may significantly oruniquely affect small governments,including tribal governments, it musthave developed under section 203 of theUMRA a small government agency plan.The plan must provide for notifyingpotentially affected small governments,enabling officials of affected smallgovernments to have meaningful andtimely input in the development of EPAregulatory proposals with significantFederal intergovernmental mandates,and informing, educating, and advisingsmall governments on compliance withthe regulatory requirements.

The EPA has determined that today’sfinal decision would not include aFederal mandate that may result inestimated costs of $100 million in any1 year to either State, local, or tribalgovernments, in the aggregate, or to theprivate sector. Accordingly, EPA hasdetermined that the provisions ofsection 202 of the UMRA do not applyto this rulemaking. With regard to

section 203 of the UMRA, EPA hasdetermined that this rule contains noregulatory requirements that mightsignificantly or uniquely affect smallgovernments. This rule requiresreporting of the Air Quality Index onlyin MSAs with populations greater than350,000, and therefore does not affectsmall governments.

D. Paperwork Reduction ActToday’s final decision does not

establish any new informationcollection requirements beyond thosewhich are currently required under theAmbient Air Quality SurveillanceRegulations in 40 CFR part 58 (OMB#2060–0084, EPA ICR No. 0940.15).Therefore, the requirements of thePaperwork Reduction Act do not applyto today’s action.

E. Executive Order 13045: Children’sHealth

Executive Order 13045, entitled‘‘Protection of Children fromEnvironmental Health Risks and SafetyRisks’’ (62 FR 19885, April 23, 1997),requires Federal agencies to ensure thattheir policies, programs, activities, andstandards identify and assessenvironmental health and safety risksthat may disproportionately affectchildren. To respond to this order,agencies must explain why theregulation is preferable to otherpotentially effective and reasonablyfeasible alternatives considered by theagency. In today’s final decision, EPAidentified children as one of thesensitive groups which may be atincreased risk of experiencing theeffects of concern following exposure toO3, PM2.5 and NO2.5. The AQI categories,descriptors, and health effects andcautionary statements as proposed, forthe first time reflect consideration of theincreased health risk to children whichmay result from such exposures.Promulgation of the proposed AQI isone potentially effective alternative thatwas considered. However, based oncomments that the public may not beaware that healthy, active children areincluded in the sensitive groups for O3,PM2.5 and NO2, we have adopted theadditional requirement that reportingagencies must include a pollutant-specific statement of the sensitivegroups when an index value of 100 isexceeded. For example, when reportingan AQI value of 110 for ozone, thereporting agency must include astatement that children and people withasthma are the groups most at risk.Whenever the AQI value is above 100for a pollutant, and children are one ofthe sensitive groups for that pollutant,the AQI report must include a statement



that children are at risk. Therefore,today’s action does comply with therequirements of E.O. 13045.

F. Executive Order 12848:Environmental Justice

Executive Order 12848 requires thateach Federal agency make achievingenvironmental justice part of its missionby identifying and addressing, asappropriate, disproportionately highand adverse human health orenvironmental effects of its programs,policies, and activities on minoritiesand low-income populations in theUnited States.

The nature of today’s action is toinform the general public, includingminorities and low-income populations,about the nature of the air pollution inthe areas in which they live. Today’saction establishes a uniform tool forStates to use to develop programs whichwill caution particularly sensitivepeople to minimize their exposures andeducate the public about general healtheffects associated with exposure todifferent pollution levels. States mayalso use information established as partof the AQI to trigger programs designedto reduce emissions to avoidexceedances of the NAAQS. Therefore,today’s action will help facilitate publicparticipation, outreach, andcommunication in areas whereenvironmental justice issues are present.

G. Executive Order 12875: EnhancingIntergovernmental Partnerships

Under Executive Order 12875, EPAmay not issue a regulation that is notrequired by statute and that creates amandate upon a State, local or tribalgovernment, unless the Federalgovernment provides the fundsnecessary to pay the direct compliancecosts incurred by those governments, orwe will consult with those governments.If EPA complies by consulting,Executive Order 12875 requires us toprovide to OMB a description of theextent of our prior consultation withrepresentatives of affected State, localand tribal governments, the nature oftheir concerns, copies of any writtencommunications from the governments,and a statement supporting the need toissue the regulation. In addition,Executive Order 12875 requires us todevelop an effective process permittingelected officials and otherrepresentatives of State, local and tribalgovernments ‘‘to provide meaningfuland timely input in the development ofregulatory proposals containingsignificant unfunded mandates.’’

Today’s rule implementsrequirements set forth in section 319 ofthe Act and thus is required by statute.

This rule does not establish a whollynew requirement but rather modifiesexisting reporting requirements whichState and local governments have beenimplementing for approximately 20years. While these changes aresignificant in many ways, they are notexpected to result in a significantincrease in reporting burdens.Nonetheless, EPA engaged in extensiveconsultation with State and localgovernments in the development of theproposed and final rules, and thisconsultation is discussed anddocumented elsewhere in today’s noticeand in the notice of proposedrulemaking.

H. Executive Order 13084: Consultationand Coordination With Indian TribalGovernments

Under Executive Order 13084, EPAmay not issue a regulation that is notrequired by statute, that significantly oruniquely affects the communities ofIndian tribal governments, and thatimposes substantial direct compliancecosts on those communities, unless theFederal government provides the fundsnecessary to pay the direct compliancecosts incurred by the tribalgovernments, or EPA will consult withthose governments. If EPA complies byconsulting, Executive Order 13084requires us to provide to OMB, in aseparately identified section of thepreamble to the rule, a description ofthe extent of our prior consultation withrepresentatives of affected tribalgovernments, a summary of the natureof their concerns, and a statementsupporting the need to issue theregulation. In addition, Executive Order13084 requires us to develop aneffective process permitting electedofficials and other representatives ofIndian tribal governments ‘‘to providemeaningful and timely input in thedevelopment of regulatory policies onmatters that significantly or uniquelyaffect their communities.’’

Today’s rule implementsrequirements specifically set forth bythe Congress in section 319 of the Actwithout the exercise of any discretionby us. Accordingly, the requirements ofsection 3(b) of Executive Order 13084do not apply to this rule.

This rule governs the reporting of airquality by States for MSAs and, in somecases, areas that are significantlyaffected by transport of pollutants fromMSAs. In extensive public andintergovermental coordination effortsduring the development of the proposal,EPA received no information whichwould suggest that the rule will imposenew requirements on Indian tribalgovernments nor will it significantly or

uniquely affect communities of Indiantribal governments. To the extent thatair pollution from upwind MSAssignificantly affects any lands withinIndian country, this impact is not aresult of, or affected by, today’s rule andwould be addressed under existingrequirements governing theimplementation of air quality standards.

I. National Technology Transfer andAdvancement Act

Section 12(d) of the NationalTechnology Transfer and AdvancementAct of 1995 (NTTAA), Public Law 104–113, section 12(d) (15 U.S.C. 272 note)directs EPA to use voluntary consensusstandards in its regulatory activitiesunless to do so would be inconsistentwith applicable law or otherwiseimpractical. Voluntary consensusstandards are technical standards (e.g.,materials specifications, test methods,sampling procedures, and businesspractices) that are developed or adoptedby voluntary consensus standardsbodies. The NTTAA directs EPA toprovide Congress, through OMB,explanations when the Agency decidesnot to use available and applicablevoluntary consensus standards. Thisaction does not involve technicalstandards. Therefore, EPA did notconsider the use of any voluntaryconsensus standards.

J. Congressional Review Act

The Congressional Review Act, 5U.S.C. 801 et seq., as added by the SmallBusiness Regulatory EnforcementFairness Act of 1996, generally providesthat before a rule may take effect, theagency promulgating the rule mustsubmit a rule report, which includes acopy of the rule, to each House of theCongress and to the Comptroller Generalof the United States. The EPA willsubmit a report containing this rule andother required information to the U.S.Senate, the U.S. House ofRepresentatives, and the ComptrollerGeneral of the United States prior to thepublication of the rule in the FederalRegister. This rule is not a ‘‘major rule’’as defined by 5 U.S.C. 804(2).

IV. References

CEQ, (1976) A Recommended Air PollutionIndex, report prepared by the FederalInteragency Task Force on Air QualityIndicators, Council on EnvironmentalQuality, Environmental ProtectionAgency, and Department of Commerce.

EPA, (1994) Measuring Air Quality: ThePollutant Standards Index, U.S.Environmental Protection Agency, Officeof Air Quality Planning and Standards(MD–10), Research Triangle Park, NC,27711, EPA 451/K–94–001.



EPA, (1999a) Guideline for Public Reportingof Daily Air Quality—Air Quality Index(AQI), U.S. Environmental ProtectionAgency, Office of Air Quality Planningand Standards, Research Triangle Park,NC, 27711, EPA–454/R–99–010.

EPA, (1999b) Guideline for Developing anOzone Forecasting Program, U.S.Environmental Protection Agency, Officeof Air Quality Planning and Standards,Research Triangle Park, NC, 27711, EPA–454/R–99–009.

EPA, (1999c) The Air Quality Index, U.S.Environmental Protection Agency, Officeof Air Quality Planning and Standards,Research Triangle Park, NC, 27711, inpreparation.

Science Applications InternationalCorporation, (1998) Report of EightFocus Groups on the Ozone Map, thePollutant Standards Sub-index forOzone, and the Ozone Health EffectsBooklet, Science ApplicationsInternational Corporation, McLean, VA.

U.S. Department of Commerce, (1998)Statistical Abstract of the United States,U.S. Bureau of the Census.

Whitfield, R.G.; Biller, W.F.; Jusko, M.J.;Keisler, JM (1996) A probabilisticassessment of health risks associatedwith short-term exposure to troposphericozone. Report prepared for U.S. EPA,OAQPS. Argonne National Laboratory;Argonne, IL.

Wolff, G.T., (1995) Letter from Chairman ofthe Clean Air Scientific AdvisoryCommittee to the EPA Administrator,dated November 30, 1995. EPA–SAB–CASAC–LTR–96–002.

List of Subjects in 40 CFR Part 58Environmental protection, Air

pollution control, Intergovernmentalrelations, Reporting and recordkeepingrequirements.

Dated: July 23, 1999.Carol M. Browner,Administrator.

Accordingly, 40 CFR part 58 isamended as follows:

PART 58—AMBIENT AIR QUALITYSURVEILLANCE

1. The authority citation for part 58continues to read as follows:

Authority: 42 U.S.C. 7410, 7601(a), 7613,and 7619.

2. Section 58.50 is revised to read asfollows:

§ 58.50 Index reporting.

(a) The State shall report to thegeneral public through prominent noticean air quality index in accordance withthe requirements of appendix G to thispart.

(b) Reporting is required by allMetropolitan Statistical Areas with apopulation exceeding 350,000.

(c) The population of a MetropolitanStatistical Area for purposes of indexreporting is the most recent decennialU.S. census population.

3. Appendix G to part 58 is revised toread as follows:

Appendix G to Part 58—Uniform AirQuality Index (AQI) and DailyReporting

General Requirements

1. What is the AQI?2. Why report the AQI?3. Must I report the AQI?4. What goes into my AQI report?5. Is my AQI report for my MSA only?6. How do I get my AQI report to the

public?7. How often must I report the AQI?8. May I make exceptions to these reporting

requirements?

Calculation

9. How does the AQI relate to air pollutionlevels?

10. Where do I get the pollutantconcentrations to calculate the AQI?

11. Do I have to forecast the AQI?12. How do I calculate the AQI?

Background and Reference Materials

13. What additional information should Iknow?

General Requirements

1. What Is the AQI?

The AQI is a tool that simplifies reportingair quality to the general public. The AQIincorporates into a single indexconcentrations of 5 criteria pollutants: ozone(O3), particulate matter (PM), carbonmonoxide (CO), sulfur dioxide (SO2), andnitrogen dioxide (NO2). The scale of theindex is divided into general categories thatare associated with health messages.

2. Why Report the AQI?

The AQI offers various advantages:a. It is simple to create and understand.b. It conveys the health implications of air

quality.c. It promotes uniform use throughout the

country.

3. Must I Report the AQI?

You must report the AQI daily if yours isa metropolitan statistical area (MSA) with apopulation over 350,000.

4. What Goes Into My AQI Report?

i. Your AQI report must contain thefollowing:

a. The reporting area(s) (the MSA orsubdivision of the MSA).

b. The reporting period (the day for whichthe AQI is reported).

c. The critical pollutant (the pollutant withthe highest index value).

d. The AQI (the highest index value).e. The category descriptor and index value

associated with the AQI and, if you chooseto report in a color format, the associatedcolor. Use only the following descriptors andcolors for the six AQI categories:

TABLE 1.—AQI CATEGORIES

For this AQI Use thisdescriptor

And thiscolor 1

0 to 50 .............. ‘‘Good’’ ............. Green.

51 to 100 .......... ‘‘Moderate’’ ....... Yellow.

101 to 150 ........ ‘‘Unhealthy forSensitiveGroups’’.

Orange.

151 to 200 ........ ‘‘Unhealthy’’ ...... Red.

201 to 300 ........ ‘‘VeryUnhealthy’’.

Purple.

301 and above ‘‘Hazardous’’ .... Ma-roon.1

1 Specific colors can be found in the mostrecent reporting guidance (Guideline for PublicReporting of Daily Air Quality—Air QualityIndex (AQI)).

f. The pollutant specific sensitive groupsfor any reported index value greater than 100.Use the following sensitive groups for eachpollutant:

When this pollutant has an index value above 100 * * * Report these sensitive groups * * *

Ozone ....................................................................................................... Children and people with asthma are the groups most at risk.

PM2.5 ......................................................................................................... People with respiratory or heart disease, the elderly and children arethe groups most at risk.

PM10 .......................................................................................................... People with respiratory disease are the group most at risk.

CO ............................................................................................................. People with heart disease are the group most at risk.

SO2 ........................................................................................................... People with asthma are the group most at risk.



When this pollutant has an index value above 100 * * * Report these sensitive groups * * *

NO2 ........................................................................................................... Children and people with respiratory disease are the groups most atrisk.

ii. When appropriate, your AQI report mayalso contain the following:

a. Appropriate health and cautionarystatements.

b. The name and index value for otherpollutants, particularly those with an indexvalue greater than 100.

c. The index values for sub-areas of yourMSA.

d. Causes for unusual AQI values.e. Actual pollutant concentrations.

5. Is My AQI Report for My MSA Only?

Generally, your AQI report applies to yourMSA only. However, if a significant airquality problem exists (AQI greater than 100)in areas significantly impacted by your MSAbut not in it (for example, O3 concentrationsare often highest downwind and outside anurban area), you should identify these areasand report the AQI for these areas as well.

6. How Do I Get My AQI Report to the Public?

You must furnish the daily report to theappropriate news media (radio, television,and newspapers). You must make the dailyreport publicly available at one or moreplaces of public access, or by any othermeans, including a recorded phone message,a public Internet site, or facsimiletransmission. When the AQI value is greaterthan 100, it is particularly critical that thereporting to the various news media be asextensive as possible. At a minimum, itshould include notification to the media withthe largest market coverages for the area inquestion.

7. How Often Must I Report the AQI?

You must report the AQI at least 5 days perweek. Exceptions to this requirement are insection 8 of this appendix.

8. May I Make Exceptions to These ReportingRequirements?

i. If the index value for a particularpollutant remains below 50 for a season oryear, then you may exclude the pollutantfrom your calculation of the AQI in section12.

ii. If all index values remain below 50 fora year, then you may report the AQI at yourdiscretion. In subsequent years, if pollutantlevels rise to where the AQI would be above50, then the AQI must be reported asrequired in sections 3, 4, 6, and 7 of thisappendix.

Calculation

9. How Does the AQI Relate to Air PollutionLevels?

For each pollutant, the AQI transformsambient concentrations to a scale from 0 to500. The AQI is keyed as appropriate to thenational ambient air quality standards(NAAQS) for each pollutant. In most cases,the index value of 100 is associated with thenumerical level of the short-term standard(i.e., averaging time of 24-hours or less) foreach pollutant. Different approaches aretaken for NO2, for which no short-termstandard has been established, and for PM2.5,for which the annual standard is theprincipal vehicle for protecting against short-term concentrations. The index value of 50is associated with the numerical level of theannual standard for a pollutant, if there isone, at one-half the level of the short-termstandard for the pollutant, or at the level atwhich it is appropriate to begin to provideguidance on cautionary language. Highercategories of the index are based onincreasingly serious health effects andincreasing proportions of the population thatare likely to be affected. The index is relatedto other air pollution concentrations throughlinear interpolation based on these levels.The AQI is equal to the highest of thenumbers corresponding to each pollutant.For the purposes of reporting the AQI, thesub-indexes for PM10 and PM2.5 are to beconsidered separately. The pollutantresponsible for the highest index value (thereported AQI) is called the ‘‘critical’’pollutant.

10. Where Do I Get the PollutantConcentrations To Calculate the AQI?

You must use concentration data frompopulation-oriented State/Local Air

Monitoring Station (SLAMS) or parts of theSLAMS required under 40 CFR 58.20 foreach pollutant except PM. For PM, you needonly calculate and report the AQI on days forwhich you have measured air quality data(e.g., particulate monitors often report valuesonly every sixth day). You may useparticulate measurements from monitors thatare not reference or equivalent methods (forexample, continuous PM10 or PM2.5 monitors)if you can relate these measurements bystatistical linear regression to reference orequivalent method measurements.

11. Do I Have to Forecast the AQI?

You should forecast the AQI to providetimely air quality information to the public,but this is not required. If you choose toforecast the AQI, then you may consider bothlong-term and short-term forecasts. You canforecast the AQI at least 24-hours in advanceusing the most accurate and reasonableprocedures considering meteorology,topography, availability of data, andforecasting expertise. The document‘‘Guideline for Developing an OzoneForecasting Program’’ (the ForecastingGuidance) will help you start a forecastingprogram. You can also issue short-termforecasts by predicting 8-hour ozone valuesfrom 1-hour ozone values using methodssuggested in the Reporting Guidance,‘‘Guideline for Public Reporting of Daily AirQuality.’’

12. How Do I Calculate the AQI?

i. The AQI is the highest value calculatedfor each pollutant as follows:

a. Identify the highest concentration amongall of the monitors within each reporting areaand truncate the pollutant concentration toone more than the significant digits used toexpress the level of the NAAQS for thatpollutant. This is equivalent to the roundingconventions used in the NAAQS.

b. Using Table 2, find the two breakpointsthat contain the concentration.

c. Using Equation 1, calculate the index.d. Round the index to the nearest integer.

TABLE 2.—BREAKPOINTS FOR THE AQI

These breakpoints Equal these AQIs* * * CategoryO3 (ppm)

8-hourO3 (ppm)1-hour 1

PM2.5(µg/m3)

PM10(µg/m3) CO (ppm) SO2 (ppm) NO2 (ppm) AQI

0.000–0.064 ............. ...................... 0.0–15.4 0–54 0.0–4.4 0.000–0.034 (2 ) 0–50 Good.0.065–0.084 ............. ...................... 15.5–40.4 55–154 4.5–9.4 0.035–0.144 (2 ) 51–100 Moderate.0.085–0.104 ............. 0.125–0.164 40.5–65.4 155–254 9.5–12.4 0.145–0.224 (2 ) 101–150 Unhealthy for sen-

sitive groups.0.105–0.124 ............. 0.165–0.204 4 65.5–150.4 255–354 12.5–15.4 0.225–0.304 (2 ) 151–200 Unhealthy.0.125–0.374 ............. 0.205–0.404 4 150.5–250.4 355–424 15.5–30.4 0.305–0.604 0.65–1.24 201–300 Very unhealthy.(3 ) ............................ 0.405–0.504 4 250.5–350.4 425–504 30.5–40.4 0.605–0.804 1.25–1.64 301–400(3 ) ............................ 0.505–0.604 4 350.5–500.4 505–604 40.5–50.4 0.805–1.004 1.65–2.04 401–500 Hazardous.

1 Areas are generally required to report the AQI based on 8-hour ozone values. However, there are a small number of areas where an AQIbased on 1-hour ozone values would be more precautionary. In these cases, in addition to calculating the 8-hour ozone index value, the 1-hourozone index value may be calculated, and the maximum of the two values reported.



2 NO2 has no short-term NAAQS and can generate an AQI only above an AQI value of 200.3 8-hour O3 values do not define higher AQI values (≥ 301). AQI values of 301 or higher are calculated with 1-hour O3 concentrations.4 If a different SHL for PM2.5 is promulgated, these numbers will change accordingly.

ii. If the concentration is equal to abreakpoint, then the index is equal to thecorresponding index value in Table 2.However, Equation 1 can still be used. Theresults will be equal. If the concentration is

between two breakpoints, then calculate theindex of that pollutant with Equation 1. Youmust also note that in some areas, the AQIbased on 1-hour O3 will be moreprecautionary than using 8-hour values (see

footnote 1 to Table 2). In these cases, youmay use 1-hour values as well as 8-hourvalues to calculate index values and then usethe maximum index value as the AQI for O3.

II I

BP BPC BP Ip

Hi Lo

HI Lop Lo Lo=

−−

−( ) + (Equation 1)

Where:Ip = the index value for pollutantp

Cp = the truncated concentration ofpollutantp

BPHi = the breakpoint that is greater than orequal to Cp

BPLo = the breakpoint that is less than orequal to Cp

IHi = the AQI value corresponding to BPHi

Ilo = the AQI value corresponding to BPLo.iii. If the concentration is larger than the

highest breakpoint in Table 2 then you mayuse the last two breakpoints in Table 2 whenyou apply Equation 1.

Example

iv. Using Table 2 and Equation 1, calculatethe index value for each of the pollutantsmeasured and select the one that producesthe highest index value for the AQI. Forexample, if you observe a PM10 value of 210µg/m3, a 1-hour O3 value of 0.156 ppm, andan 8-hour O3 value of 0.130 ppm, then dothis:

a. Find the breakpoints for PM10 at 210 µg/m3 as 155 µg/m3 and 254 µg/m3,corresponding to index values 101 and 150;

b. Find the breakpoints for 1-hour O3 at0.156 ppm as 0.125 ppm and 0.164 ppm,corresponding to index values 101 and 150;

c. Find the breakpoints for 8-hour O3 at0.130 ppm as 0.125 ppm and 0.374 ppm,corresponding to index values 201 and 300;

d. Apply Equation 1 for 210 µg/m3, PM10:

150 101

254 155210 155 101 128

−−

−( ) + = .

e. Apply Equation 1 for 0.156 ppm, 1-hourO3:

150 101

0 164 0 1250 156 0 125 101 140

−−

−( ) + =. .

. .

f. Apply Equation 1 for 0.130 ppm, 8-hourO3:

300 201

0 374 0 1250 130 0 125 201 203

−−

−( ) + =. .

. .

g. Find the maximum, 203. This is the AQI.The minimal AQI report would read:

v. Today, the AQI for my city is 203 whichis very unhealthy, due to ozone. Childrenand people with asthma are the groups mostat risk.

Background and Reference Materials

13. What Additional Information Should IKnow?

The EPA has developed a computerprogram to calculate the AQI for you. Theprogram works with Windows 95, it promptsfor inputs, and it displays all the pertinentinformation for the AQI (the index value,color, category, sensitive group, healtheffects, and cautionary language). The EPAhas also prepared a brochure on the AQI thatexplains the index in detail (The Air QualityIndex), Reporting Guidance (Guideline forPublic Reporting of Daily Air Quality) thatprovides associated health effects andcautionary statements, and ForecastingGuidance (Guideline for Developing anOzone Forecasting Program) that explains thesteps necessary to start an air pollutionforecasting program. You can download theprogram and the guidance documents atwww.epa.gov/airnow.

[FR Doc. 99–19433 Filed 8–3–99; 8:45 am]BILLING CODE 6560–50–P


Air qual. epa

Education

Transcript of Air qual. epa