1987 rules and regulations - us epa€¦ · 24634 federal register / vol. 52, no. 126 / wednesday,...

24634 Federal Register / Vol. 52, No. 126 / Wednesday, July 1, 1987 / Rules and Regulations

ENVIRONMENTAL PROTECTIONAGENCY

40 CFR Part 50

[AD-FRL 3141-9(a)]

Revisions to the National Ambient AirQuality Standards for ParticulateMatter

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

SUMMARY: In 1971, EPA promulgatedprimary and secondary nationalambient air quality standards forparticulate matter, measured as "totalsuspended particulate matter" or "TSP."The primary standards were set at 260jig/m s , 24-hour average not to beexceeded more than once per year, and75 /g/m s , annual geometric mean. Thesecondary standard, also measured asTSP, was set at 150 #Lg/m s , 24-houraverage not to be exceeded more thanonce per year. In accordance withsections 108 and 109 of the Clean AirAct, EPA has reviewed and revised thehealth and welfare criteria upon whichthese primary and secondary particulatematter standards were based.

On March 20, 1984 (49 FR 10408), EPAproposed changes in the standardsbased on its review and revision of thecriteria. Today's notice announcesEPA's final decisions regarding thesechanges. The final decisions include: (1)replacing TSP as the indicator for'particulate matter for the ambientstandards with a new indicator thatincludes only those particles with anaerodynamic diameter less than or equal'to a nominal 10;micrometers (PM1a), (2).replacing the 24-hour primary. TSPstandard with. a 24-hour PMo standardof 150 fg/m with no more than oneexpected exceedance per year; (3)replacing the annual primaryiTSPstandard with a, PM1o. standard of 50. Ig/M3 , expected annual arithmetic mean;and (4), replacing the, secondary TSPstandard with, 24-hour' and annual PMostandards that are identical in allrespects. to the primary, standards.

Today's notice. also, announces a newFederal Reference Method formeasurement of PMo in the ambient air.The method is contained in a newAppendix J to Part 50. This notice alsoannounces a new Appendix K to Part 50,which provides rules for applying thestatistical form of the revised standards.In addition, certain clarifying changes toAppendix B and Appendix G are set out.

Related notices published elsewherein today's Federal Register set out finalregulations concerning Ambient AirMonitoring Reference and Equivalent

Methods (40 CFR Part 53), Ambient AirQuality Surveillance (40 CFR Part 58),Regulations for Implementing RevisedParticulate Matter Standards (40 CFRPart 51) with associated guidelines,Approval and Promulgation ofImplementation Plans (40 CFRPart 52),and Prevention of SignificantDeterioration (Parts 51 and 52).EFFECTIVE DATE: This action is effectiveJuly 31, 1987.ADDRESSES: A docket (No. A-82-37)containing information related to EPA'sreview and revision of the particulatematter standards is available for publicinspection between 8:00 a.m. and 3:00p.m. on weekdays at EPA's CentralDocket Section, South ConferenceCenter, Room 4, 401 M St., SW.,Washington, DC. A reasonable fee maybe charged for copying. The informationin the docket constitutes the completebasis for the decisions announced' in thisnotice. For the availability of related*information see SUPPLEMENTARYINFORMATION.FOR FURTHER INFORMATION CONTACT:Mr. John Haines, Strategies and AirStandards Division (MD-12), U.S.Environmental Protection Agency,Research Triangle Park, N.C. 27711,telephone. (919) 541-5531 (FTS 629-5531].SUPPLEMENTARY INFORMATION:

Availability of Related InformationThe revised criteria document,, Air

Quality Criteria for Particulate Matterand Sulfur Oxides (three volumes, EPA-600/8--82-029af-cf, December, 1982;Volume I NTIS #PB-84-120401, $24.95paper copy and $6.50 microfiche;Volume 11 NTIS #PB-84-120419,.$48.95paper copy and $6.50 microfiche;Volume III NTIS.#PB-84-120427, $48.95paper copy and $13.50 microfiche, theSecond Addendum to Air QualityCriteria for Particulate Matter andSulfur Oxides (1982]: Assessment ofNewly Available Health Effects'Information, (EPA/600/8-86-020-FNTIS #PB-87-176574, $24.95 paper copy,and $6.50 microfiche), the 1982 staffpaper,. Review of the National AmbientAir-Quality Standards for ParticulateMatter: Assessment of Scientific andTechnical Information-OAQPS StaffPaper (EPA-450/5-82-01, January, 1982;NTIS #PB-82-177874, $24.95 paper copyand $6.50 microfiche), and the staffpaper addendum, Review of theNational Ambient Air Quality Standardsfor Particulate Matter: UpdatedAssessment of Scientific and TechnicalInformation (EPA-450/ 5-86-012,,December 1986; NTIS #PB-87-176871,$18.95 paper copy and $6.50 microfiche)are available from: U.S. Department ofCommerce, National Technical

Information Service, 5285 Port RoyalRoad, Springfield, Virginia.22161 (add$3.00 handling charge per order). Alimited number of copies of otherdocuments generated in connection withthis standard review, such as the controltechniques document, can be obtainedfrom: U.S. Environmental ProtectionAgency Library (MD-35), ResearchTriangle Park, N.C. 27711, telephone(919) 541-2777 (FTS 629-2777).

Table of ContentsI. Background

A. Legislative Requirements Affecting ThisRule1. The Standards2. Related Control Requirements

B. Particulate Matter and Original Stand-ards for TSP

C. Development of Revised Air QualityCriteria for Particulate Matter

D. Review of the Standards: Developmentof Staff Paper

E. Proposed Revisions to the StandardsF. Supplemental Criteria Revisions and

Standards Review Following ProposalIIU. Summary of Public Comments

A. Comments on 1984 ProposalB. Comments on Subsequent Notice

1. Rationale for the Primary StandardsA. Pollutant IndicatorB. Averaging Time and Form of the Stand-

ards1. 24-hour Standard2. Annual Standard

C. Level of the Standards1. 24-hour Standard2. Annual Standard

IV., Rationale for the Secondary StandardsA. Soiling and NuisanceB. Other Welfare Effects

V. Federal Reference MethodA. Specific Changes to Appendix JB. Designation of Reference Methods for

PM10C. Technical Change to Appendix G

VI. Summary of Salient Public Commentsand Agency ResponsesA. Health Effects Criteria and Selection of

the Primary Standards1. Indicator for the Primary Standards2. Interpretation of Community Epidemi-

ological Studies3. Margin of Safety

B. Secondary Standards1. Soiling and Nuisance2. Visibility

C Averaging Time and Form of the Stand-ards1. Expected Exceedances for the 24-hour

Standard2. Expected Arithmetic Mean for the

Annual StandardVII. Regulatory and Environmental Impacts

A. Regulatory Impact AnalysisB. Impact on Small Entities

VIII. Other Reviews

ReferencesAddendum I-CASAC Review and Closure of

the. 198ZCriteria Document for ParticulateMatter/Sulfur Oxides and the 1986 SecondAddendum to the Criteria Document

Federal Register / Vol. 52, No. 126 / Wednesday, July 1, 1987 / Rules and Regulations

Addendum 1l-CASAC Review and Closureof the 1982 OAQPS Staff Paper forParticulate Matter and the 1986 Addendumto the Staff Paper

Addendum Ill-Executive Summary of the1986 Addendum to the Staff Paper

Part 50-National Primary and SecondaryAmbient Air Quality Standards

Appendix J-Reference Method for theDetermination of Particulate Matter asPM1 o in the Atmosphere

Appendix K-Interpretation of the NationalAmbient Air Quality Standards forParticulate Matter

I. Background

A. Legislative Requirements AffectingThis Rule

1. The Standards

Two sections of the Clean Air Actgovern the establishment and revision ofnational ambient air quality standards(NAAQS). Section 108 (42 U.S.C. 7408)directs the Administrator to identifypollutants which may reasonably beanticipated to endanger public health orwelfare and to issue air quality criteriafor them. These air quality criteria are toreflect the latest scientific informationuseful in indicating the kind and extentof all identifiable effects on publichealth or welfare that may be expectedfrom the presence of a pollutant in theambient air.

Section 109 (42 U.S.C. 7409) directs theAdministrator to propose andpromulgate "primary" and "secondary"NAAQS for pollutants identified undersection 108. Section 109(b)(1) defines aprimary standard as one the attainmentand maintenance of which, in thejudgment of the Administrator, based onthe criteria and allowing for anadequate margin of safety, is requisite toprotect the public health. A secondarystandard, as defined in section 109(b)(2),must specify a level of air quality theattainment and maintenance of which.in the judgment of the Administrator,based on the criteria, is requisite toprotect the public welfare from anyknown or anticipated adverse effectsassociated with the presence of thepollutant in the ambient air. Welfareeffects are defined in section 302(h) (42U.S.C. 7602(h)) to include effects onsoils, water, crops, vegetation, man-made materials, animals, wildlife,weather, visibility, climate, damage toand deterioration of property, hazards totransportation, and effects on economicvalues and on personal comfort andwell-being.

The U.S. Court of Appeals for the D.C.Circuit has held that the requirement foran adequate margin of safety forprimary standards was intended toaddress uncertainties associated withinconclusive scientific and technical

information available at the time ofstandard setting. It was also intended toprovide a reasonable degree ofprotection against hazards that researchhas not yet identified. Lead IndustriesAssociation v. EPA, 647 F.2d 1130, 1154(D.C. Cir. 1980), cert. denied, 101 S. Ct.621 (1980): American Petroleum Institutev. Castle, 665 F.2d 1176, 1177 (D.C. Cir.1981), cert. denied, 102 S. Ct. 1737 (1982).Both kinds of uncertainties arecomponents of the risk associated withpollution at levels below those at whichhuman health effects can be said tooccur with reasonable scientificcertainty. Thus, by selecting primarystandards that provide an adequatemargin of safety, the Administrator isseeking not only to prevent pollutionlevels that have been demonstrated tobe harmful, but also to prevent lowerpollutant levels that he finds pose anunacceptable risk of harm, even if thatrisk is not precisely identified as tonature or degree.

In selecting a margin of safety, EPAhas considered such factors as thenature and severity of the health effectsinvolved, the size of the sensitivepopulation(s) at risk, and the kind anddegree of the uncertainties that must beaddressed. Given that the "margin ofsafety" requirement by definition onlycomes into play where no conclusiveshowing of harm exists, such factors,which involve unknown or only partiallyquantified risks, have their inherentlimits as guides to action. The selectionof any particular approach to providingan adequate margin of safety is a policychoice left specifically to theAdministrator's judgment. LeadIndustries Association v. EPA, supra,647 F.2d at 1161-62.

Section 109(d) of the Act (42 U.S.C.7409(d)) requires periodic review and, ifappropriate, revision of existing criteriaand standards. The'process by whichEPA has reviewed the original criteriaand standards for particulate matterunder section 109(d) is described inSections I.C. and I.D. of this notice.

2. Related Control RequirementsStates are primarily responsible for

ensuring attainment and maintenance ofambient air quality standards once EPAhas established them. Under section 110of the Act (42 U.S.C. 7410), States are tosubmit, for EPA approval, Stateimplementation plans (SIPs) thatprovide for the attainment andmaintenance of such standards throughcontrol programs directed to sources ofthe pollutants involved. Other Federalprograms provide for nationwidereductions in emissions of these andother air pollutants through the FederalMotor Vehicle Control Program under

Title I1 of the Act (42 U.S.C. 7501 to7534), which involves controls forautomobile, truck, bus, motorcycle, andaircraft emissions, and through thedevelopment of New SourcePerformance Standards under section111 (42 U.S.C. 7411) and NationalEmission Standards for Hazardous AirPollutants under section 112 (42 U.S.C.7412).

B. Particulate Matter and OriginalStandards for TSP

"Particulate matter" is the genericterm for a broad class of chemically andphysically diverse substances that existas discrete particles (liquid droplets orsolids) over a wide range of sizes.Particles originate from a variety ofstationary and mobile sources. Theymay be emitted directly or formed in the-atmosphere by transformations ofgaseous emissions such as sulfur oxides,nitrogen oxides, and volatile organicsubstances. The chemical and physicalproperties of particulate matter varygreatly with time, region, meteorologyand source category, thus complicatingthe assessment of health and welfareeffects. The characteristics, origins,concentrations, and potential effects ofparticulate matter are discussed in moredetail in the staff. paper (SP) (EPA,1982a), in the revised criteria document(CD) (EPA, 1982b), in the criteriadocument addendum (CDA) (EPA,1986a) and in the staff paper addendum(SPA) (EPA, 1986b). The executivesummary of the staff paper a'ddendum isreprinted in Addendum Ill to this notice.

OnApril 30, 1971 (36 FR 8186), EPApromulgated the original primary andsecondary NAAQS for particulatematter under section 109 of the CleanAir Act. The reference method formeasuring attainment of these standardsis the "high-volume" sampler (40 CFRPart 50, Appendix B], which collectsparticulate matter up to a nominal sizeof 25 to 45 micrometers (/±m) (so-called"total suspended particulate," or "TSP").Thus, TSP is the current indicator for theparticulate matter standards. Theexisting primary standards forparticulate matter (measured as TSP)are 260 j±g/m s , 24-hour average not to beexceeded more than once per year, and75 g/m 3, annual geometric mean. Thesecondary standard (measured as TSP)is 150 pg/m 3, 24-hour average not to beexceeded more than once per year. Thescientific and technical bases for thesestandards are contained in the originalcriteria document, Air Quality Criteriafor Particulate Matter (DHEW, 1969).

24635


C. Development of Revised Air QualityCriteria for Particulate Matter

In 1976, as a result of internal Agencyreview and the recommendations of acommittee of EPA's Science AdvisoryBoard, EPA decided to revise theexisting criteria document forparticulate matter. Because of competingpriorities regarding revision of othercriteria documents, and because of theneed to complete additional research onparticulate matter, the process wasscheduled to commence in 1979. Withthe endorsement of the Clean AirScientific Advisory Committee (CASAC)of EPA's Science Advisory Board, EPAdecided to review and revise the criteriadocument for particulate matterconcurrently with that for sulfur oxidesand to produce a combined particulatematter/sulfur oxides (PM/SO,, criteriadocument. On October 2. 1979 (44 FR56731), EPA announced that it was in theprocess of revising the criteria documentand reviewing the existing air qualitystandards for possible revisions.

In developing the revised criteriadocument, EPA has provided a numberof opportunities for review and commentby organizations and individuals outsidethe Agency. Three drafts of the revisedparticulate matter/sulfur oxides criteriadocument, prepared by EPA'sEnvironmental Criteria and AssessmentOffice (ECAO), were made available forexternal review on April 11, 1980 (45 FR24913), January 29, 1981 (46 FR 9746),and October 28, 1981 (46 FR 53210). EPA.received and considered numerous andoften extensive comments on each ofthese drafts. CASAC held three publicmeetings to review successive drafts ofthe document on August 20-22, 1980 (45FR 5164, August 4, 1980), July 7-9, 1981(46 FR 31746, June 17, 1981), andNovember 16-18, 1981 (46 FR 53210,October 28, 1981). These reetings wereopen to the public and were attended bymany individuals and representatives oforganizations who provided criticalreviews and new information forconsideration. In accordance withCASAC recommendations made afterthe first review meeting, five additionalpublic meetings were held at whichEPA, its consulting authors andreviewers, and other scientifically andtechnically qualified experts selected byEPA discussed the various chapters ofthe draft document and suggested waysof resolving outstanding issues (45 FR74047, November 7. 1980; 45 FR 78224,November 25, 1980; 45 FR 76790,November 20, 1980; 45 FR 80350,December 4, 1980:46 FR 1775, January 7,1981).

The comments received on thesuccessive drafts of the revised criteria

document were considered in the finaldocument, issued simultaneously withthe proposal of revisions to thestandards. A summary of EPA'sresponses to the comments on the threeexternal review drafts of the documentsis in the public docket (Docket No. A-82-37). Transcripts of the three CASACmeetings are also in the docket. Inaccordance with its establishedprocedures, CASAC prepared a"closure" memorandum to theAdministrator indicating its satisfactionwith the final draft (December, 1981) ofthe criteria document and outlining keyissues and recommendations. Theclosure memorandum, dated January 29,1982, stated that the EPA office thatprepared this document was"responsive to Committee advice aswell as to comments provided by thegeneral public . . ." The closurememorandum further stated that thecriteria document "fulfills therequirements set forth in section 108 ofthe Clean Air Act, which requires thatthe criteria document 'shall accuratelyreflect the latest scientific knowledgeuseful in indicating the kind and extentof all identifiable effects on publichealth or welfare' from sulfur oxidesand particulates in the ambient air." TheCASAC closure memorandum on thecriteria document is reprinted in itsentirety in Addendum I to this notice.Following closure, minor technical andeditorial refinements were made to thecriteria document for printing (EPA,1982b).

A number of scientific and technicalissues were raised during the publicreview process. With respect to theparticulate matter portions of thecriteria document, the major Issuesincluded the relationship among variousmeasures of particulate matter airquality, the implications of particledeposition and other studies forselecting a particulate matter indicator,and the development and application ofcriteria for deciding whichepidemiological studies are mostappropriate for use in revising airquality standards. A summary of theseand other major scientific issues, as wellas CASAC's conclusions, is included inthe closure memorandum on the criteriadocument (Addendum I).D. Review of the Standards:Development of Staff Paper

In the evolving process of revising thenational ambient air quality standards,EPA has found it useful to prepare adocument that helps bridge the gapbetween the scientific review of healthand welfare effects contained in thecriteria document and the judgmentsrequired of the Administrator in setting

ambient standards. This document,known as the staff paper, has becomean important element in the standardsreview process, providing anopportunity for public comment onproposed staff recommendations beforethey are presented to the Administrator.

In the spring of 1981, EPA's Office ofAir Quality Planning and Standards(OAQPS) prepared the first draft of thestaff paper, Review of the NationalAmbient Air Quality Standards forParticulate Matter: Assessment ofScientific and Technical Information.This draft staff paper, based on the thenexistent draft of the revised criteriadocument, evaluated and interpreted theavailable scientific and technicalinformation most relevant to the reviewof the air quality standards forparticulate matter and presented staffrecommendations on alternativeapproaches to revising the standards.This and a second draft of the paperwere reviewed at two CASAC meetingson July 7-9, 1981 (46 FR 31746, June 17,1981), and November 16-18, 1981 (46 FR53210, October 28, 1981]. Numerouswritten and oral comments werereceived on the drafts from CASAC,representatives of organizations,individual scientists, and otherinterested members of the public. Asummary of major revisions made inresponse to comments on the first draftis contained in an October 31, 1981 letterto CASAC (Padgett, 1981). Following thesecond CASAC meeting, the staff madefurther revisions in response tocomments and prepared an executivesummary that was reviewed by CASACmembers before preparation of theclosure memorandum on the staff paper.In January, 1982, EPA released the finalOAQPS staff paper (EPA, 1982a), whichreflects the various suggestions made byCASAC and members of the public. TheJanuary 29, 1982, CASAC closurememorandum states that the staff paper"has been modified in accordance withrecommendations made by CASAC," isconsistent with the criteria document,and provides the Administrator "withthe kind and amount of technicalguidance that will be needed to makeappropriate revisions to the standard."This closure memorandum is reprintedin Addendum II to this notice.

A number of major issues were raisedduring the public review process. Themore important issues are outlinedbelow.

1. Substantial discussion concernedthe maximum size of particles (orparticle size fraction) to be used inmeasuring particulate matter forregulatory purposes. Some groupsfavored retaining TSP as an indicator,

Federal Register / Vol. 52,. No. 126 / Wednesday, July 1, 1987

others called for alternative size-specificstandards with nominal "size cuts"('1e"; see discussion in Section III.A.)of 15 pm, 10 G6mm, 5-7 G6mm, and 2.5G6mm. After CASAC closure on thestaff paper and criteria document,comments were received from one groupfavoring a so-called "Do" of 10 Atm(approximately equivalent to a nominalsize cut [Dal of 6 um).

2. Much attention was focused onselecting the level of the primarystandards and on the question of whichhealth effects studies were mostappropriate for this purpose. Significantcriticisms were received on the majorepidemiological studies of particulatematter exposures, highlighting theirlimitations for use in standard setting. Ina number of comments, specificsuggestions for standards were made.

3. With respect to secondarystandards, most attention focused on thepossible need for a fine (-2.5 G6m)particle standard designed to protectvisibility.

These and other major issues arediscussed more fully in the executivesummary of the staff paper and in latersections of this notice. CASAC'sdiscussion of these issues and itsrecommendations are contained in theclosure memorandum on the staff paper(Addendum II).

E. Proposed Revisions to the Standards

On March 20, 1984 (49 FR 10408) EPAproposed a number of revisions to theprimary and secondary particulatematter standards. The proposedrevisions, based on the revised criteria.included:

(1) Replacing TSP as the indicator forparticulate matter for the primarystandards with a new indicator thatincludes only those particles with anaerodynamic diameter less than or equalto a nominal 10 micrometers (PM~o);

(2) Changing the level of the 24-hourprimary standard to a value to beselected from a range of 150 to 250 Mg/m 3 and replacing the deterministic formof the standard, which permitted notmore than one observed exceedance ofthe standard per year, with a statisticalform that would permit one expectedexceedance per year;

(3) Changing the level of the annualprimary standard to a value to be .selected from a range of 50 to 65 f~g/m3,and changing the form from an annualgeometric mean to an expected annualarithmetic mean: and

(4) Replacing the current 24-hoursecondary TSP standard by an annualTSP standard selected from a range of,70 to 90 fg/m3, expected annualarithmetic mean.

The Administrator expressed aninclination to select the primarystandards from the lower portions of theabove ranges. With respect to thesecondary standards, the Administratorwas inclined to select the final standardfrom the upper portion of the range, butalso called for comment on thealternative of using PMo as theparticulate matter indicator for thesecondary standards and making thesecondary standards identical in allrespects to the primary standards. Theproposal notice sets forth the rationalefor these and other proposed revisionsof the particulate matter NAAQS andbackground information related to theproposal.

F. Supplemental Criteria Revisions andStandards Review Following Proposal

Following publication of the proposal,EPA held a public meeting inWashington, D.C. on April 30, 1984 toreceive comments on the proposedstandards revisions. A transcript of themeeting has been placed in the publicdocket (Docket No. A-82-37). After theclose of the original public commentperiod (June 5, 1985), the CASAC met onDecember 16-17, 1985 to review theproposal and to discuss the relevance ofcertain new scientific studies on thehealth effects of particulate matter thathad emerged since the Committeecompleted its review of the criteriadocument and staff paper in January,1982. A transcript of this meeting is alsoavailable in the Docket. Based on itspreliminary review of these new studies,the Committee recommended that theAgency prepare separate addenda to thecriteria document and staff paper for thepurpose of evaluating the relevant newstudies and discussing their potentialimplications for standard-setting. TheAgency announced its acceptance ofthese recommendations on April 1, 1986(51 FR 11058). On July 3, 1986, EPAannounced (51 FR 24392) the availabilityof the external review draft documententitled: Second Addendum to AirQuality Criteria for Particulate Matterand Sulfur Oxides (1982): Assessment ofNewly Available Health EffectsInformation. At the same time, theAgency announced a supplementarycomment period on the March 20, 1984proposal to provide the public anopportunity to comment on theimplications of the new studies andaddenda for the final standards. OnSeptember 16, 1986, EPA announced (51FR 32878) the availability of the draftstaff paper addendum entitled Review ofthe National Ambient Air QualityStandards for Particulate Matter: -Upda ted Assessment of Scientific andTechnical Information. CASAC held a

public meeting on October 15-16, 1986 toreview both the criteria documentaddendum and the staff paperaddendum. At this meeting, CASACmembers as well as representatives ofseveral organizations, provided criticalreview of both EPA documents. Atranscript of the CASAC meeting hasbeen placed in the public docket (A-82-37).

The CASAC sent a closure letter onthe criteria document addendum to theAdministrator dated December 15, 1986,which concludes "that this 1986Addendum along with the 1982 CriteriaDocument, previously reviewed byCASAC, represent a scientificallybalanced and defensible summary of theextensive scientific literature on thesepollutants" (Lippmann, 1986b). Theclosure letter on the criteria documentaddendum is reprinted in Addendum I ofthis notice. The Committee sent theirclosure letter on the staff paper,addendum to the Administrator datedDecember 16, 1986, stating "TheCommittee believes that this documentprovides you with the kind and amountof technical guidance that will beneeded to make appropriate revisions tothe standards" (Lippmann, 1986c). Theclosure letter on the staff paperaddendum, which also discusses majorissues addressed by the CASAC and theCommittee's recommendationsconcerning these issues, is reprinted inAddendum If to this notice. The finaladdenda to the criteria document (EPA,1986a) and the staff paper (EPA, 1986b),which include revisions to reflectcomments from CASAC and the public,are available from the address listedabove. Where there are differencesbetween the 1982 Criteria Document andstaff paper and the more recentaddenda, the addenda supersede theearlier document. The executivesummary of the staff paper addendum isreprinted in Addendum III to this notice.

II. Summary of Public Comments

The following discussion summarizesin general terms the comments receivedfrom the publicand from governmentalagencies regarding the proposedrevisions to the indicator, form,averaging times, and levels of theprimary and secondary standards. Manyof these comments had been made * 'previously by the-public during public,"deliberations on drafts of the criteriadocument and staff paper and were.reviewed and addressed-by EPA in.revisions to those documents. Salientcomments on all aspects of the proposaland Agency responses to thosecomments are summarized by categoryin Section VI of this notice. A more..

/ Rules and Regulations ; 4 24637/ Rules and Regulations .... 24637


detailed description of individualcomments and Agency responses hasbeen entered in the public docket (No.A-82-37).A. Comments on 1984 Proposal

Extensive written comments werereceived during the original commentperiod on the proposal, which closedJune 5,1985. Of some 312 writtensubmissions, 153 were provided byindividual industrial concerns orindustry groups, 93 by State, local, andFederal government agencies andorganizations, 32 by environmental andpublic interest groups, and 34 byindividual private citizens.' Thecomments on the key elements of theproposed standards are summarizedbelow:

(1) Indicator for the PrimaryStandard: The overwhelming majority ofthe comments received on this issuefavored a size-selective indicator for thePM standard. Of the 147 writtencomments received on this issue, 108supported the PMo indicator proposedby the Agency. Most of the remainingcomments were in support of alternativesmaller particle size indicators includingPM0 (28 comments) and PM2.5 (8comments). The principal support forPM0 came from mining and relatedindustries.

(2] Levels of the Primary Standards:Comments on the proposed levels forthe two primary standards were morepolarized than those on the indicator.Most industry comments favoredselecting the level of the standards atthe upper end of the proposed ranges orabove, while most of the remainingcommenters favored standard levels atthe lower bound of the ranges, and insome cases lower. Additional commentsfrom individual citizens, environmentalgroups, and government agencies urgedthat the level of protection afforded bythe current particulate matter standardsbe maintained or strengthened.

(3) Secondary Standards: Of the 105written comments received on theproposed secondary standard, 44supported retaining TSP as the indicatorand 61 opposed the use of TSP. Most ofthe latter commenters supported theproposed alternative of making thesecondary standards equal in allrespects to the primary standards.

IThis numerical distribution of comments in eachcategory should be compared with caution. Forexample, the American Iron and Steel Institute andthe American Petroleum Institute submittedcomments on behalf of 63 and 230 individualcompanies respectively, in lieu ofhaving each oftheir member companies send separate comments.Similarly, comments from interest groups such asNRDC represent the views of a number ofindividuals.

Industry commenters were virtuallyunanimous in opposing a TSP secondarystandard, while a majority (35 of 47) ofgovernment agency comments on thisissued favored retaining the TSPindicator. Some of the lattercommenters, however, recommendedtesting attainment of the TSP standardwith PMo monitors. Environmentalgroups commenting on this issue favoredretaining the TSP indicator.

(4] Form of the Standards: A majorityof the 52 comments received on thissubject supported some kind ofstatistical 24-hour standard, but anumber of industry and State and localagency commenters raised concernswith aspects of the specific formproposed. The principal concern wasthat the proposed form could result inmisclassification of areas as non-attainment. Some industry andgovernmental commenters favoredalternative forms for the 24-hourstandards including multipleexceedance (9 comments) and percentile(8 comments) forms. These forms wouldpermit five or more exceedances peryear of the 24-hour standard.Environmental groups and othergovernment agencies opposed multipleexceedance forms. Of 38 submissionsfrom industry and government agencies,26 favored a geometric mean for theannual standard over the proposedarithmetic mean.

(5) Federal Reference Method: Whilemost of the comments generallysupported the performance-basedapproach to the Federal ReferenceMethod, many commenters favoredmore stringent specifications for PM1osamplers to ensure accurate and reliableperformance under all ambient samplingconditions. Other comments andrecommendations addressed specificrequirements of Appendix J such as flowcalibration and measurement, flowregulation, filter media, humidity controland sampler maintenance.

B. Comments on Subsequent NoticeAs discussedearlier in this notice,

EPA announced an additional publiccomment period on July 3, 1986 toaddress the implications of newscientific studies on the health effects ofparticulate matter [51 FR 24392].Approximately 20 additional writtensubmissions were received by the closeof this comment period on November 17,1986, 17 of which were provided onbehalf of industry groups or companies,2 from environmental groups, and 1 froma state agency. Much of the materialrelated to evaluations of specific studiesand their treatment in the staff paperaddendum. The industry comments,which included submissions from

consulting scientists and analysts,generally found that the new studiessuffered from deficiencies that precludeplacing much weight on them instandard setting. These commentersconcluded that their originalrecommendations (summarized above)with respect to the standards remainedvalid. The two environmental groups feltthat the findings in these new studiesnecessitated standards below the lowerbounds of the proposed ranges.

III. Rationale for the Primary Standards

In selecting primary standards forparticulate matter, the Administratormust specify: (1) the particle sizefraction that is to be used as anindicator of particulate pollution; (2) theappropriate averaging times and form(s)of the standards; and (3) the numericallevels of the standards. Thesespecifications must be consideredcollectively in evaluating the margin ofsafety afforded by particulate matterstandards. Based on the assessment ofrelevant scientific and technicalinformation in the criteria document andaddendum, the staff paper and staffpaper addendum (hereinafter "SP" and"SPA," respectively) outline a number ofkey factors to be considered in makingdecisions in each of these areas (SP,Section VI; SPA, Section IV). Both thestaff and CASAC maderecommendations to focus considerationon a discrete range of options. In mostrespects, the Administrator has adoptedthe recommendations and supportingreasons contained in the staff paper andaddendum and the CASAC closurestatements (Friedlander, 1982;Lippmann, 1986c). Rather thanreiterating those discussions at length,the following discussion of thestandards revisions focuses primarily onthose considerations that were mostinfluential in the Administrator'sselection of particular options, or thatdiffer in some respect fromconsiderations that influenced the staffand/or CASAC recommendations.

A. Pollutant Indicator

Based on the staff assessment of theavailable scientific information, EPAconcludes that (1) a separate particulatematter standard (as opposed to acombination standard for particulatematter and SO) remains a reasonablepublic health policy choice, and (2)given current scientific knowledge anduncertainties, a size-specific (rather than,chemical-specific) indicator should beused. In assessing the information in thecriteria document, the staff reachedseveral conclusions summarized here(see SP, pp. 71-75):

Federal Register / Vol. 52, No. 126 / Wednesday, July 1, 1987 / Rules and Regulations - 24639

(1) Health risks posed by inhaledparticles are influenced both by thepenetration and deposition of particlesin the various regions of the respiratorytract, and by the biological responses tothese deposited materials. Smallerparticles penetrate furthest in therespiratory tract. The largest particlesare deposited in the extrathoracic (head)region with somewhat smaller particlesdepositing in the tracheobronchialregion. Still smaller particles can reachthe deepest portion of the lung, thealveolar region.

(2) The risks of adverse health effectsassociated with deposition of typicalambient fine and coarse particles 2 inthe thorax (tracheobronchial andalveolar regions of the respiratory tract)are markedly greater than thoseassociated with deposition in theextrathoracic (head) region. Maximumparticle penetration to the thoracicregion occurs during oronasal or mouthbreathing.

(3) The size-specific indicator forprimary standards should representthose particles small enough topenetrate to the thoracicregion (boththe tracheobronchial and alveolarregions). The risks of adverse healtheffects from extrathoracic deposition oftypical ambient particulate matter aresufficiently low that particles depositingonly in that region can safely beexcluded from the indicator.

* Considering these conclusionstogether with other information on airquality composition, respiratory tract -deposition and health effects, the needto provide protection for sensitiveindividuals who may breathe by mouthand/or oronasally, and the similarconvention on particles penetrating thethoracic region recently adopted by theInternational Standards Organization(ISO, 1981), the staff recommended thatthe size-specific indicator includeparticles of diameters less than or equalto a nominal 10 jim "cut point."P The

2 Particles in ambient air usually occur in twosomewhat overlapping size distributions, fine(diameter less than 2.5 j~m) and coarse (diameterlarger than 2.5 pm). The two size fractions tend tohave different origins and composition (SP.Appendix D).

3 The more precise term is 50% cut point or 50%diameter (D.). This is the aerodynamic particlediameter for which the efficiency of particlecollection is 50%. Larger particles are not excludedaltogether, but are collected with substantially.decreasing efficiency and smaller particles arecollected with Increasing (up to 100%) efficiency.Ambieit samplers with this cut point provide areliable estimate of the total mass of suspendedparticulate matter of aerodynamic size less than orequal to 10 pm. See additional discussion regardingthe Federal Reference Method in section V belowand in the accompanyingnotice revising 40 CFRPart 53.

factors considered in the original staffrecommendations for a 10 j m cut pointare outlined in the staff paper (SP, pp.75-79). This indicator is referred to as"thoracic particles" (TP) in the 1982 staffpaper; it is now generally referred to as"PMo." Such an indicator isconservative withrespect to healthprotection in that it includes all of theparticles small enough to penetrate tothe sensitive alveolar region, andincludes approximately the sameproportion of larger particles as wouldbe expected to reach thetracheobronchial region. It placessubstantially greater emphasis oncontrolling smaller particles than does aTSP indicator, but does not completelyexclude larger particles from all control.

The assessment of more recentinformation on respiratory tractdeposition in the criteria document andstaff paper addenda reinforces theconclusions reached in the original staffassessment. In particular, the staff paperaddendum found that: (1) the recentdata do not provide support for anindicator that excludes all particleslarger than 10 pim in diameter, 4 (2) theanalysis used to support an alternativeindicator with a nominal size cut of 6G6mm (Swift and Proctor, 1982)significantly underestimated thoracicdeposition of particles larger than 6 pi.min diameter under natural breathingconditions; (3) the PM10 indicatorgenerally includes a similar or largerfraction of the range of particles that can 'deposit in the tracheohronchial region,

-although it appears to be somewhat lessconservative iii this-regard thanpreviously thought with respect to laige(G6<10 pm) particle deposition underconditions of natural mouthbreathing;and (4) the studies of tracheobronchialdeposition generally involved adultsubjects; recent information indicatingeven greater tracheobronchialdeposition of particles in children thanin adults provides an additional reasonfor an indicator that includes particlescapable of penetration to thetracheobronchial region (SPA, p. 36).Consideration of these and the earlierconclusions led the staff to reaffirm itsrecommendation for a PMo indicator(SPA, pp. 36-37). The CASAC alsorestated its recommendation for PMio inits review of the proposal and theclosure letter to the. Administrator(Lippmann, 1986 a, c). .

The Administrator accepts therecommendations of the staff andCASAC and their underlying rationale

'The American Mining Congress (AMC. 1982)had recommended such an indicator, with a "DW of10 Am. EPA estimated that the "DIa" of thisindicator would be.6 pm.

and has decided to replace TSP as theparticle indicator for the primarystandards with a new indicator thatincludes only those particles less than anominal 10 Om in diameter, as specifiedin the Federal Reference Method(Appendix J to 40 CFR Part 50) beingpromulgated today. In defining thestandards for particulate matter, thisnew indicator is termed PMo.

B. Averaging Time and Form of theStandards

-Few comments on the proposedstandards contested the need for both24-hour and annual primary standardsfor particulate matter. EPA's assessmentof more recent scientific informationfound that the new data confirm theneed for both short- and long-termstandards.' The alternative of a singleaveraging time would not provideadequate protection against potentialeffects from both long- and short-termexposures without being undulyrestrictive. The forms for the 24-hourand annual standards are discussedbelow.

1. 24-hour Standard

EPA proposed that the 24-hourstandard be stated in a statistical formthat uses more than one year of dataand accounts for variations in samplingfrequency in order to predict the actualnumber of exceedances to be expectedin an average year, When used with anappropriate standardlevel, thestatistical form can provide improvedhealth protection that is less sensitive tochanges-iri sampling frequency than thedeterministic form, and also can offer, a--more stable target for control. programs.Recognition of the limitations of thedeterministic form has led EPA topromulgate a statistical form for theozone standard (44 FR 8202).

The interpretation of the statisticalform of the particulate matter standardis detailed in Appendix K of theproposed regulation. The standardwould be attained when the expectednumber of exceedances of the 24-hourstandard level is no more than one per,year. The expected number of.exceedances per year is equivalent tothe long-term average number ofexceedances per year, assuming nochanges in underlying emissions.Generally, the determination of theexpected number of exceedance will bebased on three consecutive years ofdata. -. . .. , . .- . .. • - ,. .

As- a result of EPA's evaluations ofevidence submitted and commentsreceived during the public reviewprocess, the following changes have


been made to Appendix K of theproposed rule regarding:

(1) Data Capture Requirements-Appendix K to the proposed standardscontained minimum data capturerequirements for determining attainmentof the standards. The amount of datarequired varies with the samplingfrequency and the number of years ofrecord. The Ambient Air QualitySurveillance regulations (40 CFR Part 58)proposed in 1984 and being promulgatedtoday require that sampling beperformed every day or every other dayin areas where there is a substantialprobability of nonattainment of thestandards. The proposed Appendix K tothe standards, however, would havepermitted states to demonstrateattainment of the standards with only 12samples per calendar quarter, even inthose areas where everyday or everyother day sampling is required.Commentters have argued that, for thesame reasons that everyday or everyother day sampling is required in areaswith a substantial probability ofnonattainment, 12 samples per quarterare not sufficient to establish attainmentin those areas. These commenters alsoargued that 75 percent data capture isachievable at all sampling frequencies.EPA agrees, and therefore the final rulerequires that 75 percent of the requiredsamples must be captured each calendarquarter to establish attainment of theNAAQS.

Additional criteria for situations inwhich less than 3 years ofrepresentative data are available arealso contained in the final rule. Thesecriteria are intended to permit areas todetermine their air quality status in areasonable time frame during the periodin which new PMo monitoring isinitiated, while minimizing theprobability of errors in classification.Appendix K specifies that the variousdata requirements do not apply whenthe data available establishesnonattainment unambiguously.Furthermore,' data not meeting thevarious criteria may also be sufficient toshow attainment; however, suchexceptions will have to be approved bythe Regional Administrator inaccordance with established guidance.

(2) Exceedance Calculations-EPA ismodifying the formulas used to accountfor incomplete data in the estimation ofthe expected number of exceedances peryear. In the proposal, these calculationswere based on the assumption that thefraction of missing values that wouldhave exceeded the standard level isidentical to the fraction of measuredvalues above that level for the entirecalendar year. In the final rule, these

calculations will be required on aquarterly basis, thereby taking intoaccount possible seasonal differences inexceedance rates as well as differencesin sampling frequency or data capture.The estimated annual number ofexceedances is defined as the sum of theestimated exceedances for eachcalendar quarter. This change willaccommodate situations in whichsampling frequency has been increasedto everyday according to therequirements of Part 58.13, andsituations in which the RegionalAdministrator has granted a waiver ofincreased sampling frequencyrequirements for part of the calendaryear under provisions of thosemonitoring regulations.

(3) Interpretation of the FirstObserved Exceedance-EPA isadditionally modifying Appendix K withrespect to the treatment of the firstobserved exceedance in order to reducethe chance of misjudging attainmentstatus. Under the aforementionedformulas which adjust for incompletedata, a single observed exceedancecould cause a site to fail the test forattainment, even if the true expectednumber of exceedances is less than orequal to one. Such an occurrence isespecially likely if sampling isperformed less frequently thaneveryday. In order to reduce the chancesof occurrence of this situation, the finalrule contains a provision that the firstobserved exceedance shall not beadjusted for incomplete sampling if (1)everyday sampling had not beenrequired previously by 40 CFR 58.13, (2)there was only one observedexceedance in the calendar quarter, and(3) sampling frequency has beensubsequently increased for the next 4calendar quarters in accordance with 40CFR 58.13. The associated reduction inmisclassification errors is discussed in"Revising the National Ambient AirQuality Standards for ParticulateMatter-A Selective SamplingMonitoring Strategy" which has beenplaced in the public docket.

With this change, the first observedexceedance can be interpreted as theonly true exceedance which hasoccurred in the calendar quarter. Thisassumption is believed to be reasonablesince incomplete sampling is permittedonly in areas for which stateimplementation plans are not initiallyrequired and in areas in whichmaximum PMo concentrations areestimated to be less than 80 percent ofthe level of the standard. If an area istruly in nonattainment, additionalexceedances would be expected duringthe subsequent year of everyday

sampling. If, however, everydaysampling is not initiated as required bythe monitoring regulations, all observedexceedances shall be adjusted forincomplete sampling and accordinglyconsidered in the evaluation of PM1o airquality status.

2. Annual Standard

The Administrator has decided tochange the form of the annual standardfrom the current annual geometric meanto a statistical form expressed as anexpected annual arithmetic mean. Theexpected annual arithmetic mean isequivalent to the long-term arithmeticaverage concentration level, assumingno changes in underlying emissions. Theexpected arithmetic mean is moredirectly related to the available healtheffects information than is the annualgeometric mean, which is the. currentform of the standard. Because thearithmetic mean concentration isproportional to the sum of the dailymeans, it reflects the total cumulativedose of particulate matter to which anindividual is exposed. Therefore, it is anappropriate indicator to protect againstany health effect that depends on totaldose. It is also a reasonable indicator forprotecting against health effects thatdepend on repeated short-term highconcentrations; short-term peaks havean influence on the arithmetic mean thatis proportional to their frequency,magnitude, and duration. The geometricmean, on the other hand, deemphasizesthe effect of short-term peakconcentrations, and is heavilyinfluenced by days of relatively cleanair. For these reasons, the staff andCASAC recommended the change to anarithmetic mean.

The interpretation of the statisticalform of the standard is detailed inAppendix K to the proposed regulation.Under the statistical form, the expectedannual arithmetic average is determinedby averaging the annual arithmeticaverages from three successive years ofdata. The current deterministic form ofthe standard does not adequately takeinto account the random nature ofmeteorological variations. In general,annual mean particulate matterconcentrations will vary from one yearto the next, even if emissions remainconstant, due to the random nature ofmeteorological conditions that affect theformation and dispersion of particles inthe atmosphere. If only one year of datais considered, compliance with thestandard and, consequently, emissioncontrol requirements, may bedetermined on the basis of a year withunusually adverse or unusuallyfavorable weather conditions. The


problem of year-to-year variability is,however, reduced by averaging threeyears of data.

C. Le vel of the Standards

The original staff paper and CASACrecommendations set forth a frameworkfor determining the levels for theproposed particulate matter standardsthat would protect public health with anadequate margin of safety. Thediscussion that follows relies heavily onthat framework and on the supportingmaterial in the staff paper and itsaddendum as well as the CASACclosure letters. The essential steps inthis framework are summarized here.

1. Assessment of the quantitativeepidemiological studies.

The criteria document and itsaddendum identify a small number ofcommunity epidemiological studies thatare useful in determining concentrationsat which particulate matter is likely toaffect public health. The staff used thesequantitative studies to examineconcentration-response relationshipsand to develop numerical "ranges ofinterest" for possible PMo standards.

A number of uncertainties associatedwith use of these studies must beconsidered in selecting an appropriatemargin of safety. As discussed in thestaff paper and the criteria document,and the addenda to those documents,epidemiological studies are generallylimited in sensitivity and suhject toinherent difficulties involvingconfounding variables. Moreover, manyof the quantitative studies wereconducted in times and places wherepollutant composition may have variedconsiderably from current U.S.atmospheres. Most also have usedBritish Smoke 5 or TSP as particleindicators. None of the publishedstudies used the proposed PMoindicator. Thus, assumptions must be

5British Smoke (BS) is a pseudo-mass indicatorrelated to small particle (aerodynamic diameter lessthan a nominal 4.5 pjm) darkness. This particulatematter indicator was widely used in British andother European studies. See the criteria documentfor a more detailed treatment of BS (CD. pp. 1-88 toI-90 and 14-8 to 14-11).

used to convert the various results tocommon (PMo) units (SP, pp. 96-100;SPA pp. 9-11).

2. Identification of additional marginof safety considerations.

The criteria document identifies anadditional substantial body of scientificliterature that, while not providingreliable concentration-responserelationships for ambient exposures,does provide important qualitativeinsights into the health risks associatedwith human exposure to particles. Thisliterature includes both quantitative andqualitative epidemiological studies,controlled human exposure experiments,and animal toxicological studies. Thestaff assessed this literature to identifyadditional factors and uncertainties thatshould be considered in selecting themost appropriate margin of safety (SP,pp. 100-101; 107-111, SPA pp. 52-53; 59).

3. Selection of the levels that might beconsidered to provide an adequatemargin of safety.

The intent of the margin of safetyrequirement was to direct theAdministrator to set air qualitystandards at pollution levels belowthose at which adverse health effectshave been found or might be expected tooccur in sensitive groups. Experiencewith the requirement has shown that thescientific data are often so inconclusivethat it is difficult to identify withconfidence the lowest pollution level atwhich an adverse effect will occur.Moreover, in cases such as the presentone, the evidence suggests that there is acontinuum of effects, with the risk,incidence, or severity of harmdecreasing, but not necessarilyvanishing, as the level of pollution isdecreased.

In the absence of clearly identifiedthresholds for health effects, theselection of a standard that provides anadequate margin of safety requires anexercise of informed judgment by theAdministrator. The level selected willdepend on the expected incidence andseverity of the potential effects and onthe size of the population at risk, as wellas on the degree of scientific certainty

that the effects will in fact occur at anygiven level of pollution. For example, if asuspected but uncertain health effect issevere and the size of the population atrisk is large, a more cautious approachwill be appropriate than would be if theeffect were less troubling or the exposedpopulation smaller.

EPA staff originally recommended arange of potential standards for theAdministrator's consideration (SP, pp.111-114). The recommended range wasbelow the levels at which the staff, withthe concurrence of CASAC, hadconcluded from the available data thatadverse health effects were "likely," butin the domain where the data suggestedthat such effects were "possible." TheAdministrator proposed refined rangesof standard levels that were based onthe original staff and CASACrecommendations. After considerationof the new scientific evidence containedin the criteria document addendum, thestaff revised its recommendations forranges of standards (SPA, pp. 60-62).The Administrator has considered therevised assessments and therecommendations of CASAC (Lippmann,1986b) in making his final decision onthe standard levels. The rationales forthe levels of the 24-hour and annualstandards are presented below.

1. 24-Hour Standard

The revised staff assessment of theshort-term epidemiological data issummarized in Table 1; particulatematter levels are expressed in both theoriginal (British Smoke ["BS"] or TSP)and PMo units. The "effects likely" rowin Table 1 denotes concentration rangesderived from the criteria document andits addendum at or above which aconsensus judgment suggests greatestcertainty that the effects studied wouldoccur, at least under the conditions thatoccurred in the original studies. In the"effects possible" range, the staff foundcredible scientific evidence suggestingthe existence of adverse health effectsin sensitive populations, but substantialuncertainty exists regarding theconclusions to be drawn from suchevidence.

TABLE 1.-UPDATED STAFF ASSESSMENT OF SHORT-TERM EPIDEMIOLOGICAL STUDIES

(After Table 4-1, SPA)

Measured British smoke levels (as jig/mn) (24-hr. avg.) Measured TSP levels (pg/mn) (24/ Equivalent PM-ohr. avg.) Levels (jg/m)Effects/St'udy Daily Mortality in Aggravation ofLondon Aronchitis o Combined range Small, reversible declines in lung Combine range

Lnobrnhts2function in children 3.4 Cmin'ag

1000 250-*500"?1 1 <250*

250-500 .................................................................< 250 220-420 3-200-250 4

350-600140-350

Effects Likely ..............................Effects Possible .........................

24641,


TABLE 1..--UPDATED STAFF ASSESSMENT OF SHORT-TERM EPIDEMIOLOGICAL STUDIES-Continued

(After Table 4-1, SPA)

Measured British smoke levels (as fig/m 3) (24-hr. avg.) Measured TSP levels (,Ug/m 3) (24/ Equivalent PM-,ohr. avg.) Levels (Ag/m 3)

Effects/Study Daily Mortality in Aggravation of Combined range Small, reversible declines in lungLondon I bronchitis = function in children 6 Combine range

No Significant Effects Noted ......................................................................................................... 125* -160 < 125

*Indicates levels used for upper and lower bound of range.Various analyses of daily mortality encompassing the London winter of 1958-59, 14 winters from 1958-72, in aggregate and individually.

Early winters dominated by high smoke and SO2 from coal combustion with frequent fogs. From 1982 CD: Martin and Bradley (1960); Ware et al.,(1981); Mazumdar et al. (1981). From 1986 CD Addendum: Mazumdar et al. (1982); Ostro (1984); Schwartz and Marcus (1986). Later studiesshow association, across entire range of smoke, with no clear delineation of "likely" effects or threshold of response possible.

2 Study of symptoms reported by bronchitis patients in London, mid-50's to early 70's; Lawther et al. (1970).3 Study of pollution "episodes" in Steubenville, Ohio, 1978-80; Dockery et al. (1982).4 Study of 1985 pollution episode in ljmond, The Netherlands; Dassen et al. (1986).5.(a) Conversion of BS readings to PM10 levels: Assumes for London conditions and BS readings in the range 100-500 t.g/m 3,

BS<PM1o<TSP.. Precise conversions are not possible. Uncertainty in measurements of BS and conversion relationships preclude quantitativeestimates of range for lower BS levels. The upper bound assumption (PMo=TSP=BS+ 100 pg/m3) overestimates PM0 levels, while the lowerbound assumption (PMo=BS) understates PM1o levels.

(b) Conversion of TSP to PMio for Dockery et a1 results: Based on analysis of particle size fraction relationships In Steubenville (Spengler etal. 1986). The lower bound TSP of 220 pgg/m was the peak reported for the Spring 1980 study. A PM, 1/TSP ratio of about 0.8 occurred at anearby site on days surrounding this peak. Using, lower bound of PM1o/PM10 ratio from later year (0.8), the PM1o to TSP ratio estimate used In0.64. The 160 pgi/m 3 reflects peak level in Fall 1980 from episode with no significant functional decline noted.

(c) Conversion of Dassen et al. results to PMo: Both PM indices (Respirable Suspended Particles [RSP] and TSP) reached, similar levels.Results suggest TSP levels, too low, but PMto levels unlikely to be much higher than RSP. Thus RSP=PMo assumed for conditions of, higherconcentrations in this study. The 125 gIg/m 3 entry reflects an excursion occurring 2 days prior to date on which no decrements noted.

The data do not provide evidence ofclear thresholds in expnsed populations.Instead, they suggest a continuum ofresponse for a given number of exposedindividuals with both the likelihood(risk) of any effects occurring and theextent (incidence and severity) of anypotential effect decreasing withconcentration. This is particularly truefor the, statistical analyses of dailymortality in London. Substantialagreement exists that wintertimepollution episodes produced prematuremortality in elderly and ill. populations,but the range and nature of associationprovide no clear basis for distinguishingany particular lowest "effects likely"levels or for defining a concentrationbelow which no association remains.The recent lung function studies inchildren also provide evidence of effectsat concentrations in the range listed inTable 1, but the relationships are notcertain enough to derive "effects likely"levels for PM~o. The lung functionstudies do, however, suggest levelsbelow which detectable functionalchanges are unlikely to occur in exposedpopulations. Following CASACrecommendations, the staff used thecombined range listed in the "effectspossible" row as a starting point fordeveloping alternative standards.

The original range proposed by theAdministrator, drawn from the 1982 staffanalysis, was 150 to 250 Ag/m3 PMo, 24-hour average with no more than oneexpected exceedance per year. Thelower bound of this range was derivedfrom the original assessment of the

London mortality studies. As a result ofits updated assessment of reanalyses ofthe London mortality and more recentU.S. morbidity studies, the staff reducedthe level of the lower bound of the rangeof interest to 140 pg/m a (SPA, p. 51),while noting that the difference betweenit and original lower bound (150 ipg/m 3 )is within the range of uncertaintyassociated with converting themorbidity study results from TSP toPM10.

As indicated in Table 1, the study ofLawther et al. (1970) judged to provideevidence that health effects are likely atparticulate matter concentrations above250 Ag/m 3 (as BS). The effects observedin this study (related to aggravation ofbronchitis) are of concern both becauseof their immediate impact and becauseof the potential for inducing longer-termdeterioration of health status in asignificant sensitive group. There wereapproximately 6.5 million bronchitics in,the U.S. in 1970 (DHEW, 1973). Based onthe uncertain conversion betweensmoke and PMo outlined in Table 1, thelowest "effects likely" level derivedfrom the Lawther study (250 ,g/m s asBS) should be in the range of 250 to 350Ag/m 3, in PMo units.

The assessment of this study formedthe basis for the upper bound of therange of PMo standards proposed by theAdministrator In 1984. Considering thisstudy alone, a PMo standard of 250 A*g/m3 might appear to contain some marginof safety, even for the sensitivebronchitics studied, because Itincorporates a conservative British

Smoke/PMio conversion factor andbecause of differences betweenexposure conditions in the British studyand current U.S. air quality (SP, pp. 100-101). Because bronchitics are identifiedas a group particularly sensitive toparticulate pollution, a standard of 250:pg/m3 (as PMzo) also might providesome margin of safety for other, lesssensitive, groups. Nevertheless, thisstudy of bronchitics in London hasinherent limitations in sensitivity thatpreclude derivation of unequivocal"effects thresholds" at 250 Ag/m3 as BS,and by extension PMo. The criteriadocument notes that associationsbetween pollution and health statuspersisted at lower BS concentrations inselected, more sensitive individuals.Although the lead author of the studyobjects to attaching any importance tothese latter findings (Lawther, 1986),EPA, with CASAC concurrence, finds nobasis for asserting that this studydemonstrates a population threshold at250 p.g/m3.

In evaluating the margin of safety for,a 24-hour standard, it is also importantto consider the London mortalitystudies. A standard at the upper portionof the proposed range (250 jig/m 3)would he well below the levels (500 to1000 pg/m s as BS) of the historicalLondon episodes in which the scientificconsensus indicates that pollution wasresponsible for excess mortality (CD,Table 14-7). The portions of thepopulation at greatest risk of prematuremortality associated with particulatematter exposures in such episodes

Federal Register / Vol. 52, No. 126 / Wednesday, July 1, 1987 / Rules and Regulations 24843

include the elderly and persons withpre-existing respiratory or cardiacdisease. Although the extent of lifeshortening (days, weeks, or years)cannot be specified, the seriousness ofthis effect strongly justifies a margin ofsafety for it (below the consensuseffects levels) that is larger than thatwarranted for the effects on bronchitics.

The staff assessment of the severalreanalyses of London mortality suggests,however, that the risk of prematuremortality to sensitive individualsextends to concentrations substantiallylower than those which occurred in the"episodes." The more recent analyses(Mazumdar et al., 1982; Ostro, 1984;Shumway et al., 1983) provide noobjective support for a populationthreshold below which such a risk nolonger exists. Although the risk toindividuals may be small atconcentrations of 250 1Lg/m 3 and below,the number of people exposed to lowerconcentrations given current U.S. levelsis substantially larger than the numberexposed to higher levels (SPA, Table 2-1). The increased number of individualsexposed increases the risk that effectswill occur in the total populationexposed.

Differences in the composition ofparticles and gases among U.S. citiesand between current conditions in theU.S. and those in London at the time themortality and morbidity data weregathered add to the complexity ofassessing the risk associated withparticulate matter in the U.S. In the caseof the mortality studies, however, thestaff found that at least one of the morerecent studies (Ozkaynak and Spengler,1985) provides qualitative support for anassociation between daily mortality andparticle concentrations in nearlycontemporary U.S. atmospheres (SPA,pp. 43-44).

The 1982 assessment of the mortalitystudies and related factors prompted theAdministrator to consider standardlevels that extended from 250 Ag/mSdown to the lower bound of the originalstaff range of interest (150 gg/m 3) andeven lower. The more recent analyses ofthe London mortality data provideadditional evidence that serious adversehealth effects may occur at particulateconcentrations below 250 1g/m 3 . Theseanalyses have addressed a number ofthe uncertainties associated with theearlier studies, and have reinforced theAdministrator's concern that a 24-hourstandard at the upper end of theproposed range may not provide anadequate margin of safety. However,given the uncertainties in convertingfrom BS to PMo measurements,particularly at lower concentrations,

and the possible differences inparticulate composition betweenLondon at the time the data weregathered and the contemporary U.S., itis difficult to use these studies to set aprecise level for a PM1o standard (SPA,pp. 49-51).

Given these difficulties, it is importantto examine contemporary studies thatutilize gravimetric measurements ofparticulate concentrations. The stafffound the studies of Dockery et al. (1982)and Dassen et al. (1986) to beparticularly useful. The Dockery studyobserved physiologically small butstatistically significant decreases in lungfunction in a group of children exposedto peak PMo levels of 140-250 pg/m 3 .The decrements persisted for 2-3 weeksfollowing the exposures. The study alsosuggested the possibility of largerresponses in a subset of the children,including those with existing respiratorysymptoms. The Dassen study recordedsimilar decrements in children in theNetherlands following exposure to PMolevels estimated at 200 to 250 gg/m 3, butno observable effects two days afterexposure to PMo levels estimated at 125tg/m3. The particle composition, atleast in the Dockery study, is morerepresentative of contemporary U.S.cities and the associated aerometryprovides a more reliable estimate ofPM1o levels than do the measurementsused in the London studies. It isreasonable to-expect that the effectsobserved (small reversible reductions inlung function in children] are, in mostcases, more sensitive to air pollutionthan those observed in the Londonstudies. These effects are, of themselves,of uncertain significance to health, butmight be associated with aggravation ofrespiratory syiptoms in children withpreexisting illness (SPA, p. 47). Long-term examination of respiratory healthin the same community studied byDockery et al. (1982) suggests that thechildren in that community have ahigher incidence of respiratory illnessand symptoms than children incommunities with lower particle levels,but the data show no evidence for anypersistent reduction in lung function(Ware et al., 1986). Uncertainties withrespect to the effects of other pollutants(e.g., S02), the consistency of thechanges, and exposures precludespecifying unequivocal "effects likely"levels based on this study. The staffassessment therefore suggests thatshort-term lung function effects inchildren are possible across a range of140-250 l±g/m3 or more as PM 0 (SPA, p.50).

In making a decision on a finalstandard level, the Administrator also

considered information from the morequalitative studies of PM assessed bythe staff (SP, pp. 101-103; SPA, pp. 52-53). These suggest increased risks forsensitive groups (asthmatics) and risksof potential effects (morbidity in adults)not demonstrated in the morequantitative epidemiological literature.The qualitative studies do not provideclear information on effects levels, butdo justify consideration of effects ofparticulate matter that have not beensufficiently investigated.

Based on the scientific assessment atthe time, the Administrator in 1984expressed an inclination to select a 24-hour level from the lower portion of theproposed range of 150-250 jg/ms. Thepresent Administrator finds that theupdated scientific assessment supportsthe original inclination and, if anything,suggests an even wider margin of safetyis warranted. The recent analyses ofdaily mortality are of particular concernin this regard. The Administrator has,therefore, decided to set the finalstandard at the extreme lower bound ofthe range originally proposed; that is, at150 p.g/M 3 . This standard provides asubstantial margin of safety below thelevels at which there is a scientificconsensus that particulate matter causespremature mortality and aggravation ofbronchitis. Such a margin is necessarybecause of the seriousness of theseeffects and because of the recentanalyses of daily mortality that suggestadverse effects may occur at particulatematter levels well below the consensuslevels. The standard is in the lowerportion of the range where sensitive,reversible physiological responses ofuncertain health significance arepossibly, but not definitely, observed inchildren. Using a conservativeassessment of lung function/particlerelationship from Dockery et al., achange in concentration frontbackground levels (-20 jLg/m 3) to 150j~g/m 3 would produce lung functionchanges of at most 10 to 15% in less than5% of exposed children (SPA, p. 48).Based on the results of Dassen et al.(1986), it appears unlikely that anyfunctional changes would be detectedone or two days following suchexposures (SPA, p. 50). Thus, themaximum likely changes in lungfunction appear to present little risk ofsignificant adverse responses.

.Standards set at a somewhat higherlevel would, however, present anunacceptable risk of prematuremortality and allow the possibility ofmore significant functional changes.Furthermore, a standard level of 150 zg/m3 is fully consistent with the


-recommendations of CASAC on the 24- term epidemiological studies are'subject can be identified for all effectshour standard (Lippmann, 1986c). to additional confounding variables that categories. Evidence exists of effects at2. Annual Standard -reduce their sensitivity and make their lower levels-the "effects possible.

interpretation more difficult than that of levels"-but the evidence isThe updated staff assessment of short-term studies. The "effects likely" inconclusive and effects are difficult to

important long-termepidemiological levels are derived from the criteria detect In the available epidemiologicaldata is summarized in Table 2. Long- document, but again, no clear thresholds studies.

TABLE 2.-UPDATED STAFF ASSESSMENT OF LONG-TERM EPIDEMIOLOGICAL STUDIES (AFTER TABLE 4-2, SPA)

Measured BS Measured TSP levels (pg/m3) Equivalentlevels,(as pig/ PM10 levels

m 3) Increased (,ig/m 3)respiratory Increased

Effects/Study Increased disease, Increased respiratory Reducedlungrespiratory symptoms, respiratory symptoms and function in Combineddisease, small symptoms in illnesses in childreni

reduced lung reduction in adults i children range Creduce lungrange 5function in lung function children 4children I in adults 2

Effects likely ...................... 230-300 "180 ................................. .......... >180 >80-90Effects possible ................. <230 "130-180 60-150(110) 60-114 ............................. 60-180 40-90No significant 6 effects

noted ....... .. ....................... 80-130 .......................................................... 40-114 <60 <40

*Indicates levels Used for upper and lower bound of range.':Study conducted in 1963-65 in Sheffield, England (Lunn et al., 1967). BS levels (as pg/m 3) uncertain.2 Studies conducted in 1961-73 in Berlin. NH (Ferris et al., 1973, 1976). Effects likely level (180 fg/m3) based on uncertain 2-month average.

Effects in lung function. were relatively small.3 Study conducted in 1973 in two Connecticut towns. (Bouhuys et al., 1978). Exposure estimates reflect 1965-73 data In Ansonia. Median

value (110 g/ms ) used to indicate long-term concentration. No effects on lung function, but some suggestion of effects on respiratory symptoms.4 Study conducted in 1976-1980 in 6 U.S. cities (Ware et al., 1986). Exposure estimates reflect 4-year averages across cities. Comparable

pollution/effects gradients not noted within cities.. Conversion of TSP to PMo equivalents for Berlin, Ansonia studies based on estimated ratio of PM,o/TSP for current U.S. atmospheres

(Pace, 1983). The estimated ratio ranged between 0.45 and 0.5. Conversion for six-city study based on site-specific analysis of particle size data(Spengler et' al., 1986).

6 Ranges reflect gradients in which no significant effects were detected for categories at top. Combined range reflects all columns.

Based on a recent assessment ofPM,6/TSP ratios in areas with elevatedTSP levels, the updated staff assessmentrevised the "effects likely" levels fromthe Ferris et al. (1973) study to 80 to 90fg/m 3 as PMo (SPA' p. 58). Because oflimitations in sampling duration as wellas the conversion to PMo, this estimateis particularly uncertain. As indicated inthe table, effects are possible at lowerconcentrations. Of greatest concern isthe possibility of long-term deteriorationof the respiratory system in exposedpopulations, the potential for which isindicated by lung function (mechanicalpulmonary) changes and increasedincidence of respiratory disease. One setof studies (Ferris et al., 1973, 1976),provides some evidence for a "noobserved effects" level for these effectsat or below 60 to 65 jig/m (130 jig/n 3

as TSP) while another study (Bouhuys etal., 1978), suggests some possibility ofsymptomatic responses in adults at long-term median levels at or below about 50to 55 pg/m3 as PMo. The importance ofthese symptomatic responses, whichwere unaccompanied by lung functionchanges, to long-term respiratory healthis unclear.- The most important recent study oflong-term effects is an ongoing

* examination of six U.S. cities (Ware etal., 1986). The study indicates thepossibility of increased respiratorysymptoms and illnesses in children atmulti-year levels across a range of 40 toover 58 jig/m as PMo, but found noevidence of reduced lung function atsuch concentrations. This study did notfind similar gradients in symptoms andillness within some of the cities, whichhad somewhat smaller localizedpollution gradients. The results of aseparate series of studies of long andintermediate term (2 to 6 weeks)exposures in a number of U.S.metropolitan areas (Ostro, 1987;Hausman et al., 1984) are moresupportive of the possibility of effectswithin cities (respiratory related activityrestrictions in adults) at comparableU.S. exposure levels. The results ofthese more recent studies are generallyconsistent with the earlier U.S. studieslisted in Table 2 (SPA, 57). In particular,the finding of symptomatic responses inchildren with no change in lung function(Ware et al., 1986) is consistent withsimilar findings in adults (Bouhuys et al.,1973) at estimated long-term PMo0 levelsdown to 50 jg/m. However, theinformation available to support theexistence of significant adverse effects

at annual PMo levels below 50 g/m3 -especially when 24-hour levels aremaintained below 150 p.g/m-is quitelimited and uncertain.

Because of the uncertainties in (SP,pp. 104-110; SPA, 54-59), as well as thelimited scope and number of, these long-term quantitative studies, it isparticularly important to examine theresults of qualitative data from anumber of epidemiological, animal, andambient particle composition studieswhen evaluating what constitutes anadequate margin of safety for an annualstandard. These studies justify concernfor serious effects not directly evaluatedin the studies listed in Table 2. Sucheffects include damage to lung tissuescontributing to chronic respiratorydisease, cancer, and prematuremortality (SP, pp. 109-111). Substantialsegments of the population may besusceptible to one or more of theseeffects (SP, p. 46). Although thequalitative data do not provide evidencefor major risks of these effects at-currentannual particulate matter levels in mostU.S. cities, the Administrator believes -that the seriousness of the potential.effects and the large population at riskwarrant caution in setting the standard;


Based on the then current scientificassessment, the Administrator proposedin 1984 to select the annual standardlevel from a range of 50 to 65 Ag/m 3 . Inthe proposal, the Administrator favoreda standard in the lower portion of therange. The more recent evidence,although subject to substantialuncertainty, serves to reinforce thisinclination. In light of the updatedassessment and in accordance with therecommendation of CASAC, theAdministrator has decided to set thelevel of the annual standard at the lowerbound of the original range, 50 pgg/m s ,expected annual arithmetic mean. Thisstandard provides a reasonable marginof safety against the serious effect oflong-term degradation in lung function,which has been judged likely atestimated PMo levels above 80-90 #Lg/.ms and for which there is some evidenceat PMo levels above 60 to 65 Ag/m 3 .Such a standard also providesreasonable protection against the lessserious symptomatic effects for whichsome studies provide evidence at PMolevels down to 50 g/mS* Although somesmall risk of increased respiratorysymptoms may exist at thisconcentration, the available data arecurrently inconclusive on this point.Moreover, the staff and CASAC haverecommended that the combinedprotection afforded by both 24-hour andannual standards be considered inselecting the final standard level. In thisregard, analyses of air quality datashow that implementation of the 24-hourstandard will substantially reduceannual levels in a number of areas tobelow 50 Ag/m3, adding to theprotection afforded by the annualstandard in areas with higher 24-hourpeak to mean ratios (SPA, p. 61; Freas,1986). Based on the present evidencewith respect to risks associated withannual exposures, the Administratorfinds that the annual and 24-hourstandards announced today provide anadequate margin of safety.

IV. Rationale for the SecondaryStandards

Section 109(b)(2) of-the Clean Air Actstates that secondary NAAQS should beset at a level requisite to protect thepublic welfare from any known oranticipated adverse effects associatedwith the presence of an air pollutant inthe ambient air. The criteria documentand staff paper examined the effects ofparticulate matter on such aspects ofpublic welfare as visibility and climate,man-made materials, vegetation, andpersonal comfort and well being. Eachaspect is discussed in some detail inthose documents. The followingdiscussion of the rationale for the

secondary standards focuses primarily.on considerations that were mostinfluential in the Administrator'sdecision or that differ in some respectfrom, or expand upon, -considerationsthat influenced the staff and/or CASACrecommendations.

A. Soiling and NuisanceAt high enough concentrations, both

large and small particles may soilhousehold and other surfaces, orotherwise become a nuisance. Botheffects can result in increased cleaningcosts and decreased enjoyment of theenvironment (SP, p. 140). Efforts tocontrol particulate matter in U.S. citiesfrom 1970 to 1978 were estimated tohave produced substantial economic

.benefits because of reduced soiling andnuisance (CD, p. 1-51). The staff papertherefore recommended consideration ofsoiling and nuisance generated by dustand other particles in setting asecondary standard (SP, p. 141).

In proposing secondary standard(s)for particulate matter, the Administratorfirst examined whether the pollutantindicator (PMio), averaging times andform, and range of levels of theproposed primary standards wouldprovide adequate protection againstsoiling and nuisance. This examinationwas complicated by uncertainties in thescientific data base that largely precludeaccurate quantification of the extent ofeffects associated with specific particlesizes and concentrations or deposition,and by the fact that the protectionafforded by primary standards dependsupon the particular combination oflevels chosen within the ranges thatwere proposed for the primarystandards. The Administrator proposeda separate indicator and range ofsecondary standards, while alsosoliciting comment on the alternative ofmaking the secondary standardsidentical in all respects to the proposedprimary standards for PMo. In so doing,the Administrator noted that "dependingon the exact levels of primary standardschosen, the combined requirements formeeting both 24-hour and annualprimary standards for PM1o might beconsidered adequate to protect againstpossible adverse effects relating tosoiling and nuisance from all relevantparticle sizes." (49 FR 10418].

The decision to adopt the specificrevised primary standards discussed insection IV above permits a moredefinitive assessment of the protectionafforded by those standards againstpotential adverse welfare effects. Inaddition, information submitted in thepublic comments, the review of theMarch 20, 1984 proposal by the CASAC,and further analysis of the welfare

effects information by Agency staff haveamended the basis for the final decisionon the secondarystandards. The basisfor the original p'i6osal and the'implications of the more recent findingsare summarized below.

The Administrator originally proposed(1) to retain TSP as the indicator for thesecondary standard and (2) to select thestandard level from a range of 70-90 pg/in s , expected annual arithmetic meap.Given the nature of the evidenceavailable, the Administrator expressedan inclination to select the level for thestandard from the upper portion of therange.

The proposal noted that both PMoand TSP could be useful indicators for asecondary standard for soiling andnuisance. PM1o is useful because in aqualitative sense: (1) Particles smallerthan 10 Am in diameter are more likelythan larger particles to penetrateindoors; they are also more likely thanlarger particles to soil vertical surfaces(SP, pp. 136-137) and (2) due to the.characteristic size distributions andorigins of particles in the atmosphere(SP, pp. 14-19), control of particles lessthan 10 gAm in diameter would also limitthe concentration of larger particles. TheTSP indicator was proposed, however,because of the lack of data'permittingclear distinctions among size rangeswith respect to soiling and nuisance, themore inclusive nature of TSP, and thefact that most of the availableinformation relating soiling andnuisance to air pollution used TSP as anindicator.

Information submitted in the publiccomments expanded on some of thelimitations of TSP as an indicator thatwere noted in the preamble, namely: (1)The collection efficiency of the highvolume sampler, which measures TSP,decreases rapidly for particles withdiameters in excess of 25-40 A.m; thus,the TSP measurement itself can omit asubstantial fraction of the very largeparticles that can make a substantialcontribution to soiling of horizontalsurfaces; and (2) because the collectionefficiency of the high volume samplervaries more with windspeed than doPMlo samplers, TSP may be a lessreliable indicator of elevated "concentrations of larger particles thanPM1o.

In light of these considerations, theCASAC in reviewing the March 20proposal package concluded that itcould find no convincing scientificsupport for maintaining TSP as anindicator for the secondary stahdards(Transcript of December 16, 1985CASAC meeting,. pp. 56-71; Docket No.A-82-37).

24645

24646 Federal Register / Vol. .52; No. 126 / Wednesday, July- 1, .1987 / Rules and Regulations

.In developing a range of levels for thesecon dary standard, EPA found that the.available data base provides compellingevidence that elevated levels ofparticulate matter can produce adversewelfare effects, but provides littlequantitative information on, -.concentration-effects relationships.Physical damage and economic studiestend to show no obvious welfare effects"thresholds" for soiling. With time,particulate matter may accumulate onsurfaces even at low concentrations. Atvery low concentrations, however, theamounts of particulate matter may bevirtually invisible to the human eye orbe so slight as to be ignored by mostpeople (Carey, 1959; Hancock et al.,1976). Up to a point, the buildup ofparticles on surfaces may notbegenerally regarded as a social problembecause it is removed'by rain-or routinecleaning and maintenance beforesubstantial accumulation can oqcur.Moreover, even if an accumulation islarge enough to be noticed, it is not •necessarily considered to be a problem.Thus, the critical judgment for selectinga standard level is to determine aparticulate matter concentration at or.above which the soiling effect becomesimportant enough that it should beregarded as an "adverse" effect undersection 109(b)(2) of the Act.

-The available information suggeststhat the public does make a distinctionbetween concentrations at whichparticulate pollution is merelynoticeable and higher levels at which itis considered a nuisance. A study of theresponse of a panel of-human subjects todust on surfaces concluded that thelevel of dustiness that is found to beobjectionableis higher than the levelthat can be perceived or discriminated(Hancock et al., 1976). It is not, however,possible to derive unique. ambientconcentration thresholds for adverseeffects from this kind of study. A moredirect study of perception of airpollution as a nuisance (CD, p. 9-67)suggested that people considered airpollution a nuisance in areas whereannual levels were at or somewhatabove the level of the current annualprimary TSP standard (75 jug/ms, annualgeometric mean). The upper bound of: .the proposed range of interest (90 #g/m3TSP), expected annual arithmetic mean,was derived by taking that-level andmaking appropriate conversions toaccount for the expected arithmeticmean form.

The lower bound of the proposedrange (70 ig/m 3) was supported by arough analysis of economic benefits of.reduced outdoor soiling that. might beassociated with decreased TSP levels in

U.S. cities (CD, p. 10-73). During the1ublic comment period, one of theauthors of the analysis that formed thebasis for these estimates submitted a'more recent analysis which called theearlier analysis into question (Watsonand Jaksch, 1984). The author claimedthat estimates of benefits from reducedTSP concentrations were significantlyoverstated because they did not takeinto account the extent to which thepublic could perceive improvementsassociated with reduced concentrations.Other commenters indicated that theunderlying experimental data suggesteda threshold for economic soiling effectsat an annual TSP level of about 150 ;g/M 3

.

EPA staff examined the underlyingexperimental data used in the originalanalysis. This staff examination(Haines, 1987) has been placed in therulemaking docket. The staff found thatof 27 household cleaning activitycategories examined in the underlyingexperiment (Booz, Allen, Hamilton,1970), 6 (5 outdoor) were statisticallysignificantly associated with particulatematter across some concentrationgradient. In further comparing areaswith differing concentrations of in TSP,it was found that the number ofsignificant associations decreased withdecreasing TSP levels. The staffconcluded that these data provide noconvincing evidence to supportestimates of significant economicbenefits from reducing PM levels below90 to 100 ltg/m3.. Following the original inclination of

the Administrator and the more recentfindings, an annual TSP level of 90 pg/m3 was used as a benchmark in ananalysis to determine whether theprimary particulate matter NAAQSwould protect against soiling andnuisance (SP, Table 2-1). An earlierversion of these results was presented atthe December 16, 1985 CASAC meeting.The analytical approach, assumptions,and limitations of the methodology usedin the analysis are discussed in aseparate report, which has been 'placedin the rulemaking docket (Pace et al.

* 1986). The results indicated that thecombined implementation of the primary24-hour and annual PMo standardsannounced above would substantiallyreduce TSP levels to the extent that only6 counties nationwide would experienceannual mean TSP levels in excess of 90pg/m s and none would exceed 100 1g/in 3

.

In short, EPA has determined thatthere is.no convincing evidence ofsignificant adverse soiling and nuisanceat TSP levels below 90-100 pug/m3, andthat the primary standards promulgated

today would permit few, if any, areas to'sustain'TSP levels above 90-100 Pg/mn.On the basis of these determinations,the Administrator concludes that asecondary standard different from theprimary standards is not-requisite toprotect the public welfare against soilingand nuisance. This conclusion issupported by the CASAC'sdetermination that there is no scientificsupport for a TSP-based secondary 'standard. (Transcript of December 161985, CASAC meeting, p. 71; Docket No.A-82-37). Therefore, the Administratorhas decided to set 24-hour and annualsecondary PMo standards that are equalin all respects to the primary standards.

B. Other Welfare Effects

The other welfare effects ofparticulate matter of principal interestare impairment of 'visibility, potentialmodification of climate, andcontribution to acidic deposition. Allthree of these effects are believed to berelated to regional-scale levels of fine'particles, and control programs deaignedto ameliorate them would likely involveregion-wide reductions in emissions ofsulfur oxide (SP, p. 147; Friedlander,1982).

Because of the likely overlap betweencontrol measures designed to protectvisibility and control measures designedto address acidic deposition, EPA, in itsMarch 20,1984, notice of proposed.rulemaking on the particulate matter.standards, did not propose a secondarystandard designed to protect visibility.Instead, the Agency decided to deferaction pending development of.compatible strategies to address both of;these related regional air qualityproblems.

Since publication of the notice ofproposed rulemaking, EPA hascontinued to gather information onacidic deposition and on visibility, andto analyze the potential impact onvisibility of strategies designed tocontrol acidic deposition. In particular,EPA has received the report of anInteragency Task Force on Visibility. Inlight of the Task Forcesrecommendations as well as otherinformation gathered by the Agency,EPA is now reassessing its position with.regard t0consideration of a secondaryfine particle standard for.visibility. Inparticular, the Agency is consideringwhether, given the time that would berequired to develp, propose,'promulgate, and implement a visibilitybased standard, it would now'beappropriate to: proceed withconsideration of a visibility based -standard in parallel with work on aciddeposition, so that compatible strategies

Federal Register / Vol. 52, No. .126 / Wednesday, July 1, 1987 / Rules and Regulations

for dealing with the two problems canbe developed at the implementationstage., :

Accordingly, EPA is publishingelsewhere in today's Federal Registeran advance notice of proposedrulemaking soliciting public comment onthe appropriateness of a separatesecondary fine particle standarddesigned to protect visibility, and on anumber of issues that would have to beresolved in proposing such a standard.

The Administrator also concurs withthe staff suggestions that a separatesecondary particle standard is notneeded to protect vegetation or toprevent adverse effects on personalcomfort and well-being (SP, pp. 143-144).V. Federal Reference Method

The reference method for themeasurement of atmospheric particulatematter as PM1o, promulgated today asAppendix J to 40 CFR Part 50, is basedon selection of PM1o particles by inertialseparation, followed by filtration andgravimetric determination of the PMomass on the filter substrate. The particlesize discrimination characteristics ofreference method samplers (or samplerinlets) are prescribed as performancespecifications in amendments to 40 CFRPart 53, promulgated elsewhere intoday's Federal Register.

The requirements in Appendix J aregenerally prescribed as functional orperformance specifications in order toallow sampler manufacturers flexibilityIn designing or configuring their PMosamplers. Sampler shape, inletgeometry, operational flow rate, degreeof automation, and other samplercharacteristics or features are specifiedonly in terms of required function orperformance.

While most of the comments receivedon Appendix I generally supported theperformance-based approach tospecifying PMo reference methods,many commentors felt that the samplerperformance specifications in theproposed Appendix J and 40 CFR Part 53were not adequate to ensure accuratecollection of PM10 under all conditionsof ambient sampling. In response to suchcomments, the sampler performancespecifications in Part 53 and thecorresponding references to suchrequirements in Appendix I have beenrevised. Other comments were receivedon various requirements of Appendix Jsuch as flow calibration andmeasurement, flow regulation, filtermedia, filter equilibration, and samplermaintenance. Specific changes toAppendix J resulting from thesecomments and from review of otherpertinent information are discussedbelow.

A. Specific Changes to Appendix ISection 3.0 has been revised to specify

that all samplers should be capable ofmeasuring 24-hour PMjo massconcentrations of at least 300 pgg/m3while maintaining the operating flowrate within specified limits.

In Section 4.0 the term"reproducibility" has been changed to"precision" and the specification forPMo samplers has been changed from15 percent or better to 7 percent or 5 g.g/M, whichever is higher. The particlesize for 50 percent samplingeffectiveness in Section 5.0 has beenchanged from 10± 1 micrometers to10±0.5 micrometers. These changes area result of corresponding changes in thePM1o sampler performancespecifications in 40 CFR Part 53,promulgated elsewhere in today'sFederal Register. Refer to the Part 53action for further discussion of thesechanges.

In Section 6.0 the subsection onnonsampled particulate matter has beenremoved. The design of particle sizediscriminating inlet systems for PM osamplers essentially precludes thetransport of windborne particulatematter to the particle collection filter-during periods when the sampler is idle..'Although windborne particles couldpotentially enter a PMo sampler's airinlet opening during idle periods, theywould have to take a tortuous path withseveral changes in direction to reach thecollection filter.

References to "automatic flowcontroller" throughout Appendix J havebeen changed to "flow control device".The latter term is less restrictive andmore clearly allows the use of any typeof flow regulation device, provided thatthe other flow-related requirements ofAppendix J are met. In particular,Section 7.1 has been changed to requirethat a PM o sampler have a flow controldevice capable of maintaining thesampler's operating flow rate within thelimits specified for the sampler inlet.The requirement that the flow controldevice have aflow rate adjustmentcapability has been removed to allow.for the use of certain types of flowcontrollers (e.g., Venturi-type criticalflow devices) that regulate flow at aconstant but unadjustable rate. Flowcontrollers of this type generally employa fixed-geometry orifice and control thesampler's flow rate without any movingparts or electronic components. Therequirement that the flow control devicebe disabled during calibration has also:been removed because it is, onlyapplicable to certain types of devices(e.g., electronic flow controllers).Sampler-specific operational

requirements such as this are betteraddressed in the sampler manufacturer'sinstruction manual.

Subsection 7.1.6 has also beenchanged to explicitly require that theinstruction manual associated with thesampler include detailed procedures forcalibration, operation, and maintenanceof the sampler. Since much emphasis isplaced on the role of the samplermanufacturer's instruction manual inAppendix J, it is important that itcontain detailed information on allaspects of sampler operation. Theinstruction manual for each designatedreference method would be reviewedand approved as part of the Part 53reference method designation process.

The filter alkalinity specification inSubsection 7.2.4 has heen changed from<0.005 milliequivalents/gram of filter to<25 microequivalents/gram of filter. Inaddition, the method used for thealkalinity determination has beenchanged to a newly developed, moresensitive, and more reliable method. Thechange in the magnitude of thespecification results from the change inprocedures (alkalinity measurementsare approximately 5 times higher withthe new method), and from the changein the measurement units.

:Section 7.3 includes specifications andother requirements for the flow ratetransfer standard used during samplercalibration. The specifications for thereproducibility and resolution of theflow rate transfer standard have heenremoved and replaced with an accuracyspecification. The revised Section 7.3requires that the flow rate transferstandard be capable of measuring thesampler's operating flow rate with anaccuracy of ±2 percent. An accuracyspecification, stated in this context, ismore meaningful and useful thanspecifications for reproducibility andresolution. In addition, the requirementthat the flow rate transfer standardinclude a means to vary the sampler'sflow rate during calibration is notappropriate for all types of samplersand/or flow rate transfer standards andhas been removed, This is anotherexample of a sampler-specificrequirement that is better addressed inthe sampler manufacturer's instructionmanual.

The humidity requirement for the filterconditioning environment in Section 7.4has been changed from a singlespecification of 45±5 percent relativehumidity (RH) toseparate specifications:for humidity range (20 percent to 45percent RH] and humidity control (±5percent RH). Under the revisedrequirements, filters may be equilibratedat any.preselected humidity betWeen20

Z4647


and 45 percent RH, provided that thehumidity is controlled to within 5percent RH. Language has also beenadded to Section 9.0 to require that thesame temperature and humidityconditions be used for both pre- andpost-sampling filter equilibration.

The calibration and operationalprocedures for PM 0 samplers varyconsiderably depending on the type ofsampler (e.g., high-volume, medium-volume, low-volume)and the type offlow control and flow measurementdevices employed in the sampler.Accordingly, the calibration andprocedure sections of Appendix J(Sections 8.0 and 9.0) have been revisedsubstantially to be more general innature. The revised procedures serve toillustrate the steps involved in thecalibration and operation of a PM~osampler, and place more emphasis onthe sampler manufacturer's instructionmanual and the Quality AssuranceHandbook for specific guidance.*A new section on sampler •maintenance has been incorporated'intoAppendix J to explicitly, require thatPM1o samplers be maintained in strictaccordance with the proceduresprovided in the sampler manufacturer'sinstruction manual. The performance ofsome PM1o samplers may be adverselyaffected by the buildup of substantialquantities of non-PMo particulatematter within the sampler inlet. Suchsamplers may require periodic cleaningand other maintenance to ensureaccurate collection of PMo particulatematter. This new section has beenadded as Section 10.0. and thecalculations and references sectionshave been renumbered accordingly.

When temperature and pressurecorrections to sampler flow indicatorreadings are required, corrections basedon existing temperature and pressure atthe time the readings are taken (or dailyaverage values during the samplingperiod in some cases) are preferable.However, incorporation of site orseasonal average temperatures andbarometric pressures into the samplercalibration to avoid daily temperatureand pressure corrections is alsoallowed. When temperature andpressure corrections to flow indicatorreadings are required, existingtemperature and pressure at the time thereadings are taken (or daily averagevalues during the sampling period insome cases) must be used. Likewise, thecalculations section has been changed.to require that the average barometricpressure and average ambienttemperature during the sampling periodbe used to calculate Qtd. Site or .seasonal average values for temperature

and barometric pressure may berequired in the adjustment of the set-point of certain types of flow controldevices (e.g., mass flow controllers). Siteor seasonal average values fortemperature and pressure are used inthese cases to ensure that the deviationsin actual volumetric flow rates, resultingfrom daily changes in temperature andpressure at the monitoring site, arecentered about the sampler inlet'sdesign flow rate.

Other minor wording changes havebeen made throughout Appendix J toclarify the requirements.B. Designation of Reference Methods forPMo

Before a method for PM1o is approvedas a PM1o reference method, it mustmeet the requirements of Appendix Jand be tested and designated as areference method In accordance withthe provisions of 40,CFR Part 53. Testingof candidate reference methods willgenerally be conducted by the samplermanufacturers. A notice will bepublished in the Federal Register in,accordance with Part 53 whenever anapplication for a PM1o reference methoddetermination is received by EPA.Likewise, a notice of designation andother information pertinent to thedesignation will be published in theFederal Register each time a PMoreference method is approved for use.PMo sampler manufacturers arerequired to provide sampler purchaserswith an operation or instruction manualcontaining detailed procedures for thecalibration, operation, and maintenanceof the sampler. Additional guidance andrecommendations regarding filter media,type of analytical balance required formass determinations, and otherrequirements of the method should alsobe provided In the manual. Part 53requires submission of the manual aspart of a manufacturer's application fora reference method determination. Theinstruction manual will be reviewed fortechnical accuracy and consistency withthe requirements of Appendix J andmust be approved as part of therequirements for designation of themethod as a reference method.

C. Technical Change to Appendix GBecause the high-volume method

described in Appendix B will continueto be used in conjunction with AppendixG ("Reference Method for theDetermination of Lead in SuspendedParticulate Matter Collected fromAmbient Air") and for other purposesthat may be specified, EPA ispromulgating the technical changes toAppendix G as proposed. Under thefinal rule the reference 10 in* Appendix G

has been deleted and section 5.1.1 of theAppendix has been revised to read asfollows:

"High Volume Sampler. Use andcalibrate the'sampler as described inAppendix B to this Part." The Appendixhas also been revised to specify moredirectly that the high-volume methoddescribed in Appendix B is to be used inconjunction with the reference methodfor lead.

VI. Summary of Salient PublicComments and Agency Responses

An overview of public comments onthe major aspects of the March 20, 1984proposal are presented in Section II. Themost important comments on specificissues are categorized and summarizedbelow together with Agency responses.A more comprehensive compilation ofcomments and Agency responses iscontained in a separate Response toComments Document that has beenplaced in the Docket,(No. A-82-37).

A. Health Effects Criteria and Selectionof the Primary Standards

1. Indicator for the Primary Standards

Comments: PMe rather than PMoshould be used as the indicator for theprimary standards because PM. moreaccurately reflects particle deposition inthe thoracic regions, provides an amplemargin of safety in protecting health,and puts less emphasis on coarseparticles that are relatively inert thandoes PMo.

Agency Response: EPA considered themajor analysis (Swift and Proctor, 1982)and preliminary arguments (AMC, 1982)in support of a PMs indicator indeveloping the 1984 proposal. AlthoughEPA deferred judgment pendingadditional analysis and review, thedecision to propose PM1 o and not PM.was based, in part, on reservationsconcerning the PMs indicator. Thelikelihood that the available data frommouthpiece studies overstated thoracicdeposition during "natural" breathingwas recognized in a qualitative sense byCASAC (cf. July 1981 transcript, p. 581;Docket No. A-82-37) and presented asone reason for recommending PMorather than PM1. or TSP as an indicator.The 1982 staff paper reflected thisargument in recommending 10 prm ratherthan 15 ftm as the cutpoint for theindicator (SP, pp. 76-77). The criteriadocument addendum points out thatassumptions used in the quantitativeanalyses used to support PM. (Swift andProctor, 1982) appear to underestimatethoracic particle deposition; thisunderestimation would reduce anymargin of safety associated with an


indicator derived from these data.Extension of the Swift and Proctoranalysis itself suggests thatapproximately 10 to 20% of 10 pmparticles could penetrate to the thoracicregion, rather than the 0% penetrationimplied by some commenters whoargued for a "Do" at 10 pm.

The Swift and Proctor analysis as wellas several more recent analyses andexperimental studies of particledeposition are reviewed in the criteriadocument and staff paper addendum.The more recent assessments tend tosupport the original proposal of PMIo.The criteria document addendumcompares the work of Miller et al. (1986),using the more recent deposition data,with the Swift and Proctor analysis andconfirms that the latter understatesdeposition of particles larger than 6 pmin individuals who habitually breathethrough the mouth.

The more recent data also show somefraction of particles of 10 pm and largercan penetrate as far as the alveolarregion (CDA, Figure 2-1). The riskassociated with deposition of insolublecoarse particles in this region is ofparticular concern because of slowclearance time (CDA, p. 2-6). Althoughremoval in the tracheobronchial regionis more rapid, deposition of coarseparticles in the tracheobronchial regionmay be associated withbronchoconstriction and alteration ofclearance mechanisms (SP, Table 5-2).The 1982 staff paper took these factorsinto account in the originalrecommendation for a 10 pm indicatorthat included all of the fine and aportion of the coarse fraction.

After considering these updatedassessments, the EPA staff reaffirmedits original recommendation of PMo asan indicator for the standards (SP, p. 32).In reviews of the March 20,1984proposal and of the criteria documentand staff paper addenda, the CASACalso reaffirmed its recommendation forPMo as an indicator (Lippmann 1986a,c). The majority of public comments onthis issue also favored PMo.

In summary, EPA finds that thepresently available record clearly favorsthe PMo indicator over the alternativePMK indicator.

Comments: Some commenterssuggested that while PM,0 represents animprovement over TSP, the fine fraction(<2.5 pm) is of relatively greaterconcern to health than the coarsefraction (2.5 to 10 pm). Such commenterssuggest that a PM2 .5 standard isneeded-in addition to or, in somecomments, instead of a PM,0 standard.

Agency Response: The possibility of afine particle indicator for the primarystandard was examined in the staff

paper (pp. 68-70). This suggestion isbased in part on the recognition thatambient particle mass and volume aredistributed such that a rough division..minimum" at about I to 3 pm separatesthe "fine" (smaller) and "coarse"fractions. Each fraction has somewhatdistinct chemical and physicalproperties and sources. The staff,however, noted a number of difficultiesin using fine particles (less than anominal 2.5 pm) alone instead of PMoas the indicator for the primarystandards. These include:

(1) Substantial overlap can occurbetween the two modes and in somecases the division minimum candisappear. Moreover, despite thediffering origins and chemistries of themodes, each is chemicallyheteorgeneous. The respiratory tract, ineffect, alters the ambient distribution,with a mixture of fine and coarse modesbeing deposited in both thetracheobronchial and alveolar regions.Indeed, the 2.5 pm !'cut" is within thesize range of maximum efficiency foralveolar deposition (2 to 4 pm). Themixing of these size fractions in therespiratory tract and the heterogeneitywithin each fraction therefore blurs thedistinction between the fractions interms of health effects.

(2) Coarse dusts have been associatedwith responses such asbronchoconstriction, altered clearanceand alveolar tissue damage (SP, Table5-2). Given current Information, it wouldbe premature to ascribe all of the effectsin the British, U.S., and otherepidemiological studies to the finefraction, or to any single chemical entitywithin that fraction.

EPA believes that a separate fineparticle standard in addition to the PMostandard is not warranted for thefollowing reasons:

(1) Fine mass typically comprises onthe order of 40 to 70% of PM,0 .Therefore, the PM,0 standards providesubstantial limits on fine mass, and

(2) The limited epidemiological datapresently available must provide theprincipal basis for any particulatematter standard. Because these data donot separate the effects of fine andcoarse fractions, it is most reasonable touse these data to support a single set ofstandards.

(3) To the extent that emerginginformation suggests additionalprotection may be necessary, it may bemore appropriate to consider theaddition of chemical-specific (e.g., acidaerosols) standards rather than a fineparticle standard in future primarystandard revisions.

2. Interpretation of CommunityEpidemiological Studies

Comments: A number of commenterstook issue with EPA's interpretation ofthe various analyses of Londonmortality data. These commenterssuggest that (a) the London data can beused to show only an association ofexcess mortality with highconcentrations of pollution duringunique episodes in which BS and SOlevels exceeded 500 to 1000 pjg/m3, (b) anumber of the analyses suffer frommethodological flaws precluding validconclusions, (c) the conclusion thateffects may be possible at low pollutionlevels (e.g., <250 pg/m s) or that there isa continuum of association with noidentifiable threshold is not supportable,(d) the results of Mazumdar et al. (1982)and Ostro (1984) are more consistent:with the hypothesis that particulatematter is acting as a surrogate for someother causal agent rather than as acausal agent itself, and (e) it isbiologically implausible that mortalitycould be affected by particulate matterat levels below those shown by Lawtheret al. (1970) to produce morbid effects in

,sensitive populations.Agency Response: EPA's'assessment

of the various London mortality analysisIs discussed at length in the criteriadocument, the staff paper, and theaddenda to these documents. The 1982criteria document found that in thecontext of historical London exposures,these data indicate clear increases indaily mortality occurred with BS andSO2 concentrations in excess of 1000pg/m with some indications of likelyincreases in daily mortality at levels ofboth pollutants in the range of 500 pg/M3 or more (CD, Table 14-7). Theseoriginal conclusions on likely effectslevels, based largely on the Martin andBradley (1960) and Ware et al. (1981)analyses, appear reasonably consistentwith the original assessment of thesedata by the original British investigatorsand the 1969 criteria document. From there-examination of these data by Ware etal. (1981) and the analysis of subsequentLondon winters by Mazumdar et al.(1981), the criteria document alsoconcluded small increases in dailymortality might occur at levels below500 pg/m s . The more recent analyses ofthese data by Mazumdar et al. (1982),Ostro (1984), and Shumway et al. (1983)all serve to reinforce the possibility thateffects were associated with particulatematter at concentrations below 500 jLg/in s . A number of commenters, however,including some of the original Britishinvestigators (Holland et al., 1985),object to this latter suggestion.

24649


EPA has carefully examined thesestudies and the various criticisms ofthem submitted as comments on theproposal. In order to respond fully tothese criticisms, EPA conducted moresophisticated reanalyses of the originalLondon data to further determine thedegree of reliance that can be placed onthe published results (Schwartz andMarcus, 1986, CDA, Appendix A). Eachof these studies does suffer fromlimitations and uncertainties delineatedin EPA's updated assessment (SPA pp.17-23; 39-44); these limitations precludedefinitive conclusions withrespect tocausality as well as identification ofclear "no observed effects" levels.Nevertheless, EPA maintains its originalinterpretation, supported by its externalscience advisors, that these data at leastsuggest the possibility of effects ofparticulate matter at BS levels as low as150 pg/m 3 and possibly even lower.None of the difficulties in statisticalmethodology or alternative mechanismscited by commenters provide anadequate explanation for the consistent.finding of association betweenparticulate pollution and mortality atlevels below 500 g/m33 (as BS). Theassociation was found for the majorityof 14 winters (analyzed individually)spanning a period when pollution inLondon and indoor heating practicesshowed marked changes, and includingwinters in which BS levels did notexceed 250 pg/m s . The relativeconsistency of the results from year-to-year despite these changes suggests thatthe observed effect is not explained byindoor air pollution or by long-termdemographic shifts in the population.The findings were consistent amongdifferent investigators, and persistedafter taking SO2,.temperature, and otherweather variables into account, andafter correcting for autocorrelationstructure.,

The principal arguments for thesuggestion by some (includingMazumdar et al., 1982) that smoke maybe acting as a surrogate for some moretoxic pollutant or related non-pollutionvariable are: (1) The coefficients in theregression equations appear to increasewith decreasing pollution across the 14winters, (2) surrogate behavior iscommonly observed in statisticalanalyses, (3) the work of Lawthersuggests a threshold for morbidity ataround 250 jg/m as BS; hencemortality would not be expected atlower levels. While the possibility ofsurrogate behavior remains, the abovearguments do not demonstrate thatsmoke acts as a surrogate for non-pollution variables. The trend towardhigher coefficients with lower pollution

is not clearly consistent in theMazumdar and Ostro regressions. Theexistence of higher coefficients in lateryears, however, prompted these authorsto suggest some plausible alternative tonon-pollution surrogates, including: (a)The possibility that the composition ofpollution changed with time, with anincrease in more toxic components, and(b) because the gravimetric mass ofparticles in the range under 10 jum maynot have declined as much as did theblack carbon content detected in thesmoke measurement (Lodge, 1986),coefficients related only to smoke mightbe expected to increase. An additionalpossibility suggested by Schwartz andMarcus is that the effect of higherpollution episodes in earlier winters wasblunted by public awareness (and hencereduced exposure) or by a tendency forthe most susceptible individuals tosuccumb on an early day of a multi-daypollution episode.

The use of the Lawther morbidity dataas a threshold for mortality isquestionable. The London mortality datainvolve an unequivocal endpoint in arelatively large population (severalhundred per day) over a 14 year period.As pointed out by Roth et al. (1986),although the bronchitic populationstudied was clearly susceptible, theeffects indicator used by Lawther was arelatively insensitive one. Moreover, thethreshold was determined not byrigorous analysis, but by visualexamination of strip chart data.Although the principal author stronglyobjects (Lawther, 1982), the criteriadocument points out that the data do notclearly indicate an effects threshold at250 ;g/mg. Furthermore, the simplecorrelation results provided by Lawtheret al. (1970) suggest the possibility that amore sophisticated analysis jointlyincorporating pollution and weatherfactors might have found increasedmorbidity occurring at lower levels. Therecent findings of small changes inpulmonary function at lower particulatematter levels in the U.S. and theNetherlands (See Table 1) support thenotion that 250 pg/m s (in this case asPMo) is not a reliable effects threshold.

Comments: The derivation of theproposed range of levels for the annualprimary standard is without scientificbasis. In particular, limitations in thetwo major series of studies usedpreclude finding effects of particulatematter at the lower TSP levels shown. Inaddition, the conversion of the results ofthese studies to PMzo uses aninappropriately low PM1o/TSP ratio.

Agency Response: EPA's assessmentof studies used to derive the range oflevels for the primary standard (Ferris et

al., 1973,1976;) and Bouhuys et al., 1978)(CD, pp. 14-44 to 46 and SP, pages 61-62and 104 -107) was reviewed by CASACand found to be an appropriate basis fordeveloping revised standard levels(Friedlander. 1982). The assessmentclearly points out the limitations andstrengths associated with the uses ofthese studies.

The Ferris et al. work (See Table 2above) involved a "longitudinal"tracking of lung function and respiratoryillness in adults vs. pollution over a 12year period in Berlin, NH, a small townin which a pulp mill was a majorpollution source. As commenters note,the "effects likely" level drawn from thefirst year of this study is particularlyuncertain, as it is based on very limitedaerometry. This level, however, was notimportant in developing the range forthe proposed standard. Because of theseriousness of the effect (a prolongeddecrement in lung function), the by thendecreased concentration observed in thefirst followup study (130 p.g/m s as TSP).was used in developing the upper boundof the range of proposed annualstandards. This concentration wasbased on a full year of monitoring.Based on the historical record, there canbe little doubt that pollution declined inthis community from 1981 to 1967, theyear of the first follow-up. The nature ofthe particular pollution source (a pulpmill) in this study, together with afinding of very low British smoke level.indicates that a variety of particles, notjust products of combustion, may be.associated with adverse effects.Although commenters have suggestedthat other pulp mill emissions may havebeen responsible for the effects, ambientlevels of the gaseous effluents from suchsources (reduced sulfur compounds andSO2 ) have not been shown to causereduced lung function.

Estimating PMo levels from this studyby using typical national average PMo7TSP ratios does not-as somecommenters argued--clearly understatePMo levels. These commenters arguedthat high PMo/TSP ratios (e.g., 0.8)should be used because sites in theeastern U.S. tend to have higher ratios.The data on PMo/TSP ratios, however,also show a general tendency for lowerratios to occur in industrialized areaswith high TSP concentrations (Pollack,et al., 1985). Moreover, air qualitymeasurements taken in the 1960'sdocument the presence of substantialquantities of larger size particles, asevidenced by high dust fall levels andlow soiling indexes (Kenline, 1962). Thelatter author concludes that this wouldbe expected "if the majority of particlespresent had diameters of 10 microns or


greater . . . ... EPA therefore believesthat the use of ratios characteristic ofindustrialized areas with high particleconcentrations is justified and does notcontribute to any excess margin ofsafety in the estimated effects levels.

The Bouhuys et al. (1978) study (seeTable 2 above) was used to set thelower bound for the proposed standardrange, which is the level at which thefinal standard is being promulgated. Thestudy found a difference in three of fiverespiratory symptoms but no differencesin lung functions between twoConnecticut towns (Ansonia andLebanon) that had a historically large(but currently small) difference in levelsof particulate matter. Although theauthors believed that air pollution didnot play a role in the observeddifferences in symptoms, the datapresented do not demonstrate that thedifferences were due solely to otherfactors associated with the conduct ofthe study. Moreover, the finding ofexcess respiratory symptomsunaccompanied by a persistent changein lung function is not unique. Similarfindings were also obtained in the Ferris(1973) follow up study and the morerecent six city study results (Ware et al.,1986).

Some commenters argued that theestimated TSP levels derived for theBouhuys study were too low. EPAdisagrees. The staff took the medianTSP values reported by Bouhuys et al.over the previous several years as therelevant exposure level for this studybecause (1) the current gradient inpollution appeared to be too small toresult in such effects, and (2) it isunreasonable to attribute all of theobserved gradient in effects amongurban and rural residents, as measuredin 1973, to the maximum historicalconcentrations reported 8 to 10 yearsprior to that time. EPA's position issupported by the observations of Ferriset al. (1973, 1976), which show anapparent measurable reduction insymptoms and improved lung functionafter only a five to six year decline inpollution. This decline suggests that anygradient in effects due to pollution eightto ten years ago would be diminishedrelative to effects that may beassociated with the more recent past.The median value used by EPA for theBouhuys study is, In fact, also relativelyclose to the weighted average of all TSPobservations reported for Ansonia forthe seven years preceding the Bouhuyset al., (1978) measurements, which weretaken in 1973 (Lounsbury, 1986),

The approach used to convert the TSPmeasurements in this study to PMoequivalents was also questioned. The

staff rejected use of the limited (15 days)particle size data for Ansonia asunrepresentative because of questionsconcerning their quality and becausethey were taken in 1973, afterparticulate matter concentrations hadbeen reduced to lower levels (SP, p. 62).Absent reliable site-specific particle sizedata, the staff used the median PMo/TSP ratio seen at other sites in theeastern U.S. with higher than averagePMo levels. Because the long-term ratiocan vary between 0.3 and 0.65 amongsuch sites, such estimates areadmittedly uncertain. Nevertheless, thestaff examination of historical airquality and source data associated withthe Bouhuys et al. study found nofactors that would make the ratiounusually high or low relative to otherhigh concentration sites In the easternU.S. The analysis by Spengler et al.(1986) of trends in particle size ratiosfrom the 1970's to the present in sixeastern cities suggests that the ratio ofPMo to TSP in early years with higherTSP levels tends to be comparable to orsomewhat lower than the current ratios.

The basis for the final ambientstandard is considerably strengthenedby the recent results from the six-citiesstudy (Ware et al., 1986). This work alsosuggests an increased risk of respiratoryillness and symptoms, but no differencesin lung function, in children across agradient of pollution that extends toconcentrations below those observed inthe previous studies. The results aretherefore qualitatively consistent withboth of the earlier studies. In addition,the associated aerometry permitssubstantially better estimates ofhistorical PMo data. Taken together,these studies provide substantialsupport for an annual standard of 50 jg/M

3.

3. Margin of Safety

Comments: The Agency hasincorporated an unrecognized three-foldmargin of safety in the 24-hourstandards through the means used toconvert British Smoke measurementsinto PMo.

Agency Response: British Smokemeasurements collect particles smallerthan about 4.5 microns in diameter(PM4 .0 ) on a substrate and then measuretheir absorption of light. Because themeasurement depends on lightabsorption, it is sensitive only to thedark, "sooty" component of theparticulate matter. EPA has relied ongravimetric calibrations, performedduring the earlier years of the mortalityand morbidity studies, that related theBritish Smoke measurements topartidulate mass concentrations that

included light-colored as well as darkparticles.

The commenters note that the dark,sooty component of the particulatematter in London today constitutes only40% as large a fraction of the totalparticulate mass as it did during theperiod of the studies on which EPA hasrelied. They argue that the use of thosestudies to set standards forcontemporary particulate pollutiontherefore Introduces an error of a factorof 2.5 (1/0.4). Multiplying this by atypical ratio of PMo to PMs of 1.2(Lodge, 1986), the commenters arrive atan alleged error of a factor of threearising from the Agency's use of theBritish Smoke measurements.

The commenters rely on the unstatedassumption that it is only the darkfraction of particulate pollution thataffects human health, and that, since thedark fraction has declined since the timeof the studies, the particulate matter inthe atmosphere today is less dangerousthan that present at the time of thestudies. EPA disagrees with thisassumption and believes that a moreplausible and prudent assumption is thateffects on health depend on the massconcentration of particles and not ontheir color.

Although it is possible that dark,carbonaceous particles were primarilyresponsible for the observed effects onhuman health in the London studies, thishas not been documented, and there isno evidence to support the assumptionthat light-colored particles have nosignificant effect on human health. EPAstaff has compared the composition ofparticulate matter in historical Londonand in the current U.S, and hasconcluded that, given the variety ofparticle types present in the U.S., thereis no clear basis for imputing higheracute toxicity to the historical Londonparticles (SP pp. 21-22, 100).

The commenters support theirargument with the assertions that thedecrease in the dark, sooty fraction ofparticulate matter in London has beenaccompanied by the elimination ofpollution-related health effects, and thatcurrent excursions of fine particle massin excess of 250 pg/m 3 have not beenassociated with health effects in Londonor elsewhere. EPA finds these assertionsto be unsupported. The studies ofmortality in London over a 14-yearperiod of declining pollution from 1958through 1971 found that the relationshipbetween pollution and mortalitypersisted throughoutthe period and that,in fact, the regression coefficientsassigned to mortality appeared toincrease over the period. (Mazumdar etal., 1982; Ostro, 1984). Moreover,

24651I

I


continuing studies In the contemporary revised form, in particular, makes directU.S. and Europe have suggested health comparison of the relative stringency ofeffects at PMo levels below 250 Pg/m3 proposed range with the current TSP(Dockery et al., 1982; Ozkaynak and standard inappropriate.Spengler, 1985; Dassen at al., 1986). EPA agrees that the analyses of

For these reasons, EPA concludes that mortality In London justify caution Init is reasonable and prudent to use the selecting a 24-hour standard level, andmass concentration estimates derived that the recent studies of lung functionfrom historical British Smoke provide a useful basis for selecting themeasurements to set ambient standards level. EPA does not, however, believefor current U.S. atmospheres under the that these studies compel a standardassumption that current U.S. particles more stringent than the one chosen, Asare equal in toxicity to those found in discussed in Section III.C.1 above,London at the time of those uncertainties in estimating PMlomeasurements. Any margin of safety equivalents of low British Smokeinherent in the British Smoke/PMo concentrations in the later years of theconversion for the earlier years when London studies make it difficult to usegravimetric calibrations were available the studies to set a precise level for ais more likely to be on the order of a PMo standard. Therefore, it is importantfactor of 1.2 (the ratio of PM. to PMo to examine the more contemporaryestimated by Lodge, 1986) rather than studies of lung function that permit athe factor of three suggested by the more direct estimation of PM1o effectscommenters. For particulate levels lower levels. In considering these studies inthan those observed in the e'arlier years, conjunction with the London mortalityEPA has supplemented the London and other relevant health studies, EPAstudies with the more contemporary finds that a 24-hour standard of 150 pg/American and European studies using m3 provides an adequate margin ofdirect gravimetric measurements. safety. EPA does not agree withComments: Several commenters bommenters suggestions that it isexpressed concerns that the margin of necessary to prevent any detectablesafety for the range of levels proposed changes in lung function. As discussedfor the 24-hour standard is insufficient. in Section III.C.1, a standard of 150 1kg/Commenters based these concerns on: ms will clearly prevent lung function(a) Calculations suggesting that even the decrements that might be considered tolower bound may be less stringent than be indicative of adverse effects in wellthe current standards, (b) evidence from over 95% of children exposed; in fact thethe more recent studies of lung function evidence suggests that even reversibledecrements in children and the analyses lung function changes (FEVo.75) in excessof London mortality data, and (c) of 10% are unlikely at this level. EPAvarious studies found to be mainly of therefore believes that the standardqualitative value. In general, such provides an adequate margin of safety.commenters felt that, in view of the Some commenters favoring standardsavailable evidence, the standard should below the lower bounds of the proposedbe set at levels at or below the lower ranges relied on studies or analysesbound of the proposed ranges. found by EPA and CASAC to be of little

Agency Response: The overriding quantitative value for establishingconsideration in selecting a standard is ranges of concern. EPA considered ahow well it protects public health, not number of such studies in selecting aits relative stringency as compared to margin of safety (e.g., SPA 52-53; SPthe previous standard. EPA believes that 109-111), but in EPA's judgment they dostandards chosen provide an adequate not provide a sufficient basis formargin of safety irrespective of the establishing standards at levels belowrelationship to the former TSP those derived from the morestandards. Nevertheless, EPA has quantitative studies summarized incompared the stringency of the revised Tables 1 and 2 above.standards with that of the existing Comments: Some commenters arguedstandards by estimating the number of that in selecting annual standards muchareas that would be expected not to greater weight be given to the results ofattain each set of standards. By this Ware et al. (1986), which suggest ameasure, the new PMo standards are possible gradient of effects atequivalent to or somewhat more concentrations extending to the loweststringent than the TSP standards (SP, levels observed in the six cities studiedTable 2-1). Commenters who calculated (25 jg/m3).or asserted otherwise often did not take Agency Response: EPA disagrees.all of the aspects of the standards into EPA staff found that the pollution andaccount. The margin of safety is a effects gradient in the three cleanestfunction not only of level, but also of the cities to be too small to provide anyindicator and form of the staiidards. The - -strong suggestiQn of effects at such

levels. Moreover, the lack of consistencyfor "within city" effects in this studyargue against placing undue reliance onthe suggestion of effects at levelsoutside of the range suggested by theother long-term studies of interest(Ferris et al., 1973,1976, Bouhuys et al.,1978). In addition, the 24-hour standardprovides an increased margin of safetyagainst annual exposures at levelsbelow 50 jg/m3, in areas where long-term exposures are dominated byrepeated short-term Peaks (Freas, 1986).

B. Secondary Standards

1. Soiling and Nuisance

Comments: The Agency shouldmaintain a secondary TSP standard.Some commenters felt that the proposedsecondary annual TSP standard isinadequate, and that the current 24-hourTSP standard should be retained.

Agency Response: As discussed inSection IV.A. above, the CASAC foundlittle scientific support for maintaining asecondary TSP standard. It follows thatlittle data exist to support maintainingthe present level or an alternative levelfor a 24-hour standard designed to

protect against soiling and nuisance.Nevertheless, the changes made in thefinal standard result in both a 24-hourand annual secondary PMo standard.Analysis of the relative protectionafforded by the 24-hour PMo standardindicate that it is relatively morestringent than the upper portion of theproposed range for an annual TSPstandard. Thus, the final standardsshould provide more protection thanthat afforded by the proposed TSPalternative toward which theAdministrator was initially inclined. Asdetailed above, the data do not provideconvincing evidence of significantsoiling and nuisance effects atconcentrations below that permitted bythe primary standards.

2. Visibility

Comment: A secondary fine particlestandard is needed to protect againstvisibility impairment and related effects.

Agency Response: The Administratordeferred judgment with respect to asecondary fine particle standard inorder to examine the relationshipbetween control programs for regionalvisibility and the related problems ofacid deposition. The initial phase of thatexamination has now been completed(EPA, 1985). Based on the availableinformation, the Administrator hasdecided to issue an Advance Notice ofProposed Rulemaking on a secondaryfine particle standard in a separatenotice in today's Federal Register.

Federal Register /.Vol. 52, No. 126 / Wednesday, July 1, 1987 / Rules and Regulations

C. Averaging Time and Form of theStandards

1. Expected Exceedances for the 24-hourStandard

Comment: Several commenters wereopposed to the proposed statistical formand either favored the current simplerdeterministic form or preferred amultiple exceedance or percentile formof the 24-hour standard. Others.supported the proposal to adopt a singleexpected exceedance statistical form.Many of the opposing commenters wereconcerned that the adjustment forincomplete sampling could cause areaswith less than one actual exceedanceper year to be misclassified asnonattainment and that the method issensitive to spurious highconcentrations. Those in favor ofadopting a single exceedance statisticalform recognized the need to account formissing data and argued that this formprovides proper health protection..Agency Response: EPA has carefullyreviewed these comments and hasdecided to maintain the basic proposedstatistical form for-the 24-hour standardbut has made some technical changesand clarifications in response toreviewers comments. The Agencybelieves that a single exceedance formfor the primary standards and theproposed adjustments for incompletesampling appropriately reflect the healthbasis for the standard. When sampling,is performed less frequently than everyday, the number of observedexceedances of the standard level willobviously be, in general, fewer than theactual number of exceedances. If, forexample, sampling is performed onlyevery sixth day, as is permitted by theAir Quality Surveillance regulations (40CFR Part 58) being promulgated today,then, on average, the number ofobserved exceedances will only be one-sixth of the actual number ofexceedances. To fail to correct for thiseffect would be irrational and wouldseriously degrade the health protectionafforded by the standards. The Agencybelieves that adequate procedures forhandling spurious high concentrationsare provided in the "Guideline on theIdentification and (Use of Air QualityData'Affected by Exceptional Events",EPA-450/4-86--007. Moreover, singlehigh concentrations will not necessarilycause a location to fail the test forattainment. Appendix K has beenmodified so that the first observedexceedance is not adjusted forincomplete sampling, if the samplingfrequency is promptly increased toeve'y day in accordance with 40 CFRPart 58.13. Accordingly, sites samplingonce in six days must observe at least

two exceedances in order to fail the testfor attainment.' Sites sampling every.other day or every day must recordthree or four exceedarices over a three-year period in order to fail the test. Thischange reduces the chances formisclassifying a site as, nonattainment.

Although a multiple exceedande formof the 24-hour'standard could reducesampling requirements, such a formwould reduce the level of healthprotection by allowing particulate levelsto exceed, on multiple days, the levelsthat the Administrator has determinedto pose an unacceptable health risk. Ananalysis of alternative numbers ofexceedances found that, in the long run,the single exceedance form providedmuch more consistent health protectionthan did the percentile formrecommended by some commenters(Biller, 1984; 1986).

In response to comments regarding'the potential for seasonal variation inparticulate matter concentrations, aswell as possible intrayear changes insampling frequency as described in Part58 of this Chapter, the Agency hasdecided to require that adjustments forincomplete sampling be performed on aquarterly basis instead of a yearly basis.

2. Expected Arithmetic Mean for theAnnual Standard

Comment: Many commenters favoredretaining the geometric mean to describeannual average particulate matterconcentrations but several supportedthe proposed use of the arithmetic mean.Those opposed to the proposed methodnoted that the geometric mean is a morestable statistic and is less sensitive tooccasional high readings. In addition,opposing commenters were concernedthat a change to an arithmetic meanincreases the stringency of the annualstandard and that the arithmetic meandoes not properly relate to healtheffects.

Response: As discussed above, EPAhas decided to adopt annual primaryand secondary standards in terms ofexpected annual arithmetic mean PMo.The Agency believes that the annualarithmetic mean is a more appropriateindicator for a long-term primary airquality standard than is the geometricmean. It provides a better estimate oftotal exposure and, with its multiple-year averaging, more appropriatelytakes into account year-to-yearfluctuations in meteorology. Asdiscussed in the rationale, the effect ofaveraging multiple years of data in orderto estimate the expected annual valueas well as the use of the arithmeticmean were both considered in settingthe concentration level of the standard.The use of the arithmetic mean does not

necessarily increase the stringency ofthe, standard level; the stringencydepends at the combination of the form,indicator, and level. Holding all. elseequal. however, the arithmetic form isrelatively more protective in areassubject to multiple elevations in 24-hourconcentrations. EPA views this as adesirable characteristic:

VII. Regulatory and EnvironmentalImpacts

A. Regulatory Impact Analysis

Under Executive Order 12291, EPAmust judge whether a regulation is a"major" regulation for which aRegulatory Impact Analysis (RIA] isrequired. At the time of the proposal, theAgency judged the proposed revisions tothe particulate matter NAAQS to be amajor 'action, and made available to thepublic a draft analysis entitled:Regulatory Impact Analysis of theNational Ambient Air Quality Standardsfor Particulate Matter-Draft (EPA,1983). The draft RIA was based oninformation developed by several EPAcontractors (inter alia., Argonne, 1983;Mathtech, 1983) and provided estimatesof costs, benefits, and net benefitsassociated with alternative standards.

In announcing the availability of thedraft RIA, the Agency stated thatneither the RIA nor the contractors'reports were considered in developingthe proposed revisions. Subsequent tothe release of the draft RIA, the publicand other governmental agencies raiseda number of questions regarding theunderlying data bases and analysesdiscussed in the draft RIA. In responseto these questions, the Agency modifiedthe costmodel used and made other,more limited, changes to the benefitsanalyses. The number and extent of thechanges were constrained, however, bythe underlying model structure and theavailable data. The Agency hascarefully evaluated the revised analysisand has concluded that despite thesignificant improvement made,fundamental questions remain withregard to certain aspects of themethodology used, particularly withrespect to the emission reduction/airquality improvement relationship whichaffects the subsequent cost and benefitcalculations. Consistent with its pastpractice, the Agency has not consideredthe final Regulatory Impact Analysis ofNational Ambient Air Quality Standardsfor Particulate Matter (EPA, 1986c) inreacfhing decisions on the finalstandards.

The final RIA has been submitted tothe Office of Management and Budge et(OMB) for review under Executiv'e

24653


Order 12291. Comments from OMB andEPA's responses to those commentshave been placed in the docket.

Reporting Requirements

This final rule does not contain anyinformation collection requirementssubjectto OMB review under thePaperwork Reduction Act of 1980 U.S.C.3501 et seq.

B. Impact on Small Entities

Under the Regulatory Flexibility Act, 5U.S.C. 600-612; EPA must prepare initialand final regulatory flexibility analysesthat assess the impact a proposed orfinalrule will have on small entities,which include small businesses, smallnot-for-profit enterprises, andgovernmental entities with jurisdictionover populations of less than 50,000. Therequirement of preparing such ananalysis is waived, however, if theAdministrator certifies that the rule willnot have a significant economic Impacton a substantial number of smallentities.

The national ambient air qualitystandards do not have a direct impacton small businesses or enterprisesbecause the standards themselves donot contain emission limits or Otherpollution controls. Rather, 'such controlsare contained in State implementation'plans promulgated under section 110 ofthe Act, 42 U.S.C. § 7410. The States aregiven considerable discretion inselecting a mix of controls to attain andmaintain the ambient standards, and theimpact on small entities depends onhow the States choose to exercise theirdiscretion.

Nonetheless, EPA conducted ananalysis of the impact of a hypotheticalcontrol strategy, designed to minimizecosts, on entities in the industries thatwould be most affected under thathypothetical control strategy. Thatanalysis, discussed in the notice ofproposed rulemaking, 49 FR at 10422,indicated that less than 20% of theentities in those industries would beaffected by the proposed standards.

During the public comment period,EPA received no comments on theregulatory flexibility analysis. On thebasis of that analysis, the Administratorcertifies that the revisions beingpromulgated today will not have asignificant impact on a substantialnumber of small entities..

VIII. Other Reviews

* This final rule was submitted to theOffice of Management and Budget(OMB) for review. Comments from'OMBand.EPA's responses to these comments:have been placed In the docket. , ,

List of Subjects in 40 CFR Part 50

Air pollution control, Carbonmonoxide, Ozone, Sulfur oxides,Particulate matter, Nitrogen dioxide,Lead.

Dated: June 2, 1987.:Lee M. Thomas,Administrator.

ReferencesAMC [American Mining Congress] (1982).

American Mining Congress Position Paperon Particle Size Issue. American MiningCongress, Washington, DC. June 17,1982.Docket *A-79-29, II-D-81a.

Argonne (1983). Costs and Air QualityImpacts of Alternative National AmbientAir Quality Standards for ParticulateMatter, Technical Support Document,prepared for U.S. EPA, Ambient StandardsBranch. Research Triangle Park, NC,January 1983. Docket #A-79-29, II-A-35.

Biller, W.F. (1984). Investigation of PM1oNational Ambient Air Quality StandardsBased on Multiple Exceedances. Preparedfor Strategies and Air Standards Division.Office of Air Quality Planning andStandards, U.S. Environmental ProtectionAgency, Research Triangle Park, NC:contract no. 68-02-3875. Docket #A-82-37,IV-A-2.

Biller, W.F. (1986). Letter to Henry C.Thomas, U.S. EPA, Office of Air QualityPlanning and Standards. Re: MultipleExceedance PM Standards. December 18,1986. Docket #A-82-37, IV-D-344.

Booz, Allen and Hamilton, Inc. (1970). Studyto Determine Residential Soiling Costs ofParticulate Air Pollution. APTD-0715, U.S.Department of Health. Education andWelfare, National Air Pollution ControlAdministration, Raleigh, NC.

Bouhuys, A., G.J. Beck. and J.B. Schoenberg(1978). Do present levels of a ir pollutionoutdoors affect respiratory health? Nature276: 466-471.

Carey, W.F. (1959). Atmospheric deposits inBritain: A study of dinginess. Int. 1. Air Poll.2:1-26.

Dassen. W., B. Brunekreef, G. Hoek. P.Hofschreuder, B. Staatsen, H. deGrout, E.Schouten, and K. Biersteker (1986). Declinein children's pulmonary function during anair pollution episode. J. Air. Pollut. ControlAssoc. 36:1223-1227.

DHEW [U.S. Department of Health,Education, and Welfare] (1969). Air QualityCriteria for Particulate Matter. U.S.Government Printing Office, Washington,DC. AP-49.

DHEW (1973). Prevalence of Selected ChronicRespiratory Conditions, United States-1970.DHEW Publication No. (HRA) 74-1511.Series 10, Number 84, Rockville, MD.

Dockery. D.W., J.H. Ware, B.G. Ferris, Jr., F.E.Speizer, N.R. Cook. and S.M. Herman(1982). Change in pulmonary function inchildren associated with air pollutionepisodes.. J. Air Pollut. Contri. Assoc.32:937-942.

'EPA [U.S. Environmental Protection Agency](1982a). Review of the National AmbientAir Quality Standards for ParticulateMatter. Assessment of Scientific and

Technical lnformation-OAQPS Staff Paper.Office of Air Quality Planning andStandards, Research Triangle Park, NC.EPA-450/5-82-0O1.

EPA (1982b). Air Quality Criteria forParticulate Matter and Sulfur Oxides.Environmental Criteria and' AssessmentOffice, Research Triangle Park, NC. EPA-600/8-82-029a-c.

EPA (1983). Regulatory Impact Analysis onthe National Ambient Air QualityStandards for Particulate Matter. Office ofAir Quality Planning and Standards.Research Triangle Park. NC.

EPA (1985). Developing Long-term Strategiesfor Regional Haze: Findings and - "Recommendations of the Visibility TaskForce. Research Triangle Park. NC.

EPA (1986a). Second Addendum td AirQuality Criteria for Particulate Matter andSulfur Oxides (1982): Assessment of NewlyAvailable Health Effects Information.Environmental Criteria and AssessmentOffice. Research Triangle Park, NC. EPA600/-8&-o20-F.

EPA (1986b). Review of the National AmbientAir Quality Standards for ParticulateMatter. Updated Assessment of Scientificand Technical Informat ion. Office of AirQuality Planning and Standards. ResearchTriangle Park, NC. EPA-450/5-86-012.

EPA (1986c). Regulatory Impact Analysis ofthe National Ambient Air Quality-Standards for Particulate Matter, SecondAddendum. Office of Air Quality Planningand Standards. Research Triangle Park,N.C.

Ferris, B.G., Jr. and D.O. Anderson (1962). Theprevalence of chronic respiratory diseasein a New Hampshire town. Am. Rev.Respir. Dis. 86:165-177.

Ferris, B.G., Jr., I.T.T. Higgins, M.W. Higgins,and J.M. Peters (1973). Chronic non-specificrespiratory disease in Berlin, NewHampshire, 1961-1967. A follow-up study.Am. Rev. Respir. Dis. 107:110-122..

Ferris, B.G., Jr., H. Chen. S. Puleo, and R.L.H.Murphy, Jr. (1976). Chronic non-specificrespiratory disease in Berlin, NewHampshire, 1967-1973. A further follow-upstudy. Am. Rev. Respir. Dis. 113:475-485.

Freas, W. (1986). Summary of PMtoProbability Simulation Results.Memorandum to John Bachmann, AmbientStandards Branch. December 10, 1986.Docket #A-82-37, IV-B-15..

Friedlander, S.K. (1982). CASAC Review andClosure .of the Staff Paper for ParticulateMatter. Memorandum to EPAAdministrator Anne M. Gorsuch, January1982. Appendix E in Docket item A-79-29,II-A-3.

Haines, J.H. (1987). Reanalysis of the Booz,Allen, and Hamilton Soiling Data, -Memorandum to John Bachmann, AmbientStandards Branch. Jan. 9,1987. Docket #A-82-37, IV-A-4.

Hancock, R.P., N.A. Esmen. and C.P. Furber(1976). Visual response to dustiness. J. AirPollution Control Association 26i54-57. -

Hausman, J.A.; B.D. Ostro, and D.A. Wise(1984). Air Pollution and Lost-Work.Cambridge, MA: National Bureau of-Economic Research, NBER Working Paperno.-1263. - - .

Federal Register / Vol. 52, No. 126 '/ Wednesday, July 1, 1987 / Rules; and Regulations

Holland,. W.W., R.11 Waller, and A.V. Swan(1983). Letter to Lester Grant. U.S. EPA,Environmental Criteria Assessment Office.July 14,1983. Docket #A-79-29, I-D-105.:

Holland, W.W., C. Florey, P.J. Lawther, andA.V. Swan (1985). Letter .to John Haines,U.S. EPA, Office of Air Ouality Planningand Standards. May 28, 1985. Docket #A-82-37. IV-D-247.

ISO [International Standards Organization](1981). Size definitions for particlesampling. Am. Ind. Hyg. Assoc. J. 42:64--68a.

Kenline, P. (1962). In quest of clean air forBerlin, New Hampshire. U.S. Department ofHealth, Education and Welfare R.A. TaftSanitary Engineering Center, Cincinnati,Ohio.

Lawther, P.J., R.E. Waller, and M. Henderson(1970). Air pollution and exacerbations ofbronchitis. Thorax 25:525-539.

Lawther, P.J. (1982). Letter to John Bachmann,U.S. EPA, Office of Air Quality Planningand Standards. February 26, 1982. Docket#A-79-29, II-D-76.

Lawther, P.J. (1986). Letter to John Bachmann,U.S. EPA, Office of Air Quality Planningand Standards, August 22, 1986. Docket#A-82-37, IV-D-319.

Lippmann, M. (1986a). Letter from MortonLippmann, CASAC Chairman, to EPAAdministrator Lee Thomas. January 2,1986.Docket #A-82-37, IV-D-315.

Lippmann, M. (1986b). Letter from MortonLippmann, CASAC Chairman, to EPAAdministrator Lee Thomas. December 15,1986. Docket #A-82-37, IV-D-339.

Lippmann, M. (1986c). Letter from MortonLippmann, CASAC Chairman, to EPAAdministrator Lee.Thomas. December 16,1986. Docket #A-82-37, IV-D-338.

Lodge, J.P. (1986). A rational basis fornational ambient air standards forparticulate matter. Aerosol Sci. Tech. Vol 5#4: 459-468.

Lounsbury, S. (1986). Long-term median TSPvalues from Ansonia, Connecticut.Memorandum to John Bachmann, AmbientStandards Branch. December 31, 1986.Docket #A-82-37, IV-B-12.

Lunn, I.E., J. Knowelden, and A.J. Handyside(1967). Patterns of respiratory illness inSheffield infant school children. Br. J. Prey.Soc. Med. 21:7-16.

Martin, A.E., and W.H. Bradley (1960).Mortality, fog and atmosphere pollution-an investigation during the winter of 1958-1959. Mon. Bull. Minist. Health Lab. Serv..19:56-73.

Mathtech, (1983). Benefit and Net BenefitAnalysis of Alternative National AmbientAir Quality Standards for Particulate,-Matter, Volumes 1-5. Prepared forEconomics Analysis Branch, Strategies andAir Standards Division, Office of AirQuality Planning and Standards, U.S. EPA,Research Triangle Park, N.C., Mathtech,Inc., Princeton, N.J. Docket #A-79-29, II-A-6.

Mazumdar, S., H. Schimmel, and I. Higgins(1981). Daily mortality, smoke and SO2 inLondon, England 1959-1972. Proceedings ofthe Proposed SO, and Particulate StandardSpecialty Conference. Air Pollution ControlAssociation, 'Atlanta. Georgia.

Mazumdar, S., H. Schimmel, and I. T. Higgins(1982). Relation of daily mortality to. Qir. !,

pollution: an analysis of 14 London winters,1958/59-1971/72. Arch. Environ. Health37:213-220."

Miller, F.J., T.B. Martonen, M.G. Menache,D.M. Spektor, and M. Lippmann (1986).Influence of breathing mode and activitylevel on the regional deposition of inhaledparticles and implications for regulatorystandards. Cambridge, United Kingdom:Inhaled particles VI: accepted forpublication.

Ostro, B.D. (1984). A search for a threshold inthe relationship of air pollution tomortality: a reanalysis of data on Londonwinters. EHP Environ. Health Perspect. 58:397-399.

Ostro, B.D. (1987). Air pollution andmorbidity revisited: a specification test. J.Environ. Econ. Manage. 14(1): 87-98.

Ozkaynak, H. and J.D. Spengler (1985).Analysis of health effects resulting frompopulation. exposures to acid precipitationprecursors. EHP Environ. Health Perspect.63: 45-55.

Pace, T.G., U.S. EPA, Monitoring and DataAnalysis Division (1983). The Use of TSPData to Estimate PMo Concentrations.Technical Memorandum to John Bachmann.Ambient Standards Branch, ResearchTriangle Park, N.C. Docket #A-79-29, I1-B-19.

Pace, T.G., E.L. Meyer,.N.H. Frank, and S.F.Slea (1986). Procedures for EstimatingProbability of Nonattainment of a PMoNAAQS Using total Suspended Particulateor'PMo Data. U.S. EPA, Office of Air 'Quality Planning and Standards, ResearchTriangle Park, N;C. EPA-450/4-86-017.

Padgett, J., U.S. EPA, Strategies and AirStandards Division (1981). Letter to Dr..Sheldon Friedlander, Chairman, Clean AirScientific Advisory Committee. October 30,1981. Docket #A-79-29, Il-C-3.

Pollack, A.K., A.B. Hudischewskj and A.D.Thrall (1985). An Examination of 1982-1983Particulate Matter Ratios and Their Use inthe Estimation of PMo NAAQS AttainmentStatus. Systems Application, Inc., SanRafael, CA. U.S. EPA, Office of Air OualityPlanning-and Standards,'Research TrianglePark, NC, EPA-450/4-85--OO.

Roth, H.D., R.E. Wyzga, and A.J. Hayter(1986). Methods and Problems in EstimatingHealth Risks from Particulates in Aerosols.(S.D. Lee, T. Schneider, L.D. Grant, P.J. 'Verkek, eds). Lewis Publishers, Chelsea,MI, pp. 837-957.

Schwartz, J.H. and A. Marcus (1986).Statistical Reanalyses of Data RelatingMortality to Air Pollution During LondonWinters 1958-1972. U.S. EPA, ResearchTriangle Park, N.C.

Shumway. R.H., R.Y. Tai, LP. Tal, and Y.Pawitan (1983). Statistical analysis of dailyLondon mortality and associated weatherand pollution effects. Sacramento, CA:California Air Resources Board; contract-No. AI-154-33.

Spengler, J.D., S.L.K. Briggs, and H. Ozkaynak(1986).Relationships Between TSPMeasurements and Size-FractionatedParticle Mass Measurements in Six CitiesParticipating In the HarvardAir'PollutionHealth Study.- Prepared for Office'of AirQuality Planning and Standards, Office ofPolicy Analysis, U.S. EnvironmentalProtection Agency, December 5, 1986.

Swift, D.L. and D.F. Proctor (1982), Humanrespiratory deposition of particles duringoronasal breathing. Atmos. Environ.16:2279-2282.

Ware, J., LA. Thibodeau, F.E. Speizer, S.:'Colome, and B.G. Ferris, Jr. (1981).Assessment of the health effects of.atmospheric sulfur oxides and particulate'matter. evidence from observationalstudies. Environ. Health Persp. 41:255-276.

Ware, J.H., B.G. Ferris, Jr., D.W. Dockery, J.D.-Spengler, D.O. Strain, and F.E. Speizer(1986). Effects of ambient sulfur oxides andsuspended particles on respiratory healthof preadolescent children. Am. Rev. Respir.Dis. 133: 834-842.

Watson, W.D. and J.A. Jaksh (1984).Production of Household Cleanliness inPolluted Environments. U.S. Department ofthe Interior, Geological Survey. Docket#A-82-37, IV-D-309.

Addendum I-CASAC Review andClosure of the 1982 Criteria Documentfor Particulate Matter/Sulfur Oxides andthe 1986 Second Addendum to theCriteria Document

January 29, 1982.Subject: CASAC Review and Closure of the

Criteria Document for Sulfur Oxides/Particulate Matter

From: Sheldon K. Friedlander, Chairman,Clean Air Scientific Advisory Committee(CASAC).

To: Anne M. Gorsuch, AdministratorOn November 16, 1981, the Clean Air

Scientific Advisory Committee of the ScienceAdvisory Board completed its third review of*the air quality criteria document for sulfuroxides/particulate matter (SOx/PM). TheCommittee notes with satisfaction the*improvements made in the quality of thedocument during the course of previousCASAC reviews on August 20-22,1980 andJuly 7-9,1981. The .staff of the EnvironmentalCriteria and Assessment Office, directed .byDr. Lester Grant, have proven responsive toCommittee advice as well as to commentsprovided by the general public, and deserveto be commended for the high quality of thedocument.:1

The purpose in writing you is to summarizethe Committee's major conclusions to assistyou in reviewing the scientific data andassociated studies relevant to theestablishment of revised ambient air qualitystandards for sulfur dioxide and particulatematter as required by law. This letter furtheradvises you of the Committee's conclusionthat the criteria document fulfills therequirements set forth In Section.108 of theClean Air Act as amended, which requiresthat the document "shall accurately reflectthe latest scientific knowledge useful inindicating the kind and extent of allidentifiable effects on public health or.welfare" from sulfur oxides and particulatesin the ambient air.'

The Committee is preparing a separateletter to you summarizing' the conclusions ofits reviews 'of'the Draft Staff Paper forParticulate Matter. ln'addition, CASAC will'prepare a'similar report on'theDraft Staff ' 'Paper for Sulfai Oxides once that document

24655


becomes available and its review iscompleted.

Major Scientific Issues and CASA CConclusions in the SO1 PM CriteriaDocument Review

Chapter 1: Executive Summary.In general, the revised draft Executive

Summary critically synthesizes the keypoints of information discussed at lengthin the individual chapters. Itsconclusions and interpretations ofscientific data, studies, and issues areconsistent with those presented in eachchapter. Relationships among individualchapters are clearly defined;redundancies that do appear arereasonable given the complexity of thesubject.

The quality of the Executive Summarywould be further improved if more -specific statements and/or tables wereadded to clarify-certain importantinterrelationships. These include thedifferences in chemical compositionassociated with each of the severalSignificant size ranges of particulatematter, and the health effects associatedwith the respiratory tract depositionpatterns of particulate matter in the.several size ranges and differentchemical compositions. Quantitativehealth effects information useful indefining specific concentrations orranges of concentrations of size-specificand/or chemical specific PM associatedwith the occurrence of health effectsshould also be highlighted. In view ofevidence that total thoracic(tracheobronchial and alveolar) particledeposition is of public health concern, itwould also be helpful to include adiscussion of the likely equivalencyamong British Smokeshade (BS), TotalSuspended Particles [TSP), and sizeselective particle aerometricmeasurements that would sample orindex atmospheric concentrations ofthose sized particles identified withtracheobronchial or alveolar deposition.

Chapter 2: Physical and ChemicalProperties of SO./PM.

This chapter is well written andaddresses the important issues relevantto a criteria document. It presents agood summary of current knowledge ofthe factors affecting the physics andchemistry of sulfur dioxide and thepathways and kinetics of itstransformation into sulfuric acid. It alsoprovides a good summary of particlecharacteristics, dynamics, andhygroscopic growth.

Chapter 3: Techniques for theCollection and Analysis of SOJPM.

The revised chapter provides anexcellent summary of the measurementof sulfur oxides and particulates.Especially important is the discussion of

the capabilities of the variousmeasurement techniques and the profileof pollutants in the ambient air whichthese measurements yield. The chaptercorrectly notes that British Smoke (BS),Coefficient of Haze (COHS), and TotalSuspended Particulate, (TSP)measurements do not adequately reflectkey physical or chemical properties ofparticulate matter in the contemporaryambient air. Precise Interconversionamong units of BS, COHS. and TSP isnot possible. In the context of aparticulate standard, British Smoke isapplicable only to a "sooty" smokeaerosol. It may not be a valid healtheffects indicator for the aerosolcompositions observed in recentsummertime episodes in the UnitedStates and Europe. Thus, it is unlikelythat BS can provide a sensitive index ofhazard for today's air pollution.

Chapter 4. Sources and Emissions.Both natural and man-made sources

emit sulfur dioxide and particulatematter Into the ambient air. Given thelimitations of Our ability to derive -

reliable estimates from both types ofsources, the criteria document presentsan adequate discussion of currentknowledge.• Chapter 5: EnvironmentalConcentrations and Exposure.

This chapter is largely acceptable inits present form. Most of the commentsand suggestions which were made forprevious drafts have been' effectivelyIncorporated. The most importantomission from the chapter is informationrelated to chemical composition withrespect to particle size. Abundantinformation of this type is available forsulfates and some trace metals. Giventhe strong dependence of depositionrates and light scattering on particlesize, it might have been worthwhile torefer to this literature in Chapter 5 or todirect attention to other documentchapters (e.g.. Chapter 2) where suchrelationships are discussed.

Chapter 6: Atmospheric Transport.Transformation and Deposition.

This chapter is concise, well-written,and effective in communicatingInformation related to the current statusof mathematical models for air pollution.The utility of various models is clearlydiscussed, and the inadequacy ofcurrent models for quantitativeextrapolation is pointed out. Topicswhich had been omitted from theprevious draft of this chapter have beenadded to other chapters withoverlapping content. The chapter is nowacceptable as written.

Chapter 7: Acidic Deposition.The Committee has recognized the

desirability of incorporating existinginformation on acidic deposition in the

present criteria document. Chapter 7provides an abbreviated but adequatesummary of the contribution of sulfuroxides and particulates to the formation,transport, and effects of acidicdeposition. The Committee hasconcluded that Chapter 7 is ascientifically adequate summary withthe conditional understanding that EPAis preparing a Critical AssessmentDocument for Acidic Deposition for itsreview that recognizes and incorporatesinformation on causes, effects, and databases for all of the various pollutantsrelevant to acidic deposition. CASAChas been briefed several times byAgency officials regarding the status ofthis document. The Committee looksforward to the submission of thisintegrated assessment for its criticalreview.

Chapter 8: Effects on Vegetation.In response to CASAC

recommendations and public comments,this chapter on vegetation effects hasbeen greatly improved compared toearlier drafts reviewed by theCommittee. It now includes a moreconcise and interpretive criticalevaluation of those few key studiesyielding quantitative dose-effect ordose-response information of most usefor criteria development and standard-setting purposes. It also reasonablyincludes tables in the appendices whichsummarize studies of particulates andsulfur dioxide related vegetation effectsthat are of less utility for criteriadevelopment and standard setting.

The Committee concurs with Chapter8 evaluations which point to the lack ofdose-response data.to establishquantitative evidence of deleteriouseffects on vegetation from particulatesat presently encountered U.S. ambientair concentrations. In contrast toparticulates, much clearer evidenceexists by which to define quantitativeexposure-effect relationships for sulfurdioxide effects on vegetation.Laboratory experiments in particularhave demonstrated the greater relativetoxicity to vegetation from high short-term exposures of sulfur dioxide. This isespecially important in view of the factthat ambient air concentrations of sulfurdioxide from point sources oftenfluctuate widely and result in highintermittent short-term exposures ofplants to sulfur dioxide concentrationsagainst a background of longer-term butmuch lower annual average sulfurdioxide levels. Also of much importanceare differences in the relative sensitivityof various plant species to sulfur dioxideexposures. The degree of sensitivitydepends in part on factors such as phaseof growth at time of exposure, ambient


temperature and humidity levels, andplant water content. Among studiesjudged to be most useful for quantitativecriteria development and standardsetting are those of Dreisinger (1965;1967) and Dreisinger and McGovern(1970) which demonstrate visible injuryto white pine (a commercially importantspecies in some U.S. areas) whennatural stands of the tree in southernCanada were exposed for 4 hours to 0.30ppm or for 8 hours to 0.25 ppm sulfurdioxide emitted from a nearby smelter.Roughly similar exposure-effectrelationships were observed in studiesreported by Jones et al. (1974) andMcLaughlin (1981) on the effects ofsulfur dioxide from a southeastern U.S.power plant on a wide variety of naturalspecies in the vicinity of the pointsource. In these studies some crop andgarden species showed visible injuryeffects with 3 hour exposures to 0.6-0.8ppm sulfur dioxide, while certain othercrop species (potato, cotton, corn,peach) did not show visible injury atlevels below 0.8 ppm. In contrast, achamber study by Hill et al. (1974)suggests that plants common to thesouthwestern U.S., with markedly lowermoisture content and under generallylower ambient air humidity levels, maybe able to withstand much higherambient sulfur dioxide concentrations(up to 11 ppm for two hours) withoutvisible injury.

Chapter 9: Effects on Visibility andClimate.

The technical aspects of this difficultproblem are well characterized. Thechapter does a good job of discussingthe physics and public awareness ofvisibility. The relationship between fineparticle mass concentrations andvisibility has been well established. Thecriteria document thus provides anexcellent technical basis for Agencydecision-making on these issues.

Chapter 10: Effects on Materials.This chapter adequately discusses the

currently available scientificinformation concerning the effect ofparticulate matter and sulfur oxides onman-made materials. This includescritical assessments of available dataconcerning pertinent materials damagefunctions, uncertainties associated withexisting characterizations of suchfunctions, and limitations regardingestimation of monetary costs and/orbenefits associated with the occurrenceor control of such damage.

Chapter 11: Respiratory Depositionand Biological Fate of Inhaled Aerosolsand Sulfur Dioxide.

This chapter is very much improvedcompared to earlier drafts reviewed byCASAC and is now a comprehensiveand more informative summary of

existing knowledge relevant to a criteriadocument. The existing knowledge inthis area is, in many cases, incomplete.For example, a potentially veryimportant factor is the influence of theintegrity of lung epithelial barriers (bothairway and alveolar) on deposition andclearance. To enhance the chapter'scomprehensiveness, this issue should bediscussed more sufficiently in thecriteria document, despite the paucity ofavailable data.

Chapter 12: Toxicological Studies.This chapter is quite comprehensive

as it describes essentially alltoxicological studies relevant-to acriteria document on sulfur oxides andparticulates. Also, it providescommentary on many studies and theSignificance of their findings to potentialhuman health effects. In addition, thepresentation of the information is morepolished than the previous draft becauseof improved editing.

Chapter 13: Controlled HumanStudies.

This Is a chapter which thoroughlydiscusses the published material oncontrolled human experiments. The

,scientific criteria for good studiesdiscussed at the beginning of thechapter cannot be overemphasized.While not all studies meet these criteria,the Committee recognizes that EPA musttake account of the available literatureand believes the studies cited in thechapter have been appropriatelyselected and discussed. Overall thechapter is well-written and directedtoward addressing those questions towhich answers are needed. One of themost important criteria for good hmtianclinical studies is that they be double-blind. Unfortunately, most of the studiesin the literature were not so performed.This factor is especially significantwhen sensitive population groups, suchas asthmatics, are under study.

The chapter is also improved by thediscussion of exposures administeredthrough the nose and mouth duringcontrolled studies. It appropriately notesthat caution should be used in anyattempted extrapolation of observedquantitative exposure/effects resultingfrom such protocols, particularly whencompared to results that might beexpected under ambient exposureconditions. The chapter identifiesadditional research results from studiesusing either face mask or open chamberoronasal breathing that would betterresolve this issue, and it discussesexisting studies in a balanced andthorough fashion.

Chapter 14: Epidemiological Studies.The current draft of this chapter

represents considerable change andimprovement over previous drafts

reviewed by CASAC. Followingdiscussion with the Committee, EPA hasapplied a set of guidelines for decidingwhich epidemiological studies are mostappropriate for use in revising ambientair quality standards.

More specific comments on thechapter include the following: (1) theintegration of Chapter 14 with Chapter 3has advanced the "real world"understanding concerning theapplication of epidemiological methods;(2)the epidemiological studies providingthe most useful quantitativeconcentration/response information forrevising the 24-hour ambient particulatestandard include: Lawther'et al, 1958and 1970; 'Martin and Bradley 1960;Martin 1964; Ware et al, 1981; andMazumdar et al, 1981; (3) theepidemiological studies providing themost useful quantitative concentration/response information for revising theannual ambient particulate standardinclude: Ferris and Anderson 1962; Lunnet al, 1967; Ferris et al, 1971 and 1976;and Bouhuys et al, 1978; and (4) thestudies by Lave and Seskin, 1970, andMendelsohn and Orcutt, 1979 suggest anassociation between chronic exposure,to high concentrations of sulfates andincreases in the level of mortality, butthey do not indicate any threshold orsafe level from such exposures, and theyare not refined enough to provideestimates of the quantitative effect ofsulfate concentrations on mortality.

Summary

The Committee made numerouscomments of an editorial nature. Theseremarks, as well as a more detailed'discussion of the recommendations andreview provided above, are included inthe transcripts of the three CASACmeetings held to review this document.With the understanding that the advisedchanges will be incorporated in the finalcriteria document, the Committee issatisfied that the air quality criteriadocument for sulfur oxides/particulate •

matter is scientifically adequate for usein standard setting.December 15, 1986.The Honorable Lee M. Thomas,Administrator. U.S. Environmental Protection

Agency, Washington, DC 20460Dear Mr. Thomas: The Clean Air Scientific

Advisory Committee (CASAC] has completedits review of two documents related to thedevelopment of National Ambient AirQuality Standards (NAAQS) for ParticulateMatter and Sulfur Oxides. These twodocuments are the 1982 Air Quality Criteriafor Particulate Matter and Sulfur Oxides, andthe 1986 Second Addendum to Air QualityCriteria for Particulate Matter and SulfurOxides (1982), both prepared by the Agency's

24657

24658 Federal Register / Vol. 52, No. 126 1 Wednesday, July 1, 1987 1 Rules and Regulations

Environmental Criteria and AssessmentOffice fECAO).

The Committee was impressed with theefforts of the staff of ECAO in preparing awell written. Integrated and thorough reviewof recent relevant scientific studies. TheCommittee unanimously concluded that this1986 Addendum, along with the 1982 CriteriaDocument previously reviewed by CASAC,represent a scientifically balanced anddefensible summary of the extensivescientific literature on these pollutants.

Several important issues are discussed inthe 1986 Addendum which the Committeebelieves should be emphasized. These issueswere raised during our review of recentstudies which relate primarily to guidance atthe lower bounds of the ranges for thestandards. These studies include the recentreanalyses of the London mortality data, twoepisodic lung function studies in the UnitesStates and the Netherlands. and thecomparison of respiratory symptoms andpulmonary function levels of children livingin six U.S. cities. Further discussion of thesestudies and reanalyses, as well as a moredetailed discussion of the basis for theCommittee's conclusions, are contained inthe attached report.

The Committee also reviewed the StaffPapers for particulate matter and for sulfuroxides at the October 15-16, 1986 meeting,and is preparing separate reports reflectingits conclusions and recommendations oneach of these two documents.

Thank you for the opportunity to presentthe Committee's views on these importantpublic health issues.

Sincerely,Morton Lippmann, Ph.D..Chairman. Clean Air Scientific AdvisoryCommittee.

cc: A. James Barnes, Lester Grant, VaunNewill, Craig Potter, Terry Yoste.

Summary of Major Scientific Issues andCASAC Conclusions on the 1986Addendum to the 1982 ParticulateMatter/Sulfuir Oxides (PM/SOJ CriteriaDocument

The Committee concentrated itsreview on newer studies and analyseswhich relate primarily to guidance onthe lower limit of the proposed rangesfor the standards. In general, theCommittee believes the CriteriaDocument Addendum has appropriatelysummarized and interpreted the designs,analyses and conclusions of studies thatshould be considered in the standardsetting process. The following is a briefchapter by chapter summary of issuesthat the Committee wishes toemphasize, or which require furtherclarification.

Chapter 1: Introduction

In general, this chapter provides anexcellent summary of the physical andchemical properties and ambientmeasurement methods for PM and SOL.However, the chapter could bestrengthened by inclusion of adiscussion of direct reading monitors forparticulate mass concentrationsincluding beta attenuation, lightscattering, or other techniques which

may be the dominant measurementtechniques in the States in the future.This was discussed at the December1985 CASAC meeting, with emphasis onthe need to move to automated andcontinuous monitoring for particles.

Chapter 2: Respiratory Tract Depositionand Fate

The presentation in this chapter couldbe expanded by clarifying thediscussion concerning the concept ofimpaired lungs and the deposition thatwould occur there as opposed to that Innormal subjects. Further. the discussionof broncho-constriction being protective(Svartengren et al., 1984) and thediscussion of other types of alteredbreathing patterns could be madeclearer, perhaps by reorganizing thisinformation by specific points.

Chapter 3: Epidemiology Studies

We wish to emphasize several studiosand-analyses discussed at the October1986 CASAC meeting. One of thesestudies (Dassen et al.) should beintegrated into this chapter, as wasrecognized by Agency staff in theirremarks at the October 1986 meeting.

.(1) The two episodic lung functionstudies show a consistency of results inSteubenville, Ohio (Dockery et al.) andljmond. Netherlands (Dassen et al.),lending credence to reported effects of amixture of PM and sulfur oxides (SO)on respiratory function in children. Thisis consistent with the earlier work ofStebbings. These studies provide arelatively sensitive indication ofpossible short term physiologicalresponses of uncertain healthsignificance to PM. The roles ofexposure times and duration offunctional decrement need betterdefinition.

(2) The London mortality studies,including recent analysis by Agencystaff, provide strong evidence thatparticulate matter is more closelyassociated with daily mortality thansulfur dioxide concentrations. Thecriteria document should recharacterizedistinctions made between "likely" and"possible" effects levels for establishingupper bounds.

(3) The Six-Cities etudy has reportedthat cough and bronchitis are twice asprevalent in children living in cities withPMo in the range of 40-60 jkg/m 3 , incomparison to cities with a range of 20-30 pg/m s .

Chapter 4: Controlled Human ExposureStudies of SO2 Health Effects

Although this chapter was well done,the Committee suggests that it be.strengthened by modifying its existing,discussions and by addition of furtherdiscussion and tabular materialconcerning short term exposure effectspresented by Drs. Horstman andFolinsbee at the October 1986 CASACmeeting.

ConclusionThe 1986 Addendum to the 1982 Air

Quality Criteria Document on PM/SO.was prepared by EPA at the request ofCASAC for the purpose of updating theknowledge of recent scientific studiesand analyses. The Committeecommends the Agency staff for itsefforts in preparing a concise and wellwritten document. The Addendumsummarizes key findings from the earlierdocuments and provides a reasonablycomplete summary of newly availableinformation concerning particulatematter and sulfur oxides, with majoremphasis on evaluation of human healthstudies published since 1981. TheCommittee unanimously concludes thatthis 1980 Addendum, with theincorporation of the changes notedabove, represents a scientificallybalanced and defensible summary of theextensive .scientific literature on thesepollutants. These documents fulfill therequirements under section 108 of theClean Air Act as amended, whichrequires that the document(s)"... shallaccurately reflect the latest scientificknowledge' useful in indicating the kindand extent of all identifiable effects onpublic health or welfare . . . "fromparticulate matter and sulfur oxides inthe ambient air.

Addendum II-CASAC Review andClosure of the 1982 OAQPS Staff Paperfor Particulate Matter and the 1986Addendum to the Staff Paper

January 29, 1982.Subject: CASAC Review and Closure of the

OAQPS Staff Paper for ParticulateMatter

From: Sheldon K. Friedlander, Chairman.Clean Air Scientific Advisory Committee

To: Anne M. Gorsuch. AdministratorThe Clean Air Scientific Advisory

Committee (CASACI recently completed itssecond and final review of the documententitled Review of the National Ambient AirQuality Standards for Particulate Matter:Assessment of Scientific and TechnicalInformation, OAQPS Staff Paper. TheCommittee notes with satisfaction theimprovements made in the scientific qualityand the completeness of the staff paper. Ithas been modified in accordance with therecommendations made by CASAC in Julyand November 1981. This document is alsoconsistent in all significant respects with thescientific evidence presented and interpretedin the combined criteria document for sulfuroxides and particulate matter. It hasorganized the data relevant to theestablishment of particulate primary andsecondary ambient air quality standards in alogical and compelling way, and theCommittee believes that it provides you withthe kind and amount of technical guidancethat will be needed to make appropriaterevisions to the standards.

CASAC has prepared this closurememorandum to inform you more specificallyof its major findings and conclusionsconcerning the various scientific issues andstudies discussed in the staff paper. Inaddition, the Committee's review of thescientific evidence leading to the particulate


standard revision leads to a discussion of itsown role in the process for setting thestandard.

CASA C Conclusions andRecommendations on Major ScientificIssues and Studies Associated With theDevelopment of Revised NAA QS forParticulates

1. Based upon the review of availablescientific evidence, a separate generalparticulate standard remains areasonable public health policy choice.

2. CASAC reaffirms its initialrecommendation of July 1981 toestablish a 10 micrometer cut point for arevised primary particulate standard.This recommendation is based upon arecognition of the periodic, andsometimes frequent, tendency of bothhealthy and sensitive populations tobreathe through their mouths and/ororonasally. This practice increases theamount of particulate matter that canpenetrate into the thorax because thelarger particles are not filtered in theoronasal passages. Deposition ofparticulates into this region is of specialconcern to those individuals with pre-existing respiiatory problems andchildren. In addition, the collection ofparticles of less than 10 micrometerdiameter size more closely resemblesparticles passing into the thoracic regionof the human body than the collection oflarger sized particles. Furthermore,monitors equipped for a 10 micrometercut are less wind dependent and canprovide a more accurate profile of thecontemporary ambient air than samplerswhich measure total suspendedparticles.

CASAC's recommended size cut isalso similar to proposals of otherscientific associations. For example. 88%of the national members of the AirQuality Committee of the InternationalStandards Organization recently votedfor a particulate cut point at 10micrometers for sampling particleswhich can deposit in the lungs.

The CASAC recommendation is basedupon available scientific data. Otherindividuals and groups have discussedthe possibility of establishing a revisedparticulate standard at a size cutconsiderably less than 10 micrometers.However, for the current revision of thestandard, the scientific data morereadily support a 10 micrometer size cut.

3. CASAC reached several majorconclusions concerning the revision ofthe 24-hour and annual particulatestandards. At the upper bound of theproposed ranges of 150-350 Jlg/m a forthe 24-hour and 55-110 pg/m3 for theannual averages, detectable healtheffects occur in the populationsevaluated in the epidemiological studies.

Since the upper end of these rangescontain little or no margin of safety, itwould be appropriate to consider lowervalues for revising the 24-hour andannual standards. In addition, the statedranges are based solely on quantitativeevidence reported in epidemiologicalstudies. A final decision on a revisedstandard should also incorporateinformation generated throughcontrolled human, animal toxicology,and from other less quantitativeepidemiological studies discussed in thecriteria document.

There is an absence of a clearlydefinable exposure-responserelationship for particles, as amplydiscussed in the criteria document andthe staff paper. In addition, becauseairborne particles are heterogeneous incomposition, the potential toxic effectsof individual constituents should beconsidered in setting the standard; Thus,compared to margins of safety set forpollutants such as ozone and carbonmonoxide, where exposure-responserelationships are better established andsmall margins of safety are morejustifiable, CASAC believes you shouldconsider a revised standard with awider margin of safety.

4. The Committee reached generalagreement that the annual particulatestandard should consist of an arithmeticmean. It is recommended that the 24-hour standard include a statistical formand that the number of exceedances isset in relation to the revised standardlevel.

5. During the past decade, the linkbetween visibility and fine particle massconcentrations has-been convincinglydocumented. Visibility is a sensitiveindicator of accumulated man-madepollutants in the ambient air. The publiccares about visibility and is willing topay something for clean air. However,the quantitative basis for establishing apsychological, economic, transportationor any other welfare cost associatedwith visibility impairment has not beenestablished. In addition, controlsrequired to achieve a given visibilitystandard are not known due to thecomplexities of pollutant transport andtransformation.

Defining acceptable levels of visibilityis a social/policy judgment as well as ascientific decision, but science canprovide some guidance. The upper endof the 8-25 jug/m 3 range for fineparticles (those particles with adiameter size of less than 2.5micrometers) would tend to maintain thestatus quo for the eastern United Statesand some western urban areas, butwould permit air quality degradation forlarge areas in the west includingnational parks. Also, it is highly

uncertain that the recommendedthoracic particle ranges for the primarystandard will protect visibility. The 8-25pg/m3 range for fine particles suggestedfor visibility protection is a seasonaland spatial average, unlike peak valueswhich will be recommended for theprimary standard.

The strongest case for a visibilityrelated standard is one that linksemissions of nitrogen oxides and sulfurdioxide with the interrelated aspects ofacidic deposition, possibleclimatological effects, and visibility.Each of these three air quality issues isrelated to the fine particles whichoriginate both as primary particulateemissions and as secondary aerosolsfrom atmospheric conversions of sulfurdioxide and nitrogen oxides emitted asvapors; In terms of a control strategy toprotect public welfare, it may be moreefficient to consider a common standardlinked to fine particles than to establisha separate set of controls for each ofthese problems and pollutants.

6. The Committee's evaluation ofscientific data and studies in the criteriadocument and the staff paper lead it toconclude that there is no scientificjustification for the establishment of aparticulate standard for the specificprotection of vegetation.

7. The Committee discussed whateffect elimination of a Total SuspendedParticulate (TSP) standard would haveon the environment. The soiling andnuisance aspects of TSP are essentiallylocal air quality problems because suchcoarse particles are not transportedgreat distances. This contrasts withvisibility or oxidant related problemswhich are distinctly issues of long rangepollution transport. Individuals whoserve on the Committee made variousrecommendations regarding retention orelimination of a secondary standard forTSP, but no clear consensus evolved.

The Process for Setting the AmbientParticulate Standard

In its report of September 21, 1981,CASAC made several majorrecommendations relating to the processfor setting ambient air standards. TheCommittee is aware that your staff isanalyzing its report and is awaiting aresponse.

A major underlying assumption of theCommittee's recommendations was theneed to make more explicit therelationship between the scientificevidence in the criteria document andthe staff paper and the eventualselection of a numerical level forindividual standards. The Committeestrongly believes in the need to clarifythe standard setting process by

2A59

24660 'Federal Register / Vol. 52, No. 126 / Wednesday, July 1, 1987 / Rules and Regulations

identifying the key studies that willshape the determination of a standard.Intensive evaluation of such studies byCASAC and the public will considerablyincrease your ability to set ascientifically supportable standard.

The Committee is greatly encouragedby your decision to improve the formatand content of OAQPS scientific issuestaff papers. In the, Draft Staff Paper forParticulate Matter key studies areidentified and their implications forsetting primary and secondarystandards are discussed. Moreimportantly, the inclusion of numericalranges and their supporting rationaleenable the Committee and the public tocritically examine the staff's proposeduse of the studies. This led to a markedimprovement in the quality of the publicdialogue concerning the scientific basisfor revising the standard. CASACcommends your effort and recommendsthat all staff papers developed forambient air standards contain numericalranges.

CASAC recognizes that your statutoryresponsibility to set standards requirespublic health policy judgments in 'addition to deiermination of a strictlyscientific nature. While the Committee iswilling to-further advise you.on theparticulate standard, we see no need. inview of the already extensive commentsprovided, to review the proposedparticulate standards prior to theirpublication in the Federal Register. Inthis instance, the public comment-periodwill provide sufficient opportunity forthe Committee to provide any additionalcomment or review that may benecessary.December 16, 1986.The Honorable Lee Thomas,Administrator, U.S. Environmental Protection

Agency, Washington, DC 20460.Dear Mr. Thomas: The Clean Air Scientific

Advisory Committee (CASAC) has completedIts review of the 1986 Addendum to the 1982Staff Paper on Particulate Matter (Review ofthe NAAQS for Particulate Matter:Assessment of Scientific and TechnicalInformation). prepared by the Agency's Officeof Air Quality Planning and Standards(OAQPS).

The Committee unanimously concludesthat this document is consistent in allsignificant respects with the scientific -evidence presented and interpreted in thecombined Air Quality Criteria Document forParticulate Matter/Sulfur Oxidesand its 1986Addendum, on which the CASAC recentlyissued its closure letter. The Committeebelieves that this document provides youwith the kind and amount of technicalguidance that will be needed to makeappropriate revisions to the standards. TheCommittee's major findings and conclusionsconcerning the various scientific issues andstudies discussed in the Staff Paper

Addendum are contained in the attachedreport.

Thank you for the opportunity to presentthe Committee's views on this importantpublic health issue.

Sincerely,Morton Lippmann, Ph.D.,Chairman, Clean Air Scientific AdvisoryCommittee.cc: A. James Barnes, Gerald Emison, VaunNewill, John O'Connor, Craig Potter, TerryYosie.

Summary of Major Scientific Issues andCASAC Conclusions on the 1986 DraftAddendum to the 1982 ParticulateMatter Staff Paper

The Committee found the technical

discussions contained in the Staff PaperAddendum to be acceptable with minorrevisions.

Particle Size Indicator

The CASAC reaffirms its January 29,1982 recommendation that a particlesize indicator that includes only thoseparticles less than or equal to a nominal10 urn aerodynamic diameter, termedPMo, is appropriate for regulation ofparticulate concentrations. Thisjudgment is based on analysis of theearlier available data, and the analysisof the recent scientific studies discussedin the 1986 Addendum to the:Air QualityCriteria for Particulate Matter/SulfurOxides and the 1986 Addendum to theParticulate Matter Staff Paper.

Implications of London MortalityStudies

Further analyses of the Londonmortality studies, including recentanalysis by Agency staff, suggest that:

(1) the data provide no evidence for athreshold for the association betweenairborne particles and daily mortality ora change of coefficient with changes inparticle composition;

(2] mortality effects can be associatedwith PM alone (with or without sulfuroxides);

(3] there is no reliable quantitativebasis for converting British Smoke (BS)readings to PMo gravimetric mass atlow (<100-200 )g/m) BS levels, andhence the mortality data are not readilyuseful for establishing a lower bound for24-hour PM1o NAAQSA, although thesuggestion of mortality at relatively lowPM levels must be given seriousconsideration in selecting a margin ofsafety.

Interpretation of Lung Function Studiesfor 24-hour Standard

Although the lung functiondecrements observed in children duringand after air pollution episodes are ofuncertain health significance, the two

episodic lung function studies (Dockeryet al., 1986; Dassen et al., 1986) areconsistent with each other and theearlier work of Stebbings. They providea relatively sensitive indication ofpossible short term physiologicalresponses. Given the difficulty inderiving a lower limit from the mortalitystudies, these lung function studies canbe useful in determining lower boundsfor a 24-hour PMo standard.

Interpretation of the Six Cities Study forAnnual Standard

In general, the Committee felt that thesix cities data are useful in establishingthe lower bound of the range for theannual standard. In addition, thefollowing are suggested by the data:

(1) Cough and bronchitis, as defined inthis study, are about twice as prevalentin children living in cities with PMo inthe range of 40-60 Ag/m 3 in comparisonto cities with 20-30 1kg/m 3;

(2) Because factors other thanparticulate matter may affect the inter-city differences, it is difficult todetermine whether these associationsshould be designated as "likely" healtheffects;

(3) The results are consistent with theOstro studies in terms of morbidityresponses at long-term averageparticulate matter exposures withincurrent particulate ambient air qualitystandards; and:

(4) The results are consistent with theBouhuys study in terms of symptomswithout changes in pulmonary function.

Ranges for 24-hour and AnnualStandards for PMlo

In its January 2, 1986 letter to theAdministrator, the CASAC noted that itspreliminary analyses of the more recentdata do not indicate-the need forfundamental changes in the stricture ofthe proposed particle standards;however, the Committee pointed outthat these new data suggest the need tofocus consideration on standards at orperhaps below the low ends of theranges proposed in the March 20, 1984Federal Register Notice. The ranges ofinterest then proposed were 150-250 )Ag/m3 for 24-hour standard, and 50-65 gLg/m 3 for annual standard.

Since then, EPA staff have proposedupdated ranges of interest for both the24-hour standard (140-250 gg/m 3), andthe annual standard (40-65 gg/m 3),based on short-term and long-termepidemiological data, respectively. TheCommittee finds these ranges of interestreasonable, given the scientific data andrelated uncertainties; however, a finaldecision should also weigh evidencefrom clinical and toxicological studies

Federal Register I Vol. 52, No. 126 / Wednesday, July 1, 1987 / Rules and Regulations

as well. The Committee agrees with EPAstaff that selection of final standardsmust include consideration of thecombined protection afforded by the 24-hour and annual standards takentogether.

The Committee recommends that youconsider setting the revised standards atthe lower ends of the proposed rangesfor both the 24-hour and annualstandards. The Committee recognizesthat the exact levels to be chosen for the24-hour and annual standards representa policy choice, influenced by the needto include a margin of safety. Given theuncertainty in the supporting scientificdata, the Committee cannot distinguishthe health effects that may be observedat different levels near the lower bound,such as the health significance of settingthe 24-hour standard at 140 tg/mcompared to 150 g/im '.

Addendum 11-Executive Summary ofthe 1986 Addendum to the Staff Paper

Review of the National Ambient AirQuality Standards for ParticulateMatter: Updated Assessment ofScientific and Technical Information-Addendum to the 1982 OAQPS StaffPaper (EPA, 1986b).

Executive Summary

This paper evaluates and interpretsthe updated scientific and technicalinformation that the EPA staff believesis most relevant to decision making onrevised primary (health) nationalambient air quality standards (NAAQS)for particulate matter and is anaddendum to the 1982-particulate matterstaff paper. The paper assesses thefactors the.staff believes should beconsidered in selecting the pollutantindicator and level for the primaryparticulate matter standards, updatingand supplementing previous staffconclusions and recommendations Inthese areas to incorporate more recentinformation. This assessment isintended to help bridge the gap betweenthe scientific review contained in theEPA criteria document addendum"Second Addendum to Air QualityCriteria for Particulate Matter andSulfur Oxides (1982): Assessment ofNewly Available Health EffectsInformation" and the judgmentsrequired of the Administrator in makingfinal decisions on revisions to theprimary NAAQS for particulate matterthat were proposed in March 1984 (49 FR10408). The staff paper and thisaddendum are, therefore, importantelements in the standards reviewprocess and provide an opportunity forpublic comment on proposed staffrecommendations before they arepresented to the Administrator.

Particulate matter represents a broadclass of chemically and physicallydiverse substances that exist as discreteparticles (liquid droplets or solids)ranging in size from molecular clustersof 0.005 micrometers (p.m) to coarseparticles on the order of 1000 in. Themajor chemical and physical propertiesof particulate matter vary greatly withtime, region, meteorology and sourcecategory, complicating the assessmentof health and welfare effects as relatedto various indicators of particulatepollution. The original measurementmethod for the particulate matterNAAQS was the "hi volume" sampler,which collects particles of sizes up to anominal 25-45 jtm (so-called "TotalSuspended Particulate" or TSP). EPAhas proposed to replace this particulatematter indicator with one that includesonly particles with aerodynamicdiameters smaller than a nominal 10 pm,termed "PM1o". Although a large numberof PMo monitors are now in place,reliable and consistent data are, atpresent, limited. Data from 39 sites inEPA's IP network show long-term urbanPMo levels range between 25 and 75 .g/m3 and maximum 24-hour values rangefrom 50 to 175 Ag/m. Higher values arelikely as more data become available.Both fine (<2.5 p±m) and coarse (>2.5um) particles are substantialcomponents of PMo mass, with atendency for higher coarse contributionsin western U.S. locations with higherconcentrations. National estimates ofPMo levels are derived from applyingmeasured PM1oTSP ratios to the widerTSP data seL This analysis (for 1983-85data) estimated that 193 countiesexceeded the lower bound of the rangesproposed for PM1o standards (150 pg/ ma24 hour, 50 pgg/m 3 annual) while 136counties had sites that exceeded thecurrent primary TSP standards.

Particle Indicator

Based on an examination of airquality composition. respiratory tractdeposition, and health effects andrelated considerations, the 1982 staffpaper recommended adoption of the sizespecific indicator (PMo) proposed in1984. The present staff assessment of themore recent information on respiratorytract deposition contained in the criteriadocument addendum reinforces theconclusions reached in the original staffassessment in 1982. The staff finds thatthe recent data do not supportalternative indicators that have-beensuggested, which exclude all particleslarger than 10 pim. The PMo indicator isgenerally conservative over the range oftracheobronchial deposition.

Recent information suggestingenhanced tracheobronchial particle

deposition for children relative to adultsprovides an additional reason for anindicator that includes particles capableof such penetration. Given theseconsiderations and its earlierconclusions, the staff reaffirms itsrecommendation to replace TSP as theparticle indicator for the primarystandards with a new indicator thatincludes only those particles smallerthan a nominal 10 pim in aerodynamicdiameter (PMo). The previouslydeveloped effectiveness criteria forsamplers are acceptable for regulatorypurposes.

Level of Standards,

The major scientific basis for selectingPM standards that have an adequatemargin of safety remains communityepidemiological research, withmechanistic support from toxicologicaland controlled human investigations.The limitations of epidemiologicalstudies for these purposes must,however, be recognized. Such studies,while 'epresenting real world'conditions, can only provideassociations between a complexpollutant mix measured at specificlocations and times and a particular setof observable health points. Difficultiesin conducting and interpretingepidemiological studies limit thereliance that can be placed on theresults of any single study. None of theavailable studies have used PMo as adirect measure of pollution, requiring-where appropriate--further conversionof results to estimated PMo units.

The 1982 criteria document and thecriteria document addendum identify alimited set of epidemiological studiesmost useful for developing quantitativeconclusions regarding the effects ofparticulate matter. This updated staffassessment incorporates the previousevaluation of the earlier studies as wellas the present assessment of morerecent studies.

The updated staff assessment of theshort-term epidemiological data issummarized in Table 1; levels areexpressed in both the original (Britishsmoke-"BS" or TSP) and PMo units.The "effects likely" row denotesconcentration ranges derived from thecriteria document and its addendum ator above which a consensus judgmentsuggests greatest certainty that someeffects would occur, at least under theconditions that obtained in the originalstudies. The data do not, however, showevidence of clear population thresholdsbut suggest a continuum of responsewith both the risk of effects occurringand the magnitude of any potential,effect decreasing with concentration.

24661

24662 Federal Register / Vol. 52; No. 126 / Wednesday, July 1, 1987 / Rules and Regulations

This is particularly true for the nature of association provide no clear effects are possible in the range listed instatistical analyses of daily mortality in basis for distinguishing any particular Table 1, but the relationships are notLondon. Substantial agreement exists' lowest "effects likely" levels or for certain enough to derive "effects likely"that wintertime pollution episodes defining a concentration below which levels for PMo. They do suggest levelsproduced premature morta lity in elderly no association remains,.The recent lung below which detectable functionaland ill populations, but the range and function studies in children suggest that changes are unlikely to occur.

Table 1. Updated Staff Assessment of Short-Term Epidemiological Studies

Measured British smoke levels (as Ag/m 3) Measured TSP levels (pg/m 3 Equivalent PM~o

Effects/study (24-hr. avg.) (24-hr. avg.) levels (g/m 3)

Daily mortality Aggravation of Combined Small, reversible declines in Combinedin London I bronchitis 2 range lung function in children 3,4 range 5

Effects Likely .......................... 1000 250*-500* 250-500 ............................................................... 350-600Effects Possible .................................... ? <250* < 250 220*-420 3-200-2504 140-350No Significant Effects Noted ............. ............... ................................. ..... 125.4_160 3 <125

* Indicates levels used for upper and lower bound of range.

Various analyses of daily mortality encompassing the London winter of 1958-59, 14 winters from 1958-72, in aggregate and individually..Early winters dominated by high smoke and SO2 from coal combustion with frequent fogs. From 1982 CD: Martin and Bradley (1960); Ware etal., (1981): Mazumdar et al. (1981). From 1986 CD Addendum: Mazumdar et al. (1982); Ostro (1984); Shumway et-al., (1983);, Schwartz andMarcus (1986). Later studies show association across entire range of smoke, with no clear delineation of "likely" effects or threshold ofresponse possible.

2 Study of symptons reported by bronchitis patients in London, mid-So's to early 70's; Lawther et al. (1970)..' Study of pollution "episodes" in Steubenville, Ohio, 1978-80; Dockery et al. (1982).4 Study of 1985 pollution episode in ljmond, The Netherlands; Dassen et al. (1986).5 (a) Conversion of BS readings to PM10 levels: Assumes for London conditions and BS readings in the range 100-500 Ag/rn 3. BS <PM~o <

TSP Precise conversions are not possible. Uncertainty in measurements of BS and conversion relationships preclude quantitative estimates ofrange for lower BS levels. The upper bound assumption (PM~o = TSP = BS + 100 jug/m a ) overestimates PMo levels.(b) Conversion of TSP to PM,0 for Dockery et al, results: Based on analysis of particle size fraction relationships. in Steubenville(Spengler et al. 1986) The lower bound TSP of 220 Lg/m was the peak reported for the Spring 1980 study. A PMs/TSP ratio of about 0.8occurred at a-nearby site on days surrounding this peak. Using lower bound of PMo/ PM, ratio from later year (0.8), the PMo to TSP ratioestimate used is 0.64. The 160 sug/m 3 reflects peak level in Fall 1980 from episode with no significant functional decline noted.( (c) Conversion of Dossen et al. results to PMjo: Both PM indices (Respirable Suspended Particles [RSP) and TSP) reached similar levels..Results suggest TSP levels too low, but PMo levels unlikely to be 'much higher than RSP. Thus RSP = PMo assumed for conditions of higherconcentrations in this study. The 125' A.g/ms entry reflects an excursion occurring 2 days prior to date on which no decrements noted.

Based on this staff assessment of theshort-term epidemiological data, therange of 24-hour PMo levels of interestare 140 to 250 pg/m3. The upper end ofthe range 'reflects the judgment of theAdministrator with regard to themaximum level proposed in 1984 for a24-hour standard, based on his.consideration of the earlier criteria andassessments. Although the recentinformation provides additional supportfor the possibility of effects at lowerlevels, it does not demonstrate thatadverse effects would occur withcertainty at a PMo concentration of 250pg/m3. This level, therefore, remains anappropriate upper bound. The recentdata suggest that the range of levelsunder consideration of alternativestandards can be reduced to 140 g.g/mS,although the original lower bound of 150pg/m3 is within the range of uncertainty

associated with expressing the data asPMo. Neither the studies used to derivethis range nor the more qualitativestudies of effects in other sensitivepopulation groups (e.g. asthmatics) oreffects in controlled human or animalstudies provide convincing scientificsupport for health risks of consequencebelow 140 L~g/m3 in current U.S.atmospheres. These qualitative data, aswell as factors such as aerosolcomposition and exposurecharacteristics, should also beconsidered in evaluating margins ofsafety associated with alternativestandards in the range of 140 Ag/m 3 to150 ;g/m3 .

The amended staff assessment of themore quantitative long-termepidemiological data is summarized inTable 2. Long-term studies are subject toadditional confounding variables that

reduce their sensitivity and makeinterpretation more difficult. The mostimportant new study shows a gradientof responses in children among six U.S.cities that follows the measured gradient.in particulate matter, but responsecomparisons for locations withsomewhat smaller pollution gradientswithin some of these cities do not followthe same patterns. The results of aseparate series of studies on long andintermediate term (2-6 weeks)exposures in a number of U.S. cities(Ostro, 1983, 1987; Hausman et. al, 1984)is more supportive of the possibility ofwithin city effects as comparable U.S.exposure levels. Thus some risk ofeffects is possible at levels somewhatbelow those suggested by the 1982assessment, but it is uncertain given the'potential for confounding present inthese more recent studied.

Federal Register / Vol. •52, No. 126 / Wedne.sday, July 1, 1987 / Rules and Regulations 24663

Table 2. Updated.Staff Assessment of Long-Term Epidemiological Studies

:Measured Measured TSP Levels (pg/m3 • : _ EquivalentBS levels (as . . PM o levels

jig/rm 3) Increased (Ag/im)respiratory

Effects/study Increased disease, Increased n Reducedre sa respiratory respiratory lung Combined"iao small rsiaoy symptoms

duction syiptoms in function in range Combinedre:.. .ducedasen reduction in a ls3 and illnesses children 4 •range 5reduced lung lun adults a nd ildnesesfunction in unoni in children 4 ciden4rnefuncton in function inchildren i adults 2

Effects likely ...................... 2307300 180" .................. .................. >180 >80-90Effects possible ........ .............................. . <230 130-180" .60-150(110) 60"-114 .......................... 60-180 40-90No significant 6 effects noted............... : 80-130 .................................................... 40-114 <60 <40

* Indicates levels used for upper-and lower bound of range.

Study conducted in 1963-65 in Sheffield, England (Lunn et al.. 1967). BS levels (as jig/in') uncertain.2Studies conducted in 1961-73 in Berlin, N.H. (Ferris et al., 1973, 1976). Effects likely level (180 jig/m 3) based on uncertain 2-month

average. Effects in lung function were relatively small.3 Study conducted in 1973 in two Connecticut towns. (Bouhuys et al., 1973). Exposure estimates reflect 1965-73 data in Ansonia. Median

value (110 pig/m 3) used to indicate long-term concentration. No effects on lung function, but some suggestion of effects on respiratorysymptoms.

4 Study. conducted in 1976-1980 in 6 U.S. cities (Ware et al., 19861. Exposure estimates reflect 4-year averages across cities. Comparablepollution/effects gradients not noted within cities.

5Conversion of TSP to PMo equivalents for Berlin, Ansonia studies based on estimated ratio of PM1o/TSP for current U.S. atmospheres(Pace, 1983). The estimated ratio ranged between 0.45 and 0.5. Conversion for six-city study based'on site-specific analysis of particle sizedata (Spengler et al.,1986}.• 6 Ranges reflect gradients in which no significant effects were detected for categories at top. Combined range reflects all columns.

Based on this updated assessment ofthe long-term epidemiological data, thestaff recommends that the range ofannual PMo levels of interest be 40 to 65.Lg/m 3. The upper end. of the range

reflects the judgment of theAdministrator with regard to themaximum level proposed for an annualstandard, based on his consideration ofthe earlier criteria and assessment. Thestaff concludes that this level remains auseful upper bound. The recent dataprompt consideration of a standard levelbelow the previous lower bound (50 lig/m 3) to values as low as 40 ILg/m 3.Uncertain data from one recent study of.six cities suggest that at this level somerisk may remain of respiratory effects inchildren, but no detectable increases inpulmonary function are expected inchildren or adults.

When evaluating margins of safety foran annual standard, it is particularlyimportant to examine the results ofqualitative data from a number ofepidemiological, animal, and air qualitystudies. These suggest concern foreffects not directly evaluated in the.studies used to develop the ranges. Sucheffects include damage to lung tissuescontributing to chronic respiratorydisease, cancer, and prematuremortality. The available scientific datado not suggest major risks for theseeffects categories at current ambientparticle levels in most U.S. areas.Nevertheless, the risk that both fine andcoarse particles may produce theseresponses supports the need to limit

long-term levels of PMo for a variety ofaerosol compositions.

When selecting final standard levels,consideration'should be given to thecombined protection afforded by the 24-hour and annual standards takentogether. For example, a 24-hourstandard at 150 lig/m'wouldsubstantially reduce annual levels in anumber of areas below 50 gg/m '

adding to the protection afforded by anannual standard in areas with higher 24-hour peak to annual mean ratios.

Because of different form, averagingprocedures, size range, and limited PM1odata, precise comparison between theabove ranges of PMo standards and thecurrent primary TSP standards is notpossible. A staff analysis of PMio/TSPratios applied to recent TSP data showsthat the revised lower bounds, takentogether, would result in standardsclearly more stringent than the currentstandards. In various analyses,standards at the lower bound of theprevious range (150,50) have appeared torange from more stringent toapproximately comparable to thepresent primary standards. Standards atthe upper end of the range could,however, result in about a four-folddecrease in the number of areasexceeding the primary standards.

PART 50-NATIONAL PRIMARY ANDSECONDARY AMBIENT AIR QUALITYSTANDARDS

For reasons set forth in the preamble,Part 50 of Chapter 1 of Title 40 of the

Code of Federal Regulations is amendedas follows:

1. The authority citation for Part 50continues to read as follows:

Authority: Secs. 109 and 301(a); Clean AirAct, as amended (42 U.S.C. 7409, 7601 (a)).

2. Section 50.6 is revised to read asfollows:

§ 50.6 National primary and secondaryambient air quality standards for particulatematter.

(a) The level of the national primaryand secondary 24-hour ambient airquality'standards for particulate matteris 150 micrograms per cubic meter (jg/M3 ), 24-hour average concentration. Thestandards are attained when the'expected number of days per calendaryear with a 24-hour averageconcentration above 150 p.g/m s, asdetermined in accordance withAppendix K to this part, is equal to orless than one.

(b) The level of the national primaryand secondary annual standards for.particulate matter is 50 micrograms percubic meter (jug/m"), annual arithmeticmean. The standards are attained whenthe expected annual arithmetic meanconcentration, as determined inaccordance with Appendix K to thispart, is less than or equal to 50 jig/m 3 .

(c) For the purpose of determiningattainment of.the primary and • -secondary standards, particulate mattershall be measured in the ambient air as *PMb (particles with an aerodynamic


diameter less than or equal to a nominal10 micrometers) by:(1) A reference method based on

Appendix I and designated inaccordance with Part 53 of this chapter,or

(2) An equivalent method designatedin accordance with Part 53 of thischapter.

§50.7 [Removed and reserved]3. Section 50.7 is removed ant

reserved.4. In Appendix G, reference 10 is

removed and reserved and section 5.1.1is revised to read as follows:

5.1.1 High-Volume Sampler. Use andcalibrate the sampler as described inAppendix B to this Part.

5. Appendix I is added and reserved.

Appendix I [Reserved]

6. Appendix J is added to read asfollows:

Appendix I-Reference Method for theDetermination of Particulate Matter asPM1o in the Atmosphere

1.0 Applicability.1.1 This method provides for the

measurement of the mass concentration ofparticulate matter with an aerodynamicdiameter less than or equal to a nominal 10micrometers (PM,0) in ambient air over a 24-hour period for purposes of determiningattainment and maintenance of the primaryand secondary national ambient air qualitystandards for particulate matter specified in§ 50.6 of this chapter. The measurementprocess is nondestructive, and the PMosample can be subjected to subsequentphysical or chemical analyses. Qualityassurance procedures and guidance areprovided in Part 58, Appendices A and B, ofthis chapter and in References 1 and 2.

2.0 Principle.2.1 An air sampler draws ambient air at a

constant flow rate into a specially shapedinlet where the suspended particulate matteris inertially separated into one or more sizefractions within the PMo size range. Eachsize fraction in the PM,0 size range is thencollected on a separate filter over thespecified sampling period. The particle sizediscrimination characteristics (samplingeffectiveness and 50 percent cutpoint) of thesampler inlet are prescribed as performancespecifications in Part 53 of this chapter.

2.2 Each filter is weighed (after moistureequilibration) before and after use todetermine the net weight (mass) gain due tocollected PMio. The total volume of airsampled, corrected to EPA referenceconditions (25° C, 101.3 kPa), is determinedfrom the measured flow rate and thesampling time. The mass concentration ofPMo in the ambient air is computed as thetotal mass of collected particles in the PM~osize range divided by the volume of airsampled, and is expressed in micrograms perstandard cubic meter (p.tg/std m3). For PMtosamples collected at temperatures andpressures significantly different from EPA

reference conditions, these correctedconcentrations sometimes differ substantiallyfrom actual concentrations (in microgramsper actual cubic meter), particularly at highelevations. Although hot required, the actualPMo concentration can be calculated fromthe corrected concentration, using theaverage ambient temperature and barometricpressure during the sampling period.

2.3 A method based on this principle willbe considered a reference method only if (a)the associated sampler meets therequirements specified in this appendix andthe requirements in Part 53 of this chapter,and (b) the method has been designated as areference method in accordance with Part 53of this chapter.

3.0 Range.3.1 The lower limit of the mass

concentration range is determined by the.repeatability of filter tare weights, assumingthe nominal air sample volume for thesampler. For samplers having an automaticfilter-changing mechanism, there may be noupper limit. For samplers that do not have anautomatic filter-changing mechanism, theupper limit is determined by the filter massloading beyond which the sampler no longermaintains the operating flow rate withinspecified limits due to increased pressuredrop across the loaded filter. This upper limitcannot be specified precisely because it is acomplex function of the ambient particle sizedistribution and type, humidity, filter type,and perhaps other factors. Nevertheless, allsamplers should be capable of measuring 24-hour PMao mass concentrations of at least 300pg/std m3 while maintaining the operatingflow rate within the specified limits.

4.0 Precision.4.1 The precision of PMo samplers must

be 5 gg/m3 for PMo concentrations below 80Ag/m 3 and 7 percent for PMo concentrationsabove 80 pg/m s, as required by Part 53 of thischapter, which prescribes a test procedurethat determines the variation in the PM,oconcentration measurements of identicalsamplers under typical sampling conditions.Continual assessment of precision viacollocated samplers is required by Part 58 ofthis chapter for PMo samplers used in certainmonitoring networks.

5.0 Accuracy.5.1 Because the size of the particles

making up ambient particulate matter variesover a wide range and the concentration ofparticles varies with particle size, it isdifficult to define the absolute accuracy ofPMo samplers. Part 53 of this chapterprovides a specification for the samplingeffectiveness of PMio samplers. Thisspecification requires that the expected massconcentration calculated for a candidatePMo sampler, when sampling a specifiedparticle size distribution, be within :-10percent of that calculated for an idealsampler whose sampling effectiveness isexplicitly specified. Also, the particle size for50 percent sampling effectivensss is requiredto be 10±0.5 micrometers. Other :specifications related to accuracy apply toflow measurement and calibration, filtermedia, analytical (weighing) procedures, andartifact. The flow rate accuracy of PMiosamplers used in certain monitoring networksis required by Part 58 of this chapterto beassessed periodically via flow rate audits.

6.0 Potential Sources of Error.6.1 . Volatile Particles. Volatile particles

collected on filters are often lost duringshipment and/or storage of the filters prior tothe post-sampling weighing 3. Althoughshipment or st6rage of loaded filters issometimes unavoidable, filters should bereweighed as soon as practical to minimizethese losses.

6.2 Artifacts.. Positive errors in PM~oconcentration measurements may result fromretention of gaseous species on filters 'Such errors include the retention of sulfurdioxide and nitric acid. Retention of sulfurdioxide on filters, followed by oxidation tosulfate, is referred to as artifact sulfateformation, a phenomenon which increaseswith increasing filter alkalinity 6. Little or noartifact sulfate formation should occur usingfilters that meet the alkalinity specification insection 7.2.4. Artifact nitrate formation,resulting primarily from retention of nitricacid, occurs to varying degrees on many filtertypes, including glass fiber, cellulose ester,and.many quartz fiber filters ', 9 *o. Lossof true atmospheric particulate nitrate duringor following sampling may also occur due todissociation or chemical reaction. Thisphenomenon has been observed on Teflon"filters 8 and inferred for quartz fiberfilters 11 12. The magnitude of nitrate artifacterrors in PMo mass concentrationmeasurements will vary with location andambient temperature; however, for mostsampling locations, these errors are expectedto be small.

6.3 Humidity. The effects of ambienthumidity on the sample are unavoidable. Thefilter equilibration procedure in section 9.0 isdesigned to minimize the effects of moistureon the filter medium.

6.4 Filter Handling. Careful handling offilters between presampling andpostsampling weighings is necessary to avoiderrors due to damaged filters or loss ofcollected particles from the filters. Use of afilter cartridge or cassette may reduce themagnitude of these errors. Filters must alsomeet the integrity specification in section7.2.3.

6.5 Flow Rate Variation. Variations in thesampler's operating flow rate may alter theparticle size discrimination characteristics ofthe sampler inlet. The magnitude of this errorwill depend on the sensitivity of the inlet tovariations in flow rate and on the particledistribution in the atmosphere during thesampling period. The use of a flow controldevice (section 7.1.3) is required to minimizethis error.

6.6 Air Volume Determination. Errors inthe air volume determination may result fromerrors in the flow rate and/or sampling timemeasurements. The flow control deviceserves to minimize errors in the flow ratedetermination, and an elapsed time meter(section 7.1.5) is required to minimize theerror in the sampling time measurement.

7.0 Apparatus.7.1 PMo Sampler.7.1.1 The sampler shall be designed to:a. Draw the air sample into the sampler

inlet and through the particle collection filterat a uniform face velocity.


b. Hold and seal the filter in a horizontaiposition so that sample air is drawndownward through the filter.

c. Allow the filter to be installed andremoved conveniently.

d. Protect the filter and sampler fromprecipitation and prevent Insects and otherdebris from being sampled.

e. Minimize air leaks that would causeerror in the measurement of the air volumepassing through the filter.

f. Discharge. exhaust air at a sufficientdistance from the sampler inlet to minimizethe sampling of exhaust air.

g. Minimize the collection of dust from thesupporting surface.

7.1.2 'The sampler shall have a sample airinlet system that, when operated within aspecified flow rate range, provides particlesize discrimination characteristics meeting allof the applicable performance specificationsprescribed in Part 53 of this chapter. Thesampler inlet shall-show no significant winddirection dependence. The. latter requirementcan generally be satisfied by an inlet shapethat is circularly symmetrical about a verticalaxis.

7.1.3 The sampler shall have a flowcontrol device capable of maintaining thesampler's operating flow rate within the flowrate limits specified for the sampler inlet overnormal variations in line voltage and filterpressure drop.

7.1.4 The sampler shall provide a meansto measure the total flow rate during thesampling period. A continuous flow recorderis recommended but not required. The flowmeasurement device shall be accurate to -2percent.

7.1.5 A timing/control device capable ofstarting and stopping the sampler shall beused to obtain a sample collection period of24 ±1 hr (1,440 -60 min). An elapsed timemeter, accurate to within ±15 minutes, shallbe used to measure sampling time. This meteris optional for samplers with continuous flowrecorders if the sampling time measurementobtained by means of the recorder meets the-±15 minute accuracy specification.

7.1.6 The sampler shall have anassociated operati6n or instruction manual asrequired by Part 53 of this chapter whichincludes detailed instructions on thecalibration, operation, and maintenance ofthe sampler.

7.2 Filters.7.2.1 Filter Medium. No commercially

available filter medium is ideal in all respectsfor all samplers. The user's goals in samplingdetermine the relative importance of variousfilter characteristics (e.g., cost, ease ofhandling, physical and chemicalcharacteristics, etc.) and, consequently,determine the choice among acceptablefilters. Furthermore, certain types of filtersmay not be suitable for use with somesamplers, particularly under heavy loadingconditions (high mass concentrations).because of high or rapid increase in the filterflow resistance that would exceed thecapability of the sampler's flow controldevice. However, samplers equipped withautomatic filter-changing mechanisms mayallow use of these types of filters. Thespecifications given below are minimumrequirements to ensure acceptability of the

filter medium for measurement of PMo massconcentrations. Other filter evaluationcriteria should be considered to meetindividual sampling and analysis objectives.

7.2.2 Collection Efficiency. >99 percent.as measured by the DOP test (ASTM-2986)with 0.3 pm particles at the sampler'soperating face velocity.

7.2.3 Integrity. -15 pg/m3 (assumingsampler's nominal 24-hour air samplevolume). Integrity is measured as the PMioconcentration ,equivalent corresponding tothe average difference between the initialand the final weights of a random sample oftest filters that are weighed and handledunder actual or simulated samplingconditions, but have no air sample passedthrough them (i.e., filter blanks). As aminimum, the test procedure must includeinitial equilibration and weighing, installationon an inoperative sampler, removal from thesampler, and final equilibration andweighing.

7.2.4. Alkalinity. <25 microequivalents/gram of filter, as measured by the proceduregiven in Reference 13 following at least twomonths.storage in a clean environment (freefrom contamination by acidic gases) at roomtemperature'and humidity.

7.3 * Flow Rate Transfer Standard. Theflow rate transfer standard must be suitablefor the.sampler's operating flow rate andmust be calibrated against a primary flow orvolume standard that is traceable to theNational Bureau of Standards (NBS). Theflow rate transfer standard must be capableof measuring the sampler's operating flowrate with an accuracy of -2 percent.

7.4 ilter Conditioning Environment.7.4.1 Temperature range: 150 to 30* C.7.4.2 Temperature control: -3 ° C.7.4.3 Humidity range: 20% to 45% RH.7.4.4 Humidity control: -5% RH.7.5 Analytical Balance. The .analytical

balance must be suitable for weighing thetype and size of filters required by thesampler. The range and sensitivity requiredwill depend on the filter tare weights andmass loadings. Typically, an analyticalbalance with a sensitivity of 0.1 mg isrequired for high volume samplers (flow rates>0.5 ms/min). Lower volume samplers (flowrates <0.5 m3/min) will require a moresensitive balance.

8.0 Calibration.8.1 General Requirements.8.1.1 Calibration of the sampler's flow

measurement device Is required to establishtraceability of subsequent flowmeasurements to a primary standard. A flowrate transfer standard calibrated against aprimary flow or volume standard shall beused to calibrate or verify the accuracy of thesampler's flow measurement device.

8.1.2 Particle size discrimination byinertial separation requires that specific airvelocities be maintained in the sampler's airinlet system. Therefore, the flow rate throughthe sampler's inlet must be maintainedthroughout the sampling period within thedesign flow rate range specified by themanufacturer. Design flow rates are specifiedas actual volumetric flow rates, measured atexisting conditions of temperature andpressure (Q.). In contrast, massconcentrations of PM~o are computed using

flow rates corrected to EPA referenceconditions of temperature and pressure (Q.,d).'

8.2 Flow Rate Calibration Procedure. :8.2.1 PMo samplers employ various types

of flow control and, flow measurementdevices. The specific procedure used for flowrate calibration or verification will varydepending on the type of flow controller andflow indicator employed. Calibration in termsof actual volumetric flow rates (Q.) isgenerally recommended, but other measuresof flow rate (e.g., Qstd) may be used providedthe requirements of section 8.1-are met. Thegeneral procedure given here is based onactual volumetric flow units (Q.) and servesto illustrate the steps involved in thecalibration of a PM,0 sampler. Consult thesampler manufacturer's instruction manualand Reference 2 for specific guidance oncalibration. Reference 14, provides additionalinformation on the use of the commonly usedmeasures of flow rate and theirinterrelationships.

8.2.2 Calibrate the flow rate transfer.standard against a primary flow or volumestandard traceable to NBS. Establish acalibration relationship (e.g., an equation orfamily of curves) such that traceability to theprimary standard is accurate to within 2percent over the expected range of ambientconditions (i.e., temperatures and pressures)under which the transfer standard will beused. Recalibrate the transfer standardperiodically.

8.2.3. Following the samplermanufacturer's instruction manual, removethe sampler inlet and connect the flow ratetransfer standard to the sampler such that thetransfer standard accurately measures the.sampler's flow rate. Make sure there are noleaks between the transfer standard and thesampler.

8.2.4 Choose a minimum of three flowrates (actual m/min), spaced over theacceptable flow rate range specified for theinlet (see 7.1.2) that can be obtained by •suitable adjustment of the sampler flow rate.In accordance with the samplermanufacturer's instruction manual, obtain orverify the calibration relationship betweenthe flow rate (actual m3/min) as indicated bythe transfer standard and the sampler's flowindicator response. Record the ambienttemperature and barometric pressure.Temperature and pressure corrections tosubsequent flow indicator readings may berequired for certain types of flowmeasurement devices. When such correctionsare necessary, correction on an individual ordaily basis is preferable. However, seasonalaverage temperature and average barometricpressure for the sampling site may beincorporated into the sampler calibration toavoid daily corrections. Consult the samplermanufacturer's Instruction manual andReference 2 for additional guidance.

8.2.5 Following calibration, verify that thesampler is operating at its design flow rate(actual m3 /min) with a clean filter in place.

8.2.6 Replace the sampler inlet.9.0 Procedure.9.1 The sampler shall be operated in

accordance with the specific guidanceprovided in the sampler manufacturer'sinstruction manual and in Reference Z The

24665

24666 Federal Register / Vol. 52, No. 126 / Wednesday, July 1, 1987/ Rules and Regulations

general procedure given here assumes thatthe sampler's flow rate calibration is basedon flow rates at ambient conditions (Q.) andserves to illustrate the steps involved in theoperation of a PMo sampler.

9.2 Inspect each filter for pinholes,particles, and other imperfections. Establish afilter information record and assign anidentification number to each filter.

9.3 Equilibrate each filter in theconditioning environment (see 7.4) for at least24 hours.

9.4 Following equilibration, weigh eachfilter and record the presampling weight withthe filter identification number.

9.5 Install a preweighed filter in thesampler following the instructions providedin the sampler manufacturer's instructionalmanual.

9.6 Turn on the sampler and allow It toestablish run-temperature conditions. Recordthe flow indicator reading and, if needed, theambient temperature and barometricpressure. Determine the sampler flow rate(actual m3/min) in accordance with theinstructions provided in the samplermanufacturer's instruction manual. NOTE.-No onsite temperature or pressuremeasurements are necessary if the sampler'sflow indicator does not require temperatureor pressure corrections or if seasonal averagetemperature and average barometric pressurefor the sampling site are incorporated into thesampler calibration (see step 8.2.4). Ifindividual or daily temperature and pressurecorrections are required, ambienttemperature and barometric pressure can beobtained by on-site measurements or from anearby weather station. Barometric pressurereadings obtained from airports must bestation pressure, not corrected to sea level,and may need to be corrected for differencesin elevation between thi sampling site andthe airport.

9.7 If the flow rate is outside theacceptable range specified by themanufacturer, check for leaks, and Ifnecessary, adjust the flow rate to thespecified setpoint. Stop the sampler.

9.8 Set the timer to start and stop thesampler at appropriate times. Set the elapsedtime meter to zero or record the initial meterreading.

9.9 Record the sample information (sitelocation or identification number, sampledate, filter identification number, andsampler model and serial number).

9.10 Sample for 24±1 hours.9.11 Determine and record the average

flow rate (Q.) in actual m3/min for thesampling period in accordance with theinstructions provided In the samplermanufacturer's instruction manual. Recordthe elapsed time meter final reading and, ifneeded, the average ambient temperature andbarometric pressure for the sampling period(see note following step 9.6).

9.12 Carefully remove the filter from thesampler, following the samplermanufacturer's instruction manual. Touchonly the outer edges of the filter.

9.13 Place the filter In a protective holderor container (e.g., petri dish, glassineenvelope, or manila folder].

9.14 Record any factors such asmeteorological conditions, construction

activity, fires or dust storms, etc., that mightbe pertinent to the measurement on the filterinformation record.

9.15 Transport the exposed sample filterto the filter conditioning environment as soonas possible for equilibration and subsequentweighing.

9.16 Equilibrate the exposed filter in theconditioning environment for at least 24hours under the same temperature andhumidity conditions used for presamplingfilter equilibration (see 9.3).

9.17 Immediately after equilibration,reweigh the filter and record thepostsampling weight with the filteridentification number.

10.0 Sampler Maintenance.10.1 The PMo sampler shall be

maintained in strict accordance with themaintenance procedures specified in thesampler manufacturer's instruction manual.

11.0 Calculations.11.1 Calculate the average flow rate over

the sampling period corrected to EPAreference conditions as Q.,& When thesampler's flow indicator is calibrated inactual volumetric units (Q.), Q,~t is calculatedas:Qtd= Q. X (P.,/T.,)(Tej/Pa)whereQd =average flow rate at EPA reference

conditions, std m3/min;Q. =average flow rate at ambient conditions,

m3/min:P.,= average barometric pressure during the

sampling period or average barometricpressure for the sampling site, kPa (ormm Hg);

Ta =average ambient temperature during thesampling period or seasonal averageambient temperature for the samplingsite, K;

Ttd= standard temperature, defined as 298 K;P,.= standard pressure, defined as 101.3 kPa

(or 760 mm Hg).11.2 Calculate the total volume of air

sampled as:V.td = Qtd X twhereV"= total air sampled in standard volume

units, std m3 ;t =sampling time, min.

11.3 Calculate the PM~o concentration as:PMlo = (Wf-Wi ) 1O/VtdwherePMo = mass concentration of PMo, pg/std

ma:

Wf, Wi= final and initial weights of filtercollecting PM1o particles, g;

106 =conversion of g to pfg.Note.-If more than one size fraction in the

PMo size range is collected by the sampler,the sum of the net weight gain by eachcollection filter [Y(Wf-W)] is used tocalculate the PMio mass concentration.

12.0 References.1. Quality Assurance Handbook for

Air Pollution Measurement Systems,Volume I, Principles. EPA-600/9-76-005,March 1976. Available from CERI, ORDPublications, U.S. EnvironmentalProtection Agency, 26 West St. ClairStreet, Cincinnati, Ohio 45268.

2. Quality Assurance Handbook forAir Pollution Measurement Systems,Volume II, Ambient Air SpecificMethods. EPA-600/4-77-027a, May 1977.Available from CERI, ORD Publications,U.S. Environmental Protection Agency,26 West St. Clair Street, Cincinnati,Ohio 45268.

3. Clement, R.E., and F.W. Karasek.Sample Composition Changes inSampling and Analysis of OrganicCompounds in Aerosols. Int. J. Environ.Analyt. Chem., 7:109, 1979.

4. Lee, R.E.; Jr., and J. Wagman. ASampling Anomaly in the Determinationof Atmospheric Sulfate Concentration.Amer. Ind. Hyg. Assoc. 1., 27:266,1968.

5. Appel, B.R., S.M. Wall, Y. Tokiwa,and M. Haik. Interference Effects inSampling Particulate Nitrate in AmbientAir. Atmos. Environ., 13:319, 1979.

6. Coutant, R.W. Effect ofEnvironmental Variables on Collectionof Atmospheric Sulfate. Environ. Sci.Technol., 11:873, 1977.

7. Spicer, C.W., and P. Schumacher.Interference in Sampling AtmosphericParticulate Nitrate. Atmos. Environ.,11:873, 1977.

8. Appel, B.R., Y. Toklwa, and M.Haik. Sampling of Nitrates in AmhientAir. Atmos. Environ., 15:283, 1981.

9. Spicer, C.W., and P.M. Schumacher.Particulate Nitrate: Lahoratory and FieldStudies of Major Sampling Interferences.Atmos. Environ., 13:543,1979.

10. Appel, B.R. Letter to Larry Purdue,U.S. EPA, Environmental Monitoringand Support Laboratory. March 18, 1982,Docket No. A-82--37, 11I--1.

11. Pierson, W.R., W.W. Brachaczek,T.J. Korniski, T.J. Truex, and J.W. Butler.Artifact Formation of Sulfate, Nitrate,and Hydrogen Ion on Backup Filters:Allegheny Mountain Experiment. J. AirPollut. Control Assoc., 30:30, 1980.

12. Dunwoody, C.L. Rapid Nitrate LossFrom PMo Filters. I. Air Pollut. ControlAssoc., 36:817, 1986.

13. Harrell, R.M. Measuring theAlkalinity of Hi-Vol Air Filters. EMSL/RTP-SOP-QAD-534, October 1985.Available from the U.S. EnvironmentalProtection Agency, EMSL/QAD,Research Triangle Park, North Carolina27711.

14. Smith, F., P.S. Wohlschlegel, R.S.C.Rogers, and D.J. Mulligan. Investigationof Flow Rate Calibration ProceduresAssociated With the High VolumeMethod for Determination of SuspendedParticulates. EPA-600/4-78-047, U.S.Environmental Protection Agency,Research Triangle Park, North Carolina27711, 1978.

7. Appendix K is added to read asfollows:


Appendix K-Interpretation of theNational Ambient Air Quality Standardsfor Particulate Matter

1.0 General.This appendix explains the computations

necessary for analyzing particulate matterdata to determine attainment of the 24-hourand annual standards specified in 40 CFR50.6. For the primary and secondarystandards, particulate matter is measured inthe ambient air as PMio (particles with anaerodynamic diameter less than or equal to anominal 10 micrometerslby a referencemethod based on Appendix I of this part anddesignated in accordance with Part 53 of thischapter, or by an equivalent methoddesignated in accordance with Part 53 of thischapter. The required frequency ofmeasurements is specified in Part 58 of thischapter.

Several terms used throughout thisappendix must be defined. A "daily value"for PMo refers to the 24-hour averageconcentration of PMo calculated or measuredfrom midnight to midnight (local time). Theterm "exceedance" means a daily value thatis above the level of the 24-hour standardafter rounding to the nearest 10 pg/m3 (i.e.,values ending in 5 or greater are to berounded up). The term "average" refers to anarithmetic mean. All particulate matterstandards are expressed in terms of expectedannual values: expected number ofexceedances per year for the 24-hourstandard and expected annual arithmeticmean for the annual standards. The"expected annual value" is the numberapproached when the annual values from anincreasing number of years are averaged. inthe absence of long-term trends in emissionsor meteorological conditions. The term "year"refers to a calendar year.

Although the discussion In this appendixfocuses on monitored data, the sameprinciples apply to modeling data, subject toEPA modeling guidelines.

2.0 Attainment Determinations.2.1 24-Hour Primary and Secondary

Standards.Under 40 CFR 50.6(a) the 24-hour primary

and secondary standards are attained whenthe expected number of exceedances per yearat each monitoring site is less than or equalto one. In the simplest case, the number ofexpected exceedances at a site is determinedby recording the number of exceedances ineach calendar year and then averaging themover the past 3 calendar years. Situations inwhich 3 years of data are not available andpossible adjustments for unusual events ortrends are discussed in Sections 2.3 and 2.4.Further, when data for a year are incomplete.it is necessary to compute an estimatednumber of exceedances for that year byadjusting the observed number ofexceedances. This procedure, performed bycalendar quarter, is described in Section 3.The expected number of exceedances is thenestimated by averaging the individual annualestimates for the past 3 years.

The comparison with the allowableexpected exceedance rate of one per year ismade in terms of a number rounded to thenearest tenth (fractional values equal to orgreater than 0.05 are to be rounded up: e.g.,

an exceedance rate of 1.05 would be roundedto 1.1, which is the lowest rate fornonattainment).

2.2 Annual Primary and SecondaryStandards.

Under 40 CFR 50.6(b), the annual primaryand secondary standards are attained whenthe expected annual arithmetic mean PM[oconcentration is less than or equal to thelevel of the standard. In the simplest case, theexpected annual arithmetic mean isdetermined by averaging the annualarithmetic mean PMo concentrations for thepast 3 calendar years. Because of thepotential for incomplete data and thepossible seasonality in PMo concentrations,the annual mean shall be calculated byaveraging the four quarterly means of PM,,oconcentrations within the calendar year. Theformulas for calculating the annual arithmeticmean are given in Section 4. Situations inwhich 3 years of data are not available andpossible adjustments for unusual events ortrends are discussed in Sections 2.3 and 2.4.The expected annual arithmetic mean isrounded to the nearest I pg/m3 beforecomparison with the annual primarystandard (fractional values equal to orgreater than 0.5 are to be rounded up).

2.3 Data Requirements.40 CFR 58.13 specifies the required

minimum frequency of sampling for PMo. Forthe purposes of making comparisons with theparticulate matter standards, all dataproduced by National Air MonitoringStations (NAMS), State and Local AirMonitoring Stations (SLAMS) and other sitessubmitted to EPA in accordance with the Part58 requirements must be used, and aminimum of 75 percent of the scheduled PMosamples per quarter are required.

To demonstrate attainment of either theannual or 24-hour standards at a monitoringsite, the monitor must provide sufficient datato perform the required calculations ofSections 3 and 4. The amount of datarequired varies with the sampling frequency.data capture rate and the number of years ofrecord. In all cases, 3 years of representativemonitoring data that meet the 75 percentcriterion of the previous paragraph should heutilized, if available, and would suffice. Morethan 3 years may be considered, if alladditional representative years of datameeting the 75 percent criterion are utilized.Data not meeting these criteria may alsosuffice to show attainment; however, suchexceptions will have to be approved by theappropriate Regional Administrator inaccordance with EPA guidance.

There are less stringent data requirementsfor showing that a monitor has failed anattainment test and thus has recorded aviolation of the particulate matter standards.Although it Is generally necessary to meet theminimum 75 percent data capturerequirement per quarter to use thecomputational formulas described In Sections3 and 4, this criterion does not apply whenless data is sufficient to unambiguouslyestablish nonattainment. The followingexamples illustrate how nonattainment canbe demonstrated when a site fails to meet thecompleteness criteria. Nonattainment of the24-hour primary standards can be establishedby (a) the observed annual number of

exceedances (e.g. four observed exceedancesin a single year), or by (b) the estimatednumber of exceedances derived from theobserved number of exceedances and therequired number of scheduled samples (e.g.two observed exceedances with every otherday sampling). Nonattainment of the annualstandards can be demonstrated on the basisof quarterly mean concentrations developedfrom observed data combined with one-halfthe minimum detectable concentrationsubstituted for missing values. In both cases.expected annual values must exceed thelevels allowed by the standards.

2.4 Adjustment for Exceptional Eventsand Trends.

An exceptional event is an uncontrollableevent caused by natural sources ofparticulate matter or an event that is notexpected to recur at a given location.Inclusion of such a value in the computationof exceedances or averages could result ininappropriate estimates of their respectiveexpected annual values. To reduce the effectof unusual events, more than 3 years ofrepresentative data may be used.Alternatively, other techniques, such as theuse of statistical models or the use ofhistorical data could be considered so thatthe event may be discounted or weightedaccording to the likelihood that it will recur.The use of such techniques is subject to theapproval of the appropriate RegionalAdministrator in accordance with EPAguidance.

In cases where long-term trends inemissions and air quality are evident,mathematical techniques should be appliedto account for the trends to ensure that theexpected annual values are notinappropriately biased by unrepresentativedata. In the simplest case, if 3 years of dataare available under stable emissionconditions, this data should be used. In theevent of a trend or shift in emission patterns.either the most recent representative year(s)could be used or statistical techniques ormodels could be used in conjunction withprevious years of data to adjust for trends.The use of less than 3 years of data, and anyadjustments are subject to the approval of theappropriate Regional Administrator inaccordance with EPA guidance.

3.0 Computational formulas for the 24-hour standard.

3.1 Estimating Exceedances for a year.If PMo sampling is scheduled less

frequently than every day, or if somescheduled samples are missed, a PMo valuewill not be available for each day of the year.To account for the possible effect ofincomplete data, an adjustment must bemade to the data collected at each monitoringlocation to estimate the number ofexceedances in a calendar year. In thisadjustment, the assumption is made that thefraction of missing values that would haveexceeded the standard level is identical tothe fraction of measured values above thislevel. This computation is to be made for allsites that are scheduled to monitorthroughout the entire year and meet theminimum data requirements of Section 2.3.Because of possible seasonal imbalance, thisadjustment shall be applied on a quarterly

24667


basis. The estimate of the expected numberof exceedances for the quarter is equal to theobserved number of exceedances plus anincrement associated with the missing data.The following formula must be used for thesecomputations:

e,=v,+1[(v/nJ)X(N 0-nq)]=vxNIn 0 Ill

wheree.= the estimated number of exceedances for

calendar quarter q, v.= the observed number of exceedances for

calendar quarter q;N,=the number of days in calendar quarter

q, . In1 =the number of days in calendar quarter q

with PMo, and.=the index for calendar quarter, q=l, 2, 3 or

4.The estimated numberof exceedances for a

calendar quarter must be rounded to thenearest hundredth (fractional values equal toor greater than 0.005 must be rounded up).

The estimated number of exceedances forthe years, e, is the sum of the estimates foreach calendar quartei. ,

If the estimated exceedances for the other 3calendar quarters in the year were 2.30, 0.0and 0.0, then, using formula [21, the estimatednumber of exceedances for the year is2.36+2.30+0.0+0.0 which equals 4.66 or 4.7.If no exceedances were observed for the 2previous years, then the expected number ofexceedances is estimated by (1/3)X(4.7+0+0)=1.57 or 1.6. Since 1.6 exceedsthe allowable number of expectedexceedances, this monitoring site would failthe attainment test.

,Example 2

In this example, everyday sampling wasinitiated following the first observedexceedance as required by 40 CFR § 58.13.Accordingly, the first observed exceedancewould not be adjusted for incompletesampling. During the next three quarters, 1.2exceedances were estimated. In this case, theestimated exceedances for the year would be1.0+1.2+0.0+0.0 which equals 2.2. If, asbefore, no exceedances were observed forthe two previous years, then the estimatedexceedances for the 3-year period wouldthen be (1/3) X(2.2+0.0+0.0)=0.7 and the

monitoring site would not fail the attainmenttest;

3.2 Adjustments for Non-ScheduledSampling Days.

If a systematic sampling schedule is usedand sampling is performedon days inaddition to the days specified by thesystematic sampling schedule, e.g., duringepisodes of high pollution, then anadjustment must be made in the formula forthe estimation of exceedances. Such anadjustment is needed to eliminate the bias inthe estimate of the quarterly and annualnumber of exceedances that would occur ifthe chance of an exceedance is different forscheduled than for non-scheduled days, aswould be the case with episode sampling.The required adjustment treats thesystematic sampling schedule as a stratifiedsampling plan. If the period from onescheduled sample until the day preceding thenext scheduled sample is defined as asampling stratum, then there is one stratumfor each scheduled sampling day. An averagenumber of observed exceedances iscomputed for each bf these sampling strata.With nonscheduled sampling days, theestimated number of exceedances is definedas

le=- e, [21'0 ---1 " " " ,

The estimated' number of exceedances for asingle year must be' rounded to one decimal-place (fractional values equal to or greater'than 0.05 are to be rounded up).'The expectednumber of exceedances is then estimated byaveraging the individual annual estimates forthe most recent 3 or more representativeyears of data. The expected number ofexceedances must be rounded to one decimalplace (fractional values equal to or greaterthan 0.05 are to be rounded up).

The adjustment for incomplete data willnot be necessary for monitoring or modelingdata which constitutes a complete record, i.e.,365 days per year.

To reduce the potential for overestimatingthe number of expected exceedances, thecorrection for missing data will not berequired for a calendar quarter in which thefirst observed exceedance has occurred if: (a)there was only one exceedance in'thecalendar quarter, (b) everyday sampling issubsequently initiated and maintained for 4calendar quarters in accordance with 40 CFR§ 58.13 and, (c) data capture of 75 percent isachieved during the required period ofeveryday sampling. In addition, if the firstexceedance is observed in a calendar quarterin which the monitor is already samplingevery day, no adjustment for missing datawill be made to the first exceedance if a 75percent data capture rate was achieved in thequarter in which it was observed.

Example 1During a particular calendar quarter, 39 out

of a possible 92 samples were recorded, withone observed exceedance of the 24-hourstandard. Using formula [1], the estimatednumber of exceedances for the quarter ise. = 1 X 92/39 =2.359 or 2.36

Wheree.= the estimated number of exceedances for

the quarter.N0 =the number of days in the quarter,mn-= the number of strata with samples

during the quarter,v,=the number of observed exceedances In

stratum j, andkj= the number of actual samples in stratum J.

Note that if only one sample value isrecorded in each stratum, then formula [3]reduces to formula [1].

Example 3

A monitoring site samples according to asystematic sampling schedule of one sampleevery 6 days, for a total of 15 scheduledsamples in a quarter out of a total of 92possible samples. During one 6-day period,potential episode levels of PM1 o weresuspected, so 5 additional samples weretaken. One of the regular scheduled sampleswas missed, so a total of 19 samples in 14sampling strata were measured. The one 6-day sampling stratum with 6 samplesrecorded 2 exceedances. The remainder ofthe quarter with one sample per stratumrecorded zero exceedances. Using formula[3], the estimated number of exceedances forthe quarter ise,=(92/14)X(2/6+0+.. .+0)=2.19

4.0 Computational Formulas for AnnualStandards.

4.1 Calculation of the Annual ArithmeticMean.

An annual arithmetic mean value for PMois determined by averaging the quarterlymeans for the 4 calendar quarters of the year.

The following formula is to be used forcalculation of the mean for a calendarquarter:

n _3E=(l/nx -- xi

where5,= the quarterly mean concentration for

quarter q, q=1, 2, 3, or 4,na = the number of samples in the quarter,

andXi the ith concentration value recorded in

''the quarter.

The quarterly mean, expressed in Ag/m3,must be rounded to the nearest tenth(fractional values of 0.05 should be roundedup).

The annual mean is calculated by using thefollowing formula:

4'1

0J=1

where'x-=-the annual mean, andK,=the mean for calendar quarter q.

The average, of quarterly means must berounded to the nearest tenth (fractionalvalues of 0.05 should be rounded up).

The use of quarterly averages to computethe annual average will not be necessary for

p.=(N m/ ) x. (vm/. )


monitoring or modeling data which results in Example 4a complete record, i.e., 365 days per year. Using formula (41, the quarterly means are

The expected annual mean is estimated as Ulate forea calenda quarter. If thethe average of three or more annual means. calculated for each calendar quarter. If theThis multi-year estimate, expressed in jig/m 3, quarterly means are 52.4, 75.3, 82.1, and 63.2shall be rounded to the nearest integer for /g/m 3, then the annual means iscomparison with the annual standard(fractional values of 0.5 should be roundedup).

T= (1/4)X(52.4+75.3+82.1+63.2= 68.25 or 68.3

4.2 Adjustments for Non-scheduledSampling Days.

An adjustment in the calculation of theannual mean is needed if sampling isperformed on days in addition to the daysspecified by the systematic samplingschedule. For the same reasons given in thediscussion of estimated exceedances (Section3.2), the quarterly averages would becalculated by using the following formula:

=(a/m.x n (xu/kj) 16j=l i=l

wherei= the quarterly mean concentration for

quarter q, q=1, 2, 3, or 4,xu = the ith concentration value recorded in

stratum j,

kj=the number of actual samples in stratum j,and

m,= the numberof strata with data in thequarter.

If one sample value is recorded in eachstratum, formula [6] reduces to a simplearithmetic average of the observed values asdescribed by formula [4].

Example 5

During one calendar quarter, 9observations were recorded. These sampleswere distributed among 7 sampling strata,with 3 observations in one stratum. Theconcentrations of the 3 observations in thesingle stratum were 202. 242. and 180 ,g/m3.The remaining 6 observed concentrationswere 55. 68, 73, 92. 120. and 155 p.g/m.Applying the weighting factors specified informula [6], the quarterly mean is

i,= (1/7) X [(1/3) X (202+ 242+180) + 55 +68+73+92+120+155 =110.1

Although 24-hour measurements arerounded to the nearest 10 9g/m3 -fordeterminations of exceedances: of the 24-hourstandard, note that these values are rounded

to the nearest 1 pg/m 3 for the calculation.ofmeans.

[FR Doc. 87-13707 Filed 6-30-87; 8:45 am],BILLING CODE 6560-50-M

24669

1987 rules and regulations - us epa€¦ · 24634 federal register / vol. 52, no. 126 / wednesday,...

Documents