Revisions to Ambient Air Monitoring Regulations

Federal RegisterOct 17, 2006

Ask Donna

What actually matters in this document.

Text

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 53 and 58

[EPA-HQ-OAR-2004-0018; FRL-8227-2]

RIN 2060-AJ25

Revisions to Ambient Air Monitoring Regulations

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Final rule.

SUMMARY:

The EPA is issuing final amendments to the ambient air monitoring requirements for criteria pollutants. The purpose of the amendments is to enhance ambient air quality monitoring to better serve current and future air quality management and research needs. The final amendments establish limited ambient air monitoring requirements for thoracic coarse particles in the size range of PM

10−2.5

to support continued research into these particles' distribution, sources, and health effects. The ambient air monitoring amendments also require each State to operate one to three monitoring stations that take an integrated, multipollutant approach to ambient air monitoring. In addition, the final amendments modify the general monitoring network design requirements for minimum numbers of ambient air monitors to focus on populated areas with air quality problems and to reduce significantly the requirements for criteria pollutant monitors that have measured ambient air concentrations well below the applicable National Ambient Air Quality Standards. These amendments also revise certain provisions regarding monitoring network descriptions and periodic assessments, quality assurance, and data certifications. A number of the amendments relate specifically to PM

2.5

, revising the requirements for reference and equivalent method determinations (including specifications and test procedures) for fine particle monitors.

DATES:

This final rule is effective on December 18, 2006.

ADDRESSES:

The EPA has established a docket for this action under Docket ID No. EPA-HQ-OAR-2004-0018. All documents in the docket are listed in the

http://www.regulations.gov

index. Although listed in the index, some information is not publicly available, e.g., confidential business information or other information whose disclosure is restricted by statute. Certain other material, such as copyrighted material, will be publicly available only in hard copy. Publicly available docket materials are available either electronically in

http://www.regulations.gov

or in hard copy at the Revisions to the Ambient Air Monitoring Regulations Docket, EPA/DC, EPA West, Room B102, 1301 Constitution Ave., NW., Washington, DC. The Public Reading Room is open from 8:30 a.m. to 4:30 p.m., Monday through Friday, excluding legal holidays. The telephone number for the Public Reading Room is (202) 566-1744, and the telephone number for the Air Docket is (202) 566-1742.

Note:

The EPA Docket Center suffered damage due to flooding during the last week of June 2006. The Docket Center is continuing to operate. However, during the cleanup, there will be temporary changes to Docket Center telephone numbers, addresses, and hours of operation for people who wish to visit the Public Reading Room to view documents. Consult EPA's

Federal Register

notice at 71 FR 38147 (July 5, 2006) or the EPA Web site at

http://www.epa.gov/epahome/dockets.htm

for current information on docket status, locations, and telephone numbers.

FOR FURTHER INFORMATION CONTACT:

For general questions concerning the final amendments, please contact Mr. Lewis Weinstock, U.S. EPA, Office of Air Quality Planning and Standards, Air Quality Assessment Division, Ambient Air Monitoring Group (C304-06), Research Triangle Park, North Carolina 27711; telephone number: (919) 541-3661; fax number: (919) 541-1903; e-mail address:

weinstock.lewis@epa.gov

. For technical questions, please contact Mr. Tim Hanley, U.S. EPA, Office of Air Quality Planning and Standards, Air Quality Assessment Division, Ambient Air Monitoring Group (C304-06), Research Triangle Park, North Carolina 27711; telephone number: (919) 541-4417; fax number: (919) 541-1903; e-mail address:

hanley.tim@epa.gov

SUPPLEMENTARY INFORMATION:

I. General Information

A. Does this action apply to me?

Categories and entities potentially regulated by this action include:

Category

NAICS code

Examples of regulated entities

Industry

334513, 541380

Manufacturer, supplier, distributor, or vendor of ambient air monitoring instruments; analytical laboratories or other monitoring organizations that elect to submit an application for a reference or equivalent method determination under 40 CFR part 53.

Federal government

924110

Federal agencies (that conduct ambient air monitoring similar to that conducted by States under 40 CFR part 58 and that wish EPA to use their monitoring data in the same manner as State data) or that elect to submit an application for a reference or equivalent method determination under 40 CFR part 53.

State/territorial/local/tribal government

924110

State, territorial, and local, air quality management programs that are responsible for ambient air monitoring under 40 CFR part 58 or that elect to submit an application for a reference or equivalent method determination under 40 CFR part 53 or for an approved regional method approved under 40 CFR part 58 appendix C. The proposal also may affect Tribes that conduct ambient air monitoring similar to that conducted by States and that wish EPA to use their monitoring data in the same manner as State monitoring data.

North American Industry Classification System.

This table is not intended to be exhaustive, but rather provides a guide for readers regarding entities likely to be regulated by this action. This table lists the types of entities that EPA is now aware could potentially be regulated by this action. Other types of entities not listed in the table could also be regulated. To determine whether your facility or Federal, State, local, or territorial agency is regulated by this action, you should carefully examine the requirements for reference or equivalent method determinations in 40 CFR part 53, subpart A (General

Provisions) and the applicability criteria in 40 CFR 51.1 of EPA's requirements for State implementation plans. If you have questions regarding the applicability of this action to a particular entity, consult the person listed in the preceding

FOR FURTHER INFORMATION CONTACT

section.

B. Where can I obtain a copy of this action?

In addition to being available in the docket, an electronic copy of this final action will also be available on the Worldwide Web (WWW) through the Technology Transfer Network (TTN). Following the Administrator's signature, a copy of the final amendments will be placed on the TTN's policy and guidance page for newly proposed or promulgated rules at

http://www.epa.gov/ttn/oarpg

. The TTN provides information and technology exchange in various areas of air pollution control.

C. Public Comments on Proposed Amendments

EPA received approximately 20,000 public comments on the proposed amendments to the ambient air monitoring regulations during the 90-day comment period. These comments were submitted to the rulemaking docket and also during public hearings held in Chicago, Illinois; Philadelphia, Pennsylvania; and San Francisco, California (71 FR 8228, February 16, 2006). Public comments on the proposed amendments were submitted by States, local governments, Tribes, and related associations; energy, mining, ranching, and agricultural interests and related associations; vendors, laboratories, and technical consultants; health, environmental, and public interest organizations; and private citizens. The EPA has carefully considered these comments in developing the final amendments. Summaries of these comments and EPA's detailed responses are contained in the

Response to Comments

document included in the docket.

D. Judicial Review

Under section 307(b)(1) of the Clean Air Act (CAA), judicial review of the final amendments is available only by filing a petition for review in the U.S. Court of Appeals for the District of Columbia Circuit by December 18, 2006. Under section 307(d)(7)(B) of the CAA, only an objection to the final amendments that was raised with reasonable specificity during the period for public comment can be raised during judicial review. Moreover, under section 307(b)(2) of the CAA, the requirements established by the final amendments may not be challenged separately in any civil or criminal proceedings brought by EPA to enforce these requirements.

E. Peer Review

The EPA sought expert scientific review of the proposed methods, technologies, and approach for ambient air monitoring by the Clean Air Scientific Advisory Committee (CASAC). The CASAC is a Federal advisory committee established to review scientific and technical information and make recommendations to the EPA Administrator on issues related to the air quality criteria and corresponding NAAQS. CASAC formed a National Ambient Air Monitoring Strategy (NAAMS) Subcommittee in 2003 to provide advice for a strategy for the national ambient air monitoring programs. This subcommittee, which operated over a 1-year period, and a new subcommittee on Ambient Air Monitoring and Methods (AAMM), formed in 2004, provided the input for CASAC on its consultations, advisories, and peer-reviewed recommendations to the EPA Administrator.

In July 2003, the CASAC NAAMS Subcommittee held a public meeting to review EPA's draft National Ambient Air Monitoring Strategy document (dated September 6, 2002), which contained technical information underlying planned changes to the ambient air monitoring networks. The EPA continued to consult with the CASAC AAMM Subcommittee throughout the development of the proposed amendments. Public meetings were held in July 2004, December 2004, and September 2005 to discuss the CASAC review of nearly 20 documents concerning methods and technology for measurement of particulate matter (PM); data quality objectives for PM monitoring networks and related performance-based standards for approval of equivalent continuous PM monitors; configuration of ambient air monitoring stations;

and other technical aspects of the proposed amendments. These documents, along with CASAC review comments and other information are available at:

http://www.epa.gov/ttn/amtic/casacinf.html

“Station” and “site” are used somewhat interchangeably in this notice of final rulemaking. When there is a difference (which will be apparent from context), “site” generally refers to the location of a monitor, while “station” refers to a suite of measurements at a particular site.

F. How is this document organized?

The information presented in this preamble is organized as follows:

I. General Information

A. Does this action apply to me?

B. Where can I obtain a copy of this action?

C. Public Comments on Proposed Amendments

D. Judicial Review

E. Peer Review

F. How is this document organized?

II. Authority

III. Overview

A. Summary of Concurrent Final Action on Revisions to the National Ambient Air Quality Standards for Particulate Matter

B. Summary of Changes to Ambient Air Monitoring Regulations

C. Significant Dates for States, Local Governments, Tribes, and Other Stakeholders

D. Implementation of the Revised Monitoring Requirements

E. Federal Funding for Ambient Air Monitoring

IV. Discussion of Regulatory Revisions and Major Comments on Proposed Amendments to 40 CFR Part 53

A. Overview of Part 53 Regulatory Requirements

B. Requirements for Candidate Reference Methods for PM

10−2.5

C. Requirements for Candidate Equivalent Methods PM

2.5

and PM

10−2.5

D. Other Changes

V. Discussion of Regulatory Revisions and Major Comments on Proposed Amendments to 40 CFR Part 58

A. Overview of Part 58 Regulatory Requirements

B. General Monitoring Requirements

1. Definitions and Terminology

2. Annual Monitoring Network Plan and Periodic Network Assessment

3. Operating Schedules

4. Monitoring Network Completion for PM

10−2.5

and NCore Sites

5. System Modifications

6. Annual Air Monitoring Data Certification

7. Data Submittal

8. Special Purpose Monitors

9. Special Considerations for Data Comparisons to the National Ambient Air Quality Standards

C. Appendix A—Quality Assurance Requirements for State and Local Air Monitoring Stations and Prevention of Significant Deterioration Air Monitoring

1. General Quality Assurance Requirements

2. Specific Requirements for PM

10−2.5

, PM

2.5

, PM

, and Total Suspended Particulates

3. Particulate Matter Performance Evaluation Program and National Performance Audit Programs

4. Revisions to Precision and Bias Statistics

5. Other Program Updates

D. Appendix C—Ambient Air Quality Monitoring Methodology

1. Applicability of Federal Reference Methods and Federal Equivalent Methods

2. Approved Regional Methods for PM

2.5

E. Appendix D—Network Design Criteria for Ambient Air Quality Monitoring

1. Requirements for Operation of Multipollutant NCore Stations

2. Requirements for Operation of PM

10−2.5

Stations

3. Requirements for Operation of PM

2.5

Stations

4. Requirements for Operation of PM

Stations

5. Requirements for Operation of Carbon Monoxide, Sulfur Dioxide, Nitrogen Dioxide, and Lead Monitoring Sites

6. Requirements for Operation of Ozone Stations

7. Requirements for Operation of Photochemical Assessment Monitoring Stations

F. Appendix E—Probe and Monitoring Path Siting Criteria for Ambient Air Monitoring

1. Vertical Placement of PM

10−2.5

Samplers

2. Ozone Monitor Setback Requirement from Roads

G. Sample Retention Requirements

H. Deletion of Appendices B and F

VI. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review

B. Paperwork Reduction Act

C. Regulatory Flexibility Act

D. Unfunded Mandates Reform Act

E. Executive Order 13132: Federalism

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

G. Executive Order 13045: Protection of Children From Environmental Health and Safety Risks

H. Executive Order 12898: Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations

I. Executive Order 13211: Actions That Significantly Affect Energy Supply, Distribution, or Use

J. National Technology Transfer Advancement Act

K. Congressional Review Act

II. Authority

The EPA rules for ambient air monitoring are authorized under sections 110, 301(a), and 319 of the Clean Air Act (CAA). Section 110(a)(2)(B) of the CAA requires that each State implementation plan (SIP) provide for the establishment and operation of devices, methods, systems, and procedures needed to monitor, compile, and analyze data on ambient air quality and for the reporting of air quality data to EPA. Section 103 authorizes, among others, research and investigations relating to the causes, effects, extent, prevention and control of air pollution. Section 301(a) of the CAA authorizes EPA to develop regulations needed to carry out EPA's mission and establishes rulemaking requirements. Uniform criteria to be followed when measuring air quality and provisions for daily air pollution index reporting are required by CAA section 319.

III. Overview

A. Summary of Concurrent Final Action on Revisions to the National Ambient Air Quality Standards for Particulate Matter

Elsewhere in this

Federal Register

, EPA is finalizing revisions to the National Ambient Air Quality Standards (NAAQS) for particulate matter (PM). These revisions were proposed on January 17, 2006 (71 FR 2620). For a detailed explanation of these revisions, see that preamble elsewhere in this

Federal Register

The EPA is finalizing the PM

2.5

NAAQS revisions as proposed. With regard to the primary standards for fine particles (generally referring to particles less than or equal to 2.5 micrometers (μm) in diameter, PM

2.5

), EPA is revising the level of the 24-hour PM

2.5

standard to 35 micrograms per cubic meter (μg/m

), providing increased protection against health effects associated with short-term exposure (including premature mortality and increased hospital admissions and emergency room visits). The EPA is retaining the level of the annual PM

2.5

standard at 15 μg/m

, continuing protection against health effects associated with long-term exposure (including premature mortality and development of chronic respiratory disease). The EPA is also finalizing the proposed revisions in the conditions under which spatial averaging of the annual primary PM

2.5

NAAQS is permitted, and placing these conditions in appendix N of 40 CFR part 50 rather than in appendix D of 40 CFR part 58.

With regard to secondary PM standards, EPA is revising the current 24-hour PM

2.5

secondary standard by making it identical to the revised 24-hour PM

2.5

primary standard, retaining the annual PM

2.5

and 24-hour PM

secondary standards, and revoking the annual PM

secondary standard. This suite of secondary PM standards is intended to provide protection against PM-related public welfare effects, including visibility impairment, effects on vegetation and ecosystems, and materials damage and soiling.

The EPA is finalizing the proposed Federal reference method (FRM) for PM

2.5

. This action in essence codifies certain desirable features that have already been in widespread use as elements of approved equivalent methods or national user modifications.

The EPA is not finalizing the proposed NAAQS for PM

10−2.5

, for reasons explained in the accompanying preamble to the revisions to the NAAQS. As a result, EPA is not finalizing a number of related provisions (notably those which would have prescribed which monitors could have been used for comparison with that proposed NAAQS) proposed as amendments to 40 CFR part 58. The EPA is, however, finalizing the proposed FRM for PM

10−2.5

(see appendix O to 40 CFR part 50). This FRM is based on paired filter-based samplers for PM

2.5

and PM

and it will serve as the standard of reference for measurements of PM

10−2.5

concentrations in ambient air. This should provide a basis for approving Federal Equivalent Methods (FEMs) and promote the gathering of scientific data to support future reviews of the PM NAAQS. Because it is a filter based system, this method can itself be used to provide speciated data. The reference measurement from the PM

10−2.5

FRM is also important in the development of alternative PM

10−2.5

speciation samplers such as dichotomous samplers. The EPA will be issuing guidance to ensure the use of a consistent national approach for speciated coarse particle monitors as soon as possible.

In conjunction with the above NAAQS revisions and FRM provisions, as part of this final monitoring rule, as described below EPA is finalizing certain provisions which support collection of additional high quality data on ambient concentrations of PM

10−2.5

. These data should be useful in improving the understanding of PM

10−2.5

air quality and in conducting future reviews of the PM NAAQS.

As explained in the preamble to the NAAQS revisions, EPA is revoking the annual NAAQS for particles generally less than or equal to 10 μm in diameter (PM

). However, EPA is retaining the 24-hour PM

NAAQS as a standard for short-term exposure to thoracic coarse particles, rather than revoking that standard in all but 15 areas as proposed. This change from the NAAQS revision proposal necessitates that the final monitoring rule restore certain PM

monitoring provisions that were proposed for removal.

B. Summary of Changes to Ambient Air Monitoring Regulations

This rule, in most respects, finalizes the proposals put forth in the January 17, 2006, notice of proposed rulemaking (71 FR 2710). This final rule will facilitate monitoring program changes envisioned in the draft National Ambient Air Monitoring Strategy which was fully described in the proposal. These final changes, which apply to the monitoring program for all of the criteria pollutants, will reduce the required scale of monitoring for pollutants for which most areas have reached

attainment. The changes are intended to better focus monitoring resources on current air quality challenges. The changes will also allow States and local monitoring agencies more flexibility to design their monitoring programs to reflect local conditions.

In amendments to 40 CFR part 53 (Reference and Equivalent Methods), this final rule incorporates the proposed criteria for approval of Federal equivalent methods (FEM) for PM

2.5

, with some modifications to the method testing requirements and approval criteria in response to persuasive public comments. The modifications will require a more robust set of testing conditions and closer performance matching of candidate FEMs to FRMs. The EPA is also finalizing the rule with some strengthening revisions to the proposed criteria for approved regional methods (ARMs) for PM

2.5

. The new criteria for PM

2.5

FEMs and ARMs will facilitate the commercialization and EPA approval of continuous PM

2.5

mass monitors, allowing them to be substituted for many of the currently operating filter-based FRMs, which will support additional monitoring objectives and reduce annual monitoring costs.

In other amendments to 40 CFR part 53, EPA is adopting FEM approval criteria for PM

10−2.5

, with some revisions from the proposal that will provide for approval and use of methods that can meet multiple monitoring objectives. The new FEM performance criteria for PM

10−2.5

will facilitate approval of filter-based methods for direct sampling of PM

10−2.5

concentrations that can be chemically speciated using post-sampling laboratory analysis. The FEM criteria are also expected to encourage commercialization of highly time-resolved continuous methods. The EPA is hopeful that the PM

2.5

and PM

10−2.5

FEM criteria together will result in the approval and commercialization of methods that provide equivalent measurements of PM

2.5

, PM

, and PM

10−2.5

from a single instrument.

In amendments to 40 CFR part 58 (Ambient Air Quality Surveillance), this final rule, as proposed, requires States to establish and operate a network of NCore multipollutant monitoring stations. The EPA intends the NCore network to consist of approximately 75 stations, of which the rule requires between 62 and 71 such stations. These stations must be operational by 2011. Most States, as well as the District of Columbia, Puerto Rico, and the Virgin Islands, will be required to operate a single station. California, Florida, Illinois, Michigan, New York, North Carolina, Ohio, Pennsylvania, and Texas will be required to operate two or three NCore stations. For these States, the selection between two or three stations will be part of the development and approval of the NCore monitoring plan that is due by July 1, 2009. The EPA also plans to negotiate with a number of States, local agencies, and/or Tribes to operate additional NCore stations on a voluntary basis, bringing the total number of stations to about 75. By approving some required stations to be in rural areas and by negotiating for additional voluntary sites in rural areas, EPA expects that about 55 NCore sites will be in urbanized areas and about 20 in rural areas. The rural sites are intended to be sited away from any large local emission sources, so that they represent ambient concentrations over an extensive area. The NCore stations must perform the types of pollutant measurements that were proposed, with three exceptions. PM

10−2.5

measurements may be made on a 1-in-3 day schedule rather than the proposed every day schedule, NO

measurements may be waived by the EPA Administrator based on certain criteria, and as explained later in this section, PM

10−2.5

chemical speciation will be required in addition to PM

10−2.5

mass concentration measurements.

refers to a broad class of nitrogen-containing reactive compounds in ambient air, explained in more detail in sections V.E.1 and V.E.7 of this preamble.

The EPA estimated that the proposed rule would have required States to operate about 225 PM

10−2.5

monitors based on the population and estimated PM

10−2.5

concentrations of metropolitan statistical areas (MSAs) with populations of 100,000 or more. In addition, PM

10−2.5

monitors were proposed to be required at NCore stations; some monitors likely would have satisfied both of these requirements. Because EPA is not adopting a NAAQS for PM

10−2.5

, the final monitoring rule does not include the proposed requirement for the broad network of PM

10−2.5

monitoring stations in MSAs over 100,000 population. However, the final monitoring rule does require PM

10−2.5

monitors at the required NCore multipollutant monitoring stations. The data gathered from these stations should be useful in improving understanding of PM

10−2.5

air quality and in conducting future reviews of the PM NAAQS. The EPA anticipates that due to natural variations among the cities and rural areas where the NCore stations will be sited, the NCore PM

10−2.5

monitors will represent a range of concentrations and nearby emission source types, and that many but not all will be in well populated locations.

The EPA is not adopting the proposed population-based and population density-based siting requirements for PM

10−2.5

monitors, or any part of the proposed five-part suitability test for PM

10−2.5

monitoring sites, which as proposed would have controlled whether PM

10−2.5

data from a monitoring site could be compared to the proposed PM

10−2.5

NAAQS. These proposed requirements were tied to the establishment of a PM

10−2.5

NAAQS with a qualified PM

10−2.5

indicator based on a determination of whether ambient mixes of coarse particles are or are not dominated by coarse particle emissions from enumerated types of sources. Since EPA is not adopting this part of the proposal, these issues are now moot. In the absence of a PM

10−2.5

NAAQS, our goal nevertheless will be to locate PM

10−2.5

monitors in a manner that satisfies an objective of the proposed rule, which was to focus most monitoring resources on population centers.

This final rule contains a requirement for PM

10−2.5

speciation to be conducted at NCore multipollutant monitoring stations. The EPA had proposed a requirement for PM

10−2.5

speciation in 25 areas, with the areas required to have this monitoring selected based on having a Metropolitan Statistical Area (MSA) population over 500,000 and having an estimated design value of greater than 80 percent of the proposed PM

10−2.5

NAAQS. This would have concentrated the PM

10−2.5

speciation monitoring in areas that have high populations and high exposures to PM

10−2.5

. Since EPA is requiring PM

10−2.5

monitoring at NCore primarily for scientific purposes, it is more appropriate to have monitoring in a variety of urban and rural locations to increase the diversity of areas for which chemical species data will be available to use in scientific studies. The EPA had already proposed to require chemical speciation for PM

2.5

at NCore stations. The collocation of both PM

10−2.5

and PM

2.5

speciation monitoring at NCore stations is consistent with the multipollutant objectives of the NCore network and will support further research in understanding the chemical composition and sources of PM

, PM

10−2.5

, and PM

2.5

at a variety of urban and rural locations. The EPA will work with States to ensure that PM

10−2.5

speciation monitors employ the latest in speciation technology to advance the science so that future regulation will provide more targeted protection against the effects only of those coarse particles

and related source emissions that prove to be of concern to public health.

Because the 24-hour PM

NAAQS is being retained in all parts of the country, this final rule retains the existing minimum monitoring network design requirements for PM

. These longstanding requirements are based on the population of a MSA and its historical PM

air quality. For any given combination of these two parameters, a range of required monitors is prescribed, with the required number to be determined as part of the annual monitoring plan. The EPA estimates that once States and Regional Administrators have considered how current population data and recent PM

air quality affect the required number of PM

monitors in each area, between 200 and 500 FRM/FEM monitors will be required, compared to about 1,200 in operation now. While States may of course choose to continue to operate monitors in excess of the minimum requirements, EPA notes that many PM

monitors have been recording concentrations well below the PM

NAAQS and are candidates for discontinuation at a State's initiative. States may choose to retain PM

monitors that are recording concentrations below the PM

NAAQS level to support monitoring objectives other than attainment/nonattainment determinations, such as baseline monitoring for prevention of significant deterioration permitting or public information.

This final rule changes the requirements for the minimum number of monitors for PM

2.5

and ozone (O

) monitoring networks. In response to comments, the final requirements require more O

and PM

2.5

monitoring in more polluted areas and more monitors in CSAs than was proposed. While this final rule requires fewer monitors than are now operating for O

and PM

2.5

, as did the pre-existing monitoring rule, EPA does not intend to encourage net reductions in the number of O

and PM

2.5

monitoring sites in the U.S. as a whole. The surplus in the existing networks relative to minimum requirements gives States more flexibility to choose where to apply monitoring resources for O

and PM

2.5

. For PM

2.5

, this final rule requires that sampling be conducted on a daily basis for monitors that have recently been recording the highest concentrations in their area and have been recording concentrations very near the 24-hour NAAQS, to avoid a bias in attainment/nonattainment designations that can occur with less frequent sampling. Pursuant to this provision, EPA estimates that about 50 sites now sampling less frequently will be required to change to daily sampling.

As proposed, minimum monitoring requirements for carbon monoxide (CO), sulfur dioxide (SO

), and nitrogen dioxide (NO

) are eliminated in this final rule. Minimum requirements for lead (Pb) monitoring stations and Photochemical Assessment Monitoring Stations (PAMS) are reduced to those that were proposed. For all five criteria pollutants, however, existing monitoring sites (except those already designated as special purpose monitors) cannot be discontinued without EPA Administrator (for PAMS or NCore stations) or Regional Administrator (for all other types of monitoring) approval. Regional Administrator approval is also required for discontinuation of O

, PM

2.5

, and PM

sites even if they are in excess of minimum network design requirements. While the rule requires EPA approval, such approvals should be facilitated where appropriate by rule provisions which clearly establish certain criteria under which discontinuation will be approved. These criteria are the same as those proposed with four minor changes explained in detail in section V.B.5, System Modifications. These criteria are not exclusive, and monitors not meeting any of the listed criteria may still be approved for discontinuation on a case-by-case basis if discontinuation does not compromise data collection needed for implementation of a NAAQS. Specific monitoring for these pollutants may currently be required in individual SIPs; this monitoring rule does not affect any SIP requirements for such specific monitoring.

Appendix A to this final rule includes most of the proposed revisions to the quality system for ambient air monitoring. In particular, the proposed requirement for States to ensure a program of adequate and independent audits of their monitoring stations is included in this final rule. One way, but not the only way, a State can satisfy this requirement is to agree that EPA will conduct these audits using funds that otherwise would have been awarded to the State as part of its annual air quality management grant. A small number of changes to the proposed quality system requirements reflect public comments on details of the proposed revisions. Also, because the objective of PM

10−2.5

monitoring is to better understand PM

10−2.5

air quality and to support health effects studies, rather than to provide data for use in nonattainment designations, and because there consequently will be a much smaller network of required PM

10−2.5

monitors than proposed, the quality system for PM

10−2.5

in this final rule differs from the proposed system in that it aims to quantify data quality at the national level of aggregation rather than at the level of individual monitoring organizations as had been proposed. Another change from the proposal is that a provision has been added allowing the EPA Regional Administrator to waive the usual quality system requirements for special purpose monitors when those requirements are logistically infeasible due to unusual site conditions and are not essential to the monitoring objectives.

The EPA is finalizing the proposed provisions regarding when data from special purpose monitors (SPMs) can be compared to a NAAQS, with minor clarifications. In summary, the final rule provides that if an ozone or PM

2.5

SPM operates for only two years or less, EPA will not use data from that monitor to make attainment/nonattainment determinations. This limitation is inherent in the form of these NAAQS, which require three years of data for a determination to be made. For the other NAAQS pollutants, as a policy matter, EPA will not use only two years of data from a SPM to voluntarily redesignate an area to nonattainment. This limitation is possible because as established in Section 107(d)(1) of the Act, the only time EPA is obligated to redesignate areas as attainment or nonattainment is after it promulgates or revises a NAAQS. Under an existing standard, voluntary redesignations are at the Administrator's discretion: EPA has no legal obligation to redesignate an area even if a monitor should register a violation of that standard (see CAA Section 107(d)(3)). In particular, in the case of PM

10,

EPA stated in section VII.B of the preamble to the NAAQS rule (printed in today's

Federal Register

) that because EPA is retaining the current 24-hour PM

standards, new nonattainment designations for PM

will not be required under the provisions of the Clean Air Act. The same is true for CO, NO

, SO

, and Pb. However, all valid data from a SPM will be considered in determining if a previously designated nonattainment area has subsequently attained the NAAQS. See also section V.B.8 below.

This final rule advances, to May 1, the date each year by which monitoring organizations must certify that their submitted data is accurate to the best of their knowledge. However, this requirement will take effect one year later than proposed, in 2010 for data collected in 2009.

This final rule retains the current requirement for an annual monitoring plan and finalizes most of the new

substantive and procedural requirements that were proposed for these plans. One change is that some required new elements proposed for the annual plan have instead been shifted to the 5-year network assessment, to reduce the annual plan preparation burden and to allow these elements to be prepared more carefully. The first 5-year network assessment has been postponed by one year, to July 1, 2010.

The proposed requirements regarding probe heights for PM

10−2.5

monitors, increased O

monitor distance from roadways (for newly established O

stations), data elements to be reported, and PM filter retention are included in this final rule.

This final rule also removes and reserves the pre-existing appendix B, Quality Assurance Requirements for Prevention of Significant Deterioration (PSD) Air Monitoring, and appendix F, Annual SLAMS Air Quality Information, of 40 CFR part 58 because they are no longer needed.

C. Significant Dates for States, Local Governments, Tribes, and Other Stakeholders

Only State governments, and those local governments that have been assigned responsibility for ambient air monitoring by their States, are subject to the mandatory requirements of 40 CFR part 58.

The following summary of applicable requirements is presented in chronological order, as an aid for States in planning their activities to comply with the rule. States are required to comply with pre-existing requirements in 40 CFR part 58, until the compliance date for each new requirement is reached.

Throughout this preamble, “States” is meant to also refer to local governments that have been assigned responsibility for ambient air monitoring within their respective jurisdiction by their States. This preamble also uses “monitoring organization” to refer to States, local agencies, and/or Tribes conducting monitoring under or guided by the provisions of 40 CFR part 58. This final rule applies the same requirements to the District of Columbia, Puerto Rico, and the Virgin Islands as apply to the 50 States. Other U.S. territories are not subject to this final rule.

The following provisions in 40 CFR part 53 and part 58 are effective on December 18, 2006:

• The criteria and process for EPA Administrator approval of FRMs, FEMs, and ARMs or where applicable Regional Administrator approval of ARMs. Manufacturers of continuous PM

2.5

and PM

10−2.5

instruments may apply for designation of their instruments as FRMs or FEMs starting today. The EPA is eager to receive such applications as soon as manufacturers can collect and analyze the necessary supporting data. State, local, and Tribal monitoring agencies may seek approval of their PM

2.5

continuous monitor as ARMs beginning today, either independently or in cooperation with instrument manufactures.

• The revised quality system requirements, except that full quality assurance practices, if not waived, are not required until January 1, 2009 for SPMs which use FRM, FEM, or ARM monitors.

• The new minimum requirements (or absence of minimum requirements) for the number of monitors for specific NAAQS pollutants and for PAMS stations, if the new minimum allows a State to discontinue a previously required monitor. See below for the compliance date of the new minimum requirements in situations in which the final requirement is greater than the currently operating network.

• The criteria for EPA Regional Administrator approval for removal of monitors that are in excess of minimum required, if a State seeks such removal.

• The criteria for use of data from SPMs in determinations of attainment/nonattainment.

• The elimination of the requirement for reporting of certain PM

2.5

monitor operating parameters.

• The revised requirement for separation between roadways and O

monitors, for new O

monitors whose placement has not already been approved as of December 18, 2006.

• The new specification for probe heights for PM

10−2.5

monitors.

The new requirement to archive all PM

10c

and PM

10−2.5

filters for 1 year begins with filters collected on or after January 1, 2007. However, EPA expects few if any monitoring agencies to be operating PM

10c

or PM

10−2.5

filters this early, so most will be affected later.

As explained in section IV.B of this preamble, the term “PM

10c

” refers to a PM

Federal reference method (FRM) that is designated as a PM

10c

FRM under the final NAAQS rule appearing elsewhere in today's

Federal Register.

In essence, it would be a PM

2.5

FRM with the inertial fractionator used to separate out particles larger than 2.5 microns removed so that all PM

is collected. Unlike other PM

instruments, a PM

10c

instrument must control flow to a specified flow rate of 16.67 liters/minute at local conditions of temperature and pressure. A PM

10−2.5

FRM consists of a PM

2.5

FRM and a PM

10c

FRM of the same model. See also 71 FR 2720.

The requirement to submit mass data on blank PM

2.5

filters begins on January 1, 2007.

The required date to begin daily PM

2.5

sampling at certain PM

2.5

monitoring sites is January 1, 2007. The EPA believes this will affect about 50 PM

2.5

monitoring sites. The EPA will notify the affected States directly.

This final rule does not change the schedule for reporting ambient air quality data to the Administrator, via the Air Quality System (AQS). However the rule now explicitly requires that associated quality assurance data be submitted along with ambient concentration data. The first submission affected will be the one due on June 30, 2007 for data collected in January through March of 2007.

As presently is the case, States must submit an annual network plan by July 1 of each year. The next plan is due July 1, 2007.

States whose PM

2.5

, PM

, or O

networks do not meet the revised requirements of this final rule regarding the number of monitors in a given MSA or CSA are required to submit a plan for adding the necessary additional monitors by July 1, 2007 and to begin operating the new monitors by January 1, 2008. The EPA believes that this will only affect O

and PM

2.5

monitoring in fewer than ten locations each. The EPA will notify these States directly.

A plan for the implementation of the required NCore multipollutant monitoring stations, including site selection, is due by July 1, 2009. States must implement the required NCore multipollutant stations by January 1, 2011, including PM

10−2.5

monitoring.

States will be required to submit earlier certification letters regarding the completeness and accuracy of the ambient concentration and quality assurance data they have submitted to the Air Quality System (AQS) operated by EPA, starting May 1, 2010 for data collected during 2009. Until then, States are required to submit these letters by July 1 of each year.

Network assessments are required from States every 5 years starting July 1, 2010.

Under the Tribal Authority Rule (TAR) (40 CFR part 49), which implements section 301(d) of the CAA, Tribes may elect to be treated in the same manner as a State in implementing sections of the CAA. However, EPA determined in the TAR that it was inappropriate to treat Tribes in a manner similar to a State with regard to specific plan submittal and implementation deadlines for NAAQS-related requirements, including, but not limited to, such deadlines in CAA sections 110(a)(1), 172(a)(2), 182, 187, and 191. See 40 CFR 49.4(a). For example, an Indian Tribe may choose, but is not required, to submit implementation plans for NAAQS-related requirements, nor is any Tribe required to monitor ambient air. If a Tribe elects to do an implementation plan, the plan can contain program elements to address specific air quality problems in a partial program. The EPA

will work with the Tribe to develop an appropriate schedule for making any appropriate monitoring system changes which meet the needs of each Tribe.

Indian Tribes have the same rights and responsibilities as States under the CAA to implement elements of air quality programs as they deem necessary. Tribes can choose to engage in ambient air monitoring activities. In many cases, Indian Tribes will be required by EPA regions to institute quality assurance programs that comply with 40 CFR part 58 appendix A, utilize FRM, FEM, or ARM monitors when comparing their data to the NAAQS, and to insure that the data collected is representative of their respective airsheds. For FRM, FEM, or ARM monitors used for NAAQS attainment or nonattainment determinations, quality assurance requirements of 40 CFR part 58 must be followed and would be viewed by EPA as an indivisible element of a regulatory air quality monitoring program.

D. Implementation of the Revised Monitoring Requirements

After promulgation, EPA will assist States in implementing the amended requirements using several mechanisms. The EPA will work with each State to develop approvable monitoring plans for its new NCore multipollutant monitoring stations, including PM

10−2.5

monitoring. For example, EPA will negotiate the selection of required new monitoring sites (or new capabilities at existing sites) and their schedules for start up as well as plans to discontinue sites that are no longer needed. The EPA will negotiate with each State its annual grant for air quality management activities, including ambient monitoring work. Once States have established a new monitoring infrastructure to meet the new requirements, EPA will review State monitoring activities, submitted data, and plans for further changes on an annual basis.

The EPA's support for and participation in enhancing the national ambient air monitoring system to serve current and future air quality management and research needs will extend beyond ensuring that States meet the minimum requirements of this final monitoring rule. The EPA will work with each State or local air monitoring agency to determine what affordable monitoring activities above minimum requirements would best meet the diverse needs of the individual air quality management program as well as the needs of other data users. The EPA may also work with the States, and possibly with some Tribes, to establish and operate PM

10−2.5

speciation sites in addition to those required by this final rule. The EPA also plans to work with the States, and possibly with some Tribes, to establish and operate sites that will measure only PM

10−2.5

concentrations in rural and less urbanized locations, in addition to the PM

10−2.5

monitors required at NCore sites.

An important element of implementing the new requirements will be EPA's role in encouraging the development and application of FEMs, and the development of a sampler or samplers that can provide a direct measurement of PM

10−2.5

for collection of filters used in chemical speciation and for continuous methods that measure both PM

2.5

and PM

10−2.5.

The EPA has determined that continuous monitoring of PM

2.5

has many advantages over the filter-based FRM. This final rule makes it more practical for manufacturers and users of continuous PM

2.5

instruments to obtain designation for them as FEMs or ARMs. To ensure objectivity and a sound scientific basis for decisions, EPA's Office of Research and Development will review applications for FEM and ARM designations based on the criteria in this final rule and will recommend approval or disapproval to the Administrator. For agencies seeking use of an ARM already approved in another monitoring network, the applicable Regional Office will conduct a review, most often as part of the EPA approval of an annual monitoring plan, based on the criteria in this final monitoring rule.

The EPA will also provide technical guidance documents and training opportunities for State, local, and Tribal monitoring staff to help them select, operate, and use the data from new types of monitoring equipment. The EPA has already distributed a technical assistance document on the precursor gas monitors

that will be part of the NCore multipollutant sites and EPA has conducted multiple training workshops on these monitors. Additional guidance will be developed and provided on some other types of monitors with which many State monitoring staff are currently unfamiliar, and on network design, site selection, discontinuation of sites, quality assurance, network assessment, and other topics. While Tribes are not subject to the monitoring requirements of this final rule, these technical resources will also be available to them directly from EPA and via grantees, such as the Institute for Tribal Environmental Professionals and the Tribal Air Monitoring Support Center.

Technical Assistance Document (TAD) for Precursor Gas Measurments in the NCore Multipollutant Monitoring Network. Version 4. U.S. Environmental Protection Agency. EPA-454/R-05-003. September 2005. Available at:

http://www.epa.gov/ttn/amtic/pretecdoc.html.

The EPA will also continue to support the National Park Service's operation of the IMPROVE monitoring network, which provides important data for implementing both regional haze and PM

2.5

attainment programs.

The number of sites in the IMPROVE program may vary, depending on EPA's enacted budget and the data needs of the regional haze and PM

2.5

attainment programs.

Additional information on EPA/National Park Service IMPROVE (Interagency Monitoring of Protected Visual Environments) Visibility Program is available at:

http://www.epa.gov/ttn/amtic/visdata.html.

The EPA will also continue to operate the Clean Air Status and Trends Network (CASTNET), which monitors for O

, PM, and chemical components of PM in rural areas across the nation.

EPA is in the process of revising CASTNET to upgrade its monitoring capabilities to allow it to provide even more useful data to multiple data users. The EPA expects that about 20 CASTNET sites will have new capabilities similar to some of the capabilities required at NCore multipollutant sites.

Additional information on CASTNET is available at:

http://www.epa.gov/castnet/.

This final rule includes a requirement that States must ensure a program of adequate and independent audits of their monitoring stations. One way, but not the only way, a State can satisfy this requirement is to agree that EPA will conduct these audits using funds that otherwise would have been awarded to the State as part of its annual air quality management grant. In anticipation of the possible inclusion of this requirement in this final rule, EPA has been working with monitoring organizations to determine which of these organizations prefer this approach. The EPA expects that, for 2007, nearly all monitoring organizations will request that EPA conduct these audits. For those that chose another acceptable approach, EPA will conduct limited cross-checks of equipment, calibration standards, auditor preparation, and audit procedures to ensure that their audit programs are adequate.

The EPA recognizes that characterizing and managing some air quality problems requires ambient concentration and deposition data that cannot be provided by the types of monitoring required by the monitoring activities addressed in today's final rule. These problems include near-roadway exposures to emissions from motor

vehicles and mercury deposition. The EPA is actively researching these issues and developing concepts for monitoring programs to address them, but these issues are outside the scope of this final rule.

The Air Quality System (AQS) is the data system EPA uses to receive ambient air monitoring data from State, local, Tribal, and other types of monitoring organizations and to make those data available to all interested users. AQS is based on a particular data structure and uses particular data input formats including data elements and defined values for categorical data. The existing AQS data structure and input formats are for the most part consistent with a number of changes made in this final rule to pre-existing terminology and requirements, but some changes will be needed in AQS to re-establish full consistency with requirements in the monitoring rule. The changes to AQS will likely, in turn, require some modifications to data preparation tools and practices at monitoring agencies. The EPA will prepare and implement a plan for making these changes, and will advise AQS users of the ramifications while doing so. Generally, the compliance deadlines in the rule are such that monitoring agencies are not required to immediately comply with any changes in rule provisions that would affect data transfer formats and procedures. Monitoring agencies, for the present, should continue to follow pre-existing AQS formats and procedures until notified.

E. Federal Funding for Ambient Air Monitoring

EPA has historically funded part of the cost to State, local, and Tribal governments of installation and operation of monitors to meet Federal monitoring requirements. Sections 105 and 103 of the CAA allow EPA to provide grant funding for programs for preventing and controlling air pollution and for some research and development efforts respectively. Eligible entities must apply for section 103 grants. Eligible entities must provide nonfederal matching funds for section 105 grants. The EPA's enacted budget specifies overall how much State and Tribal Air Grant (STAG) funding is available for these grants.

In recent years, EPA has received special authority through appropriations acts to use section 103 grant funding for establishing and operating PM

2.5

-related monitoring stations. Funding for other types of monitoring has been included in the grants awarded under section 105. Grants to Tribes for air quality management work, including ambient monitoring, have been awarded under section 103 with the overall amount for these funds established by the enacted budget.

During the public comment period for this rulemaking EPA received a large number of comments addressing funding issues. Most of these comments expressed opposition to the Administration's proposed EPA budget for fiscal year 2007, which included a proposal to provide PM

2.5

monitoring support through section 105 grant funding, as is done for all other criteria pollutants. (As of today, the Congress has not enacted a 2007 budget for EPA.) Commenters stated that if funding for monitoring were reduced as proposed, State and local agencies would have less flexibility than desired in designing and operating their monitoring programs, and that the proposed requirements for new PM

10−2.5

and NCore networks and for adequate and independent audits of monitoring stations would be burdensome. Some commenters requested that the proposed new requirements not be included in this final rule for this reason.

The EPA understands these concerns. However, the CAA requirements from which this final rule derives

are not contingent on EPA providing funding to States to assist in meeting those requirements. Accordingly, the comments regarding funding are not directly relevant to the content of this final rule. Nevertheless, EPA recognizes that resources always have been and will remain a practical consideration for establishing and operating monitoring programs. The EPA will continue to work with States in this regard, in particular as EPA determines how to allocate enacted funding among States and among types of monitoring so as to achieve the best possible environmental outcomes. Several provisions of this final rule reduce minimum requirements, which will provide flexibility for States to reduce some of their pre-existing costs.

Section 103(c)(2) of the Clean Air Act [42 U.S.C.A. 7403(c)] provides that the Administrator shall conduct a program for sampling air pollution that includes the establishment of a national network to monitor air quality and to ensure the comparability of air quality data collected in different states. Section 110(a)(2)(B) [42 U.S.C.A 7410(a)] provides that each State implementation plan shall provide for establishment and operation of appropriate devices, methods, systems, and procedures necessary to monitor, compile, and analyze data on ambient air quality and upon request make such data available to the Administrator. Section 182(c)(1) [42 U.S.C.A. 7511a(c)(1)] states that the Administrator will promulgate rules for enhanced monitoring for ozone, oxides of nitrogen, and volatile organic compounds in serious ozone areas.

Several commenters stated that EPA should not use STAG funds for the improvement or operation of Federal monitoring networks such as CASTNET. The EPA does not intend to use STAG funds from fiscal year 2007 or beyond in this way.

IV. Discussion of Regulatory Revisions and Major Comments on Proposed Amendments to 40 CFR Part 53

A. Overview of Part 53 Regulatory Requirements

Various appendices to 40 CFR part 50 define certain ambient air monitoring methods as Federal reference methods which may be used to determine attainment of the National Ambient Air Quality Standards (NAAQS), and which form the benchmark for determining equivalency of other methods which may also be used to determine attainment. Under 40 CFR part 53, EPA designates specific commercial instruments or other versions of methods as Federal reference methods (FRMs). To be so designated, a particular FRM must be shown, according to the procedures and requirements of part 53, to meet all specifications of both the applicable appendix of part 50 as well as applicable specifications and requirements of part 53.

To foster development of improved alternative air monitoring methods, EPA also designates—as Federal equivalent methods (FEMs)—alternative methods that are shown to have measurement performance comparable to the corresponding FRM. Part 53 contains explicit performance tests, performance standards, and other requirements for designation of both FRMs and FEMs for each of the criteria pollutants. In addition, States' air surveillance monitoring networks are required, under 40 CFR part 58, appendix C, to use only EPA-designated FRMs, FEMs, or ARMs at SLAMS sites. A list of all methods that EPA has designated as either FRMs or FEMs for all criteria pollutants is available at

http://www.epa.gov/ttn/amtic/criteria.html.

Elsewhere in today's

Federal Register

, EPA is promulgating a new Federal reference method for measurement of mass concentrations of thoracic coarse particles (PM

10−2.5

) in the atmosphere, to be codified as appendix O to 40 CFR part 50. Although, as explained earlier, EPA is not at this time adopting any NAAQS for PM

10−2.5

, EPA believes an FRM for PM

10−2.5

is still highly desirable to aid in a variety of needed

research studies.

This new FRM is defined as the standard of reference for measurement of PM

10−2.5

concentrations in ambient air. It will be an acceptable and readily available PM

10−2.5

measurement method for new NCore multipollutant monitoring sites to be located at approximately 75 urban and rural locations. Availability of an approved FRM for PM

10−2.5

will also help provide consistency among PM

10−2.5

measurements used in future health studies of the adverse health effects associated with exposure to thoracic coarse particles. Lastly, the PM

10−2.5

reference method will provide the basis for development of speciation samplers capable of providing an improved understanding of the compositions of different ambient mixes of thoracic coarse particles, so that this composition can be related to both health effects and to particle sources. Associated with this new reference method, EPA is also establishing related amendments to 40 CFR part 53 to extend the designation provisions of FRMs and FEMs to methods for PM

10−2.5

. These amendments set forth explicit tests, performance standards, and other requirements for designation of specific commercial samplers, sampler configurations, or analyzers as either FRMs or FEMs for PM

10−2.5

, as appropriate.

Henderson, R. Clean Air Scientific Advisory Committee (CASAC) Review of the EPA Staff Recommendations Concerning a Potential Thoracic Coarse PM Standard in the

Review of the National Ambient Air Quality Standards for Particulate Matter: Policy Assessment of Scientific and Technical Information

(Final PM OAQPS Staff Paper, EPA-452/R-05-005). September 15, 2005.

http://www.epa.gov/sab/panels/casacpmpanel.html.

Henderson, R. Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory Committee to the Honorable Stephen L. Johnson, Administrator, U.S. EPA. Clean Air Scientific Advisory Committee Recommendations Concerning the Proposed National Ambient Air Quality Standards for Particulate Matter. March 21, 2006.

http://www.epa.gov/sab/pdf/casac-ltr-06-002.pdf.

As noted in section VI.A of the preamble to the NAAQS revisions published elsewhere in this

Federal Register

, EPA recognizes that the FRM, while providing a good standard of performance for comparison to other methods, is not itself optimal for routine use in PM

10−2.5

monitoring networks. Alternative methods are needed that provide a more direct measurement of ambient PM

10−2.5

concentrations. Methods are also needed that collect samples of PM

10−2.5

that are more physically separated for analysis of chemical species. Also, automated, continuous-type methods provide many operational advantages to ease monitoring burdens, reduce on-site service requirements, and eliminate off-site sample filter support services, as well as to provide measurement resolution of 1 hour or less and near real-time reporting of monitoring data. Therefore, EPA is interested in encouraging the development of alternative monitoring methods for PM

10−2.5

by focusing on the explicit test and qualification requirements necessary for designation of such types of methods as FEMs for PM

10−2.5

. In fact, EPA anticipates that alternative FEMs will eventually provide most of the PM

10−2.5

monitoring data obtained in the States' monitoring networks.

Further, EPA recognizes that the potential benefits of automated/continuous monitoring methods apply as well to FEMs for PM

2.5

. Accordingly, as proposed, EPA is also establishing new requirements in part 53 for designation of continuous FEMs for PM

2.5

. See 71 FR 2721. The PM

2.5

and PM

10−2.5

FEM provisions parallel each other in many respects so inclusion now is both appropriate and conforming.

The new requirements for approval of automated/continuous FEMs can accommodate a wide range of potential PM

10−2.5

or PM

2.5

continuous measurement technologies. Ambient air testing of a candidate technology at diverse monitoring sites is required in order to demonstrate that the level of comparability to collocated Federal reference method measurements is adequate to meet established data quality objectives (DQOs).

This final rule also modifies somewhat certain existing requirements for designation of alternative, non-continuous methods for PM

2.5

. As explained in section IV.B of this preamble, the modified requirements will be fully consistent with the more advanced new requirements for both continuous and non-continuous FEMs for PM

10−2.5

B. Requirements for Candidate Reference Methods for PM

10−2.5

No comments were received related specifically to the PM

10−2.5

FRM designation requirements. These provisions are adopted as proposed. Because of the nearly complete similarity between the specifications for the new PM

10−2.5

reference method and for the existing PM

2.5

reference method, the designation requirements for PM

10−2.5

reference methods are essentially the same as those for PM

2.5

reference methods. As set forth in the new appendix O to 40 CFR part 50, the PM

10−2.5

reference method specifies a pair of samplers consisting of a conventional PM

2.5

sampler and a special PM

sampler. The PM

2.5

sampler must meet all requirements for a PM

2.5

reference method in 40 CFR part 50, appendix L, as well as additional requirements in part 53. However, the PM

sampler required by the method is not a conventional PM

sampler as described in 40 CFR part 50, appendix J; rather, it is a sampler specified to be identical to the PM

2.5

sampler of the pair, except that the PM

2.5

particle size separator is removed. This special PM

sampler is identified as a “PM

10c

” sampler to differentiate it from conventional PM

samplers that meet the less exacting requirements of 40 CFR part 50, appendix J. In view of the similarity of the PM

10−2.5

FRM requirements to those of the PM

2.5

FRM, the new requirements will allow a PM

10−2.5

sampler pair consisting of samplers that have already been shown to meet the PM

2.5

FRM requirements (except for the PM

2.5

particle size separator in the case of the PM

10c

sampler) to be designated as a PM

10−2.5

reference method without further testing.

C. Requirements for Candidate Equivalent Methods for PM

2.5

and PM

10−2.5

As pointed out in the preamble to the proposed rule (71 FR 2721), EPA believes very strongly that provisions to allow designation of Federal equivalent methods provide an important incentive to encourage the commercial development of innovative new and advantageous alternative methods for monitoring air pollutants. However, it is also important to show conclusively that any new candidate method will produce measurements comparable to those of the FRM and will have performance characteristics that are adequate to meet DQOs. At the same time, the testing that is necessary to show comparable and adequate performance must not be so burdensome that it undermines incentives for new method development.

Because of the complex nature of particulate matter, it is also complex to test the performance of PM monitoring methods. For methods for PM

2.5

, EPA defined three classes of candidate FEMs (Classes I, II, and III) based on the extent to which the method differs from the FRM, so that the nature and extent of the performance and comparability testing necessary can be more closely matched to the nature of the candidate method. See 40 CFR 53.3(a)(2)−(4). In this final rule, as proposed, EPA is extending these same class definitions and tiered testing requirements to apply to PM

10−2.5

candidate FEMs as well.

Class I methods are limited to minor deviations from the FRM; Class II covers

integrated-sample, filter-based, gravimetric methods deviating more significantly from the FRM; and Class III methods (originally) included all other methods not categorized as Class I or II. The three classes are described in more detail in the proposal preamble (71 FR 2721). As proposed, the definition of Class III FEMs is narrowed to include only continuous or semi-continuous analyzer methods having 1-hour or less measurement resolution, which are the Class III methods that by far hold the most potential for monitoring applications and FEM designation. The EPA has thus avoided the restrictions and complexity that would be necessary to accommodate the wide variety of other types of non-Class I or II methods that are unlikely to be economically and commercially practical. Also, the continuous operational nature of such Class III methods gives rise to a statistical advantage that allows more tolerant limits of adequate comparability, relative to a method that is not operated continuously, to achieve a similar limit of uncertainty in the monitoring data.

Class III continuous methods appear to offer many potential benefits for use in routine field monitoring networks. These automated analyzers eliminate most, if not all, of the pre- and post-weighing of sample filters, require less frequent on-site service, may be less costly to operate, and offer near real-time, electronic reporting of hourly (or less) mass concentration measurements (similar to data reporting that is common for gaseous pollutant monitors). The EPA is accordingly adopting the proposed Class III FEM provisions for PM

10−2.5

and PM

2.5

in today's rule, with some changes in response to comments.

Continuous methods, by nature, tend to have somewhat different performance characteristics from those of the corresponding filter-based FRMs, so the comparability and performance testing requirements must be adequately comprehensive and discriminating without being excessively burdensome. The Class III FEM requirements being promulgated today are based predominantly on demonstrating an adequate degree of comparability between candidate method measurements and concurrent, collocated Federal reference method measurements under a representative variety of site conditions. Many issues and much technical input were carefully considered during the development of the requirements, including peer review by the Ambient Air Monitoring and Methods Subcommittee of the Clean Air Scientific Advisory Committee. The salient Class III FEM requirements were summarized in the proposal preamble (71 FR 2722-2724). Not unexpectedly, a considerable number of comments were received in connection with the specifics of the proposed Class II and Class III requirements. The more significant of these comments are addressed below, after a summary of the proposal regarding requirements for Class II and Class III methods. Remaining comments are addressed in the

Response to Comments

document.

Class II candidate FEMs, although not offering the operational advantages of continuous Class III methods, are nevertheless important as well. Class II methods encompass the dichotomous and virtual impactor types of methods that can provide a more direct, gravimetric, filter-based measurement of PM

10−2.5

than available with the FRM. These methods are also most likely to fulfill the substantial need for collecting PM

10−2.5

samples that are physically separated from other particle sizes, or nearly so, for chemical species analysis. New requirements for Class II FEMs for PM

10−2.5

are being established in this final rule, and some of the previously established requirements for Class II FEMs for PM

2.5

are being changed somewhat to make them more consistent with the corresponding new requirements for PM

10−2.5

Class II FEMs and to incorporate some minor technical improvements.

The proposed Class II FEM requirements, as outlined in the proposal preamble (71 FR 2721-2725), were based on daily sampling; therefore, Class II equivalent methods used for determining compliance with the PM

2.5

NAAQS would generally have been restricted to daily sampling. However, in response to concerns about method performance in relatively clean areas, EPA has strengthened the additive bias (intercept) requirement. With this tighter performance criteria and considering that Class II methods are filter-based samplers, a minimum of a one-in-three day sample frequency will be appropriate to meet the network data quality objectives. Class II methods are also expected to be used for collecting samples used in chemical species analysis, which would not require daily operation. The character of the test sites specified for Classes II and III tests for both PM

2.5

and PM

10−2.5

are similar, so concurrent testing for PM

2.5

and PM

10−2.5

methods of both classes can be carried out, substantially reducing the testing burden for candidate FEMs that measure both PM

2.5

and PM

10−2.5

or for testing multiple candidate methods simultaneously.

Of particular note to instrument manufacturers, this final rule allows applications for Class II candidate FEMs for both PM

10−2.5

and PM

2.5

to optionally substitute the more extensive Class III comparability field tests in subpart C for some or all of the rather extensive and arduous laboratory wind tunnel tests, loading test, and volatility test of subpart F to which a Class II candidate FEM sampler may otherwise be subject. Such a substitution of test results may be particularly important when the special facilities necessary for the wind tunnel tests or other tests are not available. Concurrent testing of multiple methods under the Class III requirements may also help to reduce overall testing costs.

In regard to the proposed testing requirements for Class III (continuous) FEMs for PM

2.5

and PM

10−2.5

, EPA specifically solicited comments related to the adequacy of the number and location of the test sites required for the field tests to determine comparability of a candidate method to the respective FRM. See 71 FR 2722. By definition, a designated FEM is generally qualified for use at any monitoring site in the U.S. (with the possible exception of some areas with extreme conditions), so the test requirements for comparability need to represent a wide variety of possible site conditions. The EPA proposed that candidate methods be tested within three general geographical areas: (1) The Los Angeles area in winter and summer seasons, (2) eastern U.S. in winter and summer, and (3) western U.S. in winter only (for a total of five 30-day test campaigns). Each proposed test site area was selected for representing particular and diverse typical site conditions.

In response to several comments addressing this issue, a fourth test site—in the U.S. Midwest, with tests required in the winter season only—has been added to the requirements to further increase the geographical diversity. However, the requirement for a winter test campaign in the eastern U.S. has been withdrawn while the requirement for a summer test campaign in the eastern U.S. has been retained, so the total number of required test campaigns (five) is unchanged. Comparability testing of a candidate method is costly, rendering it impractical to test a candidate method under all possible combinations of site and seasonal conditions that might be encountered in national PM monitoring networks. The EPA considers the specified complement of five test campaigns in the four specified geographical areas and two seasons to be reasonable to conduct and adequately representative of the diversity of site and seasonal PM

monitoring conditions across the U.S. As noted above, the two test site areas specified for testing candidate Class II FEMs are compatible with the test sites for candidate Class III methods, which will significantly reduce testing costs by allowing Class II and III candidate methods to be tested simultaneously at the same test site. Also, the test sites have been relabeled for ease of referencing east and west sites.

Some commenters expressed concern that the Class III comparability test standards might be inadequate because a candidate method that had an unacceptable seasonal bias (such as has been noted for some continuous methods) could be found acceptable, because in pooling test data from summer and winter seasons the biases would compensate. The EPA finds that the associated minimum correlation requirement of the regression test should adequately avoid that situation. Further, in the revised test requirements, summer and winter tests at the same site, where the data are pooled, are required at only one of the four required tests sites.

Another issue concerning the proposed testing requirements for Class III (continuous), as well as Class II candidate equivalent methods for PM

2.5

and PM

10−2.5

, was the specific acceptance criteria for the regression analysis statistics—particularly the additive bias (intercept) parameter—of the comparison between collocated measurements obtained with the candidate and FRM methods. As proposed, the upper and lower limits for the regression intercept were specified as functions of the corresponding slope, with the acceptable combinations of slope and intercept represented by the area inside a trapezoid or a hexagon shape plotted on a slope-intercept coordinate system (Figures C-2 and C-3 in proposed revised subpart C of part 53 at 71 FR 2768-2769). These acceptance limits were based on statistical considerations related to the uncertainty allowable in making correct NAAQS attainment decisions for PM

2.5

(or similar comparisons of PM

10−2.5

concentrations to non-regulatory benchmarks). Several commenters were concerned that the range of acceptable intercepts proposed for Class II and III FEMs, although appropriate for DQOs related to attainment (or similar) decisions, may allow excessive measurement bias for FEMs used for other PM monitoring applications—especially those applications that require measurements of concentrations well below the level of the NAAQS.

In response to these comments and in deference to potential use of FEMs for a variety of applications, EPA has somewhat strengthened the range of allowable intercepts for those candidate FEMs. For Class III FEMs, new fixed limits of ±2.0 μg/m

for PM

2.5

methods and ±7.0 μg/m

for PM

10−2.5

methods have been added. For Class II FEMs for PM

10−2.5

, the fixed intercept limit has been reduced from ±7.0 to ±3.5 μg/m

. (The intercept requirements proposed for candidate Class II PM

2.5

methods were re-examined and found to be appropriate as proposed.) The more restrictive intercept limits will reduce the maximum allowable measurement bias and are represented by smaller hexagonal acceptance areas, as specified in 40 CFR part 53, subpart C revised Table C-4 and as illustrated in revised Figures C-2 and C-3 of this final rule.

Nevertheless, EPA wishes to point out that, because of the design of the equivalent method comparability tests (which require no low-level test concentrations) and the nature of the regression analysis, a seemingly high positive or negative intercept resulting from the regression analysis of the test data is not necessarily indicative or likely to be characteristic of the actual measurement errors or bias of the candidate method relative to the FRM at low or very low concentrations. This situation may be particularly true when the concentration coefficient of variation (CCV) for the FEM test data (see 40 CFR 53.35(h)) is relatively low, resulting in greater uncertainty in the predicted additive bias (and in the multiplicative bias (slope) as well).

Class III FEMs will generally provide 1-hour concentration measurements (in addition to the required 24-hour measurements), and EPA asked for comments on whether the FEM provisions should include any specific requirements for 1-hour precision, and if so, whether a specific standard of performance should be specified and how it should affect FEM designation. See 71 FR 2723. Of the few comments received on this issue, most agreed with EPA that 1-hour precision is an important descriptor associated with a Class III candidate method and that 1-hour FEM test data should be submitted in a Class III FEM application so that the short-term precision can be determined, but no specific standard should be set for the precision parameter in connection with the FEM designation qualifications. A few commenters suggested that a precision performance parameter based on a running average of a few (e.g., 3 to 5) hours should be established and regulated, however, to preserve flexibility, EPA believes that precision estimates are better included in method-specific quality assurance guidance (to be used by instrument operators as they believe appropriate) rather than as a formal part of the FEM provisions. Therefore, no changes were made to the proposed requirement that FEM applicants submit the 1-hour FEM test data, and there is no designation requirement based on 1-hour precision or any other particular 1-hour based performance statistic.

The EPA also asked for comments on the adequacy and appropriateness of the proposed test requirements for Class II FEMs. See 71 FR 2724. Some commenters suggested that the proposed Class II tests were inadequate because there was more variation in the PM at different sites than could be represented in the tests—particularly in regard to chemical compositions—and suggested that continued FEM designation should be conditioned on a mandatory periodic reassessment of local-agency comparisons to FRM measurements. The EPA recognizes that data produced by all FEMs operated in monitoring networks under 40 CFR part 58 should meet the data quality objectives (DQOs) of 40 CFR part 58, appendix A, section 2.3.1 on a continuing basis. The operational requirements of appendix A will help ensure this. Moreover, EPA can invoke designation cancellation procedures for the method designation under 40 CFR 53.11 (Cancellation of reference or equivalent method designation) if EPA observes that DQOs are not being maintained for a particular designated Class II equivalent method (or for any FEM or FRM). However, EPA believes that designation cancellation should be initiated by EPA when necessary, rather than have designations conditioned on specific periodic reassessments as commenters suggested. Other commenters suggested that the test sites be approved by both EPA and the STAPPA/ALAPCO Monitoring Committee, but EPA believes that would be cumbersome and unnecessary.

D. Other Changes

EPA proposed several other relatively minor changes to various provisions of subparts A, C, E, and F of part 53. See 71 FR 2724-2725. Organizational changes in subpart C consolidate the provisions for various types of methods, making them easier to understand. Other changes clarify or simplify some existing provisions for PM

and PM

2.5

Class I and II FEM testing and implement minor technical improvements to test protocols, with little, if any, impact on the nature or efficacy of the tests. Minor changes are made to subparts A, E, and F to incorporate the new PM

10−2.5

provisions and some new definitions,

make a few administrative adjustments, and incorporate a few minor technical changes. These changes are described more completely in the proposal preamble (71 FR 2724), and they are being adopted as proposed, as no comments were received pertinent to these minor changes.

After considering all comments carefully, EPA determined that no further changes should be made to the proposed new or revised FRM and FEM requirements. The EPA is thus adopting the proposed new or revised requirements and provisions for Federal reference and Federal equivalent methods for PM

2.5

and PM

10−2.5

, modified to incorporate the changes described above.

V. Discussion of Regulatory Revisions and Major Comments on Proposed Amendments to 40 CFR Part 58

A. Overview of Part 58 Regulatory Requirements

Part 58 of 40 CFR, Ambient Air Quality Surveillance, contains requirements for ambient air monitoring programs operated by States (or designated local agencies). As proposed, the structure of part 58 remains much the same as the 1997 version. Proposed subparts A through G, containing 40 CFR 50.1 through 50.61, provide definitions of terms; require the operation of certain numbers and types of monitors by certain dates; require the use of certain monitoring methods, quality system practices, and sampling schedules and frequencies; require annual plans describing a State's monitoring network and planned changes to it; provide criteria for EPA approval of planned changes; require data submission and certification that submitted data is accurate to the best of the knowledge of responsible State official; address special rules regarding special purpose monitors; provide rules for comparing monitoring data to applicable National Ambient Air Quality Standards (NAAQS); require reporting of the Air Quality Index (AQI) to the public in some areas; and provide for monitoring directly by EPA if a State fails to operate required monitors. As proposed, part 58 also includes appendices A, C, D, E, and G which were referenced by various numbered sections in subparts A through G. These appendices contain many detailed requirements, as well as considerable explanatory or background material and non-binding advice. Appendix A addresses quality system requirements, appendix C addresses monitoring methods and equipment, appendix D mostly addresses the number of required monitors and their placement within a metropolitan or other area, appendix E addresses the details of monitoring station layout, and appendix G addresses AQI reporting. (Subpart B of the 1997 version was proposed to be removed. Subpart F was already reserved in the 1997 version. No amendments were proposed to the part 58 requirements for reporting of the AQI and the associated appendix G.)

To aid in understanding the provisions of the final part 58 and their relationship to the 1997 and proposed provisions, the following discussion for the most part follows the order of the final part 58, addressing each affected numbered section and then the appendices.

B. General Monitoring Requirements

1. Definitions and Terminology

The EPA proposed to discontinue the use of the term “National air monitoring stations (NAMS)”. See 71 FR 2720. Previously, this term was used to designate Federal reference method (FRM) and Federal equivalent method (FEM) monitors which were operated to meet set requirements for the number (and, for some pollutants the type of location) of monitors and which required EPA Administrator approval for changes, as distinguished from “State and local air monitoring stations (SLAMS)” which referred to additional FRM and FEM monitors for which generally there was no minimum number, for which siting was more at the State's discretion, and for which changes were approved by the Regional Administrator.

The EPA proposed a new definition for “National Core (NCore)” stations.

The definition of “State or local air monitoring stations (SLAMS)” was proposed to be modified to include NCore, Photochemical Air Monitoring Systems (PAMS), and all other State or locally operated stations (such as PM

2.5

speciation stations) that have not been designated as a special purpose monitor or monitoring station (SPM). This change was proposed for convenience in referencing these types of monitors together because some provisions in the rule apply to all of them but not to SPMs. See 71 FR 2720. Previously, “SLAMS” referred only to FRM and FEM monitors.

The term, “Approved regional methods” (ARMs), proposed at 71 FR 2720, is added to refer to alternative PM

2.5

methods that have been approved by EPA for use specifically within a State, local, or Tribal air monitoring network for purposes of comparison to the NAAQS and to meet other monitoring objectives, but which may not have been approved as FEM for nationwide use.

The EPA proposed to adopt a new term, “Primary quality assurance organization” to clarify the working definition of the term “Reporting organization” currently utilized in section 3.0.3. of 40 CFR part 58, appendix A, Quality Assurance Requirements, and to avoid confusion with the different way “reporting organization” has come to be used in a related but distinct context (final uploading of data to the Air Quality System). See 71 FR 2778.

The EPA also proposed additional definitions to be consistent with terminology used in 40 CFR part 50, appendix O, the FRM for PM

10−2.5

. See 71 FR 2777. Modifications to the definitions of key geographical terms were proposed, as needed, to reflect changes in U.S. Census Bureau usage since the last revision to monitoring regulations.

The EPA received some questions seeking clarification of the new term “Primary quality assurance organization,” which are addressed in the

Response to Comments

document available in the docket. No other adverse comments were received on these proposed definitions, and this final rule includes all of them.

2. Annual Monitoring Network Plan and Periodic Network Assessment

The EPA proposed to consolidate current requirements for the SLAMS air quality surveillance plan and NAMS network description into elements of the annual monitoring network plan described in 40 CFR 58.10 of the proposed rule. See 71 FR 2725. The annual monitoring network plan would provide a statement of purpose for each monitor in a monitoring agency network and provide evidence that siting and operation of each monitor meet the requirements of appendices A, C, D, and E of part 58, as applicable. The EPA also proposed the addition of some required elements to the annual monitoring network plan and proposed to add a new requirement for a periodic network assessment.

The EPA received comments on a number of specific elements within the annual monitoring network plan and with regard to the network assessment requirement. The comments that were the basis for modifications to the proposed rule are discussed briefly here. Detailed responses to all comments are provided in the

Response to Comments

document available in the docket.

Comments were received on the proposed requirement for a 30-day

public inspection period before State submittal of a draft annual monitoring network plan to the Regional Administrator as well as on the proposed requirement for Regional Administrator approval of annual monitoring network plans seeking SLAMS network modifications including new monitoring sites. Some commenters requested clarification regarding what methods would be considered acceptable for making documents available for public inspection. Commenters also expressed concern that the 120 days proposed for Regional Administrator review and approval/disapproval would result in unnecessary delays.

The EPA notes the general support in the comments for the public inspection requirement. Commenters also supported the flexibility in the proposed rule which would allow monitoring agencies to design and implement appropriate ways of allowing this inspection. The EPA supports use of monitoring agency Web sites for such postings, along with other means of providing public notice including hard-copy posting in libraries and public offices. Although the public inspection requirement does not specifically require States to obtain and respond to received comments, such a process is encouraged with the subsequent transmission of comments to the appropriate EPA Regional Office for review. Therefore, EPA has modified this final rule from the proposal to specify that where the State has provided for a public comment process and provided any comments received to EPA, and the posted plan has not been substantially altered as a result of the public comments, the requirement for the Regional Administrator to obtain public comment by a separate process can be waived. The 120 days allowed for Regional Administrator review of an annual plan is a feature of the current monitoring rule, and has been kept in this final rule.

The EPA received many comments on the proposed requirement for the annual monitoring network plan to contain cost information. See 71 FR 2780. Commenters were concerned that no details were provided regarding what information would be required and how the information would be used. The accounting difficulty in calculating such cost information was also noted along with concerns regarding the administrative burden of preparing and documenting the cost estimates.

The EPA has considered the proposed requirement for cost information in the annual monitoring network plan and agrees that considerable effort would be needed to develop guidance to standardize the development of financial information and for States to collect and summarize the information for submittal. Without such standardization, cost information would be difficult to interpret. In view of these comments, EPA has deleted this element from the list of required information to be contained in the annual monitoring network plan.

The EPA proposed a new requirement that the annual monitoring network plan consider the ability of existing and proposed sites to support air quality characterization for areas with relatively high populations of susceptible individuals (e.g., children with asthma), and, for any sites that are being proposed for discontinuance, the effect on data users other than the agency itself, such as nearby States and Tribes or health effects studies. See 71 FR 2780. Several commenters noted that this requirement would be challenging to implement and involves knowledge of public health that may not be readily available to monitoring organizations. In addition, it was noted that, absent the availability of a centralized information clearinghouse, it would be difficult for States to be aware of all possible users of data for health studies or other types of research.

This new element of the annual monitoring network plan highlights the importance that EPA places on the consideration of sensitive populations when evaluating the relative value and representativeness of monitoring sites, particularly for areas where one or more NAAQS may be approached or exceeded.

The EPA acknowledges the potential challenge in obtaining information about the distribution of susceptible individuals in specific geographic areas around existing and proposed sites, and has purposely defined the requirement as a “consideration” to provide significant latitude for monitoring organizations to determine the complexity and depth of their response. In recognition of the potential complexity of preparing assessments of susceptible populations on a sub-county sized spatial scale as represented by typical monitoring sites, in this final rule EPA has moved this requirement to become a required element of the 5-year network assessment rather than the annual monitoring network plan.

See S. Rep. No. 91-1196. 91st Cong. 2d Sess. 10 (1970) (NAAQS is to be set to protect sensitive, at-risk population groups).

With regard to the proposed provision requiring States to consider the effect on data users of proposed actions to discontinue sites, EPA notes that States are already required to make their annual network monitoring plans available for public inspection and that process provides the basic framework for disseminating information about anticipated site discontinuations. The EPA recognizes that there are many potential users of air quality information and that States cannot be aware of all such users. However, to the extent that information about site shutdowns can be disseminated more widely, there are benefits to be gained by protecting key monitors that (for example) support ongoing health studies or that are the basis for long-term trend analyses, or otherwise provide information that is used by stakeholders other than the operating agency. As such, EPA has retained this provision in this final rule. The EPA will work with States and health organizations to explore options for tracking the status of key air quality sites.

The EPA received many comments in response to the proposed requirement for a network assessment to be completed every 5 years and to be submitted with the required annual network monitoring plan. Commenters acknowledged the overall value of a more complete evaluation of monitoring programs but expressed concern about the resource burden in meeting the requirement.

Network assessments are a key tool to help ensure that the right parameters are being measured in the right locations, and that monitoring resources are used in the most effective and efficient manner to meet the needs of multiple stakeholders. Network assessments can help identify new data needs and associated technologies, find opportunities for consolidation of individual sites into multi-pollutant sites, and identify geographic areas where network coverage should be increased or decreased based on changes in population and/or emissions. The EPA has already issued draft guidance to describe the possible techniques that States can use in developing their assessments, and has purposely limited the required elements to provide flexibility in the amount of resources that would be required. After consideration of the comments, EPA has retained the network assessment requirement in this final rule. In light of the concerns raised about the resource requirements needed to complete network assessments, the deadline for the first required assessment under this final rule has been delayed an additional year to July 1, 2010.

The EPA is not adopting the proposed requirement for a separate plan

establishing a network of PM

10−2.5

stations as an addendum to the annual monitoring network plan (see 71 FR 2740, 2779) since the only required PM

10−2.5

monitoring will take place as part of the NCore multi-pollutant stations, already covered by the proposed plan due July 1, 2009. The EPA has added clarifying language to this final rule requiring Administrator approval for the NCore plan due July 1, 2009 and subsequent annual monitoring network plan elements proposing modifications, consistent with the requirement for Administrator approval of NCore stations in section 3(a) of appendix D.

The proposed plan element supporting PM

10−2.5

suitability tests for NAAQS comparisons likewise is not being adopted since EPA is not finalizing the proposed PM

10−2.5

NAAQS.

The proposed prescriptive wording with reference to public hearings in the context of reviews of changes to violating PM

2.5

monitors and/or community monitoring zones (71 FR 2780) has been modified to specify that draft plans containing such proposed changes to PM

2.5

networks must be made available for public inspection and comment by States prior to submission to the EPA Regional Administrator but that States can design the process for achieving such goals.

3. Operating Schedules

The EPA proposed that manual PM

2.5

monitors at SLAMS be required to operate on a 1-in-3 day sampling frequency, except under certain conditions and when approved by the Regional Administrator. See 71 FR 2780. As discussed in section II.E.1 of the preamble to the final revisions to the PM NAAQS, published elsewhere in this

Federal Register

, commenters pointed out a potential bias in the method used to calculate the 98th percentile form of the 24-hour PM

2.5

NAAQS. As explained there, to avoid this potential bias, EPA is requiring daily sampling at design value sites that are within 5 percent of the 24-hour NAAQS for PM

2.5.

The EPA proposed that manual PM

10−2.5

samplers at SLAMS stations must operate on a daily schedule, without a requirement for any collocated continuously operated FEM PM

10−2.5

samplers. See 71 FR 2780. Numerous commenters noted that a 1-in-3 day sampling frequency was acceptable for PM

2.5

sites and said that the same sampling frequency for PM

10−2.5

would produce sufficient data for comparison to the proposed 24-hour PM

10−2.5

NAAQS averaged over 3 years. Commenters also noted the lack of currently available continuous FEM PM

10−2.5

instruments and the burdensome resource requirements associated with daily sampling requirements using the proposed filter-based FRM.

The proposed requirement for daily PM

10−2.5

sampling was based on a data quality objective system analysis that identified such a frequency as being a key factor in reducing statistical uncertainty at concentrations near the level of the proposed 24-hour PM

10−2.5

NAAQS. Since EPA is not finalizing a PM

10−2.5

NAAQS but instead is requiring a more limited set of PM

10−2.5

monitors at NCore sites to support objectives other than and (obviously) not including NAAQS compliance, additional flexibility in sampling frequency requirements is appropriate. Although daily sampling of PM

10−2.5

at NCore sites remains a desirable outcome, and will become a more practical goal with the advent of continuous FEM monitors in several years, EPA has reduced the PM

10−2.5

sampling frequency requirement in this final rule to 1-in-3 days.

The EPA proposed reducing the sample frequency requirement for PM

manual methods. Reducing the sample frequency for PM

was possible since EPA had proposed to have daily sampling of PM

10−2.5

to support protection from thoracic coarse particles. As published elsewhere in today's

Federal Register

, EPA is retaining the 24-hour PM

standard and not finalizing a PM

10−2.5

standard. The EPA is also only finalizing a limited network of PM

10−2.5

monitors at multi-pollutant NCore stations for scientific purposes. Therefore, since the existing requirement for PM

sample frequency is for daily sampling for the site with the expected maximum concentration in each area, and previous assessments of the 24-hour standard demonstrates that maximizing sample frequency will minimize decision errors, EPA is retaining the existing daily sample frequency requirement for the site with expected maximum concentration in each area. This existing requirement also allows for other sites in the same area to operate on a 1-in-6 day sample frequency. Sample frequency relief is possible for expected maximum concentration sites that are significantly away from the 24-hour PM

NAAQS and in seasons exempted by the Regional Administrator.

4. Monitoring Network Completion for PM

10−2.5

and NCore Sites

The proposed requirement for specified numbers of PM

10−2.5

sites to be physically established no later than January 1, 2009 is not included in this final rule. However, by January 1, 2011, States must implement the less extensive monitoring for PM

10−2.5,

including speciation sampling, as part of the generally-applicable requirement to operate NCore multipollutant monitoring stations by that date. A plan for the implementation of the required NCore multipollutant monitoring stations, including site selection, is due July 1, 2009.

Little comment was received on the requirement for the NCore multipollutant sites to be physically established no later than January 1, 2011, and that requirement remains unchanged in this final rule as EPA continues to believe that this is practical and desirable.

5. System Modifications

In part, EPA started this rulemaking based on the recognition by EPA and leaders of State and local monitoring agencies that State/local monitoring networks should be modified to reduce some types of monitoring activity in some areas and to begin new types of monitoring. The EPA proposed rule changes to revise the minimum required number of monitors for ozone (O

), PM

2.5

, lead (Pb), and PAMS pollutants and to eliminate altogether the minimum number of required monitors for carbon monoxide (CO), sulfur dioxide (SO

) and nitrogen dioxide (NO

) in order to utilize scarce resources more productively by allowing for reductions in the number of monitoring sites where appropriate. See 71 FR 2729.

The EPA stated in the proposal that the remaining requirements for the minimum number of monitors for Pb, PM

2.5,

and O

were intended to be necessary but not always sufficient to meet the requirements in section 110(a)(2)(B) of the Clean Air Act (CAA) that State implementation plans (SIPs) provide for operation of appropriate systems to monitor, compile, and analyze data on ambient air quality. Similarly, although EPA believes that one-size-fits-all rules for the number of CO, SO

, and NO

monitors are no longer appropriate in light of the rarity of NAAQS violations for those pollutants, EPA believes that some monitoring should be continued in many areas for these pollutants. Accordingly, EPA proposed to continue to require States to propose changes in their monitoring networks, including discontinuation of monitors, and obtain EPA approval before making changes, even when the remaining minimum requirements, if any, for number of monitors would still be met after the

changes. The EPA approval would be given by the Regional Administrator, usually through approval of the annual monitoring network plan, except for changes involving NCore sites, PAMS sites, and PM

2.5

speciation trends sites which would require Administrator approval.

While local situations need to be considered individually, EPA proposed six criteria for approval of requests to discontinue monitors. See 71 FR 2749. To summarize, the six criteria addressed: (1) Any monitor which could be shown to have a low probability of future violations; (2) a CO, PM

, SO

, or NO

monitor that has been reading consistently lower than another monitor in the same area; (3) any highest reading monitor that has not indicated any NAAQS violation in the previous 5 years and for which the approved SIP provides for an alternative to continued monitoring; (4) any monitor which cannot be compared to a NAAQS because of siting considerations; (5) any monitor designed only to measure transport from upwind areas if another transport monitor were replacing it; and (6) any monitor for which logistical problems make continued operation at the current site impossible. Situations not addressed by these criteria would be considered on a case-by-case basis.

The EPA received a number of comments on the proposed removal of the minimum monitoring requirements for some of the criteria pollutants, on the revision of the minimum numbers of monitors for other criteria pollutants, on the six proposed criteria for discontinuing monitors, and on the issue of discontinuing monitors more generally, mostly from State and local monitoring agency officials. This final rule provisions on minimum numbers of monitors for O

, PM

2.5

, PM

, and Pb are discussed in section V.E of this preamble. Comments on the other parts of the proposal are addressed here. A few commenters specifically endorsed all or part of these proposals, or at least the intention to facilitate reductions in unnecessary or duplicative monitoring activities. Most commenters expressed concern over the proposals.

A number of commenters appear to have interpreted the proposals as indicators of network reductions EPA intended to require monitoring agencies to make, and expressed opposition to such reductions. The EPA clarifies here that EPA believes that proposals for network modifications should generally be initiated by the monitoring agency; EPA does not intend to compel any agency to remove any monitor. The proposals related to network modifications, and the provisions in this final rule, govern only EPA's consideration of changes which monitoring agencies seek to adopt. The EPA recognizes that funding constraints may require agencies to discontinue monitors that they otherwise would operate, but this reinforces the need for EPA review and the usefulness of having criteria for discontinuance to govern that review.

A few commenters suggested that EPA include in the rule or provide via guidance specific formulas or calculation procedures regarding the estimation of the probability of a future NAAQS exceedance, which is the basis of the first of the six proposed adjudicative criteria. The EPA intends to provide guidance on this matter in the future, but we believe that binding formulas or procedures in rule form would preclude development of better general procedures and the sort of case-specific analysis of unique factors that is likely to be appropriate in some situations.

A number of commenters stated that the six proposed criteria were overly focused on whether a monitor is providing data for use in making comparisons to the NAAQS for purposes of attainment/nonattainment findings, and that decisions to remove or retain a monitor should also recognize the utility of the monitor in satisfying other required monitoring objectives. Section 1 of the proposed appendix D of 40 CFR part 58 stated that air monitoring networks must be designed to meet three monitoring objectives: (1) Providing air pollution data to the public; (2) supporting compliance with ambient air quality standards and emission strategy development; and (3) supporting air pollution research studies. Some commenters pointed out that EPA has articulated in the draft National Ambient Air Monitoring Strategy

seven objectives for the NCore multipollutant monitoring stations (overlapping in part with the three objectives in section 1 of appendix D) and stated that single-pollutant stations should be considered to be part of an overall network to meet these objectives. The EPA agrees that these two sets of overlapping objectives are important and that monitors should not be discontinued without regard to whether these objectives will continue to be met, but EPA believes the proposed criteria, along with other provisions regarding approval of annual monitoring network plans and periodic network assessments, protect the required monitoring objectives. The paragraphs below address two objectives that were most often mentioned by commenters.

“Draft National Ambient Air Monitoring Strategy,” December 2005.

Several commenters stated that ambient monitoring can serve as a continuing check on the compliance of a specific source, or sources in the aggregate, with applicable emissions limits. The EPA believes that given that factors such as wind direction, dispersion conditions, and atmospheric reactivity conditions can greatly influence the relationship between emissions and ambient concentrations, situations are infrequent in which ambient monitoring is a critical, or the most important, element of source compliance monitoring. Other EPA rules address requirements for direct emissions and compliance monitoring for many types of sources. Ambient monitoring agencies will have the option of continuing to operate ambient monitors they feel are useful for this objective.

Some commenters stated that the ability to track trends in air quality and assess whether those trends are consistent with trends expected from the emission control program in general or from specific control measures (i.e., accountability) could be impaired if too many existing monitors are removed. The EPA believes that tracking trends is most important for O

, PM

2.5

, and PM

because these are the NAAQS with more than a few remaining nonattainment areas. For these pollutants the revised requirements in this final rule for minimum number of monitors, the new requirement for NCore multipollutant monitoring stations, and the interest of monitoring agencies in continuing these types of monitoring as indicated by the comments themselves will, in EPA's opinion, result in networks that are appropriately robust for tracking trends and assessing causal factors. The EPA believes that the availability of multiple collocated and time resolved measurements at NCore sites will be a major advantage in this work.

The

Response to Comments

document available in the docket explains in more detail how the other objectives mentioned by commenters are consistent with the six proposed criteria.

Accordingly, this final rule mirrors the proposals, with the following four exceptions:

(1) In the first criterion, which as proposed would have allowed the removal of a monitor for any criteria pollutant if it has shown attainment over the last five years and has less than a 10 percent probability of exceeding 80 percent of the NAAQS over the next three years and if it is not specifically

required by the attainment plan or maintenance plan, this final rule also conditions the removal of the last remaining SLAMS monitor in a nonattainment or maintenance area on the attainment plan or maintenance plan not having any contingency measure triggered by air quality concentrations. If a plan does have such a trigger, a plan revision to remove that trigger would have to be adopted by the State and approved by EPA. The EPA will address the requirements for such a revision at a future date.

(2) While the preamble described a sixth criterion for approval of State proposals to discontinue a monitor, having to do with logistical problems at a current site, the proposed rule text inadvertently omitted this criterion. This final rule includes it.

(3) The second and third criteria have been slightly revised to make them applicable also to the lower reading monitor of a pair that are in the same attainment area and county, and not just to the lowest reading monitor of a pair that are in the same nonattainment area or maintenance area. A commenter pointed out the need for this revision to achieve the obvious intention of the proposal.

(4) The third proposed criterion, worded to apply only to “the highest reading monitor * * * in a county,” required that a described monitor could be removed only if the approved SIP provided for a specific, reproducible approach to representing the air quality of the affected county in the absence of actual monitoring data. While EPA intended the highest reading monitor to be addressed in this third criterion, EPA did not intend to preclude the possibility that a lower reading monitor ineligible for removal under the first two criteria could be addressed also. This final rule revises the criterion to encompass any monitor not eligible for removal under the first two criteria where applicable.

6. Annual Air Monitoring Data Certification

The EPA proposed a shorter timeframe for States to submit the annual letter certifying ambient concentration and quality assurance data to the Administrator. See 71 FR 2749. Under current requirements, States have until July 1 to certify data from January 1 to December 31 of the previous year. For data collected in 2006, for example, the annual certification letter is due no later than July 1, 2007. Under the proposed requirement, the schedule for certification would be moved up 60 days, with the data certification letter required under the accelerated deadline to be due by May 1, 2009, for data collected in 2008. The EPA proposed this change to provide opportunity for an earlier start and completion for nationwide designation actions, to provide States and the public with earlier design values in time for most ozone seasons, and to support other data uses that could benefit from earlier data certification.

In response, some commenters expressed reservations about the accelerated schedule as it applies to all submitted data, while others supported the proposal for continuous instruments that collect and report hourly data but not for data requiring lab analysis for samples collected in the field. These commenters were concerned about the feasibility and cost of meeting an accelerated schedule. The EPA notes that some States have recently provided certifications for filter-based data ahead not only of the July 1 deadline, but also of the proposed May 1 deadline, when such certifications were deemed advantageous by the States for data uses such as PM

2.5

nonattainment designations. This suggests that all States could be capable of certifying data by the proposed May 1 deadline, if not earlier, if they invest in needed improvements in information technology or efficiencies in administrative procedures. Therefore, this final rule includes the proposed May 1 deadline. In recognition of the time necessary for States to adjust to the accelerated certification requirement, the implementation date has been delayed 1 year, until May 1, 2010, for data collected in 2009.

One commenter questioned the types of annual summary reports that would required to be submitted with the data certification letter, finding the proposed requirements of 40 CFR 58.15(b) unclear. The EPA notes that different reports were mentioned in the proposal to clarify the difference between SLAMS and SPM monitors (only FRM, FEM, and ARM SPM monitors are required to be certified) and to ensure that annual summary reports are provided for both types of monitors. Providing one annual summary report for certification of both SLAMS and SPM data is appropriate. An additional report providing a summary of precision and accuracy data is necessary to demonstrate that applicable monitors meet appendix A criteria.

7. Data Submittal

The EPA proposed to reduce the data reporting requirements associated with PM

2.5

FRMs to ease the data management burden for monitoring agencies. See 71 FR 2748. The following Air Quality System (AQS) reporting requirements were proposed for elimination: Maximum and minimum ambient temperature, maximum and minimum ambient pressure, flow rate coefficient of variation, total sample volume, and elapsed sample time. AQS reporting requirements were retained for average ambient temperature and average ambient pressure, and any applicable sampler flags.

The EPA also proposed a requirement for the submission of data on PM

2.5

field blank mass in addition to PM

2.5

filter-based measurements. See 71 FR 2749. Field blanks are filters which are handled in the field as much as possible like actual filters except that ambient air is not pumped through them, to help quantify contamination and sampling artifacts. This requirement only applies to field blanks which States are already taking into the field and weighing through their laboratory procedures.

Commenters supported the proposed changes to data submittal requirements and they are being finalized without modification. The requirement for reporting of field blank mass data begins with filters collected on or after January 1, 2007.

8. Special Purpose Monitors

The January 17, 2006 proposal included a background explanation of the historical distinctions between regular air monitors and special purpose monitors (SPMs) with respect to monitoring objectives, siting actions, quality assurance, and use of data. See 71 FR 2745. The EPA proposed a revision of the definition of SPM, to the effect that any SPM must be in excess of the required minimum number of monitors and that designation of a monitor as an SPM be made by the State. The EPA also proposed that States would continue to be able to choose to start and stop SPMs at will, without needing EPA approval and that States be required to submit all data from SPMs to the AQS operated by EPA. In addition, EPA proposed that States follow 40 CFR part 58 appendix A quality assurance requirements for any SPM that utilizes a FRM, FEM, or ARM instrument and which is sited consistently with the requirements of appendix E (which does not apply to SPMs on a mandatory basis). The existing rule provides that States follow these requirements only if the data from the SPM are intended by the State for use in attainment/nonattainment determinations.

The EPA also proposed that data from the first 2 years of operation of a SPM (even if using a FRM, FEM, or ARM

instrument and meeting appendix A and E requirements) would not be used by EPA in attainment/nonattainment findings for PM

2.5

or O

if the monitor stopped operating by the end of those 2 years. See 71 FR 2745. For CO, SO

, NO

, Pb, and the 24-hour PM

NAAQS, EPA proposed that data from the first 2 years of operation of a SPM would not be used by EPA for nonattainment redesignations but that such data would be considered when determining whether a nonattainment area had attained the NAAQS. The reasons for this distinction by pollutant had to do with differences in the form of the respective NAAQS and whether the EPA action in question is mandatory or discretionary. These reasons were explained in detail in the preamble to the proposal. Finally, EPA proposed that currently operating monitors not already designated as SPMs could not be designated as SPMs after January 1, 2007.

The EPA received many comments on these issues, mostly from State and local air monitoring officials but also from two industry groups. No commenter objected to the flexibility States have to start and stop SPMs. That flexibility is retained in this final rule.

Some commenters pointed out an ambiguity in the proposed requirement that data from SPMs be submitted to AQS. The EPA intended, but did not clearly state in the proposal, that this requirement apply only to SPMs that are FRMs, FEMs, or ARMs and that are operated consistently with the requirements of 40 CFR 58.11 (network technical requirements), 40 CFR 58.12 (operating schedule), and part 58, appendix A (quality assurance requirements). These would be the SPMs that produce data that will be of most interest to EPA and the public, because except for possible inconsistencies with the siting requirements of appendix E to part 58, these are the type of data which can be compared to the respective NAAQS. This final rule provides this clarification.

One commenter suggested that the specific reference to the AQS data system be made more general, to provide for the development and use of other suitable data submission systems in the future. This comment is relevant to all monitoring data, not just data from SPMs. This final rule retains references to AQS. If AQS is replaced or supplemented with approved alternatives in the future, terminology can be updated at that time.

One State official supported the proposal that SPMs be subject to the regular quality requirements of appendix A, if the SPM is a FRM, FEM, or ARM. All other commenters on this issue contended that States should be allowed more flexibility. Most of these commenters agreed that regular quality assurance practices were desirable generally, but stated that practical difficulties can arise at a specific SPM site, such that requiring regular quality assurance practices would effectively mean that the SPM could not be legally operated at all and the useful data it could have provided would be lost to users.

After considering these comments, EPA continues to believe that regular quality assurance practices are practical and of reasonable cost and feasibility in nearly all situations, as shown by successful adherence to these practices at thousands of regular monitoring stations. They are appropriate in most cases and should be the presumptive requirement. As proposed, this final rule provides for a transition period by delaying this requirement until January 1, 2009. However, EPA recognizes that unusual situations may exist in which exceptions should be allowed. For example, a State, perhaps with EPA encouragement, might operate an automated O

monitor year-round but have difficulty getting personnel and equipment to the site regularly in winter due to road conditions. This final rule allows the Regional Administrator to approve other appropriate quality assurance practices if the requirements of 40 CFR part 58 appendix A would be physically and/or financially impractical due to physical conditions at the monitoring site and the quality assurance practices are not essential to achieving the intended data objectives. This approval can be given separately, or as part of the approval of the annual monitoring plan. Approval of alternative quality assurance practices for all or part of the year does not qualify the affected data from an affected SPM for comparison to the relevant NAAQS.

Most of the comments received on the SPM proposals addressed the application of SPM data to attainment/nonattainment findings and designations. One citizen supported the proposal. About 20 commenters argued for a general, indefinitely long prohibition on the use of data from SPMs for nonattainment findings and designations, for States to have a way of blocking EPA from using particular SPM data indefinitely, or for States to be able to negotiate in advance with EPA for particular SPM data to not be used. Those commenters who explained their position generally stated that the risk of a nonattainment finding would discourage voluntary special purpose monitoring that could benefit air quality management.

In the proposal preamble (71 FR 2745, January 17, 2006), EPA stated that it understood and to some degree sympathized with the thrust of very similar input EPA had received during the development of the proposed rule, but that EPA believed that under the CAA EPA may not legally ignore technically valid data from FRM and FEM (and by implication and logical extension ARM) monitors when making attainment or nonattainment determinations. The comments have not provided EPA with any reason to change this view of our legal obligation. There are only two situations where EPA would not have to consider such data. One situation is when the data would be insufficient for making a finding because it is of insufficient duration given the averaging period or form of the relevant NAAQS. This was the basis for the proposal concerning PM

2.5

and O

for which the form of the NAAQS requires 3 years of data.

The other situation is when EPA has the discretion to simply not make a finding or to take an action, for example by taking no action to redesignate an area to nonattainment even though a SPM indicates a new violation of a NAAQS subsequent to the area's initial designation as attainment. This was the basis for the proposal concerning the CO, SO

, NO

, Pb, and PM

NAAQS. Unlike the PM

2.5

and O

NAAQS, the NAAQS for these pollutants have forms that allow a nonattainment finding based on only 1 or 2 years of data, either because the NAAQS is explicitly based on only one year of data or because a single year of data may include so many exceedances that it is certain that the average number of expected exceedances over three years will be greater than one. However, for these other NAAQS, EPA does not have a mandatory duty to make nonattainment redesignations until such time as the NAAQS are revised. In the absence of either a NAAQS revision or a State request for redesignation, the Administrator has discretion in determining whether to redesignate an area based on data from a SPM which has operated for two years or less. The EPA does regard air quality violations seriously, and does expect States to take actions to reduce air quality to healthy levels in any areas that are experiencing violations. However, EPA recognizes that there are other ways to address such violations besides redesignating an area as nonattainment. For example, EPA can work directly with a State and nearby industries to take appropriate actions to reduce emissions that are

contributing to the violation. The EPA has worked in this way with States in the past. In the case of PM

, EPA stated in section VII.B of the preamble to the NAAQS rule (printed in today's

Federal Register

) that because EPA is retaining the current 24-hour PM

standards, new nonattainment designations for PM

will not be required under the provisions of the Clean Air Act.

With respect to the second situation, applicable to the CO, SO

, NO

, Pb, and 24-hour PM

NAAQS, EPA believes it could have extended the proposed 2-year exclusion from use of SPM data in making nonattainment findings to a longer period. However, such a provision could exclude more data than appropriate and could prevent consideration of violations in making nonattainment decisions even when a SPM monitor has shown violations over 3 or more years. The EPA believes that in some and perhaps many situations like this, it would be good policy to avoid a nonattainment designation and to find other less prescriptive approaches to reducing risk to public health. EPA also believes, however, that it could be appropriate to base a nonattainment designation on such data in some other cases, where a nonattainment designation is the appropriate way to deal with a long-term nonattainment problem. Since under the final rule EPA still has the discretion not to make nonattainment redesignations based on three more years of data if EPA so chooses, EPA concludes the appropriate approach is not to universally extend the exclusion and rather rely on the Administrator's discretion to redesignate areas only in appropriate cases.

This final rule follows the proposed approach for use of data from SPMs. The EPA would like to emphasize, however, that States and other parties will have practical ways of obtaining useful information using SPMs without risk of a nonattainment redesignation. In many situations, the potential problem to be investigated, or the place under investigation, is such that a FRM, FEM, or ARM instrument meeting the siting requirements of 40 CFR part 58, appendix E is not the only suitable measurement system, and may not even be a preferred way to measure. For example, there are many commercially available PM

2.5

monitors that lack FRM, FEM, or ARM status that nevertheless would be suitable for an initial study of PM

2.5

concentrations in an unmonitored area of interest. In some other cases, 2 years may be sufficient to achieve the study objectives. Finally, under the 1997 rule (see statement at 71 FR 2719 and section 2.8.1.2.3 of appendix D to part 58 of the 1997 rule),

a SPM that is not population-oriented may not be used in comparisons to the PM

2.5

NAAQS; this may be the situation in some studies focusing on near-source impacts as well as in some studies of transport of air pollution from rural upwind areas. If the Regional Administrator has approved alternative quality assurance practices in place of the requirements of appendix A, the data from the affected SPM are not eligible for comparison to the relevant NAAQS.

EPA is recodifying this provision in section 58.30 of the final monitoring rule, but is not reconsidering or otherwise reevaluating it.

In reviewing comments about SPMs, EPA noticed that the proposed rule text for 40 CFR 58.11(d) implied that all SPMs using FRM, FEM, or ARM methods must meet appendix E siting requirements. This was not our intention, as the study objective for a SPM may require it to be located inconsistently with appendix E requirements. The implied restriction in 40 CFR 58.11(d) as proposed conflicted with an explicit statement to the contrary in 40 CFR 58.20(b) as proposed. Removing this implication is certainly in keeping with the sense of most SPM-related comments, which supported flexibility for States to operate SPMs as they choose. The promulgated version of 40 CFR 58.11(d) is drafted so as to remove this implied restriction. Data from a SPM not sited consistently with appendix E are not eligible for comparison to the respective NAAQS, unless the State has requested and EPA has approved a waiver of these criteria.

In the course of considering all the public comments on SPMs, EPA realized that the proposed restriction on designating pre-existing SLAMS monitors as SPMs after January 1, 2007 would have the effect of preventing a State from switching a monitor to SPM status even if EPA had approved the outright removal of that monitor under other provisions. This could be counter-productive. This final rule provides that if EPA has approved the discontinuation of a SLAMS monitor, the State may choose to retain the monitor and redesignate it to be a SPM. Such a monitor could be removed later without further EPA approval.

9. Special Considerations for Data Comparisons to the National Ambient Air Quality Standards

By way of background, the preamble to the proposed monitoring rule provided an explanation of when and how monitoring data are considered comparable to the respective NAAQS under existing rules and EPA policies. See 71 FR 2719-20. The EPA also proposed to relocate one of the provisions mentioned in the discussion, proposing to move pre-existing PM

2.5

rule language currently found in section 2.8.1.2.3 of appendix D to 40 CFR 58.30 of subpart D without substantive change. This relocation would provide a more prominent rule location for monitoring requirements detailing the comparability of ambient data to the PM

2.5

NAAQS. See 71 FR 2782. One commenter objected, not to this proposed rearrangement of rule language, but rather to the underlying existing (1997) requirement that PM

2.5

sites must be population-oriented to be comparable to the PM

2.5

NAAQS. This commenter stated that EPA had failed to justify any benchmark for defining an area as population-oriented. Another commenter challenged whether EPA had provided an adequate public health basis for this provision.

The EPA considers these comments to be outside the scope of the proposal. EPA noted in the preamble to the monitoring proposal that some existing regulatory language was being reprinted without change and that such reprinting was done solely for the readers' convenience to aid in viewing the proposal in a single context (71 FR 2712). EPA also stated that all of the background description of existing regulatory provisions—including the provision the commenters challenged—was presented not to reexamine any of the background provisions but rather “to facilitate informed public comment” on certain aspects of the proposal other than these background provisions. These other provisions were “requirements for the proposed PM

10−2.5

NAAQS”, “provisions for special purpose monitors”, provisions “related to the required spacing between ozone monitors and roadways”, and “certain quality assurance requirements” (71 FR at 2719). EPA thus did not seek comment on, reconsider, or otherwise reopen the pre-existing provision regarding population-oriented PM

2.5

monitors (or any of the other provisions recited in the background section). The EPA notes, however, that the pre-existing rule and this final rule do provide the same definition of population-oriented, in 40 CFR 58.1 Definitions, which while not quantified in terms of population affected has served to guide PM

2.5

monitor placement and interpretation of monitoring data since 1997.

The most controversial portion of this part of the proposal dealt with issues pertaining to the proposed NAAQS for

10−2.5

. The EPA proposed a new five-part suitability test for the comparison of PM

10−2.5

data to the proposed qualified PM

10−2.5

indicator. This test included an urbanized area population criterion, a block group population density criterion, a requirement for sites to be population oriented, an exclusion for source-influenced microscale sites, and a site-specific assessment to insure that data were dominated by certain sources of concern. See 71 FR 2736-2738. The EPA received extensive comment on the proposed PM

10−2.5

qualified indicator and on the proposed PM

10−2.5

NAAQS five-part site-suitability test. These issues are now moot since EPA is not adopting a NAAQS using a PM

10−2.5

indicator. See also section III.C of the preamble to the final rule adopting revisions to the PM NAAQS which explains why EPA did not adopt the proposed qualified indicator for thoracic coarse particles and why the proposed monitoring suitability criteria proved to be inappropriate.

C. Appendix A—Quality Assurance Requirements for State and Local Air Monitoring Stations and Prevention of Significant Deterioration Air Monitoring

A quality system provides a framework for planning, implementing and assessing work performed by an organization and for carrying out required quality assurance (QA) and quality control (QC) activities. The proposed amendments to 40 CFR part 58, appendix A were intended to provide the requirements necessary to develop quality systems for monitoring the pollutants of SO

, NO

, O

, CO, PM

2.5

, PM

and PM

10−2.5

at SLAMS stations including NCore stations, PAMS, and Prevention of Significant Deterioration (PSD) networks, and SPM stations using FRM, FEM, or ARM monitors. The proposed revisions addressed responsibilities for implementing the quality system for EPA and monitoring organizations. They also addressed adherence to EPA's QA policy, DQOs, and the minimum QC requirements and performance evaluations needed to assess the data quality indicators of precision, bias, detectability, and completeness. In addition, the proposed amendments described the required frequency of the QC requirements and performance evaluations, the data to be collected, and the statistical calculations for estimates of the data quality indicators at various levels of aggregation. The revised statistical calculations would be used to determine attainment of the DQOs. The proposed amendments also addressed required auditing programs to help determine and ensure data quality comparability across individual monitoring programs.

The EPA received some comments expressing concerns about the funding of the quality system. Funding issues are addressed in section III.E of this preamble. Substantive and procedural issues are addressed here.

1. General Quality Assurance Requirements

The EPA proposed to revise or include a number of general QA provisions that would serve to consolidate information and to ensure conformance to the QA requirements specified in EPA Order 5360.1 A2.

The EPA proposed to consolidate the QA requirements for SLAMS and PSD stations from two separate appendices, 40 CFR part 58, appendices A and B, into one single appendix A because both programs have similar QA requirements. See 71 FR 2725. The EPA received only endorsements on the proposed consolidation and therefore this final rule consolidates these appendices.

The EPA proposed to revise the part 58 appendix A to conform to the current EPA Quality Assurance Policies in EPA Order 5360.1 A2 which requires agencies that accept Federal grant funding for their air monitoring programs to have a QA program with certain elements including quality management plans (QMPs), quality assurance project plans (QAPPs), and the identification of a QA management function. EPA received three sets of comments endorsing the revision and received one comment expressing concern about the identification of the QA manager function. See 71 FR 2725. The proposed regulation would not have required that monitoring organizations identify a QA manager but would have required that they provide for a QA management function, which provides for independent oversight of the ambient air monitoring quality system. The EPA feels that the proposed language captures the essence of the requirements in EPA Order 5360.1A2, while accommodating the diverse nature of the ambient air monitoring community which is made up of large and small (local and Tribal) organizations. Consistent with the majority of positive feedback, and the need for conformance to the EPA Order, this final rule matches the proposed rule on this point.

The EPA proposed to revise the QA program by emphasizing the DQO process. See 71 FR 2725. A DQO is a qualitative and quantitative statement that defines the appropriate quality of data needed for a particular decision—for example, the data quality necessary for EPA or a monitoring organization to make data comparisons against the NAAQS. The DQOs help to establish the requirements for the data quality indicators of precision, bias, completeness, and detectability and the rationale for the acceptance criteria for these indicators. The EPA received a number of endorsements on this approach and did not receive negative comments. This final rule matches the proposed rule.

2. Specific Requirements for PM

10−2.5,

2.5

, PM

and Total Suspended Particulates

The EPA proposed to revise some of the PM

2.5

and PM

QA requirements in an attempt to provide consistency in implementation and assessment. Since PM

10−2.5

monitoring was proposed to be required, EPA included similar QA requirements for this monitoring. These requirements included the implementation of flow rate audits conducted by the monitoring organization, collocated monitoring, and performance evaluations.

The EPA proposed to make all the requirements for flow rate verifications and audits consistent among the PM

10−2.5

, PM

2.5

, and PM

methods. See 71 FR 2728. This requirement would have increased the audit frequency for PM

monitoring and decreased the audit frequency for PM

2.5

monitoring. Most commenters endorsed the proposed approach but a few commenters voiced concerns regarding the increased frequency for high-volume samplers for PM

and total suspended particulates (TSP) which operate somewhat differently and are not as easy to audit. The EPA reviewed the comments and revised the flowrate verification requirement from monthly to quarterly for the hi-volume manual instruments sampling for PM

and TSP only.

The EPA proposed to revise the sampling frequency for the implementation of the PM

2.5

Performance Evaluation Program (PEP). See 71 FR 2726. This proposed approach, based on historical PM

2.5

precision and bias data, identified the minimum number of performance evaluations required for all primary quality assurance organizations to provide an adequate assessment of bias, rather than the current requirement that a uniform 25 percent of monitors in a primary quality assurance organization be evaluated each year. The revision would establish a suitable sampling frequency of five valid audits a year for organizations with less than or equal to five monitoring sites and eight valid

audits a year for those organizations with greater than five monitoring sites. The majority of commenters approved of the PEP reduction frequency. A few commenters suggested that some primary quality assurance organizations do not need to be audited and said PEP audits should only focus on those producing inferior results. The EPA disagrees with this comment and believes that because the PEP program needs to provide a periodic estimate of bias for each primary quality assurance organization, the program must be implemented at each primary quality assurance organization.

There was also a comment suggesting further reductions to the auditing frequency or requiring the same number of audits over a longer period of time. The proposed audit cycle is based on 3 years since that is how many years of data are collected for comparison the PM

2.5

NAAQS. Therefore, the audit cycle frequency was based on the number of audit values needed to provide EPA the confidence in our bias estimates at the primary quality assurance organization over a 3 year period. Therefore, this final rule matches the proposed rule.

The EPA proposed to reduce the lower ends of concentration limits for which collocated data can be used to provide precision estimates. See 71 FR 2727. The lower ends of concentration limits would be reduced from 6 micrograms per cubic meter (μ/m

) to 3 μ/m

for PM

2.5

and PM

10c

(low-volume samplers) and from 20 μ/m

to 15 μ/m

for PM

(high-volume samplers). Statistical evaluation of 3 years of PM

2.5

and PM

data revealed comparable estimates of precision using data from both of these reduced concentration ranges, and also revealed that the addition of the data at these lower ranges will increase the level of confidence in the precision estimates. The majority of commenters endorsed the approach but there were a few commenters who were concerned that the lower concentrations, based on the statistics used to estimate precision, might lead to greater imprecision estimates. The evaluation that EPA made with the data from these lower concentrations included did not show any major increase in imprecision compared to omitting those data.

Since EPA has proposed the use of target upper confidence limits for statistical assessments and an upper confidence limit is influenced by sample size, lowering the concentration values tends to tighten or lower the confidence limits because more data points are available in the sample and therefore offsets any greater variability that might be associated with lower concentrations. Therefore this final rule matches the proposed rule.

“Proposal to Change PM

2.5

and PM

Collocation Sampling Frequency Requriements,” Mike Papp and Louise Camalier; November 2005.

http://www.epa.gov/ttn/amtic/pmgainf.html.

Based upon the decision that there is no need to implement a PM

10−2.5

monitoring program broad enough to systematically determine attainment/nonattainment with a PM

10−2.5

NAAQS, EPA has modified the proposed PM

10−2.5

collocation precision requirement and the Performance Evaluation Program (PEP) requirements in this final rule. See 71 FR 2726. The proposed quality system for PM

10−2.5

was developed for NAAQS comparison purposes and would have provided reliable precision and bias estimates at the primary quality assurance organization level of aggregation. However, EPA is not adopting a NAAQS using a PM

10−2.5

indicator at this time, so EPA is now requiring a network of PM

10−2.5

monitors only at NCore stations. The goal of these monitors will be to improve our understanding of PM

10−2.5

, support health studies for future reviews of the NAAQS, and promote improvements in the monitoring technology. States may choose to operate additional PM

10−2.5

monitors. With this in mind, the quality system need not be focused on the data quality assessments at the primary quality assurance organization level of aggregation but rather can and should be focused on understanding and controlling the data quality of each of the methods used to collect PM

10−2.5

. Also, since it is now anticipated that a primary quality assurance organization would have very few PM

10−2.5

sites, the proposal, if adopted without change, would have required almost every NCore site to have a collocated second PM

10−2.5

monitor, and the proposal would not provide for assessment of FEM precision even if FEMs are approved and deployed in place of some or most FRMs since as proposed the first collocation requirement of an FEM in a primary quality assurance organization would always be with a FRM. To avoid these undesirable outcomes, this final rule requires fewer collocated samplers than the proposal would have. Under this final rule, EPA will ensure that collocated sampling for estimating precision be implemented at 15 percent of FRMs (all FRMs aggregated) and 15 percent of the FEMs of each method designation. The number of collocated sites would thus be based on the size of the final PM

10−2.5

network. In order to provide a distribution of collocation across the United States, EPA will require, at a minimum, one collocated site in each EPA Region. The Regional Administrator shall select the sites for collocation. The site selection process will also consider selecting States with more than one PM

10−2.5

site to have one or two of the required collocations and will aim for an appropriate distribution among rural and urban sites.

For the PEP, this final rule departs from the proposal by requiring only one PEP audit at one PM

10−2.5

site in each primary quality assurance organization each year. The proposed rule would have required five or eight PEP audits for PM

10−2.5

in each organization. See 71 FR 2787, 2788. Since the PEP is already being run, at present, for the PM

2.5

network and it is expected that the PM

10−2.5

FRMs will utilize the same FRMs as the PM

2.5

samplers, the PEP audit for the PM

10−2.5

site can count towards the required number of PEP audits for PM

2.5

sites. It will be necessary to place a PM

10c

PEP sampler at the NCore site also but, this incremental requirement will not be a significant additional resource burden. When and if FEMs are implemented at some PM

10−2.5

sites, the PEP audit will be an additional audit at those particular sites and will require additional resources for auditing.

The incremental cost of placing and operating PM

10−2.5

samplers for purposes of tracking precision will also be minor in most cases. Many of the primary quality assurance organizations that will implement the PM

10−2.5

monitor at NCore sites are required to implement PM

2.5

and PM

networks. Some or most of the initial PM

10−2.5

deployments will be with manual FRM instruments, similar to the instruments used in the PM

2.5

networks and to some of the instruments used in the PM

networks. The EPA will allow collocated PM

10−2.5

monitors to be included in the primary quality assurance organization's count for required PM

2.5

and PM

collocation. In most cases, the primary quality assurance organization's collocation requirements for FRMs will not increase overall, since it is not anticipated that any one primary quality assurance organization will have many additional PM

10−2.5

sites that are not already both PM

2.5

and PM

sites. The only restriction to this aggregated collocation count will be for monitoring organizations that are operating high-volume PM

samplers. Since the PM

10c

monitor in a PM

10−2.5

FRM will be a low-volume sampler, PM

high-volume and PM

low-volume samplers cannot be aggregated together in the collocation

count and at least one collocated monitor must be identified for each type within primary quality assurance organization. Therefore, it is expected that the 15 percent collocation requirement for PM

10−2.5

FRMs will not actually increase the overall collocation burden at the majority of the primary quality assurance organizations beyond what they would have been required to implement for their PM

and PM

2.5

networks.

For any FEMs that might be used at PM

10−2.5

sites, EPA will require 15 percent collocation of each method designation or at least two collocations within each method designation. The EPA will require two collocations in order to collocate one FEM instrument with the same method designation to provide estimates of within method precision and collocate a second with an FRM to provide for an estimate of bias. These collocations would not necessarily need to be at separate monitoring sites.

3. Particulate Matter Performance Evaluation Program and National Performance Audit Programs

The EPA proposed to revise the current regulatory requirements dealing with responsibilities for independent assessments of monitoring system performance. See 71 FR 2726. These evaluations are the subject of sections 2.4 and 3.5.3.1 of the existing (1997) appendix A to 40 CFR part 58. Section 2.4 of appendix A to 40 CFR part 58 applied to all NAAQS pollutants and section 3.5.3.1 applied only to PM

2.5

The EPA proposed to revise the text of 40 CFR part 58, appendix A to cover PM

10−2.5

and also to clarify that it is the responsibility of each monitoring organization to make arrangements for, and to provide any necessary funding for, the conduct of adequate independent performance evaluations of all its FRM or FEM criteria pollutant monitors. The proposed language also clearly indicates that it is the monitoring organization's choice whether to obtain its independent performance evaluations through EPA's National Performance Audit Program (NPAP) and PM

2.5

PEP programs, or from some other independent organization. An independent organization could be another unit of the same agency that is sufficiently separated in terms of organizational reporting and which can provide for independent filter weighing and performance evaluation auditing. The proposed approach would ensure that adequate and independent audits are performed and would provide flexibility in the implementation approach.

Monitoring organizations that choose to comply with the revised provisions of appendix A to 40 CFR part 58 regarding performance evaluations by relying on EPA audits, for PM

2.5

, PM

10−2.5

, and/or other NAAQS pollutants, would be required to agree that EPA hold back part of the grant funds they would otherwise receive directly. These funds would be used by EPA to hire contractors to perform the audits and to purchase expendable supplies. To ensure national consistency and effective audits, EPA included provisions to ensure certification of data comparability for audit services not provided by EPA and for traceability of gases and other audit standards to national standards maintained by the National Institute for Standards and Technology.

The EPA received a broad range of comments on this proposed revision. The EPA received a few comments in support of these programs and one commenter felt that the PEP audits should be increased. In general, the comments expressing concern with the proposed language did not suggest that these programs were not necessary but were concerned about some technical aspects of the programs or with funding implications. Funding issues are addressed in section III.E of this preamble.

The EPA received a number of comments expressing concerns that allowing the monitoring agencies to implement the audit programs themselves or through third parties would increase the variability in the performance evaluation data. Since one of the major goals in the historically centralized and federally implemented PEP and NPAP programs has been the evaluation of data comparability, EPA is also concerned about any additional variability and its effect on data comparability. It has been EPA's practice with regard to any State which already performs these audits to perform side-by-side comparisons of EPA's equipment and procedures and the State's procedures to ensure both are producing results of acceptable quality. The EPA has successfully performed these comparisons with the California Air Resources Board's audit system. These comparisons will be expanded to include any additional States which choose to perform audits themselves or through third parties, rather than ask EPA to do so. During the comment period, EPA asked the monitoring organizations whether or not, assuming finalization of the proposed rule changes, they would continue to use the federally implemented program or perform the audits itself. For 2007, only three monitoring organizations (besides the one already implementing NPAP) opted to implement the NPAP and three monitoring organizations (besides the two already implementing PEP) opted to implement the PEP. The EPA believes it has the capability to ensure these State will implement programs will produce data of a quality comparable to the Federally implemented program.

The EPA also received comments stating concerns about the stringency of the definition of adequate and independent. Adequacy refers to the number of audits administered at any primary quality assurance organization and the technical procedures used in the audits. This final rule does not require any additional adequacy requirements above and beyond what EPA currently implements for the federally implemented program. The EPA evaluates data quality at the aggregation called “reporting organization” (which was changed to “primary quality assurance organization” in the proposal). The EPA feels that it needs to collect enough data to be able to judge data quality within each primary quality assurance organization over the same period that it uses the data for comparison to the NAAQS (3 years).

In the case of the PEP for PM

2.5

, today's action requires five audits per year for organizations with five or fewer sites and eight audits for those organizations with greater than five sites, the same as proposed. The number of audits aggregated over three years provides a reasonable estimate of bias at a primary quality assurance organization within an acceptable level of confidence. For the NPAP program addressing NAAQS for CO, SO

, Pb, and NO

, the goal is to perform audits on about 20 percent of the sites each year, but since there may be a number of high priority sites within a primary quality assurance organization that should be audited more often, it is anticipated that NPAP might audit each site within a primary quality assurance organization over about 7 to 8 years. This 20 percent goal is the current EPA practice, but was not proposed to be required by rule and, therefore, does not appear in this final rule.

There were a few comments suggesting that some primary quality assurance organizations do not need to be audited and that EPA mandatory audits for CO, SO

, Pb, and NO

should only focus on those organizations producing inferior results. The EPA continues to believe that it is important to develop an estimate of bias for each primary quality assurance organization. To do this, the audit program must be

implemented at each primary quality assurance organization. The NPAP audits using a through-the-probe approach, which is generally not how audits are performed by the primary quality assurance organizations themselves. By auditing some stations within a primary quality assurance organization each year using the through-the-probe approach, the NPAP can identify problems which the organization may not be aware of on its own. Also, EPA continues to believe that it is necessary to provide an adequate assessment of data comparability of all primary quality assurance organizations every year.

There were also comments concerning the requirement to use independent filter weighing laboratories for the implementation of the PEP. When EPA first implemented the PEP program, EPA established two independent laboratories to weigh filters for the PEP audits. Due to program efficiencies, EPA is now using one filter weighing laboratory. If primary quality assurance organizations implement the PEP themselves, they should not be able to utilize the same laboratory in which they weigh their routine sampler filters since any bias or contamination that might occur at the routine lab will also be “passed on” to the PEP filter. Because the PEP provides an estimate of bias (systematic error), it is necessary to avoid having a systematic bias occurring in the routine filter weighing lab affect both the PEP filters and the routine filters. Primary quality assurance organizations interested in implementing the PEP themselves have the option to make arrangements with other State labs, contractor labs, or utilize the PEP national lab.

The EPA believes that both the NPAP and PEP programs serve as an integral part of the overall ambient air monitoring program quality system and provide EPA and the public with independent and objective assessments of data quality and data comparability. Both programs provide the only quantitative independent assessments of data quality at a national level. Therefore, the proposed language was not changed and this final rule matches the proposed rule.

4. Revisions to Precision and Bias Statistics

The EPA proposed to change the statistics for assessment of precision and bias for criteria pollutants. See 71 FR 2727. Two important data quality indicators that are needed to assess the achievement of DQOs are bias and precision. Statistics in the current requirements of 40 CFR part 58, appendix A (with the exception of PM

2.5

) combine precision and bias together into a probability limit at the primary quality assurance organization level of aggregation. Since the standard EPA DQO process uses separate estimates of precision and bias, EPA examined separated assessment methods that were statistically reasonable and simple.

For SO

, NO

, CO, and O

, EPA proposed to estimate precision and bias on confidence intervals at the site level of data aggregation rather than the primary quality assurance organization. Estimates at the site level can be accomplished with the automated methods for SO

, NO

, CO, and O

because there is sufficient QC information collected at the site level to perform adequate assessments.

The precision and bias statistics for PM measurements (PM

, PM

10−2.5

and

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.