oversight

Air Pollution: EPA's Actions to Resolve Concerns with the Fine Particulate Monitoring Program

Published by the Government Accountability Office on 1999-08-12.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                  United States General Accounting Office

GAO               Report to the Chairman, Subcommittee
                  on VA, HUD, and Independent Agencies,
                  Committee on Appropriations, House of
                  Representatives

August 1999
                  AIR POLLUTION
                  EPA’s Actions to
                  Resolve Concerns With
                  the Fine Particulate
                  Monitoring Program




GAO/RCED-99-215
      United States
GAO   General Accounting Office
      Washington, D.C. 20548

      Resources, Community, and
      Economic Development Division

      B-282851

      August 12, 1999

      The Honorable James T. Walsh
      Chairman, Subcommittee on VA, HUD,
        and Independent Agencies
      Committee on Appropriations
      House of Representatives

      Dear Mr. Chairman:

      In July 1997, the Environmental Protection Agency (EPA) established new
      ambient air quality standards for particulate matter smaller than 2.5
      microns (PM2.5)—fine particles about 1/30th of the thickness of a human hair
      and which, according to EPA, cause tens of thousands of premature deaths
      annually. These fine particles are a complex mixture of acids, metals,
      carbon, and other potentially harmful airborne substances. Because the
      health risks of the components comprising the PM2.5 mixture are poorly
      understood, EPA is both implementing the PM2.5 standards and investigating
      scientific uncertainties associated with these components. As part of
      implementing the new standards, in 1997 EPA announced plans for a more
      than $50 million nationwide network of 1,500 PM2.5 monitoring sites to be
      deployed by December 31, 1999. EPA’s plans also called for over 90 percent
      of these sites to use monitors that measure only the total mass of fine
      particles collected over a 24-hour period. Such “mass-only” monitors,
      while required for determining areas that exceed the PM2.5 standards,
      provide limited data to address the scientific uncertainties about the
      health effects and health risks of the components comprising the PM2.5
      mixture. Other, more expensive and complex monitors are needed to help
      address these scientific uncertainties.

      In March 1998, the National Academy of Sciences (Academy) issued a
      report1 that questioned EPA’s PM2.5 monitoring plans, calling the agency’s
      plans misdirected and insufficient to address important data gaps and
      scientific uncertainties. While EPA is currently fully funding the network,
      state and local agencies are responsible for establishing and operating the
      monitoring sites. In view of the Academy’s report, you asked us to
      describe (1) EPA’s actions in response to the Academy’s concerns with the
      planned PM2.5 monitoring program and (2) the challenges that state and
      local agencies face in establishing and operating the PM2.5 monitoring
      program, as well as EPA’s response to these challenges.

      1
       Research Priorities for Airborne Particulate Matter: Immediate Priorities and a Long-Range Research
      Portfolio, Committee on Research Priorities for Airborne Particulate Matter, National Research
      Council, National Academy of Sciences (Mar. 1998).



      Page 1                           GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                   B-282851




                   EPA has taken a number of actions to address the concerns raised in the
Results in Brief   Academy’s March 1998 report, including allocating 57 percent of its
                   monitoring budget to efforts designed to better understand PM2.5 scientific
                   uncertainties—a primary concern in the Academy’s report. For example,
                   EPA reduced the number of planned mass-only monitoring sites by more
                   than 21 percent and substantially increased the number of sites and
                   frequency of analyses at the more complex monitoring sites that identify
                   the components of the PM2.5 mixture. Although one of the Academy’s
                   concerns was that the agency was moving forward rapidly with too narrow
                   a focus on mass-only monitors, EPA continued its rapid deployment of over
                   800 mass-only monitors by December 31, 1998, on the basis that 3 years of
                   monitoring data were needed to determine attainment with the standards.
                   Agency officials also explained that they deployed these monitors in
                   response to presidential and congressional directives to expedite the
                   monitoring network, as well as extensive public comments regarding the
                   thousands of premature deaths and serious illnesses annually from PM2.5
                   pollution. The officials also said that they deployed these monitors only
                   after the monitors and the network design were peer-reviewed.

                   Largely due to EPA’s rapid deployment of mass-only monitors, state and
                   local agencies have encountered problems operating nearly one-third of
                   their monitors—problems that have increased labor costs and impacted
                   the agencies’ ability to meet EPA’s data quality requirements. For example,
                   the six state and local agencies we contacted had monitors fail to operate
                   in very cold weather, filters contaminated from leaky seals, and data lost
                   due to monitor computer failures—problems they attribute to inadequate
                   field testing of these monitors before deployment. The officials explained
                   that, due to EPA-imposed time constraints, they had to purchase mass-only
                   monitors based on prototype design and manufacturer specifications
                   without full field evaluation under actual operating conditions. While
                   deployment was rapid, EPA officials said the agency’s monitor certification
                   process was followed and that EPA has taken other actions to help states
                   address these challenges, including requesting warranty extensions from
                   the affected manufacturers and increasing funding for operation and
                   maintenance activities. However, state and local agencies remain
                   concerned that the future deployment of the more complex and costly
                   speciation monitors without adequate field testing would present even
                   greater challenges. As a result, EPA has reevaluated its monitoring plans
                   and delayed deployment of the more complex and costly speciation
                   monitors an additional year—to December 2000—to allow more time for
                   field testing. While these are steps in the right direction, this report
                   contains a recommendation that EPA ensure that future monitors



                   Page 2                  GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
             B-282851




             successfully pass full field testing before large-scale deployment is
             permitted.


             Unlike most other pollutants that EPA regulates, PM2.5 is a complex mixture
Background   of airborne particles and gases that interact physically and chemically,
             vary significantly by region and locality, and can be transported hundreds
             of miles by the wind. PM2.5 particles are primarily formed by the combustion
             of coal, oil, gasoline, diesel, and wood from such sources as power plants,
             industrial facilities, cars, trucks, and wood stoves. According to EPA,
             adverse health effects have been linked statistically to exposure to the
             aggregate, or “mass,” of fine particles smaller than 2.5 microns, but many
             scientists are concerned that one or more components of the PM2.5 mixture
             may be more toxic than others or that certain physical or chemical
             characteristics of the components may be more important to human health
             than the total particle mass. As a result, EPA is faced with the dual tasks of
             both implementing the 1997 standards2 for particle mass and
             simultaneously investigating the health, exposure, and atmospheric
             uncertainties associated with the PM2.5 mixture. According to the Director
             of EPA’s Emissions, Monitoring, and Analysis Division, this was an
             ambitious technical challenge unprecedented in EPA’s history.

             Different types of monitors are required to address these different
             regulatory and research goals, and, according to EPA, all of the currently
             available monitors have limitations if used alone; however, the agency
             believes that collectively these various monitors provide appropriate and
             useful information. Table 1 provides information on the five types of
             monitors EPA plans to use in its revised PM2.5 monitoring network.




             2
              On May 14, 1999, a three-judge panel of the U.S. Court of Appeals for the District of Columbia vacated
             the PM10 standard for coarse particles and remanded to EPA the PM2.5 standards for fine particles
             (American Trucking Ass’ns, Inc. v. EPA, Nos. 97-1440 & 97-1441 (D.C. Cir. May 14, 1999)). After further
             briefing, on June 18, 1999, the court decided not to vacate the PM2.5 standards at this time. On June 28,
             1999, EPA appealed the May 14 decision to the full court of appeals and hopes to sustain the standards
             that were remanded. As of the date of issuance of this report, EPA’s PM2.5 standards are still in place,
             and EPA is continuing with the deployment and operation of the PM2.5 fine particle monitoring
             program.



             Page 3                            GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                                              B-282851




Table 1: Types of Particulate Matter (PM2.5) Monitors
                                                          Number of sites                                    Annual operation and
Monitor type               Description                    planneda                 Capital costsb            maintenance costsc
Mass-only (Compliance)     Agencies weigh filters in    1,094                      $17,000 to $22,200        $11,900 for 1-in-6 day
                           laboratory before and after                                                       sampling to $36,900 for
                           a 24-hour sampling period                                                         daily sampling
                           to collect total PM2.5 mass;
                           particles collected in
                           accordance with
                           EPA-approved method;
                           primary focus on
                           population centers.
Speciation                 Use several different inlet 301                         $23,000                   $30,200 for 1-in-6 day
                           tubes and different filters                                                       sampling to $137,200 for
                           to collect suspected                                                              daily sampling
                           harmful components of the
                           PM2.5 mixture, such as
                           acids, metals, and organic
                           carbon; subsequent
                           analysis of fine particles in
                           laboratory.
IMPROVEd                   Similar to speciation        108                        $23,000                   $30,200
                           monitors in design
                           (collects particles on three
                           different filters); monitors
                           located in national parks
                           and wilderness areas;
                           monitors collect
                           background readings and
                           transport data for PM2.5
                           program.
Continuous                 In lieu of filters, this design 137e                    $20,000                   $6,000 to $8,000
                           uses an internal analytical
                           apparatus to
                           instantaneously analyze
                           the particles passing
                           through the sampler.
Supersitef                 Uses prototype and             4 to 9g                  $1.0 to $3.5 million,     (included in prior
                           research-grade monitors,                                (includes O&M costs       column)h
                           as well as one or more of                               and data analyses)
                           the above monitors, to
                           conduct intensive fine
                           particle analyses of the
                           complex PM2.5 mixture,
                           population exposure
                           patterns, and emissions
                           source characterization
                           and identification.
Total                                                     1,500

                                                                                                              (Table notes on next page)




                                              Page 4                        GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
B-282851




a
 Figure based on revised ambient air monitoring network plans as of May 1999 (range of planned
mass-only compliance sites nationwide is 1,050 to 1,100).
b
 Capital costs for speciation, IMPROVE (defined below), and Supersite monitors are best
estimates as of May 1999, including one-time site setup costs such as those for utility hookups.
c
 Operation and maintenance (O&M) costs include the estimated costs of site operations, filters,
laboratory analysis, and data reporting (all O&M costs are estimates since 1 full year of operation
will not be completed until Dec. 31, 1999).
d
 The Interagency Monitoring of Protected Visual Environments (IMPROVE) program is a joint effort
with the Department of the Interior, Federal Land Managers, and state and local agencies.
e
Continuous and speciation monitors are generally colocated with other monitors.
f
 These sites actually use a combination of advanced monitoring techniques to understand the
complex PM2.5 mixture.
g
 Supersites will be coordinated with ongoing state and local agency monitoring efforts to optimize
data uses.
h
Detailed estimates of O&M costs for these monitoring sites are still under development.



While mass-only monitors are essential for measuring total particle
mass—which is necessary to determine if a state or local area is in
compliance with the PM2.5 mass standards—they provide limited data to
help address scientific uncertainties. For example, mass-only monitors, by
themselves, are insufficient for a comprehensive assessment of the
sources contributing to fine particle pollution. However, EPA officials
noted that data on PM2.5 mass does help with some scientific research and
that PM2.5 mass is a likely contributor to adverse health effects. Although
they are among the more precise PM2.5 monitoring devices available, in
some instances mass-only monitors underestimate the fine particle mass
because they lose unstable components of the PM2.5 mixture as the
temperature rises during the day. Also, as shown in table 1, the capital
costs for mass-only monitors vary significantly. This cost difference
depends mostly on whether the monitor’s filter must be manually changed
after a single 24-hour sampling period or whether the monitor can
electronically change filters for multiple sampling periods. A schematic of
a typical mass-only monitor is shown in figure 1. These monitors, which
are generally about 2 to 2.5 meters high, must have unobstructed air flow
for a minimum of 2 meters in all directions, with the air inlet located from
2 to 15 meters above ground level.3




3
  These monitors must meet EPA-specified design, performance, and operational requirements as
provided in 40 CFR, Part 50, App. L (July 18, 1997 Federal Register).



Page 5                           GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                                      B-282851




Figure 1: Schematic of a Typical Pm
2.5 Mass-Only Monitor                                                                         Air inlet

                                            Air temperature probea




                                                                                                                Hinged door




                                                                                    Filter
                                                                                  cassetteb
                                          Cooling fan


                                                                                                           Door seal along edges




                                          Computer/
                                          data logger




                                      a
                                      Air temperature is recorded electronically by the monitor’s computer.
                                      b
                                       The filter exchange mechanism for sequential monitors uses either a rotating filter cassette or a
                                      pneumatic tube exchange system.


                                      Source: GAO Illustration based on EPA-supplied information.


                                      Mass-only monitors are not the only monitors with limitations. The
                                      speciation and IMPROVE monitors also lose unstable, volatile components of
                                      the PM2.5 mixture as the temperature rises during the day. Continuous




                                      Page 6                           GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
B-282851




monitors are needed to obtain the more frequent readings that portray
human exposures, but these monitors also have limitations. EPA officials
said that continuous monitors are an acceptable indicator tool when used
in combination with other monitors, but they produce inaccurate readings
in some circumstances, especially in humid conditions. For example,
continuous monitors do not provide data on the variety of chemicals that
make up the PM2.5 mixture, and some also lose volatile components of the
mixture to a greater degree than do other monitors.

Because all of the currently available monitors have limitations, EPA
officials have developed another monitoring approach—known as
supersite—that describes the agency’s most intensive efforts to
understand the complex PM2.5 mixture in various regions of the nation. As
explained in table 1, supersites use prototype and research-grade
monitors, along with one or more of the other monitors, to conduct
intensive fine particle analyses of the complex PM2.5 mixture, population
exposure patterns, and emissions source characterization and
identification. Because of the high cost of supersites—from $1.0 to
$3.5 million per site—fewer than 10 supersites will be established
nationwide. Supersites may operate from less than 1 to more than 5 years.
According to agency officials, the supersites part of the PM2.5 monitoring
program is still in the developmental phase.

While fiscal year 1999 funding for the PM2.5 monitoring program was
$65.7 million (including $50.7 million for state and local grants, plus an
additional $15.0 million for the supersites program), EPA estimates
that—after establishment—ongoing expenses for the PM2.5 monitoring
program will be about $42.5 million annually. Although the overall
network is designed, overseen, and currently fully funded by EPA, state and
local air quality agencies are responsible for establishing and operating the
monitoring sites, including taking periodic samples, handling and
transporting the samples, conducting laboratory analyses, ensuring data
quality, and reporting the data to EPA and the states. Equally important,
state and local agencies are responsible for developing strategies to
achieve and maintain the PM2.5 standards, including identifying and
controlling PM2.5 emissions sources. According to EPA, the agency will
assume full funding for the PM2.5 network until such time as the Congress
determines that states should contribute to the operation and maintenance
of the network. At that point, states will be expected to provide
approximately 40 percent of the total funding needed to operate the
program, assuming a traditional grant funding approach. Under traditional
grant funding, EPA covers about 60 percent of program expenses, and state



Page 7                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                     B-282851




                     and local agencies provide matching funds sufficient to cover 40 percent
                     of program costs. Such a funding arrangement is generally known as a
                     60/40 match. While the PM2.5 monitoring program is the most expensive EPA
                     has ever asked state and local agencies to implement, the decisions to be
                     made based on the PM2.5 data are also envisioned to be very expensive. For
                     example, EPA estimates that industry will spend about $8.6 billion annually
                     to control PM2.5 emissions beginning in 2008. Full compliance with the PM2.5
                     standards is expected no later than 2012. EPA’s next review of the PM2.5
                     standards is scheduled for 2002. According to agency officials, subsequent
                     reviews should occur at intervals no longer than 5 years thereafter.


                     EPA made numerous changes in its PM2.5 monitoring program to address the
EPA Has Taken        concerns raised in the Academy’s March 1998 report. For example, in
Actions to Address   response to the Academy’s concern that EPA’s planned program was too
the Academy’s        heavily oriented toward mass-only monitors, EPA reduced the number of
                     planned mass-only monitoring sites from 1,392 to 1,094 sites, about a
Concerns             21-percent reduction. This change allowed EPA to increase the number and
                     frequency of analyses at the more expensive speciation monitoring sites
                     and to more than double the number of continuous monitoring sites
                     planned—actions that should help the agency better address the health,
                     exposure, and atmospheric uncertainties of PM2.5. After these changes, EPA’s
                     revised plans called for 57 percent of the PM2.5 monitoring budget to be
                     devoted to monitoring efforts designed to better understand PM2.5 scientific
                     uncertainties. EPA’s responsiveness to the Academy’s earlier concerns was
                     commended by several members of the Academy and the Clean Air
                     Science Advisory Committee (CASAC)4 in a public session in
                     November 1998. Table 2 summarizes eight key concerns in the Academy’s
                     1998 report and the actions EPA has taken to address them.




                     4
                     CASAC is a permanent subcommittee of the Science Advisory Board established by the Clean Air Act
                     Amendments of 1977.



                     Page 8                         GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                                                B-282851




Table 2: Summary of EPA’s Actions to Address the Academy’s Key Concerns
                                                                                                                                    Concern
Description of key concern                      Summary of EPA’s actions                                                            addressed
1. EPA is moving forward rapidly with too       EPA reduced the number of mass-only monitors planned from 1,392 to           Partially
narrow a focus on mass-only monitoring.         1,094; increased the number and frequency of analyses at 54 speciation
                                                sites; increased the number of continuous monitors from 62 to 137; added
                                                plans for up to 9 supersites (an increase of $15 million in EPA’s monitoring
                                                budget); however, EPA continued with rapid deployment of 823 mass-only
                                                monitors by Dec. 31,1998.
2. Planned network may not adequately           Changes in planned network allowed EPA to increase fiscal year 1999                 Yes
support research on health effects,             monitoring funds for health effects, exposure, and atmospheric modeling
exposure assessment, and atmospheric            research by $21.5 million (a 42-percent increase); revised plan now
modeling.                                       allocates 57 percent of $65.7 million in fiscal year 1999 monitoring funds
                                                to monitoring efforts to help address these research issues.
3. Plans to collect speciation samples once     EPA used $3.1 million of funds planned for the mass-only monitoring                 Mostly
every 6 days will not provide useful data for   program to increase the frequency of speciation sampling from one
improving health risk assessments.              sample every 6 days to daily sampling at 10 major urban sites, and from
                                                one sample every 6 days to one sample every 3 days at 44 other sites;
                                                efforts to increase sampling frequency at another 40 speciation sites is
                                                under review; regarding remaining 208 sites, EPA officials recognize that
                                                having more frequent data is better, but said all data, including samples
                                                collected once every 6 days, will be useful to scientific community.
4. EPA should make greater use of               EPA used $3.4 million of funds planned for the mass-only monitoring                 Yes
continuous monitors to help determine           program to increase the number of continuous monitoring sites from 62 to
human exposures and facilitate time-series      137, a 121-percent increase in continuous monitoring sites nationwide.
epidemiological studies.
5. Mass-only monitors will likely not measure   EPA’s revised plans call for speciation sites to use three types of filters         Yes
some important components of the PM2.5          (teflon, nylon, and quartz) to measure components of the PM2.5
mixture, such as nitrates and organic           mixture, such as nitrates, organic compounds, sulfates, metals, and
compounds.                                      other components of the PM2.5 mixture; also, 108 IMPROVE sites in
                                                national park and wilderness areas will use these three filters.
6. Future research results may indicate         EPA’s revised plans provide a better relative mix of monitoring sites to Yes
mass-only monitors are not measuring the        help the agency both implement the 1997 standards for particle mass and
most biologically important aspects of          investigate the health, exposure, and atmospheric uncertainties
particulate matter.                             associated with the PM2.5 mixture.
7. Specific objectives, operating conditions,   Objectives, conditions, number, and siting criteria for the mass-only     Yes
number, and location of monitors should be      monitoring network were peer reviewed in August 1996; first 54 speciation
independently peer-reviewed prior to            sites, 108 IMPROVE sites, and 2 initial supersites were peer-reviewed in
implementation.                                 November 1998; EPA has announced plans to have the remaining 248
                                                speciation sites peer-reviewed.
8. Interface between monitoring and             EPA established steering committee to coordinate its PM2.5 monitoring               Mostly
research on particulate matter still largely    activities with others involved in similar activities (federal, state, local, and
uncoordinated and fragmented.                   multinational associations; academia, nonprofit advisory bodies, others);
                                                efforts to establish procedures ensuring optimization of PM2.5 data
                                                collection activities continue.

                                                As shown in table 2, EPA continued with the rapid deployment of over 800
                                                mass-only monitors by December 31, 1998. Although the Academy was
                                                concerned that the agency was moving forward too rapidly with too



                                                Page 9                          GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                  B-282851




                  narrow a focus on mass-only monitors, EPA officials explained that they
                  continued with the rapid deployment of these monitors for several
                  reasons. First, they explained that the agency needed 3 years of data to
                  make reliable determinations of those areas that exceed the new standards
                  and that the agency received extensive public comments regarding the
                  estimated 15,000 premature deaths annually from fine particles. They also
                  said that EPA deployed these monitors in response to a July 1997
                  presidential directive to expedite the deployment of the network and a
                  provision in the Transportation Equity Act for the 21st Century requiring
                  compliance monitors be installed by December 31, 1999. Additionally, they
                  said that the agency deployed these monitors only after both the monitors
                  and the network design were peer-reviewed, and that the number of
                  mass-only monitoring sites is comparable to other ambient air monitoring
                  programs.

                  EPA officials pointed out that the agency had planned to deploy 1,500
                  monitoring sites over a 3-year period (beginning in fiscal year 1999), with
                  about 20 percent deployed the first year, 40 percent the second year, and
                  the remaining 40 percent the third year. However, in July 1997, EPA
                  condensed its deployment schedule from 3 to 2 years, with over half of the
                  1,500 monitoring sites to be deployed by December 31, 1998. As will be
                  discussed in the next section, some state and local agencies’ experiences
                  indicate that the rapid deployment of these monitors has not been without
                  problems. Nonetheless, even with the early problems encountered, EPA
                  officials believe the PM2.5 mass-only monitoring network will be producing
                  quality data earlier than had EPA used its originally planned 3-year
                  deployment schedule. (Additional information on the status of EPA’s
                  actions to address the Academy’s concerns is provided in app. I.)


                  State and local agencies face both near-term and long-range challenges in
State and Local   establishing and operating the PM2.5 monitoring program. These challenges
Agencies Face     include correcting various operational problems with mass-only monitors,
Implementation    resolving data gaps caused by a lack of spare monitors, ensuring the full
                  testing of future monitors before deployment, and obtaining sufficient
Challenges        state and local resources to pay for their share of the program. EPA has
                  actions under way or planned to mitigate these challenges. Among other
                  things, EPA is working with state and local agencies and affected
                  manufacturers to remedy the remaining operational problems with
                  monitors, providing funding for spare monitors, and developing a
                  deployment schedule that will provide additional time for field testing of
                  the more complex speciation monitors yet to be deployed. As of May 1999,



                  Page 10                 GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                                         B-282851




                                         the network was 91-percent complete, which EPA officials told us
                                         represents a significant amount of success, considering the enormousness
                                         of the task and their historical experiences with implementing a major
                                         new monitoring program of this magnitude. However, about 9 percent of
                                         the network that was supposed to be operational by December 31, 1998,
                                         was still not operating after more than 4 months.


Monitor Operational                      While funding for the PM2.5 monitoring program is a long-term concern of
Problems Have Affected                   state and local agencies, their most immediate challenge has been
Data Quality                             correcting the operational problems they have experienced with the
                                         mass-only monitors that were to be deployed by December 31, 1998.
                                         According to a May 1999 EPA survey, state and local agencies nationwide
                                         have had operational problems with about 30 percent of the monitors
                                         deployed to date. For those monitors that have experienced problems,
                                         figure 2 shows the types of obstacles that state and local agencies have
                                         experienced.


Figure 2: Obstacles to Pm 2.5 Monitor
Operations for the First Quarter, 1999
                                                                   •                            3% Other problems
                                                                        •                       5% Hiring problems

                                                     30%                     15%
                                                                              •                 Contract problems
                                                    •


                                                                             21%
                                                                              •                 Procurement problems
                                                          26%
                                                            •                                   Site-installation problems




                                                                                                Monitor operational problems


                                         Source: EPA’s Office of Air Quality Planning and Standards.




                                         As shown in figure 2, while monitor malfunctions have been the single
                                         largest challenge for state and local agencies, other challenges have



                                         Page 11                            GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                                          B-282851




                                          included site installation problems (such as obtaining leases to use
                                          property), state procurement delays, and state hiring limitations. Because
                                          data quality reports for first quarter monitoring activities are not due until
                                          July 1, 1999, it is too early to tell the nationwide impact of these problems.
                                          However, we contacted six state and local agencies that, collectively,
                                          operate 218 of the 823 monitoring sites deployed, or about one-fourth of
                                          the sites that were to be established by December 31, 1998, to discuss their
                                          monitoring experiences and the impact of monitor problems on data
                                          quality. As shown in figure 3, each of the six state and local agencies that
                                          we contacted have experienced significant operational problems with one
                                          or more of their mass-only monitors since January 1, 1999—problems they
                                          believe have impacted their ability to meet EPA’s PM2.5 data quality
                                          requirements for the first quarter of 1999.


Figure 3: Percentage of Monitors With
Operational Problems Potentially            100        Percentage
Impacting Data Quality During the First
Quarter, 1999
                                             80




                                             60




                                             40




                                             20




                                                0
                                                          )



                                                                           )



                                                                                             )



                                                                                                             )



                                                                                                                             )



                                                                                                                                               )
                                                          es



                                                                           es



                                                                                             es



                                                                                                             es



                                                                                                                             es



                                                                                                                                               es
                                                       sit



                                                                      sit



                                                                                         sit



                                                                                                           sit



                                                                                                                           sit



                                                                                                                                          sit
                                                    (70



                                                                     (23



                                                                                       (26



                                                                                                       (31



                                                                                                                       (18



                                                                                                                                         (50
                                                  ia



                                                                gia




                                                                                    is



                                                                                                      rk



                                                                                                                      MD




                                                                                                                                    as
                                                rn




                                                                                 no



                                                                                                  Yo




                                                                                                                                     x
                                                                or




                                                                                                                  AQ



                                                                                                                                  Te
                                             lifo




                                                                                Illi
                                                               Ge




                                                                                                  w



                                                                                                                 SC
                                                                                               Ne
                                           Ca




                                          State



                                          *South Coast Air Quality Management District (California).

                                          Source: GAO illustration based on estimates supplied by officials of six state and local agencies.




                                          Page 12                                                 GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
    B-282851




    Among the more significant problems state and local agencies have
    encountered are the following:

•   Filter cassettes jammed in very cold weather, resulting in lost data and
    causing state and local agencies to operate their monitors in manual mode,
    which increases labor costs.
•   Monitor access doors were improperly sealed, allowing dust into the
    monitor, contaminating the filter, and resulting in voided and incorrect
    filter readings.
•   Monitor cooling fans pulled in dust containing coarse mode (greater than
    2.5 microns) and other unwanted particles, potentially resulting in
    incorrect readings.
•   Software for the automated multiday samplers had problems, resulting in
    lost data because the most recent readings overwrote previous readings.
•   Troubleshooting instructions provided to state and local agencies were
    insufficient, which precludes technicians from making simple repairs
    on-site and requires monitors to be shipped back to the manufacturer, thus
    causing more data to be lost in the interim.

    Some officials in the six state and local agencies we contacted said that
    the rapid deployment and the number of monitors needed exceeded
    manufacturers’ short-term production capacity. For example, in addition
    to being nearly 2 months later than expected, officials of the California Air
    Resources Board’s (CARB) Monitoring and Laboratory Division5 told us that
    5 of the 120 monitors they received arrived at their offices already in need
    of repair. In conducting initial acceptance testing on these monitors, the
    CARB staff found that some monitors had wiring problems, and, in some
    cases, parts within the sampling tube had not been anodized, a process
    that coats the interior wall of the sampling tube so that particles pass
    through freely and do not adhere to the walls of the tube. Georgia, Illinois,
    and the South Coast Air Quality Management District (a local program in
    California) also had monitors delivered later than EPA had planned.
    According to EPA, the late deliveries were due to the length of time EPA
    needed to get final versions of the monitors and full documentation from
    the manufacturers, thus delaying testing and approval for use in the PM2.5
    monitoring program. Although New York’s monitors were received on
    time, two monitors failed initial acceptance testing and had to be sent
    back to the manufacturer for repairs; according to New York officials, data
    were lost in the meantime.



    5
     CARB’s Monitoring and Laboratory Division performed initial acceptance testing for Arizona, Hawaii,
    and district offices in California.



    Page 13                          GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
B-282851




While getting monitors installed and operating is a major step, it does not
ensure data quality. For example, a monitoring expert with California’s
South Coast Air Quality Management District—who is also a member of
CASAC and Chair of the Association of Local Air Pollution Control Officials
(ALAPCO)6 Monitoring Committee—told us he is concerned about the
quality of data being recorded by these first monitors. In March 1999,
according to this member, the District found that over 40 percent of the
field blanks—filters used to identify contamination inside the monitoring
device in the actual field environment—had shown excessive
accumulations of dust. Officials from Illinois and New York also told us
they are concerned they will not meet the data quality requirements for
several monitoring stations for the first quarter of 1999, and some
expressed concern about the second quarter of 1999 as well. Additionally,
an unknown factor, they said, is how well these monitors will perform in
hot, humid, or dusty summertime conditions. EPA officials said that, if such
additional problems occur, they will work with the state and local
agencies and the manufacturers to resolve any problems as expeditiously
as possible. Additionally, EPA officials pointed out that it is primarily due to
the agency’s extensive PM2.5 quality assurance program—instituted before
the mass-only monitors were deployed—that state and local agencies have
been able to identify some of the operational problems in the deployment
of the monitoring network.

State and local agency officials said that manufacturers have been
repairing monitors under their warranty provisions. While some said the
problems experienced are not unusual for a major new monitoring
program of this magnitude, others believed differently. For example,
Illinois officials said the equipment malfunctions were severe, and the
decision to rapidly deploy the majority of the nationwide network in the
first year now meant “massive equipment redesigns and retrofits.” While
EPA officials agreed that some monitor repairs have been necessary, they
disagreed with Illinois’ characterization of these problems as massive. The
officials noted that both EPA headquarters and regional offices have been
assisting the affected state and local agencies in resolving these early
program problems. For example, because of the magnitude of the
problems encountered, in April 1999 EPA’s contracting officer sent letters
to the mass-only monitor manufacturers requesting full warranty coverage
for an additional 3 months to ensure that the manufacturers’ repairs have
been successful. EPA officials said the problems encountered did not
surface in the agency’s initial testing and that the manufacturers have
cooperated in repairing monitors under their 1-year warranty provisions.

6
 ALAPCO is a trade association for local air pollution programs.



Page 14                          GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                           B-282851




                           EPA has also provided state and local agencies with new options for
                           meeting their data quality requirements in the interim while manufacturers
                           work to correct equipment problems, and the agency has increased its
                           funding estimates for monitor operation and maintenance activities,
                           including labor costs.


Some Data Lost Because     An additional concern relates to the lack of spare monitors. According to
Spare Monitors Were Not    state and local agency officials, EPA did not initially provide funding for
Available                  spare monitors. Therefore, as monitor operational problems have
                           occurred, data have been lost while monitors are shipped back to the
                           manufacturers for repairs. According to CARB officials, provisions should
                           have been made for about 10 percent more monitors to be used as spares,
                           given the agency’s plans for widespread deployment in the first year.
                           Similarly, New York air quality officials said that, without spare monitors,
                           they found it necessary to cannibalize two monitors in order to obtain
                           spare parts in an attempt to increase the number of valid samples.

                           EPA   officials agree that initial grants (fiscal year 1998) for the purchase of
                           PM2.5 monitors did not include funds to purchase spare monitors. In their
                           opinion, it was more important initially to get the compliance monitors set
                           up and operational at as many sites as possible in the first year, and then
                           later to buy any spare monitors that might be needed. The officials pointed
                           out that, in designing the program, they believed that spare monitors
                           would more likely be needed after the first year when, due to wear and
                           tear, monitors may need major repairs. However, after the early
                           experiences, they said EPA took steps to rectify the problem by approving
                           funds for the purchase of spare monitors in state and local agency grants
                           (fiscal year 1999) for the next round of purchases. EPA has also provided
                           state and local agencies with options for meeting data quality
                           requirements, such as allowing less frequent sampling at problem sites in
                           the interim and using valid results from similar time periods next year to
                           fill current data gaps.


Full Testing of Monitors   Most of the state and local air quality monitoring officials we contacted
Not Completed Before       believe EPA’s deployment of the mass-only monitors was too optimistic and
Widespread Deployment      that a slower schedule would have been better. In their opinion, slower
                           deployment would have allowed the identification of monitor problems
                           before the bulk of the monitors were operating in the field. They attribute
                           many monitor difficulties to EPA’s expedited deployment of mass-only
                           monitors without adequate operational testing performed under actual



                           Page 15                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
B-282851




field conditions for a full annual cycle before their widespread
deployment. Some of the officials said that they informed EPA before
deployment that the agency was moving too quickly to authorize the PM2.5
mass-only monitors and that they disliked having to purchase the monitors
based on prototype design and manufacturer specifications without seeing
a production model in operation.

Based on the PM2.5 monitoring program’s level of complexity and the
sensitivity of the measurements, the state and local agencies said they
wanted sufficient time to evaluate a few monitors over different seasons
and in actual field conditions before making large-scale purchases. For
example, CARB officials noted that their agency commented early on that
they were concerned about the adequacy of the mass-only monitors and
the lack of full field testing.7 Similarly, New York officials said that the
majority of the problems related to the deployment of the monitoring
portion of the PM2.5 program have resulted from the compressed time frame
for implementation. The officials believe that in addition to extensive
design development, laboratory testing, and preliminary all-season field
testing, the program should have been phased in gradually. Georgia,
Illinois, and Texas officials also expressed concern that the deployment
schedule was too ambitious, especially considering the sensitivity of the
new PM2.5 monitoring system and past problems experienced with
deploying new monitoring systems for other air pollutants. Most suggested
that an alternative “go slow” approach would have saved time and money
by eliminating the need for state employees to ship defective monitors
back to the manufacturer or to remove and ship defective parts back to
the manufacturer. EPA officials acknowledged that this slower approach
was preferred by the monitoring community but said other
factors—namely public health concerns—were considered in their
decision to expeditiously deploy the mass-only monitors. According to EPA,
the agency recognized state and local agencies’ concerns about the
deployment schedule, but EPA relied on testing under the monitor approval
process rather than delay the deployment in order to test monitors over a
full annual cycle of conditions.

More importantly, some state and local agency officials are concerned that
deployment of the more complex and costly speciation monitors without
adequate field testing would present even greater challenges than they
have faced to date. To help preclude similar problems with future
monitors, the officials suggested that EPA take a slower approach to

7
 Full field testing is an evaluation of the monitor under actual field conditions where temperature,
humidity, and other factors, such as season of the year, are not simulated.



Page 16                           GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                           B-282851




                           approving the remaining monitors to ensure that they operate properly
                           under all conditions and all seasons before their widespread use is
                           undertaken. According to EPA officials, the agency recently reevaluated its
                           PM2.5 monitoring plans and has extended deployment of the more complex
                           and costly speciation and other monitors an additional year—to
                           December 2000—to allow more time for field testing by state and local
                           agencies, as well as completion of additional work sponsored by EPA in
                           fiscal year 1999.


State and Local Agencies   State and local agency officials are generally pleased that EPA has been
Face Future Resource       fully funding the program so far; however, they are concerned about their
Challenges                 ability to match the federal grant in future years. With ongoing program
                           costs expected to exceed $42 million annually, the PM2.5 monitoring
                           program is the most expensive air monitoring program for a single
                           pollutant in EPA’s history, according to agency officials. Once the
                           monitoring program is established and operational, state and local
                           agencies must provide matching funds to operate the PM2.5 monitoring
                           program—a requirement that presents significant challenges to them in
                           the future, they said, given the other air quality programs for which they
                           already must provide matching funds.

                           Some state and local agencies we contacted are uncertain about the
                           amount of matching funds they will have for PM2.5 activities in future years.
                           Resource challenges, they believe, could cause them to have to reduce the
                           number of PM2.5 monitoring sites they will be able to operate when they
                           have to pay for 40 percent of the program costs. The agencies believe that
                           a technological breakthrough in monitor design is needed to reduce their
                           operating costs. Such a breakthrough occurred in the PM10 program8 when
                           a continuous PM10 monitor was developed. EPA officials are aware that this
                           is an expensive program and have said the agency is relying on the current
                           appropriations authority that provides for full federal funding of the PM2.5
                           program through its demonstration phase. The extension of this authority
                           for full funding will mitigate the near-term impact on state and local
                           agencies, they said, and provide an opportunity for advances in monitor
                           design in the intervening years, which could reduce the resource burden
                           state and local agencies will face. The development of a continuous mass
                           monitoring system is a high priority for EPA’s Office of Research and
                           Development (ORD), which has a goal of an improved monitor by the end
                           of 2001. EPA allocated about $200,000 to such efforts in fiscal year 1999,

                           8
                             The PM10 monitoring program was established in 1987 to monitor for coarse particles 10 microns and
                           smaller.



                           Page 17                          GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                  B-282851




                  with increases in funding anticipated for 2000 and 2001. According to ORD
                  officials, this is an appropriate level of federal funding because, due to
                  market incentives, private industry manufacturers are already heavily
                  involved in their own research and development of continuous PM2.5
                  monitors. EPA also noted that an important part of future budget planning
                  will be a review of the availability of new monitoring methods and other
                  scientific advancements, as well as an examination of actual costs for
                  program elements not yet fully implemented.


                  EPA  made numerous changes in its PM2.5 monitoring program to respond to
Conclusions       the concerns raised by the National Academy of Sciences; however, the
                  agency continued with the rapid deployment of over 800 mass-only
                  monitors without full field testing. This decision led to numerous
                  operational problems discovered only after the monitors were deployed,
                  resulting in some lost and questionable data and unanticipated operational
                  and maintenance expenses. While steps are under way to address current
                  problems, future problems with the other more complex and expensive
                  monitors—if they occurred—would impact data quality, increase labor
                  costs, delay needed health protections, and eventually erode public
                  confidence in the network. To help prevent similar problems from
                  occurring in the future, EPA is delaying its planned deployment of
                  speciation monitors for 1 year to allow more time for field testing. While
                  the additional field testing may identify problems before the monitors are
                  deployed, we believe that EPA should not place an arbitrary time limit on
                  its field tests and should take whatever time is necessary to ensure that
                  future monitors successfully pass full field testing before large-scale
                  deployment is permitted.


                  We recommend that the Administrator, Environmental Protection Agency,
Recommendation    ensure that all remaining monitors planned for the PM2.5 network undergo
                  and successfully pass full laboratory and full field testing and evaluation
                  under actual operating conditions to ensure that the monitors meet data
                  quality objectives before large-scale deployment of these monitors is
                  authorized.


                  We provided a draft of this report to the Environmental Protection Agency
Agency Comments   for its review and comment. The agency generally agreed with the overall
                  message of the report, noting that it provides a fair and balanced depiction
                  of EPA’s efforts to implement the PM2.5 monitoring program. Specifically, EPA



                  Page 18                  GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
B-282851




agreed with our conclusion that no arbitrary time limit should be imposed
on testing the remaining monitors planned for the PM2.5 monitoring
network; however, EPA did not comment on our recommendation. EPA also
suggested several changes to clarify information in the report, and we
incorporated these comments where appropriate. Appendix III contains
the full text of the agency’s written comments and our responses.


We conducted our review from October 1998 through June 1999 in
accordance with generally accepted government auditing standards.

As arranged with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 15 days after the
date of this report. At that time, we will send copies of this report to the
Honorable Carol M. Browner, Administrator, Environmental Protection
Agency, and other interested parties. We will also make copies available to
others upon request.

If you have any questions regarding this letter, please contact me or David
G. Wood at (202) 512-6111. Key contributors to this report were William F.
McGee, James R. Beusse, Philip L. Bartholomew, and Richard A. Frankel.

Sincerely yours,




Peter F. Guerrero
Director, Environmental
  Protection Issues




Page 19                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Contents



Letter                                                                                              1


Appendix I                                                                                         22

EPA’s Actions to
Address Concerns
With the PM 2.5
Monitoring Program
Appendix II                                                                                        28

Objectives, Scope,
and Methodology
Appendix III                                                                                       30

Comments From the
Environmental
Protection Agency
Tables               Table 1: Types of Particulate Matter Monitors                                  4
                     Table 2: Summary of EPA’s Actions to Address the Academy’s                     9
                      Key Concerns

Figures              Figure 1: Schematic of a Typical PM 2.5 Mass-only Monitor                      6
                     Figure 2: Obstacles to PM 2.5 Monitor Operations for the First                11
                       Quarter, 1999
                     Figure 3: Percentage of Monitors With Operational Problems                    12
                       Potentially Impacting Data Quality During the First Quarter, 1999




                     Page 20                 GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Contents




Abbreviations

ALAPCO     Association of Local Air Pollution Control Officials
CARB       California Air Resources Board
CASAC      Clean Air Science Advisory Committee
CENR       Committee on Environment and Natural Resources
EPA        Environmental Protection Agency
FRM        federal reference method
HUD        Department of Housing and Urban Development
IMPROVE    Interagency Monitoring of Protected Visual Environments
NARSTO     North American Research Strategy for Tropospheric Ozone
                (and Aerosols).
NAS        National Academy of Sciences
NRC        National Research Council
O&M        operation and maintenance
ORD        Office of Research and Development
PM10       coarse particles 10 microns and smaller
PM2.5      fine particulate matter smaller than 2.5 microns in diameter
SAB        Science Advisory Board
STAPPA     State and Territorial Air Pollution Program Administrators
VA         Department of Veterans Affairs


Page 21               GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix I

EPA’s Actions to Address Concerns With the
PM 2.5 Monitoring Program

               In its March 1998 report,1 the National Academy of Sciences (Academy)
               raised eight key concerns about the ambient air monitoring part of the
               Environmental Protection Agency’s (EPA) PM2.5 program. Academy
               representatives said that EPA has made significant changes to the PM2.5
               program to address many of the key concerns presented in its report.
               Information on EPA’s actions to address these concerns is discussed in this
               appendix.

               1. EPA appeared to be moving forward rapidly with too narrow a focus on
               PM2.5Federal Reference Method mass-only monitoring

               EPA has made significant efforts to broaden its monitoring program focus
               by involving the scientific community in the speciation and supersite
               network planning. For example, since the Academy’s 1998 report, EPA has
               sought expert external scientific advice on the siting, sampling, and
               measurements needed to address health, exposure, and atmospheric
               research questions and jointly sponsored a workshop with over 200
               individuals to further involve the scientific community in EPA’s planning
               for the speciation and supersite monitoring networks. As a result, the
               agency reevaluated its monitoring plans and decided to slow down the
               rollout of the speciation monitors and supersites pending greater input
               from the scientific community. EPA also reduced the number of mass-only
               monitors planned from 1,392 to 1,094 and increased the number and
               frequency of analyses at 54 speciation sites. EPA also more than doubled
               the number of continuous monitoring sites, from 62 to 137, and added
               plans for up to 9 supersites—an increase of $15 million in EPA’s PM2.5
               monitoring budget. Additionally, EPA tripled the types of analyses to be
               performed at 108 Interagency Monitoring of Protected Visual
               Environments (IMPROVE)2 sites located in national parks and wilderness
               areas, requiring these monitors to use three different filters instead of the
               previous single-filter approach. After these changes, EPA’s revised plans
               allocated 57 percent of $65.7 million in fiscal year 1999 PM2.5 monitoring
               funds to efforts designed to better understand PM2.5 scientific uncertainties.

               EPA continued to move forward rapidly with the PM2.5 mass-only monitoring
               network, largely because of extensive public comments regarding the
               estimated 15,000 premature deaths annually from fine particles, a


               1
                Research Priorities for Airborne Particulate Matter: Immediate Priorities and a Long-Range Research
               Portfolio, Committee on Research Priorities for Airborne Particulate Matter, National Research
               Council, National Academy of Sciences (Mar. 1998).
               2
                IMPROVE is a joint effort with the Department of the Interior, Federal Land Managers, and state and
               local agencies.



               Page 22                          GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix I
EPA’s Actions to Address Concerns With the
PM 2.5 Monitoring Program




July 1997 presidential directive to expedite the network, and a provision in
the Transportation Equity Act for the 21st Century calling for completion of
the compliance portion of the network by December 31, 1999. EPA officials
pointed out that, prior to these inducements, the agency had planned for
the 1,500 monitors to be deployed over a 3-year period, with about
20 percent deployed the first year, 40 percent the second year, and the
remaining 40 percent the third year. However, in July 1997, EPA condensed
its original deployment schedule from 3 to 2 years, with over half of the
1,500 monitors to be deployed by December 31, 1998. EPA officials said
they were required by regulation to establish a minimum of about 850
federal reference method (FRM) mass-only sites for a complete nationwide
network. About 823 FRM mass-only monitors were supposed to be fully
operational by December 31, 1998, with the remainder to be installed by
December 1999.

2. The relative mix of planned monitoring sites may not adequately
support health effects, exposure, and atmospheric modeling research
efforts

Since the Academy’s report, EPA has reduced the number of planned FRM
mass-only sites from 1,392 to 1,094. EPA’s revised allocation of fiscal year
1999 PM2.5 monitoring funds allocates about $28.5 million to FRM mass-only
monitoring; $17.2 million for speciation monitoring; $5.1 million for
continuous monitoring; and $15.0 million for supersite monitoring. EPA has
delayed the deployment of these supersites in order to more fully integrate
supersite planning with its research needs for health risk, exposure
assessment, and monitoring advances. Intensive supersite monitoring
efforts may run from less than 1 to more than 5 years and are estimated to
cost from $1.0 to $3.5 million per site. EPA officials believe this allocation
of funds strikes an appropriate balance between regulatory and research
needs, taking into consideration that this is the first time that EPA’s
regulatory program has ever designed an ambient air monitoring program
both to determine compliance and to address scientific uncertainties
about a pollutant.

The $15.0 million in science and technology funds EPA received in fiscal
year 1999 to conduct intensive research monitoring efforts at four to nine
supersites was to be used for scientific purposes, but EPA had tremendous
discretion in how the $50.7 million in state and local agency grant funds
was allocated. By reducing the number of FRM mass-only sites from 1,392
to 1,094, EPA was able to shift about $6.5 million to other efforts. About
half, or $3.1 million, of these funds will be used to increase the number



Page 23                     GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix I
EPA’s Actions to Address Concerns With the
PM 2.5 Monitoring Program




and frequency of analyses performed at speciation sites from sampling
once every 6 days to sampling no less often than once every 3 days at 54
speciation sites. Similarly, about $3.4 million of these funds will be used to
increase the number of continuous monitoring sites from the 62 originally
planned to 137 sites. In addition to helping satisfy regulatory program
needs, such as identifying likely PM2.5 sources and assessing control
strategy effectiveness, the more frequent speciation analyses and
additional continuous monitors will help support health effects, exposure,
and atmospheric modeling research efforts, according to EPA officials.

3. The current plans for the speciation of particulate matter once every 6
days will not provide useful data for improving risk assessments

A reduction in the number of planned FRM mass-only sites allowed EPA to
improve other aspects of the PM2.5 monitoring program. For example, EPA
will use about $3.1 million of mass-only funds to increase the frequency of
speciation sampling from 1 sample every 6 days to daily sampling at 10
major urban area sites, and from 1 sample every 6 days to 1 sample every 3
days at 44 other speciation sites. Additionally, agency officials are
considering ways to increase the sampling frequency at another 40
speciation sites located near supersites, with a goal of sampling at least
once every 3 days. With respect to the remaining 208 speciation sites
where samples will be collected once every 6 days, EPA officials recognize
that more frequent data is better, but said that all speciation
data—including samples collected once every 6 days—will still be useful
to the scientific community.

4. EPA should make greater use of continuous monitors to help determine
human exposures and facilitate time-series epidemiological studies

EPA plans to use about $3.4 million of mass-only monitoring funds to
increase the number of continuous monitoring sites from 62 to 137—a
121-percent increase in the number of continuous monitoring sites
planned. According to agency officials, these additional continuous
monitors should help determine human exposures and facilitate
time-series epidemiological studies.




Page 24                     GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix I
EPA’s Actions to Address Concerns With the
PM 2.5 Monitoring Program




5. FRM mass-only samplers will likely not measure some important
parameters (nitrates and organic compounds)

EPA officials said the FRM mass-only monitor’s limitations were recognized
early, but no single sampling device is currently able to accurately
measure airborne fine particulate matter. They noted that many factors
influenced the decision to use the FRM sampler, including the need to
balance sampler cost, ease of use, and ability to produce reproducible
measures that can be quality assured. The FRM mass-only sampler also
produces data most comparable to the historical epidemiological data
base on which the PM2.5 standards are based, and loses fewer constituents
than most other monitors, according to EPA officials. They also noted that
the speciation monitors should measure nitrates and organics, and that the
choice of the FRM mass-only monitor was peer-reviewed and approved by
the Clean Air Science Advisory Committee (CASAC)3 Technical
Subcommittee for Fine Particle Monitoring in August 1996. The agency’s
revised plans call for speciation sites to use three types of filters (teflon,
nylon, and quartz) to speciate nitrates, organic compounds, sulfates,
metals, and other components of the PM2.5 mixture. Additionally, the 108
IMPROVE sites in national park and wilderness areas will use these three
filters.

6. Future research results may indicate that the monitors are not
measuring the most biologically important aspects of particulate matter

EPA officials said that they recognize that future research results may show
that the PM2.5 monitors are not measuring the most biologically important
aspects of particulate matter. However, they believe this determination is
years away, and—as noted previously—their assessment of their 1997
regulation is that PM2.5 mass should be monitored at a minimum of about
850 sites. Additionally, they pointed out that PM2.5 may be a confounding
co-contributor to adverse health effects, and measuring PM2.5 mass may be
the most cost-effective, long-term surrogate measure for a nationwide
network. To help address this Academy concern, as noted previously, EPA
significantly revised the relative mix of sites to better support health,
exposure, and atmospheric modeling research.




3
CASAC is a permanent subcommittee of the Science Advisory Board established by the Clean Air Act
Amendments of 1977.



Page 25                        GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix I
EPA’s Actions to Address Concerns With the
PM 2.5 Monitoring Program




7. Specific objectives, operating conditions, number, type, and location of
monitors should be thoroughly and independently peer-reviewed prior to
implementation

EPA officials said they planned to have these aspects of the PM2.5 monitoring
program independently peer-reviewed prior to implementation. For
example, the basic FRM mass-only monitoring network composed of about
1,094 sites was peer reviewed and approved by the CASAC Technical
Subcommittee for Fine Particle Monitoring in August 1996. More recently,
this subcommittee also peer-reviewed and approved 54 speciation
network trends sites, 108 IMPROVE sites, and 2 test supersites in
November 1998. EPA plans for the remaining 248 sites to be peer-reviewed
before implementation.

8. Interface between monitoring and research on particulate matter is still
largely uncoordinated and fragmented

According to EPA officials, the agency coordinates its PM2.5 monitoring
program internally through a steering committee jointly headed by
representatives of the agency’s program and research offices. Externally,
EPA coordinates its PM2.5 monitoring program with key federal programs
and activities involved in similar work. These include the North American
Research Strategy for Tropospheric Ozone organization, composed of EPA,
National Oceanic and Atmospheric Administration, the Department of
Energy, and over 50 other public and private institutions; the Health
Effects Institute; as well as the IMPROVE steering committee, comprised of
EPA, National Park Service, Forest Service, Fish and Wildlife Service,
Bureau of Land Management, the Western States Air Resources
cooperative, and Northeast States for Coordinated Air Use Management.
EPA also coordinates with the State and Territorial Air Pollution Program
Administrators/Association of Local Air Pollution Control Officials
(STAPPA/ALAPCO) PM2.5 committee, the National Academy of
Sciences/National Research Council’s Committee on Research Priorities
for Airborne Particulate Matter, and the Science Advisory Board’s CASAC
Technical Subcommittee on Fine Particle Monitoring. According to agency
officials, they recognize that coordination with some other federal and
nonfederal agencies and organizations could be improved and have
recently established a formal cross-federal coordination group under the
Committee on Environment and Natural Resources to coordinate EPA’s
PM2.5 monitoring and research activities with the National Institute of
Environmental Health Sciences; the National Institute of Occupational
Safety and Health; the departments of Energy, Defense, Agriculture, and
Health and Human Services; the National Oceanic and Atmospheric


Page 26                     GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix I
EPA’s Actions to Address Concerns With the
PM 2.5 Monitoring Program




Administration; the National Aeronautics and Space Administration; and
others. EPA is still developing procedures to ensure optimization of PM2.5
data collection activities.




Page 27                     GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix II

Objectives, Scope, and Methodology


              The Chairman, Subcommittee on VA, HUD, and Independent Agencies,
              House Committee on Appropriations, asked us to describe (1) EPA’s
              actions in response to the National Academy of Sciences’ concerns with
              EPA’s planned PM2.5 monitoring program and (2) the challenges that state
              and local agencies face in establishing and operating the PM2.5 monitoring
              program, as well as EPA’s response to these challenges.

              To describe the actions EPA has taken in response to key concerns
              identified in the Academy’s March 1998 report, we obtained and reviewed
              EPA’s, states’, and local agencies’ reports, guidance documents,
              memorandums, and financial data showing the changes that were made.
              We also interviewed officials from the Academy; the Science Advisory
              Board’s Clean Air Science Advisory Committee; EPA’s Office of Air Quality
              Planning and Standards; EPA’s Office of Research and Development; and 5
              of EPA’s 10 regional offices; six state and local agencies; and the State and
              Territorial Air Pollution Program Administrators/Association of Local Air
              Pollution Control Officials (STAPPA/ALAPCO) to gain a better understanding
              of the actions the agency has taken.

              To understand the challenges that state and local air monitoring agencies
              face in establishing and operating the PM2.5 program, we interviewed
              officials from five states and the nation’s largest local program, as well as
              their respective EPA regional offices. These state and local programs were
              selected to provide a nationwide representation of the environmental
              conditions under which the PM2.5 monitors would be expected to operate.
              The selected states were California, Georgia, Illinois, New York, and
              Texas, and the local program was California’s South Coast Air Quality
              Management District. The selection of these six state and local programs
              was coordinated with EPA and STAPPA/ALAPCO, both of which agreed before
              we contacted these agencies that they represented a good cross section of
              PM2.5 monitoring program experiences. Collectively, these six state and
              local agencies operate 218 of the 823 monitoring sites that were to be
              deployed by December 31, 1998, or about one-fourth of the sites. We asked
              representatives from these programs about their experiences with the
              monitors, including sampling, sample handling and transport, laboratory
              analysis, and data reporting. We also asked if they had any
              recommendations they believe would improve the program. We also
              interviewed officials from the national organization that represents state
              and local air monitoring programs to gain nationwide perspective on the
              experience that their other members have had with the PM2.5 monitoring
              program. Additionally, we obtained information from several regional
              organizations that represent state and local air monitoring agencies to gain



              Page 28                  GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix II
Objectives, Scope, and Methodology




a regional perspective on PM2.5 air monitoring experiences and issues. We
also interviewed EPA headquarters and regional officials to identify any
ongoing or planned mitigating actions to help state and local agencies
address the challenges they face in establishing and operating their PM2.5
networks.

We conducted our review from October 1998 through June 1999 in
accordance with generally accepted government auditing standards.




Page 29                     GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Appendix III

Comments From the Environmental
Protection Agency




               Page 30   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                 Appendix III
                 Comments From the Environmental
                 Protection Agency




See comment 1.




See comment 2.




                 Page 31                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                 Appendix III
                 Comments From the Environmental
                 Protection Agency




See comment 3.




See comment 4.




See comment 5.




Now on p. 7.
See comment 6.




Now on p. 7.
See comment 7.




                 Page 32                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                    Appendix III
                    Comments From the Environmental
                    Protection Agency




Now on p. 9.
See commment 8.




Now on pp. 9-10.
See comment 9.




Now on pp. 11-12.
See comment 10.




Now on p. 13.
See comment 11.




                    Page 33                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                  Appendix III
                  Comments From the Environmental
                  Protection Agency




Now on p. 13.
See comment 12.




Now on p. 16.
See comment 13.




Now on p. 17.
See comment 14.


Now on p. 17.
See comment 15.




Now on p. 18.
See comment 16.




                  Page 34                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                  Appendix III
                  Comments From the Environmental
                  Protection Agency




Now on p. 26.
See comment 17.




                  Page 35                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
                 Appendix III
                 Comments From the Environmental
                 Protection Agency




                 The following are GAO’s comments on EPA’s letter dated July 8, 1999. They
                 are numbered in accordance with the numbered comments in EPA’s letter.


                 1. This sentence was amended to state that mass-only monitors are
GAO’s Comments   required for determining areas that exceed the PM2.5 standards, since it is
                 EPA’s own regulations that require their use.


                 2. EPA’s views have been added to this section of the report.

                 3. As suggested, we added the term “speciation” to clarify that the more
                 complex and costly monitors that have yet to be deployed are speciation
                 monitors.

                 4. We agreed with this comment and have revised the report accordingly.

                 5. We changed footnote 2 to update the status of court rulings about EPA’s
                 particulate standards and the status of EPA’s appeals.

                 6. This sentence was clarified to show that state and local agencies
                 received $50.7 million of the $65.7 million in PM2.5 monitoring funds for
                 fiscal year 1999 and that EPA’s supersites monitoring effort received
                 $15.0 million of these funds.

                 7. See comment 4.

                 8. See comment 4.

                 9. The statements suggested by EPA have not been included because this
                 section already appropriately addressed these issues.

                 10. This section was clarified to show that state and local agencies
                 nationwide have had operational problems with about 30 percent of the
                 monitors deployed and, for those monitors that have had problems, the
                 types of obstacles that state and local agencies have experienced.

                 11. See comment 4.

                 12. EPA views have been added to this section and attributed to EPA.

                 13. See comment 12.




                 Page 36                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
           Appendix III
           Comments From the Environmental
           Protection Agency




           14. See comment 4.

           15. See comment 4.

           16. EPA’s agreement with our conclusions has been noted in the agency
           comments section of the report.

           17. See comment 4.




(160462)   Page 37                   GAO/RCED-99-215 EPA’s Fine Particulate Monitoring Program
Ordering Information

The first copy of each GAO report and testimony is free.
Additional copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order
made out to the Superintendent of Documents, when
necessary. VISA and MasterCard credit cards are accepted, also.
Orders for 100 or more copies to be mailed to a single address
are discounted 25 percent.

Orders by mail:

U.S. General Accounting Office
P.O. Box 37050
Washington, DC 20013

or visit:

Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting Office
Washington, DC

Orders may also be placed by calling (202) 512-6000
or by using fax number (202) 512-6061, or TDD (202) 512-2537.

Each day, GAO issues a list of newly available reports and
testimony. To receive facsimile copies of the daily list or any
list from the past 30 days, please call (202) 512-6000 using a
touchtone phone. A recorded menu will provide information on
how to obtain these lists.

For information on how to access GAO reports on the INTERNET,
send an e-mail message with "info" in the body to:

info@www.gao.gov

or visit GAO’s World Wide Web Home Page at:

http://www.gao.gov




PRINTED ON    RECYCLED PAPER
United States                       Bulk Rate
General Accounting Office      Postage & Fees Paid
Washington, D.C. 20548-0001           GAO
                                 Permit No. G100
Official Business
Penalty for Private Use $300

Address Correction Requested