fr-
        \&
*f  OFFICE OF INSPECTOR GENERAL
        Evaluation Report
                Decline In EPA Participate Matter
                Methods Development Activities
                May Hamper Timely Achievement
                of Prog ram Goals

                Report No. 2003-P-00016
                September 30,2003
      p

-------
Report Contributors:
                      Patrick J. Milligan
                      Mark S. Phillips
                      Patrick J. O'Malley
                      Tiffine Johnson-Davis
Abbreviations

EPA
FRM
GAO
OAQPS
OIG
ORD
PM
STAPPA/ALAPCO

TSP
Environmental Protection Agency
Federal Reference Method
General Accounting Office
Office of Air Quality Planning and Standards
Office of Inspector General
Office of Research and Development
Particulate Matter
State and Territorial Air Pollution Program Administrators/
   Association of Local Air Pollution Control Officials
Total Suspended Particulates
Cover photo: Ambient air monitoring station in Wilmington, Delaware (obtained from
             Delaware Department of Natural Resources and Environmental Control web site).

-------
                     UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                                   WASHINGTON, D.C. 20460
                                                                          OFFICE OF
                                                                      INSPECTOR GENERAL
                                 September 30,2003
MEMORANDUM
SUBJECT:
                  Decline in EPA Particulate Matter Methods Development Activities May
                  Hamper Timely Achievement of Program Goals
                  Report No. 2003-P-00016

FROM:           J. Rick Beusse Is/
                  Director for Program Evaluation, Air Quality Issues

TO:               Jeffrey R. Holmstead
                  Assistant Administrator for Air and Radiation (6101 A)

                  Dr. J. Paul Oilman
                  Assistant Administrator for Research and Development (8101R)

Attached is our final report regarding the Environmental Protection Agency (EPA) Particulate
Matter methods development activities. This report contains findings regarding the decline in
methods development activities and how this decline may hamper timely achievement of
program goals. Also, the report contains corrective actions the Office of Inspector General (OIG)
recommends.  This report represents the opinion of the OIG and the findings contained in this
report do not necessarily represent the final EPA position. Final determinations on matters in
this report will be made by EPA managers in accordance with established procedures.

EPA's Office of Air and Radiation provided us with a response on September 29, 2003, that
consolidated its comments to the draft report with those from the  Office of Research and
Development (ORD).  We included the Agency consolidated response in its entirety as
Appendix B.  Independent of the Agency's consolidated response, ORD officials requested that
we include its own response as a separate attachment. Although we believe the Agency's official
consolidated response adequately reflects ORD's position, ORD officials requested that we
include the transmittal memorandum as an attachment to  our final report (see Appendix D.)

Action Required

In accordance with EPA Directive 2750, as the action official, you are required to provide this
office with a written response within 90 days of the final report date. Since this report deals
primarily with Office of Air and Radiation's Particulate Matter Program, the Assistant
Administrator of the Office of Air and Radiation was designated the primary action official.


-------
As such, he should take the lead in coordinating the Agency's response.  The response should
address all recommendations.  For the corrective actions planned but not completed by the
response date, please describe the actions that are ongoing and provide a timetable for
completion.  If you disagree with a recommendation, please provide alternative actions
addressing the findings reported. We appreciate the efforts of EPA officials and staff, as well as
external stakeholders, in working with us to develop this report. For your convenience, this
report will be available at http://www.epa. gov/oig/

If you or your staff have any questions regarding this report, please contact me at (919) 541-5747
or Patrick Milligan, Assignment Manager, at (215) 814-2326.

-------
                      Executive Summary
Purpose
             Paniculate Matter (PM) is a complex mixture of harmful solid and liquid particles
             found in ambient (outdoor) air. PM2 5 - the smallest of the two types of regulated
             airborne particles - represents "fine" particles that are less than or equal to
             2.5 microns in diameter, or about l/30th the thickness of a human hair. In addition
             to PM2 5, the Environmental Protection Agency (EPA) also regulates PM10 particles
             less than or equal to 10 microris in diameter.  Exposure to excess levels of PM can
             result in respiratory and other health-related illnesses and, in some cases, premature
             death. EPA estimates that, annually, over 15,000 deaths in the United States may
             be attributable to PM exposure.

             A fundamental component in ensuring reliable, valid, and complete PM
             measurements is "methods development" - the process of developing, evaluating,
             and validating the methods and associated instrumentation for measuring PM in
             ambient air.  These measurements are the first step in identifying those areas of the
             country where people are exposed to unhealthy levels of airborne paniculate matter.
             Accordingly, our objectives were to answer the following:

             •   Has EPA supported PM2 5 methods development activities to the extent
                necessary to achieve the short- and long-term goals of the PM2j program?

             •   Are increased methods development activities needed to ensure the timely,
                cost-effective achievement of PM program goals?
Results in Brief
             EPA has not supported PM2 s methods development activities to the extent
             necessary to fully achieve the short- and long-range goals of the PM25 program in a
             timely manner. The Agency is about 18 months behind schedule in obtaining
             sufficient PM2 5 monitoring data needed to determine areas with excess PM2 5
             levels, also known as making attainment designations, although EPA is planning to
             reduce Ihe delay to about 9 months. Once areas have been identified, State and
             local agencies can develop emissions control strategies describing the specific
             actions to be taken to bring these areas into attainment with the health-based PM2 5
             standard. Thus, the delay in obtaining sufficient ambient air data is one of several
             contributing factors that could result in millions of individuals being exposed to
             excess levels of PM longer than planned. Increased methods development activities
             are needed to address current monitor limitations and ensure that an appropriate PM
             monitoring network is in place to best enable State and local agencies to develop
             plans to effectively reduce excess levels.  Also, increased methods development

-------
             activities will be needed to address future PM monitoring needs, including
             supporting homeland security and enabling areas to meet future PM standards, such
             as the new PM "coarse" standard being developed.

             A significant reason for delays in achieving PM program goals was a gradual and
             largely unintended decline in EPA's methods development activities related to PM
             over the last decade.  According to Agency officials, the key factors that contributed
             to the gradual  decline in PM25 methods development included the loss of technical
             expertise due to retirements, a shift in Office of Research and Development
             priorities, and  the lack of a comprehensive methods development strategy to guard
             against unintended decreases in methods development support. As a result,
             methods development activities need to be increased to ensure the timely,
             cost-effective achievement of PM program goals.

             EPA is aware of the problems with measuring PM, and many in the Agency have
             worked diligently to correct those problems. Still, these problems increased State,
             local, and tribal agency labor costs and resulted in data from many PM2 5
             monitoring sites  being lost, delayed, or determined to be unusable. Consequently,
             as the Agency  moves forward, these delays may impact EPA's progress in reducing
             human exposure to unhealthy levels  of PM.

Recommendations

             We recommend that the Assistant Administrator for Air and Radiation, in
             collaboration with the Assistant Administrator for Research and Development,
             develop  a comprehensive methods development strategy for the PM program.  We
             are also making recommendations to improve the development of cost effective
             monitoring methods for the impending PM "coarse" standard, maximize
             opportunities to support and enhance EPA's homeland security efforts, address
             current technology limitations, re-evaluate the current PM2 5 data quality
             requirements, and take needed human capital actions.

Agency Comments and OIG Evaluation

             EPA's Office of Air and Radiation provided us with a response that consolidated its
             comments to the draft report with those from ORD. Although the Agency had
             some disagreement with certain issues and underlying causes, it agreed to
             implement all  of our recommendations. We included the Agency consolidated
             response in its entirety as Appendix  B. However, EPA did not agree that the
             Agency had not supported PM25 methods development activities to the extent
             necessary to achieve the short- and long-range goals of the PM25 program.
             Nonetheless, our work with external stakeholders, particularly States tasked with
             implementing  the monitoring network, suggested otherwise. EPA also provided
             several technical clarifications and comments.  We have made changes in the report
             as a result of the Agency's consolidated response, as appropriate. Our evaluation of
             the Agency's consolidated response  is in Appendix C.

-------
                    Table of Contents
Executive Summary	   i
 Chapters
     1    Introduction  	  1
     2    Declining Methods Development Support
          Hampers Timely Achievement of Program Goals	  7
     3    Limitations of Existing PM Monitors Need To Be Addressed
          By Increasing Methods Development Activities  	 15
 Appendix
     A    Details on Scope and Methodology 	23

     B    Consolidated EPA Response to Draft Report	  29

     C    OIG Evaluation of EPA Response to Draft Report	 39

     D    ORD Response to Draft Report 	  43

     E    Distribution	  47

-------

-------
                                 Chapter 1
                                  Introduction
Purpose

             Particulate Matter (PM) is ihe complex mixture of harmful solid and liquid particles
             found in ambient (outdoor) air. PM2 5 represents fine particles that are less than or
             equal to 2.5 microns in diameter.  Exposure to unhealthy levels of PM1S can result
            . in respiratory and other health-related illnesses and, in some cases, premature death.
             The Environmental Protection Agency (EPA) estimates that 15,000 deaths in the
             U.S. may be attributable to PM exposure annually.1 Reliable, valid, and complete
             measurements of PM are essential to developing control strategies and gauging the
             success of Federal, State, local, and industry activities designed to protect human
             health. A key component in ensuring reliable, valid, and complete PM
             measurements is "methods development," the process of developing, evaluating,
             and validating the methods and associated instrumentation for measuring PM.  The
             ambient air measurements generated using these methods are the first step in
             identifying those areas of the county where people are exposed to unhealthy levels
             of airborne paniculate matter. Once these areas have been identified, State and
             local agencies can develop emissions control strategies describing the specific
             actions that industry and others must take to bring areas into attainment with the
             health-based PM2 5 standard. The objectives of our evaluation were to answer the
             following:

 -  ,          •  Has the EPA supported PM2 5 methods  development activities to the extent
                necessary to achieve the short- and long-term goals of the PM25 program?

             •  Are increased methods development activities needed to ensure the timely,
                cost-effective achievement of PM program goals?

Background

             What Is Particulate Matter?

             PM includes acids, metals, gases, and other harmful airborne substances that can be
             breathed into the lungs. PM particles include elemental carbon (soot) from diesel
             engines and wood combustion; sulfate formed from sulfur dioxide emissions from
             power plants and industrial facilities; nitrates formed from nitrogen oxide emissions
             from power plants, automobiles, and other types of combustion sources; and dust
             from roads.  Some particles are large or dark enough to be seen as soot or smoke;
       'in 2002, the World Health Organization estimated that as many as 500,000 deaths may be attributable to
PM exposure annually worldwide.

-------
              others are so small they can only be detected with an electron microscope.  These
              particles vary significantly by location.  Fine particles between one-half and
              2 microns may travel thousands of miles, while larger particles - 10 microns in size
              or larger - may only travel 100 miles or less.  Very large particles (100 microns)
              may only travel a few hundred meters. Distances for all sizes are strongly
              dependent on meteorology., such as wind speed.

              Since the early 1970s, EPA has been concerned about the adverse effects of PM on
              human health and the environment. The first airborne particles to be regulated were
              Total Suspended Particulates (TSP), which included a broad range of large and
              small particles. Today, EPA no longer monitors for TSP, but instead regulates
              several smaller-sized particles.  While the larger TSP particles  are kept from the
              lung tissues by impaction in the nose, throat, and larynx, smaller particles can slip
              past those body defenses and penetrate deep into the lungs. EPA regulates two
              categories of these smaller airborne particles, as shown in Table 1.1.

                              Table 1.1: Types of Regulated Particulate Matter
Typ**
PM,0
PM2.5
Description
Particles less than or equal to 10 microns in diameter (about
one-seventh the diameter of a human hair).
"Fine" particles, which are less than or equal to 2.5 microns in
diameter (about 1/30th the diameter of a human hair).
Date Regulated
1987
1997
              *  A new PM standard - PMC (known as "coarse") - is being considered by EPA to apply to the
                 fraction of PM between 2.5 and 10 microns. EPA's current schedule should provide fora
                 proposal of this standard in late 2004 and a final standard in late 2005.

              The newer category - PM25 - was established as a National Ambient Air Quality
              Standard in 1997 as a result of a growing body of scientific evidence indicating
              that these fine particles are most damaging to health since they can penetrate the
              lung tissues easier and deeper.  When breathed, particulate matter can accumulate
              in the respiratory system.  Fine particulate matter is associated with such adverse
              health effects as heart and lung disease and increased respiratory disease, and
              symptoms such as asthma, decreased lung function, and even premature death.
              Sensitive groups that appear to be at greatest risk include the elderly, individuals
              with cardiopulmonary disease,  and children. Also, PM is a major cause of reduced
              visibility, and adversely impacts vegetation and ecosystems.

              EPA's short-term goal for the PM program is to determine which areas exceed
              EPA's standards and by how much. Longer term PM goals include: (1) identifying
              and characterizing emissions sources2; (2) developing State, local, and tribal
              control strategies; (3) gauging the success of Federal, State, local, tribal, and
        Wlien referring to emissions sources, unless otherwise stated, we are also referring to the larger category of
sources such, as mobile, industrial, etc.

-------
industry activities to protect human health; and (4) aiding in the Nation's homeland
security efforts by early detection of selected biological and chemical terrorist
attacks.

What Are the Uses of Ambient PM Air Monitoring Data?

Ambient air monitoring systems are a critical part of the nation's air program
infrastructure. Methods for measuring PM in the air are used to both set and
implement the standards. The United States spends over $200 million annually on
ambient air monitoring. Moreover, industry spends billions to implement emission
reduction strategies. Ambient air monitoring data provide an important means of
determining emission reduction progress and whether our nation is achieving its
environmental goals. Ambient air monitoring data from PM monitors  are used to
make major decisions by stakeholders, including scientists, researchers,
lawmakers, and the public. Some of the key uses of ambient PM air monitoring
data are depicted in Table 1.2.

            Table 1.2: Key Uses of Ambient PM Air Monitoring Data
• Characterize air quality and the associated health and ecosystem Impacts
4
4
*
Establish air qua% standards
Wentify areas exceeding the air quality standards
identify pollution sources



4 Activate emergency control procedures that prevent or alleviate air pofiutkm episodes
*
Devetopetrtfision reduclionand contra) strategies

* Review/evaluate State Implementation Hans for reducing PM levete
*
*
Track Agency progress In reducing PM Jevefs
Assist research efforts by providing more information about:
• composition; interaction; and sources afPM;
• length of time peopte are in contact vwthPM (exposure); artd '
* soUrce-to-dose pathways and the critfcalrotrtes of exposure.


+• Support the Agency's short- and long-term planning through the use of air quality models to project the
impacton air quattty from bnpiementatibtt of new regulations and controls.
*
Support the Air QuaiRy Index, which presents <*aify information or* many harmM ppHซtants, The AiV
Qwality index may be lound in various media, such as USA rodsyandthe W&attw Channel.
• Enable EPA to report its progress to Congress and the public, in accordance with the Government
Performance and Results Act and ฃPA's Goat 1 , "Cfean Air/
In 1999, the EPA convened a "National Monitoring Strategy Committee" of
representatives from EPA and state, local, and tribal agencies to take an overall
assessment of the nation's air monitoring networks and make recommendations for
improving network design. -They developed a National Monitoring Strategy to re-
shape the monitoring program to accommodate both national and local needs,
improve information flow to the public, and incorporate new technologies and
measurements.

-------
What Is Methods Development?

Because EPA has not yet developed a comprehensive definition of which activities
comprise methods development, we identified key PM methods development
activities through discussions with more than 50 internal and external stakeholders.
According to these stakeholders, methods development is an integrated system for
meeting the information and data needs of the many users of ambient air
monitoring data In particular, "methods development" is the process of
developing, evaluating, and validating the methods and associated instrumentation
for reliable, valid,  and complete measurements of PM in ambient air. To achieve
environmental data needs, the methods development process requires considerable
time and highly-developed expertise since it includes: research and development;
testing and analysis; operational demonstration to identify and validate state-of-the-
art air sampling and measuring technologies; measurement against standards; and
development of monitoring instrumentation. These activities  are a key component
for each type of PM regulated.

After the particular method and associated durable instrumentation is developed, it
should be field tested (validated) to ensure it (1) meets specifications,
(2) successfully operates under real world conditions, and (3) generates quality data
that will meet the users' needs. Regarding the latter activity, EPA requires that
environmental data meet the Agency's data quality objectives, wherein data must
meet certain precision, accuracy, completeness, representativeness, and
comparability parameters before users can make reliable decisions from the data.
Ideally, the newly developed method should also collect and analyze data in the
most cost-effective manner, taking into consideration life cycle costs such as
operation and maintenance expenses.

What Is the Primary Monitor Used to Measure PM25?

The filter-based mass-only monitor is the Federal Reference Method (FRM) that
EPA designated for use in measuring PM25. When the PM25 regulation was issued
in 1997, it essentially prescribed the size particle that would be measured, the
instrumentation that must be used, how the monitor must be operated, and the data
quality objectives. State and local agencies must use this FRM method when
gathering data to determine whether an area is in attainment of the PM25 standard.
Agency officials indicated that the development of a filter-based FRM was
consistent with their scientific understanding of PM, the available technology, and
the stringent requirements for methods performance that must be met to achieve
the required health protection.

By drawing  air across a teflon filter at a pre-determined flow rate, filter-based
mass-only monitors collect PM2 5 particles over a 24-hour period.  Before each
filter is placed in the monitor, the State or local agency must weigh it Then, after
24 hours in the monitoring device, the exposed filter is taken to a laboratory,
conditioned  to account for humidity and other factors, and weighed again. The
difference in the before and after weight of a filter represents  the amount of PM25
                              4

-------
             mass that gathered on the filter, and is an estimate of the amount of fine paniculate
             matter that a person would breathe in a 24-hour period.  According to the 2002
             database provided by the Office of Air and Radiation's Office of Air Quality
             Planning and Standards (OAQPS), of the 2,033 PM25 monitors nationwide, 1,616
             were FRM filter-based mass-only monitors. The FRM filter monitors are used to
             identify those areas of the country where people are exposed to unhealthy levels of
             airborne particulate matter, and determine the extent that such levels exceed the
             PM2 5 standard. The remaining monitors are used to help ascertain the source and
             composition of the harmful airborne PM.

             How Many People Live in Areas That Exceed the PM25 Standard?

             Although data limitations have slowed EPA progress in making non-attainment
             designations to date, the Agency has developed preliminary estimates of the
             designations for the United States. As shown in Table 1.3 below, as of January
             2003, EPA estimated that at least one-third of the Nation's population lived in
             areas that would eventually be classified as exceeding the PM2 5 standard.

                    Table 1.3: EPA's Preliminary Estimates of PM15 Designation Status
Designation Status
Attainment
Non-attainment
Incomplete Data (status not yet determined)
Unclassified (no monitors; attainment assumed)
Total
Estimated
Population Affected
56,000,000
101,000,000
79,000,000
49,000,000
285,000,000
Estimated Percent
of Population
20%
35%
28%
17%
100%
Scope and Methodology
             To assess whether EPA has supported method development activities to the extent
             necessary to achieve the goals of the PM program, we reviewed numerous reports.
             We also discussed the PM program with more than 50 key stakeholders, from
             EPA, selected States and local agencies, equipment manufacturers, key industry
             organizations, key health and environmental organizations, and academics.
             Additionally, we observed a PM ambient air monitoring station that included
             various types of monitors, and also visited a laboratory responsible for analyzing
             PM2 5 filters.  We conducted this evaluation in accordance with Government
             Auditing Standards, issued by the Comptroller General of the United States.
             Additional details on our scope and methodology are in Appendix A. Our
             fieldwork was conducted from November 2002 to May 2003.

-------

-------
                               Chapter  2

          Declining Methods Development Support

     Hampers Timely Achievement of Program Goals

            EPA is about 18 months3 behind schedule in obtaining sufficient PM25 monitoring
            data needed to determine areas with excess PM2 5 levels, also known as making
        •  '  attainment designations. However, EPA plans to make final designations by the
            end of 2004 which would reduce the delay from 18 months to about 9 months.
            State and local agencies encountered trouble operating about one-third of their
            monitors as they endeavored to obtain data on PM2 5 levels in their areas.
            A significant reason for Hie problems encountered was a gradual and largely
            unintended decline in EPA's methods development activities related to PM over
            the last decade. EPA has been aware of the problems with measuring PM, and
            many in the Agency have worked diligently to correct those problems. Still, these
            problems increased State, local, and tribal agency labor costs and resulted in data
            from many PM2 s monitoring sites being lost, delayed, or determined to be
            unusable.  Consequently, as the Agency moves forward, these delays may impact
            EPA's progress in reducing human exposure to unhealthy  levels of PM.

Unmet Data Needs Hampered Efforts

            Identification ofPM2S Non-attainment Areas Delayed

            State and local agencies were to begin collecting 3 years of PM2.5 air monitoring
            data in January 1999, with 3 years of data available by January 2002.  Based on the
            data, areas were to be designated in either "attainment" or "non-attainment" An
            area is considered non-attainment if the PM25 levels exceed the PM2.5 standard.4
            Areas in non-attainment must comply with Federal regulations designed to protect
            public health and the environment, including the development and implementation
            of emissions control strategies to reduce emissions of fine paniculate matter. Such
            efforts are costly and often take years to achieve program  goals. EPA estimates
            compliance with PM25  emissions control strategies will cost industry $37 billion
            annually by 2010. Therefore, the completeness, accuracy, and reliability of
            ambient monitoring data are critical.
      3Seotion 107 of the 1990 Clean Air Act required States to designate attainment upon promulgation of the
PMj,s National Ambient Air Quality Standard. Recognizing the monitoring network first needed to be deployed to
collect sufficient data, the 1998 Transportation Equity Act for the 21* Century allowed the States 3 years, until
September 2001, to gather the data. However, a complete set of data was not available until July 2003,

      ^he PM2,j standard requires that the annual arithmetic mean be less than or equal to 15 micrograms per
cubic meter of ambient air, and that the 98th percentile readings for peak concentrations over a 24-hour period be less
than 65 micrograms per cubic meter of ambient air.
                                         7

-------
Because the FRM monitors were deployed between 1998 and 2000, the 3 years of
data were staggered. Initially, EPA planned to make the first designations in 2002
and the last in 2005. However, EPA now does not plan to make the first
designations until December 2004, as much as 18 months behind schedule.
Consequently, the dates for when States in non-attainment would be required to
submit State Implementation Plans detailing their control strategies, as well as
when they would expect to be in an attainment status, were also delayed, as shown
in Table 2-1.

     Table 2.1: Implementation of PMZ9 National Ambient Air Quality Standard
                      and Related Designation Process
Activity
EPA Issues PM1S National Ambient Air Quality Standard
3 Years of PMi5 Data Available For Attainment Designations
EPA Designates PMas Non-attainment Areas
States Submit PMซ Control Strategies
States Have Up to 5 Years To Meet PM^ Standards
(5-Year Extension Possible)
Clean Atr Act
Required Dates
1997
2001
2002
2005
2007-2012
Current
Estimated Dales
1997
2003
2004
2007
2009-2014
Difficulties Encountered in Putting Monitoring Network in Place

EPA experienced difficulty in implementing the PM25 monitoring network,
resulting in incomplete, missing, or inaccurate data. For example, from 1999 to
2001, equipment malfunction and start-up problems contributed to 439 of 1,202,
or 37 percent, of the PM25 monitoring sites failing to operate in one or more of the
required 12 quarters (3 years). However, EPA officials believe that the issues and
problems encountered by the States were not atypical for the deployment of a large
network of new monitors across the country.

Public concern over the health effects of PM25 sparked presidential and
congressional directives to accelerate the PM2 5 ambient air monitoring network
implementation. As a result, EPA officials acknowledged that they hurried to
implement the PM2 5 filter-based mass-only monitoring network at a time when the
Agency had experienced a decline in resources devoted to methods development
activities, as well as a decline in methods expertise. Consequently, EPA did not
fully develop several key aspects of planning, implementing, and maintaining the
PM25 network, resulting in: (1) limited PM25 ambient air monitor options; (2) lack
of sufficient field testing; and (3) monitors that were not properly installed,
maintained, and operated.

   Limited PM2 5 Ambient Air Monitor Options. State officials told us that they
   had little involvement in the design and development of the PM2 5 ambient air
   monitors that they were expected to use to implement the PM25 network.  They
   believed they should have been continually involved through the design,

-------
                development, and selection of the new PM25 method. Largely because of the
                decline in EPA's resources and expertise for developing new methods, filter-
                based mass-only monitors  were the only viable option when ambient air
                monitoring implementation began in 1998. According to States, monitor
                manufacturers, and academics we contacted, had EPA invested more in
                methods development, the  Agency would have had more choices regarding the
                types of monitors to deploy and more knowledge about how the PM25 network
                should be configured.

                Throughout most of the 1990s, ORD had been primarily responsible for
                developing the PM2 s standard and the associated monitoring methods.
                However, OAQPS was undergoing an effort to update and improve its ambient
                air data quality objectives at the same time that ORD was developing the PM25
                standard, yet the two offices did not coordinate their efforts.  As a result of the
                tight timeframes and inadequate coordination, OAQPS officials said that the
                PM25 standard imposed by EPA's FRM standard may have been too stringent
                in relation to the quality of data needed, and that this effectively precluded
                continuous PM2 5 monitoring methods as a possible option for meeting the
                FRM equivalency determinations. OAQPS and ORD are re-evaluating the data
                quality objectives for PM25.  Such a modification could result in allowing
                certain PM2 5 continuous methods to demonstrate that they meet the PM2 s data
                quality objectives and could be used in lieu of filter monitors. This would
                provide EPA with more flexibility in implementing its monitoring program and
                provide a wider array of more useful data.

                Lack of Sufficient Field Testing. When a new monitoring method is
                developed, field testing is a vital step in ensuring that the monitor performs
                adequately under varied operating conditions before the network is fully
                deployed.  EPA and monitor manufacturers indicated time constraints forced
                EPA to purchase and deploy  filter-based mass-only PM2 5 monitors without full
                and adequate field testing under real-world conditions.  Consequently,
                malfunctions occurred that resulted in a loss of data. For example, monitors
                were only  field tested in warm conditions, and some monitors malfunctioned in
                cold climates. A General Accounting Office (GAO) report5 and our
                discussions with EPA personnel indicated insufficient field testing resulted in
                the following types of problems with one-third of the filter-based mass-only
                PM2 5 monitors:

                •  Oil freezing and water condensation causing monitor malfunctions,
                •  Filter cassettes j amming in cold weather.
                •  Improper sealing on monitor access doors allowing dust into the monitor
                   and contaminating the  filter.
      5GAO Report: Air Pollution - EPA's Actions to Resolve Concerns With the Fine Participate Monitoring
Program (August 1999, GAO/RCED-99-215).

-------
                •   Monitor cooling fans pulling in dust-containing coarse material and other
                    unwanted particles.
                •   New readings overwriting prior ones due to software malfunction.

                Our discussions with over 50 stakeholders, including ambient air monitor
                manufacturers, environmental groups, industry associations, State and local
                agency officials, and EPA, suggested that a principal underlying reason such
                problems occurred related largely to the lack of continued investment in
                methods development activities by EPA.

                Monitors Not Properly Installed, Maintained, and Operated. To maximize
                data quality and minimize post-deployment ambient air monitoring costs,
                written standard operating procedures need to be developed prior to monitor
                deployment. Due to limited resources and time constraints, EPA did not
                provide written standard operating procedures to State and local agencies prior
                to PM2 s monitor deployment. According to State and local agency officials
                contacted, such guidance would have helped ensure consistency in the
                operation of the monitors, provided direction on repairing mechanical
                malfunctions, and more fully explained how to handle  filters before and after
                sampling. Because EPA did not prepare such procedures, many States
                . developed their own, which  resulted in inconsistencies in the data collected.
                Due to insufficient training and support from EPA, State and local officials said
                they struggled with the installation, operation, and maintenance of the new
                monitors, particularly at the  outset. For example, they noted that:

                •   Insufficient troubleshooting instructions resulted in monitors needing to be
                    shipped back to the manufacturer rather than be repaired on-site.
                •   EPA did not initially provide funding for spare parts and monitors.
                •   States had difficulty establishing weighing facilities to handle  the massive
                    influx of filters to be analyzed.

             Operating in a reactive mode, EPA and State and local agencies corrected
             implementation problems as they arose. According to Agency officials, the
             deployment of the network was  extremely challenging and EPA could not predict
             in advance the issues that arose. They also said that, taking into consideration the
             magnitude of the effort and the time frames allowed, they believe their deployment
             of the PM2 5 network was quite successful. Still, the problems encountered
             increased State and local agency labor costs and resulted in data from many PM2 5
             monitoring sites being lost, delayed, or determined to be unusable.

Several  Factors Contributed to Decline in Methods Development

             EPA's reduction in methods development activities was a gradual process that
             began in the early 1990s before  the current PM25 standard was implemented.
             Although most of the reduction  was largely unintended, some was knowingly
                                          10

-------
             carried out. For example, the 1996 Exit Strategy6 written by ORD and OAQPS
             described ORD's plans to provide less support to OAQPS in areas such as routine
             technical support to sustain the network and quality assurance/quality control work,
             activities that were related to methods development However, there was general
             consensus  among the many stakeholders we contacted that PM methods
             development warrants increased attention by EPA, not less.  According to Agency
             officials, key factors that contributed to the gradual decline in PM2.5 methods
             development included the loss of technical expertise due to retirements, a shift in
             ORD priorities, and the lack of a comprehensive methods development strategy.
             Details follow.

                Attrition Resulted In Loss of In-House Expertise

                Agency officials said the loss of expertise has been occurring for years and will
                likely continue.  Many current EPA employees began their careers when EPA
                was founded in 1970 and have reached retirement eligibility.  EPA's problem
                of recruiting and retaining first-rate scientists was recently discussed in a
                January 2003 report by GAO, Major Management Challenges and Program
                Risks.  ORD has recognized the loss of expertise and its impact on methods
                development activities, and has initiated efforts to overcome it  For example,
                ORD recently hired a national PM methods expert; has developed a hiring
                strategy with plans to double the methods development staff (from
                approximately four to eight individuals); and has implemented a Post Doctorate
                Program aimed at attracting and retaining young, qualified staff. However, we
                believe more needs to be done under a formal process. GAO's Human Capital
                guide,  "Human  Capital: A Self-Assessment Checklist for Agency Leaders"
                (GAO/OGC-00-14G), September 2000, provides guidance that could help
                EPA managers improve their human capital systems, including hiring,
                retention, and succession planning.

                ORD Priorities Shifted to Other PM Research

                Many stakeholders described how ORD personnel moved toward research and
                away from the routine tasks associated with sustaining the air monitoring
                network and developing new methods.  In 1997, ORD found that airborne fine
                paniculate matter was much more of a health hazard than previously thought,
                and ORD's priorities shifted to conducting more fine particulate matter
                research, partly at the expense of methods development.  Agency officials said
                that a decline in expertise also contributed to the decision to shift focus because
                it became increasingly challenging to carry out existing duties with the
                continual loss of expertise. Another cause for ORD's shift to more traditional
                research activities, at the unintended expense of the Agency's methods
                development efforts, were the results of the National Research Council's
      "November 1996, NERL/ORD Exit Strategy - Quality Assurance Support To OAQPS/OAR
(Document is a description of decreased quality assurance support provided by ORD to OAQPS).
                                         11

-------
independent assessment of EPA's fine paniculate matter research efforts. The
Council made 10 recommendations emphasizing the need for more and better
research on PM health effects as they relate to risk, dose, and exposure to fine
paniculate matter.  Also, as part of performing this needed research, the
Council cited the need for increased methods development activities by EPA to
develop new and improved monitoring instrumentation. However, the
Council's more resounding message, as interpreted by EPA officials, was the
need to increase PM2 5 research to better understand the relationship between
fine paniculate matter exposure and human health.

Comprehensive Methods Development Strategy Lacking

EPA does not have a system or process that clearly defines the essential
activities that comprise PM methods development, nor is there an explicit
mechanism for determining  whether these activities are performed at a
sufficient level. A comprehensive methods development strategy would ensure
that methods activities are clearly defined and tracked, increase the likelihood
that needed activities are properly funded, and guard against unintended
decreases. Although there was no comprehensive methods development
strategy, we were able to identify through extensive interviews and research
some of EPA's key methods development, shown in Table 2.2.

             Table 2.2: Key Methods Development Activities
                               Activities
    Pra-ac&rely Researching New Monitoring Techniques and Instrumentation
 *  Identifying User* of ihe Data and Assessing Their Needs
 *  Developing standards and Associated Methods
 *  Designing Instrumentation and Sampling protocols
    Field Testing/Validating Instrumentation
 *  Modifying Methods, Instrumentation, and Sampling Protocols
 t  Configuring Network
 *  Deplซ?ying and Installing Instruments
     Operating, Maintaining, and Supporting the Network
    Re-assessing Networks and Seeking improvement
Both OAQPS and ORD officials agree that the activities listed above are vital
to successfully supporting the Agency's PM monitoring program, and
acknowledge that there were unintended reductions in the support for these
activities that adversely impacted the PM2 5 program. We also noted that
EPA's methods development activities for PM were not tracked as a separate
budget line item, which may have contributed to the decline in its emphasis.
We believe such tracking is needed to adequately monitor the program.
Further, even though there have been budget increases for PM research since
                           12

-------
                 1997, ORD and OAQPS confirmed that the additional funding did not go
                 toward methods development  The Government Performance and Results Act
                 of 1993 intended for Federal programs to link resources to activities that
                 produce results, and the Office of Inspector General and others have identified
                 EPA's difficulty in linking mission to management as a key management
                 challenge.7
Conclusions
             Methods development activities are a fundamental component in EPA's efforts to
             ensure that PM2 5 ambient monitors produce reliable, valid, and complete data on
             environmental conditions. We found that a significant reason that problems
             occurred in operating many of the PM25 monitors was due to the Agency's
             unintended reductions in the support for PM methods development activities.
             EPA has worked diligently to correct problems, and the Agency expects to have
             sufficient data to make non-attainment designations by December 2004. However,
             EPA needs new and different PM2.5 monitors to meet its goals. To achieve better
             results from these future PM2 s air monitoring efforts and other PM program goals,
             the Agency should develop a comprehensive strategy that would ensure that
             methods development activities are clearly defined and tracked, and properly
             funded.
Recommendations
             We recommend that the Assistant Administrator for Air and Radiation, in
             collaboration with the Assistant Administrator for Research and Development:

             2-1.    Develop a comprehensive PM methods development strategy that:

                    (a)  identifies the methods development activities critical to meeting the
                        Agency's short- and long-term PM program goals;
                    (b)  ensures that these PM methods activities are clearly defined, tracked,
                        and appropriately funded;
                    (c)  defines the roles and responsibilities of the EPA offices responsible
                        for conducting PM methods development activities; and '
                    (d)  guards against any unintended decrease in PM methods activities,
                        including the impact of lost program expertise through retirement or
                        career transition.

             2-2.    Use the EPA data quality objectives process to re-evaluate the PM2 5
                    program's data quality objectives and the associated PM25 ambient air
      7EPA's Progress in Using the Government Performance and Results Act to Manage for Results [EPA-OIG
2001-B-000001], June 13, 2001; Audit of EPA's Fiscal 2000 Financial Statements [EPA-OIG 2001-1-00107],
February 28,2001.  EPA - Major Management Challenges and Program Risks [GAO-01-257], January 2001.
                                          13

-------
                    monitoring methods requirements, and ensure that both are appropriate to
                    achieve the short- and long-term goals of the program.

             We also recommend that the Assistant Administrator for Research and
             Development:

             2-3.     Include in the ongoing revisions to ORD's Human Capital Strategy
                     explicit measures to minimize the impact of losing PM methods
                     development expertise, and that ORD use GAO's Human Capital guide,
                     "Human Capital: A Self-Assessment Checklist for Agency Leaders," to
                     help managers improve their human capital systems, particularly for
                     hiring, retention, and succession planning.

Agency Comments and OIG  Evaluation

             EPA made detailed comments to our draft report and, where appropriate, we made
             revisions. The Agency generally agreed with the recommendations in this chapter.
             With respect to Recommendation 2-1, the Agency supported the recommendation
             to develop a comprehensive methods development strategy, but commented that a
             strategy that focuses on  PN^ 5 is too narrow. We agree, and are now
             recommending that EPA develop a comprehensive paniculate matter strategy.
             The Agency's consolidated response and our evaluation of that consolidated
             response are in Appendices B and C, respectively.
                                         14

-------
                              Chapter3

              Limitations of Existing PM Monitors

             Need To Be Addressed By  Increasing

	Methods Development Activities   	

            EPA has not ensured that an appropriate monitoring network is in place to best
            enable States and local agencies to develop plans to effectively reduce excess
            levels of PM. Increased methods development activities will be needed to address
            future PM monitoring needs, such as supporting homeland security and enabling
            areas to meet future PM standards. For example, EPA is preparing to develop a
            new PM coarse (PMc) standard in 2004, and EPA is working with the Department
            of Homeland Security to develop a continuous ambient air monitoring method
            capable of identifying a biological or chemical attack.  Manufacturers and others
            told us that EPA will need to increase its support of PM methods development
            activities, improve its efforts to involve external stakeholders early in the methods
            development process, and do more to enhance the development of alternative PM
            monitors if earlier problems experienced in the PM2 5 program are to be avoided as
            the Agency mobilizes to meet these other PM program goals.

Use of Other Monitor Types Should Be Considered

            Each type of PM monitor serves a valuable purpose, including the filter-based
            mass-only monitor used as the approved FRM PM2 5 standard to protect public
            health.  However, Agency officials said that all of the types of PM monitors need
            improvement.  Many of the stakeholders we contacted said EPA needs to work
            with industry/manufacturers to improve the present limitations of PM monitors in
            three areas - monitor capability, data collection frequency, and cost. Using the
            Clean Air Scientific Advisory Committee process would provide the appropriate
            framework for such a government-industry partnership, ensuring that such
            activities are transparent to all interested and affected parties, and that such parties
            have an opportunity to provide input into the process.

            Filter-based mass-only monitors are the primary type of monitors being used to
            measure PM, but many of the stakeholders we contacted said two other types of
            PM monitors show promise, although they also have limitations. Both types -
            continuous mass monitors and speciation monitors - are available and have been
            used to develop certain types of PM data for specific applications. A general
            description of each is in Table 3.1.
                                       15

-------
                         Table 3.1: Two Alternate Types of PM Monitors
   Typs
                                    Description
Continuous
Mass
Monitors
Unlike the filter monitor, which measures the weight of a filter after a 24-hour period and
compares it to its weight prior to sampling, a continuous mass monitor analyzes and records the
concentration of the participate matter that pass through the sampling chamber, and then
mathematically calculates the mass of the particulate matter. No filter handling is involved, nor is
there a need for subsequent laboratory analysis. These monitors can record PM data as
frequently as every 5 minutes, but usually do so in 1 -hour increments. Different continuous
monitors may use different analytical techniques, depending on what is being measured.  For
example, to analyze for sulfates a flame ionization analytical technique may be used, while for
nitrates a chemiluminescence analytical technique may be used.
Speclatlon
Monitors
Speciation monitors identify the make-up of the particulate matter, which is necessary to trace the
sources of pollution and to better understand health effects of the particulate matter. These
monitors generally use several different inlet tubes and several different filters (nylon, teflon,
quartz) to collect suspected harmful components of the PM mixture, such as acids, metals, or
organic carbon. The filters are analyzed in a laboratory using many of the same analytical
techniques discussed above. Speciation data are critical to determining the toxicity of the
particulate matter and developing pollution abatement strategies.  Most chemical speciation
monitors are filter-based, but EPA is experimenting with continuous speciation instruments.
              Table 3.2 below illustrates some of the strengths and limitations of the basic
              PM monitor types currently in use.  [Note: Check mark (/) indicates monitor
              strengths; no check mark indicates a monitor limitation.]
              Table 3.2: Strengths and Limitations of Basic Types of PM Monitors
Attribute
Used for Designations
Least Expensive Capital Cost
Minimal Operational/Life Cycle Costs
Identifies pollution source and PM composition
Real time data
Filter can be stored (historical record)
Supports mapping/modeling
Supports health research
Identifies PM levels during day
Air Quality Index
FBter-Based
Mass
/
/



/
Minimal
Minimal


Continuous
Mass
Potential



/

/
/
/
/
Filter-Based
Speclatton1



/

/
/
S


Continuous
SpeclatJon2


/
/
/

/
/
/

1/ Filter-based speciation monitors measure PM constituents by weight in a specified period of time.
21 Continuous speciation monitors measure PM constituents by concentration at continual specified intervals.
              All current PM monitors, including the PM2 s FRM mass-only monitors, have a
              limitation related to volatility (evaporation) for nitrates and some organics. When
              temperatures rise, the nitrate or organic portion of the particulate matter may
              volatilize, thus understating the amount of PM2.S in the air. Monitors that rely on
              filters require that filters be refrigerated after collection to avoid some of this, but
              in high nitrate areas, volatilization still occurs.  For continuous monitors, the
              instruments actually heat the sample to remove moisture from the particulate
              matter for a more reh'able, valid,  and complete measurement; however, this heating
                                               16

-------
process causes some degree of volatilization. While volatilization may never be
entirely eliminated, attempts are being made to minimize the impact of this process
on continuous measurements.

It should be noted mat cost plays a factor in determining which PM monitors to
use.  As shown in Table 3.3, capital costs of PM25 monitors can range from
$11,000 to $85,000, and annual operating costs from $8,000 to $53,000.

           Table 3.3: Cost to Purchase and Operate PM^ PM  Monitors
Monitor Type
Filter-Based Speciation
Filter-Based Mass
Continuous Speciation
(nitrate, sulfate, and carton)
Continuous Mass
Estimated Costs ',
Capital
$20,000
$11 ,000 -$14,000
$85,000
$20,000
Annual Operation and
Maintenance
$53,000
$20,000 -$35,000
$20,000
$8,000
The filter-based mass-only equipment has the lowest initial cost but requires more
operation and maintenance resources because it is labor intensive (collection and
analysis) and requires frequent material replacement (a new filter every 24 hours).

Advanced Monitors Are Needed to Effectively Reduce PM Levels

EPA recognizes the need for other types of monitors if it is to achieve its PM
program goals.  For example, according to ORD's September 2001 Draft
Multi-year Plan for Particulate Matter Research:

     Methodology must be developed for continuous and more accurate
     measurements of PM mass and components such as nitrates, metals, and
     organic and elemental carbons.  This methodology will then be applied to
     specific PM characteristics which toxicology studies indicate may be
     responsible for adverse health effects.

Monitor manufacturers and State and EPA officials that we contacted agreed that
existing monitors need to be modified and new monitors created if EPA is to meet
the olher needs for PM data  Many also believed continuous mass and continuous
speciation monitors are among the most promising near-term technologies,
although Agency officials stated that much more research is needed. OAQPS'
National Monitoring Strategy recommends the replacement of 50 percent of PM25
filter-based monitors with continuously operating ones, and notes network
assessments should be used to determine whether goals and objectives are being
met in the most efficient way. EPA will need to ensure that all PM methods fully
address  data needs for both short- and long-term goals of all PM programs and
that, prior to implementation, the proposed methods are peer reviewed.
                             17

-------
       Also, although the filter-based mass-only monitoring network is close to obtaining
       the required 3 years of data for achieving the short-term goal of making attainment
       designations, the network will need to include the capacity to determine paniculate
       matter origin and composition on a national, regional, and local basis to effectively
       measure its progress in achieving longer term goals.  To determine where pollution
       is coming from, States must identify the sources of PM2 5  and the extent to which
       each source contributes to the overall levels.  Currently, over 93 percent (1,891 of
       2,033) of EPA's PM25 ambient air monitoring network consists of monitors that
       only measure paniculate matter mass, with less than 7 percent (142s of 2,033)
       obtaining data on particulate matter composition and origin.  In its PM2 5
       Implementation Plan, EPA said 300 speciation monitors - more than double the
       number presently deployed - would be needed to obtain  data on the composition
       and origin of PM25.

       Generally, PM25 consists of five  substances:  sulfate, nitrate, organic carbon,
       elemental carbon, and crustal material.  However, the composition of a particulate
       matter differs in various parts of the country. For example, particulate matter in
       the eastern half of the country is  comprised mainly of sulfate, while particulate
       matter in the northwest consists primarily of organic carbon. Therefore, it is
       important for States to know the  specific composition of PN^ 5  in its area so that it
       can take the most appropriate corrective actions. Particulate matter composition
       also influences the type of monitor needed. Chart 3.1 shows how the composition
       of PM2 5 particles varies across the continental United States.
                Chart  3.1 Variations of PM2.s Composition
        Relative composition of PM 2.5
        particle Is Illustrated by the
        following color scheme:
         ii Sulfate
         EH Nitrate
         fS Organic Carbon
         il Elamental Carbon
         H Crustal Material
       Source - OIG-developed chart based on data obtained from
             EPA's 1999 National Air Quality and Emissions Trends Report.
^The 142 speciation monitors are based on information provided by OAQPS in January 2003.
                                    18

-------
After an area is designated non-attainment, States will develop and submit for EPA
approval their State Implementation Plans describing the actions that will be taken
to bring areas into attainment with the PM2 5 standard. Among other things, the
State's Implementation Plan should: (1) identify the PM25 sources and the levels
contributed by each source; (2) develop, implement, and enforce new controls on
the PM25 sources targeted; and (3) determine the level of reductions needed to
bring an area into attainment within required milestones.

Volume 1 of a four-volume National Research Council Report recommended that
EPA first identify the particular components of PM and conduct studies to
understand human exposure and health effects of PM. EPA's response to this
report is essentially embodied in ORD's Multi-Year Plan for PM which described
the goals and targets of the PM program. Volume 3 of the National Research
Council report calls for EPA to increase its methods development activities to
better understand particulate matter composition and its impact on health.  The
Council stated that EPA's filter-based mass-only monitors provide limited data and
are not sufficient to address this increased need for health effects data Four of the
Council's 10 recommendations cited the need for more and improved speciation
data to better understand the composition of the particulate matter and to what
extent the public is at risk.  EPA officials noted mat while ORD was redirecting
some internal resources to respond to the National Research Council report, overall
EPA investment in methods work was being augmented by the Supersites Program.
One of the objectives of this program is to develop and test measurement methods.

New Monitors Needed for PMC Program Efforts

ORD and OAQPS are considering the need for a PMC "coarse" standard for
particulate matter between 2.5 microns and 10 microns in diameter. EPA's
schedule provides for a proposal in late 2004 and a final standard in late 2005.
EPA plans for the PMc standard and the associated ambient air monitors to replace
the current PM10 standard and associated monitors. Manufacturers told us EPA
will need to increase its support of PM methods development activities, improve
its efforts to involve them and other external stakeholders early in the methods
development process, and do more to foster the development of alternative PM
monitors. EPA will also need to ensure mat the methods fully address data needs
for both the short- and long-term goals of the PMC program and that, prior to
implementation, the proposed methods are peer reviewed.

EPA officials said there are numerous instruments capable of measuring PM^ and
ORD is conducting field test studies on these instruments. There are numerous
viewpoints on which monitoring method should be defined as the FRM for PMc.
Some officials believe that existing monitors can be modified, while others believe
a new monitor must be developed to measure coarse exclusively. Further, some
believe a monitor must be developed to measure PM2 5 and PMc simultaneously.
Some officials believe amass  weight standard like PM25 should be used to identify
how much coarse particulate matter is in the air, while others believe EPA should
regulate by composition, to identify what the toxic components are. To avoid
                             19

-------
some of the problems encountered while implementing the PM2 s standard, EPA,
State, and local agencies, and monitor manufacturers agree that the Agency needs
to take a more proactive approach to developing and implementing the PMC
standard.

Support of Homeland Security Also Requires Increased Efforts

The terrorist attacks of September 11,2001, and the subsequent anthrax releases
expanded EPA's mission, to include working with other Federal agencies in
protecting our country against the environmental and health consequences of
terrorist acts. Details of EPA's homeland security responsibilities are described in
EPA's  September 2002 Strategic Plan for Homeland Security. One of EPA's
primary roles will be providing technical expertise to identify the threat of
biological, chemical, radiological, and other terrorist attacks, as well as responding
to and recovering from such attacks.  EPA's expanded homeland security
responsibilities further demonstrate the need for increased activity in PM methods
development.

One of the homeland security goals is for EPA to work closely with other Federal
and State agencies to identify useful information already available. Under air
protection, a key source of information will be the data collected from the
Agency's existing ambient air monitoring network.  EPA is responsible for
ensuring that its existing monitoring expertise, standards, capabilities, and data are
appropriately integrated into the Agency's efforts to detect terrorist threats. The
Agency is also to make historic data available to determine trends and background
levels that will aid in establishing baselines for detection.  Monitoring data may
also provide a means of detecting anomalies in the ambient air that would indicate
further analysis is warranted. The existing ambient air monitoring network will be
valuable because there already exists an infrastructure of monitoring platforms,
electricity, and land. In addition, EPA has an established working relationship
with State and local agencies that should expedite efforts to develop a monitoring
network for homeland security purposes.

OAQPS officials said their goal is to develop a continuous ambient air monitoring
method that would result in real-time information for identifying a biological or
chemical attack. This network, in combination with other Agency activities, would
aid in tracking any plume of contamination and help to minimize exposure risks.
Agency officials said that the PM network, designed to detect and measure various
size particles found in the air, may aid in efforts to detect and measure bio-agents.
For example, anthrax spores, which typically range in  size from 2 to 6 microns, are
potentially within the range of particles that EPA measures with its PM monitors.
However, the current network does not provide data timely enough to readily
detect terrorist attacks, and speciation data takes longer to obtain.

EPA is currently evaluating many types of PM monitors and exploring potential
options for how monitors can help with homeland security efforts.  In the event of
a national incident, EPA is the lead agency responsible for decontaminating
                             20

-------
             affected areas and determining when it is safe to return.  To do so, EPA needs to
             enhance Hie deploy able component of its emergency response air monitoring
             efforts to quickly transport mobile/non-stationary monitors to contaminated areas.
Conclusions
             Monitor manufacturers, State, and EPA officials we contacted said that existing
             monitors need to be modified and new monitors created if EPA is to meet the other
             needs for PM data, such as the Agency's expanded homeland security
             responsibilities since September 11.  Presently, the Agency's monitors do not
             provide information timely enough to readily detect terrorist attacks. However,
             because of the costs associated with developing new monitoring methods and the
             associated instrumentation, manufacturers said EPA will have to take a greater
             leadership role in the development of such monitors, including methods
             development. As EPA moves to develop monitors for addressing longer-term data
             needs, insufficient attention to methods development activities not only presents a
             risk mat there will be lost or unusable data and increased operating costs, but also a
             risk that industry will be faced with unnecessary or overly stringent compliance
             costs. With billions of dollars in estimated annual costs of compliance to industry,
             EPA should ensure that the Agency's data quality objectives process is strictly
             followed in establishing the PMC FRM for the PM coarse standard, keeping in
             mind that methods that are too stringent impose unnecessary costs and methods
             that are too lenient and jeopardize data quality.
Recommendations
             We recommend that the Assistant Administrator for Air and Radiation, in
             collaboration with the Assistant Administrator for Research and Development:

             3-1.    Establish a government-industry partnership that facilitates an objective
                    assessment of PM measurement issues that need to be addressed to improve
                    PM ambient monitoring capability and data collection frequency at less
                    cost, and that provides for transparency and openness to the public.

             3-2.     Enhance the development of cost-effective methods for addressing PMc
                     program goals by:

                     (a)  ensuring that the shortcomings of the PM25 implementation process
                         will not be repeated;
                     (b)  using the EPA data quality objectives process to develop the PMC
                         standard planned for 2004 and the associated ambient air monitoring
                         requirements, and ensuring that both are appropriate to achieve the
                         short-  and long-term goals of the PMc program;
                     (c)  ensuring sufficient stakeholder involvement in developing the most
                         promising and cost effective monitors for addressing PMc program
                         goals;  and

                                          21

-------
             3-3.
(d)  ensuring that the proposed methods are externally peer reviewed prior
    to implementation.

Maximize opportunities for utilizing current and newly developed
monitors to support and enhance EPA's Homeland Security efforts.
Agency Comments and OIG Evaluation

             EPA made detailed comments to our draft report and, where appropriate, we made
             revisions. The Agency generally agreed with the recommendations in this chapter.
             With respect to Recommendation 3-1, the Agency did not agree that using the
             Federal Advisory Committee Act would be the best approach to establishing a
             government-industry partnership. EPA believed that using the Clean Air Scientific
             Advisory Committee process would be a suitable alternative. We agree that our
             draft recommendation would have unnecessarily restricted EPA's options for
             improving PM monitoring methods, and we have modified our recommendation to
             reflect this. The Agency's consolidated response and our evaluation of that
             response are in Appendices B and C, respectively.
                                        22

-------
                                                                         Appendix A
                   Details on Scope and Methodology
To assess whether EPA has supported methods development activities to the extent necessary to
achieve the short- and long-term goals of the PM25 program, we reviewed numerous reports about
the goals, objectives, and needs of the PM2S program, including the EPA's National Ambient Air
Monitoring Strategy; the National Research Council's Research Priorities for Airborne
Particulate Matter Volumes I, II, and III; EPA's Third External Review of Air Quality Criteria
for PM, and EPA's 1996 Exit Strategy, written by the ORD and OAQPS. We also reviewed a
prior GAO report, EPA's Actions to Resolve Concerns With the Fine Particulate Monitoring
Program (GAO/RCED-99-215), March 1999.  This GAO report addressed problems encountered
by State and local agencies when deploying the PM2 s filter-based mass-only monitoring network.

Additionally, to understand how various PM monitors are operated and maintained, and how and
what type of information is collected, we observed a PM ambient air monitoring station in
Wilmington, Delaware, equipped with several types of PM monitors, including one or more of the
following: (a) TSP monitor, (b) PM]0 filter-based monitor, (c) PM25 filter-based mass-only
monitor, (d) PM2 5 continuous mass monitor, and (e) PM2 5 speciation filter-based mass monitor.
To understand the PM2 5 laboratory and analytical processes, we observed a Philadelphia,
Pennsylvania, Department of Health laboratory responsible for analyzing PM25 filters.  '

To gain an understanding of the PM25 monitoring network and the role methods development
plays, we conducted interviews with 51 stakeholders.  The following table provides information
on interviews  conducted, and details on each type of stakeholder follow.
' * -, -.""-.
Internal
Interviews
Extensa
Interviews
Total
; • tatwvtewee* - •
ORD
OAQPS
EPA Regional Offices
State artf Ueal Air Protection Agencies
Air Monitor Manufeetwws
HeailfcSrtvifOftn'terttal Groups
Industrial Groitpa
Academia
AirAsBDeiatlaRS
Other fe
-------
•  Tisch Environmental, Inc.
•  URG, Inc.

We interviewed representatives from three health/environmental groups that have been actively
involved in the PM program:

•  American Lung Association, which funds scientific research on lung diseases.
•  Resources for the Future, which performs research related to improving air quality.
•  Natural Resources Defense Council, which works to strengthen and enforce the Clean Air Act

We interviewed representatives from three key industrial organizations groups that have been
actively involved in the PM program and will likely have constituent members that may be
economically impacted by PM attainment designations:

•  American Chemistry Council, which represents leading companies in the chemistry industry.
•  American Petroleum Institute, which is doing research on PM.
•  National Association of Manufacturers, the nation's largest industrial trade association.

Because academia is in the forefront of exploratory research in the area of air monitoring
instrumentation, we interviewed scientists from three universities involved in PM research:

•  University of Delaware
•  University of Maryland
•  University of Southern California.

We interviewed the Chairman of the Clean Air Science Advisory Committee subcommittee on
fine participate matter; representatives from the State and Territorial Air Pollution Program
Administrators/ Association of Local Air Pollution Control Officials (STAPPA/ALAPCO); and
the Lake Michigan Air Directors Consortium. We also interviewed personnel at the Department
of Energy and the National Oceanic and Atmospheric Administration.

As the implementers of the PM25 network, State and local air pollution control agencies are
critical to the success of the network.  We used a structured interview to discuss with these groups
the issues and challenges associated with the PM network. We interviewed officials from a total
of eight State agencies and one local agency, as well as their corresponding EPA region, to obtain
their input and views of the PM air monitoring network. We did not evaluate any of the State or
local ambient air monitoring programs' performance as part of these interviews. A table of the
State and local agencies and their corresponding regions follows.
                                          24

-------
               :New:Stefik-bepa^^
               Soikh Carolina Department of Health aridi Environmental Control
                                                                         ;HM3!:;
vR.4:*;
                                                                            -'65


We solicited input from OAQPS, Region 3, STAPPA/ALAPCO, and States regarding the State
and local agencies we should contact, but we independently made our selection after giving
careful consideration to the input received.  We considered the following items (listed in order of
importance) when selecting the nine organizations:

•  Number of people estimated to be living in counties expected to be in non-attainment for
   PMj.s. The six larger States selected had 53 of the 179  counties (30 percent) expected to be in
   non-attainment for PM2 5. Those 53 counties had an exposed population of 52 million, which
   represents 63 percent of the country's exposed population.

•  Number of PM monitors in each State: filter-based  mass-only, continuous, and
   speciation. We looked for States with the largest and more mature networks, taking into
   account those agencies monitoring the air quality with various types of instruments.  The six
   larger States that we selected operated 27 percent of the filter-based mass-only PM25 network,
   35 percent of the continuous mass monitors, and 27 percent of the speciation monitors.

•   Levels of PM2S concentrations detected from filter-based mass-only monitors.  We
   reviewed the 1999 levels of PM nationwide from the filter-based mass-only monitors. States
   with higher concentrations were considered.

•   Levels of PM2 5 emissions reported to EPA. We reviewed the States' emissions inventories,
    and considered those areas with higher emissions.  Six of the States we selected reported a
   total of 203,000 tons of emissions in 1999. This represents 30 percent of the 673,000 tons of
    emissions nationwide.

•   Coverage of various EPA Regions. For the 9 State and local agencies selected, 7 of the 10
    EPA regions  were represented.

•   Representation of various parts of the United States. In order to account for varying
    geographical and meteorological conditions, we covered six geographic areas of the country:
    Northeast, North Central, Northwest, Southeast, South Central, and Southwest.

•   PM Constituents. We identified the five main components of PM2 5 (sulfate, nitrate, organic
    carbon, elemental carbon, and crustal material).  The States selected support adequate
    representation of the various PM constituents.
                                           25

-------
*  Recommendations of State selections by OAQPS, Region 3, STAPPA/ALAPCO, and
   States.  An OAQPS official suggested we consider talking to one or two States with smaller
   networks and less resources. They explained that the "big" States usually have large budgets
   and more staff to handle new requirements imposed upon them. Our sample includes two
   such "smaller" States: Mississippi and South Carolina.

We selected California, Illinois, Ohio, Pennsylvania, New York, and Texas on the basis that they
were ranked highest in exposed population, size of monitor network, highest PM25 concentrations,
and PMZ5 emissions, hi addition, these States provided a diverse representation for EPA regional
coverage, geographic sections of the country, and variety of PM constituents.

Mississippi, a smaller State, provides representation in Region 4 and the Southeast. Additionally
Mississippi was ranked toward the middle in exposed population, network size, and emissions.
South Carolina was average in population, monitor network size, and emission levels, and was
fairly high in PM2 5 concentration level. Thus, it represents a smaller State with a mature network
and a fairly significant PM2 5 problem.

We initially were planning to contact Washington State because it had the fourth  largest
monitoring network, and it provided us with Northwest representation, where PM is mainly
composed of organic carbon. STAPPA/ALAPCO said that some local agencies are quite
advanced in their approaches to PM25 monitoring, and may offer "richness in perspective." The
Puget Sound Clean Air Agency was selected as an active Agency in Washington's air monitoring
program.

Detailed information on the States selected is highlighted in the following table.
                                          26

-------
Details on Attainment/Nonattainment Estimates, Monitor Types, and Particulate Matter Composition
(Note: these attainment and non attainment figures are based on preliminary January 2003 EPA estimates and should not
                                  be considered the Agency's final position.)

States
(ordered by Non
Attainment1 population)
ฃHฃtf0&tttA* -•• "-'
&ฃป**&• "" '
0BS0 { "",, <
fxmu&Ktm •
me am. • - ••'
d&y.: 	 -A 	 :
GEORGIA
NORTH CAROLINA
MARYLAND
TENNESSEE
ALABAMA
INDIANA
MICHIGAN
NEW JERSEY
KENTUCKY
mmmsiismmฎ
CONNECTICUT
MASSACHUSETTS
WEST VIRGINIA
DISTRICT OF COLUMBIA
ARKANSAS
*8&&B$tf>r*f ••
DELAWARE
MISSOURI
LOUISIANA
VIRGINIA
IDAHO
UTAH
IOWA
PUERTO RICO
ARIZONA
WYOMING
MINNESOTA
NEW MEXICO
WISCONSIN
MONTANA
NORTH DAKOTA
KANSAS
#s*mx<3K>tt - - --
FLORIDA
OREGON
COLORADO
OKLAHOMA
SOUTH DAKOTA
NEBRASKA
NEW HAMPSHIRE
MAINE
ALASKA
NEVADA
RHODE ISLAND
VERMONT
HAWAII

Population In
Non Attainment
21,959,496
7,573,425
6,06.794
5,476,027
5,335,171
4.695.S49
4,503,857
3,543,051
2,696,617
2,647,536
2.508,058
2,266,858
2,084,543
1,925,149
1.780,088
1,531,177
824,008
689,807
658,406
572,059
536,339
526,109
500,265
390,554
300,345
112,278
98,453
90354
60,296
52,393
51,335
35,804
31,671
29,979
19,680
18,837
5,737
3319
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Fopulatlouiu)
Attainment
10,991,3711
2,541,000
332,071
3,6S2,OK
10,602,169
9,901,776
92,115
1,280,006
1,309,805
435,623
193,653
1,194,203
• 5,288,118
4,573,812
262,542
1,211,320
1,998,838
5,341,018
305,717
0
805,886
863,930
283,335
2,928,474
2,580,061
3,354,344
851,417
1,877,321
1392,704
U44.91S
4,419,413
126,418
3,182,639
1,212,256
3,179,563
620,322
295,331
1,267,224
5,103,104
12,872,487
2,508,440
3,222,385
2,069,031
327,709
1,022,563
998,144
1,038,282
453,777
1,756,510
912,238
305,004
1,004,250
Pop ulatlon
not
Monitored
920,774
2,304,f6ป
4,134,27!
3,123,942
3,039,117
6.254.195
3,590,481
3,226,256
1,290,064
2,606,124
1,745.389
2,619,424
2,565,783
1,915,389
1,999,139
1,262,015
582,719
318.272
844,221
-
1,331,175
1.452.619
.
2,276,183
1,588,570
3,611,893
344,083
265,494
1,473.324
2,511302
659,884
331,560
1,705,169
576,811
2,164,432
263,036
341,132
1,417,875
791,617
3,109.891
912,959
1,078,876
1,381,623
427,135
688,700
237,642
236,641
173,155
241.747
136,081
303.823
207,287
Total State
Population
33,S71,64S
12,419,293
11,353,140
12,2X1,054
10,976,457
20.S5I.S20
8,186,453
8,049,313
5,296,486
5,689,283
4,447,100
6,080.485
9.938.444
8,414,350
4,041,769
4,012,012
3,405,565
6,349,097
1,808344
572,059
1,673,400
2,S44,6SS
783,600
5,595,211
4,468,976
7,078,515
1,293,953
2,233,169
2,926,324
3,808,610
5,130,632
493,782
4,919,479
1,819,046
5,363,675
902,195
642,200
2,688,418
5,S94,I21
15,982378
3,421399
4301,261
3.450,654
754,844
1,711,263
1,235,786
1,274,923
626,932
1,998,257
1,048,319
608,827
1,211,537
Percentage
In Non
Attainment
64.i3%
60.9SH
60.66%
44.54%
2*11%
21.52%
55.02%
44.02%
50.91%
46.54%
56.40%
37.28%
20.97%
22.88%
44.04%
3SJ4%
2420%
10.86%
36.41%
100.00%
20.06%
IS.56%
63.84%
6.98%
6.72%
1.59%
7.61%
4.05%
2.06%
138%
1.00%
7.25%
0.64%
1.65%
0.37%
2.09%
0.89%
0.12%
a.oo%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
Monitor TTPซ
1
4
3
9
12
27
40

7

1
1
3
7
15

6
3
4



4
2
1
5

13

31
2


2
11
4

4
10
29
7

1

1

1
3



2

!
127
52
55
57
51
S9
44
51
23
36
29
50
36
25
31
30
17
31
19
6
49
17
11
28
28
20
30
29
18
10
21
7
58
21
36
19
14
17
S3
33
41
31
32
21
20
20
17
13
10
11
8
7
!
9
1
li
6
9


8
1
4
6
2
6
5
10
4




3
2
2
5

3
1
3
3
1
2

4

6
2
3
1
/
3
1
5
2
1
1


1
1

1

1
140
56
77
75
94
12S
44
66
24
41
36
55
49
45
41
46
20
35
19
6
52
23
15
34
33
23
44
32
52
13
23
7
64
32
46
21
21
28
S3
43
42
37
34
23
21
21
20
14
11
11
11
7

Primary Competition of PM2.5
Paniculate Matter
Organic Carton, Nitrate, Sulfae
Organic Carbon, Sulfate, Cnatai Materttti
Sulfae, Organic Carton
Sulfae, Organic Carton
Sulfae, Organic Carton
Sulfae. Organic Carton, Crustal Material
Sulfate. Organic Carbon
Sulfate, Organic Carbon
Sulfate, Organic Carbon
Sullate, Organic Carbon
Sulfate, Organic Carbon
Organic Carbon, Sulfate, Crystal Material
Organic Carbon, Sulfate, Crustal Material
Sulfate, Organic Carbon
Organic Carbon, Sulfate, Crustal Material
Sulfae, Organic Carbon
Sulfate, Organic Carbon
Sulfete, Organic Carbon
Sulfate, Organic Carbon
Sulfate, Organic Carbon
Sulfate, Organic Carbon
Sulfae, Organic Careen
Suljate, Organic Carbon
Organic Carbon, Sulfate, Crustal Material
Organic Carbon, Crustal Material
Sulfate, Organic Carbon
Organic Carbon, Sulfate
Organic Carbon, Sulfate, Crustal Material
Organic Carbon, Sulfate, Crustal Material
HA
Sulfate, Organic Carbon
Organic Carbon, Sulfate, Crustal Material
Organic Carbon. Sulfate, Crustal Material
Organic Carbon, Sulfate
Organic Carbon, Sulfate, Crustal Material
Organic Carbon, Sulfate
Organic Carbon, Sulfate, Crustal Material
Organic Carbon, Crustal Material
Organic Carton, Sulfate
Sulfate, Organic Carbon
Organic Carbon, Sulfate
Organic Carbon, Crustal Material
Organic Carbon, Crustal Material
Sulfate, Organic Carbon
Organic Carbon, Sulfate, Crustal Material
Sulfate, Organic Carbon
Sulfate, Organic Carbon
NA
Organic Carbon, Nitrate, Sulfate
Sulfate, Organic Carbon
Sulfate, Organic Carbon
NA
    1 Attainment and Non Attainment figures are based on recent EPA estimates.

    * States reviewed are indicated in shaded boxes and are also bolded
                                                  27

-------
28

-------
                                                                  Appendix B

         Consolidated EPA Response to Draft Report

                              September 26, 2003

MEMORANDUM

SUBJECT:   Response to the Draft Evaluation Report: Decline in EPA Particulate
            .Matter Methods Development Activities May Hamper Timely Achievement
             of Program Goals, Assignment No. 2002-000355

FROM:      Jeffrey R. Holmstead  /s/
             Assistant Administrator for Air and Radiation  (6101 A)

TO:         J. Rick Beusse
             Director for Program Evaluation, Air Quality Issues
             Office of Inspector General

      Thank you for providing us with the opportunity to respond to the draft report from
the Office of Inspector General (OIG). We have discussed the draft report with the Office
of Research and Development (ORD) and share many of the same concerns.

      The Office of Air and Radiation (OAR) and the ORD agree with the OIG authors'
statement that air monitoring methods development is key to supporting the Agency's
regulatory agenda and science needs.  Both offices support the OIG recommendation to
work jointly to establish a methods development program plan. In developing a plan, EPA
will need to balance near-term method needs associated with tiie 1997 National Ambient
Air Quality Standard (NAAQS) for fine particles (PM2 5) and long-term methods research
to support health and exposure studies and future NAAQS.
      There appears to be a general misunderstanding throughout the report that
three years of PM2.5 data using the Federal Reference Method (FRM) are unavailable
and that this lack of data has delayed the designation of PM25 nonattainment areas by
two years. This is simply not true and we cannot concur with that conclusion. We
believe that the delays incurred in designating nonattainment areas and beginning
implementation of this important standard are due to a number of factors, including the
time it took to settle the litigation over the standard, but not to the time necessary to deploy
the network.
See Appendix C
Notel
          5 is a new pollutant for which a new FRM had to be established. The schedule
for deploying the new network was extremely ambitious and was, in fact, directed by the
President. The Congress subsequently codified the schedule into law.  Moreover, as we
believed that the health consequences of this pollutant were significant, we sought and
obtained full  funding from the Congress for state deployment of the network.
       States deployed the PM25 FRM network over two years, 1998 and 1999. Three
years of data are now available and on a schedule that is consistent with Congress'
                                      29
See Appendix C
Notel

-------
expectation and the 40 CFR 58 monitoring regulation published in 1997. The OAR has
already published draft PM2 5 design values, and we expect to have the states*
recommendations on nonattainment designations by mid-February 2004. As a result, we
believe the underlying premise of the report that "EPA has not supported PM25 methods
development activities to the extent necessary to achieve the short- and long-range goals of
the PM2 5 program" is not accurate.
       The draft OIG report does not describe paniculate matter methods
development in the context of the outcome-oriented design for EPA's particulate
matter research program in the Particulate Matter Research Program Multi-year Plan.
This design - established with participation by ORD clients - describes the research
needed to achieve short-term outcomes, respond to recommendations by the National
Research Council Committee on Research Priorities for Airborne Particulate Matter, and
help EPA achieve its strategic goals and objectives. As the Scope and Methodology
sections of the draft OIG report do not mention the Particulate Matter Research Program
Multi-year Plan, we believe that this important design context was not considered when the
OIG prepared its observations, findings, and results-in-brief.
                                                                              See Appendix C
                                                                              Note 2
       The draft report asserts that EPA provided limited monitoring options for the
FRM supporting the 1997 NAAQS. We believe the development of a filter-based
FRM was consistent with our scientific understanding of PM, the technology
available at the time, and the stringent requirements for method performance that
must be met to achieve the required public health protection.  The FRM promulgated as
part of the 1997 NAAQS was reviewed by the Clean Air Scientific Advisory Committee
(CASAC) and underwent public comment along with the NAAQS. EPA considered
comments from both of these reviews in the final decision to promulgate the FRM for
PM2.S. We believe that our work should be evaluated in light of the scientific understanding
that existed at that time, not with current scientific understanding. We have attached a
scientific and historical perspective to place the work in the proper context (Attachment A).
                                                                              See Appendix C
                                                                              Note 3
       The OAR and ORD agree there were a number of technical obstacles
encountered with the initial PM25 network deployment that lasted into 1999. These
obstacles stemmed from many sources, including methods development, equipment
vendor support, states' ability to hire and retain staff, and the compressed time
schedule EPA dictated in deploying the network. However, we believe that the issues and
problems encountered were not atypical for the deployment of a large network of new
technology monitors across the entire country. Both the ORD and OAR recognized that
problems and issues in deploying such a large network were likely and worked extremely
hard to address this issues that arose during implementation of the new monitoring
network. Nonetheless, the OAR and ORD did learn many lessons that will be useful in
developing and managing our future monitoring programs.
                                                                              See Appendix C
                                                                              Note 4
       With consideration given to these concerns, we do support the spirit of the
recommendations for forward looking methods development research.  We concur
with all but one of the recommendations contained within the draft report. We do not
concur with the use the Federal Advisory Committee Act (FACA) to establish a
government-industry partnership (Recommendation 3-1). We believe that the CASAC -
established under the FACA - provides a better alternative. Although we concur with the
                                       30
                                                                              See Appendix C
                                                                              NoteS

-------
recommendations to develop a PM methods development strategy (Recommendation 2-1)
and to maximize opportunities to enhance Homeland Security efforts (Recommendation 3-
3), we have comments regarding their implementation.  The comments are included in the
attachment to this memorandum

      I am including more detailed information on specific recommendations,
clarifications, and suggested improvements (Attachment B). Please contact Dr. Richard
Scheffe (919-541-4650) of OAR or Tim Watkins (919-541-5114) of ORD if you have any
additional questions.

Attachment

cc:    Lee Byrd, OAQPS/EMAD
      Tom Coda, OAQPS/PRRMS
      Gary Foley, NERL
      Paul Oilman, ORD
      Tim Hanley, OAQPS/EMAD
      Mike Jones, OAQPS/EMAD
      Joe Paisie, OAQPS/AQSSD
      Rich Scheffe, OAQPS/EMAD
      Peter Tsirigotis, OAQPS/EMAD
      Tim Watkins, NERL
      LydiaWegman, OAQPS/AQSSD
                                    31

-------
                                                          Attachment A
                       Scientific and Historical Perspective

       The 1997 PM NAAQS included a new standard for fine particles, or PM25,
based upon scientific evidence that demonstrated an association between ambient
PM15 levels and observed health effects. However, questions and concerns remained
regarding the true public health significance and credibility of the effects of ambient
PM. These concerns led to an increase the in the budget for the EPA PM research program
and the formation of the NRC committee on Research Priorities for Airborne Particulate
Matter.
See Appendix C
Note 2
       In their 1998 report, the NRC developed a broad source-to-exposure-to-
response framework for evaluating PM issues. The original research portfolio
suggested by the NRC included research to provide insight on how people are
exposed to PM and its components and the mechanisms by which these exposures
lead to adverse health outcomes. To accomplish this, EPA would first need to identify the
particular components of PM that were leading to adverse outcomes. EPA would then
conduct studies to understand actual human exposures to these components and to relate
these exposures to health effects observed in the real world. Subsequently, the NRC
identified the need to develop methods for measuring PM components to support these
exposure and health studies.
See Appendix C
Note 2
       Since the issuance of the 1998 NRC report, we have learned much about PM,
but we have not identified a particular component that is solely responsible for
observed adverse health outcomes. We now believe that most PM components are
likely to have a toxic effect and we also are focusing on particular sources of PM as
potential risk drivers.  As a result, our PM methods research needs to provide the
measurement techniques necessary to support investigations relating to components and
sources of PM.  If future research studies justify a change in the format of the PM NAAQS,
the methods used in those studies could be developed as a revised FRM.  The challenge
faced by the ORD is to find an appropriate balance between supporting nearer-term needs
associated with the 1997 NAAQS and longer-term research to provide the scientific basis
for future NAAQS and, given its role in supporting both of these needs, our PM methods
research program lies at the center of this challenge.
See Appendix C
Note 2
                                       32

-------
                                                                                                   I
                                                                  Attachment B


Detailed Comments on the Office of Inspector General's Draft Evaluation Report:
Decline in EPA Participate Matter Methods Development Activities May Hamper
Timely Achievement of Program Goals, Assignment No. 2002-000355

       Comments are provided in three categories:  (1) responses to recommendations
contained within the report, (2) clarification on a number of misunderstandings that have
arisen from the very complex nature of this area, and (3) specific suggestions to improve
the report.

Responses to Recommendations                                                                      I

       In general, OAR and ORD agree with the recommendations contained within the
draft report, with the possible exception of using the Federal Advisory Committee Act
(FACA) to establish a government-industry partnership. Below are our comments on the
recommendations:
                                                                            Note 6
Recommendation 2-1: OAR and ORD support the recommendation to work
jointly to develop a comprehensive methods development strategy. However,     ee  ppen
we feel that a strategy that focuses only on PM2 5 is too narrow. The EPA is
preparing for a potential PM coarse FRM and also needs to address longer
term methods development, such as identifying specific components of PM for
evaluation with health associations. The health associations are part of the methods
development work identified in the guidance set forth by theNRC in their 1998
report. The NRC report recognized that individuals develop adverse health
responses to PM in the air that they breathe (exposure). It also shows that
understanding exposure provides a critical link between regulatory monitoring and
health outcomes. PM is a complex mixture of particles of different sizes and
compositions that will effect both its toxicity and relationships between sources,
ambient concentrations, and actual human exposures.  As EPA moves forward with
future PM standards development activities, we need to be able to measure PM
mass, species, or components in the ambient air, link these measurements to sources
and actual exposure, and relate the  exposure to health effects.  As a result, a
comprehensive PM methods development strategy must address all measurements
needed to understand the entire source-to-exposure-to-effect continuum, including
measurement methods for various PM size fractions, PM composition, PM
precursors, and tracers for PM sources. Furthermore, as part of an overall strategy,
activities to develop FRMs that are used to determine compliance with the future
standards should reflect the relationship between ambient measurements and
exposures.

Recommendation 2-2: OAR and ORD support the use of Data Quality Objectives
(DQO) to ensure important decisions with data are made with an acceptable amount
of uncertainty. The OAR and ORD are actively engaged in a DQO process to
define acceptable criteria for candidate equivalent PM2 s monitoring methods.

                                33

-------
      Recommendation 2-3:  OAR and ORD agree with the recommendation for ORD to
      maintain personnel in the area of PM methods development as part of a strategic
      hiring plan.
•     Recommendation 3-1:  OAR and ORD believe that, for formal review of PM     geeA  endixC
      measurement method issues, using the CASAC provides a better alternative to    Note 5
      establishing a new government-industry partnership. CASAC has a functional   '	
      subcommittee working on particle monitoring and is expected to continue
      meeting as EPA considers possible changes to the PM networks.  The current
      process of involving the CASAC provides an unbiased opinion for methods
      development issues. The EPA can continue to informally communicate with other
      external stakeholders to ensure that their needs and issues are addressed.

      Recommendation 3-2:  OAR and ORD have learned many valuable lessons in
      developing and implementing the FRM for PM2 5, which are being applied in
      developing the FRM for PM,0.2 5. We have tried to improve our approach to
      methods development while participating in the PM Supersites program (see below)

•     and in evaluating methods for the speciation networks.  Insights from this draft
      report will allow us to develop a better PM methods development program that is an
      integral part of our other PM program.

Clarifications

*     There is a general misunderstanding throughout the report that three years of
       PM2.5 FRM data are unavailable, or at least unavailable for designation
See Appendix C
      purposes.  This is simply untrue.  The PM2 5 FRM network was deployed over     ote l
      two years, 1998 and 1999. Three years of data from the initially deployed
      monitors were available on July 1, 2002 (for the years 1999, 2000, 2001) and from
      all sites on July  1, 2003 (for the years 2000, 2001, 2002).  This July 1 schedule is
      consistent with the 40 CFR 58 monitoring regulation published in 1997.
      The goal of EPA* s air program is to protect public health and the environment
      and, for criteria pollutants, NAAQS are set based on a comprehensive
      scientific understanding of the exposures and adverse health and                 ฐ
      environmental effects associated with elevated levels measured at ambient
      monitoring sites. These NAAQS are implemented by the states and tribes using
      tools and data provided in large part by EPA.  Methods for measuring PM in air are
      used to both set and implement the standards.  This is an important distinction and
      one that is blurred throughout the report.
 See Appendix C
      The FRM is the primary tool for implementing the NAAQS. FRMs are the
      basis for the standard and are used to determine regulatory compliance. Much
      of the discussion in the report, and the title itself, is directed at the
      development of the FRM for measuring PM25. This method was developed
      from 1994 to 1998 to determine compliance with the PM25 NAAQS set in 1997.
      The development of a filter-based FRM was consistent with our scientific
      understanding of PM, the available technology, and the stringent requirements for
                                      34
See Appendix C
Note 3

-------
method performance that must be met to achieve the required public health
protection. In our judgement, it was not feasible to develop a continuous
measurement method that would meet all of the performance requirements for all
areas of the country in the time frame that was needed for implementation
monitoring. Finally, the epidemiological studies upon which the 1997 NAAQS was
founded used filter-based methods to measure PM mass. To use some other
surrogate of mass would have been inconsistent with the state of knowledge at that
time.  So, although mere are now concerns voiced over the resources required to run
a filter-based method, it is important to point out that when we developed the FRM
for PM2 5, we placed a premium on ensuring public health protection in a timely and
scientifically rigorous manner.
The EPA has not proposed a new PM NAAQS since the existing NAAQS was
issued in 1997. The report authors make several references to a proposed
"coarse" particle standard within the draft report. The EPA is currently
reviewing the PM standards and is considering the need for a coarse particle
standard; however, it is premature to describe this as a proposed standard. The
EPA's current schedule should provide for a proposal in late 2004 and a final
standard in late 2005.  The ORD, OAR, and the state and local agencies are working
together to identify and field test a variety of candidate coarse particle methods,
including both filter-based and continuous devices.
See Appendix C
Noteg
The ORD conducts research that provides the scientific underpinning of our
regulations with an overall goal of protecting public health.  In the PM area,
methods development is but one research component that provides inputs to
both developing and implementing the standards.  As an important example,
the 1997 PM2S standard was developed based on epidemiological studies that found
relatively consistent but poorly understood associations between ambient PM
concentrations and various adverse health effects. These studies raised two
important questions: (1) could PM at ambient levels actually be responsible for the
adverse health outcomes and (2) could ambient monitors be used to represent
exposure to PM mass. These were critical questions associated with the NAAQS.
In 1998, these questions were given a higher priority than methods development in
areas not associated with the FRM as mandated by Congress, the President, and the
NRC. The ORD is striving for a balanced program that develops the methods, data,
and the models for both implementing the current standard, and developing future
standards.  In the next three years, ORD has placed an  emphasis on developing the
tools that can be used by states in implementing the NAAQS. In the future, we will
emphasize research that supports setting future NAAQS and balance our activities
in methods development, ambient and exposure measurement studies, source
apportionment, health studies, and atmospheric and human exposure modeling to
address the agency's most critical data gaps.
See Appendix C
Note 9
The EPA has taken the NRC recommendations very seriously and worked very
hard to develop a PM research program that was responsive to both the spirit
and intent of all of the recommendations.  The third chapter of the initial NRC
report recommended research to develop new personal and ambient
monitoring methods for measuring particle species and sizes. The methods research
                                35
See Appendix C
Note 2

-------
      identified by the NRC, which supports health and exposure research studies, was
      and continues to be, a focal point of the ORD PM methods research program.
      However, it was not until the third NRC report that methods research was explicitly
      acknowledged as a technical support activity to the overall PM research program.
      It is important to note the role of the PM Supersites program in the EPA PM
      methods program. As ORD responded to the recommendations in the NRC
      report, the EPA PM Supersites Program was being planned and implemented
      by OAR and ORD. An important objective of the Supersites program is to
      develop and test measurement methodologies.  So, while ORD was redirecting
      some internal resources to respond to the NRC, overall EPA investment in methods
      work was being augmented by the PM Supersites program. Significant strides in
      methods development have occurred through the PM Supersites Program, which
      will be winding down by the end of 2004, including testing and inter-comparisons
      of continuous and speciation methods. As more PM Supersites data become
      available and are analyzed, EPA will be in a position to use the results to improve
      its activities relating to PM methods development.
 See Appendix C
 Note 10
Suggestions:
      Page 1, Background, paragraph 1, line Ir5. The authors list a number of
      sources but miss many important sources (e.g., diesel for elemental carbon).
      We suggest they limit this discussion to the composition and major
      components, then state these components are from both natural and
      anthropogenic sources, including combustion of fossil fuels and modem material,
      secondary pollutants (e.g., sulfates and nitrates) from combustion sources, and, for
      example, organic carbon, which is a relatively major constituent everywhere,
      virtually all the time, and is emitted directly into the air (primary) or formed in the
      air from gas phase precursor species (secondary), both from a variety of combustion
      and other sources.
 See Appendix C
 Note 11
      Page 2,.paragraph 1, line 4. Rather than say "Some particles may travel.." and
      "while others remain within ...," we suggest being more specific about
      particle size and distance.  Fine particles about 0.5-2 urn in size may travel
      thousands of miles... while larger particles, 10 um in size, may only travel on
      the order of up to 100 miles, and very large particles 100 um in size may only travel
      a few hundred meters, distances for all sizes are strongly dependent on meteorology,
      such as wind speed.
See Appendix C
Note 12
      Page 4, top.  Measurement against standards is an important missing
      component that should be mentioned. For PM, mis is an extremely difficult
      task, but one that ORD in conjunction with NIST and others are beginning to
      make progress.

      Page 4, paragraph 4, line 1. Only Teflon filters are used within the PM2.5
      FRM network.  The nylon and quartz filters are used for chemical speciation
      analyses.
See Appendix C
Note 13
See Appendix C
Note 14
                                       36

-------
Page 4, paragraph 4, line 7, measure. A more accurate word would be
'estimate.'  The mass measured on the filter is not a measure of the fine
particles that a person would breathe, it is, however, 'an estimate.'

Page 9, paragraph 3.  Several items are identified as causes for data gaps from
the PM2 5 network. Oil freezing in the WINS impactors, water condensation,
and the mentioned software malfunctions did not result in PM2 5 concentration
data loss. While these are issues of concern, they did not lead to data loss.
The WINS impactor oil crystallization issue was researched by the ORD and found
to be a non-issue for data quality. The software malfunction was due to a
manufacturer defect that did not prevent mass data from being collected.
See Appendix C
Note 15
See Appendix C
Note 16
Page 10,3rd bullet  Note that EPA funded 52 filter weighing laboratories in
FY 1998, one for each State, Puerto Rico, and D.C. hi 1999, an additional 14
more filter weighing laboratories were funded for local air monitoring
agencies. Additionally, there are several commercial laboratories available for
this function.  Although EPA does recognize the difficulty of establishing
laboratories for filter weighing, EPA does not believe that these difficulties were
due to a lack of timely funding.

Page 16, Tables 3.1 and 3.2; Pages 18 and 19.  hi several locations, the draft
report states that the chemical speciation monitoring data will be used to
identify a "source." To prevent the misinterpretation of this  statement, EPA
advises the authors to state that the data will identify "source categories," for
example, mobile sources, industrial sources, etc., that cause elevated particulate
matter levels.
See Appendix C
Note 17
 See Appendix C
 Note 18
Page 16 and 17, Tables 3.2 and 3.3. The draft report discusses the chemical
speciation network size and cost in its Chapter 3. A few corrections are
needed here, hi experimenting with the continuous speciation analyzers, EPA
has realized that the annual operation and maintenance costs for these devices
can be as expensive as for filter based methods.  As a result, EPA suggests deleting
the checkmark in Table 3.2 that suggests that these samplers have "minimal
operational/life cycle costs."  Additionally, in Table 3.3, the initial capital cost for a
set of continuous instruments (nitrate, sulfate, and carbon) is $85,000 rather than
$20,000.
 See Appendix C
 Note 19
Page 17, last paragraph.  Comment regarding continuous speciation methods is
among the most promising near-term technologies.  The Supersites Program
has been examining many continuous methods for mass and species.
Although three years ago we believed several of these were very promising,
results from the Supersites Program often suggest that we do not know what these
monitors are actually measuring and much more research is needed in this area as
well as a better understanding of their operational effectiveness.
See Appendix C
Note 20
Page 18, paragraph 1, line 11. There are currently more than 350 rural and
urban chemical speciation sites providing data to EPA. The report indicates
that only 142 are operating.
                                  37
See Appendix C
Note 21

-------
Page 18, paragraph 2, line 2, "However, the composition of a particle
differs..." Change'a particle' to 'paniculate matter.' The same for several
other locations in this paragraph -- change particle to particulate matter.

Page 20. The OAR and ORD believe there are areas of the ambient
monitoring network that can be leveraged for supporting of EPA's homeland
security goals while maintaining their primary function to support NAAQS
implementation.  However, careful attention needs to be paid in not over-
selling the leveraging that can be done in these areas.  The EPA has been working
with the Department of Homeland Security (DHS) and other federal agencies to
establish the existing biological sensing network (BioWatch). In many cases, the
BioWatch network uses or supplements existing ambient monitoring network
infrastructure, such as sample platform locations, access to electrical power, and
field staff to operate the network. The EPA is also in regular contact with the
homeland security researchers involved with the ongoing methods development
work. The DHS and the Los Alamos and Lawrence Livermore National
Laboratories are investing significant resources in developing methods for
identifying biological and chemical agents. Developing methods that detect
biological agents requires live biological agent testing, an understanding of terrorist
threat scenarios, and other factors that cannot be pursued by EPA acting alone.  The
components of the ambient monitoring network used for NAAQS should not be
used for homeland security, except in the most high priority situation. As an
example, PM filters could not be used for both air pollution program and homeland
security purposes because the biological agent analyses must be performed
immediately and will destroy the filter.  As well, the ambient network has been
established for neighborhood to regional scale analyses, while tracking a plume for
a toxic release at local scales would require a different network of monitors.  There
also would be questions about the usefulness of the ambient filters, based on the
amount of material collected and subsequent analyses for biological and chemical
agents.  Except for continuous monitors, the ambient network mostly operates on a
l-in-3 or l-in-6 day schedule, leaving large temporal gaps for homeland security
information. Finally, most ambient continuous near real-time methods are only for
the major components of mass (sulfate, nitrate, ammonium, OC, and EC).
Continuous real-time methods do not yet exist nor are they in place in the network
for potentially toxic biologically active or chemical toxins. The authors would have
to drift significantly from the evaluation's scope in order to make the report better
reflect the complexity of this cooperation between EPA and various Federal
agencies.
See Appendix C
Note 22
See Appendix C
Note 23
                                 38

-------
                                                                 Appendix C

       O/6 Evaluation of EPA Response to Draft Report
Note 1 -      We agree that the Agency's description of timeframes is consistent with 40
             CFR 58. However, the 1998 Transportation Equity Act for the 21st Century
             allowed the States 3 years, until September 2001, to gather the data.
             However, a complete set of data was not available until July 2003. We have
             added a footnote for clarification. Also, we have modified the wording in
             our report to recognize that the decline in methods development activities
             was one of several contributing factors that jeopardizes EPA's ability to
             fully achieve the short- and long-range goals of the PM25 program in a
             timely manner.  Nonetheless, our work indicated that the decline in methods
             development was a key contributing factor, and our report continues to
             reflect that position.

Note 2 -      We agree and have modified the report to clarify the distinction between
             National Research Council's recommendations in its Volume 1 report and
             the National Research Council's Volume 3 report. We did review the PM
             Research Program Multi-year Plan,  as indicated on page 17 of the draft
             report.  It should be noted that the documents in the Scope and Methodology
             section  of our report were not meant to be all inclusive. Instead, we list
             some of the key documents used to develop the facts and issues presented in
             our report.  However, we have modified our report to present a more
             balanced view of the longer-term research needs of the overall PM research
             program, as outlined in the National Research Council's report.

Note 3 -      We agree that filter-based mass-only monitors were the viable option when
             ambient air monitoring implementation began in 1998, and have added
             further emphasis to highlight this point. However, having a vigorous
             methods development program will potentially result in more monitor
             options  becoming available as EPA moves forward with its PM monitoring
             network.

Note 4 -      We agree that the deployment of the PM25 monitoring network was
             extremely challenging and that problems encountered by States were typical
             of a large network.  We have modified the report to reflect this.

Note 5 -      We agree that our recommendation should not restrict EPA to using a
             particular process, such as the Federal Advisory Committee Act, to
             accomplish the goals and objectives of the PM program.  We agree that the
             Clean Air Scientific Advisory Committee process is a suitable alternative,
             and have modified our recommendation to reflect this.
                                      39

-------
Note 6 -      We agree tiiat a methods development strategy that only focuses on PM2 5
             may be too narrow. We have modified our recommendation for EPA to
             develop a comprehensive PM strategy.

Note 7 -      We agree that methods for measuring PM in air are used to both set and
             implement the standards, and have modified our report to reflect this.

Note 8 -      We have modified our report to reflect EPA's current estimate for the
       .  •    schedule of proposing and finalizing a coarse standard.

Note 9 -      We agree with the importance for ORD to strive for a balanced program that
             develops the methods, data, and models for both implementing the current
             standard and developing future standards. However, as part of achieving
             this balanced program, both OAQPS and ORD officials have acknowledged
             that PM methods development warrants increased attention by EPA.

Note 10 -    We agree that the PM Supersites program is important in the overall EPA
             PM methods program, and have modified our report accordingly.

Note 11 -    We agree that the Agency's definition of the sources of PM is more
             complete.  However, we believe our definition is more appropriate for the
             non-scientific readers of this report.

Note 12 -    We have modified our report regarding PM travel as it relates to the size of
             the particle.

Note 13 -    We have modified the report to reflect that monitoring is also a form of
             measurement against the standards.

Note 14 -    We have modified the report to reflect that only teflon filters are used within
             the PM2 5 FRM network.

Note 15 -    We have modified the report to reflect that the PM2 5 data is not a
             measurement but instead is an estimate of the fine particulate matter that a
             person would breathe.

Note 16 -    We agree that data gaps were eventually filled; however, our work found
             that the monitor malfunctions did result in a loss of data.

Note 17 -    We have modified our report to reflect that while there were difficulties with
             establishing laboratories for filter weighing, EPA does not believe these
             issues were due to a lack of funding.

Note 18 -    We have added a footnote to clarify that, when referring to emissions
             sources, unless otherwise stated, we are also referring to the larger category
             of sources, such as mobile, industrial, etc.
                                       40

-------
Note 19 -     We changed the original EPA estimate of $20,000 (provided to us by
             Agency staff during our field work) to reflect the Agency' s most recent
             estimate of $85,000 for a set of continuous instruments (nitrate, sulfate,
             carbon).

Note 20 -     Based on our discussions with EPA and State officials, we still maintain that
             continuous mass and continuous speciation monitors are among the most
             promising near-term technologies.  However we have added that EPA
             officials also believe that much more research is needed.

Note 21 -     We agree that there are 350 rural and urban chemical speciation sites, which
             include IMPROVE and Speciation Trends Network monitors. However, our
             number of 142 speciation monitors is based on information provided by
             OAQPS in January 2003.

Note 22 -     Where appropriate, we have changed the terminology from "a  particle" to
             "paniculate matter."

Note 23 -     We agree with the Agency that the ambient monitoring network can be
             leveraged to support EPA's homeland security goals while maintaining its
             primary function to support National Ambient Air Quality Standard
             implementation. We have clarified that EPA must work closely with other
             Federal agencies to accomplish this effort.
                                       41

-------
42

-------
i
                                                                              Appendix D

                                ORD Response to Draft Report

              Note: We did not include ORD's attachments to its response because they
              were the same as the attachments in EPA's consolidated response.
                            UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                                         WASHINGTON, D.C. 20460

                                           September 26, 2003
                                                                               OFFICE OF
                                                                          RESEARCH AND DEVELOPMENT
MEMORANDUM

SUBJECT:   ORD Response to the Inspector General's Draft Report "Decline in EPA
             Paniculate Matter Methods Development May Hamper Timely Achievement
             of Program Goals"

FROM:      PaulGilman  /s/Paul Gilman
             Assistant Administrator (8101R)

TO:         J. Rick Beusse
             Director for Program Evaluation, Air Quality Issues
             Office of Inspector General

Introduction

      The purpose of this memorandum is to provide you with the Office of Research and
Development's (ORD) response to the Office of Inspector General's (OIG) draft report
"Decline in EPA Particulate Matter Methods Development May Hamper Timely
Achievement of Program Goals." We do not concur with the findings of this report, and
the underlying premise that a decline in EPA's methods development activities has led to
delays in achieving participate matter (PM) program goals. This memorandum outlines our
concerns with the findings, presents our response to the report recommendations, and
provides a scientific and historical perspective of the EPA PM research program that must
be appreciated to place this draft OIG report in the proper context.  We have also included
an attachment that provides more detailed comments on the draft report.

Response to Findings and Recommendations

      We do not concur with the overall findings in this draft report.  Specifically, we do
not agree with the statement, "EPA has not supported PM2 5 methods development activities
                                     43

-------
to the extent necessaiy to achieve the short- and long-range goals of the PM2 5 program."
The draft OIG report focuses on the short-term issue of deployment of the PM25 monitoring
network in response to the 1997 National Ambient Air Quality Standard (NAAQS), but
does not present a balanced view of the longer-term research needs of the overall PM
research program as outlined by the National Research Council's (NRC) Committee on
Research Priorities for Airborne Particulate Matter. In addition, the report indicates that
reduced emphasis on PM methods activities has led to delays in obtaining data to make
compliance determinations and, therefore, led to "millions of individuals being exposed to
excess levels of PM." This conclusion is false.  The PM25 Federal Reference Method
(FRM) was developed in time for promulgation of the 1997 PM NAAQS.  In addition,
EPA has met the schedule for obtaining data to make compliance determinations using the
FRM that was contained in the monitoring regulation that supported the 1997 NAAQS. As
a result, concluding that a decline in EPA PM methods development activities led to excess
exposures across the country is not appropriate.

       The draft OIG report does not describe PM methods development in the context of
the outcome-oriented design for EPA's PM research program documented in the
Particulate Matter Research Program Multi-year Plan. This design-established with
participation by our clients-describes the research needed to achieve short-term  outcomes,
responds to recommendations by the National Research Council Committee on Research
Priorities for Airborne Particulate Matter, and helps EPA achieve its strategic goals and
objectives.  The Scope and Methodology sections of the draft OIG report do not  mention
the Particulate Matter Research Program Multi-year Plan, indicating that this important
design context was not considered when the OIG prepared its observations, findings, and
results-in-brief.

       This draft report also indicates that a decline in EPA methods development
activities led to difficulties encountered by states in deploying the PM2 5 ambient monitoring
network caused by limited monitoring options, insufficient field testing, and insufficient
technical assistance and guidance.  We do not agree.  We believe that the issues  and
problems encountered by the states were not atypical  for the deployment of a large network
of new monitors across the entire country.  Both ORD and OAR recognized that problems
and issues in deploying such a large network were likely and worked extremely hard to
address these issues as they arose during implementation of the new monitoring  network.
Senior managers from both organizations met monthly to address issues and provide
necessary and timely resources to remedy problems brought to their attention. Overall, the
deployment of the PM25 monitoring network was extremely challenging and EPA could not
predict issues that ultimately arose.  However, we were prepared to address the  issues that
were raised, and were successful in resolving them.

       Another assertion identified in the draft report is that EPA provided limited
monitoring options for the FRM supporting the 1997 NAAQS.  We believe the
development of a filter-based FRM was consistent with our scientific understanding of PM,
the available technology, and the stringent requirements for method performance that must
be met to achieve the required public health protection. It is also important to note that the
FRM, promulgated as part of the 1997 NAAQS, was reviewed by the Clean Air  Scientific
Advisory Committee (C ASAC) and underwent public comment along with the NAAQS.

                                       44

-------
EPA considered comments from both of these reviews in its final decision to promulgate
the FRM for PM15.

       Additionally, the recommendations in this draft report do not follow from the
premise and findings discussed above. We support the spirit of the recommendations for
forward looking methods development research, and concur with all but one of the
recommendations. The one exception is the recommendation to use the Federal Advisory
Committee Act (FACA) to establish a government-industry partnership (Recommendation
3-1). In addition, while we concur with the recommendations to develop a PM methods
development strategy (Recommendation 2-1) and to maximize opportunities to enhance
Homeland Security efforts (Recommendation 3-3), we have comments regarding their
implementation. Our detailed comments on the recommendations contained in this draft
report are included in the attachment to this memorandum.

Scientific and Historical Perspective

       The 1997 PM NAAQS included a new standard for fine particles, or PM2 5, based
upon scientific evidence that demonstrated an association between ambient PM2 s levels and
observed health effects. However, questions and concerns remained regarding the true
public health significance and credibility of the effects of ambient PM. These concerns led
to an increase in the budget for the EPA PM research program and the formation of the
NRC committee on Research Priorities for Airborne Particulate Matter.

       In their 1998 report, the NRC developed a broad source-to-exposure-to-response
framework for evaluating PM issues. The original research portfolio suggested by the NRC
included research to provide insight on how people are exposed to PM and its components
and the mechanisms by which these exposures lead to adverse health outcomes.  To
accomplish this, EPA would first need to identify the particular components of PM that
were leading to adverse outcomes. EPA would then conduct studies to understand actual
human exposures to these components and to relate these exposures to health effects
observed in the real world. Subsequently, the NRC identified the need to develop methods
for measuring PM components to support these exposure and health studies.

       Since the issuance of the  1998 NRC report, we have learned much about PM, but
we have not identified a particular component that is solely responsible for observed
adverse health outcomes.  We now believe that most PM components are likely to have a
toxic effect and we are also focusing on particular sources of PM as potential risk drivers.
As a result, our PM methods research needs to provide the measurement techniques
necessary to support investigations relating to components and sources of PM. If future
research studies justify a change  in the format of the PM NAAQS, the methods used in
those studies could be developed as a revised FRM. The challenge that we face in ORD is
to seek an appropriate balance between supporting nearer-term needs associated with the
1997 NAAQS and longer-term research to provide the scientific basis for future NAAQS
and, given its role in supporting both of these needs, our PM methods research program lies
at the center of this challenge.
                                      45

-------
Conclusion

       As indicated above, ORD does not agree with the findings of this draft report. We
do not accept the underlying premise that a decline in EPA methods development activities
has led to delay in achieving EPA PM program goals. We also believe that this draft report
has not adequately captured the scientific and historical perspective of the EPA PM
research program, which must be understood to place this report in the proper context.

       If you have any questions you would like to bring to my attention, please call me on
(202) 564-6620.

Attachment

cc: Henry L. Longest II
   William Farland
   Gary Foley
   Jeffrey Holmstead
   Rob Brenner
   Jerry Kurtzwig
                                       46

-------
                                                                   Appendix E
                                Distribution

EPA Headquarters

Assistant Administrator for Air and Radiation (6101 A)                                                    I
Assistant Administrator for Research and Development (8101R)
Comptroller (2731 A)
Agency Follow-up Official (2710A)
Agency Follow-up Coordinator (2724A)
Audit Follow-up Coordinator, Office of Air and Radiation (6102A)
Audit Follow-up Coordinator, Office of Research and Development (8102A)
Associate Administrator for Congressional and Intergovernmental Relations (1301A)
Director, Office of Regional Operations (1108A)
Associate Administrator for Communication, Education, and Media Relations (1101 A)
Director, Office of Air Quality Planning and Standards (C404-04)
Director, National Exposure Research Laboratory (MD-75)
Leader, Monitoring and Quality Assurance Group (C339-02)
Office of Air Quality Planning and Standards Audit Liaison (C404-2)
National Exposure Research Laboratory Audit Liaison (MD-343-01)

EPA Regions

Regional Administrators
Regional Air Program Directors
Regional Public Affairs Offices

State and Local Air Pollution Control Agencies

New York Department of Environmental Conservation
Pennsylvania Department of Environmental Protection
South Carolina Department of Health and Environmental Control
Mississippi Department of Environmental Quality
Illinois Environmental Protection Agency
Ohio Environmental Protection Agency
Texas Commission on Environmental Quality
California Air Resources Board
Puget Sound Clean Air Agency
                                      47

-------
EPA Office of Inspector General

Inspector General (2410)
Assistant Inspector General for Program Evaluation
Assistant Inspector General for Human Capital
Assistant Inspector General for Planning, Analysis and Results
Assistant Inspector General for Congressional and Public Liaison
Counsel
Science Advisor
Product Line Directors
Editor
Human Resource Center Managers
                                       48

-------